Modern Arc Welding Technology, 2nd Edition (2018)

Modern Arc Welding Technology
This Page is Intentionally Left Blank

Modern Arc
Welding
Technology
imililil
Oxford & IBH Publishing Co. Pvt. Ltd.

New Delhi
(A Unit of CBS Publishers & Distributors Pvt Ltd)
1A
CBS
CBS Publishers & Distributors Pvt Ltd
New Delhi • Bengoluru • Chennai ♦ Kochi • Kolkata • Mumbai
Hyderabad • Jharkhand • Nagpur • Patna • Pune • Uttarakhand
Disclaimer
Science and technology are constantly changing fields. New research and experience broaden the scope of
information and knowledge. The authors have tried their best in giving information available to them while
preparing the material for this book. Although, all efforts have been made to ensure optimum accuracy of the
material, yet it is quite possible some errors might have been left uncorrected. The publisher, the printer and
the authors will not be held responsible for any inadvertent errors, omissions or inaccuracies.
eISBN: 978-93-881-7830-3
Copyright © Authors and Publisher
Second eBook Edition: 2018
All rights reserved. No part of this eBook may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage and retrieval
system without permission, in writing, from the authors and the publisher.
Published by Satish Kumar Jain and produced by Varun Jain for

CBS Publishers & Distributors Pvt. Ltd.
Corporate Office: 204 FIE, Industrial Area, Patparganj, New Delhi-110092
Ph: +91-11-49344934; Fax: +91-11-49344935; Website: www.cbspd.com; www.eduport-global.com;
E-mail: [email protected]; [email protected]
Head Office: CBS PLAZA, 4819/XI Prahlad Street, 24 Ansari Road, Daryaganj, New Delhi- I I 0002, India.
Ph: +91-11-2328925 9, 23266861, 23266867; Fax: 011-23243014; Website: www.cbspd.com;
E-mail: [email protected]; [email protected].
Branches
Bengaluru: Seema House 2975, 17th Cross, K.R. Road, Banasankari 2nd Stage, Bengaluru - 560070,
Kamataka Ph: +9 l -80-26771678/79; Fax: +91-80-26771680; E-mail: [email protected]
Chennai: No.7, Subbaraya Street Shenoy Nagar Chennai - 600030, Tamil Nadu
Ph: +91-44-26680620, 26681266; E-mail: [email protected]
Kochi: 36/ 14 Kalluvilakam, Lissie Hospital Road, Kochi - 682018, Kerala
Ph: +91-484-4059061-65; Fax: +91-484-4059065; E-mail: [email protected]
Mumbai: 83-C, pt floor, Dr. E. Moses Road, Worli, Mumbai - 400018, Maharashtra
Ph: +91-22-24902340 - 41; Fax: +91-22-24902342; E-mail: [email protected]
Kolkata: No. 6/B, Ground Floor, Rameswar Shaw Road, Kolkata - 700014
Ph: +91-33-22891126 - 28; E-mail: [email protected]
Representatives
Hyderabad
Pune
Nagpur
Manipal
Vijayawada
Patna
Preface
Welding is a small but crucial part of metallurgy i.e. the science of discovering new metals and working effi-
ciently with them. The welding specialist has to have an intimate knowledge of the properties, structure and
behaviour of each metal as also new alloys and exotic variants for specific industries and applications. When
metallurgy moves to the next phase of metalworking there are many skills and processes that need to be mas-
tered. This is why in the middle ages there were no books but there were guilds where the 'masters' taught
'know-how' through a process of 'show-how'. Today's equivalent is the knowledge volume in hard copy (book)
form or digital storage.
Adorians have for over 50 years helped to discover, process and share new horizons in metallurgy, metal-
working, welding and cutting. There was a time when this company was literally a parent institution for the
nascent growth and development of welding science and applications in India. This earned for the company the
merited encomium:
'Welders To The Nation'.
Today there are many nuclei of welding knowledge where "excellence" is the quest and "improvisation" the
password for cost effectiveness and value addition. Over the years, the ADOR Group has always understood the
trends and cost drivers of the industry and has re-configured its contributions to meet these needs.
The coilntry has a huge manufacturing base which spans a wide spectrum: from petrochemicals to iron &
steel, power, construction & cement, mining, transport, defence, nuclear, food processing and telecom. India
also has the largest reservoir of technical and trained human resources in the world. Welding is the crucial bond
between development & future growth and welding professionals can therefore look forward to playing a major
part in the global fabrication sector.
This revised and updated edition of India's first and only Handbook on Welding could not have come at a
more appropriate time. It is both a tribute and a sequel to Mr. S. V. Nadkarni's original efforts to capture India's
welding expertise in a body of knowledge. This fills the gap between generations of people, processes, products
and welding expertise.
As the leaders of India's welding industry, ADOR WELDING LIMITED fulfils its avowed mission of grow-
ing the knowledge base and encapsulating the experiential quotient of an industry that is vital to India's future
place in the industrialized nations of the world.
Contents
Preface V
1. Introduction To Welding And Allied Processes 1

2. Power Sources For Arc Welding 31
3. Manual Metal-Arc Welding 57
4. Submerged Arc Welding 187
5. Tungsten Inter-Gas Arc Welding (TIG Welding) 261
6. Metal Inter-Gas/C0 2 Arc Welding 315
7. Flux-Cored Arc Welding 357
8. Electroslag and Electrogas Welding 377
9. Welding Metallurgy 391
10. Weldability of Metals 427
11. Hardfacing by Welding 511
12. Welding Defects: Their Causes and Prevention 533
13. Testing and Inspection of Welding 555
14. Metal Cutting Processes 599
15. Welding Costs and Economics 609
16. Safety Requirements in Arc Cutting and Welding 623
17. General Hints on Welding Design 637
18. Welding Procedure Specifications 657
19. Welding Applications 681
20. Preheat and Postweld Heat Treatment 771
21. Mechanised Arc Welding 783
22. Information Technology (IT) in Welding 789
Glossary 793
Index 801
Introduction To
Welding And
Allied Processes
WELDING is recognised all over the world, today, as a would not have been possible without the use of
remarkably versatile means of metal fabrication. Weld- welding.
ing, in combination with allied processes like thermal The technique of joining metal pieces by welding
cutting, brazing, brazewelding and metal spraying has apparently originated more than 1,500 years ago, when
provided ample freedom to the modern designer to de- man learnt the art of manufacturing wrought iron,
velop metallic products in which optimum mechanical heating iron pieces to dull red and then hammering them
properties, lightness and aesthetics are harmoniously together. This became the traditional hammer-and-tong
blended. process of the village blacksmith, which today is known
A large number of welding and allied processes have as forge-welding. The 1,500-year-old Iron Pillar at the
come into industrial use in the last 30 years. Variations Qutab Minar in Delhi is made of pure iron ingots
and extensions of these processes are being developed forgewelded together.
and put to practical use from time to time. Thus the An amazing piece of early welding can be seen in
metal-joining technology is in a constant state of flux. the 13th century temple at Konarak in Orissa. The flat
The wide range and variety of these processes enables stone roof of the Black Pagoda there is supported on
the modern engineer to join almost all commercial welded beams; each beam is 25 cm2 and nearly 6 metres
metals and alloys in many different shapes and sizes long, and is fabricated from 75 mm blocks of iron by
and in thicknesses ranging from a fraction of a millimetre welding.
to over 500 mm. The well-known Damascus swords and many other
Plain carbon steels, low-alloy steels, stainless and specimens of ancient and medieval swords that are seen
heat-resisting steels, and many non-ferrous metals and in museums today were produced by hammer welding
their alloys (such as of aluminium, nickel and copper) strips of high carbon steel to low carbon steel, forging
and metals such as titanium, columbium, molybdenum the composite metal, doubling it upon itself, again
and zirconium are w e l d e d extensively. M o d e r n welding and forging, and repeating these steps until a
industrial products such as automobiles, railway rolling finely laminated strip was obtained, from which the
stock, transmission pipelines, pressure vessels, storage blade was forged and ground. Such a weapon could be
tanks, machinery, power generation equipment, offshore hardened and given a sharp cutting edge, but at the same
platforms, oil refineries and fertiliser plants, which have time it was ductile enough to resist brittle fracture in
contributed to the material prosperity of mankind, owe combat.
their development and efficient performance to welding. In the western countries, especially in the U.S.A., the
In the last few years, m a n y of the new high- U.K. and West Germany, the application of welding to
t e m p e r a t u r e metals and superalloys h a v e been the mass production of industrial items and war
fabricated by welding into critical components required equipment dates only from the second decade of the
for guided missiles, jet aircraft and nuclear power plants. present century. Welding received special impetus
These and other similar recent spectacular developments during World War II, when the urgency of providing
2 MODERN ARC WELDING TECHNOLOGY
large fleets of ships, tanks, trucks and aircraft at a short formation of a weld, the abutting surfaces grow together
notice led to the large-scale replacement of riveting by into a common mass, so that the joint has metallic
welding. During the following decades, the use of continuity. The ideal weld is one in which there is perfect
welding multiplied with amazing swiftness and today continuity between the parts joined, such that every part
this method of joining metals has a dominant place in of the joint is indistinguishable from the metal in which
modern life. the joint is made.
Having briefly touched upon the history of welding, This ideal is never achieved in practice, because the
let us revert to the present-day welding and allied microstructure and mechanical properties are not
processes. These can be conveniently grouped under identical over the entire joint and the original metal. Yet,
seven main headings as follows, each suggesting the welds which give satisfactory service can, and are being
technological principle on which they are based:
made, using the proper welding process to suit each
• Fusion welding metal, type of joint and application. The skill of the
• Electric resistance welding welding engineer consists in the recognition of the
• Solid phase welding essential requirements which a particular weld must
• Braze welding satisfy and the choice of appropriate welding process,
and procedure which will meet them.
• Brazing
• Thermal cutting and FUSION WELDING PROCESSES
• Thermal spraying These processes involve fusion of the base metal to
Each of these can be further divided into sub-groups complete the weld. Fusion welds ordinarily do not
and their variations and extensions. All of them have been require the application of pressure, and they may be
presented at a glance in the Ador Welding Limited process completed with or without the addition of filler metal.
chart (Table 1.1). A fusion weld made without the addition of filler metal
Before starting to introduce this process chart, it is is called autogenous weld. In most cases, fusion welding
necessary to explain the terms weld and welding. The involves the use of filler metal.
American Welding Society (AWS) defines weld as a In the process chart (Table 1.1), fusion welding
localised coalescence of metals or non-metals produced processes and, among them, electric arc processes have
either by heating the materials to suitable temperatures, been given the first place. This is justified by the fact
with or without the application of pressure, or by the that they are used to the largest extent in industry today.
application of pressure alone, and with or without the
use of filler material. Indian Standard IS: 812-1957 Metal-Arc Welding
defines the weld as "a union between two pieces of a This is a group of processes in which the heat required
metal at faces rendered plastic or liquid by heat or by for fusion is generated by the electric arc formed between
pressure, or both. a metallic electrode and the base metal. The electrode is
consumed in the arc and provides the filler metal for
Filler metal may be used to effect the union."
the joint.
International Organisation for Standardisation (ISO) has
The electric arc is an ideal source of welding heat.
defined welding as "an operation by which two or more
The extremely high arc temperature of over 5000°C
parts are united, by means of heat or pressure, or both,
permits it to supply a large amount of heat to a small
in such a way that there is continuity of the nature of
area. Hence the welding speed is very high, and any
the material between these parts. A filler material, the metallurgical disturbance in the base metal is restricted
melting temperature of which is of the same order as to a narrow zone.
that of the parent material, may or may not be used." Among the arc processes, manual metal-arc welding
Welding is defined by the AWS, simply as a materials is the most common, versatile and inexpensive one and
joining process used in making a weld. accounts for over 50% of the total welding in advanced
To elaborate, a weld is a localised coalescence or countries and over 80% of the total welding in India. It
union of metals. The word coalescence comes from the is a manual process, and hence depends on the skill and
verb to coalesce, whose dictionary meaning is, to grow experience of the welder. It makes use of a flux-coated
together or unite into one body. This implies that in the electrode having a core of solid metallic wire (diameter
Table 1.1
PROCESS CHART
By S. V. Nadkarni
WELDING, CUTTING, AND ALLIED PROCESSES
FUSION WELDING ELECTRIC

1 BRAZEWELDING
1 THERMAL THERMAL
SOLID PHASE BRAZING
RESISTANCE CUTTING SPRAYING
BLACKSMITH
HAMMER
ARC GAS THERMIT ELECTRON ELECTROSLAG LASER - FORGE - _E ROLL FLAME ELECTRIC PLASMA
BEAM | GAS PRESSURE ARC
CONSUMABLE | — HIGH FREQUENCY PRESSURE
GUIDE
COLD
DIFFUSION BONDING
1 1 1 1 1 1 EXPLOSIVE
METAL CARBON TIG ARC ARC ATOMIC ELECTROGAS PLASMA
FRICTION
ARC ARC SPOT STUD HYDROGEN ARC
I—FRICTION STIR
I
MICRO PLASMA — ULTRASONIC
' — MAGNETIC PULSE
MANUAL SUBMERGED MIG MAG

i r i
FLUX CORED GRAVITY FIRECRACKER
METAL ARC ARC (ccy J
AUTO CONTACTOR
RESISTANCE DIP FLAME FURNACE INDUCTION SALTBATH DIFFUSION MIG
SYNERGIC PLASMA GAS SHELF
MIG MIG SHIELDED SHIELDING
1 I I T^
RESISTANCE FLASH SPOT SEAM PROJECTION PERCUSSION HIGH FREQUENCY OXYGEN OXYGEN ARC SPARK ELECTRON LASER
BUTT | | RESISTANCE ARC EROSION BEAM
STITCH MULTIPLE SERIES ROLLER FLAME POWDER LANCING

SPOT SPOT SPOT
BUTT SEAM CARBON METAL n 1

ARC ARC AIR PLASMA
ARC ARC
FOIL BUTT SEAM
4 MODERN ARC W E U *«G TECHNOLOGY
6.3-1.6 mm, length 450-250 mm). 11 needs a powc source, automatically into the arc at a constant speed. The arc is
either a transformer supplying A " or a generator/ covered with a layer of dry granular flux which performs
rectifier /inverter supplying DC (see fig. 1.1). the same functions as the coating of a manual electrode.
The process is extremely flexible and suitable for the The arc length is automatically controlled. The power
entire range of plate thicknesses, and for almost all source can be a transformer (AC) or a generator/rectifier
commercial metals and alloys. It is used for joining (DC). It is generally of very high capacity, say 750 or
as well as for surfacing (rebuilding). It can be used even up to 3,000 amps (see Fig. 1.2).
in all welding positions. It is well-suited for site In semi-automatic welding, the operator guides the
welding. flexible welding head along the seam. In doing so he
controls the speed of travel and the line of travel.
Submerged-Arc Welding Controlling the line is difficult, because the joint is
This is mostly used as a fully-automatic process, and submerged by flux. Hence this version is usually
sometimes as a semi-automatic process. restricted to fillet welds and grooved butt welds.
The electrode is a continuous metallic wire (solid or In fully-automatic welding, the welding head is
flux-cored) in the form of a spool or a coil. It is fed mounted on a trolley which travels along the joint.
ELECTRODE COATING
POWER SOURCE CORE WIRE
SHIELDING ATMOSPHERE
ELECTRODE HOLDER ARC POOL
SOLIDIHED SLAG
ELECTRODE ARC STREAM
aa^M
ARC — PENETRATION DEPTH
GROUND CABLE BASE METAL

(A)
ELECTRODE CABLE (B)
Fig. 1.1: Manual metal-arc welding: (A) Welding circuit; (B) Welding arc in action.
TO AUTOMATIC WIRE FEED
TO WELDER POWER
i. -TO FLUX HOPPER
WELDING VEE
BA
WELDING WIRE SE METAL
GROUND CONNECTION
WELD TRAVEL
WELD BACKING PLATE
Fig. 1.2: Diagrammatic sketch of submerged arc welding

INTRODUCTION TO WELDING AND ALLIED PROCESSES 5
Alternately, the welding head is stationary, and the joint Four distinct techniques are possible in MIG welding,
moves under it. depending on electrode size, shielding gas, arc voltage
This process gives very high productivity and and welding current:
excellent weld quality. a) Spray mode of metal transfer
It is generally applied on mild steel, high tensile b) Globular mode of metal transfer
steels, low-alloy and stainless steels. It is ideal for heavy c) Short-circuiting
thicknesses. For thin sections, weld backing is necessary
to avoid burn-through. It is commonly used for the d) Pulsed transfer
fabrication of plate girders, pressure vessels, pipes and a) This technique uses heavy wires, large currents
penstocks, for surfacing and strip cladding. In strip and preferably argon-oxygen mixture. This gives high
cladding, the electrode is in the form of a strip, usually heat input, high deposition rate and deep penetration.
1.6 mm x 75 mm wide. It is preferred for heavy gauge welding.
b) When current is low, metal transfer occurs in large
MIG Welding globules. The result is poor arc stability, shallow pen-
MIG welding is an abbreviation for metal-inert-gas etration and excessive spatter. Hence this technique is
welding. It is also known as GMAW, which stands for not used.
gas-metal-arc welding. In this process, coalescence is c) This technique is the most practical at currents
achieved by an electric arc formed between the below 200 amps and with fine wires (0.8-1.2 mm diam-
workpiece and a continuous consumable solid wire eter). It permits welding thinner sections with ease, and
electrode which is fed through a gun at controlled is extremely practical for welding in all positions.
speeds. Inert gas usually argon or argon mixture flows d) Pulsed transfer is a controlled method of spray
through the gun and forms a blanket over the weld transfer welding requiring more sophisticated power
puddle to protect it from atmospheric contamination. source whereas the three types of transfers described
The welding can be semi-automatic or fully previously can be obtained with standard power sources
mechanised. In the semi-automatic version, the welder and wire feed units. This mode of transfer can be used
concerns himself only with gun-to-work distance, gun on mild and low alloy steels, stainless steel and is par-
manipulation, and welding speed. Wire-feed rates, ticularly useful with aluminium and alloys on light to
electrical settings, and gas flow are pre-set. When the medium plate sections.
equipment is completely mechanised all of these MIG is a versatile process, and is gradually replacing
variables and welding functions are performed manual metal-arc and TIG welding. Most metals can be
automatically without the need for a welder. easily welded including aluminium, carbon steels, low-
The power source is a rectifier or motor generator alloy steels, stainless steels, nickel, copper, magnesium,
giving DC. Reverse polarity is generally used (electrode titanium and zirconium. However, for carbon steels and
positive), as it gives better melting, deeper penetration
low-alloy steels, MAG or C0 2 welding is often preferred,
and better cleaning action (see Fig. 1.3).
because it avoids the use of expensive argon gas.
In welding ferrous metals (i.e. carbon steels, low alloy
steels and stainless steels), pure argon shield gives erratic
NOTE: SOMETIMES A WATER metal transfer and shallow penetration/side-wall fusion
CIRCULATOR IS USED ^ levels and hence certain gas combinations are used, e.g.:
GAS
1) Ar + 1-2% 0 2 (slightly oxidizing) for alloy steels
SUPPLY and stainless steels
SHIELDING GAS 2) Ar + 3-5% 0 2 or 5-8% C0 2 for carbon or low alloy
U steels in lower thickness range
3) Ar + 15-20% C0 2 for carbon or low alloy steels in
■CONTROLS FOR
higher thickness range
(GOVERNING WIRE
pRIVE CURRENT
FLOW AND COOUNGj MAG (C0 2 ) Welding
WELDING
MACHINE CONTACTOR
frATER IF FUSED
MAG stands for metal-active-gas arc welding. This
is a variation of MIG welding, in which an identical
equipment is used but the inert gas is replaced by carbon
Fig. 1.3: Schematic diagram of MIG/MAG (C02) welding dioxide, which is chemically active. The term gas-metal-
arc welding (GMAW) is also applied to the MAG (C0 2 ) Some flux-cored wires are specially designed to be
process. self-shielding, which means they do not require the
C 0 2 welding has a very hot coarse arc associated with external C 0 2 shield. Such wires are widely used in the
relatively high level of spatter formation. However, U.S.A. a n d J a p a n for s t r u c t u r a l a n d h a r d f a c i n g
because of excellent fusion and wetting properties, this applications. When such wires are used, the process is
provides a lower cost solution for good quality welding, described as self-shielding or gasless.
where spatter can be tolerated.
C 0 2 welding is gradually replacing manual metal- Gravity Welding
arc welding in the fabrication of structurals, pipes, This is a semi-automatic version of manual metal-arc
automotive products, storage tanks, machinery, etc. welding. The bare end of a specially designed heavy-
coated manual welding electrode (diameter 4.5-6.3 mm,
Plasma-MIG Welding length 600-700 mm) is clamped in a holder, which is fixed
This process has been dealt with later under the on one of the legs of a tripod. The striking end of the
heading Plasma Processes. electrode is placed in the root of a tee joint. When the arc
is struck, the electrode begins to consume and the arc
Flux-Cored Arc Welding length is automatically maintained due to the sleeve
This is an extension of the MIG/MAG process. The formation at the burning end, as in contact welding. At
continuous solid wire is replaced by a tubular wire the same time, the holder slides down by gravity. The
whose core is filled with flux. The equipment is the same electrode automatically retracts and cuts off the arc when
as used for MIG/MAG welding (see Fig. 1.4). a 50 mm long stub is left (see Fig. 1.5).
The flux performs the same functions as the coating
of a manual electrode. For welding alloy steels and for
hardfacing applications, suitable alloying elements are
also included in the flux. Flux-cored wires provide an
easy means of producing alloyed and hardfacing wires
of any desired composition, in small batches and at short
notice. They are a big boon to the industry, since for solid
wire one has to depend on a steel producer, rolling mill
and a large production batch.
Flux-cored arc welding is normally performed with
a C 0 2 shield to protect the weld pool from atmospheric
attack. Thus it becomes a gas-shielded process.
Fig. 1.5: Diagrammatic sketch of gravity welding
fa
DIRECT CURRENT
CONSTANT VOLTAGE TO SOLENOID VALVE FLUX CORED ELECTRODE
POWER SOURCE CURRENT CARRYING ^ ^
\ VOLTAGE CONTROL IWDERED METAL VAPOUR
CONTACT T U B E - , f n
SHIELDING FORMING MATERIALS
Q ^ n " CONTACTOR CONTROL SOLIDIFIED SLAG - | ^<L
GAS SOURCE DEOXIDISERS
VOLTMETER AND
AMMETER
N LJ/ AND SCAVENGERS
GAS SHIELD FORMED FROM
CORE MATERIALS
ARC AND METAL TRANSFER

j * - GAS OUT
GROUND CABLE ^
Fig. 1.4: Flux-cored arc welding: (A) Schematic diagram of the process; (B) Welding arc in action
The fillet size (throat thickness) ranges between of the same length as the electrode. The copper block helps
3.5mm and 5.5 mm, the length of weld per electrode being to prevent overheating of the electrode and to maintain
in the range of 600-1,000 mm. One operator can handle intimate contact between the electrode and the joint.
four tripods at a time. The process is fairly popular in
shipbuilding, where it is used for depositing long and Auto-Contact Welding
continuous fillet welds. It can be used with equal This process combines the principles of gravity
advantage in other similar structures. In Japan, this pro- welding and firecracker welding. A gravity welding
cess has been successfully used for underwater welding. electrode is clamped in the holder of an auto-contact
welding device. In this device, a spring actuated by a
Firecracker Welding lever pushes the electrode to the welding line at a low
This is also a semi-automatic version of MMA angle (10° to 50°). The device is held against the
welding. A specially designed heavy-coated electrode, workpiece by a permanent magnet fixed in the device.
which can be of any length up to 2 metres is laid on the The arc is cut off automatically with a stub length of 60
seam of a grooved butt joint or along the root of a tee m m by a simple trip mechanism. The process is
joint. It is then clamped down with a water-cooled commonly used in shipbuilding (see Fig. 1.7).
copper bar of heavy square section, which is grooved to
accommodate the electrode. The copper bar is nearly as Carbon Arc Welding
long as the electrode. The bare end is clamped in a holder In this process, an arc is struck between a non-
(see Fig. 1.6). consumable carbon electrode and the base metal or
The arc is struck at the striking end, and the electrode between two carbon electrodes, using specially designed
gradually consumes by itself, making the butt or fillet weld electrode holders. No shielding is used. Pressure and
filler metal may or may not be used. The process has
been used for brazing operations. It has almost become
obsolete with the introduction of atomic hydrogen
welding and TIG (tungsten-inert-gas) welding.
TIG Welding
TIG is an abbreviation for tungsten-inert-gas. The
process is also termed as gas-tungsten-arc welding and
designated as GTAW.
In this process, an arc is struck between a non-
consumable tungsten electrode and the base metal. The
arc is shielded by the inert argon or helium or argon-
helium mixture (see Fig. 1.8).
A filler wire may or may not be used. When it is used,
Fig. 1.6: Diagrammatic sketch of firecracker welding it is fed externally into the arc in the form of rod or strip
Fig. 1.7: Sketch of auto-contact welding device

electric arc between an electrode and the work. The weld

NOTE: SOMETIMES A WATER is made without preparing a hole in either member. Filler
CIRCULATOR IS USED
metal or a shielding gas or flux may or may not be used
(see Fig. 1.9).
The arc is produced by using either of the carbon
arc, manual metal arc, TIG and MIG processes. Arc spot
welding by the MAG (C02) process is widely used today,
for which the equipment is provided with the necessary
controls to achieve consistent spot welds.
While electric resistance spot welding requires access
from both sides of the overlapping plates, arc spot
welding can be made from one side only.
Arc Stud Welding

The word arc is often omitted, and the process is
Fig. 1.8: Schematic diagram of TIG welding simply called stud welding. This is an arc welding
process in which the arc is struck between a metal stud
by the welder. The welder also has to control the arc length or similar part and the base metal. The arc heats the
and arc travel speed. An AC power source is used for mating ends to a proper temperature after which they
welding aluminium and its alloys, while a DC source is are brought together under pressure (see Fig. 1.10).
used for all the other metals. This is an ideal process for The process has two versions:
welding non-ferrous metals and stainless steels in limited
thicknesses. TIG is also preferred for depositing the root
pass in the pressure piping, where welding from inside
is nonpossible. 3 ^SUPPLY RECTIFIER
The most commonly used gas for TIG welding is
argon which can be used on all metals. Helium-argon
mixture give deeper penetration, greater heat input and
hence faster welding. Pure helium can be used for
welding aluminium and copper alloys.
The use of pulsed current greatly extends the control
of the process allowing improved consistency, ability to
join thick to thin material and low thickness joints
without risks of burn through.
Arc Spot Welding CONTROL CABLE

In this process, coalescence at the overlapping
surfaces is produced in one spot by heating with an Fig. 1.10: Schematic diagram of arc stud welding
V
ELECTRODE Arc stud welding and capacitor discharge welding
• Arc Stud Welding: In this case, the welding end
of the stud is recessed and contains flux which is meant
A to stabilise the arc and act as a deoxidising agent. The
OVERLAPPING stud, held in a portable pistol-shaped tool called a stud-
ARC PLATES
gun, is positioned by the operator. He then actuates the
unit by pressing the trigger switch. The welding time
//MV and the final driving home of the stud to complete the
^ ] weld are controlled automatically by a timing device. A
L ceramic ferrule is used with each stud. The ferrule con-
\ SPOT WELD
Fig. 1.9: Diagram of arc spot welding centrates the heat, prevents influx of air to the molten
metal and confines the molten metal to the weld zone.
• Capacitor Discharge Welding: In this case, the stud Electrogas Welding

has a small cylindrical tip on its welding end and the arc Electrogas welding is an extension of the M I G / C 0 2
is produced by a rapid discharge of stored electrical en- process and is designed for single-pass vertical welding
ergy with pressure applied during or immediately fol- of steel plates in the thickness range of 10-38 mm. The
lowing the electrical discharge. It employs a specially joining operation consists of lining up two square butt
designed air gun. The gun is equipped with a collect edges with a 12.5 mm gap, regardless of plate thickness.
attached to the end of a piston rod which holds the stud. The wire electrode is introduced downward into the
Air pressure keeps the stud away from the plate until the cavity formed by the two plates to be joined and two
weld is to be made. At the proper time, air pressure drives fixed water-cooled dams or chill blocks. The cavity
the stud against the workpiece. As soon as the tip of the is kept free of air by the shielding gas, which usually
stud touches the workpiece, a high-ampere, low-voltage is a m i x t u r e of argon a n d carbon dioxide. Wire
discharge takes place. The current creates an arc which diameter may be 1.6, 2 or 2.4 mm, depending on the
melts the entire face of the stud and a similar area of the thickness of the plate.
work. The stud is then driven at a high velocity into the The welding head is suspended from an elevator
molten pool. mechanism, which provides automatic control of the
v e r t i c a l t r a v e l s p e e d d u r i n g w e l d i n g . This
Atomic Hydrogen Welding mechanism raises the welding head automatically
In this process, an arc is struck between two tungsten at the same rate as the advancing weld metal. The
electrodes using AC. Streams of hydrogen gas are passed welding head is self-aligning and can adjust to any
from orifices around the electrodes into the arc. Here misalignment in plate or joint. Once the equipment
the molecules of hydrogen dissociate into atoms with is positioned on the joint, welding is completely
the absorption of heat from the arc. The atoms recombine automatic. Wire feed and current are constant. At the
into molecules at a point a few millimetres away from end of the weld the process stops automatically (see
the arc and liberate intense heat. It is this point which is Fig. 1.12).
used for melting the base metal (see Fig. 1.11). This welding technique is increasingly used in
Thus, in this process, the source of heat is not the arc shipyards and in the fabrication of storage tanks and
but the hydrogen gas acting as the carrier of heat. Filler large diameter pipes.
metal may be added externally if required.
Since hydrogen is present in the arc, this process can Plasma Arc and Micro-Plasma Arc Welding
cause hydrogen-induced cracking in hardenable steels This has b e e n dealt w i t h later u n d e r Plasma
and in heavy, restrained joints in normal steels, and Processes.
porosity in high-sulphur, free-machining steels. Hence,
it has been largely replaced by the TIG process. Gas Welding
In this process, the melting of the base metal is
achieved by means of a gas flame which derives its
intense heat from the combustion of a fuel with oxygen.
The most commonly used fuel is acetylene, though
Mapp gas (stabilised methylacetylene propadiene) and
hydrogen are sometimes used. Hence the appropriate
term of this process is oxyfuel or oxyacetylene welding.
X Y TRAVEL Filler metal may or may not be used (see Fig. 1.13).
The oxyacetylene process depends on the chemical
W0RK
&L Vk SURFACE reaction which occurs in two stages.
ARC +^k&) /
Primary stage:
POSTHEATING t f t PREHEATING ZONE C2H2 + 02 = 2 CO + H2
(GOOLING) ZONE MAX HEAT INPUT acetylene oxygen carbon hydrogen
(FUSION) ZONE monoxide
The primary combustion provides the actual flame
Fig. 1.11: Diagrammatic sketch of atomic hydrogen for welding, with temperature up to 3,092° C, which is
welding
in the inner cone.
w FLUX-CORED ELECTRODE
BASE PLATE 1
DRIVE ROLLERS
GAS SHIELDING
MOLTEN WELD METAL
SOLIDIFYING WELD METAL

WATER CIRCULATION
SOLIDIFIED METAL
WATER CONNECTIONS
COMPLETED WELD
BASE PLATE 2
Fig. 1.12: Diagrammatic sketch of electrogas welding
GAS REGULATORS
HOSES
WELDING TORCH
(A)
REDUCING
GAS SUPPLY REGULATORS
OXYGEN TORCH 8 MIXER
COMBUSTIBLE
GAS
Fig. 1.13: Oxyfuel welding: (A) Schematic diagram of the process; (B) Sketch of the welding torch
Secondary stage: part of aluminium by weight is locally heated with a

2 CO + H2 + m o2 == 2 C 0 2 + H 2 0 special ignition powder, a vigorous reaction takes place
carbon hydrogen oxygen carbon water which proceeds rapidly through the mass, resulting in
monoxide from air dioxide vapour the formation of aluminium oxide and iron, and a
considerable amount of heat. The heat is sufficient to
This secondary combustion occurs at the outer melt the iron and the oxide slag.
portion of the flame. It protects the molten puddle from In carrying out a thermit welding operation, the
attack by air and helps to preheat the base metal. thermit mixture is placed in a refractory crucible above
Equipment for oxyacetylene welding consists of the pieces to be welded. In practice, additional metals
oxygen and acetylene cylinders, pressure regulators and compounds are often placed in the thermit mixture
which reduce the high cylinder gas pressure to the to alloy the iron and improve its properties. The molten
required working pressure, a torch where the two gases metal from the thermit reaction in the refractory crucible
are mixed, and hoses which connect the regulators to is guided to the joint to be welded by a sand mould,
the torch. which is fastened around the work. By virtue of its
Gas welding has limited application for industrial superheat, the thermit metal melts a portion of the base
production purposes, because it is much slower than metal with which it comes in contact. Upon solidification
arc welding processes. It is used considerably more for of this melt, the weld is completed (see Fig. 1.14).
general maintenance work including hardfacing, The process is commonly used for welding the rails
welding metals of low melting points (especially non- of railway tracks. It is also sometimes used to repair
ferrous metals) and performing such operations as heavy broken parts such as steel mill rolls.
brazing, soldering, and thermal spraying.
Electron Beam Welding
Thermit Welding In this process, fusion is achieved by focusing a high
This process utilises the intense heat developed power density beam of electrons on the area to be joined.
during the reaction between iron oxide and aluminium. Upon striking the metal, the kinetic energy of the high
When a mixture of three parts of iron oxide and one velocity electrons changes to thermal energy, causing
CRUCIBLE CAP
IGNITION POWDER
THERMIT
CRUCIBLE LINING
(A)
CRUCIBLE SHELL
SLAG SEAL ASBESTOS SEAL

THIMBLE
TAPPING PIN
SLAG TROUGH
POURING GATE
(B)
REFRACTORY GATE
RISER
MOULD BOX
— PREHEATING GATE
PREHEATING GATE —" SPINDLE TO BE
WELDED
WAX DRAIN — j g ? S PREHEATING GATE

- g ifrA*V»
Fig. 1.14: Diagrammatic sketch of thermit welding : (A) Refractory cubicle in which thermit reaction occurs;
(B) Sand mould containing the joint
the metal to melt and fuse. The electrons are emitted from Electron beam welding can join materials ranging
a tungsten filament heated to approximately 2,000°C. from thin foils to sheets of 50 mm thickness. It is
Since the filament would quickly oxidise at this particularly suited for the welding of refractory metals
temperature if exposed to normal atmosphere, the unit such as tungsten, molybdenum, columbium, tantalum,
as well as the focusing devices and workpiece are placed and metals which oxidise readily, such as titanium,
in a vacuum (see Fig. 1.15). beryllium and zirconium. It also has wide application
The major advantages of electron beam welding are: in joining dissimilar metals, aluminium, standard steels
• Its ability to produce deeper and narrower welds and ceramics.
with total heat input which is much lower than An electron beam welder is very expensive, but its
in arc welding use is justified for critical applications such as bimetallic
• Good control over weld penetration and other blades for metal-cutting saws, aircraft engine
weld dimensions components and supercharger impellers for diesel
engines.
• High-purity welds free from oxides and nitrides, Lately, non-vacuum electron beam welders have
and been developed, which give higher production rates at
• Very high welding speeds and production rates. lower costs, because no vacuum chamber is needed for
HIGH VOLTAGE RECEPTACLE
ELECTRON BEAM GUN
COLUMN HINGE
COLUMN VENT VALVE
ALIGNMENT ADJUSTMENT
SCREW
MAGNETIC LENS
OPTICAL VIEWING SYSTEM

DEFLECTION COIL
WATER CONNECTIONS
TO VACUUM SYSTEM
CHAMBER
Fig. 1.15: Schematic diagram of electron beam welding

the work and no time is lost in pumping down for each The weld metal usually consists entirely of coarse or
load. Also work size is not limited by chamber fine columnar dendrites, or of two zones: one zone of
dimensions. The non-vacuum technique, however, gives coarse columnar dendrites and the other of fine
reduced penetration and the welds are not as narrow columnar dendrites. Electroslag joints made correctly
and non-tapering as those made in a vacuum. in structural steel applications do not need postheating.
On the other hand, joints in pressure vessels which must
Electroslag Welding have optimum notch toughness, require a normalising
This process, which is similar to the electrogas treatment to refine and reorient the grains.
process described earlier, is primarily designed for With variations in technique, this process can be
making a butt weld in the vertical position in one single applied to T-joints, corner joints, girth seams in heavy-
pass, regardless of the plate thickness which can range wall cylinders, and so forth.
between 12.5 and 500 mm. The welding heat is provided
by a small quantity of flux which is converted into a Consumable Guide Method
conductive molten slag by its resistance to the electric This is a modified version of the electroslag process.
current passing between the continuously fed wire or In this method, there is no need for the copper shoes,
wires (one wire for thickness up to 100 mm, two wires wire-feeding mechanism and associated equipment to
up to 230 mm and three wires up to 500 mm) and the move upward during welding. It uses simply a metal
parent plates. The hot molten slag melts the filler wire tube extending the full-length of the weld joint to guide
and the joint surfaces, and also shields the weld pool the welding wire to the welding zone. The copper
which moves upward along the full cross-section of the moulds also are full-length. These moulds and all wire-
joint as welding progresses. There is no arc, and hence feeding equipment remain stationary, with the wire
the welding action is quiet and spatter-free. being the only moving part. The guide tube melts in the
A pair of water-cooled copper shoes fitted on each side weld pool as the pool rises and adds itself to the weld
of the joint retains the molten metal and slag pool and metal (see Fig. 1.17).
acts as a mould to cool and shape the weld surfaces. The The consumable guide is generally provided with a
copper shoes automatically slide upward as welding flux coating on the outside to:
progresses (see Fig. 1.16). • Insulate it electrically from the base plate, and
• To maintain the level of the slag pool till the end.
^ WATER OUT
Fig. 1.17: Schematic diagram of electroslag consumable
Fig. 1.16: Diagrammatic sketch of electroslag welding guide method
14 IIODERN ARC WELDING TECHNOLOGY
Laser Welding about 0.05%, by weight, of chromium oxide. The green

In this process, fusion is achieved by directing a light pumps the chromium atoms to a higher state of
highly concentrated beam to a spot about the diameter energy. Each of these excited atoms emits red light and
of a human hair. The word laser is an abbreviation of thereby excites neighbouring atoms which also give off
light amplification by stimulated emission of radiation. red light which is in phase with the collided red light
A laser beam has a higher energy concentration than wave.
even an electron beam. Since the heat input to the In other words, the red light gets continuously
workpiece is extremely small, the size of the heat affected amplified. To further enhance this effect, the parallel
zone and the thermal damage to the adjacent parts of ends of the rod are mirrored to bounce the red light back
the weld are negligible. and forth within the rod. When a certain critical intensity
Laser can be used to join dissimilar metals and other of pumping is reached (the so-called threshold energy),
difficult-to-weld metals such as copper, nickel, tungsten, the chain reaction collisions become numerous enough
aluminium, stainless steel, titanium and columbium. to cause a burst of red light. The mirror at the front end
The current application of laser welding is largely in of the rod is only a partial reflector, allowing the burst
aerospace and electronic industries, where extreme of light to escape through it.
control in weldments is required. Its major limitation is
the shallow penetration (less than 1 mm). ELECTRIC RESISTANCE WELDING PROCESSES
The technique of projecting a laser beam is interesting In describing this group of processes, the word
(see Fig. 1.18), An intense green light is thrown on a electric is usually omitted and simply the term resistance
special man-made ruby, 10 mm in diameter, containing welding is used. In these processes, the coalescence is
POWER
CAPACITOR
SUPPLY
BANK
ENERGY
STORAGE
PUMPING
SYSTEM
FLASH TUBE
LASER
itbLs*m
OUTER REFLECTOR *»£<
MATERIAL
QUARTZ
THERMAL ^
<ti>
BARRIER
FOCUSING . FOCUSING
LENS SYSTEM
+—*
COHERENT
MONOCROMATIC
RADIATION WORK
PIECES
WELDING LASER (SCHEMATIC) BASIC LASER SYSTEM
Fig. 1.18: Schematic diagram of laser welding

achieved by passing electric current through the metallic Flash Butt Welding
parts by inserting them into a circuit. The electric This process is an extension of resistance butt welding.
resistance produces heat at the junction of the parts. The parts to be joined are gripped in the clamps, and their
Finally, pressure is applied. interfaces are gradually brought into contact to complete
In the process chart (Table 1.1), resistance welding the secondary circuit. When the welding voltage of up to
processes have been placed between fusion and solid about 10 is applied at the clamps, current flows through
phase processes, because it is difficult to decide whether the initial points of contact causing them to melt. These
they are fusion processes or solid state processes as molten bridges are then ruptured and small short-lived
evident from the following examples: arcs are formed.
• Resistance butt welding is a solid state process in The platen on which the movable clamp is mounted
which melting of the joint is totally avoided by is moving forward while this takes place, and fresh con-
controlling the parameters. tacts are then made elsewhere so that the cycle of events
can be repeated. This intermittent process, during which
• Spot, seam and projection welding are also solid
much of the metal contained in the molten bridges is
state processes, except for a small molten nugget
expelled violently in a spectacular manner, is called
which is formed in the weld. The size of the nug-
flashing.
get is kept as small as possible.
Flashing is allowed to continue until the surfaces to
• In flash, percussion and high frequency welding, be joined are uniformly heated or molten. By this time
the surfaces to be joined do get fused, but the fu- the moving platen will have advanced, at an increasing
sion zone is controlled to an extremely narrow rate, to close the gap as metal is expelled, the total
zone, almost to 0.005 mm. If the fusion is not re- distance up to the point of upset being known as the
stricted, molten metal will fly off under pressure flashing allowance. At this point the rate of movement
and the weld will not take place. of the platen is rapidly increased and a high force
applied to forge the parts together and expel the molten
Resistance Butt Welding metal on the surfaces (see Fig. 1.20).
This process is also known as upset welding or Typical applications of this process are: rails, steel
simply butt welding. Here, the temperature of the joint strips, window frames and automobile rear axle casings.
is raised by the resistance to the passage of an electric Special purpose machines are designed for each of these
current across the interface of the joint. The parts to be applications.
joined (usually wires and rods) are held in clamps, one
stationary and the other m o v a b l e , which act as Spot Welding
conductors for the low-voltage electric supply and also In this process, a spot of weld is made between over-
apply force (see Fig. 1.19). lapping sheets by means of two cylindrical copper-alloy
This force or pressure is applied only after the electrodes, one on top and the other at the bottom, which
abutting surfaces have reached a temperature slightly carry a high current. The electrodes also clamp the work
below the melting point, which results in the upsetting and apply pressure when the metal at the joint gets
of the metal. Uniform and accurately mating surfaces sufficiently heated by electrical resistance (see Fig. 1.21).
are desirable to exclude air and give uniform heating. A tiny button of fused metal results at the sheet inter-
The process is commonly used during rod rolling and face which is called the nugget. The electrodes are re-
wire drawing operations to join the ends. tracted after the weld is complete. Spot welding is always
performed with a machine in which all the parameters
can be controlled. The process is used on a large scale in
automotive production, and in sheet-metal fabrication.
Stitch Welding
This is an extension of spot welding, in which a series
of overlapping spot welds are made in the same manner
SOLID CONTACT as stitching cloth. Stitch welding may be performed with
a normal spot welding machine or with a specially
frrr =v designed one, which automatically makes spot welds
Fig. 1.19: Sketch of resistance butt welding in a continuous series.
CLAMPS JOINT TO BE WELDED
MOVABLE PLATEN
WELDING FLEXIBLE LEAD

TRANSFOFH
MER
1 1
;/>'. ■J//////3
LIGHT CONTACT AC SUPPLY
<A) IB)
Fig. 1.20: Flash butt welding: (A) Sketch of joint set-up; (B) Schematic diagram of equipment
FORCE Multiple Spot Welding

This is a modification of spot welding, in which two
or more welds can be obtained from each transformer
secondary. Four typical types are shown in Fig. 1.22.
\\V^MWti&SSq z Series Spot Welding

f/////,/;,////////////A UJ
cc
cc
In series welding (see Fig. 1.22A), a portion of the
secondary current bypasses (shunts) any weld nugget
being formed. This shunt current passes through one of
the panels being welded. Generally, two welds are made
Fig. 1.21: Sketch of spot welding per transformer secondary.
UuuuJ UuuuJ
SERIES WELDING DIRECT WELDING IN SERIES
UjLfliJ
r~6 ft
nnnnrs
OVER-AND UNDER-WELDING PARALLEL SPOT WELDING
Fig. 1.22: Sketches showing four methods of multiple spot welding

Roller Spot Welding The joint edges are in intimate contact and the roller
In this process, a series of intermittent spot welds are electrodes travel directly over the seam. Figure 1.25
made using wheels or rollers as electrodes. The rollers shows the process being used in the production of
are power driven and are stopped while individual welded tubes.
welds are made. Current is passed intermittently when
the electrodes are stationary (see Fig. 1.23). Foil Butt Seam Welding
This is a modification of seam welding, in which
Seam Welding thin narrow strips of metal are introduced between one
Seam welding is similar to spot welding, except that or both of the circular electrodes and the workpiece. The
the copper-alloy electrodes are in the form of circular joint edges are held in the same plane instead of being
rollers. The overlapping sheets are held under constant overlapped. The strips help to localise the melting and
pressure between the roller electrodes, which rotate at to avoid reduction of section thickness at the joint (see
constant speed and carry current (see Fig. 1.24). Fig. 1.26).
A series of spot welds, whose nuggets overlap each The process is being used successfully on the shells
other, are formed which give the appearance of a of railcoaches.
continuous weld seam. A common application of seam
welding is in the manufacture of steel drums. Projection Welding
This is a modified method of making single or multiple
Butt Seam Welding spot welds. Projection welds are made by providing an
This is similar to seam welding, except that the embossment or projection on one or both of the contacting
sheets to be joined are in the same plane instead of being base metal surfaces to localise the pre-ssure and current
overlapped. flow at a particular point (see Fig. 1.27).
This reduces the amount of current and pressure
required to produce a spot weld between two pieces of
ELECTRODE FORCE metal. Other advantages would include:
• Maintaining proper heat balance between two
INTERMITTENT ENT workpieces of markedly different mass
WELDS • Better control of size and shape of fused area
• Making of multiple welds simultaneously between
a single set of electrodes or contact platens and
WHEEL ELECTRODESI
WITH SHAPED • Forming of welds from one side of the assembly by
PERIPHERY series welding two separated projections, provided
adequate pressure can be applied.
Fig. 1.23: Sketch of roller spot welding
Percussion Welding
Percussion (or percussive) welding is unique and
FORCE
of considerable interest, because butt welds can be made
with incredible speed, in almost any combination of
dissimilar materials and without the expulsion of a fin
or flash around the joint.
The pieces to be joined are held a short distance
apart in clamping dies which carry current and apply
pressure. The ends to be welded are prepared for
f/mffm//tiiiX;ti;;;ti;;S;: 3
accurate mating. An extremely heavy electric current is
delivered (electrical energy from discharging condensers
or capacitors) to the pieces as a very short impulse,
perhaps for only 0.001 second, and flows across the gap
FORCE between the pieces as an arc. The heat of this high-energy
Fig. 1.24: Sketch of seam welding
arc produces superficial melting over the entire end
surfaces of the bars. An instant after the arc has struck,
SIDE PRESSURE ROLLS
TRANSFORMER
TUBE BEING WELDED
Fig. 1.25: Diagrammatic sketch of butt seam welding
FORCE
BEFORE AFTER WELDING

FOIL
FOIL
Fig. 1.26: Sketch of foil butt seam welding Fig. 1.27: Sketch of projection welding
the pieces are brought together with an impact blow SOLID PHASE WELDING
(hence the term percussion) to complete the weld (see This is a group of processes in which coalescence is
Fig. 1.28). produced essentially below the melting point of the base
metals being joined, without the additiort of brazing
High Frequency Resistance Welding filler metal. In some of them pressure is used. With the
This process is mainly used in the upset butt welding exception of cold welding and ultrasonic welding, all
of tubes as a continuous process. The tube formed out of the processes involve heat.
a continuous strip passes under the electrodes, which
transfer the high frequency welding current (alternating Forge Welding
currents ranging from 200 to 4,50,000 cycles) into the This is the age-old process used by the village
material being welded and through a set of rolls, which blacksmith. It consists of heating the surfaces to be joined
provides the welding force (see Fig. 1.29). (may be on a charcoal fire or any other handy source of
The amount of upset is regulated by the relative heat), placing these surfaces in contact, and hammering
position of the welding electrodes and the rolls applying them together to form a weld. The surfaces are heated
the upset force. The required welding heat is governed below the melting point, so that they are in a plastic state.
by the current passing through the work and the speed The oxidising flame of the fire produces a liquid film of
at which the tube goes through the rolls. iron oxide on the surface, which is squeezed out more or
DRIVE SPRING COMPONENT

WIRE LEAD WELD
TERMINAL
POWER SUPPLY
PRINTED CIRCUIT
OR MOULDED BOARD (B)
Fig. 1.28: Diagrammatic sketch of percussion welding:

(A) Wire end for capacitor discharge; (B) Percussion welded assembly
assembly of the steel plate and the cladding material such

as nickel or stainless steel is formed.
The assembly or sandwich is then welded at the
periphery to exclude air as well as to hold the parts
together during rolling. In rolling, the high temperature
and working create a solid-phase weld between the steel
and cladding metal.
Production of clad plate by hot rolling a pack or
FORCE sandwich is an example of roller forge welding.
Gas Pressure Welding

This is another modern adaptation of forge
welding, in which the joint, held in compression, is
locally heated by oxy-acetylene flames to a temperature
somewhere below the melting point. When the joint is
Fig. 1.29: Sketch of high frequency resistance welding held for a sufficient length of time at this temperature
under pressure, welding takes place.
This process has been successfully adapted for the
less completely when the solid surfaces are hammered welding of rail ends.
together.
The historic Iron Pillar in Delhi consists of several High Frequency Pressure Welding
ingots of pure iron forge-welded together. This process differs from high-frequency resistance
welding described earlier. The difference is that instead
Hammer Welding / Die Welding / Roll Welding of direct contact being made with the work, the current
These are the refined variations of forge welding. is induced in the surface layer by a coil wrapped round
Hammer welding is a semi-automatic or automatic the workpiece. This causes heating. When a suitable
process in which a heavy hammer is powered by steam, temperature is reached, the weld is consolidated by a
hydraulic or pneumatic equipment and the hammer forging action of the joint (see Fig. 1.30).
blows are applied at low velocity to forge the weld. The process is also referred to as high-frequency
In die welding, the parts are heated in a furnace and induction welding. It is used in the manufacture of tubes.
the pressure is applied by means of dies. The dies also
form the work while it is hot. Magnetic Pulse Welding (MPW)
In roll welding, the pressure is applied on heated This is basically a solid state welding process, still in
parts by means of rollers. Roll welding is used mostly for its infancy, in which bonding is produced by an oblique,
the manufacture of clad steel plates and sheets. First, an high velocity collision between the two bodies to be
Cold welding is generally done on ductile metals like

WELD POINT copper and aluminium.
I— WELD SEAM
WELD ROLLS Diffusion Bonding
CURRENT In this process, union between specially prepared
VEE
mating surfaces takes place as a result of diffusion, which
occurs due to high temperature and pressure exerted
INDUCTION COIL
for a sufficiently long time. The pressure is low enough
to ensure that there is no plastic flow or deformation.
The extended time of several minutes at elevated
temperatures (which will cause oxidation) requires that
the joint is made in protective atmosphere or vacuum.
Sometimes thin inserts are placed between the mating
surfaces to speed up diffusion and ensure strong welds.
I— IMPEDER
Explosive Welding
Fig. 1.30: Sketch of high-frequency pressure welding In this process, two pieces of metal are impacted
together at an extremely high velocity of impact
welded. The principle of magnetic repulsion effect is achieved by the detonation of an explosive charge. The
iitilised in this process. result is a solid-state weld completed in microseconds
A wide range of tubular materials have been joined without any noticeable deformation (see Fig. 1.32).
using MPW technology. Dissimilar metals like steel to
copper, steel to aluminium, copper to aluminium etc. EXPLOSIVE
can be joined using no filler metal or gases. Fig. 1.31
illustrates the basic principle of the process.
— ^ r METAL PLATE
Cold Welding
In this process, welding is obtained at room
temperature solely by the application of pressure across /
the interface. The contact surfaces have to be specially
prepared by degreasing, scratch brushing and other */ EXPLOSIVE
operations. The joints are either lap or butt. For making
lap joints, indenting dies may be forced into the metal
causing deformation and flow to provide the extension ' SPACE ^VS^^TAL PLATE
of the interface. The butt method is used for joining

wires, tubes and bar stock, the parts being gripped in Fig. 1.32: Sketches of explosive welding
dies and forced together to cause lateral flow.
The process has been used to prepare clad plates
involving dissimilar metals, and in welding tubes to
tube-plates in heat exchangers.
• id ; ' issi - * isi ; ■ cs^— 2 coii

v ym?, vW , ^ 'Wm_ ■ * îir, Discharge current
Friction Welding
8.Workpiece(after)-y— * * / " " ~ " \ / " ~ * \ / " ~ * \ / ^ ~ * \ ~
r
4. Workpiece (before!
In this process, friction is employed to generate heat
. +)( +) ( + ) ( + ) between two sliding or rotating metal surfaces. The
process is usually carried out by placing the pieces to
5. Pressure v<C . . . ^ ! ^ * ? ! ^ ! ^ ^ r ? ^ T ? T ? ? r ! ^ ^ ^ ^ ^ / ^ ^ ^ be welded in chucks on a common horizontal axis. One
part is rotated while the other remains stationary.
6. WorKpiece(afterV—V " ^ V _ , ' V_„/ V _ / V_r/ 4. Workpiece (before; Pressure is applied to generate enough heat to reach a
\'• ^ v ' • ^ v ' •^P' ■ ^fc; 2 bonding temperature within a few seconds. At this point,
\r.^^" x
^ ^ ' V
^^V'' y
^\*^ 1. Discharge current rotation is stopped very quickly and pressure is
^ y
v v v
maintained or increased until welding is complete (see
Fig. 1.31: Sketch of magnetic pulse welding
Fig. 1.33).
STATIONARY CHUCK
ROTATING CHUCK
Workpiece
THRUST CYLINDER Tool

(A) shoulder
MOTOR
DIRECTION OF ROTATION
START Welding pin with

spocial profile
THRUST APPLIED Backing bar

Fig. 1.34: The friction stir welding concept
STAGE 3 BEGINS
> frequency vibratory energy to workpieces held together
under moderate static pressure (see Fig. 1.35).
m FORGE AND BRAKE The process has proved to be an economical substi-
tute for resistance welding of electrical conductors: it
welds faster, uses less power, and usually requires lower
Fig. 1.33: Diagrammatic sketches of friction welding: capital investment.
(A) Main components of the equipment; The vibrations break up and disperse oxides, surface
(B) Stages of weld formation films, and some types of insulation at the joint surface
to permit bare metal contact. Inter-atomic diffusion
To accommodate awkward or unusually long occurs across the facing surfaces and results in a
sections, an intermediate slug or disk can be rotated metallurgical bond without application of heat. The
between the sections to be joined. joints so produced possess electrical resistance and
mechanical strength almost equal to that of the parent
In applying the process, one uses a specially designed
materials.
machine called the friction welder. It helps one to
produce consistently high quality joints, each within a The equipment consists of a frequency converter
few seconds, in various similar and dissimilar metal which changes 60-cycle line power to the high frequency
combinations. (15 to 60 kHz) required by the ultrasonic head. The head
incorporates a piezoelectric transducer, which converts
Friction Stir Welding
the high frequency electrical output into vibratory
mechanical motion, and waveguides which transmit the
Friction Stir Welding (FSW) uses a new solid phase
welding technique which has been developed in the 90's vibratory waves to the tools and into the weldment.
primarily for joining aluminium alloys. FSW is a Static clamping force is applied to the head by a
derivative of conventional friction welding enabling the pneumatic or hydraulic cylinder. High production
advantages of solid phase welding to be applied to the operations use a high-speed cam-operated anvil to
fabrication of long butt and lap joints, with very little position the work.
post-weld distortion. The concept of FSW is shown in The process has been successfully used in the
Fig. 1.34 electrical equipment production in the U.S.A. Typical
applications include:
Ultrasonic Welding • Almost all commonly used armatures. On tang
Ultrasonic welding is a solid-state process for joining type armatures, the tang is bent and welded dur-
similar or dissimilar metals by application of high- ing the ultrasonic cycle. Vibratory energy im-
CLAMPING FORCE APPLIED - - ^ j - 1

THROUGH MOUNT <J^
FREQUENCY CONVERTER VIBRATION - ^ ^ A - K
TRANSDUCER
Fig. 1.35: Diagrammatic sketch of ultrasonic welding
parted to the wire scrubs away the insulation and 450°C but below the melting point of the base metal,
makes a tack weld between tang and commutator and
and between wire and tang (c) heating the filler metal till it melts and gets dis-
• Slotted commutators tributed between the closely fitted surfaces by capillary
attraction.
• Starter motor armatures
Like welding, brazing is used in the fabrication of a
• Commutator armatures for motors of electric door variety of metallic products ranging from tea kettles and
locks and windshield wiper-washers bicycle parts to heat exchangers and aerospace
• Field coils, field coil assembly, wire harness sys- structures. In recent years, hundreds of new con-
tems, wire terminal connections, and induction structional alloys have been developed, and at the same
coils time new brazing filler metals have been perfected to join
• Joining of braided brush wire to brush plate them. Today thousands of tonnes of carbon steel
• A wide variety of wire terminals components brazed with copper alloy filler metals are
produced annually by the fabrication industry.
Brazing is being used for the production of large foil
BRAZEWELDING heat exchangers, automobile radiators, turbojet engine
This process is similar to brazing which is described components, and many such critical items.
below, except that the molten filler metal is not The range of materials which can be successfully
distributed in the joint by capillary action. It gets filled brazed is very wide and includes beryllium, titanium,
up in a groove or gets piled up as a fillet weld. In other superalloys, refractory metals, dispersion strengthened
words, brazewelding is similar to normal welding, but metals, all-metal composites, dissimilar metals and
the filler metal used has melting point above 450°C but special ceramic-to-metal combinations.
below the melting point of the base metal.
A typical example of this process is what is popularly Resistance Brazing
called bronze welding, used for making joints in mild In this process, the heat is provided by the resistance
steel and cast iron. The filler rod is a special copper alloy of the parts to the flow of low-voltage, high-current
which melts at approximately 800°C. It is used in power supplied from a suitable transformer and from
combination with flux. The filler rod is melted with an the heat conduction from the high resistance facing
oxyfuel flame. The molten metal fills the V-groove in a materials on the contacting electrodes. The electrodes
butt joint or piles up as a fillet weld and makes a joint are usually of carbon, molybdenum, tungsten or steel.
without causing the base metal to melt. This process is extensively used when small areas are to
be brazed and when the material has high electrical
Brazing conductivity.
In brazing, the coalescence between metallic parts
is obtained by Dip Brazing
(a) heating them to a suitable temperature, In this case, the heat is provided by a molten chemical
(b) using a filler metal with melting point above bath or molten metal bath. In the former, the bath may
act as a flux. In the latter, the bath provides the filler MIG Brazing
metal. Conventional brazing joins materials with the aid of
The parts to be brazed are first cleaned. Filler metal melted filler, the melting temperature of which lies below
inserts are pre-placed in the joints in the case of flux that of the parent metal, which is wetted to form the joint
bath brazing. In the case of metal bath brazing, the parts but not melted. In MIG brazing, this technique is extended
and joint surfaces are protected with flux. They are then in semi-automatic or even in fully automatic mode, similar
assembled in a fixture which is usually of stainless steel. to MIG process. Pure argon is the shielding gas most
Next, they are preheated to a suitable temperature in a often used in MIG brazing although addition of 1%
furnace and finally introduced into the molten bath. oxygen is not uncommon. Many types of copper-based
welding wires are available—common amongst these are
Flame Brazing Si-Mn alloys, aluminium-bronze and phosphor-bronze.
This is the most common process which uses the Modern inverter type MIG power sources are most
standard oxyacetylene flame and equipment. suitable for this process.
Furnace Brazing THERMAL CUTTING

Here the parts are assembled in a jig, filler metal is Thermal cutting processes involve melting of the
pre-placed in the joints, flux is also applied at the joints, metal to be cut. The word thermal distinguishes them
and the entire assembly is heated in an electric or gas/ from the mechanical methods of cutting.
oil heated furnace where the temperature can be
controlled within ±5°C. Sometimes a special atmosphere Oxygen Flame Cutting
is used in the furnace instead of the flux. In this process of metal-cutting, popularly known as
Vacuum is used for special stainless steels, oxyacetylene or gas cutting, the cut is started by
superalloys, Ti alloys, etc. preheating a section of metal to the ignition temperature
with an oxyacetylene flame. Oxygen is then turned on,
Induction Brazing which ignites the iron or steel at the upper surface. This
In this case, heat is obtained from the resistance of burning (oxidation) releases heat and thereby raises the
the work to induced electric current. Parts to be heated temperature of the metal below the surface and melts
act as the short-circuited secondary of a transformer, it. The cutting oxygen now ignites the molten metal,
where the work coil, which is connected to the generator which in turn releases heat to the metal below. The same
is the primary. The heating results from the resistance action progresses downward until the metal is severed
of the parts to the currents induced in the parts (see Fig. 1.36).
themselves by the transformer action. High frequencies The slag which is mainly iron oxide (Fe304) is blown
are usually used, the most common being 10,000 Hz out of the cut by the force of the oxygen stream.
obtained from a motor-generator set.
Salt Bath Brazing

This is similar to molten flux bath dip brazing
described earlier. Here a salt of suitable melting point is
used instead of the flux.
Diffusion Brazing
Here the parts to be joined are heated to a suitable
temperature, and braze filler metal is introduced
between the contacting surfaces by preplacing or by
capillary action. Sometimes a liquid phase develops in
situ at the contacting surfaces, which acts like a filler
metal. The filler metal or this liquid phase gets diffused Fig. 1.36: Oxygen flame cutting:
with the base metal, with the result that the joint (A) Cutting tip showing arrangement of orifices;
properties are changed to approach those of the base (B) Sketch of cutting action;
(C) Drag formation
metal.
The process uses the normal gas welding equipment stream of oxygen is blown through the hollow core to
and a specially designed cutting torch, which is equipped oxidise and eject the metal (see Fig. 1.38).
with a special lever for the control of the oxygen and a The coating helps to melt, oxidise or dilute the molten
cutting tip, which has an orifice in the centre surrounded base metal. In the case of oxidation-resistant metals, the
by several smaller ones. The smaller holes provide the coating helps the molten metal to flow from the cut. The
heating flame while the centre opening allows the flow cutting is very rapid.
of cutting oxygen. This process was primarily developed for
underwater cutting. It can be effectively used by
Powder Cutting foundries and scrap yards and for dismantling work,
The normal oxygen cutting process is ineffective for since it is capable of cutting all types of metals in all
cutting metals that form refractory oxides such as positions.
aluminium, bronzes, and high-nickel alloys. To cut these
metals, iron powder is fed into the oxygen stream. A Carbon Arc Cutting
mixture of iron powder and aluminium powder is In this process, the arc formed between a non-
sometimes used for cutting brass, copper and high- consumable carbon electrode and the base metal is used
nickel alloys. The aluminium releases more heat than for cutting. This process was formerly in use along with
just the iron powder alone. Iron powder also produces carbon arc welding, but is now almost obsolete.
a rapid cutting action on stainless steels and is very
effective in getting smoother cuts in cast iron.
Powder cutting makes use of a special torch which is
equipped with a powder tube, nozzle, and powder valve.
The powder is stored in a dispenser and is carried to the
powder valve by compressed air or nitrogen where it is
fed to the flame. In operation the powder valve is opened
first and then the oxygen valve.
Oxygen Lancing
This is an oxygen cutting process in which the base
metal is first preheated with a normal cutting torch, and
then oxygen is fed by means of a long steel pipe called
oxygen lance. The lance pipe is moved suitably to
regulate the metal being cut or pierced. The pipe gets
consumed gradually in the process (see Fig. 1.37).
Lancing and another similar process called scarfing
are used to remove the surface defects of steel blooms,
slabs and billets before finish rolling in a steel mill.
Oxygen Arc Cutting

In this case, a hollow (tubular) coated electrode
makes an arc with the base metal causing it to melt. A
Fig. 1.38: Diagrammatic sketch of oxygen arc cutting

Metal-Arc Cutting
Here a manual welding electrode having a penetrating Plasma Arc Cutting
arc is used to cut the metal (see Fig. 1.39). Some arc-cutting This is dealt with later under Plasma Processes.
electrodes are also capable of chamfering and gouging
metals. Spark Erosion Cutting
This process makes use of spark-discharge machines
to perform unique shaping and piercing operations on
hardened steel.
Electron Beam Cutting/ Laser Cutting

These processes are based on the same principles as
electron beam welding and laser welding described
earlier.
THERMAL SPRAYING
Thermal spraying is a group of processes in which a
Fig. 1.39: Metal arc cutting: (A) Swinging action of metallic or non-metallic material in the form of wire,
electrode; (B) Form of cut rod or powder is heated to a molten or semi-molten state
and then sprayed in atomised form onto a surface (called
Air-Arc Cutting substrate) to form a coating. The heating is commonly
In this process, which is widely used in fabrication done with an oxyfuel gas flame, an electric arc or a
shops and foundries, an arc is formed between a non- plasma arc. A mechanical contrivance called the spray
consumable copper-coated graphite electrode and the gun is used to propel the hot material to the surface by
base metal; the molten metal is blown away with a high means of a gas jet. The plastic or molten particles flatten
velocity jet of compressed air, parallel and external to and form thin platelets when they strike the surface. The
the electrode. Since this process does not depend on bond between the spray deposit and the substrate may
oxidation, it works equally well on metals which do not be mechanical, metallurgical, chemical or a combination
readily oxidise (see Fig. 1.40). of these.
A special holder is used to clamp a carbon-graphite Thermal s p r a y i n g is also sometimes called
electrode in such a position that air emitted from orifices metallising or metal spraying, which implies that the
in the electrode holder is directed parallel to the material being sprayed is metallic.
electrode. The air then strikes the molten metal
immediately behind the arc. The electrode holder also Flame Spraying
contains an air control valve and a cable which carries In this case, the metal to be sprayed, which is in the
both the current and the air. This cable is connected to a form of wire or finely divided powder, is passed through
welding machine as well as to a source of compressed an oxyacetylene flame (see Fig. 1.41). In the case of the
air. wire, a stream of high-velocity gas (air, nitrogen or C0 2 )
ELECTRODE
ELECTRODE RELEASE
AIR AND
POWER SUPPLY
Fig. 1.40: Diagrammatic sketch of air-arc cutting torch gripping a carbon-graphite electrode
WIRE CONTROL
UNITJ / FLOW METER
AIR CONTROL UNIT
METALLISING UNIT
HOSE UNIT
OXYGEN ' \ ACETYLENE OR PROPANE
Fig. 1.41: Schematic diagram of flame spraying
is also passed over the flame to divide the molten metal when an electrical discharge takes place in a gas, and
into fine particles and carry them to the work surface. In also in an electric arc (the blinding light of an arc comes
certain cases, the stream of gas also serves to feed the from the plasma). In normal arc welding processes, the
wire into the flame. plasma is easily blown away by moving gas streams.
But in plasma processes which are listed below, the
Electric Arc Spraying plasma is contained and used effectively:
Here two current-carrying wires are continuously fed • Plasma arc welding
into a gun through wire guides. The wire ends meet in • Micro-plasma arc welding
front of the gun nozzle and form an arc. An air jet is
• Plasma-MIG welding
blown through the gun nozzle on to the arc. It ejects
molten metal particles and carries them to the work • Plasma arc cutting
surface (see Fig. 1.42). • Plasma spraying.
In all these processes, based on DC supply, a specially
Plasma Spraying designed torch, which is a modification of the TIG torch,
This process is discussed under Plasma Processes. concentrates the plasma energy and ensures its most
efficient utilisation for welding, cutting and spraying.
PLASMA PROCESSES The tip of the tungsten electrode (DC negative) is located
Plasma is considered the fourth state of matter, the within the torch nozzle, while the nozzle has a small
other three being solid, liquid and gas. Plasma is a opening which constricts the arc. As gas (usually argon)
superheated gas which has dissociated into positive ions, is fed through the arc, it becomes heated to the plasma
neutral atoms and negative electrons. Plasma is formed temperature range (30,000 to 60,000°F). The plasma tail-
TOP VIEW
SECTION A-A
n 3
7 8
Fig. 1.42: Schematic diagram of electric arc spraying gun

Insulated housing for contact piece; 2. Contact piece; 3. Heat resistant protective plate; 4. Wire guides;
5. Atomizing air nozzle; 6. Arc point; 7. Metal particles; 8. Atomizing air stream; 9. Wires.
flame issues from the torch nozzle as a jet of tremendous

velocity.
The plasma-arc is of two types: transferred arc and ORIFICE GAS
ELECTRODE
non-transfer red arc. In the former, the arc is formed
between the electrode and the workpiece. In the latter, the
arc is formed between the electrode and the constricting
orifice inside the torch (see Fig. 1.43).
SHIELDING GAS
Plasma Welding
Plasma welding is an extension of TIG welding. The
L
main difference as explained above is the constriction
of the arc column, resulting in much higher heat transfer
rate. The torch is further modified to provide an outer
sheath of cool gas around the central plasma core, as
NOZZLE
shown in Fig. 1.44. Plasma welding makes use of the | y W _ SHIELDING GAS
keyhole technique. When the plasma jet strikes the
metal, it cuts or keyholes entirely through the workpiece
producing a small hole, which is carried along the weld
seam. During this cutting action, the molten metal in
front of the arc flows around the arc column, then gets ARC PLASMA
WORK
drawn together immediately behind the hole by surface
mi*^
tension forces and reforms as a weld bead. Thus butt
welds of 12.5 mm or larger thickness are possible in a
single pass without edge preparation or fillet metal. The
process can weld carbon steels, stainless steels, copper,
brass, aluminium, titanium, monel and inconel. Fig. 1.44: Sketch of plasma welding torch
CONSTRICTING N
RIFICE GAS
SHIELDING GAS
UP/-NfEiD
(A) (B)
Fig. 1.43: Sketch of plasma arc torch: (A) Transferred arc; (B) Non-transferred arc
Micro-Plasma Arc Welding

This is a modified process using DC current range of
ELECTRODE
0.1-10 amps. It is capable of welding extremely thin sheets ORIFICE SHIELDING GAS
and foils in the thickness range of 0.05-1.6 mm. GAS
CONSTRICTING
Plasma-MIG Welding N022LE
This is an extension of plasma arc welding in which

the non-consumable tungsten electrode is replaced by a WORKPIECE
consumable continuously-fed metallic electrode (see Fig.
1.45).
CONSUMABLE WIPE |
TUNGSTEN
POWER j . ~
SUPPLY)
Fig. 1.46: Sketch of plasma arc cutting
For cutting carbon steels, nitrogen is used as a cutting

gas while oxygen is also passed into the plasma to provide
additional heat from the iron-oxygen reaction at the
cutting point. A transferred arc is used, that is, the arc is
formed between the tungsten electrode (-) and the
workpiece (+).
Fig. 1.45: Schematic diagram of plasma-MIG welding
Plasma Spraying
Plasma Arc Cutting In this process, a special plasma spray gun is used in
This process shown in Fig. 1.46, uses DC source with which a non-transferred arc forms between the tungsten
open circuit voltage of 120 or more. Hence as a safety electrode (-) and the nozzle (+). Plasma gas is passed
precaution against electric shock, it is preferably used as through the arc in the nozzle bore. The spraying material,
a mechanised process. in powder form is fed into the nozzle bore or into the
Dual gases are often used: one as a cutting gas and flame just beyond the end of the nozzle. The plasma
the other as a shielding gas. The typical combination is flame melts the particles and impels the molten material
argon/ hydrogen or nitrogen/ hydrogen for cutting onto the workpiece (see Fig. 1.47).
aluminium and stainless steels, in which argon and The gas is either n i t r o g e n or argon or 5-25%
nitrogen are for cutting while hydrogen is for shielding. hydrogen mixed with nitrogen or argon.
TUNGSTEN ELECTRODE
NON-TRANSFERRED
POWER SUPPLY ,
CENTRE GAS
GAS- BORNE POWDER
ARC ORIFICE
TRANSFERRED POWER
SUPPLY SHIELDING GAS
Fig. 1.47: Schematic diagram of plasma spraying

REFERENCES AND SELECTED BIBLIOGRAPHY
Welding Handbook, 7th Ed., Vols. 2 & 3, published by American Welding Society.
Welding Technology by J.W. Giachino, W. Weeks & G.S. Johnson, published 1968, by American Technical Society.
Welding Process Technology by P.T. Houldcraft, published 1977, by Cambridge University Press.
Some Aspect of Welding in Modern Ship Construction by V.P. Nair, Indian Welding Journal, Oct. 1977, published by The Indian
Institute of Welding.
The Welding Encyclopedia by T.B. Jefferson, published 1964, by Monticello Books, III., U.S.A.
Welding Metallurgy. Vols. 1 & 2 by Linnet, published 1965, by American Welding Society.
Advances in Welding Processes—Conference Proceedings, published 1971, by The Welding Institute, U.K.
Current Welding Processes, published 1964, by American Welding Society.
CHAPTER 2
Power Sources
For
Arc Welding
THE various arc welding processes shown in the pro- such electric power are known as power sources for arc
cess chart (Table 1.1) dhd briefly p l a i n e d in Chapter 1 welding.
require special electfic ^dWS- ftf Matively low Vdltage Many types of power sources are available to suit
and high current to produce arid kiisiain dri arc capable the various arc welding processes, and they can be
of making a good weld. Machines designed to deliver classified in several ways. A simple classification based
Table 2.1: Classification system for power sources

Method of classification Features
1) Type of basic design Rotating type Static type
Motor-driven Engine-driven Transformer Rectifier/ Inverter
generator generator
2) Method of current control Field control/ variable reactor, variable coupling, tap changer switch,
satttfdble reactor, electronic control with feedback (thyristor control,
transistor control, IGBT control, etc.}
3) Type of output current Direct current (DC)
Alternating current (AC)
Cdmbmation of both (AC) & (DC)
4) Type of static characteristics Constant ctif^ht type (CC)
(V-A curve) Constant voltage type (CV)
Combination of both (CC) k (CV) type
5) Rated current & duty cycle Capability of the power source to deliver current at specified duty cycle, say
60% duty cycle, 100% duty cycle, etc., and range of welding current.
6) Suitability for the process Capability of the power source to be used for manual metal arc welding
(MMAW), gas metal arc welding (MIG/C0 2 ), gas tungsten arc welding
(GTAW or TIG), submerged arc welding (SAW), etc.
7) Suitability fdr ho. of operators Capability of the power source to be used simultaneously by a single operator,
by double operators or multioperators.
8) Type of cooling method Natural air-cooled (NA)
Forced air-cooled (FA)
Oil-cooled natural (ON)
Oil-cooled forced (OF)
9) Type of insulation Class A, Class B, Class E, Class F, and Class H, depending on the max.
allowable temperature rise over a specified ambient temperature (usually 40°C).
on the consideration of design, rating and applications are of drooping type, while the others are of relatively
is shown in Table 2.1. The description of a power source flat type. A machine giving a drooping curve is said to
should include the appropriate feature under each have drooping characteristics and is described as a
method of classification given in this table. For example, constant current (CC) type machine, because small
a particular power source could be described as static variations in the voltage caused by variations in arc
type transformer-rectifier having saturable core type length during welding, do not significantly change the
current control and constant current (CC) characteristics, output current.
with A C / DC output, current range of 70 to 400 Amps. A machine with a relatively flat curve is described
and rating of 400 Amps, at 60% duty cycle. Thus each as a constant voltage (CV) or constant potential (CP)
method of classification explains a particular feature of type machine, because in this case small variations in
the power source but is not complete in itself to describe load voltage caused by arc fluctuations result in
its essential features. substantial changes in the current output. Normally CV
Before proceeding further, one must understand the machines are designed for DC welding. A CV power
following basic features of the arc welding power source: source giving AC output is not suitable for arc welding.
1) Static characteristic curves (Volt-Amp curves) The terms constant current and constant voltage are
2) Open-circuit voltage (OCV) technically incorrect; yet they are used and accepted by
3) Static and dynamic characteristics the welding industry. It is only recently that electroni-
cally controlled power sources have been developed,
4) Current ratings and duty cycle specifications
which are capable of giving virtually any type of curve
5) Classes of insulation
between truly constant current and truly constant volt-
6) Power factor age. These have been dealt with later in this chapter.
Drooping or constant current type power sources are
1) Static Characteristic Curves conventionally used for manual metal arc welding
These curves show the relation between load voltage (MMAW) or gas tungsten arc welding (TIG), the obvious
and load current under various load conditions. Hence reason being that with this type of characteristics, the
they are also called Volt-Amp (V-A) curves. To obtain w e l d i n g c u r r e n t r e m a i n s substantially constant,
such a curve, a pure resistive load (usually water load) irrespective of small variations in arc length and
is connected across the output terminals of the power consequent slight change in arc voltage, which are
source. The load is gradually varied from the minimum unavoidable even in the case of a skilled welder. As the
or no-load condition to the maximum or short-circuit welding current is fairly steady, the weld quality is
condition, and the voltage across the load and the consistent. In this case the short-circuit current is not
current p a s s i n g t h r o u g h the load are accurately excessive (normally less than 200% of the set current);
measured. The graph showing load voltage versus load hence at the time of arc striking and in case of accidental
current is then plotted to obtain the static characteristic situations, the current does not exceed too much and
curve. hence the weld quality remains unimpaired.
Typical curves obtained with various types of power Flat or constant-voltage type power sources are
sources are shown in Fig 2.1. It is seen that some curves conventionally used for semi-automatic and fully-
automatic processes involving a continuous electrode
fed at a constant rate, such as M I G / C 0 2 welding, flux-
cored arc welding and submerged-arc welding. The flat
type power source together with a continuous electrode
fed at a constant wire-feed speed form essentially a self-
regulating arc. The arc length and weld current are
interrelated in such a way as to correct sudden changes.
For example, arc length variation is determined by
the difference between melting rate and feeding rate of
the electrode wire. The voltage drop across the arc is
directly proportional to arc length. A small change in
arc voltage results in a very large change in the welding
WELDING CURRENT-AMPERES current. This in turn increases the melting rate of the
Fig. 2.1: Characteristic curves of welding power sources wire and quickly restores the arc length to normal.
POWER SOURCES FOR ARC WELDING 33
In M I G / C 0 2 welding, changes in arc length can OCV plays a very important role in ensuring easy arc
occur when the operator moves the welding torch starting and good arc stability. Higher the OCV, better
toward or away from the workpiece. The flat type power is the arc stability. However, high OCV poses the danger
source compensates for these changes by momentarily of electric shock to the welder and hence its value is
changing the current and the melt-off rate of the wire restricted to 100 V maximum by IS:4559, "Specification
until equilibrium is established. Thus, the arc length can for single operator rectifier type DC arc welder" and
be adjusted by setting appropriate voltage on the power other relevant standards.
source and the w e l d i n g current is controlled by Commercially available welding rectifiers have OCV
adjusting the wire-feed speed. By selecting a correct values generally ranging between 65 and 80 V, while
combination of voltage and wire-feed speed, a stable welding generators have variable OCV values in the 40-
metal transfer can be obtained which results in getting 90 V range. Welding transformers meant for heavy duty
a good quality weld. In this case, the flat characteristic applications are generally designed for OCV of 60 to 70
is not truly flat but has a slight drooping tendency V, while low-cost limited-service class transformers have
(approximately 2 to 3 V per 100 Amps.). OCV as low as 50 V With such low OCV transformers,
Power sources having both d r o o p i n g and flat it is difficult to strike and maintain an arc; they are there-
characteristics are also available. The r e q u i r e d fore suitable for electrodes having a high proportion of
characteristic can be preselected by means of a change- arc stabilisers in their coating. Some transformers are
over switch/ link board. In modern times such change designed to give OCV of 100 V in addition to the nor-
in characteristics is acheivable by electronic switching mal OCV value, so that they can operate satisfactorily
systems. These power sources are called Universal with certain basic low-hydrogen electrodes. Generally
Power Sources or Multiprocess Power Sburces, as they speaking, all general-purpose electrodes operate satis-
are equally suitable for manual welding and semi-auto- factorily with OCV of 80 V (AC as well as DC).
matic and fully automatic welding. They are described In AC w e l d i n g , OCV becomes all the more
later. Table 2.2 indicates at a glance the suitability or i m p o r t a n t , because alternating current reverses
otherwise of A C / D C and C C / C V types of power direction of flow each 0.01 second in a 50 Hz supply
sources for several common arc welding processes. system. Figure 2.2 shows sine waveforms of a typical
two-range AC power source with OCV of 80 V and 55
2) Open-circuit Voltage V, each sine waveform completing a cycle in 0.02 second.
Open-circuit voltage (OCV) is the voltage across the It is seen that at each half-cycle, the current becomes
output terminals of the power source when it is under zero. This means that at this instant the electrons and
no-load condition. Hence it is also termed as no-load ions which cause the current flow in the arc cease to
voltage. flow. In the next instant the electrons and ions must
In CV type power sources, OCV is not significant, resume their flow but in the opposite direction. To make
but in the case of CC type machines, both AC and DC, this possible, ionisation of the material in the arc must
Table 2.2: Welding processes and recommended power source type

Process Direct current Alternating
Current
CC CV CC
Manual metal arc welding Yes Not used Yes
Gas tungsten arc welding (GTAW or TIG) Yes Not used Yes
Plasma arc welding Yes Not used No
Carbon arc welding, cutting, gouging Yes Possible Yes
Stud welding Yes Possible No
Submerged arc welding (SAW) Yes Yes Yes
Gas metal arc welding (MIG) Yes Yes No
MAG (CO ) welding Usable Yes Not used
MIG/MAG, short-circuiting Not used Yes Not used
Flux-cored arc welding Yes Yes Possible
Electroslag welding Possible Yes Usable
Courtesy: Modern Welding Technology by Carey
OPEN -CIRCUIT
VOLTAGE-55 V
CURRENT
Fig. 2.2: Typical sine waves of two-range AC power source
be either maintained or re-ignited by the voltage across inductance (choke), by providing feedback controls and
the arc gap. This voltage called the 'recovery voltage' is by modification of waveform of the welding current
shown in the figure. (pulsed modulations), singly or in combination.
To make the period of arc extinction as brief as Dynamic characteristics cannot be observed with
possible, it is necessary that the recovery voltage is kept standard graphs as in the case of static characteristics.
at the highest possible value. The figure shows that the The dynamic behaviour of a power source is studied on
recovery voltage is higher with 80 OCV, firstly, because an oscilloscope with special simulated conditions in the
of the higher source voltage and secondly, because with laboratory.
the greater phase shift of the current, the zero current
occurs at a time when the recovery voltage is nearly as 4) Current Ratings and Duty Cycles
great as the open-circuit voltage. Power sources are of various current ratings and
duty cycles, and are rated by the manufacturers on the
3) Static and Dynamic Characteristics basis of their current output at specific duty cycles. Duty
Every power source has specific static characteristics cycle is defined as the percentage of a five-minute
as well as dynamic characteristics, both of which are interval that it operates at a given current setting. A 60%
important from the welding performance point of view. duty cycle means that the arc is in action for three
As explained earlier, the static characteristics describe minutes out of a five-minute time period. In other
the relation between voltage and current under various words, a cycle of five minutes comprises a period of
fixed load c o n d i t i o n s . In contrast, the d y n a m i c three minutes of welding load followed by a period of
characteristics describe the relation between voltage and two minutes of no-load operation. Continuous operation
current u n d e r changing load conditions, i.e. the at rated currents for 36 minutes out of one hour is not a
instantaneous variation in arc voltage with change in 60% duty cycle, because the rating is based on successive
welding current over an extremely short interval of time five-minute intervals.
(say, 1 microsecond). Good dynamic characteristics For heavy-duty service in the fabrication industry,
enable the power source to provide extremely rapid 60% duty cycle is taken as standard for the MMA
changes in its o u t p u t voltage and current u n d e r process. For example, if a machine is rated at 300 Amps.,
changing arc conditions. at 60% duty cycle, it means that it can be safely operated
This is of special significance for the power source, at 300 Amps, current without the danger of overheating
because the welding arc is never in real steady state, for three out of every five minutes. Power sources
but is subjected to severe and rapid fluctuations due to designed for less rigorous applications and those
constant small variations in arc length, arc voltage and involving short welding cycles and long intervals may
welding current. The arc gives rise to transients at the be rated at lower currents than 300 Amps, and at lower
time of starting, extinction and re-ignition after each duty cycles, say 50% or even 20%, taking into account
half-cycle in the case of AC welding. It also gives similar the end use and safe working. Such machines are less
transients in spray transfer M I G / C 0 2 welding. expensive than the normal ones, but if they are used at
A power source having good dynamic characteristics higher duty cycles, their insulation is likely to burn out
gives a very stable and smooth arc in spite of such prematurely due to overloading.
transients. The dynamic characteristics of a power can Sometimes welding machines are required to give
be improved by providing appropriate value of series output other than rated, as for example, when a 400 A
60% duty cycle machine is used for automatic welding 6) Power Factor
application. In this case, the user should know the safe Power factor (pf) is an important electrical characteristic
current of operation at 100% d u t y cycle. This is of a power source, which is calculated from the following
computed by the following formula which takes into formula:
account the safe thermal rating of the power source: pf = K
^
F
Kva
Desired duty = (Rated current)2 x Rated d , ^ %
cycle in % (Desired current)2 where Kva is apparent power drawn from the supply
line and Kw (Kilo-watts) is actual power used to pro-
In the above example, the automatic current limit
duce the rated load, which gets registered on the power
will be
meter. Kva means kilovolt-ampere, which indicates the
(400)2 total product of ampere times volts delivered by the
100 = 60
(Automatic current) 7 supply company. For example, if the voltage reading
and current reading on the primary side are 240 V and
60 30 Amps, respectively and the actual watts registered
(Automatic current) 2 = (400)2 x
100 are 5040.
Hence, automatic current = 400 x 0.78 pf : 5040 = 0.7.

240 x 30
= 310 Amps.
Similar calculation shows that a 300 Amps., 60% duty Low pf means unnecessary wastage and less efficient
cycle machine could be used at 373 Amps, without the utilisation of power. Power supply companies as well
danger of damaging the insulation, if the duty cycle is as plant engineers are specially interested in maintaining
reduced to 35%. high pf for all the electrical equipment. IS: 4559 has
specified that machines must have a minimum pf of 0.80
5) Classes of Insulation at 50% of the continuous hand welding current. The pf
The duty cycle rating of a power source is primarily of a welding transformer can be as low as 0.45, and this
based on the maximum allowable temperature of applies equally to a welding rectifier, which is basically
various components (e.g. primary coil, secondary coil, a transformer with a means of rectification on the
choke, etc.) in the power source. These temperature secondary side. On the other hand, the inherent pf of a
values in turn depend on the type of insulation used in motor-generator is between 0.80 and 0.90.
the manufacture of these c o m p o n e n t s . Various The pf of a welding transformer or rectifier is
insulation classes and the corresponding maximum improved by connecting a capacitor across the primary
permissible t e m p e r a t u r e s specified by IS: 4559, winding. The result is that less primary current is
"Specification for single operator rectifier type DC arc demanded while welding is being performed. In the
welder," are given in Table 2.3. absence of a capacitor, much higher line current is
Table 2.3: Limits of permissible temperaturerise
Parts Class A Class E Class B Class F Class H

Insulation Insulation Insulation Insulation Insulation
T R T R T R T R T R
°C °C °C °C °C °C °C °C °C °C
a) Windings 50 60 65 75 70 80 85 100 105 125
b) Oil
c) Uninsulated parts including The temperature rise shall in no case reach such a value
cores not in contact with that there is risk of injury to any insulating material
insulated windings on adjacent parts or to the welding plant in any respect
Note 1—T-by thermometer R = by resistance.
Note 2—The numerical values quoted for classes F and H should be considered as tentative only and may be revised when more practical experience is
available. If Class C insulation is used, the temperature rise shall be a matter of agreement between purchaser and supplier.
Note 3—For details of classes of insulation, please see IS: 1271: Classification of insulating materials for electrical machinery and apparatus in relation
to their thermal stability in service.
drawn, which is likely to overload and damage the alternating voltage is given to one of the coils, an
electrical wiring. To avoid this, one must provide wiring alternating flux is built up, which induces voltage in
of higher capacity, which again means extra expenditure. the second coil. The relation between input voltage and
induced voltage is given by the formula:
ARC WELDING TRANSFORMERS
Input voltage (VJ No. of turns on supply winding
Among welding power sources, the transformer is
Induced voltage (V9) No. of turns on induced winding
the most popular type due to its low cost, simple
construction and ease of maintenance. This welding Supply winding is called primary and induced
machine which always delivers AC output for welding, winding is tailed secondary. The relation between
is mainly a step-down type of transformer, which voltage and Current is given by the formula:
converts the high-voltage low-current industrial supply Primary Secondary Primary no. of
into low-voltage and high-current required for welding. voltage (VJ current (I2) turns (NJ
It generally operates on single-phase supply, i.e. 220 V
Secondary Primary Secondary no. of
single-phase, or two lines of 440 V three-phase supply
Typical welding transformers are shown in Fig. 2.3. and voltage (V0) current (IJ turns (N2)
Fig. 2.3a In an arc welding transformer, the number of primary
The basic principle of a transformer is as follows: turns is considerably larger than the number of secondary
If two electrically conductive coils, which are turns. This results in low output voltage and high current
electrically insulated from each other are wound on a required for welding. Heavy duty industrial machines
common core made up of a magnetic material and an are rated for 200 to 500 Amps, at 60% duty cycle and
Fig. 2.3: Three old models of welding transformers
Fig. 2.3a: New range of welding transformers

operate on 440 V mains supply, whereas those meant D) Moving core: Moving the iron core in the reactor.
for lower duty cycles are rated for 50 to 200 Amps, and E) Saturable reactor: Putting saturable reactor unit
operate on 220 V supply Heavy duty transformers used in the secondary circuit.
for automatic submerged arc welding are rated for 1,000
Diagrams explaining the principle of these methods
to 1,500 Amps, at 100% duty cycle.
are shown in Fig. 2.4.
The welding transformer essentially consists of:
Method (A) uses a tapped reactor, which does not give
• Main transformer assembly
a continuous current output demands for certain im-
• Frame portant applications. The limited number of taps re-
• Ventilating system (natural air cooling, forced air stricts the values of output current available. However,
cooling or oil cooling) this system is relatively efficient and suitable for gen-
• Current control mechanism eral fabrication and repair work.
The main transformer consists of primary and Method (B) changes the reactance of the transformer
secondary coils and a magnetic core made up from thin by changing the relative positions of the coils. Moving
laminations of special silicon-alloy steel. The laminated one coil away from the other increases the amount of
core results in low core losses and improved overall leakage flux flow between them, thereby increasing the
efficiency. The primary and secondary coils are made leakage reactance of the coils. This reduces the current
up of copper or aluminium. Aluminium coils provide output. Bringing the coils nearer results in increased
advantages of low w e i g h t and economy. As the current output. The change in positions of coils is
conductivity of aluminium is less than that of copper, brought about by a lead screw which facilitates continu-
conductors of heavier cross-section are used for ous adjustment of current. With such a design, the coils
aluminium windings. may at times get loosened and vibrate when the trans-
Natural and forced air-cooled transformers are former is in use, causing noise. The connections to coils
commonly used in industries. However, if the environ- may also create service problems because of continu-
ment is corrosive or if metallic dust particles are likely ous flexing.
to enter the coils in service, oil-cooled transformers are
preferred. Method (C) uses a different principle for changing the
Various mechanisms are used to control and adjust reactance of the coils. It employs a low magnetic resis-
the welding current to the desired value. In the early tance path which diverts part of total lines of force link-
days, a bank of very high wattage resistors was added ing with the secondary coils. The movement of this
in the output circuit. magnetic shunt causes the leakage flux to vary and
By using various parallel series combinations the thereby adjusts the output current. This method may
effective resistance could be changed resulting in also give rise to vibration of movable parts with
variation in the output current. This method of current attendent noise if the parts wear out and become loose
control is totally outdated mainly because considerable in service. Sometimes mechanical parts such as the lead
electrical power is wasted in the resistor bank and hence screw can become dirty and difficult to move. These
the efficiency is very low, and the resistor bank is bulky drawbacks can be taken care of through careful assem-
and heavy. bly and efficient workmanship in manufacture and
In the modern industrial welding transformers, through good maintenance at the user's end. On large
welding current is controlled by any of the following machines, the movement of the magnetic shunt can be
five methods: conveniently motorised.
A) Tapped choke: Tapping on a reactor in the Method (D) uses a moving iron core in the reactor in-
secondary circuit. stead of in the main core. The moving core changes the
B) Moving coils: Changing the magnetic coupling air gap which changes the reactance. The larger the air
between primary and secondary by physically gap, smaller the impedance and higher the output. The
changing the position of coils. remarks about vibration and noise mentioned in method
C) Magnetic shunt: Changing the magnetic coupling (C) above are also applicable to this method. The inside
between primary and secondary by putting a of a typical moving core type transformer is shown in
movable magnetic shunt. Fig. 2.5.
TAPPED REACTOR
tfnnfiM
SECONDARY
ELECTRODE
HOLDER
TRANSFORMER
CORE I I
WORK
LINE
2
PRIMARY (i) TAPPED REACTOR
PRIMARY ADJUSTING SCREW ADJUSTING SCREW
PRIMARY SECONDARY
TRANSFORMER
CORE ELECTRODE
LINE HC& HOLDER
ELECTRODE
HOLDER
SECONDARY
z: (ii) MOVING COIL
\
I
WORK
I
TRANSFORMER CORE
(lii) MOVING SHUNT CORE
/RHEOSTAT
W
°RK
^RECTIFIER /
ADJUSTING SCREW
AAJtv—4 ^SATURATED
REACTOR
REACTOR
PRIMARY
PRIMARY 7 *C
TRANSFORMER CORE WORK
(iv) MOVING REACTOR CORE (v) SATURABLE REACTOR
Fig. 2.4: Methods of controlling the output of welding transformers
Fig. 2.5: Interior of a typical moving-core type welding transformer

Method (E) eliminates all the moving p^rts with their transformers on different phases to achieve a somewhat
service problems, but is more expensive. In this system/ balanced condition. Even when theoretically valid, it is
secondary reactor impedance is controlled by regulat- impossible to achieve full balance in practice.
ing the saturation level of the core electrically. The sys- Though low-priced and simple in construction,
tem uses a rectifier bridge and a rheostat to control the welding transformers are robust enough to be used for
DC current in the control coil. When there is no DC cur- general-purpose welding in shops and in open site
rent flowing through the control winding, the imped- conditions. In special and critical welding applications
ance is maximum and the output is minimum. Reverse and where high welding quality is essential, DC power
is the case when the maximum DC current is flowing in sources are preferred to welding transformers. However,
the control winding. With this method, remote control there are certain applications where DC is unsuitable
of welding current is possible, i.e. the reactor current and hence transformer power sources must be used.
adjustment can be located near the welder by using an Examples are:
extension cord. The adjustment can be by the welder's • Welding of aluminium and magnesium and their
hand or foot. alloys using the GTAW process
Electrical circuit diagram of a typical heavy-duty • Tandem submerged-arc welding of heavy plate
welding transformer having maximum continuous in which the welding transformer is used as an
handwelding current of 300 Amps, at 60% duty cycle is additional power source
shown in Fig. 2.6. This machine uses moving core • Manual metal-arc welding applications where
magnetic shunt method of current control. DC poses the problem of arc blow.
As mentioned earlier, welding transformers are
usually single-phase and are connected across two lines Remote Current Control
of a three-phase supply. This introduces an imbalance In a welding operation, the welder needs to change
in the supply lines, which is unavoidable. Welding the welding current occasionally to suit a particular
transformers with three-phase input supply are also application. In site welding, where long welding cables
available. These are either constructed with scott are often used to reach the job or when the welder is up
connection or are of open-delta type, However, it should on a ladder to weld, it becomes extremely inconvenient
be noted that in such machines even with three-phase for him to walk up to the machine to change the current
input supply, the problem of line imbalance is not fully setting, The task becomes confusing when a number of
resolved. The imbalance problem can be minimised by transformers in service are grouped together. Remote
taking suitable steps while installing several control is an easy answer to this problem,
oc
oo B o
415 V 3 80 V
380/415 V
£i
\0 50><
PILOT
LAMP
B 0-» LoD
f
Fig. 2.6: Electrical circuit diagram of a heavy-duty welding transformer
Remote control can be achieved in the moving core • Welding current regulation by modern moving
a n d m o v i n g s h u n t designs by m o t o r i s i n g the core magnetic shunt design
mechanism and providing the operator with a pendent • Stepless, smooth and infinitely variable current
with push-buttons for increase/decrease. In these cases, regulation
a separate pendent cable has to run along with the
• Adjustment of welding current feasible even
welding cables from the machine to the operator and
while welding is in progress; this enables fine
this could be a disadvantage.
current control
In the most modern remote control transformer,
neither control cable nor p u s h b u t t o n p e n d e n t is • A special design provides for excellent dynamic
required. The operator keeps a small probe in his pocket. characteristics and thus ensures smooth and op-
When the current has to be changed, the operator fits timum metal transfer.
the probe in the electrode holder and touches the job.
The corresponding welding current signal is sensed by ARC WELDING GENERATORS
electronic circuitry to change the current to the desired The welding generator was the first power source
value. This form of remote control is particularly useful successfully developed for industrial use. It is a rotary
in shipyards. type machine driven by an electric motor or an internal
combustion engine. The latter is either a diesel engine
Popular Models or a petrol engine and is air-cooled or water-cooled. The
The small transformer shown in Fig. 2.3 is designed welding generator, whether driven by a motor or an
for connection to 230 V single-phase supply and gives engine, forms a single integral unit. A typical motor-
the welding current range of 50-190 Amps, and open- driven generator is shown in Fig. 2.7 and a diesel engine-
circuit voltage to 50 V. A moving shunt core method of driven generator is shown in Fig 2.8.
current control is used w h e r e b y stepless current The welding generator is a rugged and reliable DC
regulation is achieved. It is a light-weight AC welder machine which gives high quality welds and has many
weighing 33 kg. It is rated at 20% duty cycle and is meant other outstanding advantages which will be discussed
to be used intermittently for repair, maintenance jobs later. The engine-driven type is specially developed for
and light fabrication work in farms, small shops and use in the open air and at site where regular power
automobile garages. It is suitable for electrodes of 3.15 s u p p l y is not available. The w e l d i n g g e n e r a t o r s
to 2 mm diameter. commonly used in industry have outputs of 200 to 600
One of the machines depicted in Fig 2.5 is a welding Amps..
transformer for application with 2.5 to 6.3 mm electrode The welding generators which were developed and
range and working on single phase line to line 415 V manufatured earlier were basically DC generators with
AC supply. 60% duty cycle is 300 Amps., and current carbon brushes and commutators. Now welding gensets
range is 60-400 Amps.. Current is adjusted with a are available with latest Brushless Technology which
moving core mechanism to the required value. are enery efficient and also maintenance-free at the same
The transformer whose electrical circuit diagram time, as they do not have commutator and carbon
is shown in Fig. 2.6 is a typical heavy duty air-cooled brushes which are subjected to wear and tear. These sets
single-phase machine specially designed with OCV of now also have multi-process capabilities and becoming
90 V for welding with electrodes of the cellulosic and quite useful for site fabrication as well as cross-country
basic low-hydrogen types which normally require pipeline welding. Details of these sets will be discussed
such a high OCV value for good arc stability in AC later.
w e l d i n g . Its m a x i m u m c o n t i n u o u s h a n d w e l d i n g The generator works on the following principle:
current is 300 Amps., at 60% d u t y cycle, and the When a conductor moves in a magnetic field, it cuts
current range is 60-400 Amps.. Use of a capacitor of 8 the magnetic lines of force (flux) and, as a result,
KVAr is recommended. It is the best machine for heavy alternating voltage is generated in the conductor. This
continuous manual welding with 5 m m and 6.3 m m alternating voltage is then converted to direct voltage
heavy-coated iron-powder type electrodes and for TIG by means of a device called commutator and collected
welding of thick sections of aluminium, magnesium together by a set of carbon brushes to get the required
a n d their alloys. Like o t h e r m o d e r n w e l d i n g output.
transformers, it possesses the following attractive The generator is a complex piece of equipment as
features: seen in Fig. 2.9 which shows the cut-out internal view
Fig. 2.7: A typical motor-driven welding generator
Fig. 2.8: A typical diesel engine-driven welding generator
of the motor-driven type. In addition to its elaborate the main magnetic lines of force and thereby brings
electrical set-up, it is a perfectly mechanically balanced about a change in the output voltage. Additional
assembly fitted on a common shaft. magnetic poles known as interpoles are added in many
As illustrated in Fig. 2.9, the welding generator welding generators and they normally carry the welding
consists of a stationary frame called a stator, and a current. These are known as bucking field coils or series
rotating armature, called a rotor. The frame, also called fields. The welding current is passed through these coils
yoke, contains an extremely efficient magnetic circuit normally through taps or by means of a current range
consisting of various poles. Each pole is surrounded by selector switch. This provides a magnetic field in
a fine gauge wire coil and is so connected that the poles opposition to the main field, resulting in reduced output.
form a north/south combination. The power required This produces the necessary drooping effect and makes
to energise the poles is either taken from the exciter or the generator CC type. The selection of turns on the
the machine can be self-excited type, which makes use series winding decides the output current or the output
of the residual magnetism of the magnetic field. voltage. By changing the current in the field windings
The output voltage is controlled by the rheostat in by means of a rheostat, the output current can be set at
series with the main poles. This increases or decreases the desired value.
SELFHNDICATING CONTROL DIALS
SELF-LOCKING LUGS A
STARTING SWITCH' \ LIFTING
''HOOK
BURN PROOF
POLARITY SWITCH
SAFETY STARTING CONTINUOUS

PUSH BUTTON - "CURRENT CONTROL
'OB SELECTOR RHEOSTAT SAFETY ZONED" CONTROL

BOX
MOTOR STATOR
GENERATOR FIELD POLE PIECE
MOTOR FIELD COILS GENERATOR INTERPOLE
COIL
EXCITER FIELD POLE GENERATOR ARMATURE
MAIN
EXCITER FIELD
BRUSHOLDERS
COIL
BLOWER FAN
EXCITER BRUSHOLDER
EXCITER BRUSH
SEPARATE EXCITER OMMUTATOR

ARMATURE
INTAKE VENT EXHAUST VENTS

FOR FRESH AIR /
BALL BEARING \ ARC WELDED/

\ STEEL FEET /
WELDER LAMINATED
PROTECTIVE DEVICE GENERATOR FRAME
(A)
*V**" (B)
\ UT
Fig. 2.9: (A) Internal view of a motor-driven welding generator; (B) Superchallenger 400
Besides these interpoles and series windings, there necessary DC power to the main fields. It has also been
are other poles and their windings too. These are pointed out that generators of the self-excited type use
commutating poles and in many cases compensating the residual magnetism of the magnetic circuit to build
poles. Normally these provide the necessary aiding u p the necessary voltage. Such machines, unlike the
magnetic field of appropriate value to the main field to separately excited types, give a very steady output,
get a sparkless commutation. independent of the input supply voltage fluctuations
Figure 2.10 shows a typical circuit diagram of the and are therefore preferred where input supply voltage
self-excited CC type motor-drivern generator shown fluctuates erratically. The machine shown in Fig. 2.7 has
earlier in Fig. 2.7. this advantageous feature.
In a CV type generator, the differential series fields The carbon brushes which ride on the surface of the
are reversed so as to aid the main field. This arrangement commutator pick up the generated current and convert
makes the machine produce a constant voltage. it into DC. Cables fastened to the brush holders carry
As already explained, excitation for the generator the DC power from the brushes to the two output
can be obtained through a separate exciter. Alternately, terminals to which electrode cable and earth cable have
a solid state rectifier bridge can be used to provide the to be connected. In earlier designs, stabilising coils were
6 EXCITATION COILS ARE

FITTED ON POLE (1) AND
POLE (4)
5 R1 .REGULATOR
4 ST1 -REGULATOR SOCKET
3 S1 -OUTPUT TERMINALS
2 POLE POSITIONS ARE

SHOWN LOOKING FROM
COMMUTATOR SIDE
1 COLOUR CODING OF THE

CABLES IS SHOWN IN
BRACKETS
Fig. 2.10: Electrical circuit diagram of a motor-driven welding generator
placed externally; but modern machines are built with that it can be taken out of the control box and placed at
sufficient inherent inductance, thereby obviating the the point of w e l d i n g by e x t e n d i n g . t h e regulator
need to use them. connection cable. When the regulator is so placed, the
Since welding generators give DC output, provision welder can make current changes to suit electrode size
has to be made for the reversal of polarity to suit welding and welding position without having to walk up to the
conditions. In most cases, this is done through a reversing generator.
switch or by physically changing the output connections In order to obtain current values higher than those
(i.e. electrode cable and earth cable). The output terminals permitted by the rating of a single welding generator,
are marked +ve and -ve or electrode and work. The one can connect two sets in parallel. For the purpose of
reversing switch changes the direction of the main field equal distribution of the load, it is absolutely essential
and thereby changes the output polarity. to set the welding current regulators to equal values.
Engine-driven generators can be designed for any Taking as example, the case of two machines of the type
rotational speed, but speeds between 1,500 and 3,000 shown in Fig. 2.7 to be connected in parallel, the
rpm are normally preferred. Motor generators rotate at following procedure has to be followed:
a speed of either 1,500 or 3,000 rpm, depending on motor
design. The motor is preferably a three-phase induction 1) Connect the welding sets to the supply lines.
motor, squirrel cage type, since it has the advantages of
2) Short-circuit the small b r u s h sets on the
high starting torque and ruggedness. The use of three-
commutators of the machines, preferably through
phase induction motor also ensures a balanced load on
their leads from the regulator sockets.
the supply lines. This is another advantageous feature.
It is important that the direction of rotation of the 3) Interconnect (short-circuit) terminals marked MAX
motor is in accordance with the design of the generator. on the machines by welding cable.
Generators are marked with arrows indicating the 4) Set the regulators to their minimum welding
direction of rotation. If the rotation is opposite to the arrow current position.
pointer, the two leads of the supply to the induction motor 5) Interconnect (short-circuit) the terminals marked-
should be reversed. Normally these motors are provided 6) Start the welding machines.
with suitable star-delta starters to limit the starting
current. These starters also have thermal overload 7) Set the regulators to the required current value.
devices, which trip the input supply to the motor in case (Regulators must be in identical position.)
the temperature exceeds the safe limit. 8) Now use any two terminals marked MAX and -
Remote current control is possible with a generator, for welding work.
provided the current regulator is of detachable type, so The method of interconnection is shown in Fig. 2.11.
Fig. 2.11: Sketch of parallel operation of motor-driven welding generators
Popular Models The machine has maximum OCV of 90 and two

The DC motor-driven generator shown in Fig. 2.7 current ranges: 35-130 Amps, and 110-320 Amps.. It is
and whose wiring diagram is shown in Fig. 2.10 has the rated for maximum continuous handwelding current
following salient features: of 300 Amps, at 60% duty cycle and 320 Amps, at 55%
• Specially designed to yield fine dynamic and duty cycle. The motor has rated capacity of 16 KW at
static characteristics which help easy striking and 60% duty cycle and has rotational speed of 2,850 rpm.
excellent stability of arc throughout the operat- E-class insulation is used. All electrode sizes from 1.6 to
ing range. 6.3 mm can be used with this machine.
• Versatile enough to be used with all types of Both motor and generator are mounted on an
MMAW electrodes and carbon electrodes. integral shaft running between two bearings. It is
mounted on three rubber-tyred wheels to give mobility.
• Positive protection against overload and single
It is drip-proof and designed to withstand tropical
phasing is achieved through thermostats embed-
conditions.
ded in the motor winding.
The diesel engine-driven generator illustrated in Fig.
• Interlocking against direct starting of the set in 2.8 consists of a generator, designed to give welding
delta position and continuous running of the set c u r r e n t r a n g e of 50-350 A m p s , a n d m a x i m u m
in star position. continuous hand-welding current of 300 Amps, at 60%
• Current variation is continuous over two ranges. duty cycle, which is connected to a simpson P3 (1) three-
• Suitable for connection to a 400/440 V, three- cylinder diesel engine of the water-cooled type. The
phase, 50 Hz, electrical system. Models for other engine generates 33 BHP at 2,000 rpm. To ensure perfect
voltages up to 550 V and frequencies up to 60 Hz alignment, smooth running and easy maintenance, the
can also be supplied on specific orders. generator is flange-mounted and directly coupled to the
• Designed to load all three phases of the supply engine through a flexible coupling. The machine can be
equally. provided with remote control current regulator, an
ammeter and a voltmeter. The entire engine-generator
• The detachable regulator provides for remote con- is mounted on a trailer of sturdy steel construction,
trol of welding current when connected to the which travels on two or four pneumatic rubber-tyred
generator by a 2 x 1.5 mm 2 cable. wheels. It has a towing hook and is furnished with a
• The regulator weighs only 2 kg and is therefore canopy with hinged doors to protect the equipment from
suitable for remote control. unfavourable weather conditions.
BRUSHLESS WELDING SETS suitable for welding with longer welding cables
As m e n t i o n e d above b r u s h l e s s type w e l d i n g exceeding even 100 metres
generators are available both in motor generator types 6. These welding gensets are normally provided
and engine driven types. We will discuss each separately. with electrical protections such as Undervoltage/
Overvoltage/Single phasing/Overload etc. so
A. BRUSHLESS TYPE MOTOR GENERATOR SETS that these machines become a reliable welding
As the the name suggests, these welding sets are sets
driven by Induction motor. The synchronous speed of
Parallel Operations of Welding Gensets
the motor is 3000 rpm but however size of the motor is
These gensets can be used in parallel like the old type
compact compared to the motor of old technology
of sets to get higher welding current.Thus using 2
gensets, as the insulation of motors is F class. These
generators of 320 Amps, in parallel combination, user
sets are with monoblock construction in which motor
can get welding current upto 640 Amps, and thus such
and generator rotors are mounted on single shaft. Weld-
combination can be used for gouging applications.
ing generators are either on the principle of "Synchro-
Single remote control is available to control or set the
nous" or "Reluctance" types. Rotors of Reluctance gen-
current from the machine itself.
erators are winding less . Thus generator has only sta-
tor windings which are further divided in to field wind-
ing and armature winding. Field winding or the excita- B. BRUSHLESS TYPE ENGINE-DRIVEN WELONG
tion winding of the machine sets the magnetic field in SETS
the air gap. Rotor cuts this magneitic field while it is On the similar line of motor generator sets, Brushless
rotating. Thus varying magnetic field is produced due Type engine-driven sets are also available in welding
to variation in the air gap reluctance. This varying mag- current range starting from 150 Amps, to 600 Amps..
netic field induces voltage and current in the stator The working principle of these sets is same like motor
winding.The frequency of this voltage and current de- generator sets. However there are quite good amount
pends on the speed (3000 rpm) and nos of armature of varieties of types of sets available starting from
winding poles. This electrical welding power thus gen- P e r m a n e n t m a g n e t to Synchronous generators,
erated is AC but with high frequency. This AC welding A s y n c h r o n o u s I n d u c t i o n generators, Reluctance
power is rectified to D.C. which is absolutely ripple- generators etc. Out of these different types of gensets,
hee as the AC side power is of high frequency. The weld- Reluctance and Induction generators are most reliable
ing current is controlled and regulated by regulating sets. Most common ratings are 150, 175, 225, 300/320,
the field or excitation current. (Refer generator Fig.2.7) 400, 500 & 600 Amps. (Refer Fig. 2.8.)
These brushless welding generators have following Some of the machine ratings are available with C C /
a d v a n t a g e o u s features over earlier sets w i t h CV characterstics. Thus these sets with multiprocess
commutators and brushes. capabilities can be used for SMAW (Stick), TIG, GMAW
& FCAW welding processes. Some manufacturers also
1. Windingless and magnetless or rotating diodeless manufacture these welding sets with MOSFET based
rotor improves overall efficiency of the genset as technologies, in which case entire generator and current
the rotor winding losses, Brush friction and wind- regulator systems are very efficient and a Multi-operator
age losses are absent in these sets. welding sets also become possible with the same Horse
2. Overall power savings upto 30 % over old type Power of Diesel Engine.
of welding gensets Thus double operator welding set with facility of 2
3. Enclosure protection is IP44 instead of IP23, hence operators at 300 Amps, each are available with diesel
stator windings are well protected against dusty, engine of 47.5 BHP rating.
dirty and contaminated air, making them ideally
suitable for site usage Auxiliary Power Source
4. High frequency type generator gives ripple-free These engine-driven welding sets are invariably
smooth DC welding current resulting excellent built in with auxiliary power sources which are used
welding performance w i t h good d y n a m i c for operating grinding tools, portable drilling machines,
response lightening purposes etc. The requirement of such
5. Since welding current regulation system includes auxiliary power is very essential since these welding
current feedback circuit these generators are well sets are used in remote project sites where electricity is
not available. Most common ratings of auxiliary power phase) type, which converts the mains supply to a low
source are ranging from 2 KVA to 20 KVA. Smaller voltage supply on the secondary side. On a three-phase
capacity portable ED sets have Auxiliary rating from 2 machine, the primary is connected in delta or star, but
to 4 KVA, 1-Phase and medium to large rating ED sets the secondary is connected in delta, because this
have auxiliary ratings from 5 KVA to 20 KVA in 3 Phase connection is more convenient for low voltage and high
(415 volts) and 3 to 15 KVA in single Phase (230 volts) current. The method of controlling the current is usually
supply conditions. It is possible to use these auxiliary in the AC section between the transformer and the
power sources for operating small water pumps at the rectifier set. The welding rectifier can be designed for
site since these power sources are available with 3 Phase constant current (CC) as well as for constant voltage
supply. It is also possible to use smaller current ratings (CV). CC type rectifiers meant for manual welding are
inverter based Rectifiers or TIG units on these auxiliary usually rated for any value between 200 and 600 Amps.,
power sources. In this way also these ED sets become while the CV types meant for automatic welding are
multioperator in the sense that two welding points i.e. rated between 300 and 1,500 Amps.
one from main welding generator a n d other from
auxiliary power source are available from these sets. CC Type Rectifier
These modern brushless ED sets have following In this type, the current is controlled through variable
advantages over old generation ED sets i n d u c t a n c e or i m p e d a n c e . T h e v a r i o u s m e t h o d s
1. Unique high frequency brushless design makes available for varying the impedance are: (a) moving coil,
these ED sets maintenance free as there are no (b) moving shunt, (c) saturable reactor or magnetic
carbon brushes,windings and commutator on the amplifiers, (d) tapped reactor, (e) moving reactor core
rotor which are subjected to wear and tear. and (f) solid state. The methods (a) to (e) are the same
as explained under transformer current control methods
2. Windingless, Magnetless, Rotating diodeless ro-
and illustrated in Fig. 2.4.
tor assembly makes these machines reliable and
For converting AC to DC, rectifiers are used. A
ideal choice for project site usage.
rectifier is a device which conducts easily in one
3. These ED sets are with F insulation class as com- direction, while offering a high resistance in the other
pared to earlier sets with E insulation class gen- direction. Today silicon diode rectifiers are preferred to
erators selenium or germanium rectifiers, because they have
4. Welding sets being brushless are efficient and greater rectification efficiency and are not crippled by
overall operating costs are less due to reduced aging as the latter types are. Silicon diodes are smaller
fuel consumption of the engine and lesser main- than selenium diodes and are hermetically sealed and
tenance of welding generator mounted on a suitable fin for cooling. Silicon diode
5. Single bearing, closed couple type assembly con- assembly and its parts are shown in Fig. 2.12. In most
struction with life-long stainless steel flexible cou-
pling arrangement for brushless generator makes
the entire assembly of engine and generator very
compact.
6. Auxiliary power source is also brushless and cffifrgftSaaâ—i^j
mounted on the same shaft or integral with main
welding generator.
7. Choice of waterôoled or air-cooled engine is (A)
available and BHP of engine depends on current
rating of the set.
ARC WELDING RECTIFIERS

The welding rectifier, like the welding generator,
provides DC for welding b u t unlike the welding
generator, it has no moving parts. It consists of a step-
down voltage transformer with means to rectify AC to Fig. 2.12: Silicon diode rectifier: (A) Internal
DC. The transformer can be single-phase (or three- components; (B) Final finished form
recent designs, silicon diodes have been replaced by SCR rectifier has, as stated earlier, a three-phase transformer,
(Silicon Controlled Rectifiers) or thyristors. in which the saturable reactor method of current control
Figure 2.13 shows the diagram of a three-phase is preferred.
rectifier using saturable reactor method of current The smoothness of the DC output of a rectifier
control. Such a three-phase type ensures a balanced load depends on the rectifier circuit and filter circuit used.
on the supply lines. Figure 2.14 shows typical single-phase and three-phase
It is usually necessary to use an inductance (also rectifier systems. A close look at this figure shows that a
called reactor, stabiliser or filter choke) in the DC three-phase full-wave rectification provides a vastly
welding circuit to control excessive surges in load smoother output necessary for a welding power source.
current. The value of inductance required for a three- Output can be further smoothened by using inductors
phase machine is much smaller than that required for a a n d capacitors in the o u t p u t circuit. Machines
single-phase machine. incorporating all these provisions have very low ripple
Though, as stated earlier, the welding rectifier is content and therefore give far superior performance
primarily a DC machine, it can be suitably modified to compared to those using a single-phase rectification.
give AC/DC output. The AC/DC type is provided with
a switch to connect or disconnect the rectifier to the CV Type Rectifier
secondary circuit to furnish either DC or AC at the This type is usually designed to give a slightly
output terminals. It is capable of welding the entire sloping, instead of a perfectly flat, volt-ampere curve.
range of metals and alloys by the MMA process, as it While some machines are designed with one fixed slope,
enables the operator to choose either AC or DC others are provided with an adjustment to tailor the
depending on the electrode he uses. This type is even slope of the V-A curve as desired. Slope adjustment must
more versatile as a power source for TIG (argon-arc) take care that the dynamic characteristics of the machine
welding, in which AC is required for w e l d i n g are not affected. For this purpose, adjustable inductors
aluminium /magnesium and their alloys and DC is may be placed in the DC portion of the circuit.
required for ferrous metals and for copper and its alloys. In CV machines, the slope is usually achieved by
The A C / D C type incorporates a single-phase changing taps on reactors in series with the AC part of
transformer, often using the moving shunt or saturable the circuit. Slope control may be provided by carbon
reactor method of current control. The purely DC brushes, attached to a lead screw, contacting the reactor
TRANSFORH DC
AC- CONTROLS RECTIFIER
MER "OUTPUT
TRANSFORMERS SATURABLE REACTORS
I
SECONDARY COIL LOAD COIL CONTROL COIL |
PRIMARY COIL
3 PHASE
POWER
CONTACTOB
CONTROL RECTIFIER
7 — {oj A \ )
v
CONTROL TRANSFORMER ' \ VARIABLE TRANSFORMER
Fig. 2.13: Schematic diagram of a three-phase welding rectifier
RECTIFIER CIRCUIT D.C OUTPUT

A.C. INPUT o to
A A
SINGLE PHASE HALF WAVE RECTIFICATION
AAAA
SINGLE PHASE HALF WAVE RECTIFICATION
12 3 12 3 1
»-^W1N« to—4 / W Y V W ^
THREE PHASE HALF WAVE RECTIFICATION
12 3 12 3 1 S3H
t* Tw t +
THREE PHASE
p^r 0
FULL WAVE RECTIFICATION
Fig. 2.14: Sketches of simplified rectifier systems
turns. This variable reactor p r o v i d e s continuous which makes it easy for the operator to set the correct
adjustment of slope. Another method of control uses parameters and obtain good welds.
magnetic amplifiers or solid state devices to electrically
regulate output voltages. These machines may have Popular Model
either voltage taps or slope taps in addition to electrical The machine shown in Fig. 2.15 is a heavy duty three-
control. p h a s e constant current transformer-rectifier for
CV machines have a wide range of open circuit connection to 440 V three-phase 50 cycles main supply,
voltages. Tn electrically controlled machines, it may be with welding current range of 50 to 400 Amps, and
as high as 75 V. In the case of tapped or adjustable suitable for manual metal-arc welding. Its salient
transformer types, open circuit voltage may be varied features are:
from 30 or 50 V maximum to 10 V minimum. While • Excellent welding characteristics: the use of a
there is no fixed rule for the volt-ampere slope in the three-phase DC power source combined with a
welding range, most machines have slopes of from 1 to well designed transductor gives a smooth and
3 V per 100 Amps.. steady arc with minimum spatter, resulting in a
smooth weld bead.
CV Type with Automatic Choke • Easy arc striking: The machine is designed to give
In CV type machines, the value of choke decides the easy arc striking with any type of electrode.
a m o u n t of slope. In M I G / M A G w e l d i n g , the
smoothness of droplet transfer into weld puddle is • Built-in safeguards: The set is safeguarded against
determined by the choke value, and this value varies fan failure overload or inadequate cooling. Spe-
according to current a n d voltage settings. In a cial surge suppressor devices are used for pro-
conventional p o w e r source a t a p p e d choke is tecting the control circuitry from surges.
incorporated, in which case the operator has to arrive • Reliability: M a d e to latest s t a n d a r d s , the
at the optimum choke tap by trial and error. Power equipment can withstand rigorous environ-
sources i n c o r p o r a t i n g automatic choke are n o w i mental conditions.
available w h i c h set the r e q u i r e d choke value Light and compact: The equipment is very compact and
automatically. This is achieved by feedback circuitry, light in weight, easy to manoeuvre, transport and handle.
Fig. 2.15: Popular models of welding rectifiers
The machine has OCV of 80 V and m a x i m u m Transformers require least maintenance because of
continuous hand-welding current of 400 Amps, at 60% their simple construction and absence of moving parts.
duty cycle. It has B-class insulation, forced air cooling Generators call for careful maintenance, because some
and drip-proof casing. Capacitor of 9 KVA is of their m o v i n g p a r t s require periodic repair or
recommended to be used to improve the power factor. replacement. When it comes to the cost of electric power
Remote control unit can be provided, so that the welder for current output, the motor-generator is the most
can adjust the current from the work place itself. expensive, because heavy parts are continuously rotated
even w h e n w e l d i n g is not being performed.
COMPARISON OF POWER SOURCES Transformer-rectifiers and transformers operate more
It is important to know the merits and demerits of cheaply, because they have no moving parts except
the three main classes of power sources—the welding cooling fans and, what is significant, their no-load power
transformer, the welding generator and the welding loss is negligible.
rectifier—so that a correct choice can be made according Figure 2.16 compares the efficiencies of the machines
to the conditions prevailing in the fabrication shops. at various operating loads and shows that even at high
The comparative advantages of AC and DC welding duty cycles, the efficiency of the motor-generator does
will be pointed out in Chapter 3. Based on them, one not approach that of the other two types of power
would first arrive at a decision whether to use AC sources. Such efficiencies are measured at rated load
(welding transformers) or DC (welding generators or with arc or resistance loads, and therefore represent
rectifiers or inverters). actual operating conditions.
Transformers have much lower initial costs than DC In spite of these drawbacks which are relatively
power sources. Today's market prices for typical 300 insignificant, DC machines are preferred because of their
Amps, machines of standard make are as follows: ability to weld a wide range of base metals, their better
arc stability, choice of polarity (positive or negative) and
Welding transformer Rs. 20,000.00 other advantageous features mentioned under AC
Welding generator Rs. 1,00,000.00 versus DC in Chapter 3.
Welding rectifier Rs. 60,000.00 Electrode burnoff rate, i.e. the rate at which an
Welding inverter Rs. 90,000.00
electrode of a given size and type gets consumed at a
1) Initial cost
2) Cost of periodical maintenance and repair
3) Whether the mains supply is 220 V or 440 V, or
no electrical power is available
4) Whether the machine will give steady current
output in spite of the main voltage fluctuations
5) Whether the machine will cause imbalance in the
power load
6) Inherent power factor of the machine—whether
20 40 60 80 100 120 capacitor will be required to raise it
PERCENTAGE OF RATED LOAD 7) Whether the welding jobs demand a portable unit
8) Whether the welding jobs require AC or DC or
Fig. 2.16: Effect of operating load (as a percentage of
both AC and DC
rated load) on efficiency of three types of MMAW power
supplies 9) Whether the machine has adequate current rat-
ing and range to weld with the largest and small-
est size of electrode to be used on the job
specific current value (expressed in mm per minute), 10) Whether the machine will give good arc striking
has been found in many cases to be the highest with the and arc stability for the types of electrodes to be
motor-generator and lowest with the welding used
transformer. The rate for the welding rectifier is 11) Whether the welding process to be used demands
intermediate between the two. Weld metal deposition constant current or constant voltage type
rates also vary in the same manner. A study with various 12) Whether the machine will give the required weld
sizes of medium-coated rutile-type electrodes has shown radiographic quality and impact values with the
that the average deposition rate with the motor- types of electrodes to be used
generator is 8.8% greater than with the rectifier, and
13) Whether the same machine could serve several
15.1% greater than with the transformer.
welding processes which are expected to be used
Mention has been made of the low power factor (pf) in the shop
of a welding transformer. The pf of a typical transformer
can be as low as 0.45, and this applies equally to a 14) Whether remote current control arrangement will
welding rectifier, which is basically a transformer with help, and whether the machine can provide it
a means of rectification of the secondary output. Both 15) Whether the machine will withstand shop or site
of them need capacitors to raise the pf above 0.80 as environment (corrosive gases, dust, moisture,
demanded by electric supply authorities. On the other etc.)
hand, the inherent pf of a motor-generator is between
0.80 and 0.90. SPECIAL POWER SOURCES
It has also been pointed out that welding So far the conventional power sources used for
transformers and welding rectifiers are very sensitive normal welding applications have been discussed.
to fluctuations in supply voltage, while generators of Today, considerable sophistication has been introduced
proper design are affected to a negligible extent. A 5% in power source design to meet the requirements of
change in supply voltage will be immediately reflected automation, accuracy and response during welding, and
on welding current side to the extent of about 7% in the compactness and reduced weight. It is impossible to deal
case of transformers, but only 1% in the case of with all these developments here. Hence, we shall only
generators. Transformers also present the problem of cover in brief the following types of special power
load imbalance, which is absent in the case of generators sources:
and three-phase rectifiers. 1) Universal type giving AC and DC outputs at
constant current as well as constant voltage
FACTORS FOR SELECTION OF POWER SOURCE
2) Multi-operator type giving AC or DC output
The following factors must be considered in choosing a
power source from among a transformer, generator and 3) Solid state power source:
rectifier: a) thyristorised and transistorised machines
b) rectifier with slope control for MMA welding 3) Solid State Power Source
c) transistorised/IGBT based (using inverter tech- a) Thyristorised and transistorised machines: Under a
nology to make the machine compact and light broad category, these can be termed as solid state or
weight) electronically controlled machines. These are DC power
sources, incorporating many modern sophisticated
4) Special power source for TIG and plasma arc weld-
techniques. They have the following distinct advantages
ing
over the conventional ones:
5) Special power source for MIG/C02, welding • Constant output: The output remains absolutely
constant irrespective of wide variations in the
1) Universal Type main supply voltage (normally fluctuations
This is generally a DC type power source having within ± 10 to 15% do not affect the output).
facility to obtain constant current as well as constant • Flexible output characteristics: It is possible to con-
voltage type output characteristics. Its main advantage trol and vary the static characteristics of the
is that it can be used for various processes like MMAW, power source by merely changing the control cir-
MIG/C0 2 , TIG and SAW, when coupled with suitable cuitry. This helps to get the desired characteris-
control outfits and accessories. It is generally rated for tics for various applications from a single power
600 Amps., so that it can cover practically a wide source.
spectrum of applications. This power source can be of
• Fast response: It gives very fast response to changes
conventional design (transformer-rectifier) or solid-state
design with feedback system. With the latter, one can in the output conditions, which is especially ad-
obtain the variable output characteristics easily and vantageous in pulsed mode applications.
economically. In a medium sized workshop where the • Higher efficiency: Overall efficiency is higher, as
fabrication programme demands use of several welding also the power factor. Hence, generally power
processes, a single universal power source can be used factor improving capacitors need not be added.
with advantage. • Easy remote control: The output can be varied
merely by varying reference control signals
2) Multi-operator Type through control gear circuitry; hence these power
When a large number of welders are concentrated sources can be readily used for mechanised and
in a small area, as in the case of shipyards and automated applications.
construction sites, it is economical to use a multi- Compared to thyristorised machines, transistorised
operator set capable of providing current output to 3,6, ones give steadier DC output with ripple as low as 1 to
9 or 12 operators at a time, instead of using individual 2%. The latter are also claimed to possess other attractive
units. The multi-operator power source can be either a features. However, they are very expensive. Today,
transformer giving AC output or a rectifier giving DC thyristorised power sources have been industrially
output. The overall maximum current rating of such an accepted, especially for mechanised and robotic
equipment can range between 1.200 and 2,500 Amps., applications.
catering currents up to 250 or 300 Amps, to each welder. All solid state DC power sources work on the
Each welder has a separate remote control regulator to principle of feedback control. Figure 2.17 shows a
enable him to set the current as required. simplified schematic diagram of a solid state power
The economy of a multi-operator unit is based on source incorporating the feedback control system. Its
the fact that in a large group of welders, only a few are principle can be explained thus:
actually welding at any given instant; and, secondly, The input voltage is first stepped down to low
the arc time factor varies between 10 and 30%. Arc time voltage. This low voltage is rectified by a thyristor bridge
factor is defined as the percentage of time the welder is rectifier. When the operator starts welding, a small signal
actually welding against the total shift time. The number proportional to actual welding current is fed into an
of arcs that can be used with the total welding power is electronic comparator unit. The comparator compares
given by the following formula: this feedback signal with the reference signal,, which is
Total welding power proportional to the set current. The difference between
Number = the two signals makes the error signal. If the output
of arcs Ave. amp of each arc x Ave. arc time factor of each arc current is higher than the set current, the error signal is
INPUT
I CONTROL MAIN
TRANSFORMER TRANSFORMER
V REGULATOR
1 CONTROL
[ SUPPLY
t THYRISTOR BRIDGE
ELECTRONIC CONTROLLER RECTIFIER
r^
-V ERROR OR DIODE BRIDGE
y TRANSISTOR
BANK
i i
k
REF. SIGNAL
a SET
CURRENT FEEDBACK SIGNAL
o ACTUAL
CURRENT
'r 5rur»KX
OPERATOR £}*"***
CONTROL
Wl
t f RISTICS
ELDING CHARACTE NTROL
OJ
RRENT SLOPE CO
6
SHUNT
o
6
WELDING OUTPUT
Fig. 2.17: Schematic diagram of a solid state power source
negative, and if it is lower than the set current, the error is that the performance characteristics of any given
signal is positive. Depending on the error signal the power source in a particular welding application cannot
firing angle of the thyristor gets changed to make the be predicted.
error zero, which in effect makes the output current For example, a power source may give very good
come back to the set value. If the feedback signal is result in downhand welding, but may not perform
proportional to the output voltage, then output voltage satisfactorily in vertical up and overhead welding. In
also can be controlled in the same way. some cases, if the short-circuit current at a particular
In transistorised versions, base current is controlled setting is too low, the electrode may give sticking effect;
in the same way as firing angle in order to maintain on the other hand if the short-circuit current is on the
constant output. higher side, it will give a forceful and digging arc. Some
b) Slope-controlled rectifiers for manual welding: Though power sources may give good arc striking, while others
conventional d r o o p i n g or CC type transformers, may not.
generators and rectifiers are used satisfactorily for Such varying performances are due to divergent
manual metal-arc welding, their static characteristics static characteristics of the power sources. To illustrate,
vary from type to type and from unit to unit. The result let us take two power sources A and B, A being more
drooping than B. A will give relatively less current

variation (say 15 Amps.), with change in arc length. This
power source will deposit a more uniform weld bead
with uniform penetration and will be well-suited for
welding thin sheet. B will give large current variations
(say 50 Amps.) with changes in arc length, and hence it
will give burnthrough if used for welding thin sheets.
However, B will be better suited for vertical and
overhead welding, because in this case the welder can
wilfully vary the current substantially, by making subtle
changes in the arc length, thereby controlling the volume
of molten metal in the weld puddle.
It is evident that for best all-round results, the welder
should be able to adjust the slope as desired. Recently
developed thyristorised rectifiers provide the facility to
select the optimum characteristic to suit a particular
application. The welder can select truly constant current
(i.e. absolutely vertical) characteristic or add to it as
much slope as required.
In some of these thyristorised rectifiers, a booster
circuit is incorporated to i m p r o v e arc striking.
Sometimes a safety circuit is added, which results in
very low OCV of 10 to 20 V in the idle condition, and
very high OCV of 100 to 120 V just at the time of striking
the arc. The result is excellent arc striking feature.
c) Inverter based machine for MMAW: This is a
Fig. 2.18: Inverter based rectifier
revolutionary power source using the latest high
frequency transistor inverter technology. It is five times
smaller in size a n d ten times lighter t h a n the DC supply is fed to either IGBT devices or MOSFET
conventional type and yet rated for industrial duty cycle devices to invert the DC supply to AC supply but at
and high output current. very high frequency up to 100 KHZ.The high frequency
These p o w e r sources n o w have become quite supply obtained thus is fed to step down transformer
popular and well established across almost all to reduce the voltage to the welding voltage.The size of
industries.They are available in the welding current transformer becomes very compact because of the high
range starting from 120 Amps, to 600 Amps.. Machines frequency supply at the primary side. The step down
up to 220 Amps, are basically single phase machines voltage is further rectified and filtered to DC supply as
and machines above the current ratings are three phase shown in the circuit diagram.The current feedback and
machines. Apart from light weight and compactness of control circuitry ensures that w e l d i n g current is
these machines they have higher efficiency (>85%) and m a i n t a i n e d constant to the set value on the
power factor (upto .98) compared conventional diode potentiometer scale.
based or thyristor based machines smaller rating
machines upto 150 Amps, current rating have become Inverter Based Multi Process Power Source Units
very popular for the maintenance work since they are Inverter based welding power sources are available
very light weight up to 5 kgs and can be carried any with CC/CV characteristics with the welding current
where conveniently for the welding.These machines are rating from 100 Amps to 600 Amps, and these are
either IGBT based or MOSFET based and working suitable for multi-process applications like SMAW,
frequency of devices is from 15 KHZ to 100 KHZ. (Refer GMAW, FCAW, GTAW & SAW processes.The use of
Fig. 2.18) Typical circuit diagram of the machine is inverter circuit produces significant reductions in size
shown below (Fig. 2.19). The input supply is rectified and weight of the magnetic circuitry components as well
through primary rectifier and the rectified DC output as decrease their electrical losses. Faster response times
is filtered as shown in tihe circuit diagram.This filtered are generally associated with the higher switching and
mjjLLi^
INPUT
BRIDGE
ra OUTPUT
BRIDGE
RECTIFIER INVERTER TRANSFORMER RECTIFIER INDUCTOR
14>0R3<J) ■ -o +
PRIMARY •
-w— LuuJ
-o-
llNVERTER
CONTROL
CIRCUIT
Fig. 2.19: Inverter diagram showing power supply sections and voltage wave forms
control frequencies used in this technology, resulting in time, peak current, background current, etc., can be pre-
more stable arc and superior arc performance. An set in a special module, and such a readily programmed
inverter based power source is smaller, more compact, module can be fitted in the power source to weld that
requires less electrical energy than conventional welding particular job. In more advanced and sophisticated
power sources, and offers a faster response time. power sources, such programmes can be processed
through a microprocessor. The zrelevant parameters can
4) Power Sources for TIG and Plasma Arc Welding be stored in memory bits of the microprocessor and can
Though any drooping or CC type power source is be obtained as and when required to weld a particular
suitable for these processes, machines having OCV of job. Interfacing of such microcomputer controls is pos-
70-80 V are preferred, because they give a more stable sible only with electronic power sources and not with
arc and better performance. Special power sources are conventional power sources.
also available, in which the following parts are Conventional AC power sources, used for the TIG
integrated: welding of aluminium, magnesium and their alloys, give
(a) high frequency unit for automatic arc striking; sinusoidal output waveform and require DC suppressor
(b) DC suppressor capacitor bank for AC welding of capacitor bank as mentioned above. Recent research has
aluminium, magnesium and its alloys; and (c) water shown that square wave AC output gives a very stable
circulatory system to cool the torch. In the plasma and smooth arc and eliminates the need of suppressing
welding power source, a separate circuitry is added to DC. Power sources designed to give square wave AC
provide the pilot arc. In the most up-to-date TIG/ are preferably electronically controlled.
Plasma power source, controls for pre-flow and post-
flow of shielding gas, and for up-slope and down-slope 5) Power Sources for MIG/C0 2 Welding
of welding current at the start and finish respectively, Though the conventional CV or flat characteristic
are provided. In addition, facilities like pulsed current type DC power source is suitable for MIG/C0 2 welding,
and programmed parameter setting are incorporated. regular users of this process prefer the composite power
All these features are aimed at consistently perfect source with built-in wire feeder and allied controls.
welds. Recently, one knob C0 2 welding machines have been
To suit any specific TIG/Plasma welding job, the introduced in the market, which simplify the setting of
various parameters like pre-flow, post-flow, up-slope welding parameters.
The most recently developed are the pulsed-MIG/ that these power sources give negligible spatter, high
C 0 2 power sources of various designs. Some have deposition rates and good welding even at currents as
pulsed wire-feed systems, while others have the facility low as 40 Amps.. Some of these power sources are
of impulsed current at fixed frequencies like 25, 33, electronically controlled a n d p r o g r a m m e d and
150 Hz, etc. The most recent design known as Synchro interfaced with the microprocessor. Various sets of
Pulsed MIG incorporates a closed loop feedback parameters pertaining to different types of base metals,
system, which selects correct pulse frequency to get joint configurations, joint thicknesses, etc., are stored
ideal metal transfer of one drop per pulse. It is claimed in memory bits.
Welding Power Handbook by A.F. Manz, published 1973, by Union Carbide Corporation and reprinted by American Welding
Society.
Welding Handbook, Chapter 1, Vol. 2, 7th Ed., published by American Welding Society.
Modem Welding Technology by H.B. Gary, published 1979, by Prentice-Hall Inc., N.J., U.S.A.
Metals Handbook, Vol. 6, 8th Ed., published 1971, by American Society for Metals.
IS: 1851, Single Operator Type Arc Welding Transformers, published by Indian Standards Institution.
IS: 2635, DC Electric Welding Generators, published by Indian Standards Institution.
IS: 4559, Specification for Single Operator Rectifier Type DC Arc. Welder, published by Indian Standards Institution.
IS: 6008, Single Operator AC/DC Arc Welding Power Source Incorporating Solid State Rectifiers and Having Drooping
Characteristics, published by Indian Standards Institution.
IS: 8804, Plugs, Sockets and Terminal Arrangements for Manual Arc Welding Equipment, published by Indian Standards
Institution.
Manual
Metal-Arc
Welding
AS pointed out in Chapter 1, the various arc welding from perfect, because it used bare wires as electrodes.
processes shown in the process chart (Table 1.1) find In the U.S.A., the use of sul-coated and lime-washed
wide application in industry and of these, the manual bare electrodes resulted in a marginal improvement as
metal-arc welding (MMAW) is the most extensively they provided a comparatively more stable arc. Sul-
used process, because it requires low capital investment, coated wires had a special rust-coated finish, which was
is easy and economical to use and can be applied on a obtained by maintaining the coils in a spray of water
wide variety of metals and jobs. before the final drawing. Lime-washed wires had a thin
layer of lime, used as a lubricant in the lime or dry
Historical Background drawing process.
Electric arc was discovered by Sir Humphrey Davy Oscar Kjellberg of Sweden is credited with being the
in 1801. The idea of using the arc for joining metals was pioneer of modern coated arc welding electrodes. He
born 80 years later in 1881, when Auguste de Meritens first conceived the idea of applying mixtures of organic
successfully welded lead storage battery plates using a and mineral coatings on wires to effectively control the
carbon arc. It was further d e v e l o p e d by several arc and weld metal characteristics. In 1907, he was
scientists, among them a Russian named Nikolas de granted a patent on covered electrodes.
Barnardos, who used a carbon rod fitted to an insulated Development of power sources also proceeded side
handle to weld parts manually He was issued a British by side. In the pioneering days of 1880's a bank of
patent in 1887, which is considered to be the first one batteries charged with a dynamo served as the power
on arc welding. Carbon arc process was commercially source. Serious experimentation started in the first
exploited in Europe and the U.S.A. in the late 1880's decade of the 20th century. By 1907, the first welding
and the early 1890's with limited success, because it generator was introduced in the U.S. market.
operated on high voltages (100-300 V) and currents
(600-1,000 amps), and resulted in hard and brittle weld Electric Arc
metal due to carbon inclusions. The electric arc is a sustained electrical discharge over
The shift from carbon arc to metal-arc welding came a gaseous path between two poles. The arc current is
soon. In 1889, two men independently announced the carried by ionised gas called plasma, which consists of
development of arc welding using a metal rod, in which, nearly equal numbers of electrons and ions. The current
after the arc was struck, the rod progressively melted is mainly conducted by electrons, which flow out of a
and provided additional weld metal. They were: N.G. negative pole (cathode) and move towards a positive
Slavianoff, a Russian, and Charles Coffin, an American, pole (anode). In a welding arc, the plasma may be mixed
who later became President of the General Electric with other states of matter such as molten metals, slags,
Company The latter was granted a U.S. patent on the v a p o u r s , neutral and excited gaseous atoms and
process in 1889. molecules. From the practical viewpoint, a welding arc
At the start of the 20th century, manual metal-arc may be considered to be a gaseous conductor, which
welding came to be industrially accepted, but it was far converts electrical energy into heat.
Welding arc temperatures have been found to range

between 5,000 and 30,000 K, depending on the nature ELECTRODE
of the plasma and the current passing through the arc.
In covered electrodes which carry easily ionised
materials such as sodium and potassium in the coating,
maximum arc temperatures are about 6,000 K. In an inert
gas shielded arc, the axial temperature can be as high
as 30,000 K. The highly concentrated heat of the arc and
its small size make it possible to weld thick metallic
sections at fairly high speeds with excellent bonding
and minimum dissipation of heat, resulting in narrow
heat-affected zones.
In a welding arc, heat generation takes place at the
cathode, in the plasma and at the anode. Heat at the (A)
cathode is generated mainly by ionic bombardment. The
plasma heat comes from the collisions among electrons,
ions and atoms. At the anode, the incoming electrons
strike the anode surface and liberate considerable heat,
because of the energy which they took upon emission
from the cathode and the added energy which they
acquired during acceleration across the plasma under
the potential difference of the poles.
The heat liberated at the cathode and anode regions
is usually greater than that from the plasma or arc
column. Also the amount of heat generated at the
cathode and anode differ appreciably, depending on the SLAG
metals which make up the electrode and base metal,
and on the nature of the plasma. The heat distribution
(B)
between cathode and anode often determines the
melting rate of the electrode and penetration into the
base metal. The significance of this factor will be pointed Fig. 3.1: Heat balance in arc welding:
out from time to time as various welding processes and (A) Manual metal-arc process; (B) Submerged arc process
metals involved are being discussed. It m u s t be
understood that the above remarks apply to the DC arc
and not to the AC arc, because in the latter case the If the arc voltage is 20, the current is 300 amps and
polarity is not constant. the welding speed is 150 m m / m i n , the heat input rate
During welding the entire heat produced by the arc of the arc is 2,400 joules per mm or 2.4 kilojoules (kj)
is not used for heating and melting of joint edges, but a per mm.
portion is used for heating and melting the electrode The w e l d i n g arc cannot be ignited merely by
and its coating. Some portion is lost in radiation and applying to the cold poles the voltage required by the
convection. Arc heat balance in manual arc welding and arc. A conducting or ionising link has to be provided
submerged-arc welding is shown in Fig. 3.1. between them. This is usually done by either applying
The heat developed by an arc or arc energy can be a sufficiently high voltage between the poles to cause a
calculated by using the simple formula: discharge or by bringing them together to make a
W (joules) = V (volts) x A (amps) x T (seconds) contact and then drawing them apart. How these arc
heat arc voltage current time striking techniques are applied to various arc welding
If the arc is travelling at a speed of S m m / m i n , the processes will be explained while dealing with them in
heat input rate of the arc is given by the formula: detail.
V x A x 60 Once an arc has been struck, re-igniting it after a
Joules per mm length of joint momentary extinction is relatively easy. Re-ignition is
further improved if materials which are good thermal
MANUAL METAL-ARC WELDING 59
emitters are present in the electrode. A DC arc is easy to and electrodes. Welding can be carried out in all
maintain once it is ignited, but this is not the case with positions, both in shop and at site. Welded joints of
the AC arc, which is extinguished at zero current point sound quality and adequate mechanical properties can
on each reversal, twice in every cycle. For the arc to re- be obtained by using correctly designed electrodes and
ignite, the required voltage must be available at the point proper welding procedures. The process is intermittent,
of zero current. This is ensured by using a low operating because welding has to be interrupted from time to time
power factor (usually 0.3) in an AC power source to discard the unused stub and to place afresh electrode
(welding transformer) so that the current wave lags the into the holder, and also to deslag the joint, i.e. to remove
voltage wave, and thereby the full open-circuit voltage the layer of slag covering the weld. For higher
of the transformer is available at the zero current point productivity, semi-automatic or fully-automatic welding
to reignite the arc. This has been explained in detail processes are preferred.
under Open-Circuit Voltage in Chapter 2.
Arc length is an important factor in arc welding, POWER SOURCES
because it controls arc voltage. A shorter arc means less As explained in Chapter 2, power sources of the
arc voltage and increased current, and it results in constant current (CC) type h a v i n g drooping
increased weld deposition rate and welding speed. A characteristics are used for the MMAW process.
longer arc means higher arc voltage and lower current, Power sources of the constant voltage (CV) type are
which lead to lower welding productivity. When the not suitable. The size or capacity of a-power source is
arc is too long, heat is lost to the air, spatter increases selected according to the sizes of electrodes used on the
and the weld metal picks up nitrogen, with the result shop floor. A welding machine rated around 300-350
that the welded joint shows porosity and reduced tough- amps at 60% duty cycle is considered a standard size,
ness. In welding with DC, the shortest possible arc must because it can handle electrode sizes from 6.3 mm down
be used to minimise arc blow and contamination by air. to 1.6 mm. A heavy fabrication shop using 5 mm and
6.3 mm electrodes most of the time are advised to go in
THE PROCESS for a machine rated for 400-450 amps. On the other
The heart of the MMAW process, which is illustrated hand, a shop engaged in sheet metal fabrication would
in Fig. 1.1 in Chapter 1, is the power source, which do well with a 200-250 amps machine, since the largest
provides continuous and steady supply of electrical size of electrode it may use is 3.15 or 4 mm.
current of sufficient amperage and of proper voltage to Some heavy fabrication shops prefer to use 8 mm
maintain an arc. It is either (a) a transformer giving AC size normal electrodes or 5 mm/6.3 mm sizes of heavy
supply, (b) a transformer-rectifier providing DC supply, coated iron powder electrodes to achieve very high
or (c) a motor-driven or engine-driven generator giving productivity. These shops must go in for welding
DC supply. Two cables or leads extend from the two machines rated around 500-600 amps at 60% duty cycle.
output terminals of the power source: one of them is Both AC and DC are used. DC is suitable for all types
the ground cable attached to the workpiece by means of electrodes and for welding all ferrous and non-ferrous
of an earth-clamp, the other is the welding cable attached metals. AC is not suitable for certain types of electrodes
to the electrode holder. and for welding certain metals. So before deciding to
To start the welding operation, the welder fixes go in for an AC power source, one must make sure that
the bare, contact end of a flux-coated electrode into the the electrodes he is going to use and the metals he is
holder and strikes an arc by touching or scratching the going to weld are suitable for AC. The electrode
workpiece with the arc-striking end of the electrode and producer indicates the suitability of each electrode type
then withdrawing it slightly to maintain a short arc. The for AC or DC on the carton and the descriptive leaflet.
heat of the arc gradually consumes the electrode and He also indicates the current range for each size of
converts it into weld metal and slag. The welder electrode. This data should be studied before deciding
manually guides the electrode and controls its direction to buy a power source.
and arc travel speed. He also maintains a steady arc To make a choice between AC and DC, the following
length by feeding the electrode skillfully into the points must be considered:
workpiece. He ends the operation by withdrawing the 1) All types of electrodes work on DC while some
electrode suddenly and thereby breaking the arc. nonferrous types and basic low-hydrogen ferritic type
MMAW is a versatile and flexible process requiring electrodes may not give a stable arc with AC. In AC
simple equipment, a skilled welder, welder's accessories welding, because of the reversing nature of the current,
flux coatings must contain enough arc stabilisers to re- because the electrode in this case is able to carry a much
ignite the arc immediately after the current comes to higher current without overheating and the base metal
zero during each cycle. has deeper penetration than when the electrode is
2) When using small-diameter electrodes at low positive.
amperages, DC has a definite advantage over AC, both 6) DC can cause the problem of arc blow, especially
for starting and maintaining the arc. when welding is done in corners, near the ends of joints
3) For all electrode sizes, maintaining a short arc is or on structures involving several pieces coming to-
easier with DC than with AC, except when iron-pow- gether. Welding with high currents on heavy structured
der electrodes are used, may also give rise to arc blow, more so when the fit-up
4) Vertical and overhead welding on thick sections is bad. AC does not present this problem (see the sec-
is easier with DC than with AC, because a stable arc can tion on Arc Blow in Chapter 12).
be maintained at lower currents. 7) Voltage drop due to long cables is a matter for
5) Effect of polarity: DC has the advantage of two concern both in the case of DC and AC. Such voltage
polarities, which means that the electrode can be made drop added to that of the arc may overload the welding
negative or positive. Straight polarity (i.e. electrode power source and weaken the arc. AC is preferred where
negative) can be used for MMAW of all steels (except distances between the welding areas and power sources
when low-hydrogen electrodes are used), but not for are considerable as in shipyards. Long cables should
most non-ferrous metals. With straight polarity, more not be coiled excessively, because induction set up by
of the arc heat is concentrated on the electrode and con- the coils will reduce the output of the power source,
sequently melting and deposition rates are higher, weld- and may overload the transformer. It is a good policy to
ing is more rapid and the workpiece is less susceptible keep the cables as short as required for the job.
to distortion. Higher welding speed is possible. How- 8) For sheet metal welding, DC is preferable to AC.
ever, penetration is shallower and narrower. DC straight polarity minimises burn-through, because
Reverse polarity (i.e. electrode positive) is used with of the shallow penetration it causes. Bearing in mind
basic low-hydrogen electrodes and for most non-ferrous the comparative features of AC and DC, the fabricator
metals. It produces maximum penetration for a given can decide whether to go in for a welding transformer
set of conditions. This characteristic makes reverse po- for AC supply, or a welding generator / rectifier for DC
larity the better choice for root passes in groove welds supply.
made with use of backing bars, and for vertical and Other factors which also have to be considered in
overhead welding. choosing a power source have been discussed in Chap-
In AC welding, a choice of polarity is non-existent ter 2 under the section Comparison of Power Sources.
because AC combines reverse and straight polarity in
regular cycles. Hence AC gives penetration and elec- WELDING ACCESSORIES
trode melting rate intermediate between those given by MMAW requires a number of accessories, some of
DC + and DC -. The comparative bead profiles and pen- which are essential parts of the welding circuit, while
etrations are shown in Fig. 3.2. the others are needed for the welder's safety. The
It must be pointed out that these polarity effects ob- standard set of welding accessories supplied with the
served in MMAW are different from those observed in welding power source consists of:
TIG welding, because of the influence of the metal ions
• 3 m each of welding and ground cables
and fluxing compounds in the arc.
In TIG and similar n o n - c o n s u m a b l e electrode • Insulated electrode holder
welding processes, 66% of the arc heat is generated at • Ground clamp
the positive pole and 33% at the negative pole. In TIG • Chipping hammer and steel wire brush
welding, therefore, DC electrode negative is preferred,
• Handshield with coloured glass.
DIRECT CURRENT DIRECT CURRENT

STRAIGHT POLARITY^ REVERSE POLARITY
Fig. 3.2: Comparative bead profile and penetration for different current conditions
Besides these, the welder must be provided with or touched anywhere safely, because its handle as well
protective clothing to protect him from intense heat of as the jaws are completely insulated.
the arc, spatter, hot slag and hot metal. Leather gauntlet A properly designed holder is light in weight, grips
gloves are a must to protect hands from harmful rays, the electrode securely in position, receives and ejects
and to prevent burns. A leather apron is also absolutely the electrodes easily, a n d resists overheating in
essential. Additional items like skull-caps, shoulder continuous use. Holders are available in various sizes
covers and leg-pads made of leather or any other flame- according to their current-carrying capacities, their
resisting material may be required during overhead weight increasing with increasing capacity. The jaws
welding to protect the welder from hot spatter. of a holder must be in good condition and ensure good
electrical contact with the bare end of an electrode. Bad
Electrode Holder contact at the jaws causes arcing and overheating of the
The electrode holder is used to grip the electrode and holder and impairs welding performance.
provide it with the welding current. A holder can be A holder should be firmly connected to the welding
partly or fully insulated. In the former, shown in Fig. cable. A loose connection between the two may result
3.3, only the handle is insulated to ensure that the in overheating of the holder. Electrode holders should
welder's hand is not in electrical contact with the never be immersed in water for cooling.
welding circuit. When using such a holder, care must
be taken to avoid its contact with the work, because it Ground Clamp
will cause a flash. A fully-insulated holder, such as the The ground clamp connects the ground cable to the
one shown in Fig. 3.4, on the other hand, can be placed work. It is an important device because without proper
REPLACEABLE
JAWS
MECHANICAL CABLE CONNECTION
Fig. 3.3: Diagrammatic sketch of a typical partly insulated electrode holder
JAW SPREADING
LEVER
INSULATOR
HANDLE
HEAT
SHIELD
Fig. 3.4: Diagrammatic sketch of a typical fully insulated electrode holder

grounding, one fails to get a stable arc and the required metres of cable were added, the combined resistance of
heat for welding. A ground clamp should be designed the two cables would reduce the current output of the
to provide a strong positive electrical contact, to be power source. If the machine is then readjusted for the
capable of easy attachment and detachment, and to original current output, the additional load may overheat
withstand rough use. A commonly used design is shown it as well as increase its power consumption.
in Fig. 3.5. Cable sizes should be correctly selected, because they
have a definite bearing on the welding operation. Advice
Electrode Cable/Ground Cable of the manufacturer of welding equipment may be
These cables consist of a large n u m b e r of fine sought on this point. Table 3.1 derived from IS: 434,
aluminium or copper wires enclosed in a tough casing of indicates cable sizes for copper and aluminium material.
woven fabric and rubber. They are very flexible. The cable From the resistance values and current being used, the
size must be correctly chosen depending on the capacity voltage drop can be calculated. The larger size of cable
of the power source and the distance between the power is used when the voltage drop with the smaller cables
source and the work. The effect of cable length may be becomes excessive. When greater flexibility is required,
explained thus: a 10-metre cable of a particular size may two small size cables in parallel may be used instead of
satisfactorily carry the required current, but if another 10 one equivalent larger cable.
To carry the usual heavy welding current consistently,
the cable connections must be of heavy-duty design. For
connecting the cables to the secondary terminals of a
power source, lugs are attached to the cable ends by
soldering or mechanical crimping. A lug may also be
attached to the earthing end of the ground cable to
connect it to the work bench as shown in Fig. 3.6.
Handshield/ Headshield
Arc w e l d i n g d e m a n d s that a w e l d e r uses a
Fig. 3.5: Ground clamp (earthing clamp) handshield or a headshield (see Fig. 3.7). The headshield,
also called a helmet, is used only when the welder needs
Table 3.1: Specifications of welding cables as per IS:434

Copper cable specifications
Nominal hand- C/S No. of strands Resi. per km Total thickness Min. insu. resi.
welding current (mm2) & dia. of each at 20°C of covering per km at 20°C
(Amps.) wire (mm) (Ohms) (mm) (Mega Ohms)
100 16 509/0.2 1.132 2 35
150 25 796/0.2 0.7239 2 35
230 35 1,114/0.2 0.5173 2 35
400 50 1,591/0.2 0.3622 2 35
600 70 2,228/0.2 0.2586 2 35
600 (heavy duty) 95 1,344/0.3 0.1905 2.55 35
Aluminium cable specifications

Nominal hand- C/S No. of strands Resi. per km Total thickness Min. insu. resi.
welding current (mm2) & dia. of each at 20°C of covering per km at 20°C
(Amps.) wire (mm) (Ohms) (mm) (Mega Ohms)
100 27 380/0.3 1.112 2 35
150 42 589/0.3 0.717 2 35
230 58 817/0.3 0.517 2 35
400 82 1,161/0.3 0.364 2 35
600 112 1,591/0.3 0.265 2 35
600 (heavy duty) , 153 962/0.450 0.195 2.5 35
WORK BENCH
GROUND CABLE BOLTED CLAMPED TACK WELDED
Fig. 3.6: Methods of attaching ground cable
K rf„, s??r
Fig. 3.7: Shields for arc welding: (A) Headshield; (B) Handshield
Welding MMAW. C02/MAG MIG TIG Plasma Micro-plasma

process (Amps.) (Amps.) (Amps.) (Amps.) cutting welding
Glass
Shade No.
5 0.5-1.3
6 — — — — — 1.4-2.6
7 — — — — — 2.7-5.0
8 — — — — — 5.1-9.9
9 20-39 — — 5-19 — 10-15
10 40-79 40-79 80-99 20-39 — 16-30
11 80 - 1 7 4 80 - 1 2 4 100 - 1 7 4 40-99 50 -149 —
12 175 - 299 125 - 1 7 4 175 - 249 100 - 1 7 4 150 - 249 —
13 300 - 499 175 - 299 , 250 - 349 175 - 249 250 - 400 —
14 500 - 560 300 - 449 ' 350-499 250 - 399 — —
15 — 450 - 600 500 - 550 400 - 500 400 & above —
16 — Above 600 Above 550 — — —
his other hand to be free while welding. It usually has a the form of large molten drops, which frequently short-
swing mounting to enable the welder to lift the front circuit the arc. The weld metal in the molten condition
part of the helmet above his face and to put it back, becomes supersaturated with oxygen and nitrogen, and
without removing the head band from the head. Such because the weld metal solidifies and cools very rapidly
movements are necessary between weld runs to deslag in the absence of an insulating blanket of slag, most of
and inspect them. Both the handshield and headshield the oxygen and nitrogen are retained in the weld deposit.
are meant to protect the face, head and eyes from the Iron oxide, formed by the combination of iron and
blinding radiation of the arc and hot spatter, and are oxygen, gets distributed in the weld metal as tiny
provided with coloured filter glasses through which the globules. Nitrogen turns to iron nitride Fe4N. Both iron
arc must be observed. An electric arc produces a brilliant oxide and iron nitride lower ductility and toughness.
light and gives off ultraviolet and infrared rays which, In all arc welding processes, the arc is suitably shielded
though invisible, scorch the eyes and skin. One should by means of flux or inert gas primarily to prevent
never look at the arc with naked eyes within a distance contamination of weld metal by atmospheric oxygen
of 16 metres. and nitrogen.
Coloured glasses called filter glasses reduce the The MMAW electrode consists of a straight piece of
brilliancy of the arc light and screen out the ultraviolet solid metallic wire, called core wire, having a concentric
and infrared rays. They are available in various standard covering (also called coating) of flux as shown in Fig.
shades, each one being meant for a specific welding 3.8. As much as 25 to 40 mm of the wire is left bare at
condition. The recommended shades (shade numbers one end to serve as a grip in the electrode holder during
as per DIN standard) for different welding applications welding. At the other end, called the striking end, only
are tabulated below: the core face is left exposed to facilitate arc striking.
Filter glasses are of such dark shade that one cannot The electrode size refers to the diameter of its core
see through them unless the arc is present. They are wire. The electrode length is d e t e r m i n e d by the
expensive and are usually protected from spatter and convenience of the w e l d e r to strike the arc and
abuse by being placed b e t w e e n two o r d i n a r y manipulate the electrode, and by the capacity of the
transparent glass pieces of the same size. electrode to carry a reasonably high welding current
w i t h o u t getting overheated until it is consumed.
Chipping Hammer/Wire Brush Electrode sizes and lengths have been standardised by
The welder needs a chipping hammer to remove the ISO (International Organisation for Standardisation) as
slag which covers the deposited weld bead before the shown in Table 3.2.
next weld bead is laid. Unless the slag is completely Sizes 12.5,10, 8,1.25 and 1 are rarely used. Electrode
removed, the finished weld joint is likely to have slag manufacturers who are equipped to produce them
inclusions. The chipping hammer is made of hardened would do so only against specific orders.
steel and is usually double-ended. One end is shaped T h o u g h metric sizes for electrodes have been
like a chisel for general chipping, while the other end universally accepted, they are sometimes referred to in
has a sharp point to deslag the corners. After chipping, fractional inches or in swg. Table 3.3 provides correct
the weld bead needs to be brushed vigorously with a conversion while Table 3.4 gives a p p r o x i m a t e
wire brush to remove the last traces of slag. The wire equivalents used for practical purposes.
brush is also used to remove rust and dirt from joint The flux covering thickness varies with each class
surfaces before starting to weld. and brand of electrode and is usually expressed as
coating factor, which is the ratio of coating diameter to
MMAW Electrodes the core wire diameter (CF - D_ in Fig. 3.8).
MMAW is performed with flux-coated electrodes. d
As mentioned earlier, in the early days of arc welding Electrodes are often referred to as light-coated,
(between 1889 and 1910), bare electrodes were used with medium-coated and heavy-coated depending on their
DC, but the performance and the weld metal quality coating factor as shown below:
were unsatisfactory Light-coated Upto 1.30
When w e l d i n g w i t h bare electrodes, the arc Medium-coated 1.30 to 1.50
atmosphere consists of iron ions and atoms of nitrogen Heavy-coated Above 1.50
and oxygen of the air. The arc is unsteady in the absence These definitions are based on IS 814-1991
of arc stabilisers and the metal transfer takes place in specification.
FLUX COVERING
LX::(®)\D J(/////ZZ^
\
STRIKING END CORE WIRE BARE END FOR GRIPPING
Fig. 3.8: Sketch of manual metal-arc welding electrode
Table 3.2: Standard dimensions of electrodes

Size (mm) 12.5 10 6.3 3.15/3.2 2.5 1.6 1.25
Length (mm) 350 350 350 350 350 350 350 250/ 200/ 200 200 200
300 250
450 450 450 450 450 450 450 350 300/ 250
350
Welding Currents metallic oxides or by means of powdered ferro-alloys

Every electrode has a certain current range within and metals. Thereby the mechanical properties of the
which it operates satisfactorily. On a light job where weld metal are also suitably modified
overheating must be avoided, current on the lower side 6) It helps to control the weld bead profile and to
is used. For heavy work where maximum heat for obtain smooth weld surface with even ripples
adequate fusion is necessary, current at the top of the 7) It helps to reduce weld spatter
range is used. Current in the middle of the range is used 8) It insulates the electrode so that it can operate at
for normal work. Welding current increases with sufficiently high current without getting overheated
electrode size. For a given electrode size, the current 9) It makes vertical and overhead welding possible
increases with coating thickness, because as the coating by controlling the viscosity of the slag
thickness increases, a higher current is required to melt 10) It influences the operating characteristics of the
it. Table 3.5 indicates the currents required for a medium- electrode and, to some extent, its burn-off rate and the
coated electrode. depth of penetration of the weld bead into the base metal
Too low a current gives an unstable arc and results 11) The coating having a low thermal conductivity,
in lack of fusion in the welded joint. Too high a current melts slower than the core wire. Thus, in the case of
overheats the electrode, increases spatter and results in heavy-coated electrodes, the coating forms a cup around
weld metal porosity. Currents recommended by the the upper portion of the arc and prevents air being
electrode manufacturer must be used. drawn into the arc stream. The cup formation also makes
touch or contact welding possible
Functions of Flux Coating 12) The coating increases deposition efficiency and
1) It helps to strike and maintain the arc weld metal deposition rate through iron powder and
2) It generates gases which displace the oxygen and ferro-alloy additions.
nitrogen of the atmosphere and provide a gaseous shield The coating also controls the following arc
around the arc to protect the molten droplets during their characteristics:
passage across the arc as well as the molten weld puddle • Arc voltage and current intensity
3) It produces slag which further protects the mol- • Thermal and dynamic energy
ten droplets in the arc and the molten weld puddle from
• The feasibility of welding with DC or AC, and
atmospheric attack, and also provides an insulating
electrode polarity in DC
blanket over the weld bead
4) It helps to deoxidise and refine the weld metal • Process of emission and ionisation
5) It helps to modify the chemistry by providing al- • Mode of metal transfer: short-circuiting, free
loying elements for the weld metal, either by reducing flight, spray type.
Table 3.3: Conversion for swg, fractional inch Table 3.4: Approximate equivalents of electrode sizes
and mm sizes
swg fractional (in.) mm
swg fractional (in.) mm in
— 1/32 0.8
— — 0.6 0.024 18 3/64 1.2
— 1/32 — 0.031 16 1/16 1.6
— — 0.8 0.032 14 5/64 2.0
— — 0.9 0.035 12 3/32 2.5
— — 1.0 0.039 10 1/8 3.2
— 3/64 1.2 0.047 8 5/32 4.0
18 ■ —
— 0.048 6 3/16 5.0
— — 1.5 0.059 4 — 6.0
— 1/16 — 0.062 — 1/4 6.3
— — 1.6 0.063 — 5/16 8.0
16 — — 0.064
— 5/64 — 0.078
— — 2.0 0.079 Table 3.5: Welding currents vis-a-vis electrode size
14 — — 0.080
Electrode size Welding current, (Amps.)
— 3/32 2.4 0.094
— — 2.5 0.098 (mm) for light for heavy for normal
12 — — 0.104 work work work
— — 3.0 0.118 6.3 220 320 260
— — 3.15 0.124 5 180 240 210
— 1/8 — 0.125 4 140 180 165
10 — 3.25 0.128 3.15/3.2 90 130 110
— 5.32 — 0.156 2.5 55 85 70
— — 4.0 0.157 2 40 60 60
8 — — 0.160 1.6 25 40 30
— 3/16 — 0.188
6 — — 0.192
— — 5.0 0.197 • Arc stabilisers: Potassium oxalate, Potassium sili-
— 7/32 — 0.219 cate, Zirconium carbonate, Potash feldspar,
4 — — 0.232 Lithium carbonate, Titania.
— — 6.0 0.236
— —
• Gas-forming materials: Cellulose, Limestone, Wood
6.3 0.248
— 1/4 — 0.250
flour.
2 — — 0.276 • Slipping agents (for easy extrusion): Glycerin, china
— 5/16 — 0.312 clay, Kaolin clay, Talc, Bentonite clay, Mica.
— — 8.0 0.315
— —
• Binding agents: Sodium silicate, Dextrin, Potas-
3/8 0.375
— — 10.0 0.394
sium silicate, Gum arabic, Sugar, Asbestos.
— 1/2 — 0.500 • Deoxidisers and alloying elements: Ferrotitanium,
Ferrosilicon, Ferrochromium, Ferrocolumbium,
Some of these functions will be elaborated in the Ferromanganese, Electro-nickel, Electro-manga-
section on Metal Transfer in Electrodes. nese, Ferromolybdenum, Metals in powdered
form.
Common Flux Ingredients and Their Functions
• Fluxing agents: Silica, CaO (from limestone), Flu-
orspar TYPES OF ELECTRODE COATING
• Slag-formers: Rutile, Titania, Potassium titanate, Ever since the invention of the first modern covered
Ilmenite, Asbestos, Alumina, Silica flour, Iron electrode by Oscar Kjellberg of Sweden in 1907,
powder, Fluorspar, Feldspar, Manganese dioxide, considerable research has been done on electrode
Wollastonite. coverings the world over to achieve
(a) a low-cost formulation, Rutile Type

(b) most satisfactory operating characteristics, The maior constituent of this covering is rutile,
(c) all-weld tensile and impact properties matching usually exceeding 50% by weight (not taking into
those of the base metal. account cellulosic material). Rutile is a naturally-
All such development work and experience in the occurring sand, containing around 95% of titanium
production and use of electrodes have led to the ultimate dioxide. It has a natural dark-black, shining appearance,
evolution of a few standard covering types which have unlike pure titanium dioxide (called titania), which is
been coded and classified in various national and snow-white and is therefore also termed titanium white.
international specifications for electrodes. They are: Rutile is sometimes used in the covering in finely ground
form which has a dirt-brown colour.
• Cellulosic type
The covering also contains slag-forming silicious
• Rutile type materials like feldspar and china clay, and calcium and
• Acid type potassium compounds to ensure arc stability in AC
• Acid rutile type welding. Cellulose and ferro-manganese in varying
amounts are also included. The binder is sodium silicate
• Oxidising type
or potassium silicate.
• Basic type Two distinct versions of this type have emerged over
the years. One of them contains 10 to 15% cellulose and
Cellulosic Type is applied to the electrode to give a coating factor of 1.3
The major constituent of this covering is cellulose, to 1.4. The slag formed during welding is viscous and
usually exceeding 30% by weight. Other organic quick-freezing, and this renders the electrode
materials such as wood flour, charcoal, cotton, yarn, particularly suitable for welding in the vertical and
paper wrappings, starches and gums have also been overhead positions. The electrode with this covering is
used to partially replace cellulose. the most popular type in most countries, and is
The other materials generally used are titanium described as the general-purpose mild steel structural
dioxide (derived from the granular mineral rutile), electrode.
metallic deoxidisers such as ferromanganese, various The other version contains around 3% cellulose and
types of magnesium or aluminium silicates, and liquid is applied to the electrode to give a coating factor of 1.5
sodium silicate as a binder. The cellulose, which is an to 1.7. The slag formed during welding is heavy, compact
organic material, decomposes in the arc and produces a and usually self-detaching.
voluminous gas shield which protects the arc from the
atmospheric oxygen and nitrogen. The gaseous Acid Type
atmosphere produced at the arc has approximately the The major constituent of this covering is iron oxide.
following composition: Other constituents are manganese compounds
1.0% C0 2 55% CO 42% H 2 and 1.5% H 2 0 (sometimes including manganese oxide) and silica along
The amount of slag produced is relatively small. The with sufficient deoxidisers. Organic materials are absent.
weld deposit is thus covered with a thin, friable slag Potassium silicate is used as a binder. During welding,
layer, which is easily detachable. this covering produces a voluminous iron-manganese
This covering type does not have enough ionising oxide-silica fluid slag, whose metallurgical character is
compounds and hence the electrode having this covering acidic. This slag solidifies in a characteristic honeycomb
operates only on DC. For use on AC (as well as DC), structure and is easily detached.
small quantities of calcium and potassium compounds
are added to the covering and the sodium silicate binder Acid Rutile Type
is replaced by potassium silicate. This covering is a compromise between the rutile
A special feature of this type is that 4 to 6% moisture type and the acid type. Rutile or ilmenite (a granular
is deliberately retained in the coating as it contributes mineral containing titanium oxide and iron oxide)
favourably to the operating characteristics of the replaces part of the oxides of iron and manganese. The
electrode. The moisture prevents premature charring of covering is considered to be AR type only when the
the cellulose and results in a deeply penetrating, forceful, amount of titanium oxide is larger than the total amount
spray-type arc, which is used to advantage in the so- of iron oxide and manganese oxide. The titanium oxide
called stovepipe welding of cross-country pipeline. content is normally not higher than 35%. The slag
formed by this covering during welding is similar to energy reactions. The situation is even more complex
the slag of the acid type but slightly more fluid. in manual metal-arc welding where, for example, some
electrodes perform well on D C - or AC, while others do
Oxidising Type so on DC+.
This covering is composed mainly of iron oxides and Metal transfer is subject to forces of magnetic nature,
silicate with or without manganese oxides. During known as Lorentz force and pinch effect. As the welding
welding, it transforms into a heavy solid slag which has current flows through its own magnetic field, the mobile
oxidising properties. These result in a weld deposit, electrons in the arc cut lines of magnetic flux and
which is low in carbon and manganese. No attempt is experience a force directed towards the centre. In other
made to compensate the manganese loss through words, plasma from the outer regions of the arc is
addition of ferromanganese. The resultant weld metal pumped into the hot central core of the arc. The force
is therefore dead soft and has low mechanical properties. causing this effect is called Lorentz force. The pinch
It has limited use, which is restricted to sheet metal effect refers to the momentary necking of the liquid drop
fabrication. carrying the current, caused by the interaction between
the arc plasma kinetic pressure and the self-field
Basic Type pressure. The influence of Lorentz force and pinch effect
This covering has a high content of calcium is more pronounced in the case of high current density
carbonate (in the form of limestone, calcite or marble) processes such as MIG, but it is not significant in the
and calcium fluoride (in the form of fluorspar). This case of MMA welding, which is a low current density
results in a chemically basic slag, which is fairly fluid. process.
The solidified slag, which is heavy and friable has a Early investigations of metal transfer from coated
brown to dark-brown colour and a glassy appearance. electrodes were m a d e , either w i t h high speed
Use of organic materials like cellulose and minerals with photography or by deposition on a fast moving copper
chemically combined water (such as china clay, talcum plate. In the former, high speed films of the arc were
and mica) is avoided in order to keep the arc hydrogen- taken, and the number and size of the droplets seen on
free. The gaseous atmosphere formed around the arc the film were measured. Knowing the volumetric ratio
consists of CO, C 0 2 and F c o m p o u n d s . The basic of slag and metal for an electrode in question, and
character of the slag results in good transfer efficiency assuming that each drop comprised metal covered with
of the alloying elements from the coating to the weld a layer of slag, one could obtain a size distribution of
deposit. The so-called hydrogen-controlled mild steel drops in relation to the type of coating. However, the
electrodes are generally made with this type of coating, droplets actually can be entirely slag or slag and metal,
and this coating is also preferred in low-alloy high and this method cannot distinguish between the two.
tensile electrodes, in order to minimise the risk of The copper plate method is also not reliable, since
hydrogen-induced cracking or cold cracking in the base drops may well coalesce before they solidify. In fact, the
metals. drop size measured in this way does appear larger than
The covering is made in two versions: one suitable that obtained from photographs.
for DC only, and the other suitable for AC as well as One of the more recent and reliable methods is cine-
DC. The latter, unlike the former, contains sufficient radiography, in which a flash X-ray tube produces a
amount of potassium salts and uses potassium silicate shadow-graph of the core wire and liquid metal for each
as the binder to stabilise the arc. X-ray flash. Arc voltage and current are recorded, and
by means of timing marks, it is possible to locate the
METAL TRANSFER IN ELECTRODES position of the individual frames of the film on the time
Electrical a n d other characteristics of a plain scale of the voltage and current traces.
sustained arc between two refractory materials such as A similar arrangement has been used in another
g r a p h i t e or t u n g s t e n electrodes h a v e b e e n well method, in which the X-ray source is replaced by a flash
understood because of the absence of metal transfer. The lamp of greater luminosity than the arc itself, and the
arc of a metal arc welding process, however, is more electrode tip is recorded in silhouette on the films.
difficult to u n d e r s t a n d a n d explain, because the Whereas, for given electrode, light photography
emission of electrons is accompanied by the transfer of shows a spray of fine drops, the X-ray pictures generally
metal and slag, plasma formation, and the occurrence show transfer of metal in relatively massive form.
of various physico-chemical, metallurgical and electrical Occasionally there are small drops, but these usually
traverse the arc individually. Rutile coated electrodes

show a characteristic drop diameter of 2.3 mm, and more
than 65% of the transfer occurs as drops with diameters
in the range 1 and 3 mm. Such drops have a transfer
velocity of between 10 and 25 cm/sec. It must, of course,
be recognised that very small drops might not be visible
on an X-ray film, and indeed, such drops probably form
during restriking of the arc after a short-circuit.
However, the idea that metal transfer from a rutile-
coated electrode occurs in a spray of droplets is clearly
incorrect.
In a study of rutile type electrodes, the light-coated
variety exhibited gassing in the form of relatively large
bubbles which could be ascribed to the formation of Fig. 3.9: Spreading of molten slag over
carbon m o n o x i d e , since the coating h a d a low liquid electrode tip
percentage of deoxidant. Key for Figs. 3.9 to 3.13: A. Core wire; B. Coating;
In a typical case, several bubbles formed and burst D. Liquid electrode tip; E. Arc; F. Workpiece;
during the transfer of a single large drop. The bubbles G. Liquid slag; H. Drop; I. Second arc
grew to about half the electrode diameter before
bursting and apart from causing irregularities in the
vaporisation of the metal a n d / o r slag, resulting in the
form of the drop, did not modify significantly the
formation of an excess pressure at the arc root. Such
character of the transfer, which was still mainly by large
pressure deforms the liquid drop until part of it is
drops. Oxide-coated electrodes showed much more
extruded outwards to form a protrusion, still coated
severe gassing with the drop profile changing rapidly
with slag (Fig. 3.10). Continued pressure at the arc root
between one frame of the X-ray film and the next.
then causes the protrusion to detach and transfer to the
However, the characteristic drop diameter was only
weld pool (Fig. 3.11).
slightly less than for the rutile-coated type (1.9 mm as
compared with 2.3 mm diameter). Basic electrodes The second mode of transfer begins in the same way
showed the coarsest drop transfer. The drops were by the formation of a protrusion. In this case, however,
relatively undistorted, and (no doubt because of the the arc root transfers to the tip of the protrusion, which
short arc length) transferred mainly as a result of short- is then pinched off (Figs. 3.12 and 3.13). A second arc
circuits. then forms between the electrode tip and the pinched-
off drop, and the resultant increase in pressure helps to
The study showed that the metal drops passing
project the drop across the weld pool.
across the arc are always coated with a layer of slag. In
both light and X-ray pictures, one could observe
frequent bridging between electrode and workpiece
without the oscillograph registering any short-circuit. A -
The light photographs suggested that these short- B -

circuits were caused by pure slag particles, but the X-
ray pictures provided a more convincing explanation
that a layer of slag on the transferring drop insulated it
from the workpiece.
Cine radiography method has made it possible to D
describe the transfer mechanism in more detail, and to AK$îfi
distinguish two modes of transfer. In the first, the
process starts with a volume of liquid metal at the tip of
the rod which is covered with a layer of liquid slag
except at the arc root, which is smaller in diameter than
the electrode itself (Fig. 3.9). Because of the small area Fig. 3.10: Further deformation of the liquid electrode tip
of the arc root, the energy flux (and hence the through shortening of the arc and the movement of the
temperature) is very high. Therefore, there is some arc root
The fact that the liquid metal is coated with slag and
is surrounded by a cone of unmelted coating helps to
A- direct the drops towards the weld pool. However, in
the first mode of transfer (called single arc drops), the
B- neck of the protrusion receives a lateral impulse and
the drop may be projected sideways as spatter.
Thus it is suggested that while factors such as
viscosity and surface tension of the slag and interfacial
tension between slag and metal may be significant,
vaporisation pressure is the dominant force in projecting
droplets towards the workpiece. The slag plays an
important role by restricting the area and mobility of
the arc root, so that inward pressure at the arc root causes
outward movement of part of the liquid drop, as well
as by modifying surface tension and acting as a source
Fig. 3.11: Separation of slag-coated single-arc drop of vaporised material.
The fact that vaporisation pressure is the main
driving force for the ejecting metal has been confirmed
by Wegrzyn at the Polish Welding Research Institute,
by comparing the performance of uncoated rods of
killed and rimming steel in overhead welding. The metal
transfer in the case of rimming steel was satisfactory,
but with fully-killed wire the metal melted at the tip
and ran down the side of the wire. In this case, transfer
from the rimming steel wire was considered to be due
to gas evolution inside the melting drop, which gave
the liquid metal an impulse in the upward direction.
Wegrzyn has explained that the gas may be formed
within the liquid metal by rimming action: C + FeO =
Fe + CO (example, acid type). It could be hydrogen
evolved from the moisture or any ingredient in the
coating; in the case of basic low-hydrogen types, the
Fig. 3.12: Further movement of the arc root from the
gas may evolve from the slag.
position in Fig. 3.11 to the end of the protrusion
The mode of metal transfer is markedly affected by
the surface tension and viscosity of the liquid metal and
slag. The surface tension holds the droplets at the
melting electrode tip and causes them to be spherical in
shape. As the surface tension and viscosity of the liquid
metal decrease, it is easier for the metal transfer to take
place. Surface tension of the liquid metal is reduced by
increasing the welding current, as also by the formation
of oxide and slag films on the surfaces of the droplets
caused by suitable ingredients in the flux coating.
In MMA welding with all types of electrodes, except
the basic low-hydrogen type, the molten metal in the
weld pool is usually in a disturbed and effervescent
state, because it is supersaturated with hydrogen and
carbon monoxide. This affects the mode of metal transfer
more than the surface tension and viscosity. Especially
Fig. 3.13: Formation of a double-arc drop and of a when the liquid metal is exhibiting rimming (i.e. it is
second arc in series with the first
not getting deoxidised) the values of surface tension and
viscosity remain unstable. However, in the case of basic oxidised and is metallurgically semi-killed. Hence, sur-
low-hydrogen electrodes, in which the liquid metal is face tension plays a greater part in this type than in acid/
usually well deoxidised, viscosity and surface tension oxide types. The metal transfer is less of the spray type
of the metal itself and at the slag metal interface can and more of the droplet type.
determine the mode of metal transfer to a greater extent.
Metallurgical and gas reactions, which occur in the • Cellulosic type
arc as the core wire and the coating melt, appear to play This contains nearly 30% cellulose by weight, which
a dominant role in the metal and slag transfer across gives rise to large amounts of CO, C02, H 2 0 and their
the arc. Gases make the metal apparently effervesce and dissociated products in the arc. Hence, even if the coat-
boil at lower temperature than its boiling point, and this ing contains sodium and potassium compounds and
makes the metal transfer easier in positional welding. some amount of Ti02, the concentration of their ions in
The rapidly evolving gases under high pressure may the arc atmosphere is greatly reduced. Also, the
actually carry metal droplets with them. These droplets ionisation potential of the evolved gases mentioned
are often very light, because like bubbles, they carry above is four times greater than the ionisation potential
gases in their interior, which are at very high of Na and K. As a result of reduced extent of ionisation
temperature. The more the metal differs from the killed in the arc, electronic emission is impeded. The result is
state in the metallurgical sense, the larger are the forces that the arc voltage is high.
effecting its transfer.
In all types of electrodes except the basic types, and • Basic type
particularly in the oxide and acid types, hydrogen and This contains calcium carbonate, fluoride (CaF2),
oxygen penetrate into the droplets and cause the metal powders and comparatively more powerful
following reactions to take place: deoxidisers, and sometimes small amounts of alumino-
FeO + C = Fe + CO and FeO + 2H = Fe + H 2 0 silicates. Moisture is completely removed from the coat-
These reactions increase the state of effervescence ing to render it low-hydrogen. The molten metal pro-
and aid liquid metal transfer. Coverings containing large duced in the arc is well-deoxidised, killed and low in
amounts of hydrated alumino-silicates and moisture are hydrogen. Therefore, as stated earlier, the metal trans-
even more active. They usually melt faster than they fer is in comparatively large droplets. If, however, the
degas. The result is copious evolution of gas, which electrodes are not thoroughly dried and have moisture
makes the melting slag effervesce and boil and the metal in the coating, the metal transfer tends to be of the spray
transfer all the more easy. type.
Basic coverings are almost passive metallurgically
in regard to metal. Since these are usually designed to Polarity in DC Welding
be low-hydrogen, only a small amount of hydrogen In DC welding, electrodes of the cellulosic and basic
penetrates into the electrode melting tip. At the same types operate more satisfactorily on reverse polarity (i.e.
time, Si also penetrates there and deoxidises the metal. electrode +ve) than on straight polarity (i.e. electrode -
Hence, with these electrodes, metal transfer takes place ve). On the other hand, electrodes of rutile, oxide and
in large droplets. acidic types perform better on straight polarity. These
To summarise metal transfer takes place in various opposite behaviours can be explained as follows:
electrode types as follows: As already mentioned, cellulosic electrodes generate
CO, C02, H 2 0 and their dissociated products in the arc.
• Acid and oxide types Since these gases do not develop a highly conductive
Because the coating contains large amounts of ox- arc plasma, the current distribution is such that the
ides/silicates of Fe and Mn, as also of alumino-silicates, liquid metal is forced away from the arc and weld pool
the melting metal is largely rimming and very active, as massive drops and spatter. These reactions are more
because it contains large quantities of oxygen and hy- intense when the electrode is negative, and are
drogen. This lowers the surface tension of the melt and apparently caused by the concentration of current on
results in spray droplet transfer. the cold cathode. The term cold cathode is explained
later.
• Rutile type In the case of basic electrodes, CaF2, is considered to
In this coating, the Fe and Mn compounds are re- be mainly responsible for causing stable melting, when
placed by Ti02. As a result, the molten metal is not so the electrode is made positive. The explanation put forth
is that the negative fluorine ions formed in the vicinity steel wire is drawn from cast ingots, whereas non-
of the electrode anode are attracted by it and get rimming steel wire is drawn from billets or blooms.The
discharged. At the same time, the number of cations characteristic features of rimmed steel are a high oxygen
(positively charged ions) in the anode' area remains content (typically 200-400 ppm) and a rim of almost
small, because they keep moving towards the cathode pure iron on the surface.
(weld pool). The net result is that the fluorine anions It is preferred to semi-killed or fully-killed steel,
(negative ions) are not much effective in neutralising because it is cheaper to produce and also because it
the cations and thereby reducing the degree of arc confers optimum performance characteristics on the
ionisation. Hence, the electron emission from the electrode. Since it is possible to obtain any desired
cathode (weld pool) takes place with relative ease. alloy steel c h e m i s t r y of t h e w e l d m e t a l b y
On the other hand, with electrode negative, although incorporating suitable alloying elements in the flux
the fluorine anions are driven towards the anode (weld coating, the modern electrode producer uses a single
pool), their concentration in the cathode region remains grade of rimming quality steel to manufacture a wide
undiminished, because further amounts of them are range of mild steel, low-alloy steel and hardfacing
being emitted from the covering. The electron emission electrodes.
from the cathode into the region which contains a small American Welding Society's Specification for Carbon
quantity of cations is rendered difficult. Also, the Steel electrodes for shieled metal arc welding (AWS
common transfer of an electron stream and fluorine A5.1- 91) states: "The core wire for the electrodes in this
anions in the arc results in the repulsion of the former specification is usually a rimmed or capped steel having
on the material around it and consequent spatter. a typical c o m p o s i t i o n of 0.10% carbon, 0.45%
Coverings which are predominant in rutile, lime and manganese, 0.03% sulphur, 0.02% phosphorus, and
iron oxide make the electrode thermionic (i.e. they easily 0.01% silicon." Indian Standard on mild steel core wire
supply electrons to sustain the arc). Such electrodes (IS:2879-1998) stipulates that the steel shall be of
produce a more stable arc, less spatter and form smaller rimming or non-rimming quality and the chemical
droplets with electrode negative. composition (% ) shall be as follows:
Here, it may be necessary to point out that most Rimming quality steel is best produced in the basic
structural metals and their alloys form what is called a open-hearth (BOH) furnace. The word basic refers to
cold cathode: its area is rather small, but great quantities the refractory lining used in the furnace and also
of energy are generated to release the electrons needed indicates that in making the steel, a basic slag is being
to support an arc. However, metals having very high used which is predominant in CaO, MgOCaO, FeO and
melting points easily supply electrons to sustain the arc MnO. Such a slag is capable of removing any reasonable
due to their temperature; these metals are called amount of both phosphorus and sulphur from the
thermionic. Included in this category are molybdenum molten bath, an important condition in the production
and tungsten. In the cold cathode, low melting point of steel for electrode core wire.
metals can be made to supply electrons more easily by The t e r m r i m m i n g refers to the m e t h o d of
coating them with compounds that reduce the surface deoxidising the steel in the final stage of its manufacture.
work function. They then become thermionic at low After the molten steel bath has been refined and the Si,
temperatures. Mn, C, S and P contents have been reduced to the desired
levels, it is still saturated with oxygen. In making
Electrode Core Wire rimmed steel, the oxygen-bearing metal is poured into
Core wire is an important component of a manual the cast-iron ingot moulds. The ingot commences to
electrode, because it significantly influences the solidify by depositing a crust of nearly pure iron on the
electrode performance and weld metal properties. Both walls and bottoms of the mould. As a result, the
rimming and non-rimming quality steel is used for the remaining liquid steel is enriched slightly with carbon,
core wire of mild steel and low alloy steel electrodes all which in turn, reacts with some of the oxygen to form
over the world, although former is preferred. Rimming carbon monoxide gas.
% C Si Mn p" S ~ V Ti Al
Rimming 0.10 max 0.03 max 0.38-0.62 0.030 max 0.030 max 0.15 max 0.005max 0.003 max 0.012 max
Non-rimming 0.10 max 0.03 max 0.38-0.62 0.025 max 0.025 max 0.15 max 0.005max 0.003 max 0.012 max
Much of the gas rises and escapes from the top of segregation in well-made rimmed steel is ± 3 points,
the ingot with a fiery shower of sparks. The remainder i.e., if the ladle analysis shows 0.025% sulphur, some
of the CO gas is trapped in the solidifying metal and portions of wire will have sulphur as low as 0.022% and
this counteracts the solidification contraction and others will have sulphur as high as 0.028%. This means
eliminates the shrinkage pipe. This cycle of: (1) iron that when the electrode maker specifies that the steel
freezing on the mould walls, and (2) the segregated for core wire shall have 0.030% max. of sulphur, the
carbon reacting with the oxygen, will continue until an steelmaker must ensure that his heat analysis gives
outer skin or rim of iron about 75 mm thick is formed. 0.027% max. sulphur and segregation does not exceed
Then the rimming action is stopped by the loss of ± 0.0030%.
temperature and completion of solidification. Rimmed The electrode industry consumes a large tonnage of
steel is therefore characterised by a skin of nearly pure core wire. Considering that it is a vital raw material, the
iron around the outer periphery of the cross-section and production of core-wire steel in a country like India is
central segregated core, which contains practically all planned at the national level. It must be stressed that
of the carbon, phosphorus, and sulphur. This peculiar special technological considerations are necessary for
pattern of segregation persists through forging, rolling, producing this steel. To quote from the relevant research
and wire drawing. The rimming action is facilitated by paper of the Welding Research Council of the U.S.A.:
the low carbon content specified for the steel for "Steel wire which is produced for use in gas and arc
electrode core wire, since the element carbon is a definite w e l d i n g is a special commodity, which requires
impediment to the rimming action. restrictive practices in its manufacture, in order that it
Killing and semi-killing are alternative methods of will give satisfactory performance during welding. Extra
deoxidising the steel. The killing operation is commonly precautions are taken in the manufacture of the steel, in
performed by the addition of deoxidisers like silicon (in the selection of the heats of steel, in the selection and
the form of ferrosilicon) and aluminium. Killed steel is preparation of the billets and in making tests of the rods
dense and homogeneous and has controlled grain size. and the finished wire. The chemical composition of the
In making semi-killed steel, a small amount of ferrosilicon steel must be closely controlled, and the wire furnished
or aluminium is added to the metal either in the furnace, to restricted chemical limits/'
the ladle or the mould. The amount is just sufficient to Rimming quality steel makes an ideal material for
kill any rimming action, and yet leaves some dissolved electrode core wire, because it gives to the latter the
oxygen. The evolution of this oxygen upon solidification characteristic rim/core structure explained earlier. The
of the metal tends to offset shrinkage and minimises the outer rim of pure iron helps the finally drawn wire to
shrinkage cavity or pipe in the top of the ingot. The chief get a faultless surface finish and to be less susceptible
reason for the production of semi-killed steel is the high to atmospheric corrosion. It also ensures uniform
yield of usable material from the ingot. passage of electric current and heat distribution, and
It is possible to distinguish between rimmed, semi- uniform melting of the electrode in the arc. The result is
killed and killed steels by determining the amount of good arc stability, spray type or fine globular metal
deoxidiser present. The silicon content of rimmed steels transfer and a moderate amount of fine spatter from
is 0.030% max., in the semi-killed steel it varies from the weld pool. The impurities concentrated at the core
0.030 to 0.10% and in killed steel it is over 0.10%. (It is give rise to gas evolution inside the melting drop and
assumed that Al content is in traces. Al should be this gives the liquid metal an impulse in the upward
avoided in electrode core wire.) In the selection of direction in vertical and overhead welding.
electrode quality wire, it is important, therefore, to make The fact that vaporisation pressure is the main
sure that the silicon content never exceeds 0.030%. driving force for the ejecting metal has been confirmed
Another method of checking on the rimming quality by Wegrzyn at the Polish Welding Research Institute by
steel is to make a sulphur print. The section of a wire of comparing the performance of bare rods of rimmed steel
well-rimmed steel will show an outer rim of pure iron and killed steel in overhead welding. Moreover,
and a well-defined core, which has all the impurities electrodes made fro m semi-killed or killed steel core
including sulphur concentrated in it. wire have been found to possess inferior operating
Rimmed steel is susceptible to sulphur segregation. characteristics. It is necessary to hold a longer arc to
In other words, sulphur distributes itself unevenly in prevent freezing of the electrode to the workpiece. The
the steel at the ingot stage and ultimately in the wire arc is unsteady and the metal transfer is stringy or ropey
that the electrode producer receives. The average and erratic. The spatter is large.
Restriction of carbon, sulphur and phosphorus to scarcely any elongation. The failure, of course, takes
uncommonly low levels in the electrode core wire (as place through the minute proportion of liquid that was
well as in the flux ingredients) becomes necessary, to have solidified last and presumably would have
because of the extremely severe conditions under which formed a grain boundary. Instead, a fissure has opened
the weld metal is formed, and also because of its high without visible deformation of the metal, and if the
susceptibility to hot-cracking and cold-cracking. surfaces of the crack are examined at room temperature,
Weld metal formed by the fusion of the electrode in they will be found temper coloured or scaled. Since the
the electric arc is deposited in a molten state in the recess liquid metal responsible for the cracking is in the spaces
of the joint. It solidifies and cools down very rapidly, at b e t w e e n the g r o w i n g crystals, the fracture is
rates almost amounting to quenching, because the intercrystalline or intergranular. In the case of the weld
a m o u n t of w e l d m e t a l is small a n d the heat is metal, its thermal contraction as explained earlier
immediately dissipated into the adjoining parent metal imposes stress upon the weld metal, while it is still
and air. This quenching effect has a tendency to harden solidifying, and this leads to hot-cracks.
the weld metal and reduce its ductility. To minimise this Referring to steels in general, we know that sulphur
tendency, carbon in the weld metal is kept as low as is one impurity in steel which has a particularly bad
possible. The weld metal also contracts appreciably reputation for causing hot-cracking. Sulphur reacts with
during solidification and cooling. This is called thermal iron to form iron sulphide, FeS, which is soluble in
contraction or shrinkage of the weld metal, and it is molten steel. A few hundredths of one per cent sulphur
responsible for the tendency of a welded fabrication to in steel as iron s u l p h i d e increases the r a n g e of
distort. In practice, the joint assembly is held under great temperature over which solidification occurs from about
restraint with the help of tack welds, jigs and fixtures. 20°C to possibly 480°C The last trace of sulphur-rich
The volumetric change in dimensions of mild steel liquid between the crystals does not freeze until about
from its melting t e m p e r a t u r e to a t m o s p h e r i c 980°C.
temperature is approximately 6.8%, which means that The presence of these envelopes of liquid iron
a cube of molten metal would contract approximately sulphide around the metal crystals or grains at so low a
2.2% in each of its three dimensions. In a welded joint, temperature invariably results in hot cracking. In the
the molten weld metal on solidifying becomes attached case of w e l d metal, conditions are m u c h m o r e
to the parts being joined and is restrained by them from favourable for hot-cracking to occur than during the
contracting freely in at least one and sometimes two formation of steel ingots or castings, and experience and
dimensions. Since the volumetric dimension of any research data have shown that it is necessary to restrict
metal at any one temperature can vary only within its sulphur in the case of wire to 0.030% maximum, to
elastic limit, and since the maximum elastic movement eliminate hot cracking in welded joints.
of steel is approximately only one-tenth of the total It is a well-known metallurgical fact that the hot-
requisite thermal contraction, the weld metal will flow cracking tendency of sulphur can be counteracted by
plastically and triaxially, effecting most of the requisite the element manganese, because manganese has greater
volumetric contraction along the axis or axes of least affinity for s u l p h u r t h a n iron, a n d the resultant
resistance. In doing this, it will exert considerable force manganese sulphide with its high freezing point is a
on the parts being joined, but since the parts are held in harmless constituent. The electrode core wire usually
restraint, the weld metal will be greatly strained. has about 0.40% manganese and it is difficult and
Measurements have shown that residual tension of uneconomical for the steel producer to raise this figure.
yield-point m a g n i t u d e exists in the longitudinal The electrode producer usually includes u p to 10%
direction in all welds over 450mm length, and in practice f e r r o m a n g a n e s e p o w d e r in his flux coating to
we deal mostly with welds much longer than 450 mm. compensate for manganese losses in the electric arc and
Excessive strain in the weld metal can result in cracking to ensure that the weld metal deposit has sufficient Mn
at some point during the cooling cycle. to take care of sulphur at 0.030%. If the electrode
Weld metal cracking can be of two types: hot-cracks producer has to deal with substandard core wire having
and cold-cracks. Hot-cracks take place when the weld sulphur as high as 0.040%, it is logical for him to think
metal is half liquid and half solid and has no tensile of raising the ferromanganese in the flux coating from
strength. If steel is vibrated, pulled or otherwise 10% to 15% to counteract the hot-cracking tendency of
d i s t u r b e d just as the last liquid is u n d e r g o i n g the higher sulphur level. Actual tests show that when
solidification, it fails at absurdly low stresses and with this abnormal adjustment is done, the slag of the
electrode becomes extremely fluid and it is difficult to service for some time. They are caused by excessive
use the electrode in vertical and overhead positions. The hardening of the weld metal or the heat-affected zone
spatter also increases. and hydrogen present in the arc atmosphere, or by
An important point to remember is that the hot- defects introduced into the weld by bad procedure or
cracking t e n d e n c y of s u l p h u r cannot be fully workmanship, or by the many possible combinations
counteracted with manganese from the flux coating. of temperature, stress and rate of loading.
According to D. Seferian, the welding cycle is so rapid Another adverse condition to which the weld metal
that the reaction leading to the formation of manganese is subjected is the pick-up of carbon and sulphur from
sulphide in preference to iron sulphide does not proceed the parent metal. The weld metal, especially the first
to completion. Moreover, experiments have shown that pass in a joint groove, may be diluted to the extent of
the influence of sulphur becomes obviously more 40-50% by the parent metal. Thus while the pure weld
marked as the carbon content of the steel is raised. This metal may have carbon as low as 0.10% and sulphur as
is depicted in Fig. 3.14, in which the cracking tendency low as 0.030%, the weld metal obtained due to dilution
is graphically represented as a function of the three by the parent metal ( C—0.25%, S—0.060%) can in
elements, carbon, manganeselind sulphur: It will be extreme cases have carbon as high as 0.18% and sulphur
noted that a steel containing 0.10% carbon and 0.60% as high as 0.045%. When an electrode producer specifies
manganese can tolerate up to 0.080% sulphur, while a electrode core wire with carbon lower than 0.10% and
steel containing 0.12% carbon and the same amount sulphur lower than 0.030%, he is making allowance for
manganese (0.60) will crack at a sulphur level of 0.060%. maximum amount of pick-up of these elements from
At 0.16% carbon, the sulphur content must not exceed the parent metal in practice.
0.020% if cracking troubles are to be avoided. As will be Very interesting results obtained on the measurement
explained later, the weld metal can attain as much as of pick-up of various elements by the weld metal are
0.18% carbon in a welded joint due to carbon pick-up given in Table 3.6, which is based on the work carried
from the parent steel plate having 0.25% carbon. out at the National Physical Laboratory, Teddington.
Cold-cracking occurs at a fairly low temperature, and Very significant are cases C and D. In the former, the
the path of the crack is intergranular, i.e. it travels across core wire had only 0.025% sulphur but the weld metal
the grains. While hot-cracks occur during the welding had sulphur as high as 0.033%. In the latter, the core
operation, or at least before the heat has completely left wire had 0.026% sulphur but the weld metal acquired
the joint, cold-cracks develop very slowly and are 0.043% sulphur due to pick-up from the parent plate.
sometimes exposed after the weldment has been in Both these cases refer to electrodes with rutile coating.
Uncontrolled sulphur in the core wire not only causes
hot-cracks in the weld metal but it also causes severe
weld pitting and upsets the performance characteristics
of the electrode. As far back as 1950, the American Iron
and Steel Institute in the U.S.A., moved the Welding
Research Council to carry out a research programme to
find out w h y American electrode producers were
insisting upon very low sulphur in the electrode core
wire. The result of this was the publication of a paper
entitled Core wire for electrodes, which appeared in the
Welding Journal of June 1950. Following is one of the
conclusions arrived at by the authors and it supports
the experience of every electrode manufacturer:
"The most important single core-wire variable,
which affected the operating characteristics of covered
electrodes was the average sulphur content. Sulphur
0.4 0 5 0.6 0.7 0.8 0.9 1.0 1.1 contents of more than 0.040% caused the electrodes to
have deficient operating characteristics and severe weld
MANGANESE CONTENT %
pitting tendencies. Sulphur contents of between 0.030
Fig. 3.14: The influence of sulphur on weld cracking as and 0.040% produced electrodes which had severe weld
a function of C and Mn contents of steel p i t t i n g tendencies b u t acceptable operating
Table 3.6: Study of pick-up of elements in the weld metal from parent plate
Chemical composition Coating
Electrode
C Mn Si S P type
Parent metal 0.18 0.53 0.043 0.042 0.032
A Core wire 0.12 0.49 0.010 0.026 0.017 acid
Weld 0.09 0.50 0.257 0.026 0.035
B Core wire 0.07 0.34 0.010 0.027 0.030 acid
Weld 0.05 0.24 0.220 0.029 0.036
C Core wire 0.07 0.33 0.015 0.025 0.005 rutile
Weld 0.07 0.47 0.160 0.033 0.025
D Core wire 0.12 0.41 0.020 0.026 0.018 rutile
Weld 0.05 -0.23 0.160 0.043 0.026
E Core wire 0.09 0.46 0.010 0.026 0.008 ilmenite
Weld 0.07 0.24 0.243 0.013 0.034
F Core wire 0.05 0.52 0.10 0.022 0.016 rutile
Weld 0.05 0.50 0.31 0.029 0.036
characteristics (see Fig. 3.15). The best results were mention the brand name. It enables one to pick-up
obtained with core wires which had an average sulphur equivalent brands from different electrode producers.
content of about 0.030% or below." In the absence of a descriptive leaflet, one can know the
main properties of a product by merely looking at the
CLASSIFICATION AND CODING OF ELECTRODES code number mentioned on the electrode carton or on
Standardisation of MMA electrodes is techno- the electrode itself. In technical l i t e r a t u r e a n d
logically as important as that of commercial metals and discussions, the writer or speaker can use only the
alloys. Quite often standardisation of the two goes on coding to refer to a particular class of electrode.
side by side. In some cases the system of specification Various systems of electrode classification operate
adopted is nearly identical. A good example is the in different countries; but the most important is the one
specification for stainless steels and the specification for issued by the I n t e r n a t i o n a l O r g a n i s a t i o n for
stainless steel electrodes. Standardisation (ISO). Other systems which are of
Electrodes have been standardised by dividing them interest to the Indian welding engineer are described in
into classes depending upon the type of flux covering, the relevant standards issued by the Bureau of Indian
performance characteristics and either weld metal Standards (BIS), British Standards Institution (BSI), the
chemistry or all-weld mechanical properties or both. American Welding Society (AWS) and Deutsches
Each class is then coded according to a coding system Institut fur Normung (DIN). They cover some or all of
comprising a group of letters and numbers. the following groups of electrodes:
Advantages of codification can be easily appreciated. 1) Mild s.teel electrodes
It enables one to specify or identify an electrode in the 2) Low-alloy steel electrodes
purchase order or in the drawing, without having to 3) Stainless steel electrodes
Jmf_$$j WW^-
Fig. 3.15: Weld pittings in high-sulphur weld-metal

4) Surfacing electrodes These symbols are next followed by one or two letters
5) Electrodes for cast iron indicating the type of covering as follows:
6) Copper and copper-alloy electrodes
7) Nickel and nickel-alloy electrodes A = Acid (iron oxide)
8) Aluminium and aluminium-alloy electrodes AR = Acid (rutile)
They are summarised in the following sections. B = Basic
C = Cellulosic
O = Oxidising
MILD STEEL ELECTRODES
R = Rutile (medium coated)
1) ISO-2560-1973(E): "Covered electrodes for manual RR = Rutile (heavy coated)
arc welding of mild steel and low-alloy steel—Code of sym-
S = Other types
bols for identification"'.
This standard gives a coding system applicable to ISO's description of these coverings (except S) has
mild steel as well as low-alloy steel el ectrodes. It starts been provided earlier. The symbol S is reserved for any
with a prefix E, which indicates covered electrodes for new type of coating, which is not covered by the
manual arc welding. standard symbols. When iron powder is added to a
Next comes the symbol 43 or 51. The number 43 indi- standard coating, the original symbol of the coating is
cates that the all-weld tensile strength is in the range of 430- retained and symbol S is not used.
510 N/mm 2 . The symbols u p to this stage m a k e u p the
The number 51 indicates that the all-weld tensile compulsory section of the codification. The symbols that
strength is in the range of 510-610 N / m m 2 . To make follow make up the optional section of the codification.
allowance for possible variations in welding and test- Next comes the symbol for nominal electrode
ing, the upper limits of 510 and 610 N / m m 2 may be efficiency (i.e. weld deposition efficiency or the ratio of
exceeded by 40 N / m m 2 . In reality, therefore, the two w e l d metal to core wire melted, expressed as a
ranges of tensile strength are 430-550 N / m m 2 and 510- percentage). It can be either of 110,120,130,140,..., etc.
650 N / m m 2 . A value ending with 4 or less is rounded off to the lower
For each range of tensile strength, six groups are multiple of 10, while that ending with 5 or more is
specified on the basis of Charpy-V impact values and rounded off to the higher multiple of 10. Thus 115
elongation (i.e. combination of toughness and ductil- becomes 120, while 134 becomes 130. No symbol is used
ity). for a value less t h a n 105%. Next comes a digit
These groups are given the number 0 , 1 , 2, 3, 4 or 5 designating the general welding positions, for which
as shown in Table 3.7. the electrode is suitable as follows:
Table 3.7: Electrode designation as per ISO-2560

Electrode Tensile strength* Minimum elongation Temp. for minimum
designation on 1 = 5d impact value of 28 J+
N/mm2 % °C
E43 0 430 to 510 — —
E431 430 to 510 20 +20
E43 2 430 to 510 22 0
E43 3 430 to 510 24 -20
E43 4 430 to 510 24 -30
E43 5 430 to 510 24 -40
E510 510 to 610 — —
E511 510 to 610 18 +20
E512 510 to 610 18 0
E513 510 to 610 20 -20
E514 510 to 610 20 -30
E515 510 to 610 20 -40
* Upper limit tolerance: +40 N/mm2
1 1 J = 0.102 kgf-m
1: all positions b) A basic iron powder type electrode gives the fol-
2: all positions, except vertical downward lowing minimum mechanical properties
3: flat butt weld, flat fillet weld, horizontal/vertical • Tensile strength : 560 N / m m 2
fillet weld • Elongation : 22%
4: flat butt weld, flat fillet weld • Impact strength : 47 J at -20°C.
5: as in 3 and recommended for vertical downward • Nominal efficiency: 158%
Next comes the symbol for electrical characteristics,
i.e. whether the electrode operates with DC only or with It may be used for welding in all positions except
AC as well as DC, the polarity on DC, and the minimum vertical downward, direct current only. The complete
open-circuit voltage on AC necessary for sustaining the codification for the electrode will therefore be E 51 3B
arc. It is explained in Table 3.8. 160 20 (H) and the compulsory part will be E 51 3B.
The last symbol is H, which is to be used only for
hydrogen-controlled electrodes, i.e. those giving a 2) BS 639: 1976: "Covered electrodes for the manual
maximum content of 15 ml of diffusible hydrogen per metal arc welding of carbon and carbon manganese steel".
100 g of d e p o s i t e d m e t a l w h e n d e t e r m i n e d in This standard effectively combines in revised form
accordance with the standard method explained later. what was previously covered by the separate standards
BS 639 "Covered electrodes for the manual metal-arc
welding of mild steel and medium tensile steel" and BS
EXAMPLES
1719 "Classification, coding and marking of covered
a) A medium-coated rutile type electrode gives electrodes for metal-arc welding" Part 1: "Classification
the following minimum all-weld mechanical proper- and coding". The revised version is based on ISO 2560,
ties: except for a few extensions or modifications which have
Tensile strength 500 N / m m 2 been pointed out in the following paragraphs.
Elongation 23% The coding starts with the prefix E, which is followed
Impact strength 71 J at + 20°C by digits 43 and 51 as in ISO 2560 However, a minimum
3 1 J a t 0°C yield stress value is specified along with tensile strength
20Jat-20°C range as follows:
It may be used for welding in all positions. It welds
satisfactorily on alternating current, minimum open- Designation Tensile strength Min. yield stress
circuit voltage 50 V, and direct current, positive polarity. N/mm2 N/mm2
The complete codification fo^ the electrode will therefore E43 430-550 330
by E 43 2R 13 and the compulsory part will be E 43 2R. E51 510-650 360
Table 3.8: Symbols for electrical characteristics in ISO-2560

Direct current Alternating current
Symbol recommended electrode nominal open-circuit
polarity** voltage
V
0* +
1 + or- 50
2 50
3 + 50
4 + or- 70
5 70
6 + 70
7 + or- 90
8 90
9 + 90
* Symbol reserved for electrodes used exclusively on direct current.
** Positive polarity +, negative polarity -
Next comes a two-digit figure indicating the b) a figure indicating the welding positions in which
combination of elongation and impact values of the weld the electrode can be used
metal as shown in Tables 3.9 and 3.10. It should be noted c) a figure indicating the welding current and volt-
that the second digit is an additional provision for higher age conditions recommended
elongation and impact values demanded for critical d) the letter (H) for a hydrogen controlled electrode
applications, which is non-existent in ISO 2560. Next (when appropriate).
comes a letter or letters indicating the type of covering All these four parts of the coding are identical to
as given below, which is exactly as in ISO 2560: those in ISO 2560. However, in the case of electrodes
A = Acid (iron oxide) carrying suffix (H), BS 639 additionally stipulates that
AR = Acid (rutile) the manufacturer shall give on the packet and in his
B = Basic literature the recommended drying conditions for
C = Cellulosic achieving hydrogen levels within the following three
O = Oxidising ranges, as appropriate:
R = Rutile (medium coated)
RR = Rutile (heavy coated) over 10 ml, up to and
S = Other types including 15 ml
over 5 ml, up to and per 100 g of deposited
This completes the compulsory part of the coding in
including 10 ml weld metal
the same way as in ISO 2560. Then follows the optional
up to and including 5 ml
part comprising:
a) a three-digit figure (110, 120, 130, 140,...) for the BS 639 gives the following two examples to illustrate
nominal electrode efficiency (when appropriate) the method of expressing the coding:
Table 3.9: First digit for elongation and impact strength in BS 639
First Minimum elongation Temp, for impact value of 28 J

digit
E43 E51
Not specified Not specified

20 18 +20
22 18 0
24 20 -20
24 20 -30
24 20 -40
Table 3.10: Second digit for elongation and impact strength in BS 639
Second Minimum elongation Impact properties
digit Impact value Temperature
E43 E51 E43 E 51
% % J J °C
Not specified* Not specified*
22 22 47 47 +20
22 22 47 47 0
22 22 47 47 -20
Not relevant* 18 Not relevant* 41** -30
There are no electrodes designated by this digit ^0
Not relevant* 18 Not relevant* 47 -50
* There is no E 43 equivalent to an E 51 electrode that could be designated by the second digit 4 or 6.
** No one impact value should be less than 20 ].
EXAMPLE (A) 3) DIN 1913 (1984): ''Coated electrodes for the welding
Covered electrode for manual metal are welding of unalloyed and low-alloy steels"
having a rutile covering of medium thickness and The German system of coding specified in DIN 1913
deposit weld metal with the following m i n i m u m is based on ISO 2560 with extensions or modifications
mechanical properties: as in the case of BS 639. It starts with prefix E followed
by two digits 43 or 51 which indicate the following:
Tensile strength: 500 N / m m 2 ;
Elongation: 23%
Impact strength: 71J at +20°C, 37 J at 0°C, 20 J at -20°C Symbol Tensile strength range Min. yield stress
It may be used for welding in all positions. It welds N/mm2 N/mm2
satisfactorily on alternating current with a minimum 43 430-550 360*
open-circuit voltage of 50 V and on direct current with 51 510-650 380*
positive polarity. The complete classification for the * Theseare higher than in BS 639.
electrode would therefore be: E4321R13
E 43 21 R 1 3
and the compulsory part would be E 43 21 R
Covered electrode for manual metal-arc welding
Tensile strength
Elongation and impact strength
Covering
Welding positions
Current and Voltage
EXAMPLE (B) They are followed by two digits, the first digit
An electrode for manual metal-arc welding having indicating the combined levels of elongation and
a basic covering with a high efficiency and depositing Charpy-V impact strength (28 J) and the second digit
weld metal containing 8 ml of diffusible hydrogen per indicating higher Charpy-V impact (47 J). This is
100 g of deposited weld metal with the following explained in Tables 3.11 and 3.12.
minimum mechanical properties: After this, DIN 1913 makes a radical departure from
Yield stress: 380 N / m m 2 both ISO 2560 and BS 639. It provides a classification
Tensile strength: 560 N / m m 2 based on
Elongation: 22% (a) the type of coating,
Impact strength: 47 J at -20°C (b) welding positions, and
Nominal efficiency: 158% (c) welding current conditions, and then uses the
It may be used for welding in all positions except classification number to designate each type of elec-
vertical-down direct current only. trode.
The complete classification for the electrode would a) Firstly, the types of coating indicated by a letter or
therefore be as depicted below: letters are:
E 51 33 B 160 2 0 (H)
and the compulsory part would be E 51 33B
Covered electrode for manual metal-arc welding
Tensile strength*
Elongation and impact strength
Covering
Efficiency
Welding position
Current and voltage
Hydrogen controlled
Table 3.11: First digit for elongation and impact strength c) Welding current conditions are indicated by code
in DIN 1913 numbers in the same manner as in ISO 2560 and BS 639,
except that in case of 0,
First Min. elongation Temp, for min. impact
0 means DC only, + or - polarity
digit on I=5d (%) value of 28 J at °C
0+ means DC only, + polarity
0 Nil Nil 0" means DC only, - polarity
1 +20 Combining (a), (b) and (c), 12 classification numbers
2 22 0
are given and the electrode types are designated as
3 -20
4 24 -30 shown in Table 3.13.
5 -40 After the coding for the class of electrode as shown
in the above table comes a three-digit number (100,120,
130, 140, . . . etc.) indicating the deposition efficiency,
which' is to be used only when it exceeds 105%. This
Table 3.12 : Second digit for impact strength in DIN
part of the coding in identical to that used in ISO 2560
1913
and BS 639.
Second digit Temp, for min. impact value of EXAMPLES (as given in DIN 1913)
of47Jat°C a) A coated electrode which
0 Nil gives all-weld tensile strength in the range of 430-
1 +20 550 N / m m 2
2 0
3 -20 gives all-weld yield stress of more than 360 N /
4 -30 mm 2
5 -40 gives minimum elongation of 24%
gives minimum Charpy-V value of 28 J at -20°C
A — acidic and 47 J at 0°C, has acid-rutile coating
R — rutile (thin and medium coated)
belongs to class, AR7, and
RR — rutile (heavy coated)
AR ^ - acid-rutile (mixed type) gives a deposition efficiency of less than 105%,
C — cellulosic will have the coding: E43 32 AR7 DIN 1913
R(C) - rutile-cellulose (medium coated) b) A coated electrode which
RR(C) — rutile-cellulose (heavy coated) gives all-weld tensile strength in the range of 510-
B — basic 650 N / m m 2
B(R) - basic-coated with non-basic
gives all-weld yield stress of more than 380 N /
components
mm 2
RR(B) — rutile-basic (heavy coated)
NOTE: DIN 1913 defines thin coated as having a gives minimum elongation of 24%
coating diameter of up to 120% of the core wire size; gives minimum Charpy-V value of 28 J at -20°C
medium coated as having a coating diameter of 120- and 47 J at 0° C, has heavy rutile coating
155% of the core wire size; and heavy coated as having belongs to class RR11, and
a coating diameter over 155% of the core wire size.
gives a deposition efficiency between 155 and
b) The welding positions in which the electrode can
164%,
be used are indicated by a code number as follows:
1 — all positions will have the coding: E51 32 RRII 160 DIN 1913
2' — all positions except vertical DIN 1913 gives the following information on each
downward classification number:
3 — butt weld, flat Class 1: Class 1 includes thin coated acid type A I,
fillet weld, flat without mechanical property requirements, for all-
fillet weld, horizontal position welding.
4 — butt weld, flat Class 2: In this class are included thin coated acid
fillet weld, flat type A2 and rutile type R2. The mechanical property
Number 5 which exists in ISO 2560 and BS 639 is requirements are limited. Both types are for all-position
omitted here. welding.
Table 3.13: Classification numbers of electrodes in DIN 1913

Electrode Welding Current Coating Classification
type position code* condition** see (a) above number
see (b) above see (c) above
Al 1 5 thin coated A 1
A2 1 5 thin coated A 2
R2 1 5 thin coated R
R3 2(1) 2 medium coated R 3
R(C)3 1 2 medium coated R(C)
C4 1*** 0+(6) medium coated C 4
A5 2 5 heavy coated A 5
RR6 2 2 heavy coated RR 6
RR(C)6 1 2 heavy coated RR(C)
AR7 2 5 heavy coated AR 7
RR(B)7 2 5 heavy coated RR(B)
RR8 2 2 heavy coated RR 8
RR(B)8 2 5 heavy coated RR(B)
B9 1*** 0+(6) heavy coated B 9
i***
B(R)9 6 heavy coated B(R)
BIO 2 0+(6) heavy coated B 10
B(R)10 2 6 heavy coated B(R)
RR11 4(3) 5 RR with dep. eff. 3 105% 11
AR11 4(3) 5 AR with dep. eff. 3 105%
B12 4(3) 0+(6) B with dep. eff. 3 120% 12
B(R)12 4(3) 0+(6) B (R) with dep. eff. 3 120%
^Bracketed code numbers for welding positions apply only to smaller sizes and/or low levels of deposition efficiency.
**Bracketed code numbers for current conditions mean conditional qualification.
***Favoured for vertical down.
Class 3: This class includes medium coated rutile Rutile content in the coating is approximately 50%, more
type R3 and medium coated rutile-cellulose type R(C)3. than in all other electrode varieties. Good are restriking
The R(C)3 type is similar to R3, except that it has higher and a regular smooth weld bead are its special
cellulose content. The R3 type is suitable for all characteristics. The type RR(C)6 is suitable for all
positions except vertical downward, while the R(C)3 positions while RR6 type is suitable for all positions
type is suitable for all positions including vertical except vertical downward.
downward. Class 7: This class covers the heavy coated acid-rutile
Class 4: The medium coated cellulosic type C4 (mixed) type AR7 and rutile-basic type RR(8)7. The
contains a large proportion of combustible materials, coating of AR7 type contains approximately 35% rutile
especially cellulose, in the coating. It is suitable for and acidic components, preferably iron ore. Its welding
welding in all positions and is specially suited for the characteristics correspond largely to those of the acid
vertical downward position. type of class 5. When the basic components of the AR7
Class 5: This class includes heavy coated acid type type coating are doubled in quantity (up to nearly 20%)
A5. Its distinguishing feature is a high proportion of and the iron ore content is reduced to almost nil, one
iron ore in the coating. The metal transfer across the arc arrives at the rutile-basic type RR(B)7. This type gives
is in fine drops. It is used in all positions except vertical very good weld mechanical properties and consistently
downward. favourable welding performance in all positions except
Class 6: In this class are included heavy coated rutile vertical downward. The AR7 type is also suitable for all
type RR6 and heavy coated rutile-cellulose type RR(C)6. positions except vertical downward.
Class 8: In this class, the heavy coated rutile type RR8 content in the weld deposit is 1.2% or less. In dried
and the heavy coated rutile-basic type RR(B)8 are condition, both give less than 10 ml of diffusible
brought together. The type RR8 differs from the rutile hydrogen per 100 g weld deposit. Regarding welding
type of class 6 (RR6) by having a higher proportion of positions, remarks given under Class 11 hold good.
basic components in the coating, whereby the all-weld 4) IS:814-1991 "Covered electrodes for manual metal
mechanical properties improve as also the deformability arc welding of carbon and carbon manganese steel-
of the weld metal. The type RR(B)8 contains more Mn specification".
in the weld deposit as compared to type RR(B)7 (up to Main Coding Consists of the following letters and
approximately 0.90%) and thereby gives higher tensile numerals and is followed in the order stated :
values. Both RRB and RR(B)8 can be used in all positions a) a prefix letter 'E' shall indicate a covered elec-
except vertical downward. trode for manual metal are welding, manufactured by
Class 9: This class includes the heavy coated types extrusion process;
B9 and B(R)9, whose weld deposit has Mn content of b) a letter indicating the type of covering (see 5.1);
1.2% or less. B9 has a basic coating, consisting mainly c) first digit indicating ultimate tensile strength in
of alkaline carbonates and fluorspar. Small additions of combination with the yield stress of the weld metal de-
non-basic components are made to the coating of B(R)9. posit (see 5.2);
Both the types are suitable for all positions and d) second digit indicating the percentage elongation
especially for vertical downward. When used in dried in combination with the impact values of the weld metal
condition, both give less than 10 ml of diffusible deposited (see 5.3);
hydrogen per 100 g weld deposit. Both have deposition e) third digit indicating welding position(s) in which
efficiency of less than 105%. the electrode may be used (see 5.4); and
Class 10: This class includes heavy coated basic types f) fourth digit indicating the current condition in
B10 and B(R) 10. The Mn content in the weld deposit is which the electrode is to be used (see 5.5)
1.2% or less as in the types of class 9. Type B10 gives
very good all-weld mechanical properties, particularly 4.0 Additional Coding
sub-zero impact strength. Type B(R)10 has small The following letters indicating the additional
additions of non-basic components, such as rutile and properties of the electrodes may be used, if required;
zircon, just sufficient to improve the welding a) letters HI, H2, H3 indicating hydrogen controlled
performance and not to disturb the basicity of the slag, electrodes (see 5.6)
which is responsible for the high all-weld mechanical b) letters J, K and L indicating increased metal re-
properties. Both the types are suitable for all positions, covery as 'Effective Electrode Efficiency' as per IS 13043:
except vertical downward. In dried condition, both give 1991. Specification in the followings range (see 5.7):
less than 10 ml of diffusible hydrogen per 100 g weld J =110-129 percent;
deposit. Both have deposition efficiency of less than K = 130 -149 percent; and
120%. L = 150 percent and above.
Class 11: This class includes high efficiency electrodes c) letter X indicating the radiographic quality (see 5.8).
with iron powder additions to the coating. RR11 is heavy
coated rutile type, while AR11 is heavy coated acid-rutile 4.1 Type of Covering
type, both giving deposition efficiency of 105% or more. Type of covering shall be indicated by the following
Mn content in the weld deposit is 1.0% or less. The letters:
positions in which they can be used depend on the A—Acid
electrode size and the level of deposition efficiency, B—Basic
though they are primarily designed for butt and fillet C—Cellulosic
welds in the downward position and for fillet welds in R—Rutile
the horizontal position. RR—Rutile, heavy coated
Class 12 : This class includes high efficiency iron S—Any other type not mentioned above.
powder types B12 and B(R)12. Type B12 has a basic
coating with appreciable iron powder additions, while 4.2 Strength Characteristics
B(R)12 has a basic coating with small additions of non- The combination of ultimate tensile strength and
basic components and appreciable amount of iron yield strength of the weld metal deposits shall be
powder. The deposition efficiency is 120% or more. Mn indicated by the digits 4 and 5 (see Table 3.14).
4.3 Elongation and Impact Properties 4.4.1 Where an electrode is coded as suitable for
The combination of percentage elongation and vertical and overhead positions it may be considered
impact properties of all-weld metal deposited for the that sizes larger than 4 mm are not normally used for
two tensile ranges (see Table 3.14) shall be as given in welding in these positions.
Table 3.15. 4.4.2. An electrode shall not be coded as suitable for
a particular welding position unless it is possible to use
Table 3.14: Designation of Strength Characteristics it satisfactorily in the position to comply with test
(Clauses 5.2 and 5.3) requirements of this code.
Designating Ultimate Tensile Yield strength
digit strength 4.5 Welding Current and Voltage Conditions
The w e l d i n g current and open circuit voltage
4 410 - 510 330 conditions on which the electrodes can be operated as
5 510 - 610 360 recommended by the manufacturer shall be indicated
by the appropriate designating digits as given in Table
3.16.
Table 3.15: Combination of percentage elongation 4.5.1 For the purpose of coding an electrode or any
and impact strength (Clause 5.3) of the current conditions under 5.5 shall be of size 4 m m
Designation Percentage Elongation Impact strength or 5 mm and shall be capable of being operated at that
digit (min) on 5.65/So in joules (min) at QC condition satisfactorily within the current range
(1) (?) (3) recommended by the manufacturer.
For tensile range
410 - 510 N/mm 2 4.6 Hydrogen Controlled Electrodes
0 No elongation and impact requirements The letters H I , H2 and H3 shall be included in the
1 20 47J/+27C classification as a suffix for those electrodes which will
2 22 47J/+0UC give diffusible hydrogen per 100 gm when determined
3 24 47J/-20°C in accordance with reference method given in IS 11802 :
4 24 27J/-30°C 1986 as given below:
For tensile range HI = Up to 15 ml diffusible hydrogen
510 - 610 N/mm 2 H2 = Up to 10 ml diffusible hydrogen
0 No elongation and impact requirements H3 = Up to 5 ml diffusible hydrogen
1 18 47J/+27C
2 18 47J/ +0gC 4.7 Increased Metal Recovery
3 20 47J/-20eC
The letters J, K and L, shall be included in the
4 20 27]/ -30UC
5 20 27]/ -40CC classification as a suffix for those electrodes which have
6 20 27J/ -46SC appreciable quantities of metal powder in
5.7.1 The Metal Recovery shall be determined as
4.4 Welding Position Effective Electrode Efficiency (E) as per the method
The welding position or positions in which the given in IS 13043:1991.
electrode can be u s e d as r e c o m m e n d e d by the
manufacturer shall be indicated by the appropriate 4.8 Radiographic Quality Electrodes
designating digits as follows: The letter 'X' shall be included in the classification
as a suffix for those electrodes w h i c h d e p o s i t
1) All positions radiographic quality welds.
2) All positions except vertical down
3) Flat butt weld, flat fillet weld and horizontal/ver- TYPE OF FLUX COVERING
tical fillet weld Type-1 Acid (A)
4) Flat butt weld and flat fillet weld Electrodes of acid type have a medium or thick cov-
ering and produce an iron oxide, manganese oxide, silica
5) Vertical down, flat butt, flat fillet and horizontal
rich slag (with some titania in some cases), the metal-
and vertical fillet weld
lurgical characteristic of which is acidic. The covering
6) Any other position or combination of positions confains, besides oxides, of iron or manganese (with
not classified above. some titania in some cases), a fairly high percentage of
Table 3.16: Welding current and voltage conditions (Clause 4.5)

Digit Direct current recommended Alternating Current
electrode polarity Open Circuit Voltage,
V
(1) (2) (3)
00 + Not recommended
1 + or- 50
2 - 50
3 + 50
4 + or- 70
5 - 70
6 + 70
7 + or- 90
8 - 90
9 + 90
1) Symbol 0 reserved for electrodes used exclusively on direct current.
2) Positive polarity+, Negative polarity -.
Notes :
1. The frequency of the alternating current is assumed to be 50 or 60 HI. The open circuit voltage necessary when electrodes are used on direct current
is closely related to the dynamic characteristics of the welding power source. Consequently no indication of the minimum open-circuit voltage for direct
current is given Annex E. of IS 814-1991.
ferro-manganese and / o r other deoxidisers. The slag These electrodes must be stored in a reasonably dry
generally solidifies in a characteristic honeycomb struc- place and should be dried before use, according to the
ture and is easily detached. recommendation of the manufacturer. This ensures that
This type of electrode usually has a high fusion rate the weld metal will have a low hydrogen content and
and may be used w i t h high current intensities. there is a less risk of underbead cracking when welding
Penetration can be good, particularly if the covering is steel likely to show a marked hardening in the heat
thick. These electrodes are most suitable for welding in affected zone.
flat position but can be used in other positions and can
be operated both on A.C. and D.C. Type-3 Cellulosic (C)
Covering of the cellulosic type contains large
Type-2 Basic (B) quantity of combustible organic substances, so that the
Electrode of basic type usually have a covering d e c o m p o s i t i o n of the latter in arc p r o d u c e s a
containing appreciable quantities of calcium or other voluminous gas shield. The amount of slag produced is
basic carbonates and fluorspar so that metallurgically small and the slag is easily detached.
they are basic in character. This type of electrode is characterised by a highly
There is a medium quantity of dense slag, which penetrating arc and fairly high fusion rate. Spatter losses
often has a brown to dark-brown colour and a glossy are fairly high and the weld bead is somewhat coarse,
appearance. It is easily detached, and as it rises to the with unevenly spaced ripples. These electrodes are
surface of the weld very quickly, slag inclusions are not usually suitable for welding in all positions. Generally
likely to occur. This type of electrode gives an arc of this type of electrodes are suitable for use on D.C. with
average penetration, and is suitable for welding in all electrode positive, but some electrodes are also available
positions. This type of electrode is used both on A.C. which are suitable for use on A.C.
and D.C. where D.C. positive polarity is generally
preferred for critical applications. Type-4 Rutile (R)
As the weld metal is highly resistant to hot and cold These electrodes have a covering containing a large
cracking, these electrodes are particularly suitable for quantity of rutile or components derived from titanium
welding heavy sections and very rigid mild steel struc- oxide. The electrodes have smooth arc characteristics
tures. They are also recommended for welding low alloy and normally produces very little spatter and are
steels and steels, carbon and sulphur content of which are comparatively easy to use. This type of electrode can be
higher than those of mild steel of good weldable quality. generally used in all positions and can be operated both
on A.C. and D.C. The slag detachability is generally means 60,000 psi (lbs/sq in). Similarly, 70 means
good. minimum weld metal tensile strength of 70 ksi or 70,000
lbs/sq. in. The actual stipulated minimum tensile
iype-5 Rutile, Heavy Coated (RR) strength values and the associated yield strength values
Generally similar in characteristics to rutile type but vary according to the type of covering as shown in Table
having a higher coating ratio (above 1.5). Application- 3.17. The impact strength requirements are given in
wise it is usually preferred for flat and horizontal vertical Table 3.18. The third digit indicates the position in which
position though welding in other positions can also be the electrode can be used satisfactorily, as follows:
possible.
1 F, V, OH, H
2 F, H-fillet
iype-6 Other Types (S)
4 F, OH, H, V-down
Electrodes with other type of coverings may range
from rarely used types such as oxide or acid fluxes, to The last two digits taken together indicate current
newly developed flux systems. No general guidance on conditions (type of current) and the type of covering.
special covering electrode characteristics is possible, so Table 3.19 shows the complete classification, as also the
potential users should seek the manufacturers advise. significance of the last two digits. Table 3.20 gives the
5) AWS A5.1/5.1M 2004 - "Specification for Carbon chemical composition limits of the weld metal.
Steel Electrodes for Shielded Metal Arc Welding ". The AWS standard stands distinctly apart from the
The American coding system starts with the prefix four discussed earlier, especially on the following points:
E, which designates an electrode. Then comes a two- 1) No attempt has been made to follow the pattern
digit number 60 or 70. The number 60 designates tensile of ISO 2560. The traditional system based on foot-pound
strength of at least 60 ksi of the weld deposit. 60 ksi system (60..., 70...) is still maintained.
Table 3.17: Tension test Requirements a,b,c,
AWS classification Tensile s►trength Yield strength at 0.2% Offset Elongation
Percentage in 4x
A5.1 A5.1M A5.1(ksi) A5.1M(MPa) A5.1 (ksi) A5.1M (MPa) Diameter Length
E6010 E4310 60 430 48 330 22
E6011 E4311 60 430 48 330 22
E6012 E4312 60 430 48 330 17
E6013 E4313 60 430 48 330 17
E6018 E4318 60 430 48 330 22
E6019 E4319 60 430 48 330 22
E6020 E4320 60 430 48 330 22
E6022d E4322d 60 430 Not specified Not specified
E6027 E4327 60 430 48 330 22
E7014 E4914 70 490 58 400 17
E7015 E4915 70 490 58 400 22
E7016 E4916 70 490 58 400 22
E7018 E4918 70 490 58 400 22
E7024 E4924 70 490 58 400 17 e
E7027 E4927 70 490 58 400 22
E7028 E4928 70 490 58 400 22
E7048 E4948 70 490 58 400 22
E7018M . E4918M Notef Notef 53-728 370-500* 24
Notes
a. See Table 4 for sizes to be tested.
b. Requirements are in thewelded
as-\ condition with aging as specified 12.2
in ,'.
c. Single values are minimum
d. A transverse tension test,>specified
at in 12.5 ana ' bend test, as specified in Section 13
I a longitudinal guided required.
are
e. Weld metal from electrodesidentified as E7024-1 (E4924-1) shall have elongation of 22% minimum.
/ Tensile strength of this weld metal is a nominal 70 ksi (490 MPa).
g- For 3/32 in (2.4 mm) electrodes, the maximum yield strength shall be 77 ksi (530 Mpa).
Table 3.18: Charpy V-Notch Impact Requirements

AWS classification Limits for 3 out of 5 specimens3
A5.1 A5.1M Average.min Single value, min
E6010, E6011, E6018 E4310, E4311, E4318
E6027, E7015, E4327, E4915 20ft.lbfat-20°F 15ft.lbfat-20°F
E7016b, E7018b E4916b, E4948b (27J at -30°C) (20J at -30°C)
E7027, E7048 E4927, E4948
E6019 E4319 20ft.lbfatO°F 15ft.lbfatO°F
E7028 E4928 (27J at -20°C) (20J at-20°C)
E6012, E6013. E4312, E4313
E6020, E6022. E4320, E4322 Not specified Not specified
E7014, E7024b E4914, E4924b
AWS Classification Limits for 5 out of 5 specimens0
A5.1 A5.1M Average, min Single Value, Min.
E7018M E4918M 50ft.lbfat-20°F 40ft.lbfat-20°F
(67J at-30°C) (54J at-30°C)
Notes:
a. Both the highest and lowest test values obtained shall be disregarded in computing the average. Two of these remaining three values shall equ
exceed 20 ft-lb (27 J).
b. Electrodes with the following optional supplemental designations shall meet the lower temperature impact requirements specified below:
Electrode Charpy V-Notch Impact Requirements,
AWS classification Designation Limits for 3 out of 5 specimens (Refer to footnote a above)
A5.1 A5.1M A5.1 A5.1M Average, min Single value, min

E7016 E4916 E7016-1 E4916-1 20ft.lbf at -50°F 15ft.lbfat50°F
E7018 E4918 E7018-1 E4918-1 (27J at -45°C) (20Jat-45°C)
E7024 E4924 E7024-1 E4924-1 20ft.lbfatO°F 15ft.lbfatO°F
(27Jat-20°C) (20J at -20°C)
c. Allfivevalues obtained shall be 'used
in computing the
average. Four of thefivevalues shall equal, or exceeds,
50ft-lbf(67J)
2) Lower limit of tensile strength for 60 series is set Apparently, ISO 2560 and the various national stan-
at 420 N/mm 2 , as against 430 in ISO 2560 and 410 in IS dards based on it have put forward a universal coding
815. No upper limit is specified. system, in which all possible electrodes could fit. The
3) Unlike other standards, the AWS standard lays AWS standard has, on the other hand, considered the
down the mechanical property requirements for each types which are in general industrial usage in the U.S. A.
class of electrode. Requirements for impact strength are and then brought out a system to fit them.
unusual. Either they are not required or they are quite AWS A5.1 has provided description of electrode clas-
stringent: 27 J at -29°C (except 27 J at-18°C for 7028). sifications in the Appendix. Following are the extracts:
No provision is made for impact requirement at -40°C
4) In the types of covering, acid rutile (AR), Description and Intended Use of Electrodes
oxidising (O) and other types (S) are not included.
5) No separate symbols are provided for deposition 5.1 E6010 [E4310] Classification
efficiency and for low-hydrogen property, but they are 5.1.1 E6010 [E4310]electrodes are characterized by
indicated through distinctive classification numbers. a deeply penetrating, forceful, spray type arc and readily
However, the exact deposition efficiency value is not removable, thin, friable slag which may not seem to
conveyed. completely cover the weld bead. Fillet welds usually
6) Under current conditions, OCV values are not in- have a relatively flat weld face and have a rather coarse,
dicated for AC. unevenly spaced ripple. The coverings are high in
cellulose, usually exceeding 30 percent by weight. The penetration in fillet welded joints. The coverings are
other materials generally used in the covering include high in titania, usually exceeding 35 percent by weight,
t i t a n i u m dioxide, metallic d e o x i d i z e r s such as and usually are referred to as the "titania" or "rutile"
ferromanganese, various types of m a g n e s i u m or type. The coverings generally also contain small
aluminum silicates, and liquid sodium silicate as a amounts of cellulose and ferromanganese, and various
binder. Because of their covering composition, these siliceous materials such as feldspar and clay with
electrodes are generally described as the high-cellulose sodium silicate as a binder. Also, small amounts of
sodium type. certain calcium compounds may be used to produce
5.1.2 These electrodes are recommended for all satisfactory arc characteristics on deen (electrode
w e l d i n g positions, particularly on multiple pass negative).
applications in the vertical and overhead welding 5.3.2. Fillet welds tend to have a convex weld face
positions and where welds of good soundness are with s m o o t h even ripples in horizontal w e l d i n g
required. They frequently are selected for joining pipe position, and widely spaced rougher ripples in the
and generally are capable of welding in the vertical vertical welding position which become smoother and
position with either uphill or downhill progression. more uniform as the size of the weld is increased.
5.1.3 The majority of applications for these Ordinarily, a larger size fillet must be made in the vertical
electrodes is in joining carbon steel. However, they have and overhead welding positions using E6012[E4312]
been used to advantage on galvanized steel and on some electrodes compared to welds with E6010[E4310] and
low alloy steels. Typical a p p l i c a t i o n s i n c l u d e E6011[E4311] electrodes of the same diameter.
shipbuilding, buildings, bridges, storage tanks, piping, 5.3.3 The E6012[E4312] electrodes are all-position
and pressure vessel fittings. Since the applications are electrodes and usually are suitable for welding in the
so widespread, a discussion of each is impractical. Sizes vertical welding position with either the upward or
larger than 3/16 in. (5.0 mm) generally have limited use downward progression. However, more often the larger
in other than flat or horizontal-fillet welding positions. sizes are used in the flat and horizontal welding
5.1.4 These electrodes have been designed for use positions rather than in the vertical and overhead
with deep (electrode positive). The maximum amperage welding positions. The larger sizes are often used for
that can generally be used with the larger sizes of these single pass, high-speed, high current fillet welds in the
electrodes is limited in comparison to that for other horizontal welding position. Their ease of handling,
classifications due to the high spatter loss that occurs good fillet weld face, and ability to bridge wide root
with high amperage. openings under conditions of poor fit, and to withstand
5.2 E6011 [E4311] Classification high amperages make them very well suited to this type
5.2.1 E6011 [E4311] electrodes are designed to be of work. The electrode size used for vertical and
used with ac current and to duplicate the usability overhead position welding is frequently one size smaller
characteristics a n d mechanical p r o p e r t i e s of the t h a n w o u l d be u s e d w i t h an E6010[E4310] or
E6010[E4310] classification. Although also usable with E6011[E4311] electrode.
deep (electrode positive), a decrease in joint penetration 5.3.4 Weld metal from these electrodes is generally
will be noted when compared to the E6010 [E4310] lower in ductility and may be higher in yield strength
electrodes. Arc action, slag, and fillet weld appearance (1 to 2 ksi [0.7 MPa to 1.4 MPa] than weld metal from
are similar to those of the E6010 [E4310] electrodes. the same size of either the E6010 [E4310] or E6011 [E4311]
5.2.2 The coverings are also high in cellulose and electrodes.
are described as the high-cellulose potassium type. In 5.4 E6013[E4313] Classification
addition to the other ingredients normally found in 5.4.1 E6013[E4313] electrodes, although very simi-
E6010 [E4310] coverings, small quantities of calcium and lar to the E6012[E4312] electrodes, have distinct differ-
potassium compounds usually are present. ences. Their flux covering makes slag removal easier
5.2.3 Sizes larger than 3/16 in [5.0 mm] generally and gives a smoother arc transfer than E6012[E4312]
have limited use in other than flat or horizontal - fillet electrodes. This is particularly the case for the small
welding positions. diameters 1/16 in, 5/64 in, and 3/32 in [1.6 mm, 2.0
5.3 E6012 [E4312]Classification mm, and 2.5 mm]. This permits satisfactory operation
5.3.1 E6012 [E4312] electrodes are characterized by with lower open-circuit ac voltage. E6013 [E4313] elec-
low penerating arc and dense slag,which completely trodes were designed specifically for light sheet metal
covers the bead. This may result in incomplete root work. However, the larger diameters are used on many
Table 3.19: Electrode Classification

AWS classification
A5.1 A5.1M Type of covering Welding position3 Type of currentb
E6010 E4310 High cellulose sodium F, V, OH, H-fiUet deep

E6011 E4311 High cellulose potassium F, V, OH, H-fiUet ac or deep
E6012 E4312 High titania sodium F, V, OH, H-fiUet ac or dcen
E6013 E4313 High titania potassium F, V, OH, H-fiUet ac, deep, or dcen
E6018c E4318c Low-hydrogen potassium, iron powder F, V, OH, H-fiUet ac or dcen
E6019 E4319 Iron oxide titania potassium F, V, OH, H-fiUet ac, deep, or dcen
E6020 E4320 High iron oxide H-fiUet ac or dcen
F ac, deep, or dcen
E6022d E4322d High iron oxide F, H-fiUet acor dcen
E7027 E4327 High iron oxide, iron powder H-fiUet ac or dcen
F ac, deep, or dcen
E7014 E4914 Iron powder, titania F, V, OH, H-fiUet ac, deep, or dcen
E7015 E4915 Low hydrogen sodium F, V, OH, H-fiUet deep
E7016c E4916c Low hydrogen potassium F, V, OH, H-fiUet ac or deep
E7018c E4918c Low-hydrogen potassium, iron powder F, V, OH, H-fiUet ac or deep
E7018M E4918M Low-hydrogen iron powder F, V, OH, H-fiUet deep
E7024c E4924c Iron power, titania H-fiUet, F ac, deep, or dcen
E7027 E4927 High iron oxide, iron powder H-fiUet ac or dcen
F ac, deep, or dcen
E7028c E4928c Low -hydrogen potassium, iron powder H-fiUet, F ac or deep
E7048 E4948 Low-hydrogen potassium, iron powder F, OH, H-fiUet, V-down ac or deep
Notes:
a. The abbreviations, F, H-fillet, V, V-down, and OH indicate the welding positions as follows: F = Flat, H-fillets = Horizontal fillet, V = Vertical,
Progression upwards (for electrodes 3/16 in (5.0 mm) and under, except 5/32 in (4.0 mm) and under for classifications E6018 (E4318), E7014 (E4914),
E7015 (E4915), E7016 (E4916), E7018 (E4918), E7018M (E4948M) E7048 (E4948). V-down = Vertical, progression downwards (for electrodes 3/16 in
(5.0 mm) and under, except 5/32 in (4.0 mm) and under for classifications E6018 (E4318); E7014 (E4914), E7015 (E4915), E7016 (E4916), E7018
(E4918), E7018M (E4918M), E7048 (E4948), OH = Overhead (for electrodes 3/16 in (5.0 mm) and under, except 5/32 in (4.0 mm) and under for
classifications E6018 (E4318), E7014 (E4914), E7015 (E4915), E7016 (E4916), E7018 (E4918), E7018M (E4918M), E7048 (E4948).
b. The term "deep" refers to direct current electrode positive (dc, reverse polarity). The term "dcen" refers to direct current electrode negative (dc,
straight polarity).
c. Electrodes with supplemental elongation, notch toughness, absorbed moisture, and diffusible hydrogen requirements may be further identified as
shown in Tables 2, 3,10, and 11 of AWS A5.1/5.1M-2004
d. Electrodes of the E6022 (E4322) classification are intended for single-pass welds only.
of the same applications as E6012[E4312] electrodes and bead appearance. The arc action tends to be quieter and
provide low penetrating arc. The smaller diameters pro- the bead surface smoother with a finer ripple. The us-
vide a less penetrating arc than is obtained with ability characteristics of E6013[E4313] electrodes vary
E6012[E4312] electrodes. This may result in incomplete slightly from brand to brand. Some are recommended
penetration in fillet welded joints. for sheet metal applications where their ability to weld
5.4.2 Coverings of E6013[E4313] electrodes contain satisfactorily in the vertical welding position with down-
rutile, cellulose, ferromanganese, potassium silicate as ward progression is an advantage.
a binder, and other siliceous materials. The potassium Others, with a more fluid slag, are used for horizontal
compounds permit the electrodes to operate with ac at fillet welds and other general purpose welding. These
low amperages and low open-circuit voltages. electrodes produce a flat fillet weld face rather than the
5.4.3 E6013[E4313] electrodes are similar to the convex weld face characteristic of E6012[E4312]
E6012[E4312] electrodes in usability characteristics and electrodes. They are also suitable for making groove
Table: 3.20 Chemical Composition Requirements for Weld Metal

Weight Percent b Combined limited for
AWS classification
Mn + Ni +Cr
UNS d
A5.1 A5.1M Number C Mn Si P S Ni Cr Mo V +Mo +V
E6010 E4310 W06010 \

E6011 E4311 W06011
E6012 E4312 W06012
E6013 E4313 W06013 |\ 0.20 1.20 1.00 N.S. N.S. 0.30 0.20 0.30 0.08 N.S.
E6019 E4319 W06019
E6020 E4320 W06020
E6027 E4327 W06027 i
E6018 E4318 W06018 0.03 0.60 0.40 0.025 0.015 0.30 0.20 0.30 0.08 N.S.
E7015 E4915 W07015 0.15 1.25 0.90 0.035 0.035 0.30 0.20 0.30 0.08 1.50
E7016 E4916 W07016 0.15 1.60 0.75 0.035 0.035 0.30 0.20 0.30 0.08 1.75
E7018 E4918 W07018 0.15 1.60 0.75 0.035 0.035 0.30 0.20 0.30 0.08 1.75
E7014 E4914 W07014 0.15 1.25 0.90 0.035 0.035 0.30 0.20 0.30 0.08 1.50
E7024 E4924 W07024 0.15 1.25 0.90 0.035 0.035 0.30 0.20 0.30 0.08 1.50
E7027 E4927 W07027 0.15 1.60 0.75 0.035 0.035 0.30 0.20 0.30 0.08 1.75
E7028 E4928 W070281

0.15 1.60 0.90 0.035 0.035 0.30 0.20 0.30 0.08 1.75
E7018M E4918M W07018 0.12 °' 4 ° 0.80 0.030 0.020 0.25 0.15 0.35 0.05 N.S.
to
L60
NOTES:
a. SAE/ASTM Unified Numbering System for Metals and Alloys.
b. Single values are maximum. N.S. means Not Specified.
welds because of their concave weld face and easily 5.5.2 The iron powder also permits the use of higher
removable slag. In addition, the weld metal is definitely a m p e r a g e s t h a n are u s e d for E6012[E4312] a n d
freer of slag and oxide inclusions than E6012[E4312] E6013[E4313] electrodes. The amount and character of
weld metal and exhibits better soundness. Welds with the slag permit E7014[E4914] electrodes to be used in
the smaller diameter E6013[E4313] electrodes often meet all positions.
the G r a d e 1 r a d i o g r a p h i c r e q u i r e m e n t s of this 5.5.3 The E7014 [E4914] electrodes are suitable for
specification. welding carbon and low alloy steels. Typical weld beads
5.4.4 E6013[E4313] electrodes usually cannot are smooth with fine ripples. Joint penetration is
withstand the high amperages that can be used with approximately the same as that obtained with E6012
E6012[E4312] electrodes in the flat and horizontal [E4312] electrodes (see A 7.3.1), which is advantageous
welding positions. Amperages in the vertical and when welding over a wide root opening due to poor fit.
overhead positions, however, are similar to those used The face of fillet welds tends to be flat to slightly convex.
with E6012[E4312] electrodes. The slag is easy to remove. In many cases, it removes
5.5 E7014 [E4914] Classification itself.
5.5.1 E7014 [E4914] electrode coverings are similar
to those of E6012 [E4312] and E6013 [E4313] electrodes, 5.6 Low Hydrogen Electrodes
but with the addition of iron powder for obtaining 5.6.1 Electrodes of the low-hydrogen classifications
higher deposition efficiency. The covering thickness and E6018 [E4318] , E7015 [E4915], E7016 [E4916], E7018
the amount of iron powder in E7014 [E4914] are less [E4918], E7018M [E4918M], E7028 [E4928]and E7048
than in E7024 [E4924] electrodes (see A7.10). [E4948]) are made with inorganic coverings that contain
minimal moisture The covering moisture test such as easy to remove. The weld face is convex, although a
specified in AWS A4.4M, Standard Procedure for fillet weld face may be flat.
Determination of Moisture Content of Welding Fluxes and 5.6.5.4 E7015 [E4915] electrodes u p to and
Welding Electrode Flux Coverings, converts hydrogen- including the 5/32 in. (4.0 mm) size are used in all
bearing compounds in any form in the covering into welding positions. Larger electrodes are used for groove
water vapor that is collected and measured. The test welds in the flat welding position and fillet welds in
thus assesses the potential hydrogen available from an the horizontal and flat welding positions.
electrode covering. All low hydrogen electrodes, in the 5.6.5.5 Amperages for E7015 [E4915] electrodes are
as-manufactured condition or after conditioning, are higher than those used with E6010 [E4310] electrodes
expected to meet a maximum covering moisture limit of the same diameter. The shortest possible arc length
of 0.6 percent or less. should be maintained for best results with E7015 [E4915]
5.6.2 The relative potential of an electrode to electrodes. This reduces the risk of porosity. The
contribute to diffusible hydrogen in the weld metal can necessity for preheating is reduced; therefore, better
be assessed more directly, but less conveniently, by the welding conditions are provided.
diffusible hydrogen test. The results of this test, using 5.6.6 E7016 [E4916] Classification
electrodes in the as-manufactured condition or after 5.6.6.1 E7016 [E4916] electrodes have all the
conditioning, permit the addition of an optional characteristics of E7015 [E4915] electrodes, plus the
supplemental diffusible hydrogen designator to the ability to operate on ac. The core wire and coverings
classification designation. are very similar to those of E7015 [E4915], except for
5.6.3 In order to maintain low hydrogen electrodes the use of a potassium silicate binder or other potassium
with minimal m o i s t u r e in their coverings, these salts in the coverings to facilitate their use with ac. Most
electrodes should be stored a n d h a n d l e d w i t h of the preceding discussion on E7015 [E4915] electrodes
considerable care. Electrodes which have been exposed applies equally well to the E7016 [E4916] electrodes.
to humidity may absorb considerable moisture and their 5.6.6.2 Electrodes designated as E7016-1 [E4916-
low hydrogen character may be lost. Then conditioning 1] have the same usability and weld metal composition
can restore their low hydrogen character. as E7016 [E4916] electrodes except that the managenese
5.6.4 Low hydrogen electrode coverings can be content is set at the high end of the range. They are
designed to resist moisture absorption for a considerable i n t e n d e d for w e l d s r e q u i r i n g a lower transition
time in a humid environment. The absorbed moisture t e m p e r t a t u r e t h a n is n o r m a l l y available from
test (see Section 17) assesses this characteristic by E7016[E4916] electrodes.
determining the covering moisture after nine hours 5.6.7 E6018[E4318] and E7018[E4918]
exposure to 80° F (27°C), 80 percent relative humidity Classifications.
air. If, after this exposure, the covering moisture does 5.6.7.1 E7018[E4918] electrode coverings are
not exceed 0.4 percent, then the optional supplemental similar to E7015[E4915] coverings, except for the
designator, "R", may be a d d e d to the electrode addition of a relatively high percentage of iron powder.
classification designation. The coverings on these electrodes are slightly thicker
5.6.5 E7015 [E4915] Classification than those of the E7016[E4916] electrodes.
5.6.5.1 E7015 [E4915] electrodes are low hydrogen 5.6.7.2 E7018[E4918] low hydrogen electrodes can
electrodes to be used with deep (electrode positive). The be used with either ac or deep. They are designed for
slag is chemically basic. the same applications as the E7016[E4916] electrodes.
5.6.5.2 E7015 [E4915] electrodes are commonly As is common with all low hydrogen electrodes, a short
used for making small welds on thick base metal, since arc length should be maintained at all times.
the welds are less susceptible to cracking. They are also 5.6.7.3 In addition to their use on carbon steel, the
used for welding high sulfur and enamelling steels. E7018[E4918] electrodes are also used for joints
Welds made with E7015 [E4915] electrodes on high involving high-strength, high carbon, or low alloy steels.
sulfur steels may produce a very tight slag and a very The fillet welds made in the horizontal and flat welding
rough or irregular bead appearance in comparison to positions have a slightly convex weld face, with a
welds with the same electrodes in steels of normal sulfur smooth and finely rippled surface. The electrodes are
content. characterized by a smooth, quiet arc, very low spatter,
5.6.5.3 The arc of E7015 [E4915] electrodes is and medium arc penetration. E7018[E4918] electrodes
moderately penetrating. The slag is heavy, friable, and can be used at high travel speeds.
5.6.7.4 Electrodes designated as E7018-1[E4918-1] welding position and groove welds in the flat welding
have the same usability and weld metal composition as position only, whereas E7018 [E4918] electrodes are
E7018[E4918] electrodes, except that the manganese suitable for all positions.
content is set at the high end of the range. They are 5.6.9.2 The E7028[E4928] electrode coverings are
i n t e n d e d for w e l d s r e q u i r i n g a lower transition much thicker. They make up approximately 50% of the
t e m p e r a t u r e t h a n is n o r m a l l y available from weight of the electrodes. The iron content of E7028
E7018[E4918] electrodes. [E4928] electrodes is higher (approximately 50 percent
5.6.7.5 E6018[E4318] electrodes posssess operating of the weight of the coverings). Consequently, on fillet
and mechanical property characteristics similar to welds in the horizontal position and groove welds in
E7018[E4918] except at a lower strength level. The the flat welding position, E7028 [E4928] electrodes give
electrode coating and low hydrogen characteristics are a higher deposition rate t h a n the E7018 [E4918]
also similar. This electrode is desirable where matching electrodes for a given size of electrode.
or undermatching weld deposit is required. Electrodes 5.6.10 E7048[E4948] Classification. Electrodes of the
that meet this classification may also be suitable for E7048[E4948] classification have the same usability,
buffer layer application in cladding operations. composition, and design characteristics as E7018[E4918]
5.6.8 E7018M[E4918M] Classification electrodes, except that E7048[E4948] electrodes are
5.6.8.1 E7018M[E4918M] electrodes are similar to specifically designed for exceptionally good vertical
E7018-1H4R[E4918-1H4R] electrodes, except that the welding with downward progression(see Table 3.19).
testing for mechanical properties and for classification 5.7 E6019[E4319] Classification
is done on a groove weld that has a 60 degree included 5.7.1 E6019[E4319] electrodes, although very similar
angle and, for electrodes up to 5/32 in [4.0mm], welded to E6013 and E6020[E4313 and E4320] electrodes in their
in the vertical position with upward progression. The coverings, have distinct differences. E6019[E4319]
impact test results are evaluated using all five test values electrodes, with a rather fluid slag system, provide
and higher values are required at -20° F (-30°C). The deeper arc penetration and produce weld metal that
maximum allowable moisture-in-coating values in the meets a 22 percent minimum elongation requirement,
"as-received" or reconditioned state are more restrictive meets the Grade 1 radiographic standards, and has an
than that r e q u i r e d for E7018R[E4918R]. This average impact strength of 20 ft-lbf (27 J) when tested
classification closely corresponds to MIL-7018-M in the at 0°F (-20°C).
United States military s t a n d a r d MIL-E-22200/10 5.7.2 E6019[E4319] electrodes are suitable for
specification, with the exception that the absorbed multipass welding of up to 1 in (25 mm) thick steel. They
moisture limits on the electrode covering and the are designed for use with ac, dcen, or deep. While 3/16
diffusible hydrogen limits on the weld metal are not as in (5.0 mm) and smaller diameter electrodes can be used
restrictive as those in MIL-E-22200/10. for all welding positions (except vertical welding
5.6.8.2 E7018M [E4918M] is intended to be used position with downward progression), the use of larger
with deep type current in order to produce the optimum diameter electrodes should be limited to the flat or
mechanical properties. However if the manufacturer horizontal fillet welding position. When welding in the
desires, the electrode m a y also be classified as vertical welding position with upward progression,
E7018[E4918] provided all the requirements of E7018 weaving should be limited to minimize undercut.
[E4918] are met.
5.8 E6020[E4320] Classification
5.6.8.3 In addition to their use on carbon steel, the
5.8.1 E6020[E4320] electrodes have a high iron oxide
E7018M[E4918M] electrodes are used for joining carbon
steel to high strength low alloy steels and higher carbon covering. They are characterized by a spray type arc
steels. Fillet welds made in the horizontal and flat produce a smooth and flat, or slightly concave weld face
welding positions have a slightly convex weld face, with and have an easily removable slag.
a smooth and finely rippled surface. The electrodes are 5.8.2 A low viscosity slag limits their usability to
characterized by a smooth, quiet arc, very low spatter, horizontal fillets and flat welding positions. With arc
and medium arc penetration. penetration ranging from medium to deep (depending
5.6.9 E7028[E4928] Classification upon welding current), E6020[E4320] electrodes are best
5.6.9.1 E7028[E4928] electrodes are very much like suited for thicker base metal.
the E7018[E4918] electrodes. However, E7028[E4928] 5.9 E6022 [E4322] Classification. Electrodes of the
electrodes are suitable for fillet welds in the horizontal E6022 [E4322] classification are recommended for single
pass, high-speed, high current welding of groove welds considerable portion of the electrical energy passing
in the flat welding position, lap joints in the horizontal through the electrode is used to melt the covering and
welding position, and fillet welds on sheet metal. the iron powder it contains. These electrodes are well
The weld face tends to be more convex and less suited for thicker base metal.
uniform, especially since the welding speeds are higher. 5.12 E7027 [E4927] Classification. E7027 [E4927]
5.10 E7024 [E4934] Classification electrodes h a v e the same usability and design
5.10.1 E7024 [E4924] electrode coverings contain characteristics as E6027 [E4327] electrodes, except they
large amounts of iron powder in combination with are intended for use in situations requiring slightly
ingredients similar to those used in E6012 and E6013 higher tensile and yield strengths than are obtained with
[E4312 and E4313] electrodes. The coverings on E7024 E6027 [E4327] electrodes. They must also meet chemical
[E4924] electrodes are very thick and usually amount composition requirements (see Table 3.20). In other
to about 50% of the electrode, resulting in higher respects, all previous discussions for E6027 [E4327]
deposition efficiency. electrodes also apply to E7027 [E4927] electrodes.
5.10.2 The E7024 [E4924] electrodes are well suited
for making fillet welds in the flat or horizontal position. Testing of Electrodes
The weld face is slightly convex to flat, with a very All electrode standards describe in great detail the
smooth surface and a very fine ripple. These electrodes procedures for executing all-weld tensile and impact
are characterized by a smooth, quiet arc, very low tests. Some of t h e m also describe m e t h o d s for
spatter, and low arc penetration. They can be used with determining weld deposition efficiency and hydrogen
high travel speeds. Electrodes of these classifications can in the weld deposit. The tensile strength, yield strength
be operated on ac, deep, or dcen. and elongation values obtained in the tensile test, and
5.10.3 Electrodes designated as E7024-1 [E4924-1] the values obtained in the other two tests provide the
have the same general usability characteristics as symbols for the coding of an electrode. All-weld metal
E7024[E4924] electrodes. They are intended for use in means weld deposit which is not diluted by the base
situations requiring greater ductility and a lower metal. In all-weld tensile and impact tests, the test
transition temperature than normally is available from specimens are so prepared that the area which is
E7024 [E4924] electrodes. subjected to test is pure, undiluted weld metal. The
5.11 E6027 [E4327] Classification electrode standards also prescribe supplementary tests
5.11.1 E6027 [E4327] electrode coverings contain which are not related to the code symbols, but are meant
large amounts of iron powder in combination with to evaluate the performance of an electrode and its
ingredients similar to those found in E6020[E4320] suitability for welding certain grades of steel. These tests
electrodes. The coverings on E6027[E4327] electrodes in various combinations are used for the quality control
are also very thick and usually amount to about 50 of p r o d u c t i o n batches a n d their acceptance by
percent of the weight of the electrode. consumers as indicated in the standards. The various
5.11.2 The E6027[E4327] electrodes are designed for tests included {^) in each standard are indicated in Table
fillet or groove welds in the flat welding position with 3.21.
ac, deep, or dcen and will produce a flat or slightly Procedures for all-weld tensile and impact tests laid
concave weld face on fillet welds in the horizontal down in the five standards are described and compared
position with either ac or dcen. in Chapter 11, Testing and Inspection. Transverse tensile,
5.11.3 E6027[E4327] electrodes have a spray-type impact and bend tests are also described in the same
arc. They will operate at high travel speeds. Arc chapter. The remaining tests are described in the
penetration is m e d i u m . Spatter loss is very low. following paragraphs.
E6027[E4327] electrodes produce a heavy slag which is
honeycombed on the underside. The slag is friable and Deposition Efficiency
easily removed. The method for determining this parameter is
5.11.4 Welds p r o d u c e d w i t h E6027 [E4327] described in ISO 2401, "Determination of the efficiency,
electrodes have a flat to slightly concave weld face with metal recovery and deposition coefficient", which is
a smooth, fine, even ripple, and good wetting along the reproduced in BS:639-1976 and IS:815-1974 with minor
sides of the joint. The weld metal may be slightly inferior alterations. A test plate of carbon steel (0.25% maximum
in radiographic soundness to that from E6020[E4320] carbon), 300 mm x 75 mm x 12 mm thick, is cleaned
electrodes. High amperages can be used, since a and weighed accurately. Welds are deposited on the
Table 3.21: Standard tests for electrodes

Type of test ISO 2560 BS639 DIN 1913 IS:814a AWS A5.1
All-weld tensile and :impact ii ii u ii ii
Transverse bend X ii ii ii ii
Transverse tensile X X ii u ii
Deposition efficiency xb ii X ii X
Diffusible hydrogen xc u xe ii xd
Chemical composition of weld metal X X X X ii
Weld soundness test (radiography) X X X X ii
Fillet weld X X X X ii
Deep penetration X X X ii X
" IS : 814 covers classification, coding specification and testing.

b
ISO 2401 describes this test.
c
ISO 3690 describes the method.
d
Quoting moisture test described in AWS is a requirement of the code for all low hydrogen electrodes. Additionally, E 7018M class of electrodes have
to meet the diffusible hydrogen test requirement of 4 ml max.1100 gms of weld metal. For other low hydrogen electrodes, the test is required only whm
diffusible hydrogen designator is added to the AWS code.
e
DIN 8572 describes the method.
plate with five electrodes, until a stub length of 3690, "Determination of hydrogen in weld metal aris-
approximately 50 mm remains in each case. All slag and ing from the use of covered electrodes for welding mild
spatter is removed from the plate after each electrode and low-alloy steels", and reproduced in the Appendi-
has been used. The interpass temperature is not allowed ces of BS:639-1976 and IS814 -1991 with minor alter-
to exceed 100°C. The plate with welds deposited on it is ations.
weighed after it has cooled to the ambient temperature. The principle of the method is as follows: A single
The mass of the core wires of the five electrodes is weld bead from an electrode under test is deposited on
obtained by weighing five core wires of the same a test plate and rapidly quenched. The specimen so
diameter and length as used in the electrodes. Covering produced is maintained at room temperature for 72
is also completely removed from the five stub ends and hours to release its content of diffusible hydrogen, which
they are weighed together accurately. is collected over mercury and measured volumetrically.
The nominal deposition efficiency (DE) is calculated If the test has to be delayed, the test plate is stored at -
from the formula: 78°C or lower until required for preparation and
analysis.
m2-mlx 100
DE% Following materials are required:
ra„ mA a) Test piece assembly prepared from semi-killed
where steel having 0.20% maximum carbon and 0.05% maxi-
mx - mass of plate before welding (±lg) mum sulphur. These must be degassed at 650°C for one
m2 = combined mass of plate and the weld metal hour.
deposited on it (±lg) b) Electrode of 4 mm size in the case of normal type
and of 3.15 mm size in the case of iron powder type
m, = mass of the five whole core wires (± lg)
m4 = mass of the five stub ends (±lg) having deposition efficiency higher than 130%. It must
be dried at 250°C for two hours.
Determination of Hydrogen in Weld Deposit c) A copper jig as shown in Fig. 3.16 for aligning
and clamping the test piece.
The total hydrogen present in the weld joint after
d) Hydrogen collecting apparatus. A typical gas
welding is the sum of two distinct fractions: the diffus-
burette for the measurement of cold extracted gas is
ible hydrogen and the residual hydrogen. The former
fraction escapes from solid metal at or near room tem- shown in Fig. 3.17. It fulfils the following requirements:
perature and diffuses to the surface. The latter fraction • Mercury is used as the confining liquid.
does not diffuse but remains within the structure of the • It is possible to maintain the sample under
solid metal. The method of determining diffusible hy- vacuum for a brief period to remove last traces
drogen in a weld deposit has been detailed in Draft ISO of foreign gases.
M12B0LT
SECTION A-A
-1 MM
COPPER FOIL
QUICK RELEASE CLAMP

50 MM
Fig. 3.16: Test piece assembly for hydrogen determination
9 MM SPHERICAL JOINT
HIGH VACUUM HOLLOW KEY
SINGLE OBLIQUE STOPCOCK 4 MM CORE
ATMOSPHERE
VACUUM
0.10 ML BURETTE GRADUATED TO
0.05 ML. ABOUT 30 MM LENGTH \ | | HIGH VACUUM HOLLOW KEY
PER ML, OR 10-11 MMO.D. DOUBLE OBLIQUE
STOPCOCK 4MM BORE
34/35 JOINT
WELD SPECIMEN
RUBBER BAND
120 MM 140 MM
STRAIGHT
SECTION
PERMANENT MERCURY
MAGNET BOROSILICATE
GLASS
R 80 MM
Fig. 3.17: Diffusible hydrogen collecting apparatus

• The volume of collected gas can be measured to M 2 is the mass of the sample after removal from appa-
an accuracy of 0.05. ratus (g)
M1 is the mass of the sample before deposition of the
Procedure weld (g).
The test piece assembly consists of run-on and run-
off pieces and a central sample section of 30 m m length, While the above-described method uses mercury for
which may be divided into 7.5 m m x 4 sections as the collection of diffusible hydrogen, the Japanese
shown in Fig 3.16 or 15 m m x 2 sections or 15 mm x 1 Industrial Standard JIS-Z-3113 uses glycerine and the
and 7.5 m m x 2 sections. A bead of 100 m m overall German standard DIH 8572 uses paraffin. The Japanese
length is deposited along the centreline of the assembly. method of measuring hydrogen under glycerine as
The p a r a m e t e r s for w e l d i n g are d e t a i l e d in the applicable to submerged-arc welds is described in
standard. Chapter 4. AWS also are averse to the use of mercury,
Three seconds after the arc is extinguished, the jig is as they consider it a potential health hazard, and are
released and the test piece assembly quenched and working on a standard based on glycerine. The French
cleaned in a prescribed manner. The sample pieces are standard (NFA-A-81-305) uses mercury as a collecting
broken apart while still cold and stored at solid carbon liquid but the quantity of hydrogen is linked to the total
dioxide temperature. All these operations must be fused metal (i.e. weld deposit plus fused base metal).
completed within the prescribed limited seconds. The In this method, therefore, one must determine the
samples may be stored in this manner for a period of weight of fused metal. To do this, the fused metal cross-
three days before analysis. section area together with the deposited metal is
When analysis is to be carried out, the quenched measured on the ends of the sample and the weight of
sample is washed in alcohol, then in ether and then dried the deposited metal multiplied by the average ratio:
in a blast of dry nitrogen to avoid condensation of
atmospheric humidity. These operations also must be Wt. of fused Wt. of weld fused metal area
completed within the prescribed time limit in seconds. _ x
The blanket of dry nitrogen is maintained while the metal area deposit deposited metal area
sample is transferred to the outer limb of the burette
When the other m e t h o d s are used, it becomes
(Fig. 3.17) and held in position clear of the mercury
necessary to calibrate them against the ISO procedure.
surface by a magnet. The outer limb of the burette is
The IIW (International Institute of Welding) has
then evacuated to make a vacuum. The sample is then
established the following correlation between the
transferred through the mercury air-lock to its final
glycerine and mercury methods:
position in the measuring limb of the burette. Time limit
in seconds is also set for these steps.
H - K , x 0.79-1.73
The sample is maintained under reduced pressure g1 H
g
at 25 ± 5°C (298 ± 5°K) for a period of 72 hours when
Gas Chromatography Method
'he final volume is measured. The barometric pressure
Gas chromatography is an instrumental analytical
is recorded. The sample is removed from the apparatus,
method, developed in Japan, by which a gas mixture is
thoroughly brushed so as to remove any oxide skin, and separated through separation columns containing an
weighed to the nearest 10 mg. The volume Vh (ml) at absorbent. A thermal conductivity detector is also used
STP of diffusible hydrogen per 100 g of deposited weld to measure the volume of separated gas by means of a
metal is calculated from the following formula: potentiometer. The method can be used within a wide
range of gas volume from 10~2 to 10 ml.
= Vg (B-H) 273 100 ml per 100 g of depo- The diffusible hydrogen measurement m e t h o d
h
760 298 (M2-M1) sited weld metal makes use of gas chromatography principle. First, the
where test piece is surrounded by argon gas in the collector
V is the volume of gas in the burette after 72 hours (ml) vessel, then the hydrogen which is released is introduced
B is the barometric pressure (mm of Hg) into a thermal conductivity detector by carrier gas.
H is the head of mercury at which V is measured This method may be used over a wide range of
(mm of Hg) diffusible hydrogen content from 10"1 to 102 ml/100 g.
The IIW has recognised that the gas chromatography process. The radiographic soundness requirements for
method is as reliable as the mercury method for various classes of electrodes are as shown:
determining hydrogen in welds, and the results
obtained are nearly identical. With gas chromatography, AWS classification Radiographic standard
it is possible to measure very accurately even lower E6020
hydrogen contents than 2 ml/100 g, which is not E7015
possible with the mercury or the glycerine method. This E7016 Grade I
method is therefore ideally suited for welds deposited E7018
by the gas-shielded processes, such as MIG/C0 2 , in E7048
which the hydrogen levels can be extremely low. The
gas chromatography instrument is marketed by a E6010
Japanese firm, and it is very expensive. E6011
E6013
The residual hydrogen is determined on the same
E7014 Grade II
specimen after completing the measurement of
E7024
diffusible hydrogen.
E6027
The specimen is carefully cleaned with solvent to E7027
remove the last traces of the collecting fluid and E7028
transferred to the apparatus, which determines the
residual hydrogen by hot extraction in either vacuum E6012 Not required
or inert carrier gas. E6022
An extraction temperature of approximately 700° C
is recommended, while the extraction time depends on Grades I and II are based on the sizes and numbers
the dimensions of the specimen. The specimen can also of porosity and/or inclusions in 150 mm length of weld.
be fused, but in this case a somewhat higher level of They are illustrated and detailed in the AWS standard.
residual hydrogen may be recovered. Residual
hydrogen contents are also expressed in terms of either Fillet Weld Test
deposited metal or fused metal. This test, according to AWS A5.1, "is included as a
means of demonstrating the usability of an electrode.
This test is concerned with the appearance of the weld,
Chemical Composition Test i.e. weld surface contour and smoothness, undercut,
AWS A5.1 describes the procedure of obtaining overlap, size, and resistance to cracking. It also provides
samples of undiluted weld metal for chemical analysis. an excellent and inexpensive method of determining the
A weld pad, 40 x 40 x 13 mm in the case of 4 mm size adequacy of root penetration (one of the important
and 50 x 50 x 13 mm in the case of 6.3 mm size electrode usability considerations for an electrode)."
is built up on a mild steel plate. It is deposited in layers In this test, a fillet weld is deposited in a tee joint
in the flat position. The width of each pass in each layer between two plates 75 mm wide and 300 or 460 mm
is kept within one-and-half to two-and-half times the long, depending on the electrode. The welding is done
diameter of the core wire. in the horizontal, vertical upward or overhead position
After depositing each layer, the pad is immersed in as required. The whole length of the electrode is
water for 30 seconds and then dried before welding is consumed without interruption, leaving a stub end 50
resumed. The surface of each layer is thoroughly mm long. Examined visually, the test specimen shall be
cleaned. Samples are collected by drilling or any other free of cracks, overlap, trapped slag, and surface
mechanical means, their minimum distances from porosity, and shall be substantially free of undercut. An
surface of base plate being 6 mm. infrequent, short undercut, up to 0.8 mm depth is
allowed. A macro-section, 25 mm wide, is removed at a
Weld Soundness Test point approximately 25 mm ahead of the crater end. Its
This test involves radiographic examination of the surface is polished, etched and weld size, leg lengths,
all-weld test assembly, after removal of the backing strip and convexity (if the weld is convex) are determined to
and weld ripples or surface irregularities on both the the nearest 0.5 mm by actual measurement as shown in
face and the root of the weld by any suitable mechanical Fig. 3.18.
Typical dimensional requirements are: throughout the entire length of the weld, provided that
at no point this lack of fusion exceeds 25% of the smaller
leg of the fillet."
Fillet weld Max. Max. difference
size convexity between
Deep Penetration Test
mm mm fillet leg lengths, mm
According to IS:814, deep penetration electrodes
4.0 1.2 1.2 c a r r y i n g the suffix, P m u s t meet the following
6.4 1.6 2.4 requirements:
8.0 2.0 3.2 a) The electrode must provide complete penetration
when deposited in two passes, one from each side, in a
The r e m a i n i n g t w o joint sections are b r o k e n close square butt joint between two plates of thickness
longitudinally through the fillet weld. The fractured equal to twice the core diameter of the electrode.
surfaces are visually examined and they should show b) The electrode must provide a minimum penetra-
no cracks. Lack of root fusion is assessed in the following tion of 4 mm beyond the root when it is deposited in a
manner: "Incomplete fusion at the root of the weld shall close square tee joint between two plates, each of thick-
not be greater than 20% of the total length of weld. There ness equal to twice the core diameter of the electrode,
shall be no continuous area of incomplete fusion greater in the flat and horizontal-vertical positions.
than 25 m m in length as m e a s u r e d along the c) From the butt joint welded as in (a), transverse
longitudinal weld axis, except for electrodes of the bend, tensile and impact specimens must be taken out
E6012, E6013, and E7014 classifications. Fillet welds and subjected to standard tests. The results are assessed
made with electrodes of the E6012, E6013, or E7014 in the same way as in the case of normal penetration
classifications m a y exhibit incomplete fusion electrodes.
As stated before, the deep penetration electrode has
been omitted in the 1976 edition of BS 639, since this
type is hardly used now and is no longer recognised in
ACTUAL THROAT
ISO documents. IS 814 is presently under revision, and
CONVEXITY it is expected that deep penetration electrodes will have
LEG OF no place in the new revision.
FILLET—i
THEORETICAL
Quality Control
THROAT
Specifications such as BS 639, IS 814 and AWS A5.1
specify quality control, testing and other requirements
that have to be satisfied by production batches of
SIZE OF
WELlP
electrodes. IS 814, for example, lays down requirements
in respect of the following:
SIZE OF
a) Tolerance on core wire diameter and length
WELD LEG OF FILLET b) Tolerance for flux covering
CONVEX FILLET c) Initial tests comprising:
i) All-weld tensile test
ii) All-weld impact test
THEORETICAL iii) Transverse tensile test
LEG OF T "
FILLET THROAT iv) Transverse bend test
v) Transverse impact test
vi) Test for determining the quantity of diffusible
SIZE OFl hydrogen for hydrogen controlled electrodes
WELC""^
only
SIZE OF.
d) Periodic check tests comprising all-weld tensile
WELD LEG OF
test and all-weld impact test
"FILLET"7
CONCAVE FILLET e) Production control
f) Packing and storage
Fig. 3.18: Dimensions of fillet welds in fillet weld test g) Test certificate
h) Marking on electrode or bundle or package. This This is the most comprehensive standard on the
may include the ISI certification mark, whose signifi- subject among all national standards. The classification
cance is explained as follows: system is the same as for mild steel electrodes, i.e., the
"The use of the ISI certification mark is governed by prefix E designates the electrode; the first two digits (or
the provisions of the Bureau of Indian Standards three digits of a five-digit number) designate the
(certification marks) Act and the rules and regulations minimum tensile strength of the weld metal in 1,000
made thereunder. The ISI mark on products covered by psi.
an Indian Standard conveys the assurance that they have The third digit (or fourth digit of a five-digit number)
been produced to comply with the requirements of that indicates the welding positions. The last two digits,
standard under a well-defined system of inspection, taken together, indicate the current conditions as well
testing and quality control, which is devised and as the coating type exactly as in the mild steel electrode
supervised by ISI and operated by the producer. ISI classification. In addition, a letter suffix such as Al, Bl,
marked products are also continuously checked by ISI Cl, etc., designates the chemistry of the deposited metal.
for conformity to that standard as a further safeguard. Each classification has to meet specific chemical analysis
Details of conditions under which a licence for the use as well as certain mechanical properties.
of the ISI certification m a r k m a y be g r a n t e d to The standard divides the electrodes into nine major
manufacturers or processors, may be obtained from the groups as shown in Table 3.22 on the next page.
Bureau of Indian Standards (BIS). Mechanical Properties are given in Tables 3.23 and
3.24
LOW-ALLOY STEEL ELECTRODES
These electrodes are designed for the welding of low- 2) IS: 1395-1982. "Low and medium alloy steel covered
alloy steels commonly used in structurals, pipings, electrodes for manual metal arc welding"
pressure vessels, machinery, and in the form of castings. The standard covers five groups of electrodes
They give weld deposits of steel containing small consisting of (a) carbon-molybdenum, (b) chrome-
amounts of alloying elements such as Mo, Cr, Ni and V, molybdenum, (c) nickel, (d) manganese-molybdenum,
which impart certain special properties to the weld and (e) all other alloys. The classification starts with the
metal to match those of the base metal as closely as prefix E, followed by two digits which indicate the
possible. Weld metal chemistry usually differs markedly minimum tensile strength of the weld metal as follows:
from the composition of the base metal, though in some
cases they come close to each other. Digits 41 49 55 63 68 76 83
Low-alloy steel electrodes cover a wide tensile range TS,min. MPa 410 490 550 630 680 760 830
from 480 N / m m 2 minimum to 830 N / m m 2 maximum. Next comes a letter symbol indicating the type of
They are mostly designed with basic low-hydrogen type flux covering as follows:
covering, with or without the addition of iron powder,
though electrodes in the lower tensile range may also Letter Covering type Letter Covering type
be made with cellulosic, rutile or acidic type coating. C Cellulosic O Oxidising
The standard coverings employed for electrodes of vari- R Rutile (medium coated) B Basic
ous tensile levels are indicated in the following AWS
specification. Then comes any of the letters A, B, C, D, G and M
The electrodes are normally designed with the followed by a digit a n d / o r a digit and a letter L to
standard low-carbon rimming steel core wire and indicate the chemical composition group and sub-group.
alloyed flux covering. This gives the producer the The letter L indicates carbon content less than 0.05%.
flexibility to manufacture a wide range of alloy-steel The lettering system is similar to the one used in AWS
electrodes, according to the market demand from just A5.5, except for the following variations:
one grade of core wire. Alloyed core wires are being a) IS includes B6, B7 and B8, which stand for 5 Cr/
used by some producers, especially by companies VL MO, 7 Cr/Vi Mo and 9 C r / 1 Mo respectively. AWS
engaged in the p r o d u c t i o n of steels as well as standard has included these three types under stainless
electrodes. steel electrodes in AWS A5.4, and coded them as E502,
E7Cr and E505 respectively.
1) AWS A5.5-1996, "Specifications for low-alloy steel b) IS does not include NM and D3.
electrodes for shielded metal arc welding" c) In IS, G of the AWS standard is indicated as Gl.
Table 3.22: Chemical composition requirements for undiluted weld metal, (Wt Per cent ■*)
8
AWS UNSa Mn Si Ni Cr Mo Additional Elements
classification numberd Type Amt.
i
Carbon-molybdenum-steel-electrodes s
3
E7010-A1 W17010 0.12 0.60 0.40 0.03 0.03 - - 0.40-0.65
%
E7011-A1 W17011 0.12 0.60 0.40 0.03 0.03 - - 0.40-0.65 6
E7015-A1 W17015 0.12 0.60 0.60 0.03 0.03 - - 0.40-0.65
E7016-A1 W17016 0.12 0.60 0.60 0.03 0.03 - - 0.40-0.65 i
E7018-A1 W17018 0.12 0.60 0.80 0.03 0.03 - - 0.40-0.65
E7020-A1 W17020 0.12 0.60 0.40 0.03 0.03 - - 0.40-0.65
i
X
E7027-A1 W17027 0.12 1.00 0.40 0.03 0.03 - - 0.40-0.65
5
E8016-B1 W51016 0.05-0.12 0.90 0.60 0.03 0.03 - 0.40-0.65 0.40-0.65 i
E8018-B1 W51018 0.05-0.12 0.90 0.80 0.03 0.03 - 0.40-0.65 0.40-0.65
E8016-B2 W52016 0.05-0.12 0.90 0.60 0.03 0.03 _ 0.40-0.65 0.40-0.65

E8018-B2 W52018 0.05-0.12 0.90 0.80 0.03 0.03 - 0.40-0.65 0.40-0.65
E7015-B2L W52115 0.05 0.90 1.00 0.03 0.03 _ 1.00-1.50 0.40-0.65

E7016-B2L W52116 0.05 0.90 0.60 0.03 0.03 - 1.00-1.50 0.40-0.65
E7018-B2L W52118 0.05 0.90 0.80 0.03 0.03 - 1.00-1.50 0.40-0.65
E7015-B3 W53015 0.05-0.12 0.90 1.00 0.03 0.03 _ 2.00-2.50 0.90-1.20

E7016-B3 W53016 0.05-0.12 0.90 0.60 0.03 0.03 - 2.00-2.50 0.90-1.20
E7018-B3 W53018 0.05-0.12 0.90 0.80 0.03 0.03 - 2.00-2.50 0.90-1.20
E8015-B3L W53115 0.05 0.90 1.00 0.03 0.03 _ 2.00-2.50 0.90-1.20

E8018-B3L W53116 0.05 0.90 0.80 0.03 0.03 - 2.00-2.50 0.90-1.20
E8015-B4L W53415 0.05 0.90 1.00 0.03 0.03 - 1.75-2.25 0.40-0.65
E8016-B5 W51116 0.05 0.40-0.70 0.30-0.60 0.03 0.03 _ 0.40-0.60 1.00-1.25 0.05
E8015-B6e W50215 0.05-0.10 1.0 0.90 0.03 0.03 0.40 4.0-6.0 0.45-0.65
E8016-B6e W50216 0.05-0.10 1.0 0.90 0.03 0.03 0.40 4.0-6.0 0.45-0.65
E8015-B61/ W50205 0.05 1.0 0.90 0.03 0.03 0.40 4.0-6.0 0.45-0.65
E8016-B61/ W50206 0.05 1.0 0.90 0.03 0.03 0.40 4.0-6.0 0.45-0.65
E8018-B6Le W50208 0.05 1.0 0.90 0.03 0.03 0.40 4.0-6.0 0.45-0.65
E8015-B7e W50315 0.05-0.10 1.0 0.90 0.03 0.03 0.40 4.0-6.0 0.45-0.65
E8016-B7e W50316 0.05-0.10 1.0 0.90 0.03 0.03 0.40 4.0-6.0 0.45-0.65
E8018-B7e W50318 0.05-0.10 1.0 0.90 0.03 0.03 0.40 4.0-6.0 0.45-0.65
MANUAL METAL-ARC WELDING
in in in
o o o
odd
> > >
\£ }£ £
en en en
odd
in in in
rH rH rH
odd
R RRRR£RRRR~° S
8888^8888SSSS
csicNHrsicsifoHcncofoddd
o o o o o o o o o o o o o
<5Gd<5e5<5<5e5<5<5<5e5^
enenenenenenenenenenenenen
O O O O O O O O O O O O O
d d d d d d d d d d d d d
in in o
<NJ f s j ^ J
rH T-H i-J
LnmminLninininminooo
^ ^ ^ ^ ^ ^ ^ ^ ^ ^ o o d
(NrînintncscNininincNCNQO
r H r H O O O r - l r H O O O r H r H O
d d d d d d d d d d d d d
voooinvooovoooinvoooôooo
O O r H r H r H Q Q r H r H i - H O O C k
£££££££££££££
H H H H H M N N N ( N r t ( O r t
iooiftvoiixiAiooiwoo
u w u w w u w u w
Table 3.22: (Contd.) o
AWS UNSa C Mn Si Ni Cr Mo Additional Elements o

classification number type o
m
a
z
Nickel-Steel Electrodes >
3D
O
E8016-C4 W21916 0.10 1.25 0.60 0.03 0.03 1.10-2.00
m
E8018-C4 W21918 0.10 1.25 0.80 0.03 0.03 1.10-2.00 i-
D
E9015-C5L W25018 0.05 0.40-1.00 0.50 0.03 0.03 6.00-7.25 Z
o
H
Nickel-Molybdenum Steel Electrodes m
o
E8018-NM1 W21118 0.10 0.80-1.25 0.60 0.02 0.02 0.80-1.10 0.10 0.40-0.65 V 0.02 O
Cu 0.10 o
-<
Al 0.05
Manganese-Molybdenum Steel Electrodes
E8018-D1 W18118 0.12 1.00-1.75 0.80 0.03 0.03 0.90 0.25-0.45

E9015-D1 W19015 0.12 1.00-1.75 0.60 0.03 0.03 0.90 0.25-0.45
E9018-D1 W19018 0.12 1.00-1.75 0.80 0.03 0.03 0.90 0.25-0.45
E10015-D2 W10015 0.15 1.65-2.00 - 0.60 0.03 0.03 0.90 0.25-0.45
E10016-D2 W10016 0.15 1.65-2.00 0.60 0.03 0.03 0.90 0.25-0.45
E10018-D2 W10018 0.15 1.65-2.00 0.80 0.03 0.03 0.90 0.25-0.45
E8016-D3 W18016 0.12 1.00-1.80 0.60 0.03 0.03 0.90 0.40-0.65
E8018-D3 W18018 0.12 1.00-1.80 0.80 0.03 0.03 0.90 0.40-0.65
E8018-D3 W18018 0.12 1.00-1.80 0.80 0.03 0.03. 0.90 0.40-0.65
General Low-Alloy Steel Electrodes
EXX10-G 1.00 min 0.80 min 0.50 min 0.30 min 0.20 min V O.lOmin
Cu 0.20min
EXX11-G l.OOmin 0.80min 0.50min 0.30min 0.20min V O.lOmin
Cu 0.20min
Cu 0.20min
Cu 0.20min
Cu 0.20min
Cu 0.20min
E7020-G l.OOmin 0.80min 0.50min 0.30min 0.20min V O.lOmin
Cu 0.20min
E7027-G - - l.OOmin 0.80min - - 0.50min 0.30min 0.20min V O.lOmin
Cu 0.20min
Military Similar Electrodes
E9018Mi W21218 0.10 0.60-1.25 0.80 0.030 0.030 1.40-1.80 0.15 0.35 V 0.05
E10018Mi W21318 0.10 0.75-1.70 0.60 0.030 0.030 1.40-2.10 0.35 0.25-0.50 V 0.05
E11018Mi W21418 0.10 1.30-1.80 0.80 0.030 0.030 1.25-2.50 0.40 0.25-0.50 V 0.05
E12018M1 W22218 0.10 1.30-2.25 0.60 0.030 0.Q30 1.75-2.50 0.30-1.50 0.30-0.55 V 0.05
E12018MU W23218 0.10 0.80-1.60 0.65 0.015 0.012 3.00-3.80 0.65 0.20-0.30 V 0.05
Pipeline Electrodes
E7010-P1 W17110 0.20 1.20 0.60 0.03 0.03 1.00 0.30 0.50 V 0.10
E8010-P1 W18110 0.20 1.20 0.60 0.03 0.03 1.00 0.30 0.50 V 0.10
Weathering Steel Electrode
E7018-W1J W20018 0.12 0.40-0.70 0.40-0.70 0.025 0.025 0.20-0.40 0.15-0.30 V 0.08
Cu 0.30-0.60
E8018-W2J W20118 0.12 0.50-1.30 0.35-0.80 0.03 0.03 0.40-0.80 0.45-0.70 Cu 0.30-0.75
Notes:
(a) Single values are maximum, except where specified otherwise, (b) Weld metal shall be analysed for those elements for which specific values are shown. Other element
listed without specified values shall be reported, if intentionally added. The total of these latter unspecified elements and all other elements not intentionally added shall n
exceed 0.50%. (c) The suffixes Al, B3, C3 etc. designate the chemical composition of the electrode classification, (d) SAE/ASTM Unified Numbering System for Metals and
Alloys, (e) The E8015-B6 and E8015-B6L electrodes were formerly classified as E502-15 in AWS A5.4-81, Specification for Covered Corrosion-Resisting Chromium and
Chromium Nickel Steel Welding Electrodes. The E8016-B6 and E8016-B6L were formerly classified as E502-16 in A5.4-81. The E8018-B6 and E8018-B6L were not
formerly classified but were produced to the E502 composition ranges in A5.4-81 but with the EXX18 covering of this specification. Similarly, the E80XX-B7(L) classifi-
cations were formerly classified as E7Cr-XX in A5.4-81; and the E80XX-B8(L) classifications were formerly classified as E505-XX in A5.4-81. (f) The letter "XX" used in
the classification designation for EXX13-G in this table stand for various tensile-strength levels (80, 90,100,110 and 120 ksi) of weld metal, (g) The letters "XX" used in
the classification designations for all electrodes except EXX13-G in this table stand for the various tensile-strength levels (70, 80, 90,100,110 and 120 ksi) of electrodes, (h)
In order to meet the alloy requirements of the "G" group, the undiluted weld metal shall have the minimum of at least one of the elements listed in this table. Additional
chemical requirements may be agreed to between supplier and purchaser, (i) These classifications are intended to be similar to types of electrodes covered by MIL-E-2220
1 and MIL/E-22200/j. In AWS A5.5-81, E7018-W 1 was designated E7018-W, and E8018-W2 was designated E8018-W.
d) In IS, E9018-M, E10018-M, El 1018-M and E12018- The complete range of electrodes covered by the
M have been coded as E63 B-Ml, E68 B-M2, E76 B-M3 standard is indicated through the respective IS
and E83 B-M4 respectively. E12018-MI is excluded. classifications is given further in Table 3.25.
e) W is not included in IS. To explain the coding let us take the example
The next is a digit symbol indicating the positions in E55RB2L23Fe, which indicates the following character-
which the electrode can be used. istics:
Symbol Welding positions
1 All positions 3) BS 2493-1971, "Low-alloy steel electrodes for manual
2 All positions except vertical down metal-arc welding"
3 Flat butt, flat fillet, horizontal/vertical This standard deals with four groups of electrodes,
4 Flat butt, flat fillet namely, molybdenum, chromium-molybdenum, nickel
5 Similar to 3, and recommended for and manganese-molybdenum types.
vertical downward
Then comes the digit which indicates the current The electrode coding consists of two codes:
conditions as follows:
1) Composition code
mbol DC, polarity AC, min. • The symbol Mo for molybdenum steel electrodes
0 + • The symbol CrMo for chromium-molybdenum
1 + or- 50 steel electrodes with a prefix figure indicating the
2 - 50 chromium content
3 + 50 • The symbol Ni for nickel steel electrodes with a
4 + or- 70 prefix indicating the nickel content
5 - 70 • The symbol MnMo for manganese-molybdenum
6 + 70 steel electrodes
7 + or- 90
8 - 90
9 + 90 2) Covering code
• The letter B for basic type covering
Finally comes the suffix which indicates the presence • The letter R for rutile type covering
of iron powder and the metal recovery percentage: • The letter C for cellulosic type covering
For hydrogen-controlled electrodes the letter H and
Fe—Iron powder covering giving metal recovery of min.
for iron powder electrodes the letter J or K according to
110%
the definitions in BS1719, Part I, shall be added after
J—Iron powder covering giving a metal recovery of 110
the letter for the type of covering. The letter J indicates
to 130%
metal recovery (i.e. nominal electrode efficiency or
K—Iron powder covering giving a metal recovery of
deposition efficiency) of 110 to 130%. The letter K
130 to 150% indicates metal recovery of more than 130%. Since BS
L—Iron powder covering giving a metal recovery of 1719: Part I is obsolete, these suffixes as also the entire
over 150% coding system is due for revision.
E 55 R B2L 2 3 Fe
Iron powder electrode
Metal recovery 110% minimum
DC (+) or AC 50 V open-circuit voltage
All positions except vertical down
Weld metal composition 0.05 max. C/1.0-1.5 Cr/0.40-0.65 Mo
Rutile coated electrode
Minimum tensile strength 550 MPa
Electrode for manual metal-arc welding
MANUAL METAL-ARC WELDING 1 0 5
Table 3.23: Tension test requirements a,b

Tensile strength Yield strength, at 0.2% offset Elongation
AWS classification ksi MPa ksi MPa Percent Postweld condition
E7010-P1 70 480 60 415 22 AW

E7010-A1 70 480 57 390 22 PWHT
E7010-G 70 480 57 390 22 AW or PWHT
E7011-A1 70 480 57 390 22 PWHT
E7011-G 70 480 57 390 22 AW or PWHT
E7015-X 70 480 57 390 25 PWHT
E7015-B2L 75 520 57 390 19 PWHT
E7015-G 70 480 57 390 25 AW or PWHT
E7016-X 70 480 57 390 25 PWHT
E7016-B2L 75 520 57 390 19 PWHT
E7016-G 70 480 57 390 25 Aw or PWHT
E7018-X 70 480 57 390 25 PWHT
E7018-B2L 75 520 57 390 19 PWHT
E7018-C3L 70 480 57 390 25 AW
E7018-W1 70 480 60 415 25 AW
E7018-G 70 480 57 390 25 AW or PWHT
E7020-A1 70 480 57 390 25 PWHT
E7020-G 70 480 57 390 25 AW or PWHT
E7027-A1 70 480 57 390 25 PWHT
E7027-G 70 480 57 390 25 AW or PWHT
E8010-P1 80 550 67 460 19 AW
E8010-G 80 550 67 460 19 AW or PWHT
E8011-G 80 550 67 460 19 AW or PWHT
E8013-G 80 550 67 460 16 AW or PWHT
E8015-X 80 550 67 460 19 PWHT
E8015-B3L 80 550 67 460 17 PWHT
E8015-G 80 550 67 460 19 AW or PWHT
E8016-X 80 550 67 460 19 PWHT
E8016-C3 80 550 68 to 80 470 to550 24 AW
E8016-C4 80 550 67 460 19 AW
E8016-G 80 550 67 460 19 AW or PWHT
E8018-X 80 550 67 460 19 PWHT
E8018-B3L 80 550 67 460 17 PWHT
E8018-C3 80 550 68 to 80 470 to 550 24 AW
E8018-C4 80 550 67 460 19 AW
E8018-NM1 80 550 67 460 19 AW
E8018-W2 80 550 67 460 19 AW
E8018-G 80 550 67 460 19 AW or PWHT
E9010-G 90 620 77 530 17 AW or PWHT
E9011-G 90 620 77 530 17 AW or PWHT
E9013-G 90 620 77 530 14 AW or PWHT
E9015-X 90 620 77 530 17 PWHT
E9016-G 90 620 77 530 17 AW or PWHT
E9016-X 90 620 77 530 17 PWHT
E9016-G 90 620 77 530 17 AW or PWHT
E9018M 90 620 78 to 90 550 to 620 24 AW
E9018-X 90 620 77 530 17 PWHT
E9018-G 90 620 77 530 17 AW or PWHT
E10010-G 100 690 87 600 16 AW or PWHT
Table 3.23:(Contd.)
AWS classification ksi MPa ksi MPa Per cent Postweld condition
E100U-G 100 690 87 600 16 AW or PWHT
E10013-G 100 690 87 600 16 AW or PWHT
E10015-X 100 690 87 600 16 PWHT
E10015-G 100 690 87 600 16 AW or PWHT
E10018M 100 690 88 to 100 610 to 690 16 AW
E10018-X 100 690 87 600 16 PWHT
E10018-G 100 690 87 600 16 AW or PWHT
E11010-G 110 760 97 670 15 AW or PWHT
E11011-G 110 760 97 670 15 AW or PWHT
E11013-G 110 760 97 670 13 AW or PWHT
E11015-G 110 760 97 670 15 AW or PWHT
E11016-G 110 760 97 670 15 AW or PWHT
E11018-G 110 760 97 670 15 AW or PWHT
E11018M 110 760 98 to 110 680 to 760 20 AW
E12010-G 120 830 107 740 14 AW or PWHT
E12011-G 120 830 107 740 14 AW or PWHT
E12013-G 120 830 107 740 11 AW or PWHT
E12015-G 120 830 107 740 14 AW or PWHT
E12016-G 120 830 107 740 14 AW or PWHT
E12018-G 120 830 107 740 14 AW or PWHT
E12018M 120 830 108 to 120 745 to 830 18 AW
E12018M1 120 830 108 to 120 745 to 830 18 AW
Notes:
(a) See Table 5 for sizes to be tested.
(b) Single values are minimum, except as othenoise specified.
(c) The letter suffix "X" as used in this table represents the suffixes (A1,B1, B2, etc.) except for those classification which are tested in the as-welded condition.
(d) "AW" signifies as-welded with aging when it is specified in 11.2, "PWHT" signifies postzueld heat treated as specified in 8.4.2 and in Table 7, except
that the "G" designated classifications, marked as "AW or PWHT" in this table, may have weld metal tested with or without PWHT as agreed between the
supplier and purchaser. For 3/32 in. (2.4 mm) electrodes, the upper value for the yield strength may be 5 ksi (35 MPa) higher than the indicated value.
Table 3.24: Charpy V-Notch impact requirements

Limits for 3 out of 5 Specimens0
AWS _*assification Average, min.1 Single Value, min.b
E7018-W1 20ft.lbfatO°F 15ftlbfatO°F
E8018-W2 (27J at -18°C) (20J at -18°C)
E12018M1 50ft.lbfatO°F 40ft.lbfatO°F
(67J at -18°C) (54J at -18°C)
E7010-P1 20ft.lbfat20°F 15 ft. lbfat20°F
E8010-P1 (27J at -29°C) (20J at -29°C)
E8018-NM1
E8016-C3 20ft.lbfat40°F 15 ft. lbfat40°F
E8018-C2 (27J at -40°C) (20J at -40°C)
E8016-D3, E8018-D1
E8018-D3, E9015-D1
E9018-D1, E9018-D3 20ft.lbfat60°F a 15 ft. lbfat60°F
E10015-D2, E10016-D2 (27J at -51°C) (20J at -51°C)
E10018-D2
E7018-C3L
E8016-C4 E818-C4
Table 3.24 Contd.
Table 3.24: (Contd.)
Limits for 3 out of 5 Specimens0

AWS Classification Average, min.b Single Value, min.b
E9018-M,E10018M 20ft.lbf at 60 °F 15 ft. lbf at 60 °F

E11018M,E12018M (27J at -51°C) (20J at -51°C)
E8016-C1 20ft.lbf at 75 °Fa 15 ft. lbf at 75 °Fa
E8018-C1 (27J at -59°C) (20J at -59°C)
E7015-C1L
E7016-C1L 20ft.lbfatl00°F a 15 ft. lbf at 100 °Fa
E7018-C1L (27J at -73°C) (20J at -73°C)
E8016-C2
E8018-C2
E7015-C2L
E7018-C2L (27J at -10PC) (20Jat-101°C)
(27J at -115°C) (20J at -115°C)
EXXXX-A1
EXXXX-BX
EXXXX-BXL Not specified Not specified
EXXXX-C
Notes :
a. These classifications are tested in the postweld heat treated conditions. No thermal treatment shall be performed on the test specimens ofall other classifications.
b. Impact test values shall be recorded to "nearest whole unit" of energy absorbed in foot-pounds in accordance with the rounding-off method specified
in ASTME29
c. Both the highest and the lowest test values obtained shall be disregarded in computing the average value. Two of these three remaining values shall
equal or exceed the minimum average value listed; one of these three remaining values may be lower than minimum average value, but shall not be less than
the minimum single value listed. The average of the three remaining values shall not be less than the minimum average value listed.
Table 3.25: Range of low-alloy electrode classification in IS 1395

Group I Group II Group III Group IV Group V
C-Mo Cr-Mo Ni Mn-Mo All other alloys
E49C-A1 E55 R-Bl E55 B-Cl E63 B-Dl EX C - G l

E49 R - A l E55 B-Bl E55 B-C2 E68 B-D2 EXR-G1
E49 O - A l E55 R-B2 E55B-C3 ExO-Gl
E49 B-Al E55 R-B2L EXB-G1
E55 B-B2 E63 B-Ml
E55 B-B2L E68 B-M2
E63 R-B3 E76 B-M3
E63 B-B3L E83 B-M4
E63 B-B3
E63 R-B3L
E55 B-B4
E55 B-B4L
E55 B-B5
E55 B-B5L
E41 R-B6
E41 B-B6
E41 R-B7
E41 B-B7
E41 R-B8
E41 B-B8
Note: The letters X under Group V stand to show the strength levels of various electrode types.
The codings, equivalent AWS classifications, and interpass temperature as indicated in Table 3.30 for the
weld deposit compositions are specified as shown in respective electrode class has to be maintained. After
Table 3.26. completion of welding, the assembly has to be stress-
Mechanical properties on all-weld specimens must relieved for one hour at the temperatures indicated in
be in accordance with those specified in Table 3.27. the table. Radiography for the soundness test may be
The values given in Table 3.27 are specified for all- carried out before or after stress-relief, but tension and
weld test specimens that have been heat-treated in impact tests have to be carried out after stress-relief.
accordance with the conditions given in Table 3.28. The following classes are subjected to tension and
Values for welded joints in actual fabrications require impact tests in the as-welded condition:
separate consideration. E7010-G, E7018-W, E8010-G, E8016-C3, E8018-C3,
E8018-NM, E8018-W.E9010-G, E9018-M, E10018-M,
Testing and Quality Control E11018-M, E12018-M, E12018-M1
AWS A5.5 stipulates chemical analysis of weld metal, Radiographic requirements are as shown in Table
soundness test involving radiography, all-weld tension 3.31.
test, impact test, fillet weld test, and coating moisture test. Coating moisture test: This test has been described
A common test assembly is prepared for soundness fully in AWS A5.1 and AWS A5.5. The method consists
test, all-weld tension test and impact test. Electrodes of passing oxygen over the sample of flux coating
under test have to be used in the as-received condition, (approximately 4 g) in a nickel boat placed in a high-
except for low-hydrogen electrodes (classes EXX15-X, temperature ceramic type combustion tube, which is
EXX16-X and EXX18-X). Low-hydrogen electrodes must then heated to approximately 1,000° C. Liberated water
be dried at 260 to 427°C for two hours immediately prior is collected in a weighed absorption U-tube filled with
to testing. During welding of the assembly, preheat and anhydrous magnesium perchlorate and weight of water
Table 3.26: Coding of low alloy steel electrodes as per BS 2493

Composition Nearest Chemical composition
code and equipment
covering AWS Si Mn Ni Cr Mo S P
code classification max. max. max. max.
min. max. mm. max. mm. max. mm. max.
Molybdenum Steel Electrodes
MoB* E70XX-A1 0.10 0.50 0.75 1.20 0.40 0.70 0.035 0.030
MoC E70XX-A1 0.10 0.30 0.35 0.40 0.70 0.035 0.030
MoR E70XX-A1 0.10 0.30 0.35 0.40 0.70 0.035 0.030
Chromium-Molybdenum Steel Electrodes
lCrMo B* E80XX-B2 0.10 0.50 0.75 1.20 1.0 1.5 0.40 0.70 0.035 0.030
lCrMo R— 0.10 0.30 0.35 1.0 1.5 0.40 0.70 0.035 0.030
2CrMo B* E90XX-B3 0.10 0.50 0.75 1.20 2.0 2.5 0.90 1.20 0.035 0.030
2CrMo R — 0.10 0.30 0.35 2.0 2.5 0.90 1.20 0.035 0.030
5CrMo B E502 0.10 0.50 0.50 1.00 4.0 6.0 0.40 0.70 0.035 0.030
7CrMo B E7Cr 0.10 0.60 0.50 1.00 6.0 8.0 0.40 0.70 0.035 0.030
9CrMoB E505 0.10 0.60 0.50 1.00 8.0 10.0 0.90 1.20 0.035 0.030
Nickel Steel Electrodes
1MB E80XX-C3 0.10 0.80 1.20 0.80 1.10 0.35 0.030 0.030
2NiB E80XX-C1 0.10 0.80 1.20 2.00 2.75 0.030 0.030
3NiB E80XX-C2 0.10 0.80 1.20 2.80 3.50 0.030 0.030
Manganese-Molybdenum Steel Electrodes
MnMoB E90XX-D1 0.10 0.80 1.20 1.80 0.25 0.45 0.035 0.030
*The Mn : Si ratio shall be at least 2:1.
Table 3.27: Mechanical properties as per BS 2493

Composition Tensile 0.2% proof Elongation (min.) Average (min.) of three
code strength (min.) stress (min.) for gauge length impact values
L0 = 5.65OS0, i.e. 5d
N/mm2 N/mm2 % J
Mo 490 400 22
lCrMo 560 470 17
2CrMo 630 540 15
5CrMo 415 210 20
7CrMo 415 210 20
9CrMo 415 210 20
INi 560 470 17 30 at-^0±l°C
2Ni 560 470 17 30 at—60±1°C
3Ni 560 470 17 30 at—75±1°C
MnMo 630 540 15 30 at—50 ±1°C
Table 3.28: Heat treatment conditions as per BS 2493

Composition Minimum Maximum Post-weld tempering conditions
code pre-heating interpass
temperature temperature Temperature range Time at temperature
°C °C °C h
Mo 100 200 630 to 670 1
lCrMo 100 200 630 to 670
2CrMo 200 300 670 to 710
5CrMo 200 300 840 to 870*
7CrMo 200 300 840 to 870*
9CrMo 200 300 840 to 870*
INi 50 100 None None
2Ni 100 150 590 to 620 1
3Ni 100 150 590 to 620 1
MnMo 100 200 600 to 640
* Furnace cool at rate not exceeding 55°C per hour down to 575°C and air cool
Table 3.29: Base metal for weld test assemblies

AWS classification Base metals ASTM specification UNS number*
All except E(X)XXYYM(1) Carbon steel A29 Grade 1015 or equiv. G10150
All except E(X)XXYYM(1) Carbon steel A29 Grade 1020 or equiv. G10200
All except E(X)XXYYM(1) Carbon steel A29 Grade D or equiv. K02702
All except E(X)XXYYM(1) Carbon steel A29 Grade A or equiv. K01700
All except E(X)XXYYM(1) Carbon steel A29 Grade B or equiv. K02200
All except E(X)XXYYM(1) Carbon steel A29 Grade C or equiv. K02801
All Carbon steel A36 or equivalent. K02600
All Carbon steel A131 Grade B or equiv. K02102
*SAE/ASTM Unified Numbering System for Metals and Alloys.
determined by the increase in the weight of the U-tube. producer or supplier is expected to guarantee that the
The moisture content is expressed as a percentage of electrodes in all sizes conform to the specification. On
the weight of the coating. The laboratory apparatus used the important question of quality control of electrode
for this test is shown in Fig. 3.19. production, BS 2493 lays down the following:
The limits of coating moisture content permissible "The manufacturer's quality control system shall
in the case of low-hydrogen electrodes of various classes e n s u r e that all electrodes s u p p l i e d satisfy the
are given in Table 3.32. requirements of this standard. The system shall provide
BS:2493 specifies chemical analysis, all-weld tensile for the ready detection of discrepancies and for timely
and impact tests, longitudinal bend test and fillet weld and positive corrective action. The system shall provide
test. Tensile and impact test assembly as well as the records such that the identification number of each batch
longitudinal bend test piece have to be subjected to of electrodes shall p r o v i d e references to the
preheating, interpass t e m p e r a t u r e , and p o s t w e l d manufacturer's own documented records. From these,
tempering conditions as shown in Table 3.30. certificates may be available, showing the results of the
It must be noted that while the AWS standard refers various tests and inspections that were performed on
to the postweld heat treatment of low-alloy steel that batch of electrodes during manufacture.
w e l d m e n t s as stress-relief, the BS refers to it as "Tests and inspections shall be performed during
tempering, thus implying that the resulting changes are manufacture, so that continuous quality control of
not only mechanical but also metallurgical. production is maintained and any departure from the
IS:1395 includes chemical analysis, radiographic test accepted quality will be immediately detected. From
all-weld tensile and impact tests, moisture content test each batch of electrodes, a sufficient quantity of finished
and fillet weld test. Impact tests are required only for electrodes shall be taken to allow execution of the quality
E55 B-Cl, E55 B-C2, E55 B-<:3, E63 B-DI, E68 B-D2, control tests."
E63 BMI, E68 B-M2, E76 B-M3 and E83 B-M4 classes. A BS 2493 refers to initial tests comprising chemical
few classes are exempted from radiographic test. analysis, all-weld tensile and impact tests, longitudinal
All the three above-mentioned standards specify also bend tests and fillet weld tests, to which low-alloy steel
the tolerances on electrode size a n d length, on electrodes must be subjected. It further states that the
concentricity of coating, and on the exposed core at the producer must be able to guarantee that the electrodes
striking and grip ends. They also refer to requirements in current production are capable of satisfying all the
of packing, marking or labelling and identification. The test requirements.
"SCHWARTZ" TYPE
ANHYDROUS MAGNESIUM
PERCHLORATE
ABSORPTION TUBE
OXYGEN
FLOW METER
SPRAY TRAP W
If
^ C0NC. H 2 S0 4
DRYING TOWER
Fig. 3.19: Apparatus for determining moisture in flux coating

IMMt 3.30: Preheat, interpass, and postweld heat treatment temperatures

Preheat and interpass temperatures Postweld heat treatment temperatures
AWS classification op oC °F °C
E7010-A1
E70U-A1
E7015-A1
E7016-A1
E7018-A1
E7020-A1
E7027-A1
E8018-D1
200 to 225 93 to 107 1150 ± 25 620 ± 14
E9015-D1
E9018-D1
E10015-D2
E10016-D2
E10018-D2
E8016-D3
E8018-D3
E9018-D3 /
E8016-B1
E8018-B1 \
E8015-B2
E8016-B2
E8018-B2
E7015-B2L
E7016-B2L
325 to 375 163 to 191 1275 ± 25 690 ± 14
E7018-B2L >
E9015-B3
E9016-B3
E9018-B3
E8015-B3L
E8018-B3L
E8015-B4L
L
E8015-B6
E8016-B6
E8018-B6
E8015-B6L
E8016-B6L
E8018-B6L
350 to 450 177 to 232 1375 ± 25 740 ± 14
E8015-B7
E8016-B7
E8018-B7
E8015-B7L
E8016-B7L
E8018-B7L
Table 3.30 Contd.

AWS classification op oC op oQ
E8015-B8
E8016-B8
E8018-B8
400 to 500 205 to 260 1375 ± 25 740 ± 14
E8015-B8L
E8016-B8L
E8018-B8L
E9015-B9
E9016-B9 450 to 550 232 to 288 1375 ± 25 740 ± 14
E9018-B9
E8016-C1
\
E8018-C1
E7015-C1L
E7016-C1L
E7018-C1L
200 to 225 93 to 107 1125 ± 25 605 ± 14
E8016-C2
E8018-C2
E7015-C2L
E7016-C2L
E7018-C2L
/
E9015-C5L 200 to 225 93 to 107 1125 ± 25 605 ± 14
E8010-G
E8011-G \
E8013-G
E9010-G
E9011-G
E9013-G
E10010-G
E10011-G 325 to 375 163 to 191 See Note a
E10013-G
E11010-G
E11011-G
E11013-G
E12010-G
E12011-G
E12013-G /
Table 3.30 Contd.


AWS classification op oQ op oQ
E7010-G
E7011-G
E7015-G
E7016-G
E7018-G
E7020-G
E7027-G
E8015-G
E8016-G
E8018-G
E9015-G
E9016-G
E9018-G 200 to 225 93 to 107 See Note a
E10015-G
E10016-G
E10018-G
E11015-G
E11016-G
E11018-G
E12015-G
E12016-G
E12018-G /
E7010-P1
E7018-C3L
v
E7018-W1
E8016-C3
E8018-C3
E8016-C4
E8018-C4
E8018-NM1 200 to 250 93 to 121 Not Specifiedb
E8018-W2
E9018M
E10018M
E11018M
E12018M
E12018M1
E8010-P1 / 325 to 375 163 to 191 Not specifiedb
Notes:
a. The need and specific valuesforpostweld heat treatment ofweld test assemblies made with these "G" electrodes shall be as agreed between supplier
b. Postweld heat treatment is not required for those classifications listed as "as-welded" in Table 3.23
Table 3.31: Radiographic requirements as per AWS A5.5 IS 1395 stipulates the tests to be performed under
(a) initial testing, (b) periodic checking and (c)
AWS classification3 Radiographic standard bc production control. Radiographic test requirements,
EXX15-X, EXX16-X, Grade 1 impact test requirements, covering moisture content
EXX18-X, E7020-X requirements and preheat and stress-relief temperatures
EXX18M, E12018M1 indicated in the standard are similar to those given in
EXX10-X, EXX11-X, Grade 2 AWS A5.5. The methods of moisture determination
XX13-G, E7027-X given in the AWS and Indian standards are identical.
Notes:
a. The letters "XX" used in the classification designations in this table, STAINLESS STEEL ELECTRODES
stand for the various strength levels (70, 80, 90, 100,110 and 120) of elec-
trodes. The letter suffix "X" as used in this table stands for the suffixes Al, Stainless steel (S.S.) electrodes are normally used for
Bl, B2 etc. (see Table 2). the welding of corrosion-resisting and heat-resisting Cr
b. See Figure 6. Refer radiographic acceptance standards of AWS A5.5
c. The radiographic soundness obtainable under actual industrial con- and Cr-Ni steels in the form of plates, tubes and castings.
ditions employedfor the various electrode classification is discussed in A6.10.1 Some of them are used also for the welding of austenitic
in the Annexere of AWS A5.5 standards. manganese steel, stainless steel-clad plates and
Table 3.32: Moisture content limits in electrode coverings

AWS classification Electrode designation3 Limit of moisture content, % by wt.max
As-received or reconditionedb As exposedc
E7015-X E7015-X,E7015-X-HZ
E7016-X E7016-X,E7016-X-HZ 0.4 Not specified
E7018-X E7018-X,E7018-X-HZ
E7015-X E7015-X,R,E7015-X-HZR
E7016-X E7016-X,R,E7016-X-HZR 0.3 0.4
E7018-X E7018-X,R,E7018-X-HZR
E8015-X E8015-X,E8015-X-HZ
E8018-X E8018-X,E8018-X-HZ
E8015-X E7015-X,R,E7015-X-HZR
E8016-X E7016-X,R,E7016-X-HZR 0.2 0.4
E8018-X E7018-X,R,E7018-X-HZR
E9015-X E9015-X,E9015-X-HZ
E9018-X E9018-X,E9018-X-HZ
E9018M E9018M,E9018M-HZ
E9015-X E9015-X,R,E9015-X-HZR
E9016-X E9016-X,R,E9016-X-HZR 0.15 0.4
E9018-X E9018-X,R,E9018-X-HZR
E9018M E9018M-R,E9018M-HZ
E10015-X E10015-X,E10015-X-HZR
E10016-X E10016-X,E10016-X-HZR 0.15 Not specified
E10018-X E10018-X,E10018-X-HZR
E10018M E10018M-E10018M-HZ
E10015-X E10015-X,R,E10015-X-HZR
E10016-X E10016-X,R,E10016-X-HZR
E10018-X E10018-X,R,E10018-X-HZR 0.15 0.4
E10018M E10018M-R,E10018M-HZ
E10015-G E10015-G,E10015-X-HZR
E10016-G E10016-G-E10016-X-HZR
0.15 Not specified
E10018-G E10018-G,E10018-X-HZR
E10018M E10018M-R,E10018M-HZ
E10015-G E10015-G-R,E10015-X-HZR
E10016-G E10016-G-R,E10016-X-HZR
0.15 0.4
E10018-G E10018-G-R,E10018-X-HZR
E10018M E10018M-R,E10018M-HZ
E12015-G E12015-G,E12015-G-HZ
E12016-G E12016-G,E12016-G-HZ
0.15 Not specified
E12018-G E12018-G,E12018-G-HZ
E12018M E12018M,E12018M-HZ
E12015-G E12015-G-R,E12015-G-HZ
E12016-G E12016-G-R,E12016-G-HZ 0.15 0.4
E12018-G E12018-G-R,E12018-G-HZ
E12018M E12018M-R,E12018M-HZ
E12018M1 E12018M1,E12018M1-HZ 0.10 Not specified
E12018M1 E12018M1-R,E12018M1-HZR 0.10 0.4

Notes:
a. 'HZ' in the designation stands for optional Hydrogen designator, where 'Z' stands for 4, 8 or 16 ml ofH2 per 100 g of weld metal, when tested in
accordance with one of the methods given in ANSI/AWS A 4.3. 'R' designates that the electrodes has meet the requirements ofABS moisture test.
b. The electrodes to be tested without reconditioning unless manufacturer recommends otherwise. If electrodes are conditioned, the method used, time
and temperature to be recorded.
c. The electrode samples shall be exposed in a suitably calibrated and controlled environmental chamber for 9 hours minimum at 26:7°C (+2.8, -0) and
80% RM.
dissimilar metal combinations, and for applying This is a very exhaustive standard covering 43 classes
stainless steel sheet linings to carbon steel shells. with chromium in the weld deposit ranging from 4.0 to
Austenitic stainless steel types are often preferred for 32.0% and nickel from nil to 37%. As mentioned earlier,
the welding of critical structures of carbon steel or low-
5 CrMo, 7 CrMo and 9 CrMo types, which have been
alloy steels (as for example, armoured tanks), because placed by the IS and BS under low-alloy steel groups,
of the superior mechanical properties of their weld have been included in this standard.
deposits. Moreover, use of austenitic steel weld metal In this standard, the electrodes are classified on the
in welding of difficult-to-weld steels of large thicknesses
basis of the chemistry and mechanical property require-
under heavy restraint helps prevent hydrogen-induced ments of the weld metal (see Tables 3.33 and 3.34) and
cracking, even when preheating is avoided or used to a the type of current used for welding. The classification
limited degree. This is explained in detail under system consists of suffix letter E (to indicate an
Weldability of Steels, Chapter 10. electrode), followed by various com-binations of digits
and letters to indicate the chemical composition of weld
1) AWS A5.4-92. "Specification for corrosion-resistingmetal as shown in Table 3.33, which are further followed
chromium and chromium-nickel steel covered welding by -15, -16, -17, -25 and -26 to indicate the type of
electrodes" covering type of current and welding position.
w in w w w w w w w w w w w w w w w w w w w w w w w w w w w w w G. >
w w w co co c o c o c o c o c o c o c o c o c o c o c o
co sp $ g g o o oooooooovioso^ b S S S2. &
x Z< X * X X X X X
X X X X
X X x x x X
X
35 33 00 oo go c o c o c o c o c o c o c o c o c o c o c o c o c o
3 3 3 33 3 3 3 CO CO 3z 2c
00 ►—» ►—> ►—» O O
VI CO \© v q v j O N O N ON CO O O O vO v© — 00 o 00 00 00 CO CO
»-> c o 8 H-» H - i t—» I—A h-» t—k (—* h-» h-» t - i h J ts) K) N) •-» i-»
IS O Cn O CO O CO O O O v i en o Co O VI CO O CO' ~ O CO O o o o o o
00 O
o o o 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
O O O Q O Q Q O ^ ^ ^ C O O Q ^ ^ Q ^ Q O Q Q O Q Q Q Q n
CO CO O CO H-> CO VI oo cn N) N> cn
■ i i
00 Cn 00 o o o
o o O 4 * N) 8
oo en o
4* k)
N ) h J N ) N J h J N ) N h J N>
o> o o Cn Cn
CO 4^ *■ si oo oo si o o c n c n c n c n h J N J K J K)hJO0O0O0O0O0O0 V£> O
In b b b o b
9 99 9 9 o 9 9 9 9 9 9 9 9 9 V1
N ) N ) K ) N ) K ) N ) W K ) N K ) K ) N ) W
\& t—» N) •—i I—i O i O
" ■ ■—»■ i■-*^ O O M O O Q O O O O O U I U I
b b M Vsl M b o b o o o o cn
b b o b b b b b b b b b b b O
b o o
00 vO CO CO 00 N) KJ NJ K)
CO CO N) |vJ M M O O O O
8 9 o b o o b O ggggg9999999Y,Y' 1
1—»
1—> 9 9 9 9 9 9
9 O t—» CO CO O N> N ) N> N> l l l l l l — » ►-» h-» H-* I—k H-» Q N S ] l - »
i r i i—i
VI VJ K) N) 4^. £ t 4*. 4 ^b b o o o si
o o £& cn N> K) I
b o 4*
b 4*
b 4*
o £! o b cn cn
b b b b b b
N)
.^ W
__ W
_ . N)
^ . K^) K ) O N ) O O O K ) N J O O O K ) K » O O O O O O t - i
N) CO si s i s i b b b bb bb bb b b b si b si si si b b si si ^ b b ^si sjen si sjen
cn cn cn co co
k> P b b b b o o b b b b 3
Cn
p g- oo
SI 3n
2 n n 8^J (n 3 o 1
S5! 3
3 |. | 3 |
I5* 5' 5' I
o o O ►—* >— l o <p p CD p p H - » O O O O O O O O O O C O I - » C
cn cn cn b b cn C n c nCn
c ncn
c ncn
c cn c n b b b b c n c n c n c n ^ c n c n c n c n c n
n c ncn
i
i i N> K) K) a N)
cn cn cn cn cn cn c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n r
cn
o
o o p o p p p p p p p p p p p p p p p p p p p p p p p ^ ^ p cn
S£> \Q \Q
O O o o o 8£ o o o o o o o c n c n c n c n c n o o o o o o o o o o o o o
O O o o p p p p p p p p p p p p p p p p p p p p p p p p p
o o o 9
4^ 4 ^ 4 ^
r o 4^.
o
o o o o o cn
o o o o o o o o o o o o o o o o o o o o o o o o o
o o b o b b b b b b b b b b b b b b b b b b b b b b b b b b b b
K> CO •-* co c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o
cn
o o o
o
9
o
8 IS I
co co o o o o o o o o o o o o p p o p p p p p p p p p p
9s N VI V I V I b b V | V | V | V | V | V I V I V | V | V I V | V | V I V 1 V 1 V I V | V I V I V | V 1 V | V 1 V I V |
cn cn cn c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n
b b
ADO1ONH031 0NIO13M ObV NU3Q0W 9 U
E385-XX W88904 0.03 19.5-21.5 24.0-26.0 4.2-5.2 1.0-2.5 0.75 0.03 0.02 - 1.2-2.0
E410-XX W41010 0.12 11.0-13.5 0.7 0.75 1.0 0.90 0.04 0.03 - 0.75
E410NiMo-XX W41016 0.06 11.0-12.5 4.0-5.0 0.40-0.70 1.0 0.90 0.04 0.03 - 0.75
E430-XX W43010 0.10 15.0-18.0 0.6 0.75 1.0 0.90 0.04 0.03 - 0.75
E502-XX h W50210 0.10 4.0-6.0 0.4 0.45-0.65 1.0 0.04 0.04 0.03 - 0.75
E505-XX h W50410 0.10 8.0-10.5 0.4 0.85-1.20 1.0 0.90 0.04 0.03 - 0.75
E630-XX W37410 0.05 16.00-16.75 4.5-5.0 0.75 0.25-0.75 0.75 0.04 0.03 - 3.25-4.00
E16-8-2-XX W36810 0.10 14.5-16.5 7.5-9.5 1.0-2.0 0.5-2.5 0.60 0.03 0.03 - 0.75
E7Cr-XX h W50310 0.10 6.0-8.0 0.4 0.45-0.65 1.0 0.90 0.04 0.03 - 0.75
E2209-XX W39209 0.04 21.5-23.5 8.5-10.5 2.5-3.5 0.5-2.0 0.90 0.04 0.03 0.08-0.20 0.75
E2553-XX W39553 0.06 24.0-27.0 6.5-8.5 2.9-3.9 0.5-1.5 1.0 0.04 0.03 0.10-0.25 1.5-2.5
Notes:
a) Analysis shall be made for the elements for which specific values are shown in the table. If however, the presence of other elements is indicated in the course of routine analysis, further
analysis shall be made to determine that the total of these other elements, except iron, is not present in excess of 0.50 per cent.
b) Single values are maximum percentages.
c) Classification suffix-XX may be -15, -16, -17, -25 or -26, See Section A8 of the Appendix for an explanation.
d) SAF /ASTM Unified Number System for Metals and Alloys.
e) Vanadium shall be 0.15 to 0.30 per cent.
f) Titanium shall be 015 percent max.
g) Tungsten shall be from 1.25 to 1.75 per cent.
h) This grade also will appear in the next revision ofAWS A5.5, Specification for Low Alloy Steel Electrodes for Shielded Metal Arc Welding. It will be deleted from A5.4 at the first revision
of A5.4 following publication of the revised A5.5.
Table 3.34: Mechanical property requirements for all-weld metal as per AWS A5.4
AWS Tensile strength, min Elongation in 2 in. gauge Heat treatment
classification ksi MPa length, min. %
E209-xx 100 690 15 None
E219-xx 90 620 15 None
E240-xx 100 690 15 None
E307-XX 85 590 30 None
E308-XX 80 550 35 None
E308H-XX 80 550 35 None
E308L-XX 75 520 35 None
E308Mo-XX 80 550 35 None
E308MoL-XX 75 520 35 None
E309-XX 80 550 30 None
E309L-XX 75 520 30 None
E309Cb-XX 80 550 30 None
E309Mo-XX 80 550 30 None
E309MoL-XX 75 520 30 None
E310-XX 80 550 30 None
E310H-XX 90 620 10 None
E310Cb-XX 80 550 25 None
E310Mo-XX 80 550 30 None
E312-XX 95 660 22 None
E316-XX 75 520 30 None
E316H-XX 75 520 30 None
E316L-XX 70 490 30 None
E317-XX 80 550 30 None
E317L-XX 75 520 30 None
E318-XX 80 550 25 None
E320-XX 80 550 30 None
E320LR-XX 75 520 30 None
E330-XX 75 520 25 None
E330H-XX 90 620 10 None
E347-XX 75 520 30 None
E349-XX 100 690 25 None
E383-XX 75 520 30 None
E385-XX 75 520 30 None
E410-XX 75 450 20 a
E410NiMo-XX 110 760 15 c
E430-XX 65 450 20 d
E502-XX 60 420 20 b
E505-XX 60 420 20 b
E630-XX 135 930 7 e
E16-8-2-XX 80 550 35 None
E7Cr-XX 60 420 20 b
E2209-XX 100 690 20 None
E2553-XX 110 760 15 None
Notes:
(a) Heat to 1350 to 140& F (730 to 76&C), hold for one hour, furnace cool at a rate ofl00°F (60>C) per hour to 60O>F (315°C) and air cool to ambient.
(b) Heat to 1550 to 160&F (840 to 870°C) hold for two hours, furnace cool at a rate not exceeding 10WF (55°C) per hour to 1100°F (595°C) and air cool to
ambient, (c) Heat to 1100 to 1150>F (595 to 620»C), hold for one hour and air cool to ambient, (d) Heat to 1400 to 145&F (760 to 790PC), hold for two hours,
furnace cool at a rate not exceeding 10&F (55°C) per hour to 1100°F (595°C) and air cool to ambient. (e)Heat to 1875 to 1925°F (1025 to 1050°C), hold for
one hour, and air cool to ambient, and then precipitation harden at 1135 to U65°F (610 to 630°C), hold for four hours, and air cool to ambient.
Classification as to Usability that of the -15 designation and so that description also
Five basic usability classifications are provided in applies here. The electrode differs from the -15 type in
this specification. that the core wire may be of a substantially different
The type of covering applied to a core wire to make composition, such as mild steel, that may require a much
a shielded metal arc welding electrode determines the higher welding current. The additional alloys necessary
usability characteristics of the electrode. The following to obtain the required analysis are contained in the cov-
discussion of covering types is based upon terminology ering which will be of greater diameter than the corre-
commonly used by the industry: no attempt has been sponding -15 type. These electrodes are recommended
made to specifically deifne the composition of the for welding only in the flat and horizontal positions.
different covering types. Usability Designation - 26: This slag system is very
Usability Designation -15: The electrodes are usable similar in composition and operating characteristics to
with deep (electrode positive) only. While use with that of the - 16 designation and so that description also
alternating current is sometimes accomplished, they are applies here. The electrode differs from the -16 type in
not intended to qualify for use with this type of current. that the core wire may be of a substantially different
Electrode sizes 5/32 in. (4.0 mm) and smaller may be composition such as mild steel that may require a mach
used in all positions of welding. higher welding current. The additional alloys necessary
Usability Designation -16 : The covering for these to obtain the required analysis are contained in the
electrodes generally contains readily ionizing elements, covering which will be of much larger diameter than
such as potassium, in order to stabilize the arc for the corresponding - 16 type. These electrodes are
welding with ac Electrodes sizes 5/32 in. (4.0 mm) and recommended for welding only in the flat and horizontal
smaller may be used in all positions of welding. positions.
Usability Designation - 17 : The covering of these In stainless steel electrodes, the core-wire chemistry
electrodes is a modification of the -16 covering in that and weld-metal chemistry need not necessarily be
considerable silica replaces some of the titania of the - identical. For example, an E308 Mo (19/9/2.5 Mo)
16 covering. Since both the -16 and the - 17 electrode electrode or an E310 (25/20) electrode may have 304
coverings permit ac operation, both covering types were S.S. type (18/8) core wire and flux covering alloyed with
classified as - 16 in the past because there was no requisite proportions of chromium, and nickel, and with
classification alternative untill this revision of ANSI/ molybdenum if necessary. Hence, it would be incorrect
AWS A5.4. However, the operational differences to verify an electrode class by analysing its core-wire
between the two types have become significant enough chemistry: one must get at the undiluted all-weld metal
to warrant a separate classification. and determine its composition. •
On horizontal fillet welds, electrodes with a - 17 Three types of flux coverings are generally used as
covering tend to produce more of a spray arc and a finer shown in Table 3.35.
rippled weld-bead surface than do those with the -16 The lime-type works only on DC+, hence it carries
coverings. A slower freezing slag of the - 17 covering the AWS designation-15. Since the lime slag is viscous,
also permits improved handling characteristics when the bead is more convex, and the electrode can be used
employing drag technique. The bead shape on in the vertical down position also. The titania type works
horizontal fillets is typically flat to concave with - 17 on AC and DC and carries the designation-16. The slag
covered electrodes as compared to flat to slightly convex is fluid and hence the weld bead is flat or concave. The
with -16 covered electrodes. When making fillet welds bead flatness minimises final surface grinding, but then
in the vertical position with upward progression the the electrode is a little more difficult to operate in V and
slower freezing slag of the - 17 covered electrodes OH positions. The lime-titania type is a compromise
requires a slight weave technique to produce the proper between the two, and may carry -15 or -16 designation
bead shape. For this reason, the minimum leg-size fillet depending on the particular brand.
that can be properly made with a -17 covered electrode In Table 3.33, a few classes carry suffix L, which
is larger than that for a - 16 covered electrode. While indicates that the carbon content in the deposit is
these electrodes are designed for all - position operation restricted to 0.04% maximum, in order to obtain
electrode sizes 3/16 in. (4.8 mm) and larger are not resistance to intergranular corrosion caused by carbide
recommended for vertical or overhead welding. precipitation, which can occur when the carbon content
Usability Designation - 25: This slag system is very is higher and the welded joint is exposed to corrosive
similar in composition and operating characteristics to media in service. The same end can be achieved by a
Table 3.35: Chemical constituents of flux coverings

Covering type CaC0 3 CaF2 Ti02 NaSiO. KSi04 Alloy + other
4
ingredients
Lime 37 33 3 15 12
Lime-titania 30 15 25 5 10 15
Titania 25 20 35 15 5
slight reduction in the carbon level (from 0.15 maximum low welding currents used. Since the workpiece is made
to 0.12 maximum) and addition of columbium (also negative, less heat is generated on it and yet sufficient
called niobium) as a stabiliser. Some of the columbium- penetration is obtained by virtue of the low thermal
stabilised classes carry the suffix Cb. Titanium can also conductivity of stainless steels.
serve as a stabiliser and is sometimes used in stabilised Stainless steel electrodes are also manufactured with
S.S. wrought materials. But it cannot be used in the standard low-carbon unalloyed steel core wire, all
electrodes, because TI gets oxidised in its passage across the alloying elements being incorporated in the coating.
the arc. The phenomenon of intercrystalline corrosion Unlike normal stainless steel electrodes, which as
and the mechanism by which C and Cb affect it will be explained above have to be operated at low currents
explained in Chapter 10 while discussing the weldability due to the high electrical resistivity of the stainless steel
of stainless steels. It may be mentioned here that the core, these synthetic electrodes can be operated at high
low-carbon alloy is not so strong at elevated tempera- currents to obtain high deposition rates. However, the
ture, as the Cb-stabilised alloy. weld deposit of these electrodes can have inconsistent
Two classes carry suffix H, indicating high carbon chemistry on the microscopic and sometimes
content. These are meant for the welding of high alloy, macroscopic scale, due to small variations in the coating
heat and corrosion resistant castings of the same general thickness or damage to the coating. Also during
composition. welding, drops of ordinary steel given off as spatter can
Stainless steel electrodes get rapidly overheated adhere to the stainless base metal. Both these factors
during welding, hence the recommended welding can reduce the corrosion resistance of welded joint.
currents are about 20 to 30% lower for a given size than These types are therefore recommended only for joining
for ordinary mild steel electrodes. This is because of the stainless steel to ordinary steel and for fabricating
high electrical resistivity of stainless steels, which is stainless steel, which will be exposed to mild corrosion
nearly 5-6 mil room temperature for martensitic and conditions.Table 3.36 gives comparative data on
ferritic steels and 7-8 mQ for austenitic steels as electrode sizes and recommended welding currents.
compared to about 1-2 mil for mild steel. It is also for In Table 3.33 chromium-nickel types starting from
the same reason that stainless steel electrodes are shorter E307 down to E349 are austenitic stainless steel
in length than ordinary steel electrodes. electrodes. The rest are either ferritic or air-hardening
The practice of using stainless steel electrodes on DC types.
with electrode positive is also based on sound technical
reasoning. In DC welding, there is high liberation of heat Ferrite in Weld Deposits
at the anode, which helps to promote fusion of the Fully austenitic S.S. weld deposits are susceptible to
electrode, which would otherwise be slow due to the cracking or Assuring; electrode producer, therefore,
Table 3.36: Welding currents for different electrode types

E6013 type E308-16 E309Mo-16*
S.S. core wire
Size mm Amps Size mm Amps Size mm Amps
5x450 180-250 5x350 150-180 5x450 190-290
4x450 140-190 4x350 100-140 4x450 150-225
3.2 x 450 100-140 3.2 x 350 80-100 3.2 x 350 110-165
2.5x350 60-90 2.5 x 350 50-75 2.5 x 350 70-110
*This synthetic type gives 160% deposition efficienqj.
introduce small controlled quantities of delta ferrite in mm for 6.4 mm size electrode. Minimum distance of
the austenitic by suitably modifying the weld metal sample from the surface of base plate must be 6.4, 8.0
chemistry. Austenite formation is encouraged by nickel, and 9.5 mm respectively. Pads which are too hard for
carbon and manganese. On the other hand, chromium, machining may be annealed as indicated under Table
molybdenum, silicon and columbium are known to be 3.34.
ferrite formers. Nitrogen inclusion lowers the ferrite 2) All-weld tension test
content appreciably. When the proportions of these For all-weld assembly, one may use steel of matching
elements are known, the amount of delta ferrite in the type or any structural/pressure vessel quality steel. In
weld metal can be determined by referring to Schaeffler the latter case, all the fusion faces and backing bar must
diagram or DeLong diagram or WRC-1992 diagram, be buttered with two layers of weld deposit from the
which will be explained in detail in Chapter 10. electrode under test.
Generally, ferrite in austenite is of help when the Some of the electrode classes require preheat and
welds are restrained and the joints are large. Ferrite interpass temperature to be maintained during the
increases the weld strength level. In most cases it has welding of test assembly as indicated:
no significant effect on corrosion resistance, except for
316 and 316L, where it can be detrimental in some media. Interpass temperature
It is generally regarded as detrimental to toughness in °C
cryogenic service, a n d as u n d e s i r a b l e in h i g h AWS classification min. max.
temperature service, where it can transform into the E 200 series 16 150
brittle sigma phase. E 300 series 16 150
Percent ferrite, or ferrite number (FN) to be more E 400 series 150 260
specific, can be measured on a relative scale by means E 500 series 150 260
of various magnetic i n s t r u m e n t s , since ferrite is E 16-8-2 16 150
magnetic and austenite is non-magnetic. For accurate E7Cr 150 260
results it is necessary that the instrument is calibrated E630 16 150
with standards available from a few international
sources. Normally FN is controlled in various classes of The annealing heat treatment required for some
electrodes as shown in Table 3.37. classes is indicated under Table 3.34.
3) Fillet weld test
Test Requirements This is similar to the test described earlier under AWS
1) Chemical analysis A5.1. In the case of 3.2 and 4.0 mm sizes, it has to be
For chemical analysis of all classes with suffix L and carried out in the horizontal, vertical and overhead
E630 class, the base metal must be any steel of 0.03% positions. For 4.8 and 6.4 mm sizes, it has to be done in
maximum carbon. For the rest, it can be a steel with the horizontal position only.
0.25% maximum carbon. The multi-run pad size will 4) Ferrite test
be 25 x 25 x 13 mm for 2.4 mm electrode size and below, The weld pad is built up between two copper bars
40 x 40 x 16 mm for 3.2 to 4.8 mm size and 50 x 50 x 19 laid parallel on the base plate of matching type by
Table 3.37: Ferrite content limits for S.S. weld metals

Class FN Class FN
E308 4 to 10 E312 20 min.
E308L 4 to 10 E16-B-2 5 max.
E308 Mo 4 to 10 E316 2 max.
E308 MoL 4 to 10 E316L 2 max.
E309 4 to 10 E317 2 to 5
E309L 4 to 10 E317L 2 to 5
E309 Mo 4 to 10 E318 4 to 10
E310 Nil E320 Nil
E310H Nil E330 Nil
E310 Cb Nil E347 4 to 10
E310 Mo Nil E349 20 min.
depositing single weld bead layers, one on top of the obtained must be averaged to a single value for
other to a minimum height of 13 mm. If carbon steel is conversion to ferrite number.
used as a base plate, the minimum height of the pad The Appendix of A5.4 has provided technical notes
should be 16 mm. Other data are given in Table 3.38 on the standard classifications for the guidance of
and Fig. 3.20. electrode users. They are summarised below:
- Each pass must be cleaned prior to depositing the E209. The nominal composition (wt%) of this weld
next weld bead. The maximum interpass temperature metal is 22 Cr, 11 Ni, 5.5 Mn, 2 Mo, and 0.20 N. Electrodes
must be 95°C. Between passes, the weld pad may be of this composition are most often used to weld AISI
cooled by quenching in water not sooner than 20 Type 209 (UNS S20910) base metals. The alloy is a
seconds after the completion of each pass. The last pass nitrogen-strengthened austenitic stainless steel
must be air cooled to below 430°C prior to quenching exhibiting high strength with good toughness over a
in water. The completed weld pad must be draw filed wide range of temperatures. Nitrogen alloying reduces
to provide sufficient finished surface to make the the tendency for intergranular carbide precipitation in
required ferrite reading. A total of six ferrite readings the weld area by inhibiting carbon diffusion and thereby
must be taken on the filed surface along the longitudinal increasing resistance to intergranular corrosion.
axis of the weld pad with an instrument calibrated in Nitrogen alloying coupled with the molybdenum
accordance with the procedures specified in AWS A4.2. content provides superior resistance to pitting and
"Standard Procedures for Calibrating Magnetic crevice corrosion in aqueous chloride-containing media.
Instruments to Measure the Delta Ferrite Content of Type E209 electrodes have sufficient total alloy content
Austenitic Stainless Steel Weld Metal". The six readings for use in joining dissimilar alloys, like mild steel and
the stainless steels and also for direct overly on mild
steel for corrosion applications.
Table 3.38: Welding data for ferrite test E219. The nominal composition (wt.%) of this weld
metal is 20 Cr, 6 Ni, 9 Mn, and 0.20 N. Electrodes of this
Electrode Welding Dimensions of deposit composition are most often used to weld AISI Type 219
diameter current (approx. mm) (UNS S21900) base metals. This alloy is a nitrogen-
(mm) (amp) L
w
strengthened austenitic stainless steel exhibiting high
1.6 35-45 6 32 strength with good toughness over a wide range of
2.0 45-55 6 32 temperatures. Nitrogen alloying reduces the tendency
2.4 65-75 8 38 for intergranular carbide precipitation in the weld area
3.2 90-100 10 38 by inhibiting carbon diffusion and thereby, increases
4.0 120-140 13 38
resistance to intergranular corrosion. Nitrogen alloying
4.8 165-185 15 38
6.4 240-260 18 38 also improves resistance to pitting and crevice corrosion
in aqueous chloride containing media. The E219
nmznt SHALL'
BE MEASURED IN|
THIS AREA-
1
FREE OF ARC
STARTS AND
COPPER ICOPPERI COPPER BAR CRATERS
BAR BAR
APPROX
1x1x3
3/8MIN| BASE PLATE W
■—1MIN — _2.1/2 MIN
ALL DIMENSIONS IN INCHES
Fig. 3.20: Details of weld pad for ferrite test

electrodes have sufficient total alloy content for use in A carbon content of 0.04 percent max has been shown
joining dissimilar alloys, like mild steel and the stainless to be a d e q u a t e in weld metal, even t h o u g h it is
steels, and also for direct overlay on mild steel for recognized that similar base metal specifications require
corrosion applications. a 0.03 percent limitation. This low carbon alloy, however,
E240. The nominal composition (wt.%) of this weld is not as strong at elevated t e m p e r a t u r e as the
metal is 18 Cr, 5 Ni, 12 Mn and 0.02N. Electrodes of this columbium-stabilised alloys or 304H.
composition are most often used to weld AISI Type 240 E308Mo. These electrodes are the same as E308,
and 241 base metals. These alloys are n i t r o g e n except for the addition of molybdenum. E308Mo
strengthened austenitic stainless steels exhibiting high electrodes are recommended for welding ASTM CF8M
strength with good toughness over a wide range of stainless steel castings, as they match the base metal with
temperatures. Significant improvement in resistance to regard to chromium, nickel, and molybdenum. They
wear in particle-to-metal and metal-to-metal (galling) may also be used for welding wrought materials such
applications is a desirable characteristic when compared as Type 316 stainless when increased ferrite is desired
to the more conventional austenitic stainless steels like beyond that attainable with E316 electrodes.
Type 304. Nitrogen alloying reduces the tendency for E308MoL. These electrodes are recommended for
intergranular carbide precipitation in the weld area by welding ASTM CF3M stainless steel castings, as they
inhibiting carbon diffusion and thereby increasing match the base metal with regard to chromium, nickel,
resistance to intergranular corrosion. and molybdenum. E308MoL electrodes may also be
Nitrogen alloying also improves resistance to pitting used for welding wrought materials such a$ Type 3161
and crevice corrosion in aqueous chloride-containing stainless when increased ferrite is desired beyond that
media. In addition, weldments in alloys AISI 240 and attainable with E3161, electrodes.
AISI 241 when compared to Type 304, exhibit improved E309. The nominal composition (wt.) of this weld
resistance to transgranular stress corrosion cracking in metal is 23.5 Cr, 13 hlL Electrodes of this composition
hot aqueous chloride-containing media. The E240 are commonly used for welding similar alloys in
electrodes have sufficient total alloy content for use in wrought or cast form. They are used for welding
joining dissimilar alloys, like mild steel and the stainless dissimilar metals, such as joining Type 304 to carbon
steels, and also for direct overlay on mind steel for steel, welding the clad side of Type 304 clad steels, and
corrosion and wear applications. applying stainless steel sheet linings to carbon steel
E307. The nominal composition (wt.%) of this weld shells. Occasionally, they are used to weld Type 304 and
metal is 19 Cr.9.8 Ni, 4 Mn. Electrodes of this similar base metals where severe corrosion conditions
composition are used primarily for moderate strength exist requiring higher alloy weld metal.
welds with good crack resistance between dissimilar E309L. The composition of this weld metal is the
steels such as austenitic manganese steel and carbon same as that deposited by E309 electrodes, except for
steel forgings or castings. the restricted carbon content. The 0.04 percent max
E308. The nominal composition (wt.%) of this weld carbon content of these weld deposits reduces the
metal is 19 Cr, and 10 Ni. Electrodes of this composition possibility of intergranular carbide precipitation and
are most often used to weld base metal of similar thereby increases the resistance to intergranular
composition such as AISI Types 301, 302, 304, and 305. corrosion w i t h o u t the use of stabilizers such as
E308H. These electrodes are the same as E308 except columbium (niobium) and titanium. This low carbon
that the allowable carbon content has been restricted to alloy, however, is not as strong at elevated temperature
the higher portion of the E308 range. Carbon content in as the columbium-stabilised alloys or high carbon
the range of 0.04 - 0.08 provides higher tensile and creep content Type 309 deposits.
strengths at elevated temperatures. These electrodes are E309Cb. The composition of this weld metal is the
used for welding Type 304H base metal. same as Type 309, except for the addition of columbium
E308L. The composition of the weld metal is the same (niobium) and a reduction in the carbon limit. The
as E308, except for the restricted carbon content. The columbium (niobium) provides resistance to carbide
0.04 percent max carbon content of weld metal deposited precipitation and thus increases intergranular corrosion
by these electrodes reduced possibility of intergranular resistance and also provides higher strength in elevated
carbide precipitation a n d thereby increases the temperature service. E309Cb electrodes are used also
resistance to intergranular corrosion without the use of for welding Type 347 clad steels or for the overlay of
stabilizers such as columbium (niobium) or titanium. carbon steel.
E309Mo. The composition of this weld metal is the should be limited to service temperature below 800QF
same as that deposited by E309 electrodes, except for (420QC) to avoid formation of secondary brittle phases.
the addition of molybdenum and a small reduction in E316. The nominal composition (wt.%) of this weld
the carbon limit. These electrodes are use for welding metal is 18.5 Cr, 12.5 Ni, 2.5 Mo. These electrodes are used
Type 316 clad steels or for the overlay of carbon steels. for welding Type 316 and similar alloys. They have been
E309MoL. The composition of these weld metal is used successfully in certain applications involving special
the same as that deposited by E309Mo electrodes, except base metals for high-temperature service. The presence
for the restricted carbon content. The lower carbon of molybdenum provides creep resistance at elevated
content of the weld metal reduces the possibility of temperatures. Rapid corrosion of Type 316 weld metal
intergranular corrosion. may occur when the following three factors co-exist:
E310. The nominal composition (wt.%) of this weld (1) The presence of a continuous or semicontinuous
metal is 26.5 Cr, 21 Ni. Electrodes of this composition network of ferrite in the weld metal microstructure
are most often used to weld base metals of similar (2) A composition balance of the weld metal giving
composition. a chromium-to-molybdenum ratio of less than 8.2 to 1
E310H. The composition of this weld metal is the (3) Immersion of the weld metal in a corrosive me-
same as that deposited by E310 electrodes, except that dium
carbon ranges from 0.35 to 0.45 percent. These electrodes Attempts to classify the media in which accelerated
are used primarily for welding or repairing high alloy corrosion will take place by attack on the ferrite phase
heat and corrosion resistant castings of the same general have not been entirely successful. Strong oxidizing and
composition which are designated as Type HK by the mildly reducing environments have been present where
Alloy Castings Institute. The alloy has high strength at a number of corrosion failures were investigated and
t e m p e r a t u r e s over 1700° F (930 Q C). It is not documented. The literature should be consulted for
recommended for high sulphur atmospheres or where latest recommendations.
severe thermal shock is present. Long time exposure to E316H. These electrodes are the same as E316, except
temperatures in the approximate range of 1400 to 1600 that the allowable carbon content has been restricted to
F (760 to 870QC) may induce formation of sigma and the higher portion of the E316 range. Carbon content in
secondary carbides which may result in reduced the range of 0.04 to 0.08 provides higher tensile and creep
corrosion resistance, reduced ductility, or both. strengths at elevated temperatures. These electrodes are
E310Cb. The composition of this weld metal is the used for welding 316H base metal.
same as that deposited by E310 electrodes, except for E316L. This composition is the same as E316, except
the addition of columbium (niobium) and a reduction for the restricted carbon content. The 0.04 percent max
in carbon limit. These electrodes are used for the welding carbon content of weld metal deposited by these
of heat resisting casting, Type 347 clad steels, or the electrodes reduces the possibility of intergranular
overlay of carbon steels. carbide p r e c i p i t a t i o n a n d thereby increases the
E310Mo. The composition of this weld metal is the resistance to intrgranular corrosion without the use of
same as that deposited by E310 electrodes, except for stabilizers such as columbium (niobium) or titanium.
the addition of molybdenum and a reduction in carbon These electrodes are used principally for welding low
limit. These electrodes are used for the welding of heat carbon, molybdenum-bearing austenitic alloys. Tests
resisting castings, Type 316 clad steels, or for the overlay have shown that 0.04 percent carbon limit in the weld
of carbon steels. metal given adequate protection against intergranular
E312. The nominal composition (wt.%) of this weld corrosion in most cases. This low carbon alloy, however,
metal is 30 Cr, 9 Ni. These electrodes were originally is not as strong at elevated temperatures as Type E316H.
designed to weld cast alloys of similar composition. E317. The alloy content of weld metal deposited by
They have been found to be valuable in welding these electrodes is somewhat higher than that of E316
dissimilar metals, especially if one of them is a stainless electrodes, p a r t i c u l a r l y in m o l y b d e n u m . These
steel, high in nickel. This alloy gives a two-phase weld electrodes are usually used for welding alloys of similar
deposit with substantial a m o u n t s of ferrite in an composition and are utilized in severely corrosive
austenitic matrix. Even with considerable dilution by environments (such as those containing halogens) where
austenite-forming e l e m e n t s , such as nickel, the crevice and pitting corrosion are of concern.
microstructure remains two-phase and thus highly E317L. The composition of this weld metal is the
resistant to weld metal cracks and fissures. Applications same as that deposited by E317 electrodes, except for
the restricted carbon content. The 0.04 percent max elevated temperature. Repairs of defects in alloy castings
carbon content of weld metal deposited by these and the welding of castings and wrought alloys of
electrodes reduces the possibility of intergranular similar composition are the most common applications.
carbide precipitation a n d thereby increases the E330H. The composition of this weld metal is the
resistance to intergranular corrosion without the use of same as that deposited by E330 electrodes, except that
stabilizers such as columbium (niobium) or titanium. carbon ranges from 0.35 to 0.45 percent. These electrodes
This low carbon alloy, however, is not as strong at are used primarily for the welding and repairing of high
elevated temperatures as the columbium (niobium)- alloy heat and corrosion resistant castings of the same
stabilised alloys or the standard Type 317 weld metal general composition which are designated HY by the
with a higher carbon content. Alloy Castings Institute. This composition can be used
E318. The composition of this weld metal is the same to 2100 F (1160 °C) in oxidizing atmospheres and at 2000
as that deposited by E316 electrodes, except for the R (1090 °C) in reducing atmospheres. However, high-
addition of c o l u m b i u m (niobium). C o l u m b i u m s u l p h u r e n v i r o n m e n t s m a y adversely affect
(niobium) provides resistance to intergranular carbide performance at elevated temperature.
precipitation and t h u s increased resistance to E347. The nominal composition (wt.%) of this weld
intergranular corrosion. These electrodes are used metal is 19.5 Cr, 10 Ni with Cb (Nb) or Cb (Nb) plus TA
primarily for w e l d i n g base metals of similar added as a stabilizer. Either of these additions reduces
composition. the possibility of intergranular chromium carbide
E320. The nominal composition (wt. %) of this weld precipitation and thus increases resistance to
metal is 20 Cr, 34 Ni, 2.5 Mo, 3.5 Cu, with Cb (Nb) added intergranular corrosion.
to improve resistance to intergranular corrosion. These These electrodes are usually used for welding
electrodes are primarily used to weld base metals of c h r o m i u m - n i c k e l alloys of similar composition
similar composition for applications where resistance stabilized either w i t h c o l u m b i u m (niobium) or
to severe corrosion is required for a wide range of t i t a n i u m . Electrodes d e p o s i t i n g t i t a n i u m and a
chemicals including sulfuric and sulfurous acids and stabilizing element are not commercially available
their salts. These electrodes can be used to weld both because titanium is not readily transferred across the
castings and wrought alloys of similar composition are in shielded metal are welding. Although columbium
without postweld heat treatment. (niobium) is the stabilizing element usually specified
A modification of this grade without columbium in Type 347 alloys, it should be recognized that tantalum
(niobium) is available for repairing castings which do also is present. Tantalum and columbium (niobium) are
not contain c o l u m b i u m . With this modified almost equally effective in stabilizing carbon and in
composition, solution annealing is required after providing high-temperature strength. This specification
welding. recognizes the usual commercial practice of reporting
E320LR (Low Residuals). Weld metal deposited by c o l u m b i u m (niobium) as the sum of columbium
E320LR electrodes has the same basis composition as (niobium) plus tantalum. If dilution by the base metal
that deposited by E320 electrodes; however, the produces a low ferrite or fully austenitic weld metal
elements C, Si, P, and S are specified at lower maximum deposit, crack sensitivity of the weld may increase
levels, and Cb (Nb) and Mn are controlled within substantially.
narrower ranges. These changes reduce the weld metal Some applications, especially those involving high-
fissuring (while maintaining the corrosion resistance) temperature service, are adversely affected if the ferrite
frequently encountered in fully austenitic stainless steel content is too high. Consequently, a high ferrite content
weld metals. Consequently, welding practices typically should not be specified unless tests prove it to be
used to deposit ferrite-containing austenitic stainless absolutely necessary.
steel weld metals can be used. Type 320LR weld metal E349. The nominal composition (wt.%) of this weld
has a lower minimum tensile strength than Type 320 metal is 19.5 Cr, 9 Ni, 1 Cb (Nb), 0.5 Mo, 1.4 W. These
weld metal. electrodes are used for welding steels of similar
E330. The nominal composition (wt.%) of this weld composition such as AISI Type 651 or 652. The
metal is 35 Ni, 15.5 Cr. These electrodes are commonly combination of columbium (niobium), molybdenum,
used where hear - and scale-resisting properties above and tungsten with chromium and nickel gives good
188 °F (980 °C) are required. However, highsulfur h i g h t e m p e r a t u r e r u p t u r e strength. The chemical
environments may adversely effect performance at composition of the weld metal results in an appreciable
content of ferrite which increases the crack resistance specification. (Excessive chromium will result in
of the weld metal. lowered ductility.) Welding with E430 electrodes usually
E383. The nominal composition (wt.%) of this weld requires preheat and post-heat. Optimum mechanical
metal is 28 Cr, 31.5 Ni, 3.7 Mo, 1 Cu. These electrodes properties and corrosion resistance are obtained only
are used to weld base metal of a similar composition to w h e n the w e l d m e n t is heat-treated following the
itself and to other grades of stainless steel. Type E383 welding operation.
weld metal is r e c o m m e n d e d for s u l p h u r i c and E502. The nominal composition (wt.%) of this weld
phosphoric acid environments. metal is 5 Cr, 0.5 Mo. These electrodes are used for
The elements C, Si, P, and S are specified at low welding base metal of similar composition, usually in
maximum levels to minimize weld metal hot cracking the form of pipe or tubing. The alloy is an air-hardening
a n d fissuring (while m a i n t a i n i n g the corrosion material; therefore, when welding with these electrodes,
resistance) frequently encountered in fully austenitic preheat and postweld heat treatment are required.
stainless steel weld metals. E505. The nominal composition (wt.%) of this weld
E385. The nominal composition (wt.%) of this weld metal is Cr, 1 Mo. These electrodes are used for welding
metal is20.5 Cr, 25 Ni,5 Mo, 1.5 Cu. These electrodes are base metal of similar composition, usually in the form
used primarily for welding of Type 904L materials for of pipe or tubing. The alloy is an air-hardening material
the handling of sulphuric acid and many chloride- and, therefore, when welding with these electrodes,
containing media. E385 electrodes also may be used to preheat and postweld heat treatment are required.
join Type 317L material where improved corrosion E630. The nominal composition (wt.%) of these
resistance in specific media is needed. E385 electrodes electrodes is 16.4 Cr, 4.7 Ni, 3.6 Cu. These electrodes are
also can be used for joining Type904L base metal to other primarily designed for welding ASTM A564, Type 630,
grades of stainless. The elements C, Si, P and S are and some other precipitaion-hardening stainless steels.
specified at lower maximum levels to minimize weld The weld metal is modified to prevent the formation of
metal hot cracking and fissuring (while maintaining ferrite networks in the martensite microstructure which
corrosion resistance) frequently encountered in fully could have a deleterious effect on mechanical properties.
austenitic weld metals. Dependent on the application and weld size, the weld
E410. This 12 Cr alloy is an air-hardening steel. metal may be used either as-welded, welded and
Preheat and postheat treatments are required to achieve precipitation hardened, or welded, solution treated and
welds of adequate ductility for many engineering precipitation hardened.
purposes. The most common application of these E16-8-2. The nominal composition (wt.%) of this
electrodes is for welding alloys of similar compositions. weld metal is 15.5 Cr, 8.5 Ni, 1.5 Mo. These electrodes
They are also used for surfacing of carbon steels to are used primarily for welding stainless steel, such as
resist corrosion, erosion, or abrasion. Types 16-8-2, 316, and 347, for high-pressure, high-
E410NIMo. These electrodes are used for welding temperature piping systems. The weld deposit usually
ASTM CA6NM castings or similar materials, as well as has a Ferrite Number not higher than 5 FN. The deposit
light gauge Type 410, 410S, and 405 base metals. Weld also has good hot ductility properties which offer
metal deposited by E410 electrodes are modified to relative freedom from weld or crater cracking even
contain less chromium and more nickel than weld metal under high-restraint conditions. The weld metal is
deposited by E410 electrodes. The objective is to usable in either the as-welded or solution-treated
eliminate ferrite in the microstructure, as ferrite has a condition. These electrodes depend on a very carefully
deleterious effect on mechanical properties of this alloy. balanced chemical composition to develop their fullest
Final post-weld heat treatment should not exceed 1150QF properties. Corrosion tests indicate that Type 16-8-2
(620SC). Higher temperatures may result in rehardening weld metal may have less corrosion resistance than Type
due to untempered martensite in the microstructure 316 base metal depending on the corrosive media.
after cooling to room temperature. Where the weldment is exposed to severe corrodents,
E430. The weld metal deposited by these electrodes the surface layers should be deposited with a more
contains between 15 and 18 Cr (wt.%). The composition corrosion resistant weld metal.
is balanced by providing sufficient chromium to give E7Cr. The nominal composition (wt.%) of this weld
adequate corrosion resistance for the usual applications metal is 7 Cr, 0.5 Mo. These electrodes are primarily used
and yet retain sufficient ductility in the heat-treated in welding base metal of similar composition. The Cr
condition to meet the mechanical requirements of the base metal usually is furnished as tubing, pipe, or
MANUAL METAL-ARC WELDING 1 27
casting. This alloy is an air-hardening material and followed by digits L, H, Nb and W. L indicates a low
requires the use of both preheat and postweld heat carbon content version, H indicates a high carbon
treatment for satisfactory welding and service. content version, Nb indicates the presence of niobium
E2209. The nominal composition (wt.%) of this weld as a stabiliser and the symbol W indicates the presence
metal is 22.5 Cr, 9.5 Ni, 3 Mo, 0.15 N. Electrodes of this of tungsten.
composition are used primarily to weld duplex stainless The composition code must be followed by letter B
steels which contain approximately 22 percent of or R, the former indicating basic lime-fluorspar covering
chromium. Weld metal deposited by these electrodes for DC+ welding and the latter indicating rutile type
has "duplex" microstructure consisting of an austenite- covering for AC or DC+. After this, the letters MP (metal
ferrite matrix. Weld metal deposited by E2209 electrodes powder) must be used for those electrode types whose
combines increased tensile strength with improved core wire is not of Cr-Ni austenitic or Cr-steel and in
resistance to pitting corrosive attack and to stress which the bulk of the alloying elements are supplied
corrosion cracking. through the coverings.
E2553. The nominal composition (wt.%) of this weld For chemical analysis, drillings must be taken from
metal is 25.5 Cr, 7.5 Ni, 3.5 Mo, 2 Cu and 0.17 N. These either the fractured parallel portion of the all-weld
electrodes are used primarily to weld duplex stainless tensile test specimen or from the centre-line of the all-
steels which contain approximately 25 percent of chro- weld test assembly, remote from tile test specimen or
mium. Weld metal deposited by these electrodes has a from a weld pad.
"duplex" microstructure consisting of an austenite- For all-weld tensile test, the procedure of making
ferrite matrix. Weld metal deposited by E2553 electrodes up the welded assembly is similar to that described
combines increased tensile strength with improved under low-alloy steel electrodes. During welding, the
resistance to pitting corrosive attack and to stress i n t e r p a s s t e m p e r a t u r e for straight chrome-steel
corrosion cracking. electrodes has to be controlled between 150 and 260°C
and for all others between 20 and 150°C. While Cr-Ni
2) BS: 2926-1970, "Specification for chromium-nickel types do not require any postweld heat treatment, the
austenitic and chromium steel electrodes" straight chrome-steel types demand that the tensile
This standard covers 20 types of electrodes as listed specimen be heat-treated as described below:
in Table 3.39 which also gives the nearest equivalent
AWS classification and weld metal chemistry for each Composition code Postweld heat treatment
type. The first two are straight chrome types, while the
840-870°C for 2 hours
rest are Cr-Ni austenitic types. As pointed out before, 5
13 Furnace-cool to 600°C at cooling rate
Cr, 7 Cr and 9 Cr types are not included, as they are
of 55°C/hour (max.) then air cool
covered in the standard on low-alloy steel electrodes.
Columbium is referred to as Niobium and is indicated 760-790°C for 4 hours
by the symbol Nb. Wherever it is present, its minimum 17 Furnace-cool to 600°C at cooling rate
content has to be ten times the carbon content, to en- of 55°C/hour (max.) then air cool
sure freedom from intercrystalline corrosion, when the
welded joint is exposed to corrosive atmosphere in ser- The all-weld tensile values which must be satisfied
vice. A footnote to the table mentions that the maxi- are shown in Table 3.40. It must be noted that the
mum nickel content may be increased when a low-delta elongation values are based on the factor 1 = 5d as against
ferrite content is required in the weld deposit to meet l=4d in the case of the AWS standard.
severe corrosive conditions, or to prevent formation of The standard also prescribes a fillet weld test, more
the sigma phase in high temperature service. The prob- or less similar to the one described in AWS A5.4.
lem of sigma phase will be explained later. As in the case of other BS standards on electrodes,
this one lays down the requirements for production
The BS coding system starts with the composition
which include, a m o n g other things, initial tests
code as shown in Table 3.39, which is based on proposals
comprising analysis, all-weld tensile test and fillet weld
from the International Institute of Welding and gives
test, and production control.
an immediate indication of the chemical composition
of the deposited weld metal. It consists of a figure 3) IS: 5206-1983: "Specification for covered electrodes
indicating Cr content, another figure indicating Ni for manual metal arc welding of stainless steel and other simi-
content and a third figure indicating Mo content, lar high alloy steels"
O C j l C J l C J l W W O J W v £ ) v £ ) v £ ) v £ > v £ ) v O v O v > 0
OO hO hO N3 v£> vO vO fr
to jo p p O ro ro ro N> w w to N> S|
^ ^ w w w
H z r ?
m W W trj tTj w w w w w w w w w
w w W W W W t o t o O J t o 4t o t o t o 4 * . 4 *
H-* O O O I-A (—* H-* h- 4 ^ O O t O H-*
o O v£> vO v£> ^ J O N O N O N V I O O O O O O £*• 2. 3 7>-s- 5
n . 19 ft
a- 3
o o 3
o w
ON CJl
5'
n
p p p p p p p p - p p p p p p p p p p p 3
l l ft)
H-* 4*> ^ O O ^ O O H - O O H - O X
o K ) CJl 0 4 ^ 0 0 0 4 ^ 0 0 0 4 ^ 0 0 0 0 0
H^ O O O O
o bo vq ^q ^q o o o o o o o o o o o o o
o o 0 0 0 0 0 0 0 0 0 0 0 0 0
CJl CJl 5'
u i a i u i ( j i u i a i b i u i b i u i b i a i ( j i
3
to 3
o C J I O O O C J I C J I C J I C J I C J I C J I C J I C J I C J I C J I C > I C J I C > I O O ft)
i - i h O N ) r o M h o h J h O h - * > - > i - i ^ > - * i - » i - * i - > > - * i - > H ^ 3
00 O \ ^ ^ ^ M N ) N ) N ) ^ N S l v i N M ( X 0 0 ( X ( J l H
CJl c j i o o o o o o o o o o o o o o o o o o
5'
n
M ' N J M N 3 M l s i | s J N J N ) | s J W N 3 N ) N N ) N J N 3 h ' H
t—* 0 0 0 0 0 0 0 0 C j l C J l C j l C j l O O O O O O > - k | - i | - * 0 0 t 0
O c j i o o o o o o o o o o o o o o o o o c j i ^
^
00 00 00 00 v£> vO v£> 3
o o o
o o o o 5'
o o o o o o o o o o o o o
l >-* l h-» O O
w.
CJI w.
CJI w.
CJI ~4
-^^ - 4* ►— •— I — *
o o
CJl T ^ ^ T ^ T ^ O O O O O O - ^ ^ ! ^ ^ ^
o 0 0 0 0 0 0 0 * * * * * * 0 0 0 0 0
N> N> N> N> N> N> N>
9
o O CJl CJI CJl CJl CJl CJl
2
o
4* K3 to CJI CJl CJl tO to tO 3
o CJI o CJl on CJl CJl CJl CJl ft)
X
1—»
O o o o o 3
X X X X X
n n n n n 5'
3
O ft)
X
ho 3
o 5"
4*
O
O O O O O O O O O O O O O O O O O O O
o o o o o o _ _
o _
o _
o _o _ o o o _ o _ o _ o _ o _ o_ o 75
to to to to to o
CJl o o o o CJl W W W W W W W t O t O t O t O t O t O t O
CjlCjlCjlCjlCjl- C j l C j l C j l C j l C j l C j l C j l O O
o o o o o o o o o o o o o o o o o o o o
o o o o
to to to to 2 o4 ^ 4oi .o4 ô4 i o. 4 ô4 ô4 io4 so 4o^ 4 o^ 4 ô4 ô4 ô4 ^
o o o o o k k
o O O O O O O O O O O O O O O
A 0 O 1 O N H 0 3 1 0NICH3M OUV NU3QOW g<J |.
Table 3.40: All-weld tensile requirements as per BS 2926 symbols for metal recovery (4.1.4) and for welding
characteristics (4.1.5) are also compulsory.
Composition Tensile strength . Elongation on 3.2 First Part - Symbol for the Product - The prefix
code min. ( N / m m 2 ) \=5d (%)
letter 'E' shall indicate the suitability of the covered
13 480 20 electrode for manual metal are welding.
17 480 20 3.3 Second Part - Symbol for Chemical Composition
19.9 560 30 - The symbols use for the chemical composition of the
19.9 L 510 30 deposited weld metal are given in Table 1. (In Appendix
19.9 Nb 560 25 A is given a comparison of classification in American,
19.12.3 560 25 British, German, Japanese and ISO Standards.)
19.12.3 L 510 25 3.4 Third Part - Symbol for Type of Covering - The
19.12.3 Nb 560 25 type of covering is symbolized by a letter as follows:
19.13.4 560 25
19.13.4 L 510 25
a) B for basic,
19.13.4 Nb 560 20 b) R for rutile, and
23.12 560 30 c) S for other types.
23.12 Nb 560 25 3.4.1 Basic (B) - Electrodes of the basic type have a
23.12.2 560 30 rather thick covering containing considerable quantities
23.12 W 560 20 of calcium or other basic carbonates together with
25.20 560 25 fluorspar or other fluor compositions. The slag is very
25.20 H 560 * fluid and rise quickly to the surface of the weld. Slag
25.20 Nb 560 25 inclusions are therefore not likely to occur. After
17.8.2 560 25 solidification, the slag generally has a brown to dark-
20.9.5 620 20 brown colour and a glossy appearance. It is, except for
root-passes, easy to detach. Basic electrodes give a
*This high carbon electrode is intended for service at elevated tempera-
tures, where creep resistance is of primary importance. Room temperature medium penetration and they can, as a rule, be used for
elongation measurements have little meaning in this case. For guidance, an welding in all positions as well as for pipe welding. Basic
elongation of about 5% may be expected. electrodes should be welded with a short are in order
to give good quality welds.
When manufactured, the basic electrodes are baked
The standard covers 32 types, whose requirements at a temperature sufficient to result in low water content
of chemical composition and of all-weld metal test are of the covering. In order to avoid porosity, basic
closely similar to the equivalent types mentioned against electrodes should be stored in such a manner as to
each in Table 3.42. prevent absorption of moisture and be re-dried before
3.1 The electrodes are classified on the basis of use according to the r e c o m m e n d a t i o n s of the
chemical analysis of weld metal, type of covering, metal manufacturer. Stainless steel electrodes with basic
recovery, welding characteristics of the electrode and coating are particularly used for the welding of thick
also the special characteristics of the electrode. The plates, highly restrained joints and joints in vertical and
classification is divided into six parts. overhead positions.
3.1.1 First part is a symbol for the type of product. 3.4.2 Rutile (R) - The covering of stainless steel
3.1.2 The second part is a symbol for the chemical electrodes of this type contains large amounts of rutile
composition of the deposited weld metal. or components derived from minimum dioxide, oxides
3.1.3 The third part is a symbol for the type of may include basic components. The electrodes are easy
covering. to ignite, have a stable are and low spatter losses. They
3.1.4 The fourth part is a symbol for metal recovery. are easy to weld in all positions and slag detachability
3.1.5 The fifth part is a symbol for the welding and bead appearance are good. They deposited weld
characteristics of the electrode. metal quality is also good.
3.1.6 The sixth part is a symbol 'X' indicating that When manufactured, stainless steel electrodes with
the electrode has some special properties (synthetic, rutile covering are generally baked at high temperatures.
fully nustenitie, etc.). This part is always compulsory In order to avoid porosity they should be stored under
when the electrode has special feature. In this case the dry conditions or if they have absorbed moisture, before
use according to the r e c o m m e n d a t i o n , of the

Table 3.41: Current condition
manufacturer.
3.4.3 Other Types (S) - This symbol is used for Symbol Direct current Alternating current
electrodes that cannot be defined as basic or rutile ones. recommended minimum open circuit
polarity voltage
3.5 Fourth Part - Symbol for Metal Recovery - If the
recovery is less than 110 percent no symbol is used. The (I) (2) (3)
Of —
higher value found by the method for determining the
1 + or- 50
recovery, rounded to the nearest multiple of 10 are used 2 - 50
as symbol for metal recovery, for example 110,120,130, 3 + 50
140, 150, etc. For the method of determining the 4 + or- 70
recovery of electrode "see Appendix D of IS: 815-1974". 5 - 70
6 - 70
3.6 Fifth Part - Symbols for Welding Characteristics 7 + or - 90
3.6.0 - The w e l d i n g characteristics shall be 8 - 90
symbolized by a two-digit number. The first digit 9 + 90
indicates the welding positions for which the electrode fSymbol reserved for electrodes used exclusively on direct current Posi-
is recommended and the second digit indicates the tive polarity +, negative polarity -
welding current conditions.
3.6.1 The general welding positions for which the
electrode is recommended are symbolized by a digit, as
follows : 3.7 Symbol for Special Properties
3.7.1 The symbol 'X' indicates that the electrode has
Digit Positions
one or more special properties. Synthetic electrodes, for
1 All positions example, could be characterized in this way. There are
2 All positions except vertical downward also other examples. Normally, austenitic deposited
3 Flat butt weld, flat fillet weld, horizontal/ weld metals of the 19.12.2 L type have a ferrite content
vertical fillet weld of 3 to 7 percent. For special purpose, fully austenitic or
4 Flat butt weld, flat fillet weld low ferrite variants are also developed. Such electrodes
5 Same as 3 and recommended for vertical could be marked by the symbol 'X'.
downward 3.7.2 The special properties should not, however, be
Note - Electrodes covered in this standard are not normally used in the codified. The symbol 'X' only recommends the user to
vertical downward position. study the electrode manufacturer's catalogue for further
information.
3.6.2 The welding current and open circuit voltage 3.8 Instructions for use of Classification Symbols -
are symbolized by a digit corresponding to the In order to promote the use of this identification, the
characteristics of the welding equipment required in codification is split into sections.
order to ensure working conditions free of incidents such 3.8.1 Compulsory Section - This section includes the
as instability or interuptions of the arc. Refer Table 3.41. symbols for the type of product, for the chemical
The-open circuit voltage necessary for striking the composition of the deposited weld metal and for the
arc varies according to the diameter of the electrode . A type of covering, and a symbol 'X' indicating that the
reference diameter is required for symbolization. Table electrode has some special properties, that is, the
2 applies to electrode diameters greater than or equal symbols defined in 4.2, 4.3, 4.4 and 4.7.
to 2.5 mm. If electrodes of smaller diameter are used, a 3.8.2 Optional Section - This section includes the
higher voltage may be necessary. symbols for the metal recovery and for the welding
The frequency of the alternating current is assumed characteristics of the electrode.
to be 50 or 60 Hz. The open-circut voltage necessary 3.8.3 Examples of Classification
when electrodes are used on direct current is closely re- Example 1 :
lated to the dynamic characterisitcs of the welding power Covered electrode for manual electric are welding,
source. Consequently no indication of the minimum having a rutile covering and depositing weld metal of
open-circuit voltage for direct current can be given. the following chemical analysis.
Constituent Percent Constituent Percent either of 110, 120, 130, 140, 150, etc. Next is the digit
which symbolises the general welding positions as
C 0.8 Max Si 0.9 Max follows:
Cr 18 to 21 S 0.03 Max
Ni . 8 to 11 P 0.04 Max Symbol Positions of welding
Mn 2.5 Max 1 All positions
It may be used for welding in all positions. To be 2 All positions, except vertical downward
connected on direct current to the positive pole; on 3 Flat butt weld, flat fillet weld, horizontal /
alternating current welding, requiring an open-circuit vertical fillet weld
voltage of 70 V. Has metal recovery of 120 percent. 4 Flat butt weld, flat fillet weld
The complete codification for the electrode will be: 5 As 3 and recommended for vertical
E 19.9 R 120 16 downward
and the compulsory part will be The next symbol is a digit indicating the type of
E 19.9 R current and, in the case of AC, the minimum open circuit
Example 2 : voltage (OCV) of the welding equipment necessary to
Covered electrode for manual electric arc welding, ensure a stable arc during welding. This is explained in
having a basic covering and depositing weld metal Table 3.44.
having the following chemical analysis: Referring to this table, it is stated that the OCV
Constituent Percent Constituent Percent necessary for striking the arc varies according to the
electrode diameter. The values given in the table apply
c
Cr
0 04 Max
17 to 20
Si
Mn
0 .9 Max
2.5Max to electrodes of 2.5 mm diameter or larger. For electrodes
Ni 11 to 14 S 0 03 Max of smaller diameters, a higher voltage may be necessary.
Mo 2.0 to 2.5 P 0.04 Max Further, it is stated that "the OCV necessary when
It may be used for welding in all positions except, electrodes are used on direct current is closely related
vertical downward. To be connected on direct current to the dynamic characteristics of the welding power
source. Consequently no indication of the minimum
welding to the positive pole; on alternating current
open-circuit voltage for direct current can be given."
requiring an open circuit voltage of 70 V. Refer to
manufacturer's r e c o m m e n d a t i o n s for the special Finally, suffix symbol X is to be used to indicate
characteristics of the electrodes. special or abnormal properties. These properties need
The complete codification for the electrode will be : not be codified. X only recommends that the electrode
producer's catalogue be studied for further information.
E 19.12.2.L R 14 X
For example, it could be used for synthetic electrodes
and the compulsory part will be:
and for 19.9 types with nil ferrite.
E19.12.2LRX
The standard has given the following two examples
Fillet weld test similar to that described in the AWS
to illustrate the coding system:
standard is also specified. A guide to the classification
1) Covered electrode for manual electric arc welding,
and selection is provided in the Appendix.
having a rutile covering and depositing weld metal of
4) ISO 3581-1976(E): "Covered electrodes for manual
the following chemical analysis:
arc welding of stainless and other similar high-alloy steels—
Code of symbols for identification" C 0.08
The ISO codification system starts with the letter E Cr 19
and is followed by a symbol for the chemical Ni 10
composition of the deposited weld metal as shown in It may be used for welding in all positions. To be
Table 3.43. Next comes the symbol for the type of connected on direct current welding to the positive pole;
covering, which is B for basic, R for rutile and 5 for other on alternating current welding, requiring an open-circuit
types. These symbols make up the compulsory part of voltage of 70 V Metal recovery: 120%.
the codification. After this comes the symbol for metal The complete codification for the electrode will
recovery (metal recovery is determined by the method therefore be
described in ISO 2401). If the recovery is less than 110%, E 19 9 R 120 16
no symbol is used. The higher values are rounded to and the compulsory part will be
the-nearest multiple of 10. Hence this symbol can be E 19 9 R
Table 3.42: Chemical compositions of all-weld metal

Classification Carbon Chromium Nickel Molyb- Niobium Manganese Silicon Phosphur Sulphur Copper
denum Tantalum
E13 0.12 11-14 . 0.5 _ 1.0 0.90 0.04 0.03 >

E13.4 0.07 12-15 3.5 0.49-0.70 - 1.0 9.90 0.04 0.03 -
E17 0.10 15-18 - 0.5 - 1.0 0.90 0.04 0.03 -
E17.4 0.10 15-18 3-5 0.9-1.25 - 1.0 0.90 0.04 0.03 -
E19.9 0.03 18-21 8-11 0.5 - 2.5 0.90 0.04 0.03 -
E19.9-L 0.04 18-21 8-11 0.5 - 2.5 0.90 0.04 0.03 -
E19-9Nb 0.08 18-21 8-11 0.5 8xCmin to 2.5 0.90 0.04 0.03 -
1.2 max
E19.9-LNb 0.04 18-21 8-11 0.5 8xCmin to 2.5 0.90 0.04 0.03 -
1.2 max
E19.12.2 0.08 17-20 11-14 2.2-5 - 2.5 0.90 0.04 0.03 -
E19.12.2-L 0.04 17-20 11-14 2.2-5 - 2.5 0.90 0.04 0.03 -
E19.12.2Nb 0.08 17-20 11-14 2.2.5 8xCmin to 2.5 0.90 0.04 0.03 -
1.2 max
E19.12.3 0.08 17-20 10-14 2.5-3.5 - 2.5 0.90 0.04 0.03 -
E19.12.3-L 0.04 17-20 10-14 2.5-3.5 - 2.5 0.90 0.04 0.03 -
E19.12.3Nb 0.08 17-20 10-14 2.5-3.5 8xC min to 2.5 0.90 0.04 0.03 -
1.2 max
E23.12 0.15 22-26 11-15 0.5 - 2.5 0.90 0.04 0.03 -
E23.12-L 0.04 22-26 11-15 0.5 - 2.5 0.90 0.04 0.03 -
E23.12-Nb 0.12 22-25 11-15 0.5 8xC min to 2.5 0.90 0.04 0.03 -
1.2 max
E23.12.2 0.12 22-25 11-15 2.3 - 2.5 0.90 0.04 0.03 -
E16.8.2 0.10 14.5-16.5 7.5-9.5 1.2 - 2.5 0.50 0.03 0.03 -
E18.8Mn 0.20 17-20 7-10 0.5 - 5.8 0.90 0.04 0.03 -
E18.15.3 L 0.04 16.5-19.5 13-16 2.5-3.5 - 2.5 0.90 0.04 0.03 -
E25.20 0.20 24-28 18-22 0.5 - 2.5 0.75 0.03 0.03 -
E25.20 L 0.04 24-28 18-22 0.5 - 2.5 0.75 0.03 0.03 -
EE25.20 Nb 0.12 24-28 18-22 0.5 8xC to min 2.5 0.75 0.03 0.03 -
1.2 max
E25.20.2 0.12 25-28 20-22 2-3 - 2.5 0.75 0.03 0.03 -
E25.20C 0.25.0.45 24-28 18-22 0.5 - 2.5 0.90 0.03 0.03 -
E25.25.2 Nb 0.10 24-27 24-26 2-2.5 8xC min to 2.5 0.90 0.03 0.03 -
1.2 max
E20.25.5Lcu 0.04 19-22 24-26 4-6 - 2.5 0.90 0.03 0.03 1-3
E23.27.3 LcuNb 0.04 21-25 25-29 2.5-4.3 8xC min to 2.5 0.90 0.03 0.03 2.5-3.5
1.2 max
E25.4 0.15 24-27 4-6 0.5 - 2.5 0.90 0.04 0.04 -
E29.9 0.15 28-32 8-12 0.5 - 2.5 0.90 0.04 0.03 -
E18.36 0.25 14-19 33-35 0.5 - 2.5 0.90 0.04 0.03 -
Note 1: Analysis shall be made for the elements for which specified values are shown in the table. If the however, the presence of other elements is
indicated in the course of routine analysis, further analysis shall be made to determine that the total of these other elements, except iron, is not present in
excess of 0.7 percen t.
Note 2: Single values shown are maximum percentages except where otherwise specified.
*Carbon shall be analysed to the nearest 0.01 percent.
iChromium shall be 1.9 x M min when specified.
Table 3.43: Symbolisation as per ISO 3581

Symbol C max. Cr. Ni Mo Other elements
13 0.12 11 to 14
13.1 0.07 12 to 15 0.8 to 1.5
13.4 0.07 12 to 15 3 to 6 1.0 max
17 0.10 15 to 16
17.0.1 0.25 15 to 18 1 to 1.5
30 0.10 27 to 30
19.9 0.08 18 to 21 8 to 11
19.9 L 0.04 18 to 21 8 to 11
19.9 Nb 0.08 18 to 21 8 to 11 Nb
19.9 L Nb 0.04 18 to 21 8 to 11 Nb
19.12.2 0.08 17 to 20 11 to 14 2 to 2.5
19.12.2.L 0.04 17 to 20 11 to 14 2 to 2.5
19.12.2.Nb 0.08 17 to 20 11 to 14 2 to 2.5 Nb
19.12.3 0.08 17 to 20 10 to 14 2.5 to 3.5
19.12.3 L 0.04 17 to 20 10 to 14 2.5 to 3.5
19.12.3 Nb 0.08 17 to 20 10 to 14 2.5 to 3.5 Nb
19.13.4 0.08 17 to 21 11 to 15 3.5 to 5.5
19.13.4 L 0.04 17 to 21 11 to 15 3.5 to 5.5
19.13.4 Nb 0.08 17 to 21 11 to 15 3.5 to 5.5 Nb
22.12 0.15 20 to 23 10 to 13
23.12 015 22 to 26 11 to 15
23.12 L 0.04 22 to 26 11 to 15
23.12 Nb 012 22 to 25 11 to 15 Nb
23.12 W 0.20 22 to 25 11 to 15 W2to4
23.12.2 0.12 22 to 25 11 to 15 2 to 3
16.8.2 0.10 14.5 to 16.5 7.5 to 9.5 lto2
17.8.2 0.10 16.5 to 18.5 8 to 9.5 1.5 to 2.5
18.8 Mn 0.20 17 to 20 7 to 10 Mn5to8
18.15.3 L 0.04 16.5 to 18.5 13 to 16 2.5 to 3.5
25.20 0.20 24 to 28
25.20 0.04 24 to 28 18 to 22
25.20 Nb 0.12 24 to 28 18 to 22 Nb
25.20.2 0.12 25 to 28 20 to 22 2 to 3
25.20 C 0.25/45 24 to 28 18 to 22
25.25.2 Nb 0.10 24 to 27 24 to 26 2 to 2.5 Nb
18.20.2 CuNb 0.10 17 to 20 19 to 22 2 to 2.5 Cu 1.8 to 2.2, Nb
20.25.5 L Cu 0.04 19 to 20 24 to 26 4 to 6 Cu 1 to 3
23.27.3CCuNb 0.04 21 to 25 25 to 29 2.5 to 4.3 Cu 2.5 to 3.5, Nb
25.16 C Mn 0.25/46 23 to 26 14.5 to 17 Mn5to8
20.9 Nb 0.13 18 to 21 8 to 10 0.35 to 0.65 Nb
29.9.3 0.10 18.5 to 21 8 to 10 2 to 4
25.4 0.15 24 to 27 4 to 6
29.9 0.15 28 to 32 8 to 12
18.36 0.25 14 to 49 33 to 36
17.12 Si 0.15 17 to 19 11 to 13 Si 3.8 to 4.8
24.14 Si 0.12 22 to 25 13 to 15 Si 1.5 to 2.2
16.25.6 0.12 14 to 17 23 to 25 5 to 7
expansion requirement of 0.38 mm opposite the notch

Table 3.44: Symbol for current conditions as in the Charpy-V notch test at -196°C as specified in
per ISO 3581 Section 8 of the ASME Boiler and Pressure Vessel Code.
Symbol Direct current Minimum open-circuit Weld deposits of E320, E320LR and E330 can also meet
recommended (AC) voltage (V) this lateral expansion requirement.
polarity The note further points out that E308L, E309L and
0 + E316L deposits can also be used at -196°C, provided
1 + or - 50 carbon content is kept very low, nitrogen content is kept
2 50 below 0.06% and ferrite below 3 FN. Fully austenitic
3 + 50 E316L appears to be the preferred composition. Lime-
4 +or- 70 coated electrodes are known to give superior lateral
5 - 70 expansion values as compared to titania covered
6 + 70 electrodes.
7 + or - 90 BS 2926 is not very explicit on the subject of ferrite.
8 - 90 In a footnote to the chemical composition table, it
9 . + 90 mentions that w h e n a low delta-ferrite content is
required for the classes 19.12.3, 19.12.3 L, 19.12.3 Nb,
19.13.4, 19.13.4 Land 19.13.4 Nb, the maximum nickel
2) Covered electrode for manual electric arc welding,
content may be increased. It also carries the following
having a basic covering and depositing weld metal
clause:
having the following chemical analysis:
"Ferrite content At the time of placing the order, the
C 0.1
purchaser shall indicate whether he requires the delta
Cr 17
ferrite content of the deposited weld metal to be
Ni 9
determined. If so, the method to be used and the
Mn 7
acceptable limits shall be the subject of agreement
It may be used for welding in all positions except
between the purchaser and the manufacturer."
the vertical downward one. To be connected on direct
IS:5206 makes no mention of ferrite. ISO 3581 makes
current welding to the positive pole; on alternating
the following reference:
current welding, requiring an open-circuit voltage of
Normally, austenitic deposited weld metals of the
70 V. Nominal efficiency: 150%. Mild steel core wire.
19.9 types have a ferrite content of 5 to 10%. For special
The codification for the electrode will therefore be
purposes, fully austenitic or low ferrite variants are also
E 18 8 Mn B 150 26 X
developed. Such electrodes could be marked by the
In the Annexure, ISO 3581 gives the comparison of
symbol X.
ISO, AWS, BS and DIN symbolisation systems as shown
in Table 3.45.
SURFACING ELECTRODES
Comparison The application of these electrodes is indicated by
Comparing the four standards on stainless steel the word surfacing, which is defined by the American
electrodes discussed above, it will be observed that AWS Welding Society as, the deposition of filler metal
A5.4 is the most exhaustive and up-to-date. Among (material) on a base metal (substrate) to obtain desired
other things, it explains the significance of ferrite in weld properties or dimensions. Since this broad definition can
deposits and the meaning of ferrite number (FN). It include processes like electroplating and metal spraying,
provides detailed procedure for preparing the weld pad the correct term would be, weld surfacing electrodes,
for the ferrite test. It also indicates the FN levels to be because according to the AWS, weld surfacing means,
expected in the various stainless steel weld deposits. deposition of weld metal, composed of one or more
AWS A5.4 has also given a technical note on stringer or weave beads, on a surface to obtain desired
"Stainless Steel Weld Metals for Cryogenic properties or dimensions. The most sought after
Applications" in the Appendix. It points out that E310 properties are resistance to frictional wear, impact,
deposits, which are ferrite-free have been successfully abrasion, erosion, oxidation and corrosion. Electrodes
used to join 9% Ni steel for service temperature of - designed for such wear-resistant applications are
196°C. These deposits are capable of meeting the lateral commonly referred to as, hardfacing electrodes, in spite
Table 3.45: Comparison of ISO, AWS, BS and DIN symbols

Symbol Corresponding symbolization in national standards
AWS BS DIN
13 410 13 13
13.1
13.4 13.4
17 430 17 17
17.0.1 30
19.9 308 19.9 19.9
19.9 L 308 L 19.9 L 19.9 nC
19.9 Nb ,347 19.9 Nb 19.9 Nb
19.9 L Nb
19.12.2 316
19.12.2 L 316 L
19.12.2 Nb 318
19.12.3 19.12.3 19.12.3
19.12.3 L 19.12.3 L 19.12.3 nC
19.12.3 Nb 19.12.3 Nb 19.12.3 Nb
19.13.4 317 19.13.4 19.13.4
19.13.4 L 19.13.4 L
1913.4 Nb 19.13.4 Nb
22.12 22.12
23.12 309 23.12
23.12 L 23.12 nC
23.12 Nb 309 Cb 23.12 Nb 23.12 Nb
23.12 W 23.12 W
23.12.2 209 23.12.2
16.8.2 16.8.2
17.8.2 17.8.2
18.8 Mn 18.8 Mn 6
18.15.3 L 18.15. nC
25.20 310 25.20 25.20
25.20 L
25.20 Nb 310Cb 25.20 Nb
25.20.2 310 Mo
25.20.C 25.20 H
25.25.2 Nb 25.25.2 Nb
18.20.2 Cu Nb 18.20.2 CuNb
20.25.5 L Cu
23.27.3 Cu Nb
25.16 C Mn
20.9 Nb 349
20.9.3 20.9.3
25.4 25.4
29.9 312
18.36 330 18.36
17.12 Si
24.14 Si
16.25.6
of the fact that hardness is not always a true measure of b) 2.5% carbon
wear resistance. The metallurgy of hardfaced deposits, c) Very high alloy
and processes and techniques of hardfacing are dealt
with in Chapter 11. II. COBALT BASE ALLOYS
It is apparent that surfacing electrodes are meant to A) Low alloy
be used as much in the design of new components (i.e. B) High alloy
using inexpensive base metals or blanks and applying III. CARBIDES
wear-resistant deposits at suitable points or areas) as in A) Inserts
the reclamation of used or worn-out components in B) Composite
order to reduce their wear and tear in service. C) Powder
Obviously, surfacing electrodes cover an extremely
wide range. A fair idea can be gathered from the IV. COPPER BASE
following list prepared by the AWS (ref. AWS A5.13- A) Copper-zinc
80), which, it must be noted, refers to surfacing filler B) Copper-silicon
metals which can be in the form of bare rods for gas C) Copper-aluminium
and TIG welding, and coated electrodes for arc welding.
V. NICKEL BASE
List of Surfacing Filler Metals A) Nickel-copper (Monel)
B) Nickel-chromium (Nichrome)
I. FERROUS C) N i c k e l - c h r o m i u m - t u n g s t e n molybdenum
A) Hardenable alloys (Hastelloy)
1) Carbon steels D) Nickel-chromium-boron
a) Low (0.19% C max.) The Welding Institute of UK has categorised
b) Medium (0.20-0.60% C) surfacing filler metals and also summarised their mair
c) High (0.61-1.50% C) properties as shown below:
2) Low-alloy steels
a) Low carbon Group 1: Steel
b) Medium carbon 1) Carbon steel: H i g h tensile and compressive
c) High carbon strength. High impact strength. Low abrasion resistance.
d) Cast-iron types Machinable. Heat-treatable.
3) Medium alloy steels 2) Low alloy steels: Wide range of properties varying
a) Medium carbon from low abrasion and high impact resistance to high
b) High carbon abrasion and low impact resistance. Heat-treatable for
c) Cast-iron types softening or hardening. Machinable up to 300 HV
4) Medium-high alloy 3) Martensitic chromium steel: Chromium increases
a) Low carbon heat resistance as compared with low-alloy type. Heat-
b) Medium carbon treatable.
c) High carbon 4) High speed steels: Retain hardness and cutting
d) Cast-iron types (1.5% C min.) properties of tools at temperatures of 500-600° C. Lower
carbon types have greater toughness. Heat treatable for
B) Austenitic steels softening or hardening. Machinable when annealed.
1) Chromium and Cr-Ni 5) Austenitic stainless steel: High impact resistance
a) Low carbon and low abrasion resistance. Abrasion resistance in-
b) High carbon, low nickel creased by work hardening. Heat resistant and corro-
c) High carbon, high nickel sion resistant to wide range of media depending on com-
2) High manganese position. Used extensively as buttering layer between
carbon or low-alloy steels and deposit of the austenitic
C) Austenitic—not usually heat-treated manganese type.
1) High chromium iron 6) Austenitic manganese steels: High impact resistance,
2) High-alloy iron greater than that of the austenitic stainless type, and re-
a) 1.7% carbon sistant to crack propagation in service. Work harden.
7) Austenitic chromium-manganese steels: Similar to in corrosion resistant applications. Silicon bronze and
the austenitic manganese type, but can be deposited brasses are used for corrosion resistance only.
directly on carbon or low-alloy steel without an austen-
itic stainless buttering layer. Group 6: Tungsten Carbide
Extremely high wear resistance, where abrasion
Group 2: Irons under high stress occurs, as in rock drill bits and rock
1) Austenitic irons: High abrasion resistance. crushers.
2) Martensitic irons: High abrasion resistance. A close look at the existing standards will provide
3) High-chromium austenitic irons: Very high abrasion thorough understanding of surfacing electrodes. The
resistance. Oxidation resistant. most comprehensive among them is AWS A5.13-2000
4) High-chromium martensitic irons: Very high abra- "Specification for solid surfacing welding rods and
sion resistance. Oxidation resistant. electrodes''. As the title suggests, the standard covers
5) High-chromium complex irons: Very high abrasion rods for use with gas and TIG processes and electrodes
resistance. Oxidation resistant. for use with the manual metal-arc process. The former
carry the prefix JR and the latter the prefix E in their
Group 3: Nickel Alloys classification. We shall restrict our discussion to the
1) Nickel: Welding or surfacing of cast irons also for electrodes.
buttering cast iron. Soft and machinable. The standard covers 44 electrode types divided into
2) Nickel copper: Welding or surfacing of cast irons. six groups. The range is far from exhaustive. This has
Soft and machinable. been explained by stating that, no attempt has been
3) Nickel iron: Welding or surfacing of cast irons, also made to classify all filler metals and that only those
for buttering cast iron. Machinability lower than nickel which have approximate industrial standardisation and
and nickel-copper, but can be improved by use of pre- for which technical data was available have been
heat of 150-300° C. covered. Information on each type made available in
4) Nickel-molybdenum-chromium-tungsten: Oxidation the standard is summarised in the following paragraphs.
and corrosion resistant and resistant to erosion by hot
gases. Iron-Base Electrodes
5) Nickel-chromium-boron: Oxidation resistant. Good EFe1 and EFe2 Electrodes
resistance to corrosion by steam, salt water and salt Characteristics. Deposits made with these electrodes
spray. Low impact resistance. are a machinery grade steel suitable for application on
6) Nickel-molybdenum-iron: Resistant to corrosion by carbon and alloy steels. With care, they can be applied
HCI or salt spray. crack-free. Deposits are machinable with carbide tipped
tools. Deposit hardness generally is in the range of 25 -
Group 4: Cobalt Alloys 50 HRC with EFe2 electrodes providing weld metal with
1) Cobalt-chromium-tungsten low-alloy. Combinatioiy the higher hardness. These deposits contain sufficient
of heat, corrosion and oxidation resistance under im- alloy to attain full hardness without the need of heat
pact stresses, such as occur in exhaust valves. treatment. Abrasion resistance is comparable to heat-
2) Cobalt-chromium-tungsten medium alloy. Higher treated steels of equal hardness.
abrasion resistance than low-alloy type, but IOWCT re- Applications. These electrodes are used for restore
sistance to impact or thermal shock. / worn machinery parts to their original dimensions.
3) Cobalt-chromium-tungsten high alloy. Highest abra- Deposit surfaces are suitable for metal-to-metal rolling
sion resistance and lowest shock resistance of this alloy and sliding contact, such as occurs on large, low speed
group. gear teeth, shafts etc. High compressive strength makes
4) Cobalt-chromium-tungsten nickel: Good corrosion these materials suitable as a base for more abrasion-
resistance and high temperature resistance with excel- resistant materials.
lent thermal shock resistance.
EFe3 Electrodes
Group 5: Copper Alloys Characteristics. Weld metal deposited by these
Copper-aluminium alloys are used for bearing, electrodes is an air-hardening tool steel type with high
corrosion resistant and wear resistant surfaces. Tin room temperature hardness (55-60 HRC). Deposits can
bronzes are used for bearing surfaces and occasionally be applied crack-free with careful procedures. The
deposits cannot be machined and generally are ground Abrasion resistance of the weld deposit is improved
when finishing is required. with some sacrifice in resistance to impact. Deposits air
Applications. EFe3 electrodes are used to overlay harden, and a two-layer deposit can be expected to have
surfaces and edges requiring high hardness and crack- a hardness of 60 HRC or higher. Stress-relief cracks
free deposits, such as the edges of tools and dies. (checks) typically occur through the overlay. Deposits
Deposits are compatible w i t h m a n y tool steels. cannot be machined.
A l t h o u g h generally u s e d for metal-to-metal Applications. EFe7 electrodes are u s e d for
applications, EFe3 weld metal performs well in earth overlaying surfaces that require good low-stress
abrasion applications where high impact is encountered. abrasion resistance. Applications include cement chutes,
fan blades, bulldozer blades, and other parts and
EFe4 Electrodes equipment used for earthmoving or construction.
Characteristics. These electrodes will h a v e a Carbon and alloy steels, tool steels, and stainless steels
graphitic (black) coating and are suitable for application are compatible base metals.
on cast iron. Although the deposited metal is relatively
brittle, crack-free deposits can be made with controlled EFeMn Series Electrodes. (EFeMn-A through
procedures. Deposits can be machined providing they EFeMn-F)
are slow cooled from an annealing temperature. Characteristics. Deposits made with EfeMn series
Applications. EFe4 weld metal is used to rebuild electrodes nominally contain 14% manganese, although
worn cast iron machinery parts subject to metal-to-metal they may vary from 12-21%. This is an amount sufficient
rolling or sliding contact. Although EFe4 weld deposits to yield austenitic weld deposits. Austenite is a
are compatible with carbon and low alloy steel, EFe2 nonmagnetic, tough form of steel. To preserve the
electrodes generally are preferred for such applications. toughness, excessive heat must be avoided during
welding. Stringer beads and a block sequence are
EFe5 Electrodes recommended. The additions of other elements, such
Characteristics. EFe5 electrodes deposit a cold work as 4% nickel, are made to give more stability to the
type of tool steel. Hardness as-deposited should be in astatine; chromium, molybdenum, and vanadium are
the range of 50-55 HRC. Weld metal deposited by EFe5 also addedsingly or in combination of 0.5-8% to increase
electrodes is air-hardening and machinable only after the yield strength. Abrasion resistance is only a little
annealing. better than that of low-carbon steel unless there has been
Applications. Typical applications include those sufficient impact to cause work hardening. As deposited
requiring high compressive strength with moderate surfaces generally are no harder than HRC 20, but can
abrasion and metal-to-metal wear, such as machine work harden to HRC 55. Since deposits are difficult to
components, shafts, and brake drums. machine, grinding is preferred for finishing.
Applications. These electrodes are used for the
EFe6 Electrodes rebuilding, repair, and joining of Hadfield austenitic
Characteristics. Weld metal deposited by EFe6 manganese steel. Ability to absorb high impact makes
electrodes is a high-speed tool steel with a hardness in such deposits ideal for the rebuilding of worn rock
range of 60 HRC or higher. The deposits maintains a crushing equipment and parts subject to impact loading,
high degree of hardness to 1100QF (593QC). Weld metal such as railroad frogs.
deposited by EFe6 electrodes is air-hardening and is
machinable only after annealing. EFeMnCr Electrodes
Applications. Weld deposits may be used for metal- Characteristics. Weld metal deposited by EFeMnCr
to-metal wear applications at temperatures up to 1100 electrodes have characteristics similar to austenitic
F (593 C). Typical applications combine high tempera- manganese deposits. The high chromium content
ture service with severe abrasion and metal-to-metal imparts stainless steel qualities. These deposits cannot
wear and include shear blades, trimming dies, and be flame cut. Although care must be taken in application
punching dies. to avoid hear build-up, deposits are more stable than
FeMn series electrodes.
EFe7 Electrodes Applications. Like EFeMn t y p e electrodes,
Characteristics. EFe7 series electrodes are essentially EFeMnCr electrodes are used for rebuilding, repair and
a higher carbon modification of EFe3 electrodes. joining of equipment m a d e of Hadfield austenitic
manganese steel. EFeMnCr electrodes offer the added EFeCr-A5 Electrodes

advantage of being usable for joining austenitic Characteristics. The weld deposit contains chro-
manganese steel both to itself and to carbon steel. mium carbide in an austenitic matrix. The nonmagnetic
EFeMnCr weld metals often are used as a base for weld metal has fair machinability. Build-up should be
surfacing with EFeCr types for parts subject to both wear restricted to three layers to minimize stress relief check-
and impact. ing.
Applications. Surfaced components frequently are
EFeCr-A1A and EFeCr-A4 Electrodes used for applications involving frictional metal-to-metal
Characteristics. Weld metal deposited by these wear or earth scouring under low stress abrasion.
electrodes will contain massive chromium carbide's in
an austenitic matrix providing excellent wear resistance
EFeCr-A6 and EFeCr-A7 Electrodes
and toughness. Surface checks are typical and give some
degree of stress relief. Deposits cannot be machined Characteristics. These are a higher carbon version
and must be ground when finishing is required. To of EFeCr-A5 electrodes. The deposit contains hexagonal
assure the desired deposit composition, two layers are chromium carbides in an austenitic carbide matrix and
recommended. Additional layers invite spalling and has a hardness of 50-60 HRC. Deposits develop stress-
must be applied with caution. Electrodes are suitable relief checks. The addition of molybdenum increases
for welding on carbon, alloy, and austenitic steels as well wear resistance to high stress abrasion. The weld metal
as cast irons. The weld metal deposited by EFeCr-Al A m a y be a p p l i e d on carbon, alloy, or austenitic
electrodes generally provides greater resistance to manganese steel base metal.
impact but slightly less abrasion resistance than weld Applications. Weld metal is frequently used for
metal deposited by EFeCr-AR electrodes. applications involving low stress abrasive wear
combined with moderate impact.
Applications. Deposits frequently are used to
surface parts and equipment involved in sliding and
crushing of rock, ore, etc; such as bucket lips and teeth, EFeCr-A8 Electrodes
impact hammers, and conveyors. Very low coefficients Characteristics. EFeCr-A8 is a higher chromium
of friction develop as a result of scouring by earth version of EFeCr-A3. The deposit contains hexagonal
products. chromium carbides in an austenitic matrix and has a
hardness of 50-60 HRC. The increased chromium
EFeCr-A2 Electrodes content tends to decrease the toughness while increasing
Characteristics. The weld metal deposit contains the abrasion resistance. Maximum relief checking can
titanium carbide in an austenitic matrix. It is machinable be expected. The weld metal may be applied to carbon,
only by grinding. Build-up should be limited to three alloy, or austenitic manganese base metals.
layers to minimize relief check cracking. Applications. Weld metal is frequently used for
Applications. This weld metal group may be applications involving low stress abrasion combined
applied to both carbon steel and austenitic manganese with minimum impact.
base metal. Deposits frequently are used to hardface
mining, construction, earth moving, and quarrying
equipment subject to abrasion and moderate impact. EFeCr-EX Series Electrodes
Characteristics. This family of electrodes deposits
EFeCr-A3 Electrodes weld metal containing finely dispersed chromium
Characteristics. Filler metal deposited by EfeCr-A3 carbides plus one or more metallic carbides (vanadium,
electrodes is similar to a deposit made using EFeCr-Al A niobium- columbium, tungsten, or titanium). The
electrodes except, due to the lower manganese content, resultant deposits are not machinable, and maintain
a martensitic matrix is present, rendering the deposit their hot hardness and abrasion resistance to 1200QF
somewhat brittle. These deposits are not machinable but (650QC). Deposits stress-relief check readily
may be finished by grinding where necessary. Applications. Equipment subjected to severe high
Applications. This weld metal is a general purpose stress abrasion combined with moderate impact may
hardfacing alloy for earth abrasion applications and is be surfaced with one of the specific grades. Selection of
suitable for low stress scratching abrasion with low the specific grade will be dependent on local service
impact. conditions and the specific application.
Cobalt-Base Surfacing Electrodes While the c o b a l t - c h r o m i u m deposits soften

somewhat at elevated temperatures, they normally are
ECoCr-A Electrodes
considered immune to tempering.
Characteristics. Weld metal deposited by EcoCr-A Applications. Weld metal deposited by ECoCr-C
electrodes is characterized by a hypereutectic structure electrodes is used to build up mixer rotors and items
consisting of network of about 13% eutectic chromium that encounter severe abrasion and low impact.
carbides distributed in a cobalt-chromium-tungsten
solid solution matrix. The result is a material with a ECoCr-E Electrodes
combination of overall resistance to low stress abrasive Characteristics. Welds made using ECoCr-E electro-
wear coupled with the necessary toughness to resist des have very good strength and ductility at tempera-
some degree of impact. Cobalt alloys also are inherently tures upto 1600QF (871QC). Deposits are resistant to ther-
good for resisting metal-to-metal wear, particularly in mal shock, and oxidizing and reducing atmospheres.
high load situations that are prone to galling. The high Early applications of these types of alloys were found in
alloy content of the matrix also affords excellent jet engine components such as turbine blades and vanes.
resistance to corrosion, oxidation, a n d elevated The deposit is a solid-solution-strengthened alloy
temperature retention of hot hardness up to a maximum with a relatively low weight-percent carbide phase in
of 1200QF (650QC). These alloys are not subject to the microstructure. Hence, the alloy is very tough and
allotropic transformation and therefore do not lose their will work harden. Deposits possess excellent self-mated
properties if the base metal subsequently is heat treated. galling resistance and also are very resistant to cavita-
Applications. The alloy is recommended for cases tion erosion.
where wear is accompanied by elevated temperatures Applications. Welds m a d e using ECoCr-E
and where corrosion is involved, or both. Typical electrodes are used where resistance to thermal shock
applications include automotive and fluid flow valves, is important. Typical applications, similar to those of
chain saw guides, hot punches, shear blades, extruder ECoCr-A deposits, include guide rolls, hot extrusion and
screws, etc. forging dies, hot shear blades, tong bits, and valve trim.
Typical hardness values for multilayer welds made
ECoCr-B Electrodes using cobalt base electrodes are :
Characteristics. Weld metal deposited by ECoCr-B ECoCr-A 23-47 HRC
electrodes is similar in composition to ECoCr-A depos- ECoCr-B 34-47 HRC
its except for a slightly higher carbide content (approxi- ECoCr-C 43-58 HRC
mately 16 percent). The alloy also has a slightly higher ECoCr-E 20-32 HRC
hardness coupled with better abrasive and metal-to- Hardness values for single layer deposits will be
metal wear resistance. Impact and corrosion resistance lower because of dilution from the base metal.
are lowered slightly. Deposits can be machined with
carbide tools. Nickel-Base Surfacing Electrodes
Applications. ECoCr-B electrodes are used ENiCr-C Electrodes
interchangeably with ECoCr-A. Choice will depend on Characteristics. Undiluted weld metal of this
the specific application. composition exhibits a structure consisting of chromium
carbides and chromium borides in a nickel-rich matrix.
ECoCr-C Electrodes The nickel base and high chromium content give these
Characteristics. This alloy's deposits have a higher deposits good heat and corrosion resistance. Care should
carbide content (19%) than those made using either be taken when cooling hardfacing deposits because of
ECoCr-A or ECoCr-B electrodes. In fact, the composition a tendency to stress crack. Trîs alloy possesses excellent
is such that primary hypereutectic carbides are found resistance to low stress abrasion/
in the microstructure. This characteristic gives the alloy Applications. ENiC4~C weld metal flows very easily,
higher wear resistance, accompanied by reductions in has very high abrasion resistance, and normally takes
the impact and corrosion resistance. The higher hardness on a high polish. Typical applications include cultivator
also means a greater tendency to stress crack during sweeps, plow shares, extrusion screws, pump sleeves,
cooling. The cracking tendency may be minimized by pistons, and impellers, capstan rings, glass mold faces,
closely monitoring preheating, interpass temperature, centrifuge filters, sucker p u m p rods, etc. The deposits
and postheating techniques. have high corrosion resistance and normally require
grinding for finishing. Single layer deposits typically sheaves, wear plates, dies, etc. These alloys are also used
have a hardness of 35-45 HRC. Multilayer deposits to surface dies that form or draw titanium, low-carbon
typically have a hardness of 49-56 HRC. and stainless steels. These alloys are not recommended
for applications that require resistance to corrosion.
ENiCr-Mo-5A Electrodes The ECuSi electrodes are used primarily for surfac-
Characteristics. Undiluted weld metal deposited by ing corrosion-resistant surfaces. Copper-silicon depos-
ENiCrMo-5A electrodes is a solid-solution-strengthened its generally are not recommended for bearing service.
alloy with relatively low weight-percent carbide phase Copper-tin (ECuSn) electrodes are used primarily
produced through secondary hardening. The resultant to surface bearing surfaces where the lower hardness
deposit is tough and work hardenable. of these alloys is required, for surfacing corrosion-
Deposits have the ability to retain hardness up to resistant surfaces, and, occasionally, for applications
1400QF (760QC). Deposits are machinable with high- requiring wear resistance.
speed tool bits and have excellent resistance to high- Copper-nickel electrodes (ECuNi) are used for
temperature wear and impact. rebuilding 70/30, 80/20, and 90/10% copper-nickel
Applications. These electrodes are used to rebuild alloy or the lead side of copper-nickel clad steel.
and repair hot extrusion dies, hot forging dies, sizing Preheating generally is not necessary.
punches, hot shear blades, guide rolls, tong bits, blast Copper-nickel-aluminum electrodes (ECuNiAl) are
furnace bells, etc. used to rebuild nickel-aluminum-bronze castings or
wrought components. Typical applications are those
ENioCrFeCo Electrodes requiring a high resistance to corrosion, erosion, or
Characteristics. Weld metal deposited by these cavitation in salt or brackish water.
electrodes contain a fairly large volume fraction of ECuMuNiAl electrodes are used to rebuild or
hypereutectic c h r o m i u m carbides d i s t r i b u t e d surface cast manganese-nickel-aluminum bronze
throughout the microstructure. The alloy offers many castings or wrought material. Typical applications
of the same high-performance characteristics of deposits include those requiring excellent resistance to corrosion,
made using ECoCr-C or RNiCr-C electrodes in terms of erosion, and cavitation.
abrasive wear resistance. The reduced nickel or cobalt
content, or both lowers corrosion properties and galling APPLICATIONS
resistance. The high volume fraction of carbides makes Hardness Ranges. See Table 3.46 for typical hardness
this alloy sensitive to cracking during cooling. ranges.
Applications. Welds m a d e using ENiCr-FeCo Hot Hardness. The copper-base alloy filler metals
electrodes are preferred where high abrasion (low are not recommended for use at elevated temperatures.
impact) is a major factor. Typical applications are feed Mechanical properties, especially hardness will tend to
screws, slurry pumps, and mixer components. decrease consistently as the temperature increases
above 400QF (205QC).
Copper-Base Alloy Electrodes
Introduction. The copper-base alloy electrodes Table 3.46: Approximate Weld Deposit Hardness
classified by this specification are used to deposit (SMAW)
overlays and inlays for bearing, corrosion-resistant, or Brinell Hardness
wear-resistant surfaces. AWS classification 3000 kg load 500 kg load
ECuAl-A2 electrodes are used for surfacing bearing EcuAl-A2 130-150
EcuAl-B 140-180
surfaces between the hardness ranges of 130 to 150 HB
EcuAl-C 180-220
as well as corrosion-resistant surfaces. EcuAl-D 230-270
ECuAl-B and E C u A l - C electrodes are used EcuAl-E 280-320
primarily for surfacing bearing surfaces requiring EcuSi - 80-100
hardness in the range of 140-220 HB. These alloys are EcuSn-A - 70-85
not r e c o m m e n d e d for applications that require EcuSn-C - 85-100
resistance to corrosion. EcuNi - 60-80
ECuAl-D and ECuAl-E electrodes are used to EcuNiAl 160-200
surface bearing and wear-resistant surfaces requiring EcuMnNiAl 160-200 -
hardness in the range of 230-320 HB, such as gears, cams, Note: A As-welded condition
Impact. In general, as the a l u m i n i u m content Heat Treatment. Ordinarily, no heat treatment is

increases, impact resistance decreases rapidly. The needed in surfacing with copper-base alloy filler
impact resistance of deposits made by using ECuAl- metals.
A2 electrodes will be the highest of the copper-base alloy Welding Characteristics. To minimize dilution from
classifications. Deposits made using ECuSi electrodes the base metal w h e n surfacing with copper base
have good impact properties. Deposits made using electrodes, the first layer should be deposited using as
ECuSn electrodes have tew impact values. low an amperage as practical. Excessive base metal
Oxidation Resistance. Weld metal deposited by any dilution can result in reduced machinability and service
of the ECuAl family of electrodes forms a protective performance. The manufacturer should be consulted for
oxide coatingf u p o n exposure to the a t m o s p h e r e . specific welding parameters.
Oxidation resistance of the copper-silicon deposit is fair, Preheat. Generally, a preheat is not necessary unless
while that of copper tin deposits is comparable to the the part is exceptionally large; in this case, a 200QF (93eC)
oxidation resistance of pure copper. preheat may be desirable to facilitate the smooth flow
Corrosion Resistance. Several copper-base alloy of the weld metal. At no time should the preheat
filler- metals are used rather extensively to surface areas temperature be above 400QF (205QC) when applying the
subject to corrosion from reducing type acids, mild first layer. On s u b s e q u e n t layers, an i n t e r p a s s
alkalies, and salt water. They should not be used in the temperature of approximately 200 - 600QF (93-316QC) will
presence of oxidizing acids, such as H N 0 3 , or when simplify deposition of the weld metal.
sulphur compounds are present. Filler metals producing
deposits of higher hardness may be used to surface areas Tungsten Carbide Electrodes
subject to corrosive action as well as erosion from liquid Characteristics. Tungsten carbide covered electrodes
flow for such applications as condenser heads and contain 60 percent by weight tungsten carbide granules.
turbine runners. The WCI carbide is a mixture of WC and W2C. The WC2
Abrasion. None of Jfte copper-base alloy deposits is carbide is macrocrystalline WC. Hardness of the matrix
r e c o m m e n d e d for* use w h e r e severe abrasion is of the deposit can be varied from 30 HRC to 60 HRC
encountered in servê. d e p e n d i n g on w e l d i n g technique. H a r d n e s s of
Metal-to-Metal JMear. Copper-aluminum deposits individual carbide particles typically is about 2400
with hardnesses of 130 to approximately 320 HB are HV20. The abrasion resistance of tungsten carbide
used to overlay surfaces subjected to excessive wear deposits is outstanding.
from metal-to-metal contact. For example, ECu-Al-E Applications. Tungsten carbide deposits are applied
electrodes are used to surface dies, and to draw and on surfaces subjected to sliding abrasion combined with
form stainless and carbon steels and aluminum. limited impact. Such applications are encountered in
All of the copper-base alloy filler metals classified earth drilling, digging, and farming. Specific tools that
by this specification are used to deposit overlays and may require this type of a surfacing overlay include oil
inlays for bearing surfaces, with the exception of the drill bits and tool joints, earth handling augers, excavator
CuSi filler metals. Silicon bronzes are considered poor teeth, farm fertilizer applicator knives, and cultivator
bearing alloys. Copper-base alloy filler metals selected shares.
for a bearing surface should produce a deposit of 50-75
HB under that of the mating part. Discontinued Classifications
Mechanical Properties in Compression. Deposits of Some classifications have been discontinued from
the ECuAl filler metals have high elastic limits and one revision of this specification to another. This results
ultimate strengths in compression ranging from 25 00 - either from charges in commercial practice or changes
65 000 psi (172-448 MPa) and 120 000-171 000 psi (827- in the classification system used in the specification. The
1179 MPa), respectively. The elastic limit of ECuSi classifications that have been discontinued are listed in
deposits is around 22 000 psi (152 MPa) with an ultimate Table 3.47, along with the year in which they were last
strength in compression of 60 000 psi (414 MPa). The included in the specification.
ECuSn deposits will have an elastic limit of 11 000 psi
(76 MPa) and an ultimate strength of 32 000 psi (221 Composite Surfacing Electrodes
MPa). Surfacing electrodes are sometimes designed to have
Machinability. All of these copper-base alloy mild steel tubular core wire, filled internally with metal
deposits are machinable. and alloy powders and coated with flux. They are called
composite electrodes. In rare cases, uncoated composite wearing surfaces of mining, quarrying, digging and
electrodes are used for surfacing. earth moving equipment; and a multitude of parts,
AWS A5.21-80 describes seven standard classes of where the roughness of the weld deposit (as it wears) is
composite electrodes: EFe5-A, EFe5-B,EFe5-C,EFeMn- not a handicap, but where the high abrasion resistance
A,EFeMn-B, EFeCr-Al and a tungsten carbide type. The is needed.
first six are equivalents of those h a v i n g similar WC deposit has a hard matrix, formed when the
classification in AWS A5.13-80, which have already been mild steel sheath melts and dissolves high amounts of
dealt with in the preceding section. The tungsten carbide tungsten and carbon. The matrix serves as a compe-
type needs elaboration. tent support for the hard granules, which it anchors in
According to AWS A5.21-80, the tungsten carbide place. This matrix has characteristics that range from
(WC) composite electrode consists of a rimming quality those of air hardening tungsten steels to cast iron struc-
steel tube (0.1 OC/0.45 Mn/0.01 Si) filled with fused tures containing considerable secondary tungsten iron
tungsten carbide granules of controlled mesh size in carbides.
such a manner that they make up 60% of the weight of Because of tungsten, the matrix of WC deposits
the core wire. The chemical requirements of the granules retains its hardness up to 538°C, considerably better than
are 3.6 to 4.2% carbon and 94% minimum tungsten. ordinary hardened steels as shown in Fig. 3.21.
There are five distinct classifications depending on the The deposits are relatively brittle and vulnerable
mesh size of the granules. They are: EWC-12/30, EWC- to sudden impact and tensile stresses. However, they
20/30, EWC-30/40, EWC-40 and EWC-40/120. These have high compression strength. When exposed to
various grades underline the fact that the size (and shape temperatures above 538°C, the WC granules will
also) of the carbide granules has significant influence oxidise to form yellow tungstic oxide. The matrix
on weld deposit properties including surface roughness can rust and corrode like ordinary steel. The deposits
when abraded. are not applicable to metal-to-metal wear conditions.
Weld overlays from WC electrodes offer highest They are unmachinable and have to be finished,
abrasion resistance among all available hardfacing w h e n required, using silicon carbide or diamond
materials. They are typically used to armour the cutting grinding wheels. Their response to heat treatment
teeth and gauge holding surfaces of rock drill bits; the is poor.
Table 3.47: Discontinued Electrode and Rod Classifications, a

Last A5.13(ASTM A 399) LastA5.13(ASTMA399)
AWS classification publication date AWS classificatic publication date
RFeCr-A2 1956 ERCuAl-A3C 1980
EFeCr-A2 1956 RCuAl-Ch 1980
ECuZn-E 1956 RCuAl-Dh 1980
RCuAl-B 1970 RCuAl-Eh 1980
RCuSn-E 1970 ERCuSn-A 1980
ECuSn-E 1970 RCuSn-Dh 1980
RFe5-A 1980 RNiCr-Ah 1980
RFe5-B 1980 RNiCr-Bh 1980
RFeCr-Al 1980 RNiCr-Ch 1980
RCoCr-Ab 1980 EFe5-A 1980
RCoCr-B 1980 EFe5-B 1980
RCoCr-Ch 1980 EFe5-C 1980
RCuZn-E 1980 EfeCr-Al 1980
ERCuSi-Ac 1980 ENiCr-A 1980
ERCuAl-A2c 1980 ENiCr-B 1980
Notes:
a. See A8. discontinued Classifications (in Annex A) for information on discontinued classification.
b. These AWS classifications have been transferred to AWS A5.21.2001 with the revised prefix of "ER" for electrode/rod made from solid
stock or prefix of'ERC"for electrode/rod made from metal or flux cored composite stock.
c. These AWS classification have been transferred to AWS A5.21.2001 without a change in the classification designation for solid bare
electrodes and rods or with the prefix "ERC" for electrodes/rod made from metal offlux cored stock.
IS: 7303-1974 "Specification for covered electrodes for E F e - l C h a s a high p e r c e n t a g e of carbon a n d

surfacing of metal by manual metal arc welding" chromium to give air-hardening martensitic deposit
The Indian standard covers 26 types classified into having excellent resistance to abrasion, friction and
eight groups as shown in Table 3.48. The equivalent AWS moderate impact. The typical deposit chemistry is 0.50
class had been indicated wherever it exists. Weld metal C/0.30 Mn/7.0 Cr/0.70 Mo/0.60 V and hardness is in
chemistry is indicated only for the types which have no the 550-600 Brinell range. Typical applications are metal
AWS equivalents. cutting and forming tools, punches, dies, drilling bits,
The following notes cover the types for which no shears and croppers, oil expellers, crusher hammers,
AWS equivalents exist, the others having been covered mine rails, caterpillar treads, crane wheels, conveyor
earlier. bucket, mixer blades, dipper teeth, cane cutting knives,
etc. The electrode can have a rutile or basic low-
hydrogen type coating, the latter being preferred for
iHlNDICATES LOSS OF HARDNESS DURING 2 MINUTES UNDER LOAD hardenable base metals and where sudden impact
loading may be anticipated.
The above three types together constitute the bulk
of the hardfacing electrode demand in India.
EFe-3C, 4C and 5C are variations of EFe-lC and are
i n t e n d e d for similar a p p l i c a t i o n s . EFe-6C is an
interesting alloy. With 12% Cr content and high carbon
level, the air-hardened deposit is wear-resistant and also
corrosion-resistant to many atmospheric and industrial
process environments. Typical applications are surfacing
of turbine blades and valve seats.
EFeCr-A electrodes are suitable for surfacing appli-
cations on mild steel, carbon steels, low-alloy steels and
austenitic manganese steels, where resistance to the com-
bined effects of abrasion, impact and corrosion is neces-
sary. The weld, as deposited, has a Brinell hardness of
about 250 HB, which rapidly rises to 500 HB under im-
Fig. 3.21: Apparent hot hardness of hard surfacing pact in service. They are used for surfacing application
alloys at 1000°F (540°C) on shovel tracks, coal mining cutters, dipper teeth, rock
crushers, tractor grousers, mill hammers, sand and
EFe-A is designed to give a moderately hard, wear- p u m p impellers, valve seats, etc. EFeCr-B and C are
resistant alloy steel deposit of 250 Brinell, which is easily basically similar, and hence the description of EFeCr-
machinable. Typical deposit chemistry is 0.12 C/0.35 Al under AWS A5.13-70 will more or less apply to both.
Mn/0.6 C r / 1.1 Mo. Typical applications are gears, ECrNi-Mn-A weld deposit is semi-austenitic and
shafts, pinion teeth, pulleys, couplings, spindles and work-hardens under impact. Its high compressive
axles. Rutile coating is generally preferred for this type. strength and its ability to withstand repeated impacts
EFe-B has higher carbon and alloy contents and gives with minimum deformation and no cracking make it
a harder deposit of 350 Brinell. Typical chemistry is 0.20 an excellent build-up material. Weld metal as deposited,
C/0.65 Mn/3.0 Cr. The electrode can have rutile or basic has a hardness of 250 Brinell, which rises to 400 Brinell
low-hydrogen coating, the latter being preferred for with heavy peening. Under repeated heavy impact in
surfacing of hardenable base metals, such as rails. The service, it increases further to 500 Brinell. Typical
weld is machinable w i t h carbide tools. Typical applications are rebuilding of dipper teeth, shovel
applications are conveyor parts, cams, crawler parts, tracks, rock crushers, coal mining cutters, charging rams,
brake shoes, latch bars and plates, drive sprockets and tractor grousers dipper lips, pump housings, conveyor
rails. Indian Railways have been using this type rolls, conveyor buckets, crusher mantles, screw flights,
regularly for resurfacing rail ends. In this application, a truck chains, mill hammers, scrapper blades, dredger
buffer layer from a basic low-hydrogen type mild steel cutter teeth, ploughshares, shovel drive sprockets,
electrode (E7016 class) is provided to reduce the scarifier teeth, pulveriser plows, sand p u m p impellers,
brittleness of the EFe-B deposit. ingot lifting tongs etc.
Table 3.48: Classification of hardfacing electrodes as per IS 7303

IS class Equivalent AWS class Weld metal chemistry*
EFe-A x 0.2 C/3.0 Mn/3.0 Cr/1.5 Mo
EFe-B x 0.2 C/3.0 Mn/5.0 Cr/1.5 Mo
Efe-lC x 0.3-0.6 C/0.6-1 Mn/5-9 Cr/0.5 V/0.5 Mo
EFe-2C EFe5-B
EFe-3C x 0.6-1.0 C/3.0 Mm/5.0 Cr/1.5 Mo
EFe-4C x 0.3-0.6 C/3.0 Mn/3-9 Cr/1.5 Mo
EFe-5C X 0.6-1.5 C/3.0 Mn/3-9 Cr/2.0 Mo
EFe-6C X
0.3-0.7 C/4.0 Mn/9-14 Cr/2.0 Mo
EFeMn-A EFeMn-B but
without Mo
EFeMn-B EFeMn-A
EFeMn-C EFeMn-B
EFeCr-A x 0.15 C/2.0-3.5 Mn/2-4 Ni/16-19 Cr
EFeCr-B EFeCr-Al
EFeCr-C x 3-5 C/1.0 Mn/2.5-4.5 Ni/26-32 Cr
ECrNiMn-A x 0.15 C/2-4 Mn/4-6 Ni/17-19 Cr
ECrNiMn-B X 0.15 C/4-6 Mn/8-11 Ni/18-20 Cr
ECoCr-A ECoCr-A
ECoCr-B ECoCr-B
ECoCr-C ECoCr-C
ENiCr-A ENiCr-A
ENiCr-B ENiCr-B
ENiCr-C ENiCr-C
ECuAl-A ECiAl-A
ECuAl-B ECuAl-D
ECuAl-C ECuAl-E
ECuSn-A ECuSn-A
*Single values are the maximum percentage
ECrNiMn-B austenitic stainless steel deposit has similar to that of ECrNi Mn-B mentioned above.
excellent heat resistance up to 900°C It is also resistant Another interesting development is the incorporation
to corrosion by the atmosphere, sea water and weak of titanium carbide (up to 11% by volume) into the
acids. It is specially suited for making weld joints and deposits of high chromium cast iron for outstanding
repairing cracks in austenitic manganese .steel and for abrasion resistance and improved resistance to impact.
joining austenitic manganese steel to mild steel. Another TiC is one of the hardest carbides that can be
important application is the building up of railway incorporated in manual electrode, and has been claimed
points and crossings. It is also used for depositing buffer to provide exceptionally long wear-life to parts under
layers on hardenable steels before hardfacing. severe abrasion conditions.
Owing to steeply rising cost of cobalt, several cobalt-
Non-Standard Types * base electrodes of ECoCr classification have been
As stated earlier, the range of surfacing electrodes modified with Ni substitution for replacement on some
used in industry is very wide and many of them are not applications. For example, an electrode having deposit
covered by national standards. A few outstanding ones chemistry of 23 Ni/12Co/26Cr/3 Mo/29Fe/3.5 W has
among them need to be described here. been developed, which compares well with the well-
A 14 Mn/14 Cr steel electrode is fairly popular in established cobalt-based electrode having deposit
the U.S.A. as capable of providing deposits having chemistry of 60 Co/28 Cr/W 4 (Stellite 6) in abrasive
maximum service life under conditions of extreme wear and hot hardness. At room temperature, the
abrasion and impact. The weld deposit is austenitic and hardness of the new version is lower, but metal-to-metal
work hardens to 50-55 Re. The range of applications is wear resistance is superior. Typical uses suggested for
the new alloy are fluid valves, extruder screws, mixers also determine volume loss due to wear from the
and bearing surfaces. Similarly, a 50 Co/30 Cr/12 W measured weight loss. The test set-up is explained in
(Stellite 1) is being replaced in many applications, such Fig. 3.22(B).
as high-abrasion extruder screws, rock bits, bearing For the cavitation-erosion test, which is detailed in
surfaces, slurry pumps and low-impact forming dies ASTM G-32, a horn vibrator with a small bolt-like
by an alloy 12 Co/27 C r / 7 Mo/23 Fe/3 W/bal. Ni. sample is immersed in deionised water as shown in Fig.
3.22(C). The ultrasonic horn moves the sample 0.05 m m
Wear Resistance Tests at 20 kHz.
There are no standard tests for evaluating and
comparing the wear-resistant properties of deposits, ELECTRODES FOR CAST IRON
except the conventional hardness test(Brinell, Rockwell,
etc.), but producers of surfacing alloys have devised 1) AWS A5.15-90: "Specification for welding electrodes
their own tests. For metal-to-metal wear resistance test, and rods for cast irons"
a weld specimen is fixed in a jig and rests on a ring of This standard covers gas welding rods as well as
case-hardened 4620 steel. In the test, the wheel spins at MMA welding electrodes for welding cast iron, but we
80 revolutions per minute. shall restrict our discussion to the latter. The standard
When a sliding distance of 220 metres is covered, has four distinct groups of electrodes, depending on the
the test is completed. The difference in weights of the core wire chemistry, and a total of nine types. The
specimen before and after the test is a measure of wear. Appendix describes in detail the characteristics of each
To determine the resistance of weld deposits to electrode type. The following is a summary of the
scratching abrasion, one may use the test described in Appendix:
ASTM G65-81; "Standard Practice for Conducting Dry
Sand/Rubber Wheel Abrasion Tests".The standardised Group 1: Cast Iron
test equipment and materials include the rubber wheel The electrode, designated ECI, uses as core wire the
(a steel disc covered by a layer of chloro-butyl rubber; standard cast iron rod specified for gas welding, whose
Durometer A58-62 h a r d n e s s ) , abrasive ( r o u n d e d composition is 3.5 C/3.0 Sl/0.70 Mn/0.60 P. The heavy
particles of quartz grain sand, AFS 50/70 Test Sand with flux covering gives a light slag, which is easily
less than 0.5% moisture) and specimen size 25.4 m m x removeable. Preheating in the temperature range of 150-
76.2 mm x 12.7 mm. 750°C, depending on the size of the casting, is necessary
After cleaning and weighing the test specimen, it is to ensure a machinable deposit of around 200 Brinell
seated firmly in the holder, while weights are added to hardness, which has a refined close-grained structure.
the lever arm [see Fig. 3.22(A)] which will press the
specimen against the wheel. Next, the revolution Group 2: Steel
counter is set to the number prescribed in the procedure: The electrode, designated ESt, uses the standard core
A for hard materials; B, C and D for softer alloys. The wire of mild steel electrodes, whose chemistry 0.10C/
sand flow, preset to 0.25 to 0.35 kg per minute is then 0.45 Mn/0.03 Si. The coating usually is the basic low-
started, the wheel rotation is actuated, and the lever arm hydrogen type, though the standard does not mention
is lowered to force the specimen against the wheel. After this. The electrode is used without preheat to repair
the wheel has completed the preset n u m b e r of small cracks and pits in castings. Low current is used to
revolutions, the specimen is removed and weighed. The minimise weld dilution; the beads should be short and
loss of weight in grams is converted to volume loss, widely separated to distribute the heat. Each bead
giving the extent of abrasion in cubic millimetres. should be peened lightly. The deposit is mild steel, but
To determine the resistance to metal-to-metal wear, it h a r d e n s from carbon p i c k - u p a n d becomes
the specimen is held in contact with a rotating ring. The unmachinable. In using this electrode for repairing large
test block and ring can be of the same or dissimilar cracks on sizeable castings, it is advisable to employ
materials. Specimens attain sliding velocities up to 22 studs which key the weld to the unaffected base metal
metre/min. Load can vary from 4.5 to 95 kg in 4.5 kg below the fusion zone. Studs are usually 6 to 16 mm in
increments. During testing, a friction-force load cell diameter, projecting 5 to 6.5 m m above the surface to be
connected to a recorder records the force. At the end of welded, and screwed in to a depth, at least equal to their
the test, technicians wipe wear particles from the diameter. The cross-sectibnal area of the studs should
specimens, degrease, and weigh the specimens. One can be 25 to 35% of the area of the weld surface.
Group 3: Copper Alloy

SAND Three types are included here. ECuSn-A and ECuSn-
y
rs2sas^
C have core wires of tin bronze, the former having 4-
6% tin and the latter having 7-9% tin. The third,
designated ECuAl-A2, has aluminium bronze core wire
with Al content of 9-11%. All the three, with the same
designations, are mentioned in AWS A5.6 as electrodes
RUBBER WHEEL
designed for welding copper alloys.
ECuSn types are based on the concept of braze
TEST SPECIMEN welding of cast irons, which employs an oxyacetylene
flame and copper-zinc alloys. If, in arc welding, copper-
zinc alloy is used, zinc would evaporate and cause
"~~|^-—-— F excessive weld porosity. Hence CuSn alloys are
@ preferred. These electrodes can be used at very low
W ^ ^ yr
w
£ * ^ S A N D PARTICLES,
METAL WATER DEBRIS
currents and high travel speeds, resulting in minimum
distortion and least fusion zone cracking due to the fact
that less hard, brittle white cast iron is formed in this
critical area. For best results, preheat and interpass
temperatures of 200°C are recommended. The A-type
I. J
gives a deposit hardness of 70-85 Brinell, and the C-
type 85-100 Brinell. The desposits can provide good
bearing and corrosion resistanct surfaces.
ECuAI-A2 electrode has a relatively low melting
■^TEST SPECIMEN
point and high deposition rate at lower amperage, while
the UTS and YS of the deposit is almost double those of
® the copper-tin deposits. These properties are an
advantage in joining new higher strength cast irons
-TEST RING successfully. The electrode is also well suited for
(B) building on bearing, corrosion-resistant and wear-
resistant surfaces. Preheat temperatures similar to CuSn
electrodes are recommended.
Group 4: Nickel Alloy

ULTRASONIC TRANSDUCER There are four types in this group, and they are
classified on the basis of weld deposit chemistry and
not on the basis of core wire chemistry, as done in the
earlier three groups. The details are:
HORN VIBRATOR
AWS class Weld deposit chemistry
ENi-CI2.0 max. C/85 min. Ni/8.0 max. Fe/2.5 max. Cu
TEST ETWe-CI 2.0 max. C/45-60 Ni/2.5 max. Cu/bal. Fe

ISPECIME! ENiCU-A 0.45 C/ 3-6 Fe/ 50-60 Ni/35-45 Cu
ENiCU-B 0.45 C/ 3-6 Fe/60-70 Ni/ 25-35 Cu
(O These four types are widely used for welding cast

PEtpNigEfr WATER
iron without preheating. Welds are usually machinable.
Fig. 3.22. Sketches of standard wear-resistance tests: (A) ENi-Cl produced with pure nickel core wire is
Scratching abrasion test; (B) Metal-to-metal wear test; (C) widely used for reclamation and repair of castings, as
Cavltation-erosion test. also for joining cast iron to other ferrous and non-ferrous
materials. Satisfactory welds can be produced when the COPPER AND COPPER-ALLOY ELECTRODES
P content of the cast iron is not high. 1) AWS A5.6-84 "Specification for copper and copper-
ENiFe-Cl which has ferro-nickel alloy core wire, has alloy arc-welding electrodes"
the same applications as the above-mentioned type, and This is a comprehensive standard covering nine
is to be preferred for welding nodular cast iron and classes based on the chemistry of undiluted weld metal
castings having phosphorus levels higher than 0.20%. as shown in Table 3.49.
It must also be preferred for making welds on heavy The s t a n d a r d stipulates all-weld tension test,
and highly restrained sections, and on high strength and transverse guided side-bend test and groove usability
engineering grades of cast iron. test. The tensile requirements are given in Table 3.50.
ENi-Cu types have monel core wire, and are used in Hardness test is not specified but the Appendix
m a n y of the same applications as the t w o types provides typical hardness values of the weld deposits
mentioned above. as reproduced in Table 3.51.
The hardness of nickel and nickel alloy deposits The following descriptions and intended uses of the
depends largely on the dilution by base metal, and can various classes have been extracted from the Appendix:
be controlled within reasonable limits by using low
currents and directing the arc on the molten puddle. • ECu: The core wire is deoxidised copper. The
Single layer welds may have a hardness as high as 350 electrode is used for welding all commercial
Brinell but multi-layer welds can be controlled to give grades of copper and for overlaying. It may be
hardness within 175-200 Brinell range. used for clad restoration on copper clad vessels,
The standard does not call for any tests, but only for but care must be taken to minimise dilution.
the chemical analysis of the core wire or the weld deposit Preheats up to 540°C maybe required, depending
as the case may be. The 1956 edition contained a on the mass of the copper material.
usability test, but it was withdrawn from the present • ECuSi: This electrode has already been described
edition, because the test was not searching enough and as a surfacing electrode under AWS A5.13. It is
created a false impression regarding the merits of the used primarily for welding brasses and copper,
various filler metals. A new test is under consideration and occasionally for joining of copper with
by the AWS-AFS Committee on Welding Iron Castings. dissimilar metals including ferrous alloys and for
the fabrication of silicon bronze.
2) IS: 5511-1991 "Specification for covered electrodes for • ECuSn-A & C: These two classes also have been
manual metal arc welding of cast iron" discussed as surfacing electrodes under AWS
This Indian standard has the same range as the AWS A5.13, and as electrodes for welding cast iron
standard, except that there are two versions of the cast under AWS A5.15. They are primarily meant for
iron type, designated Fe C-l and Fe C-2. The first one welding phosphor bronze, but are also useful for
has the cast iron core similar to that of ECI of the AWS welding brasses and for joining brasses to car-
specification, while the latter has a mild steel core-wire bon steel and cast iron. The weld metal has a ten-
and a coating rich in graphite and ferro-silicon, so that dency to flow sluggishly and hence preheat and
the weld deposit is nearly cast iron, but with lower levels interpass temperature of 200°C is recommended
of carbon and silicon. on heavy sections. Grade C is superior to grade
All the types are classified on the basis of core-wire A in respect of deposit hardness and tensile and
composition. The standard also provides a complete yield strengths.
coding system, which starts with a prefix letter M (to
denote metal-arc welding electrode), followed by three • ECuNi: This electrode is primarily meant for
digits. The first digit specifies the type of covering, the welding 70/30, 80/20 and 90/10 copper-nickel
second digit the welding positions and the third digit alloys in wrought or cast form, and are also use-
the welding current conditions. Four types of coverings ful for welding clad side of CuNi clad steel and
are specified: for welding nickel or NiCu alloys.
(1) basic lime-fluorspar type, (2) graphite type, (3) • ECuAl-A2 (iron bearing): This electrode is men-
alkali and alkaline earth chlorides and fluorides, (4) any tioned as a surfacing electrode under AWS A5.13
other type. and as an electrode for welding cast iron under
(1) is meant for steel electrode; (2) is for cast iron AWS A5.15. It is designed primarily for joining
types and nickel alloys; (3) is for copper alloys. aluminium bronzes of similar composition, high
Table 3.49 Chemical composition requirements for undiluted metal
Composition, weight, percent ab
AWS UNS Common Cu Zn Sn Mn Fe Si Ni d P AI Ph TI Total
classification number 0 name including other
Ag elements
Ecu W60189 Copper Remainder f f 0.10 0.20 0.10 f f 0.10 0.02 0.50
ECuSi W60656 Silicon Remainder f 1.5 1.5 0.50 2.4 f f 0.10 0.02 0.50
bronze to
(copper 4.0
silicon)
EcuSn-A W60518 Phosphor Remainder f 4.0-6.0 f 0.25 f f 0.05 0.01 0.02 0.50
bronze to 0.35
(copper-tin)
EcuSn-C W60521 .... do — Remainder f 7.0-9.0 f 0.25 f f 0.05 0.01 0.02 0.50
to 0.35
ECuNi e W60715 Copper Remainder f f 1.00- 0.40- 0.50 29.0 0.020 - 0.02 0.50 0.50
nickel 2.50 0.75 -33.0
(70/30)
EcuAl-A2 W60614 Aluminium Remainder f f f 0.50- 1.5 f - 6.5 to 0.02 0.50
bronze 5.0 9.0
EcuAl-B W60619 Aluminium Remainder f f f 2.5- 1.5 f - 7.5to 0.02 0.50
bronze 5.0 10.0
ECuiNiAl W60632 Nickel Remainder f f 0.50- 3.0- 1.5 4.0- - 6.0 to 0.02 0.50
Aluminium 3.5 6.0 6.0 8.5
bronze
ECuMnNiAl W60633 Manganese Remainder f f 11.0- 2.0- 1.5 1.0- 5.0 to 0.02 0.50
Nickel 13.0 6.0 2.5 75
Aluminium
bronze
Notes:
a. Analysis shall be made for the elements fir which specific values or an "f are shown in this table. If however, the presence of other elements is indicated in the course of routine analysis,
further analysis, shall be made to determine that the total of these other elements is not present in excess of the limited for total other elements in the last column in the table.
b. Single values shown are maximum.
c. SAE/ASTM Unified Number System for Metals and Alloys.
d. Includes cobalt.
e. Sulphur shall be restricted to 0.015 percent for the EcuNi classification.
f. Those elements must be included in total of other elements.
based on the aforementioned AWS standard and in-

Table 3.50 Mechanical property requirements
cludes the following classifications: ECu, ECuSi, ECuSn-
AWS Tensile strength, min Elongation in 4 x D A, ECuSn-B, ECuNi, ECuAI-Al, E C u A l - A 2 and
classification ksi MPa gauge length, Percent, min ECuAl-B. Under ECuAl, it covers three types, namely
Ecu 25 170 20 Al (iron free), A2 and B with Al contents of 8-9%, 9-
EcuSi 50 350 20 11% and 11-12% respectively.
EcuSn-A 35 240 20 Al-type is meant primarily for the fabrication of an-
EcuSn-C 40 280 20
nealed aluminium-bronze plate, sheet, and strip and for
EcuNi 50 350 20
the repair of castings having a similar compostion. It
EcuAI-A2 60 410 20
does not include ECuNiAl and ECuMnNiAI. The stan-
EcuAI-B 65 450 10
EcuNiAI 72 500 10 dard does not specify all-weld metal tension test or us-
ECuMnNiAI 75 520 15 ability test. The only tests called for are chemical analy-
sis, transverse tension test and transverse guided bend
test (root and face).
Table 3.51: Hardness range of weld deposits
as per AWS A5.6 NICKEL AND NICKEL ALLOY ELECTRODES
AWS classification Brinell hardness AWS A5.11-97 "Specification for nickel and nickel alloy
ECu 20 to 40 Rockwell F covered welding electrodes"
ECuSi 80 to 100 (500 Kg load) This is a comprehensive standard covering six
ECuSn-A 70 to 75 (500 Kg load) groups of electrodes, namely, ENi (pure nickel), ENiCu
ECuSn-C 85 to 100 (500 Kg load) (Monel), ENiCrFe (Inconel), ENiMo (Hastelloy) and
ECuNi 60 to 80 (500 Kg load) ENiCrMo (Hastelloy with high Cr) and EniCrCoMo
(Hastelloy with high Cr and Co). Details are given in
ECA1-A2 130 to 150 (3,000 Kg load)
Table 3.52.
ECuAl-B 140 to 180 (3,000 Kg load)
ENi-1 is a pure nickel electrode with approximately
ECuNiAl 160 to 200 (3,000 Kg load)
2.5% Ti addition to deoxidise the weld metal and
ECuMnNiAI 160 to 200 (3,000 Kg load)
eliminate porosity. It is primarily designed to weld pure
nickel used in food-processing equipment, caustic
strength copper-zinc alloys, silicon bronzes, man- handling equipment and piping, chemical containers
ganese bronzes, some nickel alloys, many ferrous for shipping, and electrical parts. Nickel is highly
metals and alloys, and combinations of dissimi- corrosion-resistant to caustic alkalis, dry halogen gases,
lar metals. halogen hydrides and organic compounds generally. It
ECuAl-B: This type also is mentioned as a sur- is also used for joining nickel to steel and surfacing steel
facing electrode under AWS A5.13. Its deposit has with nickel. In dissimilar metal application, dilution of
weld metal with steel must be kept minimum by
higher hardness and tensile and yield strengths
directing the arc on the weld puddle or towards the
than A2 type. It is mainly used for repairing cast-
nickel-base metal rather than the steel member. The
ings besides being used for obtaining wear-resis-
electrode is used with DC+. Electrodes of 3.2 mm and
tant and corrosion-resistant bearing surfaces.
smaller sizes can be used in all positions. Higher sizes
ECuNiAl: It is used for joining or repair welding are suitable only for flat and horizontal positions
of cast or wrought nickel-aluminium bronze ma- (Typical: Huntington Alloys' Nickel 141).
terials. It is also used as surfacing electrode for
ENiCu-7 is a monel electrode designed to weld all
high resistance to corrosion, erosion and cavita-
grades of NiCu alloys widely used for chemical plant
tion from salt brackish waters.
and in engineering generally, where a combination of
ECuMnNiAI: It is used for joining or repair weld- corrosion-resistance, good strength and toughness is
ing of cast or wrought manganese-nickel-alu- required. The alloys have excellent resistance to sea or
minium bronze materials. It is also used as a sur- brackish water, chlorinated solvents, glass-etching
facing electrode for excellent resistance to corro- agents, sulphuric and many other acids, and alkalis.
sion, erosion and cavitation. They are also well suited to join NiCu alloys to steel, to
3) IS: 8666-1977 "Specification for copper and copper- weld the clad side of NiCu alloy clad steel and to surface
alloy covered electrodes for manual metal-arc welding" is steel with monel. For these applications, the electrode
3.52.A: All-weld metal tension test requirements DC electrode + (Typical: Huntington Alloys' Inconel
132).
Tensile Strength, min ENiCrFe-2 is primarily designed for welding of
AWS ksi MPa Elongation* dissimilar alloys, such as austenitic and ferritic steels to
Classification percent, min each other or to high-nickel alloys. It is also used for
welding Incoloy 800 to itself. Incoloy 800 is used
ENi-I 60 410 20
extensively for high-temperature applications, because
ENiCu-7 70 480 30
ENiCrFe-1 80 550 30 of its strength and its excellent oxidation and
ENiCrFe-2 80 550 30 carburisation resistance. For example, it is used for the
ENiCrFe-3 80 550 30 pyrolysis of hydrocarbons, as in cracking or reforming
ENiCrFe-7 80 550 30 operations in the petroleum industry. DC electrode +
ENiCrFe-4 95 650 20 (Typical: Huntington Alloys' Incoweld A).
EniCrFe-9 95 650 25 ENiCrFe-3 is a high-Mn version of ENiCrFe-2,
EniCrFe-10 95 650 25 specifically meant for use in nuclear applications. The
EniMo-1 100 690 25
high Mn content lowers corrosion resistance of the
EniMo-3 100 690 25
deposit in environments other than hot water, but
EniMo-7 100 690 25
EniMo-10 100 690 25 imparts cracking resistance to heavy, highly restrained
EniMo-8 95 650 25 welds. It is specially designed for overlaying of Inconel
EniMo-9 95 650 25 deposit onto steel to nuclear energy standards, with
EniCrCoMo-1 90 620 25 particular reference to radiographic quality. It is also
EniCrMo-1 90 620 20 used for welding Inconel 600 to itself or to carbon or
EniCrMo-2 95 650 20 stainless steels, and for welding the clad side of Inconel-
EniCrMo-3 110 760 30 clad steel. DC electrode + (Typical: Huntington Alloys'
EniCrMo-4 100 690 25 Inconel 182).
EniCrMo-5 100 690 25
25
ENiCrFe-4 is an AC/DC electrode with higher yield
EniCrMo-7 100 690
EniCrMo-10 100 690 25 and tensile strengths than the earlier classes, and is
EniCrMo-13 100 690 25 meant for welding 9% nickel steel. (Typical: Huntington
EniCrMo-6 90 620 35 Alloys' Incoweld B.)
EniCrMo-9 90 620 25 It should be noted that all ENiCrFe classes contain
EniCrMo-11 85 585 25 small amounts of niobium (Nb) in the weld deposit,
EniCrMo-12 95 650 35 which are added to control the harmful effects of sulphur
EniCrMo-14 100 690 30 on ductility.
*The elongation the guage length equal to
shall bedetermined from ENiCrFe-2 and ENiCrFe-3 are ideal electrodes for
four times the guage diameter. welding dissimilar alloys (as compared to 25/20,25/12
stainless steels), because with their high Ni content, the
weld deposit solid solution is capable of accepting heavy
is suitable in the as-welded and stress-relieved dilution of iron and chromium without the danger of
conditions. Welding positions, suitable for ENi-1 apply duplex structures being formed. Also, elements such as
to this electrode also. A well designed electrode can meet Cu, Ni, Mn, Mo, Co, C, Si, V, etc., are taken into solution
stringent radiographic requirements in all positions. DC without the danger of separate phases being formed.
electrode + (Typical: Huntington Alloys' Monel 190). Thus the diluted weld metal has excellent stability at
ENiCrFe-1 is meant for welding Inconel alloy 600 to all temperatures. The weld metal is extremely resistant
itself. This alloy resists oxidation at elevated to "hot shortness", and does not suffer from carbon
temperatures and has good high-temperature migration effects. Moreover, because Inconel type weld
mechanical properties. It also has useful resistance to metal has coefficients of thermal expansion intermediate
corrosion by many chemical products, especially food between those of austenitic stainless steels and ferritic
acids. For these reasons it is widely used for heat- steels, minimum thermal stresses are generated in the
treatment and furnace equipment as well as for chemical joints in elevated temperature service.
and food-processing plant. Because of its resistance to ENiMo-1 and ENiMo-2 are meant for the welding
Cl-ion, stress-corrosion cracking and corrosion by high of NiMo alloys, among which the best known are the
purity water, alloy 600 is often used in nuclear reactors. so-called Hastelloy alloys. Hastelloy-B is typical of this
Table 3.52: Chemical composition requirements for undiluted weld metal
Weight percent k
Nb(cb) Other
AWS UNS C Mn Fe Si Cu Ni d Co AI Ti Cr plus Mo W Elements
classification number 0 Ta Total
ENi-1 W82141 0.10 0.75 0.75 0.03 0.02 1.25 0.25 92.0 ~ 1.0 1.0 0.50
min to 4
ENiCu-7 W84190 0.15 4.0 2.5 0.02 0.015 1.5 Rem 62 - 0.75 1.0 0.50
to 69
ENiCrFe-1 W86132 0.08 3.5 11.0 0.03 0.015 0.75 0.50 62 - 13 1.5- - 0.50
min to 17 to4
ENiCrFe-2 W86133 0.10 1.0 12.0 0.03 0.02 0.75 0.50 62 (e) 13 0.5 0.5 0.50
to 3.5 min to 17 to 3 to 2.5
ENiCrFe-3 W86182 0.10 5.0 10.0 0.03 0.015 1.0 0.50 59 (e) 1.0 13 1 - 0.50
to9.5 min to 17 to 2.5
ENiCrFe-4 W86134 0.20 1.0 12.0 0.03 0.02 1.0 0.50 60 - 13 1.0 1.0 0.50
to 3.5 min to 17 to 3.5 to 3.5
ENiCrFe-7 W86152 0.05 5.0 7.0 0.03 0.015 0.75 0.50 Rem (e) 0.50 0.50 28.0 1.0 0.5 0.50
tol2.0 to31.5 to2.5
ENiCrFe-9 W86094 0.15 1.0 12.0 0.02 0.015 0.75 0.50 55.0 - - - 12.0 0.5 2.5 1.5 0.50
to4.5 min tol7.0 to3.0 to 5.5
ENiCrFe-10 W86095 0.20 1.5 12.0 0.02 0.015 0.75 0.50 55.0 - - - 13.0 1.0 1.0 1.5 0.50
to3.5 min tol7.0 to3.5 to 3.5 to3.5
EniMo-1 W80001 0.07 1.0 4.0 0.04 0.03 1.0 0.50 Rem 2.5 - - 1.0 26.0 0.60
to 7.0 to 30.0 1.0 0.50
EniMo-3 W80004 0.12 1.0 4.0 0.04 0.03 1.0 0.50 Rem 2.5 - - 2.5 23.0 0.60
to 7.0 to 5.5 to 27.0 1.0 0.50
EniMo-7 W80665 0.02 1.75 2.25 0.04 0.03 0.2 0.50 Rem 1.0 - - 1.0 26.0 -
to 30.0 1.0 0.50
EniMo-8 W80008 0.10 1.5 10.0 0.02 0.015 0.75 0.50 60.0 - - - 0.5 17.0 -
min to 3.5 to 20.0 2.0 0.50
18.0 - to 4.0
EniMo-9 W80009 0.10 1.5 7.0 0.02 0.015 0.75 0.3 62.0 -
2.0 0.50
to 1.3 min to 22.0
to 4.0
EniMo-10 W80675 0.02 2.0 1.0 0.04 0.03 0.2 0.50 Rem 3.0 1.0 27.0 -
3.0 0.50
to 3.0 to 3.0 to 32.0
EniCrCoMol W86117 0.05 0.3 5.0 0.03 0.015 0.75 0.50 Rem 9.0 21.0 1.0 8.0 - 0.50
to 0.15 to 2.5 to 15.0 to 26.0 to 10.0
EniCrMo-1 W86007 0.05 1.0 18.0 0.04 0.03 1.0 1.5 Rem 2.5 21.0 1.75 5.5 1.0 0.50
to 2.0 to 21.0 to 2.5 to 23.5 to 2.50 to 7.5
EniCrMo-2 W86002 0.05 1.0 17.0 0.04 0.03 1.0 0.50 Rem 0.50to 20.5 - 8.0 0.2 0.50
to 0.15 to 20.0 to 2.50 to 23.0 to 10.0 to 1.0
EniCrMo-3 W86112 0.10 1.0 7.0 0.03 0.02 0.75 0.50 55.0 (e) 20.0 3.15 8.0 - 0.50
min to 23.0 to 4.15 to 10.0
EniCrMo-4 W80276 0.02 1.0 4.0 0.04 0.03 0.2 0.50 Rem 2.5 14.5 - 15.0 0.35 3.0 0.50
to 7.0 to 16.5 to 17.0 to 4.5
ENiCrMo-5 W80002 0.10 1.0 4.0 0.04 0.03 1.0 0.50 Rem 2.5 14.5 - 15.0 0.35 3.0 0.50
to 7.0 to 16.5 to 17.0 to 4.5
ENiCrMo-6 W86620 0.10 2.0 10.0 0.03 0.02 1.0 0.50 55.0 - 12.0 0.5 5.0 1.0 0.50
to 4.0 min to 17.0 to 2.0 to 9.0 to 2.0
ENiCrMo-7 W86455 0.015 1.5 3.0 0.04 0.03 0.2 0.50 Rem 2.0 0.70 14.0 - 14. 0.5 0.50
to 18.0 to 17.0
ENiCrMo-9 W86985 0.02 1.0 18.0 0.04 0.03 1.0 1.5 Rem 5.0 21.0 0.5 6.0 1.5 0.50
to 21.0 to 2.5 to 23.5 to 8.0
ENiCrMo-10 W86022 0.02 1.0 2.0 0.03 0.015 0.2 0.50 Rem 2.5 20.0 - 12.5 0.35 2.5 0.50
to 6.0 to 22.5 to 14.5 to 3.5
ENiCrMo-11 W86030 0.03 1.5 13.0 0.04 0.02 1.0 1.0 Rem 5.0 28.0 0.3 4.0 1.5 0.50
to 17.0 to 2.4 to 31.5 to 1.5 to 6.0 to 4.0
ENiCrMo-12 W86032g 0.03 2.2 5.0 0.03 0.02 0.7 0.50 Rem - 20.5 1.0 8.8 - 0.50
to 22.5 to 2.8 to 10.0
ENiCrMo-13 W86059 0.02 1.0 1.5 0.015 0.01 0.2 Rem 22.0 - 15.0 - - 0.5
to 24.0 to 16.5
ENiCrMo-14 W86026 0.02 1.0 5.0 0.02 0.02 0.25 0.50 Rem 0.25 19.0 - 15.0 - 3.0 0.50
to 23.0 to 17.0 4.4
Notes:
(a) The weld metal shall be analyzed for the specific elements for which values are shown in this table. If the presence of other elements is indicated in the course of the work, the amount of those
elements, shall be determined to ensure tliat their total does not exceed the limit specified for "other elements total" in the last column of the table, (b) Single values are maximum, except where
otherwise specified. .Rem - remainder, (c) ASTM/SAE Unified Numbering System for Metals and Alloys, (d) Including incidental cobalt. Rem = remainder, (e) Cobalt—0.22 Maximum, when
specified by the purchaser, (ft Tantalum—0.30 maximum when specified by the purchaser (g) UNS number formerly was W86040.
group (0.05 C/61 Ni/2.5 Co/28 M o / 1 C r / 5 Fe), which special applications, but they are outside the scope of
is primarily used for its excellent corrosion resistance, the standard.
especially to hydrochloric and other acids. ENiMo-1 is Ni and Ni-alloy electrodes call for special
designed for the welding of Hastelloy-B, but is also used consideration in use. Ni-alloy weld metal has less
for welding the clad side of Hastelloy-clad steel and for fluidity and does not spread. Hence, it must be placed
welding Hastelloy alloys td other metals. ENiMo-3 is where required and weaving must not exceed three
designed to match Hastelloy alloy W and is also used times the electrode diameter. Wider beads may expose
for welding of dissimilar metal combinations of Ni-base, the molten metal to the atmosphere. Normally lower
Co-base and Fe-base alloys. Both the classes are currents are necessary. Use of excessive current to
normally used in the flat position. promote fluidity may overheat the electrodes and result
ENiCrMo group of electrodes is used for the welding in poor arc stability and loss of protection of the weld
of NiCrMo alloys, represented by Hastelloy alloys C, F, puddle towards the end of the weld run, due to spalling
G and X. Alloy C has good corrosion resistance and high- of the coating. It is advisable to bake the electrode in an
temperature properties. Alloy C-276 is similar to alloy oven before use,
C, but has lower carbon and silicon contents to reduce
the formation of grain-boundary precipitates and enable ALUMINIUM AND AL-AUOY ELECTRODES
the alloy to be used in the as-welded condition. Alloy F AWS A5.3-99 deals with Al and Al-alloy electrodes.
resists corrosion by both oxidising and reducing agents. It covers three classes of electrodes as shown in Table
Alloy G has excellent resistance to hot sulphuric, 3.53.
phosphoric and nitric acids. Alloy X has high strength The standard stipulates chemical analysis of core
and oxidation resistance at temperatures up to 1,200°C. wire, reduced section tension test and transverse guided
ENiCrMo-1,-2 and -5 are also used for welding the root- and face-bend tests.
clad side of NiCrMo clad steel and for welding the alloy The covering consists mainly of alkaline chlorides
to other metals. Classes 1 and 2 are designed to weld and fluorides, which are hygroscopic. Absorption of
Hastelloy G and X respectively. Class 3 has similar moisture from the air can be quite rapid and the
applications but is better suited for surfacing steel with deterioration of the covering can occur alter only a few
NiCrMo alloy. It may also be used for welding dissimilar hours exposure to atmospheric humidity. For this
combinations of iron- and nickel-base alloys. Class 4 is reason, original cartons must be stored in a dry, clean
the extra-low carbon variety meant to be used for room. Electrodes in an open carton or exposed otherwise
welding low-carbon NiCrMo alloys, such as Hastelloy must be reconditioned by heating in an oven at 175-
alloy C-276, to themselves and to other metals and clad 200° C for one hour prior to use. Even traces of moisture
side of low-carbon NiCrMo clad steel. Class 6 is intended in the coating or on the base plate can cause weld
for welding cryogenic grades of steel but can also be porosity. Dirt, grease or other contamination on the
used in other applications. electrode or joint surface can also contribute to porosity.
Electrodes with ENiCrCoMo-1 classification are The slag formed over the weld bead is not easily
used for w e l d i n g Nickel-Chromium-Cobalt- removable. Though a major portion can be removed by
Molybdenum alloys to themselves and to steel and for mechanical means, such as a rotary wire brush or slag
surfacing steel w i t h N i c k e l - C h r o m i u m - C o b a l t - hammer, the final traces can only be removed by
Molybdenum weld metal. The electrodes are also used steaming or hot water rinse. Complete removal of the
for applications where optimum strength and oxidation slag is absolutely necessary, because it is corrosive and
resistance is required above 820°C u p t o 1150°C, can cause pittings in the joint area. Multiple-pass
especially when welding on base metals of Ni-Fe-Cr welding should therefore be avoided as far as possible.
alloys. It is usually advisable to preheat the joint area to 150-
The AWS standard stipulates the following tests: 200° C. Since aluminium has high heat conductivity,
chemical analysis, all-weld tension test, transverse preheating helps to maintain the fluidity of the weld
guided bend test and radiography test (termed usability p u d d l e a n d to avoid porosity d u e to too r a p i d
test). Surfacing weld bend test is demanded for ENi-1, solidification of the weld. Preheating also reduces
ENiCrFe-2, ENiFeCr-3, ENiFeCr-4, ENiCrMo-3 and distortion. One difficulty encountered in metal arc
E n i C r M o - 6 . A d d i t i o n a l tests such as corrosion welding, caused by interruption of the arc, is the
resistance, scaling resistance, or strength at elevated or formation of a fused flux coating over the end of the
sub-zero temperatures may be required for certain electrode. Re-establishing a satisfactory arc is impossible
Table: 3.53 Chemical composition requirements for core wire

ab
Weight ]Percent
AWS UNS Si Fe Cu Mn Mg Zn Ti Be Other Elements Al

classification designation 0 Each Total
EllOO A91100 (d) (d) 0.05-.0 20 0.05 0.10 0.0008 0.05 0.15 99.00 min e
E3003 A93003 0.6 0.7 0.05-0.20 1.0-1.5 0.10 0.0008 0.05 0.15 Reminder
E4043 A94043 4.5-6.0 0.8 0.30 0.05 0.05 0.10 0.20 0.0008 0.05 0.15 Reminder
Notes:
a) The core wire, or the stock from which it is made, shall be analyzed for the specific elements for which values are shown in this table. If the presence
of other elements is indicated in the course of work, the amount of those elements shall be determined to ensure that they do not exceed the limits specified for
"Other Elements." b) Single values are maximum, except where otherwise specified, c) SAE/ASTM Unified Numbering System for Metals and Alloys, d)
Silicon plus iron shall not exceed 0.95 percent, e) The aluminium content for unalloyed aluminium is the difference 100.00 percent and the sum of all other
metallic elements present in amounts of 0.010 percent or more each, expressed to the second decimal before determining the sum.
during the welding operation. Hence an electrode needs

Table 3.53 A tension test requirements
to be handled and operated with care.
Tensile strength, min. £ Bending the electrode and hitting its striking end
AWS Psi MPa hard against the base plate to strike the arc, must be
classification discouraged. These actions may cause the flux to peel
EllOO 12 000 80 off, more so in the case of basic low-hydrogen types,
E3003 14 000 95 w h i c h are b a k e d at high t e m p e r a t u r e s d u r i n g
E4043 14 000 95 m a n u f a c t u r e . Use of currents higher than those
recommended by the manufacturer will overheat the
Note: a. Fracture may occur either the base metal or the weld metal
electrodes and cause the coating to decompose or
until this formation is r e m o v e d ; this is usually disintegrate. Cellulosic electrodes (E6010, E6011, E6012)
accomplished by striking the electrode against the work are markedly sensitive to overheating. The coating gets
or any other surface. charred and in this condition it cannot provide adequate
El 100 class electrodes produce weld metal of high gaseous shielding; this results in porosity, reduced
ductility, good conductivity and slightly lower tensile penetration, higher nitrogen contents in the weld metal
strength compared to the other two. It is used for and reduced ductility and toughness. When very high
welding pure aluminium. currents are used, welders are compelled to discard
E3003 class electrodes also produce weld metal of longer lengths of stubs, thereby further increasing
high ductility and higher tensile strength than El 100. electrode cost. Contamination of electrodes by oil, grease
They are used to weld aluminium alloys 1100 and 3003. and shop-floor dirt must be avoided, as they promote
E4043 class electrodes p r o d u c e weld metal weld metal porosity and cracking.
containing approximately 5% Si, which provides Special care is necessary in the storage of electrodes.
superior fluidity. They are, therefore, preferred for Sometimes they need to be re-dried or re-baked in an
general purpose welding. They can be used to weld the oven before use, according to the recommendations of
6XXX series Al alloys, the 5XXX series Al alloys (up to the electrode producer. Moisture and water contents of
2.5% Mg content), and Al-Si alloy castings, as well as Al electrode coatings have an important bearing on
alloys 1100,1350 (EC), and 3003. With the advent of gas- performance and weld metal quality, and the electrode
shielded arc welding processes (TIG and MIG), use of producer's recommendations are intended to ensure
aluminium MMA electrodes is restricted to minor that these are maintained at optimum levels. It is
structural and repair work. i m p o r t a n t , therefore, to k n o w the desirable and
undesirable effects of water in coatings, and the
HANDLING AND STORAGE OF ELECTRODES influence of storage under different climatic conditions.
Weight for weight, an electrode is several times more All classes of electrodes except the basic low-
expensive than the corresponding alloy in wrought hydrogen types, as a rule, carry varying amounts of
form. Its cost rises further due to stub and spatter loss water in the flux coating. A part of the water may be
present in a chemically- combined form (as water of In organic (or cellulosic) types of coating, therefore,
crystallisation) with certain coating ingredients such as a moderate moisture content is held to be beneficial. If
clay, which cannot be driven out by the normal drying these coatings are baked too dry, the weld is very likely
procedures used by the electrode maker. The rest is to contain pits and holes, especially along the edge of
present as hygroscopic water, partly retained by the the weld. If these are not removed by chipping, they
alkaline silicate used as the binder, and partly as free often result in weld defects which are readily observed
moisture, which has originated from the atmosphere in radiographs. According to one investigator, this
and settled into the pores of the coating. Hygroscopic difficulty can be prevented by re-humidifying, or in
water can be driven out mostly by heating the electrode some cases by immersing the electrodes in water for a
at 150-450° C for half an hour, while chemically few minutes. However, porosity also occurs in the weld
combined water can only be released by heating to if the coating absorbs too much moisture as in storage
1,000-1,200° C. The chemically-combined water is in a damp place. Although such welds may appear
constant for a coating, while the hygroscopic water perfect on the surface, radiographs w o u l d reveal
varies as the relative humidity of the atmosphere in the porosity in the weld deposits. The difficulty can be
storage area. overcome by re-drying the electrodes according to the
The electrode designer deliberately introduces a manufacturer's recommendations.
controlled quantity of water in the flux coating partly The above observations illustrate the fact that for best
through certain hydroxides and silicates, which contain performance and weld metal quality, the electrode
"water of crystallisation", and partly as moisture should neither be too dry nor too moist. Cellulosic type
physically bonded with the silicate binder, which is electrodes of AWS class E6010/E6011 are made to carry
contrived by restoring to not-too-severe drying of the about 5 to 10% water by the electrode maker, before they
finished electrode. This water plays a critical role in the are packed and sealed. Electrodes of the class E6012 may
electrodes of the AWS classes E6010, E6011, E6012 and carry with advantage 3 to 5% water. This water, of
E6013, which contain significant amounts of organic course, refers to the total water content, chemically-
compounds (mainly cellulose) in the coating. The first combined plus hygroscopic, and is expressed as a
two classes may contain 25 to 40% by weight, and the percentage of the weight of the flux coating.
last two 5 to 10% by weight. These organic compounds
make the slag quick-freezing and the electrode easy to Influence of Atmospheric Humidity
manipulate in all positions. They further give adequate Electrodes which are exposed to the atmosphere
protection to the arc stream by forming an inert have been found to absorb and give u p moisture
atmosphere by their combustion, and also provide the according as the relative humidity in the air rises and
necessary drive for ease of manipulation and good falls. The relative humidity is the percentage of moisture
penetration. Being fully combustible, they reduce the in the air, compared with the quantity necessary to
total quantity of slag formed. This is the reason why saturate it. When this value is high, the coating absorbs
such electrodes are popular as general-purpose all- water vapour from the atmosphere, which is held by
position electrodes. the binder and in the capillary spaces between the
For the instantaneous and complete combustion of powdered grains. This action takes place until the
these organic compounds in the arc, oxygen is necessary equilibrium is reached between the coating and the
and this gas is provided by the dissociation of the water. atmosphere surrounding it and may proceed in either
The oxygen released by the water oxidises the carbon direction. It is possible to demonstrate this by weighing
of the organic compounds to the inert carbon monoxide an electrode periodically with a delicate balance, when
gas, in the familiar water-gas reaction. But for this its weight will be found to fluctuate with the change in
oxygen, the weld metal and the slag would be sluggish atmospheric humidity.
and the arc would carry no force. The oxygen from the An investigation was carried out a few years ago at
water further oxidises a part of the ferromanganese the Material Laboratory, New York Naval Shipyard, in
added to the flux as a deoxidiser, and also a portion of the U.S.A. on the variations of moisture content in
iron from core, weld and base metal. The oxides of electrode coatings due to the atmospheric humidity over
manganese and iron so formed then enter the slag and a period of months. The values of moisture content (%)
influence its fluidity. A fluid slag ensures that it does in the coating have been graphically represented, against
not get entrapped in the weld metal and that it spreads variations in the relative humidity, as reproduced in Fig.
out evenly on the entire weld metal bead. 3.23. The electrodes investigated here are the cellulosic
type, class AWS E6011. The electrodes, as received, were However, there is a form of attack which can take place
found to contain water in the range of 2.5-7.5% by on electrode coating due to continuous moist storage
weight, existing both as free water and as combined conditions over a period of months, and which results
water of crystallisation. The graph clearly indicates that in the formation of a white or crystalline fur on the
the six sample electrodes, exhibited similar tendencies surface of the coating. This fur is produced by a chemical
to gain or lose weight throughout the two-month period. reaction between the carbon dioxide in the atmosphere
In addition, there appeared to be a significant agreement and the sodium silicate of the binder, producing crystals
between the covering moisture content and the storage of sodium carbonate and silica powder. The rate of
area humidity conditions. reaction appears to be extremely slow in the absence of
The normal effects of varying moisture contents are water, and is favoured by conditions of fluctuating
not visible on the electrode, nor are they permanent, humidity.
and can be counteracted by restoring the stock of In the continuous presence of excessive moisture,
electrodes to the p r o p e r moisture content range sodium carbonate and silica will be formed, and as the
originally designed by the manufacturer. Alternately, reaction is irreversible under these conditions, these will
the stocks may be maintained in sealed, air-tight remain as a deposit. Further additions to the deposit
containers. However, this severe and costly mode of will be made every time moisture is present in excess.
packing has hardly been found necessary on technical The deposit so accumulated does not appear to have
grounds, except in the case of basic low-hydrogen any serious effect on the welding quality of most
electrodes, because the permissible moisture content electrodes, but its presence in excess may lead to rusting
range for ordinary commercially produced electrodes of the core wire and disruption of the coating. The
is sufficiently wide to meet the conditions found in the presence of the deposit may, however, be taken as an
stores, which are dry and afford free ventilation. indication of unsatisfactory storage conditions.
One of the inevitable results of the introduction of
organic compounds and water-bearing compounds into
the electrode coatings, is the formation of hydrogen in
the arc as a result of their decomposition. This hydrogen
is absorbed by the weld metal in the molten state, and
is released under pressure as the weld metal cools and
contracts. All the hydrogen may not be released, but a
part of it may collect at suitable voids or slag inclusions
in the weld metal, causing minute invisible cracks in
such regions. There is consequently an overall reduction
in the ductility and toughness of the weld metal and
the phenomenon is termed Hydrogen Embrittlement.
It is generally believed that hydrogen, which is
evolved from substances in the electrode coating, is
dissociated into atomic hydrogen by the intense heat of
co r^ r- 10 _,
*- «-
CD
CM
CD
CN
CO
£J az
« <
cc
<
the arc atmosphere. In this state, hydrogen is readily
ui ui ^ 5 5
U_
UJ
u. U. "- ^- ^ soluble in iron and thus diffuses into the molten weld
Fig. 3.23: Variation of moisture percentage in electrode metal from the arc atmosphere. The diffusion rates of
coatings with atmospheric humidity atomic hydrogen are higher in ferrite than in austenite
and depend also upon the permeability characteristics
of each particular steel. In the absence of phase changes,
the diffusion rates tend to increase with temperature.
However, they are noticeable even at ordinary-room
temperature. When the atomic hydrogen enters porosity,
ADOR"WELDING f cavities, microcracks, etc., which although often
extremely small, are present in cast steels and weld
deposits, it tends to change into the molecular form. In
Vacuum packed electrodes-limits this state, hydrogen does not diffuse because of the large
the influence of atmospheric humidity size of the molecule, and thus it remains in the steel.
This so-called precipitation of molecular hydrogen is 120°C for one hour and weighed again. The difference in
believed to cause severe multi-axial stresses. weight indicates the water content in the flux. By simple
If the stress system permits ductility, the steel can calculation, this can be expressed as a percentage.
withstand these severe internal stress concentrations. Determination of total moisture content (hygroscopic
However, if the steel is hard and brittle, or if the stress plus chemically combined) calls for elaborate apparatus
system does not permit plastic deformation, failure may and procedure, such as that described in AWS A5.5
occur, particularly if a high degree of restraint has in- "Specification for low-alloy steel covered arc-welding
troduced severe shrinkage stresses. Thus the degree of electrodes". This method was summarised earlier while
embrittlement is a function of the amount and nature discussing this standard. It was also mentioned, that
of the hydrogen present, and the magnitude, distribu- the standard stipulates that the maximum moisture
tion and nature of the internal stresses in the weldment. content of coverings of low-hydrogen electrodes, when
Therefore, classes of electrodes other than the low- supplied in hermetically sealed containers only, shall
hydrogen ones should not be used on high-carbon and be 0.4% by weight for E70XX class, 0.2% for E80 class
low-alloy steels, as the hydrogen causes what is called and some of the E90 class and 0.15% for E90, E100 and
underbead cracking at the junction of the weld metal other higher tensile classes. No reference is made to
and the base metal, which is referred to as the heat- eletrodes packed otherwise, but it would not be incorrect
affected zone. This will be dealt with in depth while to state that such electrodes be held at a temperature of
discussing the weldability of steels. 350 to 425° C for two hours prior to testing (as stipulated
for other tests in the standard).
Low-Hydrogen Electrodes and Their Storage
Low-hydrogen electrodes, as the name implies, are Electrode Storage
those in which the coatings are absolutely free from Electrodes packed effectively in hermetically sealed
organic and water-bearing compounds or any other tin containers or plastic bags can be stored for any length
ingredients which can be a source of hydrogen. Even of time without fear of deterioration. But electrodes,
the electrodes are finally baked at a temperature especially the non-low-hydrogen types, are often
between 350-425°C to remove the last traces of moisture supplied in simple cardboard cartons or in plastic bags
adhering to the silicate binder. which may not be perfectly air-tight. Such electrodes
The low-hydrogen electrodes must be stored under have a minimum shelf-life of six months under normal
conditions which will prevent the absorption of mois- storage, which can be extended up to two years and
ture. Objectionable quantities of moisture can be ab- longer through correct dry storage and special care. It
sorbed by low-hydrogen electrode coating during hu- has been pointed out that the equilibrium moisture
mid weather, even in a 24-hour period, if the electrodes content of the coating varies with the relative humidity
are unprotected. The manufacturers, being aware of this (RH) of the atmosphere. The moisture pick-up is rapid
problem, usually pack their electrodes in small hermeti- when the relative humidity exceeds 70%.
cally sealed containers (cans). Electrodes packed other- The RH is the percentage of moisture in the air,
wise need to be stored in a heated room. In fact, even compared with the quantity required to saturate it at
when low-hydrogen electrodes have just been removed the same temperature. The quantity of moisture that the
from hermetically sealed containers, pinholes and po- air can hold rises as its temperature rises, so that if there
rosity may occur unless they are re-baked. is no change in the total moisture content of the air, the
Excessively long e x p o s u r e of l o w - h y d r o g e n relative humidity falls as the temperature rises, and
electrodes to h u m i d a t m o s p h e r e m a y result in increases as the temperature falls. When the temperature
impairment of the electrode beyond the mere absorption falls below the point where the relatives humidity
of m o i s t u r e . P r o l o n g e d e x p o s u r e to the h u m i d becomes 100%, moisture is deposited. The temperature
atmosphere will ultimately destroy the binder through at which this happens is known as the dew point.
the reaction of carbon dioxide. In many places the RH of the air during certain
months or the whole year is around 80-90%. In such
Determination of Moisture in Coating conditions, it is imperative that the electrodes are stored
Determination of hygroscopic water in an electrode in a continuously heated storage room to bring down
coating is simple. A quantity of flux is dislodged by bend- the RH percentage. The importance of maintaining heat
ing the electrode 180°C across the middle and weighed all the time may be illustrated by pointing out that air
immediately. The flux is then powdered and heated to at 25QC and RH 70% will deposit moisture, if it is cooled
to below 18°C On the other hand, if its temperature is

raised to 32°C its RH falls to 50%.
Regular consumers of electrodes are advised to
maintain special electrode storage rooms, which are
dehumidified to 50% triaxdifiiirn RH or are kept heated
5 to 10°C above the ambient terhperature. When
electrodes are received in wooden cases, only the cartons
removed from them must be stored in the room aftd not
the cases, considering that wood retains moisture or
water picked up during transit and idkes longer to dry.
Sometimes, and especially under site conditions,
electrodes taken out from a container or from a drying
oven may lie exposed to moist air for several minutes
or hours. Electrode producers therefore strive to design
their coatings in such a way that moisture pick-up, even
under highly humid conditions is low. This factor is
mainly determined by the grade of silicate binder (often Fig. 3.24: Electrically heated electrode drying oven
a combination of several alkaline silicates) and the
baking temperature employed during production.
Moisture pick-up rate of low-hydrogett electrodes,
therefore, varies from brand to brand. Marty electrode The following points should be noted:
producers have successfully developed low-hydrogen 1) It is obvious that the same electrode drying oven
coatings, which are practicality insensitive to cannot be Used for non-hydrogen-controlled electrodes
atmospheric humidity; and they produce data on the and hydrogen-controlled electrodes, because the tem-
actual moisture pick-up after exposure for several hours perature ranges are quite different.
or days to the combination of low temperature and high 2) If hydrogen-controlled electrodes of different
RH. strength levels are being dried in the same oven, one
should naturally select the highest temperature appli-
Redrying of Electrodes cable. This means that electrodes which are meant to be
As pointed out earlier, electrddes other than low- dried at 250°C can be dried at 380°C and those to be
hydrogen types need re-drying; only if they have dried at 380°C can be dried at 425°C However, for con-
become moist or wet in transit or during storage. This venience and economy, one may wisely have different
abnormal condition can be checked by holding four or ovens for different grades of electrodes, in order to en-
five electrodes horizontally between the two palms and sure correct drying and avoid possible mix-up.
rattling them. Moist electrodes will give a dull sound 3) Baking beyond 350°C for more than two hours
while dry ones will emit a metallic tinkle. A surer and baking beyond 400°C for more than one hour should
method is to weld and observe the performance. A moist be avoided, because this causes oxidation of the ferro-
electrode will give a weak arc and more spatter than alloys and makes the coating brittle. In addition, the
usual and the metal will transfer in large globules across oxygen content of the weld deposit is increased and the
the arc. Low-hydrogen dee'tfddes, Uftless they have been manganese and silicon levels are decreased.
packed in hermetically sealed containers, must be re- 4) The time limit given in the redrying schedule
dried as a rule according to the schedule prescribed by should not be exceeded, because otherwise the strength
the electrode manufacturer. Attempting to dry an of the coating deteriorates. Hence the necessity of having
electrode by short-circuiting it with the base plate is not a holding oven, in which the electrodes are transferred
advisable. It hardly removes any moisture, but overheats after the time limit. There is no limit to the time for
the core wire. The correct method is to heat or bake small holding the electrodes at 150°C.
quantities of electrodes at a time in ait electric oven, such 5) It should be borne in mind that if the door of the
as the one shown in Fig. 3.24. oven is opened every now and then, either for inserting
The t e m p e r a t u r e / t i m e schedules generally or taking out the electrodes, they will not get dried to
prescribed for various electrode types are given in Table the extent desired. Here also, the holding oven plays a
3.54. useful part, because once the electrodes are transferred
Table 3.54: Redrying schedules for electrodes

Electrode class/type Temperature / time Holding temperature
Mild steel/low-alloy steels
Cellulosic type (E6010/6011) 70-80°C/l hour
Rutile-cellulose types (E6012/6013) 100/110°C/1 hour
Acid types (E6020) 150/180°C/ 1 hour
Low-hydrogen types (E7016/7018)
To reach 0.6% moisture content 230/260°C/2 hours Then transferred to
To reach 0.4% moisture content 370/400°C/l hour a holding oven at
or 125-150°C
300/350°C/2 hours
To reach 0.2% moisture content 425°C/ lhour

Stainless steels
Rutile type 120/150°C/2 hours
Semi-basis 180/200°C/2 hours
Copper and its alloys
Basic type 200/250°C/2 hours
Aluminium 175/200°C/1 hour
Cast iron
Nickel or Monel core 150/180°C/1 hour
Mild steel core 150/200°C/1 hour
Nickel and its alloys 200/250°C/2 hours
to this oven, it can be opened repeatedly for drawing WELDING PROCEDURES AND TECHNIQUES
out the required electrodes for use.
AWS defines welding procedure, as the detailed
6) Electrodes which have become very moist or wet
methods and practices including all joint welding
should not be transferred at once into the oven, which
procedures involved in the production of a weldment.
is at a high temperature. The sudden expulsion of mois-
It is very important that before starting to weld, a
ture from the coating will result in its cracking. It is ad-
welding procedure is drawn up, which will ensure
visable that such electrodes are first dried at about 100°C
acceptable quality welds at the lowest overall cost.
for half-hour or so, before they are shifted to the baking
Procedures become more stringent and costly as
oven.
criticality of the job increases. For example, fabrication
7) One should never use these ovens for drying of a pressure vessel conforming to ASME code requires
things other than electrodes. It is important to set a time defect-free welds capable of meeting special mechanical
limit for the exposure of LH electrodes to the atmosphere and non-destructive testing requirements demanded by
after they have been taken out of the primary or hold- the code. This will mean use of high quality electrodes,
ing oven and before they are used up on the job. This is skilled and certified welders, moderate currents and
necessary, because the electrodes keep on picking up travel speeds and welds with little or no porosity or
moisture from the air, more so in humid weather, un- undercut. A commercial quality vessel on the other hand
less the coating has been designed to be moisture-resis- may be fabricated with a more liberal procedure and
tant. The following guidelines may be followed: less skilled welders.
Class of electrode Humid weather Dry weather To define and draw u p a welding procedure, one
(90% RH) (50% RH) may use a standard procedure sheet such as shown
below. The sheet can be best prepared by the welding
E70XX 4 hours 8 hours engineer in consultation with the welding foreman or
E80XX 2 hours 8 hours shop-floor supervisor. It simplifies the welders' tasks
E90XX 1 hour 4 hours and prevents last minute confusion and faulty work.
E100XX/110XX 0.5 hour 2 hours The preparation of such a sheet provides an opportunity
to check on what means and materials are available in in an edge joint, and it covers a part or the whole of the
the shop, or have to be specially provided to meet the edge widths.
job requirements. The sheet also helps to qualify the Design of w e l d joints is based on several
welders before they are put on the job. Such sheets serve considerations, some of which are:
as references for the future. Important codes demand a) Manner of stress—tension, shear, bend, torsion
that such procedure sheets are p r e p a r e d and the b) Whether loading is static or dynamic; whether
procedures qualified by completing representative fatigue is involved
w e l d e d joints a n d subjecting t h e m to r e q u i r e d c) Whether subjected to corrosion or erosion
destructive and non-destructive tests. d) Joint efficiency, which is defined as the ratio of
the strength of the joint to that of the base metal, ex-
Typical Procedure Sheet for MMA Welding pressed as a percentage
a) Welding procedure number e) Economy: amount of weld metal required to com-
b) Related specification a n d / o r drawing number plete the joint and whether high deposition processes
c) Material to be welded; specification number or and procedures can be used
composition f) Construction factors: accessibility, control of dis-
d) Metallurgical condition of material tortion and shrinkage cracking, production of sound
e) Type of weld welds.
f) Preparation of parts: Tables 3.55-3.60, reproduced from BS: 499: Part 1:
(i) Angle of bevel 1965, explain and illustrate the various types of joints
(ii) Root face and welds used in welded structures.
(iii) Root radius Some common terms pertaining to joints and welds
g) Cleaning before welding which must be clearly understood are defined below,
h) Set-up of joint (gap, included angle, tolerance on
alignment, etc.)
i) Particulars of backing strip or bar (A) (B)
j) Welding position and direction
k) Make, type and classification of electrode
1) Electrical supply and electrode polarity
m) Size of electrode for each run
6 ^
=m
n) Length of run per electrode
o) Current for each run (D) (C)
p) Open circuit voltage
q) Arc voltage
r) Preheating procedure A1\ ib
s) Time between runs
t) Number and arrangement of runs
u) Welding sequence
t TV
v) Technique for depositing each run
w) Method of inter-run cleaning (E)
(F)
x) Mechanical working of runs
y) Preparation of root before welding reverse side -4 J(
z) Postweld heat treatment.
Types of Joints
There are six common types of joint, namely, butt,
tee, cruciform, lap, corner and edge. These are illustrated
in Fig. 3.25, which also illustrates three main types of Fig. 3.25: Main types of joints and welds: (A) Square butt
weld, namely, butt, fillet, and edge. A typical butt weld joint and butt weld; (B) Square tee-joint and fillet welds;
is shown in the butt joint. A fillet weld is approximately (C) Cruciform joint with four fillet welds; (D) Lap joint with
triangular in transverse cross-section, and is used in tee, single fillet weld; (E) Full-open corner joint with fillet welds;
cruciform, lap and corner joints. An edge weld is a weld (F) Edge joint with edge weld
and some of them are illustrated in Figs. 3.26-3.29, which Root radius: The radius used to form the shape of a J-
are reproduced from BS: 499. or V- groove weld joint.
Root of joint: That portion of a joint to be welded, Angle of bevel: The angle formed between the
where the members approach closest to each other. In prepared edge of a tneinber and a plane perpendicular
cross-section, the root of the joint may be either a point, to the surface of the member.
a line or an area (see Fig. 3.26). Included arigle: The total included angle of the groove
Root of weld: The points, as shown in cross-section, between parts to be joined by a groove weld.
at which the back of the weld intersects the base metal Land: It means root face.
surfaces. Tde of weld: The junction between the face of a weld
Root opening (gap): The separation between the and the base metal (see Fig. 3.28).
members to be joined at the root of the joint; Weld face: The exposed surface of a weld on the side
Root face: That portion of the groove face adjacent to from which welding was done (see Fig. 3.28).
the root of the joint (see Fig. 3.27). Leg of a fillet weld: The distance from the root of the
Root penetration: The depth that a weld extends into joint to the toe of the fillet weld (see Fig. 3.28).
the root of a joint, measured on the center line of the Throat of a fillet weld (see Fig. 3.29).
root cross-section. Theoretical throat: The distance from the beginning
Root reinforcement: Reinforcement of weld at the side of the root of the joint perpendicular to the hypotenuse
other than that from which welding was done. of the largest right triangle that can be inscribed within
the fillet weld cross-section.
Actual throat: The shortest distance from the root of
a fillet weld to its face.
WELD W I D T H
Fig. 3.26: Roots of typical weld preparations WELD WIDTH

ELD F A C |
TOES
TOES
WELD FACE. LENGTH)
Fig. 3.28: Terms pertaining to welds
ROOT RADIUS
ESIGN THROAT THICKNESS—t
| ROOT FACE
GAP-
+5*
INCLUDED ANGLE
ANGLE OF BEVflTCl
ROOT FACE
GAP*JL- GAP
=dlU
ROOT RADIUS
-LAND
Fig. 3.27: Terms pertaining to weld preparations Fig. 3.29: Actual and design throat thicknesses of welds
Effective throat (design throat): The minimum distance b) Current-type and amount: The various factors which
minus any reinforcement from the root of a weld to its must be considered in choosing AC or DC, and the
face. polarity in DC, are explained in earlier chapter. Current
values to be used are indicated under Welding Currents
Welding Parameters in this chapter earlier.
To devise a welding procedure, one must choose Where previous experience is not available, the safest
correct welding parameters, i.e. electrode size, current course is to follow the manufacturer's recommendation
characteristics and value, welding speed, arc length, regarding the type of current, polarity in the case of DC
angle of electrode, welding position and welding and the amount of current to be used.
technique. The following notes are meant to help one to c) Welding speed: By welding speed is meant the arc
arrive at an acceptable procedure. travel speed. For a given electrode size and current, the
a) Electrode size: Each size has a specific current speed is higher with the stringer bead and lower with
capacity range, which is indicated on the package by the weave bead. The wider the weave, lesser is the speed.
the electrode producer. Use of currents above the range In the case of a stringer bead, increase of welding
will cause the covering to overheat and break down, speed under constant arc voltage and current makes the
resulting in increased spatter and inferior weld quality. bead narrower and increases penetration until an
Use of currents below the recommended range will give optimum speed is reached at which penetration is
a weak arc and insufficient penetration. maximum. Increasing the speed further will cause a
The size of electrode to be used is determined by reduction in the penetration. Too high a speed of travel
joint thickness, edge preparation and welding positions also results in undercutting, more so when this is
Thicker materials call for larger diameter electrodes to coupled with current on the high side. Too low a speed
ensure efficient melting of the base plate and the may cause overlapping and overwelding. The travel
electrode and proper fusion. The largest possible size speed should be somewhere between the maximum
which produces a weld of the required quality and she without underwelding and the minimum without
at the highest possible speed should always be used to overwelding. Fillet welding affords a wider latitude with
get high productivity. regard to travel speed, but it should be suitably adjusted
For vertical and overhead welding, smaller diameter to obtain the required size of fillet weld.
electrodes have to be used to restrict the size of the weld Electrode melt-off rate is one of the most important
puddle, since there is a tendency for the molten metal factors influencing arc speed. With high-deposition iron
to flow out of it due to the force of gravity. The largest powder type electrodes, one can use higher currents to
size which an average welder can manage in these obtain higher melt-off, and considerably increase the
positions is 4 mm diameter in the case of non-iron- speed of travel to obtain a weld bead of a given definite
powder type electrode (say E6013), and 3.15 m m size. In sheet metal welding, the speed of travel is kept
diameter in the case of an iron-powder type (E7018). A fairly high to avoid burn-through, but filling the crater
skilled welder can weld satisfactorily in vertical and p r o p e r l y as the electrode progresses, d e m a n d s
overhead positions with 5 mm diameter electrodes of additional skill from the welder.
E6013 as well as E7018 class. d) Arc length: Generally speaking, arc length should
The electrode size is also dictated by the be as short as possible. Too long an arc dissipates the
consideration of accessibility to the root of the joint. In heat into the air, increases spatter and slows down the
a V-groove for example, electrodes small enough to give speed of welding. It also results in nitrogen pick-up by
correct arc length and to reach the root have to be used the weld metal. In DC welding, a long arc is rîore
for the initial passes, followed by larger sizes to complete susceptible to arc blow than a short arc. It is also
the weld. In a T-joint, on the other hand, a larger important that the arc length is maintained consistently.
diameter electrode (6 mm or 8 mm) can be used for the Arc control depends entirely on the welder's skill,
initial pass, since the access to the root is easy. manual dexterity and stamina.
In some cases, the electrode size has to be restricted An average welder usually develops enough skill to
to avoid the possibility of burn-through, caused either control one particular type of electrode. To be able to
by bad fit-up (large gap at the root) or thinness of the use various types, he must have an all-round knowledge
material. In some metals and alloys, the weldability about electrodes and the effect of different arc lengths
considerations require that the heat input is restricted on weld deposit quality. For example, cellulosic types
by using electrodes of smaller sizes than normally used. require slightly longer arc than the rutile types, and the
Table 3.55 : Types of butt joints and welds

Sketch Type of Type of Preparation Remarks
joint weld(s)
^•^.
<
u2—— l Butt Bull Close square Full penetration
Butt Butt Close square Full penetration

135* to
inclusive
5G: Butt Butt Raised edges
Butt Butt Close square Partial penetration

i <P i
L Butt Butt Open square with
backing bar
Full penetration
W Butt Butt Close square Full penetration.

Welded from both
sides
Butt Butt Single-V with root Full penetration

faces
rx/n
ML
Butt Butt Single-V with Full penetration

backing strip
V Butt Butt Single-V with root

faces
Full penetration.
Sealing run used
Butt Butt Double-V with Full penetration.

root faces Welded from both
sides
Butt Butt Double-V with Partial penetration.

wide or deep Welded from both
root faces sides
V Butt Butt and Single-V with Partial penetration.

fillet wide or deep Welded from both
root faces sides
Butt Butt Double-V with Full penetration.

root faces Welded from both
sides
1 I
1 \r J
1
Butt Fillet Close square Welded from both
sides
1
/ /
i r'
r/
o> - Butt Butt Single-bevel with Full penetration
root face
Butt Butt Single-bevel with Full penetration.

root face, thicker Sealing run used
plate tapered
Butt Compound Single-bevel with Full penetration.

of butt root face Sealing run used
and fillet
Butt Compound Single-bevel with Full penetration.

of butt root face Welding from both
and fillet sides
basic low-hydrogen types require extremely short arc. freedom from undercut and slag inclusions. This is more
In vertical and overhead welding, the welder constantly so in vertical and overhead welding and in girth welding
varies the arc length to control the weld puddle under of pipes. Welders should receive training u n d e r
the prevailing conditions. In the case of heavy-coated experienced welding instructors in controlling the
iron powder type electrodes, the tip of the coating can electrode angle a n d other aspects of electrode
be dragged along the joint without the danger of short- manipulation, to attain the standard demanded by
circuiting. The heavy coating causes a sleeve to form at important fabrications.
the burning end. The arc length is then automatically
controlled by the sleeve depth and not by the welder. Welding Positions
Electrode manipulation and welder fatigue are thus The four recognised positions of welding are: Flat or
reduced to a minimum. downhand, horizontal, vertical and overhead. They are
e) Angle of'electrode: The angle at which the electrode shown in Fig. 3.30. The four sketches on the left refer to
is held with respect to the w o r k d e t e r m i n e s the fillet welds made in tee joints, while the four sketches
uniformity of fusion and weld bead contour, and on the right refer to butt welds. The angle and direction
Table 3.56: Types of T-joint and welds
a
joint weld(s)
T Butt Single-bevel with Full penetration

root face
Fillet Close square Welded from both
L—iX sides
Butt Double-bevel Full penetration.

Welded from both
sides
Compound Double-bevel Full penetration.

of butt Welded from both
and fillets sides
Butt Double-bevel both Partial penetration.

wide or deep Welded from both
root face sides
45 # to 90# Butt Single-bevel Full penetration

inclusive
over 5* and less than 45V //

Fillet Edge prepared as
necessary
Table 3.57 : Types of erucifqrm joints and welds
Sketch Type of Type pf Preparation Remarks

joint weld(s)
Butt
czsu;i=3
Cruciform Double-bevel Welded from both
sides
Cruciform Fillet Close square Welded from both

^ - * \ sides
X7
Table 3.58: Types of lap joints and welds

joint weld(s)
± Lap Fillet Square edge
r
0* to 5*
Lap Plug Hole(s) of any

shape in one
plate
L3ap Fillet Hole(s) of any Welded all round

shape in one
plate
Lap Fusion spot None

^
in which the electrode is held is also indicated in each and a line drawn through the same root intersecting
ease. the weld surface at a point equidistant from either toe
Definitions of welding positions are not as simple of the weld. It is illustrated in Fig. 3.32. Example of slope
as they appear to be. They involve the terms weld slope and rotation are given in Fig. 3.33.
and weld rotation. Weld slope is defined as the angle
between the line of the root of a weld and the horizontal. The welding positions are defined as follows:
It is shown in Fig. 3.31. Weld rotation is defined as the • Downhand or flat: A position in which the weld
angle between the u p p e r p o r t i o n of the vertical slope does not exceed 10° and the weld rotation
reference plane passing through the line of a weld root, does not exceed 10°. It is illustrated in Fig. 3.34.
Table 3.59: Types of corner joints and welds

joint weld(s)
u
mft J\ Corner Butt Single-bevel with Full penetration
root face
Corner Fillet Edges prepared as

TT" necessary
OVER 30" A
AND LESS THAN 135*
Corner Fillet Square edge Welded from both

fVi sides
Corner Fillet Square edge Partially-lapped

J*-^ corner. Welded
-M from both sides
Table 3.60: Edge joint and weld

joint weld(s)
Edge Edge Square edge Edges fully

0 # to 30*
covered
INCLUSIVE
Inclined: A position in which the weld slope ex- • Overhead: A position in which the weld slope does
ceeds 10° but not 45° and in which the weld rota- not exceed 45° and the weld rotation is greater
tion does not exceed 90°. It is illustrated in Fig. than 90°. It is illustrated in Fig. 3.38.
3.35.
Horizontal-Vertical: A position in which the weld WELDING TECHNIQUES
slopes does not exceed 10° and the weld rotation A welder's skill in performing a satisfactory weld
is greater than 10°, but does not exceed 90°. It is involves: striking the arc, maintaining short and steady
arc length, making a suitable weaving motion when
illustrated in Fig. 3.36.
required and traversing the joint at correct speed of arc
Vertical: Any position in which the weld slope travel. The arc is struck by hitting or scratching the
exceeds 45° and the weld rotation is greater than striking end of the electrode against the joint surface.
90°. It is illustrated in Fig. 3.37. The electrode is held at a suitable angle with respect to
FLAT HORIZONTAL FLAT VERTICAL
VERTICAL OVERHEAD OVERHEAD HORIZONTAL
Fig. 3.30: Welding positions for butt and fillet welds
motion, or it may be weaved sideways. A bead deposited

LINE OF ROOT without weaving is termed a stringer bead, while that
deposited with weaving is termed a weave bead. At the
end of the weld run, the arc is made to linger
momentarily to fill up the arc crater, and the electrode
withdrawn suddenly to extinguish the arc.
Fig. 3.31: Diagram to illustrate weld slope A welder's competence to produce defect-free welds
depends largely on his ability to maintain a correct
weaving motion according to the type of electrode, type
the work and the line of the joint, depending on the type of joint and position of welding. The commonly used
of joint and position of welding. The arc may be weaving patterns for various welding positions are
traversed along a straight line without a weaving shown in Figs. 3.39-3.42.
ROTATION OF WELD 0°
V
^ 3
ROTATION OF WELD 45* ROTATION OF WELD 90*
Fig. 3.32: Diagrams to illustrate weld rotation

i\°°
zx <=*=i -0
SECTION XX SECTION XX ELEVATION
SLOPE 0°, ROTATION 0° „ ^

SLOPE 35°, ROTATION 90*
SECTION
TIONXX XX [L 7^ SECTION XX L
_-...---_-_-• A S r% _**.-» « —T'— _. . __•
SLOPE 0°. ROTATION90° SLOPE 0 ° . ROTATION 60°
Fig. 3.33: Examples of slope and rotation
1/ o*toio:
0*tq 10° OMo 10#

^ SIDE VIEW" END VIEW
LIMITS OF SLOPE LIMITS OF ROTATION
SIDE VIEW
LIMITS OF SLOPE
Fig. 3.34: Downhand (flat) position of welding
PLAN VIEW
LIMITS OF ROTATION
0 # to4& # ^ ^ FW 0*to45*
Fig. 3.37: Vertical position of welding
SIDE VIEW LIMITS OF SLOPE END VIEW LIMITS OF ROTATION
Fig. 3.35: Inclined position of welding
0 # to 45'
i END VIEW
Fig. 3.36: Horizontal-vertical positions of welding: (A) LIMITS OF ROTATION
Horizontal-vertical fillet weld; (B) Overhead fillet weld;
(C) Horizontal-vertical butt weld
Fig. 3.38: Overhead position of welding
TACK
Fig. 3.40: Methods of weaving for horizontal weld beads
^^<^<^->w-^^4Nr^
WEAVING BACKWARD AND
FORWARD
STRINGER A
WEAVING WITH A CRESCENT

MOTION
Fig. 3.41: Methods of weaving for vertical upwards weld

beads
WEAVING MOTION WITH

LINGERING AT THE EDGES
Fig. 3.39: Methods of weaving in flat position
STRINGER A
A correctly executed weaving motion takes care of

penetration; sidewall fusion; shape, size and profile of
the bead; freedom from undercut, overlap and slag
inclusions; ease of slag removal. Weaving also reduces
or eliminates weld-metal porosity by giving a stirring
motion to the molten weld pool and releasing the
entrapped gases.
Butt Welds
Flat position: The procedure of making a butt weld
Fig. 3.42: Methods of weaving in overhead position
varies with joint thickness. For plates of thickness 1.6
Fig. 3.43: Butt weld on thin plate in flat position
mm and less, it is preferable to have a square butt joint over and the root is machined or gouged out by an oxy-
and provide a backing strip as shown in Fig. 3.43 to acetylene gouging torch, or by a carbon electrode using
prevent burn-through caused by excessive penetration, the air-arc process, till sound metal is reached. The root
which is difficult to avoid even with a 1.6 mm diameter is then covered with a sealing pass deposited with a 4
electrode used at 30 amps. If the backing strip is mild mm electrode at 180 amps.
steel, it gets welded to the joint by the penetrated weld These are typical procedures carried out with
bead. The strip may be retained as an integral part of general-purpose rutile type electrodes. Electrodes with
the joint or removed by machining. The backing strip larger diameter than indicated may be used wherever
can be of copper, which does not fuse with the convenient. For example, the root pass could be executed
penetrated weld metal and can be easily detached with a 5 mm electrode at 210 amps to get deeper
later. Thin sheets are better welded by the oxy- penetration and a better guarantee that slag inclusions
acetylene or argon-arc process, if use of backing strip are completely dissolved out.
is to be avoided, because with these processes the Butt joints in 3-12.5 mm range can also be made
welder has better control on weld penetration. with the deep penetration technique, provided the
For plates in the thickness range 3-15 mm, one excessive weld metal dilution is harmless and the
usually provides a single-V joint as shown at A in Fig. specified mechanical requirements can be met.
3.44. The sequence of weld passes with specified Procedures for two typical thicknesses are shown in
electrode sizes and currents for various thicknesses up Fig. 3.45. In A, the square butt joint (maximum joint
to 12.5 mm is indicated at B to E in Fig. 3.44. In each gap 1mm) is welded from each side with a 4 mm
case, after the V has been filled up, the joint is turned cellulosic type electrode at 160 Amps. In B, the two
j 3.15/110A
f
at
T
) {
1 .3 MM \
1.5 MM B. 3 15/110 A
. 4/180 A 5/210 A
Fig. 3.44: Butt welds on 3-15 mm thick plates in flat position

gouged out to sound metal from the second side and

the sealing pass deposited.
For thickness range between 8 and 25 mm, one uses
l I3MM /fy4/160A I the joint design shown at D. Sequence of passes for one-
(A) sided welding of 12.5 mm and 25 mm thick joints are
shown at E. For two-sided welding of 12.5 mm thick
joint, the sequence would be the same as at E, except
,
12.5 MM M V
n
R^ > 5/326 A 1 that the root is gouged out and a sealing pass is
deposited on the second side with 3.15 mm diameter
LI L\ tt_
(B) electrode at 120 A. For 25 mm thick joint to be welded
from both sides, it is preferable to have double-V joint
Fig. 3.45: Deep penetration butt welds in flat position preparation as shown at F. The welding sequence for
the larger V is shown at G. For welding the smaller V,
the root is gouged out to sound metal and the V is filled
passes have been deposited with a 5 mm diameter up with three passes of 3.15 mm/120 Amps., and three
super-heavy coated deep-penetration iron powder passes of 4 mm/190 Amps.
electrode at 300-325 Amps. It is very important that the Vertical position: The joint is commonly a single-V
penetrations of the two beads overlap at the centre of with 60° included angle, root face 1.5 mm and root gap
the joint thickness, indicating that the slag entrapped 1.5 mm. Welding proceeds from bottom to top. It may
beneath the first bead has been dissolved out by the be completed either from one side only or from both
second pass. the sides. The former procedure demands special skill
For thicknesses of 20 mm and above, one usually from the welder to obtain uniformly good root
prefers a double-V butt joint to a single-V joint to reduce penetration.
the total weld metal content and minimise angular When welding from one side, a 10 mm thick joint
distortion. A typical sequence for a 25 mm T joint with can be completed with two passes of 3.15/110 Amps.
rutile type electrode is shown in Fig. 3.46 (root gap 1.5 and a third and final pass of 4/140 Amps. For 25 mm
mm, root face =1.5 mm). thick joint, two passes of 3.15/110 Amps., two passes of
Horizontal-vertical position: Butt joints in the 3-6 mm4/140 A and five passes of 5/170 Amps, are required.
thickness range in HV position are usually prepared as For welding a 10 mm thick butt joint from both sides,
shown at A in Fig. 3.47. Procedure depends on whether the same sequence as above is used, but the root is
welding is done only from one side or from both sides. gouged out and a sealing pass with 3.15/110 Amps, is
The sequence of passes for 3 mm and 6 mm thicknesses deposited from the second side. In the case of a 25 mm
deposited from one side are shown at B and those thick joint, the double-V preparation similar to that
deposited from both sides are shown at C in Fig. 3.47. It adopted for H/V welding is used (see F in Fig. 3.47).
must be pointed out that in welding from one side, the The larger V is filled with two passes of 3.15/110 Amps.,
welder has to use special skill to achieve uniformly good two passes of 4/140 Amps, and two passes of 5/170
penetration at the root. When welding from both sides, Amps. The root is gouged out and the smaller V is filled
the root penetration is not important, because it is up with two passes of 3.15/110 Amps, and a third and
final pass of 4.140 Amps.
It may be noted that in vertical welding, currents on
6 3/290 A the lower side of the range are used for 4 and 5 mm
electrodes. Also, a welder is able to deposit a bead of
<Ta3h-5/210A
1 4/180A much larger cross-section for a given electrode size, than
2
I T^p^ " 1 8 0 A is possible in other positions. This explains the
&.« JH£--"4/
I /£Z^S 4/1-80 A comparatively smaller number of passes required to
1 /5^*Jp%~~ 5/210 A complete a joint in spite of using smaller diameter
electrodes. The arc travel speed is also comparatively
3/290 A low and this means a correspondingly high heat input
rate, which in some cases has deleterious effect on weld
Fig. 3.46: Sequence of passes for 25.4 mm t double-V
butt joint in flat position
toughness. This point is elaborated elsewhere in the
book.
1 74 MODERN ARC WELDING TECHNOLOGY
(E) 4-5 4/170 A

6-15 5/190 A
Fig. 3.47: Sequence of passes for butt welds of various thicknesses in HV position
Vertical down welding (i.e. top-to-bottom) is used The first pass is deposited with minimum weaving, the
on cross-country pipelines using special technique and arc being concentrated at the root as shown at A to
electrodes. It is described in Chapter 18. ensure good fusion. For a given electrode size, the bead
Overhead position: For overhead welding, the joint size (i.e. leg length) can be varied over a range by
form is similar to that for vertical welding. When single- varying the speed of arc travel, bearing in mind that
sided unbacked procedure is used, 8 mm thick joint may
be completed with two passes of 3.15/110 Amps, and
one pass of 4/150 Amps.; a 12.5 mm thick joint with
3 MM
two passes of 3.15/110 Amps., two passes of 4/140
Amps, and two passes of 5/170 Amps.; and a 25 mm
thick joint with two passes of 3.15/110 Amsp., two
passes of 4/140 Amps, and ten passes of 5/170 Amps.
Overhead butt welds are often made with the aid of
permanent steel backing strips. The joint fit-up in this
case is as shown in Fig. 3.48. The 5 mm thick joint shown
at A is completed with two passes of 3.15/110 Amps.
and one pass of 4/150 Amps. The 25 mm thick joint
shown at B is completed with two passes of 3.15/110
Amps., two passes of 4/150 Amps, and two passes of
5/170 Amps. Overhead welding also demands that 4
m m and 5 m m diameter electrodes are used with
currents on the lower side of the range.
Fillet Welds
Flat position: For making a fillet weld in the flat
position, the T-joint is positioned as shown in Fig. 3.49. Fig. 3.48: Butt joint preparations for overhead welding
as shown in Fig. 3.50 to ensure good root penetration.

Subsequent passes are also laid with minimum weaving
and in the sequence shown in the same figure. The
welding data are given in Table 3.62.
Vertical position: High efficiency joints in vertical
welding are achieved by welding from bottom to top.
In this case, weaving is done for all the passes including
the first one. The techniques of weaving are shown in
Fig. 3.51, and the welding data are given in Table 3.63.
Top-to-bottom welding is done in some cases by
using a rutile-iron powder type or basic iron powder
type electrode. In this technique, the electrode tip is held
in contact with the root and drawn downwards. No
weaving is possible. The weld size is limited and root
penetration is partial. Hence the technique is used where
joint strength is not important.
Overhead position: The placement of passes is shown
in Fig. 3.52, and the welding data are given in Table 3.64.
Effect of Deposition Efficiency

The procedures so far described relate to a general-
purpose rutile type electrode with deposition efficiency
Fig. 3.49: Fillet welds in flat position:
(A) Single-pass weld; (B) Multi-pass weld (DE) of approximately 93%. With electrodes of higher
DE, the sequence and number of runs for a given weld
size do not change in some cases, and only slightly in
too low a speed causes overlapping, while too high a others. The advantageous features of these electrodes
speed results in undercut. Subsequent passes are laid are (a) longer weld run per electrode, (b) less number of
with weaving, and with split passes when the width electrodes per metre of weld, (c) less time to complete
becomes very large as shown at B. The sequence of the weld.
passes for a range of fillet sizes is given in Table 3.61 For completing a 25 mm thick 60°-V butt weld in the
Horizontal position: The first pass is deposited without flat position/the number of runs required with different
weaving, the arc being directed at the root all the time electrode types is indicated in Table 3.65.
Table 3.61: Sequence of passes for fillet welds in flat Table 3.62: Sequence of passes for fillet welds in
position horizontal position
Leg length Pass Electrode size/ Leg length Pass Electrode size/
(mm) numbers current (mm) numbers current
1 4/180 A 1 4/180 A
5.0 or 5.0 or
1 5/220 A 1 5/220 A
1 5/220 A 1 4/180 A
6.0 or 6.0 or
1 6.3/300 A 1 5/220 A
1 5/220 A 3 passes 4/180 A
8.0 or 8.0 or
1 6.3/300 A 2 passes 5/220 A
1 5/220 A 4 passes 4/180 A
10.0 or 10.0 or
1 6.3/300 A 3 passes 5/220 A
12.0 1 &2 6.3/300 A 12.0 3 passes 6.3/275 A

3\
Fig. 3.50: Sequence of passes for fillet welds in horizontal position
Table 3.63 : Sequence of passes for fillet welds in

vertical-up welding
Leg length Pass Electrode size/
(mm) numbers current
1 3.15/115 A
5.0 or
1 4/145 A
1 3.15/115 A
6.0 or
1 4/145 A
2 passes 3.15/115 A
8.0 or
2 passes 4/145 A
3 passes 3.15/115 A
10.0 or
2 passes 4/145 A
1 pass 4/145 A
12.0 plus
1 pass 5/170 A Fig. 3.51: Weaving techniques for multi-pass fillet
welding in vertical postion
For completing a fillet weld of 12 mm leg length in
the H / V position, three runs are required with all the
above electrodes, when the largest compatible sizes (5 Table 3.64: Pass sequence for fillet welds in
mm or 6.3 mm) are used.
overhead position
For welding in V and O positions, the largest Leg length Pass Electrode size/
electrode size that can be used depends on its DE. mm numbers current
Electrodes of DE more than 130% are generally not 1 4/170 A
suitable for these positions. Table 3.66 gives a general 5.0 or
indication. 1 5/200 A
2 passes 4/170 A
WELDING SEQUENCE 6.0 or
In a complex job involving a number of welds, it is 2 passes 5/200 A
important to know the correct sequence in v/hich the 3 passes 4/170 A
welds are made, and the direction in which each weld 8.0 or
2 passes 5/200 A
is made. An incorrect sequence of welding causes
distortion and sometimes cracks in the weld metal 4 passes 4/170 A
10.0 or
due to stress concentration at some point in the fabrication.
3 passes 5/200 A
A correct welding sequence distributes and balances
the forces and stresses produced by weld contraction. 12.0 6 passes 5/200 A
Table 3.65: Relation between deposition efficiency and

number of runs
Electrode type DE (%) No. of runs I
Cellulosic 93 11
Rutile 93 ^
12 ^
Rutile 130 10
Rutile 140 10
Rutile 150 11
Rutile 160 8
Basic 103 11
Basic 115 12
"^n
To prevent angular distortion in a double-V or
double-U b u t t joint, the s e q u e n c e generally
r e c o m m e n d e d consists of laying w e l d i n g r u n s
alternately in opposite side of the joint. However,
strictly speaking, this sequence will give a distortion-
free welded joint only when the operation of gouging
out the root and placing a root run R is not specified as
in the case of second quality w e l d s . W h e r e the
procedure demands gouging out the root and placing a
root pass R, a distortion-free joint can be guaranteed
only by starting with a double-V or double-U, in which Fig. 3.52: Sequence of passes for fillet welds in
one V or U groove is deeper than the other as in Fig. overhead position
3.53. This is called an asymmetrical double-V joint. First
Table 3.66: Relation between deposition efficiency and largest electrode size for positional welding
Electrode Welding Weld type Largest size Current

type DE (%) position F = fillet dia. (mm) Amps.
B =Butt
Cellulosic 93 V F 5 155
B 5 140
O F 6.3 170
B 5 140
Rutile 93 V F 5 170
B 5 170
O F 5 200
B 5 170
Rutile 130 V F 4 150
B 4 150
O F 5 200
B 4 150
Basic 103 V F 5 160
B 5 170
O F 5 170
B 5 170
Basic 115 V F 4 140
B 5 185
O F 5 190
B 4 150
two runs are made in the major groove as in Fig. 3.53 lung welds 6 and 7 are made on the first side, followed
(a). Then the platen turned over, and the weld run No.l by short welds 8, 9 and 10. The welds 4, 5, 6 and 7 arc
is completely gouged out as shown in Fig. 3.53 (b). three times longer than the welds 1, 2, 3 and 8, 9,10.
Runs 3 and 4 are made in this minor groove to fill it In the fabrication of large storage tanks and ships,
completely. Then the plate is again turned over and where a number of plates die butt welded together, the
runs 5 and 6 are made in the major groove to fill it up. welder comes across an arrangement of joints as shown
Sequence for making very long butt joints is shown in Fig. 3.57. In such a case, it is wrong to weld the
in Fig. 3.54. The whole length of joint is divided into longitudinal joint first as in (a), because then this weld
two equal parts with a centre line. Then two welders will have hairline cracks at points where the longitudinal
start at the centre and simultaneously make welds 1, 2, joint meets the transverse joint. The correct welding
3, 4, etc., as shown by the arrows. It will be observed sequence is shown at (b). All the transverse joints are
that while the general direction of welding is from the welded first, and the ends of the welds are ground
centre towards the opposite free ends, each individual properly, so that the V or any other groove preparation
weld run is made in a direction towards the centre. This of the longitudinal joint remains intact. Then the
is called back-step welding. If only one welder is on the longitudinal joint is welded.
job, he can make weld No. 1 on one side, then weld No. An interesting example of welding sequence is
1 on the other side, then weld No. 2 on one side, then shown in Fig. 3.58. Here, two 150 m m channels of 3
weld No. 2 on the other side, and so on. A variation of metre length are welded together to form a rectangular
the back-step welding is skip-welding, which is shown box. Light tack welds are first made in the centre as
in Fig. 3.55. indicated by 17 and 18, and then at the two ends
In Fig. 3.56, we have shown the welding sequence indicated by 1,2 and 11,12. The welding sequence with
generally recommended for making a long double-fillet back-step method is then in the regular order of 1, 2, 3,
weld without distorting the vertical member A. Short 4, 5, 6. . . . 22. With such a sequence, the welding
welds 1, 2 and 3 are first made on one side, and then distortion is less than 3 mm.
long welds 4 and 5 are made on the opposite side. Then A simple sequence for welding a square patch to a
plate without causing distortion or cracking is shown
in Fig. 3.59. The patching plate is so dimensioned and
MAJOR GROOVE (2) inserted that gaps at 1 and 2 are wider than at the other
(a)
two sides. Welds 1 and 2 are made in the directions
indicated by the arrows at A. Due to contraction of the
welds, the patching plate is moved correctly into the
MINOR GROOVE (1) PLATE TURNED OVER FOR ROOT centre as shown at B. Welds 3 and 4 are then made as
GOUGING
shown by arrows at C to complete the job.
/
A German welding magazine has described the
(b)
sequence of welding that was used successfully in the
N. repair of a steam piston head. The head of 350 mm
diameter had the grooves on the outer face broken at
Fig. 3.53: Welding sequence for double-V joint
places. It was decided to build up the surface by welding
CENTRE LINE
V- ■I .1 .1 6 s
(((((((((Mr ((([(((((((((((^^
4
■! 3 - l 2 - ! 1 ■ ' I. 2 I . 3 I . 4 I . 5 I . 6 I - 7 I - 8
i i > i i i i I I l l I I I
GENERAL DIRECTION OF WELDING GENERAL DIRECTION OF WELDING
CENTRE LINE
I
Fig. 3.54: Back-step welding
CENTRE LINE
1
V . | 7 . | 3 . I 6 . I 2 . I 5 . I 1 . 1 i. I.5I.2 i.vn.4i.8
((((((((((((((((((
M M ! GENERAL DIRECTION OF WELDING.
GENERAL DIRECTION OF WELDING
Fig. 3.55: Skip-welding
6
m
»iiii»i»ft»fl]jflwm»»M»m»iiiJj]))j)^»))])nmHflMi»uwnn»i}miHiu]]
iLk
8 t 3x"
&a
h]iHBn)»|iiiiiiiiiMiJinminin»H»HiJ»H»iiiflJMmiiu»iwiVH)i)iijroiiir
wmftm 10
Fig. 3.56: Welding sequence for double-fillet welds
TACK WELD WELD END

RACK( GROUND
CRACK
to JtmrnmrfrMniffl!
rcp
gnnnnnmnffrMnnnninni
CRACK
TACK WELD
INCORRECT CORRECT
Fig. 3.57: Welding sequence for plate channels
3 0M
21
" 9 19 Z_JL2 5 JL5 2 L3 L
\/777>\ersss'A
12 22 10 20 8 18 6 16 4 14 ~~T
Fig. 3.58: Sequence for butt welding two long channels
and then reforming the grooves by machining as shown opposite. The weld on area 3 was next followed by a
in Fig. 3.60. weld on area 4, and so on. The repaired part has been in
The job was not as simple as it looked. It was service for over a year.
important to ensure that the distortion following Sometimes a welder is called upon to build up a
welding was negligible as any change in dimensions cutting edge on a long guillotine blade, which may be
would mean scrapping the entire part. Also, it was up to 3 metre long. A hardfacing electrode is generally
important to distribute contraction stresses evenly used for this purpose. Unless a correct welding sequence
around the circumference. To achieve this, the rim was is used/the blade will distort and the weld will crack at
divided into eight parts as shown in Fig. 3.61. A weld several places. The correct procedure for preparing two
on area 1 was followed by another on area 2 diagonally parallel cutting edges on one blade is as follows:
D
^
INSERTING (A) SIDES 1 $ 2 WELDED (B) FINISHING THE WELDING (C)
Fig. 3.59: Sequence for welding a patch
The blade is fixed in the pre-setting jig and clamped GRAVITY WELDING
down with the G clamps as illustrated in Fig. 3.62. A Gravity welding is a variation of MMA welding, in
gas burner is used to maintain a gentle heat, which is which a long, large-diameter heavy coated electrode
distributed throughout the full length of the blade before (usually 700 mm long, 5.5 mm or 6.3 mm diameter) is
commencing welding. It is inserted inside the channel
bar as shown in Fig. 3.63. The work is not allowed to be
overheated and all welding is carried out in easy stages.
Figure 3.63 illustrates the sequence of welding runs
to avoid cracking and undue distortion. The first runs
are commenced from the centre and worked outwards,
as indicated, until the blade is buttered with a 2.5 mm
electrode using approximately 98 to 110 amps. After
buttering is completed, all traces of slag are removed
and any excess lumps ground off. Fig. 3.60: Repair of piston head whose grooves at A and
B were broken
The second and final runs are commenced as before
for buttering from the centre, working outwards in
stages, until the entire buttering runs have been covered.
This step-back procedure is continued until each
successive run has completed the work. The welding is
gradually built in stages and not completed in any one
part, otherwise the unequal heat distribution will cause
surface cracking. When the welding has been completed,
the gas burner is left at half-cock for about 30 minutes
and gradually turned down. Finally, the job is covered
up and allowed to cool off slowly.
When too blades are fabricated (one for the top a n d
one for the bottom) the securing holes can be drilled,
the two plates bolted together back-to-back (welded
part outwards) and the blade can then be machined
or g r o u n d flush. If g r i n d i n g is u s e d , it is m o s t
important to use a wet grinder working slowly, avoiding
overheating during this last and important operation.
Preheating and Postweld Heating

Heat treatment of a joint before and after welding,
whenever called for, is an important step in a welding Fig. 3.61: Direction and sequence of welding for
procedure. It is dealt with in detail in Chapter 20. repairing piston head
1/2" DIA BAR 1NG

; THICK BLADE IN SLOT
L3- TUBE IN SLOT
I SECTION
OF CHANNEL
I SECTION OF
I SHEAR BLADE.
l
il ■ » I I CUTTING EDGES
PREPARED FOR
HARDFACING
8'0" —
Fig. 3.62: The set-up for hardfacing the cutting edges of a shear blade
completed the weld. This leads to increased

« i £ +1SL «-&- ?! 4» 8 » JL2» U+ productivity. Welds of consistent size and quality are
obtained. Capital investment is small.
"13' ^11 " 7 ^&\ I 5 9 16 ELECTRODE HOLDER
Fig. 3.63: Recommended sequence of weld runs ARC WELDING

ELECTRODE SLIDING BAR
operated in conjunction with a tripod gravity electrode

feeder or a spring-loaded feeder to make automatic
horizontal filler welds. With the spring-loaded feeder,
one can also make groove or fillet welds in the flat
position. In Japan, the process using the spring-loaded
feeder is called Auto-Contact Welding.
The principle of the tripod-type gravity feeder is Fig. 3.64: Sketch showing the principle of gravity
explained in Fig. 3.64. The electrode is clamped in the welding
electrode holder while the arcing end remains in contact
with the joint seam all the time. In the welding operation
the electrode gets consumed and the electrode holder
slides down along the bar The length of the fillet weld
deposited by an electrode is the sum of the projection
of the electrode (minus the stub length) on the weld
seam and the projection on the seam of the part of sliding
bar along which the electrode holder descends. The
tripod type feeder in operation is shown in Fig. 3.65.
Figure 3.66 shows the principle of the spring-loaded
feeder, in which the holder moves on a hinged joint
during the welding operation. Here the length of the
fillet weld is equal only to the projection of the electrode
(minus the stub length) on the seam. This means that
the weld length is shorter than the electrode length.
The main feature of the process is that the electrode
operates by itself without the need of a welder. This
enables the operator to start the second, third or even
fourth tripod before the electrode in the first tripod has Fig. 3.65: Tripod-type gravity feeder in operation
The idea of a gravity feed systems was developed and the end when the electrode is hot. Common sizes
by K.K. Marsden, an engineer in Elsinore Shipyard, are 5.5 mm and 6.3 mm, but higher sizes of 7.00 or 8.00
Denmark, in 1938. It did not catch on then, because mm may be used for special applications.
electrode manufacturers at that time were busy finding The E6027 and E7024 class electrodes have been
ways to produce normal electrodes speedily a n d found to be well-suited for gravity welding, though
efficiently to meet the market demand, which w a s E7018 types have also been employed.
increasing at a rate of approximately 50% a year. Gravity
welding needed longer a n d more heavily coated Tripod Feeder
electrodes, which electrode companies were least There are two versions of the tripod feeder. One
interested in developing. However, gravity welding was provides for adjustment of the angle of the inclined
being used to a limited extent during the 40's in the track. With this type, fillet welds of different sizes can
Scandinavian countries, in England and Japan and in be produced with a given size of electrode operating at
the U.S.A. It was left to the Japanese shipbuilders to the same welding current. For example, welds of lengths
perfect the gravity welding system and utilise it on a between 500 and 1,000 mm can be produced with a 700
large scale in t h e p o s t w a r years. By then, high mm long electrode.
deposition-iron p o w d e r electrodes h a d come into The other version is preset for a definite incline of
general use, which were better suited for gravity the track and for producing a weld of constant length,
welding than the large-diameter non-iron p o w d e r though by varying the welding current, the weld size
electrodes used in the pre-war days. Today gravity and length obtainable from a given electrode can be
w e l d i n g is used all over t h e w o r l d , m a i n l y in varied slightly.
shipbuilding and railway wagon shops. For production work, the preset type adjusted for a
A conventional constant current AC or DC welding specific fillet weld size is preferred. This gives no room
machine normally used for MMA welding is used for to the operator to set the tripod wrongly, and it ensures
gravity welding. However, currents u p to 350 amps are production of consistent welds.
commonly used and a duty cycle of 90% may be attained Tripod feeders are designed to be as light as possible,
at the highest production level. The power sources must since they have to be lifted and carried to different
therefore be adequately rated. The conventional 60% locations every few minutes. At the same time, they are
duty cycle welding power source must be derated to robustly constructed, so that no displacement can occur
allow for the 90% duty cycle. d u r i n g w e l d i n g . A typical t r i p o d feeder w e i g h s
Gravity welding needs heavy-coated electrodes for approximately 5 kg.
at least two reasons. The heavy coating produces a deep The tripod is not provided with a current switch as
sleeve at the arc end, which prevents short-circuiting of this would make it heavy. But it is so designed that the
the arc while the electrode end remains in contact with electrode holder automatically turns over at the end of
the plate. Secondly, the heavy coating provides sufficient the operation, and this action breaks the arc. At this
stiffness to the electrode and prevents it from sagging point, the arc voltage is switched off. When the tripod
due to resistance heating. is relocated and a new electrode has been inserted and
700 mm is the usual length of the gravity electrode. it is set against the weld seam, the voltage is switched
Longer lengths may cause differences in weld between on again w i t h t h e same switch a n d w e l d i n g
the beginning of the deposit, when the electrode is cold recommences. The switching on and off of the arc
voltage between the runs ensures that the operator is
not exposed to arc flash, provided he looks away from
i ELECTRODE HOLDER
the arcing point when switching on the current.
ARC WELDING ELECTRODE Jt-»
Procedure with Tripod Feeder
_^^ZZZ^~' U X
^ 0 « — HINGED JOINT Table 3.67 gives details of a typical w e l d i n g
procedure, using preset type of tripod feeder and 700
FILLET LENGTH PER ELECTRODE- mm long electrodes, employed in a shipbuilding yard
in the U.S.A.
Fig. 3.66: Sketch showing the principle of spring-loaded The fillet size refers to the leg length. According to
feeder for gravity welding an European manufacturer of gravity welding systems
and electrodes, the process is used mainly for making
Table 3.67: Procedure for gravity welding with tripod feeder
Electrode Welding current Fillet weld

Class Size Amps. Size Length
(mm) (mm) (mm)
E6027 5.6 260 7.1 686
E6027 5.6 290 6.4 787
E6027 6.4 260/ 8.7 686
E6027 6.4 2ft) 8.0 787
E7024 5.6 290 7.1 686
E7024 5.6 /350 6.4 787
E7024 6.4 / 290 8.7 686
E7024 6.4 / 350 8.0 787
fillet welds of 8 mm leg length and smaller. They state When using this type of feeder, the fillet weld size
that their gravity electrode of 700 mm length gives a as also the root penetration varies slightly from the start
weld length of 900 mm in 180 seconds. They also market to the end. This is due to the change in the approach
a Mini-Gravity system designed for 450 mm long angle that occurs during the melting of the electrode.
electrodes, which produces 620 mm long fillets with leg The low angle of the electrode with the welding line
lengths varying between 3.5 and 6 mm. may result in poor root fusion. In the case of DC, there
is also a variation in weld appearance and spatter levels,
Spring-loaded Feeder depending on whether the weld is progressing toward
The construction of a spring-loaded feeder is or away from the work lead connection. Also, when
explained in Fig. 3.67. In this equipment, a spring using large electrodes at currents above 300 amps,
actuated by a lever pushes the electrode to the welding blowholes often appear at the root of the fillet weld,
line at a low angle (10°-50°). As the electrode is which is sometimes attributed to the arc blow associated
consumed, the angle between the electrode and the with DC welding. A typical procedure followed in an
welding line increases and finally the arc is cut off American shipyard is given in Table 3.68.
automatically with a stub end of about 60 mm, by a
simple trip mechanism. During this process, the angle Productivity
between the workpiece and the plane in which the Gravity welding is an excellent low-investment
electrode moves, remains unchanged. The welding unit method of applying MMA welding to obtain maximum
is held against the work piece with the help of a productivity, when a large number of fillet welds are to
permanent magnet. A typical feeder weighs about 4 kg. be made in a concentrated area. The work must be laid
ELECTRODE LEVER
HOLDER
PERMANENT
MAGNET
DRIVING
ARM PASE FRAME
£PRiNfi_CASE
Fig. 3.67: Construction of spring-loaded feeder

out and a logical progression established, so that the x 50 mm cross-section and are 300-500 mm long. Several
operator moves from feeder to feeder with minimum of them are lined up in a series. Sometimes one
non-arcing time and without causing the cables to cross continuous length is used. The copper blocks serve to
each other and without being exposed to the arc flash. absorb the heat and to prevent the electrodes from
Table 3.69 gives a comparison of weld metal bending and shifting by overheating. The adhering
deposition rates using one electrode manually versus spatter can also be easily removed from the copper
two, three, four and five gravity welders to make 8 mm material. The heat may be further contained by using
size fillet weld. continuous water-cooled copper blocks.
Some users of gravity welding in Europe have The process possesses many of the advantageous
experienced the following disadvantages: features of gravity welding. Additionally, in this process,
1) Less root penetration the arc is invisible and one can execute long joints in
2) Does not satisfy some of today's ergonomic narrow confined places, which are not possible with any
demands of the conventional processes.
3) Can be monotonous to use The process has been used to make butt joints in
4) The open arc causes smoke and slight atmospheric plates 0.8-3.5 mm T using single pass; lap joints in plates
contamination. 0.8-1.5 mm T, wherein the upper plate is fully melted;
corner joints in plates 1.5-3 mm T, fillet welds in tee
Fire Cracker Welding joints 4 mm T, in the flat as well as vertical positions;
This is a simple semi-automatic welding process and edge joints in the thickness range of 4-6 mm only.
known as Elin-Hafergut process, on the European For butt welds, the electrodes are generally 2.5 to 4 mm
continent and is claimed to have been developed by Elin in diameter, for fillet welds they are 4 to 8 mm. Smaller
of Austria in 1939. and larger electrodes have been used in special cases.
It uses a heavy-coated manual electrode of The fire cracker process has been successfully applied
considerable length up to 3,000 mm. As shown in Fig. to the automatic welding of longitudinal joints of small
3.68, the electrode A is placed horizontally along the diameter pipes by a Swiss inventor through a device
seam, covered with paper B and a heavy copper block named CITOMAT.
C. The bare end of the electrode is held in an ordinary Like gravity welding, fire cracker welding has vast
electrode holder connected to an AC or DC power potential in developing countries owing to its simplicity,
source. The arc is initiated by short-circuiting the striking low investment requirement and high productivity.
end of the electrode to the workpiece by means of a
carbon rod. The arc then travels automatically until the Table 3.69 : Deposition rates with one and more gravity
entire electrode is consumed. arcs
To ensure a successful weld, the electrode must have
excellent arc stability and give spray-type metal transfer. Process Deposition rate
This is achieved with a specially formulated acidic type (E6027) kg h
coating. The power source must also supply steady Manual-one are 4.5
Gravity—two arcs 77
current and arc voltage. Equally important is the joint
Gravity—three arcs 11.8
and electrode fit-up. In most cases, suitable jigs and Gravity—four arcs 15.4
fixtures are used to ensure that the work pieces do not Gravity—five arcs 29.4
shift or buckle. The copper blocks generally have 50 mm
Table 3.68: Procedure for gravity welding using spring-loaded feeder
Electrode Welding current Fillet weld

Class Size (Amps.) Type Size Length
(mm) (mm) (mm)
E6027 5.6 240 AC 6.4 686
E6027 6.4 270 AC 8.0 686
E7024 5.6 240 AC 6.4 686
E7024 6.4 270 AC 8.0 686
Pipe Welding
Special techniques have been developed over the
years for the circumferential welding of cross-country
pipelines in the field, using manual electrodes. These C COPPER BLOCK
are discussed in the chapter on Welding Applications. B. PAPER
Weld Defects
It is not easy to get absolutely defect-free welds by
the MMA process, especially when they are multi-pass
welds. Small imperfections, such as minute slag
inclusions or stray porosities are often present and
revealed in non-destructive testing. As long as they are
not damaging to the function of the weldment they are
referred to as discontinuities, and ignored. However,
when they are serious enough to lower the joint
efficiency and interfere with the joint performance in
service, they are looked upon as defects. Until recently,
the practice was to repair all defective welds revealed Fig. 3.68: Sketch of Elin-Hafergut process
by radiography in important welded structures, often (fire cracker welding)
at exhorbitant costs. Lately, with the development of
fracture mechanics, especially linear elastic fracture the welding fabricator must strive to produce
mechanics, and COD testing, the engineer is able to weldments as nearly perfect as possible using normal
determine the significance of defects in terms of procedures and techniques.
susceptibility to fracture and decide on the size and Weld defects which commonly occur due to incorrect
extent of defects that can be tolerated in a given structure welding procedures or faulty manipulation of the
under given service conditions. Yet, the fact remains that electrode are dealt with in Chapter 12.
The Procedure Handbook of Arc Welding, 12th Ed., published 1973 by The Lincoln Electric Company, U.S.A.
Welding Process Technology by Houldcroft.
Metals Handbook, Vol. 6. 8th Ed., published 1971 by American Society for Metals.
Welding Technology by J.W. Giachino. W. Weeks and G.S. Johnson, published 1968 by American Technical Society.
Metal Transfer from Coated Electrodes by W.G. Essers et al. Met. Construction, Apr. 1971.
The Transfer of Metal from Coated Electrodes by .I.E. Lancaster. Met. Construction, Oct. 1971.
Arc Properties in Manual Welding with Coated Electrodes by J. Wegrzyn. Met. Construction, July 1980.
Some Aspects of Welding in Modern Ship Construction by V.P. Nair, Indian Wielding Journal, Oct. 1977.
Gravity Welding in the Shipbuilding Industry by G. Zoethout, Weld. & Met. Fabr. Nov. 1975.
Gravity Welding—A Variation of Shielded Metal Arc Welding by H.B. Gary, Weld.-Journal, Nov. 1979.
The Fematic Gravity Welding System by ESAB. Weld. & Met. Fabr. Nov. 1975.
Moisture Loss and Regain by Some Basic Electrodes AWS. Weld. Journal, May 1976.
Role of H2 in Arc Welding with Coated Electrodes AWS. Weld. Journal. Apr. 1961.
Weld Metal H2 Level and the Definition of Hydrogen-Controlled Electrodes published 1974 by Weld, in the World.
H2 Content of Welds by Evans and Baach. Met. Constr. Oct. 1975.
Submerged
Arc
Welding
DURING the 1930's, having recognised the potential In fully a u t o m a t i c w e l d i n g , the flux is fed
advantages of mechanised welding, several attempts mechanically to the joint ahead of the arc, the wire is
were made to mechanize the arc welding process. De- fed automatically to the welding head, the arc length is
veloping a continuous coated electrode as an extension automatically controlled and the traverse of the arc or
of the manual metal-arc welding electrode was ruled the work piece is mechanized.
out for the following reasons: (a) Since the coating is In the semi-automatic version, the wire feed and arc
non-conducting, arranging electrical contact with the length control are automatic, while the welder moves
electrode is not practicable, (b) the coating is likely to the welding gun, usually equipped with a flux-feeding
peel off when the electrode is coiled, and (c) the coating device, along the joint at a controlled rate of travel. Flux
is also likely to get crushed when fed through the feed feed may be by gravity flow, through a nozzle concentric
rolls. with the electrode from a small hopper atop the gun, or
In one of the methods attempted, the work piece was it may be through a concentric nozzle connected to an
painted with a thin slag flux, while feeding of the bare air-pressurized flux tank. Flux may also be applied in
wire and arc travel were mechanized. In another case, advance of the welding operation or ahead of the arc
automatic carbon-arc welding was tried out, in which a from a hopper run along the joint.
paper impregnated with flux was delivered to the arc D u r i n g w e l d i n g , the intense heat of the arc
as it travelled. simultaneously melts the tip of the bare wire electrode
All such methods proved unsuccessful, because they and part of the flux as shown in Fig. 4.1. The electrode
failed to provide adequate shielding to the arc. Then in tip and the welding zone are always surrounded and
1932, the idea of placing a thick layer of dry granular
flux on the joint ahead of the carbon electrode was
conceived and successfully developed in the U.S.A. and
later applied to the welding of penstocks and water
conduits in California. Submerged-arc welding was the
next logical step and the process became a commercial
success both in the U.S.A. and the U.S.S.R. by the middle
and late 1930's.
The modern submerged-arc (SA) welding is an arc
welding process, in which one or more arcs formed
between one or more bare wire electrodes and the work
piece provides the heat for coalescence. The arc is
completely submerged under a blanket of granular,
fusible flux, which adequately shields the arc from
\BASE METAL
atmospheric contamination. The process can be fully
Fig. 4.1: Mechanism of SA welding
automatic or semi-automatic.
protected by molten flux, while all of them are covered have been studied in spite of it being submerged. The
by the top layer of unfused flux. As the arc progresses findings have been detailed in the section on fluxes.
along the joint, the molten metal settles down while the
lighter molten flux rises from the puddle in the form of Advantages and Applications
slag. The weld metal, having a higher melting point, The process is extensively used in heavy steel plate
solidifies first while the slag above it takes some more fabrication, because of the following advantageous
time to freeze. The solidified slag continues to protect features:
the weld metal while it is still hot, and is capable of • Absence of smoke and arc flash, hence minimum
reacting with atmospheric oxygen and nitrogen. need for protective clothing
Figure 4.2 explains the operation of the automatic • High weld metal quality
SA process while welding a single-V butt joint. Electric
• Smooth and uniform weld finish with no spatter
current from a welding transformer, generator or
rectifier passes t h r o u g h the contact tube to the • Extremely high deposition rate and welding speed
consumable electrode wire. The wire is fed continuously • High electrode deposition efficiency
through the contact tube (nozzle) and through the flux • High arc time can be achieved through automa-
blanket to the joint by mechanically powered guide rolls. tion
The wire is coiled on a reel or in a drum. The flux, stored
• Welder's manipulative skill not needed; mini-
in a hopper, is deposited on the joint ahead of the arc
mum operator fatigue.
through a tube. After the weld metal has solidified, the
unfused flux is removed manually or by a vacuum pick- Applications cover pressure vessels, line pipe,
up system, to be screened and reused. The weld backing storage tanks, heavy structural, ships, railway wagons
plate serves to prevent burn-through; starting and run- and coaches, and surfacing and build-up work.
off tabs eliminate defects likely to occur during the start
and finish of a weld run. BASIC EQUIPMENT
In the early years, it was supposed that SAW is an The SAW equipment essentially consists of:
electric resistance process like the modern electro-slag a) A wire feeder to drive the electrode to the work
process, since the molten slag, which is electrically through the contact tube of a welding gun or welding
conducting, surrounds the electrode tip. Voltage records head
of the process, however, have shown short circuits, re- b) A welding power source to supply electric cur-
ignition peaks and other features that are typical of a rent to the electrode at the contact tube
true arc. In recent years, presence of an arc cavity c) An arrangement for holding the flux and feed-
beneath the electrode tip has been recognized (as shown ing it ahead of the arc
in Fig. 4.1) and modes of metal transfer across the arc d) A means of traversing the weld joint.
TO AUTOMATIC WIRE FEEO TO FLUX HOPPER

^DIRECTION OF TfcAVEL
TO POWER S U P P L Y ^
FLUX-DELIVERY TUBE
CONNECTOR
FLUX DAM (10F 2;
o0l CONTACT]
USED ONLY IN SPECIAL GRANULAR RUNOFF TAB
APPUCATIONS) '**
SOLIDIFIED SLAG4
WELDING
GROOVE
GROUNI
BASE
GROUND METAL
METAL WELD BACKING PLATE
Fig. 4.2: SA welding operation in a single-V butt joint

SUBMERGED ARC WELDING 189
POWER SOURCES if the current required is higher than the maximum

The welding power source can be either of the output of a single machine.
following: (a) an AC transformer of the constant current When a constant current power source is used
type; (b) a DC generator of the constant voltage or together with an automatic wire feeder, it is necessary
constant current type; or (c) a DC rectifier of the constant to maintain the arc voltage (i.e. arc length) as nearly
voltage or constant current type. constant as practicable, in order to obtain a steady
AC power calls for more elaborate and complicated welding current. This is achieved by monitoring the arc
controls and the arc starting is not easy. The main voltage to control the electrode feed speed. When the
advantage is that it minimizes arc blow, which can arc voltage (arc length) exceeds a set value, the control
become quite troublesome with DC at welding currents circuit simply increases the speed of the wire-feed
above 900 Amps. In SAW with multiple wires and motor, so that the arc length is reduced. When the arc
multiple power sources (which is described later), the voltage and hence the arc length decreases, the control
trailing arc or arcs are derived from AC because two circuit decreases the speed of the wire-feed motor,
DC arcs 100 mm or less apart are deflected by each whereby the arc is lengthened. The net result is a
other's magnetic field. Combination of one DC and one reasonably constant arc length and welding current,
AC arc is often used, because their deflections can be though minor voltage and current fluctuations caused
controlled and the combination can be easily applied to by the inertia (i.e. lack of instantaneous response) of this
a variety of fabrication jobs. system cannot be avoided.
Use of DC p o w e r for SAW has the following DC, constant voltage type: As explained in Chapter 2,
advantages: (a) easy and accurate arc start; (b) good the constant voltage power source differs from the
control of the weld bead profile, depth of penetration constant current type, in that the characteristic V-A curve
and welding speed; (c) difficult contours can be is nearly flat. In SAW, this type of power source is used
negotiated at high welding speeds. together with a constant speed wire feeder, whose drive
In using DC, electrode positive (i.e. reverse polarity) operates independently from the welding circuit. With
gives an extremely stable arc and smaller weld puddle, this system, arc length control is more precise and rapid.
which lead to better control of the weld bead profile. As soon as the arc length increases ever so slightly (i.e.
The weld penetration is also deeper. Electrode negative arc voltage increases), there is a relatively large decrease
(i.e. straight polarity) gives highest metal deposition in amperage, which decreases the melting rate of the
rates, but shallower penetration. Penetration with AC electrode. Conversely, when the arc shortens even a bit
is more than that obtained with DC electrode negative, (i.e. arc voltage decreases), the current registers a big
but less than that with DC electrode positive. DC can jump and increases the melting rate of the electrode.
present the problem of arc blow, as stated earlier. When The net result is constant arc length and arc voltage
two DC arcs operate in proximity, they flare together if without fluctuations, since the response is almost
their polarity is the same, and they flare apart if their instantaneous. Since the arc is thus inherently self-
polarity is different. regulating and can be easily started and stopped, the
Whether AC or DC, the power source should be rated control circuits are much simpler than with the constant
at 100% duty cycle and not at 60% as required for manual current power source.
welding. This is because SAW is generally a fully With this type of power source, short-circuit current
automatic process, for which the rating is usually 100% is very high. The welding current is determined by the
duty cycle. This means that the power source should be wire-feed rate. The current increases as the wire speed
designed to deliver its rated output continuously, increases and vice versa. The high short-circuit current
without exceeding the prescribed temperature limits of makes arc striking easier than with constant current
its components. Most SAW is done in the current range types. Crater filling at the end of the weld run also
of 200-1,000 Amps.s (single-wire). In rare cases, the becomes easier, because the welding current can be
current can go up to 4,000 Amps.s at 55V and be as low gradually reduced by gradually decreasing the wire-
as 150 Amps.s at 18 V. For common applications, it is feed rate.
convenient to have a power source giving a maximum For welding thinner sections (3.6 mm and lower),
current output of 500 or 1,000 Amps, at 100% duty cycle. this type of power source has a distinct advantage over
DCf constant current type: The power source is either the constant current type, because the uniformity of
a motor-generator or a rectifier. Two or more generators voltage permits higher w e l d i n g speeds. Such an
or rectifiers of the same type can be connected in parallel, advantage is not evident when welding thicker sections.
In fact, for heavy welds, a power source with more MOLTEN FLUX START
drooping characteristics is to be preferred. Arc starting is easy when a molten puddle is present.
AC, constant current type: AC power for SAW is One has simply to insert the electrode into it and switch
derived from high capacity welding transformers the welding current on. This method is always used in
having minimum open-circuit voltage of 80 V to ensure multi-wire welding. Only one electrode is started to
that the arc is not extinguished during current reversal establish the molten puddle. Then the other electrodes
at every half-cycle. Small capacity transformers may be start automatically when they are fed into it.
connected in parallel to get current output greater than
the maximum obtainable from a single unit, care being WIRE RETRACT START
taken that the output voltages and currents are in phase. This method is useful and economical when frequent
In AC w e l d i n g , w e l d quality often suffers from starts have to be made and when the start must be made
fluctuations in the primary voltage, unless steps are at a definite point. It can be used only with the constant-
taken to maintain it constant. The advantages of using current type system.
AC for the trailing arcs in multiple-arc welding have In this method, the electrode wire is lowered until it
been mentioned earlier. makes a superficial contact with the work piece. The
flux is applied and the welding current switched on.
Arc Start The low voltage between the electrode and the work
Unlike in manual welding, arc starting can be difficult piece signals the wire feeder to withdraw the electrode
in SAW, because of the flux cover. There are several tip from the work piece. The arc starts as this action
methods available and the choice is dictated by the taken place. As the wire w i t h d r a w a l c o n t i n u e s
design of the power source, the time required for starting momentarily and the arc voltage increases, the wire feed
as compared to the total welding time, the frequency of motor quickly reverses direction to feed the electrode
starts demanded by production, and the importance of into the arc. The wire feed rate increases until the wire
starting the weld at a specific point on the joint. melting rate and arc voltage stabilize to the pre-set
values.
WITH STEEL WOOL OR IRON POWDER
A rolled ball of steel wool, approximately 10 mm in HIGH FREQUENCY START
diameter, is placed at the required spot on the joint and In this method, a high frequency, high voltage
the electrode wire is lowered onto it till it is tightly generator introduced in the welding circuit causes a
compressed. Alternately, fine iron powder is sprayed spark to jump between the electrode tip and the work
in the gap between the electrode wire tip and the work piece when the gap between the two is reduced to
piece. The flux is then applied and the welding is approximately 1.6 mm. The spark produces an ionized
commenced. The steel wool or iron powder conducts path, through which the welding current can flow. This
the current from the wire to the work piece and at the technique is commonly used as a starting method for
same time melts away rapidly as the arc is formed. intermittent welding or for welding at high production
rates, where a large number of starts are required.
SHARP WIRE START
The electrode wire tip is cut mechanically to form a Stopping the Arc
sharp, chisel-like configuration, so that the current Normally for terminating the arc, the carriage travel
density at the tip increases. The electrode is then lowered is first stopped, then the wire feed is switched off and a
to make a superficial contact with the work piece. The couple of seconds later, the current is switched off. The
flux is applied and the welding is commenced. The sharp brief pause between the second and the third step
point melts away rapidly and starts the arc. prevents the wire from moving forward into the molten
pool and from sticking to the molten p u d d l e as it
SCRATCH START solidifies. It also helps to fill up the crater.
The electrode wire is lowered until the tip makes a
superficial contact with the work piece, and the flux is TYPICAL WELDING OUTFITS
fed. The carriage is started and immediately afterwards As stated earlier, SAW can be fully automatic or
the current is switched on. The motion of the wire semi-automatic. In the automatic version, the wire-feed
prevents the electrode wire from short-circuiting with rate and arc length are automatically controlled. The
the work piece. travel of the arc at desired speed along the seam is also
automatically controlled, either by (a) keeping the job

stationary and making the welding head travel along
the seam or (b) keeping the welding head stationary
and making the seam travel across the point of the arc.
In the semi-automatic version, the wire-feed rate and
the arc length are automatically controlled. The
electrode, however, is fed through a flexible cable to
the welding gun or torch. The advance of the welding
torch along the seam is manipulated by the operator.
The semi-automatic process can be effectively applied
on jobs of repetitive nature, where manual welding
operations tend to become unprofitable, provided the
relevant procedures and techniques are properly
worked out.
This process scores over the fully automatic
submerged-arc welding process by virtue of its
flexibility, which enables it to be used on complicated
work shapes without the need for tooling and fixturing Fig. 4.3: Wire feeder of a semi-automatic SAW unit
demanded by the latter. Over the years, however, the
usage of semi-automatic SAW has reduced in the wire feeder. The wire sizes are 1.6, 2.0 and 2.4
substantially. One of the main reasons appears to be mm.
the higher operator fatigue in manipulating the heavy c) Torch with flux hopper. It is a self-contained unit
welding head. with bent torch head, 3 m long hose, a flux hopper that
A typical semi-automatic outfit consists of: can take up 1 kg of flux, and a suitable coupler for con-
a) Power source nection to the wire feeder. It is shown in Fig. 4.4.
b) Wire feeder A typical tractor-mounted fully automatic SAW unit
c) Torch with flux hopper shown in Fig. 4.5 consists of:
a) Power source
d) Set of cables for connecting wire feeder to power
source, comprising control cable, welding cable b) Self-propelled tractor or carriage
with end lugs, welding cable with end lug and c) Welding head arm with wire reel
ground clAmps.. d) Welding head assembly.
e) Set of accessories for wire feeder. e) Control unit.
a) Power source: It is a DC welding rectifier giving
maximum welding current of 600 Amps.s at 60% duty
cycle and 500 Amps.s at 100% duty cycle. It has con-
stant potential characteristics, giving a constant arc volt-
age, while the current is determined by the wire-feed
rate. The exceptionally high short-circuit current makes
arc striking very simple. The rectifier has a drip-proof
casing with the following built-in components: trans-
former, silicon rectifier, reactance coil, fan, safety de-
vice against ventilation failure, control transformer and
welding contactor. All controls are arranged on the front
panel for convenient handling.
b) Wire feeder: As shown in Fig. 4.3, it consists of a
DC wire drive motor, reduction gearbox, four-roll drive
mechanism and a wire spool holder. Wire-feed rate is
continuously variable in the range of 1.2-7.5 m/min.
For infinitely step-less variation of the wire-feed speed, Fig. 4.4: Torch with flux hopper for a semi-automatic
a special solid-state thyristorised circuitry is provided SAW unit
made to swivel by approximately 270° and locked in

that position. The carriage speed is controlled by means
of solid-state thyristorised circuitry fitted in the control
unit.
c) Welding head arm: The arm, supporting the weld-
ing head, consists of vertical slide and universal clamp
assembly. The welding head can be accurately posi-
tioned by means of vertical slide adjustment. The uni-
versal clamp assembly is used to tilt the welding head
to the desired angle. The welding head can be swung
round up to maximum 45° for fillet welding, and then
locked in the desired position. The two reel flanges are
designed for standardized spool size of 300 mm ID and
100 mm width.
d) Welding head assembly: One of the main compo-
nents of the welding head assembly is the wire straight-
ening and wire-feed device, designed to straighten the
wire unwinding from the spool, and to feed it to the
jaws/nozzle via wire guide. The wire is fed forward by
Fig. 4.5: Fully automatic SAW unit comprising (a) power a DC motor (1) by means of four-roll wire-feed drive,
source, (b) tractor mounted welding head (2) a reduction gearbox, (3) a wire-feed roller. The speed
of the wire-feed roller can be infinitely varied by a knob
a) Power source: The 600-Amps.s DC constant-volt- on the control unit, to provide any desired speed from
age rectifier meant for semi-automatic welding and de- 0.1 to 7.5 m / m i n . The speed is controlled by means of a
scribed above can be used, provided the welding cur- solid-state thyristorised circuitry. After the wire has
rents used on the job are within its capacity. If currents been pulled from the reel and straightened, it reaches
higher than 600-amps are to be used, two such rectifi- the wire guide tube, on the rear end of which a jaw
ers can be connected in parallel. For heavy fabrication holder and contact jaws nozzle are fitted. The pressure
jobs, it is more convenient to use a single 1200-amps of the current-carrying jaws on the wire can be adjusted
DC rectifier having maximum current capacity of 1,200 by a regulating screw. Also mounted on the rear end of
amps at 60% and 1,000 amps at 100% duty cycle, which the wire guide tube is the flux dispenser with hopper.
like the 600-amps model, gives a constant arc voltage, The elevation between the wire tip and consequently
while the current is determined by the wire-feed rate. the hopper and the work piece, can be adjusted.
Fabricators, today, also have a choice of 800, 1000 A When welding with two wires, the lower part of the
models also. The exceptionally high short-circuit cur- welding head (wire guide tube, jaw holder, contact jaws)
rent makes arc striking very simple. The rectifier (see can be swung 90°, so as to make it possible to feed the
Fig. 4.5 a) has a drip-proof casing with built-in trans- wires either parallel or in tandem. The flux is poured
former, silicon rectifier, reactance coil, fan safety device into the flux container through a built-in sieve, from
against ventilation failure, control transformer and where it is supplied through a flexible tube to the flux
welding contactor. All controls are arranged on the front dispenser. The flux supply can be stopped at the lower
panel. end of flux container by a stopcock-tapered handle. The
b) Carriage: The four-wheeled carriage houses a DC entire wire-feed system can be s w u n g forward
motor with speed reduction gearbox, which drives the approximately 45° around the axis of the wire-feed roller
rear wheels by means of a clutch engageable in any po- (parallel to the axis of the arm). For fillet welding, the
sition. The clutch is actuated by two-position lever (two arm is in transverse position.
limit stops). The front wheels run idle. The carriage e) Control unit: This is mounted on the tractor and all
speed is adjusted by the hand wheel within a range of the controls are brought on the front panel for easy
0.2-1.5 m/min. The travel direction is pre-selected by a setting of w e l d i n g p a r a m e t e r s . The front panel
change-over switch. The mounting bracket bearing the incorporates:
welding head arm can be shifted by approximately 110 1) Digital meters to read carriage speed (m/min),
mm transverse to travel direction. The arm tube can be voltmeter and ammeter
2) Potentiometers for carriage speed and wire-feed

speed setting
3) Reverse /forward switch for carriage
4) Inching u p / d o w n for wire setting
5) Start/stop push buttons for weld sequence.
Control unit incorporates the electronic speed control
circuitry and weld sequence circuitry for wire feed and
carriage drive.
SPECIAL ATTACHMENTS/ACCESSORIES
a) Fabricated rail track: The carriage assembly is
placed on this track for welding straight seams.
b) Pilot needle/spotlight projector: In SAW, the opera-
tor cannot see whether the arc is traveling exactly along
the line of the seam, because both are covered with flux.
The pilot needle is therefore provided to check the weld-
ing head position with respect to the line of the seam. Fig. 4.7: Flux recovery unit
Use of pilot needle entails slight inaccuracy, since its
tip is unavoidably at a small distance from the refer- Modifications
ence edge. Alternately, a spotlight projector is supplied, Boom mounted unit: The tractor mounted automatic
which indicates the correct path to be traced. This is unit can be suitably smodified to facilitate mounting on
fitted into a special socket on the split transformer box. a welding column and boom or a similar fixture. The
c) Fillet welding attachment: When the automatic unit b o o m - m o u n t e d unit is specially suited for the
travels directly on the work piece instead of on a track, longitudinal and circumferential welding of cylindrical
the standard welding head proves to be too short for objects like, boiler shells, pressure vessels and penstock
fillet welds. In this case, a special attachment has to be pipes. A typical unit is shown in Fig. 4.8(a). The welding
mounted on the welding head as shown in Fig. 4.6. head is mounted on an arm, which is capable of being
d) Flux recovery unit: In SAW, only a small portion moved in the horizontal, vertical and longitudinal
of the flux fed to the seam is fused by the heat of the arc. directions. This enables exact adjustment of the electrode
The rest remains on the seam in its original condition. tip on the joint. These adjustments are done by means
The excess cloth-type filter, complete filtration precedes of a compound slide, which is mounted on the vertical
vacuum producer as shown Fig. 4.7. face of the boom. The welding head consists of guides
and rollers for the wire and current collection system,
using split jaws or solid nozzle, and hopper for flux
storage. The wire-feed mechanism consists of a wire-
feed drive operating on a low-voltage, three-phase
supply that is obtained through a control transformer
housed in a separate casing, which can be placed on the
power source. A control panel houses the ammeter,
voltmeter and control switches. The spool holder can
be mounted on the boom and is capable of taking two
spools.
The power source remains the same, i.e. a constant
voltage rectifier giving the maximum output of 1,000
amps at 100% duty cycle. The unit uses electrode wires
of 2,2.5,3 and 4 mm diameter. Using special accessories,
one can weld with 5 mm diameter single wire.
Side-beam carriage unit: In this case, the carriage
carrying the welding head, wire supply, flux if required
Fig. 4.6: Fillet welding attachment and controls for the operator, is mounted on an I-beam
thicknesses and lengths. The welding head is carried

by a carriage, which traverses a long side-beam. It is
positioned over a special nold down fixture, which
pneumatically clamps and aligns the joint edges, and is
provided with a permanent copper backing (with water-
cooling if necessary) over which the joint is clamped.
Seamers are often designed to dispense with tacking.
The cylinder is centred and aligned by means of a
pneumatically swiveled blade stop. First, the left side
of the shell plate is pressed and clamped against the
blade stop. Then the stop swings out and the right plate
edge is pressed and clamped against the edge previously
clamped.
Internal welding unit: Another modification suitable
for internal welding of cylindrical jobs of diameter as
Fig. 4 8: (a) Boom mounted SAW unit small as 300 mm is shown in Fig. 4.9(b). In this unit,
dimensions of the flux hopper are suitably modified.
The welding head is also suitably modified. The arm
on which it is mounted is extended from the carriage to
reach a maximum length of two metres inside the
cylinder. The wire spool is mounted above the carriage
and the wire travels from the spool to the head, through
a tube. This modified welding head has a maximum
Fig. 4 8: (b) Side beam carriage unit
of desired length, which is fixed horizontally between

two columns at a suitable height in relation to the job
size (see Fig. 4.8b). The welding head on the carriage
can also be adjusted for different heights, and for Fig. 4.9: (a) Weld seamer
transverse dimensional requirements. This device
performs only one function and that is to provide
horizontal travel of the welding head. The welding head
is monitored by the operator and can be adjusted to
follow joints, which may not have been perfectly
aligned. Typical applications are longitudinal joints of
cylindrical vessels, p e n s t o c k shells a n d trailer
assemblies.
Weld seamer: This standardized machine shown in
Fig. 4.9(a) combines a welding head travel device with
a work-holding device. It is meant for making one-sided
l o n g i t u d i n a l b u t t w e l d s in cylinders of v a r i o u s Fig. 4.9 (b): Internal SAW unit
SUBMERGED ARC WELDINQ 195
capacity of 600 amps as compared to 1,200 amps of the the uniform feeding of the electrode in automatic and
original head. semi-automatic equipment. The electrode must be
wound to avoid kinks, waves, or sharp bends that would
SAW CONSUMABLES interfere with feeding. The outside end of the electrode
Continuous bare wires in the form of coils and dry, (the end from which welding is begun) must be secured
granular fluxes are used in combination as consumables and identified.
for the SAW welding of mild steel, low-alloy steels, This standard also gives the classification system for
stainless steels, non-ferrous alloys and for surfacing fluxes based on the mechanical properties of the weld
applications. The bare wires are mostly solid wires, metal they deposit in combination with certain
although flux-cored wires (i.e. tubular wires carrying classifications of electrodes. The mechanical tests, in
flux in the core) are also used, either of which have to effect, evaluate the quality of a flux-wire combination.
be used in combination with fluxes. Hence, these have been taken up later under
"classification of fluxes".
ELECTRODE WIRES
As with manual electrodes, SAW wires are best 2) AWS A5.23-1980, "Specification for low-alloy steel
understood by going through the standards. electrodes and fluxes for submerged-arc welding"
1) AWS A5.17-1997. "Specification for carbon steel elec- This standard covers a wide range of low-alloy wires
trodes and fluxes for submerged-arc welding" divided into five groups as shown in Table 4.3. The
This standard gives eight classifications based on classification consists of the prefix 'E' to indicate an
wire chemistry and divided into three groups according electrode and a combination of letters and digits. The
to the Mn levels as shown in Table 4.1. first two classes, namely EL12 and EM12K, are identical
In the classification system, the prefix E designates to those same classifications in AWS A5.17-97. They are
an electrode. The letters L, M and H indicate low, included here, because they are sometimes used with
medium and high Mn contents respectively. The letter alloyed fluxes to deposit weld metals of low-alloy steel
K, that appears in some classes indicates that the compositions. In the rest of the classes, the prefixes after
electrode is made from a heat of silicon-killed steel (as E agree with those used in AWS A5.5-80, "Specification
is obvious from the Si levels). The digits 8, 12, etc., for low-alloy steel covered arc welding electrodes" for
indicate in points, the nominal carbon content of the those chemical compositions, which are similar. Suffix
electrode (for example, 8 means 0.08%). N is to be used additionally, when the electrode is
Standard electrode diameters in mm are 1.6,2.0,2.5, intended for welds in the core belt region of nuclear
3.2,4.0, 5.6 and 6.4. reactor vessels. Such an electrode must have lower limits
Since SAW electrodes are mechanically driven at high of phosphorus, vanadium and copper as follows:
speeds, they need to be correctly tempered. The standard
states, "the temper of the electrodes shall be such that P= 0.012% max. V= 0.05% max. Cu= 0.08% max.
the electrodes will feed in an uninterrupted manner in
automatic and semi-automatic welding equipment." It is not allowed to be copper-coated. EF5 and EW
SAW electrodes are supplied in the form of coils with classes are not permitted to carry the suffix N.
or without liner. When supplied without liners, the It must be pointed out that this standard deals with
dimensions and net weight are decided by mutual solid electrodes as well as flux-cored electrodes, the
agreement between purchaser and supplier. Coils with latter being referred to as composite electrodes. In the
liners must meet the standard requirements given in classification system, letter C is to be used after E to
Table 4.2. indicate a composite electrode. Composite electrodes
The usage of drums for supply of SAW wire is are classified by the chemical composition of the weld
gaining popularity, especially in the pipe-mill segment. metal they deposit with a specific flux, unlike in the case
No specific dimensions are mandated for drums. of solid electrodes. This is done, because there is no
The cast and helix of the electrodes should be such standard method for analyzing the composite electrode
that they will feed in an uninterrupted manner in the itself. The standard designations for composite
welding equipment. The electrode in each coil and drum electrodes are shown in Table 4.4. The weld metal
should be of one continuous length of material. Butt joint compositions indicated herein are also applicable to the
welds are permitted, but they must be made and solid electrodes classified in Table 4.3. Additional suffix
finished in such a manner that they do not interfere with N can be used in the same way as with solid electrodes.
Table 4.1: Composition requirements for electrodes as per AWS A5.17
Electrode Chemical composition, wt. per cent3

classification Mn Si Cub
Low manganese steel electrodes

EL8 0.10 0.25/0.60 0.07 0.035 0.035 0.35
EL8K 0.10 0.25/0.60 0.10/0.25 0.035 0.035 0.35
EL12 0.05/0.15 0.25/0.60 0.07 0.035 0.035 0.35
Medium manganese steel electrodes

EM12 0.06/0.15 0.80/1.25 0.10 0.035 0.035 0.35
EM12K 0.05/0.15 0.80/1.25 0.10/0.35 0.035 0.035 0.35
EM13K 0.07/0.19 0.90/1.40 0.35/0.75 0.035 0.035 0.35
EM15K 0.10/0.20 0.80/1.25 0.10/0.35 0.035 0.035 0.35
High manganese steel electrodes

EH14 0.10/0.20 1.70/2.20 0.10 0.035 0.035 0.35
a. Single values are maximums.

b. The copper limit includes any copper coating that may be applied to the electrode
Table 4.2: Coil with liner—standard dimensions and weights as per AWS A5.17
Electrode dia. Net wt. of coil Inside dia. of liner Width of coil, max. Outside dia. of coil, max.
mm kg mm mm mm
11 65 420
1.6 to 6.4 23 305 ± 3 120 430
27
45
2.4 to 6.4 68 as mutually 125 800
91 agreed
Standard electrode sizes (except for an additional a smooth finish, free from surface imperfections,
size of 2.8 mm) and requirements for cast, helix and corrosion products, grease or other substances, which
winding are the same as for carbon steel electrodes. would adversely affect the quality of the weld or the
operation of the welding equipment. If the wire has a
3) BS 4165:1971, "Specification for electrode wires and copper coating, it must be a uniform, well bonded,
fluxes for the submerged-arc welding of carbon steel and me- smooth coating applied over a thoroughly clean surface.
dium-tensile steel" The copper content of the coated wire expressed as a
This standard specifies requirements for electrode percentage of the wire plus the coating must not exceed
wires, solid or cored (i.e. flux-cored), for the SAW of 0.4% by weight. Unless otherwise agreed between the
carbon steel and medium-tensile steel. purchaser and the manufacturer, the tensile strength of
For wires, no chemical requirements are specified. the wire must not exceed 930 N / m m 2 for carbon steel
The foreword of the standard suggests that this has been and must not exceed 1,080 N / m m 2 for medium-tensile
done in order not to inhibit developments in flux-wire steel.
combinations. The standard wire sizes are (in mm): 1.6, The wire m u s t be s u p p l i e d in coils h a v i n g
2.0,2.5,3.2,4.0,5.0,6.0,6.3 and 8.0. The wire must have dimensions and weights given in Table 4.5. The wire
Table 4.3: Composition requirements for solid electrodes as per AWS A5.23
Electrode Chemical composition wt. per cent a - b - c

classifi-
cation C Mn Si S P Cr Ni Mo Cu V Al Ti Zr
Carbon steel
EL12 0.05/0.15 0.25/0.60 0.07 0.035 0.035 — — — 0.35 — — — —
EM12K 0.05/0.15 0.80/1.25 0.10/0.35 0.035 0.035 — — — 0.35 — — — —
Carbon-molybdenum steel
EA1 0.07/0.17 0.65/1.00 0.20 0.035 0.025 — — 0.45/0.65 0.35 — — — —
EA2 0.07/0.17 0.95/1.35 0.20 0.035 0.025 — — 0.45/0.65 0.35 — — — —
EA3 0.10/0.18 1.65/2.15 0.20 0.035 0.025 — — 0.45/0.65 0.35 — — — —
EA4 0.10/0.18 1.25/1.65 0.20 0.035 0.025 — — 0.45/0.65 0.35 — — — —
Chromium-molybd enum steel
EB2 0.07/0.15 0.45/0.80 0.05/0.30 0.030 0.025 1.00/1.75 — 0.45/0.65 0.35 — — — —
EB2H 0.28/0.33 0.45/0.65 0.55/0.75 0.015 0.015 1.00/1.50 — 0.45/0.65 0.30 0.20/0.30 — — —
EB3 0.07/0.15 0.45/0.80 0.05/0.30 0.030 0.025 2.25/3.00 — 0.90/1.10 0.35 — — — —
EB5 0.18/0.23 0.40/0.70 0.40/0.60 0.025 0.025 0.45/0.65 — 0.90/1.10 0.30 — — — —
EB6 0.10 0.40/0.65 0.20/0.50 0.025 0.025 4.50/6.00 — 0.45/0.65 0.35 — — — —
EB6H 0.25/0.40 0.75/1.00 0.25/0.50 0.030 0.025 4.80/6.00 — 0.45/0.65 0.35 — — — —
Nickel steel
ENil 0.10 0.75/1.25 0.05/0.25 0.010 0.010 0.15 0.80/1.20 0.30 0.35 — — — —
ENi2 0.10 0.75/1.25 0.05/0.25 0.010 0.010 — 2.10/2.90 — 0.35 — — — —
ENi3 0.13 0.60/1.20 0.05/0.25 0.012 0.012 0.15 3.10/3.80 — 0.35 — — — —
ENi4 0.12/0.19 0.60/1.00 0.10/0.30 0.020 0.015 — 1.60/2.10 0.10/0.30 0.35 — — — —
Other low-alloy steel
EF1 0.07/0.15 0.90/1.70 0.15/0.35 0.025 0.025 — 0.95/1.60 0.25/0.55 0.35 — — — —
EF2 0.10/0.18 1.70/2.40 0.20 0.025 0.025 — 0.40/0.80 0.40/0.65 0.35 — — — —
EF3 0.10/0.18 1.70/2.40 0.20 0.025 0.025 — 0.70/1.10 0.45/0.65 0.35 — — — —
EF4 0.16/0.23 0.60/0.90 0.15/0.35 0.035 0.025 0.40/0.60 0.40/0.80 0.15/0.30 0.35 — — — —
EF5 0.10/0.17 1.70/2.20 0.20 0.010 0.010 0.25/0.50 2.30/2.80 0.45/0.65 0.35 — — — —
EF6 0.07/0.15 1.45/1.90 0.10/0.30 0.015 0.015 0.20/0.55 1.75/2.25 0.40/0.65 0.35 — — — —
CO
EM2 0.10 1.25/1.80 0.20/0.60 0.010 0.010 0.30 1.40/2.10 0.25/0.55 0.25 0.05 0.10 0.10 0.10 c
09
EM3 0.10 1.40/1.80 0.20/0.60 0.010 0.010 0.55 1.90/2.60 0.25/0.65 0.25 0.04 0.10 0.10 0.10 m
EM4 0.10 1.40/2.80 0.20/0.60 0.010 0.010 0.60 2.00/2.80 0.30/0.65 0.25 0.03 0.10 0.10 0.10 3D
©
EW 0.12 0.35/0.65 0.20/0.35 0.040 0.030 0.50/0.80 0.40/0.80 — 0.30/0.80 — — — —
EG No requirements specified >
30
O
a) Single values are maxima, b) The copper limit includes any copper coating, which may be applied to the electrode, c) The EB6 classification is similar to, but not identical with, the ER502
classification in A5.9-81, "Specification for corrosion resisting chromium and chromium-nickel steel bare, and composite metal cored and stranded welding electrodes and welding rods". m
CO
Table 4.4: Composition requirements for weld metal (both solid and composite electrodes)
Electrode Chemical composition wt. per cent a - b ' c

classifi-
cation C Mn Si S P Cr Ni Mo Cu W, Ti, Zr
(total)
Carbon-molybdenum steel
Al 0.12 1.00 0.80 0.040 0.030 — — 0.40/0.65 0.35 —
A2 0.12 1.40 0.80 0.040 0.030 — — 0.40/0.65 0.35 —
A3 0.15 2.10 0.80 0.040 0.030 — — 0.40/0.65 0.35 —
A4 0.15 1.60 0.80 0.040 0.030 — — 0.40/0.65 0.35 —
Chromium-molybdenum steel
B2 0.15 1.20 0.80 0.040 0.030 1.00/1.50 — 0.40/0.65 0.35 —
B2H 0.25 1.20 0.80 0.040 0.030 1.00/1.50 — 0.40/0.65 0.35 0.30 V
B3 0.15 1.20 0.80 0.040 0.030 2.00/2.50 — 0.90/1.20 0.35 —
B4 0.12 1.20 0.80 0.040 0.030 1.75/2.25 — 0.40/0.65 0.35 —
B5 0.18 1.20 0.80 0.040 0.030 0.40/0.65 — 0.90/1.20 0.35 —
B6 0.12 1.20 0.80 0.040 0.030 4.50/6.00 — 0.40/0.65 0.35 —
B6H 0.25 1.20 0.80 0.040 0.030 4.50/6.00 — 0.40/0.65 0.35 —
Nickel steel
Nil 0.10 1.60 0.80 0.030 0.030 0.15 0.80/1.10 0.35 0.35 0.05
Ni2 0.128 1.60 0.80 0.030 0.030 — 2.00/2.90 — 0.35 —
Ni3 0.12 1.60 0.80 0.030 0.030 0.15 2.80/3.75 — 0.35 —
Ni4 0.14 1.60 0.80 0.030 0.030 — 1.40/2.10 0.35 0.35 —
Other low-alloy steel
Fl 0.12 0.70/1.50 0.80 0.040 0.030 0.15 0.90/1.70 0.55 0.35 —
F2 0.17 1.25/2.25 0.80 0.040 0.030 — 0.40/0.80 0.40/0.65 0.35 —
F3 0.17 1.25/2.25 0.80 0.040 0.030 — 0.70/1.10 0.40/0.65 0.35 —
F4 0.17 1.60 0.80 0.040 0.030 0.60 0.40/0.80 0.25 0.35 0.03
F5 0.17 1.20/1.80 0.80 0.030 0.030 0.65 2.00/2.80 0.30/0.80 0.50 —
F6 0.14 0.80/1.85 0.80 0.030 0.030 0.65 1.50/2.25 0.60 0.40 —
Ml 0.10 0.60/1.60 0.80 0.040 0.030 0.15 1.25/2.00 0.35 0.30 0.03
M2 0.10 0.90/1.80 0.80 0.040 0.030 0.35 1.40/2.10 0.25/0.65 0.30 0.03
M3 0.10 0.90/1.80 0.80 0.030 0.030 0.65 1.80/2.60 0.20/0.70 0.30 0.03
M4 0.10 1.30/2.25 0.80 0.030 0.030 0.80 2.00/2.80 0.30/0.80 0.30 0.03
W 0.12 0.50/1.60 0.80 0.040 0.030 0.45/0.70 0.40/0.80 — 0.30/0.75 —
G No requirements specified
a) Single values are maxima.
b) The designation for composite electrodes is obtained by placing an EC before the appropriate weld metal classification in the table above.
c) The copper limit includes any copper in the coating on the electrode the weld was deposited with.
must be wound evenly in one continuous length and Each coil or reel of wire and its outer packing must
should be freefromkinks, waves, sharp bends or twists, state (1) the name of the manufacturer, (2) trade
so that it is free to unwind without restriction. The designation of wire, (3) size and weight, (4) batch
starting end of the wire on each coil must be clearly number, and (5) recommendations for special storage,
indicated. if required.
Wire must be suitably packed to guard against dam- The standard details the tensile and impact
age, including that during transportation. When stored, requirements for flux-wire combinations. These have
the wire should be kept in the original bundle or pack- been discussed later.
age in a dry storeroom and tinder such conditions the Since there is no British standard giving chemical
wire shall, for a period of at least 12 months, be capable compositions of SA wires, the welding companies in
of giving results similar to those that it would have given that country usually use IIS/IIW symbdls with the
on the date of its dispatch from the manufacturer. modification that SD is substituted for SA. Typical wires
marketed by Oerlikon of UK are listed in Tables 4.6 and This standard prescribes the requirements of solid
4.7 with their compositions: filler wires for submerged-arc welding of structural
steels in the yield strength range of 270-490 N / m m 2 and
4) Doc. IIS/IIW-545-78 "Classification and symbolization the ultimate tensile strength range of 300-690 N/mm 2 .
of bare steel wire electrodes and fluxes for submerged-arc weld- The wires are classified based on their chemical
ing of structural steels" composition as shown in Table 4.8. They comprise
Table 4.5: Dimensions of coils and weights of wire as per BS 4165
Outer Inner diameter Inner diameter Width Weight

diameter of bound coil with of bound max., mm of wire
max., mm no former mm former, mm max., kg
300 200+5 200+1 90 13

-o
-o
420 300+5 300+1 80 25
-0
-o
420 300+5 300+1 100 30
-o
420 300+5 300+5 115 30
-0 -0
760 610±10 610±10 100 70

760 610±10 610±10 100 100
760 610±10 610±10 150 100
of carbon-manganese submerged are wires (main elements only) marketed in UK

Table 4.6: Typical chemical analysis <
Weight %
Wire
C Mn Si S (max.) P (max.) A l (max.) N (max.)
S2ULTRA 0.12 1.1 0.15 0.015 0.015 0.02 0.012

SD3 0.09 1.6 0.25 0.015 0.015 0.02 0.012
S4ULTRA 0.10 2.0 0.15 0.015 0.015 0.02 0.012
Total S and P 0.02 wt% maximum
Table 4.7: Typical chemical analysis of low-alloy wires (wt.%) marketed in UK

Wire C Mn Si Cr Ni Mo S (max.) P (max.) A l (max.) N (max.)
S2Mo 0.10 1.0 0.15 — — 0.5
SD2 2Ni 0.10 1.1 0.2 — 2.2 0.015 0.015 0.015 0.02 0.012
SD2 3Ni 0.10 1.0 0.2 — 2.7 0.015 0.015 0.02 0.012
SD2 1 Ni Cr Mo 0.11 1.1 0.2 1.0 0.1 0.5 0.015 • 0.015 0.02 0.012
SD2 1 1/4 Cr Mo 0.10 1.1 0.2 1.5 0.6 0.015 0.015 0.02 0.012
SD2 2 1/2 Cr Mo 0.10 1.0 0.2 2.4 1.0 0.015 0.015 0.02 0.012
SD2 5 Cr Mo 0.09 0.6 0.4 5.0 0.5 0.015 0.015 0.02 0.012
SD3Mo 0.10 1.5 0.2 — 0.5 0.015 0.015 0.02 0.012
SD3 INi 0.10 1.4 0.2 — 1.2 0.015 0.015 0.02 0.012
SD3 lNi 1/2 Mo 0.10 1.5 0.2 — 1.3 0.5 0.015 0.015 002 0.012
TIBOR 22* 0.09 1.2 0.10 — — 0.3 0.015 0.015 0.02 0.010
*Ti—0.04% B—0.003% **Total S and P 0.02 Wt. % maximum.
Note: Refer "Toughness Properties ofS. A. welds for applications of TIBOR 22 wire in this chapter.
Table 4.8: Chemical composition requirements for finished wire electrodes in SAW as per IIS/IIW -545
Symbol AWS C Si Mn P S Mo Ni
SA1 EL12 0.06-0.15 <0.15 0.35-0.65 — —
SAIL EL8 <0.10 <0.15 0.35-0.65 — —
SA2 EM12 0.07-0.15 <0.15 0.70-1.20 — —
SA3 0.07-0.15 <0.25 1.30-1.70 — —
SA4 EH14 0.08-0.18 <0.25 1.70-2.25 — —
SA6 0.08-0.18 <0.35 2.80-3.20 <0.03 <0.03 — —
SAISi 0.06-0.14 0.15-0.50 0.35-0.65 — —
SAlLiSi EL8K <0.10 0.15-0.50 0.35-0.65 — ■ —
SA2Si EM12K 0.07 0.15 0.15-0.50 0.70-1.20 — —

SA2LSi EM5K <0.10 0.15-0.50 0.70-1.20 — —
SA2HSi EM15K 0.15-0.20 0.15-0.50 0.70-1.20 — —
SA2Si2 EM13K 0.07-0.15 0.50-0.80 0.70-1.20 — —
SA3Si 0.07-0.15 0.25-0.50 1.30-1.70 — —
SAlMo 0.06-0.14 <0.15 0.35-0.65 0.45-0.65 —
SA2Mo EA1/EA2 0.07-0.15 < 0.15 ' 0.70-1.20 0.45-0.65 —
SA3Mo 0.07-0.15 <0.25 1.30-1.70 0.45-0.65 —
SA4Mo ED1/EA3 0.08-0.18 <0.25 1.75-2.25 0.45-0.65 —
SA5Mo ED2 0.08-0.18 <0.30 2.25-2.80 0.45-0.65 —
SA6M0 0.08-0.18 <0.35 2.80-3.20 0.45-0.65 —
SA3MoSi 0.07-0.15 0.15/0.50 1.30-1.70 0.45-0.65 —
SA3MoSiNil EF1 0.07-0.15 0.15/0.50 1.30-1.70 <0.02 <0.02 0.45-0.65 0.80-1.60
SA4Mo Nio.5 EF2 0.08-0.18 <0.25 1.75-2.25 <0.02 <0.02 0.45-0.65 0.40-0.80
SA4MoNil EF3 0.08-0.18 <0.25 1.75-2.25 <0.02 <0.02 0.45-0.65 0.80-0.60
SA2M1 ENil 0.07-0.15 <0.15 1.70-1.20 <0.02 <0.02 — 0.80-1.60
SA2N12 ENi2 0.07-0.15 <0.15 1.70-1.20 <0.01 <0.01 — 1.60-2.80
SA2NJ3 ENi3 0.07-0.15 <0.15 1.70-1.20 <0.01 <0.01 — 2.80-3.80
Note: Copper, including the
zopper
c coating, if present, should not exceed 0.35. A maximum of 0.2% chromium is
recommended
carbon steels, C-Mn steels and low-alloy steels. The table of structural steels in the yield strength range of 275-
also gives the equivalent AWS classes w h e r e v e r 490 N / m m 2 and the ultimate tensile strength range of
applicable. 335-685 N / m m 2 . The classifications covered by the
The wire diameters are (in mm): 1.2,1.6,2.0,2.5,3.20, standard are given in Table 4.10 with their chemical
4.0, 5.0, 6.3, and 8.0. Standard dimensions and weights compositions, and they include C-steel, C-Mn steels and
of coils are given in Table 4.9. low-alloy steels. Standard dimensions of rims and coils
Reeling conditions and conditions of wire finish are are given in Table 4.11.
similar to those mentioned in the AWS and British In the classification, 'A' indicates the wire for
Standards. The Cu content of the copper-coated wire submerged-arc welding and 'S' stands for solid bare
should not exceed 0.3% by weight. The tensile strength wire. The digit that follows is equal to twice the Mn
of the wire should not exceed 930 N / m m 2 for carbon content of the wire. The presence of alloying elements
steel and 1,080 N / m m 2 for medium tensile steel, unless is indicated by further suffixes like Mo, Si, Ni, etc.
otherwise agreed between the manufacturer and the
purchaser. FLUXES
SAW flux performs the same functions as the coating
5) IS:7280-1974, " Specification for bare wire electrodes of a manual electrode; additionally, it must satisfy
for submerged arc welding of structural steels" certain special conditions demanded by the nature of
In preparing this standard, assistance has been the process. The flux protects the molten pool and the
derived from Doc. IIS/IIW-385-72, which is an earlier arc against atmospheric oxygen and nitrogen by creating
version of the IIS/IIW document described above. It an envelope of molten slag. The slag also cleanses the
prescribes the requirements of solid wires for the SAW weld metal (i.e. deoxidizes it and removes impurities
Table 4.9: Wire dimensions and weights as per IIS/IIW-545

Coil type Wire dia. Coil wt. Coil width Inside dia. Outside dia.
mm kg mm mm mm
10 4.0 m a x . 10 a p p r o x . 50+5 200!05 250 approx.
25 1.2 m i n . 25 a p p r o x . 100tf 300^ 450 approx.
50 2.0 m i n . 50 a p p r o x . 100+10 300^5 600 approx.
50 3.20 m i n . 100 a p p r o x . 120 +10 600

:020 800 approx.
Table 4.10: Chemical composition of filler wires as per IS 7280
IS Chemical composition%
Classification
Carbon Silicon Manganese Molybdenum Nickel Phosphorus Sulphur
(max.) (max.)
AS-1 0.10 (Max.) 0.03 (Max.) 0.40-0.60 0.03 0.03

AS-1 Si 0.08-0.15 0.10-0.40 0.40-0.60 — — 0.03 0.03
AS-2 0.08-0.15 0.05-0.15 0.80-1.20 — — 0.03 0.03
AS-2Si 0.08-0.15 0.15-0.40 0.80-1.20 — ■ —
0.03 0.03
AS-2Mo 0.08-0.15 0.05-0.15 0.80-1.20 0.45-0.60 — 0.03 0.03
AS-2Ni 0.06-0.12 0.05-0.15 0.80-1.20 — 0.80-1.20 0.02 0.02
AS^3 0.08-0.15 0.05-0.25 1.30-1.70 — — 0.03 0.03
AS-3Mo 0.08-0.15 0.05-0.25 1.30-1.70 0.45-0.60 — 0.03 0.03
AS-3M0Ni 0.08-0.12 0.05-0.25 1.30-1.70 0.45-0.60 1.30-1.70 0.02 0.02
AS-4 0.08-0.16 0.05-0.25 1.80-2.20 — — 0.03 0.03
AS-4Mo 0.08-0.17 0.15-0.30 1.80-2.20 0.45-0.60 — 0.03 0.03
AS-6 0.08-0.17 0.20-0.30 2.80-3.20 — — 0.03 0.03
AS-6M0 0.08-0.17 0.15-0.30 2.80-3.20 0.45-0.60 — 0.03 0.03
' coating, if present, must not exceed 0.4%.

Note: Copper, including the coppei
Table 4.11: Dimensions of rims and coils as per IS 7280

Rims Coils
Internal diameter Width Net weight, Internal diameter Net weight,
mm mm max. kg mm max. kg
300 ± 5 100 ± 3 12.5 ± 1 300 ± 25 12.5 ± 1
300 ± 5 100 ± 3 25 ± 2 300 ± 25 25 ± 2
600 ± 5 100 ± 3 80 ± 5 600 ± 25 80 ± 5
such as sulphur), modifies its chemical composition and relatively small welding zone. This results in a deeply
controls the profile of the weld bead. Like the manual penetrating arc, which makes narrower and shallower
electrode coating, the SA flux can incorporate alloying welding grooves practicable, thus reducing the amount
elements, so that in combination with an unalloyed wire, of weld metal required to complete the joint. It also
it will yield suitably alloyed weld metal. The molten results in higher welding speeds. The properties of the
slag also provides favourable conditions for very high flux enable SA welds to be made over a wide range of
current densities, which, together with the insulating welding currents, voltage and speeds, each of which can
properties of the flux, concentrate intense heat into a be controlled independently of the other. Thus, one can
obtain welded joints of desired shape, chemistry and to the required size; sieved, sized and packed
mechanical and metallurgical properties by using an in bags or drums.
appropriate welding procedure. • Agglomerated fluxes: For producing an agglomer-
The SA flux should be so formulated that it does not ated flux, finely powdered ingredients are mixed
evolve appreciable amounts of gases under intense heat and ground dry in a mixer. The mix is steadily
of the welding zone. It should be granular in form and moistened by spraying with a solution of alka-
should be capable of flowing freely through the flux line silicate and the mixing is continued. The
feeding tubes, valves and nozzles of standard welding mixer blades are suitably designed to assist ag-
equipment. Its particle size should be controlled. glomeration. The silicate solution initially fills the
The flux in its solid state is a non-conductor of spaces between the pores of the particles. When
electricity, but w h e n molten it becomes a highly subsequently dried, the water evaporates, leav-
conducting m e d i u m . It is, therefore, necessary to ing the binder as bridges between particles. Like
initiate the arc by special means, which have been low-hydrogen electrodes, agglomerated fluxes
described in "Welding Arc." Once the arc is struck and have to be baked at high temperatures to remove
the surrounding flux becomes molten, the welding the last traces of water clinging to the silicate
current continues to flow across the arc, while the arc binder. After baking, the flux is graded to a speci-
provides a c o n d u c t i n g p a t h of moljten flux as it fied granule size by sieving, and packed in wa-
advances. The flux contains elements capable of terproof containers.
assisting in the initial striking of the arc a n d of
• Sintered fluxes: They are produced by grinding the
stabilizing it after initiation.
dry charge together, pressing into small balls, and
heating to 1,000-1,100°C (just below melting
Manufacture
point) in gas-fired furnaces. The solid mass pro-
The welding characteristics of a flux are determined
duced then has the characteristics of a fused flux.
to a considerable extent by its method of manufacture.
It is crushed to the desired fineness, sieved, sized
Hence, it is important to know the general techniques
and packed in suitable containers.
of flux production.
Today two main types of SA fluxes are available, Classification
depending on the method of manufacture: fused and Like electrode coatings, SA fluxes h a v e been
agglomerated. Sintered fluxes have been produced in s t a n d a r d i z e d by classifying t h e m based on their
some countries b u t their properties and w e l d i n g chemical c o m p o s i t i o n s . Doc. IIS/IIW-545-78,
characteristics h a v e been rarely r e p o r t e d in the "Classification and syrr^bolization of bare steel wire
literature. While the first two types will be dealt with at electrodes and fluxes for submerged arc welding of
great length in this chapter, only brief production details structural steels", in this chapter, has classified and
will be given about sintered fluxes. symbolized fluxes as shown in Table 4.12. The two
lettered symbols are preceded by a letter, F, B or M
• Fused fluxes: Typical ingredients are quartz, indicating the method of manufacture as follows:
manganese ore or slag, dolomite, potassium F — Fused
feldspar and clay. These minerals are ground
B — Bonded
and mixed in a definite proportion and melted.
Melting is carried out in a magnesia or fireclay M— Mechanically mixed.
crucible if heating is by gas or in a graphite cru- The symbols MS, CS, ZS, AR, AB and FB are to be
cible if electricity is used. Electric melting is applied to all non-alloying fluxes. Non-alloying fluxes
started by heating with an arc between the elec- are defined as those which contain any amount of man-
trode and crucible and continued by resistance ganese and silicon as deemed appropriate; if other al-
heating with the electrode submerged in the loying elements are present, they should not be assimi-
molten flux, which is an electrical conductor. lated into the weld metal to an extent greater than 0.25%
For small charges, induction heating in a graph- of any one element or more than 0.40% of the total of all
ite crucible is used. After fusion, the melt is alloying elements other than manganese and silicon.
solidified rapidly by quenching into water or Fluxes with metallic additions not meeting this re-
pouring into cold steel chills to produce small quirement are classified as special type fluxes and carry
fragments of flux. These are dried and crushed the symbol ST.
A more practical and useful classification is Through letters A and F, the method by which each
given below. It includes all the IIW classes except type of flux is manufactured is indicated. Table
ZS and ST. Zirconium (Zr) silicate type flux is 4.13 gives composition ranges of various flux types
excluded because it is scarcely used. It is also felt along with their chemical character and basicity
that a separate class is not strictly necessary to index range. The term basicity has been explained
indicate that the flux contains alloying additions. under "Agglomerated Fluxes."
F = Fused A = Agglomerated
Flux Types
r I
Mn - Silicate Ca - Silicate Aluminate Flouride -
Type Type Type basic Type
F A
I
X 1
Aluminate Aluminate
T i 0 2 Type Basic T y p e
A/F A/F
High Mec i u m Low
Silica Silica Silica
F F/'A A
FLUX CLASSIFICATION
Table 4.12: Symbols indicating main chemical components in the flux as per IIS/IIW-545
Symbol Main chemical components Flux type

MS MnO + Si0 2 50% min. Mn silicate
CS CaO + MgO + Si0 2 60% min. Ca silicate
ZS Zr02+Si02 30% min. Zr silicate
AR A1203 + Ti0 2 45% min. Alumina-rutile
AB A1203 + CaO + MgO 45% min. Alumina-basic
A1 2 0 3 20% min.
FB CaO + MgO + MnO + CaF2 50% min.
Si0 2 20% max. Basic fluoride
CaF 2 15% min.
ST Fluxes with metallic additions Special type
Table 4.13: Composition ranges of various flux types

Type Character Composition ranges expressed as oxides Basicity
index
Si0 2 TiO, ZrO. A1 20 00,3 MnO CaO MgO CaF 2
Mn-silicate Acid 30-40 00.10 35-45 00.10 — 00.10 0.5-1.0
Ca-silicate high silica Acid 50-60 0.3-0.5 — 25-30 0.8-12 00.02 0.5-0.8
Ca-silicate medium silica Neutral 30-40 10-15 00-05 25-30 10-15 0.5-10 0.8-1.2
Ca-silicate low silica Basic 30-40 03-05 — 35^5 00-05 10-15 1.2-1.7
Aluminate Ti0 2 Neutral 10-20 05-15 00-10 20-50 10-20 00-05 — 05-15 0.5-1.0
Aluminate basic Basic A1 2 0 3 + CaO + MgO 45% min. A1 2 0 3 = 20% min. 1.2-2.0
Fluoride basic Basic 05-12 10-15 10-15 30-40 20-30 2.5-3.5
Fused Fluxes in melting temperature. This may explain why high-

Fused fluxes are usually complex silicates with a high silica Ca-Silicate type flux can withstand very high
percentage of Si0 2 , which makes them acidic in welding currents. MnO is excluded in fluxes meant for
character. While the formulation is based largely on high current applications, but is a welcome addition in
empirical approach, certain broad chemical properties applications requiring high fluidity such as high speed
are made use of. For example, in the case of Mn-Silicate longitudinal welding of line pipe. When MnO is around
type flux, the composition can be carefully balanced 35î5%, the resulting flux becomes a Mn-Silicate type,
using the MnO-Si02 phase diagram shown in Fig. 4.10 which is well suited for the deposition of butt and fillet
as a guide to proper slag composition. Melting point of welds at moderate currents and high speeds. It is also
the flux is an important property. tolerant to rust.
The flux must remain molten until after the weld Fused fluxes have excellent chemical homogeneity
metal has solidified, so that the weld bead has a good and are unaffected by moisture. Their main
surface appearance. To ensure this, the melting point of disadvantage is the inability to take up deoxidizers or
the flux must be controlled at about 1300°C This is often ferro-alloy additions due to the high temperatures
done by adding another oxide to a binary-oxide-system. involved in their processing, which result in their
For example, the melting range of CaO.Si02 can be segregation and high losses. A further explanation for
reduced from 1544°C to less than 1300°C by adding this behaviour is the chemically unstable nature of MnO
approximately 10% A1203. and Si0 2 , i.e. they are easily reducible by alloying
Viscosity control is another important aspect of flux elements. When one or both of these oxides predominate
formulation. The slag formed during welding must be a flux, the resultant slag contains a high concentration
sufficiently viscous to prevent the ingress of atmospheric of molecular oxygen, which causes the oxidation of the
oxygen and nitrogen, and to prevent it from running alloying elements. The products of the reaction are
away from the molten weld metal. At the same time, metallic Mn or Si and oxides of the alloying elements,
the viscosity must be low enough to allow the gases to the former being added to the weld metal and the latter
rise from the molten puddle as it cools down. Silica- dissolving into the slag. Relations between Si0 2 and
rich slags show high levels of viscosity at all MnO contents in the flux and Si and Mn pick-up by the
temperatures, which is why they give good weld surface weld metal are depicted in Figs. 4.11 and 4.12. Fig. 4.12
quality and absence of undercut even at very high indicates a rapid pick-up of Si when Si0 2 content of the
welding currents (up to 2,000 amps). However, they flux exceeds 40%. The oxidation of alloying elements
need to be chemically balanced to attain the slag also results in increased amounts of non-metallic
viscosity of 200 poises in the temperature range of 1,000- inclusions in the weld metal. High levels of Si and non-
1,200°C and of around five poises at 1,400°C. Fluorspar metallic inclusions are responsible for the low toughness
(CaF2) is the most effective reagent for reducing of weld metals obtained from fused fluxes.
viscosity, and most commercial fused fluxes contain 5
to 7% of it. Higher levels of fluorspar have to be avoided Agglomerated Fluxes
because they affect arc stability. Titania, manganese The agglomeration technique of manufacturing
oxide (MnO) and lime are also effective in reducing fluxes has led to the development of fluxes with low
viscosity, but lime addition also results in a marked rise Si02 content and high CaO, MgO and CaF2 contents.
LIQUID
.u 1800b / T W O IMMISCIBLE\ H 3272
7
LIQUIDS \^ ft
CRISTOBOLITE + LIQUID D
TRIDYMITE + LIQUID
<
£1400| 2552 UJ
2 RHODONITE + LIQUID Q.
5
RHODONITE + TRIDYMITE
100i -L. ■ * ■ 1832
PERCENT 20 I 40. [ 60 10 20 30 40
MnO 2Mn0-Si0 2 MnO-Si0 2 Si0 2 Si0 2 IN FLUX-%
Fig. 4.10: Phase diagram of MnO-Si02 system Fig. 4.11: Relation of Si02 in flux to Si in weld deposit
oxygen and sulphur contents of the weld metal and

higher its notch toughness value. The oxygen contents
in welds obtained from different flux types are shown
in Fig. 4.13. Agglomerated basic fluxes of good
performance characteristics have maximum basicity
between 2.6 and 3.2. Basicity higher than 3.2 makes the
flux unworkable, mainly because of very high
solidification range and very low viscosity. A
compromise is achieved through additions of acidic
oxides. Appreciable CaF2 additions are also made to
lower the melting point. When CaF2 dissociates in the
arc, the resulting CaO neutralizes the more active Si02
through the reaction CaO + Si02 = CaSi03.
Metallic deoxidizers and ferro-alloys can be
10 20 30 40 50
incorporated into agglomerated basic fluxes, since low
MnO IN FLUX-% temperatures are involved in their manufacture. In
Fig. 4.12: Relation of MnO in flux to Mn in weld deposit addition, the chemical stability of CaO and MgO
combined with the basic nature of the slag ensures
efficient transfer of the alloying elements into the weld
metal. Use of alloyed flux (also termed active flux)
CaO and MgO are chemically stable oxides. The ratio
demands a closer control of the welding parameters,
of these oxides along with other basic oxides to acidic
especially the arc voltage, to ensure consistent weld
oxides like Si02 and A1203 is termed as basicity.
metal chemistry. An increase in arc voltage increases
Basicity of a molten flux =
the flux-wire consumption ratio and thereby results in
Basic oxides CaO+MgO+CaF2+ Na20+K20+V*(MnO+FeO) increased alloy transfer to the weld metal.
Acidic oxides Si0 2 + V*(A120_ + TiQ2 + ZrQ2) Agglomerated fluxes, normally baked at 500-600°C
have a tendency to pick up moisture when stored under
Basicity index range for each flux type is shown in humid atmosphere due to the hygroscopic properties of
Table 4.13. The higher the flux basicity, lower are the CaO and MgO. Experience has shown that by baking the
0.15 KEYS
H-HIGH DILUTION WELD
# L-LOW DILUTION WELD
ARABIC NUMERALS INDICATE
NO.OFWELD ANALYSES
o.io h
i
0.05
# # * ?6
0F15F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F16 F17 F18 F13
HIGH Si02. MEDIUM Sit>2. LOW Si02
CALCIUM SILICATE .MAWGANESE SILICATE , ALUMINIUM . J A S I C .OTHER
Fig. 4.13: Ranges of 0 2 content from welds with different fluxes

flux at 800°C during manufacture, the flux becomes level rises to 1.0% and Si to 0.45% and then levels off
insensitive to moisture. Then metallic component after the sixth p a s s . With h i g h e r M n w i r e s ,
additions cannot be made to such fluxes as they will be corresponding rises are registered, but the highest Mn
oxidized. Hence, the trend today is to use 800°C-baked, level reached is approximately 1.8%, and in all cases, a
mineral (i.e. unalloyed) fluxes and alloyed wire for the high Mn/Si ratio is maintained. This high Mn/Si ratio
welding of high tensile and low-alloy steels, so that the c o m b i n e d w i t h very l o w levels of non-metallic
hydrogen-induced cracking tendency is minimized. It has inclusions confer excellent notch toughness properties
also been established that with such combinations, on weld metals of basic agglomerated fluxes.
changes in arc voltage do not affect weld metal chemistry. Agglomerated fluxes have lower bulk density than
A g g l o m e r a t e d flux b a k e d at 800°C w o u l d be fused fluxes. This h a s the a d v a n t a g e that u n d e r
expected to crumble due to the formation of complex identical conditions of current and voltage, less flux
carbonates by chemical reaction between the silicate is melted a n d forms a slag as compared to fused
binder and atmospheric C0 2 . But this has not been found fluxes. Comparative figures for agglomerated flux
to occur. The explanation p u t forth is that the Si0 2 c o n s u m p t i o n a n d fused flux c o n s u m p t i o n for
resulting from the same reaction strengthens the bond different current and voltage settings are indicated
as in the case of C 0 2 sands for foundry moulding, and in Fig. 4.15.
this bond retains its strength at elevated temperatures. A misconception exists in some quarters that an
The basic nature of agglomerated fluxes enables one agglomerated flux separates out and consistent results
to control Si and Mn pick-up. The pick-up of these cannot be obtained. A team of German researchers has
elements in six successive layers in the case of a popular concluded, after extensive research that, even in
agglomerated flux combined with three grades of wire multilayer welds with as many as 75 runs, the weld
is shown in Fig. 4.14. It is seen that with the inexpensive deposit reveals a very consistent deposit analysis in
rimmed quality steel wire (0.50% Mn, 0.03% Si), the Mn various layers.
1.80
^ ^ (WIRE WITH Mn 1.5%
1.70 1 i
i
" | Si 0.1%)
1.60 ^ \ (WIRE WITH Mn

1
I
1 j 1.0%. Si 0.1%)
1.50 1 1y
1 1
#
y
1.40 1
1 1
J
z 1.30 1 i 1
UJ 1
i 1
1.20 i 1
1
O
U 1.10 l / r 1
i
1 1
t (WIRE WITH Mn 1
1.00 i j 0.6%. Si p. 1%)
<
0.90 i I
z 0.80 1 i
< i i 1
5
i
0,70 1 | (Mn IN BASE PLATE
i 1 0.5%)
0.60 1
1
1
i
0.50
i !
1ST 2ND 3RD 4TH 5TH 6TH 1
0.80
1 (WIRE WITH Mn
0.70 1 3 - 4 l . 5 % Si 0.1%)
0.60
1 1 T (WIRE WITH Mn
1 1
z 1 [l.0%, SiO 1%) I
o 0.50 ""—{(WIRE WITH Mn
u 0.40 I 1
jO.5%. Si0 1%)
z
o
a
0.30
1 1
1 1
1
j (Si IN BASE PLATE
•0.26%)
I
_L
0.20 1 1
1 1
1 I
\ 1
1 2 3 4 5 6 NUMBER OF LAYERS
Fig. 4.14: Mn and Si recovery with various agglomerated fluxes and wires of various compositions
microstructure. The free Al to form A1N particles may

become available from the reaction:
K
UJ 600
A1A + 3 Fe -> 2 Al + 3 FeO
560
< 520 The level of non-metallic inclusions in the welds
o generated from aluminate fluxes is lower than in the
480 FUSED FLUX A i case of acidic fluxes, but higher than those produced by
X
440 / | 49%! highly basic fluxes.
u. / \ * 1
u. 400
O / 52<H{ i
Particle Size
/32 i ! /
Z
o 360 Particle size is an important property of a SA flux.
£
5
D
</>
320
280 S
4 ,
1 ^
^2%i^ ^^ G L 0 M E H A T E 0 FLUX
1
During welding, a finer flux produces a denser and
compact burden over the flux cavity, thereby reducing
Z
o |24013%l its volume. Evolution and escape of gases from the weld
o
AMPERES 400 500 600 700 800 bead then becomes more difficult and thus results in a
VOLTS 28 30 32 34 36
characteristic aerated slag. Finer fluxes are also less
SPEED
ICM/MIN. 45 45 45 45 45 | tolerant to oil and rust, because of the inability of the
Fig. 4.15: Comparative consumption of fused and resultant gases to escape. A large increase of sub-size
agglomerated fluxes particles in a flux can give rise to a humped peaky weld
bead with increased penetration.
Fluxes for currents higher than 800 amps or for multi-
The use of agglomerated fluxes affords the greatest wire applications are supplied to a finer grain size than
advantage for hard facing applications. Since ferro- similar fluxes for lower current levels. Using coarse flux
alloys can be very easily incorporated into agglomerated at very high current can result in considerable arc
fluxes and significant alloying effects obtained, deposits instability and inferior weld bead surface. The reasons
of varying hardnesses and wear-resistant properties can for such interrelation between flux particle size and
be obtained by using a simple mild steel wire in welding current cannot be accurately explained but may
combination with variously alloyed agglomerated have something to do with the arc cavity behaviour. At
fluxes. H o w e v e r , w i t h fused fluxes, alloy w i r e s high currents, there is increased flux melting and gas
invariably have to be used, which are expensive and evolution while the viscosity of the slag is very low;
not readily obtainable. under these conditions a coarse, low density flux burden
Basic agglomerated fluxes tend to produce a highly would cause violent eruptions in the are cavity which
fluid slag of low viscosity above the current range of sometimes would rise above the flux cover, leading to
800-900 amps, which tends to flood the arc cavity the defects mentioned above.
(shown in Fig. 4.1) and extinguish the arc, and to flow
away from the joint, thus affecting the bead shape. Metal Transfer
Hence, while some fused fluxes can operate even at How does metal transfer take place in SA welding?
1,500-2,000 amps; maximum current level for basic To get at the answer, researchers have used various
agglomerated fluxes is 800 amps. novel techniques including the use of a consumable
ceramic sight tube under the flux, pointing along the
Aluminate Fluxes weld axis towards the arc to allow its direct high-speed
Aluminate type fluxes came into limelight in 1969, photography. One such study revealed that at low
when Tuliani and Boniszewski established that fluxes currents, large droplets formed on the electrode tip,
high in A1203, (40 to 50%) produced a microstructure of which was transferred to the weld pool via the slag at
good notch toughness in as-deposited weld metal as the periphery of the arc cavity. Above a critical current
opposed to other flux compositions. A theoretical (for example, 400 amps at 31C V for a 3 mm wire size
explanation for this is that small particles of aluminum w i t h electrode positive), the d r o p s began to be
nitride (A1N) are present in the molten metal in a sub- transferred directly through the arc cavity into the weld
microscopic state, which during solidification restrict pool. At 450 amps, the arc forces deformed the electrode
austenite grain growth, thus promoting the formation tip to a point from which liquid metal and slag particles
of fine acicular ferrite, which is observed in the flowed down to form spherical drops, which transferred
directly to the weld puddle. The detachment of droplets a gas-filled arc cavern into the molten weld pool. It is,
appeared to be mainly due to electromagnetic forces therefore, likely that the droplet lifetimes and volumes
rather than due to gases released at the arc tip (for are similar to those estimated for open arc systems, and
example, CO generated by the reaction between the that the submerged arc system differs principally in
carbon in the wire and the oxygen coming from MnO having larger volume and longer duration weld pools.
and Si0 2 inclusions). Much interest has centered around the composition of
Though flux formulation has been found to exert the gases in the cavern during submerged-arc welding,
little influence on the mechanism of metal transfer because of the influence they will have on slag-metal
compared to current density, differences in their surface reactions. Recent work has demonstrated that this gas
tension and electrical conductivity do have some effects. is principally carbon m o n o x i d e w h e n using
For example, basic fluxes having high CaF 2 content and agglomerated basic fluxes, and that the ratio P co /P co
generating little oxygen in the arc have higher electrical decreases with increasing basicity. Slag-metal reactions
conductivity and tend to give a smaller droplet rate but in submerged-arc welding thus resemble those in the
greater droplet mass. In addition, such a flux changes MMA process more closely than might have been
from metal transfer through the slag to free flight imagined and the Fe-C-O reaction plays an important
transfer at a lower current than a fused flux. role. As with covered electrodes, the reactions of silicon
In cases where alloyed fluxes are used, the recovery and manganese have received most attention in such
of elements from them is higher when metal transfer studies, and several attempts have been made to apply
takes place through the slag than when it is free flight; mass action theory. Efforts have also been made to
on the other hand, when alloyed wires are used, the quantify the reactions of oxygen, silicon and manganese
recovery of elements from such wire is higher with free in order to devise classification systems for submerged-
flight transfer than with transfer through the slag. This arc fluxes and wires.
may explain why sometimes the recovery of elements
with a given flux formulation changes with different Classification of Fluxes
welding parameters. Like SAW wires, fluxes are better understood by
The mechanism of metal transfer has been vividly studying the classification systems given in the relevant
observed by using X-ray cinematography in the same standards.
manner as high-speed cine method has been used for AWS A5.17-97, referred to on page 198, specifies a
the open arc systems. Figures 4.16 and 4.17 show classification system for SA fluxes based on the
longitudinal and transverse sections of the arc region mechanical properties of the weld metal they deposit
as revealed by X-ray cinematography, which confirm with a specific classification of electrode given in Table
that slag-coated metal droplets are transferred through 4.1. The system is explained further.
FLUX
i-rV ?v I*.: 1 •" • 7 £*;i V "7:/ • *"• •".*» * •'• - •"' *• - • V*. «• - * ; c
SLAG
50 mm 4(3 ^ B A S E METAL SURFACE

-ARC CAVITY
MOLTEN POOL' r
MELTING SURFACE
, SOLIDIFICATION SURFACE
I - 700 A
U«30V
DIRECTION OF WELDING
V-30CM/M1N
Fig. 4.16: Longitudinal section during SAW showing arc cavity and typical dimensions
ELECTRODE-
(A)
FLUX SLAG MOLTEN STEEL
ELECTRODE
(B)
(C)
Fig. 4.17: Schematic diagrams showing mechanism of metal transfer: (A) Low current, transfer through slag,
droplet travels towards the earth; (B) Higher current, smaller droplets, transfer through arc cavity; (C) High current,
very small drops of metal and slag run down pointed electrode
Based on the results of the impact tests of the weld lower than 70,000 psi and Charpy-V notch impact
metal, the appropriate digit from Table 4.15 is inserted strength of at least 27 J at -51 °C, when deposited with
in the flux classification of Table 4.14. Weld metal from an EM12K electrode under the conditions called for in
a specific flux-electrode combination that meets this specification.
impact requirements at a given temperature also T h o u g h the specification states that this is a
meets the requirements at all higher temperatures in classification system for fluxes, it is in reality a system
Table 4.15 (i.e., weld metal meeting the requirements for flux-wire combinations.
for digit 5 also meets the requirements for digits 4, 2, AWS A5.23-97, (see in this chapter), gives a similar
0, and Z). system of classification for SA fluxes suitable for welding
Example: F7A6-EM12K is a complete designation. It low-alloy steels when used in combination with the
refers to a flux that will produce weld metal, which in electrode wires classified in Table 4.3. Refer the diagram
the as-welded condition, will have a tensile strength no for the detail.
CLASSIFICATION SYSTEM FOR SA FLUXES

SUITABLE FOR CARBON STEEL WELDING
Indicates flux.
Indicates the minimum tensile strength (in increments of 10,000 psi [69 MPa]) of weld metal with the flux and
some specific classification of electrode deposited according to the welding conditions specified herein (see
Table 4.14).
Designates the condition of heat treatment in which the tests were conducted: "A" for as-welded and "P"
for postweld heat-treated. The time and temperature of the PWHT are specified as 610-635°C for I hour.
Indicates the lowest temperature at which the impact strength of the weld metal referred to above
meet' or exceeds 20 ft. lb (27 J). (See Table 4.15).
I Indicates electrode
FXXX-EXXX
Classification of the electrode used in depositing the weld metal referred to above (see Table 4.1).
Table 4.14: Mechanical property requirements as per AWS A5.17

Yield strength
Flux Tensile strength min.2 Elongation
classification1 psi MPa psi MPa %min2
F6XX-EXXX 60,000-80,000 430-560 48,000 330 22
F7XX-EXXX 70,000-95,000 480-660 58,000 400 22
1
The letter "X" used in various places in the classification in this table stands for, respectively, the condition of heat treatment, the toughness of the weld
metal, and the classification of the electrode as explained in the above diagram. 2 Yield strength at 0.2% offset and elongation in 2 in. (51 mm) gage length.
CLASSIFICATION SYSTEM FOR SA FLUXES

SUITABLE FOR LOW-ALLOY STEEL WELDING
Indicates flux.
i— Indicates the, minimum tensile strength (in increments of 10,000 psi [69 MPa]) of weld metal with the flux and
some specific classification of electrode deposited according to the welding conditions specified. Two digits are
used for weld metal of 100,000 psi (690 MPa) tensile strength and higher (See Table 4.16).
I Designates the condition of heat treatment in which the tests were conducted: "A" for as-welded and "P" for
postweld heat-treated. The time and temperature of the PWHT are specified (Time 1 hour, temperature as in
Table 4.18).
I Indicates the lowest temperature at which the impact strength of the weld metal referred to above meets
or exceeds 20 ft. lb (27 J). (See Table 4.17).
I Indicates electrode.
I Indicates composite electrode. Omission of "C" indicates solid electrode.
I Classification of the electrode used in depositing the weld metal referred to above (see Table
4.3).
I Indicates the chemical composition of the weld metal. One or more letters or digits are
used (see Table 4.4).
FXXX — ECXXXN — XN
I 1 The 'N' is used only when limits for P, V and Cu are satisfied as pointed
out in this chpter and footnote b of Table 4.17 applies.
Table 4.15: Impact strength requirements as per AWS A5.17

Test temperature
Digit °C Energy level min.
Z — No impact requirement
0 -18
2 -29
4 -40
5 -46 27J
6 -51
8 -62
Example: F9P0-EB3-B3 is a complete designation. It with an EB3 electrode under the conditions called for
refers to a flux that will produce weld metal, which in in this specification. The composition of the weld metal
the post-weld heat treated condition, will have a tensile will be B3 (see Table 4.4).
strength higher than 90,000 psi and Charpy-V notch BS 4165-1971, earlier in the chapter, classifies flux-
impact strength of at least 27 J at -18°C when deposited wire combinations based on tensile and impact values.
Table 4.16: Mechanical property requirements as per AWS A5.23

Yield strength
Flux Tensile strength min.a Elongation
classification psi MPa psi MPa % min.a
F7XX-EXXX-X 70,000-95,000 480-660 58,000 400 22
F8XX-EXXX-X 80,000-1,00,000 550-700 68,000 470 20
F9XX-EXXX-X 90,000-1,10,000 620-760 78,000 540
FIOXX-EXXX-X 1,00,000-1,20,000 690-830 88,000 610 16
FllXX-EXXX-X 1,10,000-1,30,000 760-900 98,000 680 15b
F12XX-EXXX-X 1,20.000-1,40,000 820-970 1,08,000 750 14b
a. Yield strength at 0.2% offset and elongation in 2 in. (51 mm) gauge length.
b. Elongation may be reduced by one percentage point for FllXX-EXXX-X and F12XX-EXXX-X weld metals in the upper 25% of the
strength range.
Table 4.17: Impact strength requirements8 as per AWS A5.23

Digit Test temperature °C Energy level min.
Z — No impact requirement
0 -18
2 -29
4 -40
5 -46
6 -51 27J
8 -62
10 -73
15 -101
a. Based on the results of the impact tests of the weld metal, the manufacturer shall insert in the classification (Table 4.16) the appropriate digit from
the table above as indicated in the diagram.
Weld metal from a specific flux-electrode combination that meets the impact requirements at a given temperature also meets the requirements at
all higher temperatures in this table (i.e., weld metal meeting the requirements of digit 5 also meets the requirements of digits 4, 2, 0, and Z).
b. Weld metals with the "N" suffix shall also have a Charpy V-notch energy level of at least 102] at 21°C.
The standard makes a distinction between values IIS/IIW-545-78, "Classification and symbolisation of
obtained by multi-run welding technique and those bare steel wire electrodes and fluxes for submerged-arc
obtained by two-run welding technique (indicated by welding of structural steels", referred to on page 225
letters M and T respectively). and page 229, provides the classification system for a
The classification system is explained by Table 4.19. flux-wire combination as follows:
IS: 3613-1974, "Acceptance tests for wire-flux com- 1) Letter F, B or M indicating the m e t h o d of
binations for submerged-arc welding of structural steels". manufacture of the flux (In Chapter 4.)
This standard is largely based on BS: 4165-1971, 2) Two letters MS, CS, ZS, AR, AB, FB or ST
except for the fact that the prefix S is replaced by the indicating main chemical composition of the flux (see
prefix SA, and the group SX T 450 is modified and Table 4.12)
extended as follows: 3) Two letters SA indicating a wire electrode for the
SAW process (see Table 4.8).
Class Impact valu eat °C 4) Combination of digits and letters indicating the
SA 1 T 450 20 chemistry of the wire (see Table 4.8)
SA 2 T 450 0 5) A two-digit number indicating the tensile strength,
SA 3 T 450 -20 min.
IT 47 J yield strength and elongation of the weld metal as given
SA 4 T 450 -40 in Table 4.20
6) A one-digit number indicating the impact strength
Also for the group SA X M 450, the IS has specified of the weld metal as given in Table 4.21
47 J instead of 45 J minimum impact value. 7) Number 0 or a combination of a digit and a letter
Table 4.18: Temperatures for preheat, interpass, and postweld heat-treatment of test assemblies*
Weld metal Preheat and interpassb Postweld heat-treatmentc

classification temperature, temperature
° F ± 25 °C ± 14 ° F ± 25 °C ± 14
Al
A2 300 149 1150 621
A3
A4
B2 300 149 1150 621
B2H 300 149 1150 621
B3 400 204 1275 691
B4 400 204 1275 691
B5 300 149 1150 621
B6 600 316 1600 732
B6H 600 316 1600 871
Nil 300 149 1150 621
Ni2 300 149 1150 621
Ni3 300 149 1125 607
Ni4 300 149 1100 593
Fl 300 149 1150 621
F2 300 149 1150 621
F3 300 149 1150 621
F4 300 149 1050d 566d
F5 300 149 1050d 566d
F6 300 149 1050d 566d
Ml
M2
M3 300 149 1125d 607d
M4
W 300 149 1150d 621d
G 300 149 1150 621
a. These temperatures are specified for materials tested and classified under this specification and are not recommendations for production use. The
specific requirements for production welding must be determined by the user. They may or may not differ from those called for here.
b. The preheat and interpass temperatures listed here shall be used for the test assemblies regardless of whether the materials are classified in the as-
welded or postweld heat-treated condition. They are required for purposes of uniformity and may or may not be indicative of those that might be
satisfactory for fabrication of any particular weldment. The fabricator must determine for himself what is required there.
c. Materials classified in the postweld heat-treated condition shall be heat-treated for 1 hour at the temperature shown for each classification. No
welding shall be done after the heat treatment.
d. These classifications normally are used in the as-welded condition.
indicating the impact strength obtained and the plate 1) Symbol: FCS-SA3-51-2-1B
thickness used in the two-run technique as given in A fused calcium-silicate-type flux used with a wire
Table 4.22. This is optional and must be used only where electrode SA3 reached a tensile strength of 520 N /mm2
applicable. and an impact value of 35 J at 0°C in an all-weld-metal
The following three examples will help to better test and an impact value of 32 J at+ 20°C in a two-run
explain the coding of flux-wire combinations: test (plate thickness 20 mm).
Table 4.19: Coding of flux-wire combination as per BS 4165

Class T.S. range Y.S. min. Eln% Impact value at °C
N/mm 2 N/mm 2 min.
SIM 300 400-600 300 22 20 1
S2M300 400-600 300 22 0 Imin. 35 J
S3M300 400-600 300 22 -20 f
S4M300 400-600 300 22 -40 J
SI M 350 460-650 350 22 20
1 min. 40 J
S2 M 350 460-650 350 22 0 I
Jf
S3 M 350 460-650 350 22 -20
S4M350 460-650 350 22 -40
SI M 450 520-700 450 19 20
1
S2M450
S3 M 450
520-700
520-700
450
450
19
19
0
-20 f
I
min. 45 J
S4M450 520-700 450 19 -40 J
SIT 300 400-600 300 22 20
1
S2T300 400-600 300 22 0 I
min. 35 J
S3T300 400-600 300 22 -20 f
S4T300 400-600 300 22 ^40 J
SIT 350 500-650 350 22 20
1
S2T350
S3 T 350
500-650
500-650
350
350
22
22
0
-20 f
I
min. 40 J
S4T350 500-650 350 22 -40 J
SI T 450 • 550-700 450 19 20 min. 45 J
S2T450 550-700 450 19 0 >
(Note: Tensile specimens are heated at 200/250°C for 8/16 hours.)
Table 4.20: Symbol for tensile properties as per IIS/IIW-545

Symbol Tensile strength* Yield stress* Elongation
N / mm2 (min.) N / mm2 (min.) (L = 5d) %
43 430-510 360 22
51 510-570 380 22
57 570-650 420 20
65 more than 650 480 18
*Upper limit tolerance: + 40 N/mm2
Table 4.21: Symbol for impact value as per IIS/IIW-545

Symbol Min. impact value of 28 J (ISO-V-notch)«jt°C
0 Not required
1 +20
2 0
3 -20
4 -30
5 -40
2) Symbol: BAB-SA2-43-2-2C 3) Symbol: BFB-SA2-51-4-3A-SA2Ni2-57-5

A bonded aluminate-basic-typefluxused with a wire A bonded basic-fluoride-type flux used with a wire
electrode SA2 reached a tensile strength of 460 N/mm 2 electrode SA2 reached a tensile strength of 530 N/mm2
and an impact value of 40 J at -20°C in an all-weld-metal and an impact value of 36 J at -30°C in an all-weld-metal
test and an impact value of 39 J at 0° C in a two-run test test and an impact value of 38 J at -20°C in a two-run
(plate thickness 36 mm). test (plate thickness 12 mm). The same flux used with a
Table 4.22: Symbol for impact values (two-run technique) as per IIS/IIW-545
Symbol Plate thickness Min. impact value of 28 J
mm (ISO-V-notch) at °C
0 not required
1A 11-13 +20
IB 20-26
1C 35 min.
2A 11-13
2B 20-26 +0
2C 35 min.
3A 11-13
3B 20-26 -20
3C 35 min.
4A 11-13
4B 20-26 -30
4C 35 min.
5A 11-13
5B 20-26 -40
5C 35 min.
wire electrode SA2Ni2 reached a tensile strength of 580 composition as the electrode is melted and deposited
N / m m 2 and an impact value of 60 J at -40°C in an all- as weld metal. Some fluxes add deliberate alloying
weld-metal test. elements such as columbium and molybdenum; others
are very active in the sense that they deplete significant
CONSUMABLES FOR STAINLESS STEEL WELDING amounts of certain elements that are readily oxidized,
AWS A5.9-81, "Specification for corrosion riesisting such as chromium. Other fluxes are less active and may
c h r o m i u m a n d c h r o m i u m - n i c k e l steel: b a r e a n d contain small amounts of alloys to offset any losses and
composite metal cored and stranded arc welding thereby p r o d u c e a weld deposit with a chemical
electrodes and welding rods" covers corrosion or heat- composition close to the composition of the electrode.
resisting chrome and chrome-nickel steel wires for use If the flux is active or alloyed, changes in the welding
with submerged-arc or gas metal-arc welding. The parameters, particularly voltage, will result in significant
specification includes steels with 4% minimum Cr and changes in the chemical composition of the deposit.
50% maximum Ni. In the classification of these wires, Higher voltages produce greater flux/metal interactions
prefixes ER are used to indicate that they can be used and, for example, in the case of an alloy flux, greater
both as electrodes in SAW and MIG welding, and as alloy pickup.
rods in TIG welding. These prefixes are followed by a "The range of change of various alloying elements
three-digit number such as 308, 316 and 410, similar to for several commercial fluxes and the corresponding
the n u m b e r i n g system for S.S. m a n u a l electrodes changes in FN are shown in Table 4.23.
indicated in Table 3.36. "Due to the variability in the change as shown in
In SAW, these wires undergo a change in chemistry Table 4.23, when close control of ferrite content is re-
due to interaction with the flux. Since in stainless steel quired, the effects of particular flux should be evalu-
welding, the weld metal chemistry vis-a-vis its ferrite ated before any production welding is undertaken."
content (ferrite n u m b e r — FN) have to be closely The specification also gives details of finish, cast and
controlled, one must have a clear understanding of how helix as also dimensions and weights of coils and spools.
fluxes influence the weld deposit composition. The
following note from the Appendix of A5.9-81 explains CONSUMABLES FOR NI AND NI-ALLOY WELDING
this: Standard Ni and Ni-alloy electrode wires are covered
"Submerged arc welds show variable gains a n d / o r in AWS A5.14-76, "Specification for nickel and nickel-
losses of alloying elements depending on the flux used. alloy b a r e w e l d i n g r o d s a n d e l e c t r o d e s " . The
All fluxes p r o d u c e some changes in the chemical classification system consists of prefixes ER followed
Table 4.23 : Variation of alloying elements and FN

Typical change from Corresponding
Element wire to deposit change in FN
Carbon Varies: on "L" grades usually a gain, + 0.01 to + 0.02%; on "L" -1 to -2
regular grades usually a loss, up to -0.02%
Silicon Always a gain: + 0.3 to + 06% +1 to + 2
Chromium Varies: -3.0 to + 1.0% -6 to + 4
Nickel Usually a loss: -3.0 to -1.0% +1 to + 3
Manganese Varies: -0.5 to + 0.5% -0.5 to + 0.5
Molybdenum Little change a deliberate addition is made to the flux
Columbium Usually a loss unless a deliberate addition: -0.2 to -0.5% up to -1
by the chemical symbol Ni to identify the electrodes as 3 and Ch. 11). The consumables are either alloyed wires
nickel base alloys, and by additional chemical symbols used in combination with neutral fluxes or low-carbon
indicating the principal alloying elements of the unalloyed wires used in combination with alloyed
classification. fluxes. There is a limitation on the use of alloyed
The wires covered in this specification are meant for hardfacing wires, because rolling and drawing of such
either of the following processes: submerged-arc, gas- hardenable alloys is not simple. The trend therefore has
shielded arc (i.e. TIG/MIG) and plasma-arc. Out of the been to use a simple low-carbon low-Mn or medium-
18 classifications covered, only the following six are Mn wire (SI or S2) and alloyed fluxes to get weld metal
indicated in the Appendix as being used with the SA compositions containing higher Mn level and varying
process: percentages of Mo, Cr and V. Such deposits have a
hardness range of 250-500 brinell. Typical compositions
Class Approx. chemistry are given in Table 4.24.
ERNiCu-7 66Ni/Cu bal. For getting weld deposits of higher alloy content,
ERNiCr-3 70 Ni/20 Cr/2.5 Cb and in fact for the entire hardfacing range, tubular flux-
ERNiCrFe-5 75 Ni/15 C r / 8 Fe cored wires suitably designed for SAW are now
ERNiCrFe-6 70 Ni/15 C r / 8 Fe available. The technology of flux-cored wire
ERNiMo-3 25Mo/5Cr/5Fe/Nibal. manufacture has extended the use of the SA process in
ERNiCrMo-3 60 Ni/22 C r / 9 Mo/3.5 Cb hardfacing applications. Hardfacing wires will be
discussed in detail in the chapter on 'Flux-cored Wire
Description and application of some of the above Welding'.
alloys or of those closely allied to them are given in
Chapter 3.
The specification also gives details of finish, cast and PROCESS PARAMETERS
helix as also dimensions and weights of coils and spools. In SAW, the weld deposit quality is determined by
the type of flux, grade of wire and the following
CONSUMABLES FOR HARDFACING parameters:
SAW is often used for hard facing, i.e. for building a) Welding current
up wear-resistant surfaces on metallic parts (see in Ch. b) Arc voltage
Table 4.24: Typical compositions of SA hard facing alloys

Weld deposit chemistry (%) Brinell value of
C Mn Si Cr Mo V weld deposit
0.10 1.94 0.60 — 0.76 — 250
0.10 2.25 0.65 — 0.76 — 300
0.10 2.10 0.63 1.06 0.54 — 350
0.13 2.02 0.65 1.78 0.36 0.18 400
0.19 2.22 0.72 2.69 0.60 0.31 450
0.22 2.26 0.85 2.85 1.10 0.32 500
c) Speed of arc travel 3) Lowering the voltage produces a stiffer arc needed
d) Size of electrode for getting penetration in a deep groove and to resist
e) Electrode stick-out arc blow on high-speed work.
f) Heat input rate. 4) An excessively low voltage produces a high,
To get optimum results, one must know their effects narrow bead with poor slag removal.
and how to select and control them properly. c) Speed of arc travel
a) Welding current For a given combination of welding current and
It controls the melting rate of the electrode and voltage, increase in the welding speed or the speed of
thereby the weld deposition rate. It also controls the arc travel results in lesser penetration, lesser weld
depth of penetration and thereby the extent of dilution reinforcement and lower heat input per unit length of
of the weld metal by the base metal. Too high a current weld. Excessively high travel speeds decrease fusion
causes excessive weld reinforcement, which is wasteful, between the weld deposit and the parent metal, and
and burn-through in the case of thinner plates or in badly increase tendencies for undercut, arc blow, porosity and
fitted joints, which are not provided with proper irregular bead shape. As the travel speed is decreased,
backing. Excessive current also produces a high narrow penetration and weld reinforcement increase. But too slow
bead and undercut. Excessively low current gives an a speed results in poor penetration, because under this
unstable arc, inadequate penetration and overlapping. condition, the weld puddle is directly under the electrode
SAW equipment is usually provided with an ammeter tip and the force of the arc is cushioned by the weld puddle.
to monitor and control the welding current. Excessively slow speeds also produce a convex bead shape
b) Arc voltage that is amenable to cracking, cause the arc to break out of
Arc voltage, also called welding voltage, means the the flux layer which can hurt the operator's eye, and give
electrical potential difference between the electrode wire a large uncontrolled weld puddle, which results in an
tip and the surface of the molten weld puddle. It is uneven weld bead with slag inclusions.
indicated by the voltmeter provided on the equipment. d) Size of electrode
It hardly affects the electrode melting rate, but it As in the case of MMA welding, the electrode size is
determines the profile and surface appearance of the selected according to the plate thickness and the desired
weld bead. As arc voltage increases, the weld bead size of weld. With increase in electrode size, welding
becomes wider and flatter, and the penetration current can be increased to get higher deposition rates,
decreases. The effects of changing voltage are explained deeper penetration and increased weld size.
as follows: At a given welding current, changing over to a larger
1) Increasing voltage: electrode results in a wider, less penetrating bead.
a) Produces a flatter and wider bead Hence, in joints with poor fit-up, a larger electrode is
b) Increases flux consumption preferred to a smaller one for bridging the root gap. For
c) Increases resistance to porosity caused by rust or a given electrode size, a high current density (i.e.
scale welding current in Amps, divided by the cross-section
d) Helps bridge gaps when fit-up is poor of the wire, expressed in Amps./mm 2 ) results in a
e) Increases pickup of alloy from the flux: this can strong, penetrating arc, while a lower current density
be used to advantage to raise the alloy content of the gives a soft arc which is less penetrating.
weld when welding is performed with alloy or e) Electrode stick-out
hardsurfacing fluxes. If excessive, it can reduce ductility It is also termed electrode extension. It refers to the
and increase crack sensitivity, particularly in the making length of the electrode, between the end of contact tube
of multiple-pass welds. and the arc, which is subject to resistance heating (also
2) Excessively high voltage: called PR heating) at the high current densities used
a) Produces a hat-shaped bead that is subject to in the process. The longer the stick-out, the greater the
cracking amount of heating and the higher the deposition rate.
b) Produces poor slag removal in groove welds Increased electrode stick-out reduces to some extent
c) In multiple-pass welds, increases the normal alloy the energy supplied to the arc, resulting in lower arc
pickup from the flux voltage and a different bead shape. Hence, when the
d) Produces a concave fillet weld that may be subject electrode stick-out is increased to obtain higher
to cracking deposition rate, the voltage setting on the equipment
e) Increases undercut on fillet welds. must be increased to maintain correct arc length.
However, with longer stick-outs, the increase in consumption decreased by 13% as electrode extension
deposition rate is accompanied by a decrease in was increased with constant head voltage (Series 1), but
penetration. Hence, longer stick-outs must be avoided remained virtually static for a constant arc voltage
when deep penetration is desired. On the contrary, it (Series 2). Surprisingly the results for positive polarity
can be used with advantage when melt-through is a (Series 3) indicated a 15% increase in flux consumption
problem while welding thin sections. It can also be used with the same increase in stick-out.
to advantage in build-up and hard facing applications. Regarding weld bead shape, reinforcement was
It must be borne in mind that with long electrode stick- found to increase with increased electrode extension.
outs, it becomes more difficult to maintain the electrode Regarding weld metal properties, increased electrode
tip correctly in line with the joint. extension at a constant head voltage (and energy)
Maximum electrode stick-outs recommended are: resulted in an increase in yield (12%) and tensile
75 mm for 2.0, 2.4 and 3.2 mm wire diameters strengths (5%), but with inconsistent changes in Charpy
125 mm for 4.0, 4.8 and 5.6 mm wire diameters behaviour. With constant arc voltage (Series 2) changes
Deposition rates can be increased by as much as 2 5 - in strength were small and inconsistent, and the Charpy
50% by increasing electrode stick-out, but the technique transition behaviour deteriorated slightly with increased
is little used in industry. This may be due to the electrode extension and energy expenditure. Macro
following factors: (a) insulated guide is needed to direct examination did not reveal any defects in the weld.
the hot wire to the weld pool; (b) penetration is reduced f) Heat input rate
by about 10%, (c) excessive overheating of the wire leads Also termed arc energy, it is calculated by using the
to electrode pulsation, arc instability and stubbing. formula:
The effect of increasing electrode extension has been
experimentally s t u d i e d at Cranfield Institute of HIR = V x A x 60
Technology, using a standard controlled arc SA welding S x 1000
unit together with a 1,200 Amps, drooping characteristic where HIR = heat input rate in kilojoules per mm
rectifier. The results are shown in Table 4.25. In Series 1 (kj/mm)
the head voltage was kept constant, in Series 2 the arc V = arc voltage A = welding current
voltage was kept constant, while in Series 3 the polarity S = arc travel speed in mm/min.
of the electrode was changed to positive while the
welding parameters and energy expenditure are nearly SAW PROCEDURES
similar. SAW, semi-automatic and fully-automatic, is used
For a fixed energy expenditure, an increase in for making butt joints in the downhand position and
electrode extension from 40 to 120 m m gave 37% for making fillet welds in T and lap joints in the
increase in deposition rate for both negative and positive downhand and horizontal-vertical positions as shown
polarity. With a fixed arc voltage the increase was in Fig. 4.18. Normally this process cannot be used in
somewhat less, even t h o u g h energy e x p e n d i t u r e vertical and overhead positions, because of the difficulty
increased by 7%. For negative polarity, flux of pre-placing the flux.
Table 4.25: Influence of electrode extension on deposition rate and flux consumption
Electrode Deposition Flux
Series extension, Electrode Energy, Operating Current, rate, consumption,
mm polarity KJ/mm voltage Amps. kg/hr kg/hr
1 40 Negative 3.60 38 V head 600 11.05 11.88
80 3.60 12.49 10.62
120 3.60 15.12 10.26
2 40 Negative 3.32 33 V arc 600 11.66 10.10
80 3.41 13.28 10.12
120 3.60 15.12
3 40 Positive 3.24 36 V head 600 8.70 10.45
80 3.24 10.75 11.20
120 3.24 12.00 12.00
160 3.24 14.10 12.90
i£
Fig. 4.18: Joints and positions suitable for SAW
It is important to bear in mind that the SAW process Weld Backing Techniques
demands accurate edge preparation and fit-up. In The various commonly used techniques involve
MMAW, irregularities in this regard are taken care of use of the following: (1) base metal backing; (2)
by the m a n u a l welder, t h o u g h they do result in structural backing; (3) weld backing; (4) backing
increased welding time and a larger consumption of strip; (5) copper backing; (6) flux backing; (7) backing
electrodes. In SAW, on the other hand, the operation is tapes.
automatic, welding currents are high and the arc is
deeply p e n e t r a t i n g . Moreover, since the joint is 1) BASE METAL BACKING
submerged under the flux, the operator is unable to The root face is kept sufficiently thick as shown in
adjust the procedure to accommodate joint irregularities. Fig. 4.19, to support the weld pool without burn-
A poor fit-up in a butt joint can cause the granular flux through. This technique is used for square or partially
to spill through the root gap. It can also give rise to burn- bevelled butt joints, for fillet welds and for plug or slot
through and slag inclusions. welds. Care has to be taken that the root faces of groove
Shops u s i n g SAW are a d v i s e d to m a k e e d g e welds are in close contact. The first pass, deposited
preparations with automatic thermal cutting equipment sometimes with lower current, acts as a backing for the
(oxy-acetylene or plasma-arc), or by machining. In the second pass deposited with higher current to get
absence of such facilities, SAW becomes a slow and through penetration.
unproductive operation with frequent interruptions and
increased proportion of weld rectification. 2) STRUCTURAL BACKING
In SAW, the weld puddle is of large size and remains In certain cases where design permits, another
in a molten condition for a long time. The welding structural member can serve as a backing for the weld,
procedure must ensure that this molten p u d d l e is as shown in Fig. 4.20. It is very important that the contact
supported and contained until it has solidified at the surfaces of the joint are clean and the contact is intimate
root of the weld. This precaution is a must when full in order to avoid porosity and slag inclusions. The weld
joint penetration has to be achieved in a butt joint. The must also provide sufficient depth of fusion in the
technique used for this purpose is termed weld backing. backing member.
SECOND PASS
ACKING PASS!
^BACKING PASS
Fig. 4.19: Base metal backing for SAW Fig. 4.20: Structural backing for SAW
3) WELD BACKING
The backing weld is deposited at lower current and
with a moderately penetrating arc using the manual
metal arc, C0 2 -shielded arc or flux-cored arc process
I
(see Fig. 4.21). It may be in one or more passes to obtain ^ ^ (B)
sufficient depth to support the submerged-arc weld. The

backing weld may be retained in the joint if it is of Fig. 4.23: Copper backing for SAW: (A) V groove butt;
(B) Square butt
suitable quality. If otherwise, it may be removed by
oxygen or arc gouging, by chipping or by machining
For high production applications, the copper bar is
after the submerged-arc welds have been deposited. The
provided with internal water circulation to maintain it
resulting groove is filled up with a submerged-arc weld.
relatively cool. The bar is usually grooved as shown in
the figure to obtain weld reinforcement on the underside
4) BACKING STRIP
of the joint. It is important to ensure that the copper bar
The backing strip is of metal that is compatible with
has sufficient mass to prevent melting of the copper
the metal being welded. The weld metal fuses into the
material, which can result in contamination of the weld
backing strip, so that it becomes an integral part of the
with copper. It must be borne in mind that mechanical
joint as shown in Fig. 4.22. In this case, it is termed a
properties of steel weld metal deteriorate when the Cu
permanent backing. In case it is intended to be a
content exceeds a certain limit.
temporary backing, it may be removed finally by
machining. Suitable root opening must be kept to ensure
6) FLUX BACKING
full penetration. It varies between 1.6 and 4.8 mm,
depending on joint thickness. It is important that the As shown in Fig. 4.24, dry granular SA flux is placed
contact surfaces between the plates and the strip are in a trough on flexible sheet material. This sheet material
clean and the contact is intimate; otherwise porosity and rests on a rubberised canvas hose, which can be inflated
leakage of molten weld metal may occur. to hold the flux tightly against the back of the joint. This
technique will be discussed in detail while describing
the one-side SAW used in Japanese shipyards.
5) COPPER BACKING
Copper backing shown in Fig. 4.23 has several
7) BACKING TAPES
advantages. Its high thermal conductivity enables it to
Ceramic back-up tapes consisting of a ceramic
extract the heat rapidly from the molten weld pool. In
material on an aluminum foil backing are available from
addition, the molten steel weld metal does not fuse with
various sources. The exposed aluminium foil edges are
the copper material. Hence, it only serves as a temporary
covered with a pressure sensitive adhesive covered with
backing. The copper backing bar is either as long as the
a removable liner. Lengths of strips are 0.5 to 1.0 metre.
joint; or it is of short length and designed to slide
These can be easily applied to joints or seams to provide
underneath the travelling arc. In still other applications,
shielding or back-up for one-side welding and root pass
it may be in the form of a rotating wheel.
back-up for two-side welds to be deposited by TIG, MIG
and other arc processes. By using these tapes, arc
gouging and further backside joint operations such as
' • 'I &\ I

Fig. 4.21: Weld backing for SAW
T
PAPER
PLATE
FLEXIBLE INSERT
SHEET (OPTIONAL)
MATERIA
TROUGH
INFLATED
BACKING STRIP HOSE
Fig. 4.22: Backing strip for SAW Fig. 4.24: A method of providing flux backing for SAW
grinding are eliminated or minimised. They avoid the

use of expensive and clumsy fixtures, back-up bars and
gas purging of weld.
Butt Welds
To make a full penetration butt weld in sheet metal kS<//STEELBACK-UP
without burn-through, steel or copper backing bar must
be used. The joint is then completed with a single weld Fig. 4.25: Joint fit-up for butt welds in sheet metal
pass deposited from one side.
With copper backing, a square butt joint without root
gap is used. The procedure data are given in Table 4.26. Plates in the thickness range of 6.4-15.9 mm and with
Joint fit-up with steel backing is shown in Fig. 4.25 square edges butted together tightly, can be
which shows that a small root opening is helpful. The conveniently butt welded with two passes, one from
procedure data are given in Table 4.27. each side as shown in Fig. 4.26. The first pass deposited
Plates up to 12.7 mm thickness and with square edges at a lower current serves as a backing for the second
can be butt welded with a single pass using a steel pass. It is important that the two passes penetrate into
backing strip. It is advisable to keep a root opening, each other sufficiently to prevent lack of fusion and slag
because when the edges are butted together tightly, the inclusions in the central region. Procedure data are
resultant weld has a high build-up. Alternatively, a provided in Table 4.29.
groove can be provided. Procedure data are given in
Table 4.28. A>*
#
,*<•
cv°'
Table 4.26: Data for SA butt welds with copper backing
Plate
thickness
Electrode
dia.
Current
amps
Voltage
V
Speed
mm/sec. T SECOND PASS
t, mm m m Electrode + ve t
BACKING PASS
1.6 2.4 350 23 50 A.
2.0 2.4 400 24 42
2.4 3.2 500 30 40 Fig. 4.26: Square butt weld in two passes, one from
3.6 3.2 650 31 30 each side
Table 4.27: Data for SA butt welds with steel backing

Plate Root Electrode Current Voltage Speed
thickness opening dia. amps V mm/sec.
t, mm g, mm mm Electrode + ve
1.6 0-0.8 3.2 450 25 45
2.0 0-08 3.2 500 27 33
2.4 0-1.6 3.2 550 27 25
3.6 0-1.6 3.2 650 28 20
Table 4.28: Extension of Table 4.27 to thicker plates

Plate Root Electrode Current Voltage Speed
thickness opening dia. amps . V mm/sec.
t, mm g, mm mm Electrode + ve
4.8 1.6 5 850 32 15
6.4 3.2 5 900 33 11
9.5 3.2 56 950 33 10
2.7 4.8 5.6 1,000 34 8
The above-described procedure can be extended to passes of large cross-sections and considerable dilution
plates of 19 mm and 25.4 mm thicknesses by providing of the w e l d metal by the base metal. They are
60° V-grooves on both the sides and sufficiently large recommended for steels of good weldability having
root face as shown in Fig. 4.27. Procedure data are given low carbon equivalent and in cases where special impact
in Table 4.30. requirements for the weld metal are not specified.
When plate thickness increases further, it becomes For welding steels of difficult weldability, or where
necessary to increase the V-groove and deposit three stringent weld metal impact requirements are specified,
passes, one from the first side and two from the second procedures involving basic type of flux, multiple passes
side as shown in Fig. 4.28. Typical procedure data for of limited cross-sections deposited with low currents,
32 mm and 38 mm plates are given in Table 4.31. a n d m i n i m u m dilution by the base metal are
It must be pointed out that the above procedures are r e c o m m e n d e d . For plates of 16, 25.4 and 38 mm
valid for fused silicate type fluxes, which are capable of thicknesses, for example, the joint fit-up is made as
taking high welding currents. These procedures are very shown in Fig. 4.29. First two passes are deposited
economical and they result in minimum number of manually with a 4 m m basic low-hydrogen type
r—Tî >, *
TPASS
2ND V _\ /
T Y 4 »
/ 9.5 MM 7
12ND PASS V
x t-X / T
Y — /
9.5 MM
JfT
.19 MM
™- .• r , 4 M M yjy' I
i1ST PASS l\ \ I I 1 , 1M. nice / X.-V*
S - l f A - T * 3-2MM / 'STPASSÂJ „ MM
L
T -*—:—*£—>^
Fig. 4.27: Parameters for two-pass 19 mm and 25.4 mm t butt welds
Table 4.29: Data for two-pass square butt weld, one from each side
Backing pass Second pass
Plate Electrode Current Voltage Speed Electrode Current Voltage Speed
Thickness dia. amps V mm / sec. dia. amps V mm / sec.
t, mm mm mm
6.4 4.0 475 29 20 4.0 575 32 20
9.5 4.0 500 33 14 4.0 850 35 14
12.7 5.0 700 35 11 5.0 950 36 11
15.9 5.0 900 36 9 5.0 950 36 9
Table 4.30: Data for 19 mm and 25.4 mm 1I butt welds
19mmf 25.4 mm t
First pass
Electrode dia., mm 5 5
Current (DC+), Amps. 700 850
Voltage, V 35 35
Speed, mm/sec 12 5.5
Second pass
Electrode dia., mm 5 5
Current (DC+), Amps. 950 1000
Voltage, V 36 36
Speed, mm/sec 6 7
:T7f7CvA
^ ^ Y
Fig. 4.28: Parameters for three-pass 32 mm and 38 mm t butt welds
6 4 MM i
Fig. 4.29: Joint fit -up for multi-pass butt weld
Table 4.31: Data for 32 mm and 38 mm t butt welds
Plate First pass Seconc pass

Thickness Electrode Current Voltage Speed Electrode Current Voltage Speed
t, mm dia. mm amps V mm/sec dia., mm amps V mm/sec.
32 5 850 35 5.5 5 1,000 36 5
38 5 1,000 36 4 5 1,000 36 4
Third pass
Electrode Current Voltage Speed
Dia., mm amps V mm/sec
5 850 35 4
5 950 34 3
electrode. With these passes serving as a backing, SA because then it is easier to get a symmetrical flat-faced
weld passes are deposited at a speed of 7 mm/sec using weld with sound root penetration. In addition, a weld of
4 mm diameter electrode, 550 Amps., 28 V. The number much larger size (19 mm maximum as against 8 mm maxi-
of SA passes for 16,25.4 and 38 mm thick joints are 5,12 mum in the horizontal position) can be deposited in a
and 26 respectively. After the vee is filled up, the manual single pass. Normally, the electrode is positioned in such
weld at the root is completely gouged out and the groove a way that its centre-line passes through the root. For
is filled up with a SA pass. increased penetration, the work can be posit-ioned at 60°
and the electrode may be slightly tilled as shown in Fig.
Fillet Welds in Tee Joints 4.30 (b). Procedure data are given in Table 4.32.
Flat position: Whenever possible, one must deposit a Welds of larger sizes require multiple passes as
fillet weld in the flat position as shown in Fig. 4.30 (a), shown in Fig. 4.31. They are deposited with 5 mm
Table 4.32: Data for downhand fillet weld in tee joint

Weld size Electrode Current Voltage Travel speed
L, nun dia. (DC-) V mm/sec
mm Amps.
4.8 3.2 525 31 23
6.4 4 625 31 17
8.0 5 725 34 11
9.5 5 800 36 9
12.7 5 875 38 6
15.9 5 925 39 4
19.0 5 975 40 3
CORRECT EXCEPTION
CENTRE LINE OF ANGLE OF TILT

WELDING ELECTRODE
CENTRE LINE OF
AND JOINT
WELDING ELECTRODE
CENTRE LINE^
OF JOINT
OFF-CENTRE
DISTANCE
Fig. 4.30: Downhand SA fillet weld in tee joint: (A) Normal alignment; (B) Special alignment for deeper penetration
Fig. 4.31: Arrangement of passes in downhand fillet welds

diameter electrode at 900-1,000 Amps., 38 V and arc 1) TWIN-WIRE SAW

speed of 4-5 mm/sec. i) Twin-wire parallel power: The system differs slightly
Horizontal position: Normally fillet welds up to 8 mm from the conventional one, in that two electrodes are
leg length (L in Fig. 4.32) can be deposited with a single fed at the same rate through a common tip or set of jaws
pass in the horizontal position. Positioning of the by means of the double set of rolls and guides provided
electrode w i t h respect to the joint d e t e r m i n e s in the equipment, the current from the single power
penetration and leg size as shown in Fig. 4.32. The source being split between them. The electrodes also
welding parameters are given in Table 4.33. share the drive motor and control of the equipment and
therefore carry identical arc voltage. Electrode diameters
Process Variations range from 1.6 to 3.2 mm, and the spacing between them
SAW has been described so far mainly as a single varies from 8 to 16 mm. DC power source with flat
wire process, but several modifications involving the characteristics (constant voltage type) has been found
use of two or more wires are in industrial use, because to be most suitable.
of their special advantages. They are: For surfacing work w h e r e least penetration is
1) Twin-wire SAW having two versions: desired, the two electrodes are made negative, while
i) Twin-wire parallel power for structural applications they are made positive,
because then they give better penetration and improved
ii) Twin-wire series power
arc stability. For a given welding current, the twin-wire
2) Multiple-wire, multiple-power SAW (two or more procedure using two smaller diameter wires gives 30-
wires each with its own power source). 50% higher weld deposition rate as compared to the
3) Single-wire SAW with auxiliary currentless wire conventional procedure, using a single larger diameter
feeding wire. This is ascribed partly to resistance heating on the
4) Single-wire SAW with auxiliary hot wire feeding thinner electrodes and partly to the crowding of the two
arcs (caused by attraction, since they are of the same
5) Single-wire SAW with metal power addition
polarity), whereby they function as one single long arc
6) SA cladding with one or more wires or strip. (see Fig. 4.33).
Table 4.33: Data for horizontal fillet welds
Weld size Electrode Current Voltage Travel speed

L, mm dia. (DC+) V mm/sec
mm Amps.
. 3.2 3.2 400 24 27
4.8 4.0 500 26 18
6.4 4.0 650 30 14
8.0 4.0 700 33 10
I
l.
L(
T
A
^ *• £^*&r
3
ffi
Fig. 4.32: Effects of electrode angle on weld size and
penetration Fig. 4.33: Twin arcs of same polarity in action
Longer stick-outs can be employed in twin-wire Further welding trials on the lines of Table 4.35 have
welding to get faster melting of the electrodes due to indicated that with the twin-electrode system used on
fR-heating effect. With 2 mm diameter electrodes (DC-), electrode negative and with longer stick-outs up to 50
stick-outs up to 50 mm have been used without mm, one can control heat input as required in the welding
additional insulating guiding equipment to get of quenched and tempered steels and other alloy steels
deposition rates much higher than those obtainable with with improved impact requirements, and at the same
a single larger conventional electrode using 130 mm time get very high deposition rates and travel speeds.
stick-out. In this system, the nozzle carrying the two electrodes
Tables 4.34 and 4.35 provide comparison between can be rotated and locked, so that the two arcs are either
the conventional single-wire system and the twin-wire in tandem (one behind the other) or side-by side. Typical
system. applications of each are shown in Fig. 4.34.
Table 4.34: Comparative data of single- and twin-electrode SAW

Weld System Wire Amps. Volts Stick- Travel
size out
mm mm
ft— 1.6 mm Single 3.2 550 + 27 19 28

3 Electrode
l — 2.5 mm
Twin 1.6x2 850 + 27 19 63
Electrode
W— 6.3 mn1 Single 4 600 + 29 19 18
Electrode
1 \ LJ.
1 j \ T4 J mm"2 Twin 2x2 950 + 29 19 30
1 Electrode
Single 3.2 575 + 30 19 30
Electrode
* 6.3 mm
Twin 1.6x2 925 + 26 16 55
Electrode
I—j.2mm 1
Table 4.35: Comparative data of single- and twin-electrode SAW
Weld System G Wire Stick- Travel Deposit
mm size Amps Volts out speed rate
mm mm mm/sec kg/hr
IHt
I t i i i
Single
electrode
4.8 4.8 1,000* 35 32 7 13.2
r-\J Twin
electrode
5.5 2x2 1,250- 44 32 13 30
—»n L, mm
Single 8 4 575- 34 25.4 9 9.5
1 9 5 mm f
electrode
T Twin 8 2x2 850- 32 25.4 1/ 17
electrode
\ Single 12.5 4.8 950+ 36 32 5 13
/ electrode
Twin 12.5 2x2 1,000- 42 32 11 22
electrode
Source: W.J., March 76
Typical procedure data experimentally developed for

butt welds in plate thicknesses of 30 mm, 40 mm and 50
mm using 2.5 mm diameter twin-wires and an
agglomerated flux are given in Table 4.36. The sequence
of passes is shown in Fig. 4.35. It was observed during
the development work that the flux showed a tendency
to boil when a single wire of 5 mm was used at currents
above 800 Amps., while with the twin-wire system, this
boiling did not occur even at 900 Amps.
ii) Twin-wire series power: The main advantages of
this system are high deposition rate and minimum
Fig. 4.34: Applications of twin arc system
dilution with the base metal, which are well exploited
in surfacing applications. Two welding heads are used
with a single DC or AC power source. AC is preferred through the weld pool and surrounding metal. Each
for ferrous metals, while DC is used for non-ferrous electrode has its own feed motor and voltage control
metals. The output power cable is connected to one and thus operates independently. The electrodes are
welding head and the return power cable is connected inclined at an angle of 45° to each other, and transverse
to the other welding head instead of to the work piece to the direction of travel.
(see Fig. 4.36). Thus, the two electrodes are in series and In DC welding, the electrodes have opposite polarity.
welding current travels from one electrode to the other The arcs therefore tend to spread away from each other,
Table 4.36: Data for butt welds with twin-wire SAW
G = 0.5 mm (max) Back-gouging is recommended

C = 4 mm before starting side B.
A = 2B
60°
Side Pass Current Voltage Speed Wire feed
amps V arc travel rate
m/min m/min
30mmf
A 1 750 33 0.50 2.30
2 850 35 0.50 2.95
B 1 900 35 0.60 3.10
40 mm t
1 750 33 0.50 2.30
2 850 35 0.60 2.95
A 3 850 35 0.60 2.95
4 850 35 0.60 2.95
B 1 900 35 0.65 3.10
2 900 35 0.60 3.10
55 mm t
1 750 33 0.50 2.30
2 850 35 0.60 2.95
3 850 36 0.60 2.95
4 850 36 0.60 2.95
5 850 36 0.60 2.95
6 850 36 0.60 2.95
7 850 36 0.60 2.95
8 850 36 0.60 2.95
1 900 34 0.60 3.10
2 900 35 0.60 3.10
3 850 36 0.50 2.95
wires. The welding speeds can be proportionately

increased to limit the heat input rate. To give an
example, the welding speed on longitudinally welded
pipe of 17.5 mm thick plate has been increased from 45
1.5 MM c m / min using a single wire to 90 for tandem, 170 for
— .-i-2.5 MM
three wire or 200 for three wire/hot wire, and 220 c m /
min in the case of a four-wire process.
Let us first discuss the two-wire system in detail. In
this system, both wires can be DC, both can be AC, or one
, JO MM DC and one AC. When both wires are DC, they can be
made positive to give the deepest penetration and
therefore the highest possible welding speed. However,
this combination gives rise to arc blow and is rarely used.
The most commonly used arrangement is: the leading
wire connected to DC and made positive and the trailing
2.5 MM wire connected to AC (see Fig. 4.37(A)). The DC weld is
Fig. 4.35: Sequence of passes for twin-wire SAW of made at a high current to get the required penetration
thick joints while the AC weld is deposited at a slightly lower
current to improve the profile and surface finish of
the weld.
Two AC arcs can result in arc deflection depending
on the phase difference between them. In this system,
the welding transformers are connected to the three-
phase line using the Scott or the closed delta connection.
The scott connection, w h i c h is very simple and
economical, provides a 90° phase difference between
the arcs (see Fig. 4.37(B)). This connection is
advantageous for making single-pass welds in thick
plates and multi-pass welds in deep grooves such as
are encountered in the fabrication of pressure vessels.
In the closed delta connection (see Fig. 4.37(C)), the
phase difference between the arcs can be controlled by
c providing a variable impedance in the return lead.

Joint set-ups and sequence of passes for making two-
Fig. 4.36: Two-wire SA series connection pass square-butt welds in plate thicknesses up to 12.7
mm with two-wire DC/AC system are shown in Fig.
resulting in a fanning of the arc zone. Penetration is 4.38. Welding data are provided in Table 4.37.
deeper under the positive electrode than under the Two-pass procedure can also be applied to square
negative one. When a wide multi-pass weld deposit is butt joints in the thickness range of 19-38 mm, provided
made, it is advisable to locate the positive electrode over the joints have double-vee grooves as shown in Fig. 4.39
the previously deposited metal, so that the entire deposit and welding procedures detailed in Table 4.38 are used.
has uniform penetration. Transverse oscillation of the Multi-power twin-arc SA process has been used with
welding head is sometimes used to minimise dilution. great success for one-sided welding of plates up to 30
mm t in Japanese shipyards. The arcs are either D C /
2) MULTIPLE-WIRE MULTIPLE-POWER SYSTEMS AC or AC/AC. Flux-copper backing (FCB) is used as
Systems involving two, three and even four wires, s h o w n in Fi^. 4.40 for e n s u r i n g complete root
each powered by an independent power source, are penetration without burn-through. The joint details are
being used in industry, mainly to increase weld also given in the figure.
deposition rates. If the mean fusion capacity of one wire The leading electrode is of 6.4 mm diameter, while
is 10 kg/hour, it will be 20 kg/hour for twin wire, 40 the trailing electrode is of 4.8 mm. Maximum root gap
kg/hour for triple wire and up to 90 kg/hour for four permitted is 0.8 mm. Tack welds, 30 mm long and
B 1. TRAIL ARC;
A. 1.DC POWER SOURCE;
2,-LEAD ARC;
2. AC POWER SOURCE
3.WELDING DIRECTION.
3. TRAIL ARC;
4. LEAD ARC
C. 1 TRAIL ARC:
2. LEAD ARC;
3.^WELDING DIRECTION
a
b*
L on
" V " r1 UuuutiJ UuuumiJ
Fig. 4.37: Two-wire SA systems: (A) DC/AC system; (B) AC/DC system using scott connection; (C) AC/AC system
using closed delta connection
Table 4.37: Data for butt welds with twin-wire DC/AC system
Joint Plate Pass Electrode 1 Electrode 2 Travel Electrode

detail T No. DC+ AC speed spacing
mm/sec.
mm mm
dia. dia.
mm Amp V Angle 0 mm Amp V Angle 0
A 7 1 4.8 950 35 0 4 650 38 12 34
2 4.8 1,100 35 0 4 700 38 12 34
16
A 9.5 1 4.8 1,150 35 0 4 750 38 12 30
2 4.8 1,200 35 0 4 800 38 12 27
19
B 12.7 1 4.8 1,100 36 0 4 650 39 12 25
2 4.8 1,200 36 0 4 650 39 12 21
Table 4.38: Data for heavy butt welds with twin-wire DC/AC system
Plate Pass Electrode 1 Electrode 2 Travel Electrode

T No. DC+ AC speed spacing A° B C D E°
mm mm/sec mm mm mm mm
dia. dia.
mm Amps. V Angle 0 mm Amps. V Angle 0
19 1 5.5 1,050 35 0 4.8 700 40 12 17 22 90 3.2 9.5 6.3 90
2 5.5 1,150 36 0 4.8 800 41 12 13 22
25.4 1 5.5 1,050 36 0 4.8 700 41 12 11 25.4 80 6.3 9.5 9.5 80
2 5.5 1,150 38 0 4.8 850 43 12 9 25.4
32 1 55 1,050 36 0 4.8 o 750 42 12 8 28.5 70 9.5 9.5 12.7 70
2 5.5 1,200 38 0 4.8 850 43 12 7 28.5
38 1 5.5 1,200 36 0 4.8 850 43 12 8 32 60 11 11 16 70
2 5.5 1,300 40 0 4.8 900 44 12 6 32
Fig. 4.38: Joint set-ups for two-pass butt welds with twin-wire DC/AC system
One-sided welding
a) Requires specialised expensive backing and plate-
holding equipment
T b) Consumption of flux and wire is higher
i
c) Demands special flux which will tolerate high
current, high voltage conditions
d) The first and last 300 mm lengths of the welded
joint have to be cut out and rewelded manually
Fig. 4.39: Joint set-up for heavy two-pass butt weld with e) Over penetrated bead sometimes requires chip-
twin-wire DC/AC system ping or grinding
f) Turning over of plates is avoided, because weld-
ing is completed from one side
g) Being a single pass, the weld is completed in mini-
mum time
h) Impact values are slightly lower, and yet they meet
the minimum requirements (for 25 mm plate t, at 0°C, 8
kgf.m as compared to 10.5 kgf.m for two-pass
technique).
The acceptable impact values obtained with tandem-
COPPER BLOO*
arc welding are explained by the fact that the trailing
electrode inhibits the formation of the oriented columnar
T T T structure by depositing a sufficiently wide bead, which
AIR HOSE
prevents the formation of a crystal junction in the centre
of the weld by disorientation of the second
Fig. 4.40: Joint set-up and flux-copper backing for twin- crystallisation front. The increased contact area of the
arc one-sided SAW weld metal with the parent metal also gives faster
cooling rates. It is also to be understood that the plates
spaced 300 mm apart, are laid manually with basic low- are better controlled with respect to bon and non-
hydrogen electrodes. The backing to the joint is provided metallic inclusions.
by a layer of flux (depth 5-6 mm) spread over copper
strips, which are made to press against the bottom of Three-wire technique
the plate by air pressure, provided by a hose below. The Both DC/AC/AC and AC/AC/AC (see Figs. 4.41
plates are clamped down firmly, employing a set of and 4.42) are commonly used in shipyards, pressure
powerful electromagnets. The welding parameters are vessel shops and pipe mills to increase productivity.
given in Table 4.39. It may be mentioned that thicknesses It is to be noted that the leading electrode is held at
up to 50 mm have been welded using three or four 90° to the plate, the central one is slightly inclined while
welding heads and wires. the trailing electrode is inclined more steeply. These
It is interesting to compare the Japanese one-sided angles of inclination control the shape of the weld bead.
single-pass technique with the conventional two-sided Three-wire SAW has been used with great success
two-pass SAW technique used by shipyards elsewhere in spiral welding of pipes of large diameters at high
in the world. speeds. In a typical installation, using the DC/AC/AC
Table 4.39: Data for twin-wire SAW with flux-copper backing
Plate Leading electrode (AC) Trailing electrode (AC)

thickness
t mm Amps. V Amps
10 1,000 35 800 47
11 1,100 35 800 47
12 1,150 35 850 47
13 1,200 35 850 47
14 - 1,250 35 900 47
15 1,250 35 950 47
16 1,250 35 950 47
18 1,250 35 950 47
20 1,250 35 1,000 47
22 1,250 35 1,000 47
24 1,250 35 1,050 47
26 1,350 35 1,100 47
28 1,350 35 1,100 47
30 1,350 35 1,200 47
Note: Electrode stick-out — 30 mm
welding heads are used; one for external welding, the

other for internal welding, both using the parameters
indicated.
4~riJrrTnJrrTiJ The outstanding feature of this D C / A C / A C system
is the ability to weld a wide range of thicknesses using
the same welding parameters of wire size, wire-feed
speed, current and arc voltage. Once a combined set of
parameters for the three wires has been established, any
thickness of metal between 6.5 mm and 32 mm can be
welded and the only variable throughout is welding
speed. The system thus requires minimum amount of
operator skill and experience.
Fig. 4.41: Three-wire DC/AC/AC SA system: (1) DC > i>
power source; (2) AC power source (transformers in 1 ,_ ¥
Scott connection): (3) Trail arc; (4) Middle arc: (5) Lead * ^ A
_1_..._|
f T
arc; (6) Welding direction 440 V 1
220 VJ220 V I ^ 440 V
< 1
^JULUJUî i i tXJJJJJJL*
system-leading electrode is DC positive, because a DC
arc is more stable than an AC arc and achieves better
penetration. The DC electrode is fed by its own welding
head. A second welding head feeds two wires supplied
from a pair of Scott connected transformers, so that the
c u r r e n t s u p p l i e d to the t w o w i r e s h a s a p h a s e
displacement of 90°. The two AC wires which provide
the filling passes give minimum penetration and a weld
reinforcement without excessive build-up.
The installation is capable of welding the thickness
Fig. 4.42: Three-wire AC/AC/AC SA system: (1) Trail arc;
range of 6.5-32 m m using the edge preparations and (2) Middle arc: (3) Lead arc; (4) Welding direction
welding parameters shown in Table 4.40. Two three-wire
Table 4.40: Data for three-wire spiral welding of pipes

Plate Preparation Current Volts Wire dia. Speed
thick: mm (see footnote) mm mm/sec
12.7 L. 900 DC 30 4.8 36-38
i I 1 C. 800 AC 38 4
T. 700 AC 40 4
16 L. 900 DC 30 4.8 27.5-3C
* I 1 C. 800 AC 38 4
T. 700 AC 40 4
19 L. 900 DC 30 4.8 15-17
C. 800 AC 38 4
T. 700 AC 40 4
25.4 L. 900 DC 30 4.8 10.5-12.
/ 1 \ C. 800 AC 38 4
T. 700 AC 40 4
19 Double-Vee Butt welds
L. 900 DC 30 4.8 21-23
C. 800 AC 38 4
T. 700 AC 40 4 %.
25.4 L. 900 DC 30 4.8 15-17
C. 800 AC 38 4
T. 700 AC 40 4
32 L. 900 DC 30 4.8 9-11
C. 800 AC 38 4
T. 700 AC 40 4
Footnote: L = leading C = centre T = trailing
Table 4.41 gives an interesting comparison of SAW Experimental work on such a system was carried out
25.4 mm thick mild steel plate with two passes, one at Cranfield Institute of Technology using 1.2 mm cold
from each side, using single-wire, parallel two-wire, wire with 1.4% Mn content. The results are given in
multi-power two-wire and multipower three-wire Table 4.42.
systems. It has been realised that welding speed is It is apparent that while the increase in deposition
increased by 300% and the welding cost is reduced by rate was quite substantial (i.e. equal to the proportion
46% by moving from a single-wire system to a three- of cold wire added), the increase in flux consumption
wire system. was marginal (4%). Regarding weld bead shape, width
remained nearly constant while reinforcement increased
3) SAW WITH COLD WIRE FEEDING and penetration decreased, both markedly. Referring
In this modified SAW process, an additional to weld metal properties, yield strength increased by
continuous wire that does not carry electric current is nearly 20%, tensile strength by 10%, but ductility
fed into the arc underneath the flux blanket in cold state, remained almost the same, and there was some
that is, without being resistance heated. The wire can improvement in Charpy behaviour.
be solid or flux-cored. This requires additional fixtures In another study it was found that addition of solid
to be mounted on the existing equipment, as indicated wire of the same grade as the electrode does not
in Fig. 4.43 and a more elaborate procedure setup. This influence the transition temperature (which is a measure
technique has not been fully exploited industrially, of toughness) in the middle of the weld, but outside the
though reportedly it has been used with success for central region the transition temperature decreases (i.e.
panel production in a continental shipyard. toughness increases) with increasing amounts of cold
Table 4.41: Comparative data for welding 25.4 mm t joint with various systems
Process Single-wire Parallel Multipower Multipower

two-wire two-wire three-wire
Run No. 1 2 1 2 1 2 1 2
Current, Lead 1,000 1,050 950 1,050
Amps Centre 1,000 1,300 1,480 1,480 800 800
Trail 850 925 700 700
Voltage, Lead 32 32 30 30
Volts Centre 36-38 37-39 35 35 38 38
Trail 36 36 40 40
Welding mm/sec 6.5 4.5 8.5 8.5 12.5 11 19 19
speed
Consumables Wire, kg 1.85 1.65 1.45 1.3
usage, per m Flux, kg 2.3 2.0 1.8 1.6
of weld
Table 4.42 (a): The influence of cold wire additions on deposition rate and flux consumption
Proportion of cold wire addition, Total deposition rate Flux consumption
% of deposited weld metal kg / hour kg / hour
0 11.05 12.63
13 12.50 13.18
37 15.10 13.80
Notes: Electrode polarity: Negative
Current: 600 amps
Head voltage: 38 V
Electrode extension: 40 mm
Travel speed: 380 mm/min
(B)
(A)
Rg. 4.43: Method of cold wire attachment to SAW head: (A) Schematic diagram; (B) Geometric relationship (1) cold
wire; (2) wire feeders: (3) SA wire; (4) work piece; (5) copper contact tube
wire. This effect is considered to be due to the influence As shown in Fig. 4.46, the hot wire is fed into the
of cold wire addition on the solidification structure and weld puddle and molten slag, which are slightly ahead
the shape of the cross-sections of the weld as shown in of the arc, its point of contact being 10-20 mm ahead of
Fig. 4.44. the main wire. Its angle of entry varies between 40 and
The best toughness values are obtained with the 60°. The hot-wire speed is first set to obtain the desired
Charpy notch, normal to the direction of solidification. deposition rate, the wire run on a plate and the voltage
Cold wire addition creates this favourable condition at and amperage adjusted until the wire reaches its melting
the ends of the weld as shown in Fig. 4.44. However, point, just when it reaches the plate surface and deposits
the orientation of the structure at the weld centre a smooth bead. Normally 9 or 10 volts are found suitable.
remains the same; this explains why toughness property The heating current drawn depends on hot-wire size,
in this region is not influenced by cold wire additions. feed rate and stick-out. The common wire diameter is
The same study showed that addition of a flux-cored 1.6 mm though 1.2 mm and 2.0 mm sizes have also been
wire containing 20% CaF 2 lowered the transition used.
temperature by more than 40° C. This finding ought to Unlike in TIG hot-wire process, the hot wire is first
enhance the industrial potential of this technique for powered and fed. The arc wire is then positioned in the
critical applications. joint, flux and welding power introduced and the arc
initiated to start the weld.
4) HOT-WIRE SAW
This process, depicted in Fig. 4.45, utilises a standard
single-wire SA equipment, in which an auxiliary wire
feeder is incorporated. The second wire is PR heated by
passing current from a separate AC power source,
having slightly drooping static characteristics and
maximum OCV of 15 volts. The use of AC for hot wire
avoids arc blow. The hot wire feeding can increase the
deposition rate by as much as 100% with only marginal
increase in heat input rate, since only one arc is
operating. There is a decrease in the penetrating power
of the arc, since some of its heat is absorbed by the
additional metal deposit.
Fig. 4.45: Schematic diagram of submerged-arc hot-wire

welding system
(A) \~ y
(B)
Fig. 4.44: Effect of cold wire addition on the shape of

weld zone and direction of solidification: (A) Without Fig. 4.46: Cross-section of SA hot-wire system in
cold wire; (B) With considerable wire addition operation
Typical procedures and applications are given below: Travel speed Dep. rate
a) Wide bead in butt joint. Attempting to complete V Amps. mm/sec. kg/h
this weld in a single pass with single or multiple arcs
results in the burning through of the steel backing bar. Arc wire 35 900
11.3
With hot wire SA process, the groove is filled up at a
high speed with minimum penetration. Note that the Hot wire 15 500
13.6
hot-wire deposition rate exceeds that of the arc wire by
50%.
d) Fillet weld—horizontal position. Deposition rate is
double that of a single wire, while root penetration is
■ir excellent.
9.5 m«A
Travel speed Dep. rate

Amps. mm/sec kg/hour
Travel speed Dep. rate

Arc wire 36 625 9.5 6.8 V Amps. mm/sec kg/h
Hot wire 9 375 9.5 10.7 Arc wire 34 600 19.5 6.8
Hot wire 9 350 19.5 6.8
b) Multi-pass butt weld in HY-80 steel The hot-wire
technique gave 50% increase in weld completion rate e) Pipe welding. The process has been found to be
with less than 10% increase in heat input rate, so that well suited for welding the longitudinal seams of small
the final heat input value was well within the maximum diameter pipes using steel backing of 4-5 mm thickness.
allowable (2.16 kj/mm) for this quenched and tempered In one experimental study it was found that satisfac-
steel. tory weld beads were obtained only at specific not wire
% values, which did not exceed 70%. Table 4.42
summarises the findings.
The study also showed that the hot-wire process was
less costly compared to conventional single-wire and
two-wire tandem processes. Table 4.43 gives the cost
comparison for 1 metre length of weld.
f) Weld overlays. Because of high deposition rate and
Travel Dep. rate Heat input shallow penetration, the process is useful for depositing
speed kg/h rate overlays 4.8 mm and thicker. Using alloyed flux-cored
V Amps, m m / s e c kj/mm hot wires, the deposit chemistry can be easily controlled
as desired. Stainless steel overlays on carbon steel have
been successfully made using agglomerated special flux
Arc wire 30 600 9.5 8 1.93
and S.S. main wire and hot wire. With the overlay
Hot wire 8 200 9.5 4 0.17
thickness greater than 4.8 mm, the dilution levels remain
around 15%. This compares favourably with the more
c) Fillet weld—flat position. Large size wide, sym- complex series-arc and strip-cladding methods.
metrical bead can be made in one pass. For 22 mm fillet The mechanical properties of the w e l d s show
size the parameters are: negligible degradation, even when the weld size is
increased by hot wire additions up to 100%. Table 4.44
shows the mechanical values of three welds in ASTM
type A302B steel, deposited with a fused neutral flux
and a wire having 0.12 C/1.80 Mn/0.50 Mo. Such good
values are difficult to explain but can be assumed to be
due to relatively fast cooling rates of the weld beads.
Table 4.42 (b): Effect of % deposition through hot wire

Pipe dimension %
Main welding Hot Bead quality
parameters wire
508 dia x 16 m m 55 6 layers, side wall wetting good
500-600 Amps 65 4 layers, bead satisfactory
30-34 V, 36 M / hour 75 lack of fusion, slag inclusion
600 dia x 32 m m 55 6 layers, good penetration
550-600 Amps, 32-34 V, 65 lack of fusion, convex beads
33 M / hour 75 slag inclusion
780 dia x 40 m m 60 burn-through, undercut
550-600 Amps, 32 V 70 10 layers, satisfactory joints
27 M / hour 80 lack of fusion
100 slag inclusion, lack of fusion
Table 4.43: Cost comparison
Process No. of Total time Total cost

layers in sec inRs.
Single-wire SAW (20 mm) 1,130 51.50
HW-SAW (20 mm) 5 900 47.80
Single-wire SAW (40 mm) 17 2,400 193.90
HW-SAW (40 mm) 10 1,830 182.80
Two-wire tandem (40 mm) 4+7 1,750 199.00
Note: HW—Hot wire
Table 4.44: Mechanical properties of SA welds with hot wire addition

Hot wire added
0% 50% 100%
Yield strength, KSI 76.0 76.3 75.3
Tensile strength, KSI 88.8 89.2 88.4
Elongation, % 25.0 25.0 25.0
Reduction of area, % 71.1 70.6 70.4
Charpy-V values, J-at -12°C 112 114 116
Note: Stress-relieved at 621°Cfor 2 hours
5) SAW WITH METAL POWDER ADDITION Alloy metal p o w d e r s are judiciously b l e n d e d

The process consists of pre-placing a homogeneous, mixtures of well-specified ingredients, which are
correctly blended alloy metal powder in the joint groove s o m e t i m e s a g g l o m e r a t e d . They are m a r k e t e d by
and making the conventional single-wire SA weld as companies in Europe, the USA, Canada and Japan.
shown in Fig. 4.47. Powders meant for general fabrication work usually
METAL POWDER
Fig. 4.47: Schematic diagram of SAW with metal powder addition

I~TA /
*10 M5-30Y ®
2
Hh- X
Fig 4.48: Fillet welding without (A) and with (B) AM using
identical conditions as follows: plate f-20 mm: electrode
diameter-4.8 mm: 750 amp: 34V-15 cm/min; S49 mm2; S-
100 mm2
contain 85-95% iron powder, alloy additions (Mn—0.5

to 5.5, Si—0.2 to 1.2, Ni—0 to 5, Mo—0 to 5, Cr—0 to 5)
and strong deoxidisers, such as Al, Ca, Mg, Ti and Zr in
varying amounts.The normal powder is fine-grained Fig, 4.49: Automatic two-side butt welding with
conventional edge preparation and medium fit-up: (a) with
capable of passing entirely through 80 mesh (Tyler). The
backing, fine AM; (b) without backing, coarse AM.
coarse variety whose grains vary in diameter between Thickness range: 15 to 45 mm
35 and 10 mesh (0.5 and 2 mm) is meant to be used in
joints of poor fit-up.
The powder, called AM, is placed manually in the
joint. The preferred current is DC, electrode positive.
Welding is done in the flat position. In multi-pass
welding, the quantitative distribution of the powder is
ensured with the help of scrapers simply cut to size and
moved along the joint.
The beneficial effect of AM addition on welding
productivity can be appreciated by examining the
downhand fillet welds shown in Fig. 4.48. Weld (B) has
been made with AM addition at twice the travel speed Fig. 4.50: Automatic two-side butt welding with square
preparation and backing (a) for AM. Conditions: f—20 mm;
as compared to weld (A). Comparing the two, it is seen
e—8 mm; electrode diameter—4.8 mm; first pass—
that the cross-sectional area of the weld metal (B) is twice 650A(DC+), 32V, 30 cm/min; second pass—750 amp
that of (A), weld size has increased by 3 mm and the (DC+), 34V, 40 cm/min.
penetration has decreased by 2 mm.
Use of AM for making typical two-pass, two-sided
butt welds in shipbuilding are shown in Fig. 4.49. The Metal powder technique has been found useful in
root gap usually is as wide as 2 mm. When fine powder one-sided welding of ship plates using refractory flux
is used, it has to be supported by an adhesive seal, such backing and tandem twin-arc SA process (similar to that
as scotch tape attached to the bottom side. For coarse described on page 262) in which, due to poor fit-up, root
powder, such support is not necessary. The metal gap varies between 2 and 10 mm and the difference in
powder actually serves in both cases as a backing for height of the bottom plates can be as much as 4 mm.
the molten weld metal. The welds are smooth and free from undercut, in spite
The ability of AM to serve as a weld backing is of such irregular fit-up.
demonstrated while making a two-pass weld in 20 mm In the fabrication of a 56-storey skyscraper in
thick square butt joint with 8 mm root gap as shown in Toronto, columns with very bad fit-up had to be welded.
Fig. 4.50. The gap is filled with AM, which is supported This was done successfully as shown in Fig. 4.51 by
by an adhesive tape. With this technique, no slag inclu- using asbestos to seal the root, filling the gap with AM,
sions occur and the welds are radiographically perfect. and using multiple-head SA welding.
Combination of AM, a strip electrode and SA process

have been used in strip cladding to gain the following
advantages:
a) Reduction of dilution
b) Increase in deposition rate
c) Economical a t t a i n m e n t of desired weld
chemistry.
The numerous benefits of high current welding with
AM addition over high current welding without such
addition, are listed below:
a) Cooler, less locally overheated molten pool
resulting in:
i) A smoother, quieter fusion
ii) An improved bead appearance
Fig. 4.51: Welding of heavy columns with very poor fit-up
(A—asbestos seal). Leading head: 25 cm ahead, flux-core
iii) A slightly smaller heat-affected zone (per melt
electrode, 600 amp, 30V, 30 cm/min. Middle head: 10 cm zone area).
ahead. 4 mm SAW electrode, 650 amp (DC+), 29 V. Trailing b) Better utilisation of the arc energy, wherein part
head: 5 mm SAW electrode. 900 amp (AC). 27 V of the arc energy instead of being used irrationally to
melt excessive base metal, is freed to melt the metal
powder and bring about:
i) A higher deposition rate
ii) A decline in the consumption of welding flux,
because some of the energy is used to melt the metal
powder
iii) Less dilution, because with an increased
deposition rate, much less energy is required to fill a
50
definite groove
iv) Smaller heat-affected zone (per weld groove
area).
c) Rational conception of metal powders, wherein
being a relatively dense metallic powder, it can by filling
in the varying voids of poorly fit up joints:
i) Minimise the ill-effects of fluctuating weld
geometry on penetration, dilution, etc.
Fig. 4.52: Welding of 85 mm and 50 mm t Germal steel
plates with specially developed AM. Welding conditions ii) If necessary, support conveniently the molten
of third pass as example: hAM = 8 mm, 4 mm diameter bath
electrode, 750 amp (DC+), 32 V, 24 cm/min. iii) Contribute to the correction of the weld
chemistry.
The technique has also been used in welding various
thicknesses of HSB 55 C steel, a German fine-grained 6) SA CLADDING
steel alloyed with Cu and Ni and giving tensile range Clad metals are used mainly in the chemical,
of 58-70 kg/mm 2 , during the fabrication of offshore petroleum and nuclear industries. They are composites
drilling rigs. Figure 4.52 shows a typical joint welded of two or more metals joined in a continuous manner
with the following consumables: an electrode with 1.5% by a metallic bond. They have to be distinguished from
Mn, a specially developed AM containing the necessary liners, in which two or more metals are joined or
alloying elements to adjust the weld chemistry, and a attached in an intermittent manner.
welding flux with an easy slag removal in the first run The most familiar example of clad metal is stainless
of the narrow V-groove. The reduced number of passes steel clad material, comprising a relatively thin layer of
and the narrow groove resulted in considerable gain in stainless steel and a thick backing layer of plain carbon
time and savings in welding consumables. At the same steel. Increasing use is being made of this material in
time, the impact values were up to the mark. large fabrication jobs, such as pressure vessels and
chemical plant, where the use of fully solid stainless steel commonly applied to components such as rollers of
would involve high costs. The thickness of the clad caterpillar tractors and similar equipment, where as a
stainless steel layer is usually small relative to that of rule, the requirements as to the quality of the overlay
the carbon steel, because it is the latter, which is designed material are not so stringent. Systems (c), (d) and (e) are
to take the major proportion of the load. In some well suited and often used for cladding, because they
applications, the stainless steel clad metal is replaced are primarily low penetration methods. In system (c),
by nickel, copper, high-nickel alloys and copper-nickel the cold wire feed increases the amount of molten metal
alloys. for a given energy of the arc and removes part of this
Clad steel plates and some standard clad steel energy. With system (d), the energy input to the parent
forgings are produced by steel mills by hot rolling the metal is adjusted by a change in the distance h. As the
sandwich assembly and supplied to fabricators. The arc burns close to the work piece, it absorbs more energy
high temperature and pressure create a solid-phase per unit of surface than when the distance is greater.
weld between the steel and the clad metal. Clad metals The proportion of molten electrode quantity to molten
are also fabricated by explosion welding. This process p a r e n t metal is also i m p r o v e d by u s i n g double
is very versatile and is used to clad large and small plates electrodes.
and forgings in a wide range of thicknesses. With systems (c) and (d), the electrodes are given an
As an alternative to the purchase of readymade clad oscillating m o t i o n of controlled a m p l i t u d e ,
plate, the fabricator resorts to SA cladding, i.e. extensive perpendicular to the welding direction while they move
SA surfacing with one or more wire electrodes, or a strip forward. The term 'oscillating motion' must be clearly
electrode usually 60 mm wide and 0.5 mm thick, on flat distinguished from the term 'weaving motion'. In the
and curved surfaces, on dished ends and on forgings as former, the electrodes carry out a parallel motion; in
required from time to time. During the surfacing, each the latter they swing about a point of suspension. The
weld bead is made to tie in or merge with the adjoining width of the weld run is twice the amplitude. Correct
weld bead. The various systems of SA cladding, shown oscillation ensures that penetration is uniform across
in Fig. 4.53 are: the whole width.
a) With one electrode in series with the work piece Table 4.45 gives a comparison of the efficiencies of
b) With two electrodes connected in parallel and in the three systems.
series with the work piece
c) With one current-carrying electrode in series with Table 4.45: Comparative efficiencies of three SA
the work piece and one non-current carrying electrode cladding systems
(filler electrode) System Electrode Flux
d) With electrodes connected in series melting speed consumption
e) With a strip electrode in series with the work kg / hour kg / flux
piece. kg / wire
These have been described earlier, under Process (c) 13 1.0
Variations. Systems (a) a n d (b) p r o d u c e d e e p e r (d) 15 0.9
penetration than that permissible for cladding. They are (e) 13 0.4
W//9////A 3> V7777777/77//X

777, 3>
/////7W/\
^Pl
////////////ft////
Fig. 4 53: Various systems of SA cladding
Cladding with Six Wires

A six-wire cladding machine has been successfully
developed abroad by selecting the proper electrode size,
spacing, mode of operation, polarity relationship,
oscillation mode and frequency, and by designing a
t
wire-feed mechanism havingsix feed rolls driven by a
common shaft, and a special electrode contact head.
This machine is capable of depositing 308 alloy
stainless steel on carbon steel base material, in a single Fig. 4.56: Bead edge configuration with all six
electrodes in line
layer, to the r e q u i r e d physical a n d chemical
specification. It deposits at the rate of 40 kg/hour and
covers an area of 125 cm 2 / in. Six coils of 1.6 m m in-line electrode is used in this position. The electrode
diameter wire electrode, each weighing approximately pattern was selected to provide minimum dilution,
230 kg, are fed into the wire drive mechanism. optimum chemistry and provide a sufficiently smooth
The electrodes are arranged in accordance with Fig. surface to permit dye penetrant and ultrasonic testing
4.54. The wire making the tie-in to the previously without further surface preparation. All other methods
deposited bead is made the positive polarity, while the of submerged-arc cladding require at least some surface
remaining five wires are made negative. The sixth p r e p a r a t i o n to perform non-destructive tests. A
electrode is set back a small distance from the in-line favourable chemistry and controlled dilution of a single
five electrodes to provide a tapered edge on the bead electrode submerged-arc process is dependent upon the
and assist the tie-in of the next bead, as shown in Fig. amount of overlap of the previously deposited bead.
4.55. Figure 4.56 depicts the edge configuration, if an With the six arc oscillating electrode method of cladding,
the dilution of the weld deposit by base material is
minimised by directing the arc force on the larger molten
POWER
weld puddle characteristic of the six arc process. One
SOURCE important additional feature of this process is the ability
of the welding head by simple manipulation of the wire
drive controls, to deposit weld metal with either one,
OSCILLATION two, three, four, five, or six electrodes.
SA Strip Cladding
This system of cladding, shown at (e) in Fig. 4.53,
has been widely applied in industry for the following
main reasons:
1) It is a simple method with regard to control of weld-
ing data
2) Efficiency is high, if dilution is considered
3) It is a very reliable and economical process pro-
ducing consistent and reproducible weld deposits
Fig. 4.54: Arrangement of electrodes in six-wire 4) The essential investment is low. In many cases,
cladding system only the submerged-arc welding head has to be ex-
changed for one suitable for use with strips.
The process is outlined in Fig. 4.57. A DC arc is
initiated between the continuously fed melted strip
< _ electrode and the parent metal. Usually DC from a
constant voltage power source is employed to maintain
a constant arc length. The arc and the welding zone are
protected against atmospheric contamination by a layer
of flux, which partially melts and forms a protective
Fig 4.55: Bead edge configuration with one slag cover. After cooling, the solidified slag can be
electrode offset detached easily from the weld deposit.
Due to the strip electrode size (usually 60 x 0.5 mm)

and the relatively low current density used (about 20- STRIP ELECTRODE
25 A m p s . / m m 2 ) , t h e s u b m e r g e d - a r c c l a d d i n g STRIP FEEDING SYSTEM

CURRENT CONTACT
equipment, commonly mounted on a manipulator, has
CLADDING FLUX
to fulfil special technical requirements. For example, FLUX HOPPER
control of electrode feeding as well as the power supply ARC CAVERN
to strip via the contacts must be efficient. Suitable
positioners must be used to ensure correct travel speed ELD DEPOSIT
of either the welding head or the work piece. The flux FAVOURABLE DROPLET RECRYSTALLISATION FRONT
TRANSFER wLkB-E9i2LAND FLOW DIRECTION
s u p p l y unit should be constructed to g u a r a n t e e WELDING DIRECTION
complete protection of welding zone as well as to enable
* PARENT METAL
the correct flux burden height to be maintained in front
and behind the electrode. Fig. 4.58: Schematic representation of SA cladding
During welding, the arc does not burn over the whole showing elements of the cladding operation
w i d t h of t h e strip electrode, n o r does it r e m a i n
stationary, but oscillates irregularly, and suddenly shifts 2) Complete fusion between parent metal and de-
from one point to another on the strip. Furthermore, posit,
two or more arcs can exist together for a short period. 3) Finely rippled bead surfaces and almost notch-
Droplets arise on the end of the electrode and are free overlapping from one pass to the other.
detached and transferred into the weld pool. Moreover, for most stainless steel cladding specifi-
During the droplet formation and transport into the cations, three main objectives have to be considered:
weld pool, the liquid metal reacts rapidly with the 1) Chemical composition within the range of the
molten liquid slag from the flux. In this short period of relevant stainless steel composition over the whole
droplet growth, about 0.04 sec depending on welding overlay.
current, and metal transfer, the important chemical 2) A minimum thickness of cladding within the
reactions, such as loss or pick-up of elements, take place. stainless steel specification.
Parts of the liquid flux/slag are also transported into 3) Suitable deposit microstructure and mechanical
the weld pool and carry out additional reactions. Due properties (e.g., 4-10% free ferrite in an austenitic ma-
to their lower density, these slag particles float along trix and sufficient ductility at the clad/metal interface
the recrystallisation front of the weld pool to the surface to satisfy a 3T side bend test).
of the deposit and solidify to form the slag cover. Figure Control of deposit chemistry. The aim of stainless
4.58 explains the process mechanism. cladding is to produce a deposit which is corrosion-
Requirements of high quality cladding are: resistant, sound and ductile. To achieve this, carbon
1) High chemical consistency of the deposit with content of the deposit must be below 0.030%, martensite
uniform dilution of parent metal (10-15%) and constant m u s t be absent a n d delta-ferrite in the austenite
layer thickness. microstructure must be around 5% to avoid cracking.
For delta-ferrite control, reference is made to Schaeffler
diagram and the Cr and Ni contents as well as Cr and
I STRIP ELECTRODE Ni equivalents are maintained within certain limits.
POWER SOURCE Here, it has to be borne in mind that appreciable Cr-
A STRIP FEEDING DEVICE
loss from the strip occurs during deposition. This loss
can be as high as 32% with an acidic flux and 11% with
CONTACTS a highly basic flux. Increasing arc voltage increases Cr-
FLUX BURDEN loss. Losses of Ni, Mo and Nb are usually below 10%,
SLAG W E L D
OyERLAY especially w i t h basic fluxes. These losses are
compensated for by using overalloyed strips a n d
smm'^wwi suitably alloyed fluxes. Typical strip compositions are
given in Table 4.46.
Another factor influencing deposit chemistry is
Fig. 4.57: Schematic representation of SA cladding with dilution by the base metal arising from the penetration,
strip electrode which is usually 1.5-2.0 mm. The effects of dilution are
Table 4.46: Typical stainless steel cladding strip compositions
Strip Chemical composition, wt %

C Cr Ni Mo Nb
max
Type 308L 0.03 19.5 10.5
Type 316L 0.03 19.0 12.5 2.5
Type 309L* 0.03 23.0 13.5
Type 347 0.05 19.5 10.0 12x%C
Overalloyed Type 316 L 0.03 20.5 13.3 3.1
Overalloyed Type 347 0.03 20.5 11.0 0.8
Overalloyed Type 347 0.03 23.5 12.8 0.7
*Overalloyed Type 308L or Type 304L
usually overcome by using two or three layers, so that d) A side bend test to confirm the absence of low
the top 2-3 m m thickness fully meets the specified ductility structures in the zone between the clad layer
chemistry requirements. and the base metal. The diameter of the mandrel to be
To understand the effects of dilution and technique four times the specimen thickness (i.e. 38 mm), and the
of overcoming them, we shall cite a case study involving specimen to be bent 180° without cracks occurring. The
strip cladding of a tank in a heavy water reactor. The first three requirements are explained in Fig. 4.59. The
tests were carried out on a 70 mm thick normalised plate last one is explained in Fig. 4.60.
having 0.17 C/1.36 Mn/0.40 Si/ 0.18 Cr/0.14 Cu and The cladding was accomplished in three layers. For
giving YS of 340 N / m m 2 and UTS of 535 N / m m 2 . The the-first layer, taking guidance from the Schaeffler
quality requirements were: diagram, a overalloyed 27 Cr/12 Ni type strip with
a) Penetration into base metal must be 1.5-2.0 mm 0.078% carbon was selected in combination with a Cr-
b) Size of machined layer must be 6.4 mm, exclud- alloyed agglomerated flux. The next two layers were
ing machining allowance of 2.0 mm minimum deposited with a standard strip (19 Cr/10.5 Ni/0.030
c ) Depth of 1.5-2.0 mm below machined clad sur- C) and the same flux as for the first layer.
face must consist of 18% C r / 8 % Ni and carbon content The chemistry of the clad layer was determined as a
below 0.06%. function of the distance from the plate surface. For each
DILUTED WELD METAL
PENETRATION
SIS 2101
Fig. 4.59: Dimensions of the cladding layer in mm
CLAD LAYER
9.5
Fig. 4.60: Bend test specimen to test cladding ductility. The specimen thickness is the most essential dimension.
mm, turnings were taken and analysed for the key Increase of voltage increases flux consumption and
elements Cr, C and Ni. Their distribution is shown loss of Cr. It also causes microslag inclusions and
graphically in Fig. 4.61. adhesion of slag particles.
The figure shows that at the original plate level Increase of welding speed means that the arc is
surface (A), the composition is 19 C r / 8 Ni/0.08 C. At directed ahead of the weld pool and on to the base metal,
2.4 mm below the clad layer (i.e. 4 mm above the plate resulting in increased dilution. It also reduces the
surface) the desired levels of Cr, Ni and C have been deposit height and width.
reached. Increasing productivity: Using standard 60 mm by 0.5
Welding parameters: Parameters for cladding are strip and a welding speed of 100 m m / min, an area of
usually such that a pass 4.5 mm high and 65 mm wide 0.32 m 2 can be surfaced in an hour. This means that a
is obtained. For 60 mm by 0.5 mm strip, current is in the conventional pressure vessel for a nuclear reactor, clad
range of 550-750 Amps, and voltage in the range of 26- with 10 tons of strip in one layer will take six weeks to
34 V, giving a deposition rate of 12-15 kg/hour. Strips complete. Therefore, attempts have been m a d e to
100 mm by 0.75 m m have been used at currents up to increase cladding output by using modified systems,
1,200 Amps, with deposition rates up to 27 kg/hour. in addition to using 120 mm wide strips.
Welding speeds are in the range of 90-170 m m / m i n . One such system, termed 'double strip cladding' (see
Electrode extension, i.e. the distance from contact tips Fig. 4.62), employs a normal strip cladding head and a
to parent plate, is within 25-30 mm. modified contact with a distance piece of 8 mm between
Current determines deposition rate, penetration, the two contact clAmps... The higher deposition rate is
dilution, weld bead size and heat input. With increasing produced by an electro-slag effect between the two
current, deposit height and width as well as dilution strips. At this point, the liquid slag is so hot that the
increase. If the current is too low, defects such as slag electrodes are molten without the assistance of an arc.
inclusions and lack of fusion occur. Thus, less energy is put into the parent metal and less
energy is necessary to melt the flux to form the slag.
The results from tests in chemical a n d nuclear
%Cx10 2 %Cr%Ni
applications are shown in Table 4.47.
—— In another system used in the U.S.A., an auxiliary
cold strip fed by its own drive rolls or placed beneath
the arc to provide additional filler metal increase
deposition rate and reduce dilution. The electrode strip
is 50-75 mm wide by 0.75-1.5 mm thick, while the cold
strip is 40 mm by 1.25-1.50 mm. The parameters are
1,200 Amps./32 V/380 mm per min travel speed. Clad
thickness can be varied between 4 and 9.5 m m by
STRIP ELECTRODE
STRIP FEEDING
POWER SOURCE
CONTACT PIECE
FLUX FEEDING
5 6 7 8 9 10 MM SOLIDIFYING POOL
LL [DISTANCE FROM PARENT

IB. PLATE SURFACE
BASE METAL
FLUX SLAG
DEPOSIT
HEAT AFFECTED ZONE
WELDING DIRECTION
Fig. 4.61: Cr, C and Ni contents as function of the distance
from original plate surface (as-welded specimen). A = Fig. 4.62: Schematic representation of double strip
original plate surface; B = surface of cladding layer after electrode cladding system for two 60 x 0.5 mm strips in
treatment; C = fusion zone combination with a special flux.
Table 4.47: Welding conditions and results of double strip cladding on carbon steel plates with a proprletory flux
Strip Current Arc voltage Weld speed Overlapping
(mm) (amp) (V) (cm/min) (mm)
2x60x0.5 1250 31 30 7
Reinforcement Width Dilution Deposit rate Cladded rate
(mm) (mm) (%) (kg / hour) (m2/hour)
4.3 60 9 28 0.85
C% Si% Mn% Cr% Ni% Nb% Delta-
Ferrite
Strip El. 0.02 0.40 2.0 23.5 12.0 0.90 16
1st layer 0.04 1.1 1.1 20.0 11.0 0.70 9
adjusting the travel speed and the feeding rate and mechanical properties of the weld by proper selection
dimensions of the cold strip. of the flux-wire combination. However, it sometimes
becomes difficult to ensure sidewall penetration and to
Narrow Gap SAW avoid slag inclusions. Also slag removal between passes,
The conventional SAW of thick plates involves single visual inspection of each individual pass and removal
or double-V preparations as mentioned earlier in the of defective weld metal and making weld repair have
section on SAW procedures. Figure 4.63 shows the to be managed. In MIG/ C 0 2 welding, arc stability,
difference between conventional and narrow gap weld effective gas shielding and the effect of magnetic fields
preparations for 75 mm thick plate. It is seen that in on the arc may cause problems.
narrow gap welding (NGW), less metal has to be In NGW by the SA process, one may use the single-
deposited with reduced welding time and less distor- centre pass method or the two-pass layer method as
tion. Also with a constant root gap, a reproducible weld shown in Fig. 4.64.
pass can be deposited over the entire thickness and thus The single-centre pass method: This requires less
the welding procedure can be readily automated. welding time but makes it difficult to control sidewall
NGW has been carried out successfully in many penetration, though in some cases this is overcome by
shops using the SA as well as the MIG/C0 2 process. With mechanical oscillation of the wire. A relatively high heat
the SA process, one can weld without spatter, produce input is generally required to ensure thorough fusion,
welds with good bead shapes and can control the but this factor may not be favourable for welding certain
Fig. 4.63: Comparison of conventional and narrow gap weld preparations for welding 75 mm thick plate: (A)
Conventional, area of preparation 1095 mm2; (B) Narrow gap, area of preparation 935 mm2
/ /
//A
/////7x;'/s I
I
Fjg. 4.64: Alternative methods of narrow gap SAW: (A) Single centre pass method; (B) Two-pass layer method
low-alloy steels, where heat input has to be restricted. specific size of grinding disc, to provide access to the
Removal of solidified slag, especially at the side walls, root of the gap for removal of defective metal.
can be an additional problem. Gap width of 15 mm minimum, enables one to use
Two-pass layer method: These drawbacks are not found grinding discs of 600-650 mm diameter and 12.7 mm t.
in the two-pass layer method, though it requires nearly Welding capability for three typical wire sizes and gaps
50% more welding time. The need for repair can be for the two-pass layer method is indicated in Table 4.48.
completely avoided. For this method, the gap width is Experience has shown that while joint edges can be
critical and depends on the wire size, as indicated in kept parallel up to 20 mm gap depth, 1° angular incline
Fig. 4.65. The gap width has to be compatible with a on each side in deeper grooves allows for shrinkage and
ensures faultless sidewall fusion.
NGW is ideally suited for the circumferential joints
I * . g - 4 5d-6 5d —«( of pressure vessels, because they can be completed
- • ) 1.5d-3.5d | » -
w i t h o u t i n t e r r u p t i o n , t h u s fulfilling the m a i n
r e q u i r e m e n t for integral a u t o m a t i o n . For this
application, system functions have to be provided as
follows:
a) A means of keeping constant the vertical distance
from the end of the contact tip to the work piece. This is
also necessary to allow for work piece ovality.
b) A means of keeping constant the horizontal
distance between the wire and one wall of the gap.
c) A means of keeping constant the horizontal
distance between the wire and the other wall of the gap
(b and c becoming operative on succeeding passes).
Fig. 4.65: Gap width control in narrow gap welding d) On completion of each 360° rotation plus some
suitable angular overlap, a means of moving the welding
Table 4.48: Welding capability for a range of wire and gap sizes
Weld metal
Wire Welding current doposition rate Time for gap
dia, Gap (low-medium-high) (low-medium-high) filling; time
mm mm amps kg / hour index no.
3.2 15 390^50-500 4.50-5.15-5.67 148-129-117
4 18 520-595-670 5.92-6.80-8.00 135-118-100
5 22 686-(790)-(900) 7.90-9.60— 124-(102)—
Note: Current values in brackets represent higher than allowable current levels, as mentioned in the section on limiting factors.
wire. i.e. the torch, from one gap wall to the opposite ignition failure. If this occurs, the wire electrode-feed
one. Functions a, b and c are relevant to a conventional speed interacts with the apparent arc voltage, causing
two-axis guiding system, where the axes are vertical and stubbing and resultant weld defect. Recently, square
horizontal, the latter being at right angles to the weld wave AC power sources similar to those used in TIG
travel direction. The only special feature is that, at the welding, with extremely rapid polarity transitions are
end of each revolution a probe changes the pre-load being preferred. It is claimed that using such a source,
direction and senses the opposite gap wall. arc outages at polarity reversals are virtually non-
Function (d) represents the simplest version of a bead existent, electrode feed and burnoff rate are uniform,
locating program to be repeated at completion of each heat input and sidewall penetration are consistent and
revolution. In spite of this simplicity, the system is still the magnetic arc blow effects associated with a DC arc
able to be self-correcting. If the gap width is not constant, do not occur. It is also claimed that with square wave
the welding wire is always correctly positioned in AC source, arc operation is possible at much closer
relation to the gap wall, thus avoiding lack of proximity to the sidewall without arc blow effects and
penetration. consequent inconsistent sidewall penetration and that
The system will function automatically until insuf- square wave AC power, unlike sine wave AC, can be used
ficient land is left for the sensing probe to operate effec- with fluxes formulated for AC welding, i.e. arc stabili-
tively. From this position onwards (i.e. approximately sers and ionisers need not be incorporated in the fluxes.
10 mm of wall remaining), the gap has to be filled by The flux has to be specially formulated to ensure that
manually guiding the welding torch. This requires that the slag is fragile and self-detachable even at high
the system be provided with an additional function: preheat temperatures, welds are free from defects and
e) When the horizontal sensor is deactivated-push a stable arc is maintained. The flux should also generate
buttons for manual guidance. Manual guidance presents minimum amounts of gas, because of the small space
no particular problems as visibility is unrestricted and in the narrow gap. The bulk density of the flux should
the task is relatively simple for an average operator. be lower than conventional fluxes, so that it is not so
In the top portion of the joint, all regions which are sensitive to the depth/width ratio of the flux layer and
below level because of misaligned or irregular gap width gases generated in the weld zone can escape easily.
have to be filled on with manually guided weld beads. Above all, the flux must yield weld metal of high impact
Additionally, one must provide a means of resistance demanded for the welded product.
preventing the cylindrical vessel from creeping on the Owing to the low degree of dilution with the base
turning rolls, so as to prevent the horizontal guiding metal and the limited energy inputs associated with
axis from reaching its limit switch. narrow gap submerged-arc welding, the weld metal
NGW demands a high level of accuracy in wall produced is characterised by high ductility and good
preparation and joint fit-up. Assembly and shrinkage notch toughness. With the exception of the capping
tolerances must be closely monitored in order to bead, each bead is affected by the thermal cycle
maintain a satisfactory gap width. Axial displacement associated with subsequent beads, resulting in the initial
of the component being welded must not exceed 0.5 mm; coarse-grained microstructure being retained only in
but since this is difficult to achieve in practice, it becomes small areas, whereas the remainder is recrystallised and
necessary to correct the electrode position constantly refined. It has been noted that, in comparison with
by seam tracking equipment. The welding equipment conventional SAW, narrow gap SAW can be carried out
must be so designed that the weld nozzle must not get using wires with lower strength or less alloying content
overheated; there should be no accidental arc strikes without impairing the mechanical properties of the
between the contact nozzle and the side walls; flux weld. This may be due to the homogeneity of the weld
should be fed uniformly; and excess flux and slag metal through the thickness of the weld due to the
remnants must be completely extracted. An automation constant degree of parent metal dilution, the refinement
system must be provided, which among other things, of the microstructure noted above, and the effects of
can maintain constant contact-tip/work piece distance the thermal cycle associated with relatively low heat
despite any work piece ovality, and constant wire/ inputs. The possibility of using lower strength, lower
sidewall distance. alloy wires may be advantageous, because it reduces
The welding power source must have a high level of the danger of hydrogen cracking, and possibly
reliability. DC power can lead to arc blow problems, solidification cracking in the weld metal, and may allow
while AC drooping type power source can cause arc re- a reduction in preheat temperature.
Literature survey shows that narrow gap SAW has on the dirty floor. Dirt must not be allowed to get into
been successfully used on a wide range of applications. the flux. The flux recovery system must be equipped to
The typical ones are listed in Table 4.49. separate and remove the dust and foreign particles from
the flux.
SA Weld Defects: Causes and Prevention Flux rendered moist during storage and handling
Weld defects arising from i m p r o p e r w e l d i n g can give rise to weld porosity.
parameters and procedures have been pointed out Joints inside tanks and vessels may sometimes
earlier in the section on Process Parameters. We shall contain mill scale dropped from the plate surfaces,
now discuss defects arising from other factors. which can contaminate the weld. The mill scale may
• Porosity also get into the flux. Flux recovery systems are
Porosity is commonly caused by rust, oil, dirt and sometimes e q u i p p e d with magnetic separators to
moisture, which may be present on the joint surfaces. remove the scale and other magnetic materials.
Rust is removed by intense wire brushing. Sometimes Flux may sometimes be trapped between the base
surfaces having rust and red scale need to be torch of the root weld bead and the bottom side of the joint,
heated to 250-350°C to drive away residual moisture in and cause porosity which may remain inside the weld
them. Degreasing solvents must be used to remove oil, or rise to the surface as large holes. This porosity can be
grease and die lubricants. It is always a good practice to avoided by either keeping a margin of 4 mm between
heat the joint edges with a gas torch about 300-600 mm the base of the weld bead and the bottom edge of the
in front of the arc during the welding operation, at least joint or letting the weld penetrate fully into the other
for the root pass or first pass. side. This is explained in Fig. 4.66.
Porosity may also be caused by welding wire which • Moisture influxes
is rusty or oily. Rusty wire should be scrapped; but the Fluxes must be used in a dry condition. The inherent
wire covered with oil, grease or dirt must be cleaned moisture content of a welding flux depends upon its
before use. general nature, i.e. fused or agglomerated, its chemistry
If the layer of flux during welding is shallow, arc a n d its m e t h o d of m a n u f a c t u r e . Most critical
flashes can be seen through it. This is an indication that applications demand the control of flux moisture and
the flux is inadequate and this may cause scattered wire cleanliness (freedom from rust, grease, oil, etc.), so
surface porosity. Insufficient flux depth commonly as to reduce the resultant weld metal hydrogen content
occurs on circumferential welds, especially when the below a certain level related to the section thickness and
diameter of the cylindrical section is small. In such cases, restraint of the fabrication being welded. Failure to
the flux must be allowed to pile up around the arc by reduce weld metal hydrogen below this level may result
mechanical support. in both, weld metal and HAZ cracking.
Since the flux is repeatedly recovered and reused, it While some fused acidic and semi-basic fluxes have
is likely to pick up contaminants, especially when the inherently low moisture contents, they are unable to
plate surfaces are not clean or the flux is allowed to spill provide weld metal with the consistent o p t i m u m
Table 4.49: Typical applications of narrow gap SAW

Applications Type of material Thickness, mm
Vessels for chemical and petrochemical industry 2.25 Cr-1 Mo 225
Reactor pressure vessels, steam generators Not available 275
Big water pipes ST52 Grade 3 120
Vessels for boiler industry, water turbines Not available 200
Vessels for the chemical and petrochemical industry 2.25 Cr-1 Mo or 1.5 Cr-0.5 Mo Not available
Pressure vessel and boiler, constructions, bridges, Carbon steel and 250
industrial machines low-alloy steels 100
200
Reactor pressure vessels (for the Navy) and boilers 2 Mn-1 Mo 120
Reactor pressure vessels, steam generators Not available 350
Nuclear reactor components 22 NiMoCr37 20MnMoNi55 600
Vessels for the boiler and nuclear industry 15NiCuMoNB5 120
Vessels for the boiler and nuclear industry Not available 200
g < 4 MM
BACKING WELD
POROSITY PRESENT! IPOROSITY ABSENT
Fig. 4.66: Effect of penetration on porosity
mechanical properties (toughness) demanded in critical ensure the supply of consistently low agglomerated flux
applications. Accordingly, semi-basic and basic moisture contents. This is based upon the use of high
agglomerated fluxes, which are hygroscopic to some final baking temperatures during manufacture (for
degree must be used to provide optimum mechanical example 800°C minimum in the case of OP41TT),
properties. It is therefore critical that the manufacturing followed by packaging hot into sealed drums or
process of an agglomerated welding flux is controlled moisture resistant plastic bags, depending on customer
to provide not only low final flux moisture content, but requirements. This final baking procedure also provides
also a high resistance to moisture re-absorption under a high resistance to subsequent moisture re-absorption.
fabrication shop conditions. This is particularly so for In Table 4.50 we reproduce the hydrogen
the fully basicfluxes,which tend to be used in the most determination results of two Oerlikon fluxes when they
demanding applications. are used for welding in the as-received condition.
Leading producers of agglomerated fluxes such as How these fluxes react to the humidity of the air in
Oerlikon of Zurich have developed manufacturing storage is shown graphically in Figs. 4.67 and 4.68. The
procedures and quality control systems designed to absorption characteristics are expressed in terms of the
Table 4.50 Hydrogen levels in welds deposited with Oerlikon fluxes in as-received condition
Flux Batch No. Hydrogen (ml / 100 gm Hydrogen
weld deposit) (ml / 100 gm fused metal)
OP41TT 028 001 3.76 1.01
018 013 4.07 1.63
OP121TT 038 027 6.67 2.67
038 028 4.69 1.88
"T—T " T - 1 — r i—i—r i—i—I—i—I—i—r [20

h- 20°c 0P12ITT 20 h 20*C
EEMM MEDIUM*
5 90% £| OP4ITT
Q.
5
Q.
Q-
o
cc
Q
cc 5
—' LLI 50%HH- ^ O
t CO 5 O
S Q
Q if K/EftV LOW
1 2 3 4 5 6 7 8I
J— I I TIME.
EXPOSURE T I DAYS
I L JLJO
Fig. 4.67: Effect of exposure of OP41TT on the resultant Fig. 4.68: Effect of exposure of OP121TT on the
weld-metal hydrogen content resultant weld-metal hydrogen content
diffusible hydrogen contributed to the weld metal by

flux exposed to varying levels of humidity, the hydrogen
being measured by the standard IIW test described at (/)
the end of this section. D
The data show that both fluxes always remain in
LL.
CL
5
CL
very-low/low hydrogen category under the usual shop z

conditions. OP41TT has better resistance to moisture, LU
o
because it is baked at 800°C during manufacture. The
II
O cc <
cc
a rsfli
presence of alloy additives in OP 121 TT precludes the >
i 10
use of a high baking temperature and the water content LU Q
is slightly higher after exposure to atmospheric CD

C/>
-1 3
D <
humidity. INITIAL
The rebaking schedule for the two fluxes in case they 5 5 A 5
have picked up objectionable levels of moisture is H/ERY LOW)
evident from Figs. 4.69 and 4.70, which show the effect 250 350 450 550
of rebaking for two hours at various temperatures upon -L
"REftJwNfl TEMPERATURE *c
the weld metal hydrogen levels. The data show that
rebaking the flux at 300°C for two hours is sufficient to Fig. 4.70: Effect of rebaking exposed OP121TT on the
resultant weld-metal hydrogen content
restore both fluxes to the very-low category, after
exposure to conditions more severe than those likely to
(c) excessive dilution of the weld with the parent ma-
be encountered in normal practice. The only special
terial, resulting in pick-up of carbon and other
precaution to be observed is that the flux must be baked
alloying elements and consequent hardening,
in shallow trays about 100 mm deep.
• Cracks (d) excessive hydrogen in the arc derived from mois-
Weld cracks are seldom encountered in mild steel ture, grease, oil, etc.
welding, when the sections are less than 10 mm thick. It is convenient to discuss fillet weld and butt weld
In thicker sections, cracks may appear in the welds due separately.
to one or more of the following factors: Fillet welds: Weld bead shape influences weld
(a) rapid cooling of the deposit, cracking in fillet welds as shown in Fig. 4.71.
(b) heavy restraint in the joint,
rr-*-i " 1 —
—' HJ 2 0
[ 0P41TT IMEDIUMM
5 2
CL
CL a.
CL
—115 2 TB. EXCESSIVE;

/A. SMALL WELD FACE AND 1
O CONCAVITY)
EXCESSIVE PENETRATION
o I ^90% cc <
E5^
u -J
-410 LU Q
' LU
CO _
L70% ^ C/> CO FREE FROM CRACKS /-TS
XI
it W ^ o
"- CL •' I «
5 O O
MNTTIAL
K/ERY LOWI
I
■^zzi.
250 350 450 550
_J
I i xREBAKING
i 1 U C & D LARGE WELD Ai
FACE AND
TEMPERATURE °C SOME CONVEXITY
Fig. 4.69: Effect of rebaking exposed OP41TT on the Fig. 4.71: Effects of bead shape on cracking
resultant weld-metal hydrogen content in fillet welds
When cracks are caused by excessive weld metal may not be compatible with the weld beads produced
dilution, welding procedure must be modified to reduce by the submerged-arc and other processes employing
weld metal penetration by taking the following steps: . high current and arc energy levels. An IIW document
a) Change electrode polarity from positive to IIA-593-83 has suggested that it is possible to perform
negative the SA weld test on one or more of the original size of
b) Increase electrode size centre test piece shown at (a) in Fig. 4.74, if it is rotated
c) Change from single electrode to twin-electrode 90° about a vertical or a yertical and horizontal axis and
method aligned as shown in Fig, 4.74 (b) and (c). and used with
d) Deposit first pass with a manual electrode instead the fixture shown in Fig. 4.73. In this case, the run-on
of with the SA process and run-off pieces have to be compatible with the new
e) Use bevel joint. The angle of bevel must be 60° cross-sections and the length has to be increased to 80
minimum to ensure that the width is greater than the mm.
depth. The root pass in the bevel may be deposited The suggested IIW procedure is as follows:
manually with a basic low-hydrogen electrode. a) Electrode size: The electrode size is linked to the
Butt welds: In butt welds also, abnormal bead shape current used and the size of the weld bead. It will be
is often the cause of cracking. Examples are given in necessary to use the welding fixture shown in Fig. 4.73.
Fig. 4.72. b) Flux drying: The flux should be dried to the
If the cracks are caused by pick-up of carbon and maker's instructions. At least I kg of flux is required for
other alloys from the base metal due to excessive triplicate welds. Drying should be done in an open dish
dilution, suitable steps may be taken as recommended placed in a drying oven set at the correct temperature.
u n d e r the 'fillet w e l d s ' to reduce dilution and A suitable dish size is 250 x 200 mm, which allows a 15
penetration. mm flux depth. The dried flux should be used immedi-
ately after it is removed from the drying oven. If there
Determination of Hydrogen in SA Welds are no drying instructions, the flux should be dried at
The standard methods of determining hydrogen 350°C for 2 hours. The flux should spend the full speci-
described in Chapter 3 are mainly applicable to welds fied time at the drying temperature and other fluxes
deposited by the m a n u a l metal arc process. The should not be placed in the oven during this time. Used
dimensions of the test piece assembly of these methods flux should not be cycled for re-use.
c) Test piece assembly: The run-on and run-off pieces
are 30 x 10 (Or 15) x 135 mm length as shown in Fig.
(A) CRACKING DUE TO
4.73. The centre sample block is aligned as shown in
Fig. 4.74 (b) or (c). The preparation, degassing and use
<ns>
LARGE WIDTH AND
LOW DEPTH of the test piece assembly is as described in the original
method. The flux is kept at a predetermined constant
depth of 30 mm by levelling off along the top of the
copper foil inserts.
d) Making the test weld : Triplicate test welds, each of
(B) CRACKING DUE TO SMALU total length 260 mm, should be made on different test
WIDTH AND HIGH DEPTH
piece assemblies. The deposit should be along the centre
line of the test piece assembly. The time spent in
deposition should be noted.
(C) FREE FROM CRACKS When welding is complete, the weld specimen is
quenched and may be stored as described in the original
method, after which it is cleaned and analysed for
hydrogen content.
(D) EXCESSIVE CRACKING DUE e) Welding conditions : The range of consumable wire
TO NARROW NECKING diameters and therefore the range of currents are being
investigated by IIW to ensure compatibility with the
welding fixture. When data are available, a table of wire
Fig. 4.72: Effects of bead shape on cracking in butt diameters and recommended welding conditions will
welds
be made available. Typical conditions are:
Wire diameter 4 mm JAPANESE GLYCERINE METHOD

Welding current 625 + 25 Amps. The Japanese method of determining hydrogen in
Polarity DC, electrode positive SA welds is described in JIS Z 3116-1980. The test plate,
which is 18 mm thick, 35 mm wide and 100 mm long, is
Voltage as recommended or 30 : I V
initially heated at 400°C for 1 hour to expel hydrogen.
Welding speed 10 ± I mm/sec A weld bead is deposited on three such test plates closely
Stick-out 30 ± 5 mm placed longitudinally as shown in Fig. 4.75. The test
Depth of flux 30 mm piece is obtained by cutting off the run-on and run-off
parts of the test material.
A small ball of wire wool can be used to promote arc The flux is redried under conditions recommended
initiation, by the manufacturer before use. Standard electrode wire
k— 3 0 -
ISPECIMEN!
1MM THICK ANNEALED COPPER
SHEET
10 GAP
-300-
E I i
2
S
•' 11
ZIIZ.~JJ'SS"SSJLZ "J~ zfcfrrrjt!ii^H

—-o
^jr^-vA-\UtvX-^"i--i-îi|| ^>
^T^^t^^jj^j^t^t^ 1=N
150
il n
yi.
la
285 LONG TESTPjECE ASSEMBLY j
Mi iPi
L
%
! i
31
DIMENSIONS IN M M
■n7" WATER COOLING TUBES
Fig. 4.73: Welding fixture for high currents (centre sample can have dimensions as shown at (b) and (c) in Fig. 4.74)
(A) HEAT INPUT £ 30 KJ/CM
(B) HEAT INPUT ^ 30 KJ/CM
(C) HEAT INPUT £ 30 KJ/CM
DIMENSIONS IN MM
Fig. 4.74: Alternative alignments of specimens in the welding fixture
2PP.
T
fjiXL* 100
:ctt<<i*«tt<ca«:?££3;&s
IRUN-ON PART TEST PIECE IRUN-OFF PART
s
APPROX 210
2C
DIMENSIONS IN MM U- 35
APPROX 40 APPROX 70
Fig. 4.75: Material for Japanese glycerine method
of 4 mm diameter is used. The test plate to be welded is The test material is quenched in ice within five
fixed with a copper coupling as shown in Fig. 4.76. The seconds of the completion of welding. Next, the run-on
welding is commenced at room temperature under the and run-off parts are removed. The test piece is cleaned
conditions of 500Amps. welding current, 33 V arc and put in the collecting apparatus based on glycerine
voltage and 30 cm/min arc travel speed. The length of displacement method. This operation has to be finished
the bead is about 210 mm, about 40 mm of it being on within 90 sec after finishing welding. Hydrogen is
the run-on part and about 70 mm on the run-off part as collected by immersing the test piece for 72 hours in
shown in Fig. 4.75. glycerine, which is maintained at 45°C. The collecting
300 j
ST ft'J0
1 4i
HP
1 i1 50| T #40
U 15
1 i 1
jlj 1 # CO1
1 *i! — | C O ^
a\
o»
ml 1
1
o
CO
DIMENSIONS IN MM
Fig. 4.76: Copper coupling for Japanese

glycerine method
\t- 521 DIMENSIONS IN MM
Fig. 4.78: Collecting burette for Japanese
glycerine method
to COLLECTING BURETTE up to 60 mm thickness. Fused, high silica type flux with
V^ around 50% Si0 2 content possessing very good welding
characteristics at the high currents was used. Typical
procedures are given below:
A
These welds had adequate tensile properties and
could meet impact values at room temperature. In those
GLYCERINE BATH
d a y s , SAW w a s u s e d on relatively u n i m p o r t a n t
o
o applications, in which very little emphasis was placed
on toughness properties of the weld metal and the heat-
affected zone.
In the 1960's as SAW began to be used extensively
in pressure vessels, bridges and in shipbuilding, there
w a s a m a r k e d s w i n g to Charpy-V testing at
DIMENSIONS IN MM UL.75-J temperatures below room temperature. It was then
found that the weld metal deposited by these high heat
Fig. 4.77: Collecting apparatus for Japanese input procedures gave poor notch impact values,
glycerine method presumably because of their coarse microstructure. It
was also realised that the notch toughness property
apparatus and shape and dimensions of the glycerine i m p r o v e d w i t h decreasing heat i n p u t a n d w i t h
bath are shown in Fig. 4.77. The volume of glycerine is increasing number of passes in a given joint. The heat
about 650 ml. The shape and dimensions of the collecting input could be lowered by decreasing welding current
burette are shown in Fig. 4.78. or by increasing welding speed for given values of
current and arc voltage. The large number of passes of
TOUGHNESS PROPERTIES OF SA WELDS relatively smaller cross-section helped to maximise the
In the early days of SA welding, i.e. during the 1940's, proportion of grain-refined structure in the weld
the inherent high current-carrying and weld deposition deposit. As an example, we have given the modified
capabilities of the process were fully exploited for procedure for 38 mm thick weld on the next paragraph.
achieving ultimate economies. Two-pass and one-pass As a consequence, SAW d e g e n e r a t e d from an
(with copper bar backing) welding procedures using originally h i g h - p e r f o r m a n c e p r o c e s s , w i t h weld
single wires up to 12 mm diameter and currents up to d e p o s i t i o n rates of 20 k g / h o u r a n d m o r e , to a
4000 Amps. AC were known and used for butt welds mechanized multilayer process depositing around 12
a) One-pass butt weld in 38 mm t plate

Electrode dia. — 9.5 mm
w
"V V AC current
Voltage
—
—
1,900-2,000 amp
40-43 V
Travel speed — 200 m m / m i n
Heat input rate — 25 k j / mm (ave)
b) Two-pass butt weld in 57 mm I plate
v -y Parameters Backing pass

at A
Finishing
pass at B
T '
-V B
Electrode dia. 8 mm 8 mm
AC current 1,600 amps 1,700 amps
i
124 mm
Voltage 38 V 37 V
Travel speed 228 m m / m i n 200 m m / m i n
A-\ Heat input rate 16 kj/min 19 kj/min
kg/hour. When more stringent impact requirements Investigations by Tuliani and Boniszewski regarding
came to be specified, fabricators were compelled to weld the as-deposited microstructure produced by different
in the parameter range of around 600 Amps./32 V/600 flux compositions had shown that of all the flux systems
mm per minute, and at deposition rates as low as 7-8
k g / h o u r (yet three times the deposition rates with
manual MAW electrodes). 2.0' n i —' '
This setback was short-lived. Almost at the same
t i m e , p r o g r e s s in h i g h d e p o s i t i o n r a t e s w a s ■ ^
1.5
achieved by developing tandem-arc multiple wire
• THREE WIRES
configurations featuring 2, 3 and 4 wires in which
heat input rate was restricted by employing very 1.0
high w e l d i n g s p e e d s , ' h e r e l a t i o n s h i p b e t w e e n
z
welding speeds obtained with 1, 2 and 3 wires in Q
butt welding a range of plate thicknesses is shown 0.5 - ONE WIRE N x ^ TWO WIRES J
^ x 1
in Fig. 4.79. For 10 mm plate thickness, the welding
speed ratio is 1:1.9:2. 7; for 20 mm it is about 1:1.9:3.0
and for 30 mm it is 1:1. 8:3. 3. In all the cases, the heat 0 1? £_ 30 ** 40
PLATE THICKNESS. MM
input is nearly constant for a given plate thickness, but the
welding power input ratio is approximately 1:2:3. Fig. 4.79: Welding speeds when using one, two and
The d e v e l o p m e n t of n e u t r a l a n d basic fluxes three electrode systems with electrodes connected to
separate power sources
(especially agglomerated types) with low Si0 2 contents
at this time also led to the improvement in weld metal
toughness. Early research studies showed that high Si0 2
(30-60%) fluxes tended to transfer Si into the weld metal
and also to introduce oxygen at the rate of 0.01% oxygen
per 0.1% Si. Si content and hence the oxygen content in
the weld metal decreased with increase in flux basicity
and CaF2 content in the flux. Grain refinement improved Electrode Current Voltage Welding speed Heat input
with decrease in Si content and increase in Mn/Si ratio. dia. amp V mm/min. rate kj/mm
Increasing flux basicity lowered the sulphur content of
the weld metal. All these factors combined to raise the Pass 1 5.5 1,050 36 228 10
Pass 2 5.5 1,100 36 305 7.8
energy absorption in ductile rupture and to decrease
Pass 3 5.5 1,000 34 203 10
the transition temperature.
used to produce C-Mn weld metal, a flux containing Charpy value of 85 J at 1.9 k j / m m fell to 30 J at 3.3 k j /
approximately 40% A1203 produced a microstructure mm and to the same low level of a fused silicate flux at
supposed to be most favourable to notch toughness in 7.6 k j / m m . In the case of an aluminate-basic flux
the as-deposited condition. The possible explanation for (basicity 1.45), the value of 75 J at 1.9 k j / m m fell to 40 J
this w a s given u n d e r "Aluminiak Fluxes" while at 3.3 k j / m m and levelled to 35 J over the further heat
discussing the aluminate fluxes. However, the A1203- input range u p to 7.6 k j / m m . A comparison of the
type flux used by them was acidic and gave a high transitional behaviour of the three fluxes (see Fig. 4.80)
percentage of Si in the weld metal, which was the reason showed that at heat input of 1.9 kj/mm, the highly basic
why optimum impact values were not obtained. Taking flux fared the best, the aluminate flux came next, and
a hint, another series of experiments were carried out then the fused-silicate flux. At heat input level of 7.6
using a flux containing 50% A1 2 0 3 , 38% of basic k j / mm, the fluxes shifted their positions; the aluminate
ingredients (CaO, MgO, CaF2 and MnO) and only 12% flux fared the best, then came the Mn-silicate flux, while
of acidic materials for the two-pass welding of EH2 the highly basic flux showed the least toughness.
quality ship plate. The results showed that the impact Such odd behaviour of the fluxes was explained
requirement could be easily met using SD3 wire (0.11 through microstructural studies of the weld deposits.
C, 1.62 Mn, 0.30 Si), and the impact values improved Fused flux gave high proportions of upper bainite (35-
further as SD3 grade wires alloyed with 0.50% Mo and 41%) and lath martensite (10-20%) over the entire heat
with 1.0 Ni/0.5 Mo were used. An important condition input range. In the case of the aluminate flux, the
in all these cases, however, was that the cooling rate of proportions of acicular ferrite and polygonal ferrite
the weld deposit must be sufficiently fast (i.e. heat input remained within the limits of 74-65% and 26-35%
rate must be restricted). respectively (while u p p e r bainite was absent and
In another interesting study of the influence of flux martensite was low) over the entire heat input range. In
basicity and heat input on weld metal notch toughness the case of the highly basic flux, while the microstructure
using SD3 wire (C-0.12, Mn-1.70), it was established at 1.9 k j / m m consisted of 70% acicular ferrite and 30%
that, at -20°C, the Charpy impact values of a fused Mn- polygonal ferrite, the microstructure at 3.3 k j / m m
silicate type flux (basicity 0.78) are uniformly very low d e v e l o p e d 12% of lath m a r t e n s i t e p l u s t w i n n e d
over the heat input range of 1.9-7.6 k j / mm. In the case martensite, and at 7.6 kj/mm, the proportion of lath
of an agglomerated highly basic flux (basicity 3.0), the martensite rose to 15%. This explained the disappointing
160
OP41TT (8 = 3.0) S
-140
- 120
3
>
O
-100
DC
Ul
Z / V B X 2 0 0 ( B = 1.45) 100
OJ - 80
> 80 BX200(B = 1.45)
CL
- 60
<
CC
X
o
- 4g,o/
- 20
• ^ * ^ LINCOLN 1 (B - 0 . 7 8 )
(A) 1 1 1 1 I I ! KB)
-60 -40 -20 0 20 40 60 -40 -20 0 20 60
TEMPERATURE <°C) TFMPERATURF CO
Fig. 4.80: Effect of heat input and flux basicity on the Charpy-V notch transitional behaviour of submerged-arc welds
in 25 mm C-Si-Mn-Nb-AI primary structural steel with SD3 wire at a 300° C interpass temperature: (A) 1.9 kJ/ mm; (B)
7.6 kJ/mm
notch toughness properties of the highly basic flux

registered at very high heat input levels.
FIRST PASS
A large number of research investigations, such as
•SECOND PASS
the ones above, which have been carried out the world
over, have led to the conclusion that the main factors
which control weld metal toughness are: flux basicity, ^--x— IMPACT SPECIMEN
wire chemistry, postweld heat treatment and, above all,
Fig. 4.81: Location of impact specimen in two-pass butt
welding heat input (expressed in k j / m m ) . These
weld
combined factors influence the impact property through
the microstructure they help to generate in the weld The question of obtaining tough single-pass welds
deposit. To have good impact strength and a low in spite of slow cooling rates was resolved in 1970 when
transition temperature, the weld metal must have a high a document was published for the Japanese Welding
proportion of the interlocking acicular ferrite structure Society, which was subsequently issued by the IIW as
and correspondingly low proportions of pro-eutectoid document 11-583-71. It dealt with the correlation
ferrite and upper bainite. between Ti, Mo, B and CaF2 contents of flux-cored wires
It is well known that increasing Mn level and making and the mechanical properties of submerged-arc welds.
Mo and Ni additions to C-Mn weld metal improves its It showed that it is possible to deposit weld metal in
sub-zero impact properties, because these elements are massive bead cross-sections and resultant low cooling
strong acicular ferrite formers. This fact is used with rates, but still exhibiting toughness values equal to those
advantage in the design of manual electrodes. In SAW, of the best multi-layer deposits. This property was a
however, it is found that the good effects of these result of the combined effect of various micro-elements
elements cannot be fully exploited, because at high input coupled with low O and N contents. B and Mo additions
rates these very elements prove detrimental to notch give h a r d e n a b i l i t y a n d hence a low austenite
toughness. It is for this reason that in SA welding of transformation temperature, producing a strong and
LPG containers, where impact strength of 41 J at -55°C tough acicular ferrite, while Ti prevents contamination
must be satisfied, the heat input rates have to be kept of the boron by N 2 or 0 2 . The total effect is to increase
within 1.0 to 2.2 k j / m m for plate thicknesses of 10-28 the proportion of acicular ferrite in the microstructure
mm in spite of using basic fluxes and expensive wires and refine the acicular ferrite grain size, thereby leading
alloyed with 2% Ni or 2.5% Ni-0.8% Cr-0.5% Mo. Yet to significant increases in sub-zero Charpy-V impact
reasonably high deposition rates are secured by values.
combining high currents (530 Amps, for 10 mm plate, The Japanese trials were made with the combination
750 Amps, for 25 mm plate) with high travel speeds of flux-cored wires of the folded type and Ca-silicate
(720-900 mm/min). type flux, which could not yield optimum impact values,
While multiple-wire a n d m u l t i p a s s w e l d i n g because the folded tubular wire was not capable of
procedures became the accepted means of getting SA taking in sufficiently large quantities of CaF2 in add it
welds of adequate toughness, studies were also under ion to the microelements. The problem was overcome
way to produce single-pass welds of fairly large cross- when a solid wire called TIBOR 22 was introduced by
section, which in spite of not having the benefit of grain Oerlikon in England to be used in combination with a
refinement, would satisfy the impact requirements. high fluoride, high basicity flux OP41TT.
These studies had practical significance, because in Typical composition of TIBOR 22 is:
shipbuilding there is a requirement for wire-flux C —0.09% Mo —0.30% S—0.015% max.
combination capable of butt welding plates of 19 mm Mn—1.2% Ti—0.04% P—0.015% max..
and 29 mm thickness with one pass on either side, and Si —0.10% B—0.003% Al—0.020% max.
at the same time capable of meeting the weld metal N—0.010% max.
impact requirement of 51 J at -10°C, the impact Constitution of OP41TT flux is:
specimens to be removed from the centre of the second Si0 2 + Ti0 2 — 10%
and final pass (see Fig. 4.81). The studies would also CaO + MgO—35% Suitable for DC + ve
interest the majority of fabricators, w h o o w n e d A1203 + MnO—20% at currents up to 800 Amps.
conventional single wire SA equipment and desired to CaF 2 —30%
get maximum productivity out of it. Basicity—3.1%
It is interesting to compare the impact values given below (vast improvement over the Al 2 0 3 -type fluxes
by this combination in butt welding 38 mm thick plate, described earlier):
type BS 4360:72 grade 50D, using 10-pass procedure at Plate : BS 4360:1972 Grade
3 k j / m m heat input rate in one case and 3-pass 50D
procedure at 6.4 k j / m m in the other. The comparison Plate preparation : 60°C Double-Vee with
given below shows that the weld toughness is hardly 5 mm root face
affected by heat input and by individual weld bead Welding conditions: Current 850 Amps.
cross-section. Voltage 35 V
a) Joint preparation: Speed 166 m m / m i n
Polarity DC + ve
Heat input 10.1 k j / m m
No. of runs 2
Charpy V-notch value: atl4°C 147 J
At -20°C 95 J
Side 1
Similar high impact values at -55°C are obtained
with 10 to 18 mm thick butt welds in LPG grade steel
Welding conditions: Current 600 Amps. using the combination of Fluxocord 35.21 and OP121TT
Voltage 30 V flux developed by Oerlikon. The former is a micro-
Speed 380 m m / m i n alloyed flux-cored wire designed on the pattern of
Polarity DC + ve TIBOR 22, while OP121TT is a flux of basicity —3.1, for
Heat input 2.8 kj/min DC and AC welding at currents up to 1,200 Amps.)
No. of passes 10 having the following constitution:
Si0 2 + Ti0 2 - 15%
Charpy V-notch test temp. °C + 20 -20 -40 60 CaO + MgO - 35%
Charpy V-notch test values, J 180 172 172 83 A1 2 0 3 + MnO - 20%
(90) (70) (38) H CaF 2 - 25%
Values in parenthesis are those obtained with S3 The sequence of passes and the impact values at —
wire having 1.5% Mn under identical welding and 5S C are indicated in Figs. 4.82 and 4.83.
testing procedures. Table 4.51 gives Charpy-V values at low
t e m p e r a t u r e s for the all-weld metal specimens
b) Joint preparation: deposited by OP41TT flux in combination with various
700
~ Side 1 grades of British wires in the as-welded (AW) and stress-
relieved c o n d i t i o n s . For p r e p a r i n g the all-weld
specimens, the standard procedure involving multipass
technique with fairly low heat input rates has been used.
Side 2
Welding conditions: Current 800 Amps. 78-80-84 J

Voltage 34 V
Speed 254 m m / m i n
Polarity DC + ve
Heat input 6.4 kj/min 116-118-120J
No. of passes 3
Charpy V-notch test values, J 170 165 168
148 67-75-81 J
Charpy V-notch test values, J 170 165 168 94-96-100J
148
Fig. 4.82: Charpy-V impact values of SAW welds at -55°
The combination also gives excellent impact values
C on cryogenic steel using tubular Oerlikon Fluxocord
at -20°C in two-pass welding of 50 mm t plate using 35.21 and OP121TT flux (t = 1 0 - 1 2 mm)
high heat input, as is evident from the example given
Table 4.51: All-weld mechanical properties from 0P41TT flux and various wires
In combi- Heat U.T.S. Y.S. Elonga- Charpy V Impact energy (j)
treat- N/mm2 N/mm2 tion
with wire ment (Lo-5d) 0°C -20°C -40°C -60°C -80°C -100°C
%
SD3 AW 580-630 >420 >22 >80 >70 >40 >20
SD3 SR 520-560 >370 >24 >120 >100 >80 >50
SD3Mo AW 630-680 500-540 >20 90-130 50-70
SD3Mo SR 570-630 460-500 >22 100-150 60-90
SD3 INi AW 560-630 450-500 >22 80-140 50-90 30-60
SD3 INi SR 520-570 390^40 >22 75-140 55-95 30-55
SD2 2Ni AW 600-650 500-540 >22 40-100
SD2 3Ni SR 585-595 410-440 >22 140-150 65-90 55-75 30-55
SD3 INiMo SR 600-680 500-560 >20 120-160 100-130 60-90 50-80
SD21NiCrMo AW 750-800 650-700 >20 50-65 20-35
SD21NiCrMo SR 800-830 690-730 >20 50-65 20-35
SD2 l 1/4 CrMo AW 730-770 660-700 >15
SD2 l 1/4 CrMo SR 590-620 500-530 >20
SD2 21/2CrMo SR 770-810 600-650 >18
SD2 5CrMo AW 820-880 740-790 >15
SD 2 5CrMo SR 800-840 690-730 >18
TiBoR 22 AW 500-530 390-420 >25 140-160 100-120 80-100
TiBoR 22 SR 450-480 360-380 >35 140-160 100-130
Note: For composition of wires, see Tables 4.6 and 4.7.
COD requirement of 25 or 0.30 mm at -10°C is specified

Table 4.52 gives COD results for samples taken from
full plate thickness w e l d e d with OP41TT flux in
117-131-141 J
combination with the wires indicated.
Case study
Plate type : BS 4360:72 Grade 50 D
68-86-98 J
Chemistry : C—0.15, Mn—1.38, Si—0.42,
S—0.018, P—0.015
Wire TiBoR 22
Flux OP41TT
64-65-72 J
Plate
2 preparation
92-94-102J
Side 1
Fig. 4.83: Charpy-V impact values of SAW welds al 55°C

on cryogenic steel using tubular Oerlikon Fluxocord
35.21 and OP121TT flux (t = 1 5 - 1 8 mm)
In addition to Charpy-V, crack opening displacement

A^K Side 2
(COD) tests are used to measure the notch toughness

property of welds. They are described in BS 5762-1979 Welding conditions: Current 700 Amps.
"Methods of COD testing". In critical fabrications used Voltage 30 V
in offshore constructions in the North Sea, minimum Speed 350 m m / m i n
Table 4.52: COD values from 0P41TT flux and various wires
In combination Heat Plate C.O.D. (mm)
with wire treatment thickness
(mm) -10°C -40°C -80°C
SD3 As-welded 60 0.58-0.65
SD3 S.R. 60 1.08-1.35
TiBoR 22 As-welded 60 0.90 0.80 0.2
TiBoR 22 S.R. 60 0.75 0.55 0.2
SD2 2Ni As welded 57 0.4-0.84
TiBoR 22
1j%MnS3 TYPE WIRE
C/>
o AS-WELDED
o
o
ONSET OF CLEAVAGE - 1 0 # C
TEMPERATURE C O
Fig 4.84: COD test curves for as-welded specimens
8
TiBoR 22 «
7 * ^ ^* V i f * Mn S3 TYPE WIRE
ok
6
STRESS-RELIEVED
CO
5
5 • ' / ^ ESTIMATED ONSET
OF CLEAVAGE -35*C
Q
O
U
4
V /
3
' y
1
¥
-80 -60 -40 -20 + 20
TEMPERATURE 'C
Fig. 4.85: COD test curves for stress-relieved specimens
Polarity DC + ve The COD test curves shown in Figs. 4.84 and 4.85
Heat input 3.54 k j / m m confirm the excellent fracture toughness of TiBoR 22/
No. of runs 6 OP41TT welds in both as-welded and stress-relieved
Interpass temp. 150°C conditions.

The Procedure Handbook of Arc Welding, 12th Ed., published 1973 by The Lincoln Electric Company, U.S.A.
Metals Handbook, Vol. 6, 8th Ed., published 1971 by American Society for Metals.
Fluxes and Slags in Arc Welding by Coe. WIRR 75/1978 of Nov. 1978.
Welding Process Technology by P.T. Houldcroft.
SAW-Combining Increased Deposition Rates with Improved Mechanical Properties by H. Baach. Advani-Oerlikon Tech. Bull.
No. 215-216, May-Aug. 1981.
Submerged-Arc Welding Fluxes by P.S. Visvanath. Advani-Oerlikon Tech. Bull. No. 218, Nov.-Dec. 1981.
Relationship between Flux Basicity and Mechanical Properties in SAW by Garland and Kirkwood. Weld. & Met. Fabr., Apr.
1976.
Design of a Special Weld Mill for Large Diameter Thin Wall Pipes by W.S. Byard et al. Met. Constr., Sept. 1970.
High Current Density SAW with Twin Electrodes by J.E. Hinkel & F.W. Forsthoefel. Weld Journal, Mar. 1976.
A New High-Efficiency Welding Procedure Using Finely Divided Filler Metal by R.J. De Haeck. Weld Journal, Oct. 1971.
SAW and Surfacing with Hot Wire Additions Weld. Journal, May 1972.
Three-Wire SAW of Line Pipe by G.D. Uttrachi & J.E. Messina. Weld. Journal, June 1968.
Multipower SAW of Pressure Vessels and Pipe by R.A. Kubli & H.I. Shrubsall. Weld. Journal, Nov 1956.
Notch Toughness Properties of Commercial SA Weld Metal by Colvin & Bush. Weld. & Met. Fabr., Feb. 1971.
Towards Improved SA Weld Metal by Garland & Kirkwood. Met. Constr., May/June 1975.
Effect of Ti and b Additions to SA Welds in SOD Plate By Still & Rogerson. Met. Constr., July 1978.
The Influence of Cold Wire and Fluoride Additions on Weld Metal Toughness in SAW by Anderson, Baggerud & Thaulow.
Weld. & Met. Fabr., Nov. 1975.
Hot-Wire Methods of Surfacing and SAW by Lee. Met. Constr., Nov. 1972.
S.S. Cladding Strip Electrodes by Almqvist & Egeman. Weld. & Met. Fabr., July 1963.
Weld Cladding with Wire Electrodes by Armstrong. Weld. & Met. Fabr., March 1971.
Influences of Fluxes on SA Cladding with Stainless Strip Electrodes by Zentner. Weld. & Met. Fabr,, Apr. 1976.
The Narrow Gap SA Welding of Steel: A Tech. Review, by Kennedy. WIRR 233/1984 of Feb. 1984.
SA Narrow-Gap Welding by Altamer. Met. Constr., Oct. 1980.
High Deposition Rate Cladding-Broad strip 180 mm and Double Strip 60 mm Wide by Neff et al. Conference Proceedings of
'Advances in Welding Processes' 1974 by the Welding Institute.
Development of Four-Station Quadruple-Electrode One-Sided Machine for SAW by Terai 259 al, Ibid.
Water Absorption and Redrying Characteristics of Two High Basic Agglomerated Fluxes by Dr. Evans. IIW Doc. IlA-410-77.
Water Content in SA Fluxes and its Influence on Weld Cracking by Dr. Willi & Baach. JWS Symp., Tokyo, Nov. 1971.
Hj Content of Welds Deposited by Different Welding Processes Metals Tech. Conf, Sydney.
CHAPTER 5
Tungsten
Inert-Gas Arc
Welding (TIG Welding)
TUNGSTEN INERT-GAS Arc Welding, as the name to stabilize the AC arc. The high-frequency ioniser,
suggests, is a process in which the source of heat is an however, could not remove the inherent imbalance
arc formed between a non-consumable tungsten elec- between the voltage on alternate half-cycles, which
trode and the workpiece, and the arc and the molten resulted in a DC component that tended to saturate the
puddle are protected from atmospheric contamination transformer. This problem was finally overcome by
(i.e. oxygen and nitrogen) with a gaseous shield of in- using large capacitors in series with the arc,
ert-gas such as argon, helium or argon-helium mixture. appropriately termed DC suppressors.
Filler metal, if required, is added externally to the arc in These successful developments of the process were
the form of a bare wire by the welder. It is often referred a boon to the fabricators of aluminium and magnesium
to in abbreviated form as TIG welding. Some authors and their alloys, especially the aircraft and chemical
prefer to call it inert-gas tungsten-arc welding. industries, because until then no suitable process was
The American Welding Society refers to the process available to weld these alloys. Manual metal-arc welding
as gas tungsten-arc-welding and has given it the letter is not convenient, because the electrode coatings tend
designation GTAW. While a few countries such as the to be hygroscopic (causing weld metal porosity) and
U.S.A. and Canada use helium and argon-helium mix- there is danger of burn-through with thin gauge
tures besides argon for arc shielding, most other material. With this process and with gas welding, there
countries including India use only argon, as helium is is always the risk of flux entrapment in the joints, which
not available. In the latter countries the process is some- can reduce their corrosion resistance. Gas welding
times called Argon arc Process for the sake of simplicity. (oxyacetylene welding) is a slow process and gives rise
The concept of using gases in place of fluxes to shield to uncontrollable distortion. The more concentrated heat
the welding arc was first tried out by Roberts and van input of the TIG process as compared to gas welding
Nuys in 1919, and by other scientists in the following enables welding speeds to be increased, distortion to
years. Besides inert gases, hydrogen and hydrocarbons be minimised, and the metallurgical quality of the weld
were also considered. A workable method of inert-gas to be improved.
welding was first developed by the Northrop Aircraft
Co. of the U.S.A. in 1940, using a simple tungsten BASIC EQUIPMENT
electrode torch and a DC welding generator. It was The basic TIG equipment is shown in Fig. 5.1. The
successfully used on thin-gauge magnesium and essential components are (a) the welding power source
stainless steel. designed to deliver AC or DC or both as required; (b)
The main target of the process was aluminium but welding torch with a tungsten electrode which is gas-
initially it was found that the refractory oxide film on cooled or water-cooled; (c) supply of inert gas; and (d)
the aluminium surface interfered with the arc unless controls.
AC was used. AC could be used successfully only when The most important component of all TIG equipment
by 1946 it was found that a spark ioniser could be made is the high frequency (HF) unit, by means of which high-
A
ELECTRODE HOLDER
TUNGSTEN
ELECTRODE
GAS
PASSAGES
WELDING INERT
MACHINE GAS
SUPPLY
SHIELOING GAS
WORKPIECE
Fig. 5.1: Schematic representation of basic TIG equipment
frequency high-voltage is superimposed on the welding In the case of DC TIG welding, once the arc is struck,
current. Due to this, the shielding gas (argon, helium or the HF superimposition can be cut-off. However, in the
their mixture) gets ionised. The electrons which become case of AC TIG welding, HF superimposition is required
free during the process of ionisation form a conducting continuously to improve arc stability. Also in the case
path between the workpiece and the tungsten electrode of AC TIG welding, filter capacitor, commonly known
as shown in Fig. 5.2. Thus the arc can be started without as DC suppressor, has to be connected in series in the
directly touching the tungsten electrode to the output circuit to improve arc stability. To appreciate its
workpiece. Starting the arc by touching the tungsten function in AC TIG welding, the phenomenon called
electrode to the workpiece must be avoided as it arc rectification must be understood.
contaminates and wears out the electrode tip, and gives Arc rectification: Electric current flows more readily
rise to tungsten inclusions in the weld deposit, thereby in one direction than in the other between the tungsten
degrading the weld metal properties. electrode and the workpiece. The difference in current
ELECTRODE (-VE)
a) Flow of electrons (- ) is from electrode to the workpiece.

PROTONS-**/ ® ' © / * f B E E ELECTRONS b) Flow of protons (+) is from the workpiece to the electrode.
c) Workpiece gets 66% of the total heat and electrode gets 33%
(A) of the total heat generated.
ELECTRODE (+VE)
a) Flow of electrons (- ) is from workpiece to the electrode.
b) Flow of protons (+) is from the electrode to the workpiece.
ROTONS-*/©
/ © / /S%_
©/ c) Workpiece gets 33% of the total heat and electrode gets 66%
»© ^ / F R E E ELECTRONS of the total heat generated.
(B) ( ^ WORKPIECE / > f l
Fig. 5.2: Electron flow and heat distribution due to polarity condition: (A) Straight polarity; (B) Reverse polarity
TUNGSTEN INERT-GAS ARC WELDING (TIG WELDING) 263
flow is considerable with some metals such as During the positive half-cycle, the flow of gas ions
aluminium, magnesium and their alloys. Secondly, this (positively charged) is from the electrode to the
resistance to current flow in one direction (similar to a workpiece and the flow of free electrons (due to
diode action) is greater when heavy oxide film is present ionisation of gas) is from the workpiece to the electrode.
on the base metal. Aluminium oxide shows this The positively charged gas ions called protons are
particular property predominantly. (In other words, in attracted to the negatively charged workpiece, and strike
one AC cycle, there is greater resistance to the flow of its surface with sufficient force to break up and dislodge
current in positive half-cycle than in negative half-cycle.) the oxide layer. It should also be noted that the oxide
This results in rectification of the AC current. breaking action with the heavier gas argon (atomic wt.
The DC component in the AC current gives rise to 40), is much more pronounced than with the relatively
unbalanced AC waveform as shown in Fig. 5.3, which lighter helium gas. This explains why argon is always
in turn leads to instability in the arc as explained in Fig. preferred as a shielding gas for welding aluminium.
5.4. Due to unbalanced AC waveform, the positive half- Thus, when the positive half-cycle duration is
cycle duration is reduced. The duration of positive half- reduced due to the unbalanced waveform, the oxide
cycle is very important for metals like aluminium and cleaning action becomes less effective leading to arc
its alloys, which form a heavy refractory oxide film on instability. In addition, the DC component arising from
the surface, because it helps to disrupt the oxide film. arc rectification may cause magnetic saturation of the
This oxide scavenging action can be explained as core of the main transformer, leading to distortion in
follows: primary currents and overheating.
The DC component can be reduced by connecting a
bank of capacitors of sufficient value in series, in the
+ VE HAtF
welding circuit. These capacitors get charged during the
negative half-cycle by the excess energy, and this
charged energy in the capacitors gets discharged during
DC COMPONENT the positive half-cycle. Thus a balanced and symmetrical
waveform is obtained. The DC component in the AC
current is actually filtered or suppressed by the capacitor
bank. That is why the capacitors bank is commonly
called filter capacitors or DC suppressor.
, 5.3: Unbalanced AC waveform due to DC
component The arc rectification phenomenon as explained above
is due to inherent properties of the base metal, and hence
it is called inherent rectification. Arc rectification can
also occur due to other factors, which can be explained
as follows:
When the arc voltage and current are passing
through the zero period, the arc is extinguished. The
voltage necessary for reignition of the arc, i.e. restriking
voltage, depends on many factors, like
1) Temperature of tungsten electrode
ARC CURRENT
2) Temperature of workpiece
RfcSTRIKING VOLTAGE 3) Surface conditions of workpiece
ARC REIGNITION /
4) Shielding gas/flow rate.
POSITIVE /
When the arc voltage becomes zero after passing
through positive half-cycle, restriking of arc takes place
-k
HALF CYCLE/
with ease. This is because of electron emission from the

ARC VOLTAGE/
/ARCEXTINGUISHED> / NEGATIVE HALF CYCLE
hot tungsten electrode having negative polarity. Also a
' \ / small percentage (1 to 2%) of thorium is added in the
OPEN CIRCUIT
VOLTAGE
tungsten electrode to improve electron emission
properties.
Fig. 5.4: Arc instability caused by DC component in AC However, when the arc voltage becomes zero after
TIG welding passing through the negative half-cycle, the electrode
is having positive polarity, and hence there is no electron the welding stops, etc., are incorporated in modern
emission. As the workpiece is at lower temperature, it control units.
cannot emit electrons readily. As a result/arc restriking Gas pre-flow is a must for good quality TIG welding,
voltage is higher when arc voltage becomes zero after as it removes air from the torch and gas hose before the
passing through negative half-cycle. Exhaustive tests arc strikes, a n d e n s u r e s complete a n d perfected
have shown that during current reversals involving shielding of the electrode and base metal even before
change from electrode negative to electrode positive, the arc is struck and during arc striking.
about 100 to 450 V are required for arc restriking. Up-slope and down-slope controls for current are
So in this case, arc restriking is not easy and the arc sophisticated features especially useful in mechanised
r e m a i n s e x t i n g u i s h e d partially or completely welding applications. Up-slope avoids the initial current
throughout the positive half-cycle. This gives rise to arc surge, and down-slope minimises the crater formation
rectification, which results in unstable, noisy and fluttery at the end of the weld bead.
arc, affecting the heat input, penetration as well as bead Gas post-flow is essential for important welding
shape. applications, as it helps to shield the weld puddle at the
Generally AC w e l d i n g machines designed for end of the weld bead from atmospheric contamination
manual metal-arc welding have an OCV of about 70 to and protects the hot tip of the tungsten electrode from
80 V, which is not sufficient to restrike the arc during oxidation until it cools, thereby extending its life. Safety
every positive half-cycle. Increasing the OCV beyond controls such as the following may also be provided:
80 V is not advisable, due to the potential danger of 1) Water pressure/flow control safety device. This en-
electric shock to the operator. sures that unless sufficient water-flow with optimum
If by means of a suitable auxiliary device, high pressure is built up for the cooling of the water-cooled
voltage pulses can be fed in the welding circuit when torch, the controls will not work.
the arc current and voltage passes through zero, the arc 2) Gas wastage reduction control: After starting the
will be rendered smooth and stable and better welding unit, if the welding operation does not start within the
action will be obtained. In practice, high voltage, high next 6-8 sec (may be due to bad earthing/negligence of
frequency (HF) pulses (radio frequency) are injected in the operator, etc.), the gas circuit gets cut off automati-
the welding circuit by using conventional spark gap cally, thereby avoiding the wastage of costly shielding
oscillator circuits. The use of HF is advantageous in two gas.
ways:
• A s the high voltage pulses are at radio frequency, Power Source
the average power in the pulses is very small. This This is either a transformer, a rectifier or a motor
eliminates any danger from high voltage shock to the generator set, depending on the application. The power
operators, and it does not affect heat input of the source characteristic is essentially drooping type, which
welding current. means that power sources designed for and used in
• The high-frequency high-voltage pulses give rise manual metal-arc welding (MMAW) can be directly
to a series of accelerated collisions between shielding used for TIG welding. For better arc stability and a
gas atoms, which result in complete ionisation of gas. smooth arc, the OCV of the power source should be
The free electrons arising from this ionisation form a between 70 and 80 V (RMS).
c o n d u c t i n g p a t h b e t w e e n the electrode a n d the
workpiece and lead to easy arc starting. TIG TORCHES
Thus for easy arc starting in the case of DC TIG The function of the TIG torch is to conduct both the
welding, HF injection system is required; and for good welding current and the inert gas to the weld zone. The
arc starting as well as for a smooth and sustained stable tungsten electrode which supplies the welding current
arc in the case of AC TIG welding, wave balancing is held rigidly in the torch by means of a collet that
system (filter capacitor unit or DC suppressor) and arc screws into the body of the torch. Collets of various sizes
stabilizing system (HF injection device) are essentially are available to suit the diameter of the electrode to be
required. These systems form integral parts of all the used. Gas is fed to the weld zone through a nozzle which
TIG controls. consists of a ceramic cup. Gas cups are threaded into
In addition, controls like gas pre-flow, up-slope for the torch head and they control the direction and
current, gas flow control valve, down-slope control for distribution of the gas. The cups can be interchanged to
current as a crater-filling device, gas post-flow even after suit a variety of gas flow rates. A gas lense is sometimes
fixed on the nozzle to eliminate turbulence of the gas TYPICAL EQUIPMENT

stream, which tends to pull in air and cause weld TIG equipment is available in India in the following
contamination. The gas lense has a permeable barrier three forms:
of concentric fine-mesh stainless steel screens that fits
into the nozzle. Fig. 5.5 shows the cross-sectional view 1) TIG control outfit
of a water-cooled torch, while Fig. 5.6 shows the 2) Complete AC and DC TIG welding system
assembled torch along with the ceramic nozzle and the 3) Composite AC/DC TIG welding outfit
gas lense. 1) Control outfit: This outfit is meant for shops in-
When gas turbulence is minimum, welding can be tending to utilise their existing AC and DC manual
done with the nozzle as far as 25 m m from the metal-arc power sources for TIG operations. The outfit
workpiece. This enables the welder to see the weld pool is either an AC/DC or a DC type. The AC/DC type must
more clearly and to reach difficult places, such as inside be purchased when aluminium, magnesium and their
corners. alloys are to be welded and it must be used in conjunc-
Torches for manual TIG welding and for automatic tion with an AC power source, i.e. a welding trans-
TIG welding have the same basic construction as former. The DC outfit must be purchased when ferrous
described above, but the manual torch is provided with metals and non-ferrous metals, except aluminium, mag-
a handle in which control switches may be incorporated, nesium and their alloys, are to be welded and it must
while the automatic torch has no handle and is designed be used along with a DC power source, such as a weld-
to be clamped in a holding bracket. ing generator or rectifier.
Fig.5.5: Cross-sectional view of a water-cooled torch
Actually, the difference between A C / DC and DC

types is that the former is p r o v i d e d with a DC
suppressor assembly. The A C / DC unit can be used for
the welding of ferrous metals by connecting it to a DC
power source, but in this case the DC suppressor is
disconnected.
The control outfit comprises:
a) A control unit with built-in contactor, HF unit,
water recirculatory unit (in case of water-cooled
models), and DC suppressor (in case of AC/DC units
WITH INSULATOR
only). The water recirculatory unit incorporates a
radiator which is cooled by a sturdy axial fan. The water
Fig. 5.6: TIG torch using several fine-mesh screens (gas line is provided with a water pressure switch for positive
lenses) to reduce turbulence and produce a smoother protection of the torch against overheating.
gas flow b) A suitable welding torch with accessories.
c) A set of accessories consisting of welding cables, gas is intended for all ferrous metals and for non-ferrous
hose, gas regulator-cum-flowmeter, welding helmet and metals other than those mentioned above.
gloves, set of tungsten electrodes, etc. 3) Composite AC/DC TIG welding outfit: This outfit is
In these control outfits, facility for gas pre-flow and principally an A C / D C transformer rectifier unit having
post-flow is provided through an electromagnetic built-in TIG welding controls. It provides easy switch-
solenoid valve to ensure perfect shielding and protection over from AC to DC operation, so that it can be used on
of the electrode tip and weld. The necessary operations ferrous as well as non-ferrous applications. The
are performed through the torch switch control. The composite outfit is therefore an AC TIG system plus a
torch switch consists of the front part and the back part. DC TIG system built into a single unit. It is preferred by
The torch, the control unit and the accessories together a fabricator who uses the TIG process regularly to weld
form a four-track On/Off system, which operates as the complete range of industrial alloys.
follows: The outfit is available in three models with welding
current ranges (in two overlapping ranges) indicated
To start welding
below, and is supplied along with a welding torch and
Press back part : Gas is On (i.e. pre-flow starts)
a set of accessories:
Press front part : Welding contactor is On
HF is On Model Current range, Amps.
Gas is On AC DC
Release front part : HF is Off (in case of DC) AD-150 25-150 25-120
AD-350 35-350 35-280
Welding operation proceeds AD-500 40-500 40-400
To stop welding
Its salient features are:
Press back part : Welding is Off
Gas flow continues (post-flow) a) Facility for gas pre-flow and post-flow through
Release back part : Gas is Off an electro-magnetic solenoid valve to ensure perfect
shielding and protection of welds. The necessary op-
Table 5.1 provides current ratings of the standard erations are performed through the four-track On/Off
models of control units. They all are designed for input system as in the case of the control outfits.
supply of 230 V, single-phase 50 cycles. A typical model b) Built-in high frequency unit and DC suppressor.
of AC/DC control unit is shown in Fig, 5.7. c) Built-in liberally designed water recirculatory sys-
2) Complete TIG welding system: This system consists tem (in case of water-cooled versions) incorporating a
of the above-mentioned control outfit and a matching radiator cooled by a sturdy axial fan.
power source. It is meant for fabricators wishing to have d) Water pressure switch is incorporated in the wa-
TIG equipment with its own exclusive power source. ter line for positive protection of torch.
The AC TIG system, comprising an A C / DC type control Table 5.2 gives complete specifications of the three
outfit and a welding transformer of matching rating, is models. Figure 5.8 shows the finished form of AD-350.
to be used for welding aluminium, magnesium and their TIG torches: They are either gas-cooled or water-
alloys. The DC TIG system consisting of a DC type cooled. The torch head is connected to cable hose
control outfit and a welding rectifier of matching rating assembly At the other end a suitable coupler is provided
Table 5.1: Ratings of standard TIG control units

Type Model Current ratings
AC, duty cycle DC duty cycle
60% 100% 60% 100%
Amps. Amps. Amps. Amps.
11/15 115 90 150 110
AC/DC 25 250 180 300 230
45/50 450 350 350 380
15 150 110
DC 30 — 300 230
50 500 380
Fig. 5.8: Finished model of 350-amp AC/DC TIG

equipment
Fig. 5.7: TIG control unit
Table 5.2 Specifications of AC/DC composite outfits
AD-150 AD-350 AD-500
Open-circuit voltage V 80/73 80/73 80/73

Operating voltage V 10-16 10-24 10-30
Max. welding current Amps. 150-120 350/280 500/400
Max. cont. current
@ 60% duty cycle Amps. 115/90 265/210 380/300
@ 100% duty cycle Amps. 90/75 200/160 285/220
Rating max. KVA 13.5 31.5 - 45
@ 60% duty cycle KVA 10.5 24 34
Insulation Class B B B
Cooling Type Forced air Forced air Forced air
Input supply
Voltage V 380/415 380/415 380/415
Phase No. .. .2 lines of 3o supply
Frequency Hz 50 50 50
Recommended switch
Fuse rating Amps. DP 35 DP 85 DE120
Torch Model BTG-15-3 BTW-30-5 BTW-50-5
Cooling (torch) Type Gas Water Water
for connecting the torch to the control unit. The torches INERT GASES
are supplied with a set of accessories. All these torches Argon is monatomic gas (i.e., its molecule consists
incorporate the switch control, which in conjunction of one atom instead of two atoms in the case of common
with the control unit and the accessories, provides the gases like oxygen, nitrogen, chlorine, etc.). It is extracted
four-track On/Off system explained earlier under from the atmosphere by liquefaction of air and refined
Control Outfits. Table 5.3 gives the specifications of to 99.9% purity. It is supplied as compressed gas in
standard torches. cylinders or as a liquid at temperatures below -184°C
INPUT SUPPLY
00
LI (choke)- RF.O FOR AD 350 ONLY. £
o
o
m
T^k- 33
z
>
33
O
£
m
r-
O
z
o
H
m
O
X
z
o
5
CD
-<
LINK POSITION CHART

j»v PRIMARY BQAftD tn*
m0 o»>
u <■■ ■ D t l
OS
*°
AUXILIARY BOARQ
,* A
\ . \
i n i u
Table 5.3: Specifications of TIG torches
Models Type of Current ratings Suitable for Max. length Standard

cooling tungsten of tungsten hose
AC, duty cycle DC, duty cycle electrode electrode length*
60% 100% 60% 100% dia. accommo- (m)
(Amps.) (Amps.) (Amps.) (Amps.) (mm) dated (mm)
A Gas 115 90 150 110 1.6, 2.4, & 3.2 75 3

B Water 270 195 300 225 1.6, 2.4, 3.2 150 5
&4
C Water 400 310 450 350 1.6, 2.4, 3.2, 150 5
4,4.8, & 6.3
*Longer hose lengths can be supplied when required.
in large insulated containers. For bulk storage, the latter Density of argon is 1.33 times that of air and ten times
form is preferred, but suitable equipment must be that of helium. After emerging from the torch nozzle,
provided to vaporise the gas for distribution on the shop argon forms a blanket over the welding area and
floor. provides an effective shield. Being lighter, helium tends
Helium is also a monatomic gas, one-tenth as light to rise in a turbulent manner around the arc. For effective
in weight as compared to argon, and is extracted from shielding, the flow rate of helium needs to be two to
natural gas which occurs in a b u n d a n c e in a few three times that of argon.
countries including the U.S.A. This explains why helium Another important feature of helium is that, it gives
is not freely available in India for industrial use. Helium appreciably higher arc voltages for a given arc length
has an extremely low liquefaction point (-272°C) and is and welding current as compared to argon as shown in
therefore commonly transported and used as a gas in Fig. 5.9. Since the arc energy is given by V x A (where V
high pressure cylinders. Helium has greater thermal is arc voltage and A is current), this means that the
conductivity and hence requires higher arc voltages and helium arc is hotter than the argon arc. Helium is
energy inputs than argon. This will be discussed in detail therefore preferred over argon for welding thick
under Welding Procedures. materials and those having high thermal conductivity
Argon purity: There have been instances where TIG (such as copper) or high melting points, and for high
welding problems have been traced to impure argon. speed mechanised welding.
The fabricator m u s t insist on s t a n d a r d argon of Figure 5.9 also shows that for welding thin materials
guaranteed purity. in the current range of 50-150 amps, argon has a
Purity requirements are laid down in IS:5760-1969, pronounced edge on helium and provides the operator
"Specification for Compressed Argon". These are: with more latitude in arc length to control the welding
Argon % min. 99.967 operation.
Oxygen ppm max. 10 Another point of distinction is that argon gives
Nitrogen ppm max. 300 equally good arc stability on AC and DC, and good
Hydrogen ppm max. 5 cleaning action with AC. Helium gives a very stable arc
Carbon dioxide and other carbon with DC, but with AC the arc stability and cleaning
compounds determined as 0 2 ppm max. 5 action are poor. Therefore, w h e n aluminium and
Water vapour, mg/1, max. 0.0056 magnesium are welded with AC, argon is always
preferred.
In the Appendix of the above standard are described Mixtures of argon and helium with up to 75% helium
the methods of determining these impurities in a content are used when some balance between the
chemical laboratory. The percentage of argon is obtained characteristics of both is desired.
by deducting from 100 the sum of the percentage of Argon-hydrogen mixture: Addition of hydrogen to
impurities. argon increases the arc voltage and provides benefits
0 50 100 150 200 250 300 350 400

ARC CURRENT A
Fig. 5.9: Voltage-current relationship with argon and helium shielding
similar to helium. Mixture with 5% H 2 sometimes helps Classification and r e q u i r e m e n t s for t u n g s t e n

to get cleaner welds with manual TIG welding. Mixture electrodes are given in AWS A5.12-80, "Specification
with 15% H 2 is used for high speed mechanised welding for Tungsten Arc Welding Electrodes". Table 5.4,
of tight butt joints in stainless steel u p to 1.6 m m r e p r o d u c e d from this specification, specifies the
thickness. It is also used for welding stainless steel beer chemistry of five standard classes of electrodes.
barrels and tube to tube-sheet joints in various stainless Tungsten electrodes are commonly supplied in the
steel types and nickel alloys. Mixture with 35% H 2 can diameter range of 0.25-6.35 mm and in lengths ranging
be used on all thicknesses of stainless steel if root gap in from 76 to 610 mm with either clean or ground finish.
the joint is kept within 0.25-0.50 mm. Excessive use of Clean finish means that after the drawing or swaging
H2 must be avoided as it causes porosity. The use of these operation, the surface impurities are removed by
mixtures is restricted to stainless steels, nickel-copper chemical cleaning. Ground finish means that the surface
and nickel alloys. On other metals, hydrogen produces impurities are removed by grinding.
unfavourable effects. Pure tungsten electrodes are less expensive, have a
Nitrogen: Nitrogen is sometimes added to argon for relatively low current carrying capacity and low
welding copper and copper alloys and pure nitrogen is resistance to contamination. They are generally used on
sometimes used as a purging gas while welding stainless less critical applications Compared to pure tungsten
steel. The term purging is explained later. Besides argon electrodes, thoriated ones have higher emissivity, better
and helium, many other inert gases including neon, current carrying capacity, longer life and greater
xenon and krypton exist in nature, but they are scarce resistance to contamination. With these electrodes, arc
and very expensive. striking is easier and the arc is more stable.
Zirconiated electrodes have properties intermediate
TUNGSTEN ELECTRODES between the above two. Sometimes in welding with AC,
Tungsten is preferred for electrode material, because they are found to work better than the others.
it is a refractory metal (i.e. highly resistant to heat), has Approximate current ranges for various classes and
a high melting point (3,410°C) and is a strong emitter of sizes of electrodes are given in Table 5.5, bearing in mind
electrons which stream across the arc path, ionise it and that the current carrying capacity of an electrode is also
thus facilitate the maintenance of a stable arc. Among affected by the design of the electrode holder, the
several refractory metals, tungsten has the best resis- extension of the electrode from the holder and the
tance to erosion in use. When carefully used, a tungsten position of welding.
electrode lasts a long time and hence it is justifiably It is evident from Table 5.5 that very high welding
called non-consumable. It, however, oxidises readily at currents can be used when the tungsten electrode is
the extremely high temperatures reached at the root of negative, while the current values have to be severely
the arc, unless it is protected with a shield of inert gas. restricted when the electrode is positive. This is because
Table 5.4: Chemical requirements of electrodes
AWS Tungsten, Thoria, Zirconia, Total other

classification min. % (by difference) /o 0/
/o
elements, max. %
EWP 99.5 — — 0.5
EWTh-1 98.5 0.8 to 1.2 — 0.5
EWTh-2 97.5 1.7 to 2.2 — 0.5
EWTh-3 98.95 0.35 to 0.55 — 0.5
EWZr 99.2 — 0.15 to 0.40 0.5
Note: EWTh-3 is a tungsten electrode with an integral lateral segment throughout its length, which contains 1.0 to 2.0 thoria. The average thoria
content of the electrode is as specified in this table
Table 5.5 : Typical current ranges for tungsten electrodes3
Direct current, Amps. Alternating current, Amps.

Straight Reverse 1Unbalanced wave Balanced wave
polarity polarity
Electrode EWP *' EWP
diameter EWTh-1 EWTh-1 EWTh-1 EWTh-1
mm EWTh-2 EWTh-2 EWTh-2 EWTh-2
EWTh-3 EWTh-3 EWP EWzr EWTh-3 EWP EWZr EWTh-3
0.26 up to 15 b up to 15 up to 15 b up to 15 up to 15 b
0.51 5-20 b 5-15 5-20 b 10-20 5-20 10-20
1.02 15-80 b 10-60 15-80 10-80 20-30 20-60 20-60
1.59 70-150 10-20 50-100 70-150 50-150 30-80 60-120 30-120
2.38 150-250 15-30 100-160 140-235 100-235 60-130 100-180 60-180
3.18 250-400 25-40 150-210 225-325 150-325 100-180 160-250 100-250
3.97 400-500 40-55 200-275 300-400 200-400 160-240 200-320 160-320
4.76 500-750 55-80 250-350 400-500 250-500 190-300 290-390 190-390
6.35 750-1000 80-125 325-450 500-630 325-630 250-400 340-525 250-525
a. All values are based on the use of argon as the shielding gas.
Other current values may be employed depending on the shielding gas, type of equipment,
and application.
b. These combinations are not commonly used.
66% of the total heat in a tungsten arc is generated at the extension is too long, even a relatively low current
the positive pole called anode and only 33% at the can cause the electrode to overheat and melt above the
negative pole called cathode as shown in Fig. 5.2. This terminus of the arc. Conversely, if the current density is
means that an electrode can carry a much higher current too low, the arc will be erratic and unstable.
without overheating when it is negative than when it is Often one can employ a small range of electrode
positive. The table also shows that the DC capacity of diameters to operate over a wide range of currents by
an electrode when it is negative is greater than its AC
capacity, and the AC capacity with an unbalanced wave
BASE METAL
is more than it is with a balanced wave.
Tungsten has very low electrical conductivity, and
therefore, heats up when current is passed through it.
When welding with tungsten electrodes, the arc tip
should be the only hot part of the electrode; the BACKING BAR
RELIEF GROOVE
remainder should be kept as cool as possible. One FOR SHIELDING GAS
method of preventing electrode overheating is to keep
the extension of the electrode from the collet short. If Fig. 5.10: A simple set-up for argon purging
Table 5.6: Tungsten electrode tip shapes and current ranges

DC, electrode -ve
Electrode Dia. at Included Steady current Pulsed current
dia, mm tip, mm angle, 0 range, Amps. range, Amps.
1.04 0.125 12 2-15 2-25
1.04 0.25 20 5-30 5-60
1.60 0.5 25 8-50 8-100
1.60 0.8 30 10-70 10-140
2.4 0.8 35 12-90 12-180
2.4 1.1 45 15-150 15-250
3.2 1.1 60 20-200 20-300
3.2 1.5 90 25-250 25-350
shaping or tapering them to a small tip diameter. Table metal or welding rod, and maintenance of an insuffi-
5.6 serves as a guide for electrode tip preparation for cient shielding gas flow are causes of contamination.
four commonly used electrode sizes with recommended g) The equipment and, in particular, the shielding
current ranges. gas nozzle should be kept clean and free of weld spat-
The following general recommendations from AWS A5. ter. A dirty nozzle adversely influences the gas shield.
12-80 will help to get the best out of tungsten electrodes: This contributes to improper gas-flow patterns and arc
a) The appropriate current (type and magnitude) wandering, which can result in poor quality. It may also
should be selected for the electrode size to be used. Too contribute to excessive electrode consumption.
high a current will cause excessive melting, dripping, Argon purging or backing: In welding thin sheets, the
or volatilisation of the electrode. Too low a current will underside of the joint can become hot enough to oxidise
cause cathode bombardment and erosion due to the low and produce a rough penetration bead. The argon gas
temperature and result in arc instability. from the torch, which adequately shields the top of the
b) The electrode should be properly broken or base metal, cannot give any protection to the lower or
ground tapered by following the supplier's suggested penetration side. For high quality work, argon backing
procedures. Improper breaking may cause a jagged end or purging is necessary to ensure smooth penetration bead
or a bent electrode, which usually results in a poorly and uncontaminated base metal. This is accomplished
shaped arc and excessive electrode overheating. by introducing argon atmosphere into the relief groove
c) The electrodes should be carefully handled and of a backing bar as shown in Figs. 5.10 and 5.12.
kept as clean as possible. In order to obtain maximum
cleanliness, they should be stored in their original pack- TIG WELDING PROCEDURES*
age until used. Stainless steels: TIG process is well suited for welding
d) The shielding gas flow should be maintained not the entire range of stainless steels, because the chemistry
only when welding, but also after the arc is broken and of the filler metal remains almost unaltered when it is
until the electrode cools. When the electrodes are prop- converted into weld metal. In the absence of flux and
erly cooled, the arc end will appear bright or polished: active gases, gas-metal reactions and slag-metal reactions
when improperly cooled, they may oxidise and will in the arc are totally absent and hence the weld is free
appear to have a coloured film on their arc end, which from non-metallic inclusions.
can, unless removed, affect the weld quality on subse- Alloying elements from the filler metal are almost
quent welds. All connections, both gas and water, should wholly transferred across the arc to the weld metal
be checked for tightness. including titanium, used as a stabiliser. (It may be
e) The electrode extension within the gas shielding mentioned here that titanium cannot be transferred across
pattern should be kept to a minimum, generally dic- the arc in flux-shielded processes, because this element
tated by the application and equipment. This is to en- has extremely high affinity for oxygen and passes off into
sure protection of the electrode by the gas, even at lower the slag as an oxide.) Because the satisfactory service
gas-flow rates. behaviour of stainless steel welds depends largely on close
f) C o n t a m i n a t i o n of the electrode should be tolerances of chemistry, TIG welding is the ideal process
avoided. Contact between the hot electrode and the base in many instances, whether a filler metal is used or not.
""Tables 5.7 to 5.15 are taken from Metals Handbook, Vol. 6 published by American society for metals.
TIG welding is best suited for metal thickness of 6.4 allowed to come into contact with the molten puddle.
mm and below. It can be used in all positions. DC from Argon is preferred to helium or argon-helium mixture.
a constant-current type power source is used with Table 5.7 shows joint designs and welding parameters
electrode negative to deposit a narrow, deep-penetration for welding austenitic stainless steel in the thickness range
weld. of 1.6-12.7 mm.
For initiating the arc, scratch starting should be Heat-resisting alloys: These alloys can be divided into
avoided as the weld gets contaminated with tungsten. four groups as follows:
High-frequency starting must be used. The electrode a) Solid-solution nickel-base alloys such as Hastelloys,
commonly used is EWTh- 2 (i.e. alloyed with 2% thoria), Inconel 600, 601 and 625.
because of its excellent current-carrying capacity and arc b) Precipitation-hardenable nickel-base alloys such as
stability. While welding, the electrode tip must not be Inconel 700, 702, 706, 722 and Rene 41 and Udimet 700.
Table 5.7: TIG welding procedures for stainless steel

Type of joint Electrode Welding curr., amps Arc travel Filler rod Argon
and weld dia., mm electrode negative speed dia. flow rate
(Ref. Page 274) F V O mm/min mm 1/min
Base metal: t = 1.6 mm

Butt, A or B 1.6 80 to 100 70 to 90 70 to 90 305 1.6 4.7
Lap, F or G 1.6 100 to 120 80 to 100 80 to 100 255 1.6 4.7
Corner, H 1.6 80 to 100 70 to 90 70 to 90 305 1.6 4.7
T or corner, E or J 1.6 90 to 100 80 to 100 80 to 100 255 1.6 4.7
Butt, A or B 1.6 6100 to 120 90 to 110 90 to 110 305 1.6, 2.4 4.7
Lap, F or G 1.6 110 to 130 100 to 120 100 to 120 255 1.6, 2.4 4.7
Corner, H 1.6 100 to 120 90 to 110 90 to 110 305 1.6, 2.4 4.7
T or corner, E or J 1.6 110 to 130 100 to 120 100 to 120 255 1.6, 2.4 4.7
Butt, A or B 1.6 120 to 140 110 to 130 105 to 125 305 2.4 4.7
Lap, F or G 1.6 130 to 150 120 to 140 120 to 140 255 2.4 4.7
Corner, H 1.6 120 to 140 110 to 130 115 to 135 305 2.4 4.7
T or corner, E or J 1.6 130 to 150 115 to 135 120 to 140 255 2.4 4.7
Butt, A or B 2.4 200 to 250 150 to 200 150 to 200 255 2.4, 3.2 7
Lap, G 2.4, 3.2 225 to 275 175 to 225 175 to 225 203 2.4, 3.2 7
Corner, H 2.4 200 to 250 150 to 200 150 to 200 255 2.4, 3.2 7
T or corner, E or J 2.4,3.2 225 to 275 175 to 225 175 to 225 203 2.4, 3.2 7
Butt, A or B 3.2 275 to 350 200 to 250 200 to 250 127 2.4 to 4.8 7
Lap, G 3.2 300 to 375 225 to 275 225 to 275 127 2.4 to 4.8 7
Corner, H 3.2 275 to 350 200 to 250 200 to 250 127 2.4 to 4.8 7
T or corner, E or K 3.2 300 to 375 225 to 275 225 to 275 127 2.4 to 4.8 7
Base metal: t: = 12.7 mm
Butt, C or D 3.2, 4.8 350 to 450 225 to 275 225 to 275 76 3.2 to 6.3 7
Lap, G 3.2, 4.8 375 to 475 230 to 280 230 to 280 76 3.2 to 6.3 7
T or corner, E or K 3.2,4.8 375 to 475 230 to 280 230 to 280 76 3.2 to 6.3 7
t (max)
-i I-
A—Square butt, no root gap B—Square butt, root gap = t
rZ-y
1.6 MM
T
C—Single-V butt with root face = 1.6 mm and no root D—Double-V butt with root face = 3 mm
gap
**&** \
E—Double fillet T joint F—Double concavefilletin lap joint
r^~nat
G—Double convex fillet in Up joint H—Square butt in corner joint
• 50*
J—Square butt in corner joint with root gap r t K—Single-V butt in corner joint
c) Fe-Ni-Cr and Fe-Cr-Ni alloys such as Incoloy 800, Aluminium alloys: These alloys are almost entirely
801, 802 and 901. welded by the inert-gas shielded processes, i.e. TIG and
d) Cobalt-base alloys such as HS-21, 25, 31 and 188. MIG, because they need no flux and give optimum weld
TIG process is commonly used for the welding of metal quality. When oxyacetylene welding or brazing
heat-resisting alloys in thicknesses less than 6.3 mm. with flux is used for joining, flux residues may remain
Argon is generally preferred, but argon-helium mixtures in difficult-to-reach places and internal surfaces which
may be used for thicker sections and for higher speeds may be totally inaccessible, resulting in premature
in thinner sections. DC with electrode negative is used. corrosion. Besides absence of flux, welding in all
Table 5.8 is intended as a guide to setting up correct positions is easy with the TIG process, because there is
procedures for welding butt, corner or T-joints. For fully no slag to be negotiated and washed out later.
penetrated T-joints, the current may be increased by 10 The thickness range commonly welded by the TIG
to 20 amps. Interpass temperature during welding is process is 1-9.5 mm for manual welding and 0.25-25 mm
not allowed to rise above 90°C. Oscillation of the for mechanised welding.
welding torch sometimes helps to prevent cracking by AC is used for Al welding as mentioned earlier,
changing the solidification pattern and to improve the because it provides the o p t i m u m combination of
surface finish of the weld bead. Edge preparation and current-carrying capacity, arc controllability and arc
joint fit-up need special care. Joint surfaces should be cleaning action. The power source is usually a single-
perfectly free from paint, grease, oil, oxide film and scale. phase welding transformer with an open-circuit voltage
When foreign matter containing sulphur and lead is of 80-100 V, which is primarily meant for manual metal-
present on the surface, heating may result in severe arc welding. The oxide layer also gives rise to some arc
cracks due to the diffusion of these elements into the rectification during the reverse polarity half of the AC
base metal. The need for cleanliness in handling the plate cycle, which results in poor arc stability and possible
material cannot be overemphasised. overheating of the transformer. To overcome this
Group (a) alloys (i.e. Ni-Cr and Ni-Mo types) are problem, filter capacitor (called DC suppressor) is
welded in the annealed condition and need no postheat inserted in the power circuit as pointed out earlier.
treatment. Alloys of group (b) need to be correctly heat The suitable electrodes are p u r e tungsten or
treated before and after welding to prevent cracking in zirconiated tungsten. The latter are preferred, because
the weld metal or HAZ. Group (c) alloys are easily they are less prone to contamination by aluminium and
welded by the TIG process without the need for preheat have slightly higher current capacities. The electrode
and postheat, p r o v i d e d m i n i m u m heat i n p u t tip should be hemispherical as shown in Fig. 5.11. The
commensurate with the base metal thickness and filler tip is prepared by using an electrode one size larger than
metal diameter is used. For cobalt-base alloys of group required for the welding current, taper grinding the tip
(d), the smallest diameter of 2% thoriated tungsten and forming the hemispherical end by welding for a
electrode that will carry adequate current should be few seconds with a current 20 amps higher than needed
used. For sound welds, argon flow rate should be on and holding the electrode vertically up. Contamination
the higher side (7 to 11.81/min), with additional argon of the electrode by aluminium occurs occasionally. Slight
for weld backing fed at the rate of 2.4 to 4.8 1/min. contamination can be burned off by using high current
Table 5.8 :: TIG welding procedijres for heat -resisting alloys

Base metal Filler wire Electrode Shielding Flow rate Current
t, mm dia, mm dia, mm gas 1/mm Amps.
0.25 0.50 1.0-1.5 Ar 5.6 to 7 10-15
0.50 0.75 1.5 Ar 5.6 to 7 15-25
075 0.75, 1.15 1.5 Ar 5.6 to 7 25-35
1.15 1.15 1.5 Ar 5.6 to 7 40-50
1.27 1.15 1.5 Ar 5.6 to 7 45-55
1.52 1.15 1.5 Ar 5.6 to 7 55-65
2.0 1.52 1.5 Ar 5.6 to 7 75-85
2.54 1.52, 2.3 2.35 Ar or He 5.6 to 7 95-105
3.3 1.52, 2.3 2.35 Ar or He 5.6 to 7 110-135
6.3 1.52, 2.3 2.35 Ar or He 5.6 to 7 130-200
and holding the arc on scrap metal. Excessive Parameters for welding butt joints in the thickness
contamination should be removed by grinding or by range of 1.6-12.7 mm are given in Table 5.9, and those
cutting off the end and reforming the tip to the shape for welding tee and lap joints are given in Table 5.10.
shown in Fig. 5.11. Argon is the usual shielding gas, but Copper and copper alloys: TIG is the most widely used
helium or argon-helium mixture is sometimes used to process for these materials, because of its intense
weld thick sections at high speeds, since it gives a more localised heat source, good control and suitability for
penetrating arc. use with or without filler metal. Sections up to 3.25 mm
thickness are welded easily, though sections up to 12.7
mm thickness can be welded satisfactorily. For
Jr^rLJL 3/4d
thicknesses above 12.7 mm, MIG is the suitable process.
In spite of the high thermal conductivity of these
materials, TIG process is able to give complete fusion
of the joints with minimum heating of the surroundings.
In welding precipitation-hardened Cu-alloys, HAZ
Fig. 5.11: Shaping of electrode tip for Al welding
must be kept narrow and this is possible with the TIG
process.
In TIG welding of aluminium alloys, a band of white Sections up to 3.25 mm thick are square butt welded
deposit appears alongside the weld bead. Chemical without using filler wire or preheat. For higher
analysis has revealed it to be aluminium oxide. It is thicknesses, both filler wire and preheat are necessary.
believed to be caused by the emission of electrons from DC with electrode negative is used, because it enables
the surface of the aluminium when it is negative one to use a smaller size electrode for a given amperage,
cathode. The electrons detach the oxide from the surface and produces maximum penetration. For beryllium
as they fly off. When the deposit band is of hairline coppers and aluminium bronzes, AC stabilised by high
width, it indicates that the shielding is just adequate. If frequency is used to avoid the formation of tenacious
the band is wider, it means that the gas flow is too much oxide films on these alloys.
and gas is being wasted. The range of wrought Cu and Cu-alloys commonly
Aluminium is a good conductor of heat and hence TIG welded are described in Table 5.11.
calls for preheat when the base metal thickness exceeds Pure copper, unless it has been deoxidised with
6.3 mm. At 6.3 mm thickness, preheat is optional phosphorus, silicon, aluminium, iron or manganese,
depending on the arc energy. For welding 9.5 mm thick contains oxygen in the form of gas or cuprous oxide,
material in the flat position, it is up to 200°C. For welding which causes porosity and reduces the strength of welds.
9.5 mm thick material in other positions and for welding For best results, one must use deoxidised coppers,
12.7 mm thick material in all positions, the preheat because they are free from cuprous oxide and contain
temperature may be as high as 315°C. residual phosphorus. Phosphorus combines with
Table 5.9: Butt welds in aluminium using AC
3 C
■ 0 to 3 mm
Vvy
(CHVO)
0 110°
| 1.6 m m V / ° t 0 4*8 Tgt
M i M i i 1 , : * ;2.4 mm
0-4.8 -H H—T 1 6.3 mm19 mm | f

0 to 4.8 mm
(FHV) i-r 38 mm
(FV)
Same as F but
Same as E-but
groove angle 110°
groove angle 90°
and inverted for (o)
(H) overhead welding
Table 5.9: Contd.

Table 5.9: Contd.
Argon
Metal Joint Welding Elect- Filler Current Number
thick- design position rode metal Nozzle Flow rate AC, of
ness, (see above) (b) diameter, diameter, diameter 1 / mm Amps. passes
mm mm mm mm
1.6 A F 1.6 2.4 9.5 9.5 70
A H,V 1.6 2.4 9.5 9.5 70
A O 1.6 2.4 9.5 11.8 60
2.4 A F 2.4 3.2 9.5 9.5 95
A H,V 2.4 2.4 9.5 9.5 85
A O 2.4 2.4 9.5 11.8 90
3.2 A F 3.2 3.2 11.0 9.5 125
A H,V 2.4 3.2 9.5 9.5 115
A O 2.4 3.2 9.5 11.8 120
4.8 B F 3.2 4.0 11.0 11.8 175 2
C H 3.2 4.0 11.0 11.8 160 2
B V 3.2 4.0 11.0 11.8 160 2
D O 3.2 4.0 11.0 14.0 170 2
6.3 B F 4.8 4.8 12.7 14.0 225 2
C H 4.0 4.8 12.7 14.0 200 2
B V 4.0 4.8 12.7 14.0 200 2
D O 4.8 4.8 12.7 16.5 215 2
9.5 B F 6.3 4.8 15.9 16.5 325 2
E F 6.3 4.8 15.9 16.5 360 2
F F 6.3 4.8 15.9 16.5 325 3
C H 4.8 4.8 15.9 16.5 250 3
E H 4.8 4.8 15.9 16.5 260 2
G H 4.8 4.8 15.9 16.5 250 4
B V 4.8 4.8 15.9 16.5 250 3
E V 4.8 4.8 15.9 16.5 260 2
F V 4.8 4.8 15.9 16.5 250 4
D o 4.8 4.8 15.9 19.0 275 3
H o 4.8 4.8 15.9 19.0 275 3
12.7 B F 6.3 4.8 15.9 16.5 350 3
E F 6.3 4.8 15.9 16.5 400 3
F F 6.3 4.8 15.9 16.5 350 4
C H 4.8 4.8 15.9 16.5 260 3
E G 4.8 4.8 15.9 16.5 270 3
G H 4.8 4.8 15.9 16.5 260 4
B V 4.8 4.8 15.9 16.5 260 3
E V 4.8 4.8 15.9 16.5 270 3
F V 4.8 4.8 15.9 16.5 260 4
D o 6.3 4.8 15.9 19.0 280 3
H o 6.3 4.8 15.9 19.0 280 4
Note: Welding speed is 200 mm/minfor all metal thicknesses except 3.2 and 4.8 mm for which it is 250 mm/min.
Table 5.10: TIG welding parameters for T and lap joints

Metal Welding Elect- Filler Argon Current Welding
thickness, position rode metal Nozzle Flow (AC) Preheat Speed, No. of
mm (a) diameter, diameter, diameter, rate, Amps. temp., mm/min passes
mm mm mm 1 / min °C
~L6 F L6 Z4 95 95 80 None 200
H,V 1.6 2.4 9.5 9.5 80 None 200
O 1.6 2.4 9.5 11.8 70 None 200
2.4- F 3.2 2.4 9.5 9.5 120 None 200
H,V 2.4 2.4 9.5 9.5 100 None 200
O 2.4 2.4 9.5 11.8 110 None 200
3.2 F 3.2 3.2 11.0 9.5 150 None 250
H,V 2.4 3.2 9.5 9.5 120 None 200
O 2.4 3.2 9.5 11.8 135 None 200
4.8 F 4.0 4.0 12.7 11.8 215 None 200
H,V 3.2 4.0 11.0 11.8 180 None 200
O 4.0 4.0 11.0 14.0 190 None 200
6.3 F 4.8 4.8 12.7 14.0 260 Optional 200 2
H,V 4.8 4.8 12.7 14.0 235 Optional 200 1
O 4.8 4.8 12.7 16.5 240 Optional 200 1
9.5 F 6.3 4.8 15.9 16.5 345 up to 200 200 2
H,V 4.8 4.8 15.9 16.5 290 up to 315 200 2
O 4.8 4.8 15.9 19.0 290 up to 315 200 3
12.7 F 6.3 4.8 15.9 16.5 375 up to 315 200 3
H,V 6.3 4.8 15.9 16.5 300 up to 315 200 3
O 6.3 4.8 15.9 19.0 310 up to 315 200 3
oxygen absorbed during welding and thus prevents the and fluxing action of silicon and the low thermal
formation of cuprous oxide. conductivity of silicon bronzes.
Copper alloys sometimes contain free-machining Iron and manganese are often present in small
additives like lead, tellurium and sulphur, which make a m o u n t s , b u t they h a v e no significant effect on
them susceptible to hot cracking during welding. weldability.
Presence of zinc in brasses and nickel silvers reduces Welding parameters for various types and sizes of
their weldability, and its adverse effect increases in joints in commercial coppers are given in Table 5.12.
proportion to the amount present. Zinc vapours are Welding parameters for various types and sizes of
toxic, and hence the welder must be protected from them joints and welding positions in silicon bronzes are given
during the welding operation. Presence of tin in tin in Table 5.13.
brasses and phosphor bronzes (up to 10% max.) has no Phosphor bronzes: These alloys in wrought form up
effect on weldability. Its vapours are also less toxic. to 12 mm thick are easily TIG welded. The process is
Beryllium, aluminium and nickel, especially the first also suitable for repairing or joining phosphor bronze
two, form tenacious oxide films on the plate material, castings. Alloys with tin content of 5% or above have
which must be scrupulously removed by cleaning before comparatively low thermal conductivity, and can be
welding. The oxide formation on the plate due to heat welded without preheat when the sections are up to 3.2
during welding must be controlled by gas shielding or mm thick. The alloy, 1.25% E, containing 1.3% tin has
by fluxes in conjunction with the use of AC. High- nearly four times the thermal conductivity of the other
strength beryllium coppers containing 1.7% and 1.9% three and demands higher current or lower travel speed
Be are more easily weldable than high conductivity Be- for welding.
Cu. Aluminium bronzes: These alloys up to 9.5 mm thick
A l u m i n i u m bronzes and silicon bronzes have are readily welded by the TIG process. The process is
excellent weldability. The latter are the most easily also used for repair welding of Al-bronze castings. Iron,
weldable among the alloys, because of the deoxidising manganese or nickel are included in the base metal or
Table 5.11: Welding of copper alloys with TIG process
Alloy Name Nominal chemistry, % Weldability

Of and ETP Coppers
Oxygen-free copper (OF) 99.95 Cu G
Electrolytic tough pitch copper (ETP) 99.90 Cu, 0.04 0 2 F
Deoxidised Coppers
Phosphorus-deoxidized copper, low-P (DLP) 99.9 Cu, 0.008 P E
Phosphorus-deoxidized copper, high-P (DHP) 99.9 Cu, 0.02 P E
Beryllium Coppers
High-conductivity beryllium copper, 0.6% 96.9 Cu, 0.06 Be, 2.5 Co F
High-strength beryllium copper, 1.75% 98.3 Cu, 1.7 Be G
High-strength beryllium copper, 1.9% 98.1 Cu, 1.9 Be G
Low-Zinc Brasses
Gilding, 95% 95 Cu, 5 Zn G
Commercial bronze, 90% 90 Cu, 10 Zn G
Red brass, 85% 85 Cu, 15 Zn G
Low brass, 80% 80 Cu, 20 Zn G
High-Zinc Brasses
Cartridge brass, 70% 70 Cu, 30 Zn F
Yellow brass, 65% 65 Cu, 35 Zn F
Muntz metal, 60% 60 Cu, 40 Zn F
Tin Brasses
Admiralty 71 Cu, 38 Zn, 1 Sn F
Naval brass 60 Cu, 39.25 Zn, 0.75 Sn F
Special Brasses
Manganese bronze A 58.5 Cu, 39 Zn, 1.4 Fe, 1 Sn, 0.1 Mn F
Aluminium brass, arsenical 77.5 Cu, 20.5 Zn, 2 Al (0.06 As) F
Nickel Silvers
Nickel silver, 65-10 65 Cu, 25 Zn, 10 Ni F
Phosphor Bronzes
Phosphor bronze, 1.25% E 98.7 Cu, 1.3 Sn (0.2 P) G
Phosphor bronze, 5% A 95 Cu, 5 Sn (0.2 P) G
Phosphor bronze, 8% C 92 Cu, 8 Sn (0.2 P) G
Phosphor bronze, 10% D 90 Cu, 10 Sn (0.2 P) G
Aluminium Bronzes
Aluminium bronze D, Sn-stabilised 89 Cu, 7 Al, 3.5 Fe (0.35 Sn) G
Aluminium bronze D 91 Cu, 6-8 Al, 1.5-3.5 Fe, 1 max Mn G
Silicon Bronzes
Low-silicon bronze B 98.5 Cu, 1.5 Si E
Low-silicon bronze A 97 Cu, 3 Si E
Copper Nickels
Copper nickel, 10% 88.6 Cu, 9-11 Ni, 1.4 Fe, 1.0 Mn E
Copper nickel, 30% 70 Cu, 30 Ni E
Note: E = excellent; G - good; F =fair.

Table 5.12:: Nominal conditions for TIG welding of commercial coppers
Work-metal Root Electrode Diameter Shielding Gas-flow Current Travel Number Preheat
thickness, opening, diameter, of welding gas (c) rate, Amps. speed, of temperature,
mm mm (b) mm rod, mm 1/mm mm/min passes °C
Butt joints-Square groove
1.6 0 1.6 not req. Argon 7.0 110 to 140 600 1 None
3.2 0 2.4 not req. Argon 7.0 175 to 225 550 1 None
3.2 3.2 2.4 2.4, 3.2 Argon 7.0 175 to 225 550 1 None
4.8 4.8 3.2 3.2 Helium 14.0 190 to 225 500 1 95
Butt joints-—160° Single-V groove, 1.6 mm Root face
6.3 1.6 max. 3.2 3.2 Helium 14.0 225 to 260 450 1 150
9.5 1.6 max. 4.8 4.8 Helium 19.0 280 to 320 2 260
Lap joints —Fillet welded (e)
1.6 0 1.6 1.6 Argon 7.0 130 to 150 500 1 None
3.2 0 2.4 2.4, 3.2 Argon 7.0 200 to 250 450 1 None
4.8 0 3.2 3.2 Helium 14.0 205 to 250 400 1 95
6.3 0 3.2 3.2 Helium 14.0 250 to 280 350 1 150
9.5 0 4.8 4.8 Helium 19.0 300 to 340 - 3 260
Outside corner joints—Square groove
3.2 1.6 max. 2.4 2.4, 3.2 Argon 7.0 175 to 225 550 1 None
4.8 4.8 max. 3.2 3.2 Helium 14.0 190 to 225 500 1 95
6.3 4.8 max. 3.2 3.2 Helium 14.0 225 to 260 450 1 150
9.5 6.3 max. 4.8 4.8 Helium 19.0 280 to 320 - 2 260
Outside corner joints—50° Single-Bevel groove, 1.6 mmi Root face
4.8 1.6 max. 3.2 3.2 Helium 14.0 205 to 250 400 1 95
6.3 1.6 max. 3.2 3.2 Helium 14.0 250 to 280 350 1 150
9.5 1.6 max. 4.8 4.8 Helium 19.0 280 to 320 . . 3 260
Inside corner joints Square groove, Fillet welded
3.2 3.2 max. 2.4 2.4,3.2 Argon 7.0 200 to 250 450 None
T-joint —Fillet welded
3.2 1.6 max. 2.4 2.4, 3.2 Argon 7.0 200 to 250 450 1 None
4.8 1.6 max. 3.2 3.2 Helium 14.0 205 to 250 400 1 95
6.3 1.6 max. 3.2 3.2 Helium 14.0 250 to 280 350 1 150
9.5 1.6 max. 4.8 4.8 Helium 19.0 300 to 340 - 3 260
Note: Using EWTh-2 electrodes, RCu, welding rod, and straight-polarity direct current
filler metal chemistry to avoid weld porosity. Since the Nickel alloys: Most of the nickel alloys are either solid-
thermal conductivity of these alloys is nearly the same solution (i.e. non-age-hardenable) type or precipitation-
as that of carbon steel, heat input requirements are not hardenable type. Examples of the former type are Monel 404
high and preheating is normally not required for sections (55 Ni, 44 Cu) and Monel 400 (66 Ni, 31.5 Cu, 0.90 Mn,
up to 9.5 mm thick. 1.35 Fe). Examples of the latter are Monel K-500 and Monel
In addition to argon, a special flux may be applied 502, which are Ni-Cu alloys with controlled amounts of Ti
to the joint edges to increase weld metal fluidity and and Al.
prevent oxidation of the base metal due to welding heat. Nickel and nickel alloy material (i.e. joint edges as
AC stabilised by high frequency is preferred. well as the surrounding metal) must be completely free
Parameters for welding Al-bronzes are given in Table from impurities such as grease, oil, paint, marking
5.14. Preheating is not required for the thicknesses covered. crayons or inks, lubricants used for forming, cutting
Table 5.13: Nominal conditions for TIG welding of silicon bronzes

Plate Current, Electrode Travel Diameter of Shielding-gas Number
thickness, Amps. diameter, speed, welding rod, flow rate, of
mm mm mm/min mm 1 / mm passes
Manual Welding
Square-groove butt joints, flat position
1.6 100 to 120 1.6 600 1.6 7.0 1
3.2 130 to 150 1.6 600 2.4 7.0 1
4.8 150 to 200 2.4 - 3.2 9.5 1
6.3 250 to 300 3.2 - 3.2, 4.8 9.5 1
6.3 150 to 200 2.4 - 3.2, 4.8 9.5 3
Square-groove butt joints, vertical and overhead positions
1.6 90 to 110 1.6 - 1.6 7.0 1
3.2 120 to 140 1.6 - 2.4 7.0 1
60° Single -V-groove butt joints, flat position
9.5 230 to 280 3.2 - 3.2, 4.8 9.5 3 to 4
12.7 250 to 300 3.2 - 3.2, 4.8 9.5 4 to 5
19.0 300 to 350 3.2 - 4.8 9.5 9 to 10
25.0 300 to 350 3.2 - 4.8, 6.3 9.5 13
Fillet-welded lap joints, vertical and overhead positions
1.6 110 to 130 1.6 500 1.6 7.0 1
3.2 1.40 to 160 1.6, 2.4 500 2.4 7.0 1
4.8 175 to 225 2.4 - 3.2 9.5 1
6.3 175 to 225 2.4 - 3.2, 4.8 9.5 3
9.5 250 to 300 3.2 - 3.2, 4.8 9.5 3
12.7 275 to 325 3.2 - 3.2, 4.8 9.5 6
19.0 300 to 350 3.2 - 4.8 9.5 12
25.4 325 to 350 3.2 - 6.3 9.5 16
Square-groove outside corner joints, flat position
1.6 90 to 110 1.6 - 1.6 7.0 1
3.2 120 to 140 1.6 - 2.4 7.0 1
50° Single'-bevel-groove outside corner joints, flat position
6.3 175 to 225 2.4 - 3.2, 4.8 9.5 3
9.5 230 to 280 3.2 - 3.2, 4.8 9.5 3
12.7 275 to 325 3.2 - 3.2, 4.8 9.5 7
19.0 300 to 350 3.2 - 4.8 9.5 14
25.4 325 to 350 3.2 - 4.8, 6.3 9.5 20
Fillet-welded square-groove inside corner joints, flat position
1.6 110 to 130 1.6 500 1.6 7.0 1
3.2 140 to 150 1.6, 2.4 500 2.4 7.0 1
4.8 175 to 225 2.4 - 3.2 9.5 1
Fillet-welded T-joints, flat position
1.6 110 to 130 1.6 500 1.6 7.0 1
3.2 140 to 160 1.6, 2.4 500 2.4 7.0 1
4.8 175 to 225 2.4 - 3.2 9.5 1
6.3 175 to 225 2.4 - 3.2, 4.8 9.5 3
9.5 230 to 280 3.2 - 3.2, 4.8 9.5 3
12.7 275 to 325 3.2 - 3.2, 4.8 9.5 7
19.0 300 to 350 3.2 - 4.8 9.5 14
25.4 325 to 350 3.2 - 4.8, 6.3 9.5 20
Note: Using zero root opening, no preheat, EWTh-2 electrodes, RCuSi-A welding rod, argon shielding gas, and straight-polarity direct current.
Table 5.14 : Parameters for TIG welding of Al-bronzes

Base metal Root gap, Electrode Filler rod Argon Amps No. of
t,mm mm dia, mm size, mm flow rate, passes
1 / mm
Square butt joints
1.6 max. 0 1.6 1.6 9.5-14 25-80 1
1.6 3.2 1.6 max. 2.4 3.2 9.5-14 60-175 1
3.2 3.2 max. 4-4.8 4 14 210 1
70°-V butt joints
9.5 0 4-4.8 4 14 210-330 4
Fillet in T-joint or square groove inside corner joint
9.5 OforT
9.5 max. 4-4.8 2.4 14 225
for corner
fluids, shop dirt, etc., as they contain low-melting metals with each pass and ultimately interferes with smooth
like bismuth and elements such as lead, sulphur and welding and gives rise to oxide stringers along the weld
p h o s p h o r u s , which induce embrittlement d u r i n g axis. These oxide inclusions are likely to induce stress-
welding or any other form of heating. The complete area corrosion cracking in service.
on both sides of the material which will get heated must For TIG welding of nickel alloys generally, DC
be thoroughly cleaned. This usually involves a series of electrode negative is recommended. The welding torch
washings with substances like acetone, methylene should be held at an angle of 90° to the work. The largest
chloride, alkaline cleaners, hot water, and alcohol possible gas-nozzle size must be selected. The distance
depending upon the nature of impurities. between the nozzle and the work piece must be as short
The solid-solution type alloys can be as easily welded as possible. The shielding gas can be argon, helium or a
by either of the TIG, MIG and manual metal-arc welding m i x t u r e of the t w o . Argon w i t h 5% h y d r o g e n is
processes as austenitic stainless steels. sometimes used for single-pass welding of pure nickel
The precipitation-hardenable alloys require special to avoid porosity. For thin sheets, proper fixturing and
considerations for welding because of their susceptibil- clamping including copper backing bars with provision
ity to cracking. Cracks can occur in the HAZ on aging or for gas purging should be provided. Typical backing
in service at temperatures above the aging temperature bar designs are shown in Fig. 5.12
due to residual welding stresses or stress induced by pre- Nickel alloy castings should be solution annealed
cipitation. Parts which have been severely formed must before welding. Preheating to 95-220°C, depending on
be annealed before welding, preferably in a controlled- the mass of the casting and light peening of the first
atmosphere furnace to minimise surface oxidation and pass immediately after deposition are recommended to
its subsequent removal. The alloys must be stress-re- reduce cracking at the junction of the weld metal and
lieved (i.e. solution treated) after welding and before pre- the cast base metal.
cipitation-hardening. This would require fairly rapid Magnesium alloys: These alloys are mostly welded by
heating of the weldment to the solution-treating tem- TIG and MIG processes TIG welding is suitable for joints
perature in a furnace, to avoid its prolonged exposure to up to 12.7 m m thick. Some of the alloys have excellent
temperatures within the precipitation-hardening range. weldability: with them 100% joint efficiency can be easily
The alloys may be welded in the aged condition, but achieved. Others have limited weldability and give joint
the weldments must be solution-treated and re-aged if efficiencies below 100% but higher than 60%.
during service they are expected to operate in the In the Mg-Al-Zn alloy group, Al content u p to
precipitation-hardening range. Precipitation-hardenable approximately 10% improves weldability by refining the
alloys are preferably welded by TIG process with grain structure, while Zn content above 1% encourages
controlled heat input. For multipass welds, stringer hot cracks during welding. Mg alloys with more than
beads must be used. On completing a pass, the oxide 1.5% Al are susceptible to stress corrosion, when subjected
film on the surface must be removed by shot-blasting to corrosive media in service. Weldments of these alloys
or grinding. If this is not done, the oxide film thickens must be stress-relieved.
AC with superimposed high frequency current is

used. The arc length should be less than 1mm. The
recommended angles of torch and filler wire are shown
in Fig. 5.13.
In butt joints, stringer beads should be deposited at
a constant speed. For fillet welds or large corner joints,
deposition may be effected with a weaving or rotary
motion. Interruption of the arc must be avoided as far
as possible. If a weld has been stopped, it should be
restarted on the previously deposited weld metal 12 mm
from the end.
Welding parameters for butt joints are given in Table
5.15.
Titanium and its alloys: TIG process is ideally suited
for the welding of these materials. At the same time,
special considerations are necessary, because of their two
peculiar characteristics, namely (a) the slightest
contamination in the joint area gives rise to weld
porosity, (b) when heated to above 540° C, they rapidly
oxidise and the oxides penetrate into the material and
embrittle it.
To obtain a weld with maximum freedom from
Fig. 5.12: Groove designs for backing bars in TIG welding: porosity, joint area of 25 mm width on either side must
(A) Standard groove for use without gas purging; (B) be meticulously cleaned. Grease and oil must be
Square-corner groove employed with gas purging
removed to the last trace with steam cleaning, alkaline
cleaning, vapour degreasing or solvent cleaning. Oxide
film must be removed by pickling. If the material has
Joint fit-up must be accurate and root openings must been exposed previously to temperatures above 540°C,
be avoided. Welding fixtures, closely spaced tack welds a thick scale is formed whose removal calls for a complex
and hold-down bars must be used to prevent distortion treatment. If the e x p o s u r e has been below this
of thin plates. To prevent burn-through, backing plates temperature, a lighter scale is formed on the surface,
of steel, magnesium, aluminium or copper, cut with a which demands a less severe cleaning treatment.
small groove may be used. To prevent embrittlement of the weld and the HAZ,
Plate edges must be perfectly free from oxide and dirt welding was previously carried out in closed chambers,
before welding. This is ensured by cleaning them which were either flow-purged or vacuum-purged with
mechanically with aluminium or stainless steel wool, inert gas. In the former type, a continuous flow of argon
aluminium oxide abrasive cloth or powered wire brushes pushes out the air. It is necessary in this case to provide
having stainless steel bristles. Some fabricators prefer to
clean the parts with special chemicals in tanks, rinse them
with hot water and dry in air.
The need for preheating is dictated more by joint
restraint than by joint thickness. Unrestrained thick joints
may be successfully welded without preheating, while
thin joints under heavy restraint may demand preheating.
High-zinc alloys require preheating to prevent weld
cracking.
Magnesium alloy castings need to be correctly
preheated and postheated. The temper dr the alloy must + .——
be controlled both before welding and after postheat
treatment. A competent authority must be consulted
Fig. 5.13: Electrode and filler wire angles for welding Mg
before undertaking repair welding of a casting.
Table 5.15 Conditions for TIG welding of butt joints in Mg alloys

Base metal Joint No. of. Electrode Current (AC), Argon flow Fillet metal
thickness, design passes diameter, Amps, (b) rate, diameter,
mm (a) mm 1 / min mm
1.2 A 1.6 35 5.5 2.4
1.6 A 2.4 50 5.5 2.4
2.0 A - 2.4 75 55 2.4
2.5 A 2.4 100 5.5 2.4
3.2 A 2.4 125 5.5 3.2
4.8 A 3.2 160 7.0 3.2
6.3 B 2 4.0 175 9.5 3.2
9.5 B 3 4.0 175 9.5 4.0
9.5 C 2 4.8 200 9.5 3.2
12.7 C 2 4.8 250 9.5 3.2
Notes: (a) A = Square butt joint without root gap;
B = 60° bevel, single-V-groove butt joint, 1.5 mm root face and without root gap;
C = 60° bevel, double-V-groove butt joint, 2.4 mm root face and without root gap;
(b) Thorium-containing alloys will require 20% higher current. With He shielding, current will be 20-30 amp less.
a transparent shielding gas cup to protect the molten Primary shielding is meant to protect the face side of a
pool and give good visibility to the welder. Trailing and weld puddle and the area immediately surrounding it.
leading shields (referred to later in out-of-chamber Secondary shielding is required to protect the solidified
welding) are not required. In the vacuum-purged and cooling weld bead and its HAZ, which lie behind
version, the work together with the TIG-torch and the the progressing arc. Back-up shielding is required by
filler wires is first loaded in the chamber, which is then the hot metal underneath the joint, i.e. on the side
evacuated completely and filled with argon. Manual or opposite the welding torch.
mechanised TIG welding is then performed inside the Satisfactory primary shielding is provided by a
chamber. The torch does not require additional inert gas correctly sized torch, having a gas cup of 12.7 to 19 mm
supply. Only water inlet and outlet hoses are required. size, or even larger if it does not diminish accessibility
The chamber is equipped with glove ports on the side, and visibility.
so that the welder can manipulate the torch and filler Secondary shielding is achieved by using a trailing
metal without admitting air to the chamber. Viewing shield such as the one shown in Fig. 5.14, which provides
ports are also provided, so that the welder can observe a diffused argon flow over the hot solidified weld bead.
his operation. Excellent welds are obtained in such Its length is adjusted according to the arc energy and
vacuum chambers, because of the total absence of the arc travel speed.
atmospheric oxygen and nitrogen. The trailing shield is not convenient for manual
From the w e l d e r ' s point of view, welding in a welding and is used more often in mechanised welding.
chamber is far from ideal, because the walls of the For manual welding, the torch is equipped with a shroud
chamber and the location of the glove ports restrict which receives an extra supply of argon. Such a torch is
visibility, movement and accessibility. To improve shown in Fig. 5.15. Secondary shielding can also be
visibility, rigid or collapsible c h a m b e r s m a d e of effected through ports in hold-down bars as shown in
transparent plastic have been used. Rigid plastic Fig. 5.16. The figure also shows how back-up shielding
chambers are flow-purged with argon, in volumes equal is provided by ports in the backing bar.
to about eight times the volume of the chamber. Unalloyed titanium is available in several grades
Collapsible chambers are first collapsed and then flow- with Ti content ranging between 98.5 and 99.5%. All the
purged with argon. They save considerable amounts of grades are easily weldable provided they are in the
gas. annealed condition. Welding them in cold-worked
To give maximum flexibility and ease of welding to condition must be avoided.
the welder, open-air techniques have been successfully Ti alloys are divided into three groups: alpha, alpha-
developed by taking care of the three p h a s e s of beta and beta. Alpha alloys are easily weldable and they
shielding: primary, secondary and back-up shielding. are welded in the annealed condition.
Alpha-beta alloys, especially those which are

--ÔRCH*
strongly beta-stabilised, give low weld ductility caused
by phase transformations in the weld zone or in the
HAZ. Weld ductility is improved to some extent by
GAS INLET using pure Ti or alpha-Ti alloy filler wires, but their use
fa
& i. does not prevent embrittlement in the HAZ. The alpha-
IL beta alloy Ti-6AI-4V and other weakly beta-stabilised
alloys suffer less embrittlement in this respect and are
\ considered to be fairly weldable.
•POROUS DIFFUSION SHEET
Most beta alloys when welded in the annealed or
Fig. 5.14: The trailing shield device for secondary heat treated condition, produce welds which are ductile
shielding which is commonly used in automatic welding but lower in strength. Their weldments are usually used
in as-welded condition. A few other beta alloys, whose
welds embrittle more severely when age hardened are
first welded in the annealed condition, the weld suitably
cold worked, and finally the entire weldment is solution
nc j ^
treated and aged. This procedure ensures adequate weld
I ■i ductility.
COPPER
■fcfl DC rectifiers are preferred to motor-generator sets,
POUTER SHROUD
SHAVINGS \ because they give better current control to ensure even
^^BN penetration. Electrode is m a d e negative to get a
narrower beed with deeper penetration. Since Ti alloys
L - kLfiCTRODE need more thorough shielding as compared to other
ADDITIONAL | B^j , metals, larger nozzle sizes are used. For 0.8 mm diameter
SHIELDING I H^ i
electrode, a 15- mm ID nozzle is taken, and for 1.6 mm
U
G A S ^
diameter electrode, a 19-mm ID nozzle is used. Use of
phenolic or other plastic nozzles should be avoided, as
w NORMAL
*. SHIELDING GAS they are likely to contaminate the weld with carbon.
The torch is provided with a concentric outer shroud.
Fig. 5.15: Sectional view of a torch nozzle equipped with Back-up shielding is also taken care of, as explained
an outer shroud containing copper shavings to provide earlier.
diffused and cooled additional gas shielding in manual Welding procedures are generally similar to those
TIG welding used for austenitic stainless steels. Preheating and
interpass temperatures are not necessary. If cracks occur,
they are to be ascribed to joint contamination. Preheating
HOLD DOWN TIG TORCH does not help to prevent them.
BAR (1 OP 2)x
.•SHIELDING GAS OUTLET Sections u p to 2.5 mm thick can be square butt
welded without filler metal. For thicker sections, the
joint is grooved and welding is done with filler metal.
When tack welding is used, it is better to deposit
long tack welds of full size, so that the main weld never
SHIELDING GAS
PORT(l OF 2) crosses over them. The main weld beads should be
blended into the ends of the tack welds. Heat input
should be minimum. If weld porosity is encountered,
SHIELDING GAS welding speed should be reduced to allow entrapped
PORT (BACKING GAsl / gases to escape from the molten puddle.
BACKING BAR Most of the welded structures in pure Ti and its alloys
BACKING GROOVE
are stress-relieved to prevent weld cracking, to reduce
susceptibility to stress-corrosion cracking and to
Fig. 5.16: Typical set-up for secondary and back-up
shielding in TIG welding of Ti
improve fatigue strength. Competent authority must be
consulted for correct procedures.
DEPOSITION OF ROOT PASS IN SMALL

DIAMETER PIPES FLEXIBLE METAL HOSE
PURGING GAS INLET / 0UTLET HOLES
When root beads of exceptional quality are to be / G A S
made in high-pressure piping of mild steel, low-alloy

steels, high-alloy steels and stainless steels, the TIG
welding process becomes the ideal choice. Outstanding
features of this process are:
1) Excellent weld quality can be obtained in almost
STEEL DISKS RUBBER DIAPHRAGM
all industrial metals.
2) Practically no post-weld cleaning required. Fig. 5.17: Arrangement for purging
3) Arc and molten pool are clearly visible to the
welder.
4) Filler metal is not transferred across the arc, hence
no spatter. V/////JJJ///77777\.
5) Welding is possible in all positions.
6) N o slag is formed; hence no danger of slag
entrapment. f))r})x)))))))i\i))))))))mf)77A
7) Root welds in pipes have smooth internal surface.
GAS
This significantly reduces turbulence in any flowing
substance inside the piping system. In many cases such Fig. 5.18: Purging with plastic gas bags
as nuclear power piping, this is a very important factor.
8) Because of the inert argon shield, the filler rod does of the joint is left open to allow the air to escape. When
not undergo any chemical change. By choosing the welding, the sealing tape around the pipe is removed
correct filler rod, it is possible to obtain a weld having in sections just ahead of the weld. Additional argon is
the same chemical, metallurgical and physical properties blown into the pipe to make up for the gas, which is
as the pipe material. lost through this opening.
In critical applications, and where highly alloyed Typical joint preparation and fit-up is shown m Fig.
steels such as stainless steels are involved, excessive 5.19. When all the aforesaid preparations are made, four
oxidation, resulting in rough underbead appearance, is evenly spaced tack welds are made around the joint.
avoided by displacing the air underneath the joint area Since these welds will become a part of the root bead,
or root with argon. The procedure is referred to as they must be made with the same care and skill as the
purging. remainder of the root bead.
Purging is achieved on short pipe lengths or tubing To weld, for example, 8 in Schedule 60 mild steel
by sealing off or capping the open ends and introducing pipe nipples, a 2.4 m m diameter thoriated tungsten
a liberal amount of argon through the cap at a slightly electrode and a 3.15 m m diameter mild steel filler rod
higher pressure than atmospheric. On tubing of large are used. The electrode should be ground to a pencil
lengths, special sealing arrangements can be employed, point with a fine grinding wheel. The electrode must be
which produce an enclosed area around the joint. Such positioned in the torch in such a way that it extends the
arrangements minimise the consumption of argon. A correct distance beyond the end of the nozzle. This is
typical set-up is shown in Fig. 5.17. Infield conditions done by placing the torch in an upright position with
where the pipe diameters and lengths are commonly
large, the joints can be sealed by inflating two plastic
balloons in the pipe as shown in Fig. 5.18. They must be
positioned far enough from the joint to prevent them
from bursting due to welding heal. Their location must
s \ 75V
be marked on the outside of the pipe, so that when the

root bead is completed, just enough heat can be applied
with a gas torch at these markings to burst the balloons.
1 6 TO 2 4 M M
In both the above arrangements, the pipe joint itself 6 MM
is taped to prevent argon, which is blown into the pipe
Fig. 5.19: Joint fit-up for root TIG welding
at a very low pressure, from escaping. A small portion
the nozzle resting on the level of the weld joint, as shown

in Fig. 5.20. While this is being done, the welding current fSA
should, of course, be shut off. The electrode is then
ir
adjusted, so that it is positioned with the end almost
flush with the bottom of the joint or the inside surfaces
of the pipe.
DC with electrode negative should be used. The
current range is 150-225 amps. The argon supply rate 15*15°
should be 20 cubic feet per hour (101/min).
Fig. 5.22: Torch oscillating procedure
When the pipe is in 5G position (i.e. horizontally
fixed, involving all-position welding), tack welds are
best placed at 8:30, 4:30, 1:30 and 11:30 positions. The in the 5G position, there is a tendency for the puddle to
regular bead is started at 6 o'clock position and carried drip, causing excessive penetration. When this occurs,
uphill to 12 o'clock position. Then the other half is corrective action should be taken by slightly increasing
welded, starting again at the 6 o'clock position and the speed of travel and the speed of the oscillating
reaching the top. movement. If possible, more filler metal than usual
The relative positions of the torch and the filler wire should be added to the puddle, which tends to chill the
are shown in Fig. 5.21. After the puddle is established puddle and to reduce its size.
and the bead is commenced, the welder must oscillate The welder must always be careful to prevent the
the welding torch from side to side as shown in Fig. electrode from making contact with the sides of the joint
5.22, allowing the arc to travel slightly beyond the edges and the molten puddle. This would not only
of the pipe. contaminate the end of the electrode but also cause the
While welding, the welder must watch the puddle arc to become unsteady. If this occurs, the weld should
at all times to see that it is built up to a uniform height be stopped and the tip of the electrode reground before
and forms a smooth bead with only traces of a ripple. restarting the weld.
As with most welding processes, when welding a pipe One problem that is occasionally encountered is that
the nozzle tends to stick or grab as it is lightly dragged
over the bevel during welding. The end of the nozzle
GLASS OR CERAMIC
gets very hot, a factor which promotes its sticking to
the heated bevel. As shown in Fig. 5.23A, the edge of
the nozzle is square, and sticking can be avoided by
grinding a radius on the edge of the nozzle as shown in
Fig. 5.23B.
Pulsed TIG Welding

This is a modified version of TIG welding, in which
Fig. 5.20: Adjusting the tungsten electrode the welding current (DC) is fed intermittently in the
form of pulses. The pulsed current alternates between
a low or background level and a high or peak level. The
duration and amplitude of both peak and background
GRIND RADIUS
IRECTION OF WELDING
(A) (B)
Fig. 5.21: Correct angles of torch and filler rod Fig. 5 23: A Standard nozzle. B: Nozzle edge rounded
currents can be varied independently to suit the job. The pipes or for depositing the root pass in thick-walled
melting takes place during the peak current period, and pipes, where the pipes have a fixed horizontal position,
the weld pool solidifies between pulses as the heat is because the same average current and voltage can be
dissipated in the job during the background current used at all points around the joint, except at the start
period. This current pulsing leads to intermittent and the finish. With the conventional (non-pulsed) TIG
melting along the joint seam, giving a series of discrete process, the parameters would have to be suitably
melt spots which overlap each other. altered as the arc moves from the flat through the vertical
The most significant feature of this process is the to the overhead position. For the same reason, the pulsed
apparent tolerance to external variables such as joint process in automated form is used for the orbital
geometry, clamping, fit-up, dissimilar thicknesses or welding of tubes to tubes, and tubes to plates in the
other factors causing variation in thermal heat sink. This fabrication of heat exchangers.
tolerance is derived from the fact that the weld pool is The a d v a n t a g e o u s features of the process are
allowed to solidify during background time, and owing summarised below:
to this the effect of heat build-up is largely overcome. • Tolerates variations in joint fit-up or clamping.
Conversely, in conventional TIG welding using fixed • Permits welding of sheets down to 1 mm with
or steady current, a delicate balance has to be maintained ease, which is normally difficult by standard TIG
by the welder, between heat input from the arc and the process.
heat sink due to the job. This has to be done to avoid
• Minimises distortions owing to controlled heat
change in the degree of melting, resulting in uneven
input.
penetration and even burn-through in some cases. Such
a manual manipulation, by either change in torch angle • Makes positional welding easy.
or w e a v i n g is difficult to achieve, especially in • Requires less operator skill as compared to the
mechanised welding and therefore leads to significant normal TIG process.
variations in joint quality. This problem is largely • Shows marked improvement in weld quality.
eliminated by the use of pulsed current technique.
• Is suited to mechanisation/automation.
A typical pulsed current waveform along with the
argon gas flow sequence is shown in Fig. 5.24. • Is ideal for critical applications such as root passes
Pulsed current with rapid current rise and current of pipes, joining dissimilar thicknesses, etc.
decay, together with a high pulse-repetition rate is used • The strong electromagnetic forces developed by
in the joining of precision parts. Pulsed current with the high pulse of current agitate the weld puddle
slower rates of current rise and fall and slower current and thereby eliminate weld porosity and ensure
pulse rates is used in mechanised welding. thorough root fusion.
The process is ideally suited for the automatic all- A typical modern pulsed TIG equipment, shown in
position welding of the circular joints of thin-walled Fig. 5.25 is a solid-state outfit with fully transistorised
current control. With the help of the electronic controller
fitted therein, it is possible to pre-set different welding
INITIAL CURRENT parameters like pre-flow time, up-slope, down-slope,
.PEAK CURRENT LEVEL
UP-SLOPE peal background current, peak time, background time,
.BACKGROUND CURRENT LEVEL
final current level and post-flow time as required.
CRATER FILLER
In this outfit, excellent waveform is achieved through
DOWN-SLOPE
solid-state transistorised circuitry. The arc is initiated
MIN CURRENT by a high frequency system with booster voltage
LEVEL circuitry, which automatically gets cut off after the arc
is formed. A special circuitry automatically cuts off the
complete system, including argon flow if the welding
operation does not commence within 10 sec of pressing
GAS^
PREFL0W the start switch. This prevents wastage of costly argon
PEAK TIME BACKGROUND
TIME gas on the shop floor.
The power source (DC) has constant current (i.e.
Fig. 5.24: Pulsed current waveform drooping) characteristics in the operating region to
ensure uniform penetration. Control panel is of modular
Important specifications of the outfit are given below:

Peak current range (DC) Amps. 10 -200
Background current range (DC) Amps. 10-200
Time duration of peak current Sec 0.1-2.5
Time duration of background current Sec 0.1-2.5
Starting up-slope time (up to peak current setting) Sec 1-5
Down-slope time (up to min. current level) Sec 3-12
Final current level Amps. 10-200
Gas pre-flow time Sec 0.1-5
Gas post-flow time Sec 0.1-10
Open circuit voltage V 55
Max. cont. welding current at 60% duty cycle Amps. 200
Max. auto, welding current at 100% duty cycle Amps. 145
Rating at 60% duty cycle KVA 13
Rating at 100% duty cycle KVA 10
Semi-automatic TIG Welding

In this process, the filler wire in continuous length is
fed mechanically and at controlled speed into the arc
area, while the welder advances the torch manually
along the joint. A motor-driven wire-feeder feeds the
wire through a flexible hose. The attachment that brings
the filler wire into the arc area is mounted on the
electrode holder, thus enabling the wire to enter the arc
from any direction. The semi-automatic TIG demands
less skill from the welder and makes his task easier.
Mechanised or Automatic TIG

The various devices and controls necessary for
mechanised TIG welding are:
a) arc voltage control (for steady arc length)
b) electrode positioning devices
c) seam tracking devices
d) oscillating devices
e) filler wire feeders
f) traversing mechanisms.
a) ARC VOLTAGE CONTROL

The commonly used arrangement is shown in Fig.
5.26. A reversible gear motor keeps the voltage constant
Fig. 5.25: Pulsed TIG equipment by raising or lowering the electrode to maintain the
desired arc length. The arc voltage measured in this
system consists of two parts: the voltage drop across
construction for easy accessibility and maintenance. Self- the arc and the drop in the electrode. The voltage drop
sufficient water recirculating system is provided for in the electrode begins to rise as soon as the arc is struck
cooling of power transistors. Safeguards are also due to resistance heating. As its value increases, the arc
incorporated against inadequate supply of cooling water voltage control shortens the arc length to maintain the
or failure of recirculating system. Equipment trolley has total arc voltage at a constant value. This is visually
space for mounting the gas cylinder to ensure complete evident from the characteristic diving of the tungsten
portability of the equipment. electrode.
deposit the weld on a sensor-identified path (whether

KKKMCES ABC
this weld is good enough is not the province of the
f
COMPARISON AMPLIFIER MOTOR
VOLTAGE
CIRCUIT CIRCUIT CIRCUIT
sensor). Imaging is in 'man view' mode, i.e. a man would
_J
(ARC LENGTH AQJUSTER)
— ERROR VOLTAGE
easily understand the TV screen display, and the output
is not in machine language.
WELOlNfiilEAO
Laser-only techniques commonly employ low
SIGNAL
f REVERSIBLE
1
as
CIRCUIT GEAR MOTOR powered lasers (0.5 mW) in a rastering range-finding
I 1 mode or a shadow generating mode. Here the process
ELECTRQPE
WORK
£ WELDING
CURRENT
SUPPLY
involves measuring, welding, remeasuring, recalibrat-
ing the weld, etc. Limitations to this technique arise from:
a) The working range of the range finder—a few mm;
Fig. 5.26: Automatic gas-tungsten arc voltage control b) The a m o u n t of data processed to provide a
system computed contour.
Currently the generation of weld contours and the
An alternative method of arc control consists of computer reselection of welding parameters is possible,
maintaining the electrode at a fixed pre-set distance from although it appears to be too slow for production lines
the work and adjusting the current, as necessary, to keep or flexible manufacturing systems.
a constant potential or a constant arc power. These Laser shadow techniques apparently promise much,
systems vary from simple m a n u a l adjustment to but can frequently be in serious error. If, for instance,
elaborate magnetic and eddy current devices. the plate surface quality changes (say, from rolled to
machine) then its reflectivity also changes and thus the
B) ELECTRODE POSITIONING DEVICES intensity of reflected laser radiation varies. Despite its
They are referred to later while describing an dependence upon surface reflectivity, the system is being
automatic pulsed TIG tube-plate welder. investigated widely.
C) SEAM TRACKING DEVICES D) OSCILLATING DEVICES

Seam tracking, w h i c h involves sophisticated These operate on mechanical or electromagnetic
technology is gaining prominence with the increasing principles to impart oscillatory motion to the tungsten
use of robotic arc welding. Seam tracking devices use electrode or the arc. The motions can be uniform,
sensors which range from simple to extremely complex. harmonic or with dwell as required.
Simple systems use a mechanical finger or electrical
capacitance measurements. These have a limited scope, E) FILLER WIRE FEEDERS
because they can track a few joint geometries and suffer Wire feeders are referred to as cold or hot, depending
from inertia/speed problems. on whether the filler wire they feed mechanically into
Commercially available seam trackers, frequently the arc is at room temperature or preheated by electric
built into robots, are of the nodding head type, using resistance method.
changes in arc voltage to indicate whether the seam is Cold wire feeder: The main components are wire drive
being followed. This system requires a very accurate m e c h a n i s m , speed controller a n d a wire g u i d e
and stable welding power source, which though very attachment, to introduce the wire into the leading edge
expensive, can be justifiably used in robotic arc welding. of weld puddle.
Such nodding heads are well suited for fillets or deep- The wire drive mechanism consists of an electric
V preparations, but they are not fully effective where motor and a set of gears which power the drive rolls
the sidewalls are not steep or high. But it must be that push the wire. The wire feed is controlled by a
accepted that arc welding even with solid-state power constant speed governor, which can be a mechanical or
sources is inherently variable and its parameters are not an electrical device. A flexible hose guides the wire from
desirable as control signals. the drive mechanism to the wire guide attachment. The
More sophisticated and specialised seam trackers wire guide attachment is attached to the torch. It
i n c l u d e vision sensors b a s e d on w h i t e light TV maintains the position at which the wire enters the weld
monitoring or laser-illuminated techniques. pool and the angle of approach with respect to the
TV sensors u s e b o t h n o r m a l light a n d laser electrode, work surface and the joint. Wire sizes in the
illumination. The sensor seeks the joint and aims to range of 0.8-2.4 m m are fed at a constant rate.
Hot wire feeders: As the name suggests, these feeders This technique also eliminates weld porosity in many
feed electric resistance heated continuous wire into the instances by driving off most of the volatile contamina-
weld pool. The feeding may be done through a torch tion and entrapped hydrogen from the filler metal sur-
provided with inert gas shielding to protect the hot wire face during 1 R (electric resistance) heating.
from oxidation. For heating, AC from a constant The hot wire TIG process has been widely applied
potential power source is used as shown in Fig. 5.27. in the U.S.A. for the fabrication of pipe and piping
The heating starts only after the wire makes contact with systems in a wide range of materials including carbon
the weld pool and completes the electric circuit. The and low-alloy steels, stainless steels, and alloys of copper
current is so regulated that the filler metal reaches its and nickel. Figure 5.28 compares the root passes
melting point as it approaches the weld. Thus the weld produced by the hot wire TIG and the conventional
metal deposition is independent of the arc heat. The fused insert manual TIG process in a U-groove joint. At
figure also shows that the hot wire is fed into the trailing (A), a fusible insert is placed in the root gap, which is
edge of the weld pool, a little behind the tungsten fused by the welder with a TIG torch without addition
electrode at an angle of about 50° with the electrode. of filler wire. The root pass so produced is only 1.78
It has been stated that AC is used for heating the mm thick and hence the subsequent passes have to be
filler wire. If DC were to be used for this purpose, the deposited at low currents to prevent burn-through. The
cost of power source would be higher, and further, the TIG hot wire joint at B has 1.6 mm root face and no root
interaction of the magnetic fields developed by the arc gap. In this case, using 230 Amps, at 11 V, and using a
and the current in the filler wire would cause the arc to welding speed of 250 mm/min a root pass can be
be blown either backward or forward, which is not deposited which is 3.55 mm thick, enabling subsequent
desirable. With AC in the wire, the magnetic effects cause passes to be made at higher currents and higher
the arc to oscillate 30° in the longitudinal direction, when deposition rates. At (A) the joint area is 1.135 sq. cm,
the heating current is not more than 60% of the arc which is reduced to 0.97 sq. cm at (B) by withdrawing
current. This amplitude of arc oscillation is well-suited the insert. The production rate of (B) joints is nearly four
for practical applications, and in order not to exceed it, times the production rate of (A) joints.
the filler wire diameter is limited to 1.2 mm. If the wire A typical welding procedure for hot wire TIG
size is larger, the heating current will exceed 60% of the welding of 6 in. Schedule 80 carbon and stainless steel
arc current and cause the arc to oscillate too widely. pipe is outlined in Table 5.16.
The hot wire technique increases the weld deposition
rate of the TIG process. For example, at the arc energy
of 4 kW (330 amps at 12 V) the maximum cold wire
deposition rate is 1.36 kg/hour. With hot wire, up to 3.6
kg/hour can be deposited and this rate can be further
increased to about 4.2 kg/hour by using oscillation.
11 MM
HIGH SPEED FILLER

WIRE FEEDER X R=*4MM
FUSIBLE INSERT
40
° J
CONTACT TUBE*
AC DC
HOT POWER
WIRE TRAVEL (TIG)
POWER
WORKPIECE ' Fig. 5.28: Comparative set-up for root pass in hot wire
Fig. 5.27: TIG hot wire system TIG and fused insert TIG welding
Table 5.16: Procedure for hot wire TIG welding of pipes

Pass Welding Voltage, Travel speed Hot wire speed Deposition
current, Amps. V cm/min cm/min rate, kg/hour
1 240 11 23 250 1.36
2 340 13 30.5 375 1.8
3 440 13 30.5 750 3.6
4 450 13 30.5 750-1,250 3.6-6.36
Table 5.17: Comparative welding speeds of hot wire and cold wire TIG welding
Sheet Cold wire Hot wire
thickness, mm Amps. speed cm/min Amps. Speed cm/min
1.6 160 62.5 300 87.5
2.4 220 37.5 360 75
3.2 250 30 410 62.5
4.8 350 17.5 450 30
6.3 400 10 480 17.5
Table 5.17 shows that stainless steel sheets can be

welded at appreciably faster rates by the hot wire TIG VERTICAL ADJUSTMENT WIRE FEEDER
as compared to the cold wire process.
F) TRAVERSING MECHANISMS
They are designed to move the electrode torch or
the welding head with respect to the joint or vice versa.
They maintain the position of the electrode with respect
to the joint within close limits and travel smoothly at a
set speed without much vibration. They are described
in detail in Chapter 20.
For welding stationary joints, the welding head or
the electrode torch can be mounted on a motorised
trolley, which runs on rails or directly on the workpiece.
For long straight joints, the welding heads and allied
equipment can be transported by means of a side beam
carriage, which is supported on the vertical face of a
flat track. The track can be attached to the side of a beam,
to construction brackets on a wall or to columns erected
on the floor. The carriage is fitted with a mechanical
clutch, so that the drive motor can be disengaged while Fig. 5.29: Automatic cold wire TIG equipment
the welding head is being positioned. For longitudinal
and circumferential welding of cylindrical sections,
combinations of welding booms and rotators can be built-in with a fine vertical adjustment for exact
used as in the case of the automatic submerged-arc positioning of the torch. A control box which houses all
welding process. the control elements for carriage and wire-feed is
separately located. All these items make up the welding
Automatic TIG Equipment head, which is suitable for use with any appropriate
An example of the a u t o m a t i c cold wire TIG standard TIG power source and torch. Filler wire in the
equipment is shown in Fig. 5.29. It consists of the form of spool is fed into the arc at a rate which is
carriage, wire-feed mechanism, spool holder to carry infinitely adjustable in the range of 0.1 to 12 metres/
filler wire spool and an adjustable torch-holding device min and the particular wire-feed setting is maintained
constant by the wire-feed m e c h a n i s m t h r o u g h a instead of the conventional preparations used for the
thyristorised electronic control. manual metal-arc or submerged-arc processes. Atypical
The equipment shown in Fig. 5.30 is specially narrow gap butt joint for 63.5 mm wall thickness along
developed to perform DC TIG welding on longitudinal with the conventional preparations is shown in Fig. 5.31.
and circumferential joints in stainless steel box structures The groove volume of the narrow gap shown at (C) is
and cylindrical shells made from sheets of up to 6 mm only 25% of the other two. A typical machine marketed
thickness. In case of circumferential joints a job rotator in the U.K., shown in Fig. 5.32, incorporates a patented
is necessary. Longitudinal joints of desired length can torch design and a telescopic shielding arrangement
be welded using carriage travel track of suitable length. which fits around the gas channel and shields the arc
Maximum length for longitudinal welding is 10 metres. effectively at weld preparation depths of 100 mm. Its
To ensure 100% radiographic quality joint, a gas purging main component is the carriage which orbits the pipe
arrangement and a backing bar are provided. accurately even under high preheat temperatures up to
300°C, which are used for the welding of low-alloy
Orbital Narrow Gap TIG Welding creep-resisting materials (such as Vi Cr-Mo-V steel).
Automatic TIG welding has been successfully The system has a guide ring of stainless steel with
applied to high pressure pipework for the modern steam two V grooves around the periphery to accept the
turbine generator system. To compensate for the low carriage wheels. It is located on the outside of the pipe
deposition rate of the process, a narrow gap joint is used by adjusting screws and will split into two halves for
Fig. 5.30: Automatic equipment for S.S. box structures and cylindrical shells
Fig. 5.31: Typical joint preparations for pipe butt joints of t= 63.5 mm (a) MMAW using a TIG root run; (b) SAW with
TIG root run and 13 mm MMA reinforcement; (c) narrow gap weld
Fig. 5.32: NGW head in use
removal after welding. The welding head and drive this defect is overcome by using an argon: 1% hydrogen
carriages are separate and are located diametrically mixture.
opposite each other by quickly detachable link arms.
This separation reduces the out-of-balance loading on Automatic Pulsed-TIG Tube-To-plate Welder
the drive motor and enables the complete assembly to In tube-to-plate welding during the fabrication of
be rotated easily around the pipe at all times. There are boilers and heat exchangers, the difference in thermal
two connecting arms between the carriage; one is spring mass between the gauge of tube and the heavy section
loaded with a release mechanism to allow for unloading plate can create problems. Excessive penetration (burn-
and loading on to the guide ring. through, of the tube) and insufficient penetration of the
The drive, carriage locates and runs on the guide ring plate must be avoided. Pulsed TIG is recommended for
and houses a high-torque drive system to all four wheels fully automatic tube-plate welding. The tube-plate joints
of the carriage a n d incorporates a quick-release and welds are generally as shown in Fig. 5.33.
m e c h a n i s m for fast m a n u a l r e w i n d of the cable A typical welding head shown in Fig. 5.34 is designed
assemblies. The torch assembly houses a fast response to weld 75 mm diameter tube of 2 mm wall thickness to
automatic voltage control system together with a wire- a 5 mm thick plate, the pitch between tube centres being
feed unit. Automatic travel speed is 75 m / m i n , running 95 mm. An interchangeable expanding mandrel locks
on precision shafts and driven by a ball screw. The wire- the head into the tube and accommodates a range of
feed guide also fits down to the bottom of the groove. tube sizes. The rotating parts move over the surface of
The torch carriage locates and runs on the guide ring the plate supported on three wheels pitched at 120°,
and carries the torch assembly on sides, allowing cross- e n s u r i n g that torch a n d wire-feed assembly are
seam adjustments which can be performed during the maintained at a pre-set height above the joint.
welding cycle. The complete unit weighs approximately Both the weld torch and wire-feed nozzle have
23 kg a n d h a s a radial a n d axial clearance of micrometer adjustment in three motions (see Fig. 5.35)
approximately 250 mm. It will cover a pipe size range and critical angles for arc and wire can then be set very
of 150-600 mm OD, but in principle the technique and accurately. A precision quick-release allows the torch to
equipment could be extended to cover the welding of be r e m o v e d for t u n g s t e n r e p l a c e m e n t w i t h o u t
thick walled pressure vessels of much greater diameter. interfering w i t h the torch setting. Electronically
Experience has shown that use of pure argon as a controlled 24 V DC motors in the head are geared to
shielding gas causes lack of side-wall fusion, and that give a maximum speed of rotation of 0.5 rev/min and a
.COVER P A S S / - B ! F ? R E . < A F T E R W E L D , N G
TUBE WELDED'
DED\ BEFORE
" t r i r n t vWELDING
OP
„ • „ r WELDED PLUG (DRILLED) SUBSEQUENTLY
BEFORE ÊLDING f r BEFORE^ '
WELDING \ / W £ L D METAL
\i
DRILL t / TUBE SHEET
BEFORE EXTRUSION
D it
AFTER EXTHUSION
Fig. 5.33: Typical joints for tube-to-plate welds
Fig.5.35: Adjustment of torch angles
for weld head rotation and the wire-feed motors are

interlocked with the power source. Output transistors
switch at 15 kHz. The speed error is amplified and
compared with a 15 kHz triangular wave generator to
produce a pulse-width modulated waveform to the
drive motor. A 20:1 speed range is possible and the
maximum motor current is electronically limited to
guard against possible stalling.
Operating sequence is fully automatic and the
operator merely observes the weld pool or the inside
wall of the tube. The 'Start' push-button initiates a pre-
purge gas period followed by HF ignition of the arc.
The head drive then rotates the torch as the weld current
builds up to the set value. When this is reached, filler
^ 4. ^ , ^p wire then feeds into the weld pool.
After a complete revolution plus the pre-set overlap
Fig. 5.34: Pulsed TIG welding head angle the wire feed automatically stops, weld current
decays to zero, and the arc extinguishes. A post-purge
maximum wire speed of 0.3 m/sec. An overlap angle period is followed by high speed reversal of the head to
for the weld over a range 0-25° may be pre-set. the start position, ready for the next weld. Operating
The arc, struck by a high frequency circuit, is parameters (current levels, pulse times, weld and wire
maintained by a pilot arc supply and pulsed weld speeds, etc.) are set on the power source panel. Remote
current is then produced from a thyristor bridge-rectifier. control facilities to stop and start the process and to trim
Controls are provided for peak and background current the weld current are provided for close supervision.
and their times of operation. Weld current is controlled
to within 1% of the set value despite variations in arc Plasma Arc Welding (PAW)
length. Build-up and decay rates are also controlled to The term plasma means a gas that has been heated
eliminate thermal shock and craters. Power amplifiers to a sufficiently high temperature, so that it is
transformed into an ionised condition (i.e. into an equal the form of a high velocity, intensely hot and collimated
number of electrons and ions) and is able to conduct an plasma jet. This plasma jet is termed as a plasma arc.
electric current. Plasma is regarded as the fourth state Figure 5.36 compares the common tungsten arc and the
of matter following the three commonly known states: plasma arc welding processes, while Fig. 5.37 compares
solid, liquid and gaseous. The electric arc with its intense the shapes and isothermal maps of the argon-tungsten
heat, generates a considerable amount of plasma. In fact, arc and the constricted plasma arc.
the electric arc consists of a relatively high current PAW process, as compared to TIG, has the following
discharge sustained through a column of plasma. advantages: (1) Greater concentration of energy, (2)
In the normal open arc welding processes including improved arc stability especially at low currents, (3)
the TIG process, the plasma jet has a low velocity and is higher heat content, (4) higher velocity of plasma, (5)
diffused over a wide area, with the result that the arc less sensitivity to variations of arc length, (6) no tungsten
loses much of its power and its zonal temperatures are contamination, (7) less welder skill in manual welding,
lowered. Plasma arc processes (these include plasma (8) solid backing a v o i d e d by a d o p t i n g keyhole
arc welding, plasma arc cutting, plasma arc surfacing technique.
and plasma arc spraying) use a specially designed Drawbacks of PAW process are: (1) Equipment costs
plasma arc torch, which is a modified version of the TIG more, (2) short life of constricting nozzle, (3) welder
torch, in which the tungsten-arc is directed through a needs deeper understanding of the process, (4) increased
nozzle through which an inert gas (argon or argon- consumption of inert gas.
helium mixture), called the carrier gas, flows. The
nozzle, located between the tungsten electrode and the Plasma Arc Operation
base metal, prevents the divergence of the arc and To understand the operation of the process, one must
constricts it into a small cross-section. This results in a know the terminology of the plasma arc torch as shown
substantial increase in the resistive heating of the arc, in Fig. 5.38 and the significance of each term.
so that the arc temperatures and the arc voltage are • Electrode: The tungsten electrode is recessed inside
raised. From the constricting nozzle, the arc emerges in the constricting nozzle, so that it cannot touch the
SHIELDING- © ELECTRODE
ORIFICE GAS
GAS
SHIELDING GAS
SHIELDINi( f G A S "SHIELDING GAS

NOZZLE NOZZLE
ARC PLASMA ARC PL

PLASMA
" " ^ ^ ^ V / v WORK p ARC
TIG PAW
Fig. 5.36: Comparison of (a) TIG and (b) PAW processes
Cathode *-)
Nonconstrieted arc £C^ Constricted arc

40ft»/h argon §§S (3/1 in)diaorif ice
4.8mm(3/16
200A 40ft»/n argon
15V 220A
Temperature K
} 10 000-14 000
|} 14 000-18 000
3j£l8 000-24 000
tt 24 000 up
Fig. 5.37: Comparison of shapes and isothermal maps of TIG and PAW arcs
ELECTRODE ORIFICE GAS
SHIELDING GAS' CONSTRICTING

OZZLE
OUTER GAS NOZZLE

PLENUM
CHAMBER
THROAT LENGTH
ELECTRODE
SETBACK
ORIFICE DIA
TORCH STANDOFF
Fig. 5.38: Plasma arc torch terminology
workpiece. Hence tungsten contamination, which surround the electrode and to enter the arc through
commonly occurs in the TIG process, does not occur in the plenum chamber. It becomes ionised in the arc to
this process. The electrode has negative polarity. form the plasma, which expands due to the heat of
Positive polarity causes excessive deterioration of the the arc and issues from the orifice or orifices of the
electrode. constricting nozzle at an accelerated speed as the
• Electrode setback: This is the distance between the plasma jet. The jet is so powerful that it is capable of
electrode tip a n d the outer b o t t o m e d g e of the cutting a metal. The higher the orifice gas flow rate,
constricting nozzle. the more penetrating is the arc. For welding, the jet
• Orifice gas: This gas, referred to earlier as carrier is controlled by holding the orifice gas flow rates within
gas, is an inert gas which is fed through the torch to the range of 1.5-15 1/min. The orifice gas alone is not
adequate to shield the weld pool from the atmosphere. suitably positioned to give an elongated arc in line with
Additional shielding gas is therefore provided through the joint, which for the keyhole technique described
an outer gas nozzle. later, gives 30-50% higher welding speed than that
• Plenum chamber: This has been referred to under obtained with single port nozzle.
'orifice gas'. It is the space between the inside wall of The nozzle is usually made of copper and is water-
the constricting nozzle and the electrode. cooled. The fact that copper with a melting point of
• Shielding gas: As mentioned under 'orifice gas', 1,083° C is able to withstand the plasma temperatures
the shielding gas is provided additionally through the of 24,000° C and above is explained thus: during
outer gas nozzle, in the same manner as in the TIG- welding, the arc column inside the nozzle is surrounded
process to protect the area of the workpiece, over which by an annular layer of cooler gas with a very steep
the arc plasma impinges. Its composition is the same as thermal gradient. This layer of cooler gas provides
the orifice gas, i.e. either argon or argon-helium mixture. thermal and electrical insulation to the inside surface
• Constricting nozzle: The arc constricting nozzle is a of the nozzle. If the smooth flow of this gas layer is
critical component of the torch. Its contour partly disturbed either by insufficient orifice gas flow or
determines the thermal energy of the arc. Its main xcessive arc current, the nozzle may suffer damage due
dimensions are 'orifice diameter' and 'throat length', to double arcing, i.e. between the electrode and the
which are shown in the above figure. The nozzle may nozzle and between the nozzle and the work.
have single or multiple ports. The most commonly used • Orifice diameter: The nozzle orifice size is a criti-
types are shown in Fig. 5.39. The central orifice may be cal dimension as stated earlier. It is rated for a
cylindrical or have a converging or diverging taper. given current at a given gas flow rate. If the gas
In the case of the single port nozzle shown at (a), the rate is reduced, the current rating at the orifice
entire orifice gas and the arc pass through the single must be proportionately decreased.
orifice. With the nozzle shown at (b), the arc and a part • Torch standoff: It denotes the distance between the
of the orifice gas pass through the central orifice, while outer edge of the constricting nozzle and the
the rest of the orifice gas flows through the two small workpiece. The torch standoff distance is not
ports. important, because changes in arc length have
While the nozzle type (a) is most widely used with negligible effect on the process results. Much
DC electrode negative, the type (b) can be used with longer torch standoffs are possible with the
advantage on certain types of joints. The type (b) can be plasma arc than with the TIG process.
CONSTRICTING NOZZLE
LECTRODE
AUXJLIAI
GAS
PASSAGES
RIFICE-
(A) SINGLE PORT (B) MULTIPLE PORT
Fig. 5.39 : Single and multiple port constricting nozzles

Plasma Arc Properties transfer rate. The plasma arc jet being cylindrical,
In the conventional TIG process, the arc plasma changes in arc length have negligible effect on the area
spreads over a large area of the workpiece and the arc of heat input and the arc intensity.
is easily deflected by weak magnetic fields. The plasma Two types of arc are used for welding as shown in
arc, on the contrary, is stiff and unidirectional and very Fig. 5.40. The transferred arc (A) is formed between the
little affected by magnetic fields. The constriction of the electrode and the workpiece. The non-transferred type
arc in the latter, results in higher plasma temperatures (B) is formed between the electrode and the nozzle.
and arc power. For example at 200 Amps., electrode In the transferred arc, the workpiece is in the circuit,
negative, and an argon flow rate of 19 1/min, a which means that it has to be electrically conducting. The
constricted arc passing through a 4.8 mm diameter transferred arc transfers greater energy to the work than
orifice shows a 100% increase in arc power (mainly the non-transferred type, and produces heat both from
because the arc voltage is doubled) and a 30% increase the anode spot on the workpiece and plasma stream. For
in temperature as compared to an open TIG arc. welding, this type of arc is generally used.
The energy of a plasma arc can vary over a wide The non-transferred arc is contained within the torch
range depending on (a) plasma current, (b) size and and is independent of the workpiece. The hot plasma
shape of nozzle orifice, (c) type of orifice gas, and (d) issuing from the nozzle orifice serves as a useful source
flow rate of orifice gas. Hence the process can be applied of heat. Non-transferred arcs are used for joining and
with equal efficiency for cutting, welding, surfacing and cutting non-conductive materials or for applications
metal spraying by suitably controlling the four factors. demanding low arc energy.
For cutting, for example, a high arc current, a small-
sized orifice and high orifice gas flow rate are used to Equipment
obtain a highly concentrated energy and a high jet The basic equipment for manual PAW is the same as
velocity. For welding, a low plasma jet velocity is for TIG welding, and is shown in Fig. 5.41. It consists of
obtained by using a large orifice, low gas flow rate and a DC power source, control console, water-cooled
low transferred arc currents. welding torch, cylinders and hoses for orifice gas and
The normal TIG arc has a conical shape and hence shielding gas, coolant circulator with pipings leading
the area of heat input to the workpiece varies as the to and back from the torch, and accessories such as on-
square of the arc length. A slight variation in arc length, off switch, gas flowmeters and remote current control.
therefore, causes an appreciable change in unit area heat More commonly, the process is used in mechanised
CONSTRICTING NOZZLE
RIFICE GAS
SHIELDING.
'GAS
TRANSFERRED '"' NON-TRANSFERRED
Fig. 5.40: Transferred and non-transferred plasma arc modes

are a high frequency power generator, an arc voltage

START-STOP
WELDING SWITCH control and filler wire-feed system. The wire-feed
TORCH system can be cold or hot as used in mechanised TIG
welding. For mechanised welding the maximum currnt
used is approximately 400 Amps.
Power sources are the same as those used for DC
TIG welding. They are usually rectifiers with current
ratings from 100 to 400 Amps, at 60 to 100% duty cycle
1 WORK h with OCV in the range of 65-80 V.
REMOTE A typical control system for mechanised high-current
CURRENT plasma arc welding consists of a main console control,
CONTROL
a junction box for gas and water hoses, an operator's
Fig. 5.41 : Typical equipment for plasma arc welding pendant control box, and a high-frequency generator.
The main control unit sequences the power source; high-
form, so that high currents can be used to achieve high frequency generator, orifice, shielding, and backing
welding speeds and increased penetration. In fact, gases; torch travel; auxiliary wire feeding; and cooling
manual PAW is restricted to currents less than 100 amps water for the torch. Travel speed and wire-feed rates
due to the intense heat of the arc. The manual torch are controlled by electronic governors. The need for
configuration is shown in Fig. 5.42 (A). A mechanised sensing and maintaining a constant arc length does not
PAW system is provided with a specially designed torch arise, because the plasma arc preserves its melting
(see Fig. 5.42 (B), which is fixed to a torch stand, so that capability in spite of variations in arc length.
it can mechanically traverse a joint or can be kept Flowmeters are provided for metering the orifice,
stationary while the joint travels. Other accessory units shielding, and backing gases. In addition, timers and a
TOURCH BARREL
HEAD
SHIELDING GAS
NOZZLE
(A) SHIELDING GAS NOZZLE
Fig. 5.42: Torches for PAW: (A) Manual; (B) Mechanised

gas flow tapering device are provided for u p and down combination of plasma gas flow, arc current and arc
sloping of the orifice gas. An adequate supply of cooling travel speed. The keyhole is formed by the plasma
water for the torch is provided by a circulation system. stream displacing the molten metal to the top bead
A flow switch and interlocking circuitry assure that surface. As the plasma torch travels along the joint seam,
cooling water is flowing to the torch before the arc can the metal being melted by the arc is pushed backward
be ignited. around the plasma stream to form the weld pool, which
solidifies as the arc progresses.
Arc Start This motion of the completely penetrated weld metal
Since in PAW the tungsten electrode is well inside allows the impurities to rise to the surface and the gases
the constricting nozzle and touch-starting against the to be expelled before solidification. The welding
workpiece is ruled out, a special method has to be conditions for the keyhole technique must be accurately
devised for initiating the arc. The welding arc is started maintained. The orifice gas flow rate, especially, has to
by first igniting a pilot arc between the electrode and be closely controlled.
the constricting nozzle, the power for this pilot arc being In the keyhole t e c h n i q u e , the molten pool is
derived from a separate power source or from the prevented from spilling by its surface tension. Hence
welding power source itself. To start the pilot arc, the no backing bar is required to support it. However, a
electrode can be made to touch the nozzle and then grooved backing bar, as shown in Fig. 5.43, is used to
retracted to form an arc. This method is suitable for low support the assembly, hold the underbead shielding gas
amperage torches. In the case of a high amperage torch, and provide- an exhaust for the plasma jet. When filler
a high frequency generator is provided, or high-voltage, metal is required, it is added to the leading edge of the
low-power pulses are superimposed on the welding pool formed by the keyhole. The molten metal flowing
circuit. The high-voltage power ionises the gas and around the keyhole then forms a reinforced weld bead.
enables it to conduct the pilot arc current. Square butt joints up to 6.4 mm t can be welded in a
In a typical system using the HF generator, the single pass this way. For heavier joints which require
constricting nozzle is connected to the positive terminal multi-pass welding, partial beveling is done and the root
of the power source through a current reducing resistor. pass of the largest size possible is deposited with the
A low-current pilot arc is first formed between the keyhole technique without using filler wire. The rest of
electrode and the nozzle by the HF generator, the the passes are then made with the normal melt-in
electrical circuit being completed through the resistor. technique with filler wire additions.
The pilot arc ionises the gas, which forms an electrical Keyhole technique presents no difficulties in straight
path between the electrode and the nozzle or between seams as well as in circumferential pipe welds when
the electrode and the work. When the power source is the metal thickness is under 3 mm. The welds can be
switched on, the main arc is initiated between the started and completed at full operating current, travel
electrode and the work or between the electrode and speed and orifice gas flow rate, because the keyhole is
the nozzle, depending on whether the circuit is arranged formed without any disturbance in the weld pool.
for transferred or non-transferred arc. The pilot arc may When the keyhole technique is used on thicknesses
then be extinguished. over 3 mm, the plasma stream tends to bore a tunnel
Procedures and Techniques

The manual PAW technique is similar to that used
for TIG welding. It is referred to as melt-in technique. PLASMA JET
Low amperage and low-orifice gas flow rate are used
and the filler rod is added manually to the weld pool.
This technique is used to weld in all positions.
The penetrating power of the plasma arc and high
speed capability of the process are fully utilised in BACKING GAS BACKING BAR
mechanised welding by adopting the so-called keyhole
technique. This technique is suitable for square butt
joints in thickness range of 1.6-6.4 mm. It consists of
producing a small weld pool with a hole penetrating Fig. 5.43: Typical backing bar for keyhole plasma arc
completely through the base metal, using a special welding
beneath the molten metal during keyhole start. Run-on

and run-off tabs have to be used on straight seams to
exclude the keyhole start and finish areas from the main 40 PLASMA ARC
seam. On circumferential joints where such tabs cannot
be used, the keyhole initiation and withdrawal is 0.030 IN DIAM ORIFICE
30
automatically programmed to give a gradual rise in
welding current and orifice gas flow rate at the start, 20
V GAS TUNGSTEN ARC
and their gradual decay at the end as shown in Fig. 5.44.
Welding e q u i p m e n t i n c o r p o r a t i n g controls to
10
programme the welding operation as described is
commercially available.
Presently PAW process is limited to thicknesses of 25
0 ? i A
AMPERES (DCSP)
? I
P
mm and lower for butt welds. When thickness exceeds Fig. 5.45: Comparison of gas-tungsten arc and plasma
25 mm, the plasma torch nozzle is not able to reach the arc volt-ampere characteristic curves below 10 amp in
root of the joint. argon
thickness the gravity effect diminishes rapidly until at

| / CURRENT
\ 1 0.38 mm it disappears. The surface tension of the molten
metal then determines its shape regardless of the
\JJ
GAS welding position.
z> The success of thin foil welding depends largely on
u < effective gas shielding and good fixturing. Fixturing in
o
z butt joints is simplified by flanging the edges. The
5 \ ^
> flanged edges act as pre-placed filler metal to fill the
gap and assure fusion across the joint. They also
STARTING WELD CLOSING
minimize warpage by providing extra stiffness. Defects
1 KEYHOLE TIME KEYHOLE
usually arise from (a) failure to maintain continuous
Fig. 5.44: Welding current and orifice gas slope control contact between the two joint edges, (b) oxidation of
cycle for starting and terminating a keyhole weld the base metal or weld metal due to improper shielding,
and (c) inadequate clamping.
The electrical circuit is shown in Fig. 5.46. It includes
Microplasma Arc Welding two separate power sources, one for the pilot arc and
For welding extremely thin sheets and foils (0.10- the other for the transferred arc. The pilot-arc power is
1.25 mm t), low current PAW is used, in which the pre-set to deliver approximately 5 amps continuously
current range may be 0.1-10 Amps, or 0.2-20 Amps. The when in use. The pilot arc is initiated by touching the
process is sometimes referred to as microplasma arc
welding. Plasma arc is preferred in place of normal TIG
POWER SOURCE AND
arc, because at such low currents the TIG arc is highly CONTROL UNIT
unstable and inconsistent. The reason for this is TORCH
explained by the V-Amp curves shown in Fig. 5.45. The ELECTRODE (-)
slope of the 6.4 mm long plasma arc is nearly flat over TRANSFERRED PILOT ARC
ARC POWER POWER
0.10-10 Amps, range, indicating very little change in SUPPLY
SUPPLY
ORIFICE GAS
arc voltage. On the other hand, in the case of the 1.3
/ SHIELDING GAS
mm long tungsten arc, the voltage registers 83% rise
CURRENT
(12 to 22 V) aL the current decreases from 10 to 0.10 <
ni
< a CONTROL
Amps. The relatively stable arc voltage of the plasma SNOZZLETm
ensures steady and controlled welding at low currents. CONTACTOR I © 'GAS LENSE
In microplasma welding of thin sheets, the molten ! NOZZLE
weld pool being extremely small, surface tension rather U—<= WORK ZJ<+>
than gravity becomes increasingly dominant as the
section thickness decreases. For example, below 0.76 mm Fig. 5.46: Microplasma arc welding equipment
electrode to the nozzle and then retracting it or with a PAW has b e e n p r o v e d to give better flexibility,
HF discharge. The main transferred arc is started by consistently high quality, and better economy. It is
closing the contactor. therefore an accepted process in shipbuilding, nuclear,
Microplasma welding can be mechanised and electronic, aerospace and many other engineering
adapted for keyhole technique down to 0.8 mm, but the industries.
success depends on the accurate measurement and Mechanised high current PAW with keyhole
control of the plasma jet velocity, which is not adequately technique has been used to obtain high welding speeds
provided by the flowmeter, but requires the use of a and consistently high quality joints in the following
pressure gauge. typical applications in the U.S.A.:
A typical microplasma welding equipment consists 1) Stainless steel and titanium tubing (longitudinal
of power source giving a current range of 0.2-20 Amps., welds)
a control module, welding torch with accessories, water 2) Girth joints in pipe fabrication
recirculatory unit with suitable connecting hoses, foot 3) Missile tankage
switch assembly with cable and plug, pressure regulator 4) Turbine engine components
for argon (shielding) along with non-return valve, Manual PAW with the melt-in technique is usually
pressure regulator for hydrogen (shielding) along with employed on very thin metal sections, because of the
non-return valve, pressure regulator for argon (plasma), ease and reliability it provides.
gas hose with suitable end connectors for hydrogen Microplasma welding is successfully used on the
(shielding gas), and gas hose w i t h suitable end following applications, because it provides a stable yet
connectors for argon (plasma). cool arc at very low currents:
1) Thin-wire mesh screen filters
Salient features of this equipment are:
2) Small-wire butt welds
• Fully transistorised, ensuring very high reliabil- 3) Relay case fabrication
ity and excellent arc stability. 4) Bellows assemblies
• Pulsed mode operation for precise control of heat 5) Thermal shields
input through accurate programming. 6) Thin-wall pressure vessels
• All controls mounted on front panel for easy 7) Vacuum tube components
accessibility. Independent stepless variation of 8) Thermocouple junctions.
all welding parameters like up-slope, peak
current, background current, down-slope, peak It is possible for highly skilled welders to use
current time, background current time, etc. microplasma keyhole technique to achieve uniform
• Gas pre-flow and post-flow facilities provided. w e l d p e n e t r a t i o n in applications such as pipe
fabrication.
• Inbuilt gas mixer unit for optimum results.
• Non-return valves provided in shielding gas- Filler Metals for TIG and Plasma Arc Welding
flow lines for maximum safety. Information on filler metals is available from the
• Being fully solid-state, requires little mainte- relevant standards of the AWS, ISI and BSI.
nance. The list of AWS specifications is given in Table 5.18.
• Very low power consumption. As indicated in Table 5.18, filler wires in some
specifications are suitable also for MIG, oxyacetylene
• Light weight, easy portability. a n d s u b m e r g e d - a r c w e l d i n g processes. In the
• Immune to input supply voltage fluctuations up classification system for all these wires, letters ER or R
to ± 5%. are used as prefix followed by a number indicating the
• Constant current output characteristics to enable chemical composition of the filler metal. E stands for
excellent arc stability and uniform penetration. electrode and indicates that the wire can be used as an
• Most modern pilot-arc starting system for instan- electrode in processes such as MIG and submerged-arc
taneous start. welding (where if forms a part of the welding circuit).
R stands for rod and indicates that it is suitable for
Applications of PAW processes such as TIG, PAW and oxyacetylene welding.
PAW can be used on all metals which are normally AWS A5.7-1977 contains the following
welded by the TIG process. On many fabrication jobs classifications of copper and copper-alloy filler rods
Table 5.18: AWS specifications for TIG filler metals

AWS Fillet metals To be used with
specification processes
A5.7 Copper and copper-alloy welding rods and electrodes TIG / MIG
A5.9 Corrosion resistant chromium and chromium-nickel steel bare
electrodes and welding rods TIG / MIG / SA
A5.10 Aluminium and aluminium-alloy welding rods and bare electrodes OA / TIG / MIG / PAW
A5.14 Nickel and nickel-alloy bare welding rods and electrodes TIG / MIG /PAW / SA
A5.16 Titanium and titanium-alloy bare welding rods and electrodes TIG / MIG
A5.18 Mild steel electrodes for gas-metal arc welding TIG / MIG /PAW
A5.19 Magnesium-alloy welding rods and bare electrodes OA / TIG / MIG
A5.24 Zirconium and zirconium-alloy bare welding rods and electrodes TIG / MIG
A5.28 Low-alloy steel filler metals for gas shielded arc welding TIG / MIG /PAW
Notes: OA - Oxyacetylene; SA = Submerged-arc; PAW = Plasma arc welding
suitable for TIG, plasma arc and MIG welding processes. carbon will typically produce a deposit of 0.04% carbon.
They are similar in composition to the manual electrodes There is also some nitrogen pick-up; a gain of 0.02%.
listed in AWS A5.6- in chapter 3 ERCu, ERCuSi-A, The change in other elements is not significant in the
ERCuSn-A, ERCuNi, ECuAl-Al, ECuAl-A2, ECuAI- undiluted weld metal.
A3, ERCuNiAl and ERCuMnNiAl. AWS A5.10-1980 deals with Al and Al-alloy filler
Properties and applications of these filler metals are wires suitable for oxyacetylene, TIG, plasma arc and
similar to those of the corresponding manual electrodes MIG processes. As shown in Table 5.19, there are 14
described in chapter no. 3. classifications; 10 of them carry the prefix ER which
AWS A5.9-1977 has listed corrosion-resisting means that they can be used as electrodes by the MIG
straight chrome and chromium-nickel steel filler wires process and as r o d s by the other processes. The
suitable for use with TIG, MIG and submerged-arc remaining carry the prefix R, indicating that they are
processes, which are similar to those listed as manual suited for the first three processes and not for the MIG
electrodes in AWS A5.4-in chapter no. 3. The process.
compositions are such that the weld deposit has the The last four are used for the repair of castings. The
capability of providing an approximately predictable rest are general purpose filler metals meant for use on
ferrite number. various standard grades of wrought Al-alloy products
The typical classifications are ER 308L, ER 309, ER as indicated in tabular form in the Appendix.
316, ER 310 and ER 347. The Appendix states that Si The Al and Al-alloy rods must have clean and oxide-
may be used as suffix to the classification to indicate free surfaces. To quote from the Appendix: "Oil, or other
that the filler metal contains 0.65-1.0% Si rather than organic materials, as well as a heavy oxide film on the
the standard 0.30-0.65% Si. Examples are: ER 308 Si, ER rod will interfere with coalescence of the weld. Because
308L Si, ER 309 Si, ER 316 Si, ER 316L Si, ER 309 Si, ER of this, it is necessary to clean the welding rod and
316 Si, ER 316L Si and ER 347 Si. The higher Si level electrode before packaging. Proper storage of welding
improves the usability of the filler metal in TIG / M I G / rods and electrodes is essential to avoid contamination,
PAW processes. If dilution by the base metal results in which may affect their performance. Packages of filler
low ferrite or fully austenitic weld metal deposit, the metal should not be left outdoors or in unheated
high-Si content weld metal is more resistant to cracking buildings, because the greater variations in temperature
than a low-Si content weld metal. and humidity increase the possibility' of condensation.
The TIG and plasma arc processes involve the least Experience has demonstrated that undesirable storage
change in the chemical composition from wire to conditions m a y a d v e r s e l y affect filler metal
deposit, and hence produces the smallest difference performance. Investigation of the effect of storage time
between the ferrite content calculated from the wire on electrode performance indicates that packaged
analysis and that measured on the deposit. There is some electrodes, stored under good conditions (dry places in
loss of carbon in gas-tungsten arc welding—about half heated buildings), are satisfactory after extended
of the carbon content above 0.02%. Thus a wire of 0.06% storage. Contamination of filler metal from handling or
Table 5.19: AWS classification of Al and Al-alloy filler wires

AWS class Approx. composition
ER 1100 99.0 min. Al
ER 2319 6.3 Cu / rest Al with controlled amounts of V, Zr
ER4043 5.3 Si / rest Al
ER 4047 12 Si/rest Al
ER 4145 10Si/4Cu/restAl
ER 5183 0.8 M n / 4.8 Mg / rest Al
ER 5356 0.1 Mn / 5 Mg / rest Al
ER5554 0.8 M n / 2.7 Mg / rest Al
ER 5556 0.8 Mn/5.1 Mg/rest Al
ER 5654 Mn trace / 3.5 Mg / rest Al
R242 4Cu/1.5Mg/2Ni/restAl
R295 4.5 Cu / rest Al
R355 5 Si / 1.2 Cu / 0.5 Mg / rest Al
R356 7Si/0.3Mg/restAl
storage may occur. In most cases, the contaminating side of nickel-chromium-iron clad steel and for surfacing
influences will dictate the cleaning method. The practice steel with nickel-chromium-iron alloy.
of giving the welding rod a rub with steel wool, just ERNiCrFe-5 is intended for welding nickel-
before welding is quite widely followed, if it has been chromium-iron alloys (ASTM B163, B166, B167 and B
exposed to the shop atmosphere for long periods of 168). The higher columbium-plus-tantalum content,
time." aimed at a minimum of four times the silicon content,
A5.14-76 lists Ni and Ni-alloy bare filler metal for is intended to minimise cracking in welds, where high
use with the TIG, Plasma arc, MIG and submerged-arc stresses are encountered, such as in thick material.
processes. The classifications covered are listed in Table ERNiCrFe-6 is intended for welding certain
5.20. dissimilar metals and for surfacing carbon steel. The
The following notes on applications are taken from deposited weld metal will precipitation harden on heat
the Appendix. treatment. The degree to which it hardens depends on
ERNiCr-3 is intended for welding nickel-chromium- the temperature and the time at temperature. The
iron alloy, for dissimilar metals, for welding the clad supplier or his technical literature should be consulted
Table 5.20: AWS classification of Ni and Ni-alloy filler wires

AWS class Approx. composition
ER Ni-1 Similar to the electrode in A5.ll
ER NiCu-7 Similar to the electrode in 5.11, with 2.2 Ti
ER NiCr-3 67.0 min Ni / 20.0 Cr / 2.5 Cb + Ta
ER NiCrFe-5 70.0 min Ni / 15.5 Cr / 8.0 Fe / 2.2 Cb + Ta
ER NiCrFe-6 67.0 min Ni / 15.5 Cr / 8.0 Fe / 3.0 Ti
ER NiCrFe-7 70.0 min Ni / 15.5 Cr / 7.0 Fe / 2.4 Ti / 0.7 Al / 1.0 Cb + Ta
ER NiFeCr-1 42.0 Ni / 20.0 min Fe / 21.5 Cr / 3.0 Mo / 1.0 Ti
ER NiMo-1 Similar to the electrode in A5.ll
ER NiMo-2 16.5 Mo / 7.0 Cr/rest Ni
ER NiMo-3 Similar to the electrode in A5.ll
ER NiMo-7 28.0 Mo / rest Ni
ER NiCrMo-1
ER NiCrMo-2
ER NiCrMo-3 Similar to the electrode in A5.ll
ER NiCrMo^
ER NiCrMo-5
ER NiCrMo-7 16.0 Cr / 15.5 Mo / 0.50 Ti / rest Ni
ER NiCrMo-8 24.5 Cr / 6.0 Mo / 1.10 Ti / 1.0 Cu / 50.0 Ni
for specific recommendations concerning the precipi- processes. It covers seven classes as shown in Table
tation hardening tendency of the weld from this filler 5.22.
metal. Strictly speaking, none of the standard grades is
ERNiCrFe-7 is used for welding nickel-chromium- suitable for TIG welding, because of high Si a n d / o r Mn
iron-titanium precipitation hardenable alloys. Prior to contents. They are really meant for MIG/CO; welding
precipitation hardening, the weldment must be solution using argon-oxygen mixture, argon-C0 2 mixture or pure
annealed at 980°C for one hour. Air cooling or water C 0 2 as a shielding gas. The filler wire composition
q u e n c h i n g follows. The w e l d m e n t can then be normally recommended for TIG welding of mild steel
precipitation hardened by heating it to 700°C and is C 0.12% maximum, Mn—0.90 to 1.45%, Si—0.20
holding for 20 hours, followed by air cooling. to 0.50%. It could be covered by the classification R70S-G.
ERNiFeCr-1 is intended for gas shielded arc welding AWS A5.19-69 deals with four classifications of
of nickel-iron-chromium-molybdenum-copper alloy magnesium alloys as shown in Table 5.23
(ASTM B423 or UNS-No. 8825). The Appendix indicates in a tabular form the various
ERNiMo-2 is intended for welding corresponding commercial Mg alloys, on which the above filler metals
base metal (ASTM B434) to itself or to other alloys and can be used and the typical weld properties that can be
for cladding steel by the TIG and MIG processes. achieved.
ERNiMo-7 is i n t e n d e d for w e l d i n g the The filler wires requires special care in storage and
corresponding type of nickel-molybdenum base metal use as in the case of Al and Al-alloy wires. The following
(ASTM B333) and for cladding steel. It can be used with recommendations given in the Appendix must be
the gas shielded arc processes. Postweld heat treatment followed:
is neither required nor prohibited. Filler metal ele aliness and freedom from surface
E R N i C r M o - 7 is i n t e n d e d for w e l d i n g nickel- oxidation is important in gas shielded-arc welding. Oil
chromium-molybdenum alloy base metal (ASTM B575), or other organic material as well as heavy oxide coatings
dissimilar metals and for cladding steel by the TIG, on the surface of the filler metal will interfere with
plasma arc and MIG processes. coalescence of the weld puddle and cause porosity or
ERNiCrMo-8 is i n t e n d e d for w e l d i n g nickel- other weld defects or both. For this reason, filler metal
chromium-molybdenum alloy base metal (ASTM is manufactured and packaged in such a manner as to
B582), dissimilar metals and for cladding steel by prevent contamination.
the TIG, plasma arc and MIG processes. Proper storage of welding rod and electrode in the
AWS A5.16-70 gives 14 classifications of Ti and user's plant is essential, if the filler metal is to remain
Ti-alloy filler wires as shown in Table 5.21. free of contamination until used. Packages of filler metal
A unique feature of these filler metals is that severe should not be left outdoors or in unheated buildings,
restrictions have been placed on C, O, H and N contents, because the greater variations in temperature and
which together are referred to as the interstitial content. h u m i d i t y increase the possibility of m o i s t u r e
The footnote to the table states that analysis for condensation. Properly protected filler metal can be
interstitial content should be made on the finished wire, stored for long periods of time without adverse effect
i.e. wire drawn to its final diameter. Its value is kept on its performance.
low, especially in the classes marked* so that the high Packages of filler metal should remain sealed until
toughness required for cryogenic applications and other ready for use. Once removed from the container, spooled
special uses can be obtained in the deposited weld metal. electrode should be kept covered, even during use. to
The Appendix gives unusually detailed information prevent surface contamination by dust, moisture, or
on the welding of Ti and Ti-alloys by covering the other air-borne foreign material. After welding is
following topics: w e l d i n g characteristics, surface completed, the electrode should be returned to its
preparation for welding, welding procedure. Inspection, original container for storage. Welding rod should also
mechanical testing, weld-metal tests for cryogenic be kept covered until ready for use. Storage conditions
applications. Effect of interstitial elements on welds in here need not be as rigorous as for spooled electrode,
alpha-beta-Ti alloys, and welding of the 1.32 metre because good welding practice always includes a
diameter 6A1-4V-Ti second stage minuteman rocket stainless steel wool rub of the rod just prior to use, to
motor case, serve as an example. remove surface oxidation or other contamination.
AWS A5.18-79 d e a l s w i t h c a r b o n steel filler The Appendix also provides detailed information on
metals for TIG, p l a s m a arc, MIG a n d M I G / C O ; the welding of Mg alloys in the same manner as AWS A5.16
Table 5.21: AWS classification of Ti and Ti-alloy filler wires
Hydro- Nitro- Alumi- Vana- Tin Chro- Iron Molyb- Colum- Tant- Palla- Tita-
AWS Carbon Oxygen gen gen nium dium mium denum bium lum dium nium
0/ 0/ 0/ 0/ 0/ 0/ 0/
classification % /o % % % %
/o /o /o /o /o /o /o
ERTi-1 0.03 0.10 0.005 0.012 — — — — 0.10 — — — — Remainder
ERTi-2 0.05 0.10 0.008 0.020 0.20 Remainder
0.10- " " " " " " "
ERTi-3 0.05 0.15- 0.008 0.020 — — — — 0.20 — — — — Remainder

0.15-
ERTi-4 0.05 0.25 0.00ft 0.020 0.30 Remainder

ERTi-0.2Pd 0.05 0.15 0.008 0.020 — — — — 0.25 — — — 0.15- Remainder
0.25
ERTi-3Al-2.5V 0.05 0.12 0.008 0.020 2.5- 2.0 — — 0.25 — — — — Remainder
3.5 3.0
ERTi-3Al-2.5V-l* 0.04 0.10 0.005 0.012 2.5- 2.0- — — 0.25 — — — — Remainder
3.5 3.0
ERTi-5Al-2.5Sn 0.05 0.12 0.008 0.030 4.7- 2.0- 0.40 Remainder
5.6 " 3.0 " " " " "
ERTi-5Al-2.5Sn-l* 0.04 0.10 0.005 0.012 4.7- — 2.0- — 0.25 — — — — Remainder

5.6 3.0
ERTi-6Al-2Cb-lTa- 0.04 0.10 0,005 0.012 5.5- — — — 0.15 0.5- 1.5- 0.5- — Remainder
TMo 6.5 1.5 2.5 1.5
ERTi-6Al-4V 0.05 0.15 :.UH)8 0.020 5.5- 3.5- — — 0.25 — — — — Remainder
6.75 4.5
ERTi-6Al-4V-l* 0.04 0.10 .,-#* 0.012 5.5- 3.5- — — 0.15 — — — — Remainder
6.75 4.5
ERT1-8AI-IM0-IV 0.05 o.r 0.03 7.35- 0.75 — — 0.25 0.75- — — — Remainder
8.35 1.25 1.25
ERTi-13V-llCr-3Al 0.05 0.12 0.008 0.03 2.5 12.5 — 10.0 0.25 — — — — Remainder
14 c; 190
*This classification offillermetal restricts the allmmble interstitial acontent

low level
toin order that the high toughness required for applications and other
cryogenic special uses can be obtained
in the deposited weld metal.
Note 1: Analysis for interstitial content shall be nvide after the welding rod or electrode has been reduced to itsfinaldiameter.
Note 2: Single values are maximum percentages.
Table 5.22: AWS classification of carbon steel filler wires

Chemical composition, weight %
AWS
Classification Mn Si Ni Cr Mo V Cu Ti Zr Al
0.90 0.401 0.05 0.02 0.05
ER70S-2 0.07 to to 1 to to to
1.40 0.70 0.15 0.12 0.15
0.06 0.90 0.45
ER70S-3 to to to
0.15 1.40 0.70
0.07 1.00 0.65 !
ER70S-4 to to to 'I 0.025 0.035 — 0.50
0.15 1.50 0.85 1
0.07 0.90 0.30 0.50
ER70S-5 to to to to
0.19 1.40 0.60 0.90
0.07 1.40 0.80
ER70S-6 to to to 1
0.15 1.85 1.15 f
0.07 1.50 0.50
ER70S-7 to to to
0.15 2.00 0.80
ER70S-G No chemical r equirerr
PART 1: Ferritic Steels

Table 5.23: AWS classification of Mg-alloy filler wires
It is convenient to discuss this part with IS 6419-1971,
AWS Class Approx. composition because they are identical and deal with wires for TIG
ERAZ61A 6.5 Al / 1.0 Zn/rest Mg and MIG/C0 2 welding.
ERAZ101A 10.0 Al / 1.0 Zn/rest Mg The British standard lists five types of wires for mild
ERAZ92A 9.2 Al / 2.0 Zn/rest Mg steel as shown in Table 5.25. The types are designated A15,
EREZ33A 2.5 Zn / 0.75 Zr/3.2 rare earth / rest Mg A16, A17, A18 and A19, depending on the chemical
composition. The same identical five types are included in
AWS A5. 28-79 deals with bare solid wires of low- IS 6419, but the designations are S-1, S-2, S-3, S-4 and S-5.
alloy steel for use with TIG, MIG and PAW processes. In explaining the characteristics of these different
The class-ifications and chemical compositions are types, we shall refer to the British designations:
reproduced in Table 5.24.
AWS A5.24-76 dealing with Zr and Zr-alloy filler A15
metals is not elaborated here, because the welding of Could be used for TIG welding or MIG with C0 2 ,
these alloys is rarely carried out. Ar plus C 0 2 or Ar plus 0 2 mixtures.
British Standard A16

The British standard on TIG/MIG filler metals is BS The deoxidant content of this wire indicates its
2901:1970 entitled Tiller rods and wires for gas-shielded probable suitability for use with TIG or MIG with C0 2 ,
arc welding'. It is available in five parts: Ar plus C 0 2 or Ar plus 0 2 mixtures. High carbon levels
(0.30%) would obviously enhance the yield and ultimate
1) Ferritic steels tensile strengths compared with other Si/Mn wires in
2) Austenitic stainless steels this series. This does, however, appear to have limited
use in industrial application possibly because of
3) Copper and copper alloys
anticipated increased susceptibility to solidification
4) Al and Al alloys and Mg alloys cracking. In fact commercial availability of this wire is
5) Nickel and nickel alloys. doubtful even though it remains in the standard.
Table 5.24 : AWS classification of low-alloy steel filler wires
Chemical composition requirements, %
AWS Carbon Manga- Silicon Phos- Sul- Nickel Chro- Molyb- Vana- Tita- Zirco- Alumi- Copper Total
Classifi- nese phorus phur mium denum dium nium nium nium other
cation elements
Chromium-molybdenum steel electrodes and rods

ER80S-B2 0.07-0.12 0.40-0.70 0.40-0.70 0.025 0.025 0.20 1.20-1.50 0.40-0.65 — 0.35 0.50
ER80S-B2L 0.05 0.40-0.70 0.40-0.70 0.025 0.025 0.20 1.20-1.50 0.40-0.65 — 0.35 0.50
ER90S-B3 0.07-0.12 0.40-0.70 0.40-0.70 0.025 0.025 0.20 2.30-2.70 0.90-1.20 — 0.35 0.50
ER90S-B3L 0.05 0.40-0.70 0.40-0.70 0.025 0.025 0.20 2.30-2.70 0.90-1.20 — 0.35 0.50
Nickel steel electrodes and rods

ER80S-NU 0.12 1.25 0.40-0.80 0.025 0.025 0.80-1.10 0.15 0.35 0.05 — 0.35 0.50
ER80S-Ni2 0.12 1.25 0.40-O.80 0.025 0.025 2.00-2.75 — — — 0.35 0.50
ER80S-N13 0.12 1.25 0.40-0.80 0.025 0.025 3.00-3.75 — — 0.35 0.50
Manganese-molybdenum steel electrodes and rods

ER80S-D2 0.07-0.12 1.60-2.10 0.50-0.80 0.025 0.025 0.15 0.40-0.60 — 0.50 0.50
Other low-alloy steel electrodes and rods

ERlOOS^l 0.08 1.25-1.80 0.20-0.50 0.010 0.010 1.40-2.10 0.30 0.25-0.55 0.05 0.10 0.10 0.10 0.25 0.50
ER100S-2 0.12 1.25-1.80 0.20-0.60 0.010 0.010 0.80-1.25 0.30 0.20-0.55 0.05 0.10 0.10 0.10 0.35-0.65 0.50
ER110S-1 0.09 1.40-1.80 0.20-0.55 0.010 0.010 1.90-2.60 0.50 0.25-0.55 0.04 0.10 0.10 0.10 0.25 0.50
ER120S-1 0.10 1.40-1.80 0.25-0.60 0.010 0.010 2.00-2.80 0.60 0.30-0.65 0.03 0.10 0.10 0.10 0.25 0.50
ERXXS-G Subject to agreement between supplier and purchaser
Table 5.25: BS classification of mild steel filler wires

Type Carbon Silicon Manganese Aluminium Sulphur Phosphorus
min. max. min. max. min. max. min. max. min. max. min. max.
% % % % % %% % % % % %
A15 — 0.12 0.30 0.90 0.90 1.60 0.04 0.40 — 0.040 — 0.040
A16 0.25 0.30 0.30 0.50 1.30 1.60 — — — 0.040 — 0.040
A17 — 0.12 0.20 0.50 0.85 1.40 — — — 0.040 — 0.040
A18 — 0.12 0.70 1.20 0.90 1.60 — — — 0.040 — 0.040
A19 0.08 0.12 0.30 0.50 1.00 1.30 0.35 0.75 — 0.040 — 0.040
A17 Sometimes wires have titanium and zirconium

Should be satisfactory for TIG welding or MIG deoxidants to partly replace silicon, which operate
welding with predominantly Ar-gas mixtures. satisfactorily under C 0 2 shielding. Such wires will not
give harmful effects if used with argon-oxygen mixtures.
A18
Suitable for TIG welding, and by far the most PART 2: Ausentic Stainless Steels
popular choice for MIG welding with C0 2 , Ar plus C 0 2 This standard lists 14 types of filler metal. The range
and Ar plus 0 2 mixtures. is much smaller than in AWS A5.9, because while the
BS deals only with austenitic stainless steels, the
A19 American Standard covers straight chrome and Cr-Ni
Suitable for TIG welding and MIG with C0 2 , Ar plus ferritic, austenitic and martensitic stainless steels. The
C 0 2 and Ar plus 0 2 mixtures. Although the silicon levels 14 BS classifications and their AWS equivalents are
are somewhat lower than other wires for use with C0 2 , shown in Table 5.26.
this is compensated for by 0.75% (maximum)
aluminium. PART 3: Copper And Copper Alloys
Generally, this wire is capable of producing porosity- The standard covers 13 types as shown in Table 5.27.
free joints on weldments, which are rusty or otherwise Type C7 is intended for welding copper using argon
contaminated. Despite this apparent advantage, the or helium as a shielding gas, while Type C8 is a modified
wires do not share the popularity of type A18. Probably version to be used with nitrogen as a shielding gas. Type
this can be attributed to their inferior metal transfer and C21 carrying minute amounts of boron is covered by a
flow characteristics, which also result in a weld of coarse British patent.
appearance.
It must be appreciated that it is difficult to quantify PART 4: AI/AI Alloys And Mg Alloys
each wire application, because factors other than the The standard covers eight types of Al and Al alloys
wire composition and shielding gas, should also be as shown in Table 5.28
considered. These include cleanliness of the weidments, The standard covers eight types of Mg alloys as
chemical composition of the weldments with respect to detailed in Table 5.29.
deoxidant levels, dilution levels and element recovery
values associated with the particular metal transfer PART 5: Nickel And Nickel Alloys
characteristic being used. Eleven types are covered as shown in Table 5.30.
Important note: Generally speaking, wires which have
high level of Si (above 0.65%) or of Si plus Al (above Indian Standards
0.80%) are designed to work with a C 0 2 shield. Since The four Indian Standards on TIG/MIG filler metals
the C 0 2 gas is reactive in the arc, the level of Si is are:
considerably reduced in the weld metal, which is an 1) IS 6419-1996: equivalent to BS 2901-1970—Part 1
important and favourable condition. If such wires are 2) IS 585*^1970: equivalent to BS 2901-1970— Part 2
used with argon or argon-oxygen mixture, the weld 3) IS 5897-1970: equivalent to BS 2901-1970— Part 4
metal has an excess of silicon and is prone to hot cracks. 4) IS 5898-1970: equivalent to BS 2901-1970— Part 3
Table 5.26: BS classification of S.S. filler wires

BS 2901: Part 2 Equivalent
Type AWS A5.9 type
308S92 ER 308L
308S96 ER308
308S93 ER 308L Si
307S96 ER347
309S94 ER309
311S94 This is ER 309 type and stabilised with Nb. There is no AWS equivalent.
310S94 ER 310
310S98 This is ER 310 with 0.35-0.45%C. There is no AWS equivalent.
313S94 This is E 310 and stabilised with Nb. There is no AWS equivalent.
316S92 ER 3216L
316S96 ER316
316S93 ER 316L Si
317S96 ER 317
318S96 ER 318
:Table 5.27: BS classification for Cu and its alloy filler wires

Type Common Approx. composition Nearest AWS
name equivalent
C7 copper 98.5 min. Cu / 0.25 Mn / 0.28 Si ERCu
C8 copper 99.4 min. Cu / Al+Ti 0.38 X
C9 silicon-bronze 1.0 M n / 3.0 S i / rest Cu ER CuSi-A
ClO phosphor-bronze 93.8 min. Cu / 5.3 Sn / P 0.20 ER CuSn-A
Cll phosphorbronze 92.3 min. Cu / 6.8 Sn / P 0.20 X
C12 aluminium-bronze 90.0 min. Cu / 6.8 Al ER CuAl-Al
C12Fe aluminium-bronze 89.0 min. Cu / 75 Al / 3.0 Fe X
C13 aluminium-bronze 86.0 min. Cu / 10.0 Al / 1.0 Fe ER CuAl-A2
C16 copper-nickel 0.35 Ti / 1.6 Fe / 11.0 Ni / 0.8 Mn / rest Cu X
C18 copper-nickel 0.35 Ti / 0.7 Fe / 31.0 Ni / 0.8 Mn / rest Cu ER CuNi
C20 aluminium-bronze 82.5 Cu / 8.8 Al / 2.5 Fe / 4.2 Ni / 1.3 Mn ER CuNiAl
C21 copper 99.8 min. Cu / 0.06 B X
C22 aluminiumbronze 7.8 Al / 3.0 Fe / 2.2 Ni / 12.5 Mn / rest Cu ER CuMnNiAl
Table 5.28 : BS classification of Al and Al alloy filler wires

Type Approx. composition Nearest
AWS equivalent
G1A 99.8 min. Al / Cu+Si+Fe+Mn+Zn : 0.2 max. ER1100
GIB 99.5 min Al / Cu+Si+Fe+Mn+Zn+ : 0.5 max. ER1100
NG21 Al / 5 Si ER4043
NG3 Al / 1.0 Mn X
NG5 Al / 3.5 Mg ER5654
NG52 Al/2.7Mg/0.8Mn/0.1Ti ER 5554
NG6 Al / 5.0 Mg / Mn + Cr : 0.1 to 0.5 ER5356
NG61 Al / 5.25 Mg / 0.8 Mn / 0.1 Cr / 0.1 Ti ER5556
Table 5.29: BS classification of Mg and its alloy filler wires

AWS equivalent
Dl 9.8 Al / 0.27 Mn / Cu+Si+Fe+Ni: 0.4 max / rest Mg ERAZ101A
D2 1.5 Mn / rest Mg X
D3 7.0 Al/0.27 M n / r e s t Mg ERAZ61A
D4 3.0 Al / 0.42 Mn / 1.0 Zn / rest Mg X
D5 1.1 Zn / 0.7 Zr / rest Mg X
D6 2.0 Zn / 0.7 Zr / 2.0 Cd / rest Mg X
D7 4.3 Zn / 0.7 Zr / 1.25 rare earth metals / rest Mg X
D8 2.0 Zn / 0.7 Zr / 3.2 rare earth metals / rest Mg EREZ33A
Table 5.30: BS classification of Ni and its alloy filler wires

AWS equivalent
NA32 Ni with 3% Ti ERNi-1
NA33 66 Ni / 33 Cu ERNiCu-7
NA34 Ni with 20% Cr X
NA35 Ni with 20% Cr / 3.0 Mn / 1.5 Fe / 0.5 Ti / Nb+Ta : 2.5 ERNiCr-3
NA36* 20 Cr / 88 Co / 2.5 Ti / 1.4 Al /rest Ni X
NA37* 18 Cr / 14 Co / 7.0 Mo / 2.2 Ti / 2.1 Al / rest Ni X
NA38* 20 Cr / 20 Co / 5.8 Mo / 2.2 Ti / 0.45 Al / 0.25 Si/rest Ni X
NA39* 15.5 Cr / 4.0 Fe / 3.0 Ti / 2.35 Mn / 67.0 min. Ni ERNiCrFe-6
NA40 21.2 Cr / 1.5 Co / 18.5 Fe / 9.0 Mo / 0.6 W / rest Ni ERNiCrMo-2
NA41 40 Ni / 21.5 Cr / 3.0 Mo / 1.4 Ti / 2.2 Cu / rest Fe ERNiFeCr-1
NA42* 44 Ni / 16.5 Cr / 3.2 Mo / 1.2 Ti / 1.2 Al / rest Fe X
*These are age-hardenable type.
Reference was made to IS 6419 while discussing BS (2.0) 500 or 1000

2901 — Part 1 on page 291. The other Indian standards (2.4)
need no elaboration as they give identical types and 2.5
compositions as contained in the equivalent British (3.2)
standards. 4.0 1000
5.0
Dimensions The preferred sizes are given in parenthesis.
Standard dimensions for TIG welding rods are
specified in ISO/R546, "Lengths and tolerances for
drawn or extruded filler rods for welding supplied in The rods must have smooth finish, and must be free
straight lengths", and they have been incorporated in from surface imperfections, corrosion products, oxides,
BS 2901 and the corresponding Indian standards. The grease and oil, which w o u l d adversely affect the
diameters and lengths are as follows. soundness and mechanical properties of the weld.
Mechanised TIG welding may call for continuous
Diameter (mm) Length (mm) wires wound on spools or rims, or in the form of coils.
(1.2) Standard dimensions for these are given in the next
(1.6) chapter on M I G / C 0 2 welding.
Metals Handbook, Vol. 6, 8th Ed., published 1971, by American Society for Metals.
Modern Welding, Chapter 11, by Althouse et al., published 1965, by The Goodheart-Willcox Co.
Pipe Welding Procedures, Chapter 6 by H. Rampaul, published 1973, by Industrial Press Inc., New York.
Sensors, Present and Future by Venn. Weld. & Met. Fabr., Oct. 1984.
GTA Hot Wire Welding, Weld ]., May 1970.
Orbital Narrow Gap TIG Welding, Met. Constr., Oct. 1977.
Automatic Pulsed-TIG Tube-Plate Welder, Weld. & Met. Febr., May 1980.
TIG Welding Troubleshooting, Weld J. Nov. 1983.
Hot Wire—for TIG. Plasma and SAW, Weld & Met. Fabr., June 1971.
TIG Welding of Al-Bronze, Weld & Met. Fabr., June 1975.
TIG Welding of Al Pipes, Weld & Met. Fabr., Dec. 1975.
Metal
lnert-Gas/C02
Arc Welding
METAL INERT-GAS ARC WELDING, as the name nium. This occurred in 1950. The replacement of argon by
suggests, is a process in which the source of heat is an the active gas C0 2 or argon-oxygen mixture took place in
arc formed between a consumable metal electrode and 1955. In 1960, the flux-cored continuous wire was
the workpiece, and the arc and the molten puddle are developed to replace the solid-metal electrode wire. The
protected from contamination by the atmosphere (i.e. years 1960-66 saw the growth of the spray transfer tech-
oxygen and nitrogen) with an externally supplied gas- nique and the development of the pulsed MIG process.
eous shield of inert gas such as argon, helium or an ar- M I G / C 0 2 welding process overcomes the restriction
gon-helium mixture. No external filler metal is neces- of using small lengths of electrodes as in manual metal-
sary, because the metallic electrode provides the arc as arc w e l d i n g a n d overcomes the inability of the
well as the filler metal. It is often referred to in abbrevi- submerged-arc process to weld in various positions. It
ated form as MIG welding. is not surprising, therefore, that the 50/50 level of the
MIG welding was primarily developed as a high relative weights of weld metal deposited by the manual
current-density, smaller-diameter metal-electrode metal-arc and MIG/ C 0 2 processes was reached in 1973
process with argon shielding for welding aluminium. in the U.S.A. and in 1978 in Europe.
Hence the term MIG was appropriate, and the process M I G / C 0 2 is basically a semi-automatic process, in
was equally applicable to deoxidised copper and silicon which the arc length and the feeding of the wire into
bronze. But when it was extended to the welding of the arc are automatically controlled. The welding
ferritic and austenitic steels, addition of 1-2% oxygen operator's job is reduced to positioning the gun at a
to argon was found necessary to get smooth metal correct angle and moving it along the seam at a
transfer. Later it was established that mild steel could controlled travel speed. Hence less operator skill is
be welded using 100% carbon dioxide (C0 2 ) or argon- required with this process as compared to the TIG and
C 0 2 mixture as a shielding gas. Since these gases are manual metal-arc processes. Yet, systematic initial
not inert, the process came to be termed as M I G / C 0 2 o r training in the setting u p of the e q u i p m e n t and
MIG/MAG welding process. MAG is an abbreviation manipulation of the gun must be provided to the
of metal active-gas, in which active gas refers to argon- operator to ensure quality M I G / C 0 2 welding. By
oxygen, carbon dioxide and argon-C0 2 mixture, which suitably adjusting the process parameters, it is possible
are chemically reactive and not inert. to weld joints in the thickness range of 1-13 mm in all
The American Welding Society refers to the process welding positions. For production welding of steel joints
as Gas Metal-Arc Welding and has given it the letter thicker t h a n 13 m m , flux-cored arc w e l d i n g or
designation GMAW. This term appears simpler, because submerged-arc welding is preferred.
it covers inert as well as active shielding gases. All the major commercial metals can be welded by
After the successful industrial adoption of the TIG the M I G / C 0 2 process, including carbon steels, low-alloy
process in 1940, it was but one step forward for the and high-alloy steels, stainless steels, aluminium,
development of the MIG process for welding alumi- copper, titanium, zirconium and nickel alloys.
MIG/C0 2 is also used in mechanised and automatic used, because it yields a stable arc and smooth metal
forms to eliminate the operator factor and to increase transfer with least spatter for the entire current range
productivity and consistency of quality. used. AC is unsuitable for this process, because the arc
is extinguished at each half-cycle and rectification of the
BASIC EQUIPMENT reverse polarity cycle makes the arc erratic. DC with
The basic MIG/C0 2 equipment is shown in Fig. 6.1. electrode negative is also not practicable, because the
It consists of a DC power source, a wire feeder, a arc becomes unsteady and erratic.
shielding gas supply, controls for governing wire drive, Constant current power sources, such as those used
current, gas flow and cooling water (if used), and a for manual metal-arc welding are not suitable for this
welding gun. The power source is usually a constant process, because they demand stringent control of
potential welding rectifier giving DC. The wire feeder electrode-feed speed and the system is not self-
pulls the wire electrode from a spool and pushes it regulating. On the other hand a constant voltage power
through the welding gun at required speed. As the wire source, when used in conjunction with a constant-speed
passes through the gun, it picks up electric current from wire feeder, provides a self-correcting arc length system
the copper contact tube, which is electrically connected as explained in Chapter 2. In this case, the arc voltage is
to the power source, and makes an arc with the adjusted by setting the output voltage on the power
workpiece. The initiation and termination of gas flow supply. The welding current is adjusted by setting the
is automatically controlled. The gun is cooled either by wire-feed speed.
the shielding gas passing through it or by a water Power sources are rated at 60% duty cycle for semi-
circulation system. The later is required only when automatic operation and at 100% duty cycle for
currents above 400 amps are used, and then controls automatic continuous operation. The maximum
are provided to start and stop the water circulation amperage used is 600 amp at 60% duty cycle, though in
automatically during the welding operation. exceptional cases automatic equipments may use power
sources of 1,000-2,000 Amps. The commonly used
Power Sources power sources cover the following ratings:
Power sources used in MIG/C0 2 , welding are either Max. Amps, at 60% duty cycle 160 200 300 400 600
welding generators or transformer-rectifiers of constant- Max. Amps, at 100% duty cycle 120 150 230 310 500
voltage type such as are used in SA welding, which give When the power source is of the constant-voltage
direct current. DC with electrode positive is always type, the arc can be started by simply feeding the
ELECTRODE
|WIRE DRIVE
WIRE REEL ^
Note : Sometimes a water circulator is used. MAY BE LOCATED'
IN WELDING GUN,
HANDLE OR AT
WIRE REEL
GAS
SUPPLY
CONTROLS FOR GOVERNING

WIRE DRIVE. CURRENT. GAS
FLOW. AND COOLING WATER.
IF USED
Fig. 6.1: Diagram of MIG/C0 2 arc welding equipment

METAL INERT-GAS/C02 ARC WELDING 317
electrode to the work. It is necessary, however, to have The above sequence of operations is shown diagram-
the wire moving as soon as it comes in contact with the matically in Fig. 6.2.
work, since the large current pulse at short circuit can
burn off the wire faster than it is fed, causing it to fuse Welding Gun
with the contact tube. This is called the burn-back The g u n is provided with (a) a contact tube to
phenomenon. The second requirement for the wire feed transmit the welding current to the electrode wire, (b) a
is that the wire should not protrude too far, by over- gas passage and a nozzle to direct the shielding gas
running when the arc is broken at the end of the welding around the arc and the weld puddle, (c) arrangement
and also should not stick into the weld pool. for cooling by shielding gas or circulating water, and
In modern M I G / C 0 2 machines, these points have (d) a switch to start and stop the welding current, wire
been taken care of with the help of an electronic feed and shielding gas flow.
controller whose controls avoid: For semi-automatic welding, the gun is shaped like
• Burning back of contact tubes a pistol and is of goose-neck design, with a gun head
angle of 45°. The gun head is connected to a cable hose
• Loss of wire through over-run assembly, whose other end has a coupler for remote
• Loss of time spent for cutting of extra-long pro- connection to a wire feeder. A typical gas-cooled gun is
truded wire shown in Fig. 6.3.
• Loss of time spent for cutting the wire stuck into In another design, a pull-type wire-feed mechanism
the weld pool. and the wire spool are mounted directly on the gun.
The required features are brought about by carefully The spool has a diameter of 102 mm and weighs 0.5 to
sequencing the wire feed, the power source On-Off 1.0 kg. The wire diameters are usually 0.8-1.2 mm. In
function and gas On-Off in relation to gun switch this case, there is no limit for the length of the cable
triggering. The sequence adopted in the electronic connecting to the power source and the gas cylinder.
controller is as under: In yet another design, a self-contained wire feeding
mechanism is mounted on the gun, which pulls the wire
1) When the gun is switched On through a flexible cable from a remote spool. It may be
a) The gas, wire feed and the power source are used in conjunction with a normal wire feeder to make
made On simultaneously it a push-pull type feeding system. In this case, the length
b) As a result, the welding operation starts of the cable between the gun and the wire feeder can be
2) The On state continues till the gun switch is re- 15 m or more. Such a gun can conveniently operate with
leased, i.e. turned Off aluminium and steel wires, whose diameters range
between 0.8 and 1.6 mm.
3) When the gun switch is released The choice of gas-cooled or water-cooled guns is
a) The wire feed stops instantaneously, because based on several factors. For intermittent welding with
of special braking circuit provided for the C 0 2 gas shield, gas-cooled gun can be used for currents
wire-feed motor up to 500 amp. For continuous welding the limit is 400
b) Weld power is turned Off after a time delay Amps. If the shield is of argon or helium, the current is
of 0.1 to 0.2 sec. This power is turned Off by limited to 200 Amps. The other factors are the metals
means of a contactor put in the input side of being welded, weld joint design and the established
the power source shop practice. Gas-cooled guns are easier to manipulate
4) The shielding gas flow is also stopped along with than the water-cooled types and are preferred for
the weld power. positional welding and welding in confined areas.
WIRE FEED V////////J///////////A

WELD POWER 7////////XW////////A
GAS V////////////////M^^^^ TIME
TORCH SWITCH PRESSED ON TORCH SWITCH RELEASED OFF
Fig. 6.2: Operational sequence in MIG/C0 2 arc welding

GAS TUBE / ELECTRODE GUIDE JUBE

COMPOSITE
rfPOSITE<CABLE. \
' NOZZLE
\ GUN SWITCH
CURRENT CONTACT TUBE
WELDING ELECTRODE
Fig. 6.3: Typical semi-automatic gas-cooled, curved-neck gas metal-arc welding gun
Shielding Gas Control wire feeder, (3) welding gun and accessories, (4) set of
C 0 2 gas is normally supplied in cylinders. These control cables and welding cables with hoses for gas
cylinders contain liquid and gaseous carbon dioxide in and water (water only for capacities of over 500 Amps.),
equilibrium at high pressure. As C 0 2 gas is withdrawn and (5) set of accessories which include gas regulator-
from the space at the top of the cylinder, liquid C 0 2 cum-flowmeter, gas preheater, welding helmet, gloves
evaporates to replace the gas removed, thus maintaining and anti-spatter silicone spray.
the gas-liquid equilibrium. Evaporation of the liquid
Some of the salient features of the equipment are:
C 0 2 causes intense cooling of the gas coming out of the
a) Thyristorised DC variable speed drive system en-
cylinder. Also it is necessary to ensure that only gaseous
C 0 2 , and not the liquid, passes into the pressure sures smooth and precise wire feeding
reducing valve. For this purpose an electrically heated, b) Easy plug-in type connections facilitate quick in-
thermostatically controlled vaporiser or gas preheater terchangeability
is provided between the cylinder valve and the reducing c) Facility is provided for pre-selection of voltage
valve. and gas-flow rates before commencing the weld-
The pressure and flow of the shielding gas to the ing operation
welding gun is controlled by using a gas pressure d) All the control circuits are m o u n t e d on one
regulator and a flowmeter. The regulator reduces the printed circuit board for ease of maintenance
high pressure of the gas from the cylinder or any other
e) Automatic choke setting avoids the possibility of
source to a safe working pressure, and maintains a
selecting wrong choke values and ensures opti-
constant delivery pressure within the operating range
mum metal transfer with better dynamic charac-
of the gas. The flowmeter controls the rate of gas flow
teristics
to the welding gun. The flowmeter is calibrated in litres
per minute. The inlet gas pressure to the flowmeter must f) Burn back control is incorporated to ensure quick
be maintained as specified by the equipment supplier, and smooth arc-striking without wastage of filler
by setting the pressure regulator suitably. wire
Gaseous mixtures in standard proportions are often g) Protection against ventilation failure, a n d
available readily fiom gas suppliers. Sometimes gas- consequent overheating and d a m a g e to the
mixing units are used on the shop floor to mix the gases equipment, is provided. Where water-cooled
from individual cylinders in the desired proportion. system is used, a built-in water circulatory unit
is provided for effective torch cooling. It has extra
TYPICAL EQUIPMENT protection against p u m p failure or inadequate
Typical modern M I G / C 0 2 equipment as shown in water flow
Fig. 6.4 and 6.4 A consists of (1) constant potential h) Effective gas flow control is provided through a
rectifier-type power source and controls, (2) separable solenoid valve, which ensures adequate shield-
Fig. 6.4: ADORMIG welding outfit Fig. 6.4A: MIGMAG welding outfit
ing of the weld puddle and minimises wastage smooth welding arc. As seen in Fig. 6.4 it is ruggedly
of gas. constructed, drip-proof, and is provided with suitable
The equipment is available in four models A, #, C and fan for adequate cooling of power elements. The
D, whose brief specifications are given in Table 6.1. machine is mounted on rubber-type wheels and has a
mounting bracket for the gas cylinder along with lifting
Power Sources lugs for shifting from job to job.
All the models have constant-potential type welding
rectifier as basic power source giving excellent short arc Wire Feeder
and spray arc transfers of weld metal during the welding Typical wire feeder, shown in Fig. 6.5 contains a
operation. With their almost horizontal (constant suitable DC motor, speed reducing gear box, 2/4 roll
voltage) characteristics, they are well suited for fully- drive, gas solenoid valve and potentiometer for
automatic as well as semi-automatic MIG/C0 2 welding. adjustment of wire-feed speed. Its salient features are:
Accurate voltage adjustments can be made by way of a) DC motor of adequate power to cover the entire
fine and coarse control voltage selection switches. range of wire feeders to feed wires of different
Automatic choke incorporated in the machine provides sizes
Table 6.1: Brief specifications of MIG/C02 models

Model Qirrent range Wire sizes in mm
up to (Amps.) Steel Aluminium
A , 150 0.8 to 1.00

B 250 0.8 to 1.20 1.20
C 400 ' 0.8 to 1.60 1.20 to 1.60
D 600 0.8 to 2.00 150 to 1.60
Their detailed specifications are given in Table 6.2.
Table 6.2: Detailed specifications of MIG/C02 models

Model A B C D
Power source Type CPR CPR CPR CPR
Current (up to) amp 150 250 400 600
Max. DC voltage V 26 35 49 56
Welding current
@ 60% duty cycle amp 115 250 400 600
@ 100% duty cycle amp 90 195 310 500
Rating:
Continuous KVA 2.4 8 17 31
@ 60% duty cycle KVA 3 10 22.5 36.5
Max. KVA 4 — — —
Insulation Class F B B B
Cooling (power source) Type Natural air Forced air Forced air Forced air
Input supply 230V / lph / 50 450V / 3 ph / 50 415V / 3ph / 50 415V / 3ph / 50
Wire Feeder Model E F F G
Drive system Type Planetary 2-roll 2-roll 4-roll
Cooling (gun) Type Gas Gas Gas Water
Fig. 6.5: Wire-feeder Model VCG-B-201 Fig. 6.5A: Wire-feeder
b) Includes two-track/four-track systems for wire Specifications of the three models are given in Table
feeding 6.3.
c) Continuously variable wire-feed speeds
Guns
d) Vertical spool position ensures a smooth wire-
These are available both in gas-cooled and water-
feed and avoids bird-nest phenomenon
cooled versions. The gun head is connected to a cable
e) Optional CO, spot welding attachment can be hose assembly. All models are fitted with a suitable
plugged in for spot welding operations coupler for connection to the right wire feeder. The guns
f) Suitable for .mounting on welding booms or for are supplied with requisite accessories. Special guns
use in conjunction with various welding aids with straight heads or those suitable for automation
with/without minor modifications. can also be supplied. They are also available with
Table 6.3: Specifications of wire feeders

Model E F G
Wire diameters:
Steel mm 0.8 to 1.2 0.8 to 1.6 0.8 to 2.0
Aluminium mm — 1.2 to 1.6 1.2 to 1.6
Wire drive motor Type DC DC DC
Drive system Type Planetary 2-Roll 4-Roll
Wire-feed speed control Type Thyristorised Thyristorised Thyristorised
Wire-feed speed m / min 1.2 to 18 1.2 to 18 1.2 to 18
suitable spot welding nozzles and soft wire/aluminium c) The gun is kept off the floor. This increases its life
welding attachments. and reduces the risk of its getting damaged. Also
there are fewer chances of accidents due to the
SPECIAL GUN HOLDER gun trailing on the ground.
Where MIG/C0 2 welding is preformed continuously d) Trouble-free wire feeding, as the gun is kept rela-
on a production scale, a special gun holder provided tively straight.
with a spring-loaded support can be conveniently used
e) Wider area of coverage, because of the swivel-
to take away the strain of the gun from the welder's
ling mechanism.
arm. It is designed to be used in conjunction with the
wire feeder. The arm of the gun holder swivels along
with the wire feeder through 360° to increase the coverage AUTOMATIC WELDING HEAD
of working area. The arm joint is spring-loaded and the A typical outfit consists of a water-cooled welding
tension can be adjusted for optimum operation. While in gun and a wire feeder, the former being mounted directly
use, the arm of the gun holder moves down as the welder below the latter. Welding currents up to 1,200 amp and
manipulates the gun (see Fig. 6.6). long arc times are used. The electrode wire is driven by
Following are the advantages of using this gun holder: feed rolls into a guide assembly, which supports and
a) Fatigue to the welder is reduced since the weight protects the electrode during the welding operation.
of the gun is taken away by the holder. This re- Electrode diameters vary from 0.8 to 6.4 mm.
sults in increased productivity. The welding head may be-mounted on a trolley and
made to traverse a track running parallel to the joint.
b) Facilitates easy manipulation of the gun.
Alternatively, it may be mounted on a beam and
operated in conjunction with specially built positioners
and fixtures. For example, by mounting the head on a
column and boom, longitudinal arid circumferential
joints of cylindrical vessels can be automatically welded
as shown in Fig. 6.7.
COMBINED SA-MIG/C02 WELDING HEAD

Since submerged arc and MIG/C0 2 welding
processes are best suited for high-speed mechanised
welding, fabricators sometimes desire to have and
inquire for a combined welding head for these two
processes. Fabricators must understand that such a
welding head cannot be conceived, because (a) wire-
feed speed range and torque requirements are different
for these two processes, and, (b) while the SA process
needs flux feeding and flux dispenser arrangements,
MIG/C0 2 demands gas shielding and water-cooling
Fig. 6.6: Gun holder with wire feeder mounted on it. arrangements.
CIRCUMFERENTIAL WELDS
FROM OUTSIDE:
• BOOM IN FIXED POSITION

\M
LONGITUDINAL WELDS
FROM OUTSIDE:
• FIXED WORKPIECE
CIRCUMFERENTIAL WELDS
FROM INSIDE:
• BOOM IN FIXED POSITION
LONGITUDINAL WELDS
FROM INSIDE:
•FIXED WORKPIECE
• BOOM MOVING AT WELDING SPEED • WORKPIECE ROTATING • BOOM MOVING AT WELDING
• WORKPIECE ROTATING
O N ROTATOR SPEED
ON ROTATOR
Fig. 6.7: Automatic MIG/CO. head mounted on column and boom
At best, the equipment manufacturer can supply a

universal-type power source, carriage unit and guide
track as common items; and separate welding heads MOLTEN
which can be interchanged as required. METAL
Some internationally known companies supply the
SA welding head and a supplementary kit consisting
of a motorgear box unit to match MIG/C02 wire-feed
speeds and other substitution parts to enable the
fabricator to convert the SA head to a MIG/C02 head.
Such an arrangement cannot be a true combined
welding head.
Fig. 6.8: Non-axial metal transfer
METAL TRANSFER
In the MIG/C02 process, the metal transfer from the
electrode tip to the weld pool across the arc is either the molten drop. The electric current flowing through
globular, spray type or short-circuiting type depending on: the electrode gives rise to several electromagnetic forces
a) the magnitude of welding current that act on the molten tip including the pinch force (P)
b) current density and the anode reaction force (R) (see Fig. 6.8). The pinch
c) shielding gas force which increases with current and electrode
diameter causes the drop to detach. With C0 2 shielding,
d) electrode extension and the electrode tip is not heated directly by the arc plasma
e) electrode chemistry. but by the arc heat conducted through the molten drop.
The molten drop grows in size and finally detaches by
Globular Transfer short-circuiting or gravity, after having overcome the
With carbon-dioxide shielding, the transfer is force R, which tends to support the drop.
globular and non-axial, whatever may be the value of
the welding current, current density and other factors. Spray Transfer
Hence there is considerable spatter. Drops become With pure argon or a mixture of C0 2 with at least
smaller in size as the current increases and they continue 80% argon or helium, the metal transfer is globular at
to be directed axially and non-axially. relatively low current density, but changes to spray type
Axial transfer means that the metal droplets move when the current density exceeds a certain value
along a line that is an extension of the longitudinal dxis depending on the wire electrodetypeand shielding gas.
of the electrode. Non-axial transfer means that the Table 6.4 indicates transition currents for various
droplets are hurled in any other direction. electrode types.
The non-axial transfer is caused by an In the spray-type transfer, the metal travels across
electromagnetic repulsive force acting on the bottom of the arc in the form of fine droplets which are axially
Table 6.4: Globular-to-spray transition currents for a variety of electrodes

Wire electrode
diameter Minimum spray arc
Wire electrode type (mm) Shielding gas current, amp
Mild steel 0.76 98% argon-2% oxygen 150
Stainless steel 0.89 99% argon-1% oxygen 170
Aluminium 0.76 Argon 95
Deoxidized copper 0.89 Argon 180
Silicon bronze 0.89 Argon 165
directed. The droplet diameter is equal to or less than In the dip transfer technique, the low arc voltage results
the electrode diameter. The arc column is fine and the in a reduced arc length and at low currents used, the
wire tip becomes pointed. The spray transfer is induced molten metal droplet gets transferred into the weld pool
by the electromagnetic forces acting on the molten by direct contact. Arc voltage has an important bearing
electrode tip as well as by the arc plasma. The pinch on metal transfer: its value is suitably chosen to give the
force P plays its part by squeezing the molten metal highest short-ciraiiting frequency, so that one gets smooth
formed at the electrode tip and pushing it towards the and steady metal transfer with negligible spatter.
weld pool. Metal vapour generated at the electrode tip Oscillographic analysis of the dip transfer arc has
also influences spray transfer. shown that the voltage reaches zero at each short-circuit
MIG/ C0 2 welding with spray-type arc is best suited and the current rises during this time to a value some
for welding relatively thick parts, because of the high two to four times greater than the arc current. At the
welding currents it demands. peak current, the short-circuit bridge is broken, the arc
re-ignites and the voltage rises to arc voltage with a
Short-circuiting or Dip Transfer reduced current (100 amp). This is a repetitive sequence
Normal C0 2 welding which is extensively used in and occurs maybe 200 times a second depending on the
industry is a high current density process (e.g. 350 amp wire size, other welding parameters and type of gas used
for 1.6 mm diameter wire), which gives a penetrating (see Fig. 6.9).
spray-type arc with free flight transfer of metal. It is For optimum performance, when the wire has short-
well suited for welding thick steel plates in the circuited with the weld pool, the short-circuited system
downhand position. should be able to draw the correct amount of current at
However, its penetrating arc and large sized molten the appropriate rate to enable the short-circuit to clear
weld pool makes it unsuitable for welding thin sheets the bridging wire and re-establish the arc. For the
and for all-position welding. For these applications, a average 1.2 mm wire, this amounts to about 300 amp in
modified technique for C0 2 welding called short- 0.005 sec. This is achieved in the case of a constant
circuiting arc or short arc or dip transfer is used. potential power source by making available an
Dip transfer technique uses small diameter wire additional inductance or choke to suitably alter its
(0.8-1.6 mm), low arc voltage (16-22 V) and low current dynamic characteristic.
(60-180 amp). While typical parameters for spray-type Selection of the correct choke is of utmost importance
arc welding are 360 amp, 34 V and 1.6 mm wire, for in dip transfer welding. Too little choke gives a fast rise
short-circuiting arc they are 120 amp, 19-21 V, 1.2 mm of current and causes explosive clearance of short-circuit
wire. called spatter. Too much choke prevents the short-circuit
ZERO
-ARCING PERIOD—*
z
o
Fig. 6.9: Oscillograms of short-circuit transfer
from clearing and the result is that the wire stubs into helium is sometimes preferred to argon for the welding
the plate. In practice, a good operator can arrive at the of thick sections of high conductivity metals such as
correct choke setting by the feel, sound and appearance aluminium, magnesium and copper alloys. Helium has
of the arc and weld pool. A good choke setting lies higher thermal conductivity than argon and hence, for
midway between the minimum, to prevent the a given welding current and arc length, it gives higher
formation of spatter and the maximum which still arc voltage and higher heat input. A disadvantage with
permits an easy arc characteristic. pure helium is that it gives a wide bead with wide
Optimum short-circuiting arc operating ranges for reinforcement as depicted in Fig. 6.10, wherein bead
three wire sizes are given in the following table: contours obtained with various shielding gases are
shown. This is overcome by using an argon-helium
Parameter Wire size (mm) mixture containing 50-80% helium. Also one can obtain
0.8 1.2 1.6 spray transfer more readily with argon shielding than
with helium shielding.
Wire-feed rate m/min 5 2.5 1.7 In replacing argon with helium, it must be
Arc Voltage, V 18 19 20 remembered that argon is nearly ten times heavier than
Average current, Amps. 100 120 160 the latter. Therefore, while argon forms a natural blanket
Response rate, kA/sec 50-150 40-130 20-75 over the weld area after leaving the welding nozzle,
helium rises in a turbulent manner and tends to disperse
The range of response rate (kA/sec) shown above is
obtained by the use of a variable choke, as mentioned Wire size Wire-feed speed
earlier, and this is also used to control the heat of the mm m/min
welding conditions. Thus a 40 kA/sec rise would be
0.8 2.5-7.5 (60-160 amp)
suitable with 1.2 mm diameter wire for welding thick
1.2 2.0-4.7 (100-175 amp)
plate in position, 60 kA/sec for the root run of a pipe
1.6 1.5-2.0 (120-180 amp)
joint in the field, and 130 kA/sec is the requirement both
for welding sheet and for the fast response necessary
for starting the arc at the beginning of a run.
Operating ranges of wire-feed rate are given in Table
6.10. The current is shown in brackets. POSITIVE ELECTRODE
ARGON-0 2 ARGON HELIUM-ARGON HELIUM

SHIELDING GASES
Fig. 6.10: Bead contour and penetration patterns for
Pure 100% argon is used for almost all metals except various shielding gases
steels. In countries where helium is available, pure 100%
into the air, because it is lighter than air. Therefore higher from element to element, depending on how readily it
flow rates are required for helium than for argon combines with oxygen. Table 6.5 gives the transfer
shielding. efficiency expressed in % recovery, which is calculated
Pure nitrogen or nitrogen mixed with argon has been from the formula:
used in Europe for welding copper, because it gives
greater heat input like helium and is inexpensive. % element in weld metal
When pure argon, pure helium or an argon-helium % recovery = x 100
mixture is used for welding ferrous metals, the flow of % element in original wire
the molten filler metal along the edges is irregular,
sometimes causing undercut, and the metal transfer is
erratic and results in considerable spatter. Small addi- Table 6.5: Transfer efficiency of alloying elements in C02
tion of oxygen or carbon dioxide to argon (usually 1- shielding
5% 0 2 or 3-10% C02) overcomes these troubles. It helps C* Al Zr Ti Si Mn Nb
to achieve axially directed spray transfer, promotes the
50-235 30-40 30-40 40 50-70 60-75 60-70
wetting and flow of the weld metal, eliminates undercut
and porosity and reduces bead penetration. Oxygen or Cr Mo Ni Cu P&S
carbon dioxide additions are also made to helium or 90-95 95-100 95 95-100 100
argon-helium mixtures, but such mixtures are used for
short-circuiting or pulsed DC techniques. * With 0.30% carbon wire, carbon in the weld metal is 0.15%.
With 0.055% C in a S.S. wire, C in the weld metal is 0.13%.
It must be remembered that the addition of 0 2 or
C0 2 to argon or helium causes the shielding gas to be
In welding with short-circuiting transfer technique,
oxidising. The use of this gas mixture may cause
special gas mixtures may be used. For example, argon-
porosity in some ferrous metals, unless suitable
deoxidisers are incorporated in the electrode. The C0 2 mixtures containing 20-50% C02 are often used to
absence of deoxidisers in alloy steel electrodes may also weld mild steel. They perform satisfactorily also for
cause some loss of alloying elements such as silicon, welding stainless steels, but they may raise the carbon
manganese, chromium and vanadium. content of the weld metal and reduce its corrosion
resistance. Hence a less reactive mixture of 90%
Carbon dioxide is widely used for the welding of
He—7.5% Ar—2.5% C0 2 is generally preferred for
mild steel, because besides being inexpensive, it gives
stainless steels in the U.S.A. For low-alloy steels,
sound weld deposits with adequate mechanical
mixtures of 60 to 70% He—25 to 35% Ar-î to 5% C0 2
properties at high welding speeds and with good joint
are used and they give welds of good impact properties.
penetration. The drawbacks of C0 2 shielding are
globular or non-axially directed spray transfer, harsh Information on the choice of shielding gases is given in
arc and considerable spatter. In C02 welding, an Table 6.6.
occasional small short-circuit current pulse is
immediately followed by a high current peak, which is PROCESS PARAMETERS
too high for the wire size being used; the resultant The following parameters must be considered in
explosive disruption of the short-circuit wire bridge drawing up a MIG/C0 2 welding procedure:
gives rise to spatter particles. Spatter can be kept 1) Electrode size
reasonably low by using a very short, uniform arc length 2) Welding current
and maintaining the electrode tip below the adjacent 3) Arc voltage
work surface.
4) Arc travel speed
Carbon dioxide becomes oxidising in the arc due to
decomposition into carbon monoxide and oxygen. On 5) Electrode extension
the basis of Mn and Si losses in the arc, a C0 2 shield 6) Electrode position
may be regarded as equivalent to a shield of inert gas 7) Welding position
containing about 10% oxygen. To ensure porosity-free
welds, the electrode must contain the appropriate 1) Electrode Size
balance of deoxidisers (such as Al, Ti, Si, Zr). Commonly used electrode sizes are (mm): 0.8, 1.0,
When alloy-steel electrodes are used with C0 2 1.2, 1.6 and 2.4. Each size has a usable current range
shielding, the transfer efficiency across the arc varies depending on wire composition and whether spray-type
Table 6.6: Gas selection for gas metal arc welding
Metal type Thickness Transfer mode Recommended shielding Advantages/description

gas
Carbon Steel Up to 14 gauge Short Circuit Argon + C0 2 Good penetration and

C + C0 2 +0 2 distortion control to reduce
potential melt-through.
14 gauge -1 /8 in Short Circuit Argon + 8 to 25% C0 2 Higher deposition rates without
Argon + He + C0 2 melt-through. Minimum distortion

and spatter. Good weld control for
out-of-position welding.
More than 1/8 in. Short Circuit Carbon Dioxide High welding speeds. Good
Argon + 15 to 25% C0 2 penetration and pool control.
Applicable for out-of-position welds.
Short Circuit Argon + 25% C0 2 Suitable for high current and

Globular high-speed welding.
Short Circuit Argon + 50% C0 2 Deep penetration; low spatter; high

travel speeds. Good out-of-position
welding.
Short Circuit Carbon Dioxide Deep penetration and fastest travel
Globular (buried speeds but with higher melt-through
arc) potential. High current mechanized
welding.
Spray Transfer Argon + 1 to 8% 0 2 Good arc stability; produces a more

fluid weld pool as 0 2 increases;
good coal-essence and bead contour.
Good weld appearance and weld pool control.
Spray Transfer Argon +5 to 20% C0 2 Fluid weld pool and oxidizing to weld
metal causes higher amounts of slag
and scale as C0 2 increases. Good arc
stability. Weld soundness and
increasing width of fusion.
Short Circuit Argon +COz+ 0 2 Applicable to both short circuiting
Spray Transfer Argon + He + CO, and spray transfer
modes. Has wide welding current range and
good arc performance. Weld pool has good
control which results in improved
weld contour.
High Current Argon +He + C02+ 0 2 Used for high deposition rate
Density Rotational Argon + C0 2 + 0 2 welding where 15 to 30 lb/h (7 to
14 kg/h) is typical special welding
equipment and techniques are some
times required to achieve these
deposition levels.
More than 14 Pulsed Spray Argon + 2 to 8 % 0 2 Used for both light-gauge and heavy
gauge Argon +5 to 20% C0 2 out-of-position weldments. Achieves
Argon +C02 + 0 2 good pulse spray stability over a wide
Argon + He + C0 2 range of arc characteristics and
deposition ranges.
Low- and- Up to 3/32 in. Short Circuit Argon +8 to 20% C0 2 Good coalescence and bead contour.
High-Alloy Helium + Ar + C0 2 Good mechanical properties.
Steel Argon +C02 + 0 2
Short Circuit Argon + 20% to 50% C0 2 High welding speeds. Good

Globular penetration and weld pool control.
Applicable for out-of-position welds.
Suitable for high current and
high-speed welding.
More than 3/32 in. Spray Transfer Argon + 2% Oz Reduces undercutting. Higher
(High Current Argon + 5 to 10% C0 2 deposition rates and improved bead
Density & Argon + C0 2 + 0 2 wetting. Deep penetration and good
Rotational) Argon + He + C0 2 + Oz mechanical properties,
Pulsed Spray Argon + 2% 0 2 Used for both light-gauge and heavy

Argon + 5 C0 2 out-of-position weldments. Achieves
Argon + C0 2 + 0 2 good pulse spray stability over a wide
Argon + He + C0 2 range of arc characteristics and
deposition ranges.
Steel, Upto 14 gauge Short Circuit Argon + 2% to 5% 0 2 Good control of melt-through and
Stainless, distortion. Used also for spray arc z-welding.
Nickel Weld pool fluidity sometimes
Nickel Alloys sluggish depending on the base alloy.
Table 6.6 Contd...

CO
Table 6.6 (Contd.) 00
o
o
m
Metal type Thickness Transfer mode Recommended shielding Advantages/description 2)
gas z
>
31
Steel, More than 14 Short Circuit Helium + 7.5 Ar + 2.5 CO. Low C0 2 percentage in He mix o
Stainless, gauge minimizes carbon pick-up, which can
m
z
Nickel cause intergranular corrosion with some ©
Nickel Alloys alloys. Helium improves wetting action H
m
and contour C0 2 percentages above o
i
5% should be used with caution on some
alloy. Applicable for all-position welding.
Spray Transfer Argon + 1 to 2% 0 2 Good arc stability. Produces a fluid but
Argon +He + C0 2 controllable weld pool; good
Helium + Ar + CO. coalescence and bead contour.
Minimizes undercutting on heavier thickness.
Stainless Steel More than 14 Pulsed Spray Argon + 1 to 2% 0 2 Used for both light-gauge and heavy
gauge Argon + He + C0 2 out-of-position weldments. Achieves
Helium + Ar + C0 2 good pulse spray stability over a wide
Argon + C0 2 + H2 range of arc characteristics and
deposition ranges.
Aluminium Upto Vi in. Spray Transfer Argon Best metal; transfer, arc stability and
Pulsed Spray plate cleaning. Little or no spatter.
Remove oxides when used with DCEP
(reverse polarity).
More than l/i in. Spray Transfer Helium + 20 to 50% Argon High heat input. Produces fluid weld
Pulsed Spray Argon + Helium pool, flat bead contour and deep
penetration. Minimises porosity.
arc or short-circuiting arc is used. Generally stated, the 2) Welding Current

electrode size increases as the base metal thickness The current within the usable range is selected
increases. However, the selection of electrode size is according to the electrode size, type of metal transfer
important and should be made with care, depending and the base metal thickness. When the current is too
on the particular application. For a given current, a low, the molten metal fails to wet the joint surface and
change to a smaller electrode size results in a faster causes lack of fusion. When the current is too high, it
droplet transfer rate and hence a higher melting rate gives rise to spatter, porosity and poor weld bead profile.
due to an increase in the current density. General data on current selection are given in Table 6.7.
This fact is made use of, for example, in multi-pass With a constant-voltage power source, the welding
mechanised welding, where for intermediate passes a current increases as the electrode-feed rate is increased,
high droplet rate at fairly low current ensures an and decreases as the electrode speed is decreased,
acceptable weld profile. A 1.2 mm size electrode is provided all the other variables are kept constant.
chosen here in preference to a 2.4 mm size electrode. Operating ranges of wire-feed rate in C0 2 welding are
The increased I2i?-heating effect due to the smaller given in Table 6.8.
electrode size also favours the weld deposit. On the other The upper limits of wire-feed rate for 0.8 and 1.2 mm
hand, for mechanised welding of a plate girder in the electrodes are dictated by the maximum available wire-
downhand position, a 2.4 mm size can be used at very feed speed of the machine, which is around 15-18 m/
high amperage to obtain heavy deposition rates. min. For larger electrodes, the upper limits of wire-feed
The 2.4 mm size is avoided for semi-automatic speed are imposed by the high arc force at the maximum
welding, because it is difficult to feed in manual guns. current, which results in a badly shaped weld and
The most versatile size is 1.2 mm. General data on excessive spatter.
electrode size and other variables in mild steel welding With increasing current, the melting rate of the
are provided in Table 6.7. electrode increases. The relations between these factors
Table 6.7: General guide to welding conditions

Arc voltage 0.8 mm dia. 1.2 mm dia. 1.6 mm dia. 2.4 mm dia.
V wire wire wire wire
Low penetration for Good control over Good fusion—
sheet root penetration and vertical plate
fusion
16-22
(Short-circuiting)
arc) ~T 60-160 amp
n
100-175
! = .amp
i
LU
&
120-180 amp
/ Not normally used
(150-200 amp)
Fast mechanised Fast mechanised H.V. Downhand
24-28
weld on corner
n mechanised butt weld
Not normally used
(Semi-shorting arc)
15(£550amp 200-300 amp
C§D
250-350 amp
(300-400 amp)
Good penetration Multi-pass Deep penetration Deep penetration,

with 200 amp set mechanized weld downhand high deposition rate,
3
ft It
thick plate
PL
30-45
(Spray type arc)
150-250 amp 200-350 amp
QD
300-500 amp
^
500-750 amp
are shown in Fig. 6.11 for four common sizes of mild For a given electrode size, increasing the amperage
steel electrodes. For each electrode size, the curve is within the usable range results in (a) increased depth
nearly linear at the lower current range. However, at and width of weld penetration, (b) increased deposition
higher currents and especially with smaller electrode rate, and (c) increased size of weld bead, provided all
sizes, the melting rate curve becomes non-linear and the other variables are kept constant.
progressively increases at a higher rate as the welding
amperage increases. This change is due to the I ^-heat- 3) Arc Voltage
ing of the electrode extension beyond the contact tube. This is a very important variable in MIG/ C 0 2
welding, mainly because it determines the type of metal
transfer by influencing the rate of droplet transfer across
the arc. The arc voltage to be used depends on base metal
thickness, type of joint, electrode composition and size,
shielding gas composition, welding position, type of
weld and other factors. Typical values for welding
various metals by drop transfer (i.e. by globular or
spray) and short-circuiting transfer using various
shielding gas compositions are given in Table 6.9, while
the arc voltage ranges for typical applications have been
indicated earlier in Table 6.7.
WELDING CURRENT (DCRP). AMP To arrive at a precise value of arc voltage for a
Fig 6 11: Melting rate of carbon-steel electrodes welding job, one is advised to carry out welding trials,
because no specific values are appropriate for all
Table 6.8: Ranges of wire-feed rate in C0 2 welding
Wire-feed speed, m/min

(current is shown in parentheses)
Wire dia. (mm) Spray-type arcs (30-45 V) Short-circuiting arcs (16-22 V)
0.8 5.0-15 (150-250 amp) 2.5-7.5 (60-160 amp)
1.2 5.0-15 (200-350 amp) 2.0-3.8 (100-175 amp)
1.6 5.0-8.8 (350-500 amp) 1.5-2.0 (120-180 amp)
2.4 3.8-7.5 (500-750 amp) 1.25-1.6 (150-200 amp)
Table 6.9: Typical arc voltages for MIG/C0 2 welding of various metals
Drop transfer Short-circuiting transfer

1.6 mm (1/16 in.) diameter electrode 0.9 mm (0.035 in.) diameter electrode
25% A r - Ar-02 Ar-02 75% Ar-
Metal Argon Helium 75% He (1-5% 0 2 ) co2 Argon (1.5% 0 2 ) 25% C 0 2 co2
Aluminium 25 30 29 — — 19 — — —
Magnesium 26 — 28 — — 16 — — —
Carbon steel — — — 28 30 17 18 19 20
Low-alloy steel — — — 28 30 17 18 19 20
Stainless steel 24 — — 26 — 18 19 21 —
Nickel 26 30 28 — — 22 — — —
Nickel-copper alloy 26 30 28 — — 22 — — —
Nickel-chromium-iron alloy 26 30 28 — — 22 — — —
Copper 30 36 33 — — 24 22 — —
Copper-nickel alloy 28 32 30 — — 23 — — —
Silicon bronze 28 32 30 28 — 23 — — —
Aluminium bronze 28 32 30 — — 23 — — —
Phosphor bronze 28 32 30 23 — 23 — — —
METAL INERT-GAS/C0 2 ARC WELDING 331
applications. Starting with a certain value of arc voltage, undercutting occurs along the edges of the weld bead,
any further increase tends to flatten the weld bead and because the deposition of filler metal is not sufficient to
further widen the fusion zone. A decrease makes the fill the path melted by the arc.
weld narrower with a pronounced reinforcement and
deeper penetration. When the value is much above the 5) Electrode Extension
usable range, porosity, spatter and undercutting may Electrode extension (also called stick-out) is the
be encountered; when it is too low the result can be distance between the end of the contact tube and the tip
porosity and overlap at the weld edges. of the electrode as shown in Fig. 6.12, The figure also
As indicated in Table 6.7, the low-voltage range of shows nozzle-to-work distance and contact tube-to-
16-22 V is used for all-positional C0 2 welding of work distance, which also have to be taken care of. As
relatively thin steel sheets using a short-circuiting arc the electrode extension is increased, the electrical
and small size electrode resulting in minimum resistance of the electrode and hence its PR- heating
penetration. The high range of 30-45 V is mostly used increases. The result is that less amperage is necessary
for automatic spray-type C0 2 welding of thick sections to melt the electrode at a given wire-feed rate. The
in the flat position, using largest size electrodes and high electrode extension is particularly important in welding
currents to achieve deep penetration and high stainless steels, where a small variation can vary the
deposition rates. The intermediate range of 24-30 V current appreciably for an otherwise similar welding
gives the characteristics of both the spray and short- procedure. This is shown in Fig. 6.13, where the current
GAS
NOZZLE
ONTACT TUBE
T
CONTACT TUBE
NOZZLE TO
TO WORK
WORK DISTANCE
i
DISTANCE
ARC LENGTH
WORK PIECE
Fig. 6.12: Electrode extension and related terms
circuiting arcs and is used for semi-automatic and

automatic welding of medium thicknesses. 250
"1—I—r—
1.2 MM DIAMETER WIRE
4) Arc Travel Speed 200
<
The linear rate (expressed in cm/min or mm/sec) at f
i~
which the arc moves along the joint, termed arc travel z 150
speed, affects weld bead size and penetration. With AC
other variables kept constant, there is a certain value of D
O
travel speed at which the weld penetration is maximum. ioa
2
If this value is lowered, the weld pool becomes larger 9
and shallower and the advancing arc impinges on this 50
weld pool rather than on the base metal. The result is a
shallower penetration and a wider weld bead. As the 3.15 64 95 12.7 15.9 19 22.2
travel speed is increased, the heat input rate of the arc ELECTRODE EXTENSlONÊ (MM)
is decreased. The result is decreased penetration and
Fig. 6.13: Relation between electrode extension and current
narrower weld bead. When the travel speed is excessive,
drawn by 0.8 mm and 1.2 mm electrodes at electrode and a narrower, more convex weld bead with deep
extensions of 6.4 to 19 mm can differ by as much as 40 penetration. Perpendicular technique is used more in
amp in 100 amp at a constant wire-feed speed of 5 m / automatic welding, and avoided in the semi-automatic
min and 18 V arc voltage. mode, because the end of the gas nozzle restricts the
Too long an electrode extension results in excess operator's view of the weld pool.
weld metal being deposited at low arc heat, giving rise
to a badly shaped weld and shallow penetration. The 7) Welding Position
arc stability is also affected. Such effects are noticeable As mentioned earlier, spray type technique is
while welding deep grooves, which are narrow enough restricted to making butt and fillet welds in the
to limit the access of the nozzle. When the electrode downhand position. When used for horizontal fillet
extension is too short, excessive spatter gets deposited welds, this technique tends to give unequal leg lengths,
on the nozzle, which can restrict the shielding gas flow convexity and undercutting. For welding thin sheets in
and cause porosity in the weld. Recommended range the downhand position, the penetrating arc of the spray-
for electrode extension is 6 to 13 mm for a short- type technique may give weld burn-through. Hence the
circuiting arc and 13 to 25 mm for drop-transfer arc. short- circuiting technique or spray transfer with pulsed
Generally the nozzle-to-work distance should be DC is preferred for such jobs.
between 16 and 19 mm. Vertical and overhead welding are usually carried
out with electrode diameters of 1.2 mm and under, using
6) Electrode Position short-circuiting metal transfer or spray transfer with
As with all arc welding processes, the position of pulsed DC.
the gun and the electrode with respect to the joint affects
the weld bead shape and penetration in a much larger WELDING PROCEDURES
measure than arc voltage or arc travel speed. The gun
angle is usually maintained within 10 to 20° on either As with other arc welding procedures, a good MIG/
side of the vertical. Depending on which way the gun is C0 2 welding procedure starts with correct edge
inclined, the technique is referred to as forehand and preparation and joint fit-up. The joint surfaces must be
backhand. The various electrode positions and free from rust, scale, grease, oil, paint and other foreign
techniques and their effects are shown in Fig. 6.14. It is materials. For making full penetration joints by welding
observed that as the electrode is changed from the with spray transfer technique from both sides, it is
perpendicular to the forehand technique, the weld bead necessary to gouge out the root from the second side
becomes shallower and wider and has less penetration. before starting to weld that side. When welding is done
Backhand technique gives a more stable arc, less spatter only from one side, suitable weld backing must be
provided. Sometimes weld backing can be avoided by
making the root pass with the short-circuiting technique
to obtain uniform penetration and depositing the fill up
w passes by high current spray transfer technique.

The welding equipment must be assembled and the
welding parameters set according to the manufacturer's
, DIRECTION OF WELDING instructions. All gas and water connections must be
absolutely leak-proof. If the shielding gas gets
contaminated with air or water, the arc becomes erratic
and pores appear on the weld.
The gun nozzle size and the shielding gas flow rate
must be correctly set according to the material being
welded and its joint design. Some joint designs demand
longer nozzle-to-work distance than normal; in such
| <QH> 1 cases one must use higher gas flow rates than those
FOREHAND TORCH BACKHAND recommended by the equipment manufacturer or
TECHNIQUE PERPENDICULAR TECHNIQUE specified in standard procedures, and a gas nozzle of
Fig. 6.14: Electrode angle and its effects on weld bead adequate size to cover the welding area. On the other
hand, smaller nozzle sizes may be used for welding in
Tfeble 6.10: Guidance on MIG/C02 welding procedure
Arc type Typical conditions and Procedure

applications
Spray-type arc 360 amp, 34 V, 1.6 mm wire. 1) Set open-circuit voltage to a little above the
Downhand welding of plate required arc voltage; e.g. 38 V.
2) Set wire-feed speed* to the recommended value for
the electrode size and material, e.g. 5 m/min.
Short-circuiting arc 120 amp, 19 V, 1.2 mm wire. 1) Set open-circuit voltage to a little above the re-
Positional welding of sheet and quired arc voltage, e.g. 20 V.
plate 2) Set wire-feed speed* to the recommended value for
the electrode size and material, e.g. 2.5 m / min.
3) Set choke (tune the circuit) to get required crisp-
ness and heat of arc.
*The wire-feed speed determines the welding current
confined areas or in the root of a thick joint. The For example, when using the short-circuiting arc, the
electrode-feed rolls and the contact tube must be contact tube may extend 3 mm beyond the end of the
compatible with the size and composition of the nozzle. Further guidance on procedures using a constant-
electrode, as recommended by the manufacturer. If the voltage power source is given in Table 6.10.
contact tube is worn out in usage, it must be replaced Following the settings of Table 6.10, trial bead welds
before the gun starts getting heated due to bad electrical should be deposited to arrive at correct arc voltage and
contact between it and the electrode. the electrode-feed rate (current). In the short-circuiting
Electrode extension is dictated mainly by whether procedure, the choke should be finally adjusted to obtain
short-circuiting or spray-type welding is being done. It good arc start and a stable arc with minimum spatter.
is set by adjusting the distance between the end of the
contact tube and the gas nozzle opening, which is PROCEDURE DATA
between 6.4 and 9.5 mm for normal spray-type welding The following data (taken from C02 Welding by A.A.
(see Fig. 6.12). In special applications, the contact tube Smith) should be used as a guide to arrive at precise
may be flush with or protruding from the gas nozzle. welding procedures.
BUTT WELDS
Wire-feed Arc travel
Plate t Joint Current Voltage rate speed
mm type amp V m/min cm/min
1) With 0.8 mm wire
Flat position
1.6
32 95 16-20 4.6 37
3.0 1 140 19-23 8.6 76

s 140 19-23 8.6 35
8.0 1 210 24-28 15.5 33

s 210 24-28 15.6 64
12.5 1&2 210 24-28 15.6 46

S 210 24-28 15.6 64
HV position
1.6 130 16-20 7.6 62
1 '\P I
3.0 1 180 20-24 12.3 109
s 180 20-24 12.3 50
8.0 &2 205 22-28 15.0 56

S 205 22-28 15.0 61
12.5
rr-2-n -A 205 22-28 15.0 44
Ts 3-3
S 205 22-28 15.0 61
1.5
Vertical position
1.6 115 17-21 6.3 51
y60.
3.0 1 165 19-23 10.7 80
W/^-l-P s 165 19-23 10.7 44
P^l
60 20-24
8.0 1 185 13 30
2 140 19-24 8.6 29
1.5 S 185 20-24 13 53
^ T
12.5 1 185 20-24 13 30

2&3 140 19-24 8.6 12
S 185 20-24 13 53
Overhead position
1.6 165 19-23 10.7 92
ifer
I ,<•> I
n
3.0 2 JO 165 9-23 10.7 26.7
^50*^
8.0 1&2 165 19-23 10.7 17.3
.2.0
7K
*• 5(M
12.5 1-3 165 19-23 10.7 13.2

2.0
2) With 1.2 mm wire

Rat position
1 155 18-22 3.2 57

S 155 18-22 3.2 22
^
«fc
T. 5 .°:> 1 205 22-26 5.1 31

s 205 22-26 5.1 35
*tfi«
*y-
12.5 1-3 245 25-29 7.0 42

S 245 25-29 7.0 48
20 -5 305 28-32 10.0 35

S 305 28-32 10.0 48
HV position
3 1 160 18-22 3.5 88
&
s 160 18-22 3.5 24
■ ^
I 195 22-26 4.7 23

.60 s 195 22-26 4.7 32
s^rY<-*-3.o
12.5 1-3 245 25-29 7.0 48

S 245 25-29 7.0 48
y 3 V * ^ \ \ 30°
20 Similar to above, 1-7 245 25-29 7.0 38

but 7 passes in S 245 25-29 7.0 48
the groove
Vertical position
3 -4 IM.S 1 160 17-21 3.5 49

~'A
I=J3^ s 160 17-21 3.5 24
1&2 160 17-21 3.5 14

S 160 17-21 3.5 24
12.5 Similar to above, 1-3 160 17-21 3.5 10

but 3 passes in S 160 17-21 3.5 24
the groove
20 Similar to above, 1-3 160 17-21 3.5 4.4

but 3 passes in S 160 17-21 3.5 24
the groove
METAL INERT-OAS/CO, ARC WEUNNO 337
Overhead position
3 1&2 155 17-21 3.25 31
-*t I- 3.0
I
aa=i
LA
60"
Similar to 1&2 155 17-21 3.25 9.2

above
12.5 Similar to above, 1-3 155 17-21 3.25

but 3 passes in
the groove
20 1-4 155 17-21 3.25 4.7

-4 H-2.0
3) With 1.6 mm wire

Flat position
3 1 190 19-23 2.4 75
s 190 19-23 2.4 27
>*-60V 1 210 20-24 2.75 24

s 210 20-24 2.75 31
n
<*£*G-
12.5 Similar to above, 1-3 300 25-29 4.4 39

but 3 passes in S 300 25-29 4.4 50
the groove
25 Similar to above, 1-6 300 25-29 4.4 17

but 6 passes in S 300 25-29 4.4 50
the groove
HV position
3 1 190 18-22 2.4 107
s 190 18-22 2.4 27
□ gS£
zp
1-3 200 19-23 2.6 68
60« 3
S 200 19-23 2.6 27
12.5 2 / H 1-4 210 19-23 2.75 44

S 210 19-23 2.75 31
Ail5c
25 1-10 210 19-23 2.75 21

,T30« s 210 19-23 2.75 31
i'tl
SjU-"*
Vertical position
3 1 180 19-23 2.25 56

c s 180 19-23 2.25 25
1 150 17-21 1.8 10

^60<y s 150 17-21 1.8 20
> * 1.5
12.5 Similar to above, 1&2 150 17-21 1.8 8.2

but 2 passes in S 150 17-21 1.8 20
the groove
25 Similar to above, 1-4 150 17-21 1.8 6.0
but 4 passes in S 150 17-21 1.8 20
the groove
Overhead position
3 165 17-22 2.0 21
7!Ut:2.0
L—ffTj? I
1-3 165 17-22 2.0 20

-it-2.o
A- A3
12.5 Similar to above, 1-5 165 17-22 2.0 17

but 5 passes in
the groove
25 Similar to above, 1 140 16-21 1.7 20
but 10 passes in 2-10 160 17-22 1.9 10
the groove
FILLET WELDS
Leg length Run Current Voltage Wire-feed rate Arc travel speed
mm numbers amp V m/min cm/min
1) With 0.8 mm wire
Flat position
3 1 150 19-23 10 62
4 1 150 19-23 10 40
5 1 150 19-23 10 27
6 1&2 150 19-23 10 40
Horizontal
3 1 145 19-22 9.6 70
4 1 145 19-22 9.6 40
5 1 145 19-22 9.6 26
6 1&2 145 19-22 9.6 37
Vertical down
3 1 115 17-21 7.1 52
4 1 115 17-22 7.1 32
5 1 115 17-22 7.1 21
6 1 115 17-22 7.1 15
Overhead
3 1 155 19-22 10.7 66
4 1 155 19-22 10.7 42
5 1 155 19-22 10.7 29
6 1&2 155 19-22 10.7 42
Note: Above welds can be made at higher speed, using one of the following combinations of current and voltage:
Amp V
170 18-23
195 23-27
205 24-28
230 26-30
340 MOOCRN ARC WEUNNQ TICMNOtOOY
2) With 1.2 mm wire

Flat position
3 1 150 20-23 3.2 56
alt. 1 200 24-28 5.2 92
4 1 150 20-23 3.2 33
alt. 1 200 24-28 5.2 53
5 1 200 24-28 5.2 35
alt. 1 350 33-37 14.2 95
6 1 200 24-28 5.2 24
alt. 1 350 33-37 14.2 67
8 1 200 24-28 5.2 14
alt. 1 350 33-37 14.2 38
10 1 200 24-28 5.2 9
alt. 1 350 33-37 14.2 25
12 1&2 200 24-37 5.2 13
alt. 1 350 33-37 14.2 17
Horizontal
3 1 150 20-23 3.2 56
alt. 1 200 24-28 5.2 92
4 1 150 20-23 3.2 33
alt. 1 200 24-28 5.2 53
5 1 200 24-28 5.2 33
alt. 1 300 30-34 10.2 68
6 1 200 24-28 5.2 24
alt. 1 350 33-37 14.2 68
8 1&2 200 24-28 5.2 28
alt. 1&2 350 33-37 14.2 77
10 1&2 200 24-28 5.2 18
alt. 1-3 350 33-37 14.2 75
12 1-3 200 24-28 5.2 19
alt. 1-3 350 33-37 14.2 53
Vertical down
3 1 115 17-20 2.0 36
alt. 1 145 18-21 3.0 53
4 1 115 17-20 2.0 21
alt. 1 145 19-22 3.0 31
5 1 115 17-20 2.0 13
alt. 1 150 19-22 3.3 22
6 1 115 17-20 2.0 9.5
alt. 1 165 20-23 3.8 18
Vertical up
8 1 150 20-23 3.3 9
10 1 150 20-23 3.3 6
12 1&2 150 20-23 3.3 8.5
Overhead
3 1 130 19-20 2.5 48
4 1 130 19-22 2.5 26
5 1 130 19-22 2.5 17
6 1 130 19-22 2.5 12
alt. 1 150 20-23 3.3 16
8 1 150 20-23 3.3 9
METAL INERTOAt/CO, AHC WEUNMO 341
10 1 150 20-23 3.3 6

12 1-3 150 20-23 3.3 12
3) With 1.6 mm wire
Flat position
6 275 25-28 3.8 31
alt. 435 35-39 9.2 74
8 275 25-28 3.8 18
alt. 435 35-39 9.2 43
10 275 25-28 3.8 12
alt. 435 35-39 9.2 28
12 1&2 275 25-28 3.8 16
alt. 1 435 35-39 9.2 20
15 1-3 275 25-28 3.8 16.5
alt. 1&2 435 35-39 9.2 26
20 1-4 275 25-28 3.8 12
alt. 1&2 435 35-39 9.2 15
Horizontal
6 1 265 22-26 3.7 30
alt. 1 400 34-37 7.2 58
8 1&2 265 22-26 3.7 35
alt. 1&2 400 34-37 7.2 68
10 1-3 265 22-26 3.7 34
alt. 1&2 400 34-37 7.2 44
12 1-3 265 22-26 3.7 24
alt. 1-3 400 34-37 7.2 47
15 1-5 265 22-26 3.7 26
alt. 1-5 400 34-37 7.2 51
20 1-10 265 22-26 3.7 30
alt. 1-8 400 34-37 7.2 47
Vertical down
6 1 210 19-22 2.7 22
8 1 210 19-22 2.7 13
Vertical up
10 1 190 19-22 2.3 7
12 1 190 19-22 2.3 5
15 1 190 19-22 2.3 3.3
20 1&2 190 19-22 2.3 4.7
Overhead
6 1 195 20-23 2.4 20
8 1 195 20-23 2.4 11.5
10 1&2 195 20-23 2.4 15
12 1-3 195 20-23 2.4 16
15 1-5 195 20-23 2.4 21
20 1-12 195 20-23 2.4 24
GMA WELDING OF VARIOUS ALLOYS thus far in this chapter apply to mild steel, unless other
alloys are specifically mentioned.
Mild StMl Filler wires suitable for mild steel welding have been
C0 2 welding is extensively used for welding mild specified in AWS A5.18 79, and they have been covered
steel, all over the world. Most of the descriptions given on page 290, while describing the filler wires for TIG
welding. Similarly mild steel filler wires specified in BS short-circuiting transfer, either C0 2 or argon-based gas can
2901:1970 (Part I) have been covered on pages 290-291. be used.
On page 291 mention has been made that IS: 6419-1971
is equivalent to this British Standard. INFLUENCE OF SHIELDING GAS
Investigations were made recently in West Germany,
Low-Alloy Steels concerning the influence of the shielding gas composition
These steels are conveniently welded by the MIG/ on the undiluted weld chemistry and all-weld metal
C0 2 process. C0 2 is used extensively as a shielding gas properties in MIG/C0 2 welding using mild steel and
for alloys of low-carbon equivalent (i.e. with CE no low-alloy steel electrodes.
higher than that of AISI 4130 steel). Electrode wires of Gas compositions used are indicated in Tables 6.11
matching or compatible alloy composition are used, and and 6.12. Weld metal analysis for each case is given in
since C0 2 is an oxidising gas, the wire is made to contain Table 6.11, while all-weld metal mechanical test values
greater amounts of deoxidisers, such as Si and Al than are given in Table 6.12.
are required for argon-based shielding gases. Two sets of welding parameters were used, one with
For welding alloy steels of high hardenability and lower current range (I.e.) and the other with higher
for important applications, C0 2 shielding is avoided current range (h.c.) as given below:
because of the high arc energy and spatter associated Filler wires: Commonly used low-alloy steel filler
with it. High arc energy may induce weld cracking. One wires (electrodes) are listed in AWS A5.28-79 and have
of the following argon-based gas mixtures is preferred: been reproduced on page 291, Chapter 5, because most
98% Ar—2%0 2 of them are suited also for TIG (and PAW) processes.
75% Ar — 25%C02 The following notes on these wires are reproduced from
80% Ar — 20%CO2 the Appendix of this standard. They should be read in
80% Ar —15% C0 2 — 5% 0 2 conjunction with Table 5.24.
Use of 100% argon is avoided as stated earlier, because It should be noted that weld properties may vary
of the irregular weld bead shape and undercutting. appreciably depending on filler metal size and current
If spray-type metal transfer is desired, C0 2 shielding used, plate thickness, joint geometry, preheat and
must be avoided, because it is incapable of giving a true interpass temperatures, surface conditions, base metal
spray arc. Instead, an argon-based gas must be used. With composition and extent of alloying with the filler metal,
Electrodes of the following chemistry were used:
Mn Si O, Cr Mo Ni
SG2 0.11 1.49 0.82 0.020 0.015 0.0095
SG3 0.09 1.65 0.99 0.012 0.013 0.007
CrMo 0.11 1.20 0.58 0.008 0.015 0.009 1.14 0.54 0.10
FK* 0.087 1.63 0.65 0.013 0.010 0.019 0.30 0.29 1.36
'Also contains 0.012 Al, 0.086 V, 0.028 Cu
Parameters l.c. h.c.

Welding current Amps. 150-165 260-300
Wire diameter mm 1.2 1.2
Wire-feed speed mm / sec 61.2 146.7
Melting rate kg / hour 1.94 4.66
Arc travel speed mm / sec 4.0 6.17
Electrode extension mm 14 21.6
Gas-feed rate 1 / min 15 15
Preheat and interpass temperature °C 120-150 120-150
Position flat flat
Table 6.11:: Weld metal chemistry

Analysis
Shielding gas Wire C Mn Sii P S o2
l.c. Kc. Tc h.c. 1Z "Tc. l.c. Tc. l.c. h.c. l.c. h.c.
% % % % % % % % % % % %
100% C0 2 SG2 0.10 0.098 1.04 0.97 0.55 0.48 0.016 0.020 0.009 0.013 0.066 0.071
82% Ar, 18% C0 2 SG2 0.11 0.10 1.29 1.21 0.70 0.65 0.019 0.019 0.012 0.011 0.044 0.036
91% Ar, 5% C 0 2
4% 0 2 SG2 0.11 0.11 1.26 1.33 0.70 0.70 0.017 0.020 0.010 0.013 0.037 0.032
95% Ar, 5% 0 2 SG2 — 0.10 — 1.27 — 0.70 — 0.015 — 0.012 — 0.025
100% C 0 2 SG3 0.095 0.085 1.15 1.20 0.72 0.71 0.014 0.016 0.008 0.010 0.065 0.070
82% Ar, 18% C 0 2 SG3 0.085 0.085 1.42 1.31 0.90 0.82 0.015 0.016 0.010 0.010 0.050 0.036
91% Ar, 5% C 0 2
4% 0 2 SG3 0.08 0.08 1.41 1.34 0.89 0.85 0.015 0.015 0.009 0.009 0.029 0.032
95% Ar, 5% 0 2 SG3 — 0.075 — 1.43 — 0.90 — 0.015 — 0.009 — 0.024
100% C 0 2 Cr-Mo 0.12 0.098 0.80 0.81 0.41 0.40 0.010 0.012 0.013 0.014 0.072 0.077
82% Ar, 18% C 0 2 Cr-Mo — 0.10 — 0.905 — 0.42 — 0.011 — 0.013 — 0.048
95% Ar, 5% 0 2 Cr-Mo — 0.105 — 1.13 — 0.58 — 0.009 — 0.013 — 0.042
100% C 0 2 FK 0.083 0.068 0.99 0.98 0.47 0.34 0.012 0.012 0.012 0.013 0.072 0.074
82% Ar, 18% C 0 2 FK — 0.078 — 1.07 — 0.44 — 0.011 — 0.014 . — 0.053
95% Ar, 5% 0 2 FK — 0.055 — 1.35 — 0.57 — 0.010 — 0.012 — 0.048
carbon
*h.c.—High
*l.c.—Low carbon
Table 6.12: All-weld metal properties

Impact Impact
value value
Shielding gas Wire Mechanical properties Redn. of (ISO-V- (ISO-V-
U.T.S. Y.P. Elongation area TestO°C) Test-20°C)
TE- h.c. Tc- KZ~ l.c. h.c. " Tc7^ f ^ c T Tc h.c. I.e. h.c.
N / m m 2 N / m m 2 N/mm 2 N / m m 2 % % % % J J J J
100% C 0 2 SG2 594 538 489 422" 24.2 26.9 64~ 68~~ 50 81 45 67
82% Ar, 18% C 0 2 SG2 644 592 525 485 22.6 22.1 68 66 78 62 61 63
91% Ar, 5% C 0 2
4% 0 2 SG2 635 598 532 464 22.7 21.7 63 45 91 90 63 58
95% Ar, 5% 0 2 SG2 — 609 — 485 — 24.9 — 66 — 95 — 35
100% C 0 2 SG3 586 562 455 435 26.8 25.6 65 68 59 88 46 64
82% Ar, 18% C 0 2 SG3 659 613 548 503 23.3 25.2 64 67 144 100 114 71
91% Ar, 5% C 0 2
4% 0 2 SG3 630 615 542 505 20.2 27.1 61 69 121 113 116 88
95% Ar, 5% 0 2 SG3 — 604 — 492 — 28.8 — 71 — 127 — 99
100% C 0 2 Cr-Mo 836 795 786 721 17.1 15.3 54 54 43 34 30 28
82% Ar, 18% C 0 2 Cr-Mo — 790 — 702 — 17.2 — 49 — 53 — 33
95% Ar, 5% 0 2 CR-Mo — 830 — 762 — 16.1 — 52 — 62 — 42
100% C 0 2 FK 757 720 700 640 16.3 17.1 59 59 55 74 40 42
82% Ar, 18% C 0 2 FK — 737 — 665 — 18.6 — 59 — 54 — 44
95% Ar, 5% 0 2 FK — 786 — 760 — 17.5 — 61 — 90 — 54
and shielding gas. For example, when filler metals of the filler metal, when used with argon-oxygen
having an analysis within the range of Table 5.24 are shielding gas but will show a considerable reduction in
deposited, the weld metal chemical composition will the content of manganese, silicon, and other deoxidizers
not vary greatly from the as- manufactured composition when used with C0 2 as the shielding gas.
• ER80S-B2 Classification. Filler metal of this classi- control porosity when welding with C0 2 as the
fication is used to weld V£Cr-ViMo, lCr-ViMo, and shielding gas, and molybdenum for increased
l-ViCr- V6M0 steels for elevated temperatures and strength. They will give radiographic quality
corrosive service. It is also used for joining dis- welds with excellent bead appearance in both or-
similar combinations of Cr-Mo and carbon steels. dinary and difficult-to-weld carbon and low-al-
The spray transfer, short-circuiting, or pulsed loy steels. They exhibit excellent out-of-position
power modes of the GMAW process may be used. welding characteristics with the short-circuiting
Careful control of preheat, interpass temperatures, and pulsed arc process. The combination of weld
and post-heat is essential to avoid cracking. soundness and strength makes filler metal of this
• ER80S-B2L Classification. This filler metal is iden- classification suitable for single and multiple-pass
tical to the type ER80S-B2, except for the low car- welding of a variety of carbon and low-alloy steels.
bon content (0.05% maximum). This alloy exhib- • ER100S-1, ER100S-2, ER110S-1, and ER120S-1
its greater resistance to cracking and is more suit- Classification. These filler metals deposit high-
able for welds to be left in the as-welded condi- strength, very tough weld metal for critical ap-
tion or when the accuracy of the PWHT (postweld plications. Originally developed for welding HY
heat treatment) operation is questionable. 80 and HY 100 steels for military applications,
• ER90S-B3 Classification. Filler metal of this clas- they are also used for a variety of structural ap-
1
sification is used to weld the 2 4Cr-l MO steels plications, where tensile strength requirements
used for high-temperature-high-pressure piping exceed 690 MPa and excellent toughness is re-
and pressure vessels. It may also be used for join- quired to temperatures as low as -50°C.
ing combinations for Cr-Mo and carbon steel. All • ERXXS-G Classification. This classification in-
gas-metal arc welding modes may be used. Care- cludes those solid electrodes and rods which are
ful control of preheat, interpass temperatures, and not included in the preceding classes. The sup-
post-heat are essential to avoid cracking. plier should be consulted for the characteristics
• ER90S-B3L Classification. This filler metal is iden- and intended use of these filler metals. This speci-
tical to thetypeER90S-B3, except for the low car- fication does not list specific chemical composi-
bon content (0.05% maximum). This alloy exhib- tion or impact requirements. These are subject to
its greater resistance to cracking and is more suit- agreement between supplier and purchaser. How-
able for welds to be left in the as-welded condi- ever, any filler metal classified ERXXS—G must
tion. meet all other requirements of the specification.
• ER80S-NH Classification. This filler metal depos- All-weld tensile and impact requirements stipu-
its weld metal similar to 8018-C3 covered elec- lated in AWS A5.28-79 are given in Table 6.13
trodes and is used for welding low-alloy high- Preheat, interpass, and postweld heat treatment
strength steels requiring good toughness at tem- (PWHT) temperatures required for the prepara-
peratures as low as -40°C. tion of tensile and impact specimens are given in
• ER80S-M2 Classification. This filler metal depos- Table 6.14.
its weld metal similar to 8018-C1 electrodes. Typi-
Stainless Steels
cally, it is used for welding 3V4% nickel steels and
MIG welding provides a method of welding stainless
other materials requiring a tensile strength of 550
steels at high speeds, because it is a semi-automatic
MPa and good toughness at temperatures as low
process with high arc times and deposition rates. Flux
as -60°C
or slag is totally absent. This minimises cleaning and
• ER80S-Ni3 Classification. This filler metal depos-finishing time and gives consistent weld-metal
its weld metal similar to 8018-C2 electrodes. Typi- chemistry. The quality of welded joints is excellent.
cally it is used for welding 3Vfc% nickel steels for Shielding gas is either pure argon, argon with 1% oxygen
low temperature service, where a tensile strength or argon with 2% oxygen. Compared to pure argon,
of 620 MPa is required. argon with 1% oxygen gives improved arc stability,
• ER80S-D2 Classification (same as E70S-1B of produces a more fluid and controllable weld puddle,
A5.18-69). Filler metals of this classification con- good coalescence and bead contour and minimises
tain a high level of deoxidisers (Mn and Si), to undercutting on heavier stainless steels. Argon with 2%
Table 6.13: Tensile and impact requirements as per AWS A5.28-79
Class Shielding UTS YP Eln. % in Condition Impact value

gas min. MPa min. MPa 50 mm, min.
ER 80S-B2 550 470 19
ER80S-B2L
PWHT Not reqd.
ER90S-B3 Ar+
ER90S-B3L 1-5% 620 540 17
o2
ER80S-Nil As-welded 27Jat-46°C
ER80S-Ni2 550 470 24 PWHT 27Jat-62°C
ER80S-Ni3 27Jat-73°C
ER80S-D2 co2 550 470 17 As-welded 27Jat-29°C
ER100S-1 690 610 to 700 16
ER100S^2 Ar+ 690 610 to 700 16
ER110S-1 2% 760 660 to 740 15 As-welded 68 J at -51°C
ER110S-2 o, 830 730 to 840 14
Table 6.14: Heat treatment temperatures as per AWS A5.28-79

Class Preheat and PWHT
interpass temp. temperature
°C °C
ER80S-B2 150 ± 15 620 ± 15
ER80S-B2L
ER90S-B3 200 ± 15 690 ± 15
ER90S-B3L
ER80S-Ni2 150 ± 15 620 ± 15
ER80S-Ni3
ER80S-D2
ER80S-NU
ER100S-1
ER100S-2 150 ± 15 None
ER119S-1
ER120S-1
oxygen provides better arc stability, coalescence and this minimum, the transfer will be in the form of large
welding speed than 1% oxygen mixture on thinner drops with spatter and arc instability. For 0.9 mm
stainless steel materials. diameter wire, the threshold current value for stable
With pure argon, there is hardly any loss of alloying spray transfer is 170 amp. For 1.15 mm diameter it is
elements between the filler wire and the weld deposit. 225 amp and for 1.6 mm diameter it is 285 amp. Nominal
Even elements like Ti are readily transferred across the data for spray arc welding is given in Table 6.16.
arc, making Type 321 stainless weld metal practical with
this process. Table 6.15 shows the transfer efficiency of SHORT-CIRCUITING TRANSFER
several elements across an argon-shielded arc. Welding with a short-circuiting arc employs lower
DC is used with electrode positive. For thickness of current, generally ranging from 50 to 225 amp, low
3.2 mm and above, current above a certain minimum, voltages of 17 to 24 V, and small diameter wires—0.80,
which will ensure spray transfer must be used. Below 1.0 and 1.2 mm being the popular sizes. The distinctive
Table 6.15: Chemical compositions of some stainless steel filler wires and deposited weld metal
Alloy Type Mn Si Ni Cr Mo Ti Cb
308L Wire 0.025 1.51 0.37 0.020 0.011 10.32 20.31 — — —
308L Weld 0.030 1.57 0.35 0.019 0.011 10.28 20.46 — — —
308 Wire 0.056 1.75 0.47 0.021 0.021 9.94 20.94 — — —
308 Weld 0.051 1.72 0.33 0.018 0.013 10.03 20.89 — — —
321 Wire 0.079 1.43 0.58 0.019 0.016 10.52 18.58 0.95 —
321 Weld 0.085 1.44 0.63 0.016 0.012 10.52 18.78 — 0.85 —
347 Wire 0.069 1.59 0.49 0.023 0.014 9.96 20.82 — — 1.18
347 Weld 0.063 1.55 0.48 0.022 0.014 9.64 20.80 — — 0.87
Table 6.16: Nominal conditions for spray arc welding of stainless steel
Plate Joint and edge preparation Electrode Current Wire-feed Welding Number
thickness, wire dia- (dcrp), speed, speed, of
mm meter, mm amp m/min cm/min passes
3.2 Square butt with backing 1.6 200-250 2.75-3.75 50.0 1
6.5 Single-V butt, 60° incl. angle,
not root face 1.6 250-300 3.75-5.0 38.0 2
1.6 mm root face 1.6 275-325 5.65-6.30 50.0 2
1.6 mm root face 2.4 300-350 1.90-2.15 12.0 3-4
1.6 mm root face 2.4 350-375 2.15-2.40 10.0 5-6
1.6 mm root face 2.4 350-375 2.15-2.40 5.0 7-8
Note: Tables 6.16-6.19 are taken from Metals Handbook, Vol. 6, published by American Society for Metals.
feature of the short-circuiting arc is the frequent shorting The minimum thickness that can be welded by spray
of the electrode wire to the work. All metal transfer takes transfer arc is 3.2 m m . The characteristic d e e p
place at arc outages, which occur at a steady rate and penetration of this process calls for butt joints with
can vary from 20 to more than 200 per second. The net narrow groove angles and large root faces. Fit-up has
result is a stable arc of low energy and heat input, ideally to be perfect. For sheets thinner than 3.2 mm and for
suited to the welding of thin sections in all positions. joints with bad fit-up, short-circuiting mode of transfer
The low heat input minimises work-metal distortion. must be used, since this gives very shallow penetration.
The short-circuiting arc is better than the spray arc for Filler wires: Standard stainless steel filler wires for
welding joints that have poor fit-up. MIG welding are covered in the following specifications:
Electrode wire extension (stick-out) should be kept AWS A5.9
as short as possible. Backhand welding is normally BS 2901: Part 2
easier on fillet welds and results in cleaner welds. IS 5856-1970
Forehand welding is used for butt welds, outside corner The same wires are also suited for use with some or
welds are made with a stra-ight movement. Typical arc all of the following processes: Oxyacetylene, TIG, PAW
voltages and gas flow rates are: and Submerged-arc.
V Litres/
They have been referred to earlier in Chapter 5.
min
Spray transfer/argon shield/1.6 mm wire 24 19
Spray transfer/argon with 1-2% 0 2 /1.6 mm wire 26 19 Aluminium Alloys
Short-circuiting/argon shield/0.9 mm wire 18 12 These alloys are most conveniently welded by the
Short-circuiting/argon with 1-2% O2/0.9 mm wire 19 12 TIG and MIG processes, and their weldabihties are the
same for both the processes. TIG welding of these alloys For best result, all the above operations should be
is described on page 258. carried out immediately before welding. Alternatively,
The range of wrought Al alloys in industrial use is welding must be done on freshly machined and freshly
very wide, and their weldability properties differ from filed surfaces.
alloy to alloy. Some are readily weldable, some others A common source of weld-metal porosity in MIG
have limited weldability, while a few are not meant to be welding of Al alloys is the invisible oxide film on the
welded. A few examples of each category are given in filler wire. In TIG welding, it is possible to remove this
Table 6.17. contaminant by giving the filler rod a rub with steel wool
The standard filler wires are covered in AWS A5.10, immediately before welding. Such pre-cleaning is not
BS 2901-Part 4 and IS 5897. They have been dealt with possible with the continuous wire in reeled or spooled
on pages 287,290 and 291 as filler wires for TIG welding. form, which is used for MIG welding. Moreover, the
The high melting and solidifying rate of the weld relatively thinner wires used in this process mean that
deposited by the MIG process can result in entrapped the ratio of exposed metal surface to the unit weight of
gas in the welds. The gas is caused by contaminants weld metal is significantly high. Producers of wires use
present on joint surfaces and filler wires. Joint surfaces specialized techniques to ensure that the spooled wires
must be free from dirt, oil, grease, paint, moisture, metal are absolutely oxide-free and they supply these spools
particles and heavy oxide film. Some alloys form a thin in thoroughly sealed plastic containers. Users are
oxide coating on the surface, while others have a thick, advised to open the containers immediately before use
dark oxide coating. The methods of removing various and to ensure that no contaminants are picked up in
contaminants are given below: the wire feeder of the MIG equipment. After a day's
use, the balance wire must be repacked in an airtight
Contaminant Method of removal
container. Any contamination of the wire from storage
Dirt Washing and scrubbing with a detergent or handling or by exposure to humid air must be
solution avoided. Wires in their original and intact packings must
Grease and oil Swabbing with an organic solvent like be stored in a dry, warm storage room.
acetone, carbon Shielding gases: Typical shielding gases used in the
tetrachloride or trichlorethylene. U.S.A. are pure argon, pure helium or Ar-He mixture.
Heavy oxide Removed with wire brush, steel wool, Argon is preferred for thinner sections (up to 19 mm f),
layer mill file or scraper. because of lower arc heat, smoother and steadier arc
Total oxide Immersion in butylalcohol-phosphoric and less weld spatter. Helium gives higher arc heat,
removal acid type solution, followed by w a s h - deeper penetration, and a weld profile which is wider
ing with water and drying with hot air. and less convex. The weld has a darker tint and the
Table 6.17: Al alloys and their weldability

Category Alloy grade* Approx. alloy composition (%)
Easily weldable 1060 NHT 99.6 (min.)

1100 NHT 0.12 Cu / 99 Al (min.)
2219 6.3 Cu / 0.3 Mn / 0.18 Zr / 0.1 V / rest Al
3004 NHT 1.2 Mn/1.0 Mg/rest Al
5050 NHT 1.4 Mg / rest Al
5456 NHT 0.6 Mn / 5.1 Mg / 0.12 Cr / rest Al
6061 HT 0.6 Si / 0.27 Cu / 1.0 Mg / 0.2 Cr / rest Al
Used for heavy 7005 HS/HT 0.45 Mn / 1.4 Mg / 0.13 Cr / 4.5 Zn / rest Al
defence structures 7039 HS/HT 0.27 Mn / 2.8 Mg / 0.2 Cr / 4.0 Zn / rest Al
Required special procedure 2014 HS/HT 0.8 Si / 4.4 Cu / 0.8 Mn / 0.5 Mg / rest Al
4032 12.2 Si / 0.9 Cu / 1.1 Mg / 0.9 Ni / rest Al
Limited weldability 2024 HS/HT 4.4 Cu / 0.6 Mn / 1.5 Mg / rest Al-
Not weldable 7075 HS/HT 1.6 Cu / 2.5 Mg / 0.3 Cr / 5.6 Zn / rest Al
*The numbers refer to the designations of the Aluminium Association of U.S.A.
NHT = Non-heat treatable. HT = Heat treatable. HS = High-strength.
Table 6.18: Data for butt welds in flat position for Al
Plate t Joint design Elect. Ar flow Wire-feed Amps./V Welding No. of

mm dia., mm rate, speed, speed, passes
1/min m/min cm/min
3.15 sq. butt, no root gap 1.2 14 4.45 110 / 20 60 1

6.4 60° V, 0-3 mm root gap,
1.6 mm root face 1.6 20 4.3 200 / 25 60 1
12.5 60° V, 0-3 mm root gap,
1.6 mm root face 2.4 24 3.3 290 / 27 40 2
spatter is slightly more than with argon. Ar-He mixtures Large-diameter electrode techniques have also been
are often used to get the best out of the two gases, developed for butt welding plates in the thickness range
especially in positional welding of thicker sections. The of 19.5-75 mm. They call for wide single or double-V
flow rates are 14-33 1/min for pure argon, 24-50 1/min grooves and use electrode sizes of 4-5.5 mm in the
for 25-75 Ar-He and 28-661/min for pure helium. These current range of 450-650 amp at 28-31 V. The resultant
values are for 1.6 mm electrode wire. welds are capable of meeting the requirements of section
Most Al-alloy welding is done with spray-type arc IX of the ASME Boiler and Pressure Vessel Code. Heavy
using normal steady current or pulsed current. Spray cranes and spheres have been welded by this method.
metal transfer is ensured by using high current densities
ranging from 80 to 480 amp/sq. mm of electrode cross- Copper and Copper Alloys
section. The advantages of spray-type arc are: deep The standard filler wires for MIG welding of Cu and
penetration, which ensures sound root fusion; stiff arc, its alloys are covered in AWS A5.7, BS 2901:1970 (Part
which helps positional \ 'elding and a narrow stable arc, 3) and IS: 5898-1970, referred to on pages 287, 291 and
which enables one to make small fillet welds in relatively 293.
thick material Pulsed-current welding enables one to Though MIG process is applicable to many Cu alloys,
use spray-type arc for the thickness range of 0.75-3 mm. it is preferred for the welding of aluminium bronzes,
It also makes the welding of thin sheets to heavier plates silicon bronzes and copper nickels in section thicknesses
easier. greater than 3 mm. For lower thicknesses, the TIG
Typical parameters for butt welds in the flat position process is preferred. Flat position welding is done with
are given in Table 6.18. spray transfer. Welding in the vertical and overhead
Backing bars of carbon steel or stainless steel are positions is also carried out with small diameter
commonly used for temporary backing of butt welds. electrodes, low currents and globular or short-circuiting
Copper and aluminium may be used when higher mode of transfer. Lead containing alloys cannot be MIG
thermal conductivity is needed. Carbon steel being welded.
magnetic can sometimes deflect the arc and interfere Coppers: MIG welding can be used for deoxidised
w i t h w e l d i n g . W h e n c o p p e r backing is u s e d , coppers to get the same weld soundness and strength
contamination of the weld metal by copper must be as is obtained with the TIG process. Coppers which have
prevented. not been deoxidised or contain insufficient amounts of
High current-density techniques can be used for deoxidiser (such as electrolytic tough pitch copper,
making butt welds in the thickness range of 6.4 to 15.8 oxygen-free copper, and low-P deoxidised copper) give
mm using square butt joints and depositing one pass greater porosity and lower strength in the HAZ with
from each side. Back gouging is not required. The the MIG process than with the TIG process. Hence the
resultant welds are of high radiographic quality. The latter process is preferred. The electrode wire must
current range of 370 to 450 amp is used at 24-25 V for conform to AWS class ECu, i.e. it m u s t have 98%
2.4 mm electrode size. In one case, 44.5 mm thick butt minimum purity and contain controlled amounts of P,
weld was made, using one pass from each side with Sn, Si and Mn as deoxidisers to prevent or minimise
450 amp, 28 V, 481/min argon-feed rate and 1.6 m m porosity. Sections thicker t h a n 6.3 m m require
diameter electrode. preheating due to the high thermal conductivity of
copper. Preheat temperatures range between 100 and weld. Preheating must be avoided and the interpass
480°C, depending on workpiece thickness. temperature must be held below 95°C The commonly
Phosphor bronzes: MIG process is commonly used for used electrode wire conforms to AWS class E Cu Si (2.8-
the thickness range of 9.5-12.7 mm. For alloys with less 4.0 Si, 1.5 maximum Sn, 1.5 maximum Mn, 0.5 maximum
than 8% Sn, AWS class ECu Sn-A electrode wire having Fe). A thin oxide film forms on the weld deposit after
4-6% Sn and 0.1-0.35% P is used. For alloys with 8-10% each pass, which must be removed with a wire brush
Sn, ECu Sn-C having 7-9% Sn and 0.05-0.35% P is used. before depositing the next pass.
These wires may also contain 0.5% Si to help minimise Copper-nickels: These alloys in thicknesses greater
weld porosity. Controlled preheating in the range of than 1.6 mm are usually MIG welded in the flat position,
100-200° C is used to increase weld metal fluidity and though welding in the vertical and overhead positions
reduce porosity. Severe preheating and large heat input is possible. Their thermal conductivity is equal to or
during welding encourages hot-cracking of the weld lower than that of mild steel. Hence no preheating is
and columnar grain growth. Hence stringer bead necessary. Post-heating is also not necessary. Interpass
technique with small weld pool and high are travel temperature during welding should not exceed 65°C.
speed must be used. The weld must be peened between The commonly used electrode wire conforms to AWS
passes. class E Cu-Ni and has 70% Cu, 30% Ni and 0.15-1.0%
Al bronzes: These alloys in the thickness range of 3 - Ti. Titanium minimises porosity and prevents oxygen
12.5 mm can be MIG welded with ease in all positions, embrittlement by acting as a deoxidiser, and makes the
because of the comparatively high tension pf the molten weld metal more fluid.
weld metal and the low thermal conductivity of the base Table 6.19 gives nominal welding parameters for
metal. While spray-type metal transfer is used for the welding 9.5 mm thick butt joints in copper and copper
flat position, welding in the vertical and overhead alloys.
position is carried out with the globular or short-
circuiting mode of metal transfer, using electrode wire MIG Spot Welding
sizes of 1.6 mm and less. Electrode wire conforming to MIG or GMA (gas-metal arc) spot welding is similar
AWS class ECu A1-A2 is normally used. It contains 9- to resistance spot welding or riveting. In this process,
11% Al, and also 1.5% Fe to make the weld deposit less two overlapping sheets are fused together by means of
susceptible to hot cracking than welds from E Cu Al- a spot weld, which penetrates completely through one
Al. Wire of AWS class E Cu Al-B (11-12% Al, 3-4.25% sheet into the other. No joint preparation is required,
Fe) also gives welds free from hot shortness, but they but the overlapped surfaces must be scrupulously
are stronger, harder and less ductile. cleaned before welding. The welding gun is held
Si bronzes: These alloys are readily welded by the stationary against the plate while the spot weld is being
MIG process, but stringer bead technique, low current made. To make the weld, the operator initiates a pre-
and high travel speed must be used to prevent excessive set timed sequence, which strikes the arc and feeds the
heat built-up, which occurs because of the low thermal wire, and finally stops the arc as well as the wire feed
conductivity, and which results in hot cracking of the after a weld of required size and penetration is
Table 6.19: Data for MIG welding of 9.5 t butt joint in Cu and Cu-alloys
Joint design Electro. Shielding Gas flow Amps./V Welding No. of Preheat
dia., mm gas rate, speed, passes temp., °C
1/min cm/min
Commercial copper
90° V, 5 mm root face, no root gap 2.4 Ar 14 500 / 27 36 — 205
Phosphor bronze
90° V, no root face, root gap 3 mm 1.6 He 16 280 / 27 — 3 or 4 95-150
Aluminium bronze
70° V, no root face, root gap 3 mm 1.6 Ar 14 285 / 28 — 2 none
Silicon bronze
60° V, no root face, root gap 3 mm 1.6 Ar 14 265 / 28 21 2 none
Copper-nickel
70° V, no root face, root gap 4.5 mm 1.6 Ar 14 280 / 28 — 2 none
completed. The arc time is expressed in cycles, one cycle materials can be welded with conventional welding
being equal to 1/.50 sec. The gun is then moved to the power sources, it has distinct advantages over resistance
next weld location. welding techniques in the fabrication and lining of
The process uses a standard MIG-welding gun, to structures and vessels.
which a special nozzle may be attached, so that the gun This method is commonly used on mild steel,
can be rested on the workpiece, gas may be allowed to stainless steel and aluminium. A typical example of mild
escape through holes or notches in the nozzle, or legs steel application is a motor-end panel, to which support
may be attached to the nozzle to maintain it at a standard bars and channel sections are spot welded, using a
distance from the workpiece, so that electrode extension portable gun with a nozzle guide tube, which is seated
remains constant. against two right-angled surfaces to make spot fillet
The following sequence of operations proceeds welds. Using 1.6 mm wire, 350 amp (DC electrode -ve),
automatically, once the torch trigger is depressed: 30V, 22 cycles arc time and shielding mixture of argon
1) The welding contactor is closed, shielding gas plus 27% oxygen at flow rate of 71/min, a fillet weld of
flows, and the filler wire advances at a pre-set low 6.4 mm leg length and 6.4 mm long is made.
speed (run-in) proportional to the welding speed. In another application, cupro-nickel sheets were
2) On contact between the wire and the workpiece, attached to mild steel water boxes in a seawater
the arc strikes and the wire accelerates to or to- distillation plant. It was possible to control penetration
wards the spot at pre-set welding speed. to maintain the iron content of the weld at a sufficiently
low level to prevent cracking.
3) The arc is maintained for a pre-set time.
4) The wire halts and the arc is allowed to continue PULSED MIG WELDING
sufficiently long to separate the wire from the
Pulsed arc MIG welding is a controlled method of
weld (burn-back), while shielding gas flows to
spray transfer welding requiring a more sophisticated
prevent excessive oxidation of the weld surface.
power source, where as the three type of transfers
The MIG spot welding process is controllable and described previously can be obtained with standard
requires m i n i m u m operator skill. Since access is power sources and wire feed units. In spray transfer,
normally required from one side of the joint and thick droplets of metal are projected from the wire tip across
the arc gap to the molten pool at a constant current. In
r dip transfer, metal is transferred to the molten pool

R
(3)-
(C) somewhat irregularly during the periods of short
circuiting. Pulsed arc welding enables droplets to be
(2)
(4)-
WORKPIECE projected across the arc gap at a regular frequency, using
Fig. 6.15 Power of source of pulsed MIG welding pulses of current in the spray transfer range from a
special supplied power source. Transfer of metal from
the wire tip to the molten pool occurs only at the period
© @®0® of pulse or peak current.
During the interval between the pulses, a low
background" current maintains the arc to keep the wire
O U b Q v tip molten, but no metal is transferred. Pulsed transfer
means that the weld metal is projected across the arc
Pulsed arc droplets gap at a high current, but the mean welding current
remains relatively low. The operator can vary the pulse
8. 300 A height and the background current to obtain full control
< of both the heat input and the amount of weld metal
§ 2QCH deposited. However, in modern power sources the pulse
p r o c e d u r e is preset or p r e - p r o g r a m m e d by the
o manufacturer to simplify use. Pulsed arc transfer can
Time be used on mild and low-alloy steels, stainless steel and
18 m/sec is particularly useful with aluminium and its alloys on
Fig. 6.15 (A) Sequence of events in pulsed metal a light to medium plate sections as dip transfer cannot
transfer be used in these applications.
The advent of electronic power sources has enabled detach a single molten drop. A low background current
controlled metal transfer to be achieved with the MIG is chosen to maintain arc stability and frequency of
process. These power sources provide a static volt-amp pulsing, and hence the mean current is proportional to
characteristic that can be tailored to process the wire feed speed. If, as originally conceived, the
requirements and a dynamic response significantly power source is operated in the constant current mode,
more rapid than that from a conventional power source. and the pulse parameters are specified in terms of peak
Feedback from the arc and the wire feeder modifies the current, then an additional voltage control of either the
power source output to control metal transfer and wire feeder or p o w e r source may be required to
maintain a stable arc. maintain arc length. This means that 'one knob' control
First development of pulsed MIG power sources of the power source can be achieved with the wire feed
used semiconductors employing series regulator or speed governing the arc parameters.
chopped (switched) secondary. However the advances Pulsed (synergic) MIG was originally conceived with
mode in inverter technology have made these designs the pulse parameters defined in terms of peak and
obsolete. In the inverter design the mains input is first background current, the theory being that current is the
rectified and then switched (by transistors) through a main parameter controlling heat input. However, recent
welding transformer at high frequency (typically 5-40 power source designs operate with constant voltage in
KHz), before being rectified again to give a DC output. the peak. This gives a measure of self adjustment during
This technique offers a considerable improvement in the pulse and offers greater arc stability, particularly
efficiency and performance. These electronic power with aluminium and cored wires. Metal transfer and
sources can be configured to p r o d u c e an o u t p u t arc stability can also be enhanced by generating a more
specifically suited to a particular process, wire, gas or complex wave form than a square pulse. A high current
joint design. Feedback from the arc controls the metal peak can induce droplet detachment and a slope on the
transfer and maintains arc stability. trailing edge of the pulse will aid the stable transition
DC operation with conventional power sources, at between peak and background levels.
low mean welding currents using dip transfer, the short Pulsed MIG welding results reduced fusion faults
circuit currents draw a high peak current which ruptures at low mean currents and considerably lower spatter
the contact explosively giving serious spatter. This is levels. Synergic pulsed MIG offers controlled heat input
followed by a period of low arc current giving a 'cold' with spray type metal transfer, permitting positional
weld and the possibility of lack of fusion faults. With welding with a small weld pool and superior weld
controlled metal transfer welding, the power source finish.
detects that a short circuit is about to occur and injects a
high current in a controlled manner. This can control MIG BRAZING
the end of the wire shorting to the liquid pool and will W h e n coated steels are fusion welded, the
minimimise spatter. One type of short-arc technique effectiveness of the coating adjacent to the weld
senses the arc current and wire feed speed. The power deteriorates due to heat and a recondition operation is
source output is adjusted to give a constant arc current often required to restore corrosion resistance or desired
per unit volume of wire. As the wire feed speed is varied, visual appearance to the affected areas. MIG-brazing
the arc current is varied in proportion. This helps to process addresses this problem.
overcome the incidence of lack of fusion associated with In MIG-brazing, like in the conventional brazing
conventional dip transfer welding. It is to be understood process, employs the technique of "wetting" the joint
that with these controlled short-arc techniques it is without melting it with the help of a continuously fed
generally the welding current that is controlled (via filler metal, the melting temperature of which is below
feedback), rather than the voltage as with a conventional that of the steel.
MIG power source. The process is carried out in either semi-automatic
In pulsed MIG welding (often known as Synergic or automatic technique similar to MIG welding. Unlike
MIG) a spray-type metal transfer is achieved at low in the conventional brazing process, there is no need at
mean welding currents by pulsing the power source all for any fluxes.
output between a low and high level. Metal is transfered Pure argon is mostly used as shielding gas in MIG
only during the peak current pulse. The peak current is brazing, although little bit addition of active gases like
chosen to be above the globular to spray transition C 0 2 or 0 2 helps improve stability and 'wetting'
current level and the duration of pulse is selected to properties of the weld metal. Argon with 1% oxygen as
Table 6.20 : Wires for MIG brazing

Wire type % Melting UTS, Hardness, Applications
Composition point, C N/mm 2 BHN
Si - Mn Alloy 1 mm 980 - 1020 350 90 MIG brazing of steels
DIN: S -CuSi3 3 Si
Balance Cu
Al - Bronze 8A1 1030 550 130 MIG brazing of dissimilar metals
DIN: Cu A18 Bal ance Cu
Phosphor-Bronze 7Sn 900 -1050 500 120 MIG brazing on ferrous and dissimilar
DIN: S-Cu Sn 6 Balance Cu metals.
shielding gas is most optimum as confirmed by studies. laid side by side, simultaneously by two welding heads,
A number of different copper based wires—Si-alloy or each with a separate power source and controls, which
aluminium bronze/phosphor bronze alloys—can be act symmetrically with respect to the joint plane. The
used for MIG brazing. Typical wire compositions, important components of the welding head are:
physical properties and applications are shown in Table a) Wire and its driving and stressing system
6.20. The wire diameters can vary from 0.8 mm tol.2 b) Torch
mm and the spool sizes from 1 kg to 15 kg.
c) Gas supply
A few users of this process have been working
without pulse mode in short-circuiting technique. This d) Welding head setting
may lead to a lot of uncontrolled spatter if the power a) The copper-coated wire is driven by rollers,
source is not well regulated. In addition, the operator which feed it to the joint as well as stress it so that it
needs to adhere to a very constant contact tip distance curves after leaving the torch end. The stressing system
to avoid abrupt parametric changes with consequent allows the wire to be fed in a range from straight to an
impairment in the weld. These factors necessitate use arc about 180 mm in diameter. The mechanical assem-
of pulsed technique for MIG brazing, with one single bly consists of a motor tachometer, gear reduction ar-
molten droplet detaching from the wire electrode for rangement, stress roller, drive rollers, skew-pressure
pulse. roller and the torch.
In order to keep zinc vaporisation from light gauge b) The torch is of special design. A typical torch con-
sheets as low as possible, MIG brazing is usually sists of a 4 mm thick copper bar, electrically insulated
performed at low power—typically 40 Amps, to 130 and cooled by a water supply. An axial hole drilled in
Amps, with travel speeds between 70-100 cm/min. The the copper bar permits wire driving, whereas a contact
background current is often 20 Amps, or even less. An tube located at the bottom of the bar transmits the elec-
inverter power source with high switching frequency trical current to the electrode wire by a rubbing contact.
of 100 KHz delivers an extremely smooth welding The contact tube is equipped with a small tip, which
current. Today power sources are available in the can be replaced when it becomes worn or if a burnback
"Synergic" mode (on knob control) for any wire/gas occurs. The tip is electrically insulated by a ceramic
combination with pre-programmed parameters. Fully shield which also serves, in case of a burnback, to extin-
digitised systems are now available for convenience of guish the arc before it reaches the water jacket of the
synergy setting and the possibility of optimising all the contact tube.
parameters relevant to the process individually. c) The arc and the weld puddle are shielded by two
tubes located at each side of the torch. These tubes are
Narrow-Gap Welding equipped with special nozzles to produce a laminar flow
The general economics of narrow-gap welding of of gas preventing aspiration of air into the weld region.
thick plates was dealt with in the chapters on SA Approaching the end of the weld, a special nozzle (called
welding and TIG welding. The MIG process is equally a gas cup) is put on the torch to give a circular gas sup-
suitable for this technique. ply, which avoids air contamination; the shielding gas
Narrow-gap MIG welding operates in the spray flow remains the same. For low-alloy steels, 80% Ar/
transfer range. Each weld layer is made up of two beads 20% C 0 2 mixture is used as a shield.
d) Each torch is carried on a support which is Twist-Arc Welding

constituted by two crossed motorised runners. This This is a modified version of narrow gap MIG
setting allows two travel directions for the welding welding developed in Japan. It uses two intertwined
head; one located in the joint plane, the other wires as the electrode (see Fig. 6.16), which as it
perpendicular. These tvv o possibilities act directly on the consumes, gives rotational movement to the welding
wire-to-work distance. arc and ensure thorough defect-free sidewall fusion.
Two transducers, referred with respect to the bottom When the gap is maintained within 12-18 mm, sound
of the joint and to the adjacent side, act upon the weld joints can be obtained by adjusting the travel speed
electronic controls of the runner motors to put the torch according to the gap size, without changing the welding
into position. Thus, the guidance of each torch is fixed current and arc voltage.
with respect to the work independently of the joint The process developed by Kobe Steel employs an
width variations, of the joint filling height, or of the automatic control system, in which the image of the
parallelism of the relative displacement of the torch with square groove in the vicinity of the welding arc is
respect to the joint side. registered by a small TV camera. Seam tracking and
The different operations are automated by three control of the welding parameters are then performed
functions: by analysing and processing the video signal from the
(1) Proximity, (2) Seam track, (3) Wire feed. camera. This automatic control system makes it possible
Proximity is the drive used to control the torch-to- to hold the electrode position within 0.5 mm of the centre
work distance by positioning the head. This function of the groove gap in real time, and frees the welding
effectively controls the arc voltage, since this voltage operator from the severity of the working environment
varies directly with the arc length or with the torch-to- by virtue of its remote-control function. Optimum
work distance. Seam track controls the cross slide drive, welding conditions are given below:
which keeps the head tracking along the work seam.
Wire feed controls the rate at which filler wire is fed to Power source DC drooping characteristic
the work. Shielding gas Ar + 20% C 0 2
Each of these functions is controlled by a separate Wire diameter 2.0 x 2.0 mm
feedback motor control system. This is a closed loop, Pitch of twist 9 - 1 0 mm
solid state control, featuring reversing with the use of a Groove width 12 ~ 18 mm
contactor, and having a precision control under varying Welding current 550 - 600 amp
loads. Arc voltage 31 ~ 32 V
Following conditions must be observed: Welding speed 400 ~ 425 mm/min
1) Oxyacetylene cutting must not be used for edge
preparation, because it decarburises the base
metal and cannot guarantee gap tolerance of ± 1
mm for 9 mm gap and +10 mm for 6 mm gap.
2) Joint faces must be accurately parallel. Tolerances
are +050 mm.
3) Joint perpendicularity with respect to the sup-
port is not very important. Tolerances are +0040
radian.
4) Electrode wire curvature between 175 and 190
mm is advisable. Buckling of the wire of 20 mm
maximum seems appropriate.
Advantages of narrow gap w e l d i n g are: (1)
reduced internal stresses and distortion, (2) better as-
welded mechanical properties of the joint, (3) better
economy for thicknesses above 50 mm. The
disadvantages are: (1) more likelihood of weld defects, Fig 6.16: General arrangement of twist arc welding
(2) difficulty of removing and rectifying defects when process: (1) twist wire; (2) arc; (3) molten pool; (4)
detected. direction of welding; (5) shielding gas
Applicability of the process to pressure vessels and metal transfer as desired. A sufficiently high wire-feed
boiler plates in the thickness range of 50-120 mm has rate causes the molten drops to touch the weld pool,
been established. resulting in a dip-transfer technique. When the wire
speed is somewhat decreased, the mode of transfer
Plasma-MIG Welding changes to free flight. The coarse drop transfer occurring
This is a modification of the TIG plasma arc process here is very different from transfer during MIG welding
, in which filler wire is continuously fed through the below the transition current: no restraining arc force is
axis of the constricted arc plasma. The apparatus, as present, because the arc is not attached to the drop.
d e v e l o p e d by a E u r o p e a n company, is s h o w n Material transfer occurs in an envelope of arc plasma
schematically in Fig. 6.17. The tungsten electrode is streaming to the workpiece. Consequently spatter is
placed away from the axis of the copper housing, which absent. With further decrease in wire speed, melting
has a water-cooled orifice for constricting the arc plasma. occurs in a more constricted zone of the arc and a fine
The wire can then be axially fed through the orifice. A spray of filler metal is carried along with the plasma,
curved arc is formed between the tip of the tungsten which resembles the spray transfer during MIG welding
electrode and the workpiece. The wire passes through above the transition current.
a considerable length of the plasma and the hottest Figure 6.17 also shows that the filler wire can be
constricted zone. Typical welding parameters and energised by connecting it to a separate power source.
melting rates of filler wire are given in Table 6.21. When this is done, it is possible to have an arc between
By suitably adjusting the wire-feed rate, one can the filler wire and the workpiece, while maintaining a
obtain short-circuiting, free flight or fine-spray type separate arc between the tungsten electrode and the
workpiece. This is the real version of the plasma-MIG
welding process and it results in higher melting rate
and improved control of penetration. The higher melting
t CONSUMABLE WIRE
TUNGSTEN
rates are evident in Table 6.20.
ELECTRODE - HIGH The special feature of this process is that, it is suitable
FREQUENCY!
[GENERATOR] for welding refractory metals like molybdenum and
tungsten.
POWER
SUPPLY HIGH SPEED WELDING (HSW)
The most important single factor in increasing
productivity is the deposition rate. The average rate is
between 3-5 kg/hour. It should always be possible to
raise this deposition rate to between 7-10 k g / h o u r
without the need for additional investment in new
equipment.
y Thus, by using unconventional parameter settings,
we can go beyond the traditional working ranges and
Fig. 6.17: Schematic representation of plasma-MIG
welding equipment thus considerably raise the productivity. This is the
thinking behind a concept for high productivity
Table 6.21: Melting rates of plasma arc welding with filler wire and plasma-MIG welding
Plasma arc Plasma arc Filler wire MIG-arc Diameter of Melting rate of
current, voltage, current, voltage, filler wire, filler wire,
amp V amp V mm g / min
110 29 — — 0.9 22
135 30 — — 0.9 28
160 32 — — 0.9 33
190 34 — — 0.9 40
190 37 100 31 1.2 85
190 38 150 32 1.2 130
w e l d i n g of thicker materials, (e.g. 15-20 m m in

Arc voltage thickness).
Figure 6.18 s h o w s the w o r k i n g ranges for
conventional GMA welding and HSW-RAPID ARC
A N D RAPID MELT. In many cases RAPID ARC
technique can more than double the welding speed,
depending on the application. The technique can also
4%>*.r'{
be performed with most existing welding equipment
and is suitable for mechanised welding as well as for
Mixed arc
Short arc
manual welding.
The RAPID MELT technique can provide very high
deposition rates — as high as 20 kg/hour. This means
the wire feed speeds of upto 50 metres per minute.
The technique, therefore, requires some investment
in the form of reliable, rapid wire feed units, and high
capacity power source. RAPID MELT is also feasible
W i r e fwM $f»ee<l for mechanised welding.
Fig. 6.18 Working ranges — Conventional GMA However, the optimal sheilding gas required for
welding and RAPID ARC/RAPID MELT welding RAPID processes are Argon rich mixtures (Typically Ar
+ 8% C 0 2 +.03% 0 2 ). Such a gas with low C 0 2 content
gives a stable arc, low spatter, low convexity and little
welding, HSW. The concept comprises two welding surface oxidation.
procedures. RAPID ARC, which aims at raising the RAPID ARC and RAPID MELT processes are being
welding speed in most plate thicknesses, and RAPID increasingly used for mechanised welding — in
MELT, which raises the deposition rate, primarily in the particular for robotic welding.
C02 Welding by A.A. Smith, published, 1970, by The Welding Institute.
Metals Handbook, Vol. 6, 8th Ed., published, 1971, by American Society for Metals.
Pulsed Current MIG Welding by Villeminot et al, Conference Proceedings of "Advances in Welding Processes", 1970, by The
Welding Institute.
Plasma-MIG Welding by Essers et al, Conference Proceedings of "Advances in Welding Processes", 1970, by The Welding
Institute.
The Twist Arc Welding Process by Kimura et al, Conference Proceedings of "Advances in Welding Processes", 1978, by The
Welding Institute.
MIG Welding in Excavators, Met. Constr., Sept. 1973.
MIG Welding of Aluminium, Weld. J., Nov. 1974.
MIG Welding of Aluminium, Weld. }., Jan. 1975.
Plasma-MIG for Welding Aluminium, Weld. & Met. Fabr., Oct. 1980.
Pulsed MIG for Welding of Aluminium, Met. Constr., Apr. 1982.
Recent Developments in Mechanised High Efficiency Gas-Shielded Arc Welding Processes WIRR 235/84, Apr. 1984.
Narrow Gap Welding by Gazes & Ducrot. Conference Proceedings of "Advances in Welding Processes", 1974, by The Weld-
ing Institute.
Flux-Cored
Arc
Welding
FLUX-CORED ARC WELDING (FCAW) is a modifi- of slag forms on the weld bead, which is easily detached
cation of MIG/C0 2 welding, in which the continuous by chipping.)The weld-metal is less liable to porosity
solid wire is replaced by a continuous flux-cored elec- and it is easier to get consistent X-ray quality welds. All
trode wire, i.e. a tubular wire filled inside with flux. these factors provide highest welder appeal to flux-
The flux performs essentially the same functions as the cored wires when compared to solid C02-shielded wires
flux coating of a manual metal-arc welding electrode. and manual electrodes.
They are: Gas consumption is also much less. While in solid
1) To provide shielding gas through chemical de- wire C0 2 welding, the rate of gas flow is 171/min, it is
composition. However, auxiliary shielding gas is used 8-12 1/min with flux-cored wires.
to ensure optimum weld-metal quality. The most striking advantage of flux-cored wires is
2) To act as deoxidisers or scavengers, which will that they eliminate the consumable goods
help purify and produce a sound weld-metal. manufacturer's continuous dependence on the
3) To form a slag, which will float on the molten steelmaker. In the case of solid wires, the steelmaker
weld-metal and protect it from the atmosphere during must first provide the billets of required chemistry.
solidification. These then must be rolled down to rods of suitable
4) To act as arc stabilisers, which will produce a diameter in coil form by a rolling mill. High cost of these
smooth welding arc and reduce spatter. operations can be justified only when the demand is
5) To add alloying elements to the weld-metal, large and regular.
which will increase strength and provide other desir- Since flux-cored wires use a standard tube material
able weld-metal properties. and achieve the required chemistry through alloy
powders introduced into the core, the producer can
Advantages of FCAW supply even small batches of wires of any desired
Flux-cored arc welding overcomes all the composition at relatively short notice. Continuous
shortcomings of manual metal-arc welding, such as low hardfacing wires can also be easily produced, which in
arc time, low deposition rate, stub loss, factors of the case of solid wires would be a difficult, sometimes
welder's skill and fatigue. The equipment provides the even impossible proposition, because foiling and
operator with automatic wire-feed and arc control, drawing would be involved.
leaving him to provide guidance and relative motion For welding low-alloy steel pipes in 5G position (in-
or travel for the torch. volving 360° welding), welders find it easier to use flux-
Flux-cored wires have several advantages over solid cored wires than solid wires. In welding high strength
wires used in the gas-shielded welding process. Spatter steels where mechanical properties are affected by heat
is considerably reduced and weld surface finish is vastly input, it becomes possible to increase alloy content as
improved, because of the arc-stabilising and slag- the wire diameter increases, so as to obtain uniform
forming compounds present in the core. (A thin layer mechanical properties over a wide range of heat input.
EQUIPMENT otherwise distort during feeding. The puslvtype wire

The basic equipment for flux-cored arc welding is feeding is usually employed, that is, the electrode is
essentially the same as for C0 2 welding: DC constant pulled from a reel by drive rolls and pushed through a
potential power source, a wire feeder, a means to supply flexible conduit to the welding gun and onwards to the
shielding gas to the arc, a welding gun and a control arc. A four-roll feeder in which all the rolls are driven
box. Typical equipment is shown in Fig. 7.1. (see Fig. 7.2) is more effective than a two-roll system,
Constant-potential power source is preferred, though the latter is also sometimes employed with
because it makes the control system simpler. The one-roll being driven, while the other is a pressure
electrode wire is fed into the arc at a required constant roll. The design allows easy adjustment of roll speed
rate, and it automatically draws the amount of current and pressure and quick changing of the rolls. It is
required to maintain the pre-set arc voltage. If wire-feed important to select rolls compatible with the wire size
rate is increased, current increases and consequently the being used.
weld deposition rate increases. For medium to large diameter electrode wires (1.6
Like MIG/C0 2 welding, flux-cored arc welding is mm and above) V-groove knurled rolls are used as
primarily a semi-automatic process, in which the gun shown at (B) in Fig. 7.2. Grooved-gear rolls (C) are better
is held by the operator and manipulated along the seam. suited for soft electrode wires. Concave smooth-faced
It can also be used as an automatic process for special rolls (D) are preferred for small diameter wires (1.2 mm
applications, where the operator continuously monitors and below) which are less liable to flatten than the larger
the operation during mechanised travel of the seam or ones.
the welding head or both. Wire feeders are normally mounted on the power
source, though there are instances where they are
Wire-Feeders several metres away from the power source. Sometimes
Drive rolls of wire feeders are specially designed to they are mounted on overhead jib cranes or booms, so
ensure that the flux-cored electrodes do not flatten, or that the welder can operate over a large area.
DIRECT CURRENT
CONSTANT VOLTAGE
POWER SOURCE
fx
CrT
-VOLTAGE CONTROL TO SOLENOID VALVE»
SHIELDING
.CONTACTOR CONTROL GAS SOURCE
VOLTMETER AND
fTTI ITTI AMMETER
WIRE FEED
(CURRENT)
ONTROL
SASJN.
AS OUT
ELECTRODE POWER CABLE
GROUND CABLE
Fig. 7.1: Schematic representation of semi-automatic flux-cored arc welding equipment

FLUX-CORED ARC WELDING 3 5 9
WIRE INLET FROM REEL
ROLL(1 (5F4 )
FLEXIBLE CONDUIT TO-
ELECTRODE HOLDER
A. FOUR-ROLL PUSH-TYPE FEEDER
ARROWS INDICATE
- * _ WATER IN
—►WATER OUT SWITCH-t"
OGAS
B. V-GROOVE D. CONCAVE HAND SHIELD

KNURLED ROLLS ROLLS.
Fig. 7.2: Electrode wire-feed system and three designs

of feed rolls Fig. 7.3: Diagrammatic sketches of guns for
semi-automatic gas-shielded flux-cored arc welding
Welding Guns
These are either gas-cooled or water-cooled. For
welding currents of 500 amp or more, water-cooled guns
must be used. Under Indian tropical conditions, it is
advisable to prefer these types for currents above 350
amp. Gas-cooled guns are kept cool by radiation of heat
to the surrounding air and by the shielding gas, which
is quite cold, as it passes through it. Gas-cooled guns
must be preferred where conditions permit, because
they are lighter in weight, more compact, and demand
less maintenance. Typical gas-cooled and water-cooled
guns for semi-automatic welding are shown in Fig. 7.3.
They may have straight or curved nozzles, the latter
vary-ing between 40 and 60°. Curved nozzles are more SIX STYLES OF
FLUX- CORED ELECTRODES
flexible and make electrode manipulation easier in some
applications.
FLUX-CORED ELECTRODE WIRES

6.3 MM-DIAM
FLUX-COVERED
ELECTRODE
O
1.6 MM-DIAM
2.4 MM-DIAM
FLUX-CORED SOLID ELECTRODE
The electrode wires meant for the welding of mild ELECTRODE
steel, medium tensile steels and low-alloy steels, and
for some of the hardfacing applications consist of a low- Relative sizes of electrodes for use at nominal
carbon steel sheath surrounding a core of fluxing and welding current of 400 amp
alloying materials. The common core elements and their Fig. 7.4: Various designs of flux-cored electrodes, and
functions are given in Table 7.1. size comparison of flux-covered, flux-cored and solid
Various forms of flux-cored wires are available and metal electrodes for use at the same nominal
six of them are shown in Fig. 7.4. The three top ones are welding current
Table 7.1: Common core elements in flux-cored electrodes

Element Usually present as Purpose in weld
Aluminium Metal powder Deoxidise and denitrify
Calcium Minerals such as fluorspar (CaF2) and Provide shielding and form slag
limestone (CaC03)
Carbon Element in ferroalloys such as Increase hardness and strength
ferromanganese
Chromium Ferroalloy or metal powder Alloying to improve creep resistance, hardness,
strength and corrosion resistance
Iron Ferroalloys and iron powder Alloy matrix in iron base deposits, alloy in nickel
base and other non-ferrous deposits
Manganese Ferroalloy such as ferromanganese or Deoxidise, prevent hot shortness by combining with
as metal powder sulphur to form MnS; increase hardness and
strength; form slag
Molybdenum Ferroalloy Alloying to increase hardness, strength, and in
austenitic stainless steels to increase resistance to
pitting type corrosion
Nickel Metal powder Alloying to improve hardness, strength, toughness
and corrosion resistance
Potassium Minerals such as potassium bearing Stabilise the arc and form slag
feldspars and silicate and in frits
Silicon Ferroalloy such as ferrosilicon or Deoxidise and form slag
silicomanganese; mineral silicates such
as feldspar
Sodium Minerals such as potassium bearing Stabilise the arc and form slag
feldspars and silicates and in frits
Titanium Ferroalloy such as ferrotitanium; in Deoxidise and denitrify; form slag; stabilise
mineral, rutile carbon in some stainless steels
Zirconium Oxide or metal powder Deoxidise and denitrify
Vanadium Oxide or metal powder Increase strength
Note: Denitrifying elements have an important function in gasless flux-cored wires. They are not required in gas-shielded wires.
the most common and their steel content is about 75- the atmosphere in long storage. The finished wire may
85% of the total weight and about 75% of the cross- also not be perfectly circular.
sectional area of the electrode. They are made by passing Seamless flux-cored wires are produced from a
low-carbon steel strip through contour-forming rolls seamless tube. They are marketed under the trade
that bend the strip into a U-shaped cross section. The names Automig FC and Automelt FC. The former is
U-shaped product is then filled with a measured amount designed for flux-cored arc welding and the latter for
of granular core material (flux) by passing it through a submerged-arc welding. The unique feature of Automig
filling device. Next, the flux-filled U-shaped strip passes FC wire is that it is perfectly circular and the core
through closing rolls that form it into a tube and tightly material is permanently sealed against moisture pick-
compress the core materials. The tube is then pulled up in storage. As a result, Automig FC gives consistent
through drawing dies that reduce its diameter and extra-low hydrogen weld metal, which is not the case
further compress the core materials. The drawing with the conventional open-seam wires. For the same
operation secures the core materials inside the tube. The reason, Automig FC wires are ideal filler metals for
electrode may or may not be baked during or between m e c h a n i s e d u n d e r w a t e r w e l d i n g . The perfectly
drawing operations, depending on its type. Additional symmetrical structure of the wire corss-section with its
drawing operations are performed to produce different consistently thick metal annulus permit equally good
sizes of electrode wire. The finished electrode wire is bending in all directions. The wire is twist-free and does
wound into a continuous coil or onto spools, as required. not tend to open along its longitudinal axis, nor is it
In these wires, the seam remains unsealed, and hence liable to deform as a result of bending forces. Automig
the flux in the core is liable to pick up moisture from FC is therefore suitable for welding on any MIG/MAG
FLUX-CORED ARC WELDING 361
welding set, and this applies equally to the lowest submerged-arc, inert gas shielded or covered electrode
diameter of 1 mm. deposits. This is attributed to the reduction of hydrogen,
The range of flux-cored wires is as wide as for phosphorus and possibly sulphur in the deposited weld
manual electrodes. It covers mild steel, medium strength by oxidation during welding.
and high strength steels, alloy steels, creep-resisting The use of C 0 2 shielding thus improves the weld-
steels, stainless steels and hardfacing applications. metal ductility and toughness. C 0 2 shielding also
provides deeper weld penetration. In general, flux-cored
SHIELDING GAS welding with C 0 2 has penetrating qualities much
Most electrode wires are designed for use with an greater than low-hydrogen iron powder electrodes, but
external shield of carbon dioxide (C0 2 ). C 0 2 gas is will not penetrate as deeply as C 0 2 welding with a solid
inexpensive, gives deep penetration and globular metal electrode wire at the same welding current range.
transfer across the arc. The function of C 0 2 in the arc is When C 0 2 is replaced by A r / C 0 2 mixture, the metal
explained below. It is important to point out that transfer is spray-type and penetration is moderate. In
alloying elements like Cr, Ni, Mo in low-alloy weld the case of wires primarily designed for 100% C 0 2
metals are not affected by the oxidising atmosphere of shielding, the Mn and Si levels in the weld metal will
the C 0 2 gas. go up, since the arc atmosphere is now less oxidising.
The C 0 2 dissociates in the welding arc into carbon The rise can be about 10% for Mn and about 30% for
monoxide and oxygen. A considerable amount of Si. This rise in Si level will have no effect on the weld-
oxygen is therefore present in the arc to oxidise metals. metal impact properties, as long as the Si content does
Molten iron also reacts with C 0 2 producing iron oxide not exceed 0.50%.
and carbon monoxide in a reversible reaction: Weld-metal d e p o s i t e d with A r / C 0 2 mixture
Fe + C 0 2 < » FeO + CO generally has higher tensile and yield strengths than
Deoxidising agents (Al, Ti, Zr, Si) are added to the weld-metal deposited with 100% C 0 2 shielding.
core to compensate for the oxidizing effect of the C0 2 . The mixture is mainly used for out-of-position
Carbon monoxide may also dissociate further into welding of pipes, especially low-alloy steel, because of
carbon and oxygen. This carbon may dissolve in the better arc characteristics and greater operator appeal.
weld-metal to increase its carbon content. Some electrode wires are designed exclusively for
Actually, the C 0 2 a t m o s p h e r e can serve as a use with a r g o n / C 0 2 mixture (80% Ar/20% C0 2 ) while
carburising or a decarburising medium depending upon many can be used with this mixture as well as with 100%
the original carbon content in the base metal and filler C0 2 . The manufacturer's catalogue must be consulted
metal. Carbon content in the weld deposit usually ranges for the selection of the appropriate shielding gas.
between 0.05 and 0.12%. If the carbon content in the
electrode metal and base metal is lower than this range, SELF-SHIELDED FLUX-CORED ARC WELDING
the weld-metal will tend to pick up carbon from the Specially formulated flux-cored electrodes have been
C 0 2 shielding atmosphere. developed in recent years, which generate a shielding
On the other hand, if the carbon content of the base atmosphere through combustion and decomposition of
materials and electrode metal is greater than 0.12%, the the flux-core compounds in the arc and make auxiliary
weld-metal will show a tendency to lose carbon. This gas shielding unnecessary. They are known as self-
loss of carbon can be attributed to such reactions as: shielding or gasless flux-cored electrodes, and the
FeO + C < » Fe + CO method of using them is called self-shielded flux-cored
When this reaction occurs, the carbon monoxide arc welding.
forms as a gas which can be trapped in the weld deposit Since the electrode wires have to operate in the
and form porosity. The tendency is avoided by keeping absence of an external gas shield, the core must contain
the level of the deoxidising elements in the cored a large amount of deoxidisers as well as denitrifiers. The
electrode sufficiently high. commonly used denitrifiers are Al, Ti and Zr (see Table
Oxygen will react with the deoxidising elements in 7.1). With Al, aluminium nitride inclusions occur in the
preference to carbon. This reaction results in a solid weld metal and they lead to low toughness. Excessive
material that floats on the weld surface instead of a gas Al (in some commercial wires Al is often more than 10%
that could form porosity. of the core material) also promotes bainitic structure in
An advantage of C 0 2 shielded arc welds is that they the weld. Ti also promotes bainite formation, while Ti-
have greater hot-cracking resistance t h a n either killed welds have high density of inclusions, which
affects toughness. Where impact requirements have to be Self-shielding wires generate large amounts of fumes
met, gasless wire must be selected with due care. A few as shown in Fig. 7.5 (C) and hence for semi-automatic
types are being marketed with guaranteed impact, but welding, fume extraction equipment is a must. In the
they are expensive since they have to contain Zr and Ni. U.S.A. and other advanced countries, special welding
Use of self-shielding electrode wires makes the entire guns with a built-in fume extraction arrangement are
equipment simple, since one can dispense with gas available.
cylinders, nozzles, valves, etc. The welding gun can also A typical gun of this type is shown in Fig. 7.5(D).
be simple. Figure 7.5 showrs the difference between the Naturally, simplicity achieved through elimination of
nozzle of a gun for a normal shielded process (A) and gas shield is nullified to a large extent through the need
for the gasless process (B). In the former, the contact for this additional device. On the other hand, self-
tube extends nearly to the end of the gas cup, so that shielding wires lead to easier automation, where
the electrical stick-out (distance from the end of the welding fumes do not matter.
contact tube to the weld puddle) is nearly the same as Self-shielding wires have made the biggest impact
the visible stick-out (distance from the end of the gas in the hardfacing field, where some amount of porosity
cup to the weld puddle). The distance from the end of the and lack of toughness in the weld deposit can be
contact tube to the end of the gas cup is usually 9.5 m m ignored.
or more. Electrical stick-out is approximately 19 r 25 mm. For general surfacing, 2.8 m m wire size is commonly
For the self-shielded method, the electrical stick-out used. For critical work, a small diameter wire with C 0 2
is usually 65-75 m m as shown at (B). shield is used. Custom-built fully-automatic hardfacing
This leads to better accessibility in narrow corners. equipment using gasless flux-cored wires have been
Since no gas cup is necessary, an insulated wire guide designed and are in use in the U.S.A. for typical
can be provided at the end of the gun, as shown, to applications like blast furnace bells and hoppers, impact
prevent the wire from touching the end or nozzle crushers, gyratory crushers and crusher rolls.
portion of the gun. The longer electrical stick-out causes
electrical current to flow through a longer length of the RECENT DEVELOPMENTS
electrode, which increases its preheat and results in Self-shielding wires were introduced in the market
higher weld deposition rate. more than 35 years ago, but in spite of being very
This method is not suited for spray-type technique, convenient to use, their popularity did not increase,
because when the fine spray is formed, the total surface because of their limited mechanical properties and their
area of the metal particles is large, and with the metal inability to operate in vertical and overhead positions.
being transferred from outside of the electrode to the These inadequacies have been recently overcome and
weld p u d d l e , the oxidation in this area cannot be today the process is widely used for all-position welding
prevented. Self-shielded welding is therefore performed in all types of code work including high rise buildings,
with globular or short-circuiting metal transfer. bridges, shipbuilding and offshore drilling platforms.
Mechanical properties, especially ductility and For the welding of offshore drilling platforms in the
impact strength at low temperatures, of welds deposited N o r t h Sea, n e w s t a n d a r d s w e r e set in 1972 a n d
by self-shielded method are considered to be lower than upgraded in 1980 both for the parent metal and the
those of welds deposited by the standard gas-shielded weld-metal. The steels to be used included BS 4360-50D
method. Lower impact strength is caused by the and BS 4360-50D modified. (The modified grade
presence of gas-forming and deoxidising elements in indicates a vacuum degassed steel which improves the
the core of the electrode wire as explained earlier. t h r o u g h thickness p r o p e r t i e s for n o d e barrel
Gasless wires are mainly used for fillet welds in the fabrication.) The steel types and the weld-metal
downhand position, and to a limited extent for butt toughness requirements since 1980 are shown in Table
welds and in vertical position. Owing to the self- 7.2. Self-shielded wires are now available, which meet
shielding property, the performance is not affected by these Charpy-impact and CTOD test requirements for
windy weather (gas-shielded wires need protection offshore platforms. A leading producer of such wires
against wind in outdoor work), and hence these wires has pointed out the following advantageous features:
possess a definite advantage for site fabrication. Good welds can be made semi-automatically at
Self-shielding wires gives less penetration compared higher deposition rates and with a higher operator factor
to gas-shielded types. Hence they will bridge wide gaps (per cent of total time actually spent depositing weld-
more easily. metal) than is possible when using MM A welding.
CONTACT
TUBE
INSULATED
CONTACT GUIDE WIRE
TUBE
FLUX-CORED
ELECTRODE ELECTRICAL
RICAL STICK-OUT
FLUX-CORED^H
STICK-CUT
ELECTRODE
Fig. 7.5: Comparative designs of (A) nozzle for use with auxiliary gas shielding, and (B) nozzle for use with self-
shielding. Fume generation in welding with self-shielding wires using (C) a normal gun, and (D) a gun having built-
in fume extraction arrangement
Table 7.2: Offshore structure requirements in 1980

Steel type Application area Condition Weld-metal property
requirements
Charpy-V BS 5762
notch CTOD, mm
BS 4360-50D Flotation tanks, brace tubulars, Normalised, fully -40°C -10° C
skid beams, deck trusses, killed 35 J 0.44
leg tubulars (50 mm) (0.26, PWHT)
BS 4360-50D Node barrel tubulars, man deck Vacuum degassed, -40°C -10°C
Modified plate, girders, lifting padeyes, normalised, with 35 J 0.44
water-tight diaphragms sulphur < 0.008 (0.26, PWHT)
API 5LGrB Secondary structural members Normalised -20°C None
< 600 mm 27 J
Further cost savings are realised by the user, since there and 22-storey buildings in Atlanta and Florida gave
is only about 60% electrode efficiency when using MMA approximately three times the weld-metal deposition
welding, while self-shielded wires are typically 80-90% rate of the other processes that can be used in the field.
efficient. The weld penetration and appearance were excellent.
In addition to increased production rates, experience The ability to complete a weld with fewer passes
has shown that repair rates for MMA welding range increased productivity. Stub losses were negligible and
around 4%, while a rate of 1.-5% or less can be expected the electrodes did not require redrying ovens for
when using self-shielded wires. The main reason is reconditioning, as would be the case with manual
fewer stops and starts with a semi-automatic wire, and electrodes".
higher penetration which allows the operator to burn There are two classification groups depending on
through any slag not removed during weld cleaning. the tensile level, namely, E6 and E7. E6 group has nine
This digging arc makes it easier to obtain good standard types, E7 group has 12. They are listed in Table
sidewall fusion and root penetration, giving a back bead 7.3 along w i t h their all-weld tensile a n d impact
with both good appearance and strength. Because of requirements in the as-welded condition.
this arc characteristic and weld pool control, training of Electrodes covered by this specification are capable
operators is simplified. of producing weld deposits that meet most radiographic
When fabricating structures as large as North Sea quality requirements.
offshore rigs, root gaps may sometimes range up to 12 The specification classifies 12 different types of
mm. It may seem contradictory to say that wires with electrodes with the following different suffixes: T-l, T-
good penetration are also good at handling poor fit-up, 2, T-3, T-4, T-5, T-6, T-7, T-8, T-10, T - l l , T-G and T-
but the nature of self-shielded wires and the pool control GS.
afforded to the operator enable wide gaps to be handled Each suffix indicates a general g r o u p i n g of
relatively well. electrodes, w h i c h contain similar flux or core
Self-shielded flux-cored wire welding has been used components and have similar performance and usability
with great success in the erection of the steel structures characteristics.
of skyscrapers in the U.S.A. The steel erector has For T-GS, T-2, T-3 and T-10, chemical requirements
described the advantageous features of the process thus: of the deposited weld-metal are not specified, because
"The mechanised welding used on 52-storey, 32-storey these types are meant for single-pass welding with
STANDARDS ON FLUX-CORED WIRES

The American Welding Society has issued the following three important standards:
AWS A5.20-95, "Specification for carbon steel electrodes for flux-cored arc welding"
AWS A5.22-95, "Specification for flux-cored Cr and Cr-Ni steel electrodes"
AWS A5.29-98, "Specification for low-alloy steel electrodes for flux-cored arc welding"
A5.20
The method of AWS classification is illustrated below:
Designates an electrode.
Indicates the minimum tensile strength of the deposited weld-metal in a test weld
made with the electrode and in accordance with specified welding condition.
1 I -Indicates the primary welding position for which the electrode is designed:
EXXT-X 0 — flat and horizontal positions
1 — all positions
-Indicates usability and performance capabilities.
Indicates a flux-cored electrode.

Note: The letter "X" as used in this figure and in electrode classification designations in this specification substitutes for specific designa-
tions indicates by this figure.
Table 7.3: AWS classes of flux-cored wires with mechanical property requirements
AWS UTS YS % Eln. in V-notch impact,
classification MPa, min MPa, min 50 mmi, min min
E6XT-C 415 330 22 n.r.
E6XT-GS 415 n.r. n.r.
E7XT-1 480 400 22 27Jat-18°C
E7XT-2 480 n.r. n.r.
E7XT-3 4S0 n.r. n.r.
E7XT-4 480 400 22 n.r.
E7XT-5 480 400 22 27Jat-29°C
E7XT-6 480 400 22 27Jat-29°C
E7XT-7 480 400 22 n.r.
E7XT-8 480 400 22 27 J at -29°C
E7XT-10 480 n.r. n.r.
E7XT-11 480 400 20 n.r.
E7XT-G 480 400 22 n.r.
E7XT-GS 480 n.r. n.r.
Notes : a) n.r. means not required; b) Single values shown are minimums; c) Classifications E6XT-GS, E7XT-2, E7XT-3, E7XT-10 and E7XT-GS are
intended for single-pass welding. Only transverse tension and longitudinal guided bend tests are required; d) For each increase of one percentage point in
elongation over the minimum, the minimum required yield strength or the tensile strength or both may decrease 7 MPafor a maximum reduction of 14 MPa
in either the required minimum yield strength or the tensile strength or both.
resultant high dilution. For the rest, the chemical • Class T-3: Electrodes of the T-3 classification are
composition of the weld-metal must be self-shielded, used on DC with positive polarity, and
C — not specified, but must be determined have a spray type transfer. The slag system is designed
Mn — 1.75% max. Cr —0.20% max. to give characteristics, which make possible very high
Si — 0.90% max. Mo — 0.30% max. welding speeds. They are used to make single-pass
Ni — 0.50% max Al —1.8% max. welds in the flat and horizontal positions on sheet metal
(for self-shielded types only) up to 4.8 mm thick.
S — 0.03% max. P — 0.04% max. • Class T-4: Electrodes of the T-4 classification are
A5.20 gives the following descriptions of the standard self-shielded, operate on DC with positive polarity, and
classes: have a globular type transfer. They give high deposi-
• Class T-1: Electrodes of the T-1 group are classi- tion rates and desulphurise the weld metal to a very
fied with C 0 2 shielding gas by this specification. How- low level, which helps make the weld deposit very re-
ever, argon-C0 2 mixtures are also used to improve us- sistant to cracking. These electrodes are designed for
ability, especially for out-of-position applications. De- low penetration, enabling them to be used on weld joints
creasing amounts of C 0 2 in the argon-C0 2 mixture will with poor fit-up, and for single and multiple-pass weld-
increase manganese and silicon in the deposit and may ing in the flat and horizontal positions.
improve the impact properties. The T-1 electrodes are • Class T-5: Electrodes of the T-5 group are de-
characterised by a spray transfer, low spatter loss, flat signed to be used with C 0 2 shielding gas (argon-C0 2
to slightly convex bead configuration, and a moderate mixtures may be used as in the T-1 types) for single-
volume of slag, which completely covers the weld bead. and multiple-pass welding in the flat position and for
Most electrodes in this group have a rutile base slag. horizontal fillets. These electrodes are characterised by
• Class T-2: Electrodes of this classification are es- a globular transfer, slightly convex bead configuration,
sentially T-1 electrodes with higher manganese or sili- and a thin slag, which may not completely cover the
con or both and are designed primarily for single-pass weld bead. Electrodes in this group have a lime-fluo-
welding in the flat position and for horizontal fillets. ride base slag. Weld deposits produced by electrodes
The higher levels of deoxidisers in these electrodes al- of this group have improved impact properties and
low single-pass welding over scaled or rimmed steel. crack resistance in comparison to the rutile types.
The arc characteristics and deposition rates are similar • Class T-6: Electrodes of the T-6 classification are
to those of the T-1 electrodes. self-shielded, operate on DC with positive polarity, and
have a spray-type transfer. They give very good low- • Class T-G: The EXXT-G classification is for new
temperature impact properties, deep penetration be- multiple-pass electrodes, which are not covered under
yond the root of the weld, and excellent deep groove any of the presently defined classifications.
slag removal. They are used for single- and multiple- • Class T-GS: The EXXT-GS classification is for new
pass welding in the flat and horizontal positions. single-pass electrodes, which are not covered under any
• Class T-7: Electrodes of the T-7 classification are other presently defined classification.
self-shielded and operate on DC with negative polar-
ity. Large sizes are used for high deposition rates and A5.29
the smaller sizes for all-position welding. The slag sys- The system of classification is similar to the one used
tem is also designed to desulphurise the weld metal to in A5.20, except that an additional suffix is used to
a very low level, which helps make the weld deposit indicate the weld-metal chemistry as has been done in
resistant to cracking. They are used for single- and mul- AWS A5.5, "Specification for low-alloy steel covered arc
tiple-pass welding. welding electrodes", Chapter 3. The A5.29 classification
• Class T-8: Electrodes of the T-8 classification are system describes low-alloy steel flux-cored electrodes
self-shielded and operate on DC with negative polar- as exhaustively as A5.5 describes low-alloy steel covered
ity. They are suitable for all-position welding. The slag electrodes and covers 52 classes as shown:
C-Mo types:
E70T5-A1, E80T1-A1, E81T1-A1
Cr-Mo types:
E81T1-B1, E80T5-B2L, E80T1-B2, E81T1-B2, E80T5-B2, E80T1-B2H.E90T1-B3L, E81T1-B1L, E80T1-B2L
E90T1-B3, E91T1-B3, E90T5-B3, E100T1-B3, E90T1-B3H, E80T5-B6, E80T5-B6L, E80T5-B8, E80T5-B8L,
Ni-steel types:
E70Tg-Nil, E80T1-N11, E81T1-Nil, E80T5-M1, E71T8-N12, E80T1-N12, E81T1-N12, E60TI-N11, E61T1-Nil,
E70T6-N1, E80T11-N13
E80T5-M2, E90T1-N12, E91T1-N12, E80T5-M3, E90T5-N13,
Mn-Mo types:
E91T1-D1, E90T5-D2, E100T5-D2, E90T1-D3,
All other types:
E80T5-K1, E70T4-K2, E71T8-K2, E80T1-K2, E90T1-K2, E91T1-K2, E80T5-K2,
E90T5-K2, E100T1-K3, E110T1-K3, E100T5-K3, E110T5-K3, E110T5-K4,
E111T1-K4, E120T5-K4, E120T1-K5, E61T8-K6, E71T8-K6, E101T1-K7,
EXXXTXG, E80T1-W,
system is also designed to produce very good low-tem- The final suffixes Al, Bl, B2, B2L, Dl, D2 and D3 in
perature impact properties in the weld metal, and to A5.29 denote weld-metal chemistry similar to those in
desulphurise the weld metal to a very low level, which A5.5. The suffixes N i l , Ni2 and Ni3 used in A5.29
helps resist weld cracking. They are used for single- and correspond respectively to the covered electrode classes
multiple-pass welding. C3, Cl and C2. Table 7.4 gives a comparison of the two
• Class T-10: Electrodes of the T-10 classification and will help the fabricator to decide which flux-cored
are self-shielded and operate on DC with negative po- electrode type to use in place of a low-alloy manual
larity. They are used for making single-pass welds on electrode he has been using.
material of any thickness in the flat and horizontal po- Classes K2 to K7 of A5.29 are e q u i v a l e n t to
sitions at high travel speeds. classes E9018-M to E12018-M of A5.5, as far as all-
• Class T - l l : Electrodes of the T - l l classification weld mechanical properties are concerned, because
are self-shielded and operate on DC with negative po- t h e y a r e m e a n t to b e u s e d on s t e e l s , w h e r e
larity, and have a smooth spray-type arc. They are suit- mechanical requirements have to be met and weld-
able for all-position welding and high travel speeds. metal chemistry is a secondary issue. Since the heat
They are used as general purpose electrodes for single- i n p u t in m a n u a l w e l d i n g a n d in flux-cored arc
and multiple-pass welding in all positions. welding is different, the chemistries of flux-cored
Table 7.4: Comparison of A5.5 and A5.29 classes

Nominal deposit chemistry AWS classification
A5.5 A5.29
0.5% Mo
C-Mo steel electrodes: E7010-A1, E7011-A1, E7015-A1, E70T5-A1, E80T1-A1, E81T1-A1
E7016-A1, E7018-A1, 7020-A1,
E7027-A1
Cr-Mo steel electrodes:
0.5% Cr-0.5% Mo E8016-B1, E8018-B1 E81T1-B1
0.05% max. C, 1.25% Cr, 0.5% Mo E8015-B2L, E8018-B2L E80T5-B2L
0.12% max. C, 1.25% Cr, 0.5% Mo E8016-B2, E8018-B2 E80T1-B2, E81T1-B2
0.10-0.15% C, 1.25% Cr, 0.5% Mo None E80T5-B2, E80T1-B2H
0.05% C max., 2.25% Cr, 1.0% Mo E9015-B3L E9018-B3L E90T1-B3L
0.12% C max., 2.25% Cr, 1.0% Mo E9015-B3, E9016-B3, E9018-B3 E90T1-B3, E91T1-B3
0.10-0.15% C, 2.25% Cr, 1.0% Mo None E90T5-B3, E100T1-B3, E90T1-B3H
0.5% Ni, 0.5% Cr, 0.5% Cu E8018W E80T1-W
Ni-steel electrodes:
0.12% C max., 1.0% Ni E8016-C3, E8018-C3 E71T8-Nil, E80T1-Nil,
E81T1-Nil, E80T5-Nil
0.12% C max., 2.25% Ni E8016-C1, E8018-C1 E71T8-Ni2, E80T1-M2,
E80T1-Ni2, E80T5-Ni2, E90T1-M2,
E91T1-Ni2
0.12% C max., 3.25% Ni E8016-C2, E8018-C2 E80T5-Ni3, E90T5-Ni3
Mn-Mo steel electrodes:
1.5% Mn-0.35% Mo E9015-D1, E9018-D1 E91T1-D1
1.8% Mn-0.35% Mo E10015-D2, E10016-D2, E10018-D2 E90T5-D2, E100T5-D2
1.4% Mn-0.5% Mo E8016-D3, E8018-D3 E90T1-D3
Ni-Mo steel electrodes:
1% Ni-0.5% Mo E8018-NM E80T5-K1
K classes and the coated M classes are not directly or core components that produce distinctive welding
comparable. characteristics and similar slag systems. The main
Looking closely at K classes in A5.29, one finds that characteristics of these types are given in Table 75.
some of them can have a multiplicity of tensile strength. Like A5.5, A5.29 gives details of base metal plates,
For example, the K-2 class is capable of producing weld- preheat and interpass temperatures, pass sequences and
metal tensile strengths of 483, 552 or 621 MPa within stress-relief temperatures wherever applicable, which
the chemistry range of the class (as indicated by classes are required for preparing welded test assemblies for
E7XXX-K2, E8XXX-K2, E9XXX-K2). This is partly due determining all-weld mechanical properties. However,
to the fact that the specification for low-alloy flux-cored there are differences in the two specifications in these
electrodes places a range on tensile strengths, as details, which must be carefully noted.
opposed to the m i n i m u m values imposed by the It must also be borne in mind that Tl type flux-cored
specification for low-alloy covered electrodes. On the electrodes are not listed for any classification where
other hand, the low-alloy covered electrodes in the M tensile and charpy-V testing are done after stress-relief.
and M - l classes must exhibit yield strength within a This is because the weld deposit of Tl type, which has
narrow range. Low-alloy flux-cored wire electrodes in rutile-type flux, loses more of its notch toughness after
the K classes m u s t meet only a m i n i m u m yield stress-relief than a T5 deposit having similar chemistry.
requirement (see Table 7.6). T5 type which has a basic flux is considered unsuitable
The A5.29 classification shows that there are only for positional w e l d i n g , while Tl type operates
four general types of low-alloy flux-cored electrodes: satisfactorily in all welding positions, w h e n wire
Tl, T4, T5 and T8. Each of these suffixes indicates a diameters are 1.6 mm or less. On the other hand, basic-
general grouping of electrodes that contain similar flux coated low-alloy manual welding electrodes (XX15-X,
Table 7.5 : Main characteristics of types in A5.29

Type Polarity Shielding gas Remarks
Tl Electrode positive co 2 ( a ) All-position capability is common in 1.6 mm and
smaller diameter; usually have rutile-based slag
systems
T4 Electrode positive None Globular transfer; usable in flat and horizontal
positions only
T5 Electrode positive co 2 ( a ) Globular transfer; usable in flat and horizontal
positions; better crack resistance and impact
properties than Tl; usually have basic slag
system
T8 Electrode negative None All position capability; good crack
resistance and impact properties
Note:<a) Argon-C02 mixtures may be used where recommended by the manufacturer.
Table 7.6 : Tensile requirements as per AWS A5.29

AWS classification UTS YS % Eln. in 50 m m
MPa min MPa min mm
E6XTX-X 410 to 550 340 22
E7XTX-X 490 to 620 400 20
E8XTX-X 550 to 690 470 19
E9XTX-X 620 to 760 540 17
E10XT-X 690 to 830 610 16
E11XT-X 760 to 900 680 15
E12XT-X 830 to 970 750 14
EXXXTX-G Properties as agreed upon between supplier and purchaser
Note: Properties of electrodes that use external gas shielding (EXXTI-X and EXXT5-X) vary with gas mixtures. Electrodes
classified under this specificaiion should not be used with gases other than those listed in Table 7.5 without first consulting the
manufacturer.
XXI 6-X classes) offer superior low impact properties strength within a deposit chemistry classification. It is
along with all-position capability. therefore a necessary condition that both the mechanical
Comparing the C groups of A5.5 and Ni groups of property code and the chemistry code must be included
A5.29, the Tl type flux-cored electrodes must have in the low-alloy flux-cored electrode classification to
higher Ni content in order to meet comparable charpy- make it complete.
V impact values. Within A5.29, only T5 types meet Another noteworthy point is that, the self-shielding
v a l u e s of 27 J at -59°C a n d -75°C. The lowest flux-cored electrodes in A5.29 are restricted to E6XTX-
temperature impact property listed either in A5.5 or X and E7XTX-X classifications. This suggests that for
A5.29, i.e. 27 J at -100°C is attained only with low-alloy welding high strength steels, auxiliary gas-shielded flux-
manual electrodes. cored electrodes must be preferred.
Mechanical property requirements of flux-cored
electrodes are shown in Table 7.6. It shows that the AWS A5.22
electrodes must meet a minimum yield strength, but This specification classifies stainless steel electrodes
must not exceed the minimum tensile strength for their on the basis of weld deposit chemistry and shielding
classification by more than 140 MPa. As pointed out medium to be used. Table 7.7 shows the shielding
earlier while referring to the K series, the deposit designations in the classification and corresponding
chemistries and the flux characteristics chosen by the current characteristics:
producer, when combined with this narrow tensile The EXXXTX-1 class uses C 0 2 shielding, which causes
strength range, can result in a multiplicity of tensile loss of oxidisable elements and slight increase in carbon
Table 7.7: Shielding and current characteristics as per AWS A5.22

AWS designations3 External shielding Current and
all classifications medium polarity
EXXXT-1 co 2 DCEP
x
(75-80%) Ar reminder C0 2 DCEP
EXXXT-4
None (self-shielded) DCEP
EXXXTX- 3
None specified Not specified
EXXXT-G
Note:" The letters XXX stand for the chemical composition (AISI type), such as 308, 316, 410 and 502.
content. The EXXXTx-3 is self-shielding type, and hence PROCESS VARIABLES

loss of oxidisable elements and nitrogen pick-up from The process variables and their effects on welding
the air occur. If this type is used at low current and with must be properly understood, in order to arrive at a
long arc length, the weld deposit can pick up excessive correct welding procedure.
nitrogen. Since nitrogen stabilises the austenite, it will
reduce the ferrite content of the weld metal. Obviously, Welding Current
to control the ferrite content of the weld deposit, the As mentioned earlier, using a constant voltage power
chemistries and Cr/Ni ratios of EXXXTX-1 and EXXXTX- source, welding current is proportional to the electrode-
3 deposits have to be different. feed rate for a given electrode size, composition and
For each classification, all-weld mechanical electrode extension. The relationship for an E70T-1
properties are specified, as also radiographic soundness general purpose mild steel electrode is shown in Fig. 7.6.
requirements. Though these electrodes are intended to With other variables kept constant, increasing the
be used on corrosion-resisting and heat-resisting steels, welding current increases weld deposit rate and
no corrosion or scale resistance tests are specified. When penetration. Low currents produce large droplet transfer
required, special tests pertinent to an intended and excessive spatter and in the case of self-shielded
application are expected to be decided upon, between electrodes, may result in nitrogen pick-up and porosity
the electrode producer and the user. in the weld deposit. Excessive currents give convex weld
beads of poor appearance.
Ferrite Control in Welds When welding current is changed, output voltage
The Appendix makes the following comments of the power supply must be suitably readjusted to
regarding control of ferrite in weld deposits: maintain optimum relationship of arc voltage to current,
The chemical composition is set up to allow adequate For a specific electrode-feed rate, an increase in electrode
latitude for the manufacturer to control the ferrite extension causes a decrease in measured welding
number (FN) of the undiluted deposit. With the current and vice versa.
EXXXTX-1 classifications using carbon dioxide shielding,
there is some minor loss of oxidisable elements and some Arc Voltage
pick-up on carbon content. With the EXXXTx-3 Arc voltage and arc length are inter-related. A long
classifications that are used without external shielding, arc means high arc voltage and a short arc means low
there is some minor loss of oxidisable elements and a arc voltage. The voltage reading on the voltmeter of the
pick-up of nitrogen, which may range from quite low power source is the sum of the voltage drops through
to over 0.20%. Low welding currents coupled with long the welding cable, the electrode extension, the arc, the
arc lengths (high arc voltages) should be avoided, workpiece and the earthing cable. When all other circuit
because they result in excessive nitrogen pick-up and elements remain constant, arc voltage will be
excessive loss in the ferrite content of the weld. proportional to the voltmeter reading.
The E307T-X, E308T-X, E308LT-X, E308MoLT-X and Too high an arc voltage caused by too long an arc
E347T-X grades are normally ferrite controlled. When will result in excessive spatter and a wide weld bead of
used with the recommended shielding gases and with irregular profile. With self-shielded electrodes, this will
reasonable and conventional welding currents and arc give rise to excessive nitrogen pick-up. Very low arc
lengths, they produce weld metal with a typical ferrite voltage caused by a very short arc gives a narrow convex
level of 4-to 14 FN. bead, excessive spatter and shallower penetration.
000
400
2 114 MM
(0.045 IN) DIAM 350
z 800
300
<
tr. 600 250
EEJ 1.59 MM
Ej d/16 IN) DIAM
Q Eg ^§J> 200
O W .^S^1.98MM 2
CO 400 f j£§r (5/64 IN) DIAM
2 38 MM 150
> ^ ^ ^«<^^ 1
4^^ >^^^^^^^-<<^^^^^^S>f7/64IN) D
'AM
100
200
« * ^ ^ ^ ^ (1/8 IN) DIAM" 50
100 200 300 400 500 600 700 800
ft — i i 1 ,, i 1 1— ■J
WELDING CURRENT. A
Fig. 7.6: Electrode feed rate versus welding current range for E70T-1 steel electrodes with C02 shielding
Electrode Extension flow is set according to the type and size of the gun
This variable is more significant in the case of flux- nozzle, distance of the nozzle from the work and draft
cored electrodes than with solid electrodes, because of conditions. For welding in still air, flow rates of 14 to 1$
the need to keep the melting and activation of the core 1/min are adequate, but in moving air and with electrode
ingredients consistent with that of the outer metallic tube extension longer than normal, flow rates of up to 26 1/
as also with arc characteristics. With other variables kept min may be required. Nozzle openings must not be
constant, too long an extension gives an unsteady arc allowed to get clogged with weld spatter.
and excessive spatter. An unduly short extension makes
the arc too long at a given voltage setting, and in the Electrode Angle
case of gas-shielded electrodes, gives rise to excessive For making butt and fillet welds in the downhand
spatter build-up in the nozzle, which interferes with the and horizontal positions, correct drag angle must be
gas flow and causes weld porosity and oxidation. As used (see Fig. 7.7), so that the arc force helps to give
stated earlier, the electrode extension or electrical stick- proper shape to the weld bead and to prevent the slag
out is usually 19-25 mm for gas-shielded electrodes and from running ahead of and getting entrapped in the
65-75 mm for self-shielded types. weld metal. With gas shielded method, the drag angle
should be between 2° and 15° and not more. With larger
Arc Travel Speed drag angle, the shielding becomes less effective. For the
As travel speed increases, the cross-section of the self-shielded method, drag angles are between 20° and
weld bead decreases a n d p e n e t r a t i o n becomes 45° and larger for thin sections. For horizontal fillet
shallower. Too low a travel speed can overheat the base welds, in addition to the drag angle, a work angle of
metal and cause burn-through in thinner plates. It also 40°^15o as shown in Fig. 7.7 at B must be used to control
results in a rough weld with slag entrapment. When the flow of the weld pool.
the travel speed is too high, the weld bead is ropy and
irregular in profile. WELDING PROCEDURES
In order to take advantage of the high current
Shielding Gas Flow densities and consequent deep penetration of FCAW,
For normal gas-shielded welding, the gas-flow rate joints are designed to have smaller groove angles,
should be correctly set. Inadequate flow results in narrower root gaps and larger root faces than are used
porous and oxidised weld. Excessive gas flow creates in manual metal-arc welding. However, care should be
turbulence and ingress of air into the gas shield. Gas taken that the root is accessible a n d permits the
TRAVEL
(DRAG)
ANGLE
WORK ANGLE
*40°to 50°
\ I / / WELDING
DIRECTION
i
ONE ELECTRODE
zzzzzzzzzzzz DIAMETER
^ W 4-
(A) (B)
Fig. 7.7: Welding electrode positions
necessary electrode manipulation during welding, and For automatic operation, the welding head may be
that a constant electrode extension can be maintained mounted on a trolley and made to traverse a track
for all the passes in the groove. running parallel to the joint. Alternately, it may be
Fillet welds made in the flat and horizontal positions mounted on a beam and operated in conjunction with
possess excellent penetration beyond the root, and hence specially built positioners and fixtures. For example, by
their sizes can be reduced by approximately 2 to 3 mm, mounting the head on a column and boom, longitudinal
w h e n design specifications and quality control and circumferential joints of cylindrical vessels can be
standards permit. automatically welded as explained in Chapter 20.
Typical FCAW procedures for welding mild steel
with E70T-1 class electrodes using DC, electrode Applications of FCAW
positive, are given in Table 7.8, which is taken from
Welding Handbook, Vol. 2, 7th Ed., p u b l i s h e d by • IN THE U.S.A.
American Welding Society. Structural Welding Code, ref. AWS Dl.l, published by
the American Welding Society has given an important
Automatic FCAW place to the flux-cored arc welding process. Many steel
Like the M I G / C 0 2 process, FCAW is used as an fabricators in the U.S.A. are using it to fabricate I-beam
automatic process for specific production applications type girders by joining flange to webs of different
and large-scale hardfacing. Typical equipment is shown structural steels. They are also using it for joining
in Fig. 7.8. To fully exploit the process, heavy electrodes stiffeners, attaching clip angles and for splicing web and
(2.4 to 4.0 mm diameter), high electrode-feed rates and flange plates.
long arc times are used. Hence the power source is Typical applications of flux-cored arc welding in
designed to provide the required high current at 100% advanced countries are, main frames on bulldozers,
duty cycle. The wire feeders and wire drive motors are bulldozer blades, gear boxes, etc., rotating frames for
of sufficiently high capacity. Welding guns are of special shovels and cranes and for power shovel booms, tractor
design. Typical guns with their nozzle details are shown frames, road rollers, motor grader frames, and other
in Fig. 7.9. The one on the left is air-cooled and designed similar heavy-duty equipment. Since the process has
for side shielding. It presents no problem of spatter been recognised as a time-saving one, it has been taken
build-up in the nozzle and can be conveniently used in up by other industries such as the container industry for
narrow, deep grooves. It is good enough for currents the manufacture of tanks of all types as well as pressure
up to 600 amp. For higher currents, the water-cooled vessels, autoclaves, etc. The metal-working industry has
type such as shown on the right must be used. also been quick to adopt the process to manufacture
Table 7.8: Typical FCAW procedures

Joint design Thickness Root gap No. of Electrode Welding power
T R passes dia.
(mm) (mm) (mm) V amp
Flat position
13 0 2 2.38 30 480
25 0 6 2.78 32 525
jr^-y. 16 5 3 2.78 32 525

25 5 6 2.78 32 525
^45-y
25 0 6 2.78 32 525
51 0 20 2.78 32 525
Horizontal position
13 3 6 1.98 28 350
25 3 18 1.98 28 350
5 0 1 1.98 28 350
13 0 3 2.78 30 450
Vertical position
13 2.4 4 1.14 22 180
25 2.4 9 1.14 22 180
3 0 1 1.14 22 180
13 0 2 1.14 22 180
DIRECT CURRENT
CONSTANT VOLTAGE
POWER SOURCE VOLTMETER AND VOLTAGE SHIELDING
AMMETER CONTROL GAS SOURCE
CONTACTOR WIRE FEED
(CURRENT)
GAS OUT
GROUND CABLE
Fig 7.8: Schematic representation of fully automatic flux cored arc welding equipment
SHIELDING
GAS
CIRCULATING
WATER
WATER-COOLED
CONCENTRIC-SHIELDED
NOZZLE ASSEMBLY
ELECTRODE
ELECTRODE
Fig. 7.9: Diagrammatic sketches of guns for fully-automatic welding with gas-shielded flux-cored electrodes
machine-tool base frames for punch presses, the crown building, the process helped to reduce the 12-month long
assemblies for bending presses, etc. The transportation schedule by six weeks. In the construction of the 10-
equipment industry also employs the process to storey 1,350 ft (415 m) high twin tower World Trade
manufacture mineral wagons, under-frames for Centre in New York City, a tremendous amount of
locomotives, automobile components such as rear axle welding has been involved. All the steel structures up
housings, truck frames, trailer frames, etc. The ship- to the ninth floor above street level and the seventh floor
building industry is using the process for the fabrication below street level are all-welded.
of tugboats, river barges, and similar weldments,
One of the first applications of flux-cored arc welding • IN EUROPE
in the U.S.A was in the field erection of a skyscraper in Figures 7.10-7.14 illustrating typical applications in
Chicago. This building — a ^5-storey, 3,800 ton, 76 ft by industry clearly show that they are used wherever high
178 ft tower — was designed for all-welded column quality requirements under production conditions have
splices and welds joining the primary beams to the to be met.
columns. The column splices were partial penetration
single-bead 45° bevel butt welds, with a penetration • IN INDIA
slightly greater than half the metal thickness. The beam- Ador Welding pioneered the use of FCAW in India
to-column joints utilized full penetration single-bevel by commencing production of Automelts FC wires in
45° butt welds between the flanges and the column. 1978, under a license agreement with Oerlikon of Zurich.
Back-up strips and a root opening were used together The range of AutoMIG FC wires, presently restricted to
with the gas-shielded flux-cored arc welding process. the welding and hardfacing of ferritic steels, is as wide
The beam web to column joint was made with manual as the range of mild steel, low-alloy and high tensile steel
covered electrodes, because the smaller diameter flux- and hardfacing manual electrodes. Most of the AutoNDG
cored wires were not available at that time. On another FC wires can be modified for use with the submerged-
arc process. The modified versions make u p the AutoMIG FC-11 has been the most popular type in
Automelts FC range, and the equivalent AutoMIG FC India. It has a rutile base and gives a soft arc and good
and Automelts FC wires carry the same two-digit number. weld finish, especially in fillet welds. It is meant to be
For example, AutoMIG FC-31 and Automelt FC-31. used for general mild steel applications. It has been
Fig. 7.10: Automelt FC welding on diesel engine chassis
Fig. 7.11: Automelt FC welding on locomotive underframe

found very suitable fw the thickness range of 4-30 mm. Tenalloy It has been used in the thickness range of 4-50
Typical applications are rubber machinery, cement mm on the following applications: excavators, sugar
machinery, material handling equipment, steel safes and machinery, heavy machine-building, mobile cranes,
almirahs, steam drums and boiler accessories. Main automobile components, valves and boiler components.
advantage is its high productivity and low-hydrogen The main considerations in its favour have been its high
characteristic, in spite of being a rutile type. productivity and the good crack resistance of the welded
AutoMIG FC-31 has a basic flux and is an excellent joints.
substitute for the manual electrodes Supabase and AutoMIG FC-18 has been used for the welding of
Corten -1 weather-resisting steel in the fabrication of
Fig. 7.13: AutoMIG FC root welding on bridge girders.

Filling runs and top layers deposited with the
Fig. 7.12: AutoMIG FC welding on flat tubes in furnaces submerged-arc process
Fig. 7.14: The largest gantry crane in Europe welded with AutoMIG FC and AutoMELT FC
electrostatic precipitators. Two engineering companies In all the above applications, pure and dry carbon
in south India have used AutoMIG FC-42 for the dioxide is used as a shielding gas. Welding is performed
welding of Tl steel, which is a high tensile quenched- in the downhand or horizontal position using standard
and-tempered steel produced in the U.S.A. The core equipment. AutoMIG FC wires are well suited for
of this electrode carries a basic flux alloyed with vertical and overhead welding, provided argon-C0 2
Cr, Ni and Mo. The weld deposit gives a tensile range mixture is used for shielding. It is expected that the wires
of 730 -790 N / m m 2 and good impact values down to will be used in these positions, especially on pipes, in
-50° C the near future.
Metals Handbook, Vol. 6, 8th Ed., published 1971 by American Society for Metals.
Flux-Cored Wire for the Welding of Steel, Weld. ]., May 1965.
Efficient Use of MIG Welding with Flux-Cored Electrodes by Lee et al., Weld. ]., July 1965.
Flux-Cored Arc Welding by Cary, Howard, B., Weld. & Met. Fabr., Nov. 1970, Dec. 1970 & Jan 1971.
Open-Arc Hardfacing by Soudometal, Weld & Met. Fabr., July 1971.
Application of Flux-Cored Welding in Germany by Weyland. Australian Weld. /., Sept./Oct. 1972.
Welding with Seld-Shielding Flux-Cored Wires, Met. Constr., Sept. 1980.
Recent Developments in Applications of Innershield, Met. Constr., Dec. 1983.
Innershield Welding, Met. Constr., Nov./Dec. 1981.
Mechanised Welding Speeds High Rise Construction, Met. Constr., June 1984.
The Development and Application of Flux-Cored Welding by Shackleton, Met. Constr., Oct. 1974.
Electroslag and
Electrogas
Welding
ELECTROSLAG WELDING A N D ELECTROGAS pool is shielded by the molten slag while in the latter, it
WELDING are two similar processes, which being quite is shielded by a gas.
different from the are welding processes discussed so It is pertinent to point out here that in submerged-
far, stand in a class by themselves. arc welding, the arc remains stable under the molten
In both these processes which are fully automatic, slag, because the arc voltage is in the range of 25-32 V
the plates to be joined are kept in the vertical position, and the slag layer is shallow. In electroslag welding, on
i.e. the axis of the weld joint is vertical. Yet the welding the other hand, the slag pool is 38-50 mm deep and it
takes place in the flat or downhand position, with offers a conductive path between the electrode and the
vertical upward travel of the weld pool. A consumable base metal. Hence the current flow is maintained after
electrode (sometimes more than one) is fed downward the arc is extinguished.
into a cavity formed by the two square edges of the plate
pieces being welded and two dams or shoes. Figure 8.1 ELECTROSLAG WELDING
illustrates the method of electroslag welding with three There are two basic versions of this process. The usual
electrodes and Fig. 8.2 illustrates electrogas welding and conventional method uses a non-consumable guide
with a flux-cored electrode. The entire weld is completed or contact tube to direct the wire electrode into the slag
in one pass and without interruption, whatever may be bath. The welding head travels steadily upward along
the thickness of the plates. with the shoes or dams as the weld is deposited. In the
The two processes differ from each other in the other version called the consumable guide method, a
manner in which the heat required for melting the wire consumable guide or contact tube is used. In this case,
and the plate edges is generated. In the electroslag the welding head remains fixed at the top of the joint,
process, an electric arc is used initially for a brief the shoes are clamped on the sides and are immovable,
moment to heat and melt the flux and to convert it into and both the guide tube and the electrode are melted
slag. The molten slag then chokes and extinguishes the by the molten slag.
arc, but the current continues to pass between the
electrode and the work via the molten conductive slag. Non-consumable Guide Method
The resistance offered by the molten slag to the electric In the conventional or non-consumable guide
current provides the heat necessary to keep the process method, a square butt joint with a gap of approximately
going. Enough heat is generated to maintain the internal 30 mm is placed, so that the axis of the weld is vertical.
temperature of the bath at approximately 1,900°C and A starting trough of full plate thickness and 75 mm
its surface temperature at approximately 1,650°C. On depth is attached to the bottom. The trough is made out
the other hand, electrogas is an arc welding process in of three scrap plate pieces tack welded together. It can
which the arc is formed between the electrode and the also be a clamp-on sump made from copper, which can
molten weld pool. In the electroslag process, the weld be used repeatedly. Run-off tabs of full plate thickness
CURVED ELECTRODE
GUIDE TUBE
ELECTRODC
MOLTEN SLAG BATH.
MOLTEN WELD POOL
WATER-COOLED SHOE
SOLIDIFIED WELD METAL
COMPLETED WELD
BASE PLATE
Fig. 8.1: Electroslag welding with three electrodes
WELDING GUN,
FLUX-CORED ELECTRODE^
BASE PLATE 1
DRIVE ROLLERS-^ MOLTEN S±.AG GAS

SHIELDING
MOLTEN WELD METAL WATER

SOLIDIFYING WELD MATAL CIRCULATION
SOLIDIFIED METAL
WATER
CONNECTIONS
BASE PLATE 2 COMPLETED WELD
Fig. 8.2: Electrogas welding with a flux-cored electrode

ELECTROSLAG AND ELECTROGAS WELDING 379
and 75 mm height are attached to the top, so that the except for some loss due to its deposition, as a thin layer
weld crater is above the top of the joint. U-shaped on the outer weld surfaces. Small amounts of flux have
restraining plates are also welded on the faces of the to be added manually to the slag bath from time to time.
joint. They should provide enough room for the water- If flux-cored wire is used, no such manual addition is
cooled shoes to move through them (see Fig. 8.3). necessary.
The welding machine is then positioned beside the When the weld is completed, the welding machine
gap between the two plates. The two dams or shoes (ap- along with the shoes is withdrawn. The starting trough
propriately called mold shoes or retaining shoes), which and the run-off tabs are removed flush with the ends of
are attached to the welding machine are designed to the joint.
move vertically upward as the machine moves. The More than one wire has to be used with increasing
dams which are water-cooled and made of copper serve thickness. If the workpiece thickness exceeds
to contain the molten weld-metal and slag, to acceler- approximately 70 mm per wire used, the wires along
ate the solidification of the weld-metal and to provide with their guide tubes need to be oscillated, i.e. moved
the desired reinforcement or outside contour to the weld horizontally back and forth in a direction parallel to the
joint. edges of the workpiece. One oscillating electrode is good
A layer of granular flux is placed at the bottom of enough for thickness up to 120 mm, two electrodes up
the trough and an arc is initiated beneath this flux layer to 230 mm and three electrodes up to 500 mm.
as is done in submerged-arc welding. As soon as a suffi-
ciently thick layer of slag is formed, the arc stops and EQUIPMENT
the process automatically becomes electroslag welding. The standard electroslag welding equipment consists
The electrode wire is fed through a wire guide, which of the following:
consists of a beryllium-copper tube for'better heat
a) Power source
resistance, with two supports brazed to it. It is curved
and narrow (usually less than 12 mm in diameter), so b) Wire feeder and oscillator
that it can travel vertically in the joint gap With sufficient c) Electrode guide tube
clearance. The tube is wrapped withj&sulating tape to d) Retaining shoes
prevent short-circuiting with the work. The guide tip is
e) Welding head with controls
maintained 50 to 75 mm above the molten slag.
The welding machine moves vertically up along with f) Travel carriage.
the shoes, and this movement is kept consistent with a) Power source: The commonly used power source
the deposition rate. The amount of slag remains constant is a constant potential welding rectifier rated for 750 or
RUNOFF TAB
(1 OF 2)
vBASE
METAL
STRONGBACK
STARTING TROUGH
SHOWN WITHOUT WEDGES AND DAMS

DIMENSIONS ARE IN INCHES
Fig. 8.3: Two 152 mm thick plates ready for electroslag welding
1,000 amp at 100% duty cycle and with open-circuit iii) Controls to oscillate the guide tubes, length of
voltage of 60 V minimum. It is similar to the one stroke and duration of dwell at each end of the
designed for submerged-arc welding. It is usually stroke
provided among other things with a contactor, a means iv) Control for the vertical rise of the head. In an au-
to control output voltage remotely, a range control, an tomatic version, an electric eye sensor is aimed
ammeter and a voltmeter. Each of the wires used in the at a point below the top of the containment shoes
operation must have a separate power source of this and adjusted to detect the top of the molten slag
description. bath. When the bath rises above the point, the
b) Wire feeder and oscillator: The motor-driven wire rise drive is activated and moves the welding
feeder is similar in design to the one used in M I G / C 0 2 head and shoe up until the bath is no longer de-
and submerged-arc processes. It is mounted on the tectable. A continuous incremental rise is ob-
welding head and feeds the wire to the molten bath tained automatically in this manner.
through the guide tube at a constant speed. The speed
f) Travel carriage: The head is mounted on a travel
is set according to the current required for the deposition
carriage and driven on a vertical track or rail. The rate
rate. It usually is in the range of 17 to 150 mm/sec for
of vertical travel is controlled as in (iv) above.
2.4 mm or 3.2 mm size solid or flux-cored electrode.
The wire coil or spool has to be of adequate size to ensure
Consumable Guide Method
that the entire joint is completed without interruption.
Figure 8.4 shows the consumable guide method of
Each wire has its own individual wire drive. Often a
electroslag welding. It is to be noted that the welding
wire straightener is provided in the feeder to remove
head incorporating the wire reel, wire feeder and other
the cast in the electrode wire, because the cast causes
components is stationary and fixed above the joint, while
the electrode to wander as it emerges from the guide
the guide tube extends the entire height of the joint. The
tube and gives rise to defects such as lack of fusion.
guide tube directs the wire to the bottom of the joint
For oscillating the g u i d e t u b e , m o t o r - d r i v e n and to the molten pool as the weld progresses upward.
mechanical systems such as a lead screw or a rack and The guide tube, which is made of steel and carries the
pinion are used. Oscillation m o v e m e n t has to be welding current, melts along with the wire and becomes
controlled with respect to travel distance, travel speed an integral part of the weld. The guide contributes 5 to
and dwell at the end of each stroke. Electronic circuitry 10% of the filler metal that goes into the joint. The guide
is used for this purpose. is insulated so that it does not short-circuit with the side
c) Guide tube: Reference has been made to this walls and retaining shoes. Sometimes the guide tube is
component earlier. Besides guiding the wire to the slag flux coated to insulate it as well as to replenish the slag
bath, it serves as an electrical contact to the wire. The bath.
bore of the tube is insulated all along, except at the tip
The water-cooled retaining shoes are also of the same
where the electrical contact is necessary.
height as the joint and are clamped on the opposite faces.
d) Retaining shoes: These have been described earlier.
For long joints, it may be more convenient to have
e) Welding head with controls: The electroslag welding
several pairs of shoes. As welding progresses, a pair of
head incorporates the wire spool or coil, wire feeder,
shoes is removed from the joint and placed above the
wire guide tube or tubes with their electrical connections
next pair. This leap frog pattern is repeated until the
and a means of attaching the head beside the joint. In
weld is completed.
special cases, it has an a r r a n g e m e n t for multiple As with the conventional method, more than one
electrode operation and oscillation mechanism. To make wire m a y be u s e d a n d they m a y be oscillated
the method portable, the wire bundle and the wire horizontally in the joint, d e p e n d i n g on the plate
feeder may be removed and placed a short distance thickness. When not oscillated, each wire can weld
away from the welding head as in semi-automatic MIG/ a p p r o x i m a t e l y 63 m m of plate thickness. W h e n
C 0 2 welding. oscillated, one wire can weld up to 130 mm, two wires
The controls consist of: up to 300 mm and three wires up to 450 mm.
i) Ammeter, voltmeter, remote control for each
power source and a remote voltage control EQUIPMENT
The equipment is the same as required by the
ii) A speed control for each of the wire drive motors
conventional method, except for the design of the
POWER SOURCE ICONTROL PANEL
MOLTEN SLAG BATH
MOLTEN WELD
POOL
RETAINING SHOE
WORK (GROUND).
LEAD
y WATER IN
WATER OUT
Fig. 8.4: Consumable guide method of electroslag welding
electrode guide tubes and the fact that the consumable c) Welding voltage
guide method does not require vertical travel of the head d) Electrode extension
and the shoes. The equipment is therefore simple and
e) Electrode oscillation
less expensive.
Guide tube: The guide tube is made from steel which f) Slag bath depth
is compatible with the base metal, and it is slightly g) Number of electrodes and spacing
longer than the joint. It usually has an outside diameter h) Root opening.
of 16 mm and inside diameter 3.2 to 4.8 mm. It is
a) Form factor: The ratio of the width of the weld pool
attached to a copper-alloy support tube which connects
to its maximum depth is termed as form factor. Width
to the welding head. The current is supplied to the
is given by the initial root gap plus the total penetration
electrode wire via the copper tube and the consumable
into the two plate edges. Depth is the distance between
steel tube.
the top of the molten weld pool and the lowest point of
The guide tube needs to be insulated against short-
the liquid-solid interface. With high form factor, the
circuit only When the weld is more than about 750 mm.
grains in the solidified weld meet at an acute angle, and
As an alternative to flux coating, insulator rings are
with low form factor (i.e. narrow width and deep weld
placed on the outside at a spacing of 350 mm and held pool) the grains meet at an obtuse angle. In the former
in place with small welds. Like the flux coating, the case, the weld metal is highly crack-resistant; while in
insulator rings also melt and help replenish the bath as the latter case, the cracking resistance is low.
the guide tube gets consumed.
b) Welding current: Increasing the current increases
the deposition rate as also the depth of the molten wTeld
PROCESS VARIABLES
pool. This leads to a decrease in the form factor and
Following variables affect the operation, weld quality lowers the resistance to cracking.
and economics of the electroslag process: c) Welding voltage: This variable is very important as
a) Form factor it controls the depth of fusion. Increase in voltage causes
b) Welding current increase in the depth of fusion and the width of the weld.
This helps to increase form factor and resistance to unnecessarily more filler metal and may cause lack of
cracking. Voltage also controls the welding operation. edge fusion. Root openings usually range between 20
If it is too low, arcing to the weld pool occurs. Too high and 40 mm.
a voltage may cause slag spatter and arcing on the top
of the slag bath. For best results, the welding voltage Typical Electroslag Welds
must be between 31 and 55 V per electrode. Figure 8.5 shows typical welds for various types of
d) Electrode extension: This factor is present in the joints. In many cases, specially designed retaining shoes
conventional electroslag process a n d not in the are required. The manner and depth effusion are shown
consumable guide method. It is the distance between by dashed lines.
the guide contact tube and the slag pool surface, and is
called dry extension. Increasing the electrode extension Circumferential Welds
increases the resistance in the dry extension and this For making circumferential welds on thick-walled
causes the power supply to reduce its current output. shells of pressure vessels, several electroslag welding
This leads to a slight increase in the form factor. If the methods and equipments, some of them patented, are
voltage setting is increased to bring the current to the available. In one method, the dams are designed to
previous level, the form factor decreases markedly. follow the curvature of the shell, and they as well as the
Electrode extension is usually in the range of 50-75 welding head are kept stationary while the shell rotates.
mm. If it is less than 50 m m , the guide tube gets When the last 450 mm of the joint length is approached,
overheated. If it is more than 75 mm, the electrode melts the electroslag process is stopped and the remaining
on the slag bath surface instead of in the bath. This welding is carried out with the manual metal-arc
makes the operation unstable. process.
e) Electrode oscillation: Oscillating the electrode helps In some cases, the circumferential weld is started and
to distribute the heat and ensure edge fusion. Oscillation closed as follows: Welding is started on a simple dam
speed is important. Increase in the oscillation speed that is tacked in place on the inside of the cylinder.
reduces weld width and consequently the form factor. Preparation for closing is made when welding is half
It must therefore be tuned with the other variables. The completed, by cutting away the starting dam and
dwell time at the end of each oscillation is also gouging out the first weld metal deposited along a
important, as it controls the fusion of the base metal predetermined surface that is tangent to the inside
and helps to overcome the chilling effect, of the retaining diameter. Tapered finishing tabs are tacked in place
shoes. outside the joint at this location, so that the weld puddle
f) Slag bath depth: The bath should be sufficiently can be brought to the surface of the cylinder as the joint
deep to enable the wire to enter into it and melt beneath is completed.
the surface. If the bath is too shallow, slag splits and
arcing occurs on the surface. If it is too deep, the weld Portable Equipment
width is reduced as also the form factor. For best results, A portable electroslag welding equipment capable
the bath depth should be around 40 mm. of welding the thickness range of 16-60 mm has been
g) Number of electrodes and spacing: For a given developed in India, specially for the site welding of blast
number of electrodes, increase in plate thickness causes furnace shells.
a slight decrease in weld width, but a considerable The equipment consists of:
decrease in pool depth, thus resulting in an improve- a) A special carriage with vertical drive mechanism.
ment of form factor. Thickness increase beyond a certain b) A guide angle at the desired curvature to facili-
limit, however, results in lack of edge fusion. The correct tate guiding of the carriage.
number of electrodes, stationary and oscillating, for a c) A set of water-cooled copper shoes.
given plate thickness has been stated earlier.
h) Root opening: A minimum root opening of 20 mm d) A clamping device, mounted on the carriage, to
is necessary to give a sizeable slag bath, good slag clamp the copper shoes firmly on to the walls of
circulation and clearance for the insulated guide tube the plates to be welded.
in the case of the consumable guide method. An increase e) A suitable power source to feed the welding cur-
in the root opening has no effect on weld pool depth, rents.
but it increases the weld width and hence the form Figure 8.6 shows the carriage mounted on the guide
factor. If the root opening is too large, it consumes rails on a test joint. A set each of guide rollers and drive
nn
XT
BUTT JOINT
CORNER JOINT
T-JOINT
nod
K
T- JOINT TRANSITION JOINT
&-
FILLET WELD
■if
Tl
L-J
CROSS WELD OVERLAY BUILDUP
E D G T J Q I NÎT
o
PLUG WELD
Fig. 8.5: Joint designs for electroslag welding—dashed line shows depth of fusion into the base metal
rollers enable the carriage to negotiate the travel. It is The p o w e r for w e l d i n g is obtained from a
provided with a thyristor drive arrangement to enable thyristorised supply-compensated rectifier power
setting of the desired travel speeds. source. It p r o v i d e s special constant current
characteristics while striking the arc and then
automatically switches over to flat characteristics. It also
has provision for remote voltage setting and adjustment.
Also due to the feedback system provided, one could
achieve up to 1% regulation in case of supply outages
of 10%.
Since the success of the electroslag welding process
depends also upon the copper shoes and their efficient
cooling, a tank with pump is provided.
A control panel is suitably located to control:
a) Wire feed setting
b) Vertical travel setting
c) Arc voltage setting, and
d) Switching On and Off operation.
Thus the entire system is simple, effective and
efficient. All the in-built systems are also provided in
the equipment, because the electroslag welding process
does not tolerate any interruption whatsoever. Each
interruption would mean gouging out the weld metal
Fig. 8.6: The electroslag welder making test joint and restarting the welding.
The equipment was used successfully at Bhilai Steel Seven such segments were then assembled vertically,
Plant. The following welding procedure was adopted: involving six joints, to be welded to a height of 12 m.
The weld portion of the blast furnace shell was 16 m All the welded joints were tested ultrasonically and
in diameter and 12 m in height and the joints had to be partly by X-ray and were found to be flawless.The flux-
welded in the vertical position. Square butt joints were wire combination, which was successfully used for this
laid keeping a gap of 25-35 mm between the two plates. job was also indigenously developed.
For a height of every metre, this gap had to be increased
by 1 m m in order to allow for shrinkage, so that Consumables for Electroslag Welding
distortion could be avoided. Before welding, the plate Fluxes: Flux is an important consumable, since it
edges were ground smooth, and the scales and burrs melts a n d p r o v i d e s the slag b a t h that gives the
were removed to enable the copper shoes to travel electroslag process its name.
smoothly. A guide angle, 75 mm x 75 mm in size, was There are no standard specifications for fluxes. They
bent to the profile of the plate of 10 m diameter, and are marketed as proprietary items. Their formulations
tack welded to one of the plates parallel to the weld usually contain oxides of Si, Mn, Al, Ti, Ca and Mg.
joint. The powered carriage of the welding equipment Calcium fluoride (fluorspar) is added to adjust fluidity
moved in this angle by means of the guide rollers. The and electrical resistivity. As the amount of calcium
welder's seat was located separately on a platform, so fluoride increases, viscosity, m e l t i n g point a n d
that it could move vertically as the welding progressed. resistivity decrease. Titanium oxide also decreases
To start with, segments of 8 m radius, 6.8 m length resistivity while aluminium oxide increases it.
and 2 m width were welded together, keeping a tall The flux must meet the following requirements:
curvature of 6-8 m in the vertical position. Six such a) It must be sufficiently conductive when molten,
segments were welded together. Figure 87 shows the to carry the welding current from the electrode to the
site welding of one such curvature joint. It took only molten weld puddle and to the surfaces of the solid base
about 5.5 hours to complete a 6-8 m long weld using metal without arcing. A flux of high resistance will draw
the following parameters: too little current, allowing the pool to cool and the wire
Electrode wire, diameter 3.15 mm to drive too deeply into the molten flux. A flux of low
Current used 600-650 amp resistance will d r a w excessive current, raising the
Voltage 46V temperature of the pool until the process stabilises with
Slag pool 30 mm a shorter length of wire immersed.
b) Flux viscosity must be in the range between the
high viscosity, that results in sluggishness and prevents
setting of small droplets of molten metal (and also leads
to slag inclusions in the weld) and the low viscosity,
that causes the flux to leak through small crevices
between the dams and the work. Viscosity must be low
enough to permit a good stirring action in the flux pool,
so that heat will be distributed uniformly to all surfaces
of the joint.
c) The flux must have a melting range lower than
that of the weld metal and the boiling point well above
the operating temperature to prevent losses and changes
in pool composition.
d) The flux must be metallurgically compatible with
the alloy being welded. For carbon steel, low-alloy steel,
and stainless steel, a basic flux is usually recommended.
e) The flux should solidify to a slag that is easily
removed from the surface of the weld.
Electrode wires: Two types of electrodes are used,
namely, solid and metal-cored. The latter is a composite
electrode consisting of a tube filled with alloying
Fig. 8.7: Welding in progress ingredients and small amounts of arc stabilisers and
fluxing agents. Standard flux-cored electrodes are not process. It is used widely in the structural field.
suitable as their use results in excessive build-up of slag Typical examples are: transition joint between
in the molten bath. different flange thicknesses, stiffeners in box columns
In selecting the wire for a particular steel, it must and wide flanges, joining of large, wide flange beams.
borne in mind that the electroslag weld suffers dilution In the machinery area, large, heavy plates required for
to the extent of 30-50% by the base metal. The mixing is heavy presses and machine tools are electroslag welded.
thorough and the weld-metal chemistry remains Other items are kilns, gear blanks, turbine rings, and
uniform throughout. rims for road rollers. Heavy pressure vessels for the
In higher alloy steels, the required mechanical chemical, petrochemical, marine and power generating
properties of the joint, i.e. weld-metal and the heat- industries are often electroslag welded, as also the
affected zone, are achieved through chemistry as well branch pipes of such vessels. In shipbuilding, the main
as postweld heat treatment. It is advisable, therefore, to hull section is welded by the conventional electroslag
ensure that the weld-metal chemistry matches the base- method. Vertical welding of the side shell, from the bilge
metal chemistry as closely as possible, so that both will area to (but not including) the sheer strake is done with
respond almost equally to heat treatment. a specially designed electroslag welder called the plate
The most popular sizes are 2.4 and 3.2 mm diameters. crawler. Longitudinal and transverse hull stiffeners are
Supplies are normally made in the form of 27 kg coils, welded with the consumable guide tube method. In steel
spools or drums. Since electroslag welds are usually foundries, massive difficult-to-cast components are
long and heavy, and no interruption of the operation is produced in smaller, better quality castings and then
tolerated, spools or drums up to 340 kg are used. electroslag welded together. Since the metallurgical
AWS specification A5.25 deals with flux-wire com- structures of a casting and an electroslag weld are
binations used for electroslag welding of carbon and similar, they respond to postweld heat treatment in a
high strength low-alloy steels. Solid electrodes are classi- similar way.
fied on the basis of their composition, as manufactured.
\
Metal cored types are classified on the basis of chemical ELECTROGAS WELDING
composition of undiluted weld-metal. For chemical \ A brief introduction of this process is given on the
composition of the latter, an undiluted ingot is deposited
first page of this chapter along with Fig. 8.2 showing
in a water-cooled copper mould using the typical the welding operation using a flux-cored wire. Figure
electroslag welding conditions given in the specification.
8.8 shows electrogas welding with a solid electrode,
Many of the solid electrodes are the same as are used
which is fed through a modified M I G / C 0 2 welding gun.
for submerged-arc welding and M I G / C 0 2 welding. The gun also serves as an electrode guide and a contact
Fluxes are classified on the basis of the mechanicaltip holder. Its width is restricted to 10 mm, so that it can
properties of a weld deposit made with a particular fit and move easily in a minimum of 17 m m root
electrode. The classification designation given to a fluxopening. Shielding with C 0 2 or argon-C0 2 mixture is
consists of a prefix FES followed by a two-digit number provided through the gun and also through pipes in
representative of the tensile-strength and impact the moveable shoe as shown. There is no molten slag
p r o p e r t y r e q u i r e m e n t s for test w e l d s m a d e in
on the weld pool to protect it. The entire protection is
accordance with this specification. This is then followedprovided by the gas shielding.
by a set of letters and numbers corresponding to the In electrogas welding with a flux-cored wire, a thin
classification of electrode used to classify the flux. slag layer is present on the top of the molten pool.
Mechanical property requirements are: two tensile Whether additional gas shielding is required or not
ranges, namely 420-550 MPa with minimum yield is dictated by the type of flux-cored electrode used.
strength of 250 MPa, and 490-650 MPa with minimum With self-shielding type wire, gas shielding is not
yield strength of 350 MPa. Three Charpy-V notch levels required.
of 20 J minimum at -18°C, 20 J minimum at -29°C or none The basic mechanical equipment is similar to that
at all. It is to be noted that these tensile and impact values
for conventional electroslag process, except for the fact
have to be obtained without postweld heat treatment. that introduction of shielding gas, when required, has
to be provided for. For supplementary gas-shielding of
Applications the weld pool, gas boxes may be mounted on the
Plates and other sections ranging in thickness from retaining shoes as shown in Fig. 8.8. Gas from these
19 to 460 mm are commonly welded by the electroslag boxes is directed out through ports which are suitably
ELECTRODE
CONDUIT
WELDING-
GUN
3*—GAS
fMxfc GAS
WATER
MOVEABLE
SHOE
FIXED SHOE
Fig. 8.8: Electrogas welding with a solid wire electrode
sized and positioned to uniformly shield the arc as well electrodes without extra gas shield are used. The latter
as the weld pool. Gas metering device is also necessary. results in increased deposition rate due to resistance
For plates less 32 mm thick, gun oscillation is not heating.
necessary. For thickness range of 32-102 m m , the Horizontal oscillation of the electrode is used when
welding gun must be oscillated horizontally over the the plate thickness exceeds 32 mm. The oscillation is so
weld pool to achieve uniform metal deposition and to controlled that the electrode guide does not get within
ensure thorough fusion of the joint edges. 10 mm distance of each retaining shoe. Oscillation speed
is 7 to 8 mm/sec and the dwell time at the ends is 1-3
Process Variables sec.
The welding variables of electrogas welding are The plates to be joined are assembled vertically with
similar to those for electroslag welding discussed earlier. square faces and a root gap of 17 mm. Plates thicker
Increase in electrode-feed speed causes increase in the than 25 mm require starting as well as run-off tabs, while
welding current and leads to an increase in weld width. plates thinner than 25 mm require only a starting plate.
A rise in arc voltage increases the depth of fusion and The arc is initiated in the same manner as in M I G / C 0 2
the weld width. Arc voltages are generally in the range or flux-cored arc welding.
of 30-55 V. With increasing current, the deposition rate C 0 2 is used as a shielding gas for flux-cored
increases as also the upward travel speed. electrodes, while 80% Ar-20% C 0 2 mixture is preferred
Electrode extension is approximately 40 mm when for solid wires. Flow rates usually range between 14 and
gas shielding is used and 60-75 mm when self-shielding 66 1/min.
Electrodes in this process, the weld-metal may contain up to 50%

AWS Specification A5.26, "Specification for of admixed base metal.
consumables for electrogas welding of carbon and high Because of the prolonged thermal cycle, the
strength low-alloy steels", covers the two forms of electroslag weld-metal structure consists of large prior
electrogas welding electrodes used in industry, namely, austenite grains that follow a columnar solidification
flux-cored and solid. The specification classifies flux- pattern. The grains are oriented horizontally at the weld
cored electrodes on the basis of (a) whether auxiliary edges and turn to a vertical orientation at the weld
shielding gas is required or not, (b) chemistry of the centre. The microstructure of low-carbon steel
deposited metal, and (c) as-welded mechanical electroslag weld usually consists of acicular ferrite and
properties of the deposited metal. Solid electrodes are pearlite grains with pro-eutectoid ferrite outlining the
classified on the basis of their chemistry as prior austenite grains. While the prior austenite grains
manufactured and as-welded mechanical properties of at the weld periphery are usually coarse, those near the
the deposited metal. The tensile and impact centre of the weld are finer. This fine-grained region
requirements are similar to those for electroslag welding appears equiaxed in a transverse cross-section; however,
electrodes. The original AWS specification must be longitudinal sections reveal its columnar nature. The
consulted for full details. relative proportions of the coarse - and fine-grained
Electrogas flux-cored electrodes differ from those regions change markedly with changes in weld-metal
designed for flux-cored arc welding in having a lower chemistry, sometimes even to the extent that only one
proportion of slag-forming compounds in their flux. is present.
They provide a thin slag layer under the retaining shoes, The prolonged high temperature and the slow
which gives a smooth weld profile. They also provide cooling rate after electroslag welding result in wide,
the necessary alloying elements. Sizes range from 1.6 to coarse-grained heat-affected zones having only
4 mm diameter. relatively soft, high-temperature transformation
Solid wires are identical to those used for MIG/ C0 2 products.
welding. Electroslag welds along with their heat-affected
zones as described above can be used in the as-welded
Applications condition in applications such as structural welding,
The common applications of electrogas welding are where special demands are not made on their notch
similar to those for electroslag welding discussed earlier. toughness and brittle fracture resistance. If higher notch
Electrogas process is, however, more suitable for the toughness values at low temperatures as demanded by
lower thickness range of 10-50 mm. When the thickness codes have to be satisfied, it becomes necessary to
exceeds 75 mm, the electroslag process must be normalise the entire fabrication, so that all traces of the
preferred. Electrogas process has been found very useful cast structure are removed and the properties of the
for the field welding of vertical joints in large storage weld-metal including notch toughness are brought in
tanks and ship hulls. Use of self-shielded flux-cored level with those of the base metal. A brief note on
electrodes makes the equipment simpler and lighter. normalising is given later.
The process is also adaptable to two-pass welding of Thus electroslag welded pressure vessels of large
double-V joints, which gives superior as-welded impact dimensions made from C-Mn and low-alloy steels are
properties as compared to electroslag welding. often normalised, i.e. heated above the critical
temperature (i.e. much above the temperatures used
METALLURGY OF ELECTROSLAG WELD-METAL in stress-relief treatment) and slowly cooled. Such heat
Electroslag weld-metal is deposited in one pass and treatment is not only an expensive operation but causes
the temperatures attained are considerably lower than problems by lowering the yield point and tensile
in arc welding processes. With conventional arc welding strength to such an extent that the standard wires are
processes, the weld of comparable thickness, on the not capable of attaining the specified minimum
other hand, is deposited with a pointed source of heat strength in higher tensile steels. The use of basic fluxes
having much higher temperature and in multiple passes. in place of acidic fluxes, also cannot help; they only
In the electroslag process, the weld-metal stays molten improve the resistance to cracking in the as-normalised
for a sufficiently long time to permit slag-refining action, condition.
which results in the escape of dissolved gases and In recent years, improvement of the structure, and
transfer of non-metallic inclusions to the slag bath. Also toughness of the electroslag welded joints have been
achieved by the use of higher currents and the aduition such as Ti, V, Ce and Zr to the parent steel. The effect is
of metal p o w d e r to the molten pool, along with further intensified by the use of fine dispersed nitrides.
o p t i m u m electrode a r r a n g e m e n t and the joints The result is that the formation of Widmannstatten
geometry. The supply of the filler metal in powdered s t r u c t u r e and p r o e u t e c t o i d ferrite at the grain
form and without need of additional current increases boundaries, which decrease notch toughness, is reduced
the weld deposition rate and reduces overheating of the and the formation of acicular ferrite is promoted. Also
molten weld pool. In conjunction with increased the distribution of non-metallic inclusions is more
nucleation, such cooling of the molten pool affects the favourable. Steels with low carbon content (around
weld structure favourably. 0.12%) and alloy additions of Cr, Mn, B, Ce, Mo and Ni
Procedure for electroslag high-speed welding with are now available,which allow electroslag welding of
powdered filler metal is shown in Fig. 8.9. It is to be plate thicknesses up to 2,000 mm, in which the HAZ is
noted that two wires are placed side by side for 10-40 sufficiently resistant to brittle fracture without
mm plate thickness range, while two such pairs (total additional normalising.
four wires) are used for 40-100 m m range. This This method is applied today not only to unalloyed
arrangement prevents lack effusion defects at the plate and low-alloy structural and pressure vessel steels, but
edges, which are likely to occur due to high welding also for the welding of forgings and steel castings up to
speeds. For metal powder addition, special devices are the largest cross-sections made of creep-resisting and
used. Using this procedure, fine-grain structural steel pressurised hydrogen-resistant QT steels as well as
of 63.5 mm thickness has been welded with two double austenitic Cr-Ni-Ti and Cr-Ni-Mo steels and Ni alloys.
wires and metal flux addition using 2,600 amp current,
51 V voltage and 23 cm/min welding speed, which is METALLURGY OF ELECTROGAS WELD-METAL
ten times faster than the conventional electroslag Unlike the electroslag process, electrogas is an arc
method. welding process, in which the welding energy density
For obtaining good as-welded notch toughness is much higher and more concentrated. Also the weld
properties in the HAZ, the steelmaking process has been pool is smaller and the solidification pattern is faster.
modified. Grain growth in the HAZ due to overheating Since both are single-pass processes with single large
has been restricted by the addition of active deoxidising weld cross-section, the weld macrostructure is similar
agents, as well as nitride- and carbide- forming elements but the grain sizes of the electrogas weld and its heat-
1 WATER COOLING
2. METAL POWDER SUPPLY
3. MAGNETIC SCREENING
4. WIRE ELECTRODE
•3 5. COPPER SHOE (WATER COOLED)
6 SLAG BATH
7 MOLTEN BATH
8. PARENT METAL
9 WELDING JOINT
(B)
Fig. 8.9: Electroslag high-speed welding with powdered filler metal: (A) plate t range 10-40 mm;
(B) Plate t range 40-100 mm.
affected zones are smaller. Consequently, as-welded for v a r i o u s reasons including grain-structure
mechanical properties of the electrogas welds are better. refinement, homogenisation, removal of residual
In electrogas welds, too, the form factor, i.e. ratio of stresses and improved machinability. The rate of heating
width to the depth of the weld pool, controls dendrite in the normalising process is not critical, except in the
pattern and resistance of the weld to cracking. A high case of parts having great variation in section size. Time
form factor and an acute-included angle must be at the maximum temperature is critical, only in that it
achieved. The postweld heat treatment requirements must be sufficient to dissolve the carbides, if present,
for electrogas welds are the same as for electroslag a n d / o r for movement of alloy atoms to obtain a desired
welds. final structure. The rate of cooling can be suitably
Under abnormal welding conditions, electrogas controlled to obtain the desired strength and hardness,
welds may contain defects like slag inclusions, porosity bearing in mind that the higher cooling rates result in
and cracks; the last are usually hot cracks located in the higher strength and hardness and lower cooling rates
centre part of the weld. result in lower strength and softness. After the steel has
cooled evenly to black heat or below Ar a , it may be
Normalising water- or oil-quenched to decrease the total cooling time.
Normalising means heating a steel to a temperature For plain carbon steels, the normalising temperatures
of about 55°C above the Ac3 i.e. above the upper critical range between 915 and 860°C; the actual temperature
line of the iron-iron carbide phase diagram followed by decreases with increasing carbon content. For alloy steels
cooling in still or lightly agitated air. Normalising is done they range between 925 and 870°C
Welding Handbook, Chapter 7, Vol. 2, 7 th Ed., published by American Welding Society.

Metals Handbook, Vol. 6, 8th Ed., published 1971 by American Society for Metals. IIW Doc. XII-A-23-1982
Review of Electroslag Welding, Weld. J., Dec. 1981.
Methods to Improve the Properties of Electroslag Welds by Vaidya et al. Weld. & Met. Fabr., May 1976.
The Development of a Consumable Guide Machine for Vertical Welding of Ship Sections by Turner & Coleman.
Conference Proceedings of "Advances in Welding Processes" 1970, by The Welding Institute.
High Speed Consumable Guide Welding of Relatively Thin Steel Plate by Haslam et al. Conference Proceedings of "Advances in
Welding Processes", 1974 by The Welding Institute.
Unusual Electroslag Welding Applications by Dorchu et al. Weld. J., Nov. 1973.
Electroslag Welding of Heavy Section 2V4 Cr-1 Mo Steel by Okumura et al. Weld. J., Dec. 1976.
Electroslag Welding Speeds Shipbuilding by Parrott et al. Weld. J., Apr. 1974.
Slag-Metal Reactions in the Electroslag Process by Patchett et al. Weld. J., Oct. 1972.
CHAPTER 9
Welding
Metallurgy
A WELDING ENGINEER MUST acquire a basic unconventional metallurgy in many important aspects.
derstanding of welding metallurgy, because it enables To have a clear understanding of welding metallurgy,
him to control two important aspects of a weldment it is necessary to acquire the basic knowledge of
which determine its suitability for service, namely, (a) conventional metallurgy, as also to appreciate the
the chemistry and soundness of the weld-metal, and metallurgical aspects peculiar to welding.
(b) the microstructure of the weld-metal and of the heat- This chapter is divided into two sections. The first
affected zones (HAZs) of the base metal. section deals with the elements of conventional
In simple words, metallurgy is the branch of science metallurgy, which are essential for an understanding of
which deals with making, shaping and treating of the second section. The second section covers the specific
metals. It can broadly be classified in two parts: process aspects of welding metallurgy.
metallurgy and physical metallurgy. Process metallurgy
involves the conversion of ores into metals, refining and I: GENERAL METALLURGY
alloying of metals, and shaping them to semi-finished Structure of Metals
and finished products by casting, forging, rolling/etc. To understand the structure of metals, it is necessary
Physical metallurgy deals with heat treatment, to follow the pattern of solidification of a metal from
mechanical testing, metallography and many other the liquid state. In the liquid state, the atoms of a metal
topics which concern the design, applica-tion, testing are in constant motion and have no orderly
and inspection of the metallic products. arrangement. As the liquid is allowed to cool and it
Welding involves knowledge on both process approaches the solidification temperature, solid
metallurgy and physical metallurgy plus many other particles begin to form at specific points as shown at
metallurgical features peculiar to its processes. Process (A) in Fig. 9.1. These are called nuclei. As solidification
metallurgy forms the basis of formulations of electrode proceeds, each nucleus grows into a large solid particle
coverings and submerged-arc welding fluxes. In fact, called grain as shown at (B). When the solidification
formation and deposition of weld-metal is similar to process is complete, the large solid grains meet as a1
the production of metal in a furnace, but with the (C). The lines along which the neighbouring grains meel
significant difference that it takes place instantly and are called grain boundaries.
on a minimum scale. The hot and cold peening of weld- A grain of a metal, when examined minutely, is found
metal is like hot forging or cold finishing mill to have a crystalline structure, in which the atoms oi
operations. Principles of physical metallurgy are each crystal are arranged in a specific geometric pattern.
utilised to control the cooling rates of the weld-metal The most common crystalline structures displayed bj
and the base metal heat-affected zones through welding metals are shown in Table 9.1 and atomic arrangements
heat input control, preheating and postheating, and to are shown in Fig. 9.2.
achieve desired microstructures and mechanical All grains in a given metal and at a given temperature
properties. However, welding metallurgy differs from have identical crystalline structure and atomic spacing
INITIAL CRYSTALS SOLID GRAINS SOLID GRAINS WITH

GRAIN BOUNDARIES
/ _ ^ / ^
-tf LIQUID -H-

^P LIQUID^s,
-44- 4f
* * *
(A) INITIAL CRYSTAL FORMATION; (B) CONTINUED SOLIDIFICATION; (C) COMPLETE SOLIDIFICATION
Fig. 9.1: Pattern of solidification of a metal
Table 9.1 : Crystal structures of common metals

Crystal structure Metals with the structure
Body centred cubic (BCC) Fe (at room temperature and again near melting point), Cr, Cb,
W, V, Ti (at high temperatures), Mo
Face centred cubic (FCC) Fe (at intermediate temperature), Cu, Au, Pb, Ni, Ag, Al
Hexagonal close packed (HCP) Co, Mg, Sn, Ti (at room temperature), Zn, Zr
<r^r
k±ZJ7*
Fig. 9.2: Common crystal structures of metals.

Right: face centred cubic (FCC): Centre: body centred cubic (BCC): Left: hexagonal close packed (HCP)
However, the orientation of the grain lattice in each a) Direct substitution: In this case, the alloying atom
grain is different from that in the neighbouring grain, directly replaces the atom of the parent metal in the
because each of them has grown independently of the lattice (see Fig. 9.3 (a)). The new atoms are said to have
other. Therefore, the orderly arrangement of the atoms dissolved into or among the old atoms. The new atoms
is disrupted at the regions of the grain boundaries. The are known as the solute; the remaining old atoms are
grain b o u n d a r y d i s o r d e r has m a n y i m p o r t a n t called the solvent; and the resulting alloy is referred to
repercussions on the metal properties. These will be as a solid solution. Solid solution results when the
pointed out later. alloying atoms and the parent metal atoms are similar
What we have discussed so far applies to pure in size and chemical behaviour. Examples are alloying
metals. However, metals are largely used in industry of silver with gold or of nickel with copper. Both are
as alloys, that is, in combination with other metals and solid solutions.
n o n - m e t a l s , b e c a u s e alloys p r o v i d e e n g i n e e r i n g b) Interstitial alloying: In this case, the alloying atom
properties, which for most applications, are superior to places itself in the space between the parent metal atoms
those of pure metals. Alloying elements are incorpora- without actually displacing any of them, as shown in
ted into the parent metal in either of the two following Fig. 9.3 (b). This type of alloying occurs when the
ways: alloying atom is small in comparison with the parent
WELDING METALLURGY 393
QQQQ QOQ
OTW^-^W
<&(pcp(p(p-e-€>
Q
< ^ > (a) (b) O (J) (b (D (D <tT(i)
Fig. 9.3: Modes of metal alloying: (a) substitutional solid solution; (b) interstitial alloying
atom. A common example is carbon in very small and low temperature, alloys must be fine-grained,
amount, alloyed interstitially with iron in mild steel. while for high temperature service they must be coarse-
In most engineering alloys, however, several alloying grained.
elements are used w h o s e atoms d o n o t dissolve Grain size is therefore another important aspect of a
completely, either interstitially or substitutionally They microstructure. Metals may be coarse-grained, fine-
consequently give rise to what is termed a multiphase grained or can have a mixture of both coarse- and fine-
alloy, in which several phases, each one having its own grains, depending on the solidification rate, alloy
characteristic crystalline structure, exist side by side. additions and other factors. Grain size control is all the
The grains, grain boundaries and phases together more important in the case of the weld-metal and it can
make up what is called the micro-structure of the alloy. be accomplished when the metallurgy of the weld-metal
All of them are distinguished and identified by suitably is correctly understood.
polishing and etching the alloy surface and examining
it under a microscope at high magnification. This Phase Transformations
procedure is referred to as metallographic examination. We have referred to the multiple phases that can
It is one of the important methods of studying the coexist in an alloy. A phase change occurs when a metal
characteristics of metals a n d alloys, since t h e melts or solidifies. Some metals undergo phase changes
microstructure of an alloy is largely responsible for its in the solid state when heated or cooled. Such changes
physical and mechanical properties and behaviour in are termed allotropic transformation. Iron, titanium,
service. The microstructure is determined not only by zirconium and cobalt are examples of metals which are
the alloy chemistry, but also by hot and cold working subject to allotropic transformation. The phase changes
and heat treatment, as will be explained later. which occur during the heating and cooling of pure iron
Reverting to grain boundaries, many significant are described under iron-carbon phase diagram.
phenomena occur in these regions. For example, since
the arrangement of atoms is irregular and interatomic Phase Diagrams
space may be larger than normal at the grain boundaries, Solidification, phase changes and other phenomena
movement of individual atoms of elements through the occurring in alloy systems are best understood by
solvent structure occurs, which results in so-called referring to phase diagrams, which are also called
segregation. This segregation often causes the formation constitution diagrams or equilibrium diagrams. The
of undesirable phases that reduce ductility of the alloy word equilibrium is used, because the diagram describes
and make it susceptible to cracking during welding or the behaviour of an alloy system under equilibrium
heat treatment. conditions, i.e. slow heating and cooling rates and long
Another u n i q u e p h e n o m e n o n of t h e grain hold times. But such conditions are rarely met with in
boundaries is that, they resist the deformation of practice, a n d hence t h e d i a g r a m s are at best an
individual grains a n d thereby increase the room approximation of how alloys actually behave in use.
t e m p e r a t u r e s t r e n g t h of a n alloy. At elevated A simple phase diagram for the copper-nickel alloy
temperatures, this resistance is lowered and the atoms system is shown in Fig. 9.4. A point to be noted is that,
in the boundaries are able to slide past one another these two elements are fully soluble in each other in all
more easily. The result is reduced strength. This points proportions and at all temperatures, both in liquid and
to the fact that for better properties at room temperature solid states. Phase d i a g r a m s are d r a w n with
MOOO
SOLID SOLUTION
200
20 40 60 80 100
NICKEL, PERCENT
Fig. 9.4: The nickel-copper phase diagram. Points along the top curve A, represent the liquidus for each alloy
composition. Points along the bottom curve B, represent the solidus for each alloy composition
temperatures on the vertical axis and alloy content on exactly like that in the copper-nickel system, i.e. all solid
the horizontal axis. The extreme left of the horizontal alloys solidify as crystals of a solid solution of copper
axis represents zero % nickel and 100% copper, while in silver called alpha-phase. Similarly all solid alloys
the extreme right represents 100% nickel and zero % represented in the extreme right portion, and having
copper. Between them are the intermediate alloys. 92-100% copper contain a single-phase called beta (a),
For all alloy compositions, only liquid phase is present which is a solid solution of silver in copper.
above the top curve marked A, while below the bottom Figure 9.5 shows that there is one particular alloy
curve B only solid phase is present. The solid alloy, composition which solidifies at a single temperature as
whatever be its composition, is single phase only, because a pure metal does, while in the case of all the other alloy
the two elements are fully soluble in each other in the
solid state.
The diagram also shows that the solid and the liquid
LIQUID
phases coexist in the region between the two curves. 1083#C
(1981#F). 2000
This means that for all alloy compositions (except 100%
Cu and 100% Ni) there is no fixed melting point, and
the melting takes place over a temperature range. For
fi+ LIQUIDI
example, when 70% Ni-30% Cu alloy (known
commercially as monel) is heated, it begins to melt at SQUPVS J -|" 150
° a
1,330°C and the melting is complete at 1,365°C.
2
We shall next take up the silver-copper phase a.
diagram, which is useful for formulating brazing alloys 1000 Z
(see Fig. 9.5). It differs from the earlier one in that certain B
parts of the diagram show two phases existing side by
side. At temperatures above the line marked liquidus,
500
all alloy compositions are fully liquid. At temperatures
below the line marked solidus, they are fully solid. The
COPPER, PER CENT
extreme left portion marked a represents alloys with
up to 9% copper. In this portion the solidification occurs Fig. 9.5: The copper-silver phase diagram
compositions, the solidification is complete at this composition at a given temperature is that of an alloy
temperature and goes through a range. This unique alloy which is completely liquid at that temperature. Thus
composition (28% Cu-72% Ag) is termed a eutectic and when the crystals have the compositions shown by
the temperature at which it solidifies (779°C) is termed points Bx and B2, the liquid has the compositions shown
the eutectic point. It is indicated by an arrow in Fig. 9.5. by points Ax and A 2 respectively. At the eutectic
In order to understand how changes take place in a temperature of 779°C, the crystal composition is that of
eutectic alloy during cooling, we shall first take up a 15 point B3, and the liquid composition is that of point A3,
Cu/85 Ag alloy and refer to Fig. 9.6, which is a magnified which is eutectic-alloy composition.
version of the relevant part of Fig. 9.5. When the liquid Further cooling below 779°C results in the
alloy has reached the liquidus temperature, solid crystals simultaneous formation of alpha and beta solid-solution
of alpha solid-solution, termed primary crystals are crystals. These two phases may be intermingled, or
formed. The early crystals do not have the same present as thin alternate plates in the structure. The
composition as the liquid alloy from which they overall structure of the final solid alloy thus consists of
crystallise, but they have the composition of the alloy the large primary crystals of alpha formed at higher
that is completely solid at the existing temperature. For temperatures and a eutectic mixture of alpha and beta.
example, at the temperature marked A1, the composition Though the liquid and solid phases formed in the
of the crystals forming will be that represented by Bv above process depart from on<e another in composition,
which is obtained by drawing an isothermal line from the equilibrium conditions require that the crystals
the point Av of the alloy over to the solidus. As cooling readjust their composition to that of the initial liquid
proceeds, the crystals will adjust their composition so alloy. The adjustments occur by diffusion of atoms and
as to meet the condition explained above. Thus at the it is assumed that the cooling rate is slow enough to
lower temperature A 2 , the alpha crystals will have the permit it. If the cooling is rapid as in the formation of
composition indicated by the point B2; and at the weld-metal, the resulting alloy may be heterogeneous
temperature A3 which is the eutectic temperature, they in composition.
will have the composition of point B3. If an original alloy of eutectic composition (28% C u -
All the crystals formed so far are richer in silver than 72% Ag) is subjected to the same exercise, both the alpha
the original alloy, and as a result, the remaining liquid and beta phases will form concurrently and its final solid
has increased proportion of copper than at the start. Its structure will consist entirely of alpha and beta mixtures.
In the case of an alloy whose composition falls between
the eutectic and 92% copper, the structure of the solid
alloy will consist of the primary crystals of beta and a
eutectic mixture of alpha and beta.
960°C (1760*F)
Allotropic Transformations of Iron
n\— Before taking up the iron-carbon phase diagram, we
shall describe the phase changes that take place in pure
tg
i ^
iron, when heated and cooled in the solid state.

SOLIDUS\
At room temperature, iron has a body-centred cubic
BV N*2
i crystal lattice known as alpha-iron (oc-iron). As it is
heated slowly, it gradually expands until 910°C, when
2 A i 779°C A3 it transforms to a face-centred cubic crystal lattice called
z
ID
1 (1434°F) gamma-iron (y-iron) and suffers about 1% contraction
in volume. This crystal form persists until 1,400°C is
H
reached, when the iron reverts to the body-centred cubic
form known as delta-iron (5-iron). This form is retained
until the melting point of 1,540°C is reached.
i i 1 . . 1
10 20 30 The crystalline forms of alpha-iron and delta-iron
COPPER, PER CENT are identical, but different names are given to indicate
that they are formed at different temperatures.
Fig. 9.6: The silver-rich portion of the copper-silver On slow cooling from 1,540°C, the same phase
phase diagram
changes take place in the reverse order and at
approximately the same temperatures. This shows that 4) In the portion enclosed by lines A3, A ^ and A:,
iron has two transformation temperatures, 910 and the lines are similar to those of a eutectic system, except
1,400°C, at which phase changes occur during heating that, only solid phases are present. This area is termed
and cooling. eutectoid region, to distinguish it from a eutectic.
5) The letter A of A3, Acm and Ax indicates the arrest
Iron-Carbon Phase Diagram point during heating and cooling. The arrest on heating
Iron is used mostly in the form of steel, which is an is termed Ac point and on cooling Ar point. The two
iron alloy with up to 2% carbon. Alloys with 2-5% arrest points during heating are indicated as Acl and
carbon are known and used as cast iron. Alloys with Ac3. On slow heating, the transformation starts at Acl
more than 5% carbon cannot be practically produced. and is complete at Ac3 point. On slow cooling, the
Iron-carbon phase diagram is therefore of special transformation starts at Ar3 and is completed at Arl. A3
significance to the student of metallurgy. It is shown in is referred to as upper transformation temperature while
Fig. 9.7. The diagram shows the phases formed in iron- A1 is called the lower transformation temperature. The
carbon alloys depending on the amount of carbon temperature range between Acl and Ac3 or Ar3 and Arl is
present and the temperature at which the alloy is termed the critical range or transformation range.
maintained. 6) The terms ferrite, pearlite, cementite and austenite
Following points must be noted in the diagram: refer to the constituents of the alloy structure.
1) Above the liquidus line, the alloy is completely Ferrite is pure alpha-iron. It is very soft, very ductile,
liquid and below the solidus line it is completely solid. has a tensile strength of about 18 tons/sq. in. (278 N /
Between these lines a mixture of solid crystals and the mm2) and a yield point of 9 tons/sq. in (139 N/mm 2 ). It
liquid phase exists. is also magnetic.
2) Carbon up to 2% goes into solid solution in iron. Pearlite is the strongest of any of the constituents of
This solution in turn forms a eutectic with iron carbide, steel. It has a tensile strength of about 56 tons/sq. in.
Fe3C, which is a compound of iron and carbon and (865 N/mm 2 ) and its yield point is about half of that.
contains 6.67% carbon. It is harder than ferrite, but cannot approach cementite
3) The pure eutectic alloy contains 4.3% carbon and in hardness. It is quite ductile. Pearlite structure when
melts at 1,130°C, the lowest melting point of any alloy seen with a powerful microscope is found to consist of
in the system. alternate bands of ferrite and cementite as shown in Fig.
9.8. At 0.80%C, 100% Pearlite structure is achieved which
is referred to as eutectoid.
3000 Cementite is the hardest of any of combinations of
carbon and iron. It is also very brittle and has very little
resistance to shock. It is probably quite strong in
compression if the stress is applied slowly, but has very
low strength in tension.
Austenite is a solid solution of iron carbide in gamma-
iron. It is stable only at high temperatures as shown in
the diagram. Below the critical range, austenite changes
into an aggregate of ferrite and pearlite or ferrite and
cementite. Austenite is extremely ductile and tough. It
is non-magnetic. It is also difficult to machine, because
it has a tendency to work-harden though it is originally
softer than pearlite and harder than ferrite.
The phase changes occurring in alloys containing 0.4
, 0.9, and 1.3% carbon during slow, continuous cooling
from above Ar, point to room temperature and depicted
in Fig. 9.7 are described below:
0.4% carbon: Above the transformation temperature
CARBON, PER CENT (see Fig. 9.9), the alloy has granular structure as at A.
As the alloy cools below A3 ferrite precipitates at the
Fig. 9.7: The iron/iron carbide equilibrium diagram
grain boundaries of the austenite as shown at Band
pearlite with a still larger evolution of heat than occurs

in the lower carbon steels. The structure is thus pure
pearlite, without ferrite or cementite.
1.3% carbon: On reaching the temperature of 860°C
on the A line, cementite which is carbide of iron (Fe„C)
v
cm ' 3 '
separates from the austenite in the same manner as fer-
rite separates in a low-carbon steel. It forms a network
enclosing grains of austenite. The austenite becomes
poorer in carbon as it cools and on reaching Arl (723°
C), it attains the eutectoid composition and transforms
into pearlite. Thus the structure at room temperature
comprises cementite and pearlite.
Fig. 9.8: Typical lamellar appearance of As mentioned earlier, these phase diagrams are
pearlite 1500 x (before reduction): Etchant: Picral based on extremely slow heating and cooling. The al-
loys used in practice are either normalised, annealed or
hardened and tempered. These operations modify the
gradually increases in volume as the temperature drops.
steel microstructures very significantly.
When A y or 723°C is reached, the ferrite surrounds the
Normalising, as explained in chapter 8, is a method
islands of austenite, now at the eutectoid level of carbon
of heat treatment in which a steel, in whatever condi-
content (0.8%). At this temperature the remaining
tion it may be to start with, is reheated to a temperature
austenite transforms to pearlite, which is a eutectoid
exceeding the upper limit of the critical range (i.e. above
mixture of ferrite and cementite. The final structure at
A3 point) and cooled freely in air.
room temperature thus comprises ferrite and pearlite
Annealing is a method of heat treatment, in which a
as shown at D.
steel in whatever condition it may be to start with, is
0.8% carbon: In this alloy, no change occurs in the reheated to a temperature exceeding the upper limit of
structure until Arl (690°C) is reached, as denoted by point the critical range and cooled slowly. This may be done
E in the diagram. The solid solution is exactly of eutec- by putting off the furnace and allowing the steel to cool
toid composition, and at Arl (690°C) it breaks down into in the furnace.
FERRITE
FERRITE -
Fig. 9.9: Structural changes (schematic) in 0.40% carbon steel during slow cooling
Hardening is a method of heat treatment, in which hypoeutectoid steels, almost all the carbon is in the
the steel is heated to above the A 3 point and cooled pearlite constituent, since ferrite is pure iron. That is
rapidly by immersing in oil or water. the reason why the relative proportions of pearlite and
The structure of normalised and of annealed steels ferrite change with the carbon content of steel.
consists of ferrite enclosing grains of pearlite in the case It is possible to estimate the amounts of pearlite and
of low-carbon steels (below 0.8%C i.e, hypo-eutectoid ferrite in a pure carbon steel by using the following
steels), and of cementite enclosing grains of pearlite in formula:
the case of high-carbon steels (C is above 0.8%). When a
section of steel is polished and etched and examined Amount of pearlite _ c
under a powerful microscope, the ferrite is seen as a Amount of ferrite 0.9-c
light-etching constituent and pearlite is seen as«a dark- where c = carbon per cent.
etching constituent. The proportion of ferrite decreases Thus a 0.3% carbon steel will have one-third pearlite
and that of pearlite increases, as the carbon content of and two-thirds ferrite and a 0.45% carbon steel will have
the steel increases. With increase of pearlite (i.e. increase half pearlite and half ferrite. These figures are true only
of carbon up to 0.8%) the hardness, tensile strength and in the case of pure carbon steels, which have been cooled
elastic limit of steel increase, while elongation, reduction slowly. For a given carbon content, the proportion of
of area, and shock-resisting properties diminish. This pearlite to ferrite is increased by a more rapid rate of
change in properties is reflected in the uses to which cooling, or by an increase in manganese. Manganese is
untreated steels of various compositions are put: also always present in commercial steels in varying
C% Mn% Approx. % of Uses

pearlite in the steel
as used
DTD 03 ID Thin sheet tinplate, etc.
0.15-0.20 0.5 20 Structural purposes, ship's plates
0.20-0.25 0.5 25 ^ Boiler plate
0.30-0.35 0.7 40 Axles
0.55-0.75 0.7 70-90 Tyres
0.50 1.0 70 Rails
0.60 1.0 80 Tramway rails
0.65-0.75 0.8 90-100 High tensile wire
The grain size of a n n e a l e d steel is u s u a l l y quantities, and this is used as one of the means to control
considerably larger than that of normalised steel. This the amount of pearlite and thereby the tensile properties
is most frequently d u e to the h i g h e r m a x i m u m of a steel.
temperature employed in annealing (this phenomenon Depending on the rate of cooling during annealing
will be explained later), but even w h e n the same or n o r m a l i s i n g , the pearlite c o n s t i t u e n t of a
maximum temperature is used, the annealed steel will hypoeutectoid steel assumes different characters, and
generally have a slightly larger grain size on account of according to this character the steel attains particular
the longer time it is maintained at that temperature. A physical properties. When the rate of cooling has been
more important difference between a steel in the moderately rapid, an extremely fine pearlite called
normalised and in the annealed condition is to be found sorbitic pearlite is formed. This type of structure is also
in the character and extent of the pearlite areas. induced by the addition of manganese. Mechanically it
While describing the iron-carbon diagram, we have is the strongest type of pearlite and confers on a steel
stated that steel with 0.8% carbon is said to be of high elastic limit and considerable hardness.
eutectoid composition and its structure is entirely The pearlite of annealed steels, cooled slowly through
pearlitic. Steels with less than 0.8% carbon are called the critical range, becomes coarsely laminated. As the
hypoeutectoid steels, while those with more than 0.8% coarseness of the lamellar structure increases, the elas-
carbon are called hypereutectoid steels. In the case of tic limit, yield point and hardness of the steel decrease.
Prolonged heating at temperatures just below Aq accompanied by lower ductility and increasing
(i.e. at 700°C) followed by slow cooling causes the brittleness.
formation of globular cementite. In this condition the For effective hardening, the steel is first heated to
steel has minimum hardness and the lowest elastic limit above A3 point and maintained long enough at this
and yield point possible for a given percentage of temperature to ensure that the structure consists
carbon. Hence this structure is aimed at where the steel entirely of solid solution of carbon in gamma-iron (i.e.
has to be severely cold-worked, as for example, in austenitic structure). Rapid quenching from this
spinning, cupping or the production of thin strip. temperature tends to retain the steel in the austenitic
The effects of hardening depend on the steel condition, which it had acquired at that high
composition and the rate of quenching. When the carbon temperature. The less drastic the quenching, the greater
is very low, hardening may be an advantage. For is the possibility of change. In carbon steels, however
example, the yield point and toughness of mild steel drastic the quenching, the gamma- to alpha-iron change
containing 0.10-0.15% carbon are greatly improved by always occurs and the microscopical constituents
water quenching from the normalising temperature observed in quenched steels are more or less advanced
instead of cooling in air. This explains why a mild steel decomposition products of the original austenitic
weld is generally so strong, where its sudden cooling condition or solid solution. Some alloy steels and the
by the surrounding base metal has the same effect as well-known austenitic stainless steels can retain the
water quenching. If the carbon is more than 0.30%, the austenitic condition at room temperature by sufficiently
increase in hardness is very appreciable. For the same drastic quenching.
rate of cooling, hardness increases with increase of The generally recognised stages of decomposition
carbon up to 1%. Increasing hardness is generally of austenite are given on the below:
Stages of decomposition of the gamma-iron-carbon solid solution

AUSTENITE
On tempering at progressively higher On cooling at progressively slower

temperatures below the critical range, rates through the critical range is
yields replaced by
l
Martensite Martensite
Troostite Troostite
Sorbite Sorbitic
(minutely pearlite
granular) (minutely
lamellar)
Fine coalesced Lamellar

carbide in a pearlite
groundmass of
ferrite
Coarse coalesced carbide Granular

in a groundmass of ferrite pearlite
Products of Quenching toughness. Tempering is a kind of heat treatment in

Martensite is the most important constituent of which a steel, in whatever condition it may be to start
hardened (untempered) steel. In steel of any given with, is reheated to a temperature not exceeding the
composition, it is the hardest of the possible constituents lower limit of its critical range, followed by either rapid
obtained by heat treatment. In its formation, the or slow cooling. Tempering of martensite is further
allotropic change from gamma- to alpha-iron has been described later in this section.
completed, but the carbon (or carbide) remains in almost The products of tempering are martensite, troostite
as highly dispersed a condition as in the solid solution, and sorbite. These may have different characteristics
austenite. It has acicular (needle-like) appearance as from those identical constituents formed in quenching.
shown in Fig. 9.10. Martensite: A mild tempering of some alloy steels
Troostite is a structureless black-etching constituent. obtained in the austenitic condition by quenching gives
The carbide, though still irresolvable, is less highly martensite, structurally similar to that produced by
dispersed than ill martensite. Troostite is produced when quenching. The change is accompanied by an increase
the rate of cooling is less rapid than that required to in hardness.
produce martensite. With less severe quenching, a Troostite: A mild tempering of quenched martensitic
structure intermediate between troostite and sorbitic carbon or alloy steels gives rise to troostite, structurally
pearlite is formed. similar to that obtained by rapid quenching, but with a
Sorbitic pearlite: If the rate of cooling is still slower slight difference. The change from martensite to troostite
(as, for example, that of large masses in oil, or of smaller is accompanied by a slight fall of hardness and an
specimens in hot water, or in an air blast), opportunity increase in toughness. Troostite is present in properly
is afforded for the separation of more ferrite or cementite hardened and tempered cutting tools, chisels, knives,
from the solid solution, though the final separation of etc.
the eutectoid, pearlite, is partially suppressed. The Sorbite is the essential constituent of hardened and
lamellar pearlite formed on slow cooling is replaced by tempered steels intended for constructional purposes.
grains of sorbitic pearlite. The sorbite in this case is not sorbitic pearlite: its
structure when resolvable is minutely granular, and not
Products of Tempering lamellar.
A hardened steel which is too brittle for use is usually
tempered to reduce hardness sufficiently and to improve Isothermal Transformation Diagrams
It is obvious from the earlier discussion on the iron-
carbon phase diagram, that it does not give information
on the transformation temperatures and cooling rates
required to produce other useful structures resulting
from hardening and tempering. A practical and useful
diagram in this respect is the isothermal transformation
d i a g r a m , which is also termed time-temperature
transformation or TTT diagram. It vividly explains the
response of a steel to increased cooling rates and indicates
the cooling rates required to transform the austenite to
pearlite, bainite or martensite as desired.
Each steel composition has its own characteristic TTT
diagram. Figure 9.11 shows the diagram of 0.80% plain
carbon steel of eutectoid composition. It was obtained
by heating several samples of the steel to approximately
850°C (above A3,) to give them a fully austenitic structure
a n d a b r u p t l y cooling t h e m d o w n to a series of
temperatures between 705°C and room temperature by
plunging them into various liquid media such as brine
solution, oil and water at the desired temperatures. Each
Fig. 9.10: Quenched martensite (500 x before specimen was kept in its bath for a specified length of
reduction, etched) showing acicular structure time, removed and cooled rapidly to room temperature.
The resulting microstructure was examined under a A striking feature of this exercise is that, the samples
microscope. which were cooled rapidly enough to get past the nose
If we take, as example, the temperature of 700°C on of the curve ended up with the structure of hard mar-
the vertical axis of the diagram and draw a horizontal tensite. This is clearly depicted in the diagram. The dia-
line, we find that the specimen held at this temperature gram also shows that the martensite transformation
commenced transformation only after eight minutes started at 230°C and was complete at 120°C These tem-
(shown by point S) and completed it after about one peratures are indicated by the letters Ms (martensite
hour (point E). The final structure was coarse pearlite start) and Mf (martensite finish) respectively. Ms and
and the steel fairly soft (hardness of 15Rc). Mf temperatures vary with the steel chemistry. As its
Next, taking 565°C, the diagram shows that the carbon content increases, Ms and Mf temperatures are
transformation started almost at once (point T) and was lowered.
complete in five seconds (point F). The transformation It is to be noted that M$ and Mf temperatures are
period was the shortest at this temperature of 565°C, denoted by lines and not curves. This indicates that
which is referred to as the nose of the curve. The martensite is not formed by normal nucleation and
resulting structure was fine pearlite and the steel harder growth and that its formation is independent of time. It
(hardness of 41Rc). is temperature dependent, and its formation takes place
4
At all temperatures below 565°C, the initial time for mostly while the temperature is falling. Thus marten-
the start of the transformation increased, the site transformation is also called as diffusion-less shear
microstructure was bainite and the hardness of the transformation. If a steel is held for a period between
sample had increased further. Ms and Mf, some bainite is formed isothermally and
°F
TRANSFORMATION
AUSTENITE
800 AT 705°C
(1300 °F)
1400
A, TEMPERATURE STARTS E JENDS
700 AUSTENITE
S
1200
w 600
B
z 500 h
o
p
400
<
300
200|
lOOh
MARTENSITE
i-HV- J_8 J I 1 I L -X.

15 30 1 2 4 8 15 30 1 2 4 J8 15
I-
SECONDS -MINUTES- +\ HOURS
TIME OF TRANSFORMATION
Fig. 9.11: The T I T diagram for the transformation of austenite in an eutectoid (0.80% carbon)
plain carbon steel
some austenite remains untransformed. When the room To explain the behaviour of the second group, the
temperature approaches, retained austenite transforms isothermal diagram for 1.97%- chromium steel is shown
slowly and partly to martensite. in Fig. 9.12.
It has already been stated that the isothermal It is seen that from the transformation range (A3, A1
diagram, which indicates transformation behaviour, points) to about 650°C, the pearlite transformation curve
varies with steel composition. Other factors those cause is similar to that for 0.80% carbon steel, except that it
such variation are austenite grain size and austenite has moved appreciably to the right. At 540°C and below,
homogeneity or heterogeneity. Referring to composition, the slopes turn sharply to the left. The structure here is
it has been observed that carbon, manganese below 1%, bainite. A point to be noted is that, in this steel bainite
silicon below 1.5%, nickel and copper shift the curves requires less time to form than pearlite while opposite
of the diagram to the right, that is, to longer times of is the case with a carbon Heel. The carbide formers,
transformation, without changing their shapes especially molybdenum, hardly affect the rate of bainite
markedly. Carbide formers like chromium, formation. To avoid this structure, nickel or manganese
molybdenum and vanadium also shift the curves to the must be added to the steel.
right and, in addition, alter their shapes to a remarkable The rate of pearlite transformation slows down as
extent. When the curves move to the right, conditions the grain size of the austenite increases. (The austenite
producing less rapid cooling can be used while still grain size is determined by the temperature above A3-
allowing martensite to form. These steels are said to have point, to which the steel is heated before cooling it. The
higher hardenability. higher this temperature, larger is the austenite grain
•C °F
800
1400
700
1200
600
1000
500
e
o§ 400
w
a*
2
P
300
200 L 400
100 h 200
0 L
«0 ~ <S V>
- « * S 8 8S'8-S
TIME, SEC. " " " S" 8" 8,8"
Fig. 9.12: Isothermal transformation diagram for a steel containing 0.3% carbon, 0.45% manganese, and 1.97%
chromium. Austenite grain size: 6 to 7, Austenitised at 1,600°F (870°C)
size.) This is because pearlite starts to nucleate from the This information is very vital for correct hardening of
grain boundaries and grows inward through the grain. steel, because the formation of pearlite or bainite has to
A coarse grain means greater distances over which the be completely avoided. The CCT diagram can also give
growth has to take place. On the other hand, bainite us the conditions required for completely avoiding the
formation is little affected by grain size. formation of martensite, which is largely responsible
Heterogeneity of austenite arises from segregation for cracking in the heat-affected zones of welded joints,
and from undissolved carbides and non-metallic and also for obtaining desirable microstructures. We
inclusions. Segregation means that some areas of the shall deal with CCT diagrams in detail in the section on
steel have lower contents of carbon and alloying heat-affected zones.
elements, while others are richer in these elements. At this point it is necessary to explain the term
Austenite in the former areas will transform more hardenability, which is sometimes confused with
readily than it would, if the steel had homogeneous hardness. The maximum hardness of a steel is primarily
composition. On the other hand, austenite in the latter determined by its carbon content. A steel is said to be
areas will transform more sluggishly. highly hardenable when it will form martensite even
Undissolved carbides and non-metallic inclusions with slow cooling in air. On the other hand, a steel which
speed u p transformation by serving as points of requires extremely high cooling rate to transform
nucleation. Carbides also tie up carbon and sometimes completely to m a r t e n s i t e is said to have low
alloying elements too, and prevent them from being hardenability. Thus hardenability is a measure of how
fully effective in retarding the transformation of the easily martensite forms on quenching. It is an important
austenite phase. factor of weldability of steel, since it determines the
Besides carbon, alloying elements also lower Ms and extent to which a steel will harden in the HAZ during
Mt points. In some steels, the Mf point may be below welding. A steel having a coarse grained micro-structure
the room temperature. In such cases, some austenite is is more prone to martensite formation over a fine-
retained in the steel after quenching. In the case of grained steel u n d e r identical conditions of
austenitic stainless steels, the Ms point is below the room transformation.
temperature and hence a fully austenitic structure results
when the material is quenched from above the Appoint. Tempering of Martensite
Ms and Afrtemperatures are also influenced by austenite Martensite formed by quenching is hard and brittle
homogeneity, but grain size has little effect, since as and therefore unsuitable for practical use. By heating
stated earlier, martensite formation does not involve the martensite to some appropriate temperature below
nucleation and growth. Aj, its ductility and toughness can be increased, while
The formation of martensite on q u e n c h i n g is there is a marginal fall in its strength and hardness. This
explained as follows: The extremely rapid cooling from heat treatment is termed as tempering, and the resulting
above A3, prevents the transformation of austenite to modified microstructure is called as tempered
ferrite and cementite. However, the face-centred cubic martensite. Tempering causes the carbon originally
lattice of the austenite reverts to body-centred cubic dissolved in ferrite to precipitate as tiny carbide
lattice and the carbon remains dissolved in the resulting particles. By correctly choosing the time and
alpha-iron instead of changing to cementite. Martensite temperature of tempering, the desired Compromise
is therefore best described as a supersaturated solution between hardness and toughness can be obtained. As
of carbon in ferrite. As the carbon content increases, the the tempering temperature increases, the steel becomes
martensite becomes increasingly hard and brittle. softer and tougher.
The TTT or isothermal diagram provides heat Quenched and tempered steels having the optimum
treatment procedure data, where isothermal cooling is combination of strength, toughness and weldability, and
used. Actually continuous cooling and not isothermal, high yield-to-tensile ratios, are being used in the
is used in the hardening of steels and, in fact, the heat- fabrication of structures and pressure vessels in
affected zones are subject to continuous cooling in increasing tonnage in recent years. We shall deal with
welding. Therefore continuous cooling transformation them in detail in the Chapter on Weldability of Metals,
(CCT) diagrams have also been derived for steels. TTT
and CCT diagrams are closely interrelated, but with the Grain Refinement
latter the slowest cooling rate which will allow-100% Fine-grained steels are superior to coarse-grained
martensite to form in a steel can be easily determined. steels in ductility, toughness and weldability. There are
several ways of obtaining fine-grained steels including solid solution. If the cooling is slow, the beta-phase
normalising, which consists of heating the steel to a precipitates at the grain boundaries. If the cooling is
temperature slightly above A3, and cooling down to rapid, there is no time for the rejection of beta and the
room temperature in still air. single-phase supersaturated structure is retained. Both
As the temperature above A r increases, the austenite these sequences are shown by routes a-b-c and a-b-d in
grain size goes on increasing. This is called grain Fig. 9.14.
coarsening. The coarsened grains persist even when the The supersaturated structure at room temperature
steel is cooled down to room temperature. High heat (8) is identical with the structure at temperature Ta (a),
inputs during welding encourage grain-coarsening in but is not stable. If the alloy with structure (5) is reheated
the HAZ. to an intermediate temperature between T2 and T4, the
excess of dissolved beta is precipitated again. But the
Precipitation Hardening precipitate is a finely divided, uniformly distributed
Some steels a n d alloys of nickel, a l u m i n i u m , beta-phase (e). This represents the correctly age-
m a g n e s i u m a n d t i t a n i u m can be h a r d e n e d or hardened structure, which confers optimum strength
strengthened by precipitation hardening (also called on the alloy. Under-aging results in low strength. Over-
age-hardening). They constitute an important class of aging causes the fine precipitation particles to grow and
today's engineering alloys. agglomerate (f), leading to a fall in strength. An alloy
To explain precipitation hardening, we refer to the can get over-aged in service if it is exposed to
phase diagram of an alloy system in Fig. 9.13 in which t e m p e r a t u r e s above the aging t e m p e r a t u r e for a
the beta (p) phase is soluble in alpha (a) phase and the sufficiently long time.
II : WELDING METALLURGY
Weld-Metal
LIQUID-na Having completed a survey of general metallurgy,
let us now take up the metallurgy of weld-metal. The
en formation of weld-metal occurs in various w a y s
i v»^"" depending on the welding process. In the electric
a d resistance welding processes, for example, the weld is
iT»i > produced from the base metal by being heated to a
1 plastic state and subjected to pressure. The resulting
Bill >
a+p
weld, in some cases, contains a minute portion of fused
metal called the nugget. In dealing with the metallurgy
of weld-metal, it is customary to refer to arc welding
processes, in which the weld-metal is formed in a
100A X superheated molten condition from the filler metal
9kp * added. During its formation, it is exposed to gases which
Fig. 9.13: Phase diagram of a precipitation hardening are present in the arc atmosphere and to slags (if flux is
system used) present in the weld pool. As the heat source moves
further, the weld-metal is cooled by the surrounding
solubility increases with temperature. Let us consider metal. The cooling rate, which influences the final
an alloy of the composition denoted by X. This alloy structure and mechanical properties of the weld-metal,
h a s t w o co-existent p h a s e s of b a n d b at r o o m may be relatively slow or extremely fast, depending on
temperature. As the alloy is slowly heated, increasing the joint section thickness and several other factors
amounts of P dissolve in a, until the temperature Tl is which will be explained later. The weld-metal itself may
reached at which all of the available a of the alloy is be a mixture of filler metal and base metal, depending
dissolved in the a solid solution. This means that at or on the depth effusion. The extent to which the filler
above Tlf the alloy is stable as a single-phase solid metal chemistry is altered by admixture with the base
solution. metal is termed weld-metal dilution. Thus various
If the alloy X is now cooled down from Tv the solid metallurgical and other p h e n o m e n a occur in the
solution becomes supersaturated and the beta-phase, formation of a weld, making it an extremely complex
which cannot remain in solution is rejected from the material to understand.
WELDING METALLURGY 405.
SOLUTIONISED
(AT T,)
'a
BALED
ANNE-
JO « + £
g FINE PRECIPITATE
PROLONGED OR /* + " 7 ^ « \
OVERHEATING /. ". 'ALr't. \
DURING AGE JT^- /V'J
a + COARSE fiî£>^
AGE OVERAOED
^ ° / HARDENED
a +0 IN SOLUTION
Fig. 9.14: Microstructural changes in an age-hardenable alloy during several stages in heat treatment
Weld-metal is often termed simply as weld. Weld is container (example: wall of an ingot mould in ingot
defined by the American Welding Society as follows: formation) or from foreign solid particles suspended in
A localised coalescence of metals or non-metals the liquid. These are examples of heterogeneous
produced either by heating the materials to suitable tem- nucleation. In cases w h e r e such heterogeneous
peratures, with or without the application of pressure, nucleation is not available, nuclei must be provided by
or by the application of pressure alone, and with or with- the liquid itself. The liquid can do this only when it has
out the use of filler material. cooled well below the normal solidification temperature.
In other words, nucleation or commencement of the
Structure solidification process requires supercooling of the liquid
In practice, the filler metal is designed to match the metal.
mechanical properties and chemical composition of the In arc welding, the molten weld pool is contained
base metal as closely as possible. Quite often, the by the surrounding solid metal, which means that a
matching of mechanical p r o p e r t i e s is the major liquid-solid interface is always present at the weld
consideration, and therefore the weld-metal chemistry fusion boundary. This interface provides an ideal
varies to some reasonable extent from the base-metal nucleation site. Hence during the solidification of weld-
chemistry. The mechanical properties, as metallurgy tells metal, there is n o h o m o g e n e o u s nucleation and
us, are determined by the structure, which in the case supercooling is negligible. As the nucleating grains
of weld-metal, is a direct result of the sequence of events grow, some of them are pinched off, causing fewer but
occurring during welding, which include gas-metal larger grains in the weld-metal. The resulting grain size
reactions, slag-metal reactions, solidification and solid- has a marked effect on mechanical properties.
state reactions which follow solidification. Generally the heat flow in the weld zone is highly
Solidification: At the start of this chapter, we had directional toward the adjacent cold metal and hence
explained how the solidification of liquid metal on the weld acquires a distinctly columnar structure, in
cooling commences at specific points called nuclei, how which the grains are long and parallel to the direction
these nuclei grow into grains or clusters of crystals and of heat flow as shown in Fig. 9.15. In the case of the
how the grains with differing orientations meet at grain deep pear-shaped weld shown on the right, it is seen
boundaries. that columnar grains growing from opposite sides meet
Thus the starting point for a grain is a nucleus. at the middle of the weld. This midplane, where groups
Nucleation m a y commence from the wall of the of intersecting columnar grains have solidified last, also
grain size also has certain relationship with heat input,

but it is disturbed by other factors such as the presence
-<jflffiSB»*- of grain-refining agents and vibration.
When solidification is extremely rapid, dendrites do
not develop fully. Under this condition, projection of
the solidifying interface into the liquid weld pool is
Fig. 9.15: Columnar structure of welds much shorter. The resulting structure is called a cell
structure. In a cellular structure, the spacings between
cells are smaller than those between dendrites and
often contains impurities and porosities. It Is therefore segregation of solutes is less extensive. Cellular and
prone to fracture at low strains. This type of dendritic growth patterns and the manner of solute
solidification defect is sometimes referred to as ingotism. distribution (i.e. segregation pattern) are illustrated in
The problem of ingotism can be overcome by adjusting Fig. 9.16.
the joint gap configuration and welding procedure, so Gas-metal reactions: The molten weld-metal produced
that the geometry of the columnar grain structure is during arc welding must be protected from oxygen and
suitably changed. nitrogen of the air, because their ingress into the weld
Similarly the coarse-grained, columnar structure renders it brittle and porous. It is a little difficult to
of the weld-metal can be improved upon by reconcile to this fact when one considers that in steel
hanging the weld-metal chemistry, bearing in mind that making and casting, steel is melted and poured in air
alloys which solidify over a narrow temperature range without any detrimental effect. A valid explanation is
develop the most columnar and coarse-grained weld- that, the gases oxygen, nitrogen and water vapour ionise
metal. Another way is to vary the welding process, so and become highly reactive in an electric arc and thus
that the pattern of turbulence in the liquid metal is become harmful. They go in solution at higher
suitably altered. temperature but try to come out when the temperature
Dendrite formation: Most weld-metals are subject to drops down causing defects like, porosity, crack, etc. in
rapid solidification, which involves micro-segregation the weld metal.
of alloying elements. During the solidification process,
the alloy elements which lower the melting point are
progressively displaced by the growing solid and are
forced to shift from the liquid-solid interface by diffusion
into and through the liquid. Since diffusion is a time
consuming process, the low-melting liquid tends to
accumulate at the interface. This accumulation impedes
growth and causes the formation of so-called dendrites,
which have a characteristic configuration of filaments
and branches. A dendrite is a special feature of the weld-
metal structure and reflects the irregular shape taken
by the liquid-solid interface during solidification. In an
etched cross-section of a welded joint, one can visualise
how the growing solid dendritic arms have pushed
themselves through the surrounding layer of low-
melting liquid.
A single grain can contain thousands of dendrite NOTE GREATER
branches. This shows that dendrite arm spacing is of I DISTANCE
very minute magnitude. The more rapid the J BETWEEN SOLUJEl

(RICH REGIONS {
solidification, the more closely spaced are the dendrites. J X
This relationship between dendrite arm spacing and LOCATION LOCATION
CELLULAR GROWTH DENDRITIC GROWTH
energy input during welding is fairly consistent, and
has been used to estimate the energy inputs used on Fig. 9.16: Schematic representation of solute
specific welded jobs, for which welding data were not distribution for cellular and dendritic growth patterns
available, by measuring the dendrite arm spacing. The
Oxygen gas is soluble to the extent of 0.2% in molten manganese, aluminium and silicon into the weld-metal,
steel, and the solubility decreases as the melt cools. through the flux or the filler wire. These elements react
Further oxygen is used up in converting iron to iron preferentially with oxygen thus:
oxide (FeO), which is a brittle non-metallic product. In Si+ 20 — > Si0 2
steel, which is basically an iron-carbon alloy, oxygen 2 Al + 30 > A1203
combines with carbon to form carbon monoxide (CO). Mn+O > MnO
Carbon monoxide bubbles out as the molten weld-metal Such oxides formed in the weld pool may rise to the
cools, but a part of it gets entrapped as a result of rapid surface as slag or remain entrapped in the weld in the
solidification. The result is weld porosity. form of complex oxides. They are recognised as non-
Nitrogen also has a similar reaction. Quantities of metallic inclusions. Such inclusions are present even in
nitrogen dissolve in the molten weld and are liberated high quality welds, but they have very little effect on
as the metal cools. Iron also chemically combines with mechanical properties, because they are extremely fine
nitrogen to form iron nitride, which gets entrapped in and widely dispersed. In badly made welds they may
the weld and renders it brittle. be present in larger amounts. The result is low impact
Protection of the arc and the weld pool from the properties, which mean reduced toughness.
atmosphere is therefore an important consideration in It is important to point out here that the quantity of
the arc welding process. In mai .*&! metal-arc welding, deoxidisers provided in the filler wires and fluxes must
such protection is given by the ilux covering. In be just sufficient to balance the oxidising potential of
submerged-arc welding, it is given by the blanket of flux the weld system. Excess of silicon and aluminium in
fully covering the arc and the weld pool. In TIG and the weld, especially the latter, leads to serious loss of
MIG processes an nert gas shield is provided around ductility and toughness, especially at sub-zero
the arc. Influ*:coreu arc welding, the shielding is done temperatures. Achieving a correct balance requires an
by the fi id gas or by flux alone. intelligent selection of the process and materials.
Rfc m r to TIG and MIG/C0 2 processes, the This is best illustrated in the case of gas-shielded
arrun^ exposure to atmosphere is strongly processes. A mild steel filler metal, which is designed
dej^ ent on the techniques adopted. When dip for use with argon—2% oxygen may produce a porous
t w er (short-circuiting) technique is used in the case weld when 100% C0 2 is used as a shield .On the other
*>f MIG/C0 2 and with the TIG process generally, the hand, a filler metal designed for use with 100% C0 2 may
liquid metal is transferred in the form of large droplets give a weld of reduced ductility and toughness when
and these droplets are never in free flight through the argon—2% oxygen is used, because of over-
high temperature arc plasma. In these cases, the deoxidation. Under-deoxidation results in porosity
exposure of the weld-metal is minimum. In MIG/C0 2 which can be easily detected, but the effects of over-
welding with spray transfer technique, the metal -passes deoxidation are more subtle and not detectable by non-
across the arc as tiny droplets, which offer a large surface destructive methods. The consequences of over-
area for reaction. Also they are superheated to deoxidation are thus potentially more dangerous.
temperatures much above the melting point. Further, Water vapour is another harmful gas to be avoided.
the arc atmosphere dissociates and ionises the gas It is introduced into the arc by moisture on the joint
molecules, making them highly reactive. surface, in the flux or in shielding gases. Water vapour
A process such as electron beam welding is capable dissociates in the arc to form oxygen and hydrogen. The
of giving high quality welds even in the case of reactive harmful effects of oxygen have already been described.
metals, mainly because welding is carried out in Hydrogen is less liable to cause porosity in the molten
vacuum. Welding of titanium is best carried out in a weld-metal, provided it has sufficient fluidity to allow
closed chamber filled with an inert gas like argon, tha gas to escape. Hydrogen in excess can cause weld-
because it completely eliminates the presence of oxygen metal embrittlement. A far more harmful effect of
and nitrogen in the vicinity of the arc. hydrogen, even when present in small quantities, is the
In conventional arc welding processes, in spite of the formation of cracks in the heat-affected zones of welded
maximum protection provided, some amount of oxygen joints in high-carbon and low-alloy steels. This will be
is present in, or available to, the liquid weld-metal. This discussed in detail in the Chapter on Weldability of
is more so in C0 2 welding. The reaction of oxygen with Metals.
carbon and the resulting porosity are generally In the case of non-ferrous alloy welds, the gases to
prevented by introducing deoxidising elements like be specially guarded against are hydrogen and water
vapour. Their solution in the molten weld-metal and slag. If the arc environment is strongly oxidising, it calls
subsequent reactions and evolution lead to weld defects. for more deoxidation and this gives rise to greater
Porosity is often encountered in a l u m i n i u m and quantity of non-metallics. With reducing conditions in
magnesium welds, because of the hydrogen introduced the arc, these impurities are p r o d u c e d in smaller
into the arc by hydrated oxides present on the surfaces amounts and the resulting weld-metal is much cleaner.
of the filler wires and joint surfaces. For this reason, MIG An i m p o r t a n t factor d e t e r m i n i n g w e l d - m e t a l
welding wires of Al and Mg are stored in sealed, cleanliness is the choice of deoxidisers and their correct
desiccated containers, while those exposed to moist air proportions in the consumables. Aluminium is well
are mechanically cleaned or vacuum heated at 150°C recognised as a powerful deoxidiser, but when used
prior to use. alone it produces a fine dispersion of solid non-metallic
In the case of copper and copper-alloy welds, any particles, which remain in the weld-metal and render it
hydrogen present in the arc reacts with oxygen in the unclean. The correct procedure is to use silicon and
molten metal to give rise to water vapour and porosity. manganese (in the form of Fe-Si and Fe-Mn) as main
Suitable deoxidisers are therefore incorporated into deoxidising agents, even though they are less powerful,
copper-alloy filler metals. Aluminium weld is not prone so that they react with the bulk of the dissolved oxygen.
to oxygen porosity, because the element itself has a A small controlled a m o u n t of a l u m i n i u m is also
deoxidising effect. The porosity in Al is caused by provided to bring the final oxygen content to a very low
hydrogen as explained above. Nickel, nickel-copper and level, so that any oxidation of carbon and consequent
other nickel-alloy welds are markedly prone to porosity. porosity are avoided. When these three deoxidisers are
Filler metals for these alloys must be provided with correctly used, they will leave 0.01-0.05% metallic
strong deoxidisers. aluminium in the weld-metal.
Welds as well as heated base-metal area of titanium Another important function of welding slags is to
alloys are embrittled by contact with a number of gases control the amounts of sulphur and phosphorus in the
such as nitrogen, hydrogen and oxygen. Welding is weld-metal. Basic slags produced from lime-fluorspar
therefore carried out in vacuum or air-tight chambers type fluxes (for example, the coating of AWS EXX 18
filled with inert gas like argon, as explained in Chapter 5. class electrode) are capable of lowering the contents of
Slag-metal reactions: Fluxes used in manual metal-arc, these harmful elements in the weld-metal and thereby
submerged-arc, flux-cored arc and electroslag processes improving its ductility and toughness. Acidic fluxes, by
are converted to liquid slag in the arc. Various chemical their very chemical nature, are unable to perform such
reactions take place between the molten weld-metal and useful function.
slag, which are very similar to those which occur when Slag-metal reactions and the non-metallics they
the metal produced from the ore is being refined. The produce can sometimes give rise to a weld defect known
slag thus plays an important role in flux-shielded as hot cracking. Hot cracking occurs during the last
processes. This fact underscores the significance of stages of w e l d - m e t a l solidification, w h e n the
correct flux formulation. interdendritic liquid has much lower freezing point than
The mechanism of slag-metal reactions has been the bulk dendrite. This small liquid region is subjected
explained in detail in the chapters dealing with the to s h r i n k a g e stresses p r o d u c e d as a result of,
respective welding processes. Here it may suffice to solidification, and it develops interdendritic cracks.
point out that the slags are designed primarily to remove Since the cracks occur at temperatures near to the bulk
the deoxidation products or non-metallics from the solidification temperature, they are rightly termed as
molten weld-metal. The quantity and type of such hot cracks. They are likely to occur whenever the
compounds are dependent on the welding consumables compositional variations in the weld-metal result in the
used. In most cases they are silicates of aluminium, production of a low-melting interdendritic liquid.
manganese and iron, which rise to the surface of the In low-alloy and stainless steel welds, hot cracking
weld pool and become part of the slag. A minute is usually caused by low-melting iron and alloy
proportion of the slag, however, remains entrapped in sulphides, t h o u g h in stainless steels silicates also
the weld as non-metallic inclusions. The ultimate promote it. It therefore becomes necessary to control the
objective is to obtain clean weld-metal, i.e. with the amount and type of such sulphides in the weld-metal.
lowest amount of non-metallic inclusions. This depends In the case of mild steel and other ferrous weld-metal,
on the quantity of non-metallics generated in the weld manganese-to-sulphur percentage ratios are kept above
pool and the extent to which they are removed with the 30 to prevent hot cracking. In austenitic stainless steels,
a fully austenitic structure in the weld-metal is more microstructure and its mechanical properties, especially
prone to hot cracking than a duplex structure consisting toughness at low temperatures. At the same time factors
of austenite with 3-8% ferrite. relating to the composition of welding consumables and
Solid-state reactions: The solid-state reactions welding procedures (especially weld cooling rates)
occurring in steel weld-metal during cooling from w h i c h p r o m o t e the formation of favourable
above A 3 point are indicated by the iron-carbon phase microstructures in the w e l d - m e t a l have been
diagram. Welds are subjected to very rapid cooling established. Sufficient knowledge of welding metallurgy
rates and, unless the carbon content is kept very low, is available today, to obtain welds of high toughness
the transformation products are likely to be martensite and ductility demanded by increasingly critical service
and other hard constituents instead of an aggregate of conditions.
ferrite and carbide. If in addition, hydrogen is present
in the arc atmosphere, it can promote cracking in the Weld Microstructure
weld-metal. The cracking defect occurs after the weld Metallurgists have always recognised that the weld
has cooled to quite a low temperature, even below microstructure is extremely complex, because of the
150°C. It is therefore termed cold cracking or delayed unique manner in which it is formed and deposited.
cracking. Yet the microstructure of a simple mild steel weld had
Cold cracking of the weld-metal is easily avoided not been adequately established until recently. How this
by resorting to one or more of the following steps: microstructure is constituted; how it is influenced by
a) Carbon content of the weld-metal is kept low chemistry and welding procedures; what is the relation
b e t w e e n w e l d microstructure and mechanical
b) The cooling rate of the weld-metal is reduced by
properties, especially toughness; what is the influence
using higher heat input or by resorting to pre-
of non-metallic inclusions, oxygen and nitrogen contents
heating on toughness; these vital questions had not been
c) Low-hydrogen consumables are used. satisfactorily answered for a long time.
Cold cracking is not much of a problem in the case Until 1975 there was no universally accepted stan-
of weld-metal, but occurs quite commonly in the heat- dard terminology to describe weld-metal microstruc-
affected zones of the base metal. Hence it will be tures, thereby making it difficult for investigators to
discussed in detail under Weldability of Metals. compare their work with others. In that year, Widgery
An important point to be considered with the weld- categorised the microstructural constituents in as-de-
metal is that the desired mechanical properties must be posited weld-metal as:
obtained in the as-welded condition, since heat 1) Proeutectoid ferrite (light etching)
treatments such as those used for steels are ruled out. 2) Intermediate lamellar products, mainly ferrite
In some instances, weldments receive a postweld heat side plates resembling upper bainite (light etch-
treatment, which mainly relieves residual stresses and ing)
softens a hardened heat-affected zone to some extent.
3) Acicular ferrite, consisting of a fine structure of
No greater changes occur, since the temperature of heat
interlocking ferrite plates (dark etching); and
treatment is much below the A3 point.
The tensile and other properties of the ferrous weld- 4) Other constituents such as carbides, martensite
metal, whose carbon level is kept low, is usually and inclusions.
obtained by solid solution strengthening and fine Shortly later, Widgery and Davey came out with a
particle strengthening effects. This explains why the method of assessing the proportion of these various
manganese content of a weld is higher than that of the constituents, which involved a systematic point count
corresponding base metal. to determine the constituents lound at each of 1,000
Most weld-metals n a t u r a l l y d e v e l o p a fine points arranged in a two-dimensional array.
solidification structure. The solidification dendrites are Dr. Evans of Oerlikon was one of the first to use
apparently large, and yet the fine dendrite branching Widgery's technique to investigate the microstructural
between the major dendrite columns produces a fine changes taking place in mild steel weld deposit as its
microstructure, which is e n c o u r a g e d by the Mn content increased from 0.60 to 1.8% and how they
strengthening mechanisms. influenced weld properties, especially toughness. He
In recent years, m u c h w o r k has been d o n e made four experimental electrodes of iron powder basic
internationally to establish relationship between weld types (AWS E7018 class) giving Mn levels of 0.65. 1.0,
1.4 and 1.8% in the weld and coded them A, B, C and D

respectively. jE-GftAlNED
Multi-run deposits were m a d e with these four COARSE-GRAINED TOP BEAD-COLUMNAR
electrodes according to the procedure described in the \ V—— , /
International Standard ISO 2560-1973. Each deposit
consisted of nine layers with three passes per layer
- \ ^#g* ~
^ 5 . ^ ^ f r ^ _ _ ^ * "^
deposited side by side. A typical cross-section, etched 1 y SSJoE^_^r y
and magnified for macroscopic effect, is shown in Fig.
9.17.
The figure reveals a very significant point that the
ill
multi-pass procedure has resulted in two distinct zones:
columnar or as-cast and a transformed zone. The latter
zone represents the portion which was heated to above
the A3, point during deposition of a subsequent pass,
J
resulting in a change from columnar structure to
equiaxed and a change in grain size. This transformed Fig. 9.18: Diagram of top bead and adjacent areas
zone itself has two distinct sub-zones: the coarse-grained
one which was heated far above the Appoint and the
fine-grained one heated a little above A3 point. Typical variation in such distribution among the four deposits
locations of these regions in and around the top bead and the average worked out to 20% columnar, 34%
extracted from the earlier figure are shown in Fig. 9.18. coarse-grained and 46% fine-grained.
The reason for using deposits made according to ISO The effect of Mn on all-weld tensile properties is
2560 for this investigation is that, all-weld tensile shown in Fig. 9.20. The tensile test results confirm that
welding consumables. The impact values obtained are
influenced to some extent by the p r o p o r t i o n a l
distribution of the columnar, coarse-grained and fine-
grained zones in the Charpy-V specimen. The zone JCOARSE- £ j | FINE-
30LUMNAR |
distribution for the four deposits in the as-welded and GRAINED GRAINED!
stress-relieved conditions (2 h/580°C), as measured in
the vertical midplane region, is shown in Fig. 9.19. The
PLATE
central 10 mm region from which the impact specimen SURFACE
is extracted is also indicated. There was a marginal
<
a, 5
O
u
<
o
PC
15
w
u
20
•STRESS-RELIEVED
Fig. 9.17: Cross-section of multi-run deposit as per ISO
2560-1973 Fig. 9.19: Zone distribution along the vertical centre line
250 T f T r—T r-
U.T.S. A H 30
C _ __ CHARPY-V
D
200
/ ^ - ~
> /?7>-
A
- i 20
g 150 - / / • /
/ / / /
2
u
Q
/ /.' /
CO
o ioo / /; /
09
< /''' /
- i 10
#AS-WELDED /,•/ /
OSTRBSS-RELIEVEDl
50
s'
0 "I -L i i i i i
MANGANESE IN WELD, % -80 -60 -40 -20 0 20
Fig. 9.20: Effect of manganese on the tensile properties TEST TEMPERATURE, °C.
of the multi-run deposits Fig. 9.21: Charpy-V notch impact results (as-welded)
manganese increases the yield strength and tensile

strength of iron-manganese alloys. For the range of
m a n g a n e s e contents investigated, solid-solution
h a r d e n i n g a n d grain refinement led to a linear
influence, an increase of 0.1% Mn increasing the tensile
parameters by 10 N / m m 2 . The tensile properties
decrease after stress-relieving, the drop being greater
in the case of yield strength and high manganese levels.
Carbide precipitation occurred at grain boundaries
during the heat treatment, but evidently no secondary
hardening occurred in the plain C-Mn weld metal
system over the time involved. Results of impact tests
in the and Charpy-V impact specimens machined out
from them are used for the standard assessment of arc
temperature range from 20° down to -80°C are shown
in Fig. 9.21. The results have been rearranged in Fig.
9.22 to highlight the effect of Mn on Charpy-V values
at different temperatures. It clearly establishes that the
maximum toughness is exhibited at the Mn level of
1.5%.
For quantitative metallographic measurements, a
portion of the top central bead of each specimen, as 1.0 1.5 2.0
shown in Fig. 9.18 was examined at x 200 magnification. MANGANESE IN WELD, %
A typical photomicrograph representing the columnar Fig. 9.22: Effect of manganese (Charpy-V)
top bead is shown in Fig. 9.23, in which light etching
proeutectoid ferrite is m a r k e d 1, light etching The results obtained on point counting are shown
intermediate lamellar products are marked 2, and in Fig. 9.24, They show that the amount of acicular
aciculai ferrite is marked 3. Carbides and martensite ferrite increases m a r k e d l y at the expense of
were not noticeable. proeutectoid ferrite as the Mn content increases. The
Fig. 9.23: Photomicrograph of top bead (columnar)
proeutectoid ferrite and the lamellar products have the

100
B c D opposite effect.
-r ~r-
• AS-WELDED While the proportion of acicular ferrite keeps on
# 90 O STRESS-RELIEVED increasing throughout, optimum toughness is observed
H at 1.5% Mn and not at higher levels. This shows that the
Z 80
w pattern of behaviour is dependent on the competing
zo 70 ACICULAR actions of Mn to:
FERRITE (3) 1) Increase the yield strength by hardening the
o 60 ferrite
u
2) Increase the acicular ferrite volume fraction and
50
3 to refine the grain size in the reheated region.
40 Stress-relief does not influence Charpy-V results.
This is explained by the fact that the decrease in
Io 30
20
toughness expected as a result of carbide precipitation
is compensated for by a softening of the ferrite.
PRO-EUTECTOID Using identical four electrodes (A, B, C, D) and
u similar t e c h n i q u e s a n d p r o c e d u r e s , Dr. Evans
10 FERRITE (1)
investigated the influence of the following factors on
0 I weld microstructure and mechanical properties:
0.5 1.0 1.5 2.0 1) Increase of interpass temperature from 20°C to
MANGANESE IN WELD, %. 300°C while making the deposit
Fig 9.24: Effect of Mn on microstructure of top bead
2) Electrode diameter ranging from 3.25 to 6 mm
proportion of intermediate lamellar component also 3) Heat input ranging from 0.6 to 4.3 k j / mm
decreases with increasing Mn. It is obvious that acicular 4) Carbon content of the weld ranging from 0.045
ferrite is c o n d u c i v e to g o o d t o u g h n e s s , w h i l e to 0.145%
5) Manganese content from 0.6% to 1.8% mained unchanged but the lath size of the acicu-
6) Silicon content of the weld ranging from 0.2 to lar ferrite increased.
0.9% b) The average width of the columnar grains in
7) Sulphur and Phosphorus contents of the weld as-deposited weld-metal increased.
ranging from 0.005 to 0.045%. c) The microstructure of the high temperature
reheated regions coarsened, the width of the fer-
8) Chromium and Nickel content
rite envelopes increasing.
9) Molybdenum, Niobium, Vanadium content
d) The grain size of the equiaxed fine-grained regions
10) Titanium and Boron content increased.
11) Effects of Aluminium, Copper, Oxygen e) The percentage amount of unrefined weld-metaJ
12) Effect of Iron powder in the central regions of the deposits increased.
13) Welding position f) The hardness variations in the bulk of the
deposit became more pronounced and
14) Effect of stress-relieving
corresponded to the wide unrefined regions.
15) Normalising for 30 min at 930°C and tempering
g) The yield and tensile strengths decreased but not
for 2 hours at 580°C to a marked extent.
16) Effect of interpass temperature h) The Charpy-V impact properties deteriorated, the
17) Effect of electrode diameter lateral displacement being dependent on the Mn
In investigations (1) to (5), intermediate lamellar content.
products have been substituted by the term bainite i) The optimum Mn level with regard to impact
(ferrite side plates). For (7) and (8), Dr. Evans has used properties remained at approximately 1.4%.
the revised weld microstructure terminology of the 3) Increase in heat input resulted in the following:
Welding Institute explained in chapter no. 9. a) The number of beads per layer varied and the
1) He found that increasing interpass temperature re- nugget and recrystallised area increased.
sulted in the following: b) The hardness of as-deposited weld-metal de-
a) Increased the width of the recrystallised zones creased.
and this tended to eliminate the columnar bands
c) The average width of the columnar grains in-
from the bulk of the deposit.
creased.
b) Coarsened the microstructures of as-deposited
weld-metal, acicular ferrite being replaced by d) The amount of proeutectoid ferrite in as-depos-
proeutectoid ferrite and, to a slight extent, ferrite ited weld-metal increased at the expense of the
side plates (bainite). acicular ferrite.
c) Increased the lath size of the acicular ferrite in e) The lath size of the acicular ferrite increased.
as-deposited weld-metal. f) The microstructure of the high temperature re-
d) Coarsened the microstructure of the high heated regions coarsened, the width of the fer-
temperature reheated regions, the width of the rite envelopes increasing.
ferrite envelopes increasing. g) The grain size of the equiaxed fine-grained re-
e) Increased the grain size of the equiaxed gions increased.
fine-grained regions. h) The amount of columnar weld-metal remaining
in the bulk of the deposit varied.
f) Lowered the yield and tensile strengths,
the loss being dependent on weld-metal i) The yield strength and the tensile strength de-
Mn content. creased.
g) Had a beneficial effect on the impact properties. j) The strengthening effect of manganese de-
h) The optimum Mn level remained at 1.4% with creased.
respect to impact properties. k) Optimum Charpy-V impact properties were
2) Increase in electrode diameter resulted in the follow- achieved at 2 kj/mm, when the notch was cen-
trally located.
ing:
a) The relative percentages of the microstructural 1) The optimum Mn level with regard to impact
components in as-deposited weld-metal re- properties remained at approximately 1.4%.
4) Dr. Evans found that the following occurred on in- i) A reversal in impact property relationships
creasing the carbon content of the weld: occurred at manganese contents above 1.2%.
a) The average width of the prior austenite grains j) Optimum impact properties were maintained at
decreased. a manganese content of about 1.4%.
b) The amount of aciciilar ferrite increased at the k) The impact properties, at the optimum, were com-
expense of the proeutectoid ferrite.
parable when considered on the basis of heat in-
c) The aspect ratio of the acicular ferrite changed, put.
and increasing amount of carbide formed be-
6) The following occurred when the four deposits A,
tween the laths.
B, C and D were stress-relieved at 580°C for two hours:
d) Grain refinement occurred in the high tempera- a) The m o r p h o l o g y of second-phase particles
ture reheated regions. changed, breakdown occurring to ferrite and car-
e) Grain refinement occurred in the low tempera- bide.
ture reheated regions.
b) Pearlite and cementite films were spheroidised.
f) Increasing amounts of second phases were pre-
cipitated in the fine-grained regions. c) Grain boundary carbides developed and subse-
quently coarsened.
h) The yield and tensile strengths increased linearly.
d) The hardness of deposited and reheated regions
i) The Charpy-V curves were tilted the upper shelf
decreased.
being depressed, whereas the lower shelf was
raised. e) The yield and tensile strength decreased.
j) The degree of scattering in the Charpy test was f) Notch toughness improved at low C and Mn
reduced. levels and deteriorated at high C and Mn levels.
k) Notch toughness improved at low Mn levels and g) At intermediate levels, no essential change in
deteriorated at high Mn levels. toughness occurred, the optimum being retained
1) At an intermediate Mn content (1.4%), the at 1.4% Mn w h e n the carbon content was
t o u g h n e s s initially i m p r o v e d a n d then between 0.07 and 0.09%.
deteriorated, an optimum occurring in the range 7) For studying the effects of normalising and tem-
0.07 to 0.09% C. pering, the all-weld metal deposits A, B, C and D were
m ) A t any specific carbon content, o p t i m u m normalised for 30 min at 930°C and a duplicate set was
toughness was encountered at 1.4% Mn. additionally tempered for two hours at 580°C. Weldments
5) Dr. Evans noted the following points on reverting from C (1.4% Mn) were also normalised at 880,980,1030
from the flat to the vertical-up welding position at and 1080°C for 30 min and further tempered as 0...23
equivalent heat-inputs: mentioned earlier.
a) The weld-metal carbon, manganese, silicon and The specimens were subjected to conventional met-
oxygen were increased. allographic examination and their carbon replicas were
additionally examined under a scanning electron mi-
b) The nitrogen content was decreased. croscope. In this and subsequent studies, Dr. Evans has
c) The bead size was greater. used the revised terminology on weld-metal micro-
d) The degree of recrystallisation, in the central structure of the Welding Institute explained in chapter
region of the deposits, was less. no. 9.
e) The columnar grains were inclined at a flatter The microstructure of the weld-metals normalised
angle to the bead surface. at 930°C is seen to be polygonal ferrite existing in
combination with dark etching second phases, which
f) The hardness of the as-deposited weld-metal was
i n c r e a s e d in a m o u n t w i t h i n c r e a s i n g Mn. The
identical.
microphases were identified as being cementite films
g) The microstructures of the as-deposited metals, (CF),_p_earlite (P) and martensite/austenite/bainite
the coarse grained regions and the fine grained (M/A/B). The total amount of second-phase increased
regions were comparable for both w e l d i n g systematically as the manganese content increased. In
positions. addition, t h e ^ o r m of the phases was found tc be
h) The yield strengths and the tensile strengths were critically dependent on the weld-metal composition.
essentially the same. At 0.6% Mn, the majority of the carbide existed as grain
boundary film. With increasing manganese, fewer c) The aspect ratio of the acicular ferrite changed
films were evident and more pearlite was observed a n d increasing a m o u n t of second-phase
together with a lighter etching duplex phase. At 1.8% wasretained.
Mn, large amounts of this latter intermediate phase, d) Although grain size in the low temperature
designated as M / A / B , coexisted with a lesser amount reheated region r e m a i n e d u n c h a n g e d , the
of pearlite. amount of second-phase increased.
After tempering of normalised low Mn weld-metal,
e) The m o r p h o l o g y of these second-phases
the cementite films tended to globularise at the grain
changed, cementite film and B/P being replaced
boundaries and partial breakdown of the pearlite
by M / A .
occurred. Similarly, the pools of M / A / B in the high Mn
deposits dissociated after tempering, with the formation f) The hardness increased non-linearly.
of fine carbide particles within the regions occupied by g) The tensile parameters increased non-linearly.
the colonies. h) The notch toughness deteriorated to an extent de-
Average grain size measurements for 1.4% Mn pendent on the manganese content.
deposit normalised at temperatures between 880 and
i) A silicon content of up to 0.5% can be tolerated
1080°C indicated that at 880°C, incomplete
when the manganese content is optimised.
normalisation had occurred. The extent of grain-
coarsening was only g r a d u a l w i t h increasing 9) The following was found on increasing sulphur and
temperature and even above 1,000°C, the grain size was phosphorus:
still by no means excessive. After normalising at the a) Sulphur increased the volume fraction of ferrite
higher temperatures, however, the pearlite colonies were with aligned MA-C in the as-deposited metal; this
large and much better defined. change in transformation is attributed to the ex-
The effects of full heat-treatment (normalising) have istence of a layer of MnS on the surface of inclu-
been summarised by Dr. Evans as follows: sions.
a) The ferrite grain size increased with increasing b) Sulphur had no effect on the grain size of reheated
temperature. weld-metal, but did change the morphology of
the second phases, from M / A to pearlitic-type
b) The ferrite grain size decreased with increasing
structures and cementite films.
manganese content.
c) Sulphur lowered the hardness, decreased the ten-
c) The total amount of second-phase increased with
sile properties and drastically reduced notch
increasing manganese.
toughness.
d) The m o r p h o l o g y of the s e c o n d - p h a s e w a s
d Phosphorus had no apparent effect on the micro-
dependent on weld-metal composition, grain
structure of either the as-deposited metal or on
boundary cementite films being replaced succes-
the reheated regions.
sively by pearlite and martensite/austenite/
bainite as the manganese content increased. e) Phosphorus increased the hardness and the ten-
sile properties, but had little effect on notch
e) The hardness of the deposits decreased.
toughness.
f) The yield and tensile strength decreased.
f) Sulphur exerts its influence through non-metal-
g) The notch toughness deteriorated but the opti- lic inclusions and phosphorus through ferrite
mum was still retained at approximately the 1.5% strengthening.
Mn level.
g) Little benefit is to be gained by reducing sulphur
h) Subsequent t e m p e r i n g lowered the tensile and phosphorus much below 0.005%.
strength and improved notch toughness to an
To understand correctly the relation between weld
extent dependent on the type of second-phase
microstructure and toughness, one must distinguish
present.
between two main mechanisms of fracture occurring in
8) Increase in weld-metal Si from 0.20 to 0.90% reweld-metal. They are (a) cleavage and (b) microvoid
sulted in the following: coalescence as shown in Fig. 9.25. Resistance of a given
a) The weld-metal oxygen decreased. weld deposit to these two mechanisms determines the
b) The amount of acicular ferrite in as-deposited s h a p e of the transition curve (toughness versus
metal increased. temperature). In order to arrive at correct solutions for
Fig. 9.25: Fracture mechanisms in weld-metal: (a) cleavage; (b) microvoid coalescence
improving weld-metal toughness, factors controlling any further increase of Mn or Mo will promote brittle
these mechanisms must be treated separately. martensite regions and lead to a fall in cleavage
COD testing is considered to be best suited for resistance.
investigating and distinguishing between the two The effects of alloying elements Mn, Ni and Mo on
fracture m e c h a n i s m s , because firstly, it assesses the development of acicular ferrite microstructures and
resistance to initiation of fracture and, secondly the the resultant toughness are very much dependent on
t o u g h n e s s values can be readily linked to the heat input rate, which determines the cooling time from
m i c r o s t r u c t u r e at the specimen crack tip by 800 to 500°C, i.e., At 800-500. Taking the example of Mn,
metallographic sectioning and to the mechanism of at low heat inputs the increase in Mn necessary to form
initiation by fractography. In Charpy-V test, it is a given proportion of acicular ferrite would be much
impossible to separate clearly the contributions of the smaller. At high heat inputs (i.e. slow cooling rates), a
two fracture mechanisms, except at the extreme ends of substantial increase in Mn would be necessary to
the transition curve, where 100% crystallinity or 100% displace the proeutectoid ferrite and upper bainite
fibrosity are developed. transformation start curves sufficiently to form acicular
Weld-metal used in critical dynamically loaded ferrite. Thus in submerged-arc welding, changing from
structures, such as offshore platforms must have high S2 wire (1% Mn) to S3 (1.5% Mn) or S4 (2% Mn) is likely
resistance to cleavage, since fracture initiation by this to be beneficial at low heat inputs but detrimental at
mechanism may lead to catastrophic failure of the high heat inputs, unless additional elements are used
structure or vessel. Today data on factors controlling to achieve sufficient hardenability to produce acicular
resistance to cleavage of weld-metals are scarce: on the ferrite.
other hand, information on factors governing microvoid The same general arguments apply to Mo and Ni
coalescence is more readily available, since upper shelf additions, which are strong acicular ferrite formers.
values from Charpy tests reflect resistance to this C o n s u m a b l e s containing 0.5% Mo, 1-3% Ni or
mechanism of fracture. combinations of Ni and Mo are often selected for high
toughness applications of C-Mn steels. At low heat
Resistance to Cleavage inputs (At 800-500°C—10 sec), Mo and Ni additions
In C-Mn weld deposits, resistance to cleavage is would be expected to be beneficial to cleavage resistance
improved and transition temperature lowered as the by promoting acicular ferrite, but at high heat inputs,
proportion of interlocking acicular ferrite is raised the situation could be reversed, unless sufficient
and what of proeutectoid ferrite and upper bainite amounts were added to develop significant proportions
is l o w e r e d . Also, the finer the plate size of the of acicular ferrite. Taking the example of Mo, the use of
acicular ferrite the better the toughness. In the heat Mo submerged-arc wires will probably be beneficial at
input range of 2-4 k j / m m , increasing the levels of heat inputs around 3 k j / m m , but they will be less
Mn, Mo and Ni. Increases the proportion of this beneficial at high heat inputs, unless other elements are
constituent and results in improved toughness. Once used to achieve adequate hardenability and to develop
the level of nearly 90% acicular ferrite is reached, acicular ferrite, e.g. Ni and Ti.
Microalloyed Weld-Metal in inhibiting upper bainite formation as mentioned

Considerable research work has been devoted to the earlier. The result is higher proportions of proeutectoid
effects of elements such as Nb, V, Ti, B and Al on weld- ferrite and acicular ferrite and improved cleavage
metal toughness. These elements may be present in resistance.
small quantities as deliberate a d d i t i o n s to the Titanium additions up to 0.04% maximum have been
consumables, or may be picked up in dilution with the found to improve Charpy-V toughness of mild steel
parent steel. welds deposited by the C 0 2 process by increasingly
The role of Nb is a little puzzling, but has been refining the microstructure and promoting more acicular
explained by the fact that Nb acts to inhibit proeutectoid ferrite. Above this level, the microstructure is further
formation at austenite boundaries, but does not prevent refined but the yield strength increases markedly and
ferrite plates growing into the grains to form upper the toughness gets reduced. In alloyed weld-metals
bainite structures. From CCT considerations, Nb pushes d e p o s i t e d by the MMA, submerged-arc or MIG
back the polygonal ferrite nose, but not that for upper processes, similar good effects have been observed, but
bainite. Thus for weld structures of low hardenability, the optimum level of Ti has been found to be 0.015% or
i.e. those in which the polygonal ferrite and upper less. In all the above investigations, the welding
bainite fractions are high and acicular ferrite low, Nb conditions and alloy contents were such as to develop
will tend to increase the proportion of upper bainite and acicular ferrite even in the absence of Ti. Titanium
acicular ferrite at the expense of polygonal ferrite at additions increased hardenability and in particular
austenite boundaries. The yield strength will also inhibited u p p e r bainite formation and promoted
increase, and the net effect would be a fall in cleavage acicular ferrite. Evidence has been gathered that Ti and
resistance. For weld structures already containing a high V additions are more effective in combination than when
proportion of acicular ferrite due either to a low heat added alone, in the same way as Nb and V are.
input or adequate alloy content at a high heat input, Nb As a microalloying element in weld-metal, boron has
will still act to decrease the polygonal ferrite fraction, stolen considerable limelight in the last five years. Its
but the resulting overall increase in acicular ferrite useful property is that, it can eliminate proeutectoid
fraction is more important than a small increase in the ferrite and promote fine-grained structures without the
bainite fraction. However the net effect of Nb additions disadvantage of significant solid solution hardening,
in this case is zero, since the improvement in acicular which one sees with Mn, Mo, etc.
ferrite content is offset by the formation of small The demand for high heat input, high deposition rate
segregated lath martensite regions. The aforesaid holds submerged-are welding procedures together with good
good as long as Nb is below 0.04%. Above this level, Nb low temperature weld-metal toughness properties in C-
has been found to markedly lower toughness even if Mn structural steels has led to the development of new
the acicular ferrite content is high. submerged-arc solid or flux-cored wire containing
While Nb inhibits the formation of proeutectoid controlled amounts of B, Ti and Mo (a patented solid
ferrite, vanadium has been found to inhibit upper bainite wire called Ti-Bor-22 contains 0.22 Mo, 0.0038 B and 0.03
formation. The effect of vanadium primarily depends Ti besides 1.50 Mn). B a n d Mo a d d i t i o n s give
on heat input and hardenability of weld-metal. In cases hardenability and hence a low austenite transformation
where high heat input is combined with low weld temperature, producing a strong and tough acicular
hardenability, the increase in V is not sufficient to ferrite, while Ti prevents contamination of the boron
achieve the hardenability necessary to form substantial by N 2 ; or 0 2 ;. The total effect is to increase the proportion
acicular ferrite; on the other hand the yield strength of acicular ferrite in the microstructure and refine the
increases with V additions, and the net result is a fall in acicular ferrite grain size, thereby leading to significant
cleavage resistance. If the initial weld-alloy content is increases in sub-zero Charpy-V impact values. Using
sufficient to develop a reasonable proportion of acicular Ti-Bor-22 wire with highly basic flux (Oerlikon OP-41TT
ferrite for the given welding procedure, the V additions of basicity = 3.0) to weld 50 mm t BS 4360 50 D plate by
increase the proportion of acicular ferrite by decreasing multipass technique at heat input of 7.9 kj/mm, Charpy-
the upper bainite fraction, and cause weld-metal V values of over 100 J at - 60°C have been reported for
toughness to improve. the weld-metal which had 84% acicular ferrite and no
Vanadium additions to Nb weld deposits have been ferrite side plates.
found to promote proeutectoid ferrite and lower yield This concept of Ti-B additions was recently extended
strength. This effect might be explained by the role of V to manual electrodes. The conventional 0.5-3.5% Ni type
electrodes which give good Charpy-V impact values densities being 3,700-4,000 k g / m 3 . Oxygen is
down to -50°C in flat position have been found to determined by vacuum fusion method.
possess inadequate COD property even at -10°C for Oxygen levels vary markedly between welding
welds deposited in vertical position, which normally processes. In MMA welding, rutile and acid-type
involves high heat input rates. On the other hand, a electrodes give deposit oxygen content of 0.04 to 0.10%,
typical Ti-B electrode without Ni and having the while basic types give 0.02-0.04%. Figure 9.26 shows a
composition given below has been reported to give, in clear inverse correlation between oxygen at 300-1000
addition to excellent Charpy-V values, COD values of ppm (parts per million) and Charpy-V impact energy
0.25 m m or more at -50°C even for the vertically at 0°C for MMA deposits. In submerged-arc welding,
deposited welds at heat input of 3.5 k j / m m . This the oxygen content range is 0.02-0.17% depending on
exceptional resistance to fracture by cleavage has been the type of flux and weld-metal dilution (see Fig. 9.27);
ascribed to the microstructure, which consists of fine for MIG welding 0.01-0.02% and for C0 2 -welding 0.04-
structures of proeutectoid ferrite and acicular ferrite, free 0.06%. One may now understand why for obtaining
of side-plate ferrite. highest toughness in gas-shielded welding, one is
This development has special relevance to site advised to use an argon-rich gas shield and vacuum
welding of LPG storage tanks, where considerable deoxidised wires a n d w h y wires h a v i n g high
amount of vertical welding is involved and adequate percentages of Mn and Si are used when welding under
toughness at -50°C has to be ensured. the highly oxidising C 0 2 shield.
Sulphur contributes directly to the inclusion volume
Resistance to Microvoid Coalescence fraction by forming sulphide inclusions. For highest
In C-Mn deposits, fracture by void nucleation,
growth and coalescence at non-metallic inclusions is OXYGEN, PPM
considered less serious with respect to catastrophic 200
500 MOO 1500
failures than fracture by cleavage. Yet if one is concerned
•
w i t h Charpy-V upper-shelf v a l u e s , one m u s t
understand the metallurgical factors controlling this
(J
•j
mechanism.
For good Charpy-V toughness; the weld-metal must
b
<
150
r ••
• .
have low non-metallic inclusion population, that is, low
volume fractions and a high mean free spacing of the
-1
• 'x
inclusions. These can be realistically determined only H 100 h
o
<
> . . •
by using replica techniques and the electron microscope. Q.
-|
To keep inclusion volume fractions low, oxygen and >
sulphur levels in the deposit must be kept minimum.
Since the solubility of oxygen in weld-metal is extremely <
x
50 Lh . -^
• \
o
low, its main effects are related to the volume fraction
and type of oxide reaction products trapped in the solid
weld by the rapid freezing over of the weld pool surface. . l i t l__J 1 _L_J 1 _j 1 1
The m a i n oxide inclusions are (Mn, Fe)Si0 3 , 0.02 0.04 0.06 0.08 0.10 0.15
OXYGEN, WT%
Si0 2 Al 2 0 3 ,Ti0 2 and (Fe Mn)0. Sulphides such as (Mn,
Fe)S, nitrides, carbides and carbonitrides are also Fig. 9.26: The effect of oxygen on Charpy-V energy in
present. MMA weld deposits at 0°C
c Si Mn Ti B YS UTS Eln.
N/mm2 N/mm2 %
0.07 0.20 1.65 0.04 0.004 510 570 30
The total oxygen level in a weld reflects the volume toughness, one must ensure that this impurity element
fraction V0 of oxide inclusions through the approximate is kept as low as possible in wires, fluxes and base
formula V0 = 5.5 (% O), based on oxide inclusion plates.
KEY
H-HIGH DILUTION WELD
L - L O W DILUTION WELD
ARABIC NUMERALS INDICATE NO. OF
WELD ANALYSES.
14
f I #(Mh fl
F15 Fl F2 F3 F4 F5 F6 F7 F8 F9 F10 F16 F17 F18 F13
HIGH Si02. MED S1O2. LOW S1O2
CALCIUM SILICATE MANGANESE * ALUMINIUM BASIC

SILICATE
Fig. 9.27: Ranges of O content from welds made with different fluxes
Nitrogen is another impurity element one must have N 2 levels as high as 0.080-0.20%, unless the wires
guard against. Moragaki et al. have studied the effect of have correct amounts of Al and Ti to play the role of
N2 in weld-metal on Charpy-V impact property at -30°C final deoxidation and nitride fixing. TIG and MIG welds
and their results are shown in Fig. 9.28. It is observed
that the impact value falls steeply when weld N, content i i —1 1 1 1 1
exceeds 100 ppm (0.010%). Ingress of N 2 into weld-metal ELECTRODE SHIELDING
SIZE OAS J
is the result of its ionisation in the arc. The degree of 0 4 AND 5
ionisation increases with arc temperature. • 4 AND 5 Ar J
20
Cellulosic and basic electrodes have the lowest ni-
trogen levels of MMA electrodes of around 0.004- . O A
0.014%, by virtue of the heavy shielding effect of the • O
gases from the decomposition of the cellulose and basic
carbonates respectively. By comparison, acid and rutile
•
8o
0
8V
coated electrode welds have higher typical mean fig- •
ures of 0.022 and 0.027% N 2 , which is attributable to the •* H
longer arc lengths and higher oxygen levels. Over-bak- ~i
ing the acid or rutile types increases nitrogen levels be-
10 OcQ
cause of reduced H 2 / H 2 0 shielding; basic types are
unaffected up to 450-500°C, because of the carbonate
- •*-
~^8 H
stability. Submerged-arc welds have nitrogen range of

NQ°
0.005-011%, most of it derived from wire and plate. The
\£
-j
C0 2 process is prone to nitrogen entrainment in the arc.
It is usually caused by some defect in the equipment or
faulty welding technique or by windy conditions. Po- 0 L 1 i i ......_i i i j 1 J
rosity occurs in C 0 2 welds when the N; content exceeds 100 200
0.020% and this happens when nitrogen in the shield- NITROQEN, PPM
ing gas is above 3% for short-circuiting arc and above
Fig. 9.28: Relation between Charpy-V notch energy
15% for spray-type arc. The so-called open arc weld absorption at -30° and nitrogen content of weld-metals
deposits given by self-shielding flux-cored wires can
have also been known to contain nitrogen in spite of the six categories of constituents as shown in Table 9.2
argon shielding, more so in draughty conditions. are looked for and quantified. For this purpose,
magnification of x 500 is adequate. Figure 9.29 gives
Wl Scheme examples of the categories. Table 9.2 also lists equivalent
Having realised the inadequacy of Widgery's descriptions used elsewhere. Each of the categories has
terminology on weld-metal microstructure referred to been given a code to help in point counting and in
earlier, the Welding Institute put forth in 1980 a revised annotating micrographs as is done in Fig. 9.29.
scheme whose objective is to quantify two main features This new technology differs from Widgery's in
of any weld microstructure, first the proportion of the the following respects:
various constituents present and second their effective 1) The term proeutectoid ferrite is replaced by more
dimensions. descriptive headings, which recognise that fer-
Initially, the prior austenite grain size is determined. rite can occur either in association with austenite
Its shape is usually revealed by the presence of ferrite grain boundaries (GF) or can be apparently in
which formed at the austenite boundaries during tragranular (PF)
transformation on weld cooling. A linear intercept
2) Replacing the term lamellar component by
technique is usually satisfactory for determining the
ferrite with aligned M-A-C which mean; ferrite
grain dimensions. Sometimes it is not possible to
containing martensite-austenite (M-A) or carbide
determine this, because the position of the grain
(C) phases
boundaries cannot be recognised easily. In the next step,
3) Identifying martensite as a separate constituent
from ferrite-carbide aggregates, such as pearlite.
The term bainite is not included, because confu-
sion surrounds its use and very special details are
required to identify the constituent.
In the third step, specifically those constituents re-
vealed by a picral etch are observed at high magnifica-
tion. They are M-A phases and/or carbides whose pro-
portion, type and distribution may be correlated with
mechanical properties. Their proportions which can of-
ten be up to 10% of the microstructure can be usefully
quantified by point counts. Further subdivision into
twinned or lath martensite is possible through trans-
mission electron microscopy.
In point counting, errors may occur due to, (a)
selection of inappropriate region or section, (b) careless
specimen preparation, (c) difficulty in classifying
constituents, or (d) statistical method used.
Heat-affected Zone
Heat-affected Zone (HAZ) is that portion of the base
metal lying next to the fusion line of weld, which had
not melted but whose mechanical properties or
microstructures have been altered by the heat of
welding. The HAZ is subjected to a complex thermal
cycle (sudden heating followed by rapid cooling), in
which all temperatures from the melting range of the
steel down to mere warming are involved, and therefore
consists of a series of graded structures ringing the weld
bead.
Fig. 9.29: As-deposited weld microstructures showing HAZ structure: A typical HAZ in a carbon steel can
the various microstructural constituents: be divided into four distinct regions according to the
Table 9.2 : Proposed scheme for the quantitative description of weld-metal microstructural constituents
using the light microscope
Orientational
(Please indicate) Transverse
Longitudinal Sketch (if required)
Plan
Other
1) Determination of prior austenite grain size—nital etch
Mean linear intercept = mm
2) Examination of constituent phases—nital etch
Preferred magnification = x 500 Magnification used = x
Microstructural constituent Code Previous descriptions Characteristic Proportion
dimension of o/
/o
constituent of
colony* um
Grain boundary ferrite, GF Proeutectoid ferrite,
ferrite veins, blocky
ferrite, polygonal ferrite
Polygonal ferrite PF Ferrite islands
Acicular ferrite AF Acicular ferrite
Ferrite with aligned M-A-C AC Ferrite side plates, upper

bainite, feathery bainite,
lamellar product
Ferrite-carbide aggregate FC Pearlite, ferrite+ interphase
carbide
Martensite M Martensite
3) Examination of microphases—picral etch
Preferred magnification range = x 1000 to x 2000 Magnification used
Microstructural constituent Code Characteristic dimension Proportion
of constituent, um %
Ferrite
Ferrite-carbide aggregate FC
M-A constituent (includes M-A

retained austenite, lath
martensite and twinned
martensite)
'e.g. Ferrite band width, acicular ferrite lath width, colony size offerrite with aligned M-A-C
* Quote to nearest whole number
maximum temperature reached. These regions along is high enough, can readily transform to brittle
with the pertinent iron-carbon phase diagram are shown martensite. This is the most potentially dangerous area
in Fig. 9.30. of the HAZ.
Region 1 which lies in contact with the fusion line Region 2 was heated slightly above A, into the fully-
was heated high into the austenite temperature range austenitic range. It is normally fine-grained, of
and is therefore extremely coarse-grained. It is moderate hardness and not likely to transform to
normally a high hardness zone and, if the cooling rate martensite.
as to avoid the formation of martensite in the HAZ.

Martensite is not acceptable, because it is brittle. In
addition, in the presence of hydrogen and restraint-
induced stresses it can cause delayed cold cracking in
the HAZ. Hydrogen is generated in the welding arc by
the dissociation of moisture present in the electrode
coating or submerged-arc flux, or by the combustion
of organic compounds which are sometimes present in
the electrode coatings of the rutile and cellulosic types.
Hydrogen is also generated from the oil and grease,
which may exist on the plate edges or on the bare
welding wire.
Hydrogen-induced cracking can be controlled by, (1)
using hydrogen-controlled electrodes and dry, clean
submerged-arc welding consumables, (2) by control of
joint design, and (3) by delayed cooling of the HAZ after
Fig. 9.30: Relation between the peak temperatures transformation to allow hydrogen to diffuse out. Here
experienced by various regions in HAZ, and how these it must be understood that removal of hydrogen above
correlate with the iron-corbon phase diagram the transformation temperature is not possible, because
the diffusion rate of hydrogen in austenite is too low.
Region 3 was heated between A2 and A3, i.e. partially The phenomenon of hydrogen- induced cracking and
into the austenite range and thus acquires a mixed methods of preventing its occurrence will be explained
structure. A small amount of martensite may form in it, at length in Chapter 10.
if the cooling rate is high enough. It is usually not very Sometimes one has to weld hardenable steels under
hard and yet may be brittle if martensite is present. restraint, where martensite formation and cold cracking
Region 4 was not heated high enough to form are likely to occur even after using best welding
martensite, but the wilding heat has served to temper procedures and hydrogen-controlled electrodes. In such
the metal to a soft condition. cases, the remedy lies in ensuring delayed cooling of
Factors controlling microstructure: The H A Z the HAZ by extending preheating into postheating
microstructure is determined by the composition of the without interruption. Postheat temperature of 200°C
steel and the welding thermal cycle. The higher the maintained for several hours (up to a maximum of ten
carbon content, the harder and more susceptible to hours) will drive off all hydrogen, while raising this
cracking will be the transformed HAZ microstructure. temperature to 600°C to stress-relieve the fabrication will
A higher content of alloying elements will increase the remove all hydrogen within an hour or so. The stress-
hardenability, enabling a hardened HAZ to be formed relief must be insisted u p o n where brittle (though
more readily; the particular alloying elements can also uncracked) martensite in the HAZ may endanger the
influence the susceptibility to cracking. The entire factor safety of the structure in service.
of carbon and the alloying elements can be expressed HAZ hardness: The HAZ hardness is usually a good
as carbon equivalent, which is given by the following indication of the presence of martensite and thus
formula: potential cracking. In a general way it can be stated that
Mn Cr + Mo + V Ni+Cu cracking rarely occurs when the HAZ hardness is 250
HV or less, but is common as the HV approaches 450.
The maximum hardness that can be tolerated in the
The cooling rate of the HAZ is dependent upon HAZ without hydrogen-induced, cracking under given
thickness of the plate being welded (an increase of which welding conditions is called critical hardness. For
increases the cooling rate), size of weld bead (process carbon-manganese structural steels, a critical hardness
energy input) and preheat temperature (increase in both of 350 HV has been suggested while welding with rutile
will decrease the cooling rate). electrodes, and 400 HV with basic well-dried hydrogen-
Hydrogen-induced cracking: Taking steel of a controlled electrodes.
particular composition and thickness, heat input rate Figure 9.31 shows three welds in a 25 mm thick 0.25%
coupled with preheat, if necessary, should be so chosen C steel plate that underwent three different thermal
NO PREHEAT PREHEATED POSTHEATED
Fig. 9.31: Weld zone hardness. Weld bead on 25 mm 0.25% carbon steel. Knoop indenter, 500 gram load,
Nital etch. (100 x)
regimes. The first weld was made without preheat and

using low heat-input rate, thus hard martensite formed
in the HAZ. The second weld was made with sufficient IFFERENTIAL
preheat to lower the cooling rate a n d eliminate TRANSFORMER
martensite. The third plate was welded with preheat COILS
and was also postheated at 600°C to duplicate the stress-
relief operation. The hardness values for the three zones MOVING
of each w e l d m e n t speak for the favourable and CORE
unfavourable microstructures formed in each case.
AS COOLING
Dilatometric study: It is possible to simulate the weld ANNULUS
HAZ microstructures of a steel in the laboratory by
dilatometric investigations. In a typical apparatus de- AS INLET
signed by the British Welding Institute (Figs. 9.32 and
9.33), small hollow specimens of the steel are heated by
high frequency induction to 1,325°C and quenched im-
mediately by a blast of argon gas. By pre-selection of
the stored volume and pressure of gas for quenching
and by passing it through standard orifices, it is pos-
sible to subject them to six basic thermal cycles, repre-
INDUCTION
sentative of those which can occur in the HAZ of welds
HEATING
over a wide range of welding conditions. COIL
Thermocouples are attached to the inside wall of the
specimen at m i d - l e n g t h , diametrically opposite TEST SPECIMEN
positions.The thermocouple output is used as one
function of an x-y recorder. SILICA
SEATIN
Changes in length of the specimen are transmitted TUBE
by a push-rod to the moving iron core of a differential
transformer. The output of the transformer is used as Fig. 9.32: Section of BWI's dilatometer
DILATOMETER
PUSH-ROD FROM SPECIMEN
PRESSURE GAUGE
DILATION CONTROLS
PEN MOVEMENT
,$3C00LING
^HEATING
^THERMOCOUPLE ^ THERMOCOUPLE
WIRES GAS CYLINDERCOLD JUNCTION H.F. FILTER |
TEMPERATURE CONTROLS CHART MOVEMENT
SECOND PEN MOVES AT CONSTANT RATE AND
DRAWS COOLING CURVE
Fig 9.33: Diagram showing relationships of ancillary equipment to dilatometer
the second function of the x-y recorder. A second pen Hardness measurements are made on each of the
on the recorder is moved at constant speed across the individual dilatometer specimens. These hardness
chart on the same axis as the length changes. Thus two values are then recorded on the CCT diagram.
lines are drawn on the same record; one records length Relationship between hardness and cooling rate at 300°C
changes against temperature, the other temperature is also represented graphically, since cooling rate at
changes against time. Thus it is possible from each 300°C is used as the most typical single parameter of
record to determine the temperature and time of start the cooling curve.
and finish of transformation. This information is used Research workers of the British Welding Institute
to construct the CCT diagram. A typical CCT diagram have produced CCT data for a range of C-Mn steels
for a 38 mm thick plate to BS: 968-1962 is shown in Fig. having carbon contents between 0.15 and 0.27%. From
9.34. the results, they have produced diagrams showing the
("GRAIN SIZETS5 I
SAMPLE FROM 1/2 IN PLATE TO B.S. 968:1962
j RECEIVED (ASTM) | CAST NO 87326
C iMn Si s P Ni 1 Cr 1 M o l rsib
I
1
I'o.i 6[1.32 B . Q 2 3.046 0.019 — |o.031 - 0. 024 1
s-
I 1 1
1LB e n l
K
800
^ N
^ X
\ Ss H4
S\ Sv
N
\ >
fid ^ \
600 1 j
\
\\Y\
400 Jr\ L^^^^
\[ \
i
[S
200
N
i \\
\I N \ N
0 U It\Y 420^414 3: 5 2 36 22 4.214
2 4 68 j 2 4 6 8 10 2 4 68 102 2 4 6 8 1Q 3
TIME TO COOL FROM 1000°C, SEC
Fig. 9.34: CCT diagram for steel to BS: 968-1982 W mm f

critical welding conditions necessary to produce an important to distinguish this phenomenon from cold
unhardened HAZ structure (i.e. microstructure with a cracking, because if it is mistaken for cold cracking and
hardness less than 350 HV) and hence to give complete preheat is used as a remedial measure, it might
freedom from hydrogen-induced HAZ cracking. They aggravate the overall problem rather than eliminate it.
have finally been able to recommend specifications for Sulphur is considered the most offensive element in
ladle analysis to be so fixed that plate compositions the hot cracking problems. Carbon and phosphorus lend
would never exceed a carbon equivalent of, say, 0.45 support to sulphur in causing hot cracking. As shown
(as given by the CE formula mentioned earlier) if the in Fig. 9.35, if the sulphur content of a high-strength
material were to be capable of welding without preheat alloy steel is near the permissible maximum of 0.04%, it
at combined joint thicknesses up to about 75 mm. is likely to develop hot cracks in the HAZ when welded
In recent years, the Welding Institute have preferred under conditions of high restraint. Hot cracking tests
to study real HAZs w h e r e possible, rather t h a n carried out on AISI4340 steel show that hot cracks under
simulated ones. This is for a number of reasons, but high restraint can be reduced to zero only when the
principally because of the difficulty of matching both combined S and P contents are below 0.02%.
thermal history (particularly peak temperature and time Carbon increases cracking tendency appreciably.
at high temperature) and austenite grain size. They Carbon has been shown to promote interdendritic
believe that it is important to reproduce both these segregation of sulphur, which increases the likelihood
factors, and not only the austenite grain size; since of concentrating lower-melting compounds in grain
particularly in microalloyed steels, reproduction of time boundary areas. Decreasing the carbon content, even
at temperature in the high temperature HAZ may also by decarburisation, reduces the hot cracking tendency.
be important with respect to the dissolution of various An annealing treatment which produces spheroidised
carbides and carbonitrides. According to WI, a carbide particles also t e n d s to reduce cracking
convenient method is to use un-notched implant propensity, apparently because the short austenitising
specimens (see page 718) of the steel under test, or in time in welding does not dissolve all the globular
some circumstances, to make a series of bead-on-plate particles. Therefore, the actual carbon content in the
welds. matrix is lower than the overall carbon analysis for the
Significance of Ms and Mf. Ms and Mf points (see steel.
Figs. 9.11 and 9.12) indicated on the isothermal
transformation and continuous cooling transformation
of a steel have an important bearing on HAZ cold
cracking. Mf is the temperature at which transformation
of austenite to m a r t e n s i t e starts, a n d Mf is the
temperature at which it is completed. These points arc SEVERE CRACKING
important in transformations occurring in the HAZ, u Q
because hydrogen is far more soluble in austenite than
in martensite. If, therefore, the Mf is sufficiently high O <
(above 260°C as is indicated by experience in welding X
<
alloy steels), the chances of cold cracking are reduced, 8
because of t w o p h e n o m e n a w h i c h occur
simultaneously: because of the lower solubility of 2
hydrogen in martensite the dissolved hydrogen in
austenite diffuses out during transformation, and at the
same time the higher Mf results in some tempering of
the martensite, which reduces the hardness and lowers NO CRACKING
residual stresses. However, many of the commercial 0.0 0.02 0.04 0.06
steels display Mf temperatures appreciably below SULPHUR. PERCENT
260°C, and therefore d e m a n d special w e l d i n g
procedures. Fig. 9.35: Typical effect of sulphur on susceptibility of a
Hot cracking in the HAZ: Weldable alloy steels having high-strength alloy steel (e.g., AISI 4340) to hot cracking
yield strengths of 56 k g / m m 2 and higher, sometimes in base metal heat-affected zone of highly restrained weld
develop hot cracking in the HAZ after welding. It is test specimen
Manganese has been found to counteract the effect certain temperature ranges. These steels usually are
of sulphur with respect to hot cracking tendency. This made with very low sulphur content to avoid undue
benefit is believed to arise from the strong tendency of segregation, which might be promoted by the high
m a n g a n e s e to form stable, higher m e l t i n g - p o i n t silicon content.
sulphides in steel. Rather than attempt to use large C h r o m i u m , m o l y b d e n u m a n d v a n a d i u m are
manganese additions as a cure-all, however, good steel believed to have a mild beneficial effect upon hot
makers attempt to reduce hot cracking tendency cracking, particularly if a high carbon content is present
through reduction of sulphur content. to aggravate the problem. These elements possibly tie
Phosphorus already has been cited as a strong up some carbon as complex carbides, thus reducing
promoter of hot cracking. It appears to act in this manner cracking susceptibility by lowering the effective carbon.
by increasing the segregation of sulphur and carbon in Nickel exerts an adverse effect u p o n hot cracking
the steel—to the point where banding of the sulphide tendency. Its effect is particularly noticeable when
inclusions and carbide particles can be observed in the significant sulphur also is present. Presumably, the
microstructure of w r o u g h t material. This b a n d e d presence of nickel affects the composition of sulphides
condition greatly increases the likelihood of hot formed in the grain boundaries. Aluminium is another
cracking, and if the banding of carbon is particularly element that seems to influence the composition and
marked, these localised areas of high carbon may be distribution of sulphide inclusions. Residual aluminium
particularly prone to cold cracking as a continuation of in the range of about 0.02/0.06% appears to foster
the hot cracks. intergranular sulphides, which noticeably increases
Silicon is believed to cause sulphur to segregate, and cracking susceptibility.
in this m a n n e r is likely to increase hot cracking. Here we have dealt with the HAZs of only carbon
Ordinarily, silicon does not have a marked effect on steels and low-alloy steels. It is equally important to
hot cracking susceptibility, because less than 0.50% know how similar microstructural and related changes
usually is present in high-strength alloy steels and the take place in the HAZs of other industrial metals and
content varies over a relatively narrow range. However, alloys as a result of welding, how far they have a
some of the very high strength (heat-treatable) alloy deleterious effect on joint properties and serviceability,
steels make use of silicon contents of about 1%, because and how these ill-effects can be controlled. These aspects
this element inhibits softening during tempering in will be dealt with in the next chapter.
Introductory Welding Metallurgy, published 1968, by American Welding Society.

Physical Metallurgy and the Design of Steels by Pickering, published 1976, by Applied Science Publishers, London.
Welding Handbook, Chapter 4, Vol.1, 7th Ed., published by American Welding Society.
Welding Metallurgy, Vols. 1 & 2, by Linnert, published by American Welding Society.
Weldability of Steels by Stout & Doty, published 1978, by Welding Research Council, New York.
Metallurgy for Engineers by E.C. Rollason, published 1973, by The English Language Book Society & Edward (Publishers)
Ltd.
Deoxidation Practice for M.S. Weld Metal by Widgery. Weld. J., Mar. 1976.
A Technique for the Chracterisation of Weld Metal Microstructure by Widgery & Davey. IIW Doc.II A- 389-1976.
Factors Controlling Weld Toughness—The Present Position Part 2 by Dolby, WIRR 14/ 1976.
A Scheme for the Quantitative Description ofFerritic Weld Metal Microstructures by Abson & Dolby. WIRB, Apr. 1980.
Weldability
of
Metals
THE TERM WELDABILITY is not easy to define, and special precautions, such as increasing the preheating
different authorities have interpreted it in different ways temperature and use of consumables of lower potential
according to their own viewpoints. Yet the following defi- hydrogen, to prevent embrittlement and possible cracking
nition worked out by the International Institute of Weld- in the HAZ.
ing and given in ISO document R 185/1967 has been The study of weldability is usually confined to the
widely accepted as standard: heat-affected zones and its primary objective is to help
A metallic substance is considered to be weldable to a achieve good properties and characteristics of the weld-
stated degree by a given process and for a given purpose, metal as well as of the HAZ. However, it must be
when metallic continuity can be obtained by welding appreciated that the suitability of a welded structure for
using suitable procedure, so that the joints comply with the intended service depends equally on the design of
the requirements specified in regard to both their local the structure as a whole, and the properties and
properties and their influence on the construction of which characteristics of the base metal away from the welds.
they form a part. The American W e l d i n g Society has defined
The definition underlines the fact that weldability is weldability in a simpler way as follows:
not an intrinsic property of a metal like density or Weldability is the capacity of a metal or combination
hardness, but is determined by several extraneous factors. of metals to be welded under the fabrication conditions
Also a metal can have several degrees or levels of into a specific and suitably designed structure to meet
weldability depending on the process, procedure and satisfactorily in the intended service.
service conditions.
For example, aluminium is not easily weldable if Carbon Steels
manual metal-arc welding is used, but is readily weldable Carbon steels represent a broad range of steels with
by the GTAW process or resistance welding process. Cast u p to 1.70% m a x i m u m carbon, 1.65% m a x i m u m
iron is considered weldable if the given purpose is to manganese and 0.60% maximum silicon. They do not
repair a crack, but is unweldable if the purpose is to contain alloying elements like Cr, Ni and Mo. They are
fabricate a pressure vessel out of it. Mild steel is normally sub-divided into four categories as:
considered to have excellent weldability, but is not
1) Low-carbon steels up to 0.15% C
suitable for sub-zero applications in the welded form,
2) Mild-carbon steels 0.15-0.29% C
because of its low notch toughness at low temperatures.
3) Medium-carbon steels 0.30-0.59% C
Referring to carbon steels and low-alloy steels, it will
4) High-carbon steels more than 0.60% C
be shown in the following p a r a g r a p h s that their
weldability is dependent on several factors including 1) Low-carbon steels: These steels are available as
hardenability, joint thickness and hydrogen levels rimmed, semi-killed or killed grade (killed with Al or Si).
generated in the welding process. As the degree of Mn ranges between 0.30 and 0.60%. They have low yield
weldability of the alloy decreases, its welding demands strength and are mainly used in sheet or coil form in cold
rolled or hot rolled condition. Commonly they are welded c) Deposit the first few passes at low currents to
by electric resistance process, though manual arc welding minimise base metal dilution
can be used for thicker sections. d) Deposit further passes with increased heat input
2) Mild-carbon steels: This group is the most common
e) Maintain minimum interpass temperature of
material in industrial use. In as-rolled condition they
250°C
give yield strength in the range of 25CM50 N / m m 2 . In
heat-treated condition, they give yield strengths up to f) Without allowing the base metal to cool below
690 N / m m 2 along with excellent notch toughness. They 235°C, give it a postweld heat treatment at 550°C
are commonly used in the form of plates, have excellent Then cool slowly under an insulation blanket.
weldability and are welded with the conventional 4) High-carbon steels: These are hard materials referred
welding processes. to as tool steels. They are used in dies, mill rolls,
Some grades of mild-carbon steels are heat-treated to railway wagon wheels, rails, etc. Welding of these
improve mechanical properties. They are known as heat- steels is avoided as far as possible, since it is diffi-
treated carbon steels. The heat treatment is either cult to guarantee crack-free welds.
normalising or quenching followed by tempering. When
heavy sections of coarse-grained steels are normalised, Low-Alloy Steels
the microstructure becomes uniformly finer (example: These steels can be divided into four groups according
ASTM A515 grade steel). Fine-grained steels in thin as to strength, heat-treated condition or service application.
well as thick sections are normalised in order to improve • Group 1: These steels have yield strength range of
their notch toughness, so that they meet the impact test 310-480 N / m m 2 They are supplied in the as-rolled or
requirements at sub-zero temperatures. Very heavy normalised condition and can be welded as simply as
sections of fine-grained steels are quenched and tempered, mild-carbon steels.
so that they acquire a microstructure whose toughness These steels are referred to as high-strength structural
matches that of the thinner normalised sections of those steels. Carbon is kept below 0.25%, while small additions
steels (example: ASTM A516 grade steel over 100 mm of alloying elements such as Cr, Cu, Mo, Ni, P, V or Zr are
thick). made to produce a finer microstructure, which results in
Normalised grades can be welded as easily as mild higher strength and increased toughness of the steel.
steel without any special precautions. In the case of Further increase in yield strength is achieved by adding
quenched and tempered steels, heat input control is 0.01 to 0.05% Cb or V. In the case of higher strength steels,
necessary when the submerged-arc process is used. For whose welded fabrications have to be stress-relieved in
example, in the case of A516 type steel, it should be below practice, the stress relief treatment results in increased
3.2 kj/mm. The important consideration in welding all strength due to precipitation of copper or alloy carbides.
these heat-treated steels is to select welding consumables Addition of nickel improves sub-zero toughness to a
and procedure that will give joints of matching strength remarkable degree.
and sub-zero impact values. For manual welding, low- Well-known ASTM grades suitable for structural
hydrogen mild-steel electrodes (AWS E7015, -16 or -18) fabrication are A242, A441, A572 and A588 and those
are suitable. For submerged-arc welding, high manganese for pressure vessel fabrication are A203, A204, A225,
(2% Mn) or manganese- molybdenum wire with neutral A353 and A537. Nickel steels are represented by ASTM
or basic flux should be preferred. A203 Gr. D (with 3.5% Ni for service at -100°C) and A353
3) Medium-carbon steels: These are high-strength (with 9% Ni for service at -196°C).
materials with comparatively high hardness. They are This group also includes a class of structural steels in
used for machinery parts and machine tools. They can which additions of Cu, Cr and sometimes P are made to
be successfully welded by using proper preheat and increase strength as well as resistance to atmospheric
postheat p r o c e d u r e s a n d l o w - h y d r o g e n w e l d i n g corrosion. It is represented by ASTM Gr A374 and A375.
consumables. A well-known example is COR-TEN which will be dealt
Typical welding procedure for 0.58% carbon steel with in detail later.
would be as follows: A basic consideration in welding these steels is the
a) Preheat to around 300°C prevention of hydrogen-induced cracking or cold
b) Use low-hydrogen consumable in perfectly dry cracking in the heat-affected zone of the welded joint.
condition (manual electrodes should be baked at The procedure for avoiding this defect is explained in
400°C for 1 hour) detail later. Generally speaking, stress-relief does not alter
WELDABILITY OF METALS 429
the strength level of the weld significantly, but improves b) 4335 V with approximately 0.35 C, 0.85 Mn, 0.25 Si,
its ductility and notch toughness. 1.8 Ni, 0.80 Cr, 0.35 Mo, 0.20 V (%).
• Group 2: It includes steels with yield strength range Welding is carried out with perfectly dry high-tensile
of 350-1240 N/mm, which are supplied in quenched steel low-hydrogen electrodes (for example AWS E10015)
and tempered (Q&T) condition. Some of them have and with preheat. The preheat temperature depends on
excellent toughness at sub-zero temperatures. Postweld various factors as pointed out later in the discussion on
treatment including stress-relief is avoided in the case of hydrogen-induced cracking. The preheat is maintained
these steels as far as possible, because it may reduce the at a temperature equal to or above the preheat temperature
toughness of the base metal as well as the weld-metal. for a long time after the welding is completed to allow the
harmful hydrogen to escape completely. The steel is then
Popular ASTM grades of this group are: cooled down if no stress-relief is called for, or placed
a) A537 Gr. B is used in pressure vessel fabrication. directly in a furnace for stress-relief.
b) A514/517 is used widely for fabricating earth-
For example, in one typical case, a 4335 V steel is
moving equipment, bridges, buildings, steel mill
preheated to 150°C and maintained at this temperature
equipment, pressure vessels and penstocks. The well-
until welding is completed. The temperature is then
known commercial steel of this category is T-l marketed
by United States Steel Corp. of the USA. raised to 315°C and maintained until the weldment is
placed in a furnace at 650° C for stress-relief. In a typical
c) A543 is intended for nuclear reactor vessels. A543
procedure with 4340 steel, preheat of 315°C is used and
Gr. B class 1 and 2 steels are better known as HY-80 and
is maintained for one hour after welding is completed,
HY-100 steels and are used for military and marine
and then the weldment is allowed to cool to room
construction and hydrospace vehicles.
temperature. Weld-metal cracking in the 4340 steel
d) A553 is 9% Ni steel, a Q&T version of A353 steel,
welded joint is minimised by using a filler metal whose
which is used at temperatures as low as -196°C (liquid
carbon and alloy content is as low as possible and the
nitrogen).
combined S and P content is below 0.025%.
e) HY-130 with yield strength range of 900-1030 N /
The procedure for quenching and tempering consists
mm2 and excellent toughness is used in hydrospace and
of heating the steel above A3 point to its austenitising
aerospace applications.
temperature, holding at that temperature for sufficient
Weldability of each of these steels will be discussed
length of time and quenching in a suitable medium such
in detail later.
as water, brine or oil. Then tempering is done at a suitable
• Group 3: This group includes steels which are
temperature in the range of 260-704°C, depending on
quenched and tempered after welding to achieve an
the strength or hardness desired. It must be noted that as
optimum combination of strength and hardness. They
the tempering temperature increases, strength and
are used in lighter sections and on applications where
hardness decrease while ductility and toughness
toughness is not important, but high hardness or high
increase. For example in the case of AISI 4340 steel, the
strength-to-weight ratio is required. Hence the carbon
austenitising temperature range is 800-843°C, the
content is kept high (0.30-0.45%) and suitable alloy
additions are made to ensure deep and complete quenching medium is oil, and the variation of strength
hardening of the section. with tempering temperature is as follows:
Weldability of these steels is not good because of their Tempering Room temp.
high hardenability, which encourages the formation of temp.,°C tensile strength,
brittle martensite in the HAZ and cracks in this region N/mm 2
due to contraction stresses. Weldability is further reduced 454 1,380-1,520
by sulphur and phosphorus. Sulphur increases the hot 495 1,240-1,380
cracking susceptibility (particularly in the weld-metal) 566 1,110-1,240
while phosphorus reduces ductility and toughness and 635 960-1,110
increases hot cracking susceptibility in the base metal as
• Group 4: This group consists of chromium-
well as the weld-metal. Hence the levels of these elements
molybdenum steels used mainly in power plants, fertiliser
have to be kept below 0.02% each.
plants and oil refineries at service temperatures up to
Typical AISI grades of this group are: 704°C. They are distinguished from other low-alloy steels
a) 4340 whose approximate composition is 0.40 C, by their remarkable oxidation resistance, resistance to
0.80 Mn, 0.28 Si, 1.8 Ni, 0.80 Cr, 0.25 Mo (%) sulphide corrosion and high-temperature strength or
creep resistance. These properties are derived from Cr the hardened HAZ of a welded joint. The postweld heat
and Mo and they improve with the increase of these treatment of welded joints in Cr-Mo steels consists of
alloying elements. heating them to a temperature below the lower critical
There are nine standard types as shown in Table 10.1 temperature (approximately 740°C), holding it for a
along with their chemical compositions. It is seen that proper length of time and cooling in still air. It not only
the Mo content is either around 0.50 or 1.0% according to tempers the HAZ martensite but also relieves the locked-
the Cr content of the steel. The 2.25 Cr-1% Mo steel has up stresses caused by welding. Air cooling does not result
the best high-temperature strength. They are usually in appreciable hardening, because the cooling takes place
supplied in the annealed or normalised and tempered from below the critical range. However, rapid cooling
condition. caused by cold draughts, if present, can introduce new
These Cr-Mo steels are air-hardening steels and stresses.
undergo microstructural transformations that depend on These steels are readily weldable with the conventional
the rate of cooling from above the u p p e r critical arc welding and electroslag welding processes. The air-
temperatures. The mechanical properties are dependent hardening property of the base metal and high alloy
on these transformations. With faster cooling rate, content of the weld-metal demand that correct welding
strength and hardness increase and ductility is reduced. procedures including preheat, postweld heat treatment,
However, because of the lower carbon content (0.15% low-hydrogen consumables, and filler metal chemistry
maximum), carbides are present in limited amounts and are used to prevent HAZ and weld-metal cracking.
this results in higher ductility at any given strength level Recommended preheating temperatures are indicated in
than that obtained from high-carbon steels. Table 10.2. It is obvious that the preheat temperature must
When these steels get h e a t e d above their be increased as the thickness and carbon and alloy
transformation temperatures in operations like rolling, contents increase to ensure crack-free joints. The welding
forging, hot forming and welding with consequent loss process can also determine the preheat condition. With a
in ductility and toughness due to their air-hardening low-hydrogen process such as TIG, or a high heat input
characteristics, they require further heat treatment to process such as electroslag, much lower preheat would
restore these useful mechanical properties. The alternate be justified or even none at all. The chemical composition
types of heat treatment applied are: of the filler metal must be the same as of the base metal,
a) Annealing except for carbon content which is lower for the weld-
b) Normalising and tempering metal. When different Cr-Mo steels are to be welded at
c) Quenching and tempering the same time, filler metal selection can be restricted by
d) Simply tempering. using the same filler metal for a variety of steels. For
(a), (b) or (c) restore proper grain size and confer example, 1.25 Cr-0.5% Mo filler metal can be used for
mechanical properties as desired, (a) and (b) provide three base metals: 0.5% Cr-0.5% Mo, 1% Cr-0.5% Mo and
enough strength to meet the design requirements of ASME 1.25% Cr-0.5% Mo. Similarly 2.25% Cr-1% Mo filler metal
Boiler and Pressure Vessel Code, (d) is employed to soften can be used for 2% Cr-0.5% Mo and 3% Cr-1% Mo steels.
Table 10.1:: Nominal chemical composition of chromium-molybdenum steels

Composition percentage
Type C* Mn S* P* Si* Cr Mo
0.5Cr-l/2Mo 0.10-0.20 0.30-0.60 0.045 0.045 0.10-0.30 0.50-0.80 0.45-0.65
lCr-l/2Mo 0.15 0.30-0.60 0.045 0.045 0.50 0.80-1.25 0.45-0.65
1.25Cr-l/2Mo 0.15 0.30-0.60 0.030 0.045 0.50-1.00 1.00-1.50 0.45-0.65
2Cr-l/2Mo 0.15 0.30-0.60 0.030 0.030 0.50 1.65-2.35 0.45-0.65
2.25Cr-lMo 0.15 0.30-0.60 0.030 0.030 0.50 1.90-2.60 0.87-1.13
3Cr-lMo 0.15 0.30-0.60 0.030 0.030 0.50 2.65-3.35 0.80-1.06
5Cr-l/2Mo 0.15 0.30-0.60 0.030 0.030 0.50 4.00-6.00 0.45-0.65
5Cr-l/2MoSi 0.15 0.30-0.60 0.030 0.030 1.00-2.00 4.00-6.00 0.45-0.65
5Cr-l/2MoTi 0.15 0.30-0.60 0.030 0.030 0.50 4.00-6.00 0.45-0.65
7Cr-l/2Mo 0.15 0.30-0.60 0.030 0.030 0.50-1.00 6.00-8.00 0.45-0.65
9Cr-lMo 0.15 0.30-0.60 0.030 0.030 0.25-1.00 8.00-10.00 0.90-1.10
* Max unless ranges are shown
Table 10.2: Recommended preheat temperatures Table 10.3: Stress-relief temperatures for Cr-Mo steels
for Cr-Mo steels
Alloy type Temperature range, °C
Thickness 12.7 mm 12.7 to 57 mm Above 57 mm
0.5Cr-l/2Mo 590-700
Cr-Mo steel Preheat temp.°C
lCr-l/2Mo 590-730
0.5Cr-l/2Mo 20 95 150 1.25Cr-l/2Mo 590-745
lCr-l/2Mo 120 150 150 2.25Cr-lMo 680-760
1.25Cr-l/2Mo 120 150 150 3Cr-lMo 680-760
2Cr-l/2Mo 150 150 150 5Cr-l/2Mo 680-760
2.25Cr-lMo 150 150 150 5Cr-l/2MoSi 680-760
3Cr-lMo 150 150 150 5Cr-l/2MoTi 680-760
5Cr-l/2Mo 150 150 150 7Cr-l/2Mo 680-760
5Cr-l/2MoSi 150 150 150 9Cr-lMo 680-760
5Cr-l/2MoTi 150 150 150
7Cr-l/2Mo 200 200 200
9Cr-lMo 200 200 200 Cr-Mo steels in lower thicknesses can be put into
service without postheat treatment, provided proper
Minor repairs on Cr-Mo steels can be carried out with preheating procedure has been applied during welding.
25 Cr-20 Ni or 25 Cr-12 Ni type austenitic stainless steel For example, 2.25% Cr-1% Mo steel welded with extra-
filler metals, when postweld heat treatment is not low carbon filler metal (C-0.03%) has enough as-welded
possible. The weld-metal has lower yield strength ductility to meet code requirements. Postheating may
compared to the matching Cr-Mo composition, but is therefore be avoidable in some cases.
extremely ductile and can resist cracking due to shrinkage Cr-Mo steel vessels are sometimes wholly annealed
stresses. Though hardness in the HAZ cannot be to obtain a softer ferritic structure in the entire base metal,
avoided, the austenitic weld-metal prevents occurrence HAZ and weld-metal. Annealing procedure is as follows:
of cold cracking in this region. The limitations of the the weldment is heated in the critical temperature range
austenitic weld-metal are: of 843-913°C, held at that temperature for one hour per
a) It may cause failure at the fusion line w h e n 25 mm thickness, cooled to 538°C at a maximum rate of
subjected to cyclic temperatures in service 28°C per hour, then furnace or air cooled to room
b) Carbon can migrate from the lower Cr base metal to temperature.
higher Cr weld-metal at high service temperatures, As far as possible, Cr-Mo steels must be preheated,
resulting in the weakening of the weldment w e l d e d a n d p o s t w e l d h e a t e d in one continuous
c) Brittle sigma-phase may develop at high service operation. Experience, however, has shown that steels
temperature with 2.25 Cr and less can be safely cooled to room
d) Difference in coefficient of thermal expansion temperature before being postweld heat treated.
between the weld-metal and base metal may give rise to For steels having higher Cr contents, an interruption
stresses at high or under cyclic temperatures. between preheat and postheat is likely to produce
Welding with austenitic weld-metal is more prevalent cracking in the weldment. If interruption is inevitable for
in the petroleum industry, where more-or-less steady some reasons, a practical step to prevent cracking consists
temperatures exist than in thermal power stations, where of raising the weldment temperature from the preheat
demand fluctuation causes cycling of temperature and temperature to above 427°C and maintaining there for
pressure. three minutes per 25 mm thickness before cooling to room
temperature. This results in removal of most of the
POSTWELD HEATTREATMENT hydrogen from the weld and the HAZ. The postweld
Cr-Mo weldments, especially in pressure pipings, heat treatment is finally carried out at any convenient
have to be stress-relieved as required by relevant codes. time.
Welded joints in pipings are given local stress-relief while
vessels and similar fabrications are wholly stress- HYDROGEN-INDUCED CRACKING
relieved. Table 10.3 gives the stress-relief temperatures The most serious problem of weldability in carbon
which are below the critical range and result in reduced and low-alloy steels is hydrogen-induced cracking in
hardness and residual stress level and increased ductility the heat-affected zone of the base metal. This type of
of the weld-metal and the HAZ. cracking often occurs after the welding has been
completed and the weldment has cooled down. It lies when it comes to plate thickness, the main criterion is
underneath the weld bead and is not easily detected. combined thickness (i.e. total thickness of the plates
Many service failures due to fatigue and brittle fracture
have been traced to pre-existing cracks in the HAZ. Such
cracks have led to the fracture of pressure vessels during
hydraulic testing and collapse of bridges and storage
tanks in service.
Hydrogen-induced cracking (HIC) is also known as:
a) Cold cracking, because it occurs when the metal
has cooled below 150°C
b) Delayed cracking, because its formation is
sometimes delayed for 20 hours or more
c) Underbead cracking, because in most cases it is
located at the HAZ underneath the weld bead.
HIC occurs in the HAZ when:
a) Sufficient hydrogen is present in the arc. Hydrogen
is derived from organic materials and minerals having
chemically combined water used in the covering of
manual electrodes. The main source, however, is moisture
which is inevitably present in all types of fluxes. Oil,
grease paint and moisture on the base metal surface or
electrode-wire surface can also generate hydrogen during
welding.
b) Shrinkage stresses, supplemented by other stresses
developed as a result of joint restraint, acting on the weld.
c) A sufficiently hard and brittle microstructure
develops in the HAZ as a result of its being heated to Fig. 10.1: Hydrogen-induced cracks in HAZs of (A) fillet
above A3 temperature and cooled rapidly. weld and (B) butt weld
The usual sites of HAZ cracking in butt and fillet
welds are shown in Fig. 10.1.
Cracks can also occur in welds if their alloy content is
high and the conditions leading to HAZ cracking as
enumerated above are present. They appear to be
transgranular under the microscope as shown in Fig.
10.2(a). At (b) is shown a root bead which has cracked
through under severe restraint.
HIC depends on the following factors:
1) Carbon equivalent of the steel
CE is given by the formula
CE ='C% + Mn% + Cr% + Mo% + V% + Cu% + Ni%
6 5 15
Higher the CE, Harder will be the HAZ with increased
risk of cracking. In normal structural steels, we are
concerned only with C and Mn. Therefore weldability
will be controlled not by C content alone, but by the joint
action ofC and Mn through the expression
C + M n
6
2) Thickness of sections Fig. 10.2: Weld-metal hydrogen cracks in (a) single-run
Thicker the plate, faster is the cooling rate of the HAZ MMA fillet weld on low-alloy steel x150, and
(b) root bead of a Y-groove welding test x 10
with increased risk of cracking. Accurately speaking,
Higher the heat input rate a n d / o r higher the preheat

—f
t1 = 25
¥3 ^
and interpass temperature, lower will be the HAZ cooling
and lesser will be the chances of cracking, and vice versa.
Heat input rate can vary widely (0.6 to 8 kj/mm), con-
sidering at one extreme a root pass with a small diameter
t1+t2=50MM
electrode and intermittent arc in a pipe of large wall
thickness, and at the other extreme a multi-wire sub-
merged-arc weld using a combined value of over 1,500
II
CO amps and a low speed of travel. Higher current and lower
speed of travel will obviously necessitate lower preheat
temperature or no preheat at all for a given set of
T conditions.
t1 = 17* t2 = 17
—I v Values of heat input rate under a few typical MMA
tl -h t2H-13= 51. MM welding conditions are given in Table 10.4.
4) Hydrogen rating of the welding consumable
H y d r o g e n rating means the level of hydrogen
k t i -»
potentially available from a consumable after any
appropriate drying treatment, which will be absorbed by
t the weld pool during welding. The four standard
t4 t2 hydrogen ratings are as follows:
-JlL
H Rating Weld hydrogen level Critical HAZ
ml/100 g of deposit hardnes HV
k- t3
1) High Above 15 390
2) Medium Above 10 and below 15 375
t1 8 t 3 = 12 MM 3) Low Above 5 and below 10 400
t2 8 t 4 = 13 M M 4) Very low Below 5 450
t 1 + t 2 + t3 + t 4 = 5 0 M M For each hydrogen rating, the critical HAZ hardness
is indicated. To ensure freedom from HIC, this value of
Fig. 10.3: Combined thickness in different joints
hardness of the HAZ must not be exceeded. It is obvious
that with decreased weld hydrogen level, higher HAZ
meeting at the joint line) and therefore we have to hardness can be tolerated and generally lesser are the
distinguish between different joint configurations. For chances of cracking.
example, a butt weld between two 25 mm thick plates is The approximate conditions under which welding
equivalent to a fillet weld between two 17 mm thick plates consumables give these ratings are indicated below:
or a cruciform joint among 12 and 13 mm thick plates, as Rating Examples
far as the rate of cooling they will impose on the weldment 1 Manual electrodes of rurile and cellulosic type; LH
is concerned. This is because in the fillet weld the heat electrodes which have become wet. Wet sub-
loss is in three directions, in the cruciform it is in four merged-arc flux and unclean wire and plate sur-
directions, whereas in the case of butt weld it is in two faces. Flux-cored wires for C0 2 -welding with open
directions. This is shown in Fig. 10.3. seam stored in open for a long time.
3) Heat input rate 2 LH electrodes taken fresh from airtight cartons,
Heat input rate, or arc energy is an important but not dried before use. Dry submerged-arc flux
parameter, which is given by the following formula: and clean wire and plate surfaces, but flux not
Arc Energy = V x A x 60 heated before use. Flux-cored wires for C0 2 -weld-
S x 1000
ing with open seam, but kept in airtight container
V = arc voltage
A = welding current 3 LH electrodes dried above 250°C for one to two
S = welding speed or arc travel speed (mm/min) hours before use. Submerged-arc flux dried at
To this heat input rate, we have to add the preheat 250°C for one hour before use, clean wire and clean
and interpass temperature, which may be maintained by plate surfaces. Solid unclean C0 2 -wires and un-
additional heating. clean flux-cored seamless C0 2 -wires
Table 10.4: Heat input rates various electrode sizes

Arc energy
KJ/mm
Run length per
Electrode dia. 450 mm electrode 3n powder With iron powder
3.2 100 2.0 2.5
1.35 1.6 1.8
180 1.2 1.4
4 150 2.2 2.5

185 1.8 2.0
235 1.4 1.6
5 150 3.5 4.0

205 2.5 3.0
260 2.0 2.35
6.3 150 5.5 6.5

200 4.0 5.0
275 3.0 3.5
4 LH electrodes dried above 400°C for one hour or cooled down to below 150°C, the temperature at which
at 350°C for two hours before use. Agglomerated HIC begins to occur. If postweld heat treatment is to be
submerged-arc flux dried at 500°C and fused flux given, the preheating operation must be extended without
dried at 250°C used with extremely clean wire and interruption to the postheat treatment. In that case, not
plate surfaces. Solid clean C0 2 -wire and clean only is the possibility of HIC completely eliminated but
flux-cored seamless C0 2 - wire. Clean solid wires the hard martensite formed in the HAZ is transformed to
used with the TIG process tough tempered martensite.
It is also possible to prevent the formation of HAZ
British research workers have linked these factors martensite in the case of higher carbon steels, showing
mathematically and produced prediction diagrams high hardenability, without recourse to postweld heat
which bring together information on carbon equivalent treatment if its isothermal transformation diagram is
and joint thickness, so that preheat levels and arc ener- available. In Fig. 10.4, it is indicated through the
gies can be selected for welding without the risk of HIC. superimposed arrow that a minimum preheat, interpass
Such diagrams are provided in BS: 5135-1974 "Metal- and postheat temperature of 360°C should be specified
Arc Welding of Carbon and Carbon-Manganese Steels" to produce a non-martensitic microstructure. Unless the
and the equivalent Indian Standard IS : 9595-1980. diagram is based on very high austenitising temperatures
The above discussion suggests several practical steps (much above the conventional 1,250°C), reflecting those
for preventing HIC cracking. In depositing a root pass in attained by the coarse-grained region of the HAZ, the
a thick-walled alloy steel piping, for example, TIG process minimum holding time should at least be twice that
could be preferred to manual metal-arc, since it is a indicated on the diagram for 100% transformation. It
hydrogen-free process. MIG/C0 2 is another process to must be noted, however, that the microstructures
be preferred to manual metal-arc and submerged-arc produced in this way are certainly harder and less tough
processes if HIC is to be overcome, because it generates than tempered martensite.
least amount of hydrogen in the arc. In cases where preheating even to 150°C is ruled out
In the case of steels in which formation of a hard, due to large job size or difficulty of access, a simple
susceptible microstructure during welding is inevitable, method to avoid HAZ cracking is to use an austenitic
HIC can be avoided by maintaining a preheat of above stainless steel manual electrode. Hydrogen is highly
150° C throughout the welding operation and for a soluble in austenitic weld-metal even at room temperature,
sufficiently long time thereafter. This allows most of the and therefore diffuses from it more slowly than through
dissolved hydrogen to escape before the weldment has ferritic weld-metal and HAZ microstructures. Any
n1 1
y
(
\
t
START OF | \
\
V
TRANSFORMATION
1
LU
s~ FINISH OF |
TRANSFORMATION
k.
CC
<
_Z
— ♦ ♦
QC
LU
u HOLDING
!♦ "
5 | \
TEMPERATURE 3 6 0 ' \
Ms 0% 100% |
[HOLDING TIME
TIME •
Fig. 10.4: Selection of postweld heating time and temperature from the transformation characteristics
of a steel to ensure the softest microstructure in the HAZ
hydrogen entering the HAZ from the weld is thus diffused N / mm2, which are alloyed with small amounts of Cr
away faster than it is replaced and hence it does not and Cu (Ni optional) to make them highly resistant to
reach dangerous levels of concentration in the HAZ. atmospheric corrosion in service. Hence they are termed
Since HIC is a delayed process, it is an usual practice weathering steels. On exposure to atmosphere, these steels
to carry out non-destructive testing of a welded develop a corrosion-resistant oxide coating of pleasing
fabrication at least 48 years after completion, to make natural colour. Hence they are used in a bare state, thus
sure that all possible hydrogen-induced cracks are saving on initial and subsequent periodical painting.
accounted for. Many imposing buildings and bridges have been
We shall now discuss in detail the weldability of the constructed in the U.S.A. with these steels. There are two
following important constructional steels referred to grades available: COR-TEN 'A' and COR-TEN 'B\ The
under Groups 1 and 2: latter was developed for use on large structures such as
1) ASTM A374/375 represented by COR-TEN steel bridges, where yield strength level of the 'A' grade is
2)A537Gr.B demanded in higher thicknesses. Tables 10.5 and 10.6
3) A514/517 represented by USS T-l steel give the mechanical properties and chemical
4) Nickel steels for cryogenic service compositions of these steels.
5) High-strength steels HY-80, HY-100 and Hy-130 In choosing a welding consumable, it is important to
bear in mind that the weld-metal must have corrosion-
1) COR-TEN steels (ASTM A374/A375): These are
resistant properties equal to those of the parent plate and
structural steels with minimum tensile strength of 480
Table 10.5: Mechanical properties of COR-TEN steels

Yield stress minimum, N / mm 2 Elon. min. Charpy-V notch
on gauge impact test
length of
Grade Tensile Up to Over Over Over Over 200 5.65 Max. Temp., Energy Thickness
strength and 12 mm 25 mm 40 mm 63 mm mm VSo bend °C min. max.,
minimum including up to and up to and up to and up to and radius average, mm
N / mm2 12 mm including including including including joules
25 mm 40 mm 63 mm 100 mm
COR-TEN'A 480 345 325 325 By agreement 19 21 lj/2T 0 27 12
COT-TEN'A 480 345 345 345 By agreement 19 21 1V2T 0 27 40
-15 27 40
Table 10.6: Chemical compositions of COR-TEN steels

5
Grade Chemical composition % ladle analysis
C Si Mn S P Cr Cu V Ni
COR-TEN'A' 0.12 max. 0.25-0.75 0.60 max. 0.05 max. 0.07-0.15 0.30-1.25 0.25-0.55 0.65 max.
COR-TEN 'B' 0.10-0.19 0.15-0.50 0.90-1.25 0.05 max. 0.04 max. 0.50-0.70 0.25-0.40 0.02-0.10
m u s t d e v e l o p a m a t c h i n g colour on s u b s e q u e n t same strength levels, it is seen that CO R-TEN 'B' contains

weathering. Basic low-hydrogen electrodes of matching Cr and Cu, and hence has a slightly higher carbon
or near-matching chemistry, satisfying this condition, are equivalent value and hence higher hardenability for a
available. Sometimes a 2.5% Ni-alloyed steel electrode is given thickness. The question therefore arises: Are the
used in applications w h e r e t o u g h n e s s at low w e l d i n g p r e c a u t i o n s defined in BS 5135-1974
temperatures is required. Exposure tests have shown that (specification for metal-arc welding of carbon and carbon-
the 2.5% Ni deposit gives a darker appearance after six manganese steels) valid for COR-TEN 'B' Experimental
months' exposure, but develops a close colour match after work has shown that they are valid, which means that
a weathering period of two years. COR-TEN 'B' and C-Mn steels of similar CE and strength
To save on costs, unalloyed C-Mn electrodes which levels have similar weldability with regard to hydrogen-
achieve the required mechanical properties can be used induced cracking.
for all the passes except the capping pass, of butt and 2) A537 Gr. B steel: This is a typical low-alloy steel
fillet welds. The final capping pass is then deposited which is m a d e to fine-grain practice and is then
with a matching or 2.5% Ni electrode. quenched and tempered to attain high strength with
In submerged-arc or M I G / 0 2 w e l d i n g , use of excellent low-temperature notch toughness (Charpy-V
standard C-Mn steel consumables is satisfactory if deep impact value of 15 ft. lb or 20 J min. at -60°C)in thicknesses
penetration technique is adopted. The weld-metal picks up to 100 mm. Small residual elements such as Cr, Ni,
up sufficient alloying elements to develop the weathering Mo and V enhance the response of the steel to heat
characteristic. If multi-pass low penetration technique is treatment, which does not produce any martensitic
used, one can fill the weld with C-Mn consumables and hardening. Applications include containment vessels,
cap it with matching or 2.5% Ni electrode. Alternately, pressure vessels and other structures for low-temperature
the entire weld can be made with COR-TEN electrode service. A537 B is used in preference to the normalised
wire. Electrode wires matching the composition of COR- A537 A grade, where critical low-temperature structures
TEN steels are available in many countries. are to be used without postweld heat treatment, because
COR-TEN 'A' is normally used in thicknesses less of its inherently superior notch toughness.
than 12 mm and can therefore be welded in the same Chemical c o m p o s i t i o n limits a n d mechanical
way as ordinary mild steel without any danger of HIC. If properties of the steel are given in Tables 10.7 and 10.8.
the thickness exceeds 12 mm, slightly more stringent For welding Gr. B steel, low-hydrogen consumables
precautions are necessary, such as the use of well-dried must be used. Preheating may be necessary depending
low-hydrogen electrodes and sufficiently high welding on plate thickness, ambient temperature and degree of
heat input. restraint. Preheating temperatures range between 40 and
The danger of HIC exists more in the case of COR- 100°C. Higher temperatures must be avoided.
TEN 'B', since it is used in heavy sections. When Welding consumables must not only match the
compared with ordinary C-Mn structural steels of the strength but also the sub-zero impact property of the
Table 10.7: Composition of A537 steels

Composition %
C Mn P S Si
0.24 (typical 0.20 max.) 0.70/1.35 (typically 0.035 max 0.040 0.15/0.50
1.00 min)
Al Cr NI Cu Mo
(for fine grain practice) 0.35 max 0.25 max 0.25 max 0.08 max
Table 10.8: Mechanical properties of A-537 steels

Tensile strength, Yield strength, Elong. in 8 in. Elong. in 2 in.
Grade Heat treatment N / mm2 min. N / mm2 min. % min. %
A Normalised 480-620 345 18 22
B Quenched and tempered 550-690 410 22
parent plate. Manual electrodes conforming to AWS class J Mn-Mo-B

E 8018 Cl (2% Ni) or E 8018 C2 (3% Ni) normally meet K Mn-MoB
these conditions in all welding positions. Type E 8018 L Cr-Mo-Cu-Ti-B
C3 (1% Ni) matches A537-B toughness properties in the M Mn-Ni-Mo-B
flat position, but may not do so in the vertical position, United States Steel supplies T-l steel in three types as
because of the comparatively high heat input rates follows:
associated with vertical welding. The slower cooling of USS T-l steel: A multi-purpose constructional alloy
the weld deposit results in lower impact strength. For the steel with a minimum yield strength of 1,00,000 psi. This
same reason, in submerged-arc welding it is advisable to low-carbon alloy steel combines high yield strength,
make multiple-pass welds at high welding speeds, in excellent toughness, excellent resistance to corrosion and
preference to fewer passes at low welding speeds. High impact/abrasion, and is weldable in all its forms. Meets
welding heat input can also lower the toughness of the all requirements of ASTM Specification A514 Grade F
heat-affected zone of the welded joint. The submerged- and A517 Grade F.
arc welding wire is usually 2.25% Ni-Cu or Ni-Mn-Mo USS T-l type A steel: A multi-purpose constructional
alloy steel. A wire carrying the designation MIL-100501 alloy steel which features the advantages of USS T-l
is used in the U.S.A. for critical applications and it has steel, but with a change in chemical composition which
the following chemistry. It provides a much higher tensile makes it more economical for many applications in plate
strength compared to the plate which is not regarded as sizes from 5 to 32 mm including. Meets all requirements
objectionable: of ASTM Specification A514 Grade B and A517 Grade B.
C—0.04% Mo—0.30% USS T-l type B steel: A multi-purpose constructional
Mn—1.6% Ni—1.7% alloy steel which offers the same advantages of strength,
Si—0.30% S&P—0.005% each toughness, resistance to impact/abrasion, and
3) USS T-l steel (ASTMA514/A517): This is a multiple-weldability, with a change in chemistry which makes it
alloy boron-containing steel with minimum yield economical in the thickness range over 32 mm through
strength of 1,00,000 psi (690 N/mm 2 ), which is used 50 mm. It is available from 5 mm through 50 mm
extensively in earth-moving and mobile equipment, including. Meets all requirements of ASTM Specification
bridges, buildings, pressure vessels, storages tanks, A514 Grade H.
penstocks, spiral cases and ships. The steel is water- Chemical compositions of the three types are given in
quenched from 890°C (minimum) and tempered at 620°C Table 10.9.
(minimum) to produce a microstructure of tempered The discussions that follow are generally applicable
bainite and tempered martensite. to all the three grades.
ASTM A514/A517 specifies several grades as
indicated below, so that within carefully chosen limits of ISOTHERMAL TRANSFORMATION OF T-1 STEEL
thickness they provide high strength and toughness with The welding metallurgy of T-l steel is better
the least amounts of costly alloying elements: understood by studying its isothermal diagram shown
A517 grade Composition type in Fig. 10.5. A notable feature is that considerable time
A Mn-Si-Cr-Mo-Zr-B elapses before transformation starts in the pearlite
B Mn-Cr-Mo-V-B transformation temperature range of 704-593°C. This
C Mn-Mo-B means that transformation at high temperature, which
D Cr-Mo-Cu-Ti-B results in an undesirable microstructure will not occur
E Cr-Mo-Cu-Ti-B unless the cooling is abnormally slow.
F Mn-Ni-Cr-Mo-Cu-V-B Another important feature is that a sufficiently long
G Mn-Si-CrMo-Zr-B time elapses before transformation starts in the ferrite
H Mn-Ni-Cr-Mo-V-B and upper bainite range of 593-510°C, to permit the
Table 10.9: Compositions of T-1 steels

Mn Ni Cr Mo V Cu Ti
T-1 0.10- 0.60- 0.15- 0.70- 0.40- 0.40- 0.03- 0.15- 0.002
0.20 1.00 0.35 1.00 0.65 0.60 0.08 0.50 0.006
T-1 type A* 0.12- 0.70- 0.20- - 0.40- 0.15- 0.03- 0.0005 0.01-
0.21 1.00 0.35 0.65 0.25 0.08 0.005 0.03
T-1 type B* 0.12- 0.95- 0.20- 0.30- 0.40- 0.20- 0.03- - 0.0005
0.21 1.30 0.35 0.70 0.65 0.30 0.08 min.
*Cu addition is made for better corrosion resistance if desired.
quenching of heavy plates up to 50 mm t with negligible temperature of 393°C. This factor makes the steel
transformation in this range. It is important to avoid the highly resistant to quench-cracking. This is important
formation of ferrite and upper bainite, because in this from the welding viewpoint, because cooling from
case, the surrounding austenite which is rich in carbon such a high temperature provides a self-tempering
may either remain as such or be transformed to high- treatment.
carbon martensite or high-carbon bainite as the steel cools The beneficial effect of quenching and tempering on
to room temperature. Thus if the cooling rate is sufficiently the mechanical properties of 12.5 mm thick steel of this
slow to permit considerable transformation in the 593- type is shown in Fig. 10.6. When hot-rolled, the steel
510°C range, the final microstructure will consist of a acquires microstructure of proeutectoid ferrite and high-
mixture of ferrite, upper bainite and retained austenite, carbon martensite, has a comparatively low yield strength
together with high-carbon martensite or high-carbon of 550 N / mm 2 and very poor Charpy-V notch value at -
bainite. Such a combination will not confer good 46°C. In as-quenched condition, the microstructure
toughness on the steel even when tempered. contains high proportions of the low-temperature
At temperatures below 482°C, a short time is transformation products, martensite and bainite, and the
required to produce a homogeneous lower bainitic steel has much higher yield and tensile strengths and
microstructure, which is essential for high toughness, better toughness. When the quenched steel tempered at
particularly after tempering. An important feature is temperatures above 593°C, the yield and tensile strengths
that martensite formation starts at a relatively high are markedly lowered, while the elongation % is almost
5 TO 103 10*
TIME. SECONDS
A—Austenite, C—Carbide, F—Ferrite, I-T—Isothermal Transformation
Fig. 10.5: Isothermal transformation diagram for a typical A514/517 steel

TEMPERING TEMPERATURE °C
100 200 300 400 SQQ 6SQ 750
200h •
TENSILE STRENGTH
160h
z
120h YIELD STRENGTH \ V
HOT ROLLED
80
40h
30
CHARPY V. NOTCH ENERGY
AT -50'f ~i
~r uj
4= u
a cc ^ H O T ROLLED
20h *
S
</> O
<< roh *^=2^^° TENSILE ELONGATION
flc O •^HOT ROLLED

-L-
0 200 400 600 800 1000 1200 1400
TEMPERING TEMPERATURE T
Fig. 10.6: Effect of quenching and tempering on the longitudinal mechanical properties of
12.5 mm thick A517F steel
doubled. The toughness as indicated by Charpy values electrodes of classification E10018-M or E10018-G as
at -46°C is also vastly improved. an alternative.
According to AWS A5.5, the electrode must meet the
WELDING OF T-1 STEEL chemical and mechanical requirements as given in
For welding T-1 steel, consumables of matching Tables 10.10 and 10.11. Electrodes conforming to
strength, ductility and toughness must be used. Manual classifications El 1015 or El 1016 have also been used. It
electrodes classified as E 11018-M or El 1018-G in AWS is generally known that the only major difference between
A5.5 'Specification for low-alloy steel covered arc XXX15 or 16 and XXX18 is the addition of iron powder
welding electrodes' are considered to be most suitable to the covering of the latter, which improves welding
for welding these steels. Some fabricators also use performance and minimises undercut.
Table 10.10: Chemical requirements for weld-metal as per AWS A5.5

Electrode C Mn P S Si Ni Cr Mo V
class
E11018-G* 1.0 min. 0.80 min 0.50 min. 0.30 min. 0.20 min. 0.10 min.
E11018-M 0.10 1.30-1.80 0.030 0.030 0.60 1.25-2.50 0.40 0.25-0.50 0.05
E19918-D2 0.15 1.65-2.0 0.030 0.040 0.80 — — 0.25-0.45 —
E10018-G* — 1.0 min. — — 0.80 min. 0.50 min. 0.30 min. 0.20 min. 0.10 min.
E10018-M 0.10 0.75-1.70 0.030 0.30 0.60 1.40-2.10 0.35 0.25-0.50 0.05
Note: Single values shown are maximum percentages, except where otherwise specified.
* To meet the alloy requirements of the G group, the weld deposit need have the minimum, as specified in the Table, of only one of the
elements listed.
Table 10.11: Mechanical requirements for weld-metal as per AWS A5.5

Electrode TS, min. YS, min Eln ., min % V-notch impact
class N / mm2 N / mm2 values, min.
E11018-C 760 665 15 Not required
E11018-M 760 672 20 27J at -51°C
E10018-D2 690 600 16 27J at -51°C
E10018-G 690 600 16 Not required
E10018-M 690 607 20 27J at -51°C
It is obvious from Table 10.11 that electrodes whose Mn: Manganese increases strength and hardness, com-
classification carries the suffix M will be preferred to bines with sulphur, and acts as a mild deoxidant to
those having the suffix G for important applications, avoid porosity. Being comparatively inexpensive, one
since in the latter, ductility requirements are low and would like to use it to the maximum, but our experi-
impact requirements are not specified. ence is that beyond 1.65% it lowers the Charpy im-
It is to be noted that the chemical compositions of the pact value.
weld-metal and parent metal differ widely. This is because Ni: Nickel increases strength, also notch toughness, but
the main objective is to match tensile strength, ductility, beyond 2% and when it approaches 2.5% hot crack-
and notch toughness properties, and this can never be ing of the weld-metal is likely to occur, though this
accomplished by matching the chemistry. It must be can be taken care of by keeping S and P on the ex-
appreciated that the morphology of weld-metal is so tremely low side.
distinctly different from cast, forged, or rolled metal that Cr: Chromium increases tensile strength and hardness,
changes must be made in the composition of weld-metal but lowers ductility and Charpy impact values, hence
to adjust its properties to match those of a given parent it is used in small quantities.
metal. In a Q and T steel, this is all the more so, since the Mo: Molybdenum increases hardenability, strength, and
weld-metal has to attain the properties in the as-welded hardness. However, beyond 0.5% it reduces Charpy
condition, which the parent metal acquired through impact values.
controlled heat treatment. Ti: Use of ferrotitanium as deoxidiser is avoided, because
To u n d e r s t a n d the basis on which high tensile Ti in the weld-metal (unless in minute quantities) low-
electrodes such as E11018 are developed, one must know ers Charpy impact value.
the effects of various alloying elements on the high tensile V: Vanadium serves as a great strengthener of weld-
weld-metal. They are described below: metal, but we avoid its use as it lowers Charpy impact
C: Carbon is an inexpensive and a potent strengthening properties severely, particularly after stress-relief.
agent, but it is kept as low as possible in the weld- The two electrodes developed by the Oerlikon group
metal. Carbon above 0.10% reduces notch toughness for welding T-l steel are detailed in Table 10.12. Type A
and causes weld-metal to crack, especially in the root is of E11008-M class and type B is of E10018-M class.
pass under heavy restraint, where the carbon content T - l steel is extremely susceptible to hydrogen-
increases further by dilution with the plate. Although induced cracking, more so because the procedure
the AWS A5.5 specifies 0.10% maximum, the electrode demands that welding is performed with limited heat
producer would like to restrict it to around 0.05%. input and preheating is restricted to relatively low
Si: Silicon is always added through the covering as a temperatures. The low-hydrogen manual electrodes are
deoxidiser to avoid weld-metal porosity. Beyond baked during manufacture at a temperature of 425°C to
0.50% it lowers the Charpy impact value. meet the AWS requirement that the coating moisture
Table 10.12: Specifications of Ador Welding electrodes for welding T-1 steel
Typical weld-metal composition % Mechanical properties
Electrode C Si Mn Ni Cr Mo S P TS, YS, %EL, Charpy-V
min. min. l=5xd at-51°C, J
A 0.06 0.50 1.45 2.15 0.35 0.40 0.015 0.020 765 675 20 38
B 0.06 0.50 1.45 1.90 0.25 0.30 0.015 0.020 715 715 22 45
content must be below 0.2%. They are then packed in thicker plates, because heat is dissipated more rapidly in
sealed airtight metal containers. Such electrodes are thick sections than in thin sections. For the same reason
suitable for use directly from the container, provided the higher arc energy can be used in thicker plates. This is
contents once opened are kept in a dry place and used indicated in Table 10.14, which gives the upper limits of
up within a few hours. If the electrodes are packaged in heat inputs for different plate thicknesses at various
cardboard containers with moisture-proof wrapping, it preheat levels.
is quite likely that they have picked up moisture during To prevent cracking in welds the following steps are
storage and require rebaking at the user's place. recommended for the conditioning and handling of low-
For submerged-arc welding, combinations of Mn-Ni- hydrogen electrodes:
Cr-Mo alloyed wires and neutral flux or carbon-steel a) As soon as an electrode container is opened, the
wires and alloyed flux specially designed for T-l steel electrodes should be removed from the container and
are available. Similarly Mn-Ni-Cr-Mo wires for MIG placed into a baking oven.
welding with argon-oxygen shield are also available. b) Electrodes should be baked according to the
A very important condition for welding T-l steel is to manufacturer's instructions, usually 427°C for one hour.
restrict heat input by controlling arc energy and The electrodes should be charged at 260°C maximum
preheating/interpass temperature. If the heat input is and heated to 427°C, with the electrodes placed no more
too low, hydrogen-induced cracks may develop in the than three layers deep on a tray. (This baking may be
HAZ or the weld-metal owing to extremely fast cooling. omitted only if the user has satisfied himself that the
If the cooling rate is too low, as a result of very high electrode coating, as ready for use, contains less than
input, the original microstructure of tempered bainite and 0.2% moisture.)
tempered martensite will transform to ferrite with high c) Whether baked or not the electrodes should be
carbon m a r t e n s i t e regions or to coarse bainite. immediately transferred to a storage or holding oven at
Consequently, the strength and toughness of this zone 121±14°C
will be adversely affected. The weld-metal also will have d) Both, the baking oven and the storage oven should
lower strength and toughness. be equipped with air-circulating systems and temperature
Suggested preheating temperatures for different controls and indicators.
welding processes are given in Table 10.13, which shows e) Electrodes should not be placed in the storage oven
that it is possible to use higher preheat temperatures in unless sufficiently dry as-received or properly baked.
Table 10.13:: Preheating data for weldincj T-1 steel using various processes
Minimum preheat or jLnterpass temp.,°C
Submerged-arc process
Plate thicknc?ss, Shielded metal-arc Gas metal-arc, gas Alloy or carbon Carbon-steel
mm process tungsten-arc or flux- steel wire wire
cored-arc processes neutral flux alloy flux
Up to 19 10 10 10 10
Over 19-45 50 50 50 94
Over 45-64 80 80 80 150
Over 64 110 110 110 200
Table 10.14 : Maximum heat input rates for T-1 steel

Plate thickness, mm
Preheat and inteirpass 4.8 6.3 12.7 19 25.4 32 45 51
temp.,°C
Max. heat input rate KJ / mm
21 1.06 1.40 2.75 4.76 Any Any Any Any

93 0.82 1.14 2.2 3.40 6.80 Any Any Any
150 0.67 0.95 1.85 3.23 4.96 6.90 Any Any
200 0.50 0.75 1.57 2.56 3.66 5.0 6.50 Any
f) The lower-strength welding electrodes (E7018 and quenching and tempering, and (e) alloying with nickel.
E7028) should be used within one hour after removal Among these, the last factor is most effective as indicated
from the storage oven. The higher strength electrodes (E in Fig. 10.7, which compares the sub-zero notch
10018 and E 12015) should be used within 30 min after toughness properties of carbon steel with four grades of
removal from the storage oven. nickel steels. It shows that with increasing amounts of
g) Electrodes removed from the storage oven and nickel in the steel, the range of applicability is extended
exposed in a clean, dry location for times longer than as follows:
those specified should be rebaked as described under
Ni content in steel Lowest service temperature
item b. Electrodes exposed to water, grease, or dirt should
be destroyed. % CQ
h) When welding is done in areas of high humidity, it 2 V4 -60
is advisable to store a suitable quantity of electrodes in a 3% -100
small portable oven, from which electrodes may be drawn 5 -130 (-162°)
one at a time by the welder. 9 -196
The stringer bead technique, in which beads are The condition of heat treatment is also important as
deposited without weaving is preferred for T-l steel shown for 3.5 Ni steel in Fig. 10.8. In the hot rolled
welding. Weaving means lower arc travel speed and condition, the steel gives widely scattered impact values.
hence increased heat input. Special care is necessary in Normalising improves the impact strength, mainly
vertical welding, where wide weaving is normally through grain refinement. Quenching and tempering is
employed. Stringer bead technique reduces distortion even more effective, because in addition to producing a
and ensures good weld-metal toughness. fine grain size, it distributes the carbide particles more
Experience and full-scale tests have established that favourably in the alloyed matrix. However, with all the
stress-relief is not necessary in welded quenched and best possible factors, the service applications of Ni steels
tempered steels, including T - l steel. In fact, a heat cannot be extended below -196°C. For such extremely
treatment in the temperature range of 510-694°C lowers low temperatures, austenitic Cr-Ni stainless steel (type
the toughness of the weld-metal as well as the HAZ, 304) is ideal, because as indicated in Fig. 10.7, this
particularly if slow cooling is resorted to. It can also cause material does not show a toughness transition range and
intergranular cracking in the coarse-grained regions of maintains superior impact strength down to the lowest
the HAZ. The cracking occurs in the early stages of the testing temperature of -270°C. This unique property of
heat treatment, before the high residual stresses have been Cr-Ni austenitic steel is attributable to its face-centred
reduced. It may be interpreted as a kind of stress rupture. cubic crystalline structure. However, its comparatively
If stress-relief must be used to retain dimensional low yield strength at room temperature is sometimes a
stability during machining or to avoid possible stress- handicap.
corrosion in service, the temperature employed must not Although Al helps to produce a deoxidised (fully-
exceed that used for tempering the steel (620°C for T-l killed) and fine-grained steel, residual Al in excess of
steel). In fact, it should be about 30°C below the tempering about 0.05% has an adverse effect on sub-zero fracture
temperature, to avoid lowering the strength of the steel. toughness. S and P should also be restricted to lower
4) Nickel steels for cryogenic service: Cryogenic service levels t h a n found in steels for n o r m a l service
is generally understood to mean service involving t e m p e r a t u r e s . C a r b o n content is also i m p o r t a n t .
temperatures below -50°C. Steels meant for cryogenic Specifications for nickel steels specify relatively low
service are termed cryogenic steels. Steels in general lose carbon levels and yet the steelmaker must keep carbon
their toughness as temperature is lowered, and sometimes percentage as low as possible, while complying with the
register a sharp fall over a narrow temperature zone called specified strength levels. For a given steel, impact strength
transition range. A cryogenic steel must have the fracture falls as thickness increases. Severe cold working (more
toughness transition temperature below the intended than 3%) can affect toughness, more so if the steel is
service temperature. In other w o r d s , it must give sensitive to strain hardening.
minimum Charpy-V impact value of 20 J or 15 ft-lbs Cryogenic steels are mostly used in the fabrication of
(sometimes more) at service temperature. containment plant for liquefied gases, in which case the
Factors which help to lower the transition temperature service temperature is the boiling point of the liquefied
of a steel are: (a) minimal carbon, (b) deoxidation or gas. Boiling points of the commonly handled commercial
thorough killing, (c) fine grain size, (d) normalising or gases are given below:
TEMPERATURE.°C
-200 -150 -10
-300 -200 -100

TEMPERATURE. *F
Fig. 10.7: Impact properties of nickel steels and stainless steel over a wide range of temperature. Boiling points of
gases are shown on abscissa
Gas Boiling point, °C and propane. Voluminous cargo tanks have been built
into ocean-going ships to transport these materials.
LPG 20 (under pressure)
The steel is welded with electrodes of AWS class
(liquefied petroleum gas)
E8016-CI or E8018-CI which deposit approximately
Ammonia -30
2.5% Ni steel, using preheat temperature of 100-125°C. If
Butane -32
the carbon level is above 0.20% or the joint is heavily
Propane -42
restrained, a 150°C preheat is advisable. The standard
Carbon dioxide -80
codes usually demand that welded pressure vessels or
Ethylene -105
pressure pipings for cryogenic service be stress-relieved
LNG (liquefied natural gas, -162
at about 625°C.
mainly methane)
Methane 3.5% Ni steel (ASTM A203/300): This steel has also
-165
Oxygen two grades D and E, the latter having higher carbon levels
-183
Nitrogen for different thicknesses. It is usually supplied in the
-196
Hydrogen normalised condition, though further improvement in
-253
Helium t o u g h n e s s can be obtained by t e m p e r i n g after
-269
normalising or tempering after quenching.
2.25% Ni steel (ASTM A203): The steel consists of two The steel is welded with electrodes of AWS class
grades. Grade A has carbon percentage limits of 0.17, E8016-C2 or E8018-C2 using preheat in the temperature
0.20 and 0.23 for plate thicknesses of up to 51 mm, over range of 100-125°C. The weldment has to be stress-
51 to 102 mm, and over 102 mm respectively. Grade B relieved at about 625°C.
has maximum carbon levels of 0.21, 0.24 and 0.25% for As mentioned earlier, the 3.5% Ni steel has been found
the same plate ranges. Grade B has higher strength levels suitable for service temperatures down to -100°C. In order
and is-naturally more difficult to weld than Grade A. to ensure its suitability for handling liquefied ethylene,
The steel is normally supplied in normalised state. It is which has a boiling point of -105°C and is used
widely used for storing and transporting of liquid butane increasingly in the manufacture of plastics, it is being
TEMPERATURE. °C
-250 -200 -150 -100 -50 0 , 50
| i i i i r *
1
1
1
1
1
1
I ! 1
1
1
100 1
1 !
< 80
>
a CC
< 1 | *A 8// <

&/ ^11 CC
f
UJ
x 60
u 5
o CC UJ
z
M
UJ
o/ O
¥-4
> 40^—tr O
>- —'
I1 W° CC
CC
< CO
i
X
CJ
20 < crtfcD [
1 Z>^\#
1
L i
1 1
u i i 1 4 I . . 1
-400 -300 -200 -100 100 200
TEMPERATURE. °F
Fig. 10.8: Effect of condition of heat treatment on impact properties of 31/2 nickel alloy steel
subjected to quenching and tempering, which shifts the Recently the steel has been upgraded to make it
transition temperature of the steel to below -105°C. suitable for service temperatures down to -162°C, so
However, improvement in the E80XX-C2 class electrode that it can be used for handling liquefied natural gas
to match this toughness level has not been possible, and (LNG). A typical steel is Armco cryonic 5. It is essentially
the specification AWS A5.5 still specifies for the weld a l o w - c a r b o n , 5% Ni steel w i t h a d d i t i o n of
deposit minimum impact value of 27 J (20 ft-lb) at -73°C approximately 0.25% Mo. Mn and Si are normal for alloy
For the present, for liquid ethylene service a stainless steels, but P and S are kept at low levels to ensure
steel (E309) or an Inconel electrode is considered more maximum notch toughness. The key feature of cryonic
suitable. 5 is a new three-step heat treatment that produces the
To obtain adequate toughness in submerged-arc excellent combination of properties. This treatment
welding, the practice has been to use a special 3.0% N i - involves conventional quenching, temperising, and
1% M n - 0 . 2 5 % Mo wire a n d m u l t i - p a s s w e l d i n g reversion annealing. These latter two treatments involve
technique with relatively low heat input rate. MIG heating to specific temperature ranges in the two-phase
welding with the same wire, but containing additional (a+y) region and cooling therefrom at specific rates. They
Al has been quite successful. refine the microstructure by controlling the amount of
5% Ni steel (ASTM A645) : This steel was first stable austenite, ferrite, and low-carbon tempered
introduced in Europe for service applications down to - martensite. This structure accounts for the combination
130°C, but did not gain acceptance in the U.S.A., of high strength, ductility, and exceptional notch
presumably because it did not offer sufficient advantage toughness.
over the 3.5% Ni steel to justify its higher cost. Also a The steel is welded with filler metals and welding
matching filler metal of 5% Ni was found to be unsuitable, procedures, which have been found suitable for 9% Ni
as the weld-metal was seriously prone to cracking. Hence steel (for example, Inconel 92 filler metal conforming to
austenitic stainless steel or nickel-base electrodes had to AWS class ERNiCrFe-b for MIG welding or electrode
be used, whose weld deposit has comparatively lower Inco-Weld B of AWS class ENiCrFe-4 for manual
strength at room temperature. welding).
9% Ni steel (ASTM A353/A553): A353 refers to the steel mm 2 ). It is used extensively in the U.S.A. for military
supplied in the double normalised and tempered equipment, particularly as submarine hull plate, because
condition and A553 to that supplied in the quenched of its exceptional fracture toughness at low temperatures
and tempered condition. Both grades show extremely and under ballistic shock loading. Typical composition
good notch toughness at -196°C, but A553 has a higher of HY-80 steel plate is as follows (%): C-0.16, Mn-0.25,
yield strength (85 ksi or 585 N / m m 2 minimum) as P-0.012, S-0.024, Si-0.22, Cr-1.30, Ni-2.5, Mo-0.35, V-
compared to A353 (75 ksi or 515 N / mm 2 minimum). 0.0015, Ti-0.004, Al-0.004, B-0.0005.
Both have the tensile strength range of 100-120 ksi (690- For welding, preheat and interpass temperatures and
825 N/mm 2 ) and require identical welding procedures. heat input should be controlled within the limits given
Normally no preheat is required but as plate thickness in Table 10.15. However, it is a common practice to preheat
increases and joint restraint becomes more severe, it is to 100-150°C wherever possible, and use stringer bead,
advisable to preheat in the temperature range of 100- multiple-pass technique. The most suitable electrodes are
200°C to prevent cracking. 9% Ni steel electrodes are not AWS class E10018-M or E11018-M specified in AWS
suitable, because they deposit weld-metal of inadequate A5.5, 'Specification for low-alloy steel/covered arc
toughness which is prone to severe cracking. Electrodes welding electrodes'. Welded joints exhibit excellent
of the E310 class (25 Cr/20 Ni stainless steel type) give toughness and matching strength in the as-welded
adequate weld-metal toughness at -196°C, but the tensile conditions. Fillet welds are sometimes made with E9018-
strength is much lower. Nickel alloy electrodes of AWS M electrodes. Stress-relief is provided only in cases where
class ENiCrFe2 or ENiCrFe3 (for example, Incoweld 'A' dimensional stability is to be achieved or stress-corrosion
or Inconel 182) are a better choice, because the weld-metal cracking is to be avoided in certain service environments.
deposited by these electrodes has approximately 95% The temperature is limited to 550°C.
joint strength efficiency, besides good notch toughness For M I G / C 0 2 welding, the Military Specification has
at -196°C. MIG welding can be satisfactorily carried out set the compositional limits as follows:
using Ni-alloy wires of AWS class ERNiCr-3 or ERNiCr- C-0.08 max., Mn-1.15-1.55, P-0.025 max., S-0.025
6. Ferritic steel wires of matching chemistry have also max., Si-0.35-0.65, Ni-1.15-1.55, Mo-0.30-0.60, V-0.10-
been found suitable for MIG welding in Japan, provided 0.20.
Ni content is raised to about 10.5% and levels of S and P Since the weld-metal contains V, stress-relief heat
are maintained very low in the parent plates (each below treatment has to be avoided, because of the embrittlement
0.015%) and in traces in the weld deposit (each below caused by its presence.
0.006%). For submerged-arc welding, a 65% Ni-27% Mo The specification also lays down that for submerged-
wire used in combination with a basic flux has been found arc welding, the flux-wire combination should give weld-
suitable. Recently submerged-arc wire of modified metal compostion A or B as detailed below:
chemistry (75% Ni-20% Mo-3% W) has been developed
in Japan, which in combination with a low silica
agglomerated flux, is claimed to deposit weld-metal of Weld-metal composition
adequate strength and toughness, matching the parant "A B~
metal. A m a n u a l electrode of s o m e w h a t similar c 0.12 max. 0.06 max.
composition has also been developed. 9% Ni steel Mn 0.80-1.25 1.00-1.50
fabrications are used in the as-welded condition without P 0.020 max. 0.010 max.
postheat treatment even for service at -196°C. S 0.020 max. 0.010 max
Whereas the hardness of 9% Ni steel in tempered Si 0.80 max. 0.50 max.
condition is in the range of C-22 to C-26 Rockwell, it Ni 0.80-1.25 1.40-1.90
increases to approximately C-38 to C-40 at a point about Cr 0.30 max. 0.10-0.30
1 mm from the fusion line in the HAZ. Despite this increase Mo 0.15-0.60 0.20-0.40
Cu 0.40-1.10 0.10-0.30
in hardness, there is no marked decrease of toughness in
V 0.05 max. 0.05 max.
the HAZ. Some fabricators insist on restricting the carbon
Ti 0.10 max.
in the parent plate to 0.08% maximum so that the HAZ
Zr 0.10 max.
hardness does not exceed C-36.
Al 0.10 max.
HY-80 steel: This is a quenched and tempered alloy
steel produced according to Military Specification in the HY-100, HY-WO(T) and HY-130 steels: These steels
U.S.A. to give a minimum yield strength of 80 ksi (550 N / developed in the U.S.A. have been described as
Table 10.15: Preheat and related data for HY-80 steel

Preheating temperatures for air carbon-arc gouging
Plate thickness, mm Minimum temperature,°C
28.5 or less 25
Over 28.5 66
Preheating for arc welding
Plate thickness, mm Minimum preheat and
Interpass temp.,°C
12.7 or less 25
Over 12.7 butt under 28.5 52
Over 28.5 95
Under no circumstances shall 150°C be exceeded
Heat input during arc welding
Plate thickness, mm Joules/mm, Maximum
12.7 or less 1,772
Over 12.7 but under 35 2,362
Over 35 3,150
'premium-toughness high-yield strength steels of the fracture toughness properties. Chemical compositions
quenched and tempered type'. Like HY-80 steel, HY- of these three steels are given in Table 10.16.
100 is used in the construction of military, marine and Electrodes of over-matching strength conforming to
commercial hydrospace vehicles. HY-130 has been AWS E120-18 are used for welding HY-100 steel. HY-
specially developed for use in the more demanding 80 welding procedures are used, but the risk of HAZ and
hydrospace and aerospace applications. w e l d - m e t a l cracking d u e to h y d r o g e n is m o r e
HY-100 is a 100 ksi (690 N / m m 2 ) yield steel based on pronounced, because of higher carbon equivalent and
HY 80 composition, but with higher Ni and Cr contents. higher strength of weld-metal. In the case of HY-100(T)
It is specified in ASTM A543 class 2. HY 130 is a steel steel, the risk of cold cracking is much less than in the
with a yield strength in the range of 130-150 ksi (895- case of HY-80 steel, because of lower CE (0.85 maximum
1030 N/mm 2 ) and is based on 5% Ni with additional Cr, for the maximum composition as compared to 0.95 for
Mo and V. A modified version of HY-100, designated by HY-80). HY-100(T) is also welded by the MIG process,
the U.S. Navy as HY-100 (T) is based on HY-130 and using the wire whose composition is given in Table 10.17.
has better weldability. It is an electric furnace vacuum- For welding HY-130, special E14018 class electrodes
degassed 3% Ni-Cr-Mo-V steel, which exhibits better have been developed in the U.S.A. For MIG welding, a
Table 10.16: Chemical compositions of high-strength QT steels

HY 100(T)
Element HY100 12.7-25.4 38-89 102-152 HY130
mm mm mm
C 0.20 max. 0.09-0.13 0.09-0.13 0.09-0.13 0.08-0.12
Mn 0.10-0.40 0.20-0.40 0.20-0.40 0.20-0.40 0.60-0.90
Si 0.15-0.35 0.20-0.35 0.20-0.35 0.20-0.35 0.20-0.35
Ni 2.25-3.50 2.5-3.0 3.0-3.5 4.0-4.5 4.75-5.25
Cr 1.00-1.80 0.95-1.25 1.20-1.60 1.40-1.80 0.40-0.70
Mo 0.20-0.60 0.25-0.35 0.35-0.45 0.45-0.60 0.30-O.65
V 0.03 max. 0.05-0.10 0.05-0.10 0.05-0.10 0.05-0.10
S 0.025 max. 0.020 max. 0.020 max. 0.020 max. 0.015 max.
P 0.025 max. 0.020 max. 0.020 max. 0.020 max. 0.010 max.
S+P 0.045 max. — — —
Ti 0.020 max. — — — 0.020 max.
Cu 0.25 max. 0.25 max.
special Mn-Ni-Cr-Mo wire described in Table 10.17 is vanadium, aluminium, tantalum and zirconium, which
used with argon-2 to 5% oxygen mixture as a shielding readily form carbides, nitrides or carbonitrides. The total
gas. Hydrogen-induced cracking in the HAZ is not a content of these elements seldom exceeds 0.15%. These
serious problem in the case of HY-130 steel, but weld- steels are usually delivered in normalised or control-
metal cold cracking is a significant problem. To avoid it,rolled condition as fine-grained steels with good
the interpass temperature during welding with E 14018 toughness properties. They are generally categorised with
electrodes should be 135°C minimum and one hour high strength low-alloy steels.
should elapse between the deposition of adjoining beads Micro-alloyed steels are actually structural C-Mn
within a layer or the beads in one layer to the next. The steels, in which minute additions of the above-
interpass temperature should be prolonged for at least mentioned alloying elements raise the yield and tensile
one hour as a postweld heat treatment. Minimum preheat strengths, lower the transition temperature and improve
and interpass temperatures for welding various ductility. These additions also produce a fine, rather
thicknesses of HY-100 and HY-130 steels are given in than coarse microstructure in these steels during air-
Table 10.18. In one particular HY-130 fabrication cooling. This fine microstructure is the cause of
programme, heat input limits were set as shown in Table improved strength and toughness properties. With their
10.19 and the procedures adopted for welding in the flat good weldability and fabrication properties, these steels
position using stringer bead technique were as given in are now widely used for structural purpose including
Table 10.20. construction of bridges, buildings, material handling
equipment, etc.
Micro-Alloyed Steels Though these steels fall under the category of alloy
Micro-alloyed steels are defined as steels treated with steels, they are not strictly so, because the alloying
small additions of elements such as niobium, titanium, elements are not subject to chemical check analysis limits
Table 10.17: Compositions of typical MIG wires for welding HY-100(T) and HY-130 steels
% C Mn Si S P Cr Mo Ni Ti Cu Al V
For HY-100(T) 0.085 1.70 0.46 0.009 0.005 0.05 0.50 2.40 0.025 — — —
For HY-130 0.11 1.60 0.34 0.008 0.007 0.62 0.89 2.75 0.010 0.34 0.02 0.02
Table 10.18: Preheat temperatures for welding HY-100 and HY-130 steels
Temperatures, °C
Plate t, mm HY-100 HY-130
steel steel
To 12.5 38 24
Over 12.5 to 15.8 52 24
Over 15.8 to 19 52 52
Over 19 to 22.2 66 52
Over 22.2 to 25.4 66 93
Over 25.4 to 35 93 93
Over 35 to 36 93 107
Table 10.19: Heat input data for welding HY 130 steel

Thickness, Preheat temp., Max. interpass Max. heat input,
mm °C temp.,°C KJ / mm
6.3 Room temp. 121 1.2
8 Room temp. 121 1.2
9.5 Room temp. 121 1.4
12.7 Room temp. 149 1.6
19 93-121 149 1.6
Table 10.20: Procedure data for MIG and MMA welding HY-130 steel
a) MIG welds
Plate thick- Fillet metal Travel Number of Shielding gas
ness, diameter, Voltage, Current, speed, passes mixture
mm mm V amp mm/sec (total)
6.3 1.2 25.5 220 4.7 2 98%A/2%0 2
9.5 1.2 26.0 270 4.2 2 98%A/2%0 2
12.7 1.2 25.0 270 5.0 4 98%A/2%0 2
19 1.2 26.0 270 4.2-6.0 10-11 98%A/2%0 2
b) MMA welds
Plate thick- Electrode Travel Number of
ness, diameter, Voltage, Current, speed, passes
mm mm V amp mm/sec (total)
63 4 225 160 3.4 to 4.2 4 to 5
8 4 22.5 160 3.4 to 4.2 4 to 5
9.5 4 22.5 160 3.4 to 4.2 6 to 7
19 4 22.5 160 3.4 to 4.2 14 to 16
and other tolerances regularly imposed upon an alloy With such a flexible type of classification of these
steel. These steels are furnished to specified mechanical steels, a large number of different brands are available
requirements for designated thickness ranges. Limits of with a variety of chemical compositions, but similar
chemical composition are generally published for mechanical properties from different manufacturers.
information only, since adjustments in carbon and other
alloying elements are made by the steel producer as Metallurgical Aspects
necessary to maintain mechanical properties in products In micro-alloyed steels, strength improvements are
of various thickness ranges. In general, these steels must achieved by judicious control of transformation. To
produce the required properties with smallest possible achieve the desired properties, steelmakers control the
alloying contents in order to compete costwise. microstructure by using special metallurgical techniques.
International Institute of Welding has divided micro- Microstructure of micro-alloyed steels consists of a
alloyed steels into three categories, depending on their mixture of ferrite grains with isolated islands of pearlite.
strength, as follows: In some cases a small amount of bainite may also be
present, which does not differ very much from pearlite in
U.T.S. Y.P. Elongation its mechanical properties. The general features of
(kgf/mm2) (kgf/mm2) (%on5d) microstructure, on which mechanical properties of micro-
min. min. alloyed steels depend, are as follows:
1) Class 36 55-65 36 22
1) The relative proportion of pearlite
2) Class 40 55-70 40 20
2) The structure of pearlite, i.e. the interlamellar
3) Class 45 58-75 45 20
spacing
Majority of the steels in the above three groups have 3) The nature of ferrite, which can be influenced by
the following base analysis: solid solution and by precipitation
Carbon 0.22% max. 4) The grain size.
Manganese 1.0-1.5% Out of these four factors, the last two, namely, nature
Silicon 0.3% approx. of ferrite and the grain size play a major role in achieving
Sulphur and Phosphorus 0.04% max. each high strength and good toughness properties.
In class 36 steel, the most common micro-alloying 1) Proportion and structure of pearlite
element is aluminium. Several of the steels are also Carbon markedly increases pearlite content, and is
niobium treated and a few have vanadium, niobium, one of the most potent economical strengthening ele-
zirconium and titanium additions. The class 40 and 45 ments. Effect of manganese on strength is similar to car-
steels are rendered fine-grained by treating with bon but to a milder degree. With increase in relative
aluminium, niobium, vanadium and titanium or a amount of pearlite, tensile strength increases, but the
combination of two or three of these elements. yield strength does not increase significantly. The tough-
ness properties ar£ worsened with increase in the amount carbides in steel. Carbides can be further classified into
of pearlite. Pearlite formers also help in the hardening of two classes. The first group includes carbides like Fe3C,
steel. Thus their role in improving yield strength is lim- Cr7C3, etc., which have high solubility in austenite upon
ited by deterioration in weldability. The practical maxi- heating, and are of prime importance in the heat treat-
mum carbon content, at which micro-alloyed steels can ment of steels. The second group includes W2C, VC, NbC,
be used in as-rolled condition without martensite or TiC, ZrC, TaC, etc., which do not dissolve in the austenite
bainite formation is approximately 0.20%. For manga- when steel is heated under actual practical conditions.
nese also, there is an upper limit depending on the over- These carbides are formed by micro-alloying elements
all composition in general and carbon in particular. added in micro-alloyed steels. The characteristic feature
With refinement of pearlite, i.e. decrease in of these carbides is that, they are built up of ions of metal-
interlamellar spacing, tensile strength, yield strength and lic elements with carbon occupying interstitial positions
toughness are improved. Addition of manganese makes and have exceptionally high melting point and hardness.
austenite transformation occur at lower temperatures, Elements like a l u m i n i u m , v a n a d i u m and
leading to formation of fine pearlite. molybdenum form nitrides and carbonitrides in the
2) Strengthening ferrite presence of nitrogen (0.02%), which are also very hard
a) By solid solution: The fine pearlite is undoubtedly and have a high melting point. These precipitated
the major source of strength, but strengthening of the particles act as hard second phase particles within the
ferrite matrix is an indispensable adjunct. At low carbon matrix, and impede the motion of dislocations through
levels, elements like silicon a n d m a n g a n e s e are the material. This leads to the formation of dislocation
distributed between cernentite and ferrite. These elements loops around the particles and decreases the effective
improve strength of the ferrite by formation of solid spacing between them. Therefore increasing amount of
solution. stress is required to push each successive dislocation
The effectiveness of the impurity (solute) additions through. For a given volume of second-phase precipitates,
depends on the size differences of solute and solvent at- a higher stress is required for finer particle size because
oms and the percentage present. If the solute atom is larger of lesser interparticle spacing.
than the solvent atoms, compressive strain-fields are set In micro-alloyed steels, 0.02 to 0.05% aluminium is
up and if it is smaller, then tensile strain-fields are set up. normally added to form aluminium nitrides. Aluminium
In general, the higher strength levels obtained by this nitride is present in a large number of morphological
method of ferrite strengthening will be accompanied by a variations, from simple geometrical forms to large two
relatively smaller loss in plasticity, as compared with dimensional forms. A fine dispersion of these precipitates
considerable loss in ductility accompanying the hard- throughout the volume of the steel gives best results.
ness increase resulting from microstructural changes. In Along with aluminium, sometimes titanium is also
steel, the order of increasing effectiveness of various ele- used. Titanium forms carbides, nitrides or carbonitrides
ments for ferrite strengthening by solid solution appears with carbon and nitrogen, which have very high melting
to be as follows: chromium, tungsten, vanadium, molyb- points. Niobium is most commonly used and is usually
denum, nickel, manganese, silicon and phosphorus. added in an amount ranging from 0.01 to 0.10%. It forms
b) By precipitation: Precipitate formation is one of the finely dispersed niobium carbide, which at high
main mechanisms that operate in the micro-alloyed steels. temperatures takes the form of thin film at the primary
Precipitated particles can work in two ways: grain boundaries. This leads to deterioration in impact
i) The direct way in which they influence the properties. Therefore niobium is usually restricted to
mechanical properties by local deformation of the ferrite 0.05%. Vanadium is added in quantities 0.05-0.2%, and
matrix and by multiplication of dislocations. forms carbides, nitrides and carbonitrides.
ii) The indirect way by which the precipitates af- 3) Grain size
fect the grain size by influencing the austenitic grain or Grain size plays a decisive part in raising the strength
act as nuclei for ferrite during the austenite to ferrite trans- of micro-alloyed steels. In these steels, fine grain size is
formation. obtained through fine precipitation, controlled rolling or
The hard precipitates in micro-alloyed steels are nor- normalising. Fine precipitated particles formed by the
mally carbides, carbonitrides or nitrides. In relation to micro-alloying elements act as nucleating centres during
carbon, all the alloying elements added to steel can be transformation, leading to heterogeneous transformation
classified into two categories—elements which facilitate with lower free energy, causing faster transformation with
formation of graphite and elements which form the stable less supercooling and fine grain size.
In the controlled rolling method, rolling temperature, such as MIG/C0 2 , submerged-arc and electroslag are
amount of reduction applied and time between passes used, grain-coarsening is likely to occur in the HAZ
are controlled. Grain size becomes finer as the leading to a loss in strength and toughness, unless heat
temperature at which substantial reductions are made is input is restricted. More recent research has revealed that
lowered, provided this temperature is not too much below the greater difficulties faced in micro-alloyed steel
Ary Furthermore, reductions should be made regularly welding are not due to microstructural changes in the
and quickly as the temperature decreases, so as to have a heat-affected zone, but due to the transfer of alloying
uniform repetition of deformation and recrystallisation elements from the base metal into the weld, which is a
of the grains, but with minimum time for grain growth. function of composition of the parent plate and heat input
Below Ar3 there is limited additional grain refinement (dilution).
and cold-worked grains begin to appear in the The major variables affecting the weldability of micro-
microstructure. This means that only partial alloyed steels can be classified into two broad categories:
recrystallisation has occurred. The cooling rate during
transformation from austenite is also important, and this 1) Cleanliness of steel and its chemical composition
is especially true when cooling immediately after plastic 2) Welding process.
deformation of austenite. In the case of niobium treated 1) Cleanliness and chemistry of steel: The cleanliness
plates thicker than 10 mm, controlled rolling procedure of steel and its chemical composition together determine
is adopted for grain refinement to improve notch the metallurgical quality of steel and affect the heat-
toughness. affected zone and general properties of the weldment.
Some micro-alloyed steels are normalised for grain Cleanliness of steel is measured in terms of inclusion
refinement. Normalising is most effective when the content. Possibility of crack formation is more if a larger
normalising temperature is as low as possible above Ac, number of non-metallic inclusions are present in the zone
temperature, and time at this temperature is short. During adjacent to weld-metal, which is subjected to high
heating and holding at normalising temperature, temperatures and subsequently to upsetting caused by
precipitation of particles and grain refinement take place. thermal expansion. Most harm is caused by sulphur.
If this temperature is higher, coarsening of precipitates Keeping a manganese-to-sulphur ratio of 20 is normally
may start. In the case of micro-alloyed steels with taken as a safeguard against hot cracking tendency. I.
aluminium, normalising leads to the formation of fine Hrivnak has made a detailed investigation on the effect of
grains of aluminium nitrides. As mentioned earlier, the aluminium, niobium and titanium on the nature and
finer the precipitates, the finer will be the grains. properties of the inclusions with electron probe micro-
The grain size plays a decisive part in improving the analysis method. He divided inclusions into three
strength and toughness properties of the steel. For the groups:
same metallurgical structure, the yield stress increases The first group includes inclusions in manganese
as the grain size decreases. This is because of the fact steels in which aluminium is added These inclusions
that grain boundaries act as a barrier to slip across the are mainly manganese sulphide type with some bordered
grain boundaries. Finer grains also help in improving with a layer of aluminates. The number of aluminium
toughness properties. sulphide inclusions are very small. Because of their high
Grain boundaries at low temperatures are usually melting point, up to fusion line, these inclusions are not
stronger than the material in the grain itself. Thus, fracture dissolved.
in most metals at low temperature is trans-crystalline. Inclusions formed in steels, in which aluminium and
This signifies that fracture goes through the interior of titanium are added, make up the second group. In these
the grains rather than along the grain boundaries. This steels, manganese sulphide bordered with aluminates
is why low temperature impact properties improve are formed and along with this (Mn, Ti) S inclusions are
markedly with finer grain size. also formed. (Mn, Ti) S inclusions have a very low melting
Weldability of micro-alloyed steels is more complex point; therefore they are unfavourable from welding point
than that of ordinary C-Mn steels, which are usually of view.
welded without hydrogen control and with no The third group consists of inclusions formed in steel
limitations on arc energy input. Micro-alloyed steels are containing aluminium, titanium and niobium. In this
satisfactorily welded by the manual metal-arc process group more sulphide inclusions of combined MnS and
without heat input restrictions, provided low-hydrogen AIS type are formed. Titanium and niobium inclusions
electrodes are used. But when high heat input processes are present in a lower proportion. These inclusions
behave like those in the previous group and render the steels of the same C Since one of the major factors causing
steel more susceptible to hot cracking. hydrogen-induced cracking is the HAZ hardness, micro-
The microconstituents present in micro-alloyed steels alloyed steels are relatively less susceptible to this type
play a major role in weldability of micro-alloyed steels, of cracking, and therefore may tolerate more hydrogen in
more so in deep penetration welding techniques the welding consumable.
associated with submerged-arc welding and electroslag The phenomenon taking place in the HAZ of micro-
welding, since a large amount of micro-constituents is alloyed steels is entirely different from that of low-alloy
transferred into the weld pool. For this reason, during and mild steels. The thermal cycle during the welding of
the last few years the upper limit of niobium content in micro-alloyed steels produces a softer microstructure in
the micro-alloyed steels has been reduced from 0.10 to the HAZ as compared to the parent plate. The magnitude
0.06%. It has been shown that the content of about 0.001% and extent of this phenomenon increases with increase
niobium in the weld increases the transition temperature in heat input. During welding, different regions in the
by about 2°C. Similarly there is an urge anông steel HAZ pass through thermal cycles, which vary both in
makers to reduce vanadium and nitrogen content in peak temperatures attained and the cooling rate thereafter.
micro-alloyed steels. In micro-alloyed steels in the immediate vicinity of the
The low impact strength shown by the weld-metal fusion zone, carbides and carbonitrides are dissolved.
when it picks up micro-alloying elements by dilution is The zone of reprecipitation will vary depending on the
explained thus: the rapid freezing of the weld pool and cooling time. Below this temperature where dissolution
the intense cooling cause strengthening of ferrite by takes places, grain coarsening will occur. Further in the
precipitation of highly dispersed carbides and nitrides. region where the peak temperature is below ACX, there
This leads to high density of dislocations and the will be modification in distribution and density of
consequent state of stress. dislocations.
2) Welding process: As mentioned earlier, the J.L. Kaae has carried out butt welded tensile tests of
weldability of micro-alloyed steels depends to a large micro-alloyed steels using different welding processes.
extent on the welding process. The welding process He observed that the yield stress, tensile strength and
influences the weldability by affecting the heat-affected reduction in area of fracture were very much the same
zone properties and the weld properties by dilution from regardless of the process, which did not indicate any
the parent plate. loss in strength of welded joint. But, the limit of
a) Effect on heat-affected zone properties: In the case of proportionality decreased in the order of manual metal-
mild and low-alloy steels, the HAZ properties are usually arc (MMA), submerged-arc and electroslag welding.
predicted by the carbon equivalent formula: Compared to MMA welding, a decrease of about 20% in
C - r x ^ f e x Cr + Mo + V __, Ni + Cu proportional limit in the case of electroslag weld assembly
+ was observed.
^ 6 5. 15
This carbon equivalent is related to the hardenability By simulating welding conditions, the mechanical
of steel, i.e. the ability of steel to form hard martensite in behaviour of the HAZ in Nb-bearing micro-alloyed steels
the HAZ adjacent to the weld. The effect of various was studied by A.H. Arson. He obtained variation in the
alloying elements mentioned in the above formula yield strength of HAZ as a function of peak temperature,
actually depends upon whether they are present in solid which is shown in Fig. 10.9. As can be seen in the figure,
solution, in which case they assist hardenability. the yield strength decreases with increasing peak
Elements which form complex carbides and do not go temperature in the HAZ because of softening. But after
into solution during the period when HAZ is attaining a minimum it increases because of
austenitised, are not considered in the formula. reprecipitation of the dissolved precipitates.
Vanadium, being present in very low concentrations in The factors affecting the HAZ in micro-alloyed steels
micro-alloyed steels and forming complex carbides, is are also responsible for variation in its impact
not to be considered in the formula. Further, carbon being properties. The transition temperature increases
taken up by complex carbide particles present in a micro- with increase in amount of pearlite and decreases
alloyed steel, also prevents it from performing its role as with decrease in grain size, which in turn is related
promoter of hardenability and as a major element for to the peak temperatures attained in the HAZ. The
establishing hardness in transformed structures. results obtained by A.H. Arson on effect of peak
Heat-affected zones of micro-alloyed steels, thus, will temperature attained during welding on transition
be comparatively softer than those of mild and low-alloy temperature for niobium-bearing microalloyed steels
transferred into the weld by dilution. Dilution depends

on the heat input. Thus with low heat input process,
50
namely manual metal-arc welding, the dilution problem
48 is minimum. On the other hand, high heat input processes
E
E like the submerged arc and electroslag cause the weld-
5s
46 metal to pick up significant amounts of micro-alloying
elements from the base metal and consequently highly
44 dispersed carbides and nitrides are formed in the weld-
metal, probably with high density. The ductility of the
42 weld-metal is thereby adversely affected.
In IIW Doc. IX-1027-77, influence of micro-additions
40
on microstructure of weld-metal is reported. The
38 microstructure was studied with the help of electron
700 800 900 1000 1100 1200 1300
PEAK TEMPERATURE C O
microscope, and not with the optical microscope, because
the latter does not reveal the significant difference caused
Fig. 10.9: Effect of peak temperature on HAZ yield by micro-alloying elements in the w e l d - m e t a l
strength of Nb-bearing micro-alloyed steel microstructure. Weld-metal containing micro-additions
have finely dispersed carbide or carbonitride. The
is shown in Fig. 10.10. With peak temperature, the dislocation density was determined by X-ray diffraction
transition temperature increases initially because of method, and it was also found that the dislocation density
precipitation hardening. But at higher peak temperatures, increases with increase in micro-additions. This increase
the transition temperature decreases because of total effect can be due to the presence of stress of the second order
of overaging of precipitates and decrease in grain size. caused by the finely dispersed carbides and nitrides of
With further rise in peak temperatures, grain size micro-additions.
increases and coarsening of precipitates takes place The presence of fine carbide and carbonitride
which cause increase in the transition temperature. precipitates along with high dislocation density affects
From the study of influence of peak temperature on the mechanical properties of the weld. Studies have been
the tensile and impact properties of the HAZ, it is clear carried out on the influence of vanadium and niobium in
that these properties deteriorate with increase in heat micro-alloyed steels on the weld mechanical properties
input, which is a function of the welding process adopted.
b) Effect of dilution: Dilution is the decrease or increase
in alloy content of a weld deposit by virtue of base metal t I
melting and its incorporation into the weld puddle. In "D
10
Z
—► T
micro-alloyed steels, the micro-alloying elements are e> # a-
? 7Z> >
+ 20 T.S
70L
30k5 220
3
< -20h 210|
60-
25j65f- 4<00
Z 5oL
o
t 1|90|
</>
z
< -80
20H534(il8k)
-100
700 800 900 1000 1100 1200 1300 1400 0 4 8 12 15 20 24
PEAK TEMPERATURE (°C) CONTENT OF VANADIUM IN WELD-METAL (% 10"2)
Fig. 10.10: Effect of peak temperature on Fig 10.11: Effect of V on mechanical properties
HAZ transition temperature of weld-metal
% i
•o z
C9 # OL I4h
LU AC H
E 12
o
; 10
01
|70L
30k5 220
z
UJ
0C
2101
|6Q- </>
I-
o
25(-65f- 2<0 0
2
50-
190
0 10 20 30 40 50 60
20K534(il8k)
CONTENT OF NIOBIUM IN WELD METAL (% 10'2)
0 10 20 30 40 50 60 Fig. 10.14: Effect of Nb on impact strength of weld-metal
Nb
Fig. 10.12: Effect of Nb on mechanical
properties of weld-metal
16
<<
O CC
O i 20
_ 14 O Z
<-> O
E
o £ en
12 + 20 C
! 2
£-
£ 10h
o 8 -60
20 40 60 80 100
CONTENT OF NIOBIUM IN WELD METAL (% 10'2)
Fig. 10.15: Effect of Nb on COD transition temperature of

< weld-metal (in two thicknesses)
Q.
0h
0 4 8 12 16 20 24
C O N T E N T OF V A N A D I U M IN W E L D M E T A L (% 10" 2 )
< <
Fig. 10.13: Effect of V on impact strength of weld-metal
o 2
Q »-
O Z
and variation in critical crack opening displacement O O
(COD) for different plate thicknesses. The results are

shown in Figs. 10.11-10.16. e <
From these graphs it can be seen that the weld
properties deteriorate with increase in the percentage of -60
microconstituents. In welding with deep penetration 5.0 10.0 15.0 20.0 25.0
processes, dilution can be up to 50% and micro-element CONTENT OF VANADIUM IN WELD METAL (% 10~2)
content in the weld will be proportionately high. Fig. 10.16: Effect of V on COD transition temperature
Therefore, in order to ensure optimum weld-metal of weld-metal (in two thicknesses)
properties, niobium and vanadium in the micro-alloyed Preparation of parent material: General cutting
steel should not exceed 0.03% and 0.1% respectively. operations and weld joint preparations as used on
Other micro-alloying elements will also behave in a common structural steels can be performed on micro-
similar fashion. alloyed steels. Gas cutting is normally used in large-scale
J. Wegrzyn has suggested a comprehensive fabrications.
weldability index, Cw, for assessing the weldability of At TISCO, Jamshedpur, studies were carried out on
micro-alloyed steels. In his formula besides carbon and the influence of gas cutting process on manganese-
manganese, micro- additions are also taken into account. vanadium-nitrogen type steels. Figure 10.18 shows the
For niobium steels variation in hardness away from the gas-cut edges. This
clearly shows that in vanadium-nitrogen type steels,
c = (%C) + (%Mn) + 3x(%Nb)
10 hardness is never excessive and the plates can be gas-cut
without any danger of introducing incipient cracks or
For vanadium-nitrogen steels deterioration in the quality of the material. The narrow
c =<%Q + eufo) + <auD +3 x(%N) region where hardness is very high is normally removed
by grinding.
w v
' 10 3
Preheat: In general, for welding micro-alloyed steels
From the above formulae it can be seen that the effect
preheating is not required, since the cooling rate after
of manganese is markedly diminished, while that of
welding does not affect the properties of micro-alloyed
micro-alloying elements and carbon is very pronounced.
steel weld and heat-affected zone. However, Table 10.21
He suggested that for weldability index up to 0.35, the
gives the recommended minimum preheat temperatures
steel can be welded with no restrictions and it is possible
for vanadium-nitrogen containing steels of various yield
to obtain good impact properties at sub-zero
strengths, to be welded by basic-coated low-hydrogen
temperatures. In 0.35 to 0.45 Cw range, a thoughtful
electrodes.
procedure should be adopted in submerged-arc and C0 2
Welding consumables: Micro-alloyed steels can be
welding, and the heat input should be limited to 35 kj/
welded by any of the arc welding processes. Manual metal-
cm. For steels having Cw above 0.45, welding should be
arc welding is used most often, mainly because of the
carried out only by manual metal-arc process, using
nature (and very often the location) of the joints to be
electrodes of matching strength and impact properties of
welded. When conditions permit, flux-cored arc, gas-
the parent plate.
shielded metal arc and submerged-arc welding may be
The effect of heat input (i.e. weld dilution) on crack
used to realise the advantages that are obtained from
opening displacement is illustrated in Fig. 10.17 for
these processes in welding other steels. Submerged-arc
vanadium-nitrogen type micro-alloyed steel.
welding often proves more economical than other arc
r 0.016 NITROGEN "1
STEEL COMPOSITION processes. Excessive travel speeds and very thin or very
0.17 VANADIUM
L 1.47 MANGANESE -*
short welds (in relation to the workpiece dimensions)
and welding sequences that induce accumulation of
shrinkage deformations are to be avoided.
o o o 3 MM FROM SURFACE
300
A x * * 12 MM FROM SURFACE
A A A
19 MM FROM SURFACE
o % PLATE THICKNESS 25 MM
> it
©
260 0 o
X
0>
» &
X
S o A O & o
200 _ J
__ 1 1 1
O
i
• i
1 2 3 4 5 6 7 8
DISTANCE FROM GAS CUT EDGE (MM)
-20
TEMPERATURE (°C) Fig. 10.18: Hardness of gas cut edge of a
Fig. 10.17: Effect of welding process on COD value Mn-V-N steel
Table 10.21: Recommended preheat temperatures for V-N steels

Thickness Preheat (°C) for steels of Y.S. (kgf /mm2)
(mm) 32 35 39 42 46
10 none none none none none
10 to 20 none none none 38 66
20 to 38 none none 38 66 —
38 to 50 66 66 93 — —
50 to 75 93 120 150 — —
Electrodes recommended for ASTM A 572 grade plates arc welding such as E 70 S-3 and E 70 S-18 (see Table
are tabulated in Table 10.22. Electrode wires of medium 10.25) are also often used for welding micro-alloyed steels.
manganese type, such as EM 5K, EM 13K, EA 2 (see Table Postzveld heat treatment: Postweld heat treatment of
10.23) have proved successful for submerged-arc welding weldment fabricated from micro-alloyed steels is required
of micro-alloyed steels. Electrode wire for flux-cored arc for some code applications, but for many it is not required.
welding of alloy steels is usually of E 70 T-l type. Low- Little data are available on the effect of thermal stress-
alloy steel electrode wire such as E 70 T-4 (see Table relief on the HAZ of micro-alloyed steels. Evidence of
10.24) are frequently used in flux-cored arc welding of secondary hardening producing an embrittled structure
high strength steels. Electrode wires of gas-shielded metal- in HAZ has been observed in higher tensile steels, but no
Table 10.22: Recommended MMA consumables for ASTM steels

Plate material AWS class of consu-
mable recommended
ASTM A 572, Gr. 42,45 E 7016, E 7018
Gr. 50, 55 E 7016, E 7018
Gr. 60, 65 E 8016-C3, E8018-C3
Table 10.23: Composition of some electrode wires for submerged-arc welding

Chemical composition (%)
AWS class
Mn Mo Cu Total other
elements
EM5K1 0.06 0.90- 0.40- — 0.035 0.030 0.15 0.50
1.40 0.70 —
EM 13K1 0.07 0.90- 0.45- — 0.035 0.030 0.15 0.50
0.19 1.40 0.70
EA22 0.07- 1.00^ 0.05 0.45- 0.035 0.025 0.30
0.15 1.30 0.65
Notes: 1. Extracted from AWS A5.17-1997 "Specification for bare carbon-steel electrodes and fluxes for submerged-arc welding"
2. Extracted from AWS A5.23-1997 "Specification for bare low-alloy steel electrodes andfluxesfor submerged-arc welding"
Table 10.24: Weld-metal composition of some flux-covered arc welding electrodes

Chemical composition (%)
AWS class
C Mn Si Cr Mo V Ni Al
3
E70 T-l — 1.75 0.90 0.20* 0.30* 0.08* 0.30*
E70 T-43 1.50 0.90 0.20* 0.30* 0.08* 0.50 1.8
* These elements may be present

3. Extracted from AWS A5.20-1995 "Specification for mild-steel electrodes forflux-coredarc welding"
Table 10.25: Composition of some electrode wires for gas-shielded metal-arc welding
Chemical composition %
AWS class
Mn Ni Mo
4
E70 S-3 0.06- 0.90- 0.45- 0.025 0.035
0.15 0.15 1.4 0.70
E70 S-1B4 0.07- 1.60- 0.5- 0.025 0.035 0.15 0.4-
0.12 0.12 2.10 0.8 0.6
4. Extracted from AWS A5.18-2001 "Specification for mild-steel electrodes for gas-metal arc welding"
data are available on micro-alloyed steels. Low impact respectively, and some other alloying elements to act as
values have been reported with C 0 2 shielded welding hardening agents. Carbon, manganese and silicon were
and submerged-arc welding of C-Mn-Nb and C-Mn-V treated as undesirable residual elements and kept below
micro-alloyed steels. It has been suggested that the 0.03, 0.10 and 0.10% respectively. Ultimately, the 18%-
phenomenon is generally associated with welds of high Ni steel emerged as the most popular, since it could be
heat input. Prolonged stress-relief of micro-alloyed steel more simply heat-treated than the others to gain the
produces a deterioration in tensile properties. optimum mechanical properties. We shall therefore deal
only with this steel in detail.
Maraging Steels The 18%-Ni steel is available in three types with
These constitute a relatively new breed of high- progressively higher levels of tensile strength, which are
strength low-alloy steels introduced in 1959. The term identified as 18 Ni (200), 18 Ni (250) and 18 Ni (300).
maraging, is derived from the words martensitic and They are specified as grades A, B and C in ASTM A538.
aging; though, strictly speaking these steels are neither The fourth type designated as 18 Ni (350) is also currently
martensitic nor age-hardenable. available. Their typical chemical compositions and
The original group of maraging steels consisted of mechanical properties are given in Tables 10.26 and 10.27.
three nickel steels containing 18%, 20% and 25% Ni The different tensile strengths are obtained mainly by
Table 10.26: Nominal compositions of 18% Ni maraging steels

Nominal composition, %
3
Grade Ni Co Cr Mo Ti Al
ASTM A 538
Gr.A (200) 18 8 4 0.2 0.1
Gr.B (250) 18 8 5 0.4 0.1
Gr.C (300) 18 9 5 0.7 0.1
18 Ni (350) 18 12 4 1.3 0.1
a
Carbon—0.03% max.
Table 10.27: Typical mechanical properties of aged 18% Ni maraging steels

Tensile Yield Elongation Reduction Fracture Charpy-V
strength, strength, in 2 in., in area, toughness, notch impact
Grade 3 ksi ksi °/ o/
/o ksi in strength,
/o
'ft-lb(RT)
ASTMA538
Gr. A (200) 218 203 10 60 140-120 35
Gr. B (250) 260 247 8 55 110 20
Gr. C (300) 297 290 7 40 73 15
18 Ni (350) 355 348 6 25 32-45 8
a
Solution treated at 816°C, aged at 482°C
SOAK AT 1500 F TO
1500 h DISSOLVE ELEMENTS
1 Hr 800
WHICH FORM HARDEN
ING COMPOUNDS
1300 h 700
1100 600
"AGE" At 900 F
TO PRECIPITATE . 1 500 p
u. 900
1 to 4 Hr HARDENING
COMPOUNDS 400
700 IN THE MARTENSITIC
STRUCTURE
_J
O
300 2
500 o
o
200
-M,
300
U
100
100 MAATENSlTE MARAGED
30 Re 52 Re -JO
0
-100
100
STEP- ANNEAL MARAGE

PURPOSE SOLUTION TREATMENT PRECIPITATION
HARDENING.
Fig. 10.19: Heat treating sequence for the 18% Ni maraging steels
adjusting the levels of Co, Mo and Ti. The sequence of r a n g e , w h i c h is a p p r o x i m a t e l y 260-300°C. This
heat treatment is shown in Fig. 10.19. The steel is heated behaviour permits the aging of martensite at 482 °C for
to 816°C to dissolve hardening elements Ti and Al and several hours without transformation to austenite.
air-cooled (annealed). At the Ms temperature of about Welding metallurgy: The HAZ of a welded joint in a
200°C for 18 Ni (250) steel and 160°C for 18 Ni (300) steel, maraging steel consists of three distinct regions as
the austenite begins to transform to martensite. The follows:
transformation to martensite is completed at about 93°C. 1) The region next to the weld interface which is heated
The transformation to martensite proceeds over a wide to a sufficiently high temperature to become fully austen-
range of cooling rates. The fully martensitic steel is then itic. On cooling it transforms to coarse martensite. It is
given a maraging treatment of one to four hours at about relatively soft but regains full hardness on subsequent
482°C, to produce a highly dispersed precipitate within aging.
the matrix. This is actually a precipitation hardening * 2) A narrow region next to the above which gets heated
process and not aging. H a r d n e s s increases w i t h in the range 590-730°C and consists of martensite with
increasing maraging time. A minute study of the fine, reverted austenite. The proportion of such austenite
microstructures has shown that Mo enters directly into increases with increasing heat input. This region is soft
particles which harden the matrix by precipitation. and does not harden during subsequent aging. Special
Cobalt, on the other hand, does not contribute much to care must be taken during welding to restrict the width
h a r d e n i n g b u t only s p e e d s u p the precipitation of this region.
hardening response. Both the elements, however, 3) A region further away which is martensitic and
influence the transformation of austenite during cooling gets aged to various extents at temperatures below 590°C.
from the annealing temperature. It can be ignored for practical purposes.
It is a peculiar characteristic of maraging steels such Hydrogen-induced cracking rarely occurs in the
as 18%-Ni type, that during heating, the transformation welding of maraging steels, mainly because the low-
from alpha- to g a m m a - p h a s e occurs in a n a r r o w carbon martensite formed in the HAZ is relatively soft in
temperature zone of approximately 650-680°C. On the the as-welded condition. Hot cracking, however, is a
other hand during cooling, the reverse transformation potential problem because the low-Mn content makes
from gamma to alpha occurs at a much lower temperature these steels p a r t i c u l a r l y sensitive to s u l p h u r
embrittlement. In a joint of poor fit-up, cracking can occur contents must be carefully regulated to ensure thorough
even when S content is around 0.005%; with good fit-up, deoxidation of the weld pool (resulting in extremely low
S levels up to 0.010% can be harmless. Hot cracking is O and N contents) as well as adequate hardening of the
also ascribed to the formation of titanium sulphide. weld-metal. For TIG welding, Ti content of 0.5% is
Susceptibility to hot cracking increases with increasing adequate, but for the submerged-arc process, Ti content
Ti content in high strength maraging steels. They should must be around 2.5% to ensure porosity-free welds. The
therefore be welded with greater care. To avoid hot fabricator must choose filler metal of correct Ti and Al
cracking and lowering of mechanical strength, maraging contents depending upon the welding process to be used.
steels should be welded without preheat and the inter- C, Mn, Si and S must be kept at a low level to ensure
pass temperature should not be allowed to exceed 120°C. adequate fracture toughness of the weld-metal. For SA
Welding of 18% Ni (250) steel: The chemical welding, specially formulated flux is necessary to ensure
compositions of typical filler metals for TIG and MIG extremely low levels of these elements in the deposit.
welding of this steel are given in Table 10.28. Manual Hydrogen, even in small amounts, causes underbead
electrodes and submerged-arc flux-wire combinations cracking. Hence TIG and MIG processes which have low-
Table 10.28: Compositions for TIG and MIG wires for 18 Ni (250) steel
Elements TIG welding rod MIG electrode
Carbon 0.025 max. 0.025 max.
Manganese 0.10 max. 0.10 max.
Phosphorus 0.010 max. 0.010 max.
Sulphur 0.010 max. 0.010 max.
Silicon 0.10 max. 0.10 max.
Chromium 0.50 max. 0.50 max.
Nickel 17.5-18.5 17.5-18.5
Cobalt 9.0-10.5 9t.0-10.5
Molybdenum 4.0-5.0 4.0-5.0
Titanium 0.40-0.60 0.90-1.25
Aluminium 0.05-0.15 0.05-0.15
Zirconium 0.005 max. 0.005 max.
Calcium 0.05 max. 0.05 max.
Boron 0.003 max. 0.003 max.
Nitrogen 25 ppm max. 25 ppm max.
Oxygen 50 ppm max. 50 ppm max.
Hydrogen 5 ppm max. 5 ppm max.
giving weld deposits of similar chemistry are also potential hydrogen and vacuum-melted filler rods are to
available in some countries. The steel is sometimes be preferred.
welded in the annealed condition, and the entire welded
structure is given a maraging treatment at 482°C for one Stainless Steels
to three hours to obtain nearly full strength at the weld Stainless steels mean those steels which do not rust
joints. If the steel has been maraged prior to welding, the in air or corrode in contact with certain corrosive media.
joint area is locally subjected to the maraging treatment. The ability of stainless steels to resist corrosion is due to
The welded joint in either case has a strength level 5 to high chromium content. Chromium forms a tight, hard,
15% lower than the base metal strength. If 100% joint continuous and highly insoluble oxide film on the metal
efficiency is desired, the weld-metal must be annealed surface, w h i c h shields it from further attack by
prior to maraging. atmosphere or corrosive substances. The stable, bright
Typical mechanical properties of welded joints a n d s h i n i n g surface of stainless steels is d u e to
obtained with the most common welding processes are chromium. The Cr content varies between 11 and 27%.
given in Table 10.29. Preheating is avoided and multi- Corrosion resistance is usually proportional to the
pass welding procedures with low heat input are used. amount of chromium present: higher the chromium
High heat input results in HAZ cracking. Ti and Al content, better is the corrosion resistance. Stainless steels
3
Table 10:29:: Typical mechanical properties of welded joints in 18% Ni (250) maraging steel
UTS YS Elong. Red. of Weld-metal Charpy-V
ksi 0.2% ksi in 1 in, % area, % hardness Re at 21°C
Unwelded base metal
Annealed 0 158 122 15 63 — —
Maraged d 276 272 9 42 (52) —
Shielded metal-arc b
As-welded 160 130 5 35 35 25
Maraged d 242 225 6 26 50 10
Gas Metal Arc
As welded 154 134 — — — —
Maraged d 235 220
Submerged-arc b
As-welded 155 130 6 24 *32 33
Maraged d 235 220 6 25 49 10
b
" Typical properties with Vi in. thick plate. All welded specimens fractured in the HAZ '. "Annealed at 816°C.
d
Maraged at 482°C for 3 hours.
which do not contain Ni or contain it in insignificant The main difference between martensitic and ferritic
quantities (below 2%) are termed straight chromium steels. steel is that, the former has higher carbon while the latter
When these steels contain more than 27% Cr, they become has higher chromium. It is difficult to draw a demarcating
much too brittle for engineering applications. line for alloy contents in ferritic and martensitic steels;
The hardenability of steel increases with Cr content there is a certain amount of overlap. But it is important to
up to a point. Beyond it, Cr makes the steel incapable of know that higher carbon contents push the steels towards
being hardened, because this element is a strong ferrite the hard martensitic class and higher chromium contents
former. At such a high level of Cr, the austenitic push them towards the softer ferritic class.
transformation is suppressed and the normal quenching When enough nickel is present in a straight chromium
does not cause the steel to harden. For steels containing steel, it provides an essentially austenitic matrix at all
14 to 17% Cr, increase of carbon gives them the ability to temperatures including room temperature. This is
be hardened by heating and quenching. Steels with more because nickel is a powerful austenitiser. The steel is
than 18-20% Cr cannot be hardened, however high is termed chromium-nickel austenitic stainless steel or
their carbon content. simply austenitic stainless steel. This type is moderately
Thus, metallurgically speaking, there are two distinct strong but extremely ductile and tough. It is the easiest
classes of straight chromium steels, martensitic and stainless steel to weld. The limits of C, Cr and Ni for this
ferritic, according to their microstructure. The two classes class are indicated in Table 10.30, which also gives the
differ widely in their physical properties and weldability. composition ranges of martensitic and ferritic stainless
Normally, the martensitic type is hard and brittle and steels. AISI classification system and detailed chemical
difficult to weld. The ferritic type is soft and ductile and analyses of these steels are given in Table 10.31.
comparatively easier to weld. But welding can change
the properties in the HAZ and so a careful procedure of Martensitic Steels
welding m u s t be applied. These two classes are The internationally recognised standard martensitic
distinguished from each other by their chemical stainless steels are AISI types 403,410,414,416,416(Se),
composition, in which the key elements are chromium 420, 431, 440A, 440B and 440C, in which the Cr content
and carbon. ranges from 11.5 to 18% (see Table 10.31). The last three
Table 10.30: Compositional limits of the three stainless steel types

Classification Carbon Chromium Nickel
Martensitic 1.0 max. 11.0-18.0 Residual
Ferritic 0.20 max. 14.0-30.0 Residual
Austenitic 0.25 max. 14.0-30.0 6.0-36.0
Table 10.31:: AISI stainless steel classification system

Chemical analyses of stainless steels,, %
AISI No. Carbon Manganese Silicon Chromium Nickel Other elements
Chromium -nickel manganese--austenitic-non- hardenable

201 0.15 max. 5.5/57 1.0 16.0/18.0 3.5/5.5 N 2 0.25 max.
202 0.15 max. 7.5/10 1.0 17.0/19.0 4.0/6.0 N 2 0.25 max.
Chromium-nickel-austenitic-non-hardenable
301 0.15 max. 2.0 1.0 16.0/18.0 6.0/8.0 —
302 0.15 max. 2.0 1.0 17.0/19.0 8.0/10.0 —
392B 0.15 max. 2.0 2.0/3.0 17.0/19.0 8.0/10.0 —
303 0.15 max. 2.0 1.0 17.0/19.0 8.0/10.0 S 0.15 min.
303Se 0.15 max. 2.0 1.0 17.0/19.0 8.0/10.0 Se 0.15 min.
304 0.08 max. 2.0 1.0 18.0/20.0 8.0/12.0 —
304L 0.03 max. 2.0 1.0 18.0/20.0 8.0/12.0 —
305 0.12 max. 2.0 1.0 17.0/19.0 10.0/13.0 —
308 0.08 max. 2.0 1.0 19.0/21.0 10.0/12.0 —
309 0.20 max. 2.0 1.0 22.0/24.0 12.0/15.0 —
309S 0.08 max. 2.0 1.0 22.0/24.0 12.0/15.0 —
310 0.25 max. 2.0 1.50 24.0/26.0 19.0/22.0 —
310S 0.08 max. 2.0 1.50 24.0/26.0 19.0/22.0 —
314 0.25 max. 2.0 1.5/3.0 23.0/26.0 19.0/22.0 —
316 0.08 max. 2.0 1.0 16.0/18.0 10.0/14.0 Mo 2.0/3.0
316L 0.03 max. 2.0 1.0 16.0/18.0 10.0/14.0 Mo 2.0/3.0
317 0.08 max. 2.0 1.0 18.0/20.0 11.0/15.0 Mo 3.0/4.0
321 0.08 max. 2.0 1.0 17.0/19.0 9.0/12.0 Ti 5 x C min.
347 0.08 max. 2.0 1.0 17.0/19.0 9.0/13.0 Cb+Ta 10XC min.
348 0.08 max. 2.0 1.0 17.0/19.0 9.0/13.0 Ta 0.10 max.
Chromium-martensitic-hardenable
403 0.15 max. 1.0 0.5 11.5/13.0 — —
410 0.15 max. 1.0 1.0 11.5/13.5 — —
414 0.15 max. 1.0 1.0 11.5/13.5 1.25/2.5 —
416 0.15 max. 1.25 1.0 12.0/14.0 — S 0.15 min.
416Se 0.15 max. 1.25 1.0 12.0/14.0 — Se 0.15 min.
420 Over 0.15 1.0 1.0 12.0/14.0 — —
431 0.20 max. 1.0 1.0 15.0/17.0 1.25/2.5 —
440A 0.60/0.75 1.0 1.0 16.0/18.0 — Mo 0.75 max.
440B 0.75/0.95 1.0 1.0 16.0/18.0 — Mo 0.75 max.
440C 0.95/1.2 1.0 1.0 16.0/18.0 — Mo 0.75 max.
•
Chromium-ferritic-non-hardenable
405 0.08 max. 1.0 1.0 11.5/14.5 — Al 0.1/0.3
430 0.12 max. 1.0 1.0 14.0/18.0 — —
430F 0.12 max. 1.25 1.0 14.0/18.0 — S 0.15 min.
430FSe 0.12 max. 1.25 1.0 14.0/18.0 — Se 0.15 min.
446 0.20 max. 1.50 1.0 23.0/27.0 N 0.25 max.
440 types are rarely welded. The standard steels are more susceptible to cracking. Cracking is avoided by
sometimes modified by adding elements like Ni, Mo, V preheating and controlling interpass temperatures. If the
and W, primarily to raise the permissible service steel contains less than 0.10% carbon, preheating can be
temperature above the 593°C limit for the standard steels. avoided. For steels with higher carbon the preheating
The carbon content is also increased in these temperature depends not only on carbon content but also
modifications, with the result that they are more prone to on the mass of the joint, degree of restraint, presence of
cracking in the hardened HAZ of weldments than the notch effect and the chemical composition of the filler
standard counterparts. metal. Postweld heat treatment is also sometimes required
The standard steels can be welded in the annealed, to obtain optimum properties.
hardened or hardened and tempered conditions. The relationship between preheat and postheat
Whatever be the prior condition, the steel when welded temperature must be clearly understood, as it has a
produces a hard martensitic zone in the HAZ, the significant influence on the resulting weld and HAZ
hardness increasing with carbon content. As hardness structure. This relationship is schematically illustrated
increases, toughness decreases and the HAZ becomes in Fig. 10.20.
EXAMPLE A EXAMPLE B
NO PREHEAT PREHEAT
NO POSTHEAT NO POSTHEAT
TIME TIME
EXAMPLE C EXAMPLE D
PREHEAT PREHEAT TOO HIGH
6*°\ COOL BELOW Mf DID NOT COOL BELOW Mf
| \ POSTHEAT POSTHEAT NOT EFFECTIVE
"T—"\"TfiTT—Al
LU
TIME TIME
LEGEND
EXAMPLE E A, CRITICAL TEMP.
PREHEAT (ABOUT 1450°F) -788°C
DID NOT COOL BELOW Mf Mf MARTENSITE FORMATION
POSTHEAT NOT EFFECTIVE TEMP. (ABOUT 600°F) -316°Q
FERRITE. CARBIDES
OOOAUSTENITE
•••MARTENSITE
-H-TEMPERED MARTENSITE
TIME
Fig. 10.20: Relationship between pre- and postheating and metallurgical structure of air-hardening welds
Example A shows the thermal cycle of the weld and welding. The temperature ranges for subcritical and full
the HAZ in a hardenable Cr steel welded without annealing for the standard martensitic steels are given
preheating. During welding, the weld-metal and adjacent in Table 10.32.
base metal are transformed to austenite. On cooling below
Mf temperature, the austenite transforms to martensite. Table 10.32: Annealing temperatures for martensitic
The final result is a highly-stressed, hard, untempered stainless steels
martensitic weld and HAZ which are highly susceptible AISI type Subcritical Full
to cracking. annealing, °C annealing, °C
Example B shows the thermal cycle when preheating
403, 410, 416 732-788 843-871
below Mf temperature is used. Upon cooling the metal
420 748-770 871-899
transforms to martensite exactly as at A, but the final 414 —
662-704
cooling to room temperature occurs at a lower rate. In 431 621-662 —
this case, the shrinkage stresses may be lower, but the 440 (A, B & C) 732-788 899
weld and HAZ structure is still hard, brittle, untempered
martensite. The only standard martensitic stainless steel filler
In example C, preheat temperature below Mf is used, metals available are E or ER 410 and 420. This limitation
so that the weld-metal and HAZ are transformed to must be borne in mind when attempting to obtain
martensite on cooling to the preheat temperature. When matching properties in the weld-metal and base metal.
transformation to martensite is complete, a postweld heat The weld deposits of 410 and 420 filler metals have low
treatment is applied to temper the martensite, so that its toughness and must be correctly heat treated to improve
ductility and toughness improve. toughness before the weldment is placed in service.
In example D, the preheat temperature is above Mf so Austenitic stainless steel filler metals are often preferred,
that when the weldment cools to the preheat temperature, because they provide an as-welded deposit of greater
no martensitic structure is formed. The final postheat toughness but of somewhat lower strength than the
tempering treatment has therefore no effect on the martensitic types.
structure, which upon final cooling to room temperature 403, 410, 416, 416(Se) steels: E/ER410 filler metal is
is hard, untempered martensite. preferred for welding these steels. The weldments are
In example E, the preheat temperature is correctly annealed or hardened and stress-relieved before use.
below Mf but the postweld heat treatment is commenced Annealing softens and imparts ductility equally to the
before the weld-metal has cooled down to the preheat HAZ and the weld, since both respond to heat treatment
temperature. Since the martensitic structure has not been in a similar manner. 403 and 410 steels are sometimes
allowed to form, the postweld tempering treatment is not welded using the austenitic filler metal types 308, 309
effective. The final structure is identical to that obtained and 310, while 416 and 416 (Se) are welded with types
inD. 308,309 and 312. Such weldments are used for service in
Thus example C depicts the correct procedure. The the as-welded condition. In these cases, the weld-metal
postweld heat treatments represented in examples C, D, is soft and ductile but the HAZ has limited ductility.
E are known as subcritical annealing treatments, since 420 steel: E/ ER 420 filler metal is preferred for welding
their temperatures are below A1 and the weldment is annealed or hardened and stress-
For steels with the carbon range of 0.10-0.20%, it is relieved before use. Great care has to be taken during
advisable to use preheat and interpass temperature of preheating and postweld heat-treatment to avoid cracking.
260° C and to cool the weldment slowly after welding is 431 steel: It is welded with E/ER 410 type filler metal
completed. For steels with carbon content between 0.21 or with either of 308, 309 and 310 austenitic type filler
and 0.50%, the same preheat and interpass temperature metals. In the former case, it is used in the annealed or
are used and the weldment is annealed after welding. hardened and stress-relieved condition, and in the latter
For steels with carbon content more than 0.50%, the same case it is used in the as-welded condition. In this case
heat treatments are used together with high heat input also, preheating and postheating must be correctly and
during welding. carefully executed.
Annealing should be done immediately after welding
w i t h o u t allowing the w e l d m e n t to cool to room Ferritic Stainless Steels
temperature. Annealing can be omitted if the weldment The standard types are: 405 and 409 (13% Cr); 430,
is to be hardened and tempered immediately after 430 (F) and 430 F(Se) (17% Cr): 446 (25% Cr) [see Table
10.31]. Type 409 is ferritic, because its carbon content is and mechanical properties desired. Postheating or post-
less than 0.08% and its minimum Ti content is equal to weld annealing, if required, is carried out in the
six times the carbon content. Ti is a strong ferrite former. temperature range of 790-840°C This range is much
Type 405 also contains only 0.08% maximum carbon and below the temperatures for austenite formation and grain
approximately 0.20% Al, which is also a strong ferrite coarsening. Cooling from the annealing temperature is
former. done in air or by water quenching. It is advisable to remove
It w a s stated earlier that ferritic steels h a v e weldments from the furnace after they have cooled down
compositions which completely suppress the austenite to approximately 593°C, in order to avoid possible
transformation and ensure a completely ferritic structure distortion caused by handling. However, cooling in the
at room temperature. This is not practically so. Variations temperature range of 566-400°C must be fairly rapid, as
in compositions within the standard limits can result in otherwise the steel becomes brittle. Such brittleness is
the formation of small amounts of austenite during termed 885 F brittleness in the U.S.A., and it increases
heating to temperatures nearer above 870°C, which with increasing Cr-content of the steel. In the case of heavy
occurs during welding. On cooling, this austenite sections, spray quenching may be necessary to bring
transforms to martensite, thus conferring on the metal them safely through this temperature range.
the duplex structure of ferrite and some martensite. The Both ferritic and austenitic stainless steel filler metals
martensite reduces both the ductility and toughness. This are used for welding these steels, but the selection must
can happen with the HAZ of ferritic steel weldments. be made with care, depending on the specific application.
The martensite can be eliminated and normal ferritic Ferritic filler metals are less ductile in the as-welded
properties restored by annealing, but annealing is a costly condition, but have the advantages of matching colour
operation and can result in considerable distortion. and appearance, matching coefficient of thermal
In the region of the H A Z which is h e a t e d to expansion and matching corrosion resistance. Austenitic
temperatures above 954°C during welding, enlargement filler metals give welded joints of high ductility, but they
of ferrite grain size occurs and this also results in reduced cannot p r e v e n t m a r t e n s i t e formation and grain
ductility and toughness. Although martensite can be coarsening in the HAZ. Also austenitic weld-metal, unless
eliminated by annealing, coarsened ferrite grain remain it is of extra-low carbon content or is stabilised with Cb,
unaffected by this treatment. Since martensite inhibits Ta or Ti, suffers a loss of corrosion resistance as a result
ferrite grain growth, martensite formation is encouraged of annealing, since the normal range of annealing
in cases where the workpiece can be annealed after temperature falls within the sensitising temperature range
welding. For example, the Cr content of a 17%-Cr steel is for austenitic steels. The phenomenon of sensitisation of
sometimes lowered to 15 or 16%, so that some amount of austenitic weld-metal will be explained later. Further,
martensite is formed which is subsequently removed by annealing treatment is also expected to relieve residual
postweld annealing. But in this case, preheat may have stresses in the weidmeiit, but this cannot be fully realised
to be used before welding, or the carbon content of the because of the difference in the coefficients of thermal
steel lowered in order to prevent cracking in the HAZ. expansion of the austenitic weld-metal and the ferritic
In cases where annealing is not possible, it is advisable base metal.
to select a ferritic steel containing sufficient amount of a It must be remembered that the corrosion resistance
strong ferrite former such as Al, Cb or Ti. An example is of ferritic stainless steels is generally less than that of
the steel type 430 (Ti), which is a 17%-Cr steel containing austenitic stainless steels and besides, the presence of
0.12% maximum carbon and minimum Ti content equal martensite or the precipitation of sigma-phase at the grain
to six times the carbon content. Ti forms stable Ti carbides boundaries can give rise to severe intergranular corrosion
and promotes ferrite formation. When such a steel is in the HAZ. Completely ferritic steels such as the type
welded, martensite is totally absent in the HAZ but grain 430 (Ti) are not prone to intergranular attack at the
coarsening cannot be avoided. Grain coarsening is welded joint. The phenomena of sigma-phase formation
controlled to some extent by using minimum heat input and intergranular corrosion are explained later. For
during welding, and ensuring rapid cooling from the maximum corrosion resistance, the welded steels must
welding temperature. be used in the annealed condition.
For w e l d i n g ferritic steels, p r e h e a t i n g in the 405 steel: This steel is welded either with a ferritic
temperature range of 150 to 230°C is recommended. The filler metal such as E/ ER 405 (Cb) or 430, or an austenitic
need for preheating and the preheat temperatures depend filler metal such as E/ER 308, 309, or 310. In the former
on composition, joint thickness, restraint, heat input rate, case, the weldment is used in the as-welded condition,
while in the latter case it is used in the annealed condition, nickel to provide an essentially austenitic matrix at all
as annealing increases ductility of the HAZs and the temperatures including the room temperature. The
weld-metal. In type 405 weld-metal, Cb is used in place commonly available types in the wrought form are
of Al to reduce hardening. Austenitic weld-metals &re indicated in Table 10.31 along with their AISI type
ductile in the as-welded condition but the HAZ has numbers and compositions.
limited ductility. The tensile strength of these steels in annealed condi-
430 steel: In this case E/ER 430 ferritic type or either of tion and in wrought sheet form ranges between 585 and
the E/ER 308, 309 and 310 austenitic types is used as 655 N / m m 2 , and yield strength in terms of 0.2% offset
filler metal. The other comments on 405 steel also apply between 240 and 310 N/mm 2 . They are extremely ductile
to this steel. and give elongation values of around 45-50%. In addi-
430 F and 430 F(Se) steels: They are welded with E/ER tion to extraordinary corrosion resistance, they exhibit
430 or either of the E/ER 308, 309 and 312. The other excellent creep and oxidation resistance at elevated tem-
comments on 405 steel also apply here. peratures. They also give high impact strength and duc-
446 steel: This steel is welded with E/ER 446 or with tile fractures at subzero temperatures down to -196°C.
either of E/ER 308,309 and 310. It must be borne in mind These facts explain the wide use of austenitic stainless
that the scaling resistance of 308 filler metal is not equal steels in critical applications involving corrosive media
to that of the base metal. and extremes of service temperature.
While the principal alloying elements of these steels
Austenitic Stainless Steels are Fe, Cr and Ni, they also carry deliberate additions of
As mentioned earlier, these steels contain enough other elements. Some of them are meant to augment the
chromium to prevent them from rusting in air or corroding austenite stabilising action of Ni, while others are meant
in contact with certain corrosive media, and enough to promote the formation of delta-ferrite in the structure
Table 10.33: Effects of alloying elements in austenitic stainless steels

Types
Element of steels Effects
Carbon All types Strongly promotes the formation of austenite. Can form a carbide with chromium that
can lead to intergranular corrosion.
Chromium All types Promotes formation of ferrite. Increases resistance to oxidation and corrosion.
Nickel All types Promotes formation of austenite. Increases high temperature strength, corrosion
resistance, and ductility.
Nitrogen XXXN ' Very strong austenite former. Like carbon, nitrogen in 30 times as effective as nickel in
forming austenite. Increases strength.
Columbium 347 Primarily added to combine with carbon to reduce susceptibility to intergranular
corrosion. Acts as a grain refiner. Promotes the formation of ferrite. Improves creep
strength.
Manganese 2XX Promotes the stability of austenite at or near room temperature but forms ferrite at
high temperatures. Inhibits hot shortness by forming MnS.
Molybdenum 316, 317 Improves strength at high temperatures. Improves corrosion resistance to reducing
media. Promotes the formation of ferrite.
Phosphorus, 303, 303Se Increase machinability, but promotes hot cracking during welding. Lowers corrosion
selenium, or resistance slightly.
1
sulphur
Silicon 302B Increases resistance to scaling and promotes the formation of ferrite. Small amounts are
added to all grade for deoxidising purposes.
Titanium 321 Primarily added to combine with carbon to reduce susceptibility to intergranular
corrosion. Acts as a grain refiner. Promotes the formation of ferrite.
Copper CN-7M Generally a d d e d to stainless steels to increase corrosion resistance to certain
environments. Decreases susceptibility to stress-corrosion cracking and provides age-
hardening effects.
and to be carbide formers. These austenite and ferrite suffers intercrystalline corrosion. The phenomenon is
forming elements are suitably balanced according to the sometimes termed weld decay.
application needs of these alloys. The effects of these It is extremely important to prevent or get rid of carbide
elements are summarised in Table 10.33. precipitation in austenitic stainless steel weldments if
Carbide precipitation: When austenitic stainless steels they are meant to be used in corrosive media. This is
are heated or cooled slowly through the temperature achieved by any of the following three methods: (1)
range of 425-870°C, the carbon precipitates from the solid solution heat treatment, (2) use of extra-low carbon grade,
solution mainly at the grain boundaries, where it com- and (3) use of a grade stabilised with columbium plus
bines intergranularly with chromium to form chromium- tantalum or with titanium.
Fig. 10.21: Intercrystalline corrosion of austenitic stainless steel weldment
rich carbide. This is termed carbide precipitation. As a 1) Solution heat treatment consists of heating the entire
result, chromium in the adjacent matrix material gets de- weldment to the temperature range of 1,025-1120°C, to
pleted and the steel suffers localised intergranular at- put the carbides back into solution and cooling it rapidly
tack when it is exposed to certain corrosive service con- by water quenching. This method is seldom used because
ditions. The severity of such attack depends on the time of the following unfavourable factors:
and temperature of exposure, as also on the composition a) At the high temperature used, the steel oxidises
and heat-treated condition of the steel. rapidly and forms an adherent surface scale, unless it is
The temperature range of 425-870°C is known as the protected from air.
sensitising temperature range, and the steel which has b) The weldment may sag or be severely distorted at
undergone carbide precipitation is said to have become the high temperature used or during rapid cooling from
sensitised. The resulting corrosive attack is referred to as it.
intercrystalline corrosion. c) Water quenching of a large workpiece from the high
During welding, a narrow band in the heat-affected temperature can be hazardous.
zone of the base metal gets heated within the temperature d) The workpiece is often too large for the furnace and
range of 425-870°C for a sufficient time to undergo cooling facilities available in the shop.
carbide precipitation. This can also h a p p e n to the 2) Extra-low carbon (ELC) stainless steels: By restricting
previously deposited weld bead as shown in Fig. 10.21. the carbon content to 0.03% maximum, the steels are
The amount of precipitation increases with the carbon rendered sufficiently immune to carbide precipitation in
content of the material. During service under corrosive the sensitising temperature range of 425-870°C. The
conditions, this band gets preferentially attacked and steels can then be safely welded and even stress-relieved
if necessary, without loss of corrosion resistance. susceptibility to hot cracking. Weld-metal, which has a
Examples are AISI304L and 316L types. However, if these completely austenitic microstructure is highly prone to
steels remain in the sensitising temperature range for an this defect, while the weld-metal with a duplex structure
extended period in service, carbides can precipitate in of delta ferrite in austenite is fairly resistant to it.
significant quantities to cause intergranular corrosion. As regards weld-metal composition, susceptibility to
These steels are usually employed for service below 425° C. microfissuring can be reduced by a small increase in
The filler metals used for welding should also be ELC types. carbon or nitrogen. Trace elements which encourage
3) Stabilised steels: These steels contain sufficient micro-fissuring, in order of their potency, are: boron,
amount of columbium plus tantalum (Cb+Ta) or titanium phosphorus, sulphur, selenium, silicon, columbium and
(Ti) in relation to carbon. These elements have greater tantalum.
affinity for carbon than chromium and hence carbides of Since the fully austenitic weld-metal is most suscep-
these elements precipitate out in the sensitizing tible to microfissuring during its cooling from the soli-
temperature range, leaving chromium intact. These steels dus down to approximately 980°C, the amount of stress
are termed stabilised steels, and they exhibit higher imposed on the weld-metal must be minimised in this
strength at elevated temperature as compared to ELC temperature range. Peening is not effective, because of
steels. They are therefore preferred for service in a the difficulty of applying it at the high temperature range.
corrosive environment in the sensitising temperature
range of 125-870°C. The filler metals used for welding Schaeff ler Diagram
should be stabilised types. It has been found that microfissuring does not occur
It must be borne in mind that Ti cannot be transferred when the austenitic weld-metal has about 3 to 5% ferrite
across the arc in manual metal-arc and submerged-arc in its microstructure. The practice generally followed is,
processes, because it gets oxidised in the arc. Hence in therefore, to select the filler metal composition in such a
these, processes, filler metals stablised with Cb+Ta are way that the deposited weld-metal has a duplex structure
used to deposit weld-metal stabilised with Cb plus Ta. It of austenite-ferrite. To determine the amount of ferrite
is only in TIG and MIG processes employing inert gas that will be formed in the austenitic weld-metal of a given
shielding, that Ti-stabilised weld can be deposited from composition, use is made of the Schaeffler diagram shown
Ti bearing filler metal to match Ti-stabilised base metal. in Fig. 10.22. In this diagram, the nickel equivalent is
Even in stabilised steels, sensitisation (i.e. plotted on the ordinate or vertical axis, and the chromium
precipitation of chromium carbide) occurs under certain equivalent on the abscissa or horizontal axis. The nickel
conditions in narrow knife-like zones immediately equivalent is calculated by using the formula given in
adjacent to the weld fusion line, not during welding but the figure, which uses the weight percentages of the
during subsequent reheating to or service at about 650°C common austenite stabilising elements nickel, carbon and
These zones exhibit less than normal corrosion resistance manganese. The chromium equivalent is similarly
and suffer service corrosion in some media. calculated using the formula given, which is derived from
4) Hot cracking: Austenitic stainless steel weld-metal the weight percentages of common ferrite stabilisers
is susceptible to minute intergranular cracking called chromium, molybdenum, silicon and columbium. The
microfissuring, which is not visible to the naked eye. It is series of dark lines in the diagram demarcate the zones
also called hot cracking, because cracking occurs at high of weld-metal micros true tures. The figure also indicates
temperature as the weld-metal cools from the solidus the positions of various standard weld-metals in the
down to about 980°C. If the level of stress imposed on the diagram. For example, E 310 weld-metal is indicated as
contracting weld-metal is high, the fissures enlarge into being fully austenitic, while E308 weld-metal is shown
visible cracks. The susceptibility to hot cracking is to be having a duplex structure with 5-7% ferrite. The
d e t e r m i n e d by: (a) the a s - d e p o s i t e d w e l d - m e t a l actual microstructures of these two deposits as seen under
m i c r o s t r u c t u r e ; (b) the w e l d - m e t a l composition, magnification are shown in Figs. 10.23 and 10.24.
especially the content of certain residual or trace elements
such as b o r o n , p h o s p h o r u s , s u l p h u r , selenium, Delong Diagram
columbium and tantalum; (c) the amount of stress A major drawback of the Schaeffler diagram is that it
imposed on the weld during cooling; and (d) the presence ignores the significant influence that nitrogen has on the
of notches. phase balance of austenitic stainless steels, since this
The first of these factors, namely the weld-metal element is a strong austenitiser. Many heats of austenitic
m i c r o s t r u c t u r e d e t e r m i n e s to a large extent the stainless steels have been found to contain appreciable
fo30
421-
40|- APPROXIMATE BOUNDARY OF
38[ AUSTENITE REGION FOR WROUGHT
361- MATERIALS-
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40
CHROMIUM EQUIVALENT (% Cr + % Mo + 15 x % Si + 0.5 x % Cb)
Fig. 10.22: The Schaeffler diagram for estimating the microstructure of stainless steel weld-metal
■ / * -
■/■ ■■ ■'■■
Fig. 10.23: Microstructure of E 310 fully

austenitic weld-metal
amounts of nitrogen. Also samples of weld-metal have Fjg 1 0 2 4 : Microstructure of E 308 weld-metal with pools
often registered either high or low nitrogen, because of 0f delta-ferrite
excessive arc length or improper welding technique. The
revised constitution diagram is k n o w n as Delong aries. As a result, the low-melting point constituents
diagram and is shown in Fig. 10.25. While Schaeffler which promote microfissuring get distributed over a
used the metallographic method for determining the ferrite larger area and thereby the likelihood of fissures in the
percentage in weld-metal, Delong has used a magnetic weld-metal is minimised.
measuring instrument in his work. Accepting that this is Different weld-metals require different levels of ferrite
a secondary measurement system, Delong has adopted to eliminate microfissuring. For E 16-8-2, E 308, E 308L,
an arbitrary standard called ferrite number (FN), to E 316 and E 316L, at least 3 FN is required. For E 309, it is
indicate the ferrite content. The ferrite number and the 4 FN; for E 318 it is 5 FN and for E 347 it is 6 FN. A high
ferrite percentage are identical at low levels of ferrite, but ferrite level is required in E 347 weld-metal to counteract
they differ widely when the ferrite levels are high. the cracking tendency promoted by a low-melting eutectic
To understand the resistance of the duplex weld-metal phase formed by columbium.
structure to microfissuring, one must trace the weld It is difficult to control Assuring in E 310 Cb weld-
solidification process. D e p e n d i n g on the relative metal because of the presence of columbium, and the fact
proportions of ferrite- forming and austenite-forming that it is not possible to obtain ferrite in the weld-metal of
elements present, the weld-metal solidifies initially as this composition. The best one can do is to restrict the
grains of delta-ferrite or as grains of austenite. As contents of Si, S and P. Like E 310 Cb, E 310, E 310 (Mo)
solidification proceeds, initial ferrite grains undergo and E 330 weld-metals also invariably deposit a fully
solid-state transformation to austenite grains, mainly in austenitic weld-metal.
the grain centres. If sufficient ferrite forming elements are The presence of ferrite lowers the corrosion resistance
present, residual pools of delta-ferrite remain in the of austenitic steel weld-metal to a marginal extent, which
microstructure. The remaining weld-metal and the grains can be ignored in most applications. There are, however,
that have solidified initially as austenite, remain fully important exceptions. For example, 18 Cr-12 Ni-Mo
austenitic down to room temperature. weld-metals such as types E 316, E 317, and E 318 suffer
When the final transformation provides a duplex localised corrosion attack on the ferrite in certain media
structure, it greatly increases the number of phase bound- in the as-deposited condition. The corrosion can be
21
5 20
X
in 19 h
o
+
z 18 h AUSTENITE
FERRITE NUMBER
#
x
o 17
CO APPROXIMATE FERRITE
+ CONTENT BY MAGNETIC
16 ANALYSIS
15 SCHAEFFLER
A + M LINE
£ ,4
< 13
> AUSTENITE
+ MARTENSJtt j
2 12 h
/
y n |~ /
10 K \y \, \S 1/1/ / A Sx ,
16 17 18 19 20
CHROMIUM EQUIVALENT (% Cr + % Mo + 1.5 x % Si + 0.5 x % Cb)
Fig. 10.25: The Delong constitution diagram for auttonltic stainless steel weld-metal
prevented by either annealing the joint or using ferrite- poses the danger of grain growth and consequent
free weld-metal. In the fabrication of urea plant, E 316 changes in mechanical properties.
type weld-metal with ferrite content of 2% maximum is Underbead cracking: There is another form of cracking
specified for the components which come into contact associated with stainless steel weldments. It is termed
with the extremely corrosive carbamate solution. underbead cracking, because it occurs in the heat-affected
It must be brone in mind that while the fully austenitic zone at the edge of the weld bead, when the base metal
weld-metal is non-magnetic, the presence of ferrite makes thickness is more than 19 mm. Type 347 stainless steel
it weakly or strongly magnetic, depending on the level of which is stabilised with Cb plus Ta is particularly prone
ferrite. If an application requires that the weld-metal be to this type of cracking, while other common stainless
perfectly non-magnetic, filler metal giving wholly steels with the exception of type 316L are less susceptible.
austenitic weld deposit must be used. Pressure vessels made out of 347 steel have revealed
Many experts hold the view that microfissuring in failures in welded joints after having been subjected to
fully austenitic welds can be ignored, because the great elevated temperature service for a long time.
bulk of fissures seem to be small, below 1.5 mm in Apparently these joints were sound when tested
maximum dimension. In a very notch tough material initially. The failures are interpreted thus: the heavy
such as austenitic stainless steel, it would require very shrinkage stresses in the welded joints cause strain-
unusual service conditions for such small defects to induced precipitation of columbium carbide, which gives
adversely affect the service life of the structure. Here one rise to underbead cracking. This cracking ultimately leads
can pertinently quote Dr. Delong from his 1974 Adams to the major failure in service. Underbead cracking can
Lecture delivered to the American Welding Society. "From be avoided or at least minimised by keeping the weld
practical viewpoint, millions of pounds of multipass fully restraint to a minimum.
austenitic weld-metal of types 310,316,316 L, and more
special types such as 320 and 330 have been used in WRC-1992 Diagram
production weldments over the past 40 years with The Cr-equivalent and Ni-equivalent constitutional
virtually no failures attributable to fissures, yet virtually diagram proposed by W. T. Delong was modified by
all of these weldments do contain fissures". The latest Welding Research Council (WRC) & Commission II of
thinking appears to be that in the overall interest of the the International Institute of Welding (IIW) to increase
chemical plant having extremely corrosive service the scope and accuracy in prediction of ferrite number
condition, one must prefer fully austenitic welds with (FN) in stainless steel weld metal and dissimilar metal
some microfissuring to perfect defect-free welds having joints. After extensive studies on measured and actual
about 5% ferrite. weld metal ferrite, WRC-1988 diagram was proposed.
When weldments having duplex weld-metal The predictions of this constitutional diagram are
microstructure are exposed to service temperatures in independent of silicon and manganese contents as the
the range of 480-900°C, various intermetallic phases are effect of these elements were found to be statistically
formed in varying amounts depending on the chemical insignificant. Later to incorporate the effect of copper in
composition, the ferrite content and the time at the Ni-equivalent (especially in case of duplex stainless
temperature. With increasing ferrite content, the steel weld metals), WRC-1988 diagram was re-defined
transformation is accelerated and the temperature at by WRC & Commission II of IIW as WRC-1992 diagram
which it occurs is lowered. The transformation occurring (Fig. 10.25a.).
in non-molybdenum bearing stainless steels is termed This constitution diagram offers more accurate FN
sigma-phase, and those occurring in Mo-bearing steels prediction in stainless steel welds and dissimilar metal
are referred to as sigma- and chi-phase transformations. joints. Also like Schaeffler diagram, the axes of the
Nucleation of these phases occurs at the austenite-ferrite WRC-1992 diagram can be extended to predict the
boundaries and growth is primarily towards the centre dilution effects in dissimilar joints and cladding
of the ferrite. The presence of these intermetallics is viewed applications.
seriously, because they lower ductility and impact
resistance of the weldment. Ductility lost by sigma-phase Measurement of Ferrite
formation can be restored to a large extent by heating the While the various Cr-equivalent & Ni-equivalent
steel at 1,038°C for a short time. For converting sigma diagrams help one to estimate ferrite, fabricators and users
completely to austenite, heating to a temperature as high of chemical plant must have a reliable method to measure
as 1,230°C is required. Heating to such a high temperature the ferrite in actual production welds. International
standard blocks, which are supplied in a box by the

American Welding Society or the Welding Institute of
Abington, England. They are prepared on behalf of
Comm. II "Arc Welding" of the IIW. The box contains
eight samples in the approximate range of 3 to 27 FN.
The special samples have been manufactured by
VEW-Bohler in Austria from Cr-Ni austenitic stainless
steel weld-metals deposited using the submerged-arc strip
cladding process. This method is employed as it has been
shown to give a relatively uniform ferrite distribution
over the length and breadth of the weld bead, by
t$ 20 22 24 ?* 28 W comparison with other welding processes. A multi-layer
C f M « Cf * M»*0.7MI> deposit was used to remove any possible influence from
Fig. 10.25a. WRC-1992 Diagram the base metal on which the weld-metal was deposited.
Welding procedure: Ferrite content as claimed by the
electrode producer in his batch report can be reproduced
Institute of Welding has been doing considerable work by the fabricator on production welds, only when correct
on this issue. welding procedure is adopted. The most important point
Metallographic measurements are helpful but they is to control heat input by using smaller size electrodes
are difficult to run accurately, owing primarily to the and stringer beads, thus allowing the weld-metal to cool
very small size of the ferrite particles. This fineness, plus down rapidly after deposition. Interpass temperature also
the inherently different corrosion responses of ferrites of must be kept on the low side. An inexperienced fabricator
varying analyses to different etching media, make an whose previous experience is restricted to mild steel,
optimum etch difficult to establish and an accurate makes the mistake of allowing his welders to use very
quantitative result difficult to obtain. The International high currents and long arc. The high currents coupled
Institute of Welding has, however, issued Doc. IIWî82- with slow speeds of arc travel overheat the weld deposit.
75 entitled "Recommended method for the metallographic The result is much higher levels of ferrite and consequent
determination of delta-ferrite in Cr-Ni austenitic weld- loss of corrosion resistance.
metal by means of normal optical microscopy and visual A long arc and inadequate shielding causes ingress
comparison with an atlas". Experts are of the opinion of nitrogen into the weld-metal. Since this element is a
that this metallographic method can be usefully adopted strong austenitiser, the resultant weld contains lesser
by a laboratory for its internal checks, but it has limitations amount of ferrite than what is to be expected.
when used as a basis for comparison and cross-checking.
At the present time, experimental methods are not Stress Corrosion Cracking (SCC)
available that give an absolute measurement of the A typical corrosion phenomenon occurs when weld
amount of ferrite in a weld-metal, either destructively or metals having residual tensile stresses are exposed to
non-destructively; hence the d e v e l o p m e n t a n d certain environment. This is known as Stress Corrosion
international use of the concept of a ferrite number (FN) Cracking (SCC). The presence of either of the stress or the
referred to earlier. The standardised procedure for environment does not cause SCC.
determining the ferrite content of a weld-metal is laid The major determinants of occurrence of SCC are, a)
d o w n in IIW document 11-904-79 "Recommended the metal and the alloy employed and its microstructural
standard method for the determination of the ferrite condition, b) the exposed environment of the weld metal
number in austenitic weld-metal deposited by Cr-Ni steel and c) the level and distribution of tensile stresses. Among
electrodes" and AWS Standard A4.2-74 "Standard the other factors; temperature, pressure, hydrogen ion
procedures for calibrating magnetic instruments to concentration (pH), cyclic pressure, fluctuating humidity,
measure the delta-ferrite content of austenitic stainless circulation and viscosity of contacting fluids and degree
steel weld-metal". of aeration are also important.
One of the presently accepted methods uses an Stainless steel fabrications are p r o n e to stress
instrument called the Magne Gage. It works on the corrosion cracking during their service life, when they
principle of the actual magnetic pull exerted by the ferrite are functioning in a corrosive environment. Fabricators
in the deposit. The instrument needs calibration with of chemical plant ought to be alive to this problem and
able to foresee it, because when failure does occur on this branches stemming from them. Figure 10.26 shows the
account, it is too late to effect satisfactory repairs. two kinds of SCC, intergranular and intragranular, as
Stress corrosion, as its name suggests, results from viewed metallographically.
the combined action of a tensile stress and corrosive Although the mechanism of SCC has not been
environment. Stresses that cause cracking arise from established beyond dispute, two proposals have been
residual cold work, welding, or thermal treatment, or they put forth for its occurrence in steels. One is that a readily
may be externally applied during service. Generally corrodible active path is present in the material, which
speaking, the chances of SCC occurring will increase as: causes the cracking. The other is that hydrogen produced
(a) the stress level rises, (b) the attacking compounds in by cathodic corrosion processes produces hydrogen
the electrolyte become more concentrated, (c) the embrittlement and cracking. The latter proposal appears
temperature increases, and (d) as exposure time increases. more general and explains most of the cases of SCC.
Almost all metals and alloys are susceptible to SCC SCC has three recognized regions of crack-
in specific environments. Alloys show a far greater propagation rate -
sensitivity than pure metals. In the case of straight chrome a) Region-I: Initiation of cracking at a stress threshold
stainless steels having more than 12% Cr, SCC is likely to followed immediately by increasing crack growth rate as
occur in the environment of steam, halides and hydrogen stress intensity rises at the frontal tips of the cracks;
sulphides. Austenitic stainless steels are likely to develop b) Region-II: Steady state crack growth as stress at the
SCC in the presence of chlorides and hydroxides. crack tips tends to stabilize at intermediate intensities
and growth rate tends to plateau; and
c) Region-Ill: Rapid crack p r o p a g a t i o n as the
unattacked load-carrying section becomes much smaller
and the stress at the crack tips sharply increases to attain
the final stage of SCC.
In austenitic stainless steels, SCC can occur in the
unwelded parent material due to residual stresses from
cold working or service stresses. In the welded products
where welding may add to the stresses, SCC can occur
both in the parent plate and weld-metal. The most
significant environment causing SCC is aqueous
solutions of chlorides at elevated temperatures. In
Fig. 10.26: Typical kinds of stress corrosion cracking: practice, it is rare for failures to occur below 70°C The Cl-
(A) Intergranular; (B) Intragranular concentration required varies with temperature, being
lower as it rises. In many cases, SCC is caused by the
SCC is a brittle-type fracture in a material exposed to concentration of originally small amounts of Cl in a
a corrosive environment and it should not be confused solution, for example by evaporation on a heated steel
with other types of localised attack such as pitting, surface.
galvanic attack, intergranular corrosion or cavitation. Stainless steel weld-metals having mixed austenite/
Even though SCC is a brittle kind of failure, the ferrite microstructure show a considerably greater
cracking hardly ever propagates rapidly enough to cause resistance to SCC than the parent plate.
a catastrophic failure, even in the case of vessels which Hot caustic solutions may also produce transgranular
operate under pressure. However, the corrosion can SCC in these materials.
penetrate a metal section in a matter of hours, if the section Intergranular SCC can also occur in the HAZs of
is under stress and the metal is markedly sensitive to the unstabilised or high carbon welded stainless steels, where
environment. The resulting leakage of the chemical may iiitercrystalline corrosion or weld decay has already set
mean shutting down of the plant. in during service. In the petrochemical equipment, SCC
Metallographic e x a m i n a t i o n s h o w s that SCC of this type may occur due to the presence of polythionic
p r o p a g a t e s either t h r o u g h the grain b o u n d a r i e s acids, w h i c h are p r o d u c e d particularly d u r i n g
(intergranular) or across the grains (intragranular or shutdowns, by an interaction between water, sulphide
transgranular). It is very rare to find both kinds of containing scales and air. This type of SCC is also possible
propagation existing concurrently. However, in both the in the high temperature, pure water environment of
cases, the main cracks have n u m e r o u s secondary boiling water nuclear reactors. In both these cases, the
preventive measure consists of avoiding the sensitisation The soaking time at temperature should be approxi-
of the stainless steel by appropriate material selection or mately three minutes for each 2.5 mm thickness. Care
welding procedure control. Another effective measure is must be taken to cool the fabrication rapidly and uni-
of course control of the environment. formly through 900-425°C range to avoid carbide pre-
Generally speaking, susceptibility to SCC can be cipitation. Thick fabrications must therefore be water
minimised by ensuring minimum residual stresses in quenched or water sprayed, though air cooling is suit-
the welded structure. Since stress-relief treatment is not able for thin sections. Solution annealing must be under-
practicable with stainless steels, one must avoid severe taken only if a furnace of adequate dimensions, efficient
cold working and adopt forming and welding procedures handling equipment and the right cooling arrangements
in such a way, that internal stresses are minimum and are available. If the cooling is non-uniform, it can give
are evenly distributed over the whole structure. rise to increased residual stresses and distortion. Thin
Three different approaches may be used to avoid fabrications must be adequately supported during high
chloride or caustic induced SCC in stainless steels: temperature exposure to prevent sagging. Unless a pro-
1) Ensure better control on environmental conditions. tective atmosphere has been used, the steel surfaces de-
Consult the pamphlet issued by the Institution of velop an oxide scale, which must be thoroughly removed
Chemical Engineers, 1978 (UK) entitled "Guide notes on by grinding, sandblasting, acid pickling or machining.
the safe use of stainless steel in chemical process plant",
which gives useful guidance on safe procedures for the Cleaning Before Welding
hydrostatic testing of austenitic stainless steels, and on Because of their high Cr content, stainless steel
precautions to be taken when lagging them externally. surfaces develop tenacious chromium oxide film which
Lagging materials sometimes contain chlorides or interferes with welding. For high quality welding, the
encourage their concentration at the metal surface. joint surfaces as also at least 12 mm on each side of the
2) When severe environmental conditions are groove must be cleaned by either of the following techniques
unavoidable, use a more resistant alloy like Incoloy or prior to welding:
Inconel or a low Ni SSC-resistant duplex ferritic austenitic 1) Brushing with a stainless steel wire brush which
alloy or a low interstitial fully ferritic one, has not been used for any other purpose
3) Use a stress-relief heat treatment. As already 2) Sand- or grit-blasting using clean sand and grit
mentioned earlier this is not easy. The minimum effective 3) Machining and grinding using chloride-free cutting
temperature is 900°C, and this introduces complications fluid
with distortion and reintroduces significant residual 4) Pickling in nitric acid solution of 10-20% strength,
stresses on cooling. Even when a perfect stress-relief has Equally important is to ensure that the surfaces are
been achieved, the service stresses themselves may he free from carbon and hydrocarbon contamination caused
adequate to promote SCC. by grease, oil, waxes, cutting fluids and primers. This is
Solution annealing: Austenitic stainless steel fabrica- achieved with solvent cleaning.
tions rarely require postweld heat treatment. Critical fab-
rications are sometimes solution annealed to redissolve Cleaning After Welding
the precipitated carbides in the HAZ, which may cause It is well known that stainless steel is rust- and
intercrystalline corrosion in service; or to relieve the corrosion-resistant only when its surface is in a polished
stresses that may lead to stress-corrosion in service. condition. This resistance is supposed to be due to a
The recommended solution annealing temperatures continuous thin film of chromium oxide formed on the
for different grades are given in Table 10.34. surface. If this continuous film is broken due to the
presence of crevices, blowholes, etc., attack by atmosphere
Table 10.34: Recommended solution annealing or by chemicals can be as rapid as on mild steel.
temperatures for austenitic stainless steels Naturally, stainless steel weld joints should be
Type Temperature, °C finished by grinding and subsequent polishing, if they
201, 202, 301, 302, 304, 304L, 305, 308 1,010-1,120 are expected to give maximum corrosion-resistance. This
309, 309S, 316 1,038-1,120 finishing operation is an expensive proposition and the
316L, 317L 1,038-1,110 degree to which it is to be pursued should be determined
317 1,066-1,120 by the nature of corrosive conditions to be met in service.
321 954-1,066 In any case, weld reinforcement on stainless steel
347,348 982-1,066 welds must be ground out. because they act as stress-
raisers and lead to stress-corrosion when exposed to avoid gouging and damaging the surrounding material.
certain corrosive atmospheres and chemicals. Especially An attachment of stops on either side of the wheel
the surfaces directly in contact must be finished to a considerably simplifies the operation and sets a limit on
degree required by service conditions. the amount of metal removed.
Joints should be so designed that they are accessible To maintain efficient cutting, the wheel should be
for finishing operations and maintain continuity of the frequently dressed a n d carefully trued to avoid
surface. hammering; rounding of the wheel edges also materially
In the following paragraphs the traditional British assists in preventing accidental gouging of the adjacent
practice has been described, though it is likely that it has area.
been refined in recent years. High speed emery bands can be used for all stages of
The first step is coarse grinding. The high coefficient grinding where the component is small enough to be
of thermal expansion combined with the low thermal held against a polishing lathe. These bands run over a
conductivity of the austenitic stainless steels necessitates hard felt pulley with a hackstand to give the required
extreme care during grinding to prevent overheating with tension. A complete range of b a n d s is available
consequent wrinkling, distortion or heat discoloration embodying many types and sizes of grit, but it is
(this latter occurs when the surface temperature rises necessary to ensure that the grit chosen is suitable for
above 300°C). With unstabilised g r a d e s , severe cutting austenitic stainless steels. Using these emery
overheating also impairs the corrosion resistance. In bands, grinding is faster and cooler than when stones
consequence, the grinding wheel must never be allowed are employed.
to dwell in any one spot and heavy pressures must be When it is intended that this rough, grinding
avoided. Otherwise, excessive heat will be generated and operation should be the final finishing operation, the
the wheel will tend to clog and become glazed. surface should be acid cleaned and washed. This
For the initial grinding, solid wheels of the rubber or treatment will remove any metallic particles and surface
plastic bonded type are generally preferred. The grinder contaminants which may have been forced into the
may be either a portable hand type where site grinding is stainless steel surface, and which if left might initiate
necessary or fixed lathe type when the component is corrosive attack.
small. The wheels should be free cutting and self- It might be noted at this stage that during the finishing
sharpening to avoid clogging, and should not be used operation, it is sometimes considered an advantage to
with any other material because of the danger of protect adjacent surfaces by pasting on stout paper or
contaminating the steel surface. The manufacturer's cardboard. This prevents scratching and damage during
recommendations with regard to wheel sizes, abrasive, handling and can easily be removed subsequently.
grit numbers and surface speeds must be followed. Rough grinding is followed by fine grinding which is
Various shapes of stones are available to suit the shape carried out with similar technique and precautions as
of the job and to permit access to difficult positions. described for coarse grinding, but using wheels of finer
The grinding is usually performed dry, but when a abrasive. The manufacturer's recommendations relating
lubricant seems desirable, the recommendation of the to wheel type and size, grit number and surface speeds
grinding wheel maker should be observed. Cooling may must be followed. It is often an advantage, however, to
also be employed where practicable by playing a water replace final grinding by the use of emery bobs consisting
spray on the reverse side of the weld, thus permitting of solid felt wheels, on the periphery of which is glued
somewhat faster grinding. The grinding operation should emery of the required grit size,
be performed in long sweeps with wheel diagonal to the Emery bobs are often preferred to fine stones, since
line of weld rather than at right angles to it, the angle they are softer than wheels of corresponding grit and
being changed with successive grit sizes. This will be reduce the danger of damage to the surrounding metal.
found easier than attempting to grind across a narrow They are also cheaper in use than stones.
weld bead, and the change in the grinding direction at The preparation of bobs requires some care. Where
successive stages is of considerable help in removing the warm glue is used, it must be made fresh each day and
marks of earlier grinding operation. When grinding very only best hide glue should be employed. It should never
thin gauge material, the use of steel or wooden backing be overheated, and should be used at the temperature
blocks may be desirable to provide a firm basis for working. recommended by the glue manufacturers (usually about
When attempting the flush grinding of welds with 60°C). The iron-free emery should be preheated to
very rapid cutting wheels, great care should be taken to approximately the same temperature as the glue. The bob
is rolled backwards and forwards in the emery to obtain to a mirror finish, but those containing titanium, and to a
an even coat and is then dried in a low-temperature drying lesser extent niobium, cannot be successfully buffed.
oven. Buffing is essentially two distinct operations; cut
The use of bobs may be introduced at the 60 or 80 grit buffing, which prepares the surface to the required
stage during finishing, but it is necessary that the smoothness for the next operation of colour buffing, by
preliminary grinding should have been carefully which the mirror finish or high lustre is obtained on the
executed. There is a limit to the amount of metal that can work.
be economically removed by means of an emery bob, and In contrast to polishing, the abrasive for buffing is not
prior to its use grinding should be taken to a stage where glued to the wheel or mop, but is fed on to it from a compo
it only remains for the bob to fine up the surface. bar, or as a paste which consists of a fine powdered
The normal procedure is to remove the grinding marks abrasive in a suitable binder.
using a 60 or 80 grit bob, then continue successively with Buffs of finely woven (82 to 94 count) calico, or calico
bobs of 120 and 180 grit. The same precautions apply in interleaved with paper, are commonly employed. These
bobbing as in grinding, regarding the need to avoid are generally unsewn, the necessary stiffness being
dwelling on one spot. Lubricants are generally used obtained from the high rotational speeds. The cutting
during the bobbing operation which consist of tallow, action decreases with increasing surface speed, but the
palm oil or stearic acid. The residual lubricant may be lustre improves and so cut buffing is usually carried out
removed from the finished surface, and the surface at lower speeds (1,650 to 1,950 surface metres/minute)
brightened by using dry whiting or Sheffield lime on a than colour buffing (2,250 to 4,200 surface metres/
soft rag mop. minute). Lubricants such as tallow, stearic acid or wax
As with grinding, the use of bobs on a lathe has to are employed. Abrasives must be of highest quality and
some extent been displaced by continuous emery bands. purity, and entirely free from iron or sulphur, if best results
These are fed with a lubricant and appear to give faster are to be obtained.
cutting than bobs.
Calico m o p s may replace bobs w h e r e a softer PH Stainless Steels
polishing operation is desired, and with these also, the Precipitation-hardening (PH) stainless steels make up
finish from the grinding stage should be as smooth as an important group of stainless steels, which combine
possible, since mop polishing will not remove coarse high strength with good corrosion and oxidation
surface defects and irregularities. resistance. In these steels, hardening is achieved by
The mops are made up from layers of calico or canvas martensite formation, by precipitation-hardening or by a
sewn together to give stiffness, and with suitable graded combination of the two.
abrasives glued to the periphery. The abrasives used are PH steels contain key elements like Cu, Cb, Ti and Al
similar to those employed on bobs. which go into solution during solution annealing or
The first or rough polishing operation, usually carried austenite conditioning. During the subsequent aging
out without a lubricant, should be continued until a clean, treatment, they produce submicroscopic precipitates
uniform and defect-free surface is obtained. Progressively which render the matrix stronger and harder.
finer grit is then used, in most cases with a lubricant There are three distinct types of PH stainless steels
consisting of tallow, palm oil or stearic acid, until the depending on the metallurgical structure they develop
required degree of surface finish is achieved. The same when cooled from the solutioning or austenitising
general precautions as outlined for grinding and bobbing temperature. They are: martensitic, semi-austenitic and
should be observed, particularly with regard to keeping austenitic (see Table 10.35).
the mop on the move to avoid local overheating. Martensitic PH steels: The chemical compositions of
The polishing wheels should be worked with an these steels are such that they provide a strong and hard
oscillatory motion to break up the polishing lines, and martensitic structure on cooling from the austenitising
the wheel direction should be frequently changed to temperature, To explain this further, when the steel is
prevent the abrasive cutting grooves. heated to approximately 1,040°C, its structure is mainly
Polishing is usually taken further than bobbing, and austenite. On quenching, the austenite transforms to
a finishing grit size of 200 to 300 is not uncommon. martensite in the temperature range of 150-100°C. During
The final step is buffing. The response of the various subsequent aging treatment, i.e. heating the steel to the
grades of 1 8 / 8 austenitic steels to buffing varies appropriate temperature in the range of 485-620°C for
considerably. The unstabilised 18/8 steel may be buffed an appropriate time, it gets further strengthened by a
Table 10.35:1Nominal compositions of typical PH stainless steels

Nominal composition, weight %
Type Designation* UNSNo. C Mn Si Cr Ni Mo Al Other elements
Martensitic 17-4 PH S17400 0.04 0.30 0.60 16.0 4.2 — — 3.4 Cu, 0.25 Cb
(moderate 15-5 PH S15500 0.04 0.30 0.40 15.0 4.5 — — 3.4 Cu, 0.25 Cb
strength)
Martensitic PH 13-8 Mo S13800 0.04 0.03 0.03 12.7 8.2 2.2 1.1 —
(high strength)
Semi- 17-7 PH S17700 0.07 0.50 0.30 17.0 7.1 — 1.2 —
austenitic PH 15-7 Mo S155700 0.07 0.50 0.30 15.2 7.1 2.2 1.2 —
Austenitic A-286 K66286 0.05 1.45 0.50 14.75 25.25 1.30 0.15 0.30 V, 2.15 Ti, 0.005 B
17-10 P — 0.10 0.60 0.50 17.0 11.0 0.30 P
*Some of these designations are registered trademarks.
precipitation mechanism. Typical heat treatments for a of hours) exceeds 316°C. For short-time service the limiting
few well-known grades of PH steels are given in Table temperature can be raised to 480°C.
10.36. Thin sections of martensitic PH steel are usually
Some of these steels have moderate tensile strength welded in the annealed condition. After welding, the
(less than 1,380 N/mm 2 ), while others considered as high maximum strength and corrosion resistance are achieved
strength, have tensile strength more than 1,380 N / m m 2 . by a solution heat treatment followed by aging. Thick
Their corrosion resistance is as good as that of the sections and highly restrained joints are commonly
common austenitic stainless steels and is greatest in the welded in the overaged condition. The welding heat
fully hardened condition. They also possess good causes the HAZ to be austenitised and transformed to
resistance to stress corrosion cracking when aged at or martensite on cooling. The HAZ is relatively soft and
above 550°C. They tend to become embrittled when the free from underbead cracking because of the extremely
service temperature for long-time exposure (thousands low carbon content of the steel. In a few cases, the required
Table 10.36: Typical heat treatments for PH stainless steels

Type of steels Austenite conditioning Aging treatment
Temperature, °C Quenching media Temperature, °C Time, hour
Martensitic
17-4 PH 1,040 0,A 480 1
or
490-620 4
15-5 PH 1,040 W 480 1
or
552-620 4
Semi-austenitic
17-7 PH 955 A (l)-59 8
PH15-7 Mo 760 A (2) 510-540 1
565-590 1.5
Austenitic
A-286 980 O 717 16
17-10 P 1,120 w 705 24
welded joint properties are obtained by welding the steel elements such as Al, Ti and P form intermetallic
in the hardened condition (no preheat necessary) and compounds and thereby increase the strength of the steel.
then subjecting the joint to a postweld aging treatment. These steels have superior high temperature
Semi-austenitic PH steels: These steels are also properties and their good corrosion resistance can be
chemically composed to give an austenitic structure if maintained up to approximately 815°C. These steels are
cooled rapidly from annealing temperature of 730-955°C, difficult to weld and some grades are considered
because the martensitic transformation temperature is unweldable because of pronounced susceptibility to cold
below room temperature. But in practice, they are air cracking in the HAZ. The steel is usually welded in the
cooled from the high temperature range, and this results solution-treated condition. To prevent cracking, welding
in the precipitation of carbon from solution in the form of is performed with least joint restraint and at low heat
chromium carbides and consequent raising of the input. It is not possible to weld these steels with matching
martensitic transformation temperature. Thus the result filler metals because of cracking trouble. Nickel-base or
of air cooling from the austenitising temperature is the common austenitic stainless steel filler metals are
transformation of the steel to martensite. The subsequent preferred.
aging treatment (a) stress-relieves and tempers the Suggested filler metals for a few typical PH steels are
martensite, (b) increases ductility, toughness and given in Table 10.37.
corrosion resistance, and (c) further hardens the steel by
causing precipitation of an intermetallic phase. Austenitic Manganese Steel
These steels are usually welded in the annealed or Austenitic manganese steel is a high-alloy steel
solution-treated condition. Sometimes they may be containing about 11-14% Mn and 1.0-1.4% C, which
welded in the hardened condition. In either case, the has an austenitic structure at room temperature and is
welding heat transforms the narrow HAZ to austenite, therefore nonmagnetic, Its advantageous characteristics
which remains austenitic under rapid cooling that are high strength, good ductility, extreme toughness,
follows. No cracking occurs in this HAZ. However, if rapid work-hardening property and good wear
optimum mechanical properties are desired in the resistance. It is also called Hadfield manganese steel.
weldment, the steel must be welded in the hardened The steel is used mainly in the form of castings but is
condition (no preheat necessary) and the weld joint must also available as hot-rolled billets, plates, bars and wire.
be given a postweld aging treatment. The wire is normally supplied in the annealed condition,
Austenitic PH steels: These steels contain high levels of while the rest are supplied in the quenched condition.
nickel compared to their chromium contents, with the Applications: Austenitic manganese steel castings are
result that they develop a fully austenitic structure on used in crushing and grinding mills, earthmoving and
quenching from the austenitising temperature. After this material handling equipment and on railway crossings.
solution treatment, hardening is achieved by aging at Gyratory and cone crusher parts, jaw crushers and
approximately 700°C for 16 to 24 hours, during which hammer mill parts are made from this steel because of its
Table 10.37: Suggested filler metals for welding PH stainless steels

Designation UNSNo. Covered electrodes Bare welding Dissimilar
wire PH stainless
steels
Martensitic types
17-4 PH S 17400 AMS5827B AMS5826 E or ER309,
and (17-4 PH) or (17-4 PH) or E or ER309Cb
15-5 PH S 15500 E308 ER308
Semi-austenitic types
17-7 PH S 17700 AMS5827B AMS5824A E or ER310,
(17-4 PH), (17-7 PH) ENiCrFe-2, or
E308, or E309 ERNiCr-3
PH 15-7 Mo S 15700 E 308 or AMS 5812C E or ER309,
309 (PH 15-7 Mo) E or ER 310
Austenitic type
A-286 K 66286 E 309 or ERNiCrFe-6 or EorER309,
E310 ERNiMo-3 E or ER 310
good resistance to severe impact and abrasive wear. Being Si content is normaly below 1%, but it is sometimes
non-magnetic, it is used for parts of electromagnets, increased to about 2% to raise the yield strength margin-
induction furnaces and other electrical products. ally and to improve resistance to plastic flow during re-
Metallurgy: Mn plays a d o m i n a n t role in the peated impact. Above 2%, this element causes a sharp
metallurgy of this steel. Since Mn has a strong affinity for fall in strength and ductility. S has no effect on (this steel,
O, S and C, it is used in the conventional C-Mn and low- but P should be kept below 0.04% if hot cracking during
alloy steels mainly as a deoxidiser, to reduce hot cracking fabrication and repair welding has to be avoided.
and to increase strength and hardness. When Mn is in Mo or Cr is sometimes added to raise the yield
free form, it forms a substitutional solid solution with strength, and Ni is added to marginally increase ductility
iron and suppresses the transformation of austenite to and to stabilise the austenite. Ti is also used in some
hard microstructures during cooling. When Mn content cases, as it forms stable carbides and thereby lowers the
is more than 10.5%, it suppresses the transformation of amount of carbon dissolved in the austenite.
austenite so effectively that the austenite state is retained Austenitic manganese steel is considered to be easily
at room temperature even with moderately fast cooling. weldable, but several factors related to its production
In production practice, the high manganese steel is can have a deleterious effect on weldability. It has already
heated to a temperature in the range of 982-1,065°C, been mentioned that Mn should be above 11%, while
where it is fully austenitic and the carbon is in solution, carbon content should not exceed 1.35% to ensure
held at that temperature for more than 30 minutes and adequate mechanical properties.
then water-quenched. If, d u r i n g the final heat treatment, the correct
It is important to maintain the Mn content m the steels austenitising temperature is not reached and some of
within the range of 11.0-14.0%, because then the tensile the carbides remain undissolved, the microstructure of
strength and ductility are maximum, though the yield the steel reveals chains and clusters of carbides together
strength is hardly ever affected by the Mn content. The with intermittent grain boundary carbide envelopes. If
tensile strength and ductility decrease rapidly as the Mn the cooling rate from the austenitising temperature is
content is lowered; at 8% Mn these properties are almost too low, carbide precipitation will occur and the
halved. Below 8% Mn level, the steel becomes very brittle mechanical properties of the steel are lower than normal.
and unusable. Steels with carbon around 1.3-1.4% are especially prone
The carbon content of the steel is fairly high, the to such carbide formation. In the case of heavy sections,
optimum level being 1.15%. Such a high level is necessary carbides may appear in the grain boundaries in the
to obtain adequate yield strength. Carbon levels higher central portion of the section, in spite of sufficiently rapid
than 1.35% create problems during heat treatment and quenching due to the low thermal conductivity of the
foundry handling due to carbide precipitation. steel.
Tool Steels
r~ i High-speed
Water-hardeningShock-resisting Cold-work
(W1,W2,W5,) (Sl,S2,S5,etc.)
Oil-hardening Medium-alloy HighC, W type Mo type

(01,02, etc.) air-hardening high Cr (Tl,T2.etc.) (Ml,M2,etc.)
(A2,A3,etc.) (D2,D3,etc.)
Hotrwork
I 1 1
Cr type W type Mo type
(H10,H11, (H21,H22, (H42)
etc.) etc.)
Castings should be welded only after they have been which lead to embrittlement. Manual metal-arc welding
correctly heat-treated to render them fully austenitic. In is therefore generally preferred for joining as well as
as-cast form, they are susceptible to cracking because the surfacing work. Semi-automatic or fully-automatic
microstructure is austenite grains surrounded by carbides welding with solid or flux-cored wires is also quite
with patches of pearlite, depending on the chemical suitable, because the smaller electrodes and higher arc
composition and cooling rate. travel speeds of this process results in low heat input.
Castings may sometimes have a decarburised skin Processes like oxyacetylene and tungsten arc welding
which is martensitic, magnetic and relatively weaker than which cause considerable heat build-up should be
the underlying metal. Such a skin results from the high avoided.
austenitising temperature, high carbon content and some As a practical guideline, it is recommended that the
loss of manganese. base metal in the vicinity of the weld deposit should not
The quenched and austenitised casting should never be heated above 315°C for more than a minute. Also one
be tempered or stress-relieved because depending on the should be able to touch the base metal 150 mm away from
t e m p e r a t u r e a n d time at t e m p e r a t u r e , carbide the weld at any time, without burning the skin. To achieve
precipitation and pearlite formation occur and the result this, a short arc, stringer bead technique and inter-mittent
is a complete loss of ductility and reduction in strength. welding are used. Skip welding can also minimise heat
Mechanical properties: Austenitic manganese steel is build-up. Sometimes for heavy surfacing work, the
used under extremely rigorous service conditions (earth- casting maybe kept partially immersed in water to increase
moving, c'rushing, and mining equipment) mainly the weld deposition rate. Multiple-pass welds are
because of its o u t s t a n d i n g t o u g h n e s s and work- preferred to single-pass welds; because the succee-ding
hardening property. pass can temper the austenite formed in the earlier pass.
A typical casting gives a Charpy-V impact value of It is very important that, before commencing welding,
136 J at 23°C and about 68 J at 73°C. Addition of Ni raises the work-hardened metal, cracks and.O other defects are
the impact value at -73°C to about 110 J. An as-rolled Ni completely removed by grinding or air carbon-arc
alloyed bar has an impact strength of 136 J at -73°C. gouging. It should also be ensured that the surface to be
Quench annealing increases it to about 195 J. welded is non-magnetic.
These high-Mn steels have a much higher coefficient Electrodes for surfacing work are usually designed to
of work-hardening as compared to that of austenitic have lower carbon content (0.5-0.8%) to avoid the embrit-
stainless steel. The initial hardness of 185-210 HB can tling effect of carbide precipitation during cooling. Mn
rise to a maximum of 550 HB under repeated impact content is sometimes as high as 16% to ensure completely
loading in service. As hardness increases, tensile strength austenitic structure in the weld deposit. P is kept below
and yield strength also increase. In fact, a range of tensile 0.03% to prevent hot cracking. Mn-Cr types (14% M n -
values can be obtained in an austenitic manganese steel 14% Cr) and Cr-Ni-Mn types (20% Cr-10% Ni-5% Mn)
by work hardening in the same manner as can be are preferred for improved corrosion and abrasion
obtained in other steels by heat treatment. resistance and for their ability to work-harden rapidly
The work-hardening is produced by deformation, under impact. The Cr-Ni-Mn type is the best choice for
whic^ results in the transformation of austenite to joining austenitic manganese steel to carbon steel and
martensite. The hard martensitic layer which is magnetic for buttering the carbon steel surface before laying a
is rather shallow and the tough austenitic structure below deposit of austenitic manganese steel. If an austenitic
offers a cushioning effect against repeated heavy impact. Mn-steel electrode having about 16% Mn and less than
As the hardened, brittle metal wears away under 0.025% P is used for these applications, the dilution with
conditions of abrasive wear and impact, a fresh surface the carbon steel should be less than 25%. For repair of
is exposed for further work-hardening. cracks in austenitic manganese steel, aust-enitic stainless
Quite often, manganese steel parts are exposed to steel electrode types 308, 309, 310 or 312 must be used.
abrasive wear alone, where impact conditions are absent. Use of austenitic Mn-steel electrodes must be avoided.
In such cases, hardness is increased by alloying, heat
treatment, age hardening or initial work-hardening. Tool and Die Steels
Welding: Austenitic manganese steel has good Tool and die steels are high-carbon alloy steels used
weldability provided minimum heat input is used during for the forming and cutting of metals. Tool steels are
welding. As already pointed out, reheating can give rise classified by the American systems AISI and SAE, into
to carbide precipitation and transformation of austenite seven major groups of which five relevant ones are given
Table 10.38: Data for major groups of tool steels

Group Approx. Properties Typical
symbols compositions applications
Wl, W2, W5 0.60/1.40 C sometimes alloyed Water-quenched for max. hard- Inexpensive tools.
with Cr and V ness. Best machinability rating.
Proof red hardness. Excellent
resistance to decarburisation.
SI, S2. S5 0.40/0.65 C, alloyed with Si, Mostly oil-quenched, but Tools punches, chisels,
Cr, W and sometimes with Mo sometimes water-quenched for pneumatic tools, expansion
maximum hardness. Good reamers.
toughness, red hardness, wear
resistance and machinability.
Tendency to decarburisation.
0.1, 02 etc. 0.90/1.45 C, Mn smaller Good resistance to deformation Thread taps, solid
amounts of Cr and W and to decarburisation, threading dies,
good machinability, fair tough- forming tools and
ness but poor red hardness. expansion reamers.
A2, A3, etc. 0.70/2.35 C, 2% max. Mn Excellent resistance to deforma- Blanking, forming, trim-
5% max. Cr, 1.0 Mo tion and to decarburistion, good ming and thread-rolling
wear resistance, fair toughness dies.
and red hardness.
D2, D3, etc. 2.25 max. C, 12.0 max. Cr, Excellent resistance to wear and Blanking and piercing dies,
sometimes alloyed with Mo, deformation. Good abrasion resis- drawing dies, thread rolling
W&Co tance. Dimensions are retained dies and master gauges.
during hardening. Not suitable
for cutting edges. Should not be
worked above 480°C.
H10, H l l , etc. 0.35/0.40 C, 3.25 min. Cr, Good red hardness and toughness, Extrusion dies, die-casting
small amounts of V, W & Mo extremely deep air-hardening, less dies, forging dies,
distortion during hardening. mandrels and not shears.
H21, H22, etc. 0.30/0.50 C, 9/18 W, 3/12 Cr Improved red hardness, reduced High temperature appli-
toughness computed to Cr type, cations such as mendrels.
butt better toughness compared to and extrusion dies for
high-speed tool steels. brass, Ni alloys & steel.
H42 0.6 C, 4.0 Cr, 6.0 W, 5.0 Mo Better toughness and greater resis-
tance to heat cracking as compared
to W type. Danger of decar-
burisation during heat
treatment.
Tl, T2, etc. 0.75/1.50 C, 20.0 max. W, Excellent red hardness, more so Cutting tools such as tool
4.0 max. Cr, sometimes 5/12.0 when Co is present, good bits, drills, reamers,
Co to raise red hardness resistance to wear and shock, broaches, taps, milling
resistant to deformation, fair cutters, saws and wood-
machinability, medium resistance working tools. Also
to decarburisation. Well suited extrusion dies and blanking
Ml, M2, etc. 0.85/1.25 C, 4.0 max. Cr, 4.5/ for cutting tool applications. punches and dies.
9.5 Mo, sometimes 5/12.0 Co
to raise red hardness
below along with their sub-groups and typical letter High t u n g s t e n content increases resistance to
symbols (indicated in brackets). Their detailed softening at high temperatures, but it also makes the steel
information is given in Table 10.38. more brittle at high hardness.
Cobalt confers high red hardness. With high Co
ROLE OF ALLOYING ELEMENTS content, special care must be taken during heat treatment
Presence of more than 0.60% carbon in tool steels to avoid decarburisation.
ensures a martensitic hardness of at least 60 Re on Vanadium added in small amounts to the water-
quenching. When carbon is in excess of the eutectoid h a r d e n i n g g r o u p i m p r o v e s t o u g h n e s s and wear
composition, it exists in the form of undissolved carbides resistance.
and promotes wear resistance. Toughness is reduced Red hardness: A tool steel which is exposed to high
with increased carbon content; hence if good toughness service temperature as in hot forging, die casting and hot
is desired, carbon is kept at a lower level. Unalloyed shearing must not soften at elevated temperatures. In other
carbon steels have low hardenability and hence they need words, it must have good red hardness. A steel which
to be water-quenched (water has the most severe does not have this property wears out rapidly in service.
quenching action) to obtain optimum hardness. The
hardenability increases with total alloy content. Alloy WELDABILITY
tool steels have sufficient hardenability to attain The weldability of a tool steel is determined by its
maximum hardness with less severe oil quenching or air chemistry and condition. The tool may be in (a) annealed
cooling. condition, (b) hardened condition or (c) hardened and
The alloying elements commonly used in tool steels tempered condition. The repair welding procedure for
are chromium, molybdenum, tungsten and vanadium. each of these conditions is detailed below:
All of them increase the eutectoid temperature of the steel a) Annealed condition: Whenever possible, tools and
by stabilising the ferrite phase. Also with increasing total dies must be welded in the annealed condition, because
alloy content, the carbon content of the eutectoid then they have maximum ductility and are less prone to
composition decreases, since these alloying elements cracking. The steel is preheated to the appropriate
have a greater affinity for carbon than iron has. Hence temperature, and welded with a matching filler metal. It
tool steels must be heated to higher temperatures to is t h e n stress-relieved, m a c h i n e d to the n e a r e s t
dissolve the carbides into austenite. dimensions, hardened and tempered and ground to final
With increasing carbon content of the austenite, its dimensions.
transformation to other microstructures becomes b) Hardened condition: Welding a tool or die in this
increasingly sluggish. A condition may then be reached, condition must be avoided if possible. It should be
where the high-carbon austenite is retained down to annealed and then welded according to the procedure
room temperature even with severe quenching. All the given in (a).
alloying elements used in tool steels except cobalt If it must be welded in the hardened condition, the job
generally decrease the a u s t e n i t e - t o - m a r t e n s i t e should be preheated and welded with matching filler
transformation. Cobalt does the opposite. metal. Then it should be tempered and ground to fine
In the water-hardening group, small addition of dimensions.
chromium improves toughness and wear resistance. In c) Hardened and tempered condition: It is safer to weld a
the shock-resisting group, higher chromium levels tool in this condition than in the hardened condition. In
increase hardenability. In the high-carbon high- this case, annealing and hardening can be dispensed
chromium group, these two elements in combination give with and one can avoid the dimensional changes these
excellent wear resistance, abrasion resistance and non- heat treatments may give rise to.
deforming properties. Chromium in combination with Typical welding and heat-treatment procedures for a
molybdenum and tungsten promotes red hardness. few types of tool steels are given in Table 10.39.
In some tool steels, silicon is 1-2%. Silicon is a ferrite The procedures of heat treatment are as follows:
s t r e n g t h e n e d b u t the h i g h Si level accelerates Annealing: The tool is g r a d u a l l y h e a t e d to $
decarburisation. temperature slightly above its transformation range, held
Mo increases hardenability. Together with chromium at that temperature long enough for the heat to seep
and tungsten, it improves red hardness. When Mo through the entire body and then cooled slowly at a
content is high, special care is required during heat controlled rate to room temperature. Care must be taken
treatment to avoid decarburisation. to prevent carburisation or decarburisation.
Table 10.39: Typical welding and heat-treating procedures for representative tool steels
Annealed base metal
Temperature,°C Hardened beise metal
Preheat and HRC°
Type Filler metal type Preheat Annealing Austenitising Tempering Quenching HRC and
and mediab postheata°C
3
postheat
W1,W2 Water-hardening 120-230 733-790 760-840 175-340 B,W 50-64 120-230 56-62
SI Hot-work 150-260 790-815 900-955 205-650 O 40-58 150-260 52-56
S5 Hot-work 150-260 775-802 870-930 175-425 O 50-60 150-260 52-56
S7 Hot-work 150-260 815-840 930-955 205-620d A,0 45-57 150-260 52-56
Ol Oil-hardening 150-205 760-790 790-815 175-260 O 57-62 150-205 56-62
06 Oil-hardening 150-205 765-790 790-815 175-315 O 58-63 150-205 56-62
A2 Air-hardening 150-260 840-870 930-980 175-540 d A 57-62 150-205 56-58
A4 Air-hardening 150-260 733-760 815-870 175-425 d A 54-62 150-205 60-62
D2 Air-hardening 370-480 870-900 980-1,025 205-540 d A 54-61 370-480 58-60
H11,H12,H13 Hot-work 480-650 840-870 995-1,040 540-650 d A 38-56 370-540 46-54
Ml High-speed 510-590 815-870 1,180-1,217 540-590 d A.O.S. 60-65 510-565 60-63
M2 High-speed 510-590 870-900 1,192-1,230 540-590 d A.O.S. 60-65 510-565 60-63
M10 High-speed 510-590 815-870 1,180-1,217 540-590 d A.O.S. 60-65 510-565 60-63
T1,T2,T4 High-speed 510-590 870-900 1,260-1.302 540-590 d A.O.S. 60-66 510-565 61-64
Notes: a) Preheat and postheat temperatures for welding
<
b) A—air cool, B—brine quench, O—oil quench, S—salt bath quench, W—water quench
c) Hardness after postheat
d) Double temper
Stress-relieving: This heat-treatment is intended to about 65°C and then tempered. Water-hardening tends
reduce the locked-up stresses caused by welding. The to distort and change the dimensions of a tool. Hence,
procedure is similar to that used for annealing but the sometimes a fine water spray is used to achieve
temperature is below the transformation range. In the superficial hardening.
case of a hardened tool, retempering serves the purpose Tempering: This treatment is intended to relieve
of stress-relieving. In this case, the temperature used stresses, prevent cracking and toughen the tool and is
should be below the tempering temperature of the steel. most effective when carried out immediately after
Austenitising: This operation, which is intended to quenching. In fact, it is advisable to remove the tool from
dissolve the carbides in austenite, involves heating the the quenching medium, while the temperature is still
tool slowly to a temperature much above the around 80°C and temper immediately.
transformation range and holding it at that temperature It must be understood that tempering at the top end
for a sufficient length of time. The treatment should be of the recommended temperature range gives maximum
done in a protective atmosphere or vacuum to avoid toughness and reduced hardness, while that at the
scaling or decarburisation. lowest end gives maximum hardness and reduced
Quenching: Quenching means rapid cooling from the toughness. Most favourable metallurgical structure is
austenitising temperature by immersing in water, brine, obtained by resorting to two or more tempering cycles.
oil, etc., or cooling in air. The quenching is most severe in Welding processes: Repair of small tools is best carried
water and least severe in air. The quenching medium to out with the manual tungsten inert-gas (TIG) process.
be used depends on alloy composition and section Also this process is ideal for the repair of high-speed
thickness. The cooling rate must be rapid enough to obtain cutting tools, because of low welding heat and absence
full hardness but not too rapid, as this may lead to of hydrogen in the arc. Manual metal-arc welding is well
cracking. suited for repair welding of small areas on large tools.
Air-hardening tools are sometimes hot quenched to For repairing large areas, automatic TIG using hot or
540-650°C range, then air-cooled or oil-quenched to cold wire-feed can be used with advantage.
Cast Iron iron. As a result of this addition the graphite nucleates

Cast iron is a very common engineering material. It is through the entire metal matrix as regular spheroids or
always used in the form of castings. It cannot be rolled, nodules. Because nodular graphite is free from the sharp
drawn or forged because it has very low ductility and is notches of flake graphite, the cast iron has improved
brittle. strength, ductility and toughness. These properties are
Cast iron is an iron-carbon alloy with more than 2.5% achieved w i t h o u t resorting to the elaborate heat
C, 1-3% Si and relatively higher levels of P and S. When treatment required to produce malleable iron castings.
it is alloyed with elements such as Cr, Cu, Mo and Ni to Because of its pronounced ductility, nodular iron is also
improve strength, hardness or corrosion resistance, it is called as ductile iron.
termed alloy cast iron.
The high carbon together with high silicon content WELDABILITY
contribute to the unique microstructure of cast iron. In Cast iron is known to have limited weldability when
this material, the metastable iron-iron carbide system and compared to carbon steels. Among the four types of cast
the stable iron-graphite system exist side by side. A part iron, nodular iron has the best weldability, while
of carbon exists in the matrix in the same manner as in malleable iron comes next in the order. Welding of grey
steel and the matrix can be ferritic, pearlitic, austenitic or iron calls for special care and skill. White cast iron is
martensitic. The rest of the carbon exists as free graphite, extremely difficult to weld.
which is distributed through the entire matrix in the form Welding of cast iron is generally restricted to the repair
of tiny particles, flakes or spheroids. The size, shape and of castings in a foundry and the salvaging of parts that
dispersal of graphite as well as the microstructure of the are fractured or worn out in service. Fabrication of
matrix determine the mechanical properties of a components is rare and is mostly carried out with nodular
particular casting. The desired characteristics are irons. N o d u l a r iron p i p i n g systems h a v e been
achieved by chemical control and by controlling the successfully fabricated by welding.
solidification and cooling rates during casting. In some The generally poor weldability of cast iron is ascribed
cases, the castings are given suitable heat treatments to to the following two factors:
achieve the desired microstructure in the matrix. 1) Because of the high carbon content in the base metal,
There are four types of cast iron: (1) grey iron, (2) white the welding cycle causes the formation of carbides in the
iron, (3) malleable iron and (4) nodular iron. zone immediately adjacent to the weld-metal and the
1) Grey iron: Most castings in use are of this type, in formation of high-carbon martensite in the remainder of
which the free carbon is in the form of graphite flakes. the heat-affected zone. Both these microstructures are
These flakes make the fractured surface appear grey and brittle and can cause cracking during welding or in
render the cast iron brittle and less ductile. The graphite service. This is applicable to all the types of cast iron.
also makes the casting soft and machinable. 2) Because cast iron has poor ductility, it is not capable
2) White iron: This type results when the casting is of local plastic deformation, which is necessary to
solidified and cooled so rapidly that the carbon has no accommodate the thermal stresses caused by welding.
opportunity to precipitate as graphite but remains in the Grey iron is prone to cracking on this account, while
form of carbides of Fe, Cr or Mo. It is hard and extremely malleable and nodular irons are less so because of their
brittle. Its fractured surface has a white crystalline better ductility.
appearance. Referring to factor (1), brittleness of the heat-affected
3) Malleable iron: This type is produced by heat-treating zone depends on how rapidly and easily the graphite
a white cast iron, which has lower levels of carbon and dissolves in the austenite during welding. In the case of
silicon and is alloyed with carbide forming elements such grey irons, the graphite flakes with their relatively large
as Cr, Mo and V. The heat treatment consists of heating surface area dissolve easily in the austenite. In nodular
in a furnace to about 925°C, holding for several hours irons, the spheroidal graphite has a low surface-to-
and cooling slowly. As a result of this treatment, the volume ratio and therefore less graphite is dissolved in
carbon dissolved in the austenite precipitates in the form the austenite. The result is that less massive carbides
of irregular nodules of graphite and the matrix develops and lower-carbon martensite are formed in the HAZ. This
a ferritic structure. The result is improved strength and further explains the superior weldability of nodular iron.
ductility. The tendency of HAZ to cracking is lessened when
4) Nodular iron: This type is produced by adding the cast iron is arc welded with low heat input, because
magnesium or cerium to a low-sulphur melt of grey cast this technique reduces the width of the hard and brittle
zone next to the weld. To reduce the HAZ hardness, one weld-metal and the HAZ. This explains why welding
must resort to the combination of preheating and slow with nickel-based electrodes results in minimum failure.
cooling after welding. The slow cooling through the The weld-metal of pure nickel electrode is softer than
austenitic transformation range reduces the amount of that deposited by the ferronickel electrode. However, the
martensite formed and thereby the hardness. ferronickel deposit is stronger and more ductile, has bet-
Recommended preheat temperature for arc welding ter tolerance for phosphorus in cast iron, and is more
ranges between room temperature (21 °C) and 300°C; for resistant to hot cracking. For welding cast iron to mild
oxyacetylene welding the range is 450-650°C. Grey iron steel, stainless steels and nickel alloys, ferronickel elec-
requires higher level of preheat, while nodular and trode is preferred. Compared to the above two, monel
malleable irons having ferritic matrix need minimum weld deposit is more susceptible to cracking due to dilu-
preheat. Nodular and malleable irons having pearlitic tion by cast iron. All these weld deposits are machinable.
matrix require higher preheat than the ferritic types. The The advantageous feature of copper-alloy electrodes
preheat may be localised to the welding zone or the entire (tin bronze and aluminium bronze) is that the weld-metal
casting may be preheated. The preheat temperature and when hot is soft and ductile because of the large amounts
extent of preheat depends on the type of cast iron, size of of alpha-phase in the microstructure. The weld-metal
casting, welding process, type of filler metal and the therefore yields without cracking as the weldment cools
amount of weld-metal to be deposited. If local preheating and relieves most of the contraction strain. The weld-
is used, steep temperature gradients within the casting metal strength increases appreciably at room temperature
must be avoided, because unequal expansion can give without significant reduction in ductility. Aluminium
rise to cracking. Preheat must be maintained until the bronze deposit is twice as strong as the tin bronze deposit.
welding has been completed and then the heated area During welding with copper-alloy electrodes, dilution
must be covered with insulation (lime, sand, mica, must be kept minimum.
asbestos, etc.) to ensure slow, gradual cooling. 2) For oxyacetylene welding, the filler rod is either
In the case of critical castings, immediately after cast iron or copper-zinc alloy. Both are used with
welding is completed, the temperature of the entire casting appropriate flux. For welding with a cast iron rod, preheat
is raised to 600-625°C and the casting held at this temperature of 600-650°C is used. When Cu-Zn alloy
temperature for sufficient length of time (one hour per 25 rod is used, the preheat temperature is 320-500°C and
mm thickness) to relieve the internal stresses. It is then welding is effected at temperatures much below the
cooled at a rate not exceeding 30° per hour until the melting point of cast iron. The process is therefore referred
casting has cooled to 370°C. If the heat treatment is carried to as brazewelding. The Cu-Zn alloy rod gives a colour
out at 900°C followed by slow cooling to 500°C, it results mismatch. Its use is not recommended when new castings
in minimum hardness and maximum stress-relief. are to be repaired in the foundry.
3) For flux-cored arc welding, a ferronickel electrode
FILLER METALS is used, which consists of a ferrous tube filled with nickel
1) For manual arc welding, the electrode is mild steel, powder and fluxing agents. The electrode may be used
pure nickel, monel, ferronickel, tin bronze or aluminium with or without external C 0 2 shield.
bronze.
The mild steel electrode is of the basic low-hydrogen Aluminium Alloys
type. The weld-metal picks up carbon from the base metal Because of its lightness, resistance to the corrosive
and hardens. Hence the welded joint is non-machinable. action of air, water, oils and many chemicals, and easy
The weld-metal has a tendency to crack under restraint. availability, aluminium is widely used in industry and
To avoid such cracking, welding is done with low heat elsewhere. Its density is about one-third that of steel or
input to minimise dilution. Also suitable preheat is used copper. Its corrosion resistance is derived from the
to slow down the cooling rate of the deposited metal and tenacious, refractory aluminium oxide film, which
thereby reduce its hardness. develops rapidly on the clean surface in air. The thermal
In the case of nickel and nickel-alloy electrodes de- and electrical conductivities of aluminium are four times
signed specifically for welding cast iron, the weld-metal those of steel. Aluminium is therefore used often as a
picks up carbon well above the solubility limit. As the substitute for copper in electrical equipment. It is non-
weld-metal solidifies, the excess carbon is rejected as magnetic. Its coefficient of thermal expansion is twice
graphite. This reaction is accompanied by volume in- that of steel. Its melting point of 649°C is half that of steel.
crease, which results in reduced residual stresses in the It has low strength but excellent ductility. The good
ductility is retained even at cryogenic temperatures. The Heat-treatable alloys: These alloys contain Cu, Mg, Zn
strength of the metal can be increased by suitable alloying, and Si singly or in various combinations. In the annealed
cold working, heat treating or by their combination. condition, their s t r e n g t h d e p e n d s on the alloy
Most of the above-mentioned properties have a composition in the same manner as non-heat-treatable
deleterious effect on the weldability of aluminium. The alloys. The solid solubility of the four elements, singly or
refractory oxide film which is difficult to avoid is in combinations, increases with increasing temperature.
insoluble in the metal and interferes with the wetting Therefore, when these heat-treatable alloys are solution
action of molten filler metals. Suitable means have to be heat-treated, quenched and then subjected to natural or
used to remove or break up the oxide layer during welding artificial aging, their strength improves considerably.
and brazing. Because of its high thermal conductivity, Additional strength can be obtained by cold working
higher heat inputs are required for fusion welding, and before or after aging.
additional preheating is required in the case of thick Heat-treatable Al-Mg-Si alloys are easily weldable.
sections. The low melting point and the fact that Many of the Al-Zn alloys have poor weldability, but there
aluminium, unlike steel, does not show a dark-red colour are others with Mg additions which have been specifically
as it approaches the melting point, make it difficult to developed for welding. Al-Cu alloys require special
judge the approach of the molten state during welding. procedures for welding and the welded joints show lower
This can lead to an oversize welded zone or burn- ductility.
through. Because of its high coefficient of thermal When a heat-treatable alloy, which has been solution-
expansion, one would expect pronounced expansion heated and aged naturally or artificially, is arc welded,
during welding. But factually, identical aluminium and some loss of strength and ductility occurs in the heat-
steel components register the same total expansion, affected zone. The HAZ shows three distinct areas: (a)
because of the comparatively low melting point of area of solution-annealed material, (b) partially annealed
aluminium which balances out the expansion factor. area, and (c) overaged area. If such a welded joint is
Aluminium and its alloys are used in wrought form solution heat-treated and aged, original strength is
(i.e. as rolled products) or as castings. The products are restored, but not ductility. If only aged, there is no
identified by standard codings or designations based on improvement in strength and ductility becomes quite low.
alloy compositions. In addition, a temper designation is For best mechanical properties, the alloy must be welded
used to indicate the condition of the product such as: as- in the annealed condition, then solution is heat-treated
fabricated, annealed, solution heat-treated, solution heat- and aged.
treated and naturally aged, etc. The weldability of an It was stated earlier that the tenacious, refractory oxide
aluminium alloy is determined by both these factors. film on the joint surface prevents good wetting between
Aluminium alloys are divided into two classes: non- it and the filler metal. The welding arc can remove this
heat-treatable and heat-treatable. oxide film as the welding proceeds, but for best results,
Non-heat-treatable alloys: These alloys contain elements precleaning by mechanical or chemical means is
such as Si, Mn and Mg which increase the strength of advisable. Suitable mechanical m e t h o d s are wire
aluminium by solid solution or through formation of brushing, scraping, filing and scrubbing with steel wool.
dispersed phases. Of these, Mg is the most effective and For brushing, stainless steel brushes must be used and
hence Al-Mg alloys have relatively high strength in the they must be maintained scrupulously clean. The
annealed condition. All non-heat-treatable alloys harden chemical cleaners are usually acidic or alkaline solutions
on cold-working, becoming at the same time stronger and or proprietory compounds.
less ductile. They can be restored to their original strength, It is equally important to keep the joint surfaces per-
ductility and softness by annealing. The annealing fectly free from moisture, grease, oils and paints. If
procedure depends on the alloy composition and temper. present, they are decomposed by the welding arc and
When a work-hardened alloy is welded, the strength in hydrogen is released which causes porosity in the welds.
the heat-affected zone gets reduced to the level of the Degreasing by dipping in a suitable solvent or by steam
annealed metal. cleaning is recommended for critical fabrications. Com-
Pure aluminium, Al-Mn alloys and Al-Mg alloys are ponents are best cleaned before assembly and fit-up.
easily welded by TIG and MIG processes. Al-Si alloys Hydrogen is the main cause of porosity in aluminium-
can also be welded by these processes, but special alloy welds. Hydrogen has high solubility in molten Al,
procedures have to be used and lower ductility may be but very low solubility in solid Al. If moisture, oil or
obtained. grease are present in the flux coating or on the filler wire
surface or on the joint surface, the hydrogen generated DEVELOPMENT OF HIGH STRENGTH WELDABLE
by the welding arc is dissolved in the molten weld pool AI-ALLOYS
and released during solidification. The high freezing rate, The role of various alloying elements in producing
especially in the case of MIG welding, prevents the high strength Al-alloys has been explained in the fol-
hydrogen from escaping out of the weld-metal and the lowing description of non-heat-treatable and heat-treat-
result is porosity. Unlike in steel, an extremely small able alloys. Among the first category, Al-5% Mg alloy is
amount of hydrogenous material can cause considerable known to be the strongest. This alloy also possesses good
porosity in Al and Al-alloy welds. weldability, because the element magnesium helps to
For fusion welding of aluminium and its alloys, improve this property. The strongest Al-alloys are heat-
preheating must be avoided as far as possible, as it in- treatable or age-hardenable materials. Among them Al-
creases the width of the HAZ and reduces mechanical Cu alloys have very high strength. In fact, the greatest
properties. Only in the case of thick sections preheat be- exploitation of Al alloys has been in the Al-Cu system,
comes necessary. Even then the duration of heating should where tensile strength as high as 432 N/mm 2 (28 T/sq.
be as short as possible, and the preheating temperature in.) is being specified and the alloys widely used in the
should not exceed 150°C. In the case of alloys containing 3 aircraft industry. Unfortunately, these alloys have poor
to 5.5% Mg, the preheat temperature must not exceed 120°C weldability, since the element copper aggravates
and the interpass temperature must never rise above 150°C. microfissuring in the weld-metal and the HAZ.
Weld-metal cracking can occur in a specific alu- A significant feature of Al alloys is that their
minium alloy if the choice of the filler metal is not correct, weldments are prone to stress corrosion embrittlement
and the result is that either the weld-metal or the HAZ in service; and hence in the development of high strength
has low ductility or strength at elevated temperature. This alloys, special care must be taken against stress corrosion.
phenomenon is referred to as hot shortness. Generally speaking, increase of total alloy content
To reduce the tendency for intergranular cracking in increases strength and at the same time increases
the HAZ, it is advisable to use a filler metal, whose melting propensity to stress corrosion. Copper addition increases
point is equal to or below that of the base metal. In other stress corrosion resistance, but also accentuates hot
words, one should use a filler metal of higher alloy cracking in welding.
content. For example, aluminium containing 0.6% Si is Among heat-treatable alloys, Al-Zn-Mg systems are
very likely to crack in the welded joint, if a filler metal of known to have the best combination of strength and
matching composition is used. The cracking can be weldability. Zinc contributes to strength while
eliminated by using a filler metal containing 5% Si. Since magnesium contributes to weldability. The level of Zn is
the latter has lower melting point, it remains more plastic 3-5% and the level of Mg is 1-3%. Higher magnesium
than the base metal and yields during cooling to relieve level makes extrusion difficult, and if both Zn and Mg
the contraction stresses that might cause cracking. levels are kept too high, microfissuring and stress
In many applications, Al-5% Mg filler wire gives corrosion susceptibility increase. The commonly used
welds of good strength and crack resistance. Al-Mg alloys alloy has 4% Zn and 2% Mg. Small Cu additions are
should never be welded with Al-Si alloy filler wires, sometimes made to improve corrosion resistance. Such
because of the formation of magnesium-silicide eutectic, Cu-bearing alloys need extreme caution in welding.
which decreases ductility and causes cracking. Similarly, Mechanical properties of some of the well-known alloys
Mg and Cu should not be allowed to be present together are given in Table 10.40.
in an aluminium weld. This means that Al-Mg filler wires In the case of most heat-treatable alloys, part of the
should never be used to weld Al-Cu alloys and Al-Cu HAZ is degraded to such a level that optimum
filler wires should not be used to weld Al-Mg alloys. In mechanical properties can be obtained only by applying
welding an Al-Si alloy base metal with an Al-Si filler a full heat treatment after welding, i.e. solution heating,
wire, their alloy contents must be so arranged that the quenching and aging. With Al-Zn-Mg alloys, however,
weld-metal has Si content either below 0.5% or above joint properties can be improved by only natural or
2%. With Si content between these limits, the crack artificial aging after welding, which hardens the whole
sensitivity increases markedly. In the case of Al-Mg of the weld zone. This is explained by the fact that the
alloys also, the Al-Mg weld-metal must not have Mg welding heat causes reversion instead of overaging in
content between 0.5 and 2% for a similar reason. In these the HAZ of these particular alloys. Natural aging may
examples, dilution factor must also be taken into account take approximately 28 days. Artificial aging involves
and properly controlled during welding. heating schedules such as:
Table 10.40: Mechanical properties of various aluminium alloys

Alloy Composition Condition 0.2% proof stress, U.T.S. Elongation on
No. % en —— o/
D\J 11 HI l, /O
N/mm2 N/mm2
1 4.5 Mg O 130 280 16
M 130 280 12
2 1.0 Mg-1.0 Si-0.8 Mn W 115 200 12
WP 240 295 7
3 4.3 Cu-0.7 Si-0.7 Mn w 245 385 11
WP 330 410 6
4 5.7 Zn-2.7 Mg-1.0 Cu WP 425 485 6
5 3.5 Zn-2.2 Mg w 200 370 17
WP 255 415 15
6 4 Zn-2 Mg WP 340 400 12
7 4.5-5 Zn-1.2 Mg w 230 385 22
WP 310 385 17
8 4.3 Zn-1.8 Mg w 245 340 16
WP 340 385 12
9 4.0-4.5 Zn-2.5-3.0 Mg WP 385 448 14
0 = annealed M = as-manufactured
W = solution-treated and naturally agedWP == solution-treated and precipitation-treated
1) 24 hours at 130°C temperature. They therefore compete with stainless steels

2) 6 hours at 160°C in cryogenic applications.
3) 16 hours at 100°C+ Al alloys such as AS ME SB 209 5083 in the annealed
2 hours at 150°C+ condition, strengthened by the solid solution hardening
2 hours at 180°C of 4.5% Mg and 0.7% Mn, possess good welding
Al-Zn-Mg alloys are welded with TIG and MIG characteristics as the microstructure is fairly insensitive
processes and the filler wire commonly used is Al-5% to welding heat. Toughness and ductility are high; low-
Mg alloy. To avoid cracking, weld-metal must have less temperature transition characteristics are absent and the
than 0.25% Cu and more than 2Vi% Mg. Sometimes toughness is retained, or even increased, at cryogenic
zirconium addition is made to the filler wire. Zr has a temperatures.
marginal beneficial effect on stress corrosion resistance, Large liquefied natural gas (LNG) storage containers,
but it prevents cracking because this element inhibits typically of 50,000 m 3 capacity, have been manufactured
recrystallisation and is a grain refiner. Its action has been with alloy 5083 (0.6 Mn, 4.45 Mg, 0.15 Cr) or its variation
found to be superior to the combined action of Ti and B. 5086 (0.45 Mn, 4.0 Mg, 0.15 Cr). Welding these containers
Filler metals of Al-Zn-Mg alloy systems are avoided as with 5183 grade filler wire (0.5-1.0 Mn, 4.3-5.2 Mg, 0.05-
they increase risk of cracking and aggravate stress 0.25 Cr) has presented no particular problems.
corrosion. The demand for larger capacity air separation units
Al-Zn-Mg alloys have been used widely in India and has led to a growth in the fabrication of Al alloys, espe-
the U.K. for military bridges, because of their high strength cially the non-heat-treatable 3.5% Mg types. Automatic
to weight ratio, good corrosion resistance and good and semi-automatic MIG are currently used extensively
weldability. On the continent and in the U.S.A., they have for this application. However, material thicknesses in
been used for applications ranging from lamp-posts to excess of 40 mm require careful consideration when de-
pressure vessels including tipper trucks, dumper trucks, signing and applying joint configurations for use with
road tankers and railway rolling stock. these welding techniques.
High radiographic and mechanical testing standards
Al Alloys for Cryogenic Applications throw even greater burdens on the features already
Al and its alloys exhibit adequate ductility and associated with the MIG welding of aluminium, such as
toughness as well as other relevant properties at the necessity for correct cleanliness, preheat temperature
t e m p e r a t u r e s as low as -270°C or liquid h e l i u m and torch angle, in order to reduce or eliminate the risk of
weld pool puckering, lack of fusion, oxide inclusions 280 amp; arc voltage range, 26-28 V; 0.9 k j / m m min
and porosity. average heat input; all welding positions.
The main pressure column-shell seams are usually
welded with fully automatic MIG. Joint configurations Copper and Copper Alloys
d e p e n d e n t u p o n material thickness vary from These materials are used primarily for resistance to
conventional square-edge and single-vee configurations oxidation and corrosion by specific process fluids (such
to double-U or modified-Us, on heavier wall sections. as fresh and salt water, ammonia-free alkaline solutions
Multi-run and alternate-side sequencing techniques are and many organic chemicals), for their thermal and
employed to minimise distortion and to improve the electrical properties, for their good formability and
mechanical properties via a desirable grain refining effect. malleability, and their ability to be used in clad form for
Particular attention is paid to inter-run cleanliness. structural purposes. Applications include reaction
Mechanical tools are used to remove oxide and to ensure pressure vessels, evaporators, distillation columns, heat
smooth weld bead profiles for subsequent passes. exchangers, pumps, valves, pipework, and numerous
Longitudinal seams for both column and dished-end ancillary components used in process industries such
fabrications generally require some form of restraint (e.g. as chemicals and petrochemicals, gases, iron and steel,
strongbacks) to help prevent any excessive distortion. plastics and textiles, fertilisers, food and beverage.
When using permanent backing strips on circumferential Copper and copper alloys can be divided into the
closing seams, the semi-automatic MIG process is used following major groups:
to complete the root passes, allowing better control and 1) Pure and low-alloy bearing coppers
defect-free fusion to be achieved. 2) Cu-Zn alloys (brasses)
Typical welding conditions for fully-automatic MIG 3) Cu-Sn alloys (phosphor bronzes)
welding are: material specifications, ASME SB 209 5083- 4) Cu-Al alloys (aluminium bronzes)
0; material thickness r a n g e , 40-50 m m ; p r e h e a t 5) Cu-Si alloys (silicon bronzes)
temperature range, 50-150°C; interpass temperature 6) Cu-Ni alloys (cupronickels)
range, 50-200°C; filler-wire size and specifications, 2.4 7) Cu-Ni-Zn alloys (nickel silvers).
mm diameter AWS ER 5556 (closely similar to 5183 Their commercial names, nominal compositions and
grade); spray transfer; 100% argon at 30-^141/min; 400- weldability ratings are given in Table 10.41.
480 amp welding current; 30-34 V arc voltage, 300-450
m m / m i n travel speed; 2.2 k j / m m minimum heat input; 1) Pure and low-alloy bearing coppers
flat welding position. Pure copper has minimum copper content of 99.3%
Semi-automatic MIG is used extensively to weld and is available as fire-refined grade, electrolytic tough-
nozzles and many other sub-assemblies. High standards pitch grade and phosphorus-deoxidised grade. The first
are possible in the vertical and horizontal-vertical two are oxygen-bearing and contain residual oxygen in
positions. As the welding currents are lower than those the form of copper-cuprous oxide eutectic globules
for the automatic process, greater operator skill is distributed throughout wrought forms of these grades.
required, especially in monitoring and controlling the While it has no detrimental effect on mechanical
necessary preheat and interpass temperatures. Nozzle properties or electrical conductivity, welding causes the
joints on heavy wall materials usually have a double copper oxide to concentrate in the grain boundaries,
modified J preparation with a 50° inclusive angle, resulting in significant loss of strength and ductility.
allowing easy access for welding and interpass cleaning. Oxygen-bearing coppers are therefore considered
Dye-penetrant testing is used regularly as an added unweldable. They can, however, be brazed using proper
precaution for checking root runs and grind backs, procedures and filler metals, though with brazing some
especially as heavy wall thicknesses increase the rate of embrittlement will occur.
cooling and subsequently, the susceptibility to cracking, Phosphorus-deoxidised copper which has 0.004 to
Typical welding conditions for semi-automatic MIG 0.04% residual P is extremely ductile and fatigue-
welding are: material specifications, ASME SB 209 5083- resistant, and can be readily welded and brazed. P in
0; material thickness range, 10-50 mm; preheat such low amounts does not interfere with welding.
temperature range, 50-150°C; interpass temperature range, Low-alloy bearing coppers carry small amounts of
50-150°C; filler-wire size and specification, 1.6 m m various elements to enhance specific properties. Silver is
diameter AWS ER 5556; spray transfer; shielding gas type added to oxygen-free copper to increase high-temperature
and flow, 100% argon at 20-221/min, current range, 180- strength, without affecting electrical conductivity and
Table 10.41: Nominal compositions and weidability ratings of copper and copper alloys
Alloy name Nominal composition Weidability
0/
/o
MMA TIG MIG
OF and ETP coppers
Oxygen-free copper (OF) 99.95 Cu NW G G
Electrolytic tough pitch copper (ETP) 99.90 Cu, 0.04 0 2 NW F F
Deoxidised coppers
Phosphorus-deoxidized copper, low-P (DLP) 99.9 Cu, 0.008 P NW E E
Phosphorus-deoxidized copper, high-P (DHP) 99.9 Cu, 0.02 P NW E E
Beryllium coppers
High-conductivity beryllium copper, 0.6% 96.9 Cu, 0.6 Be, 2.5 Co F F F
High-strength beryllium copper, 1.7% 98.3 Cu, 1.7 Be, G G G
High-strength beryllium copper, 1.9% 98.1 Cu, 1.9 Be G G G
Low-zinc brasses
Gilding, 95% 95 Cu, 5 Zn NW G G
Commercial bronze, 90% 90 Cu, 10 Zn NW G G
Red brass, 85% 85 Cu, 15 Zn NW G G
Low brass, 80% 80 Cu, 20 Zn NW G G
High-zinc brasses
Cartridge brass, 70% 70 Cu, 30 Zn NW F F
Yellow brass, 65% 65 Cu, 35 Zn NW F F
Muntz metal, 60% 60 Cu, 40 Zn NW F F
The brasses
Admiralty brass 71 Cu, 28 Zn, 1 Sn (d) NW F F
Naval brass 60 Cu, 39.25 Zn, 0.75 Sn (d) NW F F
Special brasses
Manganese bronze A 58.5 Cu, 39 Zn, 1.4 Fe, 1 Sn 0.1 Mn NW F F
Aluminium brass, arsenical 77.5 Cu, 20.5 Zn, 2 Al (0.06 As) NW F F
Nickel silvers
Nickel silver, 65-10 65 Cu, 25 Zn, 10 Ni NW F F
Nickel silver, 65-18 65Cu,17Zn,18Ni NW F F
Phosphor bronzes
Phosphor bronze, 1.25% E 98.7 Cu, 1.3 Sn, (0.2 P) F G G
Phosphor bronze, 5% A 95 Cu, 5 Sn, (0.2 P) F G G
Phosphor bronze, 8% C 92 Cu, 8 Sn, (0.2 P) F G G
Phosphor bronze, 10% D 90 Cu, 10 Sn, (0.2 P) F G G
Aluminium bronzes
Aluminium bronzes D, Sn-stabilised 89 Cu, 7 Al, 3.5 Fe (0.35 Sn) G G E
Aluminium bronzes D 91 Cu, 6-8 Al, 1.5-3.5 Fe, 1 max Mn G G E
Silicon bronzes
Low-silicon bronze B 98.5 Cu, 1.5 Si F E E
High-silicon bronze A 97Cu,3Si F E E
Copper nickels
Copper nickel, 10% 88.6 Cu, 9-11 Ni, 1.4 Fe, 1.0 Mn G E E
Copper nickel, 30% 70 Cu, 30 Ni E E E
£ = excellent, G = good, F = fair, NW = not weldable

MhAA: Manual metal-arc welding, TIG: Tungsten-inert gas welding
MIG: Metal-inert gas welding
weldability. When lead is added, it gets dispersed Zn content between 30 and 40%. The 60-40 brass is known
throughout the matrix as fine, discrete particles resulting as Muntz metal.
in enhanced machinability and reduced ductility. Addition of Zn to Cu increases strength, hardness and
Tellurium and sulphur similarly improve machinability ductility. It lowers the melting point, density, and the
by forming hard stringers in the matrix. These free- electrical and thermal conductivities. The colour of the
machining coppers have poor weldability, because the alloy also changes with zinc content. With increasing Zn,
inclusions induce cracking during welding. the alloy comes closer to gold in colour.
Coppers alloyed with small amounts of beryllium, Most brasses are solid-solution alloys with a single-
chromium or zirconium can be precipitation-hardened phase called alpha. Brasses with 40% or more zinc have
by suitable heat treatment in the same manner as heat- two microstructural phases called alpha and beta. The
treatable Al-alloys discussed earlier. Exposure to welding beta-phase improves the hot-working property of brass.
heat causes the heat-affected zone to soften by annealing During welding, Zn tends to evaporate from the molten
or overaging. Special welding procedures are necessary metal and thereby reduces weldability. Brasses therefore
to obtain adequate joint efficiencies. For best results, these cannot be arc welded. In oxycetylene welding with a
alloys must be welded in the annealed condition and reducing flame, an oxide film forms over the molten pool,
then precipitation-hardened. which suppresses the volatilisation of zinc, provided the
Copper is sometimes alloyed with small amounts of weld-metal is not overheated. Brass gas welding rods are
cadmium. Such an alloy has good electrical conductivity designed to contain about 40% Zn, so that the weld-metal
and can be strengthened by cold working. When has a good proportion of the hard and strong beta-phase,
cadmium-copper is oxyacetylene welded, the cadmium which being soft and ductile at elevated temperatures,
oxide formed can cause trouble, unless a flux containing prevents weld cracking.
sodium fluoride and fused borax is used.
Beryllium-copper is available in two grades. One of 3) Copper-tin alloys
them contains approximately 0.5% Be and 2.0% Co and Cu-Sn alloys are commercially called phosphor
has good electrical conductivity. The other one is pre- bronzes instead of plain bronzes, because they contain
cipitation-hardening type, which contains approxi- about 0.03% phosphorus, which is added during casting
mately 2.0% Be. In welding these alloys, a tenacious film to deoxidise them. The tin content varies from 1 to 10%.
of beryllium oxide can form and interfere with wetting/ ; They are primarily single-phase alloys, but their relatively
and fusion, unless the joint faces and the surrounding' high strength is sometimes ascribed to the precipitation of
base metal surfaces are kept perfectly clean. On the other an inter-metallic compound at room temperature. They
hand, Be contributes to better weldability by lowering are also tough, hard and fatigue-resistant. Cold work
the melting point, increasing the fluidity of the weld- improves these properties further.
metal and decreasing thermal conductivity. Unless Phosphor bronzes are readily weldable, and arc
welding and postweld heat treatment are done with care, welding processes are preferred to oxyacetylene welding.
weld cracking can occur. The heat-treatable type should For manual arc welding, standard phosphor-bronze
not be welded in the precipitation-hardened condition, electrodes are used. and MIG welding processes using
because of the danger of cracking in the HAZ. If multi- standard phosphor-bronze filler rods are equally suitable.
pass welding is to be done, it is preferable to weld the The base metal must be preferably welded in the stress-
alloy in the overaged condition than in the solution heat- relieved or annealed condition, because in the stressed
treated condition. Low Be-Cu can be more easily welded condition phosphor bronzes are prone to hot cracking.
with a filler metal having higher Be content. Be-Cu
repairs are better carried out with aluminium bronze 4) Copper-aluminium alloys
manual electrodes, or with Si bronze rods using the TIG They are better known as aluminium bronzes and
process. contain up to 15% aluminium with small additions of
iron, manganese, nickel and tin. Alloys with less than
2) Copper-zinc alloys 8% Al have single-phase microstructure called alpha,
These alloys, popularly known as brasses, have the while those with more than 9% Al are two-phase alpha-
largest commercial application among copper alloys. Zinc beta alloys, which are capable of either a martensitic or a
is soluble in copper to the extent of 37% at room temperature eutectoid type of transformation. The alpha-alloys cannot
and 32.5% at 925°C. Low-zinc brasses have Zn content be hardened by heat treatment and hence they can be
ranging between 5 and 20%, while high-zinc brasses have welded without preheating. Also these alloys with less
than 8.5% Al are hot-short and may develop cracks in film formed on the molten pool protects the weld-metal
the HAZ when welded under restraint. from further oxidation. Yet, for best results its formation
The two-phase alloys containing 9.5 to 11.5% Al can must be avoided by using clean, oxide-free joint surfaces
be quenched and tempered in the same manner as steels and appropriate shielding.
to obtain the best combination of strength and ductility.
Quenching from about 498°C in water and oil produces 6) Copper-nickel alloys
a martensitic-type structure, which is tempered by heating Copper and nickel are completely solid soluble in all
in the 220 -342°C temperature range. If the proportion of proportions (see Fig. 9.4), but in these alloys, nickel ranges
Al is increased beyong 11.5%, the heat-treated alloy between 5 and 30%. The alloys with 10% nickel (90/10
becomes harder and less ductile. These two-phase alloys type) and with 30% nickel (70/30 type) are most
have generally better weldability, because they have a commonly used in welded fabrications. They have
wider plastic range than the alpha alloys. medium tensile strengths, which increase with nickel
Single-phase Al bronzes with less than 7% Al are content. They are ductile and moderately tough, but they
difficult to weld, because they are hot-short and develop have low thermal and electrical conductivities. Cu-Ni
cracks in the HAZ. 7% Al alloys are sometimes used for alloys are easily weldable if many residual elements
welded fabrications such as heat exchangers, vessels and which cause embrittlement and promote hot cracking
pipings. They must be welded using correct procedures are restricted. S and P should be held below 0.02% and
to avoid cracking. Pb s h o u l d be absent. Sufficient deoxidiser or
At elevated temperatures, aluminium bronzes develop desulphuriser (such as Mn) must be present in the filler
a film of aluminium oxide on the surface, which makes metal to prevent weld-metal cracking.
them resistant to oxidation and scaling. The same
aluminium oxide interferes with welding unless it is 7) Copper-nickel-zinc alloys
scrupulously removed earlier. These are primarily copper-zinc alloys or brasses to
which nickel is added to increase strength, improve
5) Copper-silicon alloys corrosion resistance and to give the final alloy a silvery
These alloys, called silicon bronzes, find extensive appearance. Hence they are popularly called nickel
use in industry because of their high strength, good silvers. Two types are generally available: (1) single-
corrosion resistance and easy weldability. Their silicon phase alpha-alloy with 65% Cu and (2) two-phase alpha-
content varies between 1.5 to 4% and they are further beta alloy with 55-60% Cu. Their weldability is similar
alloyed with small quantities of iron, manganese, tin or to that of the brasses described earlier.
zinc. The first two increase tensile strength and hardness, Preheating: Copper and some of its alloys possess high
while the latter two improve the fluidity of molten bronze, thermal conductivity, with the result that during welding
which is a desirable feature in casting and welding. heat is rapidly conducted away from the joint to the
Silicon is soluble in copper to the extent of 3.6% at surrounding base metal. Such heat loss from the joint
room temperature and 5.3% at 815°C. It is used both as a depends on the thickness or mass of the base metal and
deoxidiser and a major alloying element. In addition to its temperature. When the heat loss is too rapid, it slows
improving strength, ductility, and malleability, silicon down the welding speed and may even cause lack of
increases hardness and work-hardening response. As fusion and porosity. Heat loss can be minimised by
silicon content increases to about 1%, there is a gradual preheating the base metal.
drop in ductility; then it increases to a maximum value at The optimum preheat temperature depends on the
4% silicon. type of alloy, metal thickness or mass and the heat input
Though these alloys have good weldability, they tend rate of the welding process. TIG or MIG welding with
to be hot-short at elevated temperatures. To eliminate the helium instead of argon as a shielding gas requires less
risk of cracking in welding, they should be cooled rapidly preheat due to the high arc voltages developed in the
through the hot-short temperature range. They should process. MIG welding with argon requires less preheat
preferably be welded in the stress-relieved or annealed than TIG welding with argon, because it generates more
condition, and sudden temperature rise during welding heat in the arc. These effects are shown graphically in
should be avoided. Fig. 10.27 for different plate thicknesses of pure copper.
Silicon bronzes develop a silicon oxide film when Copper alloys with lower thermal conductivities
heated above 200°C. This film interferes with brazing require lower preheat temperatures than those indicated
unless a suitable flux is used. In fusion welding, the oxide in the figure. Silicon bronzes and copper-nickel alloys
Table 10.42. Heating should be gradual and the weldment

should be held at the temperature for at least one hour.
Weldments of hardenable copper alloys are sometimes
annealed to homogenise the variable structures formed
in the weld-metal and the HAZ, so that the resulting
uniform metallurgical structure will respond satisfacto-
rily to the subsequent heat-treatment. Annealing tempera-
tures for various alloys are indicated in Table 10.43. It is
observed that these temperatures are considerably higher
than those used for stress-relieving. If annealing is to be
carried out, a separate stress-relief treatment is not nec-
essary, as stresses are relieved rapidly at annealing tern-
Table 10.42: Stress-relieving temperatures for

copper alloys
PLATE THICKNESS. rwt» Alloy Temp. °C
Red brass 290
Fig. 10.27: Effects of process, shielding gas, and metal
Admiralty brass 290
thickness on preheat requirements for welding copper
Naval brass 260
Aluminium bronze 345
have such low thermal conductivities that they need no Silicon bronze 345
preheating. Copper-nickel alloys 540
For welding heavy sections of phosphor bronze with
ECuSn-C electrodes, a preheat and interpass temperature Table 10.43: Annealing temperatures for Cu-alloys
of about 200°C is required to counteract the tendency of
the molten weld-metal to flow sluggishly because of its Alloy Temp, range °C
wide melting range. For TIG welding of phosphor bronze, Phosphor deoxidised copper 370-650
preheat in the range of 175-200°C is adequate. These Beryllium copper (2.0% Be) 775-800
temperatures should never be exceeded as the welds are Beryllium copper (2.5% Co-0.5% Be) 900-940
likely to develop hot cracks. Red brass 425-730
Yellow brass 425-700
Low zinc brasses need preheating in the temperature Muntz metal 425-590
range of 100-300°C for TIG and MIG welding. High zinc Admiralty brass 425-590
brasses require lower preheat with the TIG process and Naval brass 425-590
no preheat with the MIG process. Phosphor bronze 480-675
Aluminium bronzes with less than 10% Al require Aluminium bronze (7% Al) 610-900
preheat only in thick sections, but the preheat temperature Aluminium bronze (10% Al) 590-650
should not exceed 150°C. After welding, the job should Silicon bronze 480-700
be air cooled to room temperature. For Al-bronzes with Aluminium brass 425-590
Copper-nickel, 10% 590-815
10-13% Al, a preheat and interpass temperature of 260°C
Copper-nickel, 30% 650-815
is recommended for thick sections. After welding the job Nickel silver 590-760
should be cooled rapidly.
Postweld heat treatment: The need to heat-treat a welded peratures. Annealing should not be overdone as it can
fabrication depends on the base metal composition and cause grain growth, which results in reduced tensile
service application. Precipitation-hardening type alloys strength and other unfavourable effects.
are heat-treated after welding to remove the variations of
hardness in the weld area, which occur as a result of Nickel and Nickel Alloys
aging or overaging caused by the w e l d i n g heat. Nickel, like copper, has a face-centred cubic crystal
Weldments which are prone to stress corrosion cracking structure at all temperatures up to its melting point. Since
in certain environments must be stress-relieved. Stress- it has good corrosion resistance, nickel is used in food
relieving temperatures for some of the copper alloys which processing plants, caustic handling equipment, chemical
are particularly susceptible to this defect are given in containers and other similar applications.
Nickel is alloyed with other metals to improve

Table 10.44: Nominal compositions of nickel alloys
mechanical properties and to make it resistant to
corrosion and oxidation at room and high temperatures. Alloy Composition
In all nickel alloys, strengthening takes place by solid Nickel 200 99.5 Ni, 0.06 C, 0.25 Mn, 0.15 Fe
solution and the elements which contribute most strongly Monel 400 66.0 Ni, 31.5 Cu, 0.90 Mn, 1.35 Fe
are Al, Cr, Mo and W. Other elements which contribute Monel 404 55.0 Ni, 44.0 Cu
to strength less markedly are Co, Cu,Fe, Ti and V. Monel R-405 66.0 Ni, 31.5 Cu, 1.35 Fe
Improved strength at elevated temperatures is achieved Nichrome V 76.0 Ni, 20.0 Cr, 1.0 Fe, 2.0 Mn
by Mo and W additions. Nichrome 57.0 Ni, 16.0 Cr, 25.0 Fe, 1.0 Mn
There are other nickel alloys in which further Inconel 600 0.04 C, 76.0 Ni, 16 Cr, 7 Fe
strengthening occurs by precipitation-hardening. These Inconel 601 0.05 C, 60.5 Ni, 23 Cr, 14 Fe, 1.35 Al
Inconel 625 0.05 C, 61.0 Ni, 22 Cr, 9 Mo, 3 Fe, 0.2 Al, 0.2
alloys are s o l u t i o n - t r e a t e d by h e a t i n g to a h i g h
Ti, 4 Cb
temperature, and then quenched. Upon reheating to an Incoloy 800 0.04 C, 32 Ni, 20 Cr, 46 Fe
appropriate temperature for a specific time (i.e. aging), Incoloy 825 0.03 C, 42 Ni, 21.5 Cr, 30 Fe, 3 Mo
precipitation occurs and the alloy becomes stronger and Hastelloy B 0.05 C, 18 Mo, 5 Fe, 0.4 V, bal. Ni
harder. The most important phase in the precipitate for Hastelloy N 0.04 C, 7 Cr, 16.5 Mo, 5 Fe, 0.5 Co, bal. Ni
strengthening effect is what is termed gamma-prime, Hastelloy W 0.12 C, 5 Cr, 24.5 Mo, 5.5 Fe, 2.5 Co. bal. Ni
based on the compound Ni, Al. Aluminium is thus the Hastelloy C 0.08 C, 15 Cr, 16 Mo, 6 Fe, 5 W, bal. Ni
major contributor to hardening, while Ti and Cb have a Monel K-500 65 Ni, 29.5 Cu, 0.6 Mn, 1.0 Fe, 0.6 Ti, 2.73 Al
moderating influence. Nimonic 80 A 76 Ni, 19.5 Cr, 1.6 Al, 2.4 Ti
Inconel 706 0.03 C, 41.5 Ni, 16 Cr, 40 Fe, ,0.2 Al, 1.75 Ti,
2.9 Cb
Solid-solution Alloys Alloy 901 0.03 C, 42.5 Ni, 12.5 Cr, 36 Fe, 6 Co, 0.2 Al,
These include five major groups as follows: 2.8 Ti
1) Nickel-copper alloys containing about 30 to 45%
Cu. They are well-known under the trade name monels. An important point to be borne in mind during the
Examples are Monel 400, Monel 404 and Monel R-^05. welding of Ni and its alloys is that they are extremely
2) Nickel-chromium alloys containing 16 to 20% Cr. susceptible to embrittlement by S, P and low-melting-
They may be sometimes alloyed with Fe to the extent of point metals like lead. These elements are likely to be
25%. The most familiar examples are those known by the present in oil, grease, paint, marking inks and crayons,
trade names Nichrome V, Nichrome, Inconel 600, Inconel cutting fluids, shop dirt, etc. Joint surfaces and the
601 and Inconel 625. adjoining base metal must therefore be thoroughly
3) Nickel-iron-chromium alloys having 20-45% Ni, cleaned by vapour degreasing or swabbing with a solvent
13-22% Cr and balance Fe. Typical examples are Incoloy such as acetone. Foreign matter embedded in the base
800 and Incoloy 825, both recognised by their trade metal must be removed by grinding, abrasive blasting,
names, and swabbing with 10% HC1 solution, followed by a
4) Nickel-molybdenum alloys with 16-28% Mo and thorough water wash. Oxides must be removed from the
lesser percentages of Cr and Fe. Typical examples are joint surfaces by wire brushing.
Hastelloy B, Hastelloy N and Hastelloy W. Sulphur and phosphorus form low-melting eutectics
5) Nickel-chromium-molybdenum alloys. Typical with nickel, which segregate at the grain boundaries
example is Hastelloy C. during weld-metal solidification and cause hot-cracks.
Sulphur embrittlement can be prevented by addition of
Precipitation-hardenable Alloys Mn and Mg, which preferentially combine with sulphur
They include three major groups: and prevent the formation of nickel sulphide. B and Zr
1) Nickel-copper alloys. Typical example is Monel K-500. which are sometimes added to the alloy to improve hot
2) Nickel-chromium alloys. Typical example is malleability also cause hot cracking in the fusion zone
Nimonic 80A. as well as the HAZ.
3) Nickel-iron-chromium alloys containing 40-45% Carbon, which is used in small amounts as an
Ni, 13-15% Cr, 30-40% Fe plus small amounts of Al and interstitial strengthening element, has its own role in the
Ti. Typical examples are Inconel 706 and Alloy 901. weldability of nickel. During welding, the carbon in the
Nominal compositions of the above-mentioned alloys HAZ goes into solution. During service at temperatures
are given in Table 10.44. in the region of 300°C the carbon precipitates as graphite
at the grain boundaries and reduces ductility in the HAZ. Nickel and its alloys are weldable by common arc-
Hence carbon is kept at a very low level in nickel, while welding processes. Preheat is not required while
in the case of nickel alloys, carbide-forming elements such interpass temperatures during multi-pass welding is not
as Cr, Cb and Ti are used. allowed to exceed 100°C. High heat input during welding
Some Ni-Cr and Ni-Gr-Fe alloys are prone to carbide is not desirable, especially in the case ofprecipitation-
precipitation or sensitisation in the weld HAZ, which hardenable alloys. High heat input may give rise to
leads to intercrystalline corrosion in certain service carbide precipitation, undesirable phases or compounds
environments, in the same manner as in the case of or other metallurgical phenomena. These, in turn, may
austenitic stainless steels. The same remedy is used here, lead to cracking or loss of corrosion or both.
namely, either the alloy is stabilised with Ti or Cb or an Hence, while all alloys are readily weldable by TIG
extra-low-carbon grade of the particular alloy is used. and plasma-arc (PAW) processes, many of them and
Silicon also causes hot cracking, especially in high especially the precipitation-hardenable types should not
nickel-chromium alloys. High Ni-Cr castings with high be welded by the manual metal-arc (MMAW), MIG and
Si levels must be welded with filler metals containing Cb submerged-arc (SAW) processes. This is indicated in
to prevent the occurrence of hot-cracks in the weld-metal. Table 10.45. Precipitation-hardenable alloys should
Table 10.45: Arc welding process applicable to some nickel alloys

Alloy* MMAW TIG / PAW MIG SAW
Commercially pure nickel
Nickel 200 X X X X
Nickel 201 X X X X
Solid-solution nickel alloys
Monel 400 X X X X
Monel 404 X X X X
Monel R-405 X X X
Hastelloy X X X X
Nichrome V X X
Nichrome X X X
Hastelloy G X X X
RA333 X
Inconel 600 X X X X
Inconel 601 X X X X
Inconel 625 X X X X
Carpenter 20Cb3 X X X
Incoloy 800 X X X
Incoloy 825 X X X
Hastelloy B X X X
Hastelloy C X X X
Hastelloy N X
Precipitation-hardenable nickel alloys
Monel K-500 X X
Waspaloy X
Rene 41 X
Nimonic 80A X
Nimonic 90 X
M252 X
Udimet 500 X X
Inconel 718 X X
Inconel X-750 X
Inconel 706 X
Alloy 901 X
^Several of these are registered trade names.
preferably be welded in the solution-treated condition. toughness down to -240°C. Ti has low density and some
When welded in this condition, a part of the HAZ becomes of its alloys can compete with Al alloys and steels for
aged and develops welding stresses. This area may aerospace applications. It has low thermal expansion
develop cracks during subsequent heat-treatment, the and thermal conductivity, which contribute to the good
phenomenon being referred to as strain-age cracking. weldability of this metal.
Such cracking is aggravated by Al, while it is controlled Ti, like Al, forms a stable, tenacious oxide layer on the
or totally eliminated by substituting Cb for a large portion surface even at room temperature. The affinity for oxygen
of the Al. Ob retards the aging reaction, because of the a n d the surface oxide thickness increase w i t h
sluggish formation of the Cb precipitate compared to the temperature. At temperatures above 650°C, its oxidation
rapidly forming Al precipitate, and thereby maintains resistance is reduced to such an extent that the metal
sufficient ductility in the HAZ to cope with the high must be shielded from air to avoid contamination and
welding stresses without cracking. resultant embrittlement by oxygen and nitrogen.
It is advisable, therefore, that these alloys are given a Hydrogen also embrittles the metal.
solution-treatment after welding to eliminate residual At r o o m t e m p e r a t u r e , Ti h a s a close-packed
stresses, and then hardened by an aging heat treatment. hexagonal crystal structure called alpha (a) phase. This
Some of the Ni-Cr and Ni-Fe-Cr precipitation- phase is retained up to 880°C; above this temperature an
hardenable alloys contain relatively large amounts of Al allotropic change to the body-centred cubic structure
and Ti and are therefore highly prone to strain-age called beta ((3) phase occurs.
cracking. They are therefore considered unweldable. Alloying elements added to Ti are either alpha-phase
If a precipitation-hardenable alloy is welded in the stabilisers or beta-phase stabilisers, depending on the
hardened condition, resolutioningand overaging occur phase in which they have the highest solid solubility.
in the H A Z w i t h c o n s e q u e n t loss of mechanical Aluminium, having greatest solid solubility in alpha-
properties. To restore them, a postweld heat treatment phase, is a powerful alpha-phase stabiliser and its
becomes necessary. addition raises the phase transformation temperature.
For welding these alloys, TIG process with stringer Common examples of beta-phase stabilisers are Cf, Mo
bead technique is njost suitable. In multi-pass welding, a n d V. They decrease the p h a s e transformation
if any oxide film is formed on the weld surface, it must be temperature. With large additions of beta stabilisers, the
removed by grinding or sand blasting. If these films are beta-phase can remain stable down to room temperature
not removed, they become thick enough to interfere with and below.
weld fusion and to produce laminar-type oxide stringers Alloying e l e m e n t s also influence the rate of
along the weld axis, which give rise to stress corrosion transformation from one phase to the other.
cracks in service. Ti alloys can be alpha alloys, beta alloys or alpha-
Suitable filler metals for some important alloys are beta alloys, depending on their composition and thermal
indicated below: history. The alpha-to-beta phase ratio is significant in
Alloy Filler metal AWS class many ways, considering that the presence of beta-phase
Nickel 200 .. ERNi-1 (1) enhances grain refinement and hence strength, (2)
Monel 400 .. ERNiCu-7 improves hot working property, (3) makes the alloy
Monel 404 respond to heat treatment, (4) improves toughness except
Inconel 600 .. ERNiCr-3 at very low temperatures, and (5) decreases weldability
Hastelloy G .. ERNiCrMo-1 and creep strength in the case of some alloys.
Hastelloy C-4 .. ERNiCrMo-7 It is extremely difficult to avoid the contamination of
Hastelloy B-2 .. ERNiMo-7 Ti during welding by the elements C, H, N and O. If the
Hastelloy W .. ERNiMo-6 heated metal is not protected from the atmosphere, it picks
Monel K-500 .. ERNiCu-8 up O because of the preferential affinity it has for this
element. H and O are picked up if moisture is present in
Titanium and Titanium Alloys the shielding gas. H and C contamination occurs if
Pure Ti is a silver-white metal with a relatively low residual oils, cleaning agents and dirt are present on the
tensile strength but excellent ductility. With suitable joint surfaces.
alloying, the strength properties can be increased but C, H, N and O form interstitial solid solutions with Ti.
there is simultaneous decrease in ductility. Ti and some G, N and O are alpha-stabilisers, while H is beta-stabiliser.
of its alloys possess good strength, ductility and The mechanical properties of a commercial grade of pure
Ti depends on the interstitial contents of these elements, capacity to resist weld cracking under restraints imposed
especially oxygen. Among the four, N is the most effective in a welded fabrication, and (c) ability of the welded joint
strengthener of Ti, next comes O. In alpha-beta Ti alloys, to attain the strength and toughness of the parent metal.
the strengthening effect on a particular alloy is Ti and its alloys are usually joined by the TIG process.
d e t e r m i n e d largely by the relative a m o u n t and For 2.5 mm thick and thinner plates, square butt joints
distribution of the alpha-phase. are made and welded without filler metal. For thicker
Hydrogen is almost insoluble in alpha Ti at room plates, a V-groove joint is used and filled with additional
temperature, but dissolves to some extent at temperatures filler metal.
above 300°C With slow cooling, the dissolved hydrogen Plasma-arc welding is a faster process than TIG and
precipitates throughout the alpha grains and at the grain can be used on thicker sections. A 12.5mm thick Ti-alloy
boundaries resulting in loss of toughness. On the other plate can be welded in one pass with this process, using
hand, if the Ti is quenched, hydrogen precipitates as fine a square butt joint and keyhole technique.
particles and has little effect on toughness. Subsequent For thicknesses more than 3.2 mm, MIG process is
aging, however, causes agglomeration of the hydrides preferred. This process becomes increasingly economical
and decreases toughness. as the thickness increases beyond 12.5 mm.
Since hydrogen has high solubility in beta Ti, alpha- Before commencing welding, joint surfaces of Ti and
beta alloys can dissolve a substantial amount of hydrogen Ti-alloy components must be cleaned with suitable
without the formation of hydrides. However, if the amount solvents and dried. Ordinary tap water must not be used
is too high, embrittlement can occur at low strain rates. for rinsing, because chlorides and other cleaning residues
TI alloys can be divided into five major groups as: on Ti can cause stress corrosion cracking when the
1) Commercially pure Ti component gets heated during welding. Light oxide
2) Alpha alloys coating in the vicinity of the joint must be removed by
3) Near-alpha alloys pickling, water rinsing and drying. Scale formed at high
4) Alpha-beta alloys temperatures must be removed mechanically by grit
5) Beta alloys blasting or similar method followed by pickling, to ensure
Important alloys of each group and their weldability complete removal of scale and contaminated metal on
ratings are given in Table 10.46. The ratings are with the surfaces.
respect to TIG, MIG and plasma-arc processes which are The joint edges may need further special operations
mostly used for these alloys. The criteria for weldability such as draw filing or wire brushing prior to final
rating are: (a) capacity to produce defect-free welds, (b) cleaning, to prevent possible weld porosity. Sheared joint
edges require special care in this respect.
Table 10.46: Weldability ratings of titanium alloys
Cleanliness of Ti and Ti-alloy filler metals is even more
Alloy Rating* important than in the case of Al and Al-alloys. Every
Commercially pure grades A care m u s t be taken to remove the last traces of
Alpha alloys contaminants such as inclusions, dirt, oil and drawing
Ti-0.2Pd A compounds from the filler metal surface. This is more
Ti-5Al-1.5An B critical with MIG wires, with their relatively large surface-
Ti-5Al-2.5SnELI A area-to-volume ratios.
Near-alpha-alloys We shall now discuss the weldability of three typical
T1-8AI-IM0-V A alloys:
Ti-6Al-2 Cb-1 Ta-0.8 Mo A (1) Commercially pure Ti, (2) Ti-0.2 Pd alpha alloy,
Ti-6Al-4 Zr-2 Mo-2 Sn B
and (3)Ti-6 A1-4V alpha-beta alloy
Alpha-beta alloys
Ti-6A1-4V B 1) Commercially pure Ti: This material, available in
Ti-6A1-4VELI A several grades ranging in purity from 98.5 to 99.5 Ti, has
Ti-7AMMo C excellent weldability. It is very important to maintain the
Ti-7Al-6V-2Sn C iron content in the weld metal below 0.05% by using
T1-8M0 D filler wire with very low iron content, and by avoiding
Beta alloys iron contamination d u r i n g joint preparation and
Ti-13V-ll Cr-3 Al B welding. Use of steel wire brushes must be avoided.
*A—excellent, B—fair to good, C—limited to special applications, U—For best mechanical properties, the metal is welded in
welding not recommended. the annealed condition. Annealing consists of heating to
650-760°C for between 30 minutes to two hours, depend- 12.5 and 19 mm. Usually the largest nozzle size which
ing on thickness and cooling in air. Complicated compo- will give good accessibility and visibility is used. It is
nents may require stress-relieving by heating to 480- preferable to use nozzles which provide laminar flow of
590°C and cooling in air. If a component is annealed after the shielding gas, because even slight turbulence in the
welding, it need not be separately stress-relieved, because gas flow induces air to mix with the gas.
annealing automatically relieves residual stresses. Secondary shielding is provided by a trailing shield,
2) Ti-0.2 Pd alloy: Its weldability characteristics are a typical design of which is shown in Fig. 10.28. It
similar to those of pure Ti. It is normally welded in the consists of a metal chamber clamped to the torch nozzle.
annealed condition, because welding has little effect on The inert gas flows through a porous metal diffuser
the mechanical properties of the annealed material. In screen and impinges itself over the hot weld area. Its
this alloy also, special p r e c a u t i o n s against iron length varies according to the heat input and rate of arc
contamination must be taken. Complex welded structures travel. The shield is made to cover also the width of the
may need stress-relief treatment as in the case of pure Ti. HAZ on either side of the weld bead.
Secondary shielding can be incorporated into the
3) Ti-6 AI-4V alloy: This alloy can be strengthened by
welding fixture as shown in Fig. 10.29. The hold-down
solution heat t r e a t m e n t followed by aging. For
bars of the fixture are provided with inert gas passages.
solutioning, the alloy is heated to 900-970°C and then
Shielding gas flowing from the torch and the hold-down
quenched. It is then aged at between 480 and 510°C for
bars into the channel formed by the bars drives out the
about 12 hours.
air and protects the area behind the weld pool.
The annealing temperature range for this alloy is 705-
790°C, while the stress-relieving temperature range is
540-595°C. This alloy has the best weldability among WELDING DIRECTION
alpha-beta alloys. It is usually welded in the annealed

condition or the solution-treated and partially aged
INERT GAS INLET
condition. The aging is completed through a postweld
4
stress-relief treatment.
METAL CHAMBER
Filler metals can be of two grades as shown below, in
which maximum limits are set on carbon, oxygen,
hydrogen and nitrogen. Unalloyed filler metal may also
m• \ \ \ yj
be usedfor improved ductility of the weld joint. 51
POROUS METAL
ERTi-6 A1-4V ERTi-6 A1-4V-1
DIFFUSION SCREEN
Carbon 0.05 max. 0.04 max.
TORCH NOZZLE
Oxygen 0.15 max. 0.10 max.
Hydrogen 0.008 max. 0.005 max. Fig. 10.28: Torch trailing shield for gas-shielded arc
Nitrogen 0.020 max. 0.0012 max. welding of titanium and other reactive metals
PROTECTION DURING WELDING SHIELDING GAS FLOW

Gas-shielded processes are mostly used for welding THROUGH TORCH NOZZLE
HIELDING GAS DIFFUSER
TI alloys. These processes are normally designed to shield
only the molten weld pool and adjacent base metal from
I HOLD-DOWN BARS
SHIELDING GAS PORT
the atmosphere. In the case of Ti and its alloys, such
shielding is totally inadequate, because of the sensitivity
of these alloys to embrittlement by oxygen and nitrogen
when heated above 260°C. During welding, inert gas
shielding must be provided for (1) the molten weld pool
and adjacent base metal (primary shielding); (2) the |<— BASE
solidified weld-metal and HAZ which are still hot after
the arc has proceeded (secondary shielding), and (3) the
T ~T~ t
F SHIELDING GAS PORT F
heated back side of the weld joint (backing).
Primary gas shielding is provided by the TIG torch Fig. 10.29: Secondary inert gas shielding incorporated
nozzle or the MIG gun nozzle, whose size ranges between in the welding fixture
Backing gas shielding is achieved through temporary purged with inert gas, in volume equal to about seven
backing bars or rings provided with gas passages. Typical times the volume of the chamber. A collapsible chamber
backing bars for butt and fillet welds are shown in Fig. is first collapsed and then flow-purged with the inert
10.30. The bar can also be incorporated into the welding gas. A collapsible chamber needs less gas than a rigid
fixture shown in Fig. 10.29. The backing bar serves to one.
abstract heat from and accelerate the cooling of the weld, In using these metal or plastic chambers, it is
which reduces gas consumption. The bar is made of necessary to monitor the inert atmosphere continuously
copper or stainless steel and may be water-cooled. or from time to time. For continuous monitoring, a heated
Welding of critical fabrications such as those used in tungsten filament may be placed in the chamber.
aerospace applications is most successfully carried out Discoloration of the filament indicates that the purity of
in a specially designed vacuum-purged welding chamber. the inert gas has been affected. In a flow-purged chamber,
The chamber is thoroughly evacuated and then backfilled the atmosphere can be tested either by suitable gas
with argon or helium. The chamber is equipped with analysing equipment or by welding a piece of scrap
glove ports to enable the welder to handle the torch, filler titanium, and pulling the torch gradually away from the
metal and the job without admitting air into the chamber. weld pool. The weld-metal colour gives an indication of
Viewing ports are also provided, so that the welder can contamination. The pure weld is silver-bright but with
see his operation. Welding may be carried out without increasing contamination, the colours obtained are: light-
supplying shielding gas through the torch, but in the straw, dark-straw, light-blue, dark-blue, grey-blue, grey
case of heavy and long welds additional gas supply and white loose powder. Light-straw is generally
through the torch helps to improve shielding. acceptable, but for critical applications bright-silver
Welding in such a chamber has certain disadvantages. colour must be insisted upon.
Visibility, movement and accessibility are restricted. If
the job is large, the chamber has to be unloaded after Joining of Dissimilar Metals
each job is completed and repurged for Welding the next Dissimilar metals mean either (1) those that are
job. Such a procedure is time consuming and causes chemically different (example: steel, copper, aluminium)
wastage of expensive inert gas. or (2) those that are metallurgically different (example:
Transparent flow-purged plastic chambers, rigid or mild steel, stainless steel).
collapsible, are to be preferred for one-off jobs or when Dissimilar metal joining by brazing or by solid-state
the jobs are large or intricate. A rigid chamber is flow- welding processes (such as friction welding, diffusion
BACKING-GAS INLET BACKING-GAS INLET
Fig. 10.30: Weld backing bars to provide inert gas shielding: (A) For groove welds; (B) For corner or fillet welds
bonding, explosive welding) is feasible in most combi- Metallurgically, the weld-metal in dissimilar joints
nations. When fusion welding processes are used, suc- will be either a single-phase or a mixture of two or more
cess depends on several factors. phases. A phase can be an intermetallic compound
If the melting points of the two members being welded (example: Cu-Al compound), an interstitial compound
differ by more than 100°C, solidification and shrinkage (example: iron carbide or titanium carbide), or a solid
of the metal with the higher melting point will induce solution (example: copper and nickel). An intermetallic
stresses in the other metal, which is only partially compound is to be avoided. The role of an interstitial
solidified and therefore weak, and cause it to rupture. In compound is variable. Solid solutions are desirable
some cases, the problem is solved by buttering the joint because they help to produce a good weld. Copper and
surface of the higher melting base metal with a filler metal, nickel are readily weldable to each other because they
whose melting point lies between those of the two base form a series of solid solutions when melted together.
metals. The final weld is made between the buttered face The soundness and mechanical properties of a dissimilar
and the other base metal. The filler metal should be one weld joint are determined by the number, type, relative
which is suitable for the lower-melting base metal. If a proportion and metallurgical arrangement of the phases
high-melting filler metal is used, there will be excessive present in the weld-metal.
dilution from the low melting base metal. Mild steel and Weldability of a dissimilar metal combination can
aluminium are examples of two metals having widely often be predicted by examining the phase diagram of
different melting points. the metals concerned. If intermetallic compounds are
When two metals of significantly different thermal formed, one must use solid-state welding processes
conductivities are being welded, the heat input can be instead of the conventional fusion welding processes.
right for one and either too less or too much for the The following chart indicates the degrees of weldability
other. This is because of the differing rates of heat of various metal combinations.
conduction from the molten pool to the two base metals. For dissimilar metal joints, preheat and postweld heat
Proper heat balance is achieved in this case by directing treatment procedures have to be selected with care. When
the arc at the higher thermal-conductivity base metal or only one of the two metals needs to be preheated, one
by selectively preheating it. Compared to mild steel, must make sure that the preheat can be applied
copper and aluminium have very high thermal independently to the proper side of the joint. The postweld
conductivity, while austenitic steel and monel have much heat treatment can be problematic when the heat applied
less. to one metal is harmful to the other. A typical example is
If the two metals being welded have different the welding of an age-hardenable Ni-Cr alloy to
coefficients of thermal expansion, the resultant joint has unstabilised austenitic stainless steel. When aging
high internal stresses. This factor becomes more serious treatment is given to the Ni-Cr alloy, it will make the
when the joint is exposed to cyclic temperatures in service. latter sensitive to intergranular corrosion. The problem
A common example is the joining of 7SA Cr-1 Mo steel to can be overcome by using a stabilised stainless steel.
type 304 or 316 austenitic stainless steel in thermal power Alternatively, the joint face of the age-hardenable Ni-Cr
plant. The coefficient of thermal expansion of the former alloy component can be buttered with a non-age-
is about 25% less than that of the latter. In this case, it is hardenable Ni-Cr alloy and it can be age-hardened
advisable to choose a filler metal whose thermal independently. The buttered surface can then be welded
expansion coefficient is intermediate between the two. to the stainless steel component.
The filler metal must also have sufficient ductility to The problem of metallurgical incompatibility can
tolerate the high stresses. An Inconel filler metal satisfies sometimes be solved by the use of a filler metal shim or
these conditions. welding each of the metals to a compatible transition
An important factor is dilution of the filler metal by piece. For example, 321SS filler shims are used in between
the two different base metals. In fusion welding, the filler Hastelloy X and SAE 8620 steel; Hastelloy B shims can
metal and some portions of the base metal melt together be used for welding of 304 SS to monel as well as for A -
and form the weld pool. Hence the filler metal must be 286 SS to SAE 4140 steel.
capable of alloying readily with the base metals and Calculation of weld-metal composition: For successful
produce a weld-metal that is ductile and crack-free. If welding, one must be able to calculate and control the
complex phases or intermetallic compounds are formed weld-metal composition resulting from dilution from the
in the resultant weld-metal, the joint will have poor two different base metals. The average composition can
mechanical properties and will be unsuitable for service. be calculated from (1) the compositions of the filler metal
and the base metals, and (2) the ratio of the volumes of base metals is 25%. The compositions of the three alloys
base metals melted to the total volume of the weld-metal. are given below:
These volumes are equivalent to the fused areas on the
cross-section of a weld, which can be measured. These Element Cr-Mo steel Type 317 S.S. Filler metal
are shown as A and Bin Fig. 10.31. The method of Cr 19 20
calculating the dilution by two base metals—A and B, Ni 13 72
and the total dilution when filler metal FM is used is as Mo 0.5 4
follows: Fe 93.5 58.5
Dilution by metal A, % = f x 100 Alloying elements in the weld-metal arc calculated

A + B + (FM1+Fm) thus:
Cr% = 0.10(5)+0.15(19)+0.75(20) = 18.35
Dilution by metal B,% = B Ni% = 0.15(13)+0.75(72) = 55.95
— x 100
A + B + (FM1+Fm) Mo% = 0.10(0.5)+0.15(4) = 0.65
The percentage of dilution depends on the joint
Total dilution, % A+B configuration, welding process and welding procedure.
— x 100 A single deeply penetrating heavy pass and the root pass
A + B + (FM1+Fm)
in a heavy joint have the highest dilution, while passes
in the central portion of a multi-pass, multi-layer weld
For calculating the average percentage of an alloying will suffer least dilution. The dilution ranges of common
element in the diluted weld-metal, the following equation arc welding processes are as follows:
is used: Manual metal-arc 25-40%
E = (DA)(EA) + (D A )(E A ) + (l-D A B )(E m ) Submerged-arc 25-50%
L
WM
Average % of element E in the weld-metal TIG 20-40%
E = % of element E in base metal A MIG (spray transfer) 25-40%
*A =
EB = % of element E in base metal B MIG (dip transfer) 15-30%

E = % of element E in the filler metal
Effects of dilution with various elements on different
D A = % dilution by base metal A, expressed as filler metals have been pointed out below:
decimal*
% dilution by base metal B, expressed as decimal*
1) Mild steel and low-alloy steel filler metals
D. % total dilution by base metals A and B,
a) Fe—any extent of dilution is acceptable.
expressed as decimal* (* 10% = 0.10,15% = 0.15,
b) Ni, Cr, C, Si and Mn—refer to Schaeffler diagram to
20% = 0.20)
avoid the formation of martensite.
To explain this calculation, let us take an example in These filler metals should not be used on stainless
which 5 Cr-V£ Mo steel is welded to type 317 stainless steels, nickel and nickel alloys and cast heat-resisting
steel with AWS class ERNi Cr3 filler metal. Let us assume alloys.
10% dilution by base metal A and 15% dilution by base c) Cu—small amount is acceptable. With increasing
metal B, which means that the total dilution by the two amount, the weld-metal develops hot cracks. This means
vFM2 JB
Fig. 10.31: Parameters for calculating weld-metal dilution in a dissimilar metal joint
Al 2
Gold 1 5
Beryllium 5 2 5
Cobalt 3 5 2 5
Copper 2 2 1 5 2
Iron 3 5 2 5 2 2
Magnesium 5 2 5 5 5 5 3
Molybdenum 3 5 2 5 5 3 2 3
Columbium 4 5 4 5 5 2 5 4 1
Nickel 2 5 1 5 1 1 2 5 5 5
Platinum 1 5 ' 1 5 1 1 1 5 2 5 1
Rhenium 3 4 4 5 1 3 5 4 5 5 3 2
Tm 2 2 5 3 5 2 5 5 3 5 5 5 3
Tantalum 5 5 4 5 5 3 5 4 1 1 5 5 5 5
Titanium 2 5 5 5 5 5 5 3 1 1 5 5 5 5 1
Tungsten 3 5 4 5 5 3 5 3 1 1 5 1 5 3 1 2
Zirconium 5 5 5 5 5 5 5 3 5 1 5 5 5 5 2 1 5
Ag Al Au Be Co Cu Fe Mg Mo Cb Ni Pt Pb Sn Ta Ti V
2. Very desirable (Solid solubility in all combinations) 2. Probably acceptable (complex structure may exist)
3. Use with caution (insufficient data for proper evaluation) 4. Use with extreme caution (no data available)
5. Undesirable combinations (intermetallic compounds are formed)
that these filler metals must not be used on copper and c) Fe—dilution up to 30% is acceptable.
copper alloys. d) C, Si and Mn—any amount of dilution which
normally occurs inpractice is acceptable.
2) Austenitic stainless steel filler metals
a) Fe, Ni, Cr, Si, Mn and C—refer to Schaeffler diagram, 5) Monel (Ni-Cu) filler metal
to get final weld-metal structure of austenite with 4-10% a) Ni and Cu-any extent of dilution is acceptable.
ferrite. If the structure is fully austenitic, the deposit is b) Cr—dilution higher than 5% must be avoided.
sensitive to hot cracks. If the structure is martensitic, the c) Fe—tolerable dilution varies with the process and
deposit is hard and brittle. These filler metals must not type of filler metal. For example, MMA welding with
be used for nickel and nickel alloys and heat-resisting Monel 190 electrode can accept dilution with iron up to
high-Ni castings. 30%, while TIG and MIG welding with monel filler metal
b) Cu—dilution with copper must be avoided. This can tolerate dilution only up to 15%. If these limits are
means that these filler metals must not be used on copper exceeded, hot-cracking occurs. With Monel 187 electrode,
and copper alloys. only 5-10% dilution can be tolerated.
d) C, Si and Mn-Carbon content of the weld-metal
3) Straight chromium steel filler metals must not exceed 0.4%, while silicon content must not be
a) Ni, Cr, Fe, C, Si and Mn—refer to Schaeffler dia- above 1.5%. Dilution by manganese is beneficial.
gram. They are used for welding straight chrome steels
of matching composition. With higher Cr content, the 6) Fe-Ni-Cr type filler metals
filler metal is prone to grain growth and porosity; hence a) Incoloy 65 filler metal—Incoloy 135 electrode
austenitic stainless steel or nickel-alloy filler metals are i) Ni-any extent of dilution is acceptable.
sometimes preferred for welding high-Cr steels. These ii) Cr-Cr content of the weld-metal must not exceed
filler metals should not be used for welding dissimilar 35%.
metals. iii) Fe—dilution with iron should not exceed 10-15%.
b) Cu—same remarks as for austenitic stainless steel This means that Incoloy filler metals can be used for
filler metals apply here. welding to highly alloyed stainless steels, but not to plain
4) Pure nickel Filler metal carbon steels.
a) Ni and Cu—any extent of dilution with these two iv) C, Si and Mn-Si content of the weld-metal should
elements is acceptable. not exceed 1%. Other elements are under control in
b) Cr—dilution up to 33% is acceptable. common practice.
b) Inconel 82 and 92 filler metals—Inconel 182 or

Table 10.47: Base metals for piping
Incoweld-A electrode
i) Ni—any extent of dilution is accepted. Type of steel Abbreviation
ii) Cr—Cr content of the weld-metal must not exceed Carbon C-Steel
35%. Carbon-1/2% Moly C-Mo
iii) Fe—For manual welding electrodes, iron dilution W/o Chrome - M% Moly Vi Cr - Vi Mo
up to 50% can be tolerated. For TIG/MIG filler metals, 1% Chrome - Vi% Moly 1 Cr - Vi Mo
the dilution level must not exceed 25-30%. VA% Chrome - Vi% Moly VA Cr - Vi Mo
iv) C, Si and Mn-Si content of the weld-metal should 2% Chrome - lA% Moly 2 C r - 1 / 2 Mo
2V4% Chrome - 1% Moly 2VA Cr - 1 Mo
not exceed 1%. Other elements are under control in
5% Chrome - Vi% Moly 5 Cr - Vi Mo
common practice. 7% Chrome - %% Moly 7 Cr - Vi Mo
9% Chrome - 1% Moly 9 Cr - 1 Mo
Welding of Dissimilar Steels 18% Cr - 8% Ni (Type 304)
In thermal power stations, oil refineries, etc., various 18% Cr - 10% Ni-2.5% Mo (Type 316) 18-8 Stainless
combinations of carbon steel, low-alloy steels and 18% Cr - 10% Ni-Cb (Type 347)
stainless steels have to be welded, mainly in the form of 18% Cr - 12% Ni-Ti (Type 321)
pipes. In Table 10.47 these steels are listed along with
their abbreviations used for convenience. 25 chromium-12 nickel (type 309)
There are seven classes of filler metals or electrodes (alternate)
which can be used for joining these steels or their various For non-cyclic service below 430°C:
combinations. These are: 25 chromium-12 nickel (type 309)
A. Carbon steel (such as E7016 or E7018) 25 chromium-20 nickel (type 310)
B. Carbon-!* Mo (such as E7016-AI or E7018-AI) 29 chromium- 9 nickel, (type 312) or
C. VA Cr-44 Mo (such as E8016-B2 or E8018-B2) 75 nickel-16 chromium-7 iron.
D. 2VA Cr-1 Mo (such as E9016-B3 or E9018-B3) In Table 10.48 we have indicated which of the above
E. 5 Cr-44 Mo (such as E502-15) filler metal classes has to be used for various base metal
F. 9 Cr-1 Mo combinations.
G. For cyclic service and wherever post-heating is to Among the filler metals of class G, type 309 (25Cr-
be used: 12Ni) is most suitable for joining austenitic stainless steel
75 nickel-16 chromium-7 iron (Inconel type) to carbon or low-alloy steel, because it has high-enough
For non-cyclic service above 430°C: alloy contents to prevent martensite formation in the
75 nickel-16 chromium-7 iron (preferred) weld-metal after total dilution by the two base metals,
25 chromium-20 nickel (type 310) and to retain some amount of ferrite in the final austenitic
(alternate) weld-metal to prevent hot cracking. The undiluted 25-
Table 10.48: Filler metals for various base metal combinations
o o
o
o 1 o . o o
sA r-H
1
(T3
A
1 o r-i -4->
SI
I
3!
U u u u A CD
C/5 u u U U u 00
00
i
u u ^ r-i r-* CN rsl LO [>s ON T-H
C-Steel — A A A A,B B,C B,C CD CD CD G

C-Mo A B B B B B,C B,C CD CD CD G
Vi Cr-Vi Mo A B C C C C C CD CD CD G
1 Cr-fc Mo A B C C C C C CD CD CD G
VA Cr-4* Mo A,B B C C C C C CD CD CD G
2 C r - ^ Mo B,C B,C C C C D D D D D G
2VA Cr-1 Mo B,C B,C C C C D D D D D G
5 Cr-fc Mo CD C,D CD CD CD D D E E E G
7 Cr-% Mo C,D CD CD CD CD D D E F F G
9 Cr-1 Mo C,D CD CD CD CD D D E F F G
18-8 Stainless G G G G G G G G G G
12 weld-metal has about 6-10% ferrite, much more than carbon will diffuse from the base metal into the weld-
what is present in the 18-8 weld-metal. metal during postweld heat treatment or service at
One can estimate the microstructure of a weld-metal temperatures above 427°C. This phenomenon is known
joining stainless steel to a carbon or low-alloy steel by as carbon migration. Carbon migration occurs to a larger
referring to the Schaeffler diagram shown in Fig. 10.32. extent with austenitic stainless steel weld-metal than
This figure also shows the positions of weld-metals with ferritic steel weld-metal, because of the greater
deposited by various Ador Welding stainless steel solubility of carbon in the former. It should be taken care
electrodes including lnox-D2 (Type 309). If lnox-D2 is of because it affects long-time stress-rupture strength. It
used for joining mild steel and the dilution is 20%, the can also cause fatigue failure in the decarburised band
composition of the diluted weld-metal can be deduced adjacent to the weld during cyclic temperature service.
by joining the lnox-D2 point with the M.S. point on the Chromium stainless steel (type 4xx) to steel
diagram, and then marking the point X such that it divides 1) For welding straight chrome steel to carbon steel,
the line in the ratio of 20:80. This point X represents the mild steel filler metal must be preferred.
weld-metal composition. It is 19/9 and contains 2) For welding straight chrome steel to a low-alloy
approximately 8% ferrite, thus indicating a tough and steel, filler metal of the same composition as the low-
ductile weld-metal. alloy steel must be preferred.
While joining austenitic stainless steel to carbon steel, 3) For welding together two steels of different Cr
it is advisable to butter the carbon steel with 309 type contents, one may use filler metal having Cr content equal
filler metal. Then it is a question of joining stainless steel to that of either steel.
to stainless steel, and any conventional stainless steel 4) For (3) above, austenitic stainless steel filler metal
filler metal can be used. Also the problems associated can be used, provided one of the steels is not a pressure
with joint restraint and dilution by carbon steel are vessel.
eliminated. Preheat may be necessary when a chrome steel is being
Carbon steel or low-alloy steel filler metal should never welded to any other steel. It is not possible to give general
be deposited on stainless steel, as the resulting weld- rules on preheat procedure. An experienced welding
metal is hard and brittle. metallurgist must be consulted on this point. Correctly
Different grades of stainless steel are often welded welded joints between a chrome steel and an alloy steel
together to meet various service requirements. can stand service conditions at temperatures up to 220°C
Recommended filler metals for welding these (800°F).
combinations are given in Table 10.49. For optimum properties in the HAZ, the joint edges of
Carbon migration: When an alloy-steel filler metal high both the steels must be first buttered with E 309 type
in chromium is used to weld a carbon or low-alloy steel, electrode using suitable preheat. The buttered stainless
2 4 6 8 10 12 14 16 18 20 22 24 26 28 32 34 36 38 40
C H R O M - EQUIVALENT-%Cr + % M o + 1.5%Si + 0.5 x %Nb
Fig. 10.32: Schaeffler diagram with positions of weld-metals of Ador Welding stainless steel electrodes
Table 10.49: Filler metals suitable for welding joints between dissimilar austenitic stainless steels
Suitable filler metals (listed in no preferred order; prefix ER omitted)

Base metal B (type of steel being welded to base metal listed in the first column)
Base 347,347H
Metal A(a) 304L 308L 309 309S 310 310S 316,316H 316L 317 321,321H 348,348H
304,304H,305 308L 308L308L,309 308,309 308,309,310 308,309,310 308,316 308,316 308,316,317 308 308
304L 308L 308,309 308,309 308,309,310 308,309,310 308,316 308L, 316L 308,316,317 308L, 347 308L, 347
308 308,309 308,309 308,309,310 308,309,310 308,316 308,316 308,316,317 308 308,347
309 309 309,310 309,310 309,316 309,316 309,316 309,347 309,347
309S 309,310 309S, 310S 309,316 309S, 316L 309,316 309,347 309,347
310 310,316 310,316 310,317 308,310 308,310
310S 316 316 317 308,310 308,310
316,316H 316 316,317 308,316 308,316,347
316L 317 316L 316L, 347
317 308,317 308,317,347
321, 321H 308L, 347
(a) The H suffix indicates a grade used for tubing for high-temperature service.
steel surfaces can then be joined with E 308 type electrode limited strength between Al and mild steel or stainless
without preheat. steel can still be obtained by aluminising the steel surface
Nickel alloys to steel: It is easy to weld this combination by dipping the steel component into a molten aluminium
using nickel base filler alloys, which have good ductility bath held at 366 to 374°C (1275 to 1300°F) and then
and can tolerate appreciable dilution by iron. Weld-metal brazing or TIG welding with an Al-alloy filler metal.
deposited with pure nickel or nickel-chromium manual Strong joints can be obtained between Al and steel
electrodes can tolerate iron dilution up to approximately with solid-state processes such as friction welding or
40%, while the same weld-metal deposited by TIG, MIG explosion welding. A transition piece can be fabricated
or submerged-arc process can tolerate this dilution up to with either of these processes, and the ends of this piece
25%. can be welded to aluminium and steel components using
Dilution by Cr must also be controlled. Pure nickel the common fusion welding processes.
weld-metal can tolerate Cr dilution up to 30%, while the Aluminium alloys to copper: These are also incompatible
total Cr content in Ni-Cr weld-metal should not be metals due to the formation of brittle intermetallic
allowed to exceed 30%. compounds. Joints of limited strength can be obtained by
Care must also be taken that the relatively high levels initially applying a layer of silver or silver alloy to the
of sulphur and phosphorus present in the steel are not copper surface, and then welding with an Al or Al-Si
transferred to the nickel-base weld-metal by dilution. One filler metal without penetrating through the silver coating
must bear in mind that sulphur and phosphorus cause on the copper.
hot cracking in nickel and nickel alloys.
Copper alloys to steel: Copper and steel can be welded Cladding by Weld Surfacing
together by any of the common arc welding processes Clad plate consists of carbon steel or low-alloy steel
using silicon bronze (CuSi-A) or aluminium bronze plate to which a corrosion-resistant or abrasive-resistant
(CuAl-AX) filler metal. Dilution with the steel must be metal is completely bonded, surface to surface. The latter,
kept minimum. called cladding, is generally straight Cr steel, austenitic
A better method to join copper to mild or stainless stainless steel, copper or copper alloy, nickel or nickel
steel is to butter the copper component with nickel filler alloy or titanium. Cladding thickness is generally 10 to
metal and then make a joint between nickel and steel by 20% of the total clad plate thickness.
using any of the filler metals mentioned in the previous There are several methods of preparing clad plate,
paragraphs. including weld surfacing, and these are mentioned in
When Cu-Ni alloy (monel) is to be welded to carbon chapter no. 4. For weld surfacing, the most commonly
steel or stainless steel using a Cu-Ni alloy filler metal, used process is submerged-arc welding using strip or
dilution of the weld-metal by Fe or Cr should be kept to multiple wires to provide the-clad deposit. Submerged-
the lowest minimum. In fact, the total of the two elements arc cladding has been described in detail in chpter no. 4.
should not exceed 5%. A better method is to first butter Other welding processes which are also used are:
the steel surface with nickel filler metal and then complete manual metal-arc; GTAW with manual or automatic wire-
the joint with Ni-Cu or Cu-Ni filler metal, the choice feed; GMAW in semi-automatic or fully-automatic form,
being dictated by service requirements. Alternately, the in some cases with auxiliary wire-feed; and plasma-arc
Cu-Ni alloy face is first buttered with nickel filler metal welding with powders or energised wires.
and then the weld joint is made with the same filler metal. The best welding process is one which is capable of
It is easy to weld aluminium bronze to mild steel or giving minimum base metal dilution, high deposition
stainless steel using aluminium bronze filler metal. rate, and the required clad thickness in minimum number
Copper alloys to nickel alloys: Welding of these of layers. From this point of view, submerged-arc strip
combinations presents no problem, because Cu and Ni cladding with the cold strip addition is the best. The
are mutually soluble in each other. Pure nickel, Ni-Cu or dilution of the standard SA strip cladding process is 30-
Cu-Ni filler metals must be used. If a Ni alloy contains 40%, but this is brought down to 10-15% with cold strip
Cr or Fe or both, pure nickel filler metal must be used. It is addition.
advisable to first butter the Cu or Cu-Ni joint face with With the GMAW process operating in the spray-
nickel filler metal to restrict dilution by Cu. transfer mode, the dilution levels are around 30-50%.
Aluminium alloys to steel: These are incompatible With auxiliary cold-wire feeding, the dilution is reduced
metals. A fusion weld joining Al to steel is brittle due to to 20%, while the deposition rate is increased by 80-
the formation of brittle intermetallic phases. Joints of 100%.
Plasma-arc hot double-wire welding is also well denoting the weld chemistry of standard S.S. electrodes.
suited. In this case the two wires energized by a constant A105 Gr. II is widely used for nozzle and flange forgings.
potential AC power source converge at a point under the The figure shows that with 309 weld-metal, dilution
plasma-arc torch. Dilution is as low as 10%, while the must be kept below 10% to have delta ferrite in the deposit.
deposition rate is twice that of GMAW with auxiliary When the dilution coefficient is 10-40%, the deposit is
wire. fully austenitic. Dilution in the range 40-67% gives a
Metallurgy: Cladding by weld surfacing involves semi-martensitic structure, while dilution above 67%
joining of dissimilar metals, a n d hence all the results in a fully martensitic structure. With 309 Mo weld-
metallurgical principles explained earlier in the section metal, dilution below 15% ensures delta ferrite in the
on dissimilar metal joining are applicable to cladding. deposit. With 308 weld-metal, dilution must be below
As for example, the points to be observed in cladding 13% to ensure delta ferrite in the deposit.
with corrosion-resistant stainless steel are mentioned in Comparing 309/309 Mo with 308, partial martensite
chapter no. 4. is encountered in the former only when there is excessive
In stainless steel cladding, the most important dilution, i.e. 40 to 50%. With 308, the deposit becomes
consideration is to have sufficient delta ferrite in the S.S. partially martensitic when dilution exceeds 20%. This
deposit (duplex structure) after allowing for dilution by explains why martensitic cracking is often encountered
the base metal so as to eliminate fissuring. Schaeffler in 308 cladding.
diagram provides the necessary guidance and also With 310 deposit, it is never possible to obtain fissure-
information on the structure of the dissimilar weld, if the resistant duplex structure as the figure shows. It can
percent dilution is known. In the diagram, the point however tolerate 59% dilution before martensite develops
representing the base metal composition is joined to the in the deposit structure.
point representing the weld-metal by a straight line as The high Cr content of undiluted type 312 weld-metal
explained in Fig. 10.32. This line is referred to as dilution results in a delta-ferrite content of approximately 38%.
direction line. Whilst the type 310 weld-metal may result in an unsound
In Fig. 10.33, where the Schaeffler diagram is again first layer due to a complete lack of ferrite, the type 312
reproduced, a series of such lines are shown which join weld-metal is objectionable in this case, because of the
the point B representing the composition of an A 105-Gr. excessive ferrite content. High ferrite levels may lead to
II forging (0.35% C, 0.90% Mn, 0.35% Si) to several points impairment of the high temperature ductility, to a
4-6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42
CHROMIUM EQUIVALENT =*%Cr + %Mo+ 1.5 x %Si+ 0.5 x % Cb
Fig. 10.33: Schaeffler diagram showing dilution direction lines for A105 Gr. II steel and various stainless steel electrodes
reduction of the overall corrosion resistance, and to the renders it hard and brittle. On the other hand, the stainless
formation of the embrittling sigma-phase. steel, nickel and nickel-alloy, and copper and copper-
Behaviour at elevated temperatures: The various alloy electrodes used for welding the clad side can tolerate
detrimental phenomena which can occur, when stainless dilution by steel to certain extents, and the results of such
steel clad material is exposed to elevated temperature dilution can be predicted and controlled.
during fabrication stress-relief treatment, or in service Bearing the above in mind, the recommended edge
are described below. preparations for butt welding clad steels are shown in
1) The thermal coefficient of expansion of stainless Fig. 10.34. It is to be noted that in all the cases, a sufficient
steel is much higher than that of the base metal. Also the portion of the root face is made up of backing steel, and
yield strength of the stainless steel is higher at the elevated in some the cladding has been bevelled or stripped back.
temperature. At stress-relieving temperatures, the The backing steel is usually welded first; and the joint
difference in the yield strength of the stainless steel and design takes care that the first pass, when correctly
the ferritic base metal will cause strain gradient across deposited, will not penetrate into the cladding. The final
the fusion zone of the cladding, which may initiate cracks w e l d is m a d e on the clad side after giving d u e
or propagate existing austenitic or martensitic fissures. consideration to dilution effects, ferrite content in the case
2) Long exposure at elevated temperatures results in of stainless steel welds and other factors as in the case of
carbon diffusion across the fusion line. Consequently, cladding by weld surfacing.
the carbon content of the stainless steel cladding in- In the joints in which the cladding is stripped back,
creases, while decarburisation occurs in the ferritic base the entire thickness of the base plate can be welded by
metal in the vicinity of the fusion line. Increased carbon normal procedure using matching steel electrode. Finally,
content lowers the ductility and toughness of the stain- the clad weld is restored by the surfacing technique
less steel weld-metal, and makes it susceptible to employed in the original cladding operation.
intercrystalline corrosion. Severe carbon diffusion can Sometimes in the case of thin clad steel plates, a high-
occur during stress-relieving, especially during the pro- alloy filler wire or electrode compatible with the base
longed stress-relieving cycles required for heavy carbon- metal as well as the clad material is used to fill the entire
steel or alloy-steel clad vessels. For improving corrosion groove. In this case, the joint design as shown at (a) in
resistance in such cases, strong carbide formers such as
niobium may be added to the stainless steel weld-metal.
\lO*i
3) Some compositions of stainless steel cladding may
get embrittled when exposed to the temperature range of
500-925°C due to the formation of a brittle intermetallic
phase called sigma. Large amounts of sigma-phase
reduce resistance and ductility. Sigma is more likely to
occur in duplex stainless steel weld-metal, i.e. when weld-
metal contains ferrite. Sigma formation can also
sometimes occur in fully austenitic weld-metals in the
localised areas enriched in chromium. Besides Cr, other
ferrite forming elements like Nb and Mo encourage the
formation of sigma. Sigma formation can be avoided by
keeping the contents of the ferrite forming elements to the
minimum or by increasing the Ni content.
Welding Clad Steel Plate

In welding clad plate, problems associated with
dissimilar metal welding are encountered and must be
taken care of. As a first step, the joint geometry and
welding procedure must be such that the weld deposit
on the ferritic steel base metal does not get diluted by the Fig. 10.34: Plate edge preparation for butt welding of
clad metal. In other words, this deposit should not be clad steels: (A) Single-Vee or Vee-joint with a carbon
allowed to melt in the clad metal. Even the slightest steel lip;
admixture of the ferritic steel deposit with the clad metal (B) Stripped backjoint; (C) Double-bevel joint
Fig. 10.35 is used. The V-groove is filled with a root pass chemical, petrochemical, oil refining and similar
and several filling passes as shown at (b). The joint is industries. The liners consist of as wide a range of metals
turned over and the back of the root pass is brushed with and alloys as clad materials, mainly stainless steels and
a wire brush as at (c) and the finishing bead is laid as non-ferrous metals. The lining thickness is usually in
shown at (d). This procedure involves simple edge the range of 2-3 mm.
preparation and eliminates back gouging, but requires Spacing between the welds which attach the liner to
increased amount of expensive high-alloy filler metal. the backing plate varies according to design. In few cases,
Manual electrodes and bare filler wires suitable for to minimise time and expense of welding, the widest liner
welding the clad side of non-ferrous clad steel, 300 series sheets available are used and welding is done at the
stainless clad steel and 400 series stainless clad steel are edges. No intermittent welds are made to strengthen the
listed in Tables 10.50,10.51 and 10.52 respectively. attachment. In some other cases, almost 100% attachment
is ensured by making continuous overlapping welds
Welding of Applied Liners with a suitable (multiple spot or seam) electric resistance
Quite often, instead of using clad plate, liners in the welding process. But mostly, a compromise is effected by
form of sheets or strips are attached by welding to pressure using strips, 1 to 2m long and 50 to 150 mm wide, and
vessels and tanks to meet corrosive conditions in welding their edges by manual metal-arc, MIG/C0 2 or
(b)
FINISHING BEAD
4
(c)
Fig. 10.35: Technique of welding clad plate with high-alloy filler metal
Table 10.50: Electrodes and rods for the clad side of non-ferrous clad steel
For covering steel with
first pass of layer For completing the weld
Claddding Covered Bare rod or Covered Bare rod or
type electrode* electrode** electrode* electrode**
Nickel and ENi-1 ERNi-3 ENi-1 ERNi-3
low-carbon nickel
Nickel-copper ENiCu-1, -2, -4 ERNiCu-7 ENiCu-1, -2 -4 ERNiCu-7
Nickel-chromium-iron ENiCr-1 ERNiCrFe-5, -6 ENiCrFe-2, - 3 ERNiCrFe-5, -6
ERNiCrFe-2
Copper-nickel
70-30 ECuNi ERNiCu-7 ECuNi RCuNi***
90-10 ECuNi ERNiCu-7 ECuNi RCuNi***
Copper ENiCu-1, -2, -4 ECu, ERNiCu-7 NA ECu, RCu***
ERNi-3 NA ECu, RCu***
ENi-1 ECuAl-A2 NA ECu, RCu***
ECuAl-A2
* See AWS specification A5.ll or A5.6 ** See AWS specification A5.14, A5.6 or AS.7 *** Gas tungsten-arc welding process
Table 10.51: Electrodes and rods for the clad side of 300 series stainless clad steel
first pass of layer For completing the weld
Claddding type Covered Bare rod or Covered Bare rod or
electrode* electrode** electrode* electrode**
304 E309 ER309 E308 ER308
304L E309L2 ER309L1'2 E308L ER308L
E309Cb
321 E309Cb ER309Cb2 E347 ER347
347 E309Cb ER309Cb2 E347 ER347
309 E3091 ER3091 E309 ER309
310 E3101 ER3101 E310 ER310
316 E309Mo ER309Mo2 E316 ER316
316L E309MOL1'2 ER309MoL12 E316L ER316L
E317L1 ER317L
317 E309Mo ER309Mo12 E317 ER317
E3171 ER3171
317L E309MOL1'2 ER39MoL12 E317L2 ER317L2
E317L12 ER317L12
20 Cb E3201 ER3201 E320 ER320
* Refer to AWS specification A5.4
** Refer to AWS specification A5.9
Notes:
1. Two layers of weld deposit should he used. On light gauges, it may he necessary to grind off part of the first layer to make room for the second
layer.
2. Not a standard classification.
submerged-arc process. Narrow strips do not need between the edges. Spot welds may be made by the electric
intermediate strengthening welds. But if the liner panels resistance, TIG or MIG process. Plug welds are made by
are larger, intermittent spot or plug welds are used manual metal-arc or MIG process.
Fig. 10.36: Three methods of attaching liner edges

Table 10.52: Electrodes and rods for the clad side of 400 series stainless clad steel
First pass of layer For completing the weld
Claddding Covered Bare rod or Covered Bare rod or
type electrode* electrode** electrode* electrode**
405 ENiCrFe-2 1 ERNiCrFe-5 1 ENiCrFe-2 1 ERNiCrFe-5 1
ENiCrFe-3 1 ERNiCrFe-6 15 ENiCrFe-3 1 ERNiCrFe-6 15
E3091 ER3091 E3091 ER3091
E3101 ER3101 E3101 ER3101
E4302 ER4302 E430 23 ER430 23
E410 23
410 ENiCrFe-2 1 ERNiCrFe-5 1 ENoCrFe-2 1 ERNiCrFe-5 13
ENiCrFe-3 1 ERNiCrFe-6 15 ENiCrFe-3 1 ENiCrFe-6 1 3 5
E3091 ER3091 E3091 ER3091-3
E3101 ER3101 E3101 ER3101'3
E4302 ER4302 E410 23 ER41023
E4302'3
410S ENiCrFe-2 1 ERNiCrFe-5 1 ENiCrFe-2 1 ERNiCrFe-5 13
ENiCrFe-3 1 ERNiCrFe-6 15 ENiCrFe-3 1 ERNiCrFe-6 135
E3091 ER3091 E3091 ER309 13
E3101 ER3101 E3101 ER3101'3
E4302 ER4302 ER4302'3 ER41023
429 ENiCrFe-2 1 ERNiCrFe-5 1 ENiCrFe-2 1 3 ERNiCrFe-5 1
ENiCrFe-3 1 ERNiCrFe-6 15 ENiCrFe-3 1 3 ERNiCrFe-6 15
E3091 ER3091 E3091 ER3091
E3101 ER3101 E3101 ER3101
E4302 ER4302 E4302'3 ER4302-3
430 ENiCrFe-2 1 ERNiCrFeS 1 ENiCrFe-2 1 3 ERNiCrFe-5 13
ENiCrFe-3 1 ERNiCrFe6 15 ENiCrFe-3 1 3 ERNiCrFe-6 135
E3091 ER3091 E309 13 ER309 13
E3101 ER3101 E310 13 ER3101'3
E4301 ER43014 E4301'3'4 ER4301-3-4
* Refer to AWS specification A5.4 or A5.1 1.
** Refer to AWS specification A5.9 or A5.14.
Notes:
1. Welding on plate colder than 10°C is not advisable.
2. Preheat of 100°C minimum is advisable, particularly on plate over 12.7 mm in thickness.
3. Stress relief at 600-730°C (compatible with backing steel) is desirable.
4. Type 430 electrodes and rods may not deposit 16% minimum Cr.
5. Age-hardenable/consult supplier.
For optimum results, the liner must be pressed against

the shell by some mechanical means to ensure full contact. T O P WELD
ROOT PASS
Wherever possible and especially for small diameters,
the liner plate must be rolled to the proper curvature and
then applied. The backing metal surface must be free from
oxid-ation and scale, and this is best ensured by sand
blasting.
In attaching liners to backing plates, problems of
dissimilar metal welding arise and need to be taken care
of on the lines explained in the earlier pages. Dilution of
the weld-metal by the backing plate metal can result in Fig. 10.37: Plug weld in the liner
its reduced corrosion-resistance and ductility. Cracking
in the weld or in the heat-affected zone may also occur. has negligible dilution effect. In (c), a separate liner strip
The ill-effect of dilution can sometimes be overcome by is welded over the joint. In this case, the two exposed
using electrodes of higher alloy content. If high service welds are between similar liner metals and can be
temperatures and the difference in thermal expansion of soundly deposited. It is very important in the case of all
the liner and the backing plate are expected to cause these fillet welds that they have adequate root fusion
cracking, wider spacing between attachment welds may and at the same time minimum depth effusion into the
be specified. If wide fluctuations in service temperatures backing plate. In (b) and (c), the gap g should be wide
and pressures are to be encountered, the attachment enough to accommodate two full-size fillet welds as
spacing must be kept as close as 75-100 mm. Otherwise, indicated.
the consequent repeated buckling in service may give In example (b), if the liner is 18/8 stainless steel and
rise to fatigue cracks in the liner, attachment weld or heat- the backing plate is mild steel, the first two welds in (b)
affected zone. should be deposited with E 309 or E 309 Mo electrode
Welding procedure: Three alternative procedures of and the third with E 308 type electrode, which gives an
liner welding are shown in Fig. 10.36. At (a), the liners 18/8 deposit.
overlap. The internal lap weld suffers some amount of For m a k i n g a p l u g weld, a circular hole of
dilution but it is not exposed to the corrosive medium. approximately 10 mm diameter is bored in the liner as
The outer exposed weld is made between the two liner shown in Fig. 10.37. A circular root pass with good root
plates and does not present any problem of dissimilar fusion is first deposited with an electrode, which is
welding. At (b), two small fillet welds are first made to suitable for the combination. The top nugget weld is then
firmly attach the edges to the backing plate. The third d e p o s i t e d w i t h an electrode m a t c h i n g the liner
weld rests largely over the previous welds and hence composition.
Weldability of Steels by Stout & Doty, published 1978, by Welding Research Council, New York.
Welding Metallurgy, Vols. I & 2 by Linnert, published by American Welding Society.
Welding Handbook, Vol. 4, 7th Ed., published 1982, by American Welding Society.
Welding Steels without Hydrogen Cracking, published 1973, by the Welding Institute.
Weldabilily of Structural and Pressure Vessel Steels, 1970 Conference Proceedings of the Welding Institute.
Welding of Creep Resisting Steels, 1970, Conference Proceedings of the Welding Institute.
The Metallurgy and Welding of QT 35 and HY 80 Steels, 1974, Report of the Welding Institute.
Welding ofHY 100 and HY 1.10 Steels. 1973, Literature Review ol the Welding Institute.
Welding High-strength Steels. 1968, Symposium Proceedings of AISI and the American Society for Metals. Published by
Materials and Process Engineering Bookshelf.
Welding Metallurgy of Stainless and Heat-resisting Steels by Castro & de Cadenet, published 1975, by Cambridge University
Press.
Weldable Al-Zn-Mg Alloys. 1969, Conference Proceedings of the Welding Institute.
Welding and Fabrication of Non-ferrous Metals. 1972, Conference Proceedings of the Welding Institute.
CHAPTER 11
Hardfacing
by
Welding
THE TERMS SURFACING, weld surfacing a n d mainly from carbides of Ti, W, Cr, Mo, Fe, etc., present
hardfacing have been explained in the section on in the matrix. For example, we could have three alloys
hardfacing electrodes in Chapter 3. The types of wear having identical hardness of RC55, but differing as
to which metallic components may be subjected in ser- follows:
vice are: A—High ratio of carbides-to-matrix giving
1) Frictional wear or metal-to-metal wear. maximum abrasion resistance
2) Abrasive wear B—Lower ratio of carbides-to-matrix giving lesser
a) scratching abrasion abrasion resistance
C—Hard martensitic matrix with no carbides giving
b) grinding abrasion least abrasion resistance
c) gouging abrasion Alloy C, having hard and brittle martensitic matrix,
3) Wear due to repeated impact is highly susceptible to cracking under impact loading.
4) Corrosive wear Alloys A and B have austenitic matrix. Alloy A also has
limited resistance to impact loading and may tolerate
5) High temperature wear.
only light impact, while Alloy B has better resistance to
To meet such types of wear simply or in various impact loading. As ratio of carbides-to-matrix decreases,
combinations, correct choice of hardfacing deposit is impact resistance increases.
necessary. Obviously, it is i m p o r t a n t to u n d e r s t a n d the
Hardfacing by w e l d i n g has the following metallurgy of hardfacing deposits.
advantages:
1) A highly wear-resistant or corrosion-resistant de- Metallurgy of Hardfacing Deposits
posit can be placed exactly at the areas where In the chapter on Metallurgy, it was stated that upon
wear occurs in service heating, pure iron having body-centred cubic structure
2) A wide range of alloys can be deposited with ease termed ferrite, transforms to face-centred cubic structure
3) Expensive alloys can be used sparingly and eco- termed austenite. On cooling, the iron reverts from
nomically austenite to ferrite. Addition of carbon to iron in this
transformation and the resultant structures are the basis
4) Since most hardfacing alloys are brittle, the load
of hardfacing deposits.
is carried by the tough base metal
Carbon has a higher solubility in austenite than in
5) Alloys can be easily applied in situ and under site ferrite, a n d the carbon atoms being small locate
conditions. themselves between the iron atoms, i.e. interstitially.
Here it is important to bear in mind that hardness is When iron containing carbon is cooled rapidly enough,
not a true criterion of wear resistance, especially the carbon atoms interfere with the orderly movement
abrasion resistance. Abrasion resistance is derived of the iron atoms from the face-centred cubic position
to the body-centred cubic position, and the resulting deposits. This points to the fact that tungsten carbide
structure i$ a strained one that is neither face-centred granules present in composite electrodes can hardly be
cubic nor b>ody-centred cubic, but is rather a tetragonal expected to be effective. Instead of binding these
structure called martensite. Because the structure is granules together, the molten steel actually dissolves
strained by the atoms attempting to reach their proper them.
positions, there is considerable internal energy present, The carbides in a hardfacing deposit can be in the
which resists deformation. This causes the martensite form of network carbide or dispersed carbide. These
to be very hard and possess high strength. two variations are illustrated in Fig. 11.1.
Atoms of alloying elements such as Cr, Ni, Mo, Mn Thus martensite, austenite, network carbide and
and Si w h e n a d d e d to iron, replace iron a t o m s dispersed carbide form the basis of hardfacing deposits.
substitutionally and hence their effect is not so potent Each of them has distinct properties as described below,
as carbon. However, due to some difference in atom and they are judiciously combined in the hardfacing
sizes, these alloys contribute to moderate straining of electrode to obtain maximum resistance against specific
the lattice when the structure undergoes transformation. types of wear in service.
So each of the alloying elements does contribute to the
hardening of steel to a specific degree. But the effects of MARTENSITE
all are additive. Martensite, the most common and widely used phase
The martensitic transformation takes place at a in hard surfacing deposits, is resistant to all types of
comparatively low t e m p e r a t u r e called the Ms mildly abrasive conditions. It is also resistant to certain
t e m p e r a t u r e for martensitic start. Most alloying severely abrasive environments. The martensites have
elements lower the Ms and their effect is additive. With high strength and hardness, the hardness increasing
sufficient amount of alloying elements present in the with carbon content. Martensitic deposits possess
steel, the Ms can drop below room temperature. In this moderate ductility and average resistance to impact and
case, the steel retains its high temperature austenitic do not check crack readily. Martensites show to good
phase and remains austenitic at room temperature. advantage in applications of scratching abrasion, where
Examples are: 18/8 stainless steel and 12% Mn steel. the abrasive condition can be termed moderate. The
Austenite is a relatively soft, non-magnetic, ductile
form of steel that possesses fairly high tensile strength.
The one factor that enables austenite to be used
successfully in surfacing applications is its unique ability
to work-harden. When austenite is cold-worked or
deformed, it increases in hardness and tensile strength
while its ductility decreases. The more it is worked, the
harder it becomes. The mechanism of hardening has
been described as a transformation of the austenite into
martensite by mechanical means rather than by thermal
treatments. The cold-worked microstructure is indeed
similar in many respects to martensite, and the steel
becomes slightly magnetic.
When carbon in excess of its solubility in iron is
added, the carbon atoms unite with metallic atoms to
form carbides. Some elements which have higher
affinity for carbon than others, will combine with carbon
preferentially to form carbides. Ta, Ti, Nb, Zr, V, W, Cr
a n d Fe are carbide-forming elements listed in
descending order of their carbide-forming tendencies.
The action of tungsten (W) needs to be explained.
Though this element has good affinity for carbon and
should readily form tungsten carbide in steel, the
solubility of tungsten carbide in iron is so high that pure Fig.11.1: Modes of carbide distribution in hardfacing
deposits: (A) Network carbide; (B) Dispersed carbide
tungsten carbides are rarely formed in ferrous weld
HARDFACING BY WELDING 513
higher carbon martensites are also capable of giving alloying elements being Cr, Ni, Mo and W (for example,
excellent service in grinding abrasion. 3.0 C-4.5 Ni-1.5 Cr). The matrix is usually a complex
carbide containing islands of martensite with some
AUSTENITE retained austenite. These irons have a high compressive
Austenite is soft and ductile; work-hardens rapidly strength, good abrasion resistance but limited impact
under impact, and has high utility in conditions of resistance.
gouging abrasion. This tough, ductile core supporting Irons with around 15% Cr and some amounts of Mo
a highly work-hardened surface layer is very resistant or Ni have an austenitic matrix. The deposits have better
to further furrowing or gouging. Austenite is also an crack resistance than the martensitic types and excellent
important phase in high-carbon deposits containing resistance to low-stress abrasion, i.e. where impact is low
carbides and serves to impart some ductility to this type or absent (example: ploughshare). But they are unsuitable
of deposit. Austenites are excellent for build-up prior for high-stress abrasion.
to overlaying. Irons with 24-33% Cr form hard Cr7C3-type carbides
in an austenite-martensite matrix. Austenitic matrix is
NETWORK CARBIDES promoted through addition of 4-8% Mn or 2-5% Ni.
Network carbides are formed by precipitation from An example of such alloy is class EFeCr-Al whose
a high-carbon melt and form a continuous network in properties and applications are described in Chapter 3,
the solidified metal. This type of carbide is an extremely under "Surfacing Electrodes". Mo, V or W may be added
hard, brittle phase that has a very low resistance to to increase hot hardness and abrasion resistance.
impact. In these deposits, the carbide phase surrounds Proportion of martensite can be increased through
the matrix phase, which is usually austenite. Network suitable alloy addition and faster cooling rate. With a
carbides serve to increase the wear resistance of hard high level of martensite in the structure, the deposit
surfacing deposits in low stress scratching abrasion; they exhibits excellent resistance to high-stress grinding
are also effective in high stress grinding abrasion, abrasion (example: ball mill), low-stress abrasion as well
provided the carbide has a higher hardness than the as metal-to-metal wear. But then the deposit is more
abrasive. The increase in wear resistance is proportional susceptible to cracking.
to the amount of carbide present. However, as the The effect of carbon in the wide range from 0.1 to
amount of carbide increases, the properties of the 4.5% on the development of three types of microstruc-
deposit approach those of the carbide (i.e. high hardness tures (martensite, austenite and carbides) in a 6% Cr,
and low ductility). In some instances, network carbides 3% Mn, 0.4% Si steel is shown in the photomicrographs
make the deposit quite brittle and crack-sensitive, even in Fig. 11.2. The last of them illustrates the effect of in-
though only a small total amount is present. troducing TiC in the steel.
DISPERSED CARBIDES Types of Wear

Dispersed carbides are surrounded by bonding The various types of wear mentioned earlier are
metal. Relatively high percentages of dispersed carbide described in detail in the following paragraphs, and the
can be contained in a weld deposit before the strength appropriate consumables to be selected for each type of
and ductility of the deposit approach those of the wear have been indicated. Earlier in chapter 3, the
carbide, and most dispersed carbide weld deposits various standard types of hardfacing alloys and their
reflect the properties of the bonding metal. It is the properties have been described. These two sections must
properties of the bonding metal that control the use of be read together.
dispersed carbide. When properly employed, dispersed 1) Frictional wear: The metal-to-metal frictional wear
carbides improve all types of wear. is determined b y t h e comparative hardness of the
mating surfaces, their frictional coefficients and galling
IRON-BASE ALLOYS tendencies, surface films, lubricity and plasticity. There
These alloys usually have a carbon range of 2.5-5% are no weld deposits specially designed for frictional
and have characteristics of cast iron. Hence they are wear, and means other than hardfacing are available for
termed irons. minimising frictional wear.
These alloys can have martensitic, austenitic or 2) Abrasive wear: In this case the wear is caused by a
austenito-rriartensitic matrix d e p e n d i n g on alloy hard and sharp-edged non-metallic material such as
content. Martensitic irons have a low alloy content, the sand and cement. Scratching abrasion, also known as
1000 x PICRAL-HCI ETCH
MARTENSITE
WITH A LITTLE
RETAINED AUSTENITE
0.4 C
Re 55
1 MARTENSITE
0.1 C MARTENSITE
Re 40 WITH A LITTLE
RETAINED AUSTENITE
0.5 C
Re 57
MARTENSITE PARTLY TRANSFORMED

0.2 c AUSTENITE
Re 48 0.75C
Re 43
3 MARTENSITE
0.3 C
Re 56 UNSTABLE AUSTENTIE
0.9 C
Re 21
8
12
AUSTENITE
AUSTENITE WITH
1.0 C
NETWORK CARBIDE
Re 23
2.5 C
Re 43
AUSTENITE WITH 13
DISCONTINUOUS ! AUSTENITE AND
NETWORK CARBIDE W <*&< * - . - CARBIDE
1.15C 3.5 C
Re 21 Re 50
10 14
AUSTENITE WITH NEEDLE CARBIDES IN
DISCONTINUOUS AUSTENITE-CARBIDE
NETWORK CARBIDE MATRIX
1.25 C 4.5 C
Re 27 Re 66
11
AUSTENITE WITH 15
NETWORK CARBIDE DISPERSED TITANIUM
1.75 C CARBIDE IN AUSTENITIC
Re 35 MATRIX UN-ETCHED
Fig.11.2: Microstructural changes in hardfacing deposit with increasing carbon content

low-stress abrasion, is less severe than grinding and (Chapter 10 )and austenitic stainless steels possess these
gouging abrasion. The wear is more due to the velocity properties and are therefore ideally suited for heavy
than the abrasiveness of the material. Examples of high impact. Usually, varying combinations of abrasion and
velocity are an air blast or the turbulent flow of an impact have to be met in service. If impact is light, and
abrasive-carrying liquid. Sand slingers and dredge abrasion heavy, one should prefer a martensitic high-
pumps are examples of equipment involving scratching Cr iron or a hard deposit having a high ratio of carbides-
abrasion. Typical examples are fans, b l o w e r s , to-matrix (alloy A). If impact as well as abrasion are
ploughshares, coke chutes, cement chutes, brake drums. heavy, austenitic manganese base metal surfaced with
To combat low-stress abrasion, alloys with a hard matrix a martensitic iron would be the best choice.
a n d / or high ratio of carbides to matrix must be used. 4) Corrosive wear: To resist wear caused by corrosive
Grinding abrasion is indicated by the fragmentation media, the surface can be built up with a corrosion-
of small, hard, abrasive grains, usually between metal resistant deposit of appropriate chemistry. In general,
surfaces. The broken abrasive grains are sharp and can high-Cr austenitic deposits are used to resist corrosion
cause deep furrowing. The concentrated stress on the and oxidation.
metal surface caused by the sharp corners of the broken 5) High temperature wear: Wear at elevated
abrasive can cause local fracture and removal of the temperature is aggravated by oxidation and scaling as
microconsrituents of the metal. For this reason, grinding well as by reduction of hardness. All hardfacing alloys
abrasion is sometimes known as high-stress abrasion. tend to lose hardness, more or less, with increasing
Deterioration of the metal surface occurs through temperature. In applications where high temperatures
furrowing, local plastic flow, and microcracking. Ball combined with abrasion have to be met, one can choose
milling and machinery parts that rub together in a gritty from Ti-carbide containing deposits, cobalt-base alloys
environment are examples of grinding abrasion. Typical or high-Cr irons. Typical alloy compositions suitable for
examples are augers, concrete mixer blades, scraper high temperature wear are given in Table 11.1 and the
blades, screw conveyors, cement die rings, muller tyres. effects of high temperature on their hardness are shown
For resisting high stress without impact, alloys with a in Table 11.2.
hard matrix and medium-to-high ratio of carbides to
matrix must be preferred. Deposit Hardness Variables and Control
Gouging abrasion implies high stress on a gross scale We have stated earlier that hardness is not always a
and is usually associated with impact. Sometimes the true measure of the wear resistance property of a weld
forces are applied at relatively low velocity, as in the ease deposit. Even then in many applications, hardness of
of a power shovel digging in rock; in other cases, they the weld deposit forms the basis of selection of a
may be applied at high velocity, as in the base of hardfacing consumable. For example, the most common
hammers or breaker bars in an impact-type pulveriser. selling items in a hardfacing range of electrodes are
The mechanism of metal removal is similar to that designed on the basis of hardness of their weld deposits
produced by machining with a cutting tool or high-speed as follows:
abrasive wheel. Prominent gouges are cut in or torn from
the wearing surface in gouging abrasion. Power shovels Type Hardness (BHN)
digging in rock and rock-crusher operations are prime A 250-280
examples of gouging abrasion. Typical examples are jaw B 350-380
crushers, impactors, roll crushers, gyratory crushers. To C 280-320
withstand high stress with impact, alloys with medium- D 600-625
to-hard matrix a n d / o r medium-to-high ratio of carbides
to matrix must be used. It is easy to understand that hardness is not a fixed
3) Impact wear: Impact means repeated hammering property of the weld deposit. It depends on how fast
effect in service. When impact causes only elastic the weld deposit cools down from the molten condition.
deformation of the surface, it is termed as light. When it The faster the cooling rate, the harder is the weld
causes both elastic and plastic deformation, it is termed as deposit. If the base metal is a thin plate, the weld will
medium. Impact is termed to be heavy when it results in cool relatively slowly and will be relatively less hard. If
the deformation or fracture of even the strongest material. the base metal is a big mass, the weld from the same
To resist heavy impact, the alloy must be tough as electrode will cool fast and will be harder. Also taking
well as work-hardening. Austenitic manganese steel the same size of base metal and using the same electrode
Table 11.1: Typical deposit compositions for high temperature service: Deposit composition %
Type Mn Si Cr Ti Mo W Co V Ni
A 6.0 2.7 1.0 13.0 5.2 — — —
B 4.6 1.0 1.1 27.0 — 3.5 — —
C 3.0 1.3 0.81 8.3 1.5 — — —
D 2.0 — — , 29.0 — — 12.0 51
E 1.6 1.8 0.9 5.8 4.2 0.5 — — 0.4
F 0.7 — — 27.0 — — 4.5 62
G 1.75 1.0 1.5 30.0 — 1.5 — —
H 0.8 0.5 0.65 4.0 — 8.0 1.2 — 1.0
I 1.9 1.5 0.60 7.4 4.6 1.4 — —
J 4.5 0.75 0.80 28.0 — — — —
K 4.2 1.7 0.10 5.0 6.8 — — —
L 3.1 1.7 1.7 15.0 — — — —
Note: A to H are from manual electrodes; I to L are from flux-cored wires
Table 11.2: Hardness variation of hardfacing alloys with increasing temperature

Type Rockwell' C at temperature,°C
21 260 425 540 650 760
A 59 51 51 49 55 51
B 60 57 57 41 52 53
C 48 36 32 30 52 41
D 55 48 45 43 40 35
E 63 57 51 50 48 28
F 42 32 27 21 18 17
G 39 40 40 43 42 38
H 61 55 53 50 32 —
I 60 55 50 39 48 28
J 54 54 54 52 48 40
K 40 41 41 46 38 37
L 51 38 37 38 42 37
type, larger size electrode which uses very high current and by the method of weaving. A longer arc length
will give a slightly less hard weld deposit as compared means a higher arc voltage and therefore more heat in
to an electrode of smaller size, which uses lower current. the arc. This leads to increased dilution with the base
This is because in the case of larger size electrode, the metal. In addition, the longer arc may tend to decrease
base metal gets heated up more and therefore the weld the transfer efficiency of the alloying elements from the
deposit cools down more slowly. coating to the weld deposit across the arc. The weld
Another factor which affects hardness of the weld deposit will therefore have less amounts of alloying
deposit is dilution by the base metal. The first weld pass elements and this will result in lower hardness.
or layer with a hardfacing electrode gets diluted to the Increased welding speed means decreased rate of heat
extent of 25 to 40% by the base metal. In other words, input (joules per cm). This leads to faster cooling and
the weld deposit consists of 25 to 40% of base metal and higher hardness. Increased welding current means
75 to 60% of weld-metal from the electrode. If the increased rate of heat input and deeper penetration of
welding current is higher a n d the arc is m o r e the arc. The first factor leads to less hardness and the
penetrating, then there will be more dilution. Dilution second factor leads to increased dilution.
in the second layer is to the extent of 10 to 25% and in Broad weaving means decreased welding speed and
the third layer, between 3 and 10%. therefore less hard deposit. Stringer beads mean
Even when using the same size and type of electrode, increased welding speed and a harder deposit. In the
weld-metal chemistry and the deposit hardness will be stringer bead technique, it is possible to direct the arc
affected by arc length, welding speed, welding current on the weld puddle instead of directing it on the base
metal ahead of the weld puddle. By doing this, dilution crusher hammers, mine rails, caterpillar treads, crane
of the weld deposit by the base metal is reduced to a wheels, conveyor bucket, etc.
minimum. A correct understanding of the influence of base
Since hardness is a variable property, the quality of metal dilution and cooling rate on the hardfacing
hardfacing electrode is controlled through its weld deposit enables one to obtain a deposit of the right type
chemistry. As long as the weld chemistry is exactly as to suit an application. This is illustrated in the following
specified in the leaflet, it can be presumed that it will paragraphs by taking the example of types A, B and D.
give the same hardness. Any change in the hardness If one wants a surface with 600 Brinell hardness, the
value obtained in a test or on the job can then be ascribed obvious method is (a) to take a mild steel piece and
to other factors such as cooling rate, base metal dilution, surface it with type D. One can also achieve the same
etc. objective by (b) selecting a base metal of Class IV steel
The American Welding Society has recognised this (having 0.40% carbon) and then surfacing it with type
fact and has therefore classified all hardfacing B. The deposit will attain the required hardness due to
electrodes on the basis of weld deposit chemistry in carbon pick-up from the base metal.
their "Specification for surfacing rods and electrodes", In the method (a), the first layer will have lower
AWS A5.13 (see Chapter 3). hardness due to dilution by the mild steel base metal.
The electrode types A, B, C and D mentioned above Only in the third layer, will it be possible to get 600
are designed in such a way that they give a martensitic Brinell hardness, because the weld deposit is almost
deposit and the required hardness in the as-welded undiluted. In the method (b) the hardness phenomenon
condition at the normal cooling rates in air. These will be reversed. The first layer will have maximum
properties are achieved by having suitable amounts of hardness because of maximum carbon pick-up from the
carbon, molybdenum and chromium in the weld-metal. base metal. The hardness will decrease with each
The chemistry of these types is furnished below: subsequent layer, until in the fourth pass the hardness
will be just 350 Brinell, because it consists of undiluted
metal.
Type C% Cr% Mo% V7%
One could also obtain a surface of 600 Brinell
A 0.15 0.40 1.10 —
hardness (c) by using rail steel (having 0.55% carbon)
B 0.20 2.9 — —
and depositing a single layer with type A. In this case
D 0.50 7.0 0.50 0.50 also, hardness will decrease with each subsequent layer
of type A.
The increasing amount of carbon ensures a harder Though surfaces obtained by methods (a), (b) and
martensite, while the increasing amounts of chromium (c) may have identical high hardness, it is possible that
and molybdenum ensure attainment of full hardness surfaces (b) and (c) will not give as satisfactory service
even at lower cooling rates from the molten condition. as (a). In metal-to-metal wear where considerable heat
Also chromium, molybdenum and vanadium help the is likely to develop, deposits (b) and (c) will soften very
deposit to retain its hardness at higher temperatures in easily and wear out fast, because high hardness here is
service. merely due to high carbon. In the case of deposit (a),
Applications of the types A, B and D are listed below. the hardness is fairly well retained at high temperatures
It must be noted that the weld deposit of A is easily due to the high amount of chromium and reasonable
machinable, that of B is machinable with a special amounts of molybdenum. In the same way, (a) is likely
cutting tool, and that of type D is not machinable and to withstand abrasive wear better than (b) and (c)
must be finished by grinding. because of the higher chromium content.
Type A: Gears, pinion teeth, railway and tram rails, In spite these arguments, there are many average
machine parts where moderate hardness and toughness wear applications where methods (b) and (c) can be
are required, combined with easy machinability. adopted with advantage and economy.
Type B: Shear blades, rail ends and crossings, brake One may be surfacing a component of small size with
shoes, large cogwheels and other steel castings, a large diameter electrode. He will then get reduced
ploughshares, wobbler end, cams, idlers, rollers and hardness, because the entire base metal gets heated up.
track links of tractors. He should therefore change over to a small diameter
Type D: Metal cutting and forming tools, punches, electrode. Alternatively, he must keep the base metal
dies, drilling bits, shears and croppers, oil expellers, partially immersed in water to keep it cool.
It can also happen that when the first layer is being A good example of how weld deposition is controlled
deposited, the base metal is cold a n d therefore in practice is seen in the hardfacing of rail ends made
maximum hardness is obtained. As the welding is from 0.55% carbon steel.
continued the base metal gets progressively heated up The worn rail ends are ground sufficiently deep to
and the weld-metal deposit has lower hardness. In such provide a uniform depth of 10 mm. Initial preheat of
a case, care must be taken to keep the base metal cool 100°C is provided with an oxyacetylene torch. A buffer
all the time. Wherever possible, the weld deposit can layer is deposited with 5 mm diameter E 7016 type low-
be laid on different areas of the base metal which are hydrogen electrodes. By the time this layer is complete,
reasonably cool.
the rail is sufficiently heated up, so that no additional
One may attempt to deposit type D on a base metal
preheating is necessary. Final two or three layers are
having high carbon (say 0.40% carbon). In this case, the
deposited with 5 mm diameter type B electrode of the
weld deposit will pick up carbon and its hardness will
low-hydrogen type. Welding is continued without
increase to such a degree that it will crack. Such cracking
interruption to maintain enough heat in the base metal,
can be avoided by two methods:
since much of it is being continually absorbed by the
1) First deposit one or two layers with E7016 class
electrode and then deposit further layers of type D. An mass of the rail. The welding area is protected when
intermediate layer laid between the base metal and the necessary, from wind and rain as these may cause the
final layer is called a buffer layer. weld deposit to chill and crack.
2) Preheat the base metal so that cooling rate of type The surfaced rail m u s t h a v e 320-350 Brinell
D deposit is lowered and thereby the hardness is hardness. This is achieved by suitable initial trials. If
lowered to a value where the weld-metal will not crack. the deposit is harder, it means that preheat should be
If one is depositing metal on a huge mass of mild increased and heat input during welding must also be
steel, the deposit is likely to be much harder than what increased, so that the weld deposit cools down more
is indicated in the leaflet, because of extremely fast slowly and is softer. If the deposit is softer, preheat and
cooling rate. The deposit from type D is likely to crack heat input must be decreased.
because of extreme hardness. In this case, the problem
will not be solved by a buffer layer, but only by using Control of Distortion
preheating or by changing over to an electrode which A hardfaced component is subject to distortion. The
gives a less hardenable deposit. cause of distortion is explained in Fig. 11.3. The blank
Base metal having high carbon or h a r d e n a b l e (i.e. original component) is 30 cm long and cold, while
elements like Cr or Mo is likely to develop underbead the weld-metal is deposited in the molten state across
cracks, due to hydrogen from the arc. This tendency will the whole length. If the weld deposit was free and not
increase with the mass of the base metal. These cracks bonded to the base metal, it would shrink by 6.5 mm on
can cause permanent damage to the component and also cooling to room temperature. But since it is bonded to
lead to early breakdown of the hardfaced layer in the blank, it pulls the blank inside and distorts it.
service. In such cases, one must use low-hydrogen type
hardfacing electrodes. MOLTEN
Many of the hardfacing d e p o s i t s r e s p o n d to

mechanical and thermal treatments. This means that
w//////////m/m///m
their wear-resistant properties can be improved by
suitable heat treatment procedure or by peening. The w//////////.i^ • "///mm ^^SSJW^
procedure for heat treatment has to be adopted in NO BOND BONDED
accordance with the chemistry of the deposit and taking

Fig.11.3: Distortion caused by contraction of the deposit
into consideration the type of service conditions it has
to face. In fact, many engineering components such as As a result of distortion, flat jobs become curved,
automobile valves, hot working dies and hot shear cylindrical jobs contract across the diameter and rings
blades can be most efficiently produced by correct choice may become dished, or concave.
of the blank (i.e. base metal), hardfacing deposit Distortion can be controlled by taking proper
chemistry and final heat treatment. precautions. When a blank is a long section like shear
blade, it can be straightened when hot. Since further where high preheat is used. Distortion can often be
distortion takes place on cooling from a red heat to room counteracted by presetting the blank, so that the
temperature/the blank may be made slightly convex to contraction of the deposit on cooling pulls it straight.
the extent of 2.5 mm per metre to ensure flatness when Thin b l a d e s are h e a t e d a n d bent convex before
cold. It should not be straightened in the cold state as hardfacing, or alternatively each end of the blade is
this may cause the weld-metal to crack. clamped down to the top of the bench whilst a spacer is
The thicker the weld deposit and greater the length placed under the centre to raise it and make the top
of the deposit, more severe is the distortion. One must convex during depositing.
therefore use minimum amount of weld deposit. A thin Blades thicker than 25 mm may be clamped or tack-
blank distorts more than a thick blank. Distortion can welded back-to-back, so that very little straightening is
be minimised by using special welding sequences as usually required (after) hardfacing. Rings, dies, steam
explained in Chapter 3. For example, in hardfacing a valve seats, etc., may be treated in the same way,
long blade, the back step method should be preferred. although if the preheating temperature is high enough
A short length at the centre should be first deposited to cause scaling of the u n d e r n e a t h blank whilst
and similar lengths on alternate sides of the centre hardfacing the top one, a dummy should be substituted.
should then be deposited. The direction of arc travel Distortion may sometimes be reduced by annealing
for each run should always be towards the centre. Each or stress-relieving the blank before hardfacing. During
run should be of the maximum cross-section possible this operation the blank should not be exposed to cold
or permissible, so that minimum number of passes are air and the cooling must be slow.
used. When distortion is prevented by the large mass of
Distortion can be controlled by having symmetrical the blank or by clamping or tacking, the weld deposit is
welds wherever possible. For example in a shear blade, subject to severe strain during cooling, with the result
weld can be laid alternately on the fouredges of the blank that it develops cracks. In such cases preheating of the
shown in Fig. 11.4 (a). In building up a shaft, only a blank helps to reduce the tendency to distortion as well
short run of deposit should be made along one side as to weld-metal cracking. Preheating is discussed later.
parallel to the axis and the shaft should then be turned A hardfaced job has internal stresses due to unequal
over and a similar run made on the opposite side to heating during hardfacing; the deposit is in a state of
equalize the stress. If a long run were made on one side stress, because it has not able to contract completely and
only, this would cause a permanent distortion, which the b l a n k u n d e r n e a t h is u n d e r compression.
would not be removed by the next deposit on the Components meant for critical service are therefore
opposite side (see Fig. 11.4 (b)). stress-relieved before use. Examples are aero valves and
Blanks of long lengths should be well supported at steam valves. Components which have to be machined
close intervals to prevent sagging when the whole job or ground to close tolerances also need to be stress-
becomes too hot. This precaution is more necessary relieved. Leaving aside these specialised items, the
Fig.11.4: Examples of symmetrical welds to control di on: (A) Shear blade for hot steel; (B) Weld sequence for
a shaft
normal hardfaced jobs do not need any stress-relieving Preheat temperature should not be excessively high
treatment. It must be remembered that if any cracks have as the steel part may become too plastic and get
already formed in the deposit or base metal, they cannot compressed by the pull of the weld deposit. Too much
be removed by any postheat treatment. Also distortion preheat also causes scaling which interferes with
which has already occurred is not rectified by stress- welding. In the case of small steel parts which are hot
relief. only for a short time, the preheat tempierarure may go
In stress-relieving a job, one must take care that the up to 600°C Large jobs such as dies, steam valves, etc.,
hardness of the deposit is not affected by it. Most of the should be preheated only to a black heat (450°C
hardfacing deposits including those of types A, B, C and maximum), since owing to their large size they take a
D mentioned earlier will lose their hardness slightly. In long time to heat up and develop a thick scale on the
specialised applications, the hardfaced deposit is surface.
subjected to special p o s t h e a t t r e a t m e n t such as Cooling of the job after preheating and hardfacing
quenching, tempering or annealing to get the desired should be slow and uniform. If left in air, thin sections
cool faster than thick sections and this results in unequal
properties. In such cases, the chemistry of the weld
temperature at different locations. The usual method is
deposit is carefully chosen.
to cover it with dry insulating powder so that slow
cooling is ensured. Insulating materials commonly used
Preheating
are slaked lime, ashes, sand, mica dust and asbestos
Preheating of the base metal (referred to as blank) to
powder.
be hardfaced is necessary for the following reasons:
• To minimise distortion of the blank A useful preheating chart is given in Table 11.3,
• To prevent cracking of the deposit which is reproduced by the courtesy of Welding
• To prevent cracking of the blank. Engineer, and Tempil Corp. of U.S.A.
Preheating of small parts is easy. They can be heated
with an oxyacetylene torch. Sometimes the part may be Placement of Weld Deposit
heated on charcoal) fire. Large parts can be heated To make a successful hardfacing job, it is important
locally or entirely. Local preheating can be done with to know where and how much of the deposit must be
an oxyacetylene flame or by electrical method. In applied. It is not always true that thicker deposit applied
electrical method, two ways are possible: by induction over a large area will give maximum service. Correct
or by electrical resistance. Induction heating requires application of the deposit ensures efficiency as well as
special equipment. Electrical resistance method, which economy. For example, in Fig. 11.5, the excavator tooth
is described at length in Chapter 19, consists of wrapping welded at B has twice the amount of weld deposit
well-insulated special-alloy coils around the part and compared to A. Yet tooth B will fail quickly in service,
passing high current through them. The resistance because the brittle nose will first fracture as shown in
offered by the coils make them red hot. Glass wool and the figure. The soft base metal will then be exposed and
asbestos powder covers the coils, insulates the job and
prevents the exposure of the operator to the heat. The
current is raised gradually as this ensures gradual
heating of the job.
Preheating should not be too severe and rapid,
whatever the method of heating, because then the outer
surface heats up too much while the inside mass is (A) x
SOFTER STEEL WEARS AWAY
relatively cool. This can cause internal stresses and even LEAVING DEPOSIT SHARP
cracking of the base metal. Extra care is therefore
necessary in preheating base metals like dies which have
low ductility.
Large jobs can be heated entirely in a chamber
furnace to a temperature higher than the preheat
temperature, taken out and kept on the shop floor. To (B) SOFTER STEEL WEARS AWAY
protect them from cold winds, they can be immediately DEPOSIT MAY FLAKE OFF
covered with insulation like glass wool and asbestos
powder, and heated up again by oxyacetylene torches Fig. 11.5: Placement of hard deposit on excavator tooth:
(A) Correct; (B) Incorrect.
from all sides if necessary.
Table 11.3: Metal preheating data
Approximate composition, percent z
Metal group Metal designation i?f ^commended
lvt oo
C Mn Si Cr Ni Mo Cu preheat, °C m
z>
Plain carbon steel Below 0.20 upto 100
3J
O
Plain carbon Plain carbon steel 0.20-0.30 — — — — — — — 100-150
steels Plain carbon steel 0.30-0.45 — — — — — — — 150-260 $
r~
Plain carbon steel 0.45-0.80 — — — — — — — 260-430 O
Z
Carbon moly Carbon moly steel 0.10-0.20 — — — — 0.50 — — 150-260 H
steels Carbon moly steel 0.20-0.30 — — — — 0.50 — — 200-230 m
o
Carbon moly steel 0.30-0.35 — — — — 0.50 — — 260^30 X
oz
Silicon structural steel 0.35 0.80 0.25 — — — — — 150-260 5
Medium manganese steel 0.20-0.25 1.0-1.75 — — — — — — 150-260 <
Manganese SAE T 1330 steel 0.30 1.75 — — — — — — 200-320
steels SAE T 1340 steel 0.40 1.75 — — — — — — 260-425
SAE T 1350 steel 0.50 1.75 — — — — — — 320-480
12% Manganese steel 1.25 12.0 — — — — — — Usually not
required
Manganese moly steel 0.20 1.85 0.20 — — 0.35 — — 150-260
Jalten steel 0.35 Max. 1.50 0.30 — — — 0.40 — 200-320
Manten steel 0.30 Max. 1.35 0.30 — — — 0.20 — 200-320
Armco high tensile steel 0.12 Max. — — — 0.50 Min. 0.05 Min. 0.35 Min. — up to 100
Double strength = 1 steel 0.12 Max. 0.75 — — 0.50-1.25 0.10 Min. 0.50-1.50 — 150-320
High tensile Double strength = 1A steel 0.30 Max. 0.75 — — 0.50-1.25 0.10 Min. 0.50-1.50 — 200-370
steels Mayari R steel 0.12 Max. 0.75 0.35 0.2-1.0 0.25-0.75 — 0.60 — up to 150
(see also steels Otiscoloy steel 0.12 Max. 1.25 0.10 Max. 0.10 Max. — — 0.50 Max — 100-200
below) Nax high tensile steel 0.15-0.25 — 0.75 0.60 0.17 0.15 Max. 0.25 Max. Zr 0.12 up to 150
Cromansll steel 0.14 Max. 1.25 0.75 0.50 — — — — 150-200
A.W. Dyn-EL steel 0.11-0.14 — — — — — 0.40 — up to 150
Corten steel 0.12 Max. — 0.25-1.0 0.5-1.5 0.55 Max. — 0.40 — 100-200
Chrome copper nickel steel 0.12 Max. 0.75 — 0.75 0.75 — 0.55 — 100-200
Chrome manganese steel 0.40 0.90 — 0.40 — — — — 200-320
Yoloy steel 0.05-0.35 0.3-1.0 — — 1.75 — 1.0 — 100-320
Hi-steel 0.12 Max. 0.6 0.3 Max. — 0.55 — 0.9-1.25 — 100-260
SAE 2015 steel 0.10-0.20 — — — 0.50 — — — up to 150
SAE 2115 steel 0.10-0.20 — — — 1.50 — — — 100-150
21/2% Nickel steel 0.10-0.20 — — — 2.50 — — — 100-200
SAE 2315 steel 0.15 — — — 3.50 — — — 100-260
SAE 2320 steel 0.20 — — — 3.50 — — — 100-260
Nickel steels SAE 2315 steel 0.15 — — — 3.50 — — — 100-260
SAE 2320 steel 0.20 — — — 3.50 — — — 100-260
SAE 2330 steel 0.30 — — — 3.50 — — — 150-320
SAE 2340 steel 0.40 3.50 200-370
SAE 3115 steel 0.15 0.60 1.25 100-200
SAE 3125 steel 0.25 0.60 1.25 150-260
SAE 3130 steel 0.30 0.60 1.25 100-370
SAE 3140 steel 0.40 0.60 1.25 260-425
SAE 3150 steel 0.50 0.60 1.25 320-480
Medium nickel SAE 3215 steel 0.15 1.00 1.75 150-260
chromium SAE 3230 steel 0.30 1.00 1.75 260-370
steels SAE 3240 steel 0.40 1.00 1.75 370-540
SAE 3250 steel 0.50 1.00 1.75 480-590
SAE 3315 steel 0.15 1.50 3.50 260-370
SAE 3325 steel 0.25 1.50 3.50 480-590
SAE 3435 steel 0.35 0.75 3.00 480-590
SAE 3450 steel 0.50 0.75 3.00 480-590
SAE 4140 steel 0.40 0.95 0.20 320-425
Moly bearing SAE 4340 steel 0.40 0.65 1,75 0.35 370-480
chromium and SAE 4515 steel 0.15 1.80 0.25 200-320
chromium nickel SAE 4630 steel 0.30 1.80 0.25 260-370
steels SAE 4640 steel 0.40 1.80 0.25 320-425
SAE 4820 steel 0.20 3.50 0.25 320-425
2%Cr-l/2%Mosteel up to 0.15 2.0 — 0.5 200-320
Low chrome 2% Cr-1/2% Mo steel 0.15-0.25 2.0 — 0.5 260-425
moly steels 2%Cr-l/2%Mosteel up to 0.15 2.0 — 1.0 260-370
2% Cr-1% Mo steel 0.15-0.25 2.0 — 1.0 320-425
5% Cr-1/2% Mo steel up to 0.15 5.0 — 0.5 260-425
Medium chrome 5% Cr-1/2% Mo steel 0.15-0.25 5.0 — 0.5 320-480
moly steel 8% Cr-1% Mo steel 0.15 Max. 8.0 — 1.0 320-480
Plain high 12-14% Cr, Type 410 0.10 13.0 — 150-370
chromium steels 16-18% Cr, Type 430 0.10 17.0 '— 150-370
23-30% Cr, Type 446 0.10 26.0 — 150-370
18% Cr, 8% Ni. Type 304 0.07 18.0 8.0
25-12 Type 309 0.07 25.0 12.0 Usually >
33
High chrome 25-20 Type 310 0.07 25.0 20.0 do not a
8.0 Cb.lOXC require
nickel 18-8 Cb. Type 347 0.10 18.0
5o
stainless steels 18-8 Mo. Type 316 0.07 18.0 8.0 2.5 reheat but
18-8 Mo. Type 317 0.07 18.0 8.0 3.5 it may be
desirable to
3
remove chill m
r-
Cast iron 370-480 O
Irons Z
o
en
ro
CO
wear out very fast. The hard deposit will also flake away extent. To give an example, the screw of a cement
soon afterwards. In tooth A, the softer base metal wears conveyor pump was hardfaced on the periphery to a
away gradually, while the sharp edge of the hard deposit thickness of 6 m m at the delivery end, decreasing to 3
stands proud all the time. mm at the inlet end. After giving long service, the screw
In Fig. 11.6, dipper tooth A has been built up with a became unserviceable but the periphery had remained
thick deposit while tooth B is built up with a thin deposit. intact. The maximum wear had taken place on the front
The latter will give more satisfactory service, because face of the last flight at the delivery end. Maximum
the thin deposit wears out to a sharper edge as shown. efficiency was obtained in the long run by hardfacing
the face of two flights as the delivery end, and also on
the surface of the shaft as shown in Fig. 11.8
When the service life of a component is limited by
the fracture or breaking off at a position away from the
3.05 MM TO 1.5 MM
6.35 MM ON 2 FLIGHTS ON REMAINDER
4.83 MM ON SHAFT
4.83 MM ON FACE OF 2 FLIGHTS 3.05 MM ON REAR OF TWO FLIGHTS
Fig. 11.8: Control of hardfacing deposit on a cement

conveyor screw
Fig. 11.6: Hardfaced dipper tooth: (A) Incorrect; (B) hardfaced area, it is wasteful to have a hardfacing
Correct deposit which is too thick. An example of this is a
hardfaced automobile valve. A minimum of deposit on
In Fig. 11.7, two methods of hardfacing the hammer the seating is sufficient to outlast the life given by the
of a pulverising mill are shown. Though hammer A stem working in its guide. A somewhat similar case is
contains much more weld deposit than hammer C, its that of coal-cutter picks u s e d in automatic coal
brittle corner breaks away in service as shown at B and undercutter machines. A minimum of deposit as shown
consequently the hammer wears out rapidly. Hammer in Fig. 11.9 at A is sufficient to ensure efficient cutting
C shows the correct design of the deposit, which is both of coal until the cutter gets bent beyond use or gets lost
economical and long-lasting. in the mine. Too much deposit as shown at B is wasteful.
in in in
0 0 © Fig. 11.9: Hardfaced coal cutter picks: (A) Economical
and efficient; (B) Wasteful
Fig. 11.7: Hardfaced hammers: (A) and (B) Incorrect; In some cases the application of a small area of
(C) Correct hardfacing alloy will prolong the life of a hardened-steel
surface. For instance, the narrow band of deposit applied
The area w h i c h is to be hardfaced on a n e w to top and bottom of the drill bush shown in Fig. 11.10,
component can be judged by examining the extent of which is subsequently carburised a n d h a r d e n e d ,
wear on it in service. The pattern of wear will then preserves the size of the steel bore and increases its life
indicate which areas need to be built up and to what from three to six times.
The valve is again heated in a furnace to the

tempering temperature, i.e. 650-850°C and cooled in air.
This treatment relieves the contraction stresses and
ensures long life of the valve.
In some cases, the valves are surfaced in the annealed
condition, using an alloy rod which gives a hardness
slightly below 450 DPH. The entire surfaced valve is
then subjected to hardening and tempering treatment.
• Building cutting edge on mild steel shank

A cutting tool can be made from a mild steel shank
by depositing with a tool steel electrode on the edge as
shown in Fig.11.11. TIG welding using a tool steel filler
rod is also equally suitable.
1.524 MM
Fig. 1 1.10: Control of hardfacing deposit on a bush for

drilling jig
Examples of Thermal Treatment during Hardfacing

The following examples demonstrate the role of
correct heat treatment in ensuring maximum service
from the hardfaced components:
Fig. 11.11: Building a cutting tip on mild steel shank: (A)
• Auto engine valves Initial groove for locating the deposit; (B), (C) and (D):
The valves c o m m o n l y used in h e a v y - d u t y Three alternative ways of finishing by grinding
commercial petrol and diesel engines are made from
steel containing 0.5% C, 3.5% Si and 8.5% Cr. In the The following procedure is recommended to obtain
unhardened condition this steel has a hardness of 255- a most efficient cutting edge:
295 DPH. The finished valve is delivered in a heat- a) Preheat the bar to 600-700°C
treated condition. The heat treatment consists of b) Deposit in the groove with the electrode
quenching in oil or air from 1,000 to 1,050°C and then c) Cool in air (finally in air blast)
tempering in the range 650-850°C. d) Grind out the excess metal
Hardfacing of these valves becomes necessary to e) Oil quench from 1,280°C
confer extra erosion and corrosion resistance to resist f) Temper at 560°C for one hour
the products of combustion from high-octane. Arc g) Grind to finish.
welding is not recommended, because this process gives
a thick deposit and there is too much dilution from the • Rebuilding high-speed steel cutter
base metal. Oxyacetylene or TIG welding is preferred The high-speed steel cutter is first softened by
and surfacing is carried out with a filler rod made of annealing at 850°C and cooling slowly in the furnace or
Cr-Co-W-Mo alloy. under insulating powder. It is then preheated to 600°C
The component is first heated to 400-500°C range. with an auxiliary burner and this heat is maintained
The filler rod is fused over the surface by brazewelding while the deposit is laid on the tip with a tool steel
technique. The neck of the valve should not attain a electrode. The electrode should be of small size, say 2.5
temperature of more than 850°C during welding, as mm diameter. Immediately after welding is completed,
otherwise the strength of the part is affected. After the cutter is cooled slowly in the furnace or under
deposition is completed, the entire valve is brought to insulating powder. Even with this slow cooling, the
an even heat (dull-red, i.e. 600°C) with an auxiliary deposit remains hard. Secondary hardening at 580-
burner. It is then cooled slowly under dry insulating 600°C is carried out if required. The correct cutting
powder. angles are ground in position.
ALTERNATIVE METHOD it gets embrittled. A matching electrode, which deposits

The cutter is preheated to 500°C. A buffer layer is 12%-Mn steel, must be used. The electrode should be
laid on the cutting edge with a stainless steel electrode. of small diameter and welding current should be low.
Two layers are then deposited with the tool steel A wandering sequence must be followed. The jaw can
electrode. After each run, the deposit is peened. be kept partially immersed in water, so that heat is
Immediately after welding is completed, the cutter is rapidly conducted away.
transferred to a furnace or kept under insulating powder Sometimes the jaw is subjected to abrasive wear
and allowed to cool slowly. The cutting angles are instead of impact in service. This happens when the jaw
formed by grinding. is in continuous contact with gravel or sandy material.
In such a case, the jaw has no chance of work-hardening.
OXYACETYLENE METHOD The soft surface therefore wears out fast. In such a case,
The cutter is first a n n e a l e d t h o r o u g h l y by a buffer layer is first deposited with the stainless steel
maintaining it at 850°C and cooling slowly in the furnace electrode. Over it two layers are laid with a martensitic
or under insulating powder. It is then preheated to 600°C hardfacing electrode, giving a hardness range of 50-60
with an auxiliary burner and this preheat is maintained Re. In doing this it is not necessary that the heat input is
throughout. A high-speed steel filler rod is fused on to controlled. Welding can be done at normal speed using
the edge, layer by layer. After welding is completed, large diameter electrodes.
the cutter is cooled slowly under insulating powder. The
deposit is soft at this stage. The entire cutter is finally Hardfacing Processes
given a hardening and secondary hardening treatment The various hardfacing processes in use are listed in Table
as follows: 11.4, along with the types and forms of filler alloys available
Harden at 1,320-1,370°C, quench in water. for each.
Secondary harden at 580-600°C, cool in air. Oxyacetylene welding is used for small parts and
Anneal at 840-870°C, cool slowly in furnace. for precise buildup. Arc welding processes are most
commonly used, because they yield high deposition
• Hot shear blades rates a n d simplify extensive b u i l d u p on heavy
A hardfaced hot shear blade was illustrated in Fig. 11.4. components. Some of them can be used in semi-
A martensitic-type electrode is recommended for this automatic or fully-automatic form for hardfacing
application. For best service, the deposit should have a standard components on a production scale. Metal
hardness range of 44-48 Re (440-500 DPH). In order to spraying with oxyacetylene flame or plasma-arc, a
ensure this, the blade is preheated to 150-200°C. The process closely allied to welding, is also widely popular
deposit is laid in the correctly prepared groove using in the maintenance field.
correct sequence. After welding is completed, the whole
job is allowed to cool under an insulating cover. The • Oxyacetylene welding
deposit now has a hardness of 51-55 Re (550-630 DPH). The process requires inexpensive equipment and is
The blade is finally stress-relieved by tempering at 580- capable of giving smooth, precise and high quality
620°C. This treatment brings the hardnesses of the deposit deposits with minimum base metal dilution. These
down to 44 to 48Rc) and also makes it more tough. advantages are seen, for example, when applying a Co-
based filler metal on the edge of a steel component, in
• Austenitic manganese steel jaw crushers which case dilution of the cobalt alloy by iron is
Weldability of austenitic manganese steel was detrimental. With this process, W-carbide composite
discussed in Chapter 10. Its typical composition is 1.0% rods can be melted with minimum dissolution of the
C, 12-14% Mn. The austenitic condition is obtained by wear-resistant carbide particles and with controlled
quenching the casting from 1,000°C. The casting thereby particle distribution. In the case of hardenable base
becomes soft and possesses a hardness of 180-220 DPH. metals and brittle deposits, cracking can be minimised
In service, the continuous impact on the jaw causes the through preheating and slow cooling by means of the
surface to work-harden to 500-550 DPH range. flame. For high quality welding, for example, for
In rebuilding a jaw crusher, worn-out teeth have to depositing a low-melting high-Cr iron or Cr-Co-W alloy
be built up to full size by multi-pass welding. The most on a medium-carbon steel which has a higher melting
important thing here is to prevent overheating of the point, the welder must be skilled enough to know correct
base metal. When the base metal is heated above 300°C, adjustment and manipulation of the flame, and the
Table 11.4: Hardfacing processes and suitable filler alloys

Welding / Spraying Type of filler alloy
process Ferrous Ni-based Co-based W-carbide
Oxyacetylene Bare rod Bare rod Bare rod Composite*
Powder Powder Powder Powder
MMA Coated electrode Coated electrode Coated electrode Coated electrode*
GTA Bare rod Bare rod Bare rod Composite*
Bare wire Basic wire Bare wire Granule
GMA X Flux-cored wire Flux-cored wire X
Flux-cored arc Flux-cored X X X
(gas-shielded or wire
self-shielded)
SA Bare solid wire X X X
Flux-cored wire
SA with auxiliary Bare wire with X Bare wire with X
powder metal powder metal powder
Plasma-arc Powder with or Powder with or Powder with or Powder
(transferred arc) without W-carbide without W-carbide without W-carbide
granules granules granules
Hame spray Bare rod Bare rod Bare rod Composite*
(metal spraying) Powder Powder Powder Powder
Plasma spraying Powder Powder Powder Powder
(metal spraying)
*Note: Mild steel tube filled with tungsten carbide granules is termed a 'composite'.
mode of flame travel and of wire-feed. Oxyacetylene gains proficiency in controlling the deposit within a very
welding is unsuitable for heavy parts and for austenitic short time.
manganese steel, because it causes widespread heating. This is a low heat input process and hence the base
metals are not melted. Successful repairs can be made
• Oxyacetylene surfacing with powder on all types of base metal including tool steels, hardened
This is a modified oxyacetylene process, in which a steels and cast irons.
specially designed torch as shown in Fig. 11.12 is used. The process is well suited for the reclamation of worn
From a dispenser attached to the torch, finely divided or mis-machined parts. In addition, it can also be widely
powder is fed into the flame. The powder melts to give used for providing wear, corrosion- and heat-resistant
a controlled deposit and forms a strong bond to the base coatings on critical areas of vital machinery components
material by surface alloying and diffusion. for their improved performance, longer part-life and
recycling.
Some of the important applications are worn cams,
cam shafts, worn out or damaged threads, key way
repairs, glass moulds, valve seats, brazing of copper tube
joints in heat exchangers and mis-machined parts.
For surfacing operation, the powder module is
inserted into the torch and the flame lighted in the usual
way. The flame is made neutral, but with a slight excess
Fig. 11.12: Oxyacetylene powder surfacing torch of acetylene to avoid oxidation. The part to be surfaced
is first preheated with the flame to a temperature of
This process is suitable for use with a large variety approximately 320°C as indicated by a blue colour on
of powder alloys. A finger-operated control lever is the metal. During preheating, the torch tip is held 5 to 8
depressed to feed powder into the gas stream. The cm away from the surface. After this preheat, and while
process is easy to use and no special skills are required at this temperature, a coating of approximately 0.1 to
other than following user instructions. The operator 0.2 mm is sprayed over the entire surface to be coated.
(The powder will adhere to the workpiece and will have independent of the metal-spraying procedure used,
a sandy, dull appearance at this point.) since dilution with the base metal is negligible.
Next, starting at one edge of the pre-sprayed deposit, Typical powder compositions are given in Table 11.5,
heating is continued till a temperature of 1,000°C as and their properties and applications are given as
indicated by a bright-red colour is reached and the follows:
powder becomes molten. Such heating and bonding is
continued to cover the entire pre-sprayed area. APPLICATIONS
After this original bond coat has been applied, the MP-1: Easily flowing, heat- and oxidation-resistant
powder can now be deposited to the desired thickness, alloy on the basis of nickel. Resistant against metal-to-
using the spray and fuse technique. This means spraying metal wear as well as mineral wear.
a small amount of powder, then fusing by heating until Surfacing of steels and cast iron, subjected to heavy
it becomes completely molten on the workpiece, and is wear and corrosion. Valves and valve seats, conveyor
fully b o n d e d , then, s p r a y i n g m o r e p o w d e r and belts, conveyor screws, pump shafts.
repeating the process. MP-2: Highly fluid copper alloy with good metal-
Alloy powders used in this process fall into three to-metal wear resistance. Surfacing of steel and cast iron,
general groups: copper and copper alloys. Shafts, rails, sliding surfaces,
1) Ni-Cr-B-Si alloys (Nickel-base) bearing boxes.
2) Co-Cr-B-Si alloys (Cobalt-base) MP-3: Easily flowing, heat- and oxidation-resistant
3) Cu-Al alloys (bronzes) alloy with high tungsten carbide content. Extremely
1 and 2 are self-fluxing alloys. Boron and silicon are wear resistant. Surfacing of alloyed and mild steels,
crucial elements that act as fluxing agents and also subject to heavy abrasion, oxidation and corrosion.
depress the melting point. They permit fusing at Conveyor screws and belts, mixing blades, shovels.
temperatures compatible with steels, certain cast irons, MP-4: Heat-, oxidation- and corrosion-resistant
and some nickel-base alloys. The Ni-base and Co-base nickel alloy with excellent flowing characteristics. The
alloys are sometimes mechanically blended with deposit is resistant to impact and abrasion.
tungsten carbide or chromium carbide or chromium Surfacing of steels and cast iron. Edges and angles
boride (up to 50-60%) to further improve abrasion and of molds in the glass industry, sealing surfaces of
wear-resistance. Such products are referred to as pumps, pressing tools, slide seats.
composite powders. MP-5: Easily flowing, heat- and corrosion-resistant
The hardness of fused deposits ranges from about nickel alloy. Surfacing of steels, stainless steels and cast
20 to 60 Re, depending upon alloy composition. This is iron. Repair of glass molds, molds of the rubber indus-
Table 11.5: Typical hardfacing alloy powders

Desig- Base Nominal chemistry % Hardness
nation C Si Cr Fe B Ni W2C Co W Re 55-66
MP-1 Ni-base
0.75 4.3 15 4 3.1 bal.
MP-2 Cu-base Sn 11 Cu-bal. HB 90-100
MB-3 Ni-base 2.4 8.25 2.2 1.7 bal. 45 Re 65 70
with W2C
MP-4 Ni-base 0.25 3.7 7.5 1.5 1.7 bal. Re 35 40
MP-5 Ni-base 0.03 2.4 0.5 1.4 bal. Re 20 22
MP-5F Ni-base 0.03 2.4 0.5 1.4 bal. Re 20 22
(extra fine)
MP-6 Cu-Ni Cu-50, Si 1.7, B 0.9, Ni, bal. Re 34-37
MP-7 Co-base 0.06 1.6 21 2.4 2 bal. 4.5 Re 47-50
MP-11 Ni-base 0.06 3.5 1.5 1.5 bal. Re 30
MP-12 Ni-base 0.65 3.8 14 3.75 2.8 bal. Re 55
MP-13 Ni-base 1 2.7 1 0.6 bal. 70 Re 65 70
With W2C
try, filling of cavities in cast iron, surfacing of the edges

of slide rails, etc. MP 5 is also suitable for joining.
MP-5F: Heat- and corrosion-resistant nickel alloy
with very good flowing characteristics. Because of the
extra-fine grain size of the powder, specially suited for
the surfacing of edges. Surfacing of steels, stainless steels
and cast iron. Repairs of glass molds, mold of the plastic
and rubber industry, filling of cavities in cast iron,
surfacing of edges. MP 5F is also suitable for joint
welding of cast iron.
MP-6: Alloy similar to monel with high resistance Fig. 11.13: Alternative design of powder spraying torch in
against sea-water corrosion. The deposit is tough and which the powder and oxyacetylene are mixed in the flame:
free from porosity. Surfacing of steels and cast iron. 350°C maximum temperature for cold process; 1000°C
Bearing boxes, marine pumps and equipment. minimum temperature for hot process
MP-7: Alloy on the basis of cobalt. High wear-,
oxidation- and corrosion-resistant. Surfacing of mild and thereby minimising the possibilities of distortion and
alloy steels. Hot pressing and drawing tools, gas turbine thermal shock.
construction, chemical industry. This process is used mainly for cylindrical metal
parts (except pure copper) subjected to frictional wear.
M P - 1 1 : Ni-alloy w i t h very good flowing
The c o m p o n e n t s m a y operate in either rotary,
characteristics, heat- and corrosion-resistant. Very good
reciprocatory or sliding conditions. Some typical
wetting on cast iron. Surfacing of alloyed and mild
applications are shafts, bearing journals, ball bearing
steels, stainless steels and cast iron. Repair of glass
faces and pistons.
molds, molds of the rubber industry, filling of cavities
Typical powders used in this modified process are
in cast iron, surfacing of the edges of slide rails.
indicated in Table 11.6.
MP-12: Heat- and oxidation-resistant alloy on the
FM-1 is sprayed on initially to provide an adhesive
basis of nickel and chromium with good flowing
base. It is a must to ensure perfect bonding of any of the
characteristics. Very good wetting on copper.
subsequent surfacing powders to be used. Characterised
Surfacing of mild and alloy steels as well as of cast
by excellent performance, and a very ductile yet dense
iron and copper subject to heavy abrasion.
and corrosion-resistant deposit.
MP-13: Maximum resistance to severe abrasive
FM-2 has excellent corrosion- and friction-resistant
conditions.
properties. Excellent for metal-to-metal wear. Can be
Surfacing of mixer blades, conveyor screws, guides,
machined. Ideal for shaft seats, sliders, fan blades and
segments for the forming of concrete pipes, etc.
valve seats; suitable for various bronze alloyed parts.
On account of the high tungsten carbide content, a
FM-3 is highly resistant to corrosion and wear. Excel-
buffer layer of MP-4 should be deposited before starting
lent bonding characteristics. Fully machinable with 90°
the overlay with MP-13.
carbide cutting tools. Recommended for overlaying of
In an alternative design of the torch shown in
ball- and pin-bearing seats, shafts, slider bearings and bars.
Fig.11.13, the powder from the hopper mounted on the
top of the gun is made to drop by gravity directly into FM-4 p r o v i d e s an extremely corrosion- and
the flame. In this case, a special alloy powder is first abrasion-resistant coating. Bonds very well, but can be
sprayed on the workpiece to provide an adhesive basis finished only by filing or turning on a lathe with a 90°
for subsequent spray deposition. This step enables the carbide cutting tool. Applications are overlays on
preheating temperature to be kept as low as 250°C, pistons, shaft casings or similar parts subject to severe
which is much lower than with other usual processes; wear, impact or high temperatures.
Table 11.6: Typical powders i with torch shown in Fig. 11.13

Designation Base Ni Si Al Fe Cu Cr Mo B
FM-1 Ni-base 95 — 5
FM-2 Cu-base — — 9.5 1 bal. — — —
FM-3 Fe-base 15 — — bal. — 20 4 —
FM^ Ni-base bal. 2.5 0.5 10.5 2
MMA hard facing: MMA w e l d i n g is the most pool to increase deposition rate and, at the same time to
commonly used process for hardfacing, because of its reduce dilution.
low cost, great versatility, and the availability of a wide Flux-cored hardfacing: This process is gaining in
range of sizes and types of hardfacing electrodes. popularity in advanced countries for semi automatic
Welding can be done on any size and shape of job, hardfacing applications, because flux-cored wires can be
in any position and at any location. The welder can tailored or custom-built to suit the job by introducing
control heat input rate, dilution by base metal and can suitable alloys in the core, and also for some alloy
easily cover irregular areas. He can use stringer bead compositions the flux-cored wires are the only ones
technique to minimise heat input or weave technique available, because the compositions cannot be readily
to maximise heat input. He can direct the arc on the formed into wires.
molten puddle and control the size of the puddle to For most applications, auxiliary gas shielding is not
minimise dilution by the base metal. He can use a necessary, either because the wire is designed to be self-
wandering sequence to prevent over-heating of the base shielding or the deposit can tolerate some amount of
metal (as, for example, in the welding of austenitic porosity without losing its wear-resistant property. The
manganese steel). deposit is covered with a thin friable slag, which may
The range of hardfacing electrodes covers low-alloy have to be removed before the next pass is deposited
and high-alloy steels including austenitic manganese on it.
and stainless steels, Ni-, Co- and Cu-base alloys. These The equipment is similar to that used for M I G / C 0 2
have been discussed in Chapter 3. For heavy build-up process, except for the wire-feed rolls. If gas-shielding
(for example, on large forging dies) electrodes of 10 mm is not required, the torch assembly becomes light and
and 12.5 mm diameter may be used at very high currents simple and the equipment less expensive. When gas-
(400-600 amps) to increase heat input and minimise the shielding is used, it is usually C 0 2 or C0 2 -argon mixture.
risk of cracking. With pure C0 2 , the metal transfer is globular-while with
GTA hardfacing: GTA (i.e. TIG) and oxyacetylene C0 2 -argon mixture, spray transfer mode is obtained.
surfacing have many common features. For example, Compared to M I G / C 0 2 process using solid wire, flux-
in the TIG process a filler material in the form of cored w e l d i n g gives higher deposition rate a n d
wrought, cast or tubular rod can be fed externally into increased dilution with the base metal. Deposit thickness
the arc to control the deposit. Because of its intense arc, can be controlled by suitable adjustment of parameters
TIG surfacing requires very little preheating and is faster and techniques as shown below:
than the oxyacetylene process. Also, the intense arc
results in less heat buildup in the base metal, less For thick deposit For thin deposit
distortion and avoids adverse metallurgical effects of
the oxyacetylene flame on some base metals. Straight polarity Reverse polarity
An interesting application is the hardfacing of drill CO, shield Argon shield
pipe joints, in which tungsten carbide granules are fed Spray transfer Short-circuit transfer
into the arc area. The arc melts the base metal and the High current Low current
carbide granules get embedded in the fused metal Stringer bead Weave technique
matrix. The process can be semi-automatic. Similarly Long electrode extension Short electrode
cobalt-base alloys (for example, stellite) in rod form can extension
be fed continuously into the arc, while the arc travels at Small diameter electrode Large diameter electrode
a controlled speed. Dilution with this process is more Low voltage High voltage
than that obtained with the oxyacetylene process, but Low travel speed High travel speed
less t h a n that w i t h the M I G / C 0 2 process. The
hardfacing materials which can be conveniently used Fully-automatic flux-cored hardfacing installations
with the TIG process are: high-alloy steels, Cr stainless are used for production build-up of tractor rollers and
steels, Ni and Ni-base alloys, Cu and Cu-base alloys and idlers, tractor shoes, blast furnace bells, dredger cones,
Co and Co-base alloys. Fully-automatic TIG installations etc.
have been designed, in which the arc can be oscillated GMA or MIG hardfacing: This process is often used
by mechanical or magnetic means and continuous filler for cladding and buildup, but rarely for hardfacing.
wire, heated by electrical resistance, is fed into the weld Alloying elements required to produce the desired
deposit chemistry are contained in the continuous In a modified process, specially formulated alloy
electrode, which may be solid or tubular. The commonly powder is fed onto the job surface and then it is covered
used wires are high-alloy steels, stainless steels, Co- with the granular flux. The travelling arc melts the metal
based alloys, Ni-based alloys, Cu-based alloys and Ti powder, flux, electrode and the base metal together, to
and its alloys. The process is specially suited for Al- yield the hardfacing deposit. All these are prematched
bronze overlays. Shielding gases can be argon, helium with the welding parameters. The alloy powder is
or C 0 2 , which may be u s e d singly, in v a r i o u s correctly m e t e r e d to obtain the desired deposit
proportions, or with small amounts of oxygen. chemistry. The electrode may be weaved to minimise
Recently MIG hardfacing with Ni-based and Co- dilution with the base metal. The process enables one
based alloys using an inert gas shield has become to use less expensive electrode wire and to obtain high
possible, because of the availability of 1.6 and 1.2 mm deposition rates.
continuous tubular wires of these alloys. These wires In SA hardfacing, DC with electrode negative gives
have a tube of pure nickel or cobalt, and the alloying the highest deposition rate and the thickest deposit,
elements are contained in the core. In jobs where there while DC with electrode positive gives the lowest
is insufficient access to allow the use of TIG torch and deposition rate and the thinnest deposit. AC gives
separate filler rod, the MIG torch is able to deposit the intermediate deposition rate and deposit thickness. The
Ni-based or Co-based alloy (equivalent to Deloro 45 and influence of other variables on the deposit thickness is
Stellite 6) directly on the area to be hardfaced. A typical similar to that in the case of flux-cored arc welding
example is the main seat of a 12 mm nominal bore referred to earlier.
diameter globe valve. Plasma-arc hardfacing: The process (see Fig. 1.45,
Submerged-arc hardfacing: In this m e t h o d , a Chapter 1) is used with transferred arc, i.e. the arc is
continuous electrode, either solid or tubular, is used in struck b e t w e e n the t u n g s t e n electrode and the
combination with a granular flux which fully covers the workpiece. The filler metal is fed in powder form. The
arc zone. The wire-flux combination is suitably powder as well as the base-metal surface melt together
formulated to yield the desired deposit chemistry. The to form the hardfacing deposit, whose metallurgical
process can be semi-automatic or fully-automatic, but structure is similar to that of the TIG deposit. The
for most applications the latter version is used. The fully- powder can be iron-based, cobalt-based or nickel-based
automatic process uses heavier electrodes and higher alloy, to which tungsten carbide particles may be added.
currents to give much higher deposition rates than the With this process, a wide range of materials, including
semi-automatic process. Due to considerable dilution refractories, can be deposited in very thin as well as very
with the base plate and the loss or gain of different thick layers. The deposits have very smooth finish and
elements from the molten flux, the optimum properties require minimum grinding or machining. The process
of the surfacing filler-metal are attained only after two can be used in fully mechanised form, to hardface parts
or more layers are deposited. on a production basis.
Welding Handbook, Chapter 14, Vol. 2, 7 th Ed., Published by American Welding Society.
Hardfacing by Welding by M. Riddihough, 1949, Published by Ilife & Sons, U.K.
Seminar Handbook on Surface Coatings for Savings in Engineering, 1974, published by the Welding Institute.
Weld Surfacing and Hardfacing, 1980, Published by the Welding Institute.
Welding Defects:
Their Causes and
Prevention
IN WELDING, the main objective is to obtain sound, Incomplete Penetration
defect-free welded joints, but this is not always easy to This defect occurs at the root of the joint when the
attain. It has to be admitted that perfect welds are almost weld-metal fails to reach it or the weld-metal fails to
impossible to obtain on a production scale. Normal fuse completely with the root faces of the joint. As a
welds always contain minute slag inclusions or porosity result, a void remains at the root zone, which may
as revealed in non-destructive testing. Such small contain slag inclusions. In a fillet weld, poor penetration
imperfections, which cause some variation in the normal at the root zone can give rise to cracking.
average p r o p e r t i e s of the w e l d - m e t a l are called Incomplete penetration is caused by incorrect size
discontinuities. When a discontinuity is large enough of electrode in relation to the form of the joint, low
to affect the function of the joint, it is termed a defect. welding current or inaccurate joint preparation and fit-
Defects are caused by s u b s t a n d a r d w e l d i n g up. It occurs more often in vertical and overhead
consumables, faulty welding procedures, inefficient positions. A typical defect of this type in a butt joint is
w o r k m a n s h i p , lack of cleanliness, unfavourable shown in Fig. 12.1.
properties of the base metal, and extraneous agencies
such as low ambient t e m p e r a t u r e a n d h u m i d
atmosphere. Even when all these factors are taken care
of, defects can and do occur and they may be allowed
or removed and repaired as dictated by the nature of
the fabrication and its intended service.
Having realised that consistent production of
radiographically perfect welded joints is an expensive
exercise and gouging out and repairing internal welding
defects increases fabrication costs considerably, standard
codes do permit limited levels of defects based on
fracture mechanics approach, taking into consideration
the service conditions of the fabrication.
In spite of this, every fabricator must strive to prevent
the occurrence of weld defects in the first instance, and
to rectify them if they have occurred.
The following paragraphs describe typical defects
which can occur in arc welds, explain why they occur Fig. 12.1: Incomplete penetration in a single-Vee butt
and point out the ways in which they can be avoided or weld
corrected.
In a butt weld where access is possible only form

one side, adequate root penetration is ensured by using
a temporary or permanent backing and depositing the
first pass with a sufficiently high current. When a butt
joint is to be welded from both sides, care must be taken
to chip out or flame gouge the back of the first run before
depositing the backing run from the second side of the
joint. The groove formed after chipping or gouging must
reach the sound metal and be wide enough to permit
proper manipulation of the electrode and ensure full
penetration of the weld-metal while the backing run is
being deposited.
In a fillet weld, adequate root penetration is
ensured by using the correct size of electrode (4 or 5
m m diameter) a n d sufficiently high current, a n d
directing the arc towards the root during deposition of
the root pass. The use of a small electrode (3.15 or 2.5 Fig. 12.2: Undercut in the vertical leg of a horizontal-
m m diameter) at relatively low current does not vertical fillet weld
necessarily mean improved penetration. In fact, the
largest size of electrode consistent w i t h ease of 1) Keep joint surfaces clean
manipulation and joint thickness should be preferred. 2) Use adequate welding current
As far as possible, fillet welds should be made in the 3) Deslag each weld pass thoroughly
flat position using manipulators or other positioning 4) Place weld passes correctly next to each other.
devices. Lack of fusion is rectified in the same way as lack of
Rectification of this defect is a very costly penetration.
proposition, because it requires removal of the entire
thickness of the weld and rewelding. Care must be taken Undercut
that after removing the defective metal, a sufficiently This defect appears as a continuous or discontinuous
wide groove remains, in which the electrode can be groove at the toes of a weld pass and is located on the
correctly manipulated. base metal or in the fusion face of a multipass weld. It
occurs prominently on the edge of a fillet weld deposited
Lack of Fusion in the horizontal position as shown in Fig. 12.2.
Lack of fusion is defined as a condition where It is usually caused by excessive welding current. It
boundaries of unfused metal exist between the weld- may also result from too high a speed of arc travel,
metal and base metal or between adjacent layers of wrong electrode angle or excessive side manipulation.
weld-metal. The term should not be confused with lack In special cases, it is caused by damp or improperly
of penetration. formulated electrodes.
This defect is caused by the presence of scale, dirt, In the case of statically loaded structures, the
oxide, slag and other non-metallic substances, which presence of small and intermittent undercutting can be
prevent the underlying metal from reaching the melting d i s r e g a r d e d . H o w e v e r , w h e n the s t r u c t u r e s are
temperature. It is also caused by improper deslagging dynamically loaded in service, undercutting will reduce
of a weld pass, before an adjacent or overlapping pass the fatigue endurance of the welded joint and hence it
is deposited. When individual passes are badly placed should not be permitted.
or their surfaces are r o u g h , t h o r o u g h deslagging The defect is rectified by filling up the undercut
becomes difficult. In such cases, the weld should be groove with a weld pass. If undercut is deep and
chipped or gouged before the next pass is deposited. In contains slag, it should be chipped away before
many shops, welders or their helpers are provided with rewelding. If the rectification is being carried out on
pneumatic chippers for thorough deslagging. thick joints and on high-tensile steels, the welding
To prevent the occurrence of this defect, the following procedure including preheating should correspond to
steps should be taken: the recommended procedure for the particular steel.
WELDING DEFECTS: THEIR CAUSES AND PREVENTION 535
> 1
i
.. .», *
tf:.
1 "''
u=
LL WELD OVERLAP
Fig. 12.4: Slag inclusions in a single-Vee butt weld

Fig. 12.3: Overlap in a horizontal-vertical fillet weld
dispersed throughout the weld. Slag inclusions are
detected by the normal non-destructive testing methods,
Overlap while non-metallic inclusions are observed in the weld
This defect also occurs at the toes of a weld and microstructure at high magnification.
consists of weld-metal which has overflowed on to the Slag inclusion usually occurs in multi-pass welds due
base metal surface without actually fusing to the latter. to imperfect cleaning of the slag between the deposition
It can be isolated, intermittent or continuous. It occurs of successive passes as shown in Fig, 12.4. It may also
more often in fillet welds and results in an apparent be caused by heavy mill scale, loose rust, dirt, grit and
increase in the weld size. other substances present on the surface of the base metal
It is caused by an incorrect manipulation of the or bare electrode. Slag trapped in undercuts or between
electrode, whereby the weld-metal flows away from the uneven preceding runs may give rise to elongated lines
fusion zone. Use of too large an electrode in relation to of included slag w h e n a subsequent weld pass is
the welding position and excessive current coupled with deposited. The melting characteristics of the welding
too low a welding speed also promote its occurrence. consumable and particularly the viscosity of the
When a single-pass fillet larger than 75 mm in leg length resulting slag has an important bearing on inclusions.
is made in the horizontal position, the molten metal The molten slag should float freely to the surface of the
tends to sag and causes overlapping at the toe of the weld pool and be easily removable on solidification.
horizontal member as shown in Fig. 12.3. As in the case To prevent slag inclusion, take the following steps:
of undercut, slight and intermittent overlapping may 1) Use proper welding consumables
be ignored in statically loaded structures, but it should 2) Keep joint surfaces (especially gas cut surfaces)
not be permitted in dynamically loaded structures as and bare filler wires perfectly clean
overlaps act as stress-raisers. 3) Avoid undercuts and gaps between deposited
Overlap is rectified by grinding, chipping or gouging passes
out the excess unfused weld-metal. Care should be taken 4) Clean the slag thoroughly between weld passes.
to leave a smooth surface. The strength of a welded joint may be considerably
reduced if large irregular inclusions, or elongated lines
Slag Inclusion of inclusions at the weld junction are present. These
Non-metallic particles of comparatively large size sometimes give rise to radiating hair-line cracks. The
e n t r a p p e d in the weld-metal are termed as slag presence of small, isolated, globular inclusions may not
inclusion. The term should not be confused with non- however, seriously affect the static strength of a joint
metallic inclusions which result from the metal-slag and these may normally be disregarded. To rectify this
reactions in the molten weld pool and are finely defect, the portions of weld-metal which contain slag
inclusions must be removed and then filled with sound ascribed to the failure of the deoxidisers to function
weld-metal. adequately at the start, when the weld-metal is cold. It
must be remembered that the electrode core wire is
Porosity m a d e from r i m m e d steel, which is deficient in
The presence of a group of gas pores in a weld caused deoxidisers, especially silicon. The normal rutile type
by the entrapment of gas during solidification is termed coatings are relatively thin and contain some amount
as porosity. The pores are in the form of small spherical of cellulose which burns quickly to form a geseous
cavities, either clustered locally or scattered throughout shield. In the heavy-coated LH type, deoxidisers like
the weld deposit. Sometimes the entrapped gas gives Fe-Si and Fe-Ti are used, but they become effective after
rise to a single large cavity, which is termed as a the weld-metal has hotted up. LH electrodes of the E
blowhole. In some rarer cases, elongated or tubular gas 100, E 110 and E 120 series frequently contain restricted
cavities are presented. These are referred to as piping amounts of Fe-Si, because of the need to obtain adequate
or wormholes. n o t c h - t o u g h n e s s in the w e l d - m e t a l at sub-zero
It must be understood that gases are evolved by the temperatures. The amount is not sufficient to properly
chemical reactions in the welding arc. These gases may deoxidise the first increment of core wire melted upon
have high solubility in the molten weld-metal, but as arc striking.
the metal solidifies and cools, their solubility decreases To avoid starting porosity:
rapidly and they are evolved from the metal. Sometimes a) Strike the arc on a scrap plate, and then while the
if the weld-metal solidification and cooling is too rapid, tip is hot, restart the arc on the joint being welded.
the gas gets entrapped in the form of porosity. b) Strike the arc on the joint, but about 20 mm ahead
Porosity is caused by: of the actual starting point. Then move the arc back-
a) Chemically imperfect welding consumables, for ward toward the real starting point, at the same time
example, deficient in deoxidisers depositing a thin, narrow bead. Then reverse the di-
b) Faulty composition of the base metal or electrode rection and start depositing the normal weld bead with a
wire, for example, high sulphur content momentarily lower travel speed to remelt the initial
c) Oil, grease, moisture and mill scale on the joint thin weld bead. Then proceed to weld in the normal
surface manner.
d) Excessive moisture in the electrode coating or
submerged-arc flux Crack
e) Inadequate gas shielding or impure gas in a gas- Crack is defined as a discontinuity caused by the
shielded process tearing of the metal while in a plastic condition (hot
f) Low welding current or too long an arc crack), or by fracturing of the metal when cold (cold
g) Quick freezing of the weld deposit. crack). It represents a failure under stress of a metal
Puddling of the weld-metal and use of preheat or when it is behaving in a brittle manner, i.e. it is inclined
higher current allow sufficient time for the dissolved to fracture without deformation.
gases to escape from the weld-metal. If electrodes or Cracking can occur in the weld-metal, at the fusion
fluxes are moist, they should be redried in heated ovens line or in the base metal. Cracks may be classified
according to the recommendations of the consumable according to location and direction of line and they may
producer. range in size from large cracks which can be seen by the
Presence of small, finely dispersed porosity is naked eye (called macrocracks), to extremely small
normally not expected to affect the static and even fissures which are detected with the aid of a microscope
d y n a m i c p r o p e r t i e s of a w e l d e d joint. However, (called microcracks). Typical cracks occurring in a
excessive porosity, blowholes or piping must be guarded welded joint are shown in Fig. 12.5.
against, as they seriously impair these properties. Their Hot cracks: These cracks occur at temperatures above
presence is detected by the conventional NDT methods. 540°C, and when observed under the microscope, are
The defective portions must be removed and rewelded. seen to have travelled across the boundaries between
the grains (intergranular). If the crack has extended to
Starting Porosity the surface, the fractured surface is found to be coated
Some basic low-hydrogen type electrodes have a with blue scale or possibly black scale.
tendency to deposit weld-metal, which contains internal Hot cracking occurs in a solidifying metal at the end
porosity at the start of the weld bead. The defect is of the solidification range, when the last portion is still
by having enough Mn in the steel, because sulphur has

higher affinity for this element than for Fe. Oxygen and
selenium also behave like sulphur and in their cases also,
Mn can be used as a corrective. The hot cracking
tendency caused by P, S and Si increases with carbon
and alloy content of the steel, and hence high tensile
steels must have lower percentages of these elements
than in the case of mild steel. This applies as much to
the weld-metal as the base metal. Fully austenitic
stainless steel weld-metal is also prone to hot-cracking
as explained under the weldability of stainless steels,
which is the reason why controlled amount of ferrite is
introduced into it.
Fig. 12.5: Classification of cracking according to location Hot cracks can also occur in the HAZ of the base
in a weldment: (1) Weld-metal crater cracking; (2) Weld- metal when it contains complex carbides or intermetallic
metal transverse cracking; (3) Base metal heat-affected compounds of lower melting points. Such cracks have
zone transverse cracking; (4) Weld-metal longitudinal been reported in the welding of Cr-Mo steel tubing
cracking; (5) Toe cracking; (6) Underbead cracking; (7) having the following composition: 0.35% C, 1.0% Cr and
Fusion line cracking; (8) Weld-metal root cracking. 0.20% Mo.
Cold cracks: These cracks usually occur in the HAZ
of the base metal, when this zone becomes hard and
brittle due to its rapid cooling after the weld-metal has
been deposited, and sufficient hydrogen has been
absorbed by the weld-metal from the arc atmosphere.
Joint restraint also contributes to its occurrence. Cold
cracking is also referred to as hydrogen-induced
cracking and has been dealt with in detail during the
discussion of weldability of steels.
Cold cracking is prevented by using:
rsromnncGE a) A steel of lower carbon equivalent
f STRESS
STAGfc 1 STAGE 2 f b) Higher heat input during welding
c) Preheating
Fig. 12.6: Formation of hot crack in a heavy fillet weld. d) Low-hydrogen welding consumables
Stage 1 — Initial freezing pattern; Stage 2—Narrow spaces
It is to be noted that cracks, whether hot or cold, are
partly filled with shrinking liquid (shaded)
promoted by joint restraint. Joint restraint increases with
joint thickness and since large weld beads can withstand
liquid and the mass of the metal is unable to deform shrinkage stresses better than smaller ones, heavy plates
without cracking. At this stage, when vibrational or and sections should be welded with fewer passes of
contraction stresses are imposed on the metal, a fissure larger cross-section than with a large number of passes
forms. Figure 12.6 shows how a hot crack initiates in a of smaller cross-section. Bad joint fit-up often prevents
heavy fillet weld when it is subjected to high localised the deposition of a sufficiently large and regular root
contraction or shrinkage stresses indicated by the pass for fear of burn-through. The result is lack of
arrows. penetration, lack of fusion or slag inclusion, from which
While hot cracking propensity increases with small hairline cracks may be initiated. The weld itself
increasing joint restraint, it is really the presence of can develop longitudinal cracks under joint restraint.
certain undesirable low-freezing compounds formed by Cracks encountered in the first weld pass or layer must
stray elements which promotes the phenomenon. be at once cut out and properly re welded. It is wrong to
A typical example is FeS (iron sulphide). In a suppose that the cracked weld will be remelted and
solidifying steel rich in sulphur, the last trace of the repaired by the subsequent pass. On the contrary, the
liquid containing FeS will not solidify until about 980°C. crack propagates into the second layer immediately, or
Such hot shortness caused by sulphur can be avoided in due course of time.
Hydrogen-induced cold cracks are often located preparation and fit-up; (b) electrode size and welding
underneath the weld bead (termed underbead cracks) current; (c) number and locations of runs; and (d)
and hence are likely to escape detection. To prevent cold welding speed or length of run per electrode.
cracking, especially in the case of a new high-tensile In the case of a fillet weld, faulty weld size refers to
steel, suitable welding procedures must be developed undersize weld or weld of variable size as measured by
after carrying out initial trials on test plates and the m i n i m u m leg length. Faulty profile refers to
subjecting them to the standard weldability tests. excessive concavity or convexity, unequal leg lengths
Crater crack: This crack tends to occur in the crater and irregular weld face.
formed at the end of a weld bead, because the crater Excessive concavity or convexity may be caused by
cools at a faster rate than the rest of the bead, and because an incorrectly formulated electrode, or by use of
the crater solidifies from all sides towards the centre. incorrect welding current. These defects are shown in
Hence it is a form of shrinkage cracking. Crater cracking Figs. 12.7 and 12.8. Low current gives rise to convexity,
is encouraged by elements like C, S and Cb, which are while high current induces concavity. Fillet welds of
prone to segregation. To reduce the possibility of its unequal leg length result from faulty manipulation of
occurrence, the crater must be filled up properly with the electrode and holding it at a wrong angle with
the weld-metal and the electrode withdrawn gradually. respect to the joint. When a fillet weld of leg length more
For total elimination of crater faults, a crater filling than 7.5 m m is attempted in a single pass in the
device may be used, which enables the welding current horizontal position, the hot metal tends to sag at the
to be decreased in a controlled manner by the welder, base, thus resulting in a longer horizontal leg and a
as he fills the crater and cuts off the arc. In radiographic shorter vertical leg. When a joint has excessive root gap
quality welding, welders usually chip out the crater and due to bad fit-up, the effective leg length gets reduced
start the next weld bead at that point. and an undersize fillet weld is produced.
The strength of a welded joint will, u n d e r any Excessive concavity and unequal leg lengths reduce
conditions of loading, be very seriously reduced by the the throat thickness of fillet welds and hence will also
presence of cracks in a weld and under no circumstances reduce the ultimate strength of the welded joint.
should the fault be permitted. Excessively convex fillet welds may be of equivalent
Weld-metal or parent metal containing crack can only static s t r e n g t h to s t a n d a r d w e l d s . The w e l d i n g
be rectified by removing the defective portions and by conditions which tend to produce fillets of this shape,
rewelding. Precautions should be taken to ensure that however, also tend to give rise to such faults as lack of
the cracked portions are removed completely. The metal penetration and lack of fusion, which if present, will
removed should preferably extend well beyond the
visible extremities of the cracks and the surrounding
surfaces should be properly cleaned before rewelding.
It should be noted that a modified welding procedure
may be necessary to prevent a recurrence of the fault.
Faulty Weld Size and Profile

A weld, which is otherwise correctly deposited and
formed, is termed faulty when it is different in shape
a n d / o r size from that specified.
A butt weld is considered faulty if it has lack of
reinforcement, excessive reinforcement or irregular
profile. When the face of a weld lies below the parent
plate surface, the w e l d is said to h a v e lack of
reinforcement. On the other hand, when the weld-metal
is deposited in excess of the reinforcement required or
specified, the weld is considered as having excessive
reinforcement.
All these faults arise from the use of an incorrect Fig. 12.7: Excessive concavity, lack of root penetration
welding procedure and will not occur if the following and slag inclusion in a horizontal-vertical fillet weld due
factors are correctly specified and followed: (a) joint to excessive weaving
Frequent interruptions of the arc, when welding, may

also tend to cause spatter and should be avoided as far
as possible. Arc blow also causes heavy spatter.
Spatter has no measurable effect on the strength of a
welded joint. The fault may be rectified by chipping and
wire-brushing the spatter deposit from the surface of
the plate. Cost of spatter removal can be lowered by
applying anti-spatter compound on the surrounding
surfaces.
Melting Off of Plate Edge

This defect may occur when a fillet weld is deposited
in a lap or strap joint in such a way that the exposed
corner of the parent plate is melted away along the
length of the weld. This results in a reduction of the
plate thickness at the edge. It is caused by incorrect
manipulation of the electrode, by using an electrode
Fig. 12.8: Excessive convexity and lack of root penetration which is too large in relation to the parent plate
in a horizontal-vertical fillet weld due to too low a current thickness, or by insufficient lap.
Where a fillet weld is applied to the square edge of a
lap joint, the designer should specify that the weld size
affect the strength of the weld considerably. Excessive be at least 1.5 mm less than the edge thickness, in order
convexity may also be associated with overlapping at to prevent melting the exposed corner.
the toes of a weld, and due to the resulting notch effect When this defect occurs, the actual throat of the fillet
and stress concentration, the fatigue strength of a joint weld gets reduced while apparently the throat of the
will be seriously reduced where these faults are present. weld is regarded as equal in size to the plate thickness.
Both, lack of reinforcement a n d excessive Consequently, the static and fatigue strength of the
reinforcement will reduce fatigue strength. Lack of welded lap or strap joint will also get reduced.
reinforcement in butt welds may also prevent the joint The defect can be rectified by depositing additional
from developing the full strength of the parent plate weld-metal to restore the throat thickness to the correct
under static loading. size.
Excessive concavity and unequal leg lengths in fillet
welds can be rectified by adding further weld-metal. Tungsten Inclusions
To correct excessive convexity it is necessary to chip In m a n u a l TIG w e l d i n g , the occasional and
away the excess weld-metal. accidental touching of the tungsten electrode to the work
Excessive reinforcement in butt welds may be or to the molten weld-metal results in the transfer of
rectified by removing the excess metal. Lack of reinfo- tungsten particles into the weld deposit. These tungsten
rcement can be corrected by adding further weld-metal. inclusions constitute a defect, and for critical work a
limit on the size and numbers of these inclusions is
Spatter specified. The inclusions are visible to the naked eye
During the deposition of a weld, small globules of only in rare cases. Generally, they are beneath the weld
metal from the electrode may be scattered over the surface and can be detected d u r i n g radiographic
surface of the weld and adjacent parent metal, and in inspection as the inclusions appear whiter and clearer
most cases these globules adhere strongly to the surface on the film in contrast to slag inclusions or porosity,,
on which they rest. They are termed as spatter. The main which appear darker than the average darkening of the
objection to excessive spatter is the waste of expensive film in the weld-metal areas.
weld-metal and the cost of removal. This defect can be minimised by using thoriated
Spatter is usually caused by excessive welding tungsten or zirconiated tungsten electrodes in preference
current, a l t h o u g h in some cases the w e l d i n g to pure tungsten electrodes. The use of superimposed
characteristics of the electrode, or dampness of the high-frequency current for arc starting can eliminate
electrode covering may give rise to this condition. tungsten inclusions at the start of a weld.
Lamellar Tearing the welding. Examples of joint configurations which are

This defect usually occurs in plate material in susceptible to lamellar tearing are shown in Fig. 12.9.
welding certain types of joints in a large rigid structure, Susceptibility depends on the presence of non-metallic
where the restraint is high and stresses build up during inclusions distributed parallel to the plate surface. Such
NOZZLE OR
PENETRATOR
FABRICATED
FROM ROLLED
PLATE
CIRCUMFERENTIAL
STIFFENER
CYLINDRICAL
RIGID END
(C)
CRITICAL JOINTS
ggrraas—'
Fig. 12.9: Fabricated details and joint types prone to lamellar tearing: (A) Nozzle or penetrator through a rigid plate;
(B) Stiffener or rigid end in a cylindrical fabrication; (C) Rigid box section; (D) T-joint with fillet weld; (E) T-joint with
compound butt and fillet welds: (F) Corner joint with butt weld
WELDING DEFECTS! THEIR CAUSES AND PREVENTION 541
inclusions can be detected by ultrasonic testing of the conductor (see Fig. 12.11), a magnetic field or flux is set
parent plate. Lamellar tears can occur in the HAZ and up around it, with lines of force that can be represented
away from this zone in the parent plate and they by concentric circles, in planes at right angle to the
generally run parallel to the plate surface. direction of the current. The intensity of these circular
To prevent lamellar tearing: lines of force diminishes as their distance from the
1) Replace the susceptible steel with purer steel or electrical conductor increases and varies according to
with forgings or castings the shape and path of the lead. The magnetic fields
around the leads carrying the current also develop like
2) Modify the joint configuration suitably. For ex-
ample, in Fig.12.9 (D) is less susceptible than (E). In (F), or unlike polarity depending on the direction of the
the horizontal plate may be chamfered so that the stress current. In the case of like polarity, the fields will repel
is not perpendicular to the plate each other, tending to push the magnetic lines of force
away from each other, and in the case of unlike polarity,
3) Butter the susceptible area of the base metal with the fields will attract each other, tending to merge both
soft weld-metal to absorb the contraction strains. into one.
To measure the susceptibility to lamellar tearing of In actual practice, the arc stream is deviated from its
a steel plate, a short transverse tensile test specimen is proper course by the interaction of these magnetic fields,
extracted from it as shown in Fig. 12.10 and is subjected causing the arc to blow off in the direction of the
to a tensile test. The reduction in area at the fractured magnetic forces. With AC, this phenomenon is hardly
point is measured. This parameter can then be correlated noticeable because of the rapid change of direction in
with the incidence of lamellar tearing in different types the flow of the current and the consequent change in
of fabrication. the direction of the magnetic field.
The welder or the welding supervisor should
Arc Blow
familiarise himself with the nature of this trouble, so
In arc welding with DC, welders are sometimes
that when he encounters it he will not blame the
troubled by arc blow, a phenomenon in which the arc
electrode, equipment or power supply. He can do this
tends to deflect from its intended path owing to the
by making a weld deposit on the end of a bar, set up so
magnetic field created by the current flowing through
electrode, arc stream and base metal. It then becomes as to project approximately 60 cm from the edge of the
difficult for the welder to handle the molten pool and welding table, then attempting to make the deposit after
slag, and the result is excessive spatter and a weld bead having taken a few turns of the cable around the bar.
containing porosity and lack of fusion. To effectively Another way is to attempt to deposit weld-metal inside
counteract arc blow, it is essential that one should first a deep corner in a steel box. When the condition is
understand the reason for its occurrence and the various recognised, it can generally be corrected by proper
methods adopted for minimising its effects. disposition of the ground and the cable, to prevent or
neutralise the unfavourable magnetic field.
In very general terms, arc blow is created by
To be able to do this, one must understand how the
magnetic disturbances. When DC passes through a
lines of force or flux behave under different conditions.
For example, if the medium changes from steel plate to
<s
KM 4M DIRECTION
OF FLUX
I
PLATE > 25 MM THICK PLATE < 25 MM THICK ; DIRECTION OF CURRENT
FRICTION WELDED BXTENSIONS
Fig 12.10: Extraction of short-transverse cylindrical Rg 12.11: Magnetic field set-up by a current passing
tensile test specimen from plate through a conductor
air, the circular lines get d i s t o r t e d a n d tend to controlling the ground effect. In Fig. 12.14 (A), the earth
concentrate in the steel which offers less resistance. The connection is given at the start of the seam, so that the
consequent squeezing of the magnetic flux lines at the lines of force of the earthing current are behind the arc.
plate edges, when the welding arc reaches these areas At the start of the weld, the ground effect would make
is shown in Fig. 12.12. The arc then tends to veer away the arc veer forward and this would add to the forward
from the side of the flux concentration, resulting in a arc movement caused as shown in Fig. 12.12. At the other
forward blow at the start of the weld and a back blow at end of the seam, the forward veering of the arc due to
the end of the weld as indicated in the figure. This ground effect would be cancelled by the back blow
veering of the arc is observed by the welder as arc blow. caused by concentration of the flux from the arc at the
At the middle of the seam in the figure, the flux field is end of the workpiece. Thus the total arc blow would be
symmetrical and hence there is neither back arc blow considerably diminished. This is shown in Fig. 12.14 (B).
nor forward arc blow. In Fig. 12.15 (A), the earth clamp is fixed at the finish
A similar squeezing phenomenon is also caused by of the seam and hence the ground effect makes the arc
the earthing current within the workpiece, i.e. the veer backwards. Hence there would be an increase in
electrical current passing through the workpiece to the the intensity of the back blow caused by the arc flux at
earth-clamp. This is shown in Fig. 12.13. It is observed the finish of the weld as shown at (B). This means that
that the magnetic lines of force set up by the earthing to reduce excessive forward blow at the start of the weld,
current are at right angles to those formed around the the earth clamp should be at the finishing point of the
electrode, thus resulting in a concentration or squeezing seam, while to reduce excessive back blow at the end of
of the flux at X, which causes the arc to blow away from the weld, the earth clamp should be at the starting point
the earthing point as indicated. This is called ground of the seam.
effect. In general, the bad effects of arc blow can be reduced
Ground effect may add to or diminish the arc blow, to a great extent by taking the following precautions:
depending on the relative positions of the arc and the a) Weld towards a heavy tack or towards a weld
earthing. This fact enables one to control arc blow by already made.
suitably positioning the earth clamp and thereby b) Use back stepping sequence on long welds.
c) Place ground connection as far away from the
joint as possible.
ELECTRODE
d) On small pieces, place ground connection at start
TRAVEL- and weld towards a heavy tack if possible.
BACK BLOW
FORWARD BLOW-
Fig. 12.12: Forward arc blow at the start and back are
blow at the end of the weld run (A) (B)
Fig. 12.14: Effects of relative positions of the arc and

the earthing (grounding) on arc blow (welding away
from ground connection)
RECTION
ARC BLOW
^ h*
Fig. 12.13: Arc blow caused by ground effect. The magnetic (A) (B)
flux set up by the ground current combines with the flux Fig- 12.15: Effects of relative positions of the arc and
around the electrode causing a high flux concentration at the earthing (grounding) on arc blow (welding towards
(X) that blows the arc away from the ground connection ground connection)
e) Hold short arc, so that the electrode coating and their directions around the electrode and workpiece
touches the plate, directing the tip of the electrode in have been shown by arrows. In the left corner A, like
the direction opposite to that of the arc blow, so that the polarities are meeting, as can be checked with compass
arc force will counteract it. needles, and they interfere with each other. As a result
A few typical instances of arc blow and the remedial of this, the arc is pushed away at the point of contact in
measures taken to overcome it are described below. the direction of the arrow. By varying the angle of the
If a narrow piece of steel is grounded from one side electrode with respect to the workpiece, this can either
only, the magnetic blow is directed as shown in Fig. be corrected or be made worse.
12.16. In Fig. 12.17 (A), the electrode is held in such a way
This arrangement has to be regarded as two leads that angle between the electrode and the workpiece is
with currents flowing in opposite directions. The small (or acute) in the direction of the current flow. This
magnetic forces are indicated by the concentric circles position intensifies the interaction of the magnetic forces
Fig. 12.16: Arc blow in steel grounded at one end
Fig. 12.17: Effect of electrode angle on arc blow

against each other in the left corner, and in consequence towards the edge. It becomes very pronounced at the
pushes the arc stream towards the right with still greater vicinity of the edge.
force. If the electrode is held as shown in Fig. 12.17 (B), The effect of the magnetic blow on the arc, when a
i.e. having an oblique angle to the flow of current, the long seam is being butt welded, is illustrated in Fig.
interference of the magnetic forces with each other 12.18. At (a) and (b), the force pushing the arc towards
decreases. This reduces the blowing effect, and any the edge of the plates is completely overcome by a much
residual blowing effect is taken care of by the resistance stronger force pushing from the edge towards the centre.
of the arc itself to the side thrust, and the arc points This has been indicated by arrows a and b. This
straight towards the metal being welded. particular edge effect lessens when the arc approaches
The effects of magnetic blow are particularly strong at the middle point (c).
the edges of weldments, the direction of arc blow being If this inward force is not exceptionally strong, it can
always from the edge towards the centre. Here the magn- be overcome by holding the electrode closer to the plate
etic blow shown in Figs. 12.16 and 12.17 (A), whose normal and changing its angle suitably as shown in Fig. 12.19.
direction is away from the ground lead, is over-come by As the force of the arc blow varies with the job in hand,
considerably stronger magnetic forces which flow from the welder must decide in a moment's notice in which
the edge towards the centre. This edge effect on the arc angle the electrode must be held, so that the weld made
becomes more pronounced if the V-groove angle is smaller under this condition is sound. In some instances, when
and also if a large-sized electrode is used. It is evident the welding is commenced at the edge, the electrode
that in both cases the arc becomes longer and must be held quite flat to overcome the inward blow as
correspondingly its resistance to side thrust decreases. shown at (a) in Fig. 12.19. As the arc travels away from
Similarly, the arc disturbance is stronger in a deep the edge, the electrode is gradually lifted at the holder
Vee than in a shallow Vee. A similar effect is encountered end, until normal conditions are arrived at, as shown at
with fillet welds, the magnitude depending on the size (b) and (c). By choosing the proper angle, the welder is
and the shape of the part to be welded. With non- able to correct arc blow to a great extent. In this case,
magnetic metals like copper or austenitic steels, the the speed of welding must be faster than the normal to
direction of the magnetic blow is from the centre prevent any piling up of weld-metal and slag, as with
Fig. 12.18: Effect of magnetic blow on the arc in a long butt joint
Fig. 12.19: Variation of electrode angle to overcome arc blow

strong arc blow these are pushed in front of the molten

pool.
To ensure satisfactory penetration on both sides of a
V-groove, the tip of the electrode should be pressed into
the Vee. A slight weaving motion of the electrode at the
holder end only, as shown in Fig. 12.20, gives additional
assistance to achieve the requisite penetration.
Where the arc blow is strong enough to push the arc
stream sideways through the covering of the electrode,
so that the top portion of the covering is not melted in
time, as shown in Fig. 12.21, welding becomes extremely
difficult. If this happens, both the weld-metal and slag
are thrown about in splashes and proper fusion becomes
impossible. With heavy covered electrodes, the top part
of the covering does not melt in time, whereas the core
wire melts away in the arc, which is forced out
underneath. As a result the arc becomes longer and the
situation is aggravated. Fig. 12.22: Turning motion from the holder to overcome
One way of overcoming this is to press the tip of the non-uniform melting of coating
electrode into the Vee or root of the fillet and at the same
time giving a quick turning motion from the holder end become unduly long. It also ensures fusion on both sides
as illustrated in Fig. 12.22. By doing so the unmelted of the Vee. To ensure sound welding in a case like this,
covering is broken off in time, so that the arc cannot the arc must be kept as short as possible and the speed
of travel should be such that quick freezing of metal
and slag takes place. Piling up of metal and slag must
be prevented, because under such conditions it is
immediately blown in front of the electrode.
Arc blow in small weldments: When arc blow is
encountered on smaller parts, which are welded on a
grounded welding table, the arrangements as shown in
Fig. 12.23 (A) and (B) are very helpful. At (A) an
insulating material like wood, thick cardboard or
asbestos has been put under one end and the welding
has been commenced from the same end. A still better
method, but one which is not always necessary, is the
procedure shown at (B). Here the weldment is lifted off
the table and the piece of steel near the centre serves as
the ground lead. Both these measures are on the same
principle of shifting the ground lead, in order to alter
Fig. 12.20: Electrode weaving to control arc blow
the flow of current in the workpiece.
Arc blow in heavy weldments: If heavy jobs are welded
on grounded bed plates and the arc blow cannot be
corrected satisfactorily by the aforesaid methods, the
ground lead must be attached in such a way on the job
itself, that the blowing effect of the two parallel leads
with current Fiowing in opposite directions can be made
use of. This is illustrated in Fig. 12.24 (A), (B) and (C).
At (A) with the ground lead farther off and the arc
burning at the edge of the plate, the arc blow takes place
Fig. 12.21: Non-uniform melting of electrode covering from the edge towards the centre. By bringing the
due to arc blow , ground lead nearer to the electrode, the repelling forces
WOOD
(B) * STEEL
Fig. 12.23: Control of arc blow in small weldments
Fig. 12.24: Control of arc blow by shifting ground connection
begin to counteract the blow from the edge. When the ground lead is brought still closer to the electrode, the
tight distance between the ground lead and the arc is blown towards the edge of the plate as shown at
electrode is reached, i.e. the magnetic forces act upon (Q.
each other with equal strength, the arc will burn straight In places such as inside corners, webs and stiffeners
as shown at (B). This can be continued further. If the in bed plates or frames, and generally during deposition
of corner welds in steel construction, wherever arc blow 12.24 (B) and (C). For this arrangement of movable
is likely to interfere with welding and freedom of ground lead, the welder must use a helmet so that both
movement is limited, a simple arrangement of two his hands are at his disposal. The application has been
parallel lead, as described above, is needed to produce illustrated in Fig. 12.26 (A) and (B). At (B), where the
the corrective effects. This is achieved as follows: freedom of movement is very restricted, it is advisable
On a suitable length of cable fasten a piece of square to insulate the ground lead so as to prevent it from
or flat steel bar with a facility to hole it as shown in Fig. touching anywhere except at the point of contact. The
12.25 (A) and (B). The other end of the cable is held by usefulness of this type of ground lead is apparent, if for
the ground clamp, or fastened to the grounds connection some reason or other, all the methods indicated
of the generator. If an arc blow is encountered only previously do not give the desired result.
occasionally, the ground clamp itself serves this purpose. In spacious locations, for instance on plates and
If there is not enough room to press the clamp at the places where there is enough room to move the ground
proper place, the piece of square or flat bar held by the lead along, if necessary, this device can be made more
clamp will serve for the short time it is needed. An convenient to the welder as illustrated in Fig. 12.27. It
example of this clamp is shown at (C). indicates the principle, the dimensions being only
The form of this ground lead is immaterial as long approximations. The centre portion should be of heavy
as it can be brought to the right place, and for a short round or square material, so that there is sufficient
distance from the point of contact, be held parallel to weight to hold the point of contact down, and the two
the electrode. By increasing or decreasing the distance insulated legs spaced wide enough to prevent turning
between the ground lead and the electrode in line with over by twisted cables. The welder can push it along
the arc blow, the welder has it in his hand to direct the the line of welding, as long as there is the need to do so
arc stream to wherever he wants it, as shown in Fig. and then leave it.
(B)
Fig. 12.25: Modified grounding to control arc blow in narrow corners
<A) * (B)
Fig. 12.26: Positions of ground lead to control arc blow in narrow corners
under great restraint, so that they are unable to move

H N and thereby he increases stresses in the parent metal
and weld-metal. If the stresses are too high, cracking of
cM 50-60 MM T 1-
_J._ . 'V
the weld-metal may occur at a point of highest stress
concentration. In such cases, the fabricator must use
extra ductile electrodes, that is, those which give high
INSULATING MATERIAL elongation value in all-weld tensile test. The danger of
cracking due to heavy restraint in the joint increases with
joint thickness. On the other hand, light structures
distort more readily and have less locked-up stresses.
Distortion can often be controlled by adopting a
suitable sequence of welding. Welding sequence has
been dealt with in detail in Chapter 3. A correct welding
sequence which reduces distortion to a minimum does
not remove or lower the locked-up stresses. It merely
Fig. 12.27: Modified grounding device to control arc distributes them evenly across the whole structure and
blow in spacious locations thereby reduces the risk of cracking.
Distortion is generally divided into four types as
Distortion and Its Control illustrated in Fig. 12.28. Longitudinal shrinkage is the
Distortion is a serious problem in welding. It creates contraction along the length of weld bead. It is
difficulties in maintaining correct shape, dimensions and maximum along the weld bead and decreases at points
tolerances of a finished fabrication. A fabricator must away from the bead. Generally its effect is secondary in
know by experience how much distortion to expect and nature, insofar as dimensional accuracy of the finished
in what direction, and how to reduce it to a minimum. part is concerned. Only in the case of seam welding of
Distortion is caused by the unequal heating and cylindrical vessels it may lead to a reduction in diameter
cooling of a metallic body during welding. It is also at the weld. Transverse shrinkage is the shrinkage
caused by the contraction of the weld-metal during perpendicular to the weld and it may lead to the
solidification and cooling to room temperature and the development of high residual stress and also cracking
contraction of the surrounding parent metal as it cools in case of highly restrained joints. Angular distortion is
from high welding temperature. When these portions the bending transverse to the weld, due to non-uniform
contract, they try to pull the parts together and the result heating and cooling along the thickness of the plate. This
is distortion. is the main source of mismatch and dimensional
A metallic body does not distort if it is heated as a inaccuracy in large welded structures. Bowing is the
whole uniformly a n d is then cooled as a w h o l e bending in the direction of the weld, and it is the cause
uniformly and has freedom to expand and contract of bending in many built-up sections, where the position
freely in all directions. In welding, however, only the of weld is away from the centre of gravity of the section.
weld joint and its surrounding area are heated up. This
area has no freedom to expand and contract. Uneven
contraction of the weld-metal and parent metal occurs. LONGITUDINAL SHRINKAGE
This gives rise to stresses in the weld, in the two
component parts making up the joint and in the entire
fabrication. If the fabrication is held firmly in a jig that
gives no freedom of movement, the stresses will remain
in the body as locked-up stresses. But if there is some
scope for movement, the stresses will find relief in
distortion. Consequently, the shape and dimensions of TRANSVERSE SHRINKAGE WELD BEAD BOWING
the welded fabrication will have changed when it J^m*. ^ r s v —ORIGINAL POSITION
*^5r»- . — — ► . . . r j j —AFTER WELDING
returns to room temperature.
ANGULAR DISTORTION
Thus in welding, the fabricator has to choose
between distortion and locked-up stresses, as these are Fig. 12.28: Types of distortion in welding of free plates
opposite effects of welding. He holds the joining parts
WELDING DEFECTS! THEIR CAUSES AND PREVENTION 549
How distortion develops can be well understood by only a narrow zone on both the sides of weld bead gets
the example of angular distortion. Figure 12.29 repre- heated to considerably high temperatures, the thermal
sents the changes taking place in the shape of a plate stresses associated with plastic flow and deformation
when it cools after being subjected to deposition of the remain confined mainly in this small region shown by
weld bead. For the sake of simplicity, it can be assumed broken lines. The average temperature distribution in
that the plate width is small and the weld bead is de- this region will follow the pattern shown by Ta curve,
posited in the centre of the plate from end to end. Since where T is the temperature and a is the coefficient of
thermal expansion. Considering it to be the case of one-
BEFORE WELDING
dimensional stress and strain, the total strain (ej will
TT be the sum of elastic strain (eE), plastic strain (ep) and
thermal strain (Ta). Since it can be assumed to a reason-
able degree of accuracy that a plane section remains
plane, the distribution of total strain (eT) will be linear.
The actual position of the line representing the total
strain (eT) will be governed by the equilibrium of forces
generated due to elastic strain. As the plate cools down,
DURING BEAD DEPOSITION
the temperature distribution in the plate as also the dis-
tribution of total strain (eT) will change. Some strain will
exist even after complete cooling and its distribution is
also shown in Fig. 12.29. It is obvious that the top edge
of the plate has considerable amount of shrinkage while
the bottom edge has been stretched slightly. This dif-
ference gives rise to bending (angular distortion).
Figure 12.30 depicts the strain distribution for low,
medium and high heat inputs. It is obvious from this
ON COMPLETE COOLING that angular distortion increases with increase in heat
(*) CORRESPONDING STRAIN
(A) VARIOUS STAGES OF DISTORTION DISTRIBUTION input per unit length of plate; it attains a maximum
value and then decreases again with further increase in
Fig. 12.29: Development of angular distortion
heat input.
c} 8
-> 0 S-4 08-4
[1
n....
i/
r
LOW &
HEAT 1 J r
JC) S-+ p £-■ OS—■
C^_-3 MEDIUM j
HEAT H V
r ii
L
0 S—■ o s —■ OS—■
HIGH 1
HEAT |
\
D l IRING BEAD m JR1NG COOUNG ON COMPLETE
DE POSITION COOLING
Fig. 12.30: Effect of heat input on angular distortion
Transverse shrinkage is not uniform along the length are more liable to distortion. Lap joints are subject to
of the plate. It is lesser at that end of the plate, where longitudinal shrinkage, which in the case of long seams
the bead is started. may cause buckling. Transverse shrinkage and angular
The amount of transverse shrinkage in a butt joint distortion are negligible.
depends on the volume of weld-metal in the joint. It is, In welded fabrications, allowances should be made
therefore, an advantage to have as little weld-metal as for weld shrinkage. This means that the fabrications
possible. To ensure this, the V-angle and root gap must should be built oversize by an amount sufficient to cover
be as small as possible. Welders often find that the V- the shrinkage and any w e l d i n g fixtures must be
angle is 100°-120° instead of 70° and the root gap is as designed accordingly. The following shrinkage values
much as 6 mm instead of 1.6 mm. Such bad joint fit-up may be taken as a guidance:
means not only more welding but more distortion by Transverse shrinkage
shrinkage. Fillet welds : 1.6 mm per weld
Angular distortion in a butt joint depends on the type Butt welds : 0.8 mm per weld for 60°-V joints.
of groove joint as shown in Fig. 12.31. Here it is shown Longitudinal shrinkage
that the angle of distortion for a 90°-V butt weld is 7° Fillet welds : 1.6 m m per 3 m of weld
while it is only 4M>° for a 60°-V butt joint. There is almost Butt welds : 3.2 mm per 3 m of weld
no distortion in a square butt, double V butt and a Steps for the prevention of distortion must start from
double-U butt, as long as welding is carried out from the design office. The designer m u s t design the
both sides. A 2XA° distortion is obtained when deep fabrication in such a way that a minimum number of
penetration welding is done from one side in a square joints is used. To do this:
butt joint. In the case of the single-U) butt joint, the angle 1) Use folded components and rolled sections in-
of distortion is 3l/i°. stead of built-up sections. For example, use rolled sec-
Single-V preparation is always used for economical tions in preference to welded girders.
reasons in plates up to 12.7 mm thickness, and in these 2) Use longest and widest plates possible.
cases distortion is not appreciable if plates are clamped 3) Corrugating of plating can be considered to avoid
securely. For thicknesses above 12.7 mm, it is useful to welded stiffeners.
have double-V and double-U preparation not only 4) Locate weld joints in such a way that they are
because less weld-metal is required, but because it is easily accessible to the welder from both sides.
possible to balance welding by making alternative Shop people must observe the following points for
passes on each side. This point has already been made controlling distortion:
clear in Chapter 3. 1) Use minimum weld-metal. Avoid overwelding.
In fillet w e l d s , there is no q u e s t i o n of joint 2) Use correct fit-up. Avoid wide bevels and wide
preparation. The only factor that will control distortion root gaps.
is the weld size, which should be as small as required. 3) Use correct sequence (see Chapter 3).
Also a large single-pass weld made at high speed will 4) Control the shape and size of components being
cause less distortion than a number of thin passes made welded. When these are inaccurate, the fabricator has
at a low speed. to force the parts to come together for welding. Such
In lap joints, it is preferable to have double fillet forced parts contain stresses which may add up to the
welded joints than single fillet welded ones, as the latter stresses caused by welding and lead to more distortion.
c
TYPE OF
PREPARATION
1
■ ■ cia
DISTORTION 4 1/2° 0° 2 1/2° 3 1/2°
C^3^3
TENDENCY
Fig. 12.31: Influence of joint preparation on distortion

5) Divide complicated work into sub-assemblies in

such a way that each of these can be welded without
restraint. Completion of the fabrication from these sub- , i i i i ~u.
assemblies may then have to be done under restraint.
6) Try to preset the parts correctly before welding
(A) j j J j J
wherever this is possible. In a tee-joint, for example, the
vertical member may be kept slightly tilled to one side, i i i i
so that after welding this member comes back into cor-
rect alignment. In a butt weld also, the plates can be
PACKERS
Fig. 12.33: Back-to-back assembly with or without the

use of packers
electrode giving 90% deposition efficiency for, say, weld-

ing 3.2 mm thick sheets to minimise distortion.
vzmmwzzzzm 4) Distortion is minimised when the same number
and sizes of runs are deposited on both sides of a double-
V butt or a double-fillet T joint.
5) A vertical weld usually causes less distortion than
an equivalent horizontal weld, which is deposited with
rtssf/fw**!***^ the same size of electrode, because the vertical run can
be of larger size and the vertical weld can be completed
Fig. 12.32: Presetting of tee and butt joints. Dotted lines with fewer passes.
show the position they will return to after welding 6) Submerged-arc welding causes less distortion
than manual metal-arc welding, because much heavier
preset at an angle away from the sides of welding. Pre- runs can be deposited at higher welding speeds.
setting is shown in Fig. 12.32. 7) Simultaneous welding by two or more welders
7) In making long butt welds in thin sheets, the two is a good technique, provided they progress at the
edges have a tendency to close the gap and even over- same rate. For example, a vertical double-vee butt
lap at the free end of the seam, as welding proceeds weld can be weld-ed from both sides simultaneously
from the other end. In such cases, a tapered gap is kept at by two welders.
the start, the gap increasing from the welding end to the 8) Where a joint or fracture runs to a free edge, it
other. The amount of taper is determined by experience. should always be welded towards that free edge. If
8) Where two identical jobs are being carried out, it welded towards restraint or sound metal, the shrink-
is possible to tack weld or clamp them together back- age of the weld may cause a buckle, ripple, or a crack,
to-back. Welding is then carried out alternatively on each which is difficult to remove.
job. To make this technique completely successful, the Alignment: For control of distortion, it is imperative
two assemblies must be stress-relieved before they are that the plate edges are pulled into alignment and held
separated. When stress-relieving is not desirable, it is securely during welding. Various methods are available
better to insert packers between the two jobs before for doing this. Most joints can be placed into desired
clamping them together. This is shown in Fig. 12.33 (B). alignment by welding a small clip to the edge of one
Shop personnel must remember the following rules: plate and driving a steel wedge in between this clip and
1) Thinner plates distort more easily than thicker the second plate, until the edges are lined up. The clips
plates. are welded on one side only, so they may be easily
2) Distortion is minimised by depositing minimum removed later with a hammer. The arrangement is
number of runs using the heaviest electrode possible. illustrated in Fig. 12.34. Short bars serving as strong
3) An electrode giving faster deposition rate and backs may be placed at right angles to the joint to
longer length of run will produce less distortion than an maintain alignment, as in Fig. 12.35. Where less
electrode giving slower deposition and shorter weld run. transverse restraint is desired, yet restraint against
For this reason, an iron powder electrode giving 110% angular movement must be maintained, the strongbacks
deposition efficiency would be preferable to an ordinary may be placed at 45° to the joint as shown.
Weld clip along one edge only so it may be removed easily with a hammer. Drive a steel
wedge below clip until edges of plate are in alignment
This thickness is same as
Bottom plate assembly Whole assembly with wedge

Fig. 12.34: Methods of applying small clips along one side of light plate to bring edges into alignment for welding
H
A wtdgc driven in betNM««n
a yoke will putt sirongback
nd p}»»e together before tacking vT
\ |
K
Transverse strongbacks will If less transverse restraint
offer restraint against is desired, yet restraint
angular distortion against angular distortion
must be maintained, these
strongbacks may be placed at 45°
Fig. 12.35: Use of strongbacks to hold thin plates temporarily in alignment while welding
Correction: It is usually possible to rectify the material. It should then be allowed to cool, free from
distortion by one of the following methods, provided draughts.
that the product has not shrunk so much that it will be
too small to meet the dimensional requirements.
Parts of a fabrication, which have buckled or
otherwise changed their shape may be corrected by
mechanical means or hammering. Sometimes an entire
fabrication which has bowed can be straightened on a
press, provided that it is not too large; in this case, it is
necessary to insert packers between the fabrication and
the platens as shown in Fig. 12.36.
Flat fabrications that have become distorted may
sometimes be corrected by clamping them to a heavy
base plate or some other strong structure with a flat
surface, and stress-relieving the fabrication whilst it is
clamped. It is, of course, important to make sure that
the base plate is strong enough to hold the fabrication AFTER PRESSING
flat. For stress-relieving, the fabrication should be heated
slowly to 650°C, and maintained at that temperature Fig. 12.36: Use of press to correct bowing
for a period of one hour per 25 mm of thickness of the
There are various ways in which such local heating

can be applied to rectify distortion, but it is only by
experience that the best method can be selected for any
particular job. In all cases, the greatest danger is in
overshrinkaging the area being heated, as this may cause
distortion worse than that which it is desired to remove.
Overshrinking may be caused by heating too large an
area, or by heating it to too high a temperature.
Experience is the best guide in the matter of area; in
regard to temperature, it is generally advisable to restrict
the temperature of the area to approximately 700°C, i.e.
a dull-red.
<C) l_ - — -__J Where a relatively thin plate has buckled owing to
welding a stronger framework to it, the buckling can
usually be removed by heating the plate in local spots
(D) L — _J on the convex side. In such cases the distortion due to
the welding is usually fairly regular, so that the spots
Fig. 12.37: Effect of local heating (the movement shown can be arranged symmetrically, starting at the centre
is greatly exaggerated) of the buckle and working evenly towards each side of
the frame, as shown in Fig. 12.38. This technique is used
The use of the concentrated heat from a blowpipe is effectively in the fabrication of railway coach bodies.
very useful for correcting distortion. When a metal Heating in straight lines is often used as a means of
component (A) in Fig. 12.37 is heated locally, the heated correcting the angular distortion caused by the fillet
portion will tend to expand, and this expansion will be welds as shown in Fig. 12.39. The heating generally
opposed by the surrounding colder metal (B). Since the follows the line of the welded joint, but is applied to the
heated part is relatively weak, the forces opposing its members of the joint on the side opposite to that which
expansion will squeeze it out of shape, as shown at (C). has been welded. The heating may be accompanied by
On cooling, this part will shrink, and the shrinkage will water-jet cooling on the reverse side. In this way the
cause the component to deform (D). It is relatively easy shrinkage d u e to welding is counteracted by the
to treat parts of a fabrication in this way, so that the shrinkage caused by the flame-heating.
distortion caused by the welding will be balanced by In all applications of flame-straightening other than
the shrinkage of the-heated parts. Although this the spot-heating of thin panels, a wedge-shaped heated
technique often causes overall shrinkage, it is acceptable zone as shown in Fig. 12.40 should be produced. The
provided that the fabrication is built larger than finally heating should proceed from the base to the apex of the
required. wedge, penetrating evenly through the plate thickness
and maintaining an even temperature (i.e. even colour).
The shape and size of the heated zone should be marked
for the operator's guidance. Heavy-duty blowpipes
s h o u l d be used for heating. In the event of an
interruption of heating, or of the heat being lost, the
operator must allow the metal to cool and then proceed
as if commencing the heating. The straightening effect
of heating for two-thirds of the width of the plate (see
BUCKLING OF PANEL BEFORE

£Z#
SPOT HEATING
Fig. 12.38: Spot-heating to correct buckling Fig. 12.39: Correction of distortion produced by fillet weld
3.5 m length of plate. Where a maximum amount of

contraction is required, as in straightening the convexed
-Ii3EEE^
edges of plates, uniform heat penetration into the metal
from both sides should be arranged as shown in the
figure.
ZM3
one• HEATITOM PENETRATE
:
EVENLY Peening: Peening means mechanical working of the
metal using impact blows with a hammer or any other
mechanical means. The shrinkage of the weld deposit
can be counteracted by peening, which tends to stretch
iJI* the metal. Upsetting or expansion of the metal by
peening is most effective at higher temperatures, where
Fig. 12.40: Use of wedge-shaped heated zone for yield strength of the metal is rather low; yet, on the other
local heating hand, most of the distortion problem occurs at the lower
temperatures after the yield strength has been restored
Fig 12.40), the width of the heated zone being 50 mm to its higher value. A disadvantage of peening is that, it
for every 300 mm of its length, is 4.8-6.3 mm for every work-hardens the surface of the metal.
Welding Metallurgy, Chapter 13, Vol. 2, by Linnert, published by American Welding Society.
Lamellar Tearing in Welded Steel Fabrication, 2972, published by The Welding Institute.
Control of Distortion in Welded Fabrications, 1976, published by The Welding Institute.
ISO 6S20-1982-& IS 10793-1983 "Classification of Imperfections in Metallic Fusion Welds, with Explanations"
Testing and
Inspection
of Welds
WELDED STRUCTURES are subjected to a variety of chemistry of the undiluted weld-metal derived from a
standard tests to determine their suitability for the an- welding consumable, a multi-layer weld pad whose
ticipated service conditions. These tests are designed to dimensions are approximately 75 mm 1 x 10 mm b x 18
check chemical, mechanical and metallurgical proper- mm b is built on a plate and drillings or chippings are
ties and to locate defects such as incomplete fusion, slag taken out from the top portion of the pad. An alternative
inclusions, porosity, cracks, etc. These tests are also used procedure prescribed by AWS is detailed in Chapter 3.
for the qualification of welding procedures and weld-
ers, and some of them are suitable for checking the qual- Corrosion Test
ity of welding consumables and base metals. There are Weldments which are expected to function in a
also several non-standard tests to determine specific corrosive liquid, solid or gas maybe subjected to suitable
properties of the base metal and the weld-metal. Ex- laboratory type corrosion tests, which give partial but
amples are weldability tests, hot cracking tests, etc. satisfactory information on how the weldment will
Some tests are destructive in nature, because they behave in actual service conditions. Corrosion tests are
involve sectioning or fracturing of the weldment to specified in only a few codes and specifications, because
examine p r o p e r t i e s and to check on s o u n d n e s s . most alloys and welding consumables are selected for
Destructive tests are performed on sample weldments specific applications by previous experience.
made with procedures similar to those being used in In the welding field, corrosion tests are carried out
actual fabrication. They m a y be chemical tests, on welded joints of stainless steels and nickel alloys to
m e t a l l o g r a p h i c tests, mechanical tests, or any evaluate their resistance to corrosive environments in
combination thereof. service. There are various forms of corrosion tests and
There are other tests which are non-destructive in they all consist of making a welded specimen in a
nature and are performed directly on the welded relatively thin material and immersing it in a reagent at
structure. Non-destructive testing includes visual definite temperature and for definite length of time as
inspection and magnetic particle, liquid penetrant, stipulated by relevant codes or specifications. Most
radiographic, ultrasonic, eddy current, proof test and relevant a m o n g them are those designed to test
leak test techniques. resistance to pitting corrosion or intergranular corrosion.
In welding technology, intergranular corrosion is
DESTRUCTIVE TESTS alternately termed as sensitisation, carbide precipitation
Chemical Test a n d weld decay. The metallurgical reason for
It is often necessary to check the chemical intergranular corrosion in stainless steels has been
composition of the base metal, weld-metal, welding rod explained in Chapter 10 under Carbide Precipitation,
or welding electrode. For this purpose, chippings or while discussing the weldability of stainless steels.
drillings are taken out of the material and subjected to A suitable test for pitting corrosion is described in
chemical analysis in a laboratory. For determining the ASTM G48, "Standard test methods for pitting and
crevice corrosion resistance of stainless steels and related solution is used for each period). The test is applicable
alloys by the use of ferric chloride solution". only to molybdenum-bearing grades of austenitic
The various ASTM Evaluation tests for detecting stainless steels (AISI 316, 316L, 317, 317L) and detects
intergranular corrosion are summarised below: only susceptibility to intergranular attack associated
1) A 262 Practice A: This is an oxalic acid etch test. with chromium carbide precipitates. It does not detect
The test consists of electrolytically etching a polished susceptibility to intergranular attack associated with
sample in 10 weight per cent (wt%) oxalic acid solution sigma-phase or end-grain corrosion, which, so far, are
at room temperature at lamp/cm 2 for 1.5 min. In this known to lead to rapid intergranular attack only in
test, c h r o m i u m carbide or n i t r i d e is dissolved certain nitric acid environments.
preferentially and the microstructure gives an idea of 5) A 262 Practice E: This is a copper sulphate-
c h r o m i u m d e p l e t i o n w h i c h is responsible for s u l p h u r i c acid test. In this test, the specimen is
intergranular attack. If there is no chromium carbide embedded in metallic copper chips and then immersed
present, steps are produced at the grain boundaries due in boiling 6 wt% anhydrous copper sulphate and 16 wt%
to differences in the rate of etching of variously oriented sulphuric acid for 24 hours. After the test, the specimen
grains. This test is useful for screening alloys for other is bent through 180° over a mandrel of diameter equal
tests. to the thickness of the specimen. The bent specimen is
2) A 262 Practice B: This is a ferric sulphate test to be examined under low (5 to 20 x ) magnification,
applied on austenitic stainless steels. In this test, a 6) A 708: This test is similar to A 262 Practice E, except
sample of surface area 5-20 cm2 is exposed for a period that the solution of copper sulphate-sulphuric acid is
of 120 hours to boiling 600 ml solution of 50 wt % H 2 S0 4 used without the addition of copper and the test
+ 2.5 wt % Fe2 (S0 4 ) 3 in an Erlenmayer flask fitted with duration is 72 hours.
an Allihn condenser. The samples are weighed before 7) G 28: While all the above tests are meant for
and after exposure and the corrosion rate calculated austenitic stainless steel alloys, this test is meant for
from the weight loss. It detects susceptibility associated nickel alloys such as Hastelloy, Inconel, Incoloy and
with sigma-phase in chromium-nickel molybdenum Carpenter 20 Cb-3. The test procedure is exactly similar
stainless steels (316, 316L, 317, 317L), which is known to A 262 Practice B.
to lead to rapid intergranular attack only in certain nitric
acid environments. It does not detect susceptibility to Metallographic Test
end-grain attack, which is also found only in certain These tests are rarely called for in specifications but
nitric acid environments. The ferric sulphate-sulphuric they are commonly used for quality control and during
acid test does reveal susceptibility associated with a inspection to determine weld soundness, to examine the
sigma-like-phase constituent in stabilised stainless extent and distribution of non-metallic inclusions in the
steels, AISI 321 and 347. weld, to check the number and arrangement of runs, to
3) A 262 Practice C: This is a nitric acid test, which is determine the metallurgical structure in the fusion zone
also known as Huey Test. In this test, weighed samples and the heat-affected zone and to check the depth of
of 20-30 cm2 area are exposed to 600 ml of boiling 65 weld penetration.
wt% H N O , in a one litre flask fitted with reflux Samples are obtained from a test weld by sectioning
condenser for five 48-hour periods. After each 48-hour or from a production weld by trepanning. Trepanning
period the solution in the flask in changed and the means mechanical removal of the weld joint portion in
sample is weighed. The corrosion rate for each period one piece with a special tool, which leaves a recess in
and the average for the five periods is determined. The the joint which can be easily filled up by welding. From
test detects susceptibility to rapid intergranular attack these samples, macrospecimens or microspecimens are
associated with chromium carbide precipitate and with prepared by suitable etching procedures as follows:
sigma-like-phase precipitate. The latter may be formed Macrospecimens (mild steel welds): If one wants to
in molybdenum-bearing and in stabilised grades of avoid grinding and polishing of the cross-sectional
austenitic stainless steels, and may or may not be visible surface, one can directly place the specimen in boiling
in the microstructure. This test also reveals susceptibility 50% solution of hydrochloric acid in water for
to end-grain attack in all grades of stainless steels. approximately 30 min. Alternately, one may first grind
4) A 262 Practice D: This is a nitric-hydrofluoric acid and smoothen the surface with emery wheel or emery
test. In this test, samples are tested in 10% H N 0 3 - 3 % paper and then vigorously rub the surface with a piece
HF solution at 70°C for two 2-hour periods (fresh of cotton fully wetted with a solution of one part of solid
TESTING AND INSPECTION OF WELDS 557
ammonium persulphate in nine parts of water by proportionality. Beyond P, the curve deviates from the
weight. This rubbing action must be repeated until a straight line. Point E on the curve is the elastic limit.
clearly defined macrostructure appears on the weld. This means that up to this point the specimen returns
After etching, the specimens are washed thoroughly to its original dimensions when the load is removed and
in clear water and wiped dry. They are then immersed thus exhibits elasticity. As the load is increased beyond
in ethyl alcohol, removed and air dried. If a specimen is the elastic limit, there comes a point at which there is a
to be preserved, a coat of thin, clear lacquer may be sudden extension, indicated by the drop of the beam
applied, so that the surface does not rust in storage. and continued extension with a lower load. If the load
Microspecimens: These specimens have to be prepared is removed, the specimen does not recover its original
with a highly polished mirror-like surface and etched dimensions and it is said to have undergone plastic
with proper chemicals and techniques depending on the deformation or plastic flow.
type of alloy and the microstructures one is looking for. In the curve above, the upper yield point is denoted
The procedure of sample preparation and examination by Yu the highest stress before sudden extension occurs,
is complicated and requires suitable equipment and a and its value is affected by surface finish, shape of test
trained metallographer. Interpretation of the piece and rate of loading. The lower yield point, which
photomicrographs also demands a sound knowledge is normally measured in commercial testing, is denoted
of physical metallurgy. by YL, the lowest stress producing the large elongation.
Such large elongation of the metal occurs in very few
Tensile Test materials, such as wrought iron and mild steel. As the
This test is used to determine the yield point or yield load is increased, the specimen continues to extend (i.e.
strength, tensile strength or ultimate tensile stress and plastic deformation) until a constriction occurs in the
percentage elongation of a metal. A tensile specimen of gauge length and the beam drops. This corresponds to
standard dimensions machined from the metal is the maximum load (point M on the curve) on the
inserted in a tensile testing machine (see Fig. 13.2). The specimen. The load now acts on a diminishing area and
machine consists essentially of two parts: the straining produces a stress sufficient to break the specimen. This
or pulling device and an arrangement to measure and maximum stress M is the tensile strength of the metal.
register the load on a dial. A gradually increasing tensile The actual breaking stress which is lower than the stress
load is applied on the specimen and the resultant at M is never used in practice.
extension (or strain) of the specimen is observed. The Though yield strength is the primary criterion of
relation between applied stress (i.e. load divided by load-carrying ability of a metal, much engineering
cross-section) and extension or elongation is indicated design is based on tensile strength, because in
by a stress-strain curve, such as the one shown in Fig. conventional engineering materials the ratio of yield to
13.1, which is typical of a ductile mild steel. Up to point tensile strength is fairly constant. Many of the newer
P on the curve, the stress is proportional to the strain as materials, especially alloy steels, have much higher
indicated by a straight line. It is termed the limit of ratios and therefore in these cases the design is based
on yield strength or proof stress. Proof stress refers to a
point on the stress-strain curve, which represents 0.1%
Soft mild steel extension (B in Fig. 13.1).
In the tensile specimen, gauge length and parallel
/ <Cold length are standard dimensions. These are shown in Fig.
worked 13.2. Gauge length which is usually 50 mm is marked
Slightly
by two points on the specimen before testing, and the
worked final gauge length after fracture is measured. If the
original gauge length of 50 mm (l}) has increased to 60
mm (12) at the end of the test, the percentage elongation
is calculated as follows:
| Elastic L o c a , pecking
Uniform El. J,\.i 100 60-50 100 = 20%
50"
JAJ Elongation (O
0.1 set Percentage elongation is a measure of ductility of a
Fig. 13.1: Stress-elongation curves metal and indicates the safety factor when it is stressed
wei
Fulcrum ^ ^ *ht
0 2 k 6 S 10 12 1<*
Gauge length, cm
Fig. 13.3: Effect of gauge length on percentage
elongation
Fig. 13.2: Tensile testing machine and specimen

concentrations. The energy values determined in an
impact test are used for quality control of a material,
in service. Ductility may be defined as the property for comparison of different materials, for determining
which enables metals to be drawn into wire using a the variation of impact strength with temperature and
tensile force. Percentage reduction of area is another for arriving at the transition temperature. They are not
measure of ductility which can be measured in a tensile useful for engineering design calculations.
test. It is obtained by carefully fitting together the ends Transition temperature is the temperature at which
of the fractured tensile specimen and measuring the the behaviour of the material in impact testing changes
dimensions of the smallest cross- section. The difference from ductile to brittle. Fibrous appearance of the
between this area and the original cross-sectional area, fractured surface indicates ductile behaviour, while
divided by the original cross-sectional area and crystalline appearance indicates brittle behaviour. There
multiplied by 100 gives the percentage reduction of area. are many definitions of transition temperature, some of
It is important to remember that when an elongation which are:
figure is mentioned, gauge length of the specimen a) The lowest temperature at which the specimen
should be stated. Sometimes the factor between gauge exhibits a fibrous structure.
length and diameter at the parallel length is stated (for b) The temperature where the specimen fracture
example, 1 = 5d). This is because the total elongation of shows a 50% crystalline and a 50% fibrous appearance.
a test piece is made up of two separate parts: c) The temperature corresponding to the energy
a) Uniform extension, which occurs up to the maxi- value that equals 50% of the difference between values
mum load (M) and is proportional to the gauge length. obtained at 100% fibrous and zero fibrous.
b) Local extension, due to the necking. This is inde- d) The temperature corresponding to a specific en-
pendent of the gauge length, but varies with the cross- ergy value.
sectional area of the specimen. As the gauge length is There are several impact tests among which the
increased, the effect of the necking on the value of the Charpy-V Notch, Charpy-Keyhole Notch and the Izod
total elongation decreases, as shown in Fig. 13.3. tests are well known. Charpy-V test has replaced the
Impact test: This test measures the resistance of a Izod test in recent years, because it is easier to test
metal to fracture in the presence of a sharp notch (hence specimens over a range of temperature with this test.
also called notched-bar impact test). Minimum impact An impact testing machine and Izod and Charpy-V
values are specified in many codes and specifications impact specimens are shown in Fig. 13.4. To carry out
for ferritic steels, because some of them fail by brittle the test, the specimen is gripped in a vice of an impact
fracture in service, even though they exhibit normal testing machine, with the root of the notch level with
properties as determined from the standard tensile test. the top of the vice. A weighted pendulum, swinging on
Such failure is especially severe when the material is ball bearings, is raised to a standard height and allowed
used in a notched condition. Notched conditions include to strike the specimen on the same side as the notch.
restraint due to deformation in directions perpendicular The striking energy is partially absorbed in fracturing
to the major stress, multi-axial stress, and stress the specimen, and this amount is indicated by a pointer.
2 mm
Dimensions in millimetres
Six Charpy testpieces^taken Notches perpendicular
at mid depth of" surface of plate
ij*square 10 mm square
FjKil mm ^ a
rn round Wfa
Tensile
testpiece
Izod and Charpy-V
Line of cutior
Charpy V tensile testpiece
Fig. 13.4: Izod and Charpy-V machine and specimens
The impact test values are energy values and are

presently expressed in Joules (J), though formerly they
were expressed in ft-lbf and kgf m.
Fig. 13.5: Test assembly as per ISO 2560
All-Weld-Metal Tests
All-weld-metal tensile and impact tests are meant each end of the plate. The reinforcement of the total weld
to determine yield point, tensile, elongation, impact and is not allowed to exceed 3 mm. After each run, the
other values of the weld-metal deposited by a welding assembly is left in still air until it has cooled to a
consumable. All-weld-metal means weld deposit which temperature not exceeding 250°C, the temperature being
is undiluted by the base metal. In these tests, the test taken on the weld surface midway along the run. The
specimens are so prepared that the sections which are tensile test piece is machined to the dimensions shown
subjected to the test consist of pure, undiluted weld- in Fig. 13.6, care being taken that the longitudinal axis
metal. coincides with the centre of the weld, and the mid-
The procedures for preparing the specimens and thickness of the plates. The gauge length of the test piece
testing them are described in many well-recognised is 50 mm. The ends of the test piece may be of any shape
standards. Some of them dealing with manual electrodes suitable for fixing in the testing machine. The tensile
are summarised in the following paragraphs. test piece is heat-treated in an electrically heated furnace
at 250°C for a period of not less than six hours and not
1) ISO 2560-1973 (E) more than 16 hours. The purpose of the heat treatment
For all-weld tensile and impact tests, an assembly is is to remove any hydrogen from the weld-metal. The
prepared, as shown in Fig. 13.5 in the form of a butt tensile strength and the elongation are determined at
joint between two 20 mm thick plates with a single V- room temperature.
groove with a backing plate 10 mm thick and a root gap
of 16 mm. The backing plate is tack welded to the
assembly. By preparing the assembly in this way, the
influence of the parent metal (i.e. base metal) is Gauge length: 50
eliminated. R 5 miru
Electrodes of specified size are used. Each pass is
carried out at normal welding speed and with a welding
ex
current specified by the manufacturer. If the electrode
can be used with both AC and DC, AC is chosen.
Q—r\ □ T L — 1 — fe
00
.—4
Parallel length: 60
Welding is done in the flat position, each layer being
made up of one or several passes, with each pass not ! 70 min.
more than 16 mm wide. Each electrode is consumed
completely (up to a stub end of no more than 50 mm).
The direction of deposition of each layer alternates from Fig. 13.6: Tensile test piece as per ISO 2560
The impact test piece is of the Charpy-V notch type. if X6 < 16 J, the requirements are not fulfilled
Six test pieces have to be taken from the test assembly if X6 > 35 J, the requirements are fulfilled
Their longitudinal axes have to be transverse to the weld
if 16 <X6 < 35 J, a second test assembly has to be
and the upper surface 5 mm from the upper surface of
welded.
the plate. The notch must be positioned in the centre of
the weld and must cut in the face of the test piece,
perpendicular to the surface of the plate. All dimensions From this test assembly only impact test pieces are ma-
have to be in accordance with the instructions given in chined. A second series of 12 test pieces is thus obtained.
Fig. 13.7 and must be carefully checked. The testing
temperature for electrodes receiving symbol 1 defined Second series of test pieces
in the codification system has to be about 20°C (room The average X of the results for the 12 test pieces of
temperature). For symbols 2, 3 and 4, the temperatures the second series, together with the results for the six
h a v e to be 0, - 2 0 a n d -30°C respectively. The test pieces of the first series, is assessed as follows:
temperature must be controlled to within ± 1°C. An
energy level of 27 J is required at the temperature if X18 > 27 J, the requirements are fulfilled
prescribed for the respective symbol. This value is if X18 < 27 J, the requirements are not fulfilled
evaluated in the following way: If a test piece in a series, whose average value does
not fulfil the above requirements shows weld defects
First Series Of Test Pieces which could have affected the results of the test, then
The average X6 of the results of the tests on the first that test piece may be discarded and replaced without
series of six test pieces is assessed as follows: necessarily rejecting the series as a whole.
t/\
i il r'T"^
U-J ^-U=r.
£3 I Position in test assembly
sT\ r
imensions of
testpiece
27,5 J 27,5
-* *+*• •-
55
D imensions
oo
k
Zl of V-notch
025 N L - V
ny
Fig.13.7: Impact test piece as per ISO 2560
2) BS 639:1976 ii) For the second digit of impact value, six test speci-
The procedure for making all-weld tensile and mens must be prepared, three using 4 mm elec-
impact tests specified in this standard is identical to that trode and three using the largest size. When these
of ISO 2560, except for the following additional are tested at the appropriate temperature, the
conditions: average of the three impact values from each set
a) The parent plate for the test assembly must be of specimens must be at least that stated in Table
carbon s t e e l h a v i n g a m a x i m u m t e n s i l e 3.10 "second digit for elongation and impact
strength of 500 N/mm 2 . strength".
b) The temperature of the assembly must be 10-30°C If the average of the three impact values from either
before depositing the first run of weld-metal. or both sets of specimens fails to comply with the
c) Each run of weld-metal must not be less than 2 requirements of the table by an amount not exceeding
mm and not more than 4 mm thick. 15% of the required value, three additional specimens
d) The assembly should not be quenched between for each set concerned, taken from a new test piece, shall
the deposition of individual runs. be prepared and tested and the results added to those
previously obtained to form a new average that shall
e) The portion including the weld-metal may be cut
comply with the requirements. When the average of the
off from the assembly by oxyacetylene or any original three impact values from either or both sets of
other thermal method, provided care is taken to specimens is more than 15% below the required value,
ensure that the lines of cut for the tensile speci- it means the electrode has failed the test.
men are at least 20 mm from the toes of the weld
as shown in Fig. 13.8. 3) DIN 1913-1976
f) During tension test, the stress up to the yield point The procedure is exactly as laid down in ISO 2560.
must be applied at a rate not exceeding 12 N / Regarding the deposition of weld passes (or runs) and
mm 2 per second. layers, the standard states that:
g) Method of arriving at the average impact value: a) The electrode must be melted down till the stub
i) For the first digit of impact value (refer Table 3.9 is not longer than 50 mm
of Chapter 3), six impact specimens must be pre- b) The weaving width for a pass should not exceed
pared and tested at the appropriate temperature. 5 x diameter of the core wire
The average of the six values must be at least 35
c) The length of run deposited with an electrode will
J. If the average of the six impact values is greater
depend on its type
than 16 J but less than 35 J, 12 additional speci-
d) A layer may consist of one pass covering the en-
mens, taken from a new test piece must be pre-
tire width or of a number of passes laid side by
pared and tested, and the results added to those
side
previously obtained to form a new average that
must be at least 27 J. When the average of the e) Before starting a new weld layer, the assembly
original six impact values does not exceed 16 J, it must be allowed to cool in air to a temperature
means the electrode has failed the test. below 250°C.
Line of cut for Line of cut

thermal c u t t i n g _
3 mm approx. 20 mm m i n . for thermal c u t t i n g
excess weld metal i \ ^-f
Approx. 16 mm
View in direction of arrow X
Fig. 13.8: Test assembly and lines of cut as per BS 639
4) IS: 814 -1991 markedly different from those of the previous standards.
The procedure is based on ISO 2560. Attention should Hence they will be described in full detail.
be drawn to the following special points: The test assembly is prepared as shown in Fig. 13.10
a) The parent plates for the test assembly must be from steel plates conforming to one of the following
mild steel conforming to IS: 2062-1969. ASTM specifications or their equivalent:
b) While the specified width of the test plate is 100 a) Specification A285, Pressure Vessel Plates, Car-
mm minimum in ISO 2560 and in BS 639, it varies bon Steel Low and Intermediate Tensile Strength,
in IS 814 according to the size of the electrode as Grade C
follows: b) Specification A283, Low and Intermediate Ten-
Core diameter Plate width sile Strength, Carbon Steel Plates of Structural
mm mm Quality, Grade D
4 90 ± 10 c) Specification A36, Structural Steel.
5 and 6.3 120 ± 10 Electrodes m u s t be tested in the as-received
8 150 ± 10 condition, except for the low-hydrogen electrodes which
10 and 12.5 180 ± 10 must be dried at 260-370°C for two hours immediately
c) Temperature of the plate assembly must be 25- prior to testing.
29°C before commencing welding. The assembly must be welded in the flat position
d) In order to counteract shrinkage deformation, the using appropriate size of electrode and the type of
test plates should be pre-set as shown in Fig. 13.9. current depending on the classification of the electrode.
While tack welding the assembly in such a way The assembly must be pre-set or restrained during
as to obtain a level joint after welding, no pre- welding to prevent warpage in excess of 5°. A test
setting of the backing strip is required, and be- assembly that has w a r p e d more than 5° must be
fore commencing welding, care should be taken d i s c a r d e d . Welded test assemblies m u s t not be
to see that the opening on the upper side of the straightened. The test assembly must be tack welded
V-joint is not less than 28 mm. a n d p r e h e a t e d to 110 ± 14°C. Welding m u s t be
continued with an interpass temperature of not less
28 mm min.
r -i
than 110°C, nor more than 180°C as measured by
t e m p e r a t u r e i n d i c a t i n g crayons or surface
thermometers at the point specified in Fig. 13.10. The
pass sequence must be as called for in Fig. 13.10. Each
pass must include at least one start and stop within the
d = Test plate thicfoess area of weld that must meet radiographic soundness
5 = Backinq plate thickness requirements. The direction of welding to complete a
Fig. 13.9: Presetting of test plates for tack welding pass must not vary: however, the direction of welding
for different passes may be alternated. If it is necessary
e) The direction of deposition from one end to the to interrupt the welding procedure, the assembly must
other must change with each layer. be allowed to cool in still air to room temperature. When
f) Each weld run must not be less than 1.5 mm and ready to r e s u m e w e l d i n g , the assembly m u s t be
not more than 3 m m thick. preheated to a temperature of 110 ±14°C. The procedure
g) The length of run per electrode (minus the stub) used for completing the weld must be continued as
must not be less than 0.4 (1-50), where 1 is the described before. No thermal treatment is permitted
total electrode length in mm. on the test assembly s u b s e q u e n t to the w e l d i n g
h) No run should start or end in the operative part operation.
of the test piece. One all-weld tension test specimen as shown by Fig.
i) Between the completion of one run and start of 13.11 must be machined from the same test assembly.
the next run, the text assembly must be allowed As Fig. 13.11 shows, two sizes of tensile specimens are
to cool in air to a temperature below 250°C. prescribed: the larger one for electrode sizes 4 to 8 mm
(using plate thickness of 19 mm or more) and the smaller
5)AWSA5.1-91 one for electrode sizes 2.4 and 3.2 m m (using plate
The procedure and conditions for making all-weld thickness of 12.7 mm). For all electrode classifications,
tensile and impact tests prescribed in this standard are except the low-hydrogen classi fica- tions (E7015, E7016,
E7018, E7048 and E7028), the machined tensile specimen specimens from the test assembly must be obtained at
must be aged at 95 to 105°C for 48 ± 2 hours. The the test temperature specified for the classification being
specimen must then be cooled to room temperature and tested. When computing the average value of the impact
subjected to tension until rupture. The tensile specimen properties from the set of five specimens, the lowest
from the low-hydrogen classifications must not be aged value and highest value obtained must be disregarded.
prior to testing. Two of the three remaining values must be greater than
Five Charpy-V notch impact test specimens, as the specified 27 J energy level; one of the three may be
shown in Fig. 13.10 must be machined from the same lower but not less than 20 J. The computed average value
test assembly. No thermal treatment is permitted on the of the 3 values must be equal to or greater than the 27
test specimens. The impact properties of the five J energy level.
l+~ 1 mm - V 2 lenqth-
Poini ol
temperature
measurement
Impact *- A l l - w e l d - m e t e l
specimens tension s p e c i m e n
■i | * - 1 mm
(a) Test plate showing location of test specimens

.T/2
SI Equivalents
in. mm
1/16 1.6
WeW Q WeW Q 1/2 13
1 0
Section AA Section AA 5 125
(b) Orientation and (c) Location of
location of impact 10 250
all-weld-metal
specimen tenion specimen
(T) (R) (B) Split weave

Electrode Plate thickness Root Backing Full weave
size (min) opening thickness for Press Number
mm mm mm mm layer no. Layer no. per layer of layers
2.4 12.7 6.4

3.2 12.7 6.4 1 3 to top 2 5 to 7
4.0 19 6.4 1 2 to top 2 7 to 9
4.8 19 6.4 land 2 3 to top 2 6 to 8
5.6 19 13 12.7 land 2 3 to top 2 6 to 8
6.4 25 12.7 1,2 and 3 4 to top 2 5 to 7
8.0 32 12.7 1, 2 and 3 4 to top 2 5 to 7
Fig. 13.10: Preparation of test assembly as per AWS A5.1
-•G'
LL
B
K~ *- D "M
G = Gauge length
Test plate Dimensions in mm Approximate
thickness (mm) D G C B F, min area (mm)
12.7 6.4 + 0.13 25.4 + 0.13 31.8 9.5 4.6 32

19.1 12.7 + 0.25 50.8 + 0.13 57.1 19.1 9.5 129
Fig. 13.11 All-weld tensile specimen as pet AWS A5.1
Cod Test sense as yield pointer proof stress. Knowing Kk of an

COD stands for crack opening displacement. Some alloy, the designer is able to specify acceptance limits
authorities prefer to call it CTOD test which means crack for sizes and shape of defect, so that fracture will not
tip opening displacement test. This test is normally not occur at the design stress. This elastic analysis method
specified in codes and specifications, but becomes a is well suited for high strength alloys but not for
r e q u i r e m e n t for a p p r o v i n g steels a n d w e l d i n g conventional lower strength alloys, in which fracture is
consumables meant for critical structures such as often associated with local plasticity. In their case, the
offshore platforms to be erected in rough and cold seas. occurrence of this yielding brings with it the separation
Though Charpy test is being used as a guide and a of the crack surface, even at the crack tip, before crack
quality control test for toughness, its limitations from begins to p r o p a g a t e . This critical crack o p e n i n g
the design point of view are being felt with the displacement (COD = a) is regarded as a parameter for
increasing use of new alloys for which previous measuring fracture toughness of ductile materials.
experience is lacking. The concept of linear fracture COD testing, which is detailed in BS 5762:1979 is very
mechanics is now being applied to design problems to expensive and fairly complex. In U.K., for example,
give fracture-safe structure containing a defect of known consumable producers and fabricators find it more
size. economical to get their testing done by the Fracture
This concept involves a study of stresses and strains Mechanics Laboratory of the Welding Institute at
at the tips of sharp cracks with respect to the applied Abington or any other similar testing centre.
loading. Under elastic conditions, this stress field can Hence the commonly asked question: is it possible
be described in terms of a single parameter known as to establish a correlation between Charpy-V and COD
the stress intensity factor K. The critical value of K which values? The answer is no.
will cause the crack to grow under a static load is called The distinctions which must be made between COD
K , which represents the fracture toughness of the and Charpy data result from the entirely different
material. As the material thickness increases, Kc natures of the two tests. The Charpy test measures both,
approaches a minimum value Klc under plane strain the energy required to initiate fracture in a specimen
conditions, i.e. conditions of high triaxiality and high and the energy required to propagate the crack through
restraint. Klc stands for the critical plane strain stress the specimen, although it is not possible to separate the
intensity factor, and is a material property in the same two. The COD test, on the other hand, is a measure only
of the strain which can be accommodated in a specimen structures must have a high resistance to cleavage in
before fracture initiates. Fracture p r o p a g a t i o n the first instance, since fracture initiation by this
parameters are not included in this test. There are also mechanism in most situations will lead to catastrophic
other important differences in the specimen type, which failure of the structure.
should also be emphasised. These are as follows: COD test which assesses resistance to initiation of
a) The Charpy specimen is a standard size (i.e. 10 x fracture is considered to be well suited for investigating
10 x 55 mm) irrespective of the original plate thickness, and distinguishing between resistance to the two above
and contains a blunt notch 2 mm deep, whereas the COD mentioned fracture mechanisms, since the toughness
specimen is ideally made using the full thickness of the values can be readily linked to the micro~ing and to
plate and contains an extremely sharp fatigue crack as the mechanism of initiation by fracto-graphy. Such an
the notch. investigation is difficult with the Charpy-V test, since
b) The notch tip in a COD specimen is much more it is impossible here to separate clearly the contributions
highly restrained than in a Charpy specimen. At a given of the two fracture mechanisms except at extreme ends
temperature, this can give rise to different failure of the transition curve, where 100% crystallinity or 100%
initiation modes in the two tests. fibrosity is developed.
c) A through thickness notch COD in, for instance,
weld-metal will sample a w i d e variety of Microstructure vis-a-vis toughness: Important
microstructures of varying toughness, e.g. the root microstructural constituents of C-Mn and low-alloy
region may be expected to have lower toughness than steel w e l d - m e t a l s placed in order of decreasing
other regions, whereas the Charpy test may not sample transformation temperature are:
the region of worst toughness. Thus, in any comparison 1) Proeutectoid ferrite nucleated in prior grain
of test data, care should be exercised to ensure that the boundaries and usually existing as thin bands outlining
microstructures at the point of fracture initiation in the a columnar grain structure
two tests are the same. 2) Isolated regions of ferrite-carbide aggregates such
d) The Charpy test is an impact test, whereas the as pearlite
rate of loading in COD testing is very low. This again 3) Ferrite plates nucleating in a side-by-side manner
can lead to different mechanisms of failure initiation. resembling upper bainite
To explain the COD test briefly, a test piece as shown 4) Narrow interlocking ferrite plates known as
in Fig. 13.12 in which a sharp crack has been developed acicular ferrite
from a machined notch by fatigue, is subjected to 5) Small areas of retained austenite (seen with
loading at a known temperature in three point bending. electron microscope), which in certain weld-metals is
Test results are produced by an autographic recording found partly or wholly transformed to martensite.
of the applied force against the clip gauge displacement Research studies aimed at establishing relationship
measured at the test piece surface, across knife edges between weld-metal toughness and its microstructure
located on opposite sides of the notch. The measured have indicated that a high resistance to cleavage and a
clip gauge displacement at the fracture, or the onset of low transition temperature will result if the proportion
crack extension (critical clip gauge displacement), is of interlocking acicular ferrite structure is kept high and
related to the crack tip opening displacement at this that of proeutectoid ferrite and upper bainite is kept
point, using a previously determined calibration. low.
For offshore construction in the North Sea, minimum Fracture by microvoid coalescence is considered to
COD requirement of 0.44 m m at -10°C has been be of less practical importance in determining risks of
specified for the weld-metal in certain critical areas of unstable fracture than fracture by cleavage, at least in
the structures, as indicated in Table 7.2 in Chapter 7. C-Mn weld deposits. However, a thorough knowledge
Weld-metal fracture mechanism: Toughness tests on of factors which improve resistance to microvoid
weld-metals have shown that there are two principal coalescence is expected to help meet C h a r p y - V
fracture mechanisms, namely, fracture by cleavage (see requirements more easily.
Fig.13.13A) and fracture by microvoid coalescence (or Charpy-V test results depend mainly on the non-
dimpled fracture) (see Fig. 13.13B). The shape of the metallic inclusion population. Low volume fractions
transition curve of toughness versus temperature for a and a high mean free spacing of the inclusions are
given weld deposit is determined by its resistance to needed for good toughness. Low inclusion volume
these two mechanisms. Weld-metal deposited on critical fractions result principally from control of oxygen and
Bt0 8%
W±04% L
Section through
notch
» 1
_ _
|//|02%W|A[
lllO/»%W|A
1 j / 6 0 ° nominal
Width = W
Thickness = B = 0.5 W
Half loading span L = 2W
Notch width N = 0.065 W max. (if W is over 25 mm) or
= 1.5 m m max. (if W is less than or equal to 25 mm)
Effective notch length M = 0.25 W to 0.45 W
Effective crack lenght a = 0.45 W to 0.55 W
Note 1. Sawn notches may also be used.
Note 2. Where one (ormore) of the specimen surfaces is the original product surface, machining of that surface
is not required
All dimension in mm.
Fig. 13.12: Proportional dimensions of COD te^ piece as per BS 5762-1979
sulphur levels in the deposit. Oxygen levels can vary Submerged-arc 0.02-0.15%
markedly b e t w e e n w e l d s d e p o s i t e d by v a r i o u s MIG 0.01-0.02%
processes. Typical vacuum fusion analyses have given C0 2 -shielded 0.04-0.06%
the following oxygen ranges: Sulphur contributes directly to the inclusion volume-
Manual metal-arc 0.03-0.07% fraction by forming sulphide inclusions. For highest
(A) (B)
Fig. 13.13: Fracture mechanisms in weld-metal: (A) Cleavage; (B) Microvoid coalescence
toughness, tight control on sulphur in wires, fluxes and then filled up by welding with 4 mm size electrodes in
the base plate is necessary. the weld position applicable to the test piece. After
To sum up, high COD values of weld deposits can completion of welding, test pieces are cut by sawing or
be achieved by choosing the welding consumable and machining to form one transverse tensile, one face bend
welding heat input in such a way that they will result and one root bend, and three impact test specimens as
in a most favourable weld deposit microstructure. For indicated in Fig. 13.14. Gas cutting is permissible
getting high Charpy-V values, one must choose a provided a minimum of 3 mm is machined off. In all
welding process and consumables, which give the the test pieces, the upper and lower surfaces of the weld
lowest sulphur and non-metallic inclusion contents in
the weld deposit. -55-
Discard
c
Transverse Tests
Transverse tensile, impact and bend tests are often "T-Charpy
specified as quality control tests or as acceptance tests r-J test f
for electrodes. Unlike all-weld-metal tests, these tests ) pieces
are made on butt welds in which the weld-metal has
been diluted by the base metal and it has not been heat-
Transverse
tensile
Test
specimen
i
50
treated for the removal of hydrogen. Thus these tests
are meant to confirm the compatibility of electrodes with
respect to standard base metals and they are more
representative of the practical conditions of welded
fabrication than the all-weld tests.
Face bend test
Root bend test

Specimen
Specimen
i
30
According to IS 814, the test pieces are made as Discard
L +♦
shown in Fig. 13.14 by welding together two plates 15
to 20 mm thick and not less than 100 mm wide. The
edges are bevelled to give an included angle of 70°, root U 100 min «*W-
3&
face of 3 mm and maximum gap of 3 mm. The welding 30°
position can be flat, horizontal-vertical, vertical or to 20
overhead as required for the qualification of the f o to y
All dimensions in millimetres.
electrode. After the V-groove is filled up by multi-pass
welding, a groove is cut at the root on the backside to a Fig. 13.14: Method of preparing transverse tensile,
depth of 3 mm as shown in Fig. 13.15. This groove is impact and bend test specimens
Premature failure at corners of the test specimen is not

considered a cause for rejection.
Nick-Break Test
This test is applied to a welded butt joint in plate or
pipe to check weld soundness. The typical specimen
dimensions for plate and pipe are shown respectively
Fig. 13.15: Grooving in preparation for deposition of in Figs. 13.18 and 13.19. It is not necessary to remove
sealing run the weld reinforcement. The specimen is supported as
shown in Fig. 13.20 and ruptured by giving one or more
must be filed, ground or machined level with the sudden heavy blows with a hammer at the point of the
respective original surfaces of the plates. Dimensions nick or the notch as indicated. The rate of applying the
of the transverse tensile specimen are shown in Fig. force is not critical, because it has no effect on the
13.16. The bend test specimen is made 30 mm wide. Its appearance of the fractured surface.
sharp corners are rounded off. The transverse impact
specimen is machined out as shown in Fig. 13.17.
These edges may be flame cut
IS:814 specifies that the transverse tensile specimen 6 mm approx
tested in tension should give tensile strength in the range
of 410-510 or 510-610 N / m m 2 , d e p e n d i n g on the
classification of the electrode. The transverse impact test
results are assessed in the same manner as those for all-
weld impact tests. The transverse bend test specimens
are bent through an angle of 180° over a mandrel having
a diameter equal to three times the thickness of the
specimen. One test specimen is tested with the face of
the weld in tension and one with the root of the weld in Fig. 13.18: Method of preparing nick-break specimen
tension. According to IS:814, the electrode complies with from plate
the test if on completion of the 180° bend no crack or
defect at the outer surface of the test specimen is greater
than 3.0 mm measured across the test specimen or 1.5 These edges may be flame cut
mm measured along the length of the test specimen. 6 mm approx
r W i d t h to
*0R I »uit grip
T
1
OF?—
-t.Jj-
I A * P a r a l l e l Lenqth = Width
of w e l d ► 6 m m e a c h s i d e
Fig. 13.16: Dimensions of transverse tensile test

Fig. 13.19: Method of preparing nick-break specimen
specimen
from pipe
Test piece taken

from middle of Notch perpendicular
Load
plate thickness to surface of plates
15 to 20
T
All dimensions in millimetres
Fig. 13.17: Method of preparation of Charpy test
mnm
specimen Fig. 13.20: Method of rupturing nick-break specimen
The surface of the fracture is examined for the gauge lines may be given by holding the specimen in
presence of internal defects, such as lack of fusion, slag the jaws of a vice with one-third length of the specimen
inclusions and porosity This test can be conveniently projecting from the jaws and giving hammer blows. A
carried out in shops and at sites for quick assessment of similar bend is given to the other end of the specimen.
weld quality, since no elaborate equipment is necessary Both ends should be bent through the same angle, and
and the interpretation is relatively simple. they should be symmetrical with respect to the weld.
Final free bend is achieved by applying compressive
Free-Bend Test forces to the two ends to gradually decrease the distance
This test measures the ductility of the weld-metal in between them. If a crack or depression exceeding a
a butt joint in a plate or pipe in terms of elongation specified size a p p e a r s on the convex face of the
percentage. Thickness of the specimen is the same as specimen, the load is withdrawn and the specimen
that of the plate or the pipe, width is 1.5 times the removed. If no cracks or other defects appear, the
thickness and the length is anywhere between 150 mm specimen is bent completely to 180° and then taken out.
for 6 mm t and 450 mm for 50 mm t. The length is not In either case, the final gauge length is measured to
mandatory. The weld is dressed to the plate level by the nearest 0.25 mm. The difference between the final
filing, grinding or machining, bearing in mind that the and original gauge length divided by the original gauge
tool marks should be lengthwise to the specimen. It is length and multiplied by 100 gives the elongation
important that the line between the weld-metal and the percentage which is taken as a measure of weld-metal
base metal is distinctly visible, and if necessary the ductility
surface of the specimen should be etched with a suitable
reagent.
FILLET WELD TESTS
Two gauge lines are lightly marked on the specimen.
The distance between the gauge lines (gauge length) Fillet Weld-Break Test
should be 3 mm less than the width of the weld. It should This test is similar to the nick-break test and is meant
be measured to the nearest 0.25 mm. Gauge lines must to test the weld soundness. The test specimen, as shown
appear on the wider face of the weld if it has unequal in Fig. 13.22 (a) and (b), consists of two plates making a
faces. tee joint and a fillet weld laid on the outer side of the
joint. The base metal, weld-metal and the welding
The specimen is first subjected to an initial bend in a
parameters must be strictly as prescribed in the welding
device such as shown in Fig. 13.21. The face having
procedure being qualified. Tack welds at the ends of
gauge lines must face downwards. The weld should lie
the joints should be avoided as they make it difficult to
exactly at midspan of both the supports and the loading
rupture the specimen. The manner of positioning the
block. Alternatively, a bend of 30-45° away from the
specimen and direction of force are also shown at (c) in
Fig. 13.22. The force may be applied by means of a
testing machine, a press or h a m m e r blows. The
fractured surfaces are examined for internal defects
such as porosity, incomplete root penetration and slag
inclusions.
ii m i o V Transerverse Fillet Shear test

This test is used to determine the shear strength
iA) W - Weld width fj of a transverse fillet weld. The specimen dimensions
G-Gauge length // are shown in Fig. 13.23. The width of the specimen is
t-Thickness / / m e a s u r e d in m m . A tensile load is applied to the
specimen in a testing machine until it ruptures. The
maximum applied load in N is recorded. The shear
strength of the welds per linear mm is obtained by
dividing the maximum load by twice the width of the
Fig. 13.21: Method of executing free-bend test: (A) Test specimen. The shear strength of the welds (in N/mm 2 ) is
specimen; (B) Specimen given initial bend; (C) Final obtained by dividing the shear strength (in N per linear
free bend mm) by the average throat size of the welds (in mm).
t = F + 3 mm
Minimum length
of weld= L-25 mm
Fig. 13.22: Fillet-weld break specimen and method of rupturing
f
112.7 mm-*
^0 mm
""IIP * ! *
19 mmf~[
Jl f \ 9.5 mm +0
9.5 t 1.5 mm—H P* ♦ .1.5 mm
Leq dimensions
for ell welds
Fig. 13.23: Transverse fillet-weld shearing specimen
Longitudinal Fillet Shear Test Fillet Weld Tee-Bend Test

This test measures the shear strength of a This is not a standard test but is sometimes used to
longitudinal fillet weld. The dimensions of the roughly compare the ductility of double-fillet tee joints.
specimen after welding are shown in Fig. 13.24 and Details of welding the tee joint, sectioning the specimen
those after machining are shown in Fig. 13.25. The and bending it in a guided jig until the angle of bending
length of each w e l d is measured in mm. The (angle A + angle B) is 120° or until the specimen fractures
specimen is ruptured in a tensile testing machine and are shown in Fig. 13.26. The test also determines the
the maximum load is recorded. The shear strength effect of welding on the HAZ and thus predicts the
of the welds in N per linear mm and in terms of N / service performance of a tee joint.
mm 2 is calculated in the same manner as with the The following four types of results are obtained in
transverse test. the order of decreasing ductility of the welded joint:
71
±1
8" min.- -8" min.- J
t—k l/2"-H 4 Standard **5°
2 \fr" fillet welds, size F
Dimensions F-î
Size of weld F,in. [1/8 \m[3/8 [1/2I

1 Thickness t, in.,min.i3/8 tl/2 |3/i"_Tj
| ThicknessT, in.,min. 3/8 3/4 1 ~M
1 Width W, in, 3 '3 |
1
7]m
Fig. 13.24: Longitudinal fillet-weld shearing specimen after welding
1 1/2"«-, . r-*l 1/2"
£1 ►5M
C 1
Note: for other dimensions see fig. 13.24
Fig. 13.25: Longitudinal fillet-weld shearing specimen after machining
Type 0 — The specimen bends to 120° without Type 2 — A slowly progressing crack which starts at
fracture or tear of any kind. the toe of the fillet and extends eitker directly into the
Type 1 — A crack which starts in the bond zone at plate metal or follows the fusion zonefora short distance
the toe of the fillet and follows the fusion zone under and then turns into the plate metal, but does not
the weld, but does not turn into the plate metal. continue entirely across the plate metal.
Roller
support
Discard
) Bend Specimens
^ Hardness Specimen
&8end Specimens
|»- Discard
Direction
of Rollinq
[after welding TEST PLATE FOR

TEE BEND TEST
Fig. 13.26: Fillet weld guided tee-bend test
Type 3 — A sudden or sharp crack which generally is provided. For softer metals like brasses and bronzes,
starts at the toe of the fillet and extends into and entirely a load of 500 kg is applied for a similar or longer time.
across the plate metal. The diameter of the impression formed on the metal
surface is measured with a special Brinell microscope
Hardness Tests graduated in millimetres. The average of two diameters
The term, hardness of a metal, cannot be precisely at right angles to each other is determined and the Brinell
defined and has many meanings. In general, hardness hardness number (BHN) obtained by referring to a chart
usually implies a resistance to deformation, and in the indicating diameters of impression versus hardness
case of metals, it is a measure of their resistance to plastic number.
or permanent deformation. The harder a metal is, the To ensure true values, the surface being tested should
more it resists scratching, wear, penetration, machining be kept flat and fairly smooth. The specimen should be
and mechanical cutting. Tensile strength also increases firmly supported in such a way that the load is applied
with hardness. Hardness of the base metal and weld- normal to the surface. Impressions should be separated
metal of a w e l d e d joint is affected by chemical from each other by at least two diameters to avoid
composition, the metallurgical effects of welding, cold getting wrong results due to the cold work caused by
working of the metal, heat treatment and several other the previous impression.
factors. Limitations have to be placed on the hardness
of the heat-affected zone and weld-metal, because if they 2) Vickers
are too hard, they will not have sufficient ductility for In this test, the impression is made with a pyramid-
the service conditions and their corrosion resistance may type diamond indenter, using a load of 50 kg for hard
also be impaired. metals, 30 kg for normal metals and 10 kg for soft, thin
There are nearly 30 m e t h o d s of h a r d n e s s or surface hardened metals. The time of load application
measurement, but the most widely used are: is usually standardised at 10 sec. The impression appears
1) Brinell hardness as a d a r k s q u a r e on a light b a c k g r o u n d . The
2) Vickers hardness measurements are taken across the diagonals of the
3) Rockwell hardness. square, and the hardness value corresponding to the
readings is obtained from a chart or calculated by a
1) Brinell simple formula.
For Brinell hardness testing, a hydraulic machine is The Vickers testing machine is a precision instrument
used to impress a special hardened steel ball of 10 mm which must be operated with care. The application of
on to the specimen, applying a load of 3,000 kg for at the load and its withdrawal after a pre-set interval are
least 15 sec, for which a hand-operated p u m p or wheel controlled automatically. An audible click tells the
operator that the test is completed. After the indenter is should be periodically examined and replaced when
withdrawn, the measuring microscope is swung into a necessary.
position over the impression. Three knife edges are Summary: These three testing methods supplement
provided: one is fixed, another can be moved with a each other. The Brinell test which gives a large
micrometer screw connected to the counting impression is well suited for examining a large area, such
mechanism, and the third serves to give rapid readings as the face of a weld or the base metal. The Rockwell
to specified limits. The fixed knife edge is set in line and Vickers tests give small indentations and can be
with the left-hand corner of the impression, the movable used to survey the" hardness of narrow zones such as
knife edge adjusted to coincide with the right-hand the cross-section of a weld, individual weld beads or
corner of the impression, and the reading obtained the heat-affected zone. For conversion of hardness value
directly from the indicator. Then reference is made to from one system to another, Table 13.1 can be used, but
the chart to arrive at Vickers hardness number. the conversions are approximate. The values given in
In Vickers testing, the hardness number obtained is the table are average values obtained from a large
fairly constant, irrespective of the load applied. While n u m b e r of tests. In actua 1 êsts, one meets with
in Brinell testing, values above 600 Brinell are not considerable scatter.
reliable, because of the flattening of the steel ball under
the heavy loads which have to be applied on very hard Hot Cracking Tests
materials, Vickers test can be used accurately up to the For determining the hot cracking propensity of
highest hardness limit, because the diamond indenter metals, the most sensitive tests are:
shows no distortion. Also the Vickers impressions are 1) Varestraint test
much smaller than those obtained in Brinell or Rockwell 2) FISCOtest
tests, and hence the surfaces to be tested must be 3) Lehigh restraint test
prepared with much greater care. 4) Tekken Y-groove test.
1) Varestraint test
3) Rockwell
This test was developed by Savage and Lundin at
The Rockwell h a r d n e s s tester is a precision
Rennselaer Polytechnic Institute, U.S.A. in 1965. It
instrument which uses a hardened steel ball of 1.6 or
comprises a rectangular bar, of which one end is
3.2 mm diameter for softer metals and a cone-shaped
m o u n t e d in a vice and the other is attached to a
diamond penetrator for hard metals. The test is started
hydraulic plunger. The plunger bends it to a controlled
by applying a minor load of 10 kg, which helps to locate
radius at an appropriate moment, while a longitudinal
the penetrator in the surface being tested. The dial is
bead is being deposited on the side exposed to tensile
turned to the point marked Set, and then the major load
deformation. The amount of deformation required to
of 100 or 150 kg is applied. After the pointer comes to
cause cracking serves as the index of cracking sensitivity.
rest, the major load is withdrawn, leaving the minor
Both weld-metal and base metal hot cracking can be
load still on. The Rockwell hardness number, which is
investigated in this test.
based on the difference b e t w e e n the d e p t h s of
penetration at major and minor loads, is read directly Figure 13.27 shows the test being used to study the
on the dial. hot-cracking susceptibility of stainless steel plate, in
which a TIG welding torch is used to melt the plate
There are several Rockwell h a r d n e s s scales
surface.
depending on the penetrators and major loads applied.
Among them the B and C scales (RB and Rc) are the most 2) FISCO test
common ones. The B scale uses 1.6 mm steel ball and In this test devised by Mr. Schnadt of Oerlikon,
major load of 100 kg while the C scale uses the diamond Zurich, one or more small beads are laid one after
penetrator at the major load of 150 kg. There are another on top of the gap formed by two plates, which
differently coloured dial figures for the different scales. are placed butt-to-butt and clamped very rigidly all over
Several precautions have to be taken to ensure by means of a special device. The abutting edges of the
accurate results, and some of them are common to the plates on which the beads are to be deposited may be
other testing m e t h o d s . They are detailed in the prepared as required. After welding,<he plates are
Instruction Manual issued with the machine and must removed either immediately or after letting them cool
be carefully observed. The ball penetrator and the cone during a couple of days or even a few weeks. The beads
penetrator may get blunted in constant use. They are then broken by striking one plate with a hammer,
Table 13.1: Hardness conversioni table and tensile strength of steels

Vickers Rockwell Brinell Sclero- Tensile strength
scope
C B Steel Tungsten MN/m2 tpnf/in2
scale scale ball carbide
ball
iôoo 69 — — — 99 — —
950 68 — — — 97 — —
900 67 — — — 95 — —
850 66 — — 750 91 — —
800 64 — — 722 88 — —
750 62 — — 691 85 — —
700 60 — — 656 81 2320 150
650 58 — — 611 78 2160 140
600 55 — — 564 74 1990 129
580 54 — — 545 72 1930 125
560 53 — — 525 71 1850 120
540 52 — 496 507 69 1790 116
520 51 — 480 488 67 1730 112
500 49 — 465 471 66 1650 107
480 48 — 448 452 64 1590 103
460 46 — 433 433 62 1510 98
440 45 — 415 415 59 1450 94
420 43 — 397 397 57 1370 89
400 41 — 379 379 55 1310 85
380 39 — 360 360 52 1240 80
360 37 — 341 341 50 1170 76
340 34 — 322 322 47 1110 72
320 32 — 303 303 45 1030 67
300 30 — 284 284 42 970 63
280 27 — 265 265 40 910 59
260 24 — 247 247 37 830 54
240 20 98 228 228 34 770 50
220 — 95 209 209 32 690 45
200 — 92 190 190 29 630 41
180 — 87 171 171 26 590 38
160 — 82 152 152 24 530 34
140 — 75 133 133 21 460 30
120 — 67 114 114 — 390 25
100 — 56 95 95 — — —
whilst the other plate is solidly clamped in a vice. surface there is a series of closely lined, very sharp and
Fractures are then examined by eye or by microscope solid cogs which lie parallel to the opening of the
and cacked areas are entered on special form. The results apparatus. Although the plate is backed, it is freely
are expressed as explained further on. movable in the apparatus.
Test severity depends on plate thickness, shape of In its middle is a small furrow into which a copper
the chamfer, and in case of similar shapes, on the gap. or steel distance gauge is put. As a result, the gap-width
The rigid clamping of plates is achieved by means a (a = 1-6 mm) can be regulated and kept up accurately.
of the FISCO apparatus shown in Fig. 13.28(A). It allows The plates are positioned in such a way on the cogged
the mode of fastening sketched out in Fig. 13.28(B) to plate that the edges to be welded just reach the distance
be realised. gauge placed beforehand into the furrow.
The apparatus involves a solid case C in the shape The apparatus involves 10 perpendicular clamping
of an almost wholly closed U, of particular high-strength screws V1 and four horizontal clamping screws V2. The
steel. Within it there is a massive plate F, on whose upper Vj screws are designed to push the plates strongly
at 120°C. Both manual and semi-automatic welding may

be used. During the operation, amperage is checked
continuously. It ordinarily corresponds to the normal
amperage used in downhand welding. DC electrodes
may be connected to either the positive or the negative
pole.
A l t h o u g h w e l d i n g is generally executed in
downhand position, the FISCO apparatus is equally well
suited for vertical welding. Following welding, the
plates are ordinarily left to cool in the apparatus and in
still air. Cooling can, however, be speeded up by the
action of compressed air or by sprinkling cold water.
Clamping may be maintained for some hours or days
Bending load
Dimensions are In millimetre*
in order to determine the effects of time on the crack
sensitivity of beads. Normally, the test lasts about ten
Fig. 13.27: Diagrammatic sketch of the varestraint test minutes, excluding cooling time which may vary
considerably.
against the cogged plate, whereas the V2 screws cause
the plates to be pressed against the copper distance Representation Of Results
gauge. A special key serves to clamp the screws strongly After taking the workpiece out of the apparatus, and
and speedily by hand. By these simple means, an extra- breaking it in a vice, the fractured areas of the beads are
rigid clamping of the plates is achieved. examined by eye or by microscope, so as to detect the
The test data may vary as follows: cracked areas.
plate thickness 1-40 mm These areas are subsequently entered, with a red
gap-width 0-6 mm pencil in natural size, on a specially designed sheet in
chamfer shape : any order to fix their dimensions and exact positions.
The plate length (chamfer direction) is of 200 mm, Taking into account the general test conditions the
the initial width is 120 mm maximum. The width gets crack sensitivity of an electrode is expressed by the
reduced due to the fact that the same plate may be used following equation:
several times as long as its width exceeds 70 mm.
The gap-width a depends on the diameter d of the
100 x total crack length
electrode; as a rule, the maximum size of the gap equals f=- — in%
the diameter (a = d). a
total length of the four bead
In case the test is carried out with thin plates, not
reaching 6 mm width, a 8-10 mm thick plate, designed The index a represents the gap-width in mm. Thus
to take u p the pressure of the screws, is placed in for instance, the symbols f0, iv f2, f3 and f4, are used for a
Between the V1 screws and the plates to be welded. = 0 , 1 , 2, 3, and 4 mm respectively. Generally the test is
Deformation of plates during welding is thereby carried out with various gap-widths. If fa(%) as a
avoided. In such a case, the V2 screws are not utilised. function of a (mm) is entered on a graph, the crack
Square edges and Y-shaped ones are widely used, sensitivity curve of an electrode is obtained for a given
although any other shape may be used as well. In order steel quality and for the test conditions chosen. There is
to obtain agreat number of starts and ends of runs evidence that this curve always passes by the initial
(which zones are commonly known to be most sensitive point of the graph (f0=0) and rises more or less steeply
to hot cracking) four small beads, each about 40 mm with the gap width a.
long and 5 mm distant from the preceding one are The greater the value of a, the stronger the severity
deposited (used chamfer length 200 mm). of the test. As a result, test severity may be varied and
Square edges are mostly used, because they require planned in advance, which is most important.
little preparation and are wholly defined by their gap- If fa < 20%, only ends or beginnings of runs are
width a in mm. The beads are simply deposited on top cracked as a rule (crater). Values exceeding 20% cause
and along the axis of the chamfer thus formed. Before cracks to appear all over the bead. If cracks are to be
utilisation, the electrodes are always dried for two hours retained for further record, one has only to cut the edges
m^k
(B)
Fig. 13.28: Clamping mechanism for FISCO test. (A) Finished clamp; (B) Diagrammatic sketch
away with the mechanical saw. Two small matched specimen between the bottom of the slots, measured in
strips are thus obtained which may be joined by tack inches or twice the dimension X in Fig. 13.29. The
welding. threshold degree of restraint for cracking was expressed
as that width (measured to the nearest half inch) which
3) Lehigh restraint test was just sufficient to cause cracking. The groove
Doan and others at Lehigh University developed a geometry was held constant except for necessary
butt-weld cracking test for comparing quantitatively the variations in its length. For 1 inch and 1.5 inch plates,
degree of restraint at which cracking occurs in welds the groove length was 5 inches, but for 0.5 inch and 0.75
during cooling. The specimen was designed as shown inch plates, the length was reduced to 3 inches, in order
in Fig. 13.29, with slots cut in the sides and ends. By to provide sufficient restraint to cause cracking in thin
changing the length of the slots the degree of plate steel plate.
restraint is varied from one specimen to another. This Macro examination of weld cross-sections provides
restraint was expressed numerically by the width of the the most sensitive method of crack detection. However,
the detection may be aided by measurement of the

transverse contraction at the mid-length of the weld.
>.. \2n-- ■ 1 With the latter procedure it is necessary only to-section
and examine specimens exhibiting slightly, more
transverse contraction than that known to be associated
T 1/2" hole
with cracking. In general, only weld-metal cracks were
X observed with this test. However, in steels particularly
t s <fc =tf d susceptible to base metal cracking, such cracks were
PLU-x"-■ti observed upon cross-sectioning. These cracks were not

CO
3 1/2" for plat
^-ry
5 1/2" for plate ^ 1 "
present when low-hydrogen electrodes were used.
The Lehigh restraint test was found to be particularly
useful for quantitatively rating the crack susceptibility
of weld-metal as affected by the electrode Not only did
different electrode types produce welds with very
K-M ?no w,' Groove for welding
definite differences in their cracking tendency, but
distinct differences in cracking tendency were also
displayed between welds made with various brands of
J 11/4" radius
f the same electrode class. The effects of steel composition,
\ Plate ' '
centre line ^.^ heat input, and preheat on cracking tendency also were
1/4^"1P1/I6,,9ap demonstrated.
4) Tekken Y-groove test
Omit in plate<3/4"'thick This test, developed by Japanese research workers,
is a variation of the Lehigh test employing a Y-groove
Fig. 13.29: Lehigh restraint specimen with a 2mm root opening as shown in Fig. 13.30. The
specimen can be easily prepared by simple torch
Note: 2 ^-" lengths of

the groove on two ends
are filled up with weld
runs before depositing
test welds in the middle zone.
Section A-A' Section A-A 1 Section B-B'

Before welding 1 iTest weld in groove\ I Restraint weld ^
oa
Groove/
.angle/
0.D79" (2 mm)
Root opening
Fig. 13.30: Tekken Y-groove restraint weld cracking specimen

bevelling or machining, and smaller plate sizes can be

employed.
The researchers have reported that while Lehigh test
is better suited for testing the cracking tendency of
electrodes, the Tekken test is more sensitive in detecting
base metal cracking susceptibility, and could therefore
be looked upon as a weldability test. In both the Tekken
and the Lehigh specimens, virtually all cracking is found
to initiate from the root of the weld.
Transition-Temperature Tests
The fracture-mechanics concept helps to prevent the
initiation of fracture under elastic loading conditions,
but is not adequate to take care of the accidental
situation, in which a structure may be subjected to severe
plastic strain in service. Plastic flow can also occur
locally in the vicinity of stress raisers. To prevent
catastrophic failure of a structure which can result from
appreciable plastic flow, the transition temperature
concept is applied. According to it, failure will not occur
if any of the following three basic conditions is met by
the structure:
1) Sharp notches are absent
2) The brittle-ductile transition temperature of the
materials of the structure is below the service tem-
perature -30-20 -10 010 20 30 »0 50 60 70
3) Residual stress or stress caused by a load applied Testing temperature °C
to the structure is not of sufficient magnitude.
Fig. 13.31: Transition characteristics of Charpy-V
Since conditions (1) and (3) cannot always be met, specimens
special attention has to be given to the brittle-ductile
transition temperature of the base metal, heat-affected
zone and the weld-metal. The most popular types of It must be noted that Charpy results are affected by
tests used for transition-temperature determination are: the orientation of the specimen with respect to the weld
1) Charpy V-notch impact test or the rolling direction of the base plate. Figure 13.32
2) Notched-bend test shows the various orientations of Charpy specimens in
3) Drop-weight test. rolled plate or weld-metal. For weld-metal tests, No. 3
orientation is the most commonly used.
1) Charpy test 2) Notched-bend test

Charpy test method using V-notch specimens has Various notched-bend specimens have been used,
been described earlier. A series of tests are made over a but the most popular ones are the longitudinal weld
range of temperatures (60 to -40°C for common bead notched specimens developed by Lehigh and
structural steels) and values are obtained for absorbed Kinzel, which are shown in Fig. 13.33. It is to be noted
energy (J), % lateral contraction or shear % in the that Lehigh has used two relatively blunt notches, while
fracture. Higher shear % indicates ductile fracture and Kinzel has used a single sharp notch. The weld bead
low shear % represents brittle fracture. By plotting these serves to create a heat-affected zone through which the
values against test temperatures, curves are obtained notch passes. Kinzel tested his specimens using three
as shown in Fig. 13.31. point loading and constant span for the entire thickness
Transition temperature is the temperature at which range. Lehigh used support spans which varied with
there is a sudden large drop in the test value. plate thickness.
Fig. 13.32: Various orientations of Charpy specimens in rolled plate or weldment
T, plate thickness, 1/2" to 3/0"

D, notch depth, 0.080" Rolling direction
R, notch radius, 1 mm. -» ^
y ,-rn ,-rt^gfrr I II1 H i I O f f i 1 / / ]
^V—-►□"-—4<-—-*"-—4- --*"—^
Milled notch:
0.05" deep
OJOI" vertex radius
45* vertex angle
Plate thickness
L
L-/—-^7_±j/'' Bead weld

(B) Grind plate before
laying bead
Fig. 13.33: Notched-bend tests for determination of transition temperature: (A) Lehigh test specimen; (B) Kinzel test
specimen
1r-i -r-
[ 63 «5 mm
k m
fM^ \
oc * W
fa-31.7 mm->|
JL
<.
Fig. 13.34: Weld bead details for drop-weight test
1.5 mm max.
'2.0 + 0.00-0.25 mm
Fig. 13.35: Weld notch details for drop-weight test
Bend tests are made at various temperatures. At The weld bead which is a brittle material is called
higher temperatures no cracking may be observed. As the crack-starter weld. A notch is made at the centre of
the test temperature falls, partial cracking is observed. the bead length, and is placed at right angles to the weld
When the ductility transition temperature is reached, bead and parallel to the specimen. The notch dimensions
the specimen fractures completely. are shown in Fig. 13.35. It is important that the notch is
In these longitudinal bend tests, it is useful to deter- present only in the weld deposit and the cutting tool
mine the point at which the fracture initiates, because (abrasive disc, hacksaw, etc.) does not touch the plate
one can then pinpoint the weldment zones which are surface.
most susceptible to initiation of cracks in service. The specimen is mounted on an anvil provided with
a stop block. The impact load is applied by dropping a
3) Drop-weight test weight. A machine design suggested by ASTM permits
This test is designed to measure fracture initiation use of various weights and various drop heights
characteristics of structural materials heavier than 18 mm. according to the strength level of the metal. Figure 13.36
The specimen dimensions vary according to plate shows the position of the specimen on the anvil and the
thickness. For 25.4 mm t, length and width are 355 and loading conditions. Specimens are tested over a range
90 mm; for 19 mm t, the corresponding dimensions are of temperatures.
127 and 51 mm. A weld bead, 63.5 mm long and 12.5 The test is considered to be conclusive when the
mm wide, is deposited on the tension surface of the weight of the load and the height are just enough to
specimen using a special hardfacing electrode. The initiate a small cleavage crack in the crack-starter weld,
location of weld bead is shown in Fig. 13.34. It is fixed by after very slight bending of the test specimen. The test
punch marks at A, Band C. Actually two beads are laid, then evaluates the ability of the weld to withstand yield-
one from A to C and the other from B to C as indicated in point loading in the presence of the small crack. The
the figure, so that the end craters meet at pointC. weld either accepts initiation of fracture readily under
Weld sequence
1
Underbead
crack 2,
Bend-stop
E
E
s
Fracture must be developed
with 2° bend after sharp
notch is formed r-TH
-300 mm-
Fig. 13.36: Loading condition of drop-weight test
specimen
Fig. 13.37: Cruciform test specimen
these test conditions and the specimen is broken, or

initiation of fracture is resisted and the specimen bends 1) Cruciform test
the small additional amount permitted by the anvil stop The cruciform specimen is a double-tee fillet welded
without complete fracture. The highest temperature at assembly as s h o w n in Fig. 13.37, which offers
which the propagating crack reaches the edge of the progressively higher degrees of restraint on the four fillet
specimen is a critical temperature, and is called nil welds m a d e in the unit. The test was developed
ductility t e m p e r a t u r e (NDT). It represents the principally for evaluating the susceptibility of a steel to
temperature below which the steel loses all ability to u n d e r b e a d cracking in the heat-affected zone of
deform plastically. hardenable base metals, like the quenched and tempered
alloy steels. Much testing of armour plate has been
Cold Cracking Tests conducted with the cruciform test, and the experience
These are also referred to as hydrogen-induced has indicated that all conditions in the welding and
cracking tests and weldability tests. The most commonly testing procedure must be closely controlled to obtain
used tests are: reproducible results concerning the cracking
1) Cruciform test susceptibility of the base metal used.
2) Controlled thermal severity (CTS) test The cruciform specimen is welded with the assembly
3) Lehigh restraint test at a specified temperature prior to the deposition of each
4) Implant test. fillet bead. The beads are deposited in the sequence
The first three are self-restraint bearing tests, which shown, using carefully regulated welding conditions to
are aimed at characterising the welded joint rather than insure that the increasing degree of restraint on each
the base material itself. They are more suitable for successive fillet bead will be the only variable. After
arriving at appropriate welding procedure than to assess completion of the fourth weld, the specimen is held at
quantitatively the susceptibility of the base metal to cold room temperature for 48 hours and then stress-relief
cracking, owing to the considerable influence of the type heat-treated at 620°C for two hours (heating and cooling
of electrode, way of preheating, specific heat input, etc., from 620°C at the rate of 55°C per hour). The specimen
on the test results. is inspected for external signs of cracking and then is
Implant test, on the other hand, is a strain-controlled sliced into cross-sections for metallographic
test which provides a quantitative evaluation of the examination to d e t e r m i n e w h e t h e r cracking has
influence of restraint on cold cracking sensitivity of occurred in the weld-metal or the base metal heat-
weldments. affected zones.
2) Controlled Thermal Severity Test (Cts ) Test An attractive feature of the implant test is that it
The CTS assembly consists of two plates of steel needs only a small quantity of the steel to be tested, just
under test, bolted and anchor welded together by two enough to machine out a specimen of 8 mm diameter
lap welds on opposite sides as shown in Fig. 13.38. R and with a maximum length of say 150 mm. To simulate
and S are test welds which are laid on the remaining the cooling rates of thick plates, one inserts the specimen
two sides as shown. These test welds are cross-sectioned. into a backing plate of any ordinary available steel,
The weld and the HAZ are subjected to a hardness whose thickness can be varied as desired. Thus one can
survey. They are also subjected to magnetic crack dispense with large plates of various thicknesses
detection. In special cases, the cross-section is etched necessary to make cold-cracking studies of a steel by
and examined for microcracks in the HAZ at 200 classical methods.
magnification. The 8 mm diameter x 150 mm cylindrical specimen
The test welds are given a thermal severity number is called implant. It is made from the steel under
(TSN) which varies between 2 and 24 and increases with investigation, and is notched at some distance from one
plate thickness. The second test weld 5 has higher TSN of its ends. It is inserted into a hole drilled in the backing
than the first test weld R. plate so that the notched end is flush with the plate
Using different plate thicknesses, one can determine surface as shown in Fig. 13.39.
for a given welding procedure for which TSN of the In order to execute the test, a weld bead is deposited
weld cracking is present and for which TSN cracking is on the plate in one pass under carefully controlled
absent. Depending on this, one can arrive at the conditions, in line with the axis of the implant, using
weldability index letter. One can finally establish the the welding process and the consumable to be tested.
minimum preheat temperature necessary for given plate The welding conditions are noted and the
thicknesses and heat input rate to guarantee freedom penetration must be such that the notch is located in
from HAZ cracking. the heat-affected zone. After welding and before
complete cooling, the implant is put under a tensile static
3) Lehigh restraint test load as shown in Fig. 13.40 and the stress thus applied
This test has been described earlier under Hot is measured in relation to the cross-section at the notch.
Cracking Tests.
4) Implant test r^Circular

This test, developed by Mr. H. Granjon of Institutde notch
-Backing plate
Soudure, Paris, has attracted considerable world-wide
interest.
Implant —
specimen
Fig. 13.39: Implant test assembly
Test weld
Fig. 13.38: CTS test assembly Fie 13.40: Tensile loading of implant specimen
After a specified load holding time, in the event of no

fracture having occurred, the assembly backing plate/ T1 - - ,
implant is released and any cracks which may be present

at the level of the notch in the heat-affected zone are 1 100 33.3 333 33.3 100
then investigated.
When the test is used as a complementary
information test, it must be carried out using the oo If /-k /k /K A\
tj /
1
*y \
necessary number of levels of the following variables: CM
• Hydrogen content of the deposited metal (or weld-

metal) Axes of the weld seams
• Cooling time
• Stress applied. □
The information concerning susceptibility to cold U
All dimensions in mm
: — -300 A
cracking is provided by all the results obtained by tl
combination of these different levels. Fig. 13.42: - imensions of backing plate and location of
The implant geometry is shown in Fig. 13.41. four implants
Diameter of the specimen is 6 to 8 mm, preferably 8
mm. A V-notch is made perpendicular to the section, Hydrogen content related to the weight of the deposited
having an angle of 40-45°, 0.5 mm depth and 0.1 mm weld-metal in ml/100 g must be known. This is
root radius. The distance between notch and top end of determined by the method described on page in
implant is set in relation to the depth of fusion Chapter 3. IIW has categorised welding consumables
corresponding to the experimental welding condition. under four classes (very-low, low, medium and high)
The specimen must be sliding fit in relation to the hole as shown in Table 13.2.
in the backing plate (clearance 0.05-0.15 mm).
The implant specimen is machined from the plate Following measurements have tc be made for
lengthwise in relation to rolling. Minimum thickness of defining the welding procedure:
plate from which the sample can be taken is 10 mm. 1) Measurement of welding arc energy by recording
Three implants have to be tested for a given condition the current, arc voltage and speed of arc travel. Arc
(i.e. stress, cooling time and hydrogen). energy measured in kilojoules per mm is given by the
Thickness of the backing plate is 10 to 30 mm formula:
depending on the cooling time to be achieved. Its typical
dimensions are 300 x 200 mm. Four implants can be E = V x A x 60
accommodated in one plate as shown in Fig. 13.42. S x 100
The welding consumable to be tested can be a where
manual electrode, flux-wire combination for V = arc voltage
submerged-arc welding or gas-shielded electrode. A = welding current
S = arc travel speed (mm/min)
2) Measurement of cooling time by thermocouple in
\s
the HAZ, during the test or by frequent calibrations.
*► Radius of notch Minimum bead length is 100 to 150 mm. It is laid
= 0.10 mm
Ms 40°
parallel to the plate width. The initial plate temperature
is approximately 20°C. No preheat is used.
■7 t-
-% V Load is applied when the temperature is between
150 and 100°C. If progressive loading is done, the
duration must be less than one minute. Loading must
be completed before the temperature reaches 100°C The
load holding time is 16 hours minimum.
If fracture occurs, time to fracture must be measured.
The position of fracture must be checked. The specimen
must be inspected for cracks by metallographic
Fig. 13.41: Implant geometry inspection on longitudinal section. Several examinations
Table 13.2: Recommended testing variables
Cooling time in seconds 800-500°C

Hydrogen ml/100 g of weld-metal
T2 T3
T<8 8<T<12 T>12
l
A Ys Ys % Ys Ys V2Ys Ys
H1 very low <5

H, low 5 < H < 10
H, medium 10 < H < 15
H4 High H > 15
6 10 20
Recommended cooling times
maybe necessary on successive polishing planes starting (for example, half of yield stress), corresponds to an
from outside. average restraint.
To obtain information about the susceptibility to cold It is not necessary to carry out experiments for all
cracking d u r i n g w e l d i n g of a given steel, it is the proposed levels of hydrogen. Their choice depends
recommended that the levels of the testing variables on the welding processes envisaged and also on the steel
(cooling time, hydrogen content of the deposited metal behaviour if one begins to experiment at the level H4.
and stress) be chosen from among those which are The results are presented by completing the Table
indicated in Table 13.2. For the cooling times, the values 13.3 entering in each space relevant to the investigation,
6, 10 and 20 sec (between 800 and 500°C) are given, the c o r r e s p o n d i n g result (F or NF, cracked or
because they can be predetermined during a preliminary uncracked). A table will be filled in for each of the two
setting. The levels of hydrogen content which cannot stress levels experimented with. Table 13.3 illustrates,
be set in advance, must be checked at the time of the as an example, a completed table. In addition, the
test, in the actual state of the consumables. The stress welding process used and welding conditions obtaining
levels to be applied are two in number. One of them in testing the hydrogen levels and the cooling times
corresponding to the yield stress of the metal under should be given in the report.
investigation, accounts for the cracking behaviour for Implant test is being recognised by several experts
the highest possible restraint. The other, which is lower in the IIW as the most appropriate cold-cracking test.
Table 13.3: Examples of presentation of the results
NF NF NF
H2 NF NF NF
H3 F NF NF
H. F NF
NF NF NF
H2 F NF NF
F 2P NF
H, F *l*fr NF
î
The research laboratories of the British Welding Institute specified. It is an easy and inexpensive method of
have done considerable experimental work on behalf practical quality control, which can prevent occurrence
of the IIW to standardise this test method, and to of defects or detect their presence at any early stage
compare its reliability with that of controlled thermal while the welds are being deposited.
severity test. The results are published in the IIW The only tools required for this method are a
document ll-A-307-72. magnifying glass, a borescope, scales and gauges.
The introduction to this document states: "The There are three stages of visual inspection: (1) before
determination of the effects of welding and plate welding, (2) during welding, and (3) after welding.
variables on the risk of heat-affected zone (HAZ)
cracking has generally been assessed in the past using 1) Before welding
weld cracking tests such as the controlled thermal Before the parts are assembled, their surfaces should
severity (CTS) test. The principal disadvantage of such be checked for the presence of oil, grease, scale, rust and
welding tests is that the applied stress on the HAZ is a other foreign matters. Cut edges of plates must be
"variable dependent on weld penetration and root notch carefully examined for laminations. If bevelled cuts have
geometry, plate thickness and joint restraint. Mechanical been made, the angle of bevel and tlje root face must be
tests like the constant load rupture (CLR) and fracture checked. Dimensions of the parts must be checked by
mechanics test although they have control over the level m e a s u r e m e n t . After the joints are assembled for
of stress which is applied, suffer from the disadvantage w e l d i n g , the root g a p , groove angle, clearance
of employing simulated HAZ microstructures and dimensions of backing strips or rings, joint alignment
arbitrary hydrogen concentrations. It is believed that and other features which might have a bearing on the
the implant cracking test, which applies a known stress quality of the w e l d e d joint m u s t be verified for
to a real HAZ and can thus assess the susceptibility to com0ormity to the applicable specification or procedure.
HAZ hydrogen cracking by means of a mechanical test
parameter, avoids the disadvantages of the above tests 2) During welding
and, at the same time, incorporates their merits." The details to be checked while the welding is in
progress are:
NON-DESTRUCTIVE TESTING (NDT) a) Welding process parameters
b) Welding consumables
In the case of critical welded structures, destructive c) Preheat and interpass temperature
tests which verify the properties of external test d) Placement of weld runs (stringer or weaved
specimens are not adequate. To obtain increased bead), their surface quality
assurance of the high quality and serviceability of the e) Brushing, chipping, grinding or gouging
actual structure, it becomes necessary to subject all or a f) Welding sequence
certain percentage of its welded joints to non-destructive g) Distortion control
testing (NDT). h) Dimensions of finished welded joint
The various commonly used NDT methods are i) Postheating temperature and time.
described in the following paragraphs. These methods It should be ensured that all the above are in
are complementary and it is essential that the welding conformity with the relevant standard or qualified
supervisor or inspector is thoroughly familiar with the welding procedure specification.
basic physical principles of each method, so that he can It is often useful to prepare a workmanship standard
judiciously select one or more of these tests, bearing in in advance, in which portions of successive weld layers
mind the costs and the level of inspection demanded. are shown. Welders are then expected to deposit the
This understanding will also help him to use the full passes in an identical manner. Typical standards for butt
potential of the techniques, to make sure that the and fillet welds are shown in Fig. 13.43. With the
operators are satisfactorily performing the tests and to assistance of such standards, one can compare each layer
obtain correct interpretation of test results. of production weld with corresponding layers of the
workmanship standard.
Visual Inspection It must be noted that the root pass or the first layer
This is the first step of NDT. In fact, visual inspection is particularly susceptible to porosity, slag inclusion and
should be used for all types of structures, critical and hairline cracking. If the cracks are ignored, they may
non-critical, and even w h e n 100% radiography is extend to subsequent layers. It is very important,
the m e t h o d e s t a b l i s h e d a n d u n d e r s t o o d by all
Macro sample
L.
Polish & etch
this surface inspecting personnel and shop personnel concerned
with the repair. A distinctive colour must be used for
1" min the marking, which will not be confused with other
m a r k i n g s , a n d the m a r k i n g s h o u l d be able to
withstand rough handling involved in the repair work.
Tack macro sample It should last until the repair has been completed and
t o plate with inspected.
etched surface up
P o l i s h and etch this surface Magnetic Particle Inspection

Duplicate fillet ^ This inspection m e t h o d is applicable only to
weld on this
side also Macro sample magnetic m a t e r i a l s , a n d can detect surface
discontinuities even when they are too fine to be seen
min with the naked eye, defects that lie slightly below the
surface, and even discontinuities at greater depth when
special equipment is used. The defects it can detect are
surface cracks in the weld and the HAZ; laminations
Tacfc macro
sample upriqht on the prepared joint edges; incomplete fusion and
mm
*J to plate with u n d e r c u t ; subsurface cracks, porosity a n d slag
etched surface out
inclusions; and inadequate joint penetration.
Fig. 13.43: Typical workmanship standards for butt The method is based on the principle that when a
welds (top) and fillet welds (bottom) magnetic field is set up in a piece of ferromagnetic
material having defects in the path of the magnetic flux,
therefore, that this pass is thoroughly examined, minute magnetic poles are created at these defects. These
preferably with a magnifying glass, before the next pass poles have the property of attracting magnetic particles
is d e p o s i t e d . This e x a m i n a t i o n offers a n o t h e r more strongly than the surrounding surface of the piece.
opportunity to look for plate laminations, since these Thus if the piece to be inspected is magnetised by passing
tend to open up due to the welding heat. When welding a sufficiently high current through it and the areas to be
has to be done from the second side, it is necessary that inspected are covered by finely divided magnetic
the back of the root is gouged out to remove entrapped particles, these particles form a pattern or indication on
slag and to provide a sufficiently wide groove to deposit the surface, and magnetisation takes the rough shape of
the pass from the second side. This back-gouged area the defect, which becomes clearly visible when the
also needs to be inspected. surplus magnetic powder is removed. This occurs only
when the defect is perpendicular to the flux path. If the
3) After welding defect lies parallel to the flux path, no such pat-tern will
After the structure has been welded, its dimensional appear and the defect will lie undetected. The best
accuracy including possible warpage is checked with indications are obtained when the current is flowing
the drawing, using conventional measuring methods. parallel to the discontinuity, because the magnetic field
The sizes and contours of all welds are also verified by is always at right angles to the current flow.
means of weld gauges to make sure that they conform Figure 13.44 shows an example of longitudinal
to the d r a w i n g r e q u i r e m e n t s . Joint surfaces are magnetisation produced in a part to be inspected, which
examined for weld roughness, weld spatter and the could be a shaft, drum or girder. The part has been made
presence of unfilled craters, pock marks, undercuts, the core of an electromagnet by placing it within a
overlaps and cracks. Heights of weld reinforcements are solenoid. It is seen that the field runs in a direction
also measured to make sure that they are consistent with parallel to the axis of the coil. As indicated in the figure,
specified requirements. longitudinal cracks will escape detection, those at 45°
angle to the lines of force will be weakly indicated, while
Marking Repair Areas transverse cracks will be indicated distinctly.
An important detail of NDT is the proper marking Localised circular magnetisation is obtained with the
of the defective areas to be repaired by welding. The commonly used technique called prod magnetisation.
marking should be distinct and in accordance with With this technique, large parts which cannot be
magnetised as a whole are magnetised locally by pass- V) and to provide a switch in the field circuit of the
ing current through small areas by means of contacts or generator or the primary circuit of the step-down
prods. A local circular magnetic field which is at right transformer. Usually, the inspection is made in two
angles to the direction of current flow is formed as stages: firstly, with the prods in line with the weld and
shown in Fig. 13.45. Hence defects which lie parallel to secondly, with the prods placed at right angles to the
the current flow will be detected as shown in Fig. 13.46. previous line.
To prevent arcing or burning during the passage of Overall circular magnetisation is also possible in the
high current, the contact prods and the metal surface case of certain components by passing high current
must be sufficiently clean. To prevent the arc flash, it is through them. An example is shown in Fig. 13.47. In
advisable to use a low open-circuit voltage (below 16 this case only lengthwise defects can be detected.
Transverse crack
will show f
Current passing ^
through" solenoid
Solenoid
Crack at 45'
will show
Longitudinal Part
crack will '// / *— Maqnetic lines
not show y f of force *
Magnetic field
Fig. 13.45: Localised circular magnetisation of a large

Fig. 13.44: Magnetisation of the part to be inspected by
part with prods carrying high current
making it the core of a solenoid
Current —*
Fig. 13.46: Detection of crack in large plate by passing high current parallel to the defect between prods
Contact plate Contact plate
—i—
Iron powder
Current
Fig. 13.47: Detection of lengthwise crack by circular magnetisation in a part set up in an inspection unit
This technique can be applied on cylindrical objects method and to locate defects in corners, holes, keyways,
by passing the magnetising current through a conductor etc. The fluorescent magnetic particle testing should not
running through the central opening, whereby the inner be confused with fluorescent penetrant inspection which
surface as well as the outer surface of the cylinder gets will be described later.
magnetised. When the current is passed through the part Type and amount of magnetising current is an
itself, the surface of the bore does not get magnetised, important parameter. With AC, only the surface is
but only the outer surface. magnetised and hence it is useful only for locating
Overall circular magnetisation produces high field surface defects. With DC, the field extends throughout
strength and thus provides maximum sensitivity to the part. Hence this type of current is more sensitive
subsurface defects. than AC for the detection of subsurface defects.
There are two methods of operation, viz. continuous Maximum sensitivity can be obtained with half-wave
and residual. In the continuous method, the magnetising rectified single-phase current.
current is first applied to the component and then the Voltage has no effect on the magnetic field and it is
dry powder spread over the surface. The flow of current kept low to prevent arcing and overheating. The strength
is maintained while the excess powder is being removed of the magnetic field is determined by the current, and
and the defective areas are located. The current is hence the current should be high enough to reveal the
switched off only after the inspection is completed. If a defects. If the current is too high, it produces wrong
wet suspension is used for the continuous method, the patterns. The following figures provide the guidelines
suspension is first flowed over the area and the for the amperages to be used:
magnetised current is immediately applied for a fraction Longitudinal magnetisation — 3,000 to 10,000 Amps.-
of a second. Once the current has been switched off, no turns
further suspension should be applied as it will wash Overall circular magnetisation—100 to 1,000
away the lightly adhering indications. Amps, per
In the residual method which is suitable for materials 25 mm of the outside
of high magnetic retentivity, the indicating powder is diameter of the part
applied only after the magnetising current has been Prod magnetisation — 100 to 125 Amps, per
applied and then switched off. The powder is held on 25 mm of prod spacing
to the surface by the fairly strong residual field. This Dry method — prod spacing of 75 to 300 mm
method is not as accurate and sensitive as the continuous Wet method — prod spacing of 50 to 300 mm.
method. For additional information, reference should be made
We have referred above the dry and wet methods of to the following documents:
applying the inspection media. In the dry method, ASTM E-109, Method for Dry Powder Magnetic
finely-divided ferromagnetic p o w d e r is d u s t e d Particle Inspection
uniformly over the surface by means of a dusting bag, ASTM E-138, Method for Wet Magnetic Particle
atomiser or spray gun. The powder is applied at a low Inspection.
velocity with just enough motive force to drive the
particles to the desired areas. Excess powder is removed Liquid Penetrant Inspection
with a stream of air of controlled strength, which will This method uses certain types of liquids which have
not disturb the adhering powder pattern. The dry the property of entering into fine surface discontinuities
method has easy portability and is better suited for by capillary action, and remaining there when the excess
rough surfaces. liquid is removed. A suitable developer is then applied.
The wet method uses a suspension of fine magnetic Against the background of this developer, the retained
particles in water or oil, which is flowed or sprayed over penetrant stands out by contrast and reveals the defect.
the surface. Sometimes the part is immersed in the There are two distinct m e t h o d s of p e n e t r a n t
suspension. The bath needs continuous agitation to inspection, viz. fluorescent and visible dye. The first
prevent the suspended particles from settling down. method involves use of a liquid that fluoresces under
Compared to the dry method, this method is much black light, which is near ultraviolet light and just
more sensitive to fine surface defects, but less sensitive beyond the range of the visible spectrum. Fluorescent
for detecting subsurface defects. Sometimes the particles penetrants are either basically water-washable or require
are coated with the dye that fluoresces brightly under the application of an emulsifier to render them water-
ultraviolet light to further increase the sensitivity of the washable. Visible dye technique uses a penetrant which
carries a red dye that contrasts sharply with the transmitted are determined by the thickness of the
background of a white developer. Visible dye penetrants material. Wherever defects exist in the material, there
also are either directly water-washable or can be made is less material (or reduced thickness) to absorb the
so by the addition of an emulsifier. radiation. This creates a differential in absorption, which
The surfaces to be inspected m u s t be m a d e is registered on a photographic film as a shadow picture
completely free from slag, scale, grease, oil, paint, etc., called a radiograph.
and must be perfectly dry. The penetrant is applied by X-ray radiation is obtained from an X-ray machine.
brushing, spraying or dipping. A dwell time of about G a m m a - r a y radiation is obtained from chemical
15 minutes is allowed for the penetrant to enter isotopes. Radium, which is a natural emitter of gamma-
discontinuities. In the case of a directly water-washable rays, was extensively used in the early days of industrial
penetrant, the excess is rinsed off with a water spray. If radiography, but is rarely used today. Presently, artificial
the penetrant is post-emulsifiable, the emulsifier is isotopes are used. They are: cobalt 60, cesium 137,
applied over the penetrant by flowing or spraying and iridium 192 and thalium 170, mentioned in decreasing
the excess is rinsed off with water. In the case of some order of energy level. Of these, cobalt 60 and iridium
emulsifiers, a certain dwell time becomes necessary, 192 are more widely used in radiographic inspection of
which is indicated by the supplier. In the case of some weldments than the other two. Gamma-ray sources are
dye penetrants, organic solvents such as trichlorethylene light and portable and can be used more conveniently
or acetone may be used to remove the excess penetrant. at site than X-ray machines.
Developers are either in the form of dry powder, The capacity of an X-ray machine is indicated by the
suspension in water or non-aqueous wet developer. Dry peak voltage it generates to drive electrons into the tube
developer is applied by dusting on the completely dry target, which is expressed in peak kilovolts (kVP) or in
part. Suspension-type wet developers are applied by megavolts (MV). A kilo volt is one thousand volts; a
brushing or spraying. When a n o n - a q u e o u s wet megavolt is a million volts. The capacity range of
developer is being used, the surface must be perfectly industrial X-ray machines is 5 kV to 20 MV.
dry as when the dry powder is used. In the case of the The capacity or energy level of an isotope is
water suspension developer, the surface need not be dry, expressed in million electron volts (MeV). Cesium 137
but the drying becomes necessary after the application has only one energy level, b u t others may have
of the developer. numerous energy levels. In the case of X-ray machine,
After the developer has been applied, sufficient time any energy level (voltage) within its rating can be
must be allowed to draw or blot out the penetrant from selected according to the material thickness. With a
the discontinuity. In the case of the fluorescent dye, the given isotope source, a fixed energy level is obtained
part is finally examined under black light in a darkened and it cannot be varied.
room. The defects glow brilliantly against the darker It is also customary to express the strength of gamma-
background. In the case of the visible dye, the defects ray sources in curies. The curie is actually a measure of
show up as bright red markings against the white the disintegration rate of a source. It must be noted that
background of the developer. all isotopes are continuously disintegrating, which
After the completion of inspection, the part or area means that they decrease in strength or decay with time.
is cleaned with a suitable solvent. This is the reason why various sources of a given isotope
This inspection method is suitable for locating have different strengths, which are compared in terms
discontinuities which are open to the surface. It cannot of the number of curies. The term half-life applied to an
detect subsurface defects. But within its scope it is as isotope is important. It is defined as the length of time
effective as magnetic particle inspection. it takes for an isotope to decay to half its strength from
For further details, one should refer to ASTM Spec. the start of the half-life period.
E 165, 'Liquid Penetrant Inspection. The significance of half-life can be explained by
taking the example of cobalt 60, which has a half-life of
Radiographic Inspection 5.3 years. If the original strength of a cobalt source is 80
In this method. X-ray or gamma-ray radiation of curies, in 5.3 years it will be reduced to 40 curies. In the
sufficient penetrating capability is passed through the following 5.3 years, the strength will come down from
full thickness of the material being tested. Part of the 40 to 20 curies. Iridium 192 has a half-life of only 75
radiation is absorbed, and the balance penetrates the days, which means that its strength falls off rapidly and
material. The a m o u n t absorbed and the a m o u n t hence the usefulness of an iridium source lasts for a very
short period. This is also the reason why iridium has a important for the exact location of the position of any
large number of energy levels. Cesium 137, on the other repairs to be made as a result of the radiographic
hand, has a half-life of 33 years, which explains why examination.
this isotope is considered to have one energy level. Positioning of the film cassette is an important step.
The higher the energy level of the radiation source, The Film must be placed in contact with the weld, on
the shorter is the wavelength and the greater the the side opposite to that facing the source of radiation,
penetrating power. The source energy-level is therefore and it must be large enough to cover the whole of the
selected according to the thickness of the material and weld zone within the beam radiation.
its absorption capability. To locate the films in their correct positions, the face
In taking radiographs, one must aim at good contrast of the specimen on which they will be placed must be
and sharpness of image. Good contrast m e a n s a marked with the distances corresponding to these
sufficiently dark appearance of defects against the positions, which have been determined as already
lighter general background. Sharpness ensures that fine indicated. In certain cases this can be difficult, notably
details, such as the image of a hairline crack, are readily when, because of the shape or size of the specimen, it
detected and the radiograph is properly interpreted. A cannot be done by direct observation, but only by
well-contrasted and sharply defined radiograph is measurements from some well-identified datum line,
obtained by proper selection of the energy level of the as for example the edges of the specimen, the crossover
radiation and the type of film. of two welds, or other details visible on both sides.
Radiographic films of various speeds are available. Care must also be taken to centre the film on the line
The faster the film, lesser is the exposure time needed. of the weld, and in this respect it is preferable to use a
Fast films are coarse-grained and they may fail to record film that is too large rather than one that is too small;
sharp details if they are too fast. the slightly increased cost of film is always negligible
The film is sandwiched between a pair of screens, compared with the cost of repeating the radiograph
which are either fluorescent type or lead-foil type. The because of an error in centring; moreover it is sometimes
fluorescent screens glow differentially when the X-rays advisable not to limit the examination strictly to the
reach them, and this emitted light causes the exposure weld-metal itself, but to include some of the plate metal
of the film. Lead-foil screens act as intensifiers by on either side of the weld.
emitting electrons and special rays, which help to The films, with or without intensifying screens, are
produce the radiograph, and they also act as filters by always placed in cassettes which besides ensuring
absorbing scattered radiation, which otherwise would uniform contact between film and screens, serve to
fog the film. Fluorescent screens are preferred to lead protect the film from light.
screens when very thick materials are being examined The cassette containing the film must be placed as
with heavy X-ray equipment. However, most of the close as possible to the weld, so as to avoid any loss of
radiography today is done with lead-foil screens. Screens definition in the image. For this reason, in many cases,
need not be used for the radiography of thin sections. flexible cassettes, which can be better adapted to
In carrying out radiography, it is necessary to mark irregularities, curves and variations in shape are to be
the welds and identify the locations of the radiographs. preferred to stiff cassettes.
First, one must determine the maximum length of weld, However, certain precautions must be taken in using
which can be covered by one radiograph. If a long weld flexible cassettes (they must not be pressed too heavily
has to be completely examined, the total length is against the specimen), in order not to cause rapid
divided into equal parts, each of which corresponds to deterioration of the screens or the formation of pressure
the maximum length of each radiograph, less about 25 marks on the film.
m m at each e n d , to e n s u r e an overlap b e t w e e n A sheet of 2 mm thick lead is always placed behind
successive films. the cassette to protect the film from scattered radiation.
To correspond with the sections so determined, a Because of the secondary radiation emitted by the lead
series of numbers is usually punched on the parent plate itself, it may be advisable to place a sheet of tin about 1
close to the reinforcement, generally on the side facing mm thick between the lead sheet and the cassette; even
the source of radiation. Lead numbers are placed on the better are sheets of 1 mm tin and 1 m m copper.
punched numbers and their image on the film gives the The cassettes can be held in position by various
accurate location of the area w h i c h h a s been methods, depending on the shape of the specimen and
radiographed. The accuracy of this operation is very the material available (wooden supports, wood blocks,
etc.). Permanent magnets equipped with springs are with the beam of radiation perpendicular to the surface.
very useful for this purpose. It is very important that It is thus necessary to arrange the equipment in such a
the cassette does not shift in relation to the specimen manner that it fulfils this condition, taking particular
during the exposure, as otherwise there will be a loss of care that the axis of the beam of radiation is directed at
definition which will make the radiograph unaccep- the centre of the area to be examined. The correct set-
table. Special care is necessary when using gamma-rays, up may be attained by moving either the equipment or
which involve long exposure times. the specimen.
The gamma-ray or X-ray source m u s t also be Figure 13.48 shows the arrangements frequently
correctly positioned. In general, butt welds are examined used in the inspection of pipes and cylindrical shells. It
'55
! —1|/r i
c
6^ 200 mm
8
u
c *
in
<X
*F
t 0< 200 mm
2
S ^ ) Screen with
lead plates
Hollow
anode
f
<75
c
z
</>
v
u
u
I
H3
Longitudinal welds Circumferential welds
Fig. 13.48: Typical positioning for radiography of pipes and cylindrical bodies
would not be difficult to transpose these ideas to cases d) The French AFNOR type: It also consists of a series
encountered in practice. of steps of increasing thickness, each having a drilled
Radiography reveals all internal discontinuities even hole or holes of diameter equal to the thickness of the
when they are deep inside the material, but a radiograph step.
by itself does not indicate their depth location. To The image quality of the radiograph is indicated
determine the depth, the following simple method is when the radiograph shows up holes or slots of some
used: The defect is first located on a single exposure definite size of the penetrameter used. The inspector
film. The area is then double exposed after shifting the should know the penetrameter requirements applicable
source over a short distance. A lead marker, say letter S, to the particular fabrication he is inspecting.
is stuck on the surface on the source side and another For the examination of radiographs, a darkened
marker, say letter F, is stuck on the film side. If in reading room and a special viewing equipment are used. The
the double exposure film, the letter S has shifted a latter has an illuminator that provides a cold light from
distanced, da and the defect image has shifted a shorter a fluorescent lamp. The extraneous area beyond the film
distance of d2 then the depth of the discontinuity from or part of the film is masked off during examination.
the film side is given by (d 1 /d 2 ) t, where t is the section Identification of weld defects from their radiographic
thickness, provided the cassette is in contact with the images requires thorough training, long experience and
film-side surface and the letter F will not have shifted a good k n o w l e d g e of the o p e r a t i v e techniques.
at all in the double exposure. Sometimes it becomes necessary to refer to standard
Penetrameters: In taking a radiograph, it is imperative reference r a d i o g r a p h s such as the 'Collection of
that a standard penetrameter (also called image quality Reference R a d i o g r a p h s of Steel' compiled by
indicator) is placed on the part being radiographed. Of Commission V of the International Institute of Welding
the m a n y types of p e n e t r a m e t e r s available, the (IIW). For radiographic (as well as ultrasonic) inspection
following are the most common: of important fabrications and projects, the services of a
a) The German wire type: It consists of a series of seven qualified welding inspector are usually utilised.
wires of various diameters, which is standardised by a The following defects are commonly revealed in
DIN standard. The ranges of wire diameter depend on r a d i o g r a p h y : cracks, porosity, slag inclusions,
the thickness of weld being examined. incomplete fusion, lack of penetration due to inadequate
b) The API-ASME single-thickness type: It is specified joint preparation, undercut of a special type that occurs
for pressure vessel radiography. It is in the form of a in welds m a d e with backing rings, and tungsten
thin strip of uniform thickness equal to 2% of the inclusions.
thickness of the plate to be examined. Three holes are IIW has divided the radiographs into five groups,
drilled in the strip with diameters equal to two, three graded in accordance with the relative importance of
and four times the thickness of the strip, the smallest the different types of defects. The five groups, signified
hole being not less than 1.6 mm in diameter. by colours, are described in Table 13.4. The IIW has also
c) The API-ASME stepped-type: It consists of graded issued the Collections of Reference Radiographs divided
steps of metal, each 12.7 mm long, and having graded into the five groups.
thicknesses of 0.13,0.25,0.51,0.76 and 1.0 mm for testing The defects revealed in radiographs have been
steel up to 50 mm thick, and graded thicknesses from standardised, classified and explained in ISO 6520-1982,
1.0 to 3.0 mm for steel in the 50-100 mm thickness range. Classification of Imperfections in Metallic Fusion Welds,
Each step has a hole of 4.76 mm diameter in the centre. with Explanations", and the Indian equivalent standard
Table 13.4: Colour coding for radiographs

Black Homogeneous weld or a weld with a few small scattered gas cavities.
Blue Very slight imperfections as regards homogeneity in the form of one or more of the following defects: gas cavities,
slag inclusions, undercut.
Green Slight imperfections as regards homogeneity in the form of one or more of the following defects: gas cavities, slag
inclusions, undercut, incomplete penetration.
Brown Marked imperfections as regards homogeneity in the form of one or more of the following defects: gas cavities, slag
inclusions, undercut, incomplete penetration, lack of fusion.
Red Gross imperfections as regards homogeneity in the form of one or more of the following defects: gas cavities, slag
inclusions, undercut, incomplete penetration, lack of fusion, cracks.
18:10793-1983. The standard also gives reference check the cumulative dose absorbed by an individual.
numbers for all principal imperfections and the sub- An inspector should be familiar with all such safety rules
terms, and letter designations for imperfections used and must insist on their observance.
in the IIW Collections of Reference Radiographs.
Sometimes misleading and false indications appear Ultrasonic Inspection
on radiographic films, caused by carelessness during In this method of testing, ultra-high frequency sound
the exposure or by incorrect handling or processing of waves are projected in the form of a beam into the
the film. An inspector should learn to recognise and material being tested. The sound beam emanating from
evaluate such indications and be able to order re- the transducer (sender and receiver) travels through the
radiography when warranted. homogeneous medium in a straight line until it meets
While in the case of most critical fabrications, one an acoustic boundary. At the boundary, some of the
would insist on zero defects as revealed by radiography, sound is reflected in the same way as light reflects from
many standards permit a reasonable amount of defects. a smooth surface. The reflected sound is termed an echo.
Many important codes and specifications have laid The time and amplitude of the echoes received are
down acceptance standards for radiography. Most of analysed to provide the data.
them permit limited porosity, and slag inclusions in When two different media are bonded together and
varying degree. Typical examples are the ASME Boiler their acoustic properties are similar (for example,
and Pressure Vessel Code and AWS Specification for stainless steel cladding on carbon steel), the amount of
Welded H i g h w a y a n d Railway Bridges. Since sound reflected is relatively small. When, for example,
acceptance standards vary considerably, the inspector slag inclusions are present ina weldment, the consequent
must decide which particular specification will govern difference in acoustic properties results in reflections of
his inspection work. about 25% of the incident sound beam. When a metal-
Some specifications p e r m i t partial or spot gas interface exists, as in the case of a crack or incomplete
radiography, in which case only the critical locations, fusion, the acoustic properties are so widely different
or locations selected at random on a percentage basis that 100% reflection occurs. Such variations in reflections
are radiographed. If serious defects are encountered, the are shown in Fig. 13.49.
inspector may order 100% radiography. Spot The manner in which sound beam reflects from
radiography by itself will not confirm the total quality boundaries is shown in Fig. 13.50. In the case of normal
of the work, but w h e n a welder knows that this incidence, the reflected sound returns to the transducer
inspection procedure will be applied, he will take all over the incident path. The same happens with a right
the care to ensure that the welds will pass the inspection. angle corner. When the sound strikes at an acute angle,
All persons concerned with radiographic inspection, the sound is reflected away from the incident beam.
including those who work in the vicinity of an X-ray or Reflections from smooth surfaces are smooth and direct.
gamma-ray source, must remember that its radiation But in the case of rough surfaces, a large amount gets
has harmful effects on the human body, which in some randomly scattered.
respects are cumulative. Rigid safety rules and radiation The frequency of sound used for weld inspection is
exposure limits are set up by agencies such as Bhabha around 2.25 MHz (1 mega hertz = 1 million cycles per
Atomic Research Centre in Bombay, w h o are the second). At lower frequencies, the sound beam has
suppliers of gamma-ray isotopes. These must be fully greater penetrating power, but isolated defects such as
complied with. scattered porosity are likely to be missed. The higher
Non-portable radiation sources must be housed in frequencies, on the other hand, are capable of picking
isolated concrete rooms w i t h wall thickness up minute defects in a weldment.
commensurate with the source strength. When portable Ultrasonic testing equipment consists primarily of
isotopes or X-ray machines are used in shop or at site, electronic instruments to produce, receive and display
the area around is cordoned off to prevent anyone electronic signals, and a transducer which carries a small
straying into the critical area. Alternately, the piezoelectric element capable of converting electrical
radiography is carried out in night shifts when other signal generated in the instrument into ultrasound. The
workers are not around. The radiation dosage rate at element is contained in a solid block having acoustic
any particular location is checked with a suitable survey damping properties.
meter. Personnel concerned with radiography are A transducer may be straight-beam type or angle-
provided with pocket dosimeters and film badges to beam type. With the former, the transducer is in direct
^3
1 «-s
<-3
Good Bond
i * >
<-*
Slag inclusion
ml*\ Crack or
^H
incomplete fusion
Fig. 13.49: Variations in reflections due to differences in acoustic properties caused by defects at the boundary.
Jl_ X^_ t^L

NORMAL
INCIDENCE
ANGLE
INCIDENCE
CORNER
INCIDENCE
r
Inspection
surface
- Transducer
Lucite wedge
Fig. 13.50: Sound reflections at boundaries (similar to

light reflections)
contact with the part and provides normal incidence. It

provides simple test technique, but is not adequate for
weld inspection. In the angle-beam type, the transducer
is mounted at an angle behind a wedge, which is either Test material I-Ultrasonic beam
plastic or liquid, as shown in Fig. 13.51. At the interface
between the wedge and the test material, the sound Fig. 13.51: Angle beam head with transducer mounted
beam may undergo refraction and the longitudinal at an angle behind clear plastic
sound wave in the wedge may get converted into a shear
wave in the test material. The effective angle in the test converts the electrical signal to a mechanical vibration
material is important and this is controlled by proper of ultrasound. The sound then travels through the
selection of wedge material and its angle. Fixed-angle material, reflects from the boundaries and returns to the
beam units are generally used for production, tests. transducer. The transducer reconverts mechanical
The heart of the equipment is the clock. On receiving vibrations into an electrical signal, which then traverses
a signal from the clock, the pulser sends a momentary a receiver amplifier. Finally, a vertical deflection is
high-voltage electrical signal to the transducer. The produced on a cathode ray tube. The clock circuit also
transducer, which is in contact with the material surface, initiates a sweep circuit of the oscilloscope, which causes
a horizontal deflection of the trace on the cathode ray

tube proportional to time or distance.
In the resulting pattern, the horizontal deflection
which is related to time indicates the relative position of
the various reflecting surfaces. The amplitude of the
vertical deflection of these boundaries indicates the
intensity of the sound reflected from them.
Angular incidence of the sound wave through a
wedge gives rise to what are termed surface waves,
which are capable of travelling along the material surface
while penetrating only a short finite distance below the Fig. 13.53: Four transducers set at 45° angle to the weld
surface. While travelling, the surface wave will pass to locate suspected longitudinal and transverse defects.
around gentle curves with little loss of intensity, and will
even pass round sharp corners with some reflection and the transmitted signal is registered on the transducer R1.
some transmission. It will reflect from boundaries in its If there is a lengthwise crack or an area of lack of fusion
path, and very likely return to the original transducer. (B in figure), the signals will go to the transducer 1^. If
Multiple-transducer techniques: Multiple transducerstransverse cracks are present (C infigure),the signal will
consist of two or more separate elements, one for the be picked up by the transducer R3. If porosity is
transmission and the others for the reception of sound. encountered (D in figure), the beam will be scattered
For straight beam testing, it is preferable to use a dual and no discrete signal will be registered on any receiver.
transducer which has two small elements built into a Detect location: Using the ultrasonic test, one can
common test head and uses a short-delay line between determine the position, type and extent of defect within
the transducer and the test surface. Such a transducer a weld. The location of defects (reflectors) is based on
gives very high resolution and permits a test close to the the factors of direction and time. The initial test set-up
entrance surface, which is not possible with a single enables one to predict the beaming of the sound and its
straight-beam test. direction. In angle-beam tests, one can also predict the
One often comes across cases where the primary echo point of reflection. From the time base on the CRT
is not likely to return to the transmitting transducer. An display, one can determine the relative distance the
example is a double-V butt joint having inadequate joint sound has travelled over the known path. Special
preparation shown in Fig. 13.52. In this case, two angle- attachments and accessories are available to assist in
beam transducers in tandem, with their axes in line, have the location of defects. An example is a flaw location
been used. The transmitting and receiving elements ruler, which is attached to the transducer and helps to
have been so positioned that a reflected beam from the relate the CRT data to the actual test specimen. The ruler
perpendicular reflecting surface will produce a readily has to be calibrated for a specific angle, material
recognisable signal. thickness, and test distance.
Figure 13.53 shows four transducers being used to For the evaluation of defects, that have been detected,
look for possible longitudinal and transverse cracks in use is made of the vertical signals on the CRT. The
continuously welded seams of line-pipe in a production amplitude of such a signal is proportional to the sound
mill. All are positioned at 45° to the axis of the weld. T is energy returned to the transducer, which in turn, depends
the transmitting transducer. If the weld is good, there is on the size, type, orientation and shape of the reflecting
hardly any reflection of the sound (A infigure)and hence surface or defect. In a way, the amplitude of the signal
serves as a primary criterion for the acceptance or
rejection of a weld.
To obtain quantitative information, a reference block
which helps to define the various test parameters is used.
^ Possible inadequate In some cases, a test plate in which a simulated defect is
^ joint penetration
introduced is used. Vertically drilled holes, 1.5 mm
diameter and 12.5 mm deep, are often used as a reference
reflector. Horizontally drilled holes are also used as they
Fig. 13.52: Dual angle-beam transducers used for resemble defects such as slag inclusions. Square-sided
special cases or V notches also provide a reference standard.
The calibration block containing a combination of difficulties are large reinforcements, concavity of the
drilled holes and notches designed by the International root, surface grooves and general irregularities. In some
Institute of Welding is shown in Fig. 13.54. cases they are overcome by examination from two sides
The ultrasonic test equipment usually includes an of the weld. Changes in wall thickness adjacent to a weld
attenuator, calibrated in decibels, which provides linear may also adversely affect determination of defect
relationship between a single standard reflected signal location. Geometric effects must be eliminated in cases
amplitude and the actual defect signal. The attenuator where it is only possible to examine a weld from one
enables one to make such comparisons with both the side. Ultrasonic testing of lap fillet welds and partial
signals at equivalent screen heights, thereby allowing penetration butt welds is of little use, as the areas where
more detail of the defect signal to be obtained. most defects are likely to occur (the root runs of welds)
In ultrasonic testing, confusion is often created by are not accessible for examination.
absorption and scatter of the sound waves in the weld,
the surface roughness of the job, the presence of internal Eddy Current Testing
reflectors and various geometric effects. The surface In this method, eddy currents are induced in a part
roughness should be better than 20 jam and the surface by subjecting it to the influence of an alternating
shape should be such that at least 80% of the probe is in electromagnetic field. The frequency chosen to excite
contact. The geometric effects which commonly cause the field ranges between 500 and 20,000 Hz (cycles per
6.25" H
-1h - -080"
0"HK f-f—
cU
40° 50° 60° i
-♦*♦" f in"l
.10»
3.6"
Echo Surface
2 (Ground)
•125'
-6.6"
Cylinder of Lucite
12"
Echo surface .0S0"
^ , ^* ♦ ♦ " ♦ ♦ r-
* " T ^ _ Silvercoated surface Test surface
b—M«.yg==r-*i
^^n 70° 60°
Test surface
(Planed)
.10"
.16"; I
** 11 t m ^.,,,r f t l ,.l.,.|.,.,.,^
Material; LQMC 5.2
Fig. 13.54: Diagram of IIW calibration block for ultrasonic test equipment
second). The field is usually produced by an inductor Closed containers such as pressure vessels and boiler
placed close to the part under test. The inductor can be shells are proof tested by applying internal pressure with
of any shape and design depending on the geometry of compressed air or water pressure. The pneumatic test
the part and on the type of defects to be detected. using compressed air is used only for low-pressure
When discontinuities are present in the part, they testing, because of the inherent danger of explosion if
alter the magnitude and direction of the eddy currents. the job fails suddenly. The hydrostatic test using water
The alteration is detected by a detector coil and the under pressure is a high-pressure test. In this case, the
discontinuity is registered. Changes in the eddy current pressure used is the design working pressure plus a
are interpreted by various electrical variables such as certain percentage in excess. The container under
voltage, current, impedance, phase or their combin- pressure, which may also be subjected to a hammer test,
ations. These variables are electronically analysed and is visually examined for leaks.
the desired information obtained in useful form. Open containers like water or oil storage tanks are
Eddy current testing is used almost exclusively in simply tested by filling with water and visually
the mass production of elongated articles of uniform examining for leaks.
cross-section, such as tubes, pipes, bars and billets. Welded structural members designed to carry certain
In the case of welded tubular products, the inductor loads in service can be proof tested by setting them up
is in the form of coils that surround the material. If they and loading them in a testing machine. Large members
are designed to induce a circumferential current flowing may be inertly loaded with weights like sand bags to
through the entire tube wall, one can detect defects all determine whether the resultant deflections are within
over the wall of the tube. However, to inspect only the the calculations. By loading in measured excess, one can
longitudinal seam weld of the tube, detector coils are determine if any part of the structure undergoes
positioned tangent to the tube near the weld. In this permanent deformation.
case, one can obtain more accurate information on the Procedures of the proof test, if specified, must be
presence and magnitude of small defects. The inform- provided in the drawing, specification sheet or contract
ation picked up by the tangent detector coil is transmit- for the welded product.
ted to the test circuit and analysed electronically.
Since eddy currents can be induced in any conductor,
the test method may be employed equally well on Leak Testing
magnetic and non-magnetic materials. In the case of Welded products such as vessels and piping systems
magnetic materials, it is advisable to magnetically are leak tested to determine if through-thickness
saturate the part, while it is being tested, by surrounding discontinuities exist in the welded joints, which will
it with a DC solenoid. This helps to suppress the result in a loss of liquid or gas in service. A simple leak
influence of certain disturbing magnetic variables. test consists of internally pressurising the container with
To successfully develop the test, it is necessary to water or light oil, and examining the welds externally
establish the relationships between all types of natural for evidence of leakage. While water is cheap and readily
defects, harmless variables, and the artificial defects available, light oil is more effective, as it can penetrate a
created for calibration purposes. given discontinuity at half the pressure that water would
Eddy current testing can be fully automated to require. In water testing, small leaks may sometimes be
provide automatic inspection at high speed and at missed. To increase the sensitivity of the method, a water
relatively low cost. Under certain conditions, the tests soluble fluorescent dye is added to the water and the
are useful for grading and classifying defects, since the outer surface of the container is inspected with
indications produced are proportional to the actual ultraviolet light.
defect size. In pressure testing with a liquid, care must be taken
to ensure that all air within the container is completely
removed as it gets filled up with the liquid. This may be
Proof Testing achieved by providing the necessary vents in the
Important welded structures are proof tested by topmost part of the container.
subjecting them to stresses above those they are When air or inert gas is used for pressurising, a soap
expected to carry in service, but below the elastic limit solution is applied outside on the joints, and the leaks
of the metal, and making sure that they will not fail or are located by means of bubbles formed. If the container
get permanently deformed in any part. is small enough, it may be completely immersed in a
liquid. In this case, the leaks are indicated by bubbles of vessel and any leak present is detected by the presence
the escaping air or gas. In some cases, leakage of gas of the tracer gas. For critical vessels which have to be
can be simply determined by the hissing sound of fully guaranteed against leakage, helium leak detection
escaping gas or the flickering of a flame held close to is used, which can detect the presence of one part of
the leak. Pressure testing with gas is hazardous and it helium in 2,00,000 parts of air. In this case, any possible
should be carefully controlled. leak is d e t e c t e d b y m e a n s of a h e l i u m m a s s
Leak testing can be m a d e more sophisticated by spectrometer.
adding a small percentage of a tracer gas to air or some W h e n pressure tests are used on high pressure
other carrier gas, introducing the mixture into the vessel vessels such as heat exchangers, it may be necessary to
and increasing the pressure to a predetermined level, maintain the pressure for periods between 30 min and
A sensing probe is then passed over the outside of the four hours.
Welding Inspection, 1980, by American Welding Society.
Guide to Welding Inspector Qualification and Certification including AWS QC1-84
Standard for Qualification and Certification of Welding Inspectors, 1964, by American Welding Society.
Standard Methods for Mechanical Testing of Welds, by American Welding Society.
Handbook on Radiographic Apparatus and Techniques, 1973, by International Institute of Welding.
Handbook on Ultrasonic Examination of Welds, 1977, by International Institute of Welding.
IS 822-1991 Code of Procedure for Inspection of Welds.
IS 1182-1983 Recommended Practice for Radiographic Examination of Fusion Welded Butt Joints in Steels.
IS 4853-1982 Recommended Practice for Radiographic Examination of Fusion Welded Circumferential Joints in Steel Pipes.
IS 3657-1978 Radiographic Image Quality Indicators.
IS 3658-1981 Code of Practice for Liquid Penetrant Flaw Detection.
IS 4260-1976 Recommended Practice for Ultrasonic Testing of Welds in Ferritic Steel.
IS 4904-1987 Reference Block for Calibration of Ultrasonic Flaw Detectors.
IS 7343-1974 Code of Practice for Ultrasonic Testing of Ferrous Welded Pipes and Tubular Products.
IS 3703-1980 Code of Practice for Magnetic Particle Crack Detection.
Radiographs of Welds by International Institute of Welding.
CHAPTER 14
Metal
Cutting
Processes
Arc Cutting of Metals cutting process except its limitations for cutting some
The various methods used for the severing or of the common metals and alloys.
removal of metals can be divided into two main It is difficult to cut cast iron by the normal oxyfuel
categories: mechanical a n d thermal. Mechanical process, because of the presence of graphite and iron
methods are sawing, drilling, shaping, milling, etc., carbide which hinder the oxidation of the iron. Alu-
commonly used in metalworking and machine shops. minium and its alloys and stainless steels are also diffi-
Thermal cutting methods involve heat. The process cult to cut, because the oxides of aluminium and chro-
chart presented in Chapter 1 shows the break-up of the mium have melting points much higher than those of
thermal cutting processes as follows: iron oxides. These high-melting oxides, which are re-
Oxygen cutting is based on the chemical reaction fractory in nature, continuously accumulate in the kerf
between oxygen and the metal at high temperatures. as cutting proceeds and prevent exposure of fresh iron
The necessary high temperature is obtained from an to the cutting oxygen stream. Special oxygen cutting
oxyfuel gas flame, the most popular fuel gas being techniques using metal powder or chemical flux are
acetylene. The process is therefore commonly referred therefore employed for cutting these difficult metals.
to as oxyacetylene process and is widely used in all For getting clean and efficient cuts at high speeds in
metal fabrication shops. A wide range of oxy-cutting these metals, arc cutting processes are preferred.
machines are available, with a simple manual cutting Thermal cutting also includes spark erosion, elec-
torch at one end and the most sophisticated design with tron beam and^ laser, as indicated in the chart. These
an electronic tracing device at the other end. modern sophisticated processes are rarely used in the
Since this chapter is restricted to arc cutting fabrication shops and, therefore, they are not discussed
processes, nothing more will be stated about the oxygen here.
Thermal Cutting
Oxygen Oxygen Arc Spark Electron Laser

Arc Erosion Beam
Flame Powder Lancing
Carbon Metal Air Plasma

arc arc arc arc
Oxygen-Arc Cutting manner as the powdered metal or flux helps during

This method uses a specially designed oxygen-arc oxyacetylene cutting of stainless steels.
cutting electrode to cut, pierce or gouge metals. The The equipment used for oxygen-arc cutting is an AC
electrode is similar to a manual arc welding electrode or DC manual welding power source of adequate
and is coated with a flux whose functions are to provide capacity. For rapid cutting, DC with electrode positive
an insulated sleeve, to stabilise the arc and to make the is preferred. A fully insulated electrode holder is used.
products of combustion more fluid. The core wire, The common dimensions of tubular steel electrodes are
however, is in the form of a hollow tube through which 6 mm and 8 mm diameter, 450 m m length and bore
a stream of oxygen is passed and directed to the area of diameter of 1.6 mm. The coating is usually of AWS E6013
the metal being cut. A specially designed holder, capable class (rutile type).
of conveying electric current to the electrode as well as For piercing operation, the arc is struck by tapping
oxygen to the arc, is used. The process is schematically the electrode tip at the desired point. As soon as the arc
shown in Fig. 14.1. stabilises, the oxygen valve is opened. As the piercing
In this method, the necessary preheat is provided action starts, the electrode is pushed into and through
by the electric arc, while the actual removal of metal is the plate to form a hole. The coating prevents the
effected by the stream of oxygen and the coating by a shorting of the electrode core against the sides of the
combination of melting, oxidising, diluting, fluxing and hole.
mechanical ejection. In the case of oxidation-resistant The cutting operation starts with the piercing action
metals such as stainless steels and aluminium and its as described above. The electrode is then inclined at a
alloys, the cutting occurs more by melting than by suitable angle and dragged along the plate surface at a
anything else. In this case, the electrode covering helps travel speed dictated by the progress of the cut, while
the molten metal to flow away from the cut in the same applying slight pressure. Using this technique, mild steel
up to 75 mm thickness and non-ferrous metals up to
Oxygen in 12.5 mm thickness can be cut. For gouging, the electrode
is inclined steeply until it is almost parallel to the plate
surface and pointed away from the operator along the
path to be gouged. Due to the scouring action of the arc
and oxygen, a pool of slag is formed beyond the
electrode tip. The electrode is pushed rapidly over the
plate surface into this pool.
In all these operations of piercing, cutting and
gouging, the coating is constantly kept in contact with
the base metal. The coating prevents arcing of the core
with the work and automatically maintains correct arc
length.
The metallurgical effects of oxygen-arc cutting are
similar to those of manual arc welding, but the HAZ is
shallower and harder with the cutting process, because
of the h i g h e r speed of arc travel. Cast iron a n d
hardenable steels may develop cracks on the face of the
cut. Grades of austenitic stainless steels which are
subject to intercrystalline corrosion when welded by the
shielded metal-arc process become sensitised along the
cut made with the oxy-arc process.
The faces of the cuts are r o u g h e r t h a n w i t h
oxyacetylene cutting and need some amount of chipping
or grinding before fabrication by welding. The cut faces
of cast iron and hardenable steels are likely to be non-
machinable.
This method is used effectively by foundries and
Fig. 14.1: Schematic of oxygen-arc electrode in operation scrapyards to cut all types of metals in all positions. It
METALS CUTTING PROCESSES 601
is, however, more effectively applied in underwater The cut may be started at the edge of the plate, or a
cutting operations. hole may be burned through the plate, and thus the cut
For underwater cutting, a fully insulated holder may be started at any desired point.
equipped with a suitable flashback arrester must be In piercing operations, the cutting electrode is held
used. During cutting, positive p r e s s u r e m u s t be perpendicular or at a very steep angle to the surface.
maintained against the metal being cut, irrespective of High currents must be used for piercing operations.
its thickness. To keep the metal and the slag flowing out of the
cut, a short arc should be held and the electrode should
Manual Metal-Arc Cutting be kept pushing closely up in the slot, with the tip of
This is a rough and ready method of cutting which the electrode being pushed right into the molten crater
uses a normal welding electrode or specially designed and manipulated with an up and d o w n action as
cutting electrode (4, 5 or 6.3 mm diameter), at a current indicated in Fig. 14.2.
setting which is 20 to 50% higher than that used The upward motion should be very fast and the
normally for a given size. Although AC can be used, downward motion should supply a pushing effect, thus
DC with electrode negative is preferred. Sometimes it helping to force out the molten metal and slag.
helps to make the electrode slightly wet. Water in the The metallurgical and physical effects are similar to
coating reduces overheating of the electrode to some those occurring in the oxygen-arc cutting process.
extent and dissociates in the arc to render it more When no better method is readily available, metal-
penetrating. arc cutting is widely used for cutting holes into piping,
Since there is no great action tending to force the through plates for subsequent welding of extensions,
molten metal out of the cut, some provision must be attachments, etc.
made to permit the molten metal to drop or run out of For piercing, drilling and cutting operations, the
the cut, irrespective of the position of the work. This is cutting electrode is particularly effective on the difficult-
achieved by starting the cut at some convenient point, to-machine metals and alloys. Armour plates, air- and
from which the flow of the molten metal is facilitated d e e p - h a r d e n i n g steels, stainless steels, and hard
by the gravitational force. overlays are some of the materials easily cut. In ferrous
Plates 10 mm in thickness or lighter can be cut by and non- ferrous foundries, the cutting electrode is used
simply moving the electrode along the line of cut. For for the removal of riser and gates and to cut heavy scrap
heavier plates, it is necessary to work from the underside for remelting.
of the plate, so that the molten metal and slag may have
a chance to drop or run off. Carbon-Arc Cutting
In this method, cutting is performed by the are
formed between two carbon or graphite electrodes or
between one such electrode and the base metal. No
compressed air is used. At one time, this process was
u s e d by small s h o p s , garages and h o u s e h o l d
metalworkers as it involved a small investment, but
today it has almost entirely been replaced by better arc-
cutting methods.
TIG and MIG Arc Cutting

The normal TIG welding process can be converted
into a cutting process by increasing the current and rate
of inert gas flow. The moderately high velocity of the
gas, which is usually a mixture of argon and hydrogen,
blows away the molten metal to form a kerf. Using cur-
rents in the range of 200-600 Amps., stainless steel and
aluminium up to 13 mm thick can be cut with ease. The
MIG process can also be used for cutting in a similar way.
Considering the expense of inert gas and electrodes,
Fig. 14.2: Technique of manual metal-arc cutting TIG and MIG cutting processes are rarely used in
industry. Hence they are not mentioned in the chart. shielding gas. In either case, the cut quality is not as
Today, the most efficient and commonly used processes good as when a single-flow technique is used.
are plasma arc cutting and air carbon-arc cutting. The most efficient system of PAC which is capable
of giving narrow, sharply defined cuts at high speeds
Plasma Arc Cutting (PAC) involves injection of water into the plasma flame, as
The PAC torch is designed in the same manner as shown in Fig. 14.5, to further constrict it and to prevent
the plasma arc welding torch shown in Fig. 5.36. DC is the plasma from turbulent mixing with the surround-
used and the tungsten electrode is connected to the ing air. This process is termed water injection plasma
negative pole. A transferred arc is employed for cutting, cutting.
i.e. the arc is sustained between the electrode in the torch PAC equipment consists of a power source, a control
and the work. The cutting arc is initiated by a pilot arc, unit, one or more gases to function as orifice and
which is first formed between the electrode and the shielding gas, and a cutting torch. It can be manual or
constricting nozzle by the high frequency generator. The mechanised. The power source is of drooping type,
constricting nozzle, which is water-cooled, is connected giving DC with open-circuit voltages in the range of
to the positive terminal of the power source through a 120-400 V and output current in the range of 70-1,000
current limiting resistor and a pilot arc relay contact. Amps.. It may also contain the pilot arc and circuitry
The simple PAC circuitry is shown in Fig. 14.3. for high frequency generation. For cutting ferrous
This figure shows a single flow PAC torch, i.e. a torch material as thick as 75 mm and aluminium up to 90 mm,
which provides for the flow of orifice gas and not of OCV of nearly 400 V and current up to 500 Amps, may
shielding gas. The orifice gas, which is preheated by be required. Diagram of a typical outfit of this capacity
the arc plasma, expands and is ejected through the is shown in Fig. 14.4(B). Because of the inherent danger
constricting orifice at a high velocity. The metal which of electric shock to the operator, manual cutting
is melted by the arc is then blown away by the kinetic equipment uses lower OCV of 120-200 V, and current
energy of the gas stream to form the kerf. The orifice in the range of 70-100 Amps., and achieves relatively
gas is usually argon. low cutting speeds. Manual PAC can be used
Sometimes a dual-flow torch shown in Fig. 14.4(A) conveniently for cutting non-ferrous metals up to 25 mm
is used, in which provision is made for the flow of a t and carbon steel up to 12.5 mm t.
secondary shielding gas around the arc plasma. In this The control unit contains solenoid valves to turn
case, the usual orifice gas is nitrogen. The shielding gas gases and cooling water on and off as required. They
is carbon dioxide in the case of mild steel and stainless control the rate of flow of cutting gas through
steels, and argon-hydrogen mixture in the case of flowmeters, and have a water flow switch to switch off
aluminium and its alloys. The technique can be modified the operation if water-cooling is inadequate. The high
by using water as a shield in place of the auxiliary capacity automatic machines may contain features for
upslope and downslope control of current and orifice
gas flow.
Cutting torches are of several types and for each type,
nozzles with various orifice diameters are available. The
I higher the current, the larger is the orifice diameter
High
frequency required. The nozzle is designed according to the PAC
generator system used and the metal being cut. For better quality
cuts, multiple port nozzles, which have the auxiliary
Power
gas ports arranged in a circle around the main orifice,
supply Electrode are preferred to single port nozzles.
In mechanised equipment, PAC torches are mounted
Constricting on shape cutting machines similar to the commonly
nozzle
used mechanised gas cutting equipment, while the
cutting is steered by photoelectric tracing, numerical
JsVorkNWWW^NI control or computer control. With mechanised PAC, it
is possible to cut 150 mm of aluminium alloy at a speed
Fig. 14.3: Basic plasma arc cutting circuitry of 3 mm/sec, 100 mm t stainless steel also at 3 mm/sec,
and 50 mm carbon steel at 11 mm/sec.
Electrode
Orifice gas Shielding gas
Constricting
nozzle Shielding gas
nozzle
Workpiece
(A)
POWERSOURCE
CCR 320(400) lHf y
yj£&
^ 5 o c / CONTROL UNIT PENDANT UNIT
Spcci01
El El CARRIAGE
.SPEED POT
CARRIAGE
CARRIAGE V - -^ REV/FOR
ON/OFF y^TTSJOP
START
1*1 fit CABLE TO OPERATE
MAIN PENDANT CABLE
7«n
CONTACTOR
as
m
FLEXIBLE
TAP
F ^ B ^ G ' N P U T ^ B V WATER
BOX 3J50C/S "IN
_^
WATER
RESERVOIR
UNIT K 101
PRESSURE
REGULATE R
JM
CUM
FLOWMETER
WATER IN
zJ
(B)
CONTROL CABLE
CUTTING TORCH
PLUMBING^L PL-1000-(I)
BOX TRAVEL CARRIAGE
TC101
y/s/sssysTy; / / ^~M^ ">>'> \>>>>> > TT>
TRACK2MTR
INTERCONNECTION HDPC-101 ^
.WORKPIECE
Fig. 14.4: Plasma arc cutting: (A) Dual flow cutting operation; (B) Schematic diagram of heavy duty equipment
Model / Specifications
Orifice gas swirl
/ / Elect rode Tracing Width : 1500 to 2000 mm
y^ Constricting nozzle Cutting Width : 1 x 1500 to 1 x 2000 mm
(With single torch)
Cutting Width : 2 x 750 to 2 x 1250 mm
C « L K ' Ceramic insulator (With two torches)
^Workpiece
Cutting Width : 4 x 375 to 4 x 625 mm
^tHUV^ (With four torches)
Cutting Thickness : 100 mm max. with torches
150 mm max. with 2 torches
Fig. 14.5: Water in jection plasma arc cutting arrangement 200 m m max. with 1 torch
Machine Speed Range : 70-1000 m m / m i n

Supply Voltage 200 V + 10%
Safety: PAC gives pronounced arc glare, spatter,
-15%
fumes and noise. It is necessary to protect operators and
50 Hz, 1-phase AC
other persons working in the vicinity from exposure to
Power Requirements: 500 VA (approx.)
these unpleasant aspects. It is specially important that
fume and noise are controlled. One way to control fumes
Optical Tracer and Control Features
is to place the plate to be cut on a cutting table, which is
filled with water up to the bottom surface of the plate. • Automatic shut-off of motion and cutting oxygen
The plasma jet releases gases at high speeds, which strike gas if tracer loses line
the water and produce turbulence in it. All the fume Automatic line acquisition
particles get trapped in the turbulent water. Integrated control panel for all controls
The noise is controlled by using what is termed a Electronic kerf setting upto 3 mm kerf
water muffler in conjunction with the water table Command mark option
described above. The muffler is in the form of a nozzle CNC compatible — optional interface available
which is attached to the torch body and releases a curtain Patterns—silhouette or line (minimum line width
of water around the front of the torch. The water 0.8 mm)
required by the nozzle is pumped from the water table.
The water curtain around the torch and the water Optical Features
surface of the table combine to form a sound-muffling Gas economizer — preheat control
shield around the arc. The water muffler must not be Motorized torch lifting
confused with water shielding and water injection used Upto 4 torch stations
in the modified systems of PAC. Longitudinal hose conveyors
Trace Extensions — Units 1,500 mm or 3,000 mm
Optical Profile Cutting Tracing Table Extensions — Units of 1,000 mm
This is cantiportal construction, equipped with optical
tracer and controller with gas on-off controls by sole- Air Carbon-Arc Cutting
noid valves. The longitudinal and transverse axes are In this process, which is also known as air arc cutting
controlled by Pulse Width Modulated (PWM) drive sys- (AAC), an arc is struck between a carbon electrode and
tem that gives precise control and accuracy. The main the workpiece to melt the latter. The molten metal is
beam is provided with all the transverse hose convey- ejected by a powerful jet of compressed air directed at
ors, hoses and cables, gas equipment, solenoids, con- the point of melting. As' the carbon arc is advanced,
trol panel, torch stations, etc. The basic is machine com- metal is continuously removed. The process is illustrated
plete with all interconnecting cables, gas hoses and in Fig. 14.6(A).
manifold, etc. All controls are integrated in one control The process is well suited for gouging out metal from
panel for operators' convenience. The optical profile a weld root or a defective weld area, and for groove
cutter is equipped with gas cutting using Nozzle Mix preparation prior to welding. It is also used for severing.
technology suitable for oxy-acetylene oxy-LPG gases. Unlike oxyfuel gas cutting which relies on oxidation and
A choice of four models is available. removal of low-melting slags, AAC involves only
Table 14.1: Currents for air carbon-arc cutting electrodes
Electrode dia. DC electrode AC electrode AC electrode

mm with DC with AC with DC
Amps. Amps. Amps.
4.0 90-150 — —
4.8 150-200 150-200 150-180
6.4 200-400 200-300 200-250
8.0 250-450 — —
9.5 350-600 300-500 300-400
12.7 600-1,000 400-600 400-500
16.0 800-1,200 — —
19.0 1,200-1,600 — —
melting and no oxidation. The process is primarily During cutting, the carbon electrode tip is heated to
physical and not chemical; hence it can be used on all a high temperature but it does not melt. The carbon at
metals including those which produce high melting the tip, however, gets oxidised and also sublimates
refractory oxides. (changes from solid to gaseous state). Hence the
Electrode holder-
/% + Carbon
Air M electrode
jetsNw^
(B)
Air jet 1 '" **H
on-off
button
Fig. 14.6: Air carbon-arc cutting:

(A) Cutting operation; (B) Holder or torch for manual cutting
electrode is gradually consumed as cutting progresses. continuous jointed carbon electrodes and a voltage
The process can be manual, semi-automatic or fully- controlled unit to maintain constant arc length.
automatic. A constant-current motor generator or rectifier,
The AAC electrode is made from a mixture of carbon d e s i g n e d for m a n u a l metal-arc w e l d i n g , is
and graphite. Carbon is amorphous, hard and highly recommended for AAC with all sizes and types of
resistant to abrasion or erosion. Graphite, on the electrodes, provided it has an open-circuit voltage of at
contrary, is crystalline and soft, oxidises at a higher least 60 V and sufficient current output. For AC
temperature, and has a lower rate of oxidation and lower electrodes, a constant current welding transformer is
electrical and thermal conductivity. For o p t i m u m suitable.
performance, the electrode must be hard and must have Compressed air under pressure range of 560-700 kPa
a high resistance to erosion, a high oxidising (80-100 psi) is required to obtain clean cuts at optimum
temperature, a low rate of oxidation, high electrical speeds. If air pressure is inadequate, the gouges and
conductivity and low thermal conductivity. These cuts get contaminated with slag and carbon deposits.
properties are achieved through proper blending of All hoses and fittings must be of large diameters to allow
carbon and graphite. A suitable binder is used. The sufficient volume of air to reach the electrode holder.
mixture is baked at a high temperature to produce a For light duty holders, hoses and fittings should have a
dense, homogeneous graphite electrode of low electrical minimum inside diameter of 6.4 mm. For heavy duty
resistance, which is finally plated with a controlled holders, a m i n i m u m inside diameter of 9.5 m m is
thickness of high purity copper. The copper-coating recommended. For automatic holders, the minimum
increases its current-carrying capacity, extends its diameter is 12.7 mm.
operating life, reduces radiated heat and helps to obtain For manual gouging, the carbon electrode is gripped
a smooth and uniform groove. in the holder in such a way that the maximum extension
The electrodes are commonly designed for use with from the holder to the workpiece is 150 mm in the case
DC, electrode positive. AC electrodes are also sometimes of ferrous metals and 100 mm for aluminium alloys. The
available. They carry a small proportion of rare-earth air jet is turned on and then the arc is struck by touching
metals to p r o v i d e arc stabilisation w i t h AC. AC the electrode tip lightly to the work. It is not necessary
electrodes are preferred for cutting cast iron and nickel to draw back the electrode after arc striking, because
alloys. AC electrodes may be used with DC, but in this the metal is being removed and not deposited. The
case the electrode is made negative. DC electrodes are holder is held so that the electrode is properly slanted
usually available in the diameter range of 4-19 mm, with respect to the work, with the air blast behind the
while AC electrodes are available in the range of 4.8- electrode. If the air stream is correctly directed at the
12.7 mm. arc, the removal of molten metal from the kerf is fairly
For continuous automatic cutting, jointed electrodes rapid and the operator must use his skill to maintain
are used. They are provided with a female socket and a correct arc length by moving the holder fast enough in
matching male tenon. the direction of the cut. A steady progress is necessary
Suggested current ranges for the commonly used to ensure that the cut surface is smooth.
electrode types and sizes are given in Table 14.1. Gouging can be done in all positions. In the vertical
Holders are of special design. Holders meant for position, it should be in a downward direction. In the
manual gouging (see Fig. 14.6(B)) arc similar to the horizontal position, gouging can proceed from left to
heavy duty holders used for manual metal-arc welding, right or from right to left, provided the air jet is behind
but they are provided with air passages and orifices to the electrode. For overhead gouging, the electrode is
direct the air stream along the bottom side of the held nearly parallel to the holder, while the holder is
electrode, and with valves to turn the air on and off. held at an angle that will not let molten metal drip on
Semi-automatic holders are designed to be mounted on the operator's glove.
a mechanised carriage and they use standard lengths The depth, width and contour of the groove are
of electrodes. The operator feeds the electrode manually controlled by the combined effects of electrode push
and controls the arc length at the same time. Automatic angle, arc travel speed and size of the electrode. A small
holders are also mounted on mechanised carriages, and push angle and low travel speed give rise to a deep and
the automatic gouging operation is achieved by using narrow groove, while a large push angle and high speed
produce a wide and shallow groove. The groove width Advantages

is usually 3 mm greater than the electrode diameter.
1) Fast: five times as fast as chipping. Gouging a 9.5
With a given size of electrode, a much wider groove
mm groove at over 600 mm per minute.
can be obtained by weaving. The proper travel speed
depends on the type of metal to be cut, size of electrode, 2) Easily controllable: removes defects with preci-
amperage and air pressure. A good clean cut is an sion. Defects are clearly visible in the groove and
indication of the correct speed. can be followed with ease. The depth of the cut is
Data for automatic cutting are obtainable from the easily regulated and welding slag does not de-
equipment supplier. They cover the following param- flect or hamper the cutting action.
eters with respect to the U-groove size: electrode size,
3) Low equipment cost: no gas cylinder or regula-
current, voltage, electrode-feed rate and arc travel speed.
tors are necessary except in field operations.
For severing or through cutting, the electrode is held
at a steeper push angle and the metal is pierced with 4) Economical to operate: no oxygen or fuel gas re-
the tip of the electrode. The arc is then moved up and quired. The same man who performs the weld-
down through the metal with a sawing motion. ing can operate the equipment.
Metallurgical effects: Since most AAC is done with DC,
electrode positive, ionized carbon atoms are transferred 5) Easy to operate: any welder can operate the
from the electrode to the melted base metal. This carbon equipment with a few minutes of instruction, and
absorption is unavoidable and, therefore, the air blast can become proficient in a few days. The elec-
must ensure that all such molten carburised metal is trode holder contains air control valve and a ro-
removed from the kerf. If such high-carbon metal, tating nozzle, which permits changing the elec-
having up to 1.0% C, is left in the kerf, it gives a dull trode angle to suit the job while maintaining the
grey-black colour to the groove. Insufficient air flow, air jet in alignment.
irregular electrode travel and a wrong electrode angle
can leave small pools or beads of carburised metal. If 6) Compact: not much larger than a shielded metal-
the groove is correctly made, it develops a bright-blue arc welding electrode holder.
colour.
7) Versatile: can be used in any place where it is pos-
If the presence of carburised metal is suspected, it is
sible to weld. It can be operated in spaces too re-
necessary to remove it by grinding before depositing
stricted to accommodate a chipping hammer or
the weld, especially when optimum weld ductility and
an oxyfuel gas cutting torch. It requires no diffi-
toughness have to be guaranteed.
cult adjustments for use on different metals.
It has been established that copper is not transferred
from the electrode to the groove or cut surface. With 8) Cuts cleanly: resulting surface is clean and
structural steels and non-hardenable steels, the groove smooth. Welding or brazing can generally be
and cut surfaces are machinable, but in the case of cast done without further grinding or cleaning.
iron and hardenable steels they are non-machinable. In
all cases, the HAZ is usually very shallow. 9) Compared to oxyfuel cutting, the AAC process
Applications: The AAC process is commonly used to involves lower energy input. Therefore, a
cut and gouge mild, low-alloy, austenitic manganese and workpiece gouged or cut by this process distorts
stainless steels; cast iron; and alloys of aluminium, mag- less.
nesium, copper and nickel. It is also used to remove old
surfacing material before depositing new surfacing al- Safety: The process generates a high level of fumes
loy. It is well suited for gouging out the root of a weld and noise. The operator should take care that the fumes
to sound metal, before depositing a sealing pass on the do not rise up to his face. If he is working in an enclosed
second side, as also for gouging out defective weld- or semi-enclosed area, exhaust ventilation should be
metal for subsequent repair. The AAC is a simple and provided and the operator must wear a respirator. If
versatile tool, which is extensively used for cutting the cutting operation is continuous, the operator as also
out metals at construction sites and in foundries, scrap- other persons in the vicinity must be provided with
yards, shipyards/boiler shops, maintenance shops, etc. suitable ear protection.
Welding Handbook, Chapter 13, Vol. 5, 7th Ed., published by American Welding Society,
IS 8987-1978 Recommended Practices for Air-Carbon-Arc Gouging and Cutting.
IS 6431-1971 Tolerances on Dimensions of Plates Cut by Flame.
AWS C4.2 Operator's Manual for Oxyfuel Gas Cutting, 1978, by American Welding Society.
AWS C5.2 Recommended Practices for Plasma-Arc Cutting by American Welding Society.
AWS C5.3 Recommended Practices for Air-Carbon-Arc Gouging and Cutting, 1983, by American Welding Society.
CHAPTER 15
Welding
Costs and
Economics
FOR PREPARING estimates of arc-welded fabrications, cm3. Using this factor, the weight of weld-metal required
one must have technical calculation data to determine to fill the joint can be calculated.
direct welding costs based on (a) consumption of Standard tables for weight of weld-metal for
electrodes, (b) arc time, and (c) current consumption. common joints are available in engineering handbooks
and welding data books. Typical tables for fillet welds
Electrode Consumption and butt joints are given in Tables 15.1 and 15.2.
The first step is to determine the quantity of There are several m e t h o d s for calculating the
electrodes required to complete unit length of a given quantity of electrodes required to obtain a given weight
type of joint. Once this is known, the arc time and current of weld-metal. A simple and direct method is to deposit
consumption can be determined by simple calculation. weld-metal from, say, five electrodes on a plate, using
As an example, we can take a 10 mm thick single-V the mean amperage of the current range and leaving a
butt joint of length 1 m, as shown in Fig. 15.1. The cross- stub of 50 mm length, and determining its weight.
sectional area, divided into segments A, B, C and D, Alternatively one can refer to the electrode producer
can be calculated by simple arithmetic. This area for the quantity of electrodes required to deposit 1 kg
multiplied by 1 m gives the volume per metre length, of weld-metal, or the weight of weld-metal obtained
which can be expressed in cm3. Density of steel is 7.8 g / from each electrode.
Table 15.1: Weight of steel weld-metal for fillet joints
Flat Convex Concave

Size of fillet (mm) Wt. of metal (g/m)
3.15 48.3 58.9 55.9
4.80 108.7 131.4 125.3
6.30 194.8 234.0 222.0
8.0 303.5 365.4 347.3
9.5 436.4 527.0 500.0
12.7 776.1 936.0 889.4
Courtesy: Lincoln's Procedure Handbook
2 mm
Fig. 15.1: Segments of volume of weld-metal In a butt joint
Rather than providing the above-mentioned It must be remembered that the data arrived at
information, electrode producers usually mention the through mathematical calculations and welding trials
deposition efficiency (DE) of the electrode in the leaflet. can be proved wrong in actual practice for the following
DE is defined as the ratio of the weld deposit to the reasons:
weight of core wire melted, expressed as a percentage. a) Variations in the dimensions prescribed for joint
For example, if an electrode has 90% DE, 90 g of weld- preparation
metal will be obtained from every 100 g of the core wire Even small departures from prescribed dimen-
melted in the arc. DE depends significantly on the type sions in edge preparation or fit-up lead to significant
of electrode used,*and lies between 80 and 95% for increases in the volume of weld-metal required for the
common types which do not contain iron powder and joint. For example, an increase of 3° in the 60° included
do not have appreciable proportions of ferroalloys, angle in a Vee leads to an increase of 5% in cross-section.
because some loss of metal occurs through spatter and b) Inaccurate fit-up
evaporation in the arc. Values over 100% are obtained In this case also, small departures from prescribed
when sufficient iron powder is incorporated in the flux dimensions result in large increases in cross-section.
covering. For example, an increase of root gap from 0.5 to 1.5
Weights of steel wires of various standard diameters mm in a 60°-V butt joint of 10 mm size results in 16%
per metre length are given in engineering handbooks. increase in the cross-section of the weld joint
From these data, one can calculate the weight of say c) Variable weld shrinkages and weld reinforcements
400 mm length of 4 mm diameter wire, which is the In certain forms of joints, the shrinkage of the weld-
length of core wire that is melted from one electrode, 4 metal on cooling causes a decrease in the volume of
x 450 mm long, after allowing for 50 mm stub length. If the joint and thereby a decrease in the consumption of
this weight is X, and the DE of the electrode is 90%, the electrodes. In data tables, it may be that the average
weight of weld deposit yielded by the electrode is X x shrinkage of joint is already accounted for. However,
0.90. If the DE is 120%, the weight would be X x 1.20. there can be fluctuations in shrinkages and the
Knowing the weight of weld-metal required to fill a corresponding volume changes due to varying
joint and the weight of weld-metal deposited by one contraction stresses exerted on the sections being joined
electrode, the total number of electrodes required for and these can cause departures within wide limits from
the joint can be estimated by using simple arithmetic. average shrinkage referred to above.
Where repetitive welding jobs are carried out in a Above all, in tee joints, corner joints and I-joints
shop, it is easy to collect data for such jobs through as well as in multi-pass V- and X-joints, more or less
workshop experiments and job record cards. For weld reinforcement or convexity cannot be avoided.
example, a firm constantly engaged in girth welding This factor is also accounted for in data tables. However,
of cross-country pipelines can build up data for the amount of weld reinforcement or convexity
different diameters and wall thicknesses of pipes. depends to a large extent on the skill of the welder,
WELDING C08T8 AND ECONOMICS 611
co en in co I I i I
in ^ H oo ' ' ' '
CM © CO O
H If) H O
H CN
O O v O O C M ^ O O
Q O O ^ O C O Q O C N
^ C M O O ^ O ^ C M O O
T-HfOTTsOXî-iCM
oo CN Q t< © CM ©
SS3SP!
ppoqpincMpcM
I co
I
t-H CM CO *3< ^f If) SO
! O O O r J i O O O H O H
"5 rfOOCNSoÔOCM
t-H H CM CM CM CO
o
<<£ t"H 0 0
in co t-H o o
»?> © rH
ON H N . . N. 8 8 3 8 3
oo ro N (S q 55 a\s H
H N N c^ in
E
l
■o
1 &
>-*
1 •8 R
ON©CO^COrfCM00
If) If) ^ 00
i n <N Q ON —. - , ^ -►
H in oo ^ v p IN U\ 3$g
1
v* CM* CM CO
v, rHlNtxîvOT-HOO
învOrHvovdino
\ O 0 0 H r f i n N ^ «
at ^ H Tf N H ^ ( S ON
I
rH rH* CM CM
k-~-t
CO I I M I §||
in
o
3 00
I
1
j
'
I vO v O O v O N
I t-H r-i Q rn rH
CM O 00 SO CO
CO r* rf< i n vO
CM I I o ^ co o m oo
co rH CM CM CO CO "^ 8
00 O ( S ' t vO fN Tt
H O N ^ N N ^t
m oo o co so oo t-H
r H t-H t—t r H CM
cMcoiniNoqpcMTi*
fivbaNriifioNNiri
H H H (N| M
and so, especially in thin plates, there can be wide The ratio between pure arc time and total welding
departures from the values given in the tables. time is sometimes expressed as a percentage and
d) Use of excessive current referred to as operator factor (O.F.). In the above
The welder may use excessive current, due to example, when 80 pieces are consumed in a shift, O.F.
which the spatter loss increases and the DE of the is 25% and when 100 pieces are burnt, O.F. is 150 xl00/
electrode is lowered. 480= 31.25%.
e) Stubs too long In manual welding, 100% O.F. can never be reached
The welder may for various reasons such as for obvious reasons. A 60% O.F. can be maintained in
premature overheating of the electrode due to excessive straight welds in the flat position, provided the setting
current, carelessness and inaccessibility of the joint, up of the joint, positioning, precleaning, etc., are
discard stubs longer than 50 mm and thereby obtain prearranged for the welder. 60% O. F. is considered
less weld-metal per electrode. excellent, 40% is considered good, while 25% is
considered average. It is possible to reach higher O.F.
Arc Time figures in semi-automatic welding, and still higher in
Arc time is pure welding time required to melt the fully-automatic welding.
quantity of electrodes necessary for completing a
certain joint. To find this, one multiplies the number of Current Consumption
electrodes required for the joint by the melting time This refers to the cost of electrical power used for
per electrode. The melting time must of course be arc welding. The following equations help to calculate
determined for every type of electrode used and for this cost: f
every diameter of electrode used, because each of these Welding amps x arc voltage/1,000 = Arc kilowatts
has a different melting time. The melting time per Arc kilowatts/% efficiency at arc voltage = Input kilo-
electrode depends on the welding current used. Widely watts
varying welding currents are used in a shop depending Input kilowatts x rate for kwh = Power cost per hour
on the conditions. When one wants to work accurately, Power cost per hour/metre length of weld deposited
it is r e c o m m e n d e d that one carries out tests to in one arc hour = Power cost per metre
determine welding times for each electrode type and Power cost per h o u r / N o . of electrodes consumed in
diameter for current conditions used in one's own shop. one arc hour = Power cost per electrode.
The pure welding time obtained by multiplying the
number of electrodes by the melting time per electrode Labour Cost
is exclusive of the time required for slag removal, Direct labour cost of welding usually covers the
changing of electrode, setting up of the job, and other wages of the welder and the helper who helps setting
incidental but unavoidable operations. For arriving at u p the job, precleaning, etc. From the n u m b e r of
the total welding time, which includes all other time- electrodes consumed by a welder during a shift, the
consuming operations, there are two possible methods: total weight of weld-metal deposited can be calculated.
a) The accurate method consists of conducting time The combined daily wages of the welder and his helper
studies to find out incidental time expenditure in divided by the weight of wreld-metal gives the labour
welding. cost per kg of weld-metal.
b) A less accurate method but which fully satisfies
the requirements of estimation work, takes into Overhead Cost
account the incidental time expenditure by mul- Overheads cover all the costs involved in shop
tiplying the pure welding time by a so-called m a n a g e m e n t . They may i n c l u d e the costs of
weld time factor. s u p e r v i s i o n , m a i n t e n a n c e services, e q u i p m e n t
Weld time factor can be determined by noting the depreciation and repair, cost of cleaning and lighting,
number of electrodes a welder consumes in a shift of etc. Overheads may in some cases even include the total
eight h o u r s (480 min) a n d the melting time per electric bill including welding power costs. For easy
electrode. For example, if 80 pieces are consumed in a calculation, the relation between labour cost and
shift and the melting time per electrode is 1.5 min, the overhead cost is determined and the resultant factor is
total arc time amounts to 120 min, and the weld time used for calculating overhead costs. In an average shop,
factor is 480:120 = 4.0. If 100 pieces consumed, the weld the overhead cost may be three to five times the labour
time factor is 480:150 = 3.2. cost.
WELDING COSTS AND ECONOMICS 613
Thus the final cost of 1 kg of weld-metal of an 4) Maintaining accurately the chosen joint design
electrode consists of the following individual costs: during preparation of joint edges and fit-up.
1) Cost of electrode per kg of weld-metal 5) 100% inspection of welding with the aim of elimi-
2) Labour and overhead costs per kg of weld-metal nating completely defective or unacceptable
welding.
3) Power cost per kg of weld-metal
4) Equipment cost per kg of weld-metal. Very often, by going in for a more expensive
Items 3 and 4 together normally a m o u n t to a electrode, one can eventually bring down the total
maximum of 10% of the total costs. In most cases, welding costs. For example, paying 40% more for an
therefore, it suffices to compare the cost of electrode electrode becomes an economical proposition, if
and the labour and overhead costs involved in its use because of its faster melting rate and higher deposition
to judge the economics of various types of electrodes. efficiency, the labour and overhead costs can be lowered
by at least 16% when compared with the normal
Economical Evaluation inexpensive of electrode.
In selecting an electrode for a welding job, all that a
consumer usually does is to find an electrode whose Oerlikon System Of Estimation
performance characteristics appeal to the established This system, prepared by Oerlikon Elektrodenfabrik
taste of his welder, which is suitable for welding in the Eisenberg GmbH of West Germany, is explained fully
various positions as required by the job and whose in Ador Welding's Welding Handbook. It has been
weld-metal meets the quality requirements dictated by prepared specially for the designing and planning
the specification. If need be, he also compares the list engineer to help him specify the appropriate electrode
prices of the electrodes in question and finally decides and welding procedure and also to help him estimate
on the cheapest electrode, thinking that he has chosen the cost of electrodes and welding to complete a job.
the most economical one. The data will also help him to ensure that the job is
Here we w o u l d like to point out that it is an carried out most economically. It contains a large
exceedingly short-sighted policy to judge the economics number of tables which provide the data that could be
of an electrode by looking at its list price. used for preparing estimates, and these data have been
On an average, every rupee spent on welding costs arrived at through methematical calculations and
can be broken down into four categories as follows welding experiments.
(based on German conditions): The tables in the first section contain data for
25 paise on electrodes effective standard electrode requirements per metre
6 paise on electric power length of weld for various forms of joints and welding
4 paise on equipment cost positions. An effective standard electrode means an
electrode having 100% deposition efficiency and 50
65 paise on labour cost and overheads. mm long discarded stub. In the case of 3.15, 4.0 and
From these figures, it is evident that the most effec- 5.0 mm diameter electrodes, the effective standard
tive means of reducing the total welding costs are avail- electrode is assumed to have a normal length of 450
able in the item of labour cost and overheads. Labour mm. In the case of 2.5 and 2.0 mm diameter electrodes,
cost and overheads can be lowered by reducing weld- the normal electrode length is taken to be 350 mm,
ing time. This can be achieved by the following means: and in case of 1.5 mm size, the normal length is 250
1) Choosing the right electrode and the right mm.
sizes, where one must give careful consider- Since electrodes in actual use can be of abnormal
ation to new electrode development like deep lengths, can have different deposition efficiencies, and
penetration electrodes and high deposition effi- since stub ends other than 50 mm length are common
ciency electrodes of the iron powder type. in shops, tables have been provided for (1) correction
2) Training welders in the use of newer electrode factor to account for deposition efficiency, (2) correc-
types: making their jobs easier and increasing tion factor to account for stub ends, and (3) correction
their efficiency by installing suitable positioners factor to account for abnormal electrode lengths.
and manipulators. The second section gives the following data on
3) Choosing the most economical form and dimen- the Ador Welding range of mild steel and low-alloy
sions of the joint. steel electrodes:
Table 15.3: Theoretical weight off metal in 1 metre long V-joint (gram)
Plate
thickness 50° 60° 70° 80° 90° 50° 60° 70° 80° 90° 50° 60° 70° 80° 90°
Root gap Root gap Root gap
(mm) 1 mm 1.5 m m 2mm
4 86 102 117 133 157 102 177 125 149 172 117 133 141 169 180
5 125 149 172 204 235 141 165 188 212 251 164 188 212 235 274
6 180 212 242 282 329 204 227 266 306 352 227 251 290 330 376
6.35 180 227 266 305 360 220 258 298 336 385 251 282 321 369 415
7 243 274 321 376 430 258 305 352 400 462 290 329 376 430 495
8 290 352 415 479 555 330 384 446 510 590 360 415 478 540 620
9 360 430 510 595 695 400 470 550 635 730 430 510 580 675 770
9.5 400 477 565 660 775 430 518 605 700 815 470 555 650 738 855
10 440 531 630 735 860 478 571 665 775 900 518 611 705 815 940
Root gap Root gap Root gap
1.5 mm 2mm 3m m
11 565 675 790 920 1,070 611 720 840 970 1,120 700 810 930 1,060 1,210
12 665 790 930 1,080 1,260 713 840 980 1,140 1,320 810 930 1,070 1,230 1,410
12.7 738 890 1,030 1,210 1,410 795 935 1,080 1,260 1,470 890 1,030 1,180 1,360 1,570
14 880 1,050 1,240 1,450 1,700 940 1,110 1,290 1,510 1,760 1,050 1,220 1,400 1,620 1,870
15 1,000 1,190 1,410 1,660 1,940 1,060 1,250 1,470 1,720 2,000 1,180 1,370 1,600 1,840 2,120
16 1,120 1,340 1,600 1,860 2,180 1,160 1,410 1,650 1,940 2,250 1,310 1,540 1,780 2,060 2,380
17 1,260 1,510 1,800 2,100 2,470 1,330 1,580 1,850 2,170 2,540 1,450 1,710 1,980 2,300 2,670
18 1,400 1,680 2,000 2,360 2,770 1,470 1,750 2,060 2,420 2,820 1,610 1,890 2,200 2,560 2,960
19 1,550 1,850 2,200 2,600 3,050 1,620 1,930 2,280 2,670 3,130 1,770 2,080 2,430 2,820 3,270
20 1,700 2,050 2,430 2,870 3,360 1,780 2,120 2,510 2,950 3,450 1,940 2,290 2,660 3,100 3,600
The theoretical weight of metal refers to tlie weld-metal necessary to fill up the theoretical cross-section of the joint. It does not take weld reinforcement and weld shrinkage into account.
Table 15.4: Theoretical weight of metal in 1 metr# long X-jckrt (gram)
Plate
thickness 50° 60° 70° 80° 90° 50° 60° 70° 80° 90° 50° 60° 70° 80° 90°
Root gap ]Root gap> Root gap
(mm) 1 mmi 1.5 mm 2mm
8 172 204 234 266 314 204 234 250 298 344 234 266 282 338 360
10 250 298 344 408 470 282 330 379 424 502 328 376 424 470 550
12 360 424 484 564 658 408 454 532 612 704 454 502 580 660 750
12.7 376 454 532 610 720 440 516 596 672 770 502 564 642 738 830
14 486 548 642 750 860 516 610 704 800 924 580 658 750 860 990
16 580 704 830 958 1,110 660 768 892 1,020 1,180 720 830 956 1,080 1,240
18 720 860 1,020 1,190 1,390 800 940 1,100 1,270 1,460 860 1,020 1,160 1,350 1,540
19 800 954 1,130 1,320 1,550 860 1,036 1,210 1,400 1,630 940 1,110 1,300 1,476 1,710
20 880 1,062 1,260 1,470 1,720 956 1,142 1,330 1,550 1,800 1,036 1,222 1,410 1,630 1,880
Root gap> Root gap Root gap
1.5 mm 2mm 3 mm
22 1.130 1.350 1.580 1.840 2.140 1.222 1.440 1.680 1.940 2.240 1.400 1.620 1.860 2.120 2.420
24 1,330 1,580 1,860 2,160 2,520 1,426 1,680 1,960 2,280 2,640 1,620 1,860 2,140 2,460 2,820
25.4 1,476 1,780 2,060 2,420 2,820 1,590 1,870 2,160 2,520 2,940 1,780 2,060 2,360 2,720 3,140
28 1,760 2,100 2,480 2,900 3,400 1,880 2,220 2,580 3,020 3,520 2,100 2,440 2,800 3,240 3,780
30 2,000 2,380 2,820 3,320 3,880 2,120 2,500 2,940 3,440 4,000 2,360 2,740 3,200 3,680 4,240
32 2,240 2,680 3,200 3,720 4,360 2,320 2,820 3,300 3,380 4,500 2,620 3,080 3,560 4,120 4,760 m
34 2,520 3,020 3,600 4,200 4,940 2,660 3,160 3,700 4,340 5,080 2,900 3,420 3,960 4,600 5,340 5
36 2,800 3,360 4,000 4,720 5,540 2,940 3,500 4„120 4,840 5,640 3,220 3,780 4,400 5,120 5,920
38 3,100 3,700 4,400 5,200 6,100 3,240 3,860 4,560 5,340 6,260 3,540 4,160 4,860 5,640 6,540 O
O
40 3,400 4,100 4,860 5,740 6,720 3,560 4,240 5,020 5,900 6,900 3,880 4,580 5,320 6,200 7,200
CO
The theoretical necessary tofillup the theoretical cross-section
weight of metal refers to the weld-metal of the joint. It does not take weld reinforcement and weld shrinkage into account.
>
z
o
a) Melting-off and deposition efficiency values and 1) Ratio of electrode consumption by weight to the
correction factors to account for deposition effi- total weight of welded fabrication based on ex-
ciency perience.
b) Average welding currents 2) Values of weld time factor based on experience
3) Theoretical weight of metal in a V-joint
c) Pure melting time per electrode
4) Theoretical weight of metal in an X-joint.
d) Weld-metal yield per electrode The last two are reproduced in Tables 15.3 and 15.4.
e) Electric power consumption per electrode They are somewhat similar to Table 15.2 and are equally
f) Weld-length coefficient: By this term is meant the useful for the estimation of electrodes.
ratio of length of weld-pass obtained with an The Handbook finally gives typical examples of
electrode to the length of the electrode. application of the data for estimating the quantity of
electrodes and the total welding costs.
The maximum weld-length coefficient characterises
the maximum speed of arc travel, at which the slag can Calculation of Welding Cost: Typical Indicative
still flow evenly with the weld-metal and which can Examples
still guarantee an acceptable and clean welded joint. 1) Manual electrodes
By multiplying the electrode length by the maximum For comparison, we have calculated the costs of 3.15
weld-length coefficient, one arrives at the maximum mm and 6.3 mm diameter of E 6013 class electrodes
obtainable length of weld of an electrode. and of the same sizes of E 7018 class electrodes. The
The minimum weld-length coefficient refers to the basic data of these four types required for cost
minimum speed of arc travel, at which one can still calculation are given in Table 15.5.
avoid the slag running ahead of the arc. This is also a We have assumed the arc time factor to be 25%, i.e.
measure of the possibility of weaving an electrode the arc is in action for 120 min in an eight-hour shift.
during welding. By multiplying the electrode length Welding labour cost per shift is taken as Rs. 250, which
with the minimum weld-length coefficient, one arrives is the average combined daily wage of the welder and
at the m i n i m u m obtainable length of weld of an his helper. The welding overhead cost is assumed to be
electrode. 300% of the labour cost. The cost figures are shown in
Numerically, the maximum weld-length coefficient Table 15.6.
lies in most cases between 0.50 and 2.5. The minimum The results show that it is economical to use the
weld-length coefficient lies in most cases between 0.10 largest size of electrode possible, and secondly, that high
and 0.50. deposition efficiency electrodes give lower overall
The m a x i m u m a n d m i n i m u m w e l d - l e n g t h welding costs though they appear more expensive in
coefficients depend not only on the type of electrode the price list.
and its diameter, but also on the welding position, type
of joint and the welding current. This must be borne in C0 2 welding wires
mind for comparing two types of electrodes. Here we shall take up the commonly used wires of
The third section contains tables for the stub-length 1.2 and 1.6 mm diameter. The 1.2 mm wire used at 275
and electrode length correction factors stated above. It amp gives a deposition rate of 4 kg/arc hour, while the
also contains the following tables: 1.6 mm wire used at 375 amp gives a deposition rate of
Table 15.5: Basic data of two electrode types

Electrode Price per Deposition Melting Wt. of weld
_ kg. incl. efficiency time per deposit per
Type Size excise (%) electrode kg. of electrode
mm (Rs.) (Mts.) (gms)
E6013 3.2 x 450 146.2 55.0 1.5 550
E 6013 6.3 x450 121.8 55.0 2.0 550
E 7018 3.2 x 450 133.4 60.0 1.5 600
E 7018 6.3 x 450 109.0 60.0 2.0 600
Table 15.6: Cost figures of two electrode types

Elecltrode Kg of Cost Wt. of Weld-metal cost per kg
electrodes of the weld-metal
Type Size consumed electrodes deposited Material Labour Overhead Total cost
(mm) in the shift (Rs.) (gm) cost (Rs.) cost (Rs.) cost (Rs.) (Rs.)
E6013 3.2 x 450 3.1 449.6 1691.8 265.7 147.8 443.3 856.8
E6013 6.3 x 450 10.6 1291.7 5832.8 221.5 42.9 128.6 392.9
E7018 3.2 x 450 3.2 420.2 1890.0 222.3 132.3 396.8 751.4
E7018 6.3 x 450 9.6 1046.8 5760.0 181.7 43.4 130.2 355.3
5.5 kg/arc hour. The average arc time factor for the per shift are assumed to be the same as for C 0 2 welding.
MIG/C0 2 semi-automatic process is 40%, which means The cost of flux has also to be included in the costing.
that in an eight hour shift, the arc is in action for 192
min. Hence the total weld deposit in the shift with 1.2 Economics Of Welding
mm wire is 12.8 kg, and with 1.6 mm wire it is 17.6 kg. The overall economics of a welded fabrication
In C 0 2 welding, 10% of the filler wire is lost in the form depends to a large extent on the efficient use of welding.
of spatter, which means that the deposition efficiency This demands proper care of welding consumables in
of the process is 90%. Making allowance for this and storage and use, and avoiding indiscriminate use of
assuming the welding labour cost of Rs. 300 per shift weld-metal, which is a very costly item as shown by
(Rs. 200 for the welder and Rs. 100 for the helper) and the above cost figures.
the welding overhead cost as 300% of the labour cost,
the total costs can be calculated as in Table 15.7. In this CARE OF CONSUMABLES
costing, the cost of carbon dioxide gas is not included. Fabricating shops and construction projects are often
compelled to maintain stocks of consumables for six
SA Flux:Wire Combination months to two years. Long and careless storage under
Here we take two typical wire sizes: 2.5 and 4 mm. humid conditions make them unusable or usable only
The 2.5 mm wire used at 400 amp gives a deposition after p r o p e r rebaking. One often comes across
Table 15.7: Cost figures for C02, welding

co2 Kg Price per Cost of the Wt. of Weld-metal cost per kg
wire dia of wire kg. incl. electrode weld
(mm) consumed excise deposited Material Labour Overhead Total
in the (gm) cost cost cost. cost
shift (Rs.) (Rs.) (Rs.) (Rs.) (Rs.) (Rs.)
1.2 10.0 69.6 696.0 9000.0 77.3 27.8 83.3 188.4
1.6 12.5 67.3 841.0 11250.0 74.8 22.2 66.7 163.6
The cost details are given in Table 15.8.
rate of 5.4 kg/arc hour, while the 4 mm wire gives a consumable supplies d a m a g e d beyond repair in
value of 10 kg/arc hour at 750 Amps. The average arc storage. Sometimes they are wrongly or inadequately
time factor for the fully-automatic SA process is 50%, baked at the user's end, resulting in their unsatisfactory
which means that the arc is in action for four hours in performance.
an eight-hour shift. Hence the total weld deposit in a C o n s u m a b l e p r o d u c e r s s u p p l y guidelines for
shift for 2.5 mm wire is 21.6 kg and for 4 mm wire it is storage, reconditioning and handling of their products.
40 kg. The deposition efficiency of the process is 100%, These m u s t be asked for a n d i m p l e m e n t e d .
which means that the material cost of the weld-metal is Consumables must be stored in a special room which is
the same as that of the wire. The consumption ratio of either kept heated to 10-15°C above the ambient
flux:wire is usually 1:1. The labour and overhead costs temperature or dehumidified. For redrying, electric
Table 15.8: Cost figures for SAW
SAW wire kg of kg of Price per kg Price per kg Cost of Wt. of weld Weld-metal cost per kg
dia wire flux of wire incl. of flux incl. consumable metal Matl. Labour Overhead Total
(mm) consumed consumed excise excise deposited cost cost cost cost
in a shift in a shift (Rs.) (Rs.) (Rs.) (Rs.) (Rs.) (Rs.) (Rs.)
2.4 13.0 15.0 84.7 62.6 2040.4 13000.0 157.0 19.2 575 233.9
4.0 20.0 22.0 78.9 62.6 2955.7 20000.0 147.8 12.5 37.5 197.8
ovens of s t a n d a r d design m u s t be u s e d a n d possible use of back-ups and deep penetration

m a n u f a c t u r e r ' s instructions correctly applied. technique). These have been elaborated in the
P r o c e d u r e s for storing and r e d r y i n g of w e l d i n g earlier chapters while describing welding proce-
consumables are explained in earlier chapters on dures.
welding processes. e) Specifying intermittent fillet welds in place of
A substantial wastage of manual electrodes occurs continuous weld under light-load or no-load con-
through stub loss. The discarded stub length should ditions (for example, stiffeners or diaphragms do
never exceed 50 mm. It is common to see stubs of 100 to not need much welding).
150 mm length thrown around on the shop floor. What
Even after a most economical design is m a d e
a colossal waste. This usually happens because the
available by the designer, the shop can undo it by
welder uses very high current (200 amp for 4 m m
inefficient or careless implementation through incorrect
diameter) to speed up his work, whereby the electrode
edge preparation, bad fit-up and overwelding. It must
gels prematurely overheated. To solve this problem the
be borne in mind that control of cost as well as quality
welder may be given a larger diameter electrode, say 5
is a mutual concern of both designer and production
mm size in place of 4 mm. Some welders bend the
manager. The designer must be aware of what goes on
electrode, before starting to weld. This wasteful habit
in the shop, and the production manager must pass on
must be discouraged.
his problems and experiences to the designer.
DESIGNER'S RESPONSIBILITY
AT SHOP LEVEL
The designer's role in ensuring economical welding
Everyone is aware that weld-metal is very expensive
has been discussed in detail in Chapter 17. The designer
and must be used judiciously. But few realise that more
can effect considerable saving in weld-metal by:
weld-metal means more distortion, and the best way of
a) Specifying minimum joint thicknesses through controlling distortion is to use least amount of weld-
use of higher tensile materials. metal.
b) Incorporating minimum number of joints in a Overwelding commonly occurs in fillet welds,
structure by using the following forming meth- because fillet weld sizes are left to the discretion of the
ods: welder. Fabricators are rarely conscious of the fact that
i) Press brake a small increase in fillet size results in an enormous
increase in the weight of weld-metal. Some idea of the
ii) Bending rolls
wastage of weld-metal incurred can be seen from the
iii) Roll-forming graph in Fig. 15.2. For example, increasing the weld size
iv) Tangent-bending and contour-bending from 6 mm to 8 mm means 84% extra weld-metal.
v) Flanging and dishing Fillet weld sizes should be carefully specified by the
designer and strictly controlled during fabrication by
vi) Press-die forming and drawing.
using fillet g a u g e s a n d s t a n d a r d i s e d w e l d i n g
c) Specifying correct size of weld through use of procedures.
standard codes (example: AWS structural code, In butt joints, the type of joint and mode of edge
AWS Bridge code, ASME pressure vessel code). preparation determine the amount of weld-metal and
d) Specifying most economical joint preparation and distortion. In single-sided preparation, the welding is
fit-up (type of joint, included angle, root gap and not balanced about the neutral axis of the plate, and
WELDING COSTS AND ECONOMICS 6 1 9
metal consumption and distortion to the minimum. The

narrow-gap welding procedures have been described
in Chapters 4, 5 and 6.
The weld sizes specified by the designer in the
drawing are expected to have a built-in safety factor.
Yet shop people and welders resort to overwelding
under the mistaken impression that a joint becomes
stronger. Overwelding increases cost as well as locked-
up stresses and distortion.
Overwelding occurs due to the following reasons:
1) Heavy reinforcement is provided on the butt weld
FUlet size, mm 2) Fillet weld size is unnecessarily increased
Fig. 15.2: Effect of using 8 mm fillet instead of 6 mm 3) Fillet weld has high convexity
fillet 4) Fillet weld has unequal leg length.
These have been illustrated in Fig. 15.4.
angular distortion increases with plate thickness. In In butt welds, the conventional practice is to fill the
plate thicknesses above 12 mm, double-sided V-groove from one side, gouge out the root from the
preparations are economical, because of the saving in second side and deposit one or more filling passes in
weld-metal. At thicknesses above approximately 40 mm, the root groove from the second side.
a double-U preparation may achieve sufficient saving Root gouging and deposition of weld-metal at the
in weld-metal to justify additional machining costs. root can be avoided by using permanent or temporary
Figure 15.3 shows the relationship between joint type/ backing bars to support the metal at the root. Temporary
size and the weight of weld-metal required to fill it. backings are usually made of copper. They may be
The latest trend is to use narrow-gap SA, TIG and water-cooled if a long heavy weld is being deposited.
MIG welding of heavy sections in order to reduce weld- They can have a groove to provide weld reinforcement
at the bottom.
Various backing techniques and procedures are
described in the earlier chapters on welding processes
in the sections of welding procedures.
S 3 US DC DQ
smgie-y singie-u double-v cbubie-u
The deep penetration technique has been explained
in some of the earlier chapters. In this technique, an
(A) Extra reinforcement

in butt weld
Over- Fitlet
convexity oversize
(B)
o.
in fillet
Useless extra
(C) metal due 7to
unequal teg
10 20 30 10 50 60 70 80 90 length
Thickness of material, mm
Fig. 15.3: Effect of edge preparation on

weight of weld-metal Fig. 15.4: Examples of overwelding
appreciable portion of the weld-metal in the joint is The use of positioning equipment which reduces
provided by the fused base metal. Thus, this is the overhead and vertical welding to downhand welding,
simplest and most effective method of saving weld- results in lower costs where the welds are of sufficient
metal. size and length and the work is of such a nature that it
Using this technique, a fillet weld is made in which can be positioned. Welding positioners are widely in
there is appreciable penetration beyond the root. The use in productivity-conscious shops, and they make
effective throat of the weld then equals the fillet throat significant i m p r o v e m e n t s in welding output and
plus the depth of penetration. The fillet size can thus quality in p r o d u c t i o n w e l d i n g . They have been
be reduced, resulting in considerable saving of weld- described in Chapter 20.
metal [see Fig. 15.5 (A)]. In vertical and overhead welding, there is always a
With this technique a butt weld is made with square tendency for the liquid metal to fall down by gravity
edge preparation and a small root gap, depositing one and though the slag characteristics of the coating may,
pass from each side. The technique not only saves weld- to a certain extent, control the tendency, it is the freezing
metal but also saves the cost of bevelling, root gouging of the metal against the base metal which prevents it
and deposition of a sealing pass [see Fig. 15.5 (B)]. from dripping. This condition p u t s a limit to the
Specially designed manual electrodes are available, amount of heat generated at the joint, and consequently,
which can butt weld joint thicknesses up to 12 mm by to the size of the electrode and the amount of molten
this technique. The submerged arc welding process is metal which can be deposited within a certain time. It
ideally suited for d e e p p e n e t r a t i o n w e l d i n g . has been demonstrated that in making a 9.5 mm fillet
Thicknesses up to 25 mm can be easily welded with weld, the welding speed is 0.140 min per 25 mm in the
single wire, while with twin-wire technique and partial flat position, 0.231 min per 25 mm in the horizontal
bevelling, a 38 mm thick joint can be completed in three position and 0.412 min per 25 m m in the overhead
passes [see Fig. 15.5 (C)]. position. It can be deduced from these values that if
the costs are taken as 100% in the flat position, they
rise to 165% in the horizontal and to 294% in the
overhead positions.
Measurement of Productivity
Shops using arc welding continuously in their
production p r o g r a m m e can measure arc welding
productivity in terms of weld deposition rate and
operator factor.
Weld deposition rate means the weight of weld-
metal deposited per arc hour, usually expressed in k g /
hour. Operator factor (O. R), as explained earlier, is part
of the total work time in which the arc is in action, and
it is usually expressed as a percentage. It is obvious
that to achieve maximum production, both these factors
must be maintained at the highest level possible.
In manual welding, the largest diameter of electrode
and the highest welding current permissible for the job
must be used. Iron powder electrodes having high
deposition efficiencies must be preferred to normal
electrodes.
O.F. for manual welding is comparatively low,
because of the inherent drawbacks of this process such
as changing of electrode once every few minutes and
Fig. 15.5: Deep penetration techniques: (A) x = fillet throat, the physical exertion involved. Yet it can be as high as
y = pen, beyond root, effective throat = x+ y; (B) Two- 60% and as low as 20% depending on the type of work,
pass weld in a square butt joint; (C) Three-pass weld in the speed with which the job is delivered to the welder,
partial bevelled butt joint with twin-wire SA process shop environment and whether production incentives
are given or not. O.F. well above 60% is attained by electrode (minus 40 mm stub) even when it is red hot,
experienced welders in making long continuous butt thereby causing bad welds. This eventuality can be
or fillet welds in the downhand position. An average controlled by checking the melting rate of the electrode
figure for m o d e r a t e l y h e a v y a n d large w o r k is on the welder's holder from time to time, to make sure
approximately 40% in the U.S.A. and should be 30% that it has not fallen below 100 sec or whatever value
under Indian tropical conditions. Positional welding was originally fixed. Alternately, the actual welding
will lower the O.F. because of the additional strain and current may be checked with a tong ammeter.
fatigue involved. Generally speaking, O.F. depends on In cold countries, manual welders can reach fairly
the type of structure, position of welding, the make-up high weld deposition rates by using 6 mm heavy-coated
of the crew, and the experience and stamina of the iron powder type electrodes at 400-500 amp. In tropical
welder. countries like India, on the other hand, a welder finds
In manual welding, there are several insignificant- it difficult to use more than 300 Amps, under production
looking factors which affect the O.F. in a given set-up. conditions. As the electrode size and welding amperage
For example, changing the electrode length from 450 increase beyond a certain limit, the O.F. percentage
to 350 mm lowers O.F. by anything between 3 and 5%, tends to decrease.
because of the decrease in the duration of arc time Considering these shortcomings of the MMAW
between interruptions and the consequent decrease of process, our shops should look to M I G / C 0 2 and
the total arc time in a given period. Poor quality submerged-arc processes for higher productivity. With
accessories, uncomfortable stance during welding, the M I G / C 0 2 process, O.F. of 40 to 50% can be easily
positional welding and bad joint fit-up also contribute achieved, while with fully-automatic sumberged-arc
to the fall in the O.F. percentage. w e l d i n g , O.F. of well over 70% can be usually
How can one control the O.F. of a large number of maintained, because it is commonly used for making
welders employed in a shop? Some of the firms in long, continuous welds in the downhand or horizontal-
foreign countries, especially shipyards, have installed vertical position without interruption.
special electrical devices which measure up the total In the M I G / C 0 2 process, the wire diameters are
arc time of every welder during his entire shift. Some generally on the low side (2.4 mm down to 1.2 mm)
of these devices may work on the following principle: but even for small diameter wires, very high currents
A current relay is kept at the transformer end in the can be used without the danger of overheating the wire,
secondary circuit. This relay has normally open contacts because the electrical contact to the wire is hardly 20
which are connected in series with an electric clock of mm away from the arc. This results in extremely high
the self-starting type. Each time the welder strikes the current density and a penetrating arc. The wire-feed
electrode, the current relay operates and closes contacts rates are also extremely high (up to 18 metres per min)
for the clock and records the time. At the end of the and this factor leads to very high deposition rates. Also,
day one can get total welding time directly from the because of the penetrating arc and the ability of the
clock reading. Such devices not only help the electrode to reach deep and narrow grooves, the butt
management to determine productivity, but also to joint preparation can be suitably modified, needing less
calculate the incentive bonus that may be paid to a weld-metal to complete the joint. In fillet welds also,
welder under an incentive system. deep penetrations are obtainable. Root penetration of
A simpler way of determining the O.F. from the at least 4 mm can be obtained at a welding speed of 50
melting rate of the electrode and the n u m b e r of c m / m i n compared to 1.5 mm penetration with a 6 mm
electrodes consumed in a shift was explained earlier. AWS E6024 class manual electrode used at a speed of
In this system, there is a loophole that the welder may 25 cm/min. This means that the visible fillet weld size
raise the welding current beyond the stipulate^ limit can be reduced as explained earlier.
and consume a larger quantity of electrodes in less total As already pointed out in Chapter 4, the fully-
time. In that event, his discarded stubs are likely to be automatic submerged-arc process is highly economical
longer than 40 mm, because of the electrodes getting and gives highest productivity. H u m a n factor is
prematurely red-hot. To prevent such an occurrence, eliminated. Wires up to 6 mm diameter can be used
the number and lengths of stubs discarded by the with currents up to 1,200 amp or even more. Because
welder at the end of the shift must be checked. There is of the very deep penetration given by the process, butt
also the possibility that the welder uses abnormally welds up to 18.5 mm thickness can be made with square
high currents and at the same time consumes the entire edges and without any edge preparation with only two
passes, one on either side. Partial bevelling, narrower It must be remembered that the high investment cost
grooves and wider root faces can be used for heavier of the SAW equipment and the comparatively higher
plates, resulting in least consumption of weld-metal and prices of consumables are justified only if O.F. of at least
least welding time. Here also, the joint can be completed 50% is maintained. It has been estimated in the case of
in two passes. Further economies achieved with a 12.5 mm fillet weld that by reducing the O.F. of SAW
multiple wire SAW installations h a v e also been from 50% to 12%, the cost of making the weld becomes
highlighted in Chapter 4. For getting the best out of as high as with manual welding.
this process, copper or flux backing device must be used Table 15.9 gives weld deposition rates for various
wherever possible, to eliminate the need for turning arc welding processes worked out by a British authority.
over and welding on the second side, and jigs, fixtures The values of deposition rates have been corrected in
and manipulators which are properly integrated with the last vertical column for assumed operator factors of
the welding equipment and the workpiece must be 33% for manual metal-arc welding, 40% for semi-
devised. automatic welding and 50% for fully-automatic welding.
Table 15.9: Deposition rates with various welding processes
Process Electrode Current Amps. , Probable

diameter amp deposition production
in (mm) rates deposition rates
kg/hour kg/hour
Manual welding
Mild steel general purpose rutile electrode 10 600 5.4 1.8
Mild steel general purpose rutile electrode 6 315 3 1.0
Rutile iron-powder electrode 6 400 5 1.66
Low-hydrogen rutile iron powder electrode 6 600 8 2.7
Semi-automatic welding
Gas-shielding bare-wire consumable electrode 2.5 450 6.8 2.7
Carbon dioxide shielded bare wire with magnetic flux 2.5 435 6 2.4
Carbon dioxide shielded flux-cored tubular electrode 2.5 425 6.3-9 2.5-3.6
Automatic welding
Gas-shielded bare wire 2.5 450 6.8 3.4
Submerged-arc (single-wire) 6 1,200 15 7.5
Parallel submerged-arc (2-wired): each 5 1,800 (total) 22.2 11.3
Submerged-arc (single-wire) 8 2,000 27.2 13.6
Multi-power submerged-arc (two wire and two
power sources): each 6 3,000 (total) 40.8 20.5
Continuous flux-covered electrode 8 1,000 10.4 5.2
Carbon-dioxide shielded continuous flux-covered electrode; 7 1,200 16 8
Ador Welding Welding Handbook.

The Procedure Handbook of Arc Welding, Sect. 12, by Lincoln Electric Co., U.S.A.
Safety Requirements
in Arc Cutting
and Welding
ALL METAL cutting and welding operations which in- of tools. The primary side of the welding machine
volve the use of the electric arc as an intense source of should be provided with suitable wire terminals inside
heat can harm the operator's eyes and skin, and are the machine case, for the connection of conductors of
potential health and fire hazards. The other potential at least the minimum size required, corresponding to
sources of fire in these operations are hot metal and its the rating of the welding machine. Output (welding)
flying particles, and stray arcs due to short-circuiting. terminals must be such that no current-carrying parts
Further, electric welding and cutting processes expose can make accidental contact. A disconnecting switch
the welder to the danger of electric shock, and release must be provided at or near the machine, if such a
fumes, gases and dusts which can be potential health switch is not mounted as an integral part of the welding
hazards. It is imperative, therefore, that the shop man- machine. All terminal connections, tap connections and
agement be fully aware of such hazards and adopt suit- control must be clearly marked to designate their
able measures for the safety and protection of person- purpose and correct usage.
nel and property. Some of these measures are manda- In the case of a motor generator set, a suitable motor
tory according to code requirements. starter of the hand or magnetic type, having an ampere
In purchasing the equipment, as much consideration rating equal to the ampere rating of the motor, must be
should be given to its safe operation as to cost, used. It must be mounted integrally with the motor
availability and other factors. Standards have been generator and connected to operate the motor, and
developed for most of the arc welding machines, and should be capable of interrupting stalled rotor current
they cover established practice in ratings, basis of of the motor. The starter must have overload and low-
ratings, m e t h o d s of test and general m e t h o d s of voltage protection incorporated. All parts of the motor
construction, which ensure safety during installation and generator equipment must be suitably enclosed and
and operation. Safety is assured by choosing equipment protected to meet the normal service conditions.
complying with the relevant standard. Neither terminal of the welding generator must be
bonded to the frame of the welding machine.
Construction
An AC arc w e l d i n g machine m u s t be a fully Type of machine Voltage max. V
assembled, compactly mounted unit of the air-cooled a) Single-operator, DC welding machines 100
or oil-cooled type. All its control apparatus must be
enclosed except for operating wheels, levers or handles. b) Constant-voltage, multiple-operator,
Any arrangement for changing primary taps must be DC welding machines 75
enclosed with a cover, which is removable only with c) AC welding machines 100
tools. Panels or mountings for current-carrying parts
must be of suitable insulating material. Input terminals Open-circuit voltage (no-load voltage) of the arc
must be totally enclosed and accessible only by means w e l d i n g m a c h i n e s should not exceed the above
mentioned limits to avoid the danger of electric shock the electrode coverings, or any metal part of the
to the operator: electrode holder to touch either his bare skin or any
To avoid possible electric shock to welders working wet coverings on his body. Consistent use of electrode
on wet steel or in confined quarters where they may holders with well-insulated jaws, well-insulated cables,
sweat profusely, it is possible in the case of transformer- dry protective coverings on the hands and body and
type welding machines to provide reliable control insulations from g r o u n d are helpful in avoiding
equipment to automatically reduce the no-load voltage. contact. It should never be assumed that because
contact with the electrode at one time is not harmful,
Installation similar contacts at other times would also be harmless.
Installation including power supply must conform b) Electrodes must be removed from holders when
to the relevant Electricity Act. The frame or case of the not in use to eliminate danger of electrical contact with
welding machine must be efficiently earthed. For persons or conducting objects. Electrode holders when
i n d i v i d u a l w e l d i n g machines, the rated current- not in use should be so placed that possibilities of
carrying capacity of the supply conductors should be electrical contact b e t w e e n t h e m and persons or
not less than the maximum primary current of the conducting objects are eliminated.
welding machines. Welding current should preferably c) Welders and welding supervisors must check their
be returned to the welding machine by a single cable equipment regularly to see that electrical connections
from the work to the welding machine. Connection of and insulation on the holders and cable are in good
a cable from the w e l d i n g machine to a c o m m o n order. Anything that appears unsafe must be promptly
conductor or structure on which the work rests, or to reported to the proper authority, and use of such
which the work is connected is also permitted. Pipelines equipment must be discontinued until its safety has been
containing gases or inflammable liquids, or conduits assured. Repairs must be made only by qualified
carrying electrical conductors should not be used for a personnel.
ground return circuit. Chain or wire rope should not
be used to carry welding current. All earth connections Maintenance
m u s t be checked to m a k e sure that they are All e q u i p m e n t m u s t be subjected to periodic
mechanically strong and electrically adequate for the inspection and maintained in safe working order at all
required current. times. Instructions for the operation and maintenance
Before commencing operation, all connections must of equipment supplied by the manufacturer must be
be checked to make sure that they are correctly made. strictly followed. When the welding operation is
The work lead must not be loosely placed on the work stopped for an appreciable time, or when the machine
but must be firmly clamped to it. Work leads should be is to be shifted, the power supply switch of the machine
as short as possible. The welding machine frame must must be kept open. The machine should be disconnected
be grounded. from the source of power when not in use. When used
in the open, the equipment must be protected from
Operation inclement weather conditions. When not in use over a
To avoid electric shock, the following precautions period, it must be stored in a clean dry place.
must be taken:
a) Welders m u s t be explained the m e t h o d of Welding Accessories
avoiding electric shock. Voltages required for arc- All accessories used by the welder are meant to
welding are low and normally would not cause injury ensure his safety and guard his health. They must,
or severe shock. Hence, parts having these voltages therefore, meet certain minimum quality requirements
are liable to be handled carelessly. These voltages are, as laid down in IS: 2641-1989, "Specification for electrical
nevertheless, sufficiently high and u n d e r certain welding accessories" or similar standards.
circumstances may be dangerous to life. This danger Electrode holder: All parts passing through the portion
is particularly marked in very hot weather, when the of the holder which the welder grips in his hand must
welder is sweaty or when he is wet. The welder must be fully insulated against the m a x i m u m voltage
develop the habit of always keeping his body insulated encountered to ground. Where service conditions
from both the work and the metal electrode and holder. demand, even the jaws which grip the electrodes must
He must always wear shoes and gloves. Particularly, be insulated. Holders must be sufficiently robust to
he must never permit the metal part of an electrode, withstand the maximum welding current at maximum
SAFETY REQUIREMENTS IN ARC CUTTING AND WELDING 625
duty cycle, without getting excessively overheated. The be fixed such that the arc is not visible through cracks
dipping in w a t e r o f overheated holders, whether and chinks r o u n d the edges. Helmets should be
connected or not, must never be permitted as the preferred to head shields for overhead welding and in
retained moisture within the holder may cause an conditions where both hands of the welder must be free.
electric shock. Goggles: They are meant (1) to protect the eyes of
Cables and connectors: Welding cables must conform persons, other than the welder, from the rays of an
to standard requirements as specified, for example, in electric arc, and (2) to protect the eyes of persons from
IS:434 (Part 1)-1964. They must be completely insulated, small flying pieces of slag while chipping it from the
flexible and capable of handling the maximum current weld. Goggles to protect the eyes from the injurious
used on the job, taking into account the duty cycle under rays must always be fitted with opaque side shields.
which the welder is working. Minimum cable sizes The glasses of the goggles should be of a suitable grade.
required to carry various m a x i m u m currents are (Refer table on the next page). Alternately, the goggles
specified in the above standard. The cable must be free can be fitted with Crookes spectacle glass. However,
from repair or splices up to a minimum distance of 3 m goggles fitted with this glass are suitable only for
from the electrode holder. protection w h e n the arc is not in the immediate
When it becomes necessary to connect or splice neighbourhood. Glasses of greenish neutral colour are
lengths of a cable, substantial insulated connectors of a considered to afford the maximum protection for the
capacity at least equivalent to that of the cable, must be eyes. Goggles to protect the eyes from small flying
used. If connections are effected by means of cable lugs, pieces or slag must be fitted with clean non-splinterable
then these should be securely fastened together by more glass, if there is no need to protect the wearer from rays
than one bolt to give good electrical contact. The from the arc.
exposed metal parts of the lugs must be completely Screens: All electric welding operations must be
insulated. screened to prevent the rays of the arc from affecting
Head shield: These are meant to protect the eyes and other persons working in the vicinity. Where the work
face of the welder from the heat and injurious rays is done at fixed benches or in welding shops, permanent
(infrared, visible and ultraviolet) generated by the screens should be erected, as far as practicable. Where
electric arc. They should, therefore, be constructed of this is not possible, temporary screens must be used to
heat-resisting, non-ignitable material, which is also limit the radiation. All screens must be opaque, of sturdy
impervious to the harmful rays. They should be light construction to withstand rough usage, and of material
for convenience in use, and strong to withstand rough which will not readily catch fire by sparks or hot metal.
service. The minimum amount of metal must be used Protective clothing: The main items u n d e r this
in their construction, particularly on the outside of the category are leather or asbestos gloves and flame-proof
shield. For example, there should be no metal frame aprons made of leather or asbestos. For heavy work,
for the glass either on the outside or on the inside; metal fire-resistant leggings or high boots are used. In
rivets should not be used unless one end is covered by production work, a sheet-metal screen may be placed
an insulating material, as these may be the means of in front of the w e l d e r ' s legs to p r o v i d e further
causing electric shock. protection against sparks, and against molten metal in
Head shields should be fitted with an adjustable cutting operations. During overhead welding, caps or
band to fit the wearer's head, and this band must be shoulder covers made of leather should be worn. If
made from an insulating material. The sides of the helmets are being used, leather skull caps may be worn
shield must be sufficiently large, so as to give protection under them.
to the neck and side of the head. This is especially Now a days head-shields are also made available
important, where several welders weld near each other. with the capability of autofiltering the harmful Infrared
The handles must be made of a material which is a bad & Ultraviolet rays. They are coimbined with the solar
conductor of both heat and electricity. and LCD technology. That is the most advanced and
The window through which the operator views the comfortable protection device available to work in the
arc must be of the required shade No. The frame should most critical areas. The shade number is adjusted on
be such as to take protective glass with a piece of plain continuous basis from shade No. 9-to-13 within the
glass fixed in front, facing the arc so that spatter from range of plus or minus one. An Optrel Solarmatic
the arc is intercepted by the plain glass, which may be Comfort is the popular brand in this advanced category
renewed from time to time. The protecting glass must of head-shields.
Welding Process Current Range Recommended

shade No.
S.M.A.W. / M.M.A.W. Up to 160 Amps. 10
// 160 to 250 " 11
250 to 500 " 12
G.M.A.W. / F.C.A.W. Up to 160 Amps. 11
n 160 to 250 " 12
n 250 to 500 " 14
G.T.A.W Up to 50 Amps. 10
n 50 to 100 " 12
II 150 to 500 " 14
Up to 500 Amps. 12
C.A.G. 500 to 1000 " 14
Light duty works 4
O.A.W. Med. " " 5
II
Heavy " " 8
O.A.C. Light " " 4
Med. " " 5
Heavy " " 6
Note: As a thumb rule, selection offilter glass begins with a shade No. that is too dark to see the weld. These values apply until weld
puddle is clearly seen to the welder.
Welding in Confined Spaces must be removed from the holders and the holders
Quite often a welder is called upon to cut or weld carefully located, so that there is no possibility of
inside an enclosed container with one or two small accidental contact. The machine should also be
openings. Since this operation is hazardous, special disconnected from the power source.
precautions must be taken to ensure the welder's safety.
Firstly, adequate ventilation must be provided and HEALTH PROTECTION
arrangement must be made to exhaust the fumes and Welding is considered a safe occupation, because the
gases generated in the arc. In C0 2 -welding, considerable incidence of occupational illnesses and diseases among
amount of carbon dioxide can accumulate and being welders, as proved by clinical evidence, is not higher
heavy, it may not permit the ingress of fresh air unless than among other industrial workers. Yet, it must be
the C 0 2 is exhausted. The gas cylinders and welding admitted that welders are exposed to potential hygienic
machines must be left on the outside. Where a welder hazards associated with welding fumes and gases when
has to enter a confined space through a manhole or other they work in ill-ventilated areas over a prolonged
small opening, means must be provided for quickly period.
removing him in case of emergency. When safety belts Shop management must be aware of such hazards,
and life lines are used for this purpose, they should be and must provide safe and healthy working conditions
so attached to the welder's body that he is not jammed to their welders. The management must continuously
in a small exit opening. educate the welders as well as the supervisors in the
A welder is likely to sweat profusely inside the use of proper ventilation procedures, the use of safety
container. In this condition, he becomes more susceptible e q u i p m e n t , a n d the observance of safe w o r k i n g
to electric shock. The welder must therefore use fully practices. In addition, the management must ensure
insulated holders, and well-insulated gloves even while t h a t their i n s t r u c t i o n s are b e i n g s c r u p u l o u s l y
inserting the electrode into the holder. When arc welding followed.
is to be suspended for any substantial period of time, Constant inhalation of certain constituents of
such as during lunch hours or overnight, all electrodes welding fume and gas over a period can be eventually
harmful to the lungs and other organs of the human depending on the extent of exposure and the nature of
body. The ultimate toxicity of a constituent depends the beryllium compound involved. The condition is
primarily on its concentration and the physiological termed berylliosis.
response of the human body.
The constituents of welding fumes and gases are Pulmonary Irritants and Toxic Inhalants
either particulates or gases. Particulates get deposited This covers a group of particulates contained in the
in the lungs; this condition is termed pneumoconioses. welding fume, which do not produce radiographic
Pneumoconiosis can be harmless or harmful depending changes in the lungs, but are pulmonary (i.e. pertaining
on whether the particular constituent is non-toxic or to the lungs) irritants or toxic inhalants. Their examples
toxic. The harmful ones can be fibrotic or non-fibrotic. are given below:
Examples of harmless pneumoconioses are: 1) Cadmium: Inhalation of cadmium oxide during the
1) Iron oxide: The deposition of iron oxide in the lungs soldering, brazing or welding of cadmium-plated or
leads to a condition known as siderosis. Siderosis does cadmium-containing parts can cause respiratory irrita-
not cause disability. But siderosis can get complicated tion with attendant sore, dry throat and a metallic taste
by the inhalation of silica dusts to produce silicosis. The followed by cough, chest pain, and difficulty in breath-
latter is a serious respiratory disease. The differential ing. Bronchitis, pneumonitis, and pulmonary oedema
diagnosis between siderosis and silicosis is often diffi- may occur as a result of irritative action of fume. Addi-
cult, because they may occur together. tional complaints of headache, dizziness, loss of appe-
2) Tin-oxide: The inhalation of tin-oxide dust over a tite, and weight loss may be pronounced. The liver, kid-
long period causes stannosis, which is considered to be neys, and bone marrow may be injured by the presence
non-progressive and non-disabling. of the metal.
3) Aluminium: The constant inhalation of aluminium, 2) Chromium: Welding of certain chromium alloys can
its oxide and hydrate is considered harmless. In rare produce chromium trioxide fume, which is often re-
cases, complication can arise from the presence of some ferred to as chromic acid. These fumes react with water
co-inhalant such as silica. vapour to form chromic and dichromic acid. Contact
Examples of harmful pneumoconioses of the fibrotic with these fumes can produce small, painless, cutane-
type are: ous ulcers as well as dermatitis from primary irritation
1) Silicon dioxide: As stated earlier, silicosis is caused or allergic hypersensitivity. Inhalation of these fumes
by the inhalation of finely divided silicon dioxide in the will produce bronchospasm, oedema and hypersecre-
free state, which may be in a crystalline form such as tion, bronchitis, and a hyperreaction of the traches bron-
quartz. The size of the offending silica particles is chial tree similar to asthma.
important, because it determines the depth to which 3) Lead: Inhalation of lead fumes causes lead poison-
these particles penetrate into the lung and in what ing, whose signs and symptoms may include abdomi-
amounts they may be retained there. Silicosis may be nal pain with tenderness, constipation, headache, weak-
either acute or chronic. Most cases are chronic and take ness, muscular aches or cramps, loss of appetite, nau-
years to develop. sea, vomiting, weight loss, anaemia with pallor, and a
2) Asbestos: Deposition of asbestos in the lung is lead line of the gingival margin.
termed asbestosis. Prolonged inhalation of asbestos 4) Magnesium: The oxide fumes from magnesium can
fibres between 20 and 50 jam long may lead to a typical produce metal fume fever, which may result in irrita-
pulmonary fibrosis, which may be accompanied by tion to mucous membranes.
severe respiratory disability. 5) Manganese: The fumes from manganese are highly
3) Copper: Inhalation of copper fumes is known to toxic and they can produce total disablement even after
produce fibrosis in animal studies. In welders, it causes exposures as short as a few months to high fume con-
symptoms of metal fume fever. In addition, it can centrations, but disablement is more likely after pro-
produce congestion of the nasal mucous membranes longed and repeated exposures above 30 mg/m 3 . Such
and other complications. The condition is referred to as exposure is usually caused by inhalation of manganese
copperosis. dioxide dust.
An Example of harmful pneumoconioses of the non- 6) Mercury: The welding of metals coated with
fibrotic type is: protective materials containing mercury compounds
Beryllium: Inhalation of beryllium dust or fume can will produce mercury vapours. Exposure to these
lead to an acute or chronic systematic disease, vapours may produce abdominal pain, vomiting,
diarrhea, gingivitis, p n e u m o n i t i s , renal (kidney) 3) Emphysema (loss of elasticity of the lung)

damage, and circulatory or respiratory failure. 4) Chronic bronchitis
7) Molybdenum: Little is known about the effect of
5) Asphyxiation.
Mo fumes on human beings, but animal studies have
indicated a low order of toxicity. The common toxic gases associated with welding are
8) Nickel: Nickel and its compounds are known to carbon m o n o x i d e , ozone a n d nitrogen dioxide.
be toxic and to induce cancer of the lungs and sinuses. Phosgene and phosphine which are produced by the
9) Titanium: Ti0 2 dust in high concentrations can pro- thermal decomposition of chlorinated hydrocarbon
duce irritation of the respiratory tract. In a few cases, cleaning agents a n d p h o s p h a t e metal coatings
slight fibrosis has been observed in human lungs after respectively, are also toxic. Carbon dioxide is harmless
long industrial exposure to this constituent, but with- but long exposure to high concentrations of this gas can
out any disabling injury. produce serious effects. The effects of these gases are
10) Vanadium: When Vanadium is present in the described below in detail.
welding wire, the welder is exposed to Vanadium pen- Carbon monoxide: In some welding processes, carbon
toxide, which is moderately hazardous. Effects of ex- dioxide is reduced to carbon monoxide. In the case of
posure are chiefly severe irritation of the eyes, throat, carbon dioxide-shielded metal-arc welding, carbon
and respiratory tract resulting in conjunctivitis, nasal monoxide concentrations exceeding recommended
catarrh, irritation of throat (pharyngitis), bronchopneu- levels have been detected in the fumes near the arc;
monia, and mild chronic bronchitis. however, the concentration decreased rapidly with
11) Zinc: Zinc oxide fumes are formed during the distance from the arc. With adequate ventilation, the
welding, brazing, or cutting of galvanised metals. The carbon m o n o x i d e concentration in the w e l d e r ' s
inhalation of freshly formed fumes may produce a brief, breathing zone can be maintained at harmless levels.
self-limiting illness known variously as zinc chills, Because of its great affinity for haemoglobin, the
metal fume fever, brass chills, and brass founder's fe- concentration of carbon monoxide should be kept at a
ver. This condition is characterised by chills, fever, nau- low level.
sea, vomiting, muscular pain, dryness of mouth and Carbon dioxide: C 0 2 is not usually considered a toxic
throat, headache, fatigue, and weakness. There may gas. It is present in the atmosphere in a concentration
also be a slight leukocytosis. These signs and symp- of about 300 p p m and it is present in somewhat higher
toms usually abate in 12 to 24 hours with complete re- concentrations in occupied structures. The air in the
covery following. lungs contains about 5.5% of C0 2 . If the air one breathes
12) Fluorides: During welding with basic low-hydro- contains C0 2 , the concentration of carbon dioxide in
gen electrodes, the welder inhales fluoride fumes. Pro- the blood increases, and the individual is induced to
longed exposure to fluoride fumes can produce respi- breathe faster in an attempt to lower the concentration.
ratory track irritation manifested by chills, fever, dysp- H e a d a c h e s are s o m e t i m e s o b s e r v e d by p e r s o n s
nea, and cough. The particles are readily absorbed, and exposed to concentrations of carbon dioxide greater
promptly produce an increase in the urinary fluoride than 1%. After inhalation of 4 to 5% carbon dioxide,
excretion. When excessive amounts are inhaled, this the ability to hold o n e ' s breath is reduced quite
excretion lags behind the daily intake resulting in a significantly.
buildup of fluoride in the bones. If storage of fluorides Since C 0 2 is heavier than air, it tends to accumulate
continues over a sufficiently long period, the bones may in high concentrations in the lowermost part of an
show an increased radiographic density, and structural enclosed chamber or walled area. Because of this
abnormalities may eventually develop. tendency, care should be taken during the welding
within such structures to ensure that carbon dioxide
Effects of Toxic Gases does not dilute the air to the extent that it can act as an
Prolonged exposure to the various toxic gases asphyxiant.
generated during welding can produce one or more of
Ozone: Ozone is formed by electrical arcs and corona
the following effects:
discharges in air or by ultraviolet photochemical
1) Inflammation of the lung (chemical pneumoni-
reactions. The inhalation of 1 to 2 p p m of ozone for two
tis)
hours produces headache, pain in the chest, and dryness
2) Pulmonary oedema (swelling and accumulation of the upper respiratory track. Welders who have had a
of fluids) severe acute exposure at an estimated 9 ppm of ozone
plus other air pollutants developed pulmonary oedema silicon and manganese. They also contain an
(i.e. presence of fluid in the lungs). appreciable concentration of fluorides.
Nitrogen dioxide: This gas is very irritating to the eyes When galvanised sheets are welded, the resulting
and mucous membranes. Exposure to high concen- fumes have been found to contain up to 65% ZnO, and
trations may produce immediate coughing and chest lesser amounts of Fe203 Ti02, Si02 and Mn0 2 .
pain. A study has shown that the amount of nitrogen
Phosgene: Inhalation of this gas will produce contact dioxide evolved per electrode varies between 15 and
dermatitis from primary irritation. Inhalation in high 980 mg. The data indicate that the N0 2 content increases
concentrations will produce pulmonary oedema. with increasing electrode size and current. The volume
Phosphine: This gas in high concentration is irritating of carbon monoxide produced is quite low.
to the eyes, nose and skin. The acute effects of Such studies of workshop fume samples have
phosphine can be serious and they include serious shown that the following constituents are sometime
damage to the kidneys and other organs. found to be in excess of the minimum permissible (i.e.
they are above the threshold limit value or TLV) and
FUME GENERATION IN ARC WELDING therefore present health hazards to the welders:
PROCESSES Zinc, iron, copper, vanadium, fluoride, nitrogen
dioxide, sulphur dioxide (caused by oil contamination),
Manual Metal-Arc Welding ozone, manganese, chromium, lead.
In a particular study of mild steel welding with Basic coatings: Fume studies on basic electrode
various types of mild steel electrodes, the quantity of coatings have shown that a coating having a high
fumes generated and their compositions were roughly percentage of CaF2 can sometimes give a lower
as indicated in Table 16.1. proportion of fluorides in the fume than a coating
In another case study, 1,700 welding fume samples having a lower percentage of CaF2. This is because the
were collected from various locations in shipyards. concentration of fluorides is also influenced by the Ca2/
Their analyses indicated that 50% of the total fume CaC0 3 ratio, melting temperature of the coating and
content by weight was Fe203, 15% was Ti02 and the the arc current.
remainder was a mixture of acid-soluble metals such The most likely fluoride compounds formed during
as Mg, Ca, Al, Mn, Cr, Cu and Na. In yet another study welding with basic electrodes are the fluorides of
of fumes evolved from E6010, E6011, E6012, E6013, sodium, potassium and silicon. A Japanese study has
E6020 and E6030 electrodes, it was found that the indicated that the combination of potassium and
average fume sample contained 60-70% by weight fluorine increases the hazards presented by basic
of Fe203 and Ti02,10-20% Si02 and 2-12% Mn0 2 . electrodes. The principal source of potassium is the
The fumes evolved during the welding of low-alloy potassium silicate used as a binder. The Japanese team
and high-alloy steels have been found to contain developed an improved electrode using sodium silicate
measurable amounts of nickel oxide, chromium oxide, as a binder. The fume analysis of this electrode showed
and copper oxide in addition to the usual oxides of iron, that it produced lesser amounts of soluble fluorides.
Table 16.1:: Compositions of fumes from SMAW electrodes

Qty. of fumes
Electrode g/electrode Fume composition, wt%
Type size FeA Si0 2 Ti0 2 MnO CaO F
mm
General purpose, neutral 4 0.5-1.5 40-60 15-35 trace 12-16 trace —
General purpose, neutral 10 4.0-7.0 28 29 -1 2.9 trace —
High-deposition, neutral 4 1.5-3.0 42 33 0.3 10.5 trace —
Fully mineral rutile type 4 0.5 70 8-10 2.5 6.5 trace —
Cellulosic rutile type 4 0.5-1.5 54 20 3.5 6.7 trace —
Basic lime-fluoride 4 1.0-2.5 22-28 6-12 0.6 4.8-10.2 14-20 17-28
Basic lime-fluoride 5 1.5-3.0 — — trace 4.2 19 23
Basic lime-fluoride 6 2.0-4.0 35 11 0.8 4.6 19
GTAW and GMAW Welding volume of 55 m3, the concentration of CO reached a

The gases evolved in GTAW welding are ozone and dangerous level after 10 min of welding.
nitrogen oxide. Tests conducted during the welding of In C 0 2 welding, there is a more or less direct
titanium have shown that considerable quantities of relationship between welding current and the amount of
these two gases are produced. When thoriated tungsten fumes produced. Total amount of fumes produced with
electrodes are used, thorium evaporates and becomes different sizes of wires and different currents and their
air-borne. The radioactive fumes of thorium can present analyses are given in Table 16.2.
a hazard. But investigations have shown that thorium In C 0 2 welding, the concentration of CO is far higher
concentration is well below the accepted limit at a than the concentrations of ozone and nitrogen oxide;
distance of 300 mm from the arc. In GTAW welding, the but the amount of CO has been found to decrease with
quantity of fume evolved is low. increasing welding current and increasing filler wire
Comparative studies of GTAW and GMAW welding size. When C 0 2 welding is carried out in an enclosed
of Al-Mg alloys have shown that the total quantity of area, there is an overall increase in the fume and gas
fumes evolved with GMAW was 17 times the quantity levels in concentration. Also, surprisingly, there is nearly
evolved w i t h GTAW; also the q u a n t i t y of ozone five-fold increase in the amount of ozone in such areas.
produced was almost four times. Nitrogen dioxide was This increase is attributed to the repeated reflections of
in traces in both the cases. ultraviolet radiation in enclosed areas. The following
The fumes evolved during the GMAW welding data collected during studies in open sections of the
of Al alloys consist largely of A1203. When Al-5% Si filler ship structure and in semi-enclosed areas bear out these
wire is used, the fumes contain up to 12% Si0 2 and the facts:
balance is A1203. All these findings emphasise the need for more
The concentrations of nitrogen oxide produced are thorough ventilation when welding is done with the
found to be generally higher when GMAW welding M I G / C 0 2 process than with the manual metal-arc
aluminium, than when GMAW welding stainless steel. process.
The fume studies conducted on GMAW welding of
aluminium is shipyards have shown that concentrations Flux-Cored Arc Welding
of nitrogen oxide and ozone can be dangerously high Flux-cored continuous electrodes generate the
under unventilated conditions. The concentrations are highest levels of fumes. This is clearly shown in Figs.
considerably lowered when good ventilation is used. 16.1 and 16.2, in which the fume generation rates of
When C 0 2 is used as a shielding gas, the welder is manual electrodes, GMAW with C 0 2 wires and flux-
exposed to the hazard of carbon monoxide. Studies have cored wires are compared.
shown that concentrations of CO in excess of the The significant constituents of the fumes given off
recommended value can be found as far as 0.6 m from by flux-cored wires are Fe 2 0 3 , fluorides, and nitrogen
the welding arc. Even in a ventilated room with a dioxide. A comprehensive investigation carried out in
Table 16.2: Fume compositions in GMAW process by co2

Filler Arc Arc Fume concentration Fume analy;sis, wt%
wire current, voltage, g / k g of g/min Fe Mn Si0 2 (total)
dia., mm Amps. V wire
1.0 190 22 4.62 0.23 56.7 13.7 7.6
1.2 315 29 9.30 0.84 59.7 12.9 10.0
2.0 415 34 13.50 1.62 63.6 11.1 8.8
Location Fume and gas; concentrations

Fumes co2 N02 03 co2
Mg/m3 mg/m 3 mg/m3 mg/m3 vol %
Compartments 20.0-55.0 20.0-90.0 1.0-3.0 0.0-0.03 0.14-0.47
Semi-enclosed areas 40.0-90.0 80.0-140.0 2.0-4.0 0.40-0.60 0.30-0.70
the U.S.A. showed that the amount of Fe 2 0 3 produced

by E 70T-4 class electrode was greater than that of E
3.0
7018 class electrode, while the amount of Fe 2 0 3 given
• off by E 70T-2 class was less than that of E 7018 class
• electrode. The average nitrogen oxide concentrations
25 ~ • Data points were of the same order of m a g n i t u d e for all the
• electrodes evaluated. The fluoride level was the highest
c for E 70T-4 class; the fluoride level for E 70T-2 was lower
e
~a> 2.Q L- —* than that for E 70T-4, but higher than that for the E 7018
V
(9 class electrode.
ai
b >K
4J t t 1 Submerged-Arc Welding
8 ~ ft «J
(9
The health hazards produced by SAW welding
4>
8 8 O«9 ©
o o O
1 ID
W W ♦S
are similar to those arising from other arc processes.
tt. £ £ 'a. -^ o © o o
**»
Though the arc is submerged and invisible, the very
L I S f t & fc UJ
UJ
high currents used in this process can develop some
<
2 2 «? ~
psV fumes. When internal welding of pipes and vessels is
2 2 as ^ being carried out, the operator in the vicinity of the arc
05 m m o o is likely to be exposed to dangerous levels of contami-
•s r*i
* • ♦
nants in the atmosphere. The welding area must there-

V«
fore be well ventilated. Operators are also likely to be
exposed to dust when dry flux powder is handled or
G MM AW fWG/ CO.
> Flux - C C M edel cctr odes
when the excess flux is removed from the welded joint.
Fig. 16.1: Fume generation rates of SMAW, GMAW and Studies carried out in Russia have indicated that
FLAW at various current the total a m o u n t of fumes a n d the a m o u n t of
manganese oxide generated during SAW welding were
lower than those produced with coated electrodes. Also
the total fume content produced during SAW welding
(5.1 to 12.1 mg/m 3 ) was considerably lower than that
I Data Points encountered in C 0 2 welding (8.4 to 36.0 m g / m 3 ) .
124- Similar tests carried out during electroslag welding
showed that the total fume content ranged from 5.0 to
6.0 mg/m 3 . The results of these studies indicated that
the main hazard of SAW and electroslag welding
8 1-5 processes is the fluoride content of the fumes produced
v by basic encountered fluxes; fluoride concentrations
v
1 © o considerably higher than the allowable limits were
i • •-» a
E frequently encountered.
p *
51.01 o o
In a production workshop of a Polish Shipyard
v\ © EJ £ having four SAW welding units and 30/40 manual
CM f\
£*
*rt r*»
•^ welding stations, fumes were collected in the immediate
r VI
E70S-
«J
E70S
° s <M
welding area and at distances of 6 and 12 m from this
00 00 .-* area. The results of fume studies are given in Table 16.3.
© o !•
These fume studies were coupled with clinical
1
UJ UJ
•* examinations of 30 welding operators who had used
submerged-arc welding equipment for an average of
MMAW MIG/CO. Flux-Cored Electrodes five years. The most c o m m o n ailments were
inflammation of the upper respiratory tract and the
Fig. 16.2: Weights of SMAW, GMAW with C02 and alimentary tract, conjunctivitis and psychoneurosis;
flux-cored electrodes converted to fumes at various
silicotic-type lesions were noted in the lungs of older
currents
workers.
Table 16.3: Analysis of fumes collected during welding in shipyard areas
Sampling location Fume concentrations, mg/m3 Gas concentrations, mg/m3
SiOz FeO F Mn HF N0 2 CO C0 2
Production shop
Operator's working area 4.59 3.30 0.32 0.18 1.40 Trace 21.00 429.00
6 m from operator's working area 2.37 3.10 0.30 0.19 0.35 ND (1) ND 400.00
12 m from operator's working
area 1.73 2.50 0.16 0.12 Trace ND ND 400.00
(1) Not detected.
Another study has shown that gaseous fluorides, The TLVs refer to airborne concentrations of substances
mostly SiF4, are evolved when high-silica type fused and represent conditions under which all workers may
fluxes are used. In poorly ventilated areas, SiF 4 be repeatedly exposed, day after day, without adverse
concentrations can reach the threshold limit. Carbon effect. When the TLV of a constituent is exceeded, it
monoxide and carbon dioxide produced from carbon presents health hazard to the welder.
in the wire and the base metal can reach the threshold The TLVs of particulate fumes are expressed in m g /
limit in 15 to 30 minutes in similar conditions. m 3 and of gases in parts per million (ppm). The smaller
the numerical value of the TLV, the more toxic is the
Plasma Arc Welding constituent. For example, cadmium oxide which has a
The health hazards of this process are similar to TLV of 0.1 m g / m 3 is more toxic than zinc oxide having
those of GTAW w e l d i n g , except that the arc a TLV of 5 mg/m 3 . TLVs refer to all airborne particles
temperatures and the amounts of ultraviolet radiation irrespective of their size. As already pointed out, a
are much higher with the plasma arc. Additional w e l d i n g process releases a n u m b e r of different
hazards of the plasma arc process are the high voltages constituents in the fume, depending on filler and base
required to initiate the arc and the high level of noise metal compositions, fluxes and shielding gases. Along
generated by the arc. However, the noise level is with their TLVs, one must take into account (1) the total
objectionable during plasma arc cutting and not during fume levels, (2) individual constituents, and (3) the sum
welding. of individual constituents.
The concentrations of c o n t a m i n a n t s in fumes To add u p the effects of different components, the
generated during various plasma arc welding and following formula may be used:
cutting operations arc summarised in Table 16.4, along
with the threshold limit values (TLVs). A B C
In a study conducted in Romania, the concentration — + —+ — + . . .
of nitrogen oxides was measured at various distances T
X
T TX L
A B C
from the cutting plasma arc. One metre away from the where A, B, C are the concentrations of the constitu-
arc the concentration was 38 mg/m 3 , and 7 m away it ents of the fume, and TA, TB,TC are their respective TLVs.
was 28 m g / m 3 . It may be stated in general that If the sum of the fractions is more than 1, the mixture
quantities of nitrogen oxides and ozone in excess of TLV should be considered as having exceeded the limit.
can be produced during plasma arc cutting in poorly TLVs refer to time-weighted concentrations for a
ventilated areas. seven- or eight-hour workday and a 40-hour work
week. They should be used as guides in the control of
Threshold Limit Values health hazards and should not be used as fine lines
Factory inspectorate and health authorities in between safe and dangerous concentrations. Time-
advanced countries issue recommendations for the weighted averages permit excursions above the limit,
control of health hazards in welding and cutting in the p r o v i d e d they are c o m p e n s a t e d by e q u i v a l e n t
form of guidelines to the allowable concentrations of excursions below the limit during the workday. (If the
airborne contaminants in the air of workplaces. The substance also has a ceiling limit, C, excursions are not
guidelines are listed as threshold limit values (TLVs).
allowed.)
Table 16.4: Concentrations of various fumes generated at plasma arc torch operations
Contaminant Concentration TLV Remarks
mg/m 3 mg/m 3
Barium 0.23-0.72 0.5 Spheroidisation of barium compound. Operation performed
out of exhaust hood.
0.02 0.5 Operation performed in hood.
Beryllium 2.8* 2* Spraying with exhaust ventilation equipment on.
0.9*
3.3*
4.3*
Total fume 31.3-fO.l 15 Breathing zone of welder welding stainless steel,
Chromium 3.1+0.1 without exhaust ventilation.
Total fume 24.5+0.1
Chromium 2.4+0.8
Total fume 0.41 15 Cutting stainless steel.
5.7 15 Cutting monel metal.
0.7 15 Breathing zone of plasma-arc cutter cutting inconel, with exhaust
ventilation.
0.5 15 Breathing zone of plasma-arc cutter after cutting monel with
exhaust ventilation.
Nickel Minimal 1 Cutting copper-nickel plates, ventilation on.
CrA 0.03 0.1 Cutting stainless, control switch operator's exposure.
0.02 General air.
0.01 Crane operator's exposure, directly above unit.
0.09 Crane operator's exposure, up and down bay.
Fe 2 0 3 35.1 10 Cutting stainless. Plasma-arc cutter's exposure.
(tentative)
13.9 Cutting stainless, control switch operator's exposure.
F 2 20 3 35.1 Cutting stainless, operator's exposure.
F 2 20 3 16.2 General air.
F 2 20 3 11.9 Crane operator's exposure, directly above unit.
Crane operator's exposure, up and down bay.
W
Iron
19.6
Minimal 10.0
Lead Minimal 0.2
Tin Minimal 2.0 As determined by spectrographic analysis of fume from
Nickel Minimal 1.0 cutting mild and stainless steel and from spheroidisation
Bismuth Minimal of a barium compound with and without exhaust ventilation
Copper Minimal 0.1
Cobalt Minimal 0.1
Molybdenum Minimal 15.0
(insoluble compounds)
Manganese 5.0
Chromium 0.1
Barium 0.5
(soluble compounds)
•" Micrograms per cubic metre.
For all substances not bearing a C notation:
The number of times an excursion above the TLV is
TLV Excursion allowed is restricted by the need to conform with the
ppm or mg/m 3 factor
time-weighted average TLV.
0-1 3 Tables 16.5 and 16.6 reproduced from The Facts about
1-10 2 Fumes, a publication of Welding Institute, list examples
10-100 1.5 of threshold limit values specified by the factory
100-1000 1.25 inspectorate, in Technical Data Note 2/74. The list gets
Table 16.5: Examples of TLVs for fumes specified in UK

Substance listed TLV, mg/m 3
Beryllium (Be) 0.002
Cadmium oxide fume (as Cd) -* C
Calcium oxide (CaO) 5
Chromic acid and chromates (as Cr03) 0.1
Chromium, soluble chromic, and chromous salts (as Cr) 0.5
Cobalt, metal fume, and dust (Co) 0.1
Copper fume (Cu) -*
Copper dusts and mists (Cu) 1.0
Fluoride (as F) 2.5
Iron oxide fume *
Lead (Pb), inorganic compounds fumes, and dusts 0.15
Magnesium oxide fume (Mg) 10
Manganese and compounds (as Mn) 5C
Mercury (Hg) 0.05
Molybdenum (Mo): soluble compounds 5
insoluble compounds 10
Nickel (metal and soluble compounds as Ni) 1
Silver (Ag), metal and soluble compounds 0.01
Tin (inorganic compounds except SnH4 and Sn02), as Sn 2
Tin oxide E
Titanium dioxide (Ti) E
Tungsten and compounds (as W): soluble 1
insoluble 5
Vanadium (V2Os) as V: dust 0.5
fume 0.05 C
Zinc oxide (Zn) 5
Nuisance particulates 10
Welding fume *
Cadmium oxide fume 0.05 C
Copper fume 0.2
Iron oxide fume 5
Notes: C = ceiling limit; F = simple asphyxiants; E = nuisance particulates; * = notice of intended change.
revised from time to time and hence one must consult VENTILATION
the latest version. USA S t a n d a r d Z49-1, "Safety in Welding a n d
In the U.S.A., Threshold Limit Values of Airborne C u t t i n g " has specified the need for mechanical
Contaminants are issued yearly by the American ventilation as follows:
Conference of Industrial Hygienists, Cincinnati, Ohio. Para s2.l: Mechanical ventilation shall be provided
when welding or cutting . . . :
Fume Control i) I n a s p a c e o f l e s s t h a n 2 84 m 3 per welder
Generally speaking, a welder is not exposed to the ... T , . .,. , . ,L ,, .,
, i c f,. r -r /-i\ i_ i. • .1. n) In a room having a ceiling height of less than
hazards of welding fumes if (1) he is working in the _ ° ° °
open or in a large shop having good natural ventilation
i[i In
through the roof or side windows; (2) the welding ) confined spaces or where the welding space
workload is compatible with the floor space, air volume contains partitions, balconies, or other structural
and ventilation provided for each welder; and (3) he barriers to the extent that they significantly ob-
keeps his head out of the fume by adopting the correct struct cross ventilation.
position in relation to the plumes of the fume and to Para 8.3: Mechanical local exhaust ventilation when
natural draughts. required shall be by means of one of the following:
Table 16.6: Examples of TLVs for gases specified in UK

Substance listed TLV / ppm
Acetaldehyde, CH3, CHO 100
Acetylene, C2H2 F
Acrolein, CH2, CHCHO 0.1
Argon, Ar F
Carbon dioxide, C0 2 5000
Carbon monoxide, CO 50
Helium, He F
Hydrogen chloride, HC1 5C
Hydrogen fluoride, HF 3
Nitric oxide, NO 25
Nitrogen dioxide N0 2 5C
Ozone, 0 3 0.1
Phosgene, COCl2 *
Propane, C3H3 F
Trichloroethylene, CHC1:CC12 100
Phosgene 0.05
Notes: C = ceiling limit; F = simple asphyxiants; E - nuisance particulates;
* = notice of intended change.
of ventilation required to accomplish this control

Table 16.7: Ventilation requirements for various
electrode diameters velocity using a 75 mm wide flanged suction opening
are presented in Table 16.8.
Electrode diameter, Ventilation required,
mm m3 / min
EXTRACTION
4.8 or less 57 The best way of controlling fume is to use local
6.4 100
extraction, which will prevent its release into the shop
9.6 128
3.2 (flux-cored) 128 atmosphere. The extraction unit consists of flexible
ducting, connecting a movable duct to an air suction
pump. The suction inlet is placed close to the arc and
Freely movable hoods intended to be placed by the in such a position as to suck the fume away from the
welder as near as practicable to the work being welded welder. The suction inlet is repositioned from time to
and provided with a rate of air flow sufficient to time by the welder as his welding progresses. Where
maintain a velocity in the direction of the hood of 30.5 several welders are working together, it is convenient
mm in the zone of welding when the hood is at its most to connect all flexible ducts into a larger common duct.
remote distance from the point of welding. The rates The flexible ducts can be supported on overhead arms.
Table 16.8: Mechanical local ventilation requirements

Welding zone Minimum air flow Duct diameter
in. mm eft / min m 3 / min in. mm
4-6 100-150 150 4.25 3 75
from arc or torch
6-8 150-200 275 7.8 90
from arc or torch
8-10 200-250 425 12.1 115
from arc or torch
10-12 250-300 600 16.6 5Vi 140
from arc or torch
Note: Special arrangements are necessary for use with cadmium, fluoride, zinc, lead, beryllium, mercury, and other toxic materials.
For GMAW with C 0 2 welding and flux-cored with

gas shield or self-shielded arc welding, specially
designed fume-extractor guns and fume-extractor
attachments marketed by some companies may be used
with advantage. A typical fume-extractor gun is shown
in Fig. 16.3.
For protection against toxic particulate fume where
the nature of the hazard is known, the welder should
use a dust respirator with filter as shown in Fig. 16.4
(A). In some cases, adequate safety is provided by a
welding helmet supplied with cleaned compressed air
at a slight positive pressure as shown in Fig. 16.4 (B).
When a welder has to weld in a poisonous gas
Fig. 16.3: Fume-extractor welding gun
atmosphere or inside a container where exhaustion and
Fig. 16.4: Means of protection against hazardous fumes and gases: (A) Dust respirator with filter; (B) Welding
helmet fed with cleaned air; (C) Breathing apparatus
Local extraction is a must w h e n the quantity and ventilation cannot be provided, or where the nature of
toxicity of welding fume demands special precautions, the hazard is unknown, he must be provided with a
as for example, in production welding of galvanised positive pressure air-fed breathing apparatus such as
steel and welding inside a closed container. shown in Fig. 16.4 (C).
IS 818-1968 "Code of Practice for Safety and Health Requirements in Electric and Gas Welding and Cutting Operations".
IS 3016-1965 "Code of Practice for Fire Precautions in Welding and Cutting Operations".
ANSI Z49.1 "Safety in Welding and Cutting" published by American Welding Society.
AWS F4.1-80 "Recommended Safe Practices for the Preparation for Welding and Cutting of Containers and Piping That Hold Hazardous
Substances".
Fumes and Gases in the Welding Environment, 1979, by American Welding Society.
Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment, published by the American
Conference of Governmental Industrial Hygienists.
The Facts About Fume, 1976, by the Welding Institute.
Welding Fume Control with Mechanical Ventilation, 1980, by Fireman's Fund Insurance Companies, U.S.A.
ANSI Z88.2 "Practices for Respiratory Protection" published by American Welding Society.
ANSI Z117.1 "Safety Requirements for Working in Tanks and Other Confined Spaces", published by American Welding Society.
ANSI/AWS Fl.1-78 "Method of Sampling Airborne Particulates Generated by Welding and Allied Processes", by American Welding
Society.
General Hints
on Welded
Design
An engineer entering the field of welded design has To determine the load on individual members, the
usually a mechanical or materials engineering back- designer must first ascertain the work the machine is
ground and very little understanding of the factors that expected to do or the forces a structural assembly must
contribute to efficient welded design, because welding bear, a n d the service conditions that may cause
technology and weld design are not regular subjects in overload.
engineering colleges. The design must satisfy strength and stiffness
He must remember that to be a successful design requirements; at the same time, the designer must guard
engineer he must be able to design a welded structure against over-designing, which is a waste of material
that will perform its intended functions, will have ad- and increases overall weight and cost. Sometimes the
equate reliability and safety, and is capable of being safety factor can be unrealistically high.
fabricated, inspected, transported and placed in service In developing a design, the following factors are of
at minimum cost. Total cost includes costs of design, help:
materials, fabrication, erection, inspection, operation,
1) Specify as far as possible steels which do not need
repair and maintenance.
preheating and postheating.
In designing a weldment to replace a casting, the
novice usually duplicates the shape and appearance of 2) Use standard plate and bar sizes, and standard
the casting by providing protrusions, brackets, housings rolled sections wherever possible. If a standard
and other frills. The designer must appreciate that rolled section is not available, the desired sec-
weldments are different from castings. A weldment is tion can be formed from blanks flame-cut from
an integrated, functional unit having an appearance of plate.
its own. A correctly designed weldment is much lighter 3) Use minimum number of pieces to reduce assem-
than the c o r r e s p o n d i n g casting a n d has a lean, bly time and the amount of welding. Ensure that
streamlined appearance. Efficient, cost saving welded minimum scrap is generated.
designs are possible mainly because of the availability 4) Use stiffeners properly to provide rigidity at
of minimum weight of material. Put bends in a plate
(1) mechanised flame-cutting equipment that can to. increase stiffness. Use press indentations in
produce smooth cut edges; plate to act as ribs. Use corrugated sheet for ex-
(2) high capacity press brakes which enable the tra stiffness.
wider use of formed plate;
(3) a w i d e r a n g e of w e l d i n g processes a n d 5) Prefer closed tubular section or diagonal brac-
consumables; ing for torsional resistance.
(4) welding positioners which permit smooth and 6) Check with the shop that they are able to hold
less costly welds to be deposited in the downhand the close tolerances and fits, you specify in the
position. design.
7) Pick up cost-saving ideas from the shop. approach, the various design and shop factors to be
8) Consider the use of a steel casting or forging in a taken into account, the types of loading to be dealt with
complicated weldment to eliminate design prob- and the use of appropriate formulae. The formulae
lem and reduce manufacturing cost. should be applied only after the problem has been
clearly defined and carefully analysed.
9) Consider use of hardfacing alloy at the point of
In this chapter we have not dealt with design
wear rather t h a n using expensive material formulae, procedures and problems. To get information
throughout the section. and guidance on these topics, the design engineer is
10) Adopt weld-saving methods described in Chap- advised to refer to the publications listed at the end of
ter 15. Specify intermittent fillet welds under this chapter.
light-load or no-load conditions. Remember that The design engineer should know how to indicate
stiffeners or diaphragms do not need full weld- the required welding information on engineering
ing. drawings through standard welding symbols. Using
ll)Break up the welded structure into subassemblies such symbols, he can specify joint design, face and root
to enable more men to work on the job simulta- details, welding process, type of weld, length of weld,
neously, to provide better access for welding and effective t h r o a t , w e l d i n g s e q u e n c e and other
to reduce the possibility of distortion or residual requirements of welding procedure to the fabrication
stresses. shop in a concise manner. Sometimes it is not possible
to convey all the necessary instructions t h r o u g h
12) Remember the factors which help to control and symbols alone. In such cases, the designer must provide
minimise welding distortion (refer chapter 12). s u p p l e m e n t a r y notes and additional dimensional
The designer must be able to use mathematical de- details. On the whole, the designer must make sure that
sign formulae for calculating loads and their effects on there is no ambiguity in the interpretation of his
sections, and for determining the sections necessary for drawing by those for whom it is meant.
withstanding such loads. A large number of formulae, The designer is advised to refer to any of the
developed for various conditions and member types, following standards to be able to adopt a suitable
are available for use by the design engineer. These for- system of symbols for use in his blue print drawings:
mulae appear difficult to a newcomer, but by proper AWS A2.4: Symbols for W e l d i n g a n d N o n -
approach and practice, he can simplify design analysis destructive Testing
and the use of the formulae. Guesses and rule-of-thumb BS 499 (Part 2): Symbols for Welding
methods for selecting configurations and sections must IS 813 (1986): Scheme of Symbols for Welding
be avoided, as they result in excessive costs of materi- ISO 2553 (1992): Symbolic Representation on Draw-
als and fabrication. ings.
The design formulae for strength and rigidity are Among these, only the AWS standard indicates
made up of terms describing load, member and stress symbols for n o n - d e s t r u c t i v e testing. The Indian
and strain. The application of load can be steady, impact Standard was originally based on the British Standard
or repeated cyclic (fatigue). There are five basic types BS499-(Part 2). The elementary or basic symbols now
of loads: tension, compression, bending, shear and used in the ISO, IS and European s t a n d a r d s are
torsion. Properties of steel include tensile strength, identical, while those in the AWS are different.
compressive strength, shear strength, fatigue strength, In the AWS system, a complete welding symbol
modulus of elasticity (tension) and modulus of elasticity consists of the following elements:
(shear). Properties of a steel section include area, length, 1) Reference line (always shown horizontally)
moment of inertia (stiffness factor in bending), section 2) Arrow
m o d u l u s (strength factor in b e n d i n g ) , torsional
resistance (stiffness factor in twisting) and radius of 3) Basic weld symbol
gyration. Stress is expressed as tensile, compressive or 4) Dimensions and other data
shear. Strain is expressed as resultant deformation, 5) Supplementary symbols
elongation or contraction, vertical deflection or angular 6) Finish symbols
twist.
7) Tail / Fork
Before proceeding with weld design, the engineer
must acquire a basic understanding of the design 8) Specification, process, or other references.
GENERAL HINTS ON WELDED DESIGN 639
These elements have specified locations with respect the break line for the other side. This is irrespective of
to each other on or around the reference line as shown the placement of the break line either up or down. The
in Fig. 17.1. The designer need not use all elements designer must be aware of these two different systems
unless required to make the drawing more clear. and take care that his drawing is not misinterpreted.
There are two prevailing systems of placing the sym- Besides the elementary symbols and a fairly large
bol with regard to the reference line. In the first system number of illustrated examples of the use of symbols
which is followed in the AWS 2.4, the symbol is placed based on both the systems, it gives only a few sup-
below the reference line for welds on the arrow side and plementary symbols, explains how weld dimensions
placed above the reference line for the welds on other have to be shown, and gives at the end indications for
side. The ISO and also the European standards have peripheral weld, site welds and welding process to be
accomodated now the systems identical to the IS:813- used.
1986. Where the reference line is made with one con- In many organisations, the design engineer is ex-
tinuous line and coupled with another parallel break pected to provide welding procedure sheets along with
line. The information placed over the continuous ref- his designs. To prepare them, he is expected to take
erence line for welds on the arrow side and placed over help from the welding engineer and shop supervisor.
(N)
Number of spot
or projection welds
—Elements in —
this area
remain as shown
when tail and
arrow are reversed
Finish symbol 8 basic weld symbol or detail reference
Contour or profile symbol 9 Reference line
Root opening; depth of filling for plug 10 Weld-all-around symbol
and slot welds 11 Arrow connecting reference
Effective weld size
line to arrow side member of joint
Depth of preparation; size or strength
12 Site/Field weld symbol
for certain welds
13 Pitch distance (centre-to-centre spacing)
Specification, process, or other refer-
of welds
ence
14 Length of weld
Tail (tail omitted when reference is not
15 Groove angle; included angle of coun-
used)
tersink for plug welds
Fig. 17.1: Standard location of elements of a welding symbol
Yet, to qualify as an expert designer, he must possess a dimensions and distortion. The best designed jigs are
sound knowledge of welding technology. This should often fitted into positioners and rotators, so that
cover welding processes, procedures and techniques downhand welding can nearly always be used. They
and weldability of industrial metals and alloys. He is provide ample clearance so that the welder will not be
therefore advised to study this entire book. cramped, and they are also of sturdy construction to
Plate preparation: Cutting is the first step in welded withstand rough usage.
fabrication. The commonly used metal cutting methods Simple devices for holding parts together are often
are listed below, and the designer must decide in made from scrap plate, punched, cropped or sheared
consultation with the shop floor supervisor which of as required. Some devices are based on wedges, others
these will be the most economical: on screw clamping. Improvised jigs are often quickly
1) Flame-cutting built up from existing multi-purpose clamps, vices, etc.
2) Shearing In the heavier types of arc-welded fabrication, jigs of
3) Sawing the most complicated and ingenious type are to be seen
4) Punch press blanking in the manufacture of automobile frames, tractors,
agricultural machinery, railway carriage and wagon
5) Nibbling
underframes, etc.
6) Lathe cut-off (for bar and tube stock).
Toggle clamps are used in preference to nuts and
The next step is edge preparation. For single-bevel bolts in welding jigs for holding work against prepared
or single-V plate preparation, single-tip flame-cutting surfaces. These clamps can be self-locking. The biggest
torch is suitable. For double-V, a multiple-tip flame- clamp of this type can exert a working pressure of 450
cutting torch must be preferred, because the required kg. With one type of clamp, a wide range of inter-
preparation is done in one pass on the cutting machine. changeable clamping spindles is available. The vari-
When a J or U groove is to be made in a thick plate, the ous types of spindle include those of different lengths
edge planer is most suitable. For back-pass preparation, and different ends, such as, plain, swivel, pressure
air carbon-arc gouging, flame gouging or chipping must spring loaded, swivel plate, etc. Floor-type aligning and
be preferred to machining. holding tables such as shown in Fig. 17.2 are used for
Plate forming: Forming is the next step. The common heavy structural members and machinery parts.
forming methods are listed as follows. When forming
Cam clamps with roller-cam action, C clamps and
is properly used, it can reduce cost by eliminating some
pliers are also available. An improved form of C clamp
of the welds and machining operations.
will incorporate a spring-controlled ratchet and pawl,
1) Press brake enabling the clamping pad to be pushed into contact
2) Bending rolls instantaneously with work of any thickness. Another
3) Roll-forming type of C clamp incorporates toggle action. Vices are
4) Tangent-bending and contour-bending also used in jigging. A useful type of vice clamp consists
of a vice without the fixed jaw. Two such vices, moving
5) Flanging and dishing
on a slotted bar, enable a very wide size range of job to
6) Press-die forming and drawing. be clamped. Vices and clamps can be attached to slotted
Jigs and fixtures: The d e s i g n e n g i n e e r m u s t bars, which, in turn, may be hinged, enabling the vices
familiarise himself with standard welding aids such as to clamp vertically as well as horizontally, and in
positioners, rotators, booms and seamers (see Chapter combination to act as a quickly-arranged jig.
20) so that he can adapt his designs to suit proper H y d r a u l i c a n d p n e u m a t i c c l a m p s are also
exploitation of these aids. They are marketed by commonly used in jigs. Hydraulic pressure may be
companies w h o are specialised in the design and generated by a lever or screw hand-pump, or by a
manufacture of such equipment. However, he may be hydro-pneumatic intensifier, which uses compressed
called u p o n to design jigs, clamping systems and air at low pressure, 60 to 80 lbs/sq. inch, to provide
fixtures to assemble parts quickly and accurately for hydraulic pressure at 1,400 to 1,500 lbs/sq. inch. The
welding. advantages of such clamps are that equal pressure can
A quality welding job is impossible without good be applied at all points simultaneously and the pressure
fit-up. Accurate and quick fit-up is accomplished by may be determined according to the type of material
using jigging and clamping methods. Jigs also control to be welded.
♦^s^^^^^rares^
-Grouted CJ. plates
w KR*
&
i
W YA VXwi VxYA □
Concrete
*♦ 101
v
© ® ©)V /
p
b
to
VJ
©
i i V
- * | K 1" to 1 1/2"
Fig. 17.2: Floor-type aligning and holding table

Magnetic clamps are also used to position plates for completed joint. Root face should be avoided when
welding, as for instance, in the fixing of a stiffener to a backing strips are used.
flat plate. The root face is meant to prevent burn-through and
to provide additional thickness of metal as shown at
Welding Procedures (b) in Fig. 17.3. To prepare an edge with root face, two
The designer must be well aware of the factors cuts with a torch are necessary, or possibly one torch-
which make up the welding procedure. They have been cut plus machining. If the bevel cut is made with a single
dealt with in the earlier chapters on Welding Processes. torch-cut, it results in a feather edge. Such an edge is
For a new job, procedure may be selected after prone to burn-through.
welding a few sample joints and subjecting them to the Degree of bevel for V-joints is usually 30° or 35°. In
r e q u i r e d tests. For r o u t i n e jobs, s t a n d a r d i s e d confined places, the extent of bevel may be dictated by
procedures based on past experience may be adopted. the importance of maintaining correct electrode angle.
The correct procedure is one which will complete the In Fig. 17.4, 45° minimum angle is necessary to permit
job at the lowest possible cost and, at the same time, the electrode to reach the root.
meet quality requirements. J and U p r e p a r a t i o n s are preferred for heavy
As far as possible, the steel selected must be easily thicknesses, to minimise the amount of weld-metal
weldable, i.e. it should not d e m a n d preheat and required, but then it must be remembered that they
postheat operations. Most structural steels including d e m a n d considerable accurate machining. These
quenched-and-tempered steels need no such heat treat- grooves also must have root faces and they require
ment, provided that in some cases, suitable low-hydro- back-gouging. Such preparations are used mainly on
gen type electrodes are used. boiler shells and pressure vessels.
A root gap is specified to ensure accessibility to the On butt welds, a nominal weld reinforcement of 1.5
root of the joint. The bevel angle must be just wide mm is adequate. Additional build-up serves no useful
enough to permit a 4 or 5 mm diameter electrode to purpose and increases cost.
reach the root. If the bevel angle is smaller, the root Without back-gouging, a butt joint welded from
gap has to be wider to get good fusion at the root. If both sides will have incomplete penetration and slag
the root gap is too narrow, electrode of smaller size inclusions. Gouging is done by grinding, pneumatic
has to be used to reach the root. This results in slower chipping or air carbon-arc process. Gouging should be
w e l d d e p o s i t i o n rate a n d increases the risk of deep enough to expose sound metal, and the contour
hydrogen-induced cracking. If root gap is too large, should permit complete access to the electrode.
weld burn-through can occur (see Fig. 17.3(a)) unless A butt weld connecting two structural members
the gap is bridged with several small beads deposited having the same thickness develops full force in the
with smaller electrodes or a backing strip is provided. member and is called a full strength weld. In this case,
If the backing strip is of steel, the joint has to be back- it is not necessary to calculate the unit stresses in the
gouged to remove the backing material. If it is of weld or to specify its size, since a butt weld soundly
copper, such back-gouging can be avoided, because the made has equal or greater strength than the members
copper strip is easily removable. Backing strips are being joined. This is so in the case of mild steel as well
commonly used when all welding has to be done from as low alloy high-strength steel, provided a welding
one side or when the root gap is excessive. Sometimes consumable of matching strength has been used. This
the steel backing strip is retained as a part of the is permitted by recognised building and bridge codes.
(b)
Fig. 17.3: (a) Weld burn-through; (b) Root pass correctly deposited
Fig. 17.4: Relation between bevel angle and electrode angle
Fillet welds: When structural connection is to be A fillet weld may be concave, flat or convex. A con-
made with a fillet weld, it is necessary to specify its cave weld produces a smooth change in section at the
size and length. The strength of a fillet weld is based joint and is to be preferred for dynamically loaded
on the effective throat thickness, which is defined as structures. Such a weld, however, is more susceptible
the shortest distance between the root and the face to shrinkage cracks, especially in the higher carbon
of the triangular weld, and is shown in Fig. 17.5. For steels. A slightly convex or flat fillet weld is to be pre-
an equal leg fillet weld, the throat is 0.707 times the ferred generally. Excessive convexity must be avoided
normal leg size. In an unequal-legged fillet weld, the as it produces a notch-like effect, and since the excess
leg size is measured by the largest right-angle weld-metal does not contribute to strength, the cost is
triangle, which can be inscribed within it as shown increased.
in Fig. 17.6. The effective throat in this case is slightly The strength of a fillet weld depends upon the
larger than if the weld was symmetrical. But the direction of the applied load, which may be parallel or
amount of weld-metal involved is considerable. transverse to the weld. In both cases the weld fails in
Measure
minimum
throat
Legw Leg<x>
-Leg w -Leg » .
Concave fillet weld Convex fillet weld
Fig. 17.5: Fillet weld dimensions

Increased effective
throat r 1.262t
Added weld metal

CO
(a) Throat increased by 265S

Area increased by 100*
Fig. 17.6: Unequal-legged fillet weld dimensions
shear, but the plane of rupture is not the same. The weld In d e s i g n i n g m e m b e r s to w i t h s t a n d i m p a c t
will fail on the throat plane, which has the maximum loading, either of the two methods may be used:
shear stress. 1) Estimate the m a x i m u m force exerted by the
Data concerning the allowable force on the fillet moving body on the resisting member by applying an
weld and minimum weld size required to join two impact factor. Consider this force to be a static load
m e m b e r s of given thicknesses are p r o v i d e d in and use standard design formulae.
recognised building and bridge codes, and other 2) Estimate the energy (kinetic energy Ek= ^ ) that
equivalent codes dealing with welded structurals. is absorbed by the resisting members, and from this
value determine the stresses or deformation by formu-
Designing for Impact lae for impact loads on members. This method is pre-
A structure may be subjected to impact loads under ferred for more accurate results.
the following conditions: The dimensions and material properties that give the
1) A direct impact by another member or an exter- resisting member maximum resistance to an impact or
nal body energy load are quite different from those that give the
member maximum resistance to a static load. A steel
2) The sudden creation of a force on a member
having good tensile strength and ductility under static
3) The sudden moving of a force on to a member loading, may fracture if subjected to a high velocity blow.
4) The inertia of a member resisting high accelera- The two most important properties that indicate a
tions or decelerations. metal's resistance to impact loading are obtained from
the stress-strain diagram. One is the m o d u l u s of
In (2) and (3), there is sudden application of forces resilience, which is the energy the metal can absorb
without a blow being involved. It is to be noted that per unit volume when stressed to the proportional
impact loading does not necessarily involve move- limit. This is represented by the area of the triangle
ment of a mass over a long distance. OAB having its apex A at the elastic limit shown in
Usually it is very difficult to evaluate impact forces Fig. 17.7. The second property is the ultimate energy
quantitatively. The analysis of the design problem is resistance, which indicates a metal's toughness or
more of a qualitative nature and requires that all of ability to resist fracture under impact loading. This is
the factors involved and their interrelationship are represented by the total area OACD under the curve
recognised. in the figure.
Unit strain ($
!
Fig. 17.7: Parameters of impact resistance
In designing, the notch effect on energy adsorbing 3) Reduce stress concentrations to a minimum and
capacity has to be considered. For example, if a notch avoid abrupt changes in section.
produces a stress concentration of twice the average 4) Use the impact formulae as a guide to the proper
stress, then for the same maximum stress, the average design of the member, rather than as an actual
stress in the rest of the member will be reduced to half determination of the impact stress or impact
and the energy absorbed will be one-fourth of the energy deformation.
absorbed if no notch were present. 5) In general, with steel, as the speed of loading is
Notch bar impact test values are of little use to the increased, the yield strength has a noticeable increase.
design engineer and can even be misleading for the 6) Material should have a high modulus of resilience
following reasons: u = sy 2 /2E. This is the energy absorbed per unit volume.
1) The test is highly artificial in respect to severe Although a lower modulus of elasticity (E) appears to
notch condition and manner of load condition. be helpful, materials with lower values of E generally
2) The results can be altered over a wide range by have correspondingly lower values of yield strength
changing size, shape of notch, striking velocity, and (s d ), and this latter value is more important because it
temperature. is squared. Therefore, steels with higher yield strengths
3) The test does not simulate a load condition likely have higher values of modulus of resilience and are
to be found in service. better for impact.
4) The test does not give quantitative values of the 7) The material should have sufficient ductility to
resistance of the material to energy loads. relieve the stress in any area of high stress concentration.
8) The material should have high fatigue strength,
GUIDELINES although this is not considered to be so important as
Under impact loading, the member is required to high yield strength.
absorb a certain amount of kinetic or potential energy. 9) Place material so that the direction of hot rolling
It is important to observe the following: (of sheet or bar in steel mill) is in line with impact force,
1) Design the member as an energy absorbing because the impact strength in this direction is higher
system, that is, to have the maximum volume of material than if impact occurs at right angles with the direction
stressed to the highest working stress; this increases the of rolling.
energy absorbed. 10) It is important to restrict the weight of the
2) Have the maximum area subjected to the member and yet maintain proper rigidity of the member
maximum allowable stress, for any given cross-section for its particular use or service. This means light-weight,
of the member. Stress the entire length of the member well-stiffened members having sufficient moment of
to this maximum. inertia (I) should be used.
11) Where required, to build in protection against It is possible to distinguish between a plastic fracture
inertia forces caused by the rapid movement of the in a completely ductile metal and a running crack in a
member due to earthquakes, explosions, etc., it is fully brittle metal. Ductile fractures have a shear mode
important to decrease the possible acceleration a n d / o r of crystalline failure as opposed to a cleavage mode in a
deceleration of this member through some form of brittle fracture. Consequently, the ductile fracture has a
flexible support. surface texture that is either silky in appearance (as if
the two surfaces of the fracture had been rubbed or
Problem of Brittle Fracture slipped together), or is actually fibrous in appearance.
The failure of all-welded Liberty ships built in the Ductile fractures often appear to have failed mainly by
U.S.A. during World War II drew world attention for shear. Plastic flow is very much in evidence throughout
the first time to the liability of steel to brittle fracture. the entire development of a ductile fracture, that is,
Then there was the catastrophic failure of the all-welded during initiation, propagation and the final separation.
Duplessis Bridge at Three Rivers in the U.S.A. at a At no stage does the fracture run freely without energy
temperature of -30° F on February 1,1951. In the winter dissipation.
of 1962, the King's Bridge in Melbourne, Australia, Brittle or cleavage fractures have either a granular
collapsed suddenly and without resistance, though or a crystalline a p p e a r a n c e . Cleavage occurs by
periodical routine inspection had not revealed any signs separation of the cube faces of the crystals. Little plastic
of impending failure. deformation is found in the metal in the vicinity of the
Research in the field of brittle fracture of steel is being fracture. Brittle fractures usually initiate at a nucleus of
pursued vigorously since the last 35 years in many some kind, as shown in Fig. 17.8. This may be a small
countries. The research activity has received fresh crack initiated by any kind of mechanism (dislocation
impetus through the need to install gigantic oil-drilling pile-up, ductile crack propagation). The nucleus also
platforms in the North Sea and elsewhere, where could be a pre-existing flaw. It is important to point out
extreme conditions of impact loading and temperatures that when the nucleus crack represents a critical size
exist. for the particular metal and the conditions to which it
Brittle fracture usually occurs in very large structures is exposed, a running crack develops.
involving heavy plates such as bridges, ships, penstocks Unfortunately, no single test can quantitatively
and pressure vessels, and especially when the ambient evaluate or measure the ductility or bittleness of a large
temperature is around or below 0°C. The fracture section of a material. The common practice is to use the
initiates from sharp notches caused by design, or Charpy-V impact test. The steel must give a minimum
introduced by arc strikes, crack-like weld flaws, raw impact value of 15 ft-lbs at the lowest temperature, to
p u n c h e d holes, s h e a r e d e d g e s , etc. The r a p i d which it is expected to be exposed in service. Because
propagation of the crack is encouraged by the notch this test incorporates multiaxial stresses and high strain,
brittleness of the parent steel. and can be carried out at selected low temperatures, it
It is now understood that the commonly measured is considered that a value of over 15 ft-lbs assures the
mechanical properties of structural steels such as steel's ability to withstand a reasonable amount of notch-
ultimate strength, yield point and elongation, which the effect from w e l d m e n t design or questionable
designer is accustomed to use as a basis for material w o r k m a n s h i p , a n d to sustain impact loading.
selection, do not always assure freedom from brittle Furthermore, if the weldment is subjected to overstress,
failure. This is true even when the design load is not one would expect deformation and a tough, shear-type
exceeded. To ensure safe performance, the steel must fracture instead of a sudden, brittle fracture. The 15 ft-
be resistant to the initiation and propagation of a crack lbs criterion is included as a requirement in a number
under all service conditions. This resistance to brittle of codes and specifications for steel to be used in low-
crack initiation and propagation is referred to as notch- temperature service. It may be pertinent to mention here
toughness or notch-ductility. In the context of brittle that impact tests of fractured plates of Liberty ships
fracture phenomenon, the real meaning of ductility is carried out at the temperatures at which the fracture
the ability and capacity of a metal to yield plastically occurred gave a value of less than 11.4 ft-lbs.
under constraint and high local stresses, such as those Charpy-V impact tests and notch-bend tests on steels
that exist at the root of a notch. Its inability in these at varying temperatures have shown that as the testing
respects is called notch-brittleness. temperature is lowered, the material first exhibits a
APEXES OF CHEVRON PATTERNS

IN FRACTURE FACES OF MEMBERS
CONVERGE ON AREA WHERE FRACTURE
ORIGINATED. IN THIS CASE, CLOSER
EXAMINATION REVEALS FRACTURE
]B$m
INITIATED NEAR ROOT OF POORLY
PENETRATED FILLET WELD
DEPOSITED IN ACUTE-ANGLE
JOINT ON RIGHT-HAND
SIDE OF UPRIGHT
MEMBER
Fig. 17.8: Chevron pattern frequently observed on the surface of brittle fractures will indicate the direction of origin.
Apex of chevrons point toward fracture source.
ductile type of failure with the absorption of large For the same heat of steel, plates of greater rolled
amounts of energy, then passes through a transition thickness have higher transition temperatures (i.e. they
region and thereafter fails in a brittle manner with little are less notch-tough) compared to those of lower
absorption of energy. thickness, because of the different metallurgical
Every steel and weld-metal has a specific transition structures of such plates (see Fig. 17.9). This is probably
temperature. The transition t e m p e r a t u r e may be the main reason why failures are more prevalent in
defined as that temperature at which the behaviour of large, than in small structures. Thick plates however
the specimen or structure changes from desirable plastic respond more readily to normalising treatment with a
behaviour to undesirable brittle behaviour. Steels or larger i n c r e m e n t a l i m p r o v e m e n t in transition
weld-metals with lower transition temperatures are temperature, than thin plates.
considered more notch-tough, or more resistant to A close look at the chemistry of plates which went
brittle fracture than those with higher transition into the Liberty ships revealed that no attention was
temperatures. paid to the important matter of keeping carbon as low,
Rimmed steel has poor notch-toughness and is not and manganese as high as possible. Carbon content
recommended for use in critical structures subject to ranged up to 0.35% with Mn between 0.30 and 0.60%.
dynamic loading. Semi-killed steel has reasonably good Later studies have shown that carbon is the most
notch-toughness. Fully-killed steel, especially silicon- important single factor affecting the notch-toughness
aluminium killed to fine grain size, possesses the best of steels, raising the ductility transition temperature by
notch-toughness. about 1.11 -2.22°C for each 0.01% increase in carbon
For critical applications, normalised steels are content (see Fig. 17.10). Mn acts favourably; each 0.01%
preferred, because normalising frequently refines the increase in Mn content (up to about 1.5% Mn) lowers
grain size a n d r e d u c e s v a r i a t i o n s in the steel the ductility-transition temperature by about 0.42-
microstructure introduced by different mill finishing 0.56°C. M n / C ratio thus becomes an important factor
temperatures and cooling rates. These beneficial effects in e n s u r i n g the n o t c h - t o u g h n e s s of the steel.
in turn reduce both, the average transition temperature P h o s p h o r u s is another element which raises the
and the variations from it. transition temperature markedly. A typical plate of
-40 -29 -18 -7 l\A

Transition temperature, °C
Fig. 17.9: Effect of plate thickness and deoxidation practice on the keyhole charpy
ductility-transition temperature
240 rn
V
T Carl: on
200
€0
0-011 |0.1 I
160
0.22
$120
8 0.31
Ui 80 - 0.4 ^
0.53
401 -0.63.
13L67
Ol —4
-184-130-73 -18 38 93 150 204 26C
TemperatureX
Fig. 17.10: Effect of carbon on the shape of charpy-V transition curve

class B specification taken out from the Liberty ship catastrophic as if the main members were themselves
had 0.24% carbon and 0.43% manganese, giving a M n / defective.
C ratio of 1.79. The transition temperature for this steel Prior cold work such as rolling, bending, punching,
as determined by Charpy-V test was as high as +15.6°C. shearing, straightening, etc., lowers the notch ductility
The present day steel of the same class B used in ships of a steel. When the amount of prestrain is small (1-
has a typical chemistry of 0.19% carbon and 0.71% 2%), its effect is more in rimmed steel, less in low-C
manganese, which means a M n / C ratio of 3.74. Its high-Mn semi-killed steel and least in killed steel.
transition temperature is -6.7°C. Larger strains (5-10%) raise the transition temperature
The age-old practice of raising the tensile strength of all these steels appreciably.
of structural steel through increase in carbon is now Some steels (especially Bessemer steel) are prone to
given up by steelmakers, because it results in steels of strain aging as a result of cold work. Strain aging, while
low notch ductility and poor weldability. This is now not fully understood, is usually associated with the
done through increase in manganese, while restricting presence of carbon and nitrogen, that are always
carbon to a specified limit. Beyond a certain limit of present in steel. Strain aging further raises the transition
Mn (up to about 1.5% Mn), any further increase is temperature.
effected through alloying elements in small controlled Steels susceptible to strain aging may have their
quantities and through normalising. Very high tensile notch ductility seriously affected, when after being
strength coupled with good weldability and notch- subjected to cold work, they undergo operations like
toughness is achieved in low-alloy steels through galvanising, heat straightening or adjacent welding,
quenching and tempering. which heat them in the range of 200-450°C. The loss in
It is n o w easy to u n d e r s t a n d w h y in ship ductility is restored partially after stress-relief at 600-
specifications, three classes of steel A, B and C are 625°C and fully after a normalising treatment.
stipulated. Class A is for non-critical parts. Class B is Bruising, or local deformation by a blow or bump
for plates of over 12.7 to 25.4 mm t. It is a semi-killed can be extremely damaging to a steel, resulting in
steel with 0.21% maximum carbon and manganese in almost complete loss of ductility in the area and acting
the range 0.80-1.10%. Class C is for plates of over 25.4 as a source of crack initiation on subsequent critical
mm and up to 51 mm t. This steel (C—0.24% maximum, loading. Columns are vulnerable to inadvertent bruises.
Mn—0.60 to 0.90%) must be fully-killed and made to Something similar to bruising occurs in cold flanging
fine-grain practice. Plates of over 35 mm t made to the or joggling of plates, especially when the forming load
class C specifications are required to be normalised. is concentrated at the apex of the forming tool.
Further information of this topic is given in the Providing a generous radius at this point can prevent
chapter on Weldability of Steels. this unfavourable effect.
Today's specifications and codes of practice relating Punching holes for rivets or bolts makes the steel
to pressure vessels, bridges, ships and similar standard less ductile at the edge of the hole, often leaving short
fabrications clearly lay down the requirements of steels cracks there. If by chance the area is heated to 200-430°C
to insure freedom from brittle fracture. But when it is a temperature range for galvanising, during Welding or
question of a new kind of fabrication where no standard the lay-up of the overlapping plate edges, severe strain-
specifications are available and previous experience is aging can occur, if the steel is prone to this phenomenon.
lacking, that the designer and the fabricator are called All these risks are eliminated if the hole is drilled or if
upon to take cognizance of all the factors that cause or it is reamed properly after punching to remove the
prevent brittle fracture, severely cold worked layer of metal.
Fabricators must remember that once the steel has Sheared edges may also cause severe cold work and
been correctly chosen, the resistance of structures to small cracks. These will not be a problem if the edges
brittle fracture is directly dependent upon quality of are later welded. The best practice requires that sheared
w o r k m a n s h i p a n d fabrication. They m u s t also edges which are to be left exposed are machined to
remember that non-structural details and secondary remove 1.6 m m or more material, d e p e n d i n g on
members joined monolithically to the strength members thickness.
by welding should be treated with the same care as the Fabrication notches are frequent causes of brittle
main m e m b e r s themselves, particularly in areas failure. Notches may be caused by sharp corners left in
subjected to high stress. Lack of attention to such details oxy-cutting, weld defects and sudden change of cross-
and secondary members can result in failure, just as section. It is important that necessary changes of section
are made in a smooth continuous manner in regions of fit-up, resist racking in root passes of butt welds better
stress. Oxy-cut notches can be avoided by the use of than wide, thin ones.
cutting machines, templates and radius bars in place Service experience shows that many failures origi-
of free-hand cutting. nate at serious defects in welded joints (approximately
Preheating is a very effective means of reducing the 40% of the major ship failures). Many failures have
chances of cracking in welds and heat-affected zones originated in welds at points of geometrical disconti-
of large structures and improving their ductility and nuity. On the other hand, there is almost no record of
notch-toughness. Postheat treatments like stress-relief failures having initiated in sound, properly made welds
and normalising are also very effective in improving of adequate design. Even when failures have originated
the performance of structures subjected to dynamic at obvious defects in welds, the crack has not propa-
loading. While postheating is not practicable for large gated for any distance in the weld-metal, unless it was
structures like ships, bridges and vertical storage tanks, of extremely poor quality. Also, fracture does not usu-
preheating is easily possible and is used with benefit. ally propagate in the heat-affected zone of the plate.
There is substantial laboratory evidence, some on large Instead, the crack propagates through or across the
specimens, that 200°C preheat may be as effective as weld and continues on in the unaffected base plate. In
600°C postheat in improving the performance of low- a d i s p r o p o r t i o n a t e n u m b e r of cases, cracks have
carbon steel weldments when subjected to static or stopped at or near welds which were approximately
impact loading. perpendicular to the path of the fracture. These obser-
Accidental or deliberate arc strikes which do not vations indicate that sound weld-metal and the heat-
deposit any weld-metal, can have an effect as severe as affected zones adjacent to the weld in mild steel are
a notched brittle weld, more so when the steel is a high not particularly sensitive to fracture propagation.
tensile grade, as proved by laboratory tests. There is As was pointed out at the outset, conventional
additional confirmation from records of service failures. design p r o c e d u r e s by themselves are sometimes
Fabricators should consider arc strikes which do not i n a d e q u a t e w h e r e brittle failure is of concern.
contain weld-metal as highly dangerous, and should Nevertheless, the best course of action in design is to
emphasise this fact to their production personnel. If arc use conventional procedure, at the same time keeping
strikes occur, they should be completely removed and in m i n d that certain a d d i t i o n a l p r e c a u t i o n s are
the void filled up by welding using preheat if necessary. necessary if brittle failure is thought to be a problem.
Light fillet welds and thin single pass welds in heavy The designer must recognise that these precautions
plates m u s t also be discouraged unless adequate relate to greater care in the choosing of design details,
preheat is used. in the selection of materials and in procedures of
The use of low-hydrogen electrodes with proper workmanship, fabrication, supervision, inspection and
moisture control has distinct advantages from the proof testing.
viewpoint of avoiding britle failure. Weld-metal The first g e n e r a l step in e s t a b l i s h i n g design
deposited by LH electrodes has lower transition conditions, at least in so far as brittle failure aspects are
t e m p e r a t u r e t h a n c o n v e n t i o n a l electrodes of concerned, is to d e t e r m i n e the lowest operating
corresponding strength. Such electrodes are therefore temperature for the structure. If available, a record of
to be preferred in the fabrication of weldments for low- previous service experience with similar structures, and
temperature service. However, it must be noted that of the origin of cracks in such structures (if any) is most
complete elimination of the possibility of fracture is a helpful. Following this, materials and design details can
function of the proper combination of procedure, be decided upon. It is important to bear in mind at all
electrodes, materials and workmanship, and not of the times that a crack may originate at a supposedly
electrodes alone. unimportant, non-structural member. Such a crack may
Badly fitted or incorrectly prepared joints frequently propagate across a joint and cause extensive damage
result in root cracking. This is particularly true in the to the main structure. Thus the design of non-structural
welding of thick plates due to high restraint. With poor a n d s e c o n d a r y m e m b e r s , a n d their degrees of
fit-up of the joint, it is either difficult to make the weld participation is of great importance in the behaviour of
bead span the joint, or to obtain proper penetration. the structure as a whole. This is apparent in the failures
With too wide a fit-up, the resulting beads are usually that have occurred when such members are attached
wide and thin when viewed in cross-section. Thick, to the outer fibres of a girder, e.g., bilge keels or
narrow beads, which can easily be made with good joint bulwarks on ships, rubbing strips on cranebooms. The
degree of participation depends upon various factors, amount of fatigue damage. The one exception is shown
such as the relative length of the attachment and the in Fig. 17.12. This is not only one cycle of Fmax to Fmin
degree of rigidity of the structural details along the but also one cycle of ¥1 to F2,
paths of stress transfer. The designer must be aware of In fact, many structures and machines produce
this tendency for participation and choose details of stress/time curves of an entirely random nature, of
design and methods of attachment that will preclude which a typical one is shown in Fig. 17.13. This is called
serious concentrations of stress as a result of such random loading. The only way of obtaining such a type
participation. of curve is by actual measurements with strain gauges
during service. In critical design, this information is
Designing for Fatigue essential, and is far cheaper than trial and error.
Under static loading, a steel structure may be Dynamic strain recordings show both, the number
stressed to just below yield point without danger of of cycles and their magnitude. As the weld type adjacent
failure. However, when the load fluctuates, the to the gauge is presumably obvious, one obtains the three
maximum permissible stress may be very low. Such facts immediately to make an extremely accurate
reduced strength under fluctuating loads is assessment of fatigue life.
characteristic of most metals and their bolted, riveted
or welded connections.
When a structure is subject to fluctuating load due
to live loads, vibration or temperature fluctuations, and
the following conditions exist simultaneously:
1) Load or stress is very high
2) Anticipated service extends for a great number
of cycles
3) Stress fluctuates over a wide range (i.e. stress ■•Time
range is very wide)
Fig. 17.12: Curve with two cycles
The possibility of failure by fatigue must be consid-
ered at the design stage, and eliminated. The fatigue
life of the structure is then governed by (a) number of
stress cycles, (b) magnitude of stress fluctuation, and
(c) type of weld. In a critical design, each one of these 1 digging cycle of
three factors must be correctly determined. Determin- excavator
ing the number and type of stress cycles that will occur
in service is not easy. Fatigue life is defined as the num- 0)
ber of cycles that occur before failure. to
The shape of the stress/time curve between
successive upper and lower limit stresses does not affect
fatigue strength, provided that the slope of the curve
does not change sign as shown in Fig. 17.11.
Time
Each of the cycles shown in this figure or any other
odd shape one may encounter, produces the same Fig 17.13: Random loading curve
Loading frequency has little effect on fatigue

strength. For example, take a press operating once a
minute, or a machine operating 5,000 times a minute; if
the stress in each on a similar weld detail is the same
(see Fig. 17.14), they will each fail after the same number
Time of loading cycles, assuming of course that the designer
1 stress cycle chose excessive working stresses. They will not fail after
the same amount of time has elapsed. Each cycle
Fie. 17.11: Shapes of stress/time curve produces the same amount of fatigue damage.
Type
°r\A/v/\/v/i • Alternating
Time
lAAAAAA Pulsating tension
T>ME PuUoting
+k / \ / \ A / H / \ / »
Fig. 17.14: Stress and time factor in loading 0 \j\/\/\j\/\J compression
Half tension
STRESS RANGE
The stress range necessary to produce fatigue failure Half compression
decreases with increasing numbers of stress cycles.
W h e n specimens are tested in the laboratory, VS\z\y\s\s^
identical specimens are loaded at various stress ranges Fig. 17.16: Examples of fluctuating stress
until failure occurs. A convenient method presenting
the results is an S-N curve shown in Fig. 17.15. If the
horizontal scale is logarithmic, the curve becomes a -A 8 |min f
max fatigue life
straight line. This is the usual form of an S-N curve. If (cycles)
we consider the specimen represented by point A, this 5
0.5 8 0.8x10
was tested at a stress range of x tons/ inch 2 and failed
i°rwvr, -' *
after 5 x 105 cycles. In establishing an S-N curve, Fmin is
the same for each specimen and the stress range is 2x10
varied by changing Fmax.The relative position of the
fluctuating stress to zero has a considerable effect on
fatigue life. In Fig. 17.16, all the examples shown obey
0.5 -8 100x10'
the same rule. For a given stress range, as the mean
stress increases in magnitude, the fatigue life decreases.
Mean stress is the mean of the upper and lower limit
kA/WC
stresses. Stress range is the difference between the Fig. 17.17: Figures to explain effect of mean stress
maximum and minimum stresses.
To explain the effect of mean stress, curves relating EFFECT OF WELDING
to a certain type of weld are shown in Fig. 17.17. In It must be remembered that stress concentrations
each case, the stress range is 4 T/sq. inch but fatigue give rise to fatigue cracks, and that stress concentration
lives vary. arises at every abrupt change of section of the material.
A weld usually causes change of section and is the cause
of stress concentration. But the change of section must
lie transversely to the direction of stress to have any
effect, as illustrated in Fig. 17.18. Small attachments
such as clips, gussets, ribs, etc., all normally attached
by fillet welds, and not in themselves carrying any load
may result in a low allowable stress in the member to
which they are attached.
The stress concentration formed by welds, whether
fillet or butt, lying in the direction of stress (such as
web /flange weld on a beam) is due to unevenness on
their surfaces caused by ripples, start and end craters,
1 0
Cycles t o failure ' " a n d other factors. The m o s t h a r m f u l type of
Fig. 17.15: S-N curve
longitudinal weld is the intermittent weld, at the ends
of which sharp stress concentration is formed.
through the parent plate, not in weld-metal. The one

exception is a butt weld with lack of penetration; here
the mode of failure is through weld-metal.
(No effect on fatigue strength) It must be remembered that it is not welding that
reduces the fatigue strength, but the resultant shape or
geometry of the section. It is for this reason that fillet
-£ 3- welds have lower fatigue strength, simply because they
(Considerable effect on fatigue strength) are used in lap joints, and all lap joints including riveted
Fig. 17.18: Effect of change of section on fatigue strength joints have lower fatigue strength. It must also be noted
that dressing a butt weld level with the plate increases
its fatigue strength to that of unwelded plate, also that
In welds transverse to the direction of stress, stress stress-relieving the weld joint has no appreciable effect
concentration is formed by weld reinforcement and on its fatigue strength.
undercut. Partial penetration single-V butt welds and Information on allowable fatigue stresses for
double-V butt welds with lack of penetration at the different steels and types of welded joints are given in
centre should never be used. standards such as BS 153, Bridge Specifications such as
Cruciform butt welds have a geometric stress on the one issued by the American Welding Society, and
centration due to stresses in the components joined, data sheets provided by steel producers.
being coupled with those caused by the weld as shown In the case of various types of welded joints, it must
in Fig. 17.19 (A). However, this type of joint is much be remembered that the allowable stresses refer to all
stronger than fillet welds doing the same job shown at types of steel. In other words, when welded all steels
(B). A load carrying fillet weld such as this has the have the same fatigue strength. This is a fact which is
lowest fatigue strength of all types of weld. In between difficult to explain. The only time the use of high tensile
these two cases is the transverse non-load carrying fillet steel is justified in a fabrication subjected to fatigue
weld shown at (C). In this case the weld is carrying no loading is when Fmax is above the permissible stress for
load, but it does produce a stress concentration on the mild steel.
load carrying plate. The following guidelines will help a designer in
When failure occurs from a transverse butt weld, it designing a fatigue-resistant weldment:
will always originate from the weld toe and propagate 1) In general, a machine or structure is stressed to
the full maximum value for only a portion of its fatigue
life or cycles. For most of its fatigue life, it is stressed to
a much lower value, and not to its full rated capacity;
hence, most fatigue loading is not as severe as it may
first appear.
(A) Consider actual stress rather than average stress.
Reduce, if possible, the range of stress without
increasing the maximum or average stress.
2) Fatigue loading requires careful fabrication,
A smooth transition of sections.
Avoid attachments and openings at locations of
"^
(B) high stress.
Avoid sharp corners.
Use simple butt weld instead of lap or T fillet weld.
Grinding the reinforcement off of butt welds will
increase the fatigue strength. This weld will have about
u
if~r\ n sitlk.
the same fatigue strength as unwelded plate. Grinding,
however, should not be specified unless essential, since
it does add to the final unit cost.
(O
Avoid excessive reinforcement, undercut, overlap,
Fig. 17.19: Cruciform joint designs affecting fatigue lack of penetration, roughness of weld.
strength Avoid placing weld in an area which flexes.
Stress-relieving the weld has no appreciable effect 1) Changing the forced frequency, preferably to a
upon fatigue strength. lower frequency. This usually means a change in motor
Difficulties are sometimes caused by the welds or operating speed.
being too small, or the members too thin. 2) Changing the natural frequency of the member,
3) Under critical loading, place material so that the preferably to a higher frequency.
direction of rolling (of sheet in steel mill) is in line with 3) Increasing the damping capacity through efficient,
force, because the fatigue strength may be higher in steel design.
this direction than if placed at right angles to the In starting up a machine from idle state, it is safer to
direction of rolling. ensure that it does not pass through its critical speed
4) Where possible, form member into a shape that or frequency, by moving the natural frequency of the
it tends to assume under load, and hence prevent the member up to a still higher value. The natural frequency
resulting flexial movement. can be increased by:
5) Avoid operating in the critical or resonant 1) Increasing the moment of inertia (I) o/the mem-
frequency of individual member or whole structure to ber.
avoid excessive amplitude. 2) Using a material having a higher modulus of elas-
6) Perhaps consider prestressing a beam in axial ticity (E).
compression. This will reduce the tensile bending stress
3) Reducing the cross-sectional area (A) of the mem-
and lessen chance for fatigue failure, even though the
ber (similar to weight).
compressive bending stress is increased to some extent.
7) Avoid eccentric application of loads which may 4) Reducing the unsupported length (L) of the mem-
cause additional flexing with each application of load. ber.
8) Stiffeners decrease flexibility of member and When compared to an equivalent grey-iron casting,
result in better fatigue strength, unless it causes a more a steel weldment has 2.5 times higher modulus of
abrupt change of section. If the latter should happen, elasticity, and requires 40% less moment of inertia and
the stiffeners may do more harm than good. 40% less area for the same stiffness. This means that
9) A rigid-frame type of structure or statically in- the natural frequency is increased by 58%. In other
determinate type of structure may be better than a words, simply changing from-grey-iron casting to steel
simple structure, since the load is shared by other mem- weldment gives a 58% greater operating range.
bers; hence, the structure is less likely to collapse im- When grey iron and steel are stressed to the same
mediately if a fatigue failure starts in one member. This value, the former exhibits three times the damping
will result in a more gradual failure of one part, then capacity of steel. However, in an efficient redesign
another, and this would provide a better opportunity based on equal stiffness to replace the casting, the steel
to notice that a fatigue failure is in progress. member is always stressed more than the equivalent
10) Avoid biaxial a n d triaxial stresses, avoid casting.
restrained internal sections. Steel, having a higher modulus of elasticity, requires
less moment of inertia for the same stiffness, and as a
Vibration Control result, has a lower section modulus. Therefore, for the
In machinery design, quite often, one has to consider same load, the steel member will have a corresponding
steps to control vibration. While cast iron has inherent higher stress. Because of its higher stress, its relative
damping capacity, correctly designed welded steel is damping capacity will increase.
known to cope with vibration problems in many types In a machine tool, if the redesign from casting to
of machine tools more effectively. welded steel is m a d e for equivalent stiffness, the
Every member has a certain natural frequency; it increased modulus of elasticity allows a corresponding
can also be forced to vibrate at any frequency by striking decrease in the required moment of inertia. This reduces
it repeatedly. This is called forced frequency. When the weight considerably, w i t h o u t any increase in
forced frequency equals natural frequency, the member amplitude of vibration over that of the casting.
becomes resonant and the amplitude of vibration rises Experience has shown that when steel weldments
to a very high value with disastrous results. It is the are correctly designed to replace castings, the resultant
damping property of the material which enables it to reduction in weight does not create vibration problems.
absorb the energy of the vibrating force. On the contrary, steel has the following superior
The solution to a vibration problem lies in: properties for designing for improved vibration control:
1) Steel can have a damping capacity equal to or Dimensional Stability

greater than cast iron, because an equivalent steel In weldments, tensile residual stresses of the order
design under the same load is stressed higher. of the steel's yield strength may be present in the weld
2) The reduced section of the steel equivalent design area. To balance ou f these tensile stresses, the adjacent
has a greater modulus of elasticity; hence, reduced areas are stressed in compression. .The balance is
amplitude of vibration and, therefore, higher natural achieved through some movement or distortion of the
frequency. This means a greater operating range. member. After the final distortion is reached, no further
3) The design flexibility of welded steel makes it easy movement of the member occurs, If some of the tensile
to shorten unsupported lengths by means of stiffeners or compressive areas are machined out, a new unbalance
and to make other changes that increase natural of the stresses is created and a corresponding movement
frequency and decrease amplitude, even after the or distortion of the member takes place to rebalance
machine has been fabricated, machined and tested. these stresses. This is reflected in the gradual shift of
A few design ideas for reducing vibration are given the member as machining progresses. To avoid this
in Fig. 17.20. At (a), the ends of the rigid member are difficulty, one is advised to stress-relieve a weldment
welded to reduce the amplitude. At (b), a hole is before machining. Experience has shown that a properly
punched in the stiffening panel to reduce the area and stress-relieved weldment remains dead to any further
increase its natural frequency. At (c), the long panel is movement in course of time, once it has been machined
flanged to increase stiffness and natural frequency. At and placed in service, unless acted upon by an external
(d), closed sections or diagonal bracing increases applied force.
torsional resistance, which increases the frequency and Some metals, including steel, are known to creep
reduces the angle of vibration. At (e), tack-welded when stressed over a period of time. The creep factor
stiffeners increase natural frequency and, when placed comes in only when high stress and high temperature
at 45°, add resistance to twist. are present. It need not be considered in the dimensional
Fig. 17.20: A few design ideas for control of vibration

stability of a weldment, when operating under normal member may perform the dual function of bracing the
conditions. frame against torsion and of stiffening the top panel
against deflection under a bending load.
Torsional Loading When stiffeners are used correctly, they perform
A designer must guard against torsional loading, several useful functions, some of which are listed below:
which may cause the m e m b e r to twist about its 1) Rigid stiffeners or diaphragms enable panel to
structural axis. Through welding, torsionally rigid s u p p o r t a greater b e n d i n g load by r e d u c i n g its
sections can be easily developed by the use of stiffeners. unsupported length. This assumes a diaphragm welded
Where torsional loads are a problem, one can use closed at the ends and having sufficient depth to give proper
sections or diagonal bracing and make rigid end support.
connections. 2) Stiffeners added to a flexible panel increase the
combined moment of inertia, reduce its deflection and
Braces and Stiffeners increase its strength under a bending load.
To ensure efficient use of material and reduce cost, 3) Stiffeners added to webs of beams and portions
the designer can use thin plate, which is easily sheared of frames stiffen the w e b against buckling from
and formed and add stiffeners to achieve the required diagonal compressive forces.
strength a n d / o r rigidity. Ribs or diaphragms added to 4) Stiffeners or braces placed at 45° on a panel, or at
a flat panel to strengthen or stiffen it are referred to as 45° to side members of a frame, greatly increase its
stiffeners. When a frame is reinforced or braced by the resistance to twist.
addition of members crosswise or diagonal to the 5) Stiffeners a d d e d to a flat panel increase its
members, the added members are termed as braces. resistance to buckling from edge compression; this is
The two terms are often interchanged, because the same due to the increase in radius of gyration.
Welding Handbook, Chapter 1, Vol. 5, 7th Ed., published by the American Welding Society.
The Procedure Handbook of Arc Welding, by Lincoln Electric Company, U.S.A.
Design of Welded Structures, by The James F. Lincoln Arc Welding Foundation, U.S.A.
Design for Weldments, by The James F. Lincoln Arc Welding Foundation, U.S.A.
Design Ideas for Weldments, Vols. 1 & 2, by The James F. Lincoln Arc Welding Foundation, U.S.A.
Control of Steel Construction to Avoid Brittle Fracture, 1957, by Welding Research Council, U.S.A.
AWS D. 1.84, Structural Welding Code, by The American Welding Society.
Proceedings of International Conference on Structural Design and Fabrication in Shipbuilding, 1976, by The Welding Institute.
Proceedings of the Conference on Fatigue of Welded Structures, 1971, by the Welding Institute.
AWS D 8.4, Recommended Practices for Automotive Welding Design, by American Welding Society.
IS 813-1966, Scheme of Symbols for Welding.
IS 800-1962, Code of Practice for Use of Steel in General Building Construction.
IS 803-1976, Code of Practice for Design, Fabrication and Erection of Vertical Mild Steel Cylindrical Welded Oil Storage Tanks.
IS 806-1968, Code of Practice for Use of Steel Tubes in General Building Construction.
IS 6227-1971, Code of Practice for Use of Metal-Arc Welding in Tubular Structures.
IS 816-1969, Code of Practice for Use of Metal-Arc Welding for General Construction in Mild Steel.
SP:6, ISI Handbook for Structural Engineers.
IS 1024-1968, Code of Practice for Use of Welding in Bridges and Structures Subject to Dynamic Loading.
Fatigue of Welded Structures, by Gurney, published 1980, by Cambridge University Press.
CHAPTER 18
Welding
Procedure
Specifications
A WPS is a document that describes how welding is to are usually permitted on thicknesses, diameters,
be carried out in production. They are recommended welding current, materials, joint types etc.
for all welding operations and many application codes If a WPS is used in conjunction with approved
and standards make them mandatory. welding procedures then the ranges stated should be
WPS should include sufficient details to enable any in accordance with the approval ranges permitted by
competent person to a p p l y the information a n d the welding procedure. However careful consideration
produce a weld of acceptable quality. The amount of should be given to the ranges specified to ensure they
detail and level of controls specified on a WPS is are achievable, as the r a n g e s given by w e l d i n g
dependant on the application and criticality of the joint procedure standards do not always represent good
to be welded. welding practice. For example welding positions
For most applications the information required is permitted by the welding procedure standard may not
generally similar to that recorded on a Procedure be achievable or practical for certain welding processes
Qualification Record (PQR), or also known as Welding or consumables.
Procedure Approval Record (WPAR), except that ranges Typical items that should be recorded on WPS:
A) Descriptions common to all Processes

Procedure number
Process type
Consumable Size, Type and full Codification.
Consumable Baking Requirement if applicable
Parent material grade and spec.
Thickness range.
Plate or Pipe, Diameter range
Welding Position
Welding current and Polarity
Joint Fit Up, Preparation, Cleaning, Dimensions etc.
Backing Strip, Back Gouging information.
Pre-Heat (Min Temp and Method)
Interpass, If Required (Maximum Temperature recorded )
Post Weld Heat Treatment. If Required (Time and Temp)
Welding Technique (weaving, maximum runs, width etc.)
Arc Energy Limits should be stated if impact tests are required or if the material being welded is sensitive
to heat input.
658 MODERN ARC WB iiG TECHNOLOGY
B) Descriptions specific to welding process
Descriptions MIG/
MMA TIG MAG/ SUB
FCAW ARC
Arc voltage yes If Auto yes yes
Pulse parameters (pulse time and peak &c If Used If Used
backgound current)
Welding speed if mechanised yes yes yes yes

Wire configuration yes
Shielding gas (composition, flow rate) yes yes
Purge gas (composition & flow rate) If Used If Used
Tungsten electode diameter and type yes
Nozzle diameter yes yes
Type of flux codification & brand name yes
Nozzle stand off distance
(Distance from tip of nozzle to workpiece) yes
C) Sketches 2. EN 288 Part 2: European Standard For Welding

A sketch of the joint configuration is required which Procedure Specifications
should include the basic dimensions of the weld pre-
paration. Some indication of the run sequence is also ASME Sec IX
beneficial, particularly if the correct sequence is The following sub-articles are intended as a very
essential to ensure the properties of the weld are general guide to ASME Sec IX. There are certain
maintained. exceptions in the code that are not covered in these
articles, therefore it is urged to study the code carefully
D) Non Destructive Testing before u s i n g its rules to carry out any w e l d i n g
qualifications.
A WPS is primarily concerned with welding not N
D T, this activity should be covered by separate N D T Welding qualifications to ASME Sec IX are for use
procedures. with the ASME boiler and pressure vessel code or the
ASME pipework codes.
There are two widely used codes employed for fabri- The ASME boiler and pressure vessel code embraces
its own QA system, which revolves around satisfying
cation:
the rules of the code. This demonstrates that an
1. ASME Sec. IX: American Boiler and Pressure accredited fabricator has some degree of competence
Vessel Code permitting them to perform a large amount of their own
inspection without the involvement of an Independent
WELDING PROCEDURE SPECIFICATIONS 659
A TYPICAL PQR FORM

Weld Procedure Number XXX Rev. A
Procedure Qualification Record (PQR) YYY
Method of Preparation Machine and
and Cleaning Degrease
Manufacturer XYZ Parent Metal Specification Grade 304L
Stainless Steel
Location: Workshop Parent Metal Thickness 3 to 8 mm
Welding Process Manual TIG Pipe Outside Diameter 25 to 100 mm
Joint Type Single Sided Butt Weld Wall Welding Position All Positions
Welding Progression Upwards
Joint Design Welding Sequences
> N| / \ Cm™*|
1-2mm -H H- t
Run Process Size of Current Voltage Type of Wire feed Travel speed
filler metal A V current/polarity speed
1 TIG 1.2 mm 70-90 N/A DC(-) N/A N/A
2 TIG 1.6 mm 80 -140 DC(-)
&Subs
Welding Consumables Production Sequence
Type, Designation Trade Name: ER308L: Cronitig

Any Special Baking or Drying: 1. Clean weld and 25 mm borders to bright
metal using approved solvent.
Gas : Argon 99.99% purity

Gas Flow Rate — Shielding: 8 -12 LPM 2. Position items to be welded ensuring good
Gas Flow Rate — Trailing: fit up and apply purge
Gas Flow Rate — Backing: 5LPM
3. Tack weld parts together using TIG, tacks
to at least 5 mm min length
Tungsten Electrode Type/ Size: 2% Thoriated 2.4 mm Dia

Details of Back Gouging/Backing: 4. Deposit root run using 1.2 mm dia wire
Preheat Temperature: 10°C Min
Interpass temperature: 200°C Max 5. Inspect root run internally
Post Weld Heat Treatment 6. Complete weld using 1.6 mm dia wire us-
Time, temperature, method: ing stringer beads as required
Heating and Cooling Rates*: Not required
7. 100% visual inspection of completed weld
Revision History
Date Issue Changes Authorization
dd/mm/yyyy A First Issue ABC
Inspection Body, which would be required by other The main problem with the ASME code is that it
standards for similar inspections. treats all welding processes separately and imposes
This is why welding qualifications carried out under separate restrictions for each on both procedures and
the ASME code do not require Independent Inspection, approvals. It is therefore prudent to check what the
a fabricator can qualify his own procedures and welders code says about the welding process you intend to use
using the ASME code rules. before you carry out any welding tests.
However there is nothing stopping anybody using The ASME code covers a very wide range of welding
any of the rules of the ASME code without being ASME processes and applications such as hard facing, stud
approved, but to prove your competence in welding welding, brazing etc.
an Independent Inspection Body is recommended to
The code is divided into four parts
approve your welding tests.
The advantage of ASME Sec IX is that its very
workable and very comprehensive. If there is a doubt QW100 General introduction to testing requirements
raised, it is almost certain an answer would be available etc.
somewhere in the code. QW200 Procedure Qualification details.
The ASME code may at first glance appear complex QW300 Performance Qualifications (Welder Approv-
a n d very legalistic b u t once the basic rules are als) details.
understood, they are very easy to apply in practice. Its QW400 Welding Data. This is the biggest section of the
main advantage is that it requires very few welding code and covers data for both procedure and
procedures. performance qualifications. It includes :
• Material and consumable data
The main advantages of ASME welding procedures are:- • Test requirements
• All joint types approve each other, i.e. a double • Approval ranges for thickness, diameter,
sided weld approves a single sided weld and vice joint configuration and welding positions
versa. • Definitions used in the code
• No limits on diameter approval for pipe. A plate
procedures can be used to weld any diameter of
Welding Procedure Specifications (WPS) (QW 482)
pipe.
A sample form is shown on pages 669 and 670
• Materials and welding consumables are listed in
This document details the practical application of
groups, unfortunately only American grades are
the Procedure Qualification Record (PQR). It should
included.
contain enough information to give direction to the
• Any butt weld will qualify all fillet weld sizes welder and should address all variables associated with
on any parent metal thickness. the welding process defined in QW250 including non
• A welding procedure qualified in one position essential and supplementary.
qualifies all positions, unless impact texts are re- A WPS can c o m b i n e w e l d i n g processes from
quired, in which care, procedure qualified in ver- other PQRs but all the relevant variables must be
tical up position qualifies all positions. addressed including parent metal thickness. There
• The code covers wide range of welding processes is an exception to this rule for root runs from PQRs
and applications such as hard facing, stud weld- that are greater than 1.5 inches thick (38.1 mm), see
ing, brazing etc. code for details.
The main advantages of ASME welder approval tests are:- Procedure Qualification Record (PQR) (QW 483)
• There is no minimum parent thickness, only A sample form is shown on pages 671 and 672
maximums, and they are applied to the depos- This document contains details of the welding test,
ited thickness of the weld not the parent metal it must include details of all the parameters listed as
thickness. variables in tables QW250 to QW265 for each process
• There is no upper limit on diameter approval, involved and all the destructive test results.
only a lower limit. The relevant variables for each type of welding
• Any butt weld approval will qualify all fillet weld process are clearly defined in tables QW250 to QW265.
sizes on any parent metal thickness. The left hand column of each table defines the section
and paragraph where each variable and its application materials with the same metallurgical and mechanical
to the table is explained in the code. properties although this is not in strict conformance
with the code. Typically BS1501 151 430A low carbon
Welding Variables steel could be regarded as PI and stainless steels such
Variables used in a welding procedure test are divided as 316, 304 as P8.
into 3 categories :- Note P5, 9 & 10 are divided into sub groups e.g.
5A,5B etc., Treat each subgroup like a separate P
• Essential Variables have a significant affect on Number
the mechanical properties of a joint. They must Dissimilar materials are acceptable providing they
not be changed except within the limits speci- are compatible. For example PI to P8, but this does not
fied by this code. e.g. Material thickness range, cover PI to PI or P8 to P8.
Material Group etc. Note S numbers are for pipework to B31, a P number
• Non-Essential Variables have no significant covers an S number but not the converse.
affect on mechanical properties. They can be
changed without re-qualification of the PQR. Consumables
• Supplementary Variables have an affect on the The ASME code uses its own specifications for
mpact properties of a joint. They are classed as consumables SFA. which is almost identical to the AWS
non-essential if impact testing is not required. specification.
A change in c o n s u m a b l e is only permissible
All variables listed as essential, non-essential or providing it has the same F number and A number (if
supplementary should be addressed on both the WPS applicable) as the PQR.
and the PQR. If any of the variables do not apply to the
particular application then they should be specified as Thickness Limits
not applicable. Thickness limits Groove welds. See QW451 for
precise details.
Joint Configuration • When Impact tests are required the minimum
Either plate or pipe can be used for the test piece thickness approved is restricted. See QW403.6
(plate approves pipe and vice versa ref. QW211), any • More than one PQR may be required to qualify
welding position approves all positions providing no dissimilar thickness
impact tests are required ref. tables QW250 to QW265
and any joint geometry approves all geometry's, e.g. The thickness little 'V of deposited weld metal for
single V, double V, U prep, backed or unbacked. each process involved is approved from 0 to 2xt except:-
A butt or groove weld approves branch and fillet • MIG/MAG (GMAW/FCAW) dip transfer weld
welds but not the converse, ref. QW202. Non-pressure of deposited thickness less than > W approves
retaining fillet welds in pipe or plate can be tested but maximum thickness of 1.1 x t only Ref: QW255
they must be double sided if plate and at least the (QW403.10)
dimensions illustrated in QW462.4a, ref. QW202.2c. • If any Pass in a single or multipass weld > W
Pressure retaining branch welds must be qualified by then the thickness approval equals 1.1 xT
groove (butt) welds. Dissimilar Thickness QW202.4:- The thicker and
thinner part must be qualified, Except P8 and P4X the
Material Grouping thinner part can be qualified if no Impacts and test
Refer Table No. 19.1 on pages 667 coupon > 6mm thick.
Materials are assigned P numbers in QW420; a test Thickness limits for fillet welds as per QW462.4a or
in one P number approves all materials listed under QW462.4d qualify all fillet weld sizes on all base
that P number, except where impact tests are required material thicknesses and all diameters in one test.
then approval is restricted to materials listed in the
group number within the P number. Other P number Testing Requirements (Ref QW 463 for location of
groupings are permissible ref. QW424.1 for details. Ref specimens)
QW 424.1 for further details. There is no requirement for any non-destructive
It is normally permissible if the material is not listed testing such as radiography or MPI/DPI, although
in QW422 to assign it to a P n u m b e r which lists radiography is recommended for butt welds.
The testing requirement for groove welds are as follows:- Diameter and Thickness Ranges
• Two Transverse tensile tests (QW150). Diameter limits for all circular welds including
groove welds, branch welds and fillet welds is given in
• Two Root bends and Two face bends unless the QW452.3. There are no upper limits on diameters
plate thickness exceeds 3/8" then 4 side bends approved and pipe covers plate
are required. All bend tests should be done to For branch welds the diameter considered for the
QW160 using the correct former ref. QW466 to above limits is the one containing the weld preparation.
an angle of 180 degrees. Longitudinal (all weld) Thickness limits, groove welds.
bend tests are not recommended unless the The thickness limit only applies to the deposited
base/weld materials differ markedly in bending weld metal thickness not the plate thickness and any
properties. See QW 466 for exceptions and pre- groove weld approves all fillet weld sizes.
cise details. For t greater than 12.5 mm there is no restriction on
The testing requirement for fillet welds on plate is 5 the size that can be welded (providing the test weld
macro sections only, for Pipe fillet welds 4 macro deposit contains at least 3 layers of weld).
sections. No fracture test required. Thickness limits, fillet welds.
A test on plate greater th<u\ 3/16" approves all base
Welder Performance Qualification (WPQ) metal thicknesses and fillet weld sizes ref. QW452.5.
(QW484) (Note the above diameter limits apply unless the fillet
4 sample form is shown on pages 673 & 674 weld is qualified by a groove weld)
M terials
I he purpose of this test is to determine the welders Joint Configuration
ability to deposit a sound weld therefore the base Joint geometry, a double V (or U) is considered the
material is not considered as critical as it is in the PQR. same as a joint with backing and does not qualify a
HL*i $e a performance test on any material in P groups single V (or U) without backing, but a single full
1 to 11 approves all those groups and sub groups, also penetration joint without backing qualifies all joint
P34 and P4X (P40-P49). Providing a compatible configurations.
consumable exists with the same F number used in the
qualification test (QW423.1) Approval Range
Note a single sided weld is classed as a weld without Extent of approval is very well explained in
backing and a double sided weld or weld with sealing QW461.9. Take particular note of welding positions
run is classed as a weld with backing. which are also explained in QW461, for example to
qualify a fillet weld in the normal horizontal-vertical
Consumables position with a groove weld, the groove weld must be
The F number cannot be changed without re qualified in at least the 2G position. The welding
qualification of the welder except that for performance positions defined in QW461.1.& QW461.2 should be
qualification only using SMAW (MMA) F numbers up referred to in the WPS. The position designations: 1G,
to and including 4 approve all lesser F numbers for 2G, 3G, 4G, 5G, 6G (Groove Welds) and IF, 2F, 3F, 4F
double sided or welds with backing only. One (Fillet Welds) are test positions
Consumable from F41 To F45 approves any of these
consumables, except SAW. Ref. QW404.11. Period of Validity/Renewal of Qualifications (QW
Note 'A' numbers do not apply to welder approval 322.2)
tests. Provided the welder uses the process for which he
is qualified and there is no reason to question his ability
Variables then his qualification lasts indefinitely.
For each welding process there is a list of essential If the welder does not use the welding process for
variables in QW352 to QW357 and QW360 for welding which he is qualified for a period of 6 months or more
operators, these are not necessarily the same as the ones then he must perform a new test in pipe or plate, any
for the PQR. Essential variables cannot be changed. parent material, thickness and position, if successful
Explana-tions of all these variables is given in section all the welder approvals for that welding process are
IV of the code. renewed in one test.
Testing Requirements Min Deposit Size Approved:- Point Where Chemi-

Test requirements for groove welds QW452 consists cal analysis taken No upper limit QW402.16 (462.5a)
of either:- Welding Positions QW405.4 Performance Qualific-
• One face bend and one root bend except for ation approves all deposit thickness' No min.QW381
welding positions 5G & 6G which require 4
bends (Ref QW452.1 Note 4). If the plate ex- ASME Definitions, Consumables, Welding Positions
ceeds 3/8" side bends may be used. See QW ASME has adopted their own designation for welding
466 for precise details and exceptions. processes
Note:- Bend Tests can in most cases be replaced Designation Description
by Radiography {See Below}.
OFW Oxyfuel Gas Welding
• Radiography is optional and must be supple- SMAW Shielded Metal Arc Welding (MMA)
m e n t e d by b e n d tests w h e n using GMAW SAW Submerged Arc Welding
(MIG/MAG) with dip transfer (Short Circuit- GMAW Gas Metal Arc Welding (MIG/
ing Arc) or when welding some special mate- MAG)
rials. Ref. QW304. FCAW Flux Cored Wire
Note:- Ultrasonic Examination in lieu of Radi- GTAW Gas Tungsten Arc Welding (TIG)
ography is not permitted PAW Plasma Arc Welding
Test r e q u i r e m e n t s for fillet w e l d s in p l a t e ref. Straight polarity - Electrode -ve Reverse polarity = Electrode +ve
QW452.5:-
• One macro section (QW 184) and One fracture F No. (QW 432)
test (QW182).
The location where each specimen has to be taken is F Number General description
defined in QW463 1 Heavy rutile cosated iron powder electrodes :-
Radiography Ref QW 191 A5.1 : E7024
• A length of at least 6" must be examined for 2 Most Rutile consumables such as :- A5.1: E6013
plate or the entire circumference for pipe. 3 Cellulosic electrodes such as :- A5.1 : E6011
• If the pipe circumference is less than 6" then 4 Basic coated electrodes such as : A5.1 : E7016
more samples must be welded u p to a maxi- and E7018
m u m of 4. Ref QW 302.2. 5 High alloy austenitic stainless steel and duplex
Visual Examination Ref QW 302.2 & QW 190 A5.4 : E316L-16

Performance test coupons must show complete 6 Any steel solid or cored wire (with flux or metal)
joint penetration with full fusion of the weld metal 2X Aluminium and its alloys
and base metal. 3X Copper and its alloys
The w e l d e r p e r f o r m a n c e test m u s t follow a 4X Nickel alloys
properly qualified WPS. Once qualified the welder 5X Titanium
must always work within the extent of approval of 6X Zirconium
Zirconium
any properly qualified WPS and his WPQ. 7X Hard Facing Overlay
The w e l d e r w h o qualifies the PQR is a u t o - Note:- X represents any number 0 to 9
matically approved within the limits specified in
QW304, QW305 and QW303. Ref QW301.2. A No. (QW442)
These refer to the chemical analysis of the deposited
Specialist Processes weld and not the parent material. They only apply to
Such as corrosion resistant overlay or hard facing welding procedures in steel materials.
are covered in QW 453.
Procedure variables are defined with all procedure Al Plain unalloyed carbon manganese steels.
variables in QW252 and in QW380 for welder approval. A2 to A4 Low alloy steels containing Moly and
Min base thickness approved = size welded or 1", Chrome Moly
QW 453 A8 Austenitic stainless steels such as type 316.
Welding Positions Welding Positions QW461.1 and QW461.2

Note the welding progression (vertically upwards Basically there are three inclinations involved.
or downwards), must always be stated and it is an
• Flat, which includes from 0 to 15 degrees inclination
essential variable for both procedures and performance
qualifications. • 1 5 - 8 0 degrees inclination
• Vertical, 80 - 90 degrees
Welding positions for groove welds:- For each of these inclinations the weld can be rotated
Welding position Test position from the flat position to horizontal to overhead.
Flat 1G
P Material Numbers (QW420)
Horizontal 2G This is a general guide ASME P numbers and their
Vertical Upwards Progression 3G equivalent EN288 groupings. Groups referred to in the
Vertical Downwards Progression 3G Base Metal column are ASME sub groups. EN288
Overhead 4G material groups are included for comparison only.
Pipe Fixed Horizontal 5G
EN 288:
Pipe Fixed @ 45 degrees Upwards 6G EN 288 is the European Standard that covers the
Pipe Fixed @ 45 degrees Downwards 6G specification and qualification of welding procedures
for fusion welding. This standard at present consists of
Welding positions for fillet welds:- 9 parts.
Welding position Test position EN 288 Part 1: General Rules for Fusion Welding:
This part introduces the EN288 Series of standards, it
Flat (Weld flat joint at 45 degrees) IF includes definitions and a brief description of its other
Horizontal 2F parts
Horizontal Rotated 2FR EN 288 Part 2: Specification and Approval of
Vertical Upwards Progression 3F Welding Procedures for Metallic Materials: This part
Vertical Downwards Progression 3F of the standard defines the contents of a Welding Pro-
Overhead 4F cedure Specification in the form of a list of information
Pipe Fixed Horizontal 5F which should be recorded.
fillet W e l d *
K I I I K MtsvMlhiia
I irtriT'ftrttal orlzontal
Note. No Flat or Horizontal Position «tthl# Elevation
15*T©8Cf
ftitlts FfKVttliitti
I
Vertical
an TII «ur AH Positions Vertical

hJi«lc fcJcvuliun
EN 288 Part 3: Welding Procedure Tests for the Arc • Customer specification.
Welding of Steel: This is the most familiar and widely • Examiner/Test Body requirements. This is the
used method of qualifying welding procedures. It p e r s o n or o r g a n i s a t i o n a p p o i n t e d by the
involves the welding of a test piece representing the contract-ing parties to approve the procedure.
p r o d u c t i o n w e l d in all essential features. The The use of a reputable third party inspection
completed test weld is then subjected to a variety of Authority is strongly recommended for the
specific tests to ensure that the properties of the weld approval of all pro-cedures, as this may prevent
will be acceptable for most stringent applications. the test having to be repeated to satisfy future
EN 288 Part 4: Welding Procedure Tests for the Arc contract requirements.
Welding of Aluminium and Its Alloys: This is the
• The load carrying requirement of the weld and
Aluminium version of EN 288 Part 3.
the consequence of failure.
EN 288 Part 5: Approval by the Use of Approved
Consumables: An assumption can be made that a weld • Special welding requirements such as pre-heat,
carried out using welding consumbles, approved by post weld heat treatment, controlled heat input, etc.
an inspection body, in accordance w i t h the For example, pressure vessels built to PD5500
manufacturer's recommended welding practices will require all welding qualifications to be approved by a
produce welds suitable for simple fabrication work procedure test to EN 288 part 3 or part 4.
without the need for any formal testing.
EN 288 Part 6: Approval by Previouse Welding Ex- Welder Approval Tests
perience: If the same system of welding has been used The European Standards for Welder Approval are
satisfactorily on welded fabrications for many years, it EN 287 Part 1 for steels and EN 287 Part 2 for
can be assumed that the welds thus produced are ac- aluminium and its alloys. They only cover welding
ceptable. Unfortunately we know from our knowledge processes that are manual or partly mechanised .ie.
of linear elastic fracture mechanics that a fine line of- where the skill of the welder has a significant influence
ten exists between a defective weld providing accept- on weld quality. Materials to be joined can be wrought,
able service life and one that causes a serious failure. forged or cast.
Therefore this method should be used with caution. Mechanised and Automated welding is covered by
EN 288 Part 7: Approval by a Standard Welding EN1418
Procedure: Fully documented procedures may be There is an option (non-mandatory) for a job
purchased, from reputable authorities, for joining knowledge test (refer to appendix D of the standard)
materials that are relatively easy to weld such as plain to be carried out.
carbon and austenitic steels. This route will be useful
to companies which do not wish to get involved with EN 288 And EN 287: Documentation
qualifying their own procedures. This is a general guide on how the certificates for
EN 288 Part 8: Approval By A Pre-Production EN288-3 and EN287-1 should be completed.
Welding Test: If a particular joint geometry cannot sat- Application standards, particularly those in other
isfy all the requirements of EN 288 Part 3 or Part 4 be- European countries may impose additional rules and
cause of its size and testing requirements then a spe- restrictions on the qualification of welders and welding
cial test can be carried out representing the joint in all procedures.
respects and subjected to as much of the testing required
by EN 288 Part 3 or Part 4 for a similar joint type. Welding Procedure Approval Test Certificate, Form
BS EN 288-9: Welding procedure test for pipeline E1
welding on land and offshore site butt welding of A sample form is shown on pages 675
transmission pipelines: This is the most recent The numbers on the left hand side below refer to the
addition to EN 288. line numbers that appear down the left hand side of the
The chosen method of procedure approval should El form. The section and page numbers quoted below
take account of the following points: all refer to EN 288 Part 3.
• The requirements of the application standard. • 3 to 6 The Welding procedure reference number
This is the standard that specifies the welding can be any combination of numbers and letters
qualification requirements for fabricated items that are unique to your company The inspection
such as boilers, pressure vessels or pipework. authority reference number is a unique number
issued by some Inspection Authorities, it is not a Groups 4 to 9 approve other materials in the same
mandatory requirement. The Manufacturer group with the same or lower alloy content. The major
named on this form should be the organisation alloying elements should be simply summated for
that is in control of the welding operation (Ref comparison. Therefore a 316 stainless steel will approve
EN 288 Part 3 Section 8.2). a 304 but not the converse {Note this may not be per-
• 7 The code/testing standard: EN288-Part 3 mitted if it involves a significant change in consumable
composition ref 8.4.4) e.g. For a 316L stainless steel
• 8 Date of Welding: or the date when the weld
state:
was produced.
• Type 316L and lower alloyed steels in group 9.
• 9 Range of approval: this applies from line 10
For alloy steels the alloy content may be stated e.g.
to line 20 (This line is a heading).
• 2% Cr 1% Mo and lower alloyed steels in group 5.
• 10 Welding process: state process used, if more
• 13 Parent metal thickness: Section 8.3.2
than one process specify which performed the
root and which performed the capping welds. A test piece of thickness 'f between 3 to 12mm
will approve thicknesses from 3mm to 2 x t.
• 11 Joint type: Section 8.4.3. All joint configura-
Above 12mm the approval range is 0.5 x t To 2 x t
tions approve fillet welds on plate or pipe. Single
sided full penetration welds without backing or Single run welds made from one or both sides
back gouging approve all butt welded configu- for Butt, Branch and T Butt welds approve only
rations with similar root details except branch 0.8 x t to 1.1 x t. This rule does not apply to fillet
connections. Branch Connections may require welds.
separate approval depending on the require- Single Run Welds Don't Approve Multi-Run
ments of application standards and contracting Welds and Vice Versa, this rule applies to all joint
parties. types including fillet welds
There are two types of double sided welds; with Fillet welds; state the range of throat thickness;
gouging and no gouging, they should be correctly 0.75 x a to 1.5 x a, a = throat thickness (approx.
specified as they have different approval ranges. 0.7 x Leg length). The above range of parent metal
thickness also applies; refer to section 8.3.2.3. If
A Guide To Parent metal Grouping a fillet weld is qualified by a butt weld then the
• 12 Parent metal groups: Section 8.3. The parent throat thickness 'a' will equal the depth of the
material must be put into the appropriate group, weld deposit in the butt weld.
see table no. 19.2 on page 668 for details. If the • 14 Pipe Diameter: Section 8.3.2.4.
material will not fit into any of the 11 groups then • D < 168.3 approves 0.5xD to 2xD
it is classed as a group on its own and approves
no other material. If it can be assigned to more • D = > 168.3 approves 0.5 x D to Plates
than one group then it goes in the lower group. A Note for Branch connection the diameter is the
higher group number does not approve lower branch pipe. If the branch is greater than 2 / 3 of
groups except group 2 approves group 1. the main pipe diameter the joint geometry will
Impact tests One set from H.A.Z. and one set from change significantly, therefore a separate test
weld metal for material less than 50mm thick. The ac- should be considered. Branch connection may
ceptance criterion is that of the application standard, consist of a pipe to plate subject to above rule.
refer to section 7.4.4. It is generally accepted that all pipe diameters
Groups 1 to 3 only approve other materials in that cover plate all though this is not very clear in the
group with the same or lower minimum specified yield standard. Plate still covers pipe 500mm diam-
stress {Note this the minimum value quoted by the eter and above.
materials specification and not the actual value of the • 15 Filler Metal Type/Designation: Sections 8.4.4,
material. However, it is intended for this rule to be 8.4.5, 8.5.1. Specify the codification of the con-
applied precisely but rather the general philosophy sumable used and what standard it is coded to.
behind it be adopted.}, e.g. For a BS1501-151-430A state: If more than one type of consumable is involved
• Group 1 with a yield stress less than or equal define which part of the weld each is responsible
to 250 N/mm 2 . for. e.g. BS 639 E43 22 C 2.5mm (Root weld), BS
639 E43 23 R 2 to 3.2mm (Cap weld). The con- unless more than one test piece is welded, e.g. If
sumable type used must be compatible with the a plate is welded in the OH position and another
material to be welded and have the same nomi- identical plate is welded in the Vertical Up posi-
nal composition or be in the same group of ten- tion then: all intermediate positions are approved
sile properties as the type used in the qualifying providing:- The hardness survey is taken form
procedure test. MMA electrode coating types can the overhead plate and impacts from the verti-
not be changed and if impact tests are required cal one, each plate weld must be subject to Radi-
the brand of the electrode can not be changed ography, Surface Crack Detection, one Tensile
without further impact tests. and two Bend tests.
Note the range of approval for a MMA electrode • 19 Pre heat: Section 8.4.8. If no pre-heat is used
diameters is + and - one diameter, except for open state the actual ambient temperature. (Guidance
root runs (single sided without backing) then the on pre-heat may be obtained from BS 5135). If
diameter can't be changed Standard diameters there is any interpass requirements then these
are: 1.6, 2, 2.5, 3.2, 4, 5, 6 mm. should also be stated.
• 16 Gas/Flux: Section 8.5. Shielding gas and back- • 20 Post weld Heat Treatment: Section 8.4.10.
ing gas composition or submerged arc flux speci- Specify temperature range. Note if the applica-
fication. eg. Shielding and backing gas Argon tion requires post weld heat treatment then the
99.9% pure. procedure test piece should be heat treated un-
der similar conditions. A procedure with post
• 17 Type of Welding Current: Section 8.4.6. State
weld heat treatment should not be used for ap-
type used in test .e.g. AC or DC+ve or DC-ve or
plication that do not require it.
pulsed current. Note one type of welding cur-
rent does not approve another. Welding Procedure Approval Test Certificate, Form E2
• 18 Welding Positions: Section 8.4.2. This section A sample form is shown on pages 676
states that for an all position qualification the This form should state what took place on the day
hardness survey must be taken from the area of of the test, all the values stated should be actual values
lowest heat input and the impact tests from the or machine settings recorded during the test. A simple
area of highest heat input. way to record the data is to fill in an E2 form as the test
This implies that restrictions exist on heat input progresses and type it up later. A continuation sheet
even when controlled heat input is not required. may be appended if required. If more than one process
Other factors may also affect heat input such as is involved the use of another E2 sheet may be an
a change from a stringer bead to a weave which advantage.
will increase heat input significantly The material specification must be stated in full. e.g.
No hardness or impact requirements, e.g. Aus- to state 316 is not acceptable as there are several 316
tenitic stainless steels State: All positions. (Pro- austenitic stainless steels with tensile strengths ranging
viding the welding process and consumable is from 490 to 580 N / m m 2 . It is advantage to append the
capable of being used in all positions). material certificate to this form.
Hardness requirement only, apply the following Test Results, Form E3

rules :- A sample form is shown on pages 677
Test position. Approved positions This form gives summary of all the test results. It is
similar to the back of the WP form El. Laboratory report
Flat Flat , Vertical U p ; (Horizontal
for future reference should be retained.
vertical for fillet welds only)
Vertical Up Vertical Up only Welder Approval Test Certificate, Form E4
Horiz-Vertical Horizontal-Vertical, Flat a n d A sample form is shown on pages 678
Vertical Up
Overhead All Positions except vertical down. For a welder approval test it is only necessary to
Pipe Horizontal All Positions (except v.down) complete form 'E4'. The section and page numbers
quoted below all refer to EN 287 Part 1.
Hardness and Impact Requirements Approve • 2 Designation: Section 12. This is as an aid to
only the welding positions covered in the test the fabricator to keep track of welders qualifi-
cations and therefore not to be considered man- o For Diameters greater than 150 mm OD there
datory. is no upper limit but the lower limit of 0.5 x
• 5 Welding Procedure Reference Number: The D still applies.
number of the qualifying procedure or W.P.S. o The minimum diameter that can be approved
• 16 Joint Type: Section 6.3 and table 3. The is 25 mm, for diameters below 25 mm its the
approval range need only cover the range that diameter welded e.g for a welder approval
concerns the fabricator. Single sided full test in pipe 30 mm diameter butt welded the
penetration welds w i t h o u t backing or back approval range is range is 25 mm to 60 mm.
gouging approve all butt welded configurations o For branch connection D is the diameter of
and a single sided pipe butt without backing the branch.
approves a branch connection but not for the • 23 Welding position: Section 6.8 and table 7. The
procedure test. All joint configurations approve 6G position for pipe, now referred to as HL045,
fillet welds on plate or pipe. approves all positions except vertical down. A
• 17 Parent metal Group: Section 5.4, section 6.4 plate inclined at 45° is not a recognised position
and annex A in EN 287.
As a general rule:- (refer to the standard for more
Welding Positions For Groove welds:
precise details)
o W01 Unalloyed plain carbon/carbon manga- Welding position EN
nese steels with a yield stress <355 N / m m 2 Flat PA
o W02 Alloyed Steels; Cr Mo, Cr Mo V etc. Horizontal PC
o W03 Fine grained structural steels. Vertical Upwards Progression PF
Vertical Downwards Progression PG
o W04 Ferritic stainless steels.
Overhead PE
o W l l Austenitic stainless steels. PF
Pipe Fixed Horizontal
There is no equivalent to BS 4871 Group W4, Pipe Fixed @ 45 degrees Upwards HL045
High nickel alloys. Pipe Fixed @ 45 degrees Downwards JL045
• 18 Filler Metal: Section 6.5 and section 5.5. State
full designation .e.g. BS2901 Part 2 316S92. Welding Positions For Fillet welds:-
• 19 Gas/Flux: Section 5.5 and section 6.6 Welding position EN
• 21 Material Thickness: Section 5.6 Flat (Weld flat joint at 45 degrees) PA
A test piece of thickness Y between 3 to 12 mm Horizontal PB
will approve thicknesses from 3 m m to 2 x t. Horizontal Rotated PB
Above 12 mm the approval range is 5mm and Vertical Upwards Progression PF
above. Vertical Downwards Progression PG
Overhead PD
For dissimilar thicknesses Y is the thickness that
Pipe Fixed Horizontal PF
controls the size of weld deposit, usually the thin-
ner part. • 24 Gouging and backing: Section 6.3 and Table
Fillet weld leg lengths; EN287-1 does not specify 3. State if backing was used for single sided welds
any approval range for fillet weld dimensions, it and if gouging was used for double sided welds.
recommends sizes for the test in Section 7.2. {Figs. Also state if a purge or backing gas was used.
4 and 6}. However the size must be specified in • 25 Additional W.P.S.: State the W.P.S. number if
the WPS before the test begins and represent the different from the procedure number in line 5.
weld sizes to be used in production, Ref. Section
7.3. • 33 Validity of Approval: Section 10. The date
should be 2 years from the date of issue. Then
• 22 Pipe outside diameter: Section 5.6 the certificate must be re-validated by the an
o A test piece of Diameter D approves Diam- examiner test body by providing evidence of
eters in the range 0.5 x D To 2 x D volumetric tests carried out on the welders
production work. This will involve evidence of updating. Results of visual examination and / o r
radiography, ultrasonics or fracture testing in the surface N D T will n o t be sufficient. If the
form of radiographs or reports. These must be inspection b o d y is satisfied then it signs the
traceable to the welder and taken at roughly six welder approval certificate making it valid for a
monthly intervals, i.e. the basis for employer further 2 years.
Table 19.1
PNo. Base Metal
Carbon Manganese Steels, 4 Sub Groups
• Group 1 up to approx 65 ksi
1 • Group 2 Approx 70ksi
• Group 3 Approx 80ksi
2 Not Used
3 3 Sub Groups:- Typically half moly and half chrome half moly
4 2 Sub Groups:- Typically one and a quarter chrome half moly
5A Typically two and a quarter chrome one moly
5B 2 Sub Groups:- Typically five chrome half moly and nine chrome one moly
5C 5 Sub Groups:- Chrome moly vanadium
6 6 Sub Groups:- Martensitic Stainless Steels Typically Grade 410
7 Ferritic Stainless Steels Typically Grade 409
Austenitic Stainless Steels, 4 Sub groups
• Groupl Typically Grades 304, 316, 347
8 • Group 2 Typically Grades 309, 310
• Group 3 High manganese grades
• Group 4 Typically 254 SMO type steels
9 A, B, C Typically two to four percent Nickel Steels
10A,B,C,F,G Mixed bag of low alloy steels, 10G 36 Nickel Steel
10 H Duplex and Super Duplex Grades 31803, 32750
10J Typically 26 Chrome one moly
11A Group 1 9 Nickel Steels
11 A Groups 2 to 5 Mixed bag of high strength low alloy steels.
11B 10 Sub Groups:- Mixed bag of high strength low alloy steels.
12 to 20 Not Used
21 Pure Aluminium
22 Aluminium Magnesium Grade 5000
23 Aluminium Magnesium Silicone Grade 6000
24 Not Used
25 Aluminium Magnesium Manganese Typically 5083, 5086
26 to 30 Not used
31 Pure Copper
32 Brass
33 Copper Silicone
34 Copper Nickel
35 Copper Aluminium
36 to 40 Not Used
41 Pure Nickel
42 Nickel Copper:- Monel 500
43 Nickel Chrome Ferrite:- Inconel
44 Nickel Moly:- Hastelloy C22, C276
45 Nickel Chrome :- Incoloy 800, 825
46 Nickel Chrome Silicone
47 Nickel Chrome Tungstone
47 to 50 Not Used
51, 52, 53 Titanium Alloys
61,62 Zirconium Alloys
Table 19.2
EN288 Base Metal
Carbon Manganese Steels, 4 Sub Groups

• • Group 1 up to approx 65 ksi
1 • Group 2 Approx 70ksi
• Group 3 Approx 80ksi
4 3 Sub Groups:- Typically half moly and half chrome half moly
5 2 Sub Groups:- Typically one and a quarter chrome half moly
5 Typically two and a quarter chrome one moly
5 2 Sub Groups:- Typically five chrome half moly and nine chrome one moly
6 5 Sub Groups:- Chrome moly vanadium
8 6 Sub Groups:- Martensitic Stainless Steels Typically Grade 410
8 Ferritic Stainless Steels Typically Grade 409
Austenitic Stainless Steels, 4 Sub groups
• Groupl Typically Grades 304, 316, 347
9 • Group 2 Typically Grades 309, 310
• Group 3 High manganese grades
• Group 4 Typically 254 SMO type steels
7 Typically two to four percent Nickel Steels
— Mixed bag of low alloy steels, 10G 36 Nickel Steel
10 Duplex and Super Duplex Grades 31803, 32750
— Typically 26 Chrome one moly
7 9 Nickel Steels
— Mixed bag of high strength low alloy steels.
— 10 Sub Groups:- Mixed bag of high strength low alloy steels.
— Not Used
21 Pure Aluminium
22a Aluminium Magnesium Grade 5000
23 Aluminium Magnesium Silicone Grade 6000
— Not Used
22b Aluminium Magnesium Manganese Typically 5083, 5086
— Pure Copper
— Brass
— Copper Silicone
— Copper Nickel
Copper Aluminium
— Pure Nickel
— Nickel Copper:- Monel 500
Nickel Chrome Ferrite:- Inconel
— Nickel Moly:- Hastelloy C22, C276
— Nickel Chrome:-Incoloy 800, 825
— Nickel Chrome Silicone
— Nickel Chrome Tungstone
— Titanium Alloys
— Zirconium Alloys
QW-482 SUGGESTED FORMAT FOB WELDING PROCEDURE SPECIFICATIONS (WPS)

(See QW-200,1, Section IX, ASME Boiler ami Presswe Vessel Code)
Company Name„ By:.
Welding Procedure Specrficabon No.„ . Supporting PQft No4s|«
Revision No Date.
Welding Processes) Typeis)

CAatom*r»c Mamsai. fWUtchin*, or S«mrAmtoJ
JOINTS CQVM02J Details

Joint Design —
Backing (Yes) .(Noh
Sacking Material (Type)
ffefar tt» bo«fe twcfctog and iMrimnkJ
O Metal O Nonfustng Metal

D NonmetaWc D Other
Sketches, Production Drawings, Weld Symbols or Written Description
should show the general arrangement of the parts to be welded. Where
applicable, the root spacing and the details of weld groove may be
specified.
(At the option of the Mfgr, sketches may be attached to Illustrate joint
design, weld layers and bead sequence, e.g., for notch toughness proce-
dures, for multiple process procedures, etc)
►BASE METALS (GW-403)

P*#fc>. . Group No. ..toP-No,
OB
Specification type and grade .
to Specification type 9nd grade
Oft
Cham, Analysis and Mech Prop. —
to Chem. Analysis and Mech. Prop.
Thickness Range:
Base Metal; Groove
Other
•RLLEf* METALS (QW-404)

Spec No. (SPA)
AWS No. (Das*)
F-N©
A~No
Size of Filler Metals,
Weld Metal
Thickness Range:
Groove
Fillet
Electrode-Rux (Class).
RUJC Trade Name .
Consumable Insert
Other ..
•Each base metal-filler metal combination should be recorded individually

QW-482 (Back)
WPS No ft*».
P0STWELD H6AT TREATM€f*T {O*«407>
POSITIONS *CWM05J
Pmitkwkl o* Groom*. T e m p e r a t u r e Fl»r*ge —. ,—
***&<*§ Pfogre$s*Of»: U*»_ ». Ocwwft „. Time Bangfe ,—■. :
GAS <O*-408}
Percent C o m p o s i t i o n
P R £ B £ A T IQW-#QSI
G«tlc»1 {Mixture} Plow R e t *
Pr«fc*»t T « m p . Mta._
f ni*tpm* T«mp, M»x..
ftrthmt M**m*ttane« - Shrekfcng
( C o r n e o u s or special NNrttng *rf*ere a p p l i c a b l e sfcowki b e r e c o r d e d ) Trailing
Backing
Frflar Metal
Offter
le.g,, Jrtonarfcs. C o m
Tiawef m e o t f . H o i Wfwne
Weld Type Amp Volt Speed A d d m o n , Technique,
Layer b ) Process Polar Range Range Torch Ai»0e. 6 * c J
QW-483 SUGGESTED FORMAT FOR PROCEDURE QUALIFICATION RECORDS (PQR)

(See QW-200.2, Section IX, ASME Roller and Pressure Vessei Code)
Record Actual Conditions Used to Weld Test Coupon.
Company Name ,
ProcedureQuaM*c*t*Oft Record No _ D#t*_
*/*$ Ho ,
Wefdtng *>ocesstes* „ _
Type* *Mam*af, Automatic. Sem*-AutoJ
.JOINTS tOW 4021
Groove Oes«*m of 1 Cou0t%„

IFor combmatton «auaM*cat<on»# t!*e deported *e*d metal tn«* tie *«^ft*ed »o» each tmer metal or process usacM
BASE M6TAC5 <OW~4e3t TREAT|y|6f4T iCWM07J
Mater «a< Spec
Type or Grade
PNo. to P-Ho.
Thickness Q* Test Coupon .
Diameter of Test Coupon _
Other _
IQMM08)
Percent Composition
Gastesl <M<xture) Flow Rate
Shielding
Trading
F l t L E R METALS {CMr-404) Backing
$F A Specification -
AWS Ctatstt >cat*o#> . ELECTRICAL CHARACTERISTICS <QW^09»
Filler Metal F N o Current _ - _ . .., ._,.... _
Wtkt Meta* Analysts A-No. Polarity . „
S**e of Filler Metal Amps _ Vo»t5^__
Other „ ^,„ y *~ w „„„^ TO ^,, ! ,„„„ Tungsten Electrode Sire _ _ _ _ _ _ „ _ _ _ .
Weld M e t a l Thickness
POSITION iOWÔSi TECHNIQUE ^QW-4101

Position ol Groove.-.....™,,.....,.. ,,_ Travel Speed .„ ..,, , _ , _
*¥e4d Progression iUphrfi, Ocwvnr>*lH_ String or Weave Bead _ „ „ _ _
Other___ : .„...„.„,„. Osc«Hatton_ _
Multipass or Single Pass ipe* stde)^
Single or Multiple Electrode* _..
PREHEAT IQW-406) Other
Prei>eaf Temp
Inter pass Temp,. „_
Othe* _ . _ _ _
QVV-483 (Back)
PQRMo.
Teneie Test <OW 150)
Ultimate Ultimate Type of I
I Spactn*en Total toad Unit Stress Fafture*
Ho. Width Thickness Area lb PSi IxRaatjofi
Guided-Beiid Tests tQW-t*J0t
[ Type and Figure No. Result j
-A. I
^
'^^r
Tests (QttMTO)
Notch : Specimen Test IM^tVaJues^

I Specimen
j Mo. Location Sue Temp. Ft. I b l ^ JlShear ^ I ^ M i t e Drop Weight Break fY/WJ |
S ^ iP
f -•#
t »
1
_ â
F»*t Weld Test fOW t * 8
Ve*_ Pewetretioii into Pererit Metal? TetL.
Other Tests
Type of Test
Other
Welder's Name ,Cloc* No , . Stamp No -

Tests conducted by: ««_____. „ .Laboratory Test No.
We certify that the statements m this record are correct and that the test wetds were prtpated. welded, and tested m accordance with 1
requirements of Section IX of the ASME Code
Manufacturer «
&*» " By , ,
(Detail of record of tests am iiiustratrve oofy mnd may be modified to conform to the type and number of tests required by the Code)
QW-484A SUGGESTID FORMAT A FOR WELDER PERFORMANCE QUALIFICATIONS (WPQ)

(See QW-301, Section IX, ASME Rotor and Pressure Vessel Code)
Welder's name, .Identification no..
Identification of WP$ followed. G Test coupon O Production weld

Specification of base metaKs) - Thickness.
Testing Conditions and Quttrftcattoo limits
\f QW-350) (%ftmal Vatvee Reage QMeWied

Welding processies)
Type tie: manual, semi-auto) used
Backing (metal, weld metal, douWe-wekled, etc)
Q Plate O Pipe (enter diameter if pipe or tube)
Base metal P ot S*Nomber to P- or S-Number
RtJer metal or electrode specification^) (SFA) (info only)
Filler metal ot electrode classiftcationte} {info only)
Filler metal F-Numberts)
Consumable Insert (GTAW or PAW)
Rller type (solid/metal or flux coreoVpowder} (GTAW or PAW)
Deposit thickness for each process
Position qualified f2G, 6G, 3F, etc,)
Vertical progression (uphill or downhill)
Type of fuel gas {OFWJ
Inert gas backing (GTAW, PAW, GMAW)
Transfer mode (spray/globular or pulse to short rircurt-GMAW)
GTAW currem type/polarity (AC, 0C£P. DC£N>
Visual Examination of Completed Wekt (QW-302.4)

O Bend test; O Transverse root and face iGJ|0ê.3(a)j
(a)»^rLongrtudinal
Longitudinal ro<
root and face (QW-462.3<b)); D Side (QW-462.2);
Q Pipe bend specimen teyl|W462 5(c)j; O P
Q Plate bend specimen, <XHTc«kMwesistarrt overlay (QW-462-5(dft
G 9ft» [QW-462.5<b)];
'-462.5(b)]; O Macro
I test for fusion [GW^62.5{e}]
1 Type Result W 0 ^ Ty
P* *e»* Type Result I
Alternative radiograph* examination results (QW-191)

Fillet weld - fracture test IQW-1801 . Length and percent of defects .
Macro examination IQW484) Rllet size |ln.| „ . Concavity/convexity (In.) „
Other tests .„ , .,,. .,
Film or specimens evaluated by „ . Company.
Mechanical tests conducted by _. . Laboratory test no.,
Welding supervised by
We certify that the statements in this record are correct and that the test coupons were ptep&rod, welded, and tested in
requirements of Section IX of the ASME Code.
Organization.
Date By-
This form (E0Q08A) may be obtained from the Order Dept, ASME, 22 l a w Drive, Sox 2300, fmriioki, H I 07007-2300.
OW-4S48 SUGGESTED FORMAT B FOR WELDING OPERATOR PERFORMANCE QUALIFICATIONS (WOPQ)

(See QW-301, Section IX, ASME Boiler ami Pressure Vessel Code)
Welder's name. . Identification no,

Test Description \tn(knuwikfii Onfy)
Identification of WPS followed . D Test coupon O Production weld
Specification of baseroetaMs)_ . Th#cfcness.
Base metal P or S-Nurober . . to P or S-Number. „ Position (2G, 6G, 3F, e t c ) .
Q Hate D Pipe (enter diameter if pipe or tube):
Filler metal (SFA) specHtcatioo Filler metal or electrode classification .
Testing Conditions and Qualification limits When Using Automatic Welding Equipment
Welding Variables (QW-3611) Actual Values Range Quafirted

Type of welding (Automatic)
Welding process
Filler metal (EBW or LBW)
Type of laser for LBW CC02 to YAG, etc.)
Continuous drive or inertia welding (PW)
Vacuum or out of vacuum (E8W)
Testing Conditions and Qualification Limits When Using
Welding Variables (QW-361.2)

Type of welding (machine)
Welding process
Direct or remote visual control
Automatic arc voltage control (GTAW)
Automatic joint tracking
Position qualified (2G, 6G, 3F, etc.)
Consumable inserts (GTAW or PAW)
Backing {metal weld metal, etc.)
Single or multiple passes pet side
Visual Examination of Completed Weld C

O Bend test; O Transverse root and face i Q W ^ ^ J ^ l ; D Longitudinal root Bt*6 face (QW-462.3(b}L O Side (QW-462.2);
O Pipe bend specsmm, co«oston-reststant overlay (QW~462,5(c)|; Q Plate benfi specimen, corrosion-resistant overlay (QW-462.5<d)];
D Macro test for fusion fQW-462MW; Q Macro test for fusion fQW-462S|e)l
1 Type Result Type **«* Type - *~* ]
Alternatrve radiographic examination results (QW-131).

BWet weld — fracture test (QW-180) . Length and percent of defects ,
Macro examination {GW-184) FtHet size (in.\.. . Concavity/convexity (in,) m
Other tests
Rim or specimens evaluated by. .Company.
Mechanical tests conducted by _ . Laboratory test no..
Welding supervised by
We certify that the statements in this record are correct and that the test coupons were pr*p&rtxi, welded, and tested m accordance with the
requirements of Section IX of the ASME Code.
Organization.
Date - BY-
This form (E00088) may be obtained from the Order Oept. ASME, 22 Law Drive, Box 2300. FatrfiekS, KJ 07007-2300
WELDING PROCEDURE APPROVAL TEST CERTIFICATE

El
Manufacture's Welding Procedure Inspecting Authority
Reference Number XYZ124 Reference No
Manufacturer: A Fabricator
Address: Deansgate
Manchester
M60 9AP
l Code/Testing Standard BSEN 288 Pt.3
Date Of Welding: 14/11/96
1 Range Of Approval:
Welding Process: TIG [141] Root Run MMA [111] Fill &Cap
Joint Type : Butt Weld Single Sided (See Line 21 For Approval Range)
Parent Metal Group: 1 With a Yield Stress <=241 N / mm2
Parent Metal Thickness (mm) 4.55mm ,(3mm To 9.1mm Multi-Run Weld)
Pipe Outside Diameter (mm) 33mm, (16.5mm TO 66mm And Plates)
Filler Metal Type /Designation : BS2901 Part 1 : A15 IBS639:E5143B[120.2.6]H
Gas /Flux Argon (99% pure) IBasic Coating
Type otWelding Current DC-Electrode IAC
Welding Positions : No Restriction on Welding Position
Preheat: Workshop Ambient Temperature Minimum
1 Post Weld Heat Treatment: Not Carried Out
Other Information : Joint Type Approval Range Continued From Line 11.
All Butt T Butt & Fillet Weld Configurations (Note Branch Connections Not Listed in Table 7 Approval
Range)
Fillet Weld Throat Thickness Approval Range = 3.41mm To 6.82mm
Certified that test Welds were prepared , welded and tested satisfactorily in accordance with the
requirements of the code/testing standard indicated above.
Location : Workshop Name and Signature : A Surveyor
Date of issue: 03/02/97

Inspecting Authority : A Surveyor
(CEOC Member Organisation)

DETAILS OF WELD TEST

E2
Reference Number XYZ 124 Reference No.:
Manufacturer: A Fabricator
Welder's Names : John Dyson Date Of Welding : 14/11/96
Welding Process : TIG [141] And MMA [111] Location: Workshop
Joint Type : Butt Weld Single Sided Method Of Preparation and Cleaning : Machine
Parent Metal Specification : API 5L Grade B
Parent Metal Thickness (mm) 4.55mm
Pipe Outside Diameter (mm) 33 mm
Test Piece Welding Position Flat
Weld preparation details (sketch)
Joint Design Welding Sequence
70 o
^
0mm I
2 mm
2.0 mm
WELDING DETAILS
Run Process Size Of Current Voltage Type Current Wire Feed/ Heat Input
Filler Metal A V Polarity Travel Speed
1 TIG 2.4mm 112 - DC- N/A N/A
2 MMA 3.25mm 160 - AC N/A N/A
3 MMA 3.25mm 160 - AC N/A N/A
Filler Metal Carbon Manganese Steel

Type Designation Trade Name See Other Information Opposite
Any Special Baking or Drying See Other Information
Gas Flux Argon (99% pure) TIG [ 141 ]
Gas Flow Rate -Shield 12-18 LPM MurexSaffire
-Backing N/A BS2901 Parti: A 15
Tungsten Electrode Type size 2.4mm Dia ,2% Throated
Details of Back Gouging /Backiiig None MMA 1111]
Preheat Temperature: Workshop Ambient Temperature ESAB Unitrode OK 48.00
Inter Pass Temper ature N ot Recordec1 BS639 :B 143 B[120.2.6]H
POST WELD HEAT TREATMENT Dried for at least 1 hour at

Time Temperature, Method: Not Carried Out 150 °C and held in a heated
Heating and Cooling Rates quiver prior to welding
The above test place was welded : Name and Signature : A Surveyor
In the Presence of:
Note: Welding Procedure Approval to EN 288 Part 3 has been carried out and is applicable to the named
manufacturer alone. This approval is not a Standard Welding Procedure to EN 288 Part 7 and may not be
reproduced in whole or Part and used as such
TEST RESULTS
E3
Reference Number : XYZ 124 Reference No.:
Visual: Acceptable Radiography: Acceptable

Examination:
Parent Magnetic Particle Test: Acceptable Ultrasonic Examination : N/A
Temperature : Ambient 20 °C
TENSLE TESTS
1 Type No. Re Rm A%on Z% Fracture Location Remarks

N/mm2 N/mm2
1 Requirement - 414 - -
Transverse 1 487 Parent Metal Acceptable
Transverse 2 - 494 - - Parent Metal Acceptable
BEND TESTS
Former Diameter = 4T
Type No Bend Angle Elongation Result
Filet Fracture Test: N/A
Face 1 180 N/A Acceptable
Macro/Micro Examination : 1 Off
Face 1 180 N/A Acceptable
Acceptable
Rootl 180 N/A Acceptable
Root 2 180 N/A Acceptable
IMPACT TEST Type N/A Size Requirement

Notch Location/Direction Temp °C Values Average Remarks
HARDNESS TESTS location of Measurements (Sketch)

Type /Load Not Required, Rm <420 & Re <270
Values Parent Metal
- H.A.Z.
- Weld Metal
OTHER TESTS
REMARKS
Test Carried out in accordance with the requirements BSEN 288 Part 3 and BSEN 287 Part 1 of
Laboratory Report reference No. 95/0999
Test results were Acceptable
Test carried out : Name and Signature : A Surveyor

In the Presence of:

Note: Welding Procedure Approval to EN 288 Part 3 has been carried out and is applicable to the named
manufacturer alone. This approval is not a Standard Welding Procedure to EN 288 Part 7 and may not be
reproduced in whole or Part and used as such
WELDING APPROVAL TEST CERTIFICATE

E4
DESIGNATION EN287-114[11 l],T,BW,W01,wm[B],t4 55,D33,P A ss nb
Manufacturer's Welding Procedure: Inspection Authority :
Reference Number: XYZ 124 Reference No.
Welder Names : John Dyson
Identification : W55
Method of Welders Indent Photograph
Identification :
Date and Place of: N/A (if required)
Birth:
Employer: A Fabricator
Code / Testing Standard : BSEN 287 Part 1
Job Knowledge : NOT TESTED
Weld test details Range of Approval

Welding Process TIG [141]Root MMA[111] Fill TIG[141]Root MMA[lllJFill
Plate or Pipe Pipe Pipe Or Plate
Joint Type Butt Weld Single Sided All Butt And Fillet Welds 1
! Parent metal group W01 {Carbon Steel} W01 Only
Filler metal type/designation BS 2901-1 :A15 BS639 E5143 B From Group W01 From Group W01
Gas/Flux Argon Basic Coating Similar Gas AR,R,RR,RB,RCB
Auxiliaries None N/A
Material Thickness (mm) 4.55mm 3MM To 9.1mm
Pipe outside diameter (mm) 33mm 25mm To 66mm
Welding positions PA {Flat Position} PA,PB
Gouging /Backing None Backing Gouging or None
Additional information on attached sheet/or welding procedure specification No. : None
j Type of test Performed and Not required Name and Signature : A Surveyor
acceptable
Visual YES - Inspecting Authority : A Surveyor
Radiography YES -
Magnetic particle YES - (CEOC Member Organisation)
' penetrate
Macro YES - Date of issue: 03/02/97
Fracture YES Location : Workshop
Bend YES - Validity of approval unit :14/11/98
Additional tests See WPAR
PROLONGATION FOR APPROVAL(Every six Month)
BY EMPLOYER/ SUPERVISOR
PROLONGATION FOR APPROVAL BY INSPECTING AUTHORITY
Date Signature Position or Title 1 Date Signature Position or Title 1 1
CHAPTER 19
Welding
Applications
The advantages of welding having been well recognised the Design, Fabrication and Erection of Structural Steel
by designers and fabricators, there is hardly any indus- in Buildings issued by the American Institute of Steel
try today which has not benefited from welding. Weld- Construction (AISC). Among other things, these codes
ing has made a significant impact on quite a large num- provide design criteria and standard procedures for the
ber of major industries, by raising the operational effi- guidance of the designer and the fabricator.
ciency, productivity and service life of the plant and The range and efficiency of welded structures have
equipment involved. A few of them which are highly been raised in recent years by the availability of a wide
relevant to India are listed below: variety of weldable structural steels having tensile
1) Structurals: Buildings and bridges strengths from 400 to 930 N / m m 2 . The thickness range
2) Field-erected vertical storage tanks * available is also very wide. Plates 150 mm and more in
3) Pressure vessels thickness are being welded regularly Column sections
4) Pipings and pipelines weighing over 1,000 k g / m and having flanges 125 mm
5) Power generating plant thick are available as standard product in advanced
6) Chemical plant countries. The codes and specifications also specify the
7) Ships standard grades of steel, which are permitted to be used
8) Railways in structurals.
9) Automotive The welding processes used on structurals are:
10) Machinery construction. manual metal-arc, submerged-arc, MIG/C0 2 , flux-cored
In this chapter, we have described how welding has arc, electroslag, electrogas and stud welding. The AWS
been applied in these industries and what specific benefits Code permits the use of the first four processes without
have been derived. welding procedure qualification testing, provided the
w e l d i n g p r o c e d u r e s conform to certain specific
1) Structurals limitations specified therein. For the other three, the
These can be divided into two areas: structures which procedures have to be qualified as prescribed in the
are subjected to static loading or a low frequency of Code.
dynamic stresses as represented by buildings, and those Some of the s t r u c t u r a l elements which are
subjected to fluctuations in loading as represented by conveniently w e l d e d and covered in the AISC
bridges. Specification are described below:
Application of welding to buildings and bridges has Column base plates: These are required on the ends of
been promoted by the availability of codes and specifi- columns to distribute the concentrated compressive load
cations based on many years of shop and field experi- of the column over a much larger area of the material,
ence and the extensive research carried out in recent that is to support the column. Several welded designs
years. Examples are the Structural Welding Code issued can be used. Gusset plates can be added if there is a
by the American Welding Society and Specifications for moment to be transferred at the base.
Column splices: These can be designed in several ways Welding Society. For the design of road bridges against
without punching the main columns. live loads, guidance must be sought from specifications
Beam seats and framed connections: M e t h o d s of such as the Standard Specifications for H i g h w a y
designing both stiffened and unstiffened seats and tables Bridges, adopted by the American Association of State
for p r o p o r t i o n i n g these types of connections are Highway Officials (AASHO). For the design of railway
described in publications such as the manual of design bridges, reference must be made to the appropriate
published by AISC. publication issued by the railway authorities, for
Rigid connections: The two basic types used are: (1) e x a m p l e , Manual for Railway Engineering, of the
the directly connected beam, in which flanges are American Railway Engineering Association.
welded on the supporting member without using The welded components of a bridge are similar to
connecting plates; (2) the beam cut-short and attached those of a building. They are mainly girders, beams,
with additional connecting plates. stiffeners, diaphragms, floor beams and stringers.
Semi-rigid connections: The design is somewhat For field erection, various types of fit-up clamps and
similar to that for rigid connections. clips are used, instead of erection bolts. For site welding
Wind bracing connections: For w i n d bracing of a splice, most of the bevel is placed on the top side of
connections, in a simply supported structure, the top the flange to keep overhead welding to the barest
plate is suitably designed to carry the calculated wind minimum. In making a girder splice, welding sequence
m o m e n t and w e l d e d to c o l u m n a n d b e a m . The must be properly chosen to minimise distortion. In one
connection to the seat is also suitably designed. sequence, after the assembly is fitted up, the web is first
Trusses: By designing trusses for arc welding, gusset partially welded, then the flanges are completely welded
plates can be generally eliminated or reduced to the and finally the weld in the web is completed. In the site
minimum requirement for the transfer of forces. Several welding of diaphragms, floor beams and stringers too,
designs are possible, in which the web members are the welding sequence is important.
angles or channels and chords are turned out from T
and H shapes. In India
For the erection of framed structures, the usual Steel frames for buildings, factory sheds, sports
practice is to use two or more ordinary rough bolts in stadia, railway platform roofings, etc., are largely
p u n c h e d holes to position the m e m b e r s d u r i n g welded. Typical welded details which go into structurals
plumbing and welding. With the help of two bolts, shop- are c o l u m n base plates, c o l u m n splices, truss
welded seats can be made to carry the erection loads connections and crane runway girders. There are several
and the service loads. A larger number of bolts are used outstanding examples of steel-frame buildings designed
if they directly carry the erection loads. for shop welding and rivetting, and welding or bolting
In the case of some rigid frame structures, the at site. A m o n g them are the Telephone Exchange
common practice is to use welded hitch plates or angles. Building, West Bengal Government's Building and the
All the holes are contained in the hitch material, which Reserve Bank Building, all located in Calcutta.
is cut off with an oxyacetylene torch when the welding Castellated beams, produced by suitably flame-
has been partially or fully completed. cutting and welding the standard rolled beams or
Even when a structure is designed to be all-welded, channels have been quite popular in India. Castellated
bolted connections are preferred for items like roof sections are just ideal for long s p a n s and light
purlins, roof bracing, girts and secondary beams. superimposed loads, though these have been frequently
used as other structural members as well. Since, by
Bridges castellation the depth of the section increases by about
Unlike buildings, bridges are subject to dynamic and 50% as compared to the original rolled section from
repetitive live loads. While the structural members of w h i c h it has been m a n u f a c t u r e d , there is a
buildings are protected from direct exposure to weather, corresponding increase in the moment of inertia and
such is not the case with the structural elements of section modulus of the section. The increase in the
bridges. These differences justify the more conservative moment of inertia is as much as 135% and that of section
unit stresses specified for bridges. modulus by about 56%. As a result of this, there is an
For the welding of bridge members subject to static improvement in the carrying capacity of the beam due
loading, reference should be made to specifications to the section having acquired properties to resist greater
such as the Structural Welding Code of the American bending moments. Though there is an improvement
WELDING APPLICATIONS 683
here, there is no effect in the shear capacity of the section. lengthwise and crosswise ribs, constitute a form of
In fact, at times it is adversely affected and sufficient orthotropic plate (see Fig. 19.1). While due regard could
strengthening is necessary. be given to complete symmetrical arrangement as
Use of welding in bridges makes them lighter and already recognised in bridges of this size, especially of
stronger and gives them a streamlined look. Welding is the main girders, a necessary condition was to maintain
used in India both for the fabrication of new bridges a practically perfect smoothness of the vertical plates
and for the repair or strengthening of existing ones. on the outer sides of the box, and for achieving this the
The first long-span bridge, where web members have problem of asymmetrical arrangement of weld joints
welded battens and d i a p h r a g m s is probably the had to be resolved, more so as dents in these side plates
Mortakka Bridge of Western Railway. The bridge over were not allowed to exceed 0.3%.
the river Langulia, consisting of six spans of 45 m and
one span of 41.25 m lattice type girders, was designed Materials
for shop welded and site rivetted construction by the About 55% of the total steel used was of St 52 quality,
B.N. Railway. The bridge has been found satisfactory meant for all highly-stressed fabricated parts, while the
on test for deflection and oscillation. balance was of St 37, for those components which were
The rail-cum^road bridges over the river Ganga at the least stressed in service.
Mokamah and over the river Brahmaputra at Amingaon While common bridge construction practice is to
are outstanding examples of saving steel through welding. weld extensively in the shops and to prefer rivetting on
In the main trusses, top and bottom chords, end rakers site, here welding was used throughout. It was therefore
and posts are of high tensile steel rivetted construction, necessary to choose and control base materials and the
but web members and floor systems and fixed-end and filler materials (in this case electrodes) so that they
free-end bearings are of mild steel welded construction. together gave at no stage any deficiencies in completed
Shear lugs have been welded to steel roadway stringers weldments, with regard to tensile strength, ductility and
to make a composite girder of steel and concrete. homogeneity.
While the technological properties of steel St 52 are
All-welded Bridge so very well-known, it was necessary to determine its
The all-welded bridge on the Rhine at Speyer in West weldability a n d impact p r o p e r t i e s b e t w e e n the
Germany, built in 1956, is a fine specimen of welded temperature range +20°C and -40°C. The DVM impact
structure, combining lightness with streamlined test results are shown in Table 19.1.
elegance, strength with perfect symmetry. The total Next, butt welds were made with approved electrodes
weight involved—1,700 tons—meant a saving of 69% and DVM impact tests were carried out over a
in steel when compared with its predecessor, which was temperature range from -30 to +20°C for butt weld joints
built in 1937 and blown up by enemy action in 1945. and -10 to -40°C for pure weld-metal. With the exception
of the roadway lengthwise joints, all the weld joints had
Design features the root run executed with low-hydrogen type electrode
With a useful width of about 13.7 m when seen in of German classification Kb IX/XIIS, and the top runs
section, the vertical plates form an enclosed box, in with titania type electrode of German classification
combination with the top and bottom girdle plates, and TiVIIIS. The test results obtained in respect of these
this box formation distributes even eccentrically applied electrodes are reproduced in Tables 19.2 and 19.3.
traffic load uniformly over the entire cross-section. The Test results of value more than 14.1 were not
roadway forms simultaneously the top and bottom recorded. Test-pieces in tests 4 and 5 were welded at -
plates of the box, which along with inter-crossing 15°C in unrestrained condition.
Table 19.1: Impact properties of St 52 (values are in mkg/cm2)

+20°C +10°C +0°C -10°C -20°C -30°C -40°C
10.9 11.1 11.1 10.5 11.5 10.2 6.75
12.2 12.7 12.2 12.3 12.7 10.6 7.10
10.3 11.2 11.0 10.5 9.9 7.8 7.4
9.3 9.1 10.0 9.9 8.7 8.0 6.9
Finally, the electrode weld-metal and the weld joints welding on which this particular procedure was based
completed with it were thoroughly examined under were fully justified.
conditions representative of those present in actual
construction work for tendency to surface and Shop fabrication
underbead crack formation. Next to materials, the fabrication procedure is an
Although it was logical to use automatic welding important requisite to ensure a successful welded job.
process on the roadway lengthwise joints, a cross-section Long before work commenced, careful thought had to
of which is sketched at (C) in Fig. 19.1, calculations based be given to the most appropriate assembly programme
on the conditions showed that it was more and welding procedures, and corresponding working
advantageous in this case to use high efficiency iron plans had to be decided upon, which would keep
powder electrodes of German classification Fe Es VIII, contraction stresses and/or distortion to the minimum.
with a deposition efficiency of about 150%. Later Automatic welding, in this case semi-automatic Elin-
practical experience and actual calculations showed that Hafergut Process, was used on a limited scale, and in
considerations of welding technique and economics of fact only for fillet welds in cross-beams between lamella
Type Ti VIII S
Type Fe Es VIII
Ti VIIIS
Fig. 19.1: All-welded bridge on the Rhine at Speyer. (A) Normal cross-section; (B) Cross-section over the piles; (C)
Cross-section of roadway lengthwise Joint.
and vertical plates. For all the remainder, hand-welding six components, namely the two main girders with
was given preference. varying vertical plate heights from 3.20 to 6.25 m over
As is clear from the sketch of the cross-section in Fig. the pillars (spans), and two top and two bottom box
19.1, the box-forming structure of the bridge consists of plates.
Table 19.2: Impact values for butt welds from St 52 steels at temperatures +20°C and -30°C
/ / ■ , _ ■
Impact value in/inkg/cm 2 at the testing temperature of

Test +2ri°c '; -30°c
No. Butt joint centre Heat-affected zone Butt joint centre Heat-affected zone
i) over 14.10 over 14.10 13.50 13.75
over 14.10 over 14.10 6.46 8.90
over 14.10 over 14.10 14.10 11.64
Average 14.10 14.10 11.00 11.40
2) 8.80 over 14.10 9.70 7.90
9.30 over 14.10 12.20 14.10
11.72 13.75 12.40 10.70
Average 9.90 14.00 11.40 10.90
3) 9.60 over 14.10 9.90 10.60
13.72 over 14.10 5.68 11.34
over 14.10 over 14.10 11.32 8.80
Average 12.40 14.10 8.90 10.20
4) over 14.10 over 14.10 5.60 11.00
over 14.10 13.80 10.26 12.90
Stress-relieved
over 14.10 over 14.10 10.70 13.40
Average over 14.10 14.10 8.80 12.70
5) 10.90 13.60 10.00 12.60
over 14.10 over 14.10 12.12 10.68
Stress-relieved -x
over 14.10 13.90 13.40 12.90
Average 13.00 13.90 11.84 12.10
Table 19.3: Impact values in mkg/cm2 of the weld-metal at temperatures of -10°C, -20°C and -40°C
Test No. Electrode -10°C -20°C -40°C
1 Ti VIII s 11.6 9.1 8.9
2 Ti VIII s 10.4 9.0 7.6
3 Ti VIII s 10.9 8.9 7.8
4 Ti VIII s 11.1 9.2 9.1
Average Ti VIII s 11.0 9.05 8.3
5 Ti VIII s 11.1 9.2 7.8
6 Ti VIII s 10.8 9.8 6.1
7 Ti VIII s 10.2 10.4 8.1
Average Ti VIII s 10.7 9.8 7.4
1 Kb IX / XII s 12.0 14.5 12.6
2 Kb IX / XII s 13.6 11.8 13.2
3 Kb IX / XII s 16.0 16.0 12.8
Average Kb IX / XII s 13.9 14.1 12.9
4 Kb IX / XII s 15.1 16.0 14.1
5 Kb IX / XII s 14.9 16.7 14.1
6 Kb IX / XII s 16.1 16.0 12.1
Average Kb IX / XII s 15.4 16.2 13.5
On account of large lengths and widths of individual one another was checked up before their despatch to
structural components, and with a large number of weld site, so that difficulties of assembly at site were fully
joints r u n n i n g l e n g t h w i s e a n d crosswise, it w a s avoided.
impossible to fix u p in a d v a n c e the a m o u n t of
contraction that would result, and to make allowances Site assembly and welding
for it. Data on contraction allowances based on All the main girders could not be fully welded
experimental work are indeed known, but they cannot together in the shop. As there was a limit to the
be taken absolutely for constructional work, since maximum height of the vertical plates, it was necessary
contraction effects vary from job to job. So it would be to send the main girder to the erection site in two parts:
futile to attempt to make markings for boring of holes the top and the bottom, so that these two parts could be
before welding had been finished. Moreover, the vertical welded to completion there.
plates and the strapping plates, as also the top and In this work, three different joints had to be welded
bottom boxplates had to be laid out in their lengthwise up as follows:
and crosswise directions without marking off the correct 1) The central joint of the main girder
lengths; following this, the rolled beams in the upper 2) The plates of the vertical reinforcements
plates and the U-section in the bottom plates had to be 3) The strapping plates of these reinforcements.
inserted as stiffening members, and finally the cross Joining together two parts of the main girder
girders could be inserted one after another and fastening involved placing them with the outer wall resting on
of the lengthwise stiffeners. Finally, the stiffening the setting-up floor and welding the V-groove joint. In
assembly tied up in this manner was welded in the most order to place the girder finally in the correct position,
appropriate sequence to the respective top and bottom special precautions had to be taken. In addition, the
boxplate. As a last preparation for the site work, the permanent measurements of shrinkage associated with
corresponding weld joint profiles and drilling of holes welding were studied closely, so that one could change
as were necessary for site assembly and rivet joints were over at once to another welding procedure, in case
carried out. measurements carried out in a certain procedure had
More consideration was necessary on the welding shown it to be not right, and also to enable one to lay
technique for the construction of main girders, since as down fresh instructions for such constructions.
already mentioned, asymmetrical conditions were Inspection: Over 600 examinations were carried out
present and on the outside of the plates perfect flatness by X-ray radiography and ultrasonic method. The
was necessary. Referring to the condition that the dents objectionable welds amounted to only 3.4%. These were
should not exceed a value of 0.3%, it must be made clear cut out and the portions were rewelded.
that it should be considered on a relative basis, since
with the temperature variation from -20° to +30°C in a 2) Vertical Storage Tanks
year (this means the structure can have a temperature For bulk storage of water, petroleum products, oil,
range between -20° and +30°C) the flatness must be natural gas, propane, cryogenic liquids such as oxygen,
related to some nominal temperature, so as to avoid the nitrogen and hydrogen, refrigerated ammonia and other
danger of uncontrollable stress concentrations, at important products, large vertical tanks are field-welded
extremely low temperatures. As an example of the and erected. Such tanks are a common feature of steel
necessity for the fixing up of a definite temperature as a plants, fertiliser plants and oil refineries. They are
basis for determining the maximum denting, it may be usually designed and fabricated in accordance with
m e n t i o n e d that following m e a s u r e m e n t s at a codes and specifications prepared to meet the specific
temperature of +22°C and in direct sun's rays, it was n e e d s of a particular industry. Such codes and
found necessary to adjust for denting by 1.5 to 2% over specifications designed for one industry should not be
the defined mean temperature of+10°C. applied to other industries without a full understanding
As far as was possible in the finishing of main girders, of the design, material and construction aspects
unitwise fabrication was followed, i.e. units were involved.
welded together in series and laid out, so that with Most of these tanks are made from mild steel, because
further constructional elements they could be assembled under normal stress and ambient temperature, it gives
into one higher unit. The setting of the final remaining satisfactory service. However, when a tank is exposed
work was completed by welding all the joints without to low and cryogenic temperatures and sufficiently
interruption. The final fitting up of completed units into stressed, the steel may fracture in a brittle manner,
resulting in catastrophic failure. For low-temperature upon the liquid, insulating its surface and minimising
service, the metal has to be qualified through Charpy- vapour losses. Sometimes, a tank is designed to have a
V notch impact tests. For very low-temperature service, fixed roof as well as a floating roof. The fixed roof keeps
Ni-alloyed steels, austenitic stainless steels and out rain and dust and acts as a sunshade to prevent
aluminium and its alloys are preferred. heating of the stored liquid. The floating roof minimises
The tanks are designed in various ways. The most vapour losses, oxidation and contamination of the
common shape is the cylindrical tank with flat bottom product. Domed-roof tanks with flat reinforced or
and a fixed cone roof (see Fig. 19.2) as used for refinery curved bottom are designed for storing petroleum
products, tar, alcohol and other industrial liquids. When products under low pressure to prevent evaporation.
a volatile liquid such as petrol is to be stored, the fixed For storing gas or petroleum products under pressure,
conical roof is replaced by a floating roof, which floats spheres known as hortonspheres are used (see Fig. 19.3).
Roof
Diameter
Fig. 19.2: Cylindrical oil-storage tank with cone roof.
Fig. 19.3: Sphere for storing gas or petroleum products under pressure.
Each type of tank is provided with accessories such as sides simultaneously. For heavier plates, double-bevel
foundations, vents, gauges, d r a i n s , overflows or grooves are necessary. In both these cases, back gouging
manholes depending on the product stored, local is generally not required. The electrogas welding
conditions and regulations. process, in which water-cooled copper shoes contain and
A hortonsphere is of petal construction having a shape the weld deposit to a desired contour until it
number of courses and two caps. Petals and caps are solidifies, has also been used with success on vertical
normally produced in the shop by hot pressing. In a seam welds.
few cases, spherical plates have been produced by In the design of field-welded storage tanks, full
rolling in a plate-bending machine, using a modified penetration butt joints in tank shell plating subjected to
technique. The sub-assembled components are finally 100% radiography are considered as having 100% joint
assembled and welded, and the completed vessel is efficiency; when spot radiography is used, they are
stress-relieved at site. regarded as having 85% efficiency. Double lap joints
Cryogenic storage tanks designed for the storage of which are fully fillet-welded are considered as having
liquid oxygen, nitrogen, hydrogen and liquefied natural 70-75% efficiency. The areas which sustain primary
gas at atmospheric pressure and at temperatures of - stress and hence require welded joints of 100% efficiency
183, -196, -253 and -163°C respectively, are double- are: vertical joints in cylindrical tank shells, joints below
shelled insulated flat-bottom tanks such as the one the point of support in suspended bottoms and all the
shown in Fig. 19.4. joints in low-pressure tanks. Less stressed areas such as
Since the erection of a tank involves positional the circumferential seams of cylindrical tank shells must
welding, the manual metal-arc process is commonly have at least two-thirds penetration. The shell is usually
used. M I G / C 0 2 or flux-cored gas-shielded arc process connected to a flat bottom by means of continuous fillet
is equally suitable, provided the area is shielded from welds on both sides of the shell plate. The fillet size is
winds. Flux-cored self-shielded welding is better suited equal to the thickness of the bottom plate or shell plate
for welding in the open breezy conditions. M I G / C 0 2 whichever is less, but never exceeds 12.5 mm.
and flux-cored processes are particularly preferred for Flat bottom and roof plates on vertical cylindrical
lap jointed bottoms and roofs, and for fillet welds. tanks, if the thickness does not exceed 12.5 mm, are
Specially designed submerged-arc welding equipment usually connected with fillet welds applied to the top
is available for the fully-automatic welding of girth side only. Heavier plates are usually groove-welded.
seams in the 3 o'clock position. With this equipment, Dished bottoms and dome and cone roofs on tanks
full-penetration b u t t welds can be m a d e in plate subjected to low internal pressures must be designed
thicknesses up to 9.5 mm, without bevelling the plates with a connecting weld capable of carrying the resulting
and by welding from one side at a time or from both stresses.
«*sK*m «**s*-
Fig. 19.4: Cryogenic storage tank.

A large number of all-welded vertical tanks and straight chrome steel or austenitic stainless steel or nickel
hortonspheres made from mild steel have been erected and its alloys) and non-ferrous metals (Al and its alloys,
at various sites of steel plants, fertiliser plants and oil Ni and its alloys, Cu and its alloys and titanium). Their
refineries in India. At a few places, the completed choice is determined by the following properties:
hortonspheres have been stress-relieved at site in one
a) Short time yield strength, tensile strength and
stroke, by using special gas-jet burners (see Chapter 19).
ductility at room and maximum operating tem-
All-welded stainless steel vertical tanks have also been
perature
erected at a few places for storing corrosive materials
b) Notch toughness and impact resistance, espe-
like ammonium sulphate. In some cases, Al and its alloys
have replaced the more expensive stainless steels. In one cially when low-temperature service is involved
example, a non-heat treatable Al-Mg alloy has been used c) Creep strength, rupture strength and ductility for
to fabricate a fully welded tank of 3,30,000 litre capacity, service temperatures in plastic flow range
using the TIG welding process. d) Weldability and need for preheat and postheat
treatments
3) Pressure Vessels e) Thermal conductivity, expansion and shock re-
Application of welding to pressure vessels and sistance
boilers has permitted design flexibility, significant f) Resistance to fatigue loading
weight reduction and operation at higher levels of g) Resistance to corrosion, corrosion fatigue and
strength and pressure. This has resulted in improved
stress corrosion
quality and safety, longer service life and economy of
h) Structural stability in service
construction and operation. Pressure vessel technology
is developing rapidly. Industrial demand of pressures i) Age-hardening and embrittlement in service
up to 3,500 psi, temperatures from 649 down to -260°C, Metallurgical features and weldability properties of
pressure and temperature cycling, exposure to corrosive these materials are covered in Chapter 10. For more
agents, hydrogen-containing environments and neutron information, the reader must refer to appropriate
irradiation have thrown up challenges to the designer handbooks.
and the fabricator. They should therefore keep abreast A pressure vessel is usually cylindrical in shape, and
of the evolution in materials, design and welding is made from a number of shell courses and either
techniques. ellipsoidal or two radii torispherical heads called dished
Failure of pressure vessels and boilers in service can ends. Special shapes include cylinders, spheres, ellipsoids,
not only cause monetary and production losses, but tori and cones in several combinations.
could also constitute a serious health hazard. Designers In designing pressure vessels, one must consider the
and fabricators must therefore scrupulously follow maximum allowable working pressure, maximum
safety codes, standards and regulations which govern working temperature and the possible use of pressure
their construction. Examples of such publications are relief equipment. The maximum working pressure and
the ASME Boiler and Pressure Vessel Code, Rules and temperature governing the design have to be stamped
Regulations of Lloyd's Register of Shipping, Rules for on the vessel. There are also several other design criteria
Building and Classing Steel Vessels of the American to be considered, such as fluctuating and repeated
Bureau of Shipping, and the Indian Boiler Code. pressure loading and temperature, effects of platforms,
These publications are quite exhaustive and cover supports and attached equipment, and wind and
permissible materials, sizes, shapes, design, service earthquake loading.
limitations, preheat and postheat treatments, inspection Pressure vessel codes provide design criteria for each
and testing requirements, and even specify conditions metal based on tensile strength, stress to produce
for the qualification of welding equipment, welding rupture in a specified time and creep strength. The
procedures and welders. Since they are revised from designer must be aware of various design formulae as
time to time, the designer /fabricator must use the latest applicable to shells and spheres, ellipsoidal or dished
issue including the latest addenda and amendments. ends, flat covers, etc. The codes also describe various
The alloys used in pressure vessels and boilers are: types of welded nozzles and other connections to
plain carbon steels, low-alloy steels, quenched and pressure vessels. In choosing from these types, the
tempered steels, chromium steels, austenitic stainless designer must consider the convenience of radiographic
steels, clad steels (carbon and low-alloy steels clad with inspection and economy.
Designing of supports for the vessels is equally Besides the automatic welding processes mentioned
important, because they introduce direct, bending or above, m a n u a l metal-arc welding and M I G / C 0 2
secondary stresses which have to be maintained at a welding are used, preferably for welding nozzles and
safe level. attachments. MIG process in semi-automatic or fully-
Code rules provide data on efficiencies for the automatic form is sometimes used for non-ferrous and
welded joints. Butt joints welded from both sides must stainless steel pressure vessels.
be used in the design as far as possible, as they offer H e a t i n g before a n d d u r i n g w e l d i n g may be
maximum assurance of uniform or complete fusion. necessary d e p e n d i n g on the type of metal, plate
Fabrication procedure: Plates received from the steel thickness a n d rigidity of the joint. Preheating
mill must conform to the required specifications with requirements are specified in codes.
respect to chemistry a n d mechanical p r o p e r t i e s Attachments like nozzles and pads which go on the
indicated by tensile and bend tests and in some cases, dished ends are welded to the dished ends before these
by impact tests. Before starting fabrication, the edges ends are attached to the shell, because in this case all
and surfaces of all plates must be carefully inspected welding can be completed in approximately the flat
for the presence of defects such as l a m i n a t i o n s , position. Similar external connections to the shell,
blowholes, cracks and tears. The plate edges are usually however, are attached and welded after the dished ends
beveled in the flat position, but in the case of heavy have been welded to the shell and after radiographic
plates more accurate assembly is obtained if planing or inspection, in order to avoid the interference of the
burning is done after forming. The plates are usually connections with these two steps. Nozzle welding is a
formed by b e n d i n g in pinch rolls or offset rolls. very important step, and the relevant code requirements
Sometimes they are press formed with dies. These m u s t be closely s t u d i e d , both at the design a n d
operations are commonly done in the cold. Hot-forming fabrication stage. If radiographic test is specified for
may be necessary in the case of heavy plates, which have nozzle welds, the fittings must be designed to be
to be press-formed. Since offset rolls or pinch rolls cannot amenable to radiographic inspection.
properly bend the extreme ends of the plate, the ends Most boiler codes stipulate that the welded vessel
are bent to the required curvature by means of a press be subjected to a stress-relieving p o s t w e l d heat
brake prior to roll-forming. Large hydraulic presses are treatment. The temperature range to which the vessel
used for forming very thick plate sections as well as must be heated is dictated by the type of metal, and the
double-curvature sections. time at m a x i m u m t e m p e r a t u r e d e p e n d s on plate
To assemble the vessel, the mating edges of the thickness. For carbon steels, the treatment usually
cylindrical sections are brought closer and held together consists of heating the completed vessel at a uniform
with lugs and bars, and tack welded on the weld root rate to a temperature of 600-650°C, holding at that
side. The joints are then welded, preferably by the temperature for a period of one hour per 25 m m of
automatic submerged-arc process, or by the electrogas thickness and then cooling uniformly. The maximum
or electroslag process (see Chapter 8) in the case of very temperature ranges for postweld heat treatment for
heavy plates. The completed cylinder normally shows various metals are indicated in the codes (see Chapter
depression at the joint due to weld contraction, and in 10). Besides relieving locked-up stresses in the vessel
most cases the out-of-round tolerance is exceeded. In caused by c o l d - w o r k i n g a n d w e l d i n g , this heat
light plate, the out-of-roundness can be corrected by re- treatment slightly lowers the yield point and tensile
rolling. In heavy plate, correct circularity is achieved by strength and marginally increases the ductility of the
judicious selection of edge bevel, making pressing plate material due to slight spheroidisation of the
allowances during forming, and by placing supporting carbide constituent. It also softens and increases ductility
struts at the cylinder during assembly and welding. of the weld-metal and HAZs.
Double-bevel joints in place of single-bevel also help. Thick-walled vessels which have to stay in the
Dished ends are prepared for welding by cutting the furnace for long periods and those having a high
edges in a boring mill or lathe. Oxy-cutting, gouging d i a m e t e r - p l a t e - t h i c k n e s s ratio may, d u r i n g heat
and plasma arc cutting are also used. They are attached treatment, sag and distort through plastic flow induced
to the shells with or without backing strip or ring, by the weight of the vessel. This can be prevented by
depending on the welding process and procedure. providing properly spaced saddle supports capable of
Welding procedure is the same as for the circumferential providing a large bearing surface, and internal struts
joints of the vessel. across the diameter.
In the fabrication of economisers, superheaters and longitudinal seam is welded using the underlayer as
heat exchangers which are important units of an oil- backing. Shrinkage in these welds further tightens the
fired boiler in a thermal power station, parallel tubular hold of the outer layer. An important advantage of
elements have to be connected between two or more multilayer design is that the vessel does not require
headers or drums. The welds in tubing are mostly stress-relief heat treatment.
circumferential butt welds. Such welds are made in Inspection and testing: The inspection methods
shops by flash, induction or automatic MIG/C0 2 commonly used in pressure vessel fabrication are visual,
processes. At site, such welds are made by TIG for the sectioning and etching, fluorescent and liquid
root pass and manual metal-arc for subsequent passes. penetrants, magnetic particle, radiographic and
Economisers usually operate at relatively low ultrasonic. For economic reasons, such inspection is
temperatures; hence their fabrication involves welding carried out before the vessel is postweld heat treated
of plain carbon steel. Superheaters, whose function is and subjected to pressure tests. For critical applications,
to raise the temperature of the steam delivered from and especially in the case of vessels made from special
the primary steam-generating pfortion of the boiler unit, alloys, the purchaser may demand radiographic
are located in the high-temperature region of the heating examination after the final postweld heat treatment.
gas flow. Superheater parts including tubing are In some eases, the completed vessel is subjected to
therefore made from oxidation-resistant and creep- hydrostatic, pneumatic or proof testing (see Chapter 13).
resistant alloy steels (commonly alloyed with Si, Cr and
Mo), which must be welded with special welding proce- 4) Pipings and Pipelines
dures involving preheat and postweld heat treatments In the industrial world, the term piping is usually
as prescribed in boiler codes (see Chapter 10). understood to cover pipe; tubing; fittings such as tees,
An important step in the fabrication of heat elbows, flanges and reducers; valves and headers used
exchangers, whose function is to transfer heat from one in oil refineries, power stations, nuclear plants, chemical
phase of the operation to another, is the welding of tubes and petrochemical plants and other industrial plants.
to tube plates. Special welding equipment, procedures The term pipelines usually applies to long transmis-
and techniques have been developed for this sion pipelines designed to conduct liquids such as water,
application. crude oil and petrol, and gases such as natural gas.
Multilayer vessels: When sufficiently thick plates for Today, piping systems and pipelines in industry are
high-pressure application are not available, or facilities almost fully welded. Threaded joints are rarely used.
for forming them do not exist, multilayer welded vessels Flanged joints are used only where sections have to be
may be fabricated. In their design (see Fig. 19.5), the opened for internal inspection or replacement.
inner shell, usually 12.7 to 32 mm thick, is first fabricated Piping and pipelines are dealt separately in this
by rolling and welding. Successive steel plate layers, section. Penstocks are also considered to be transmission
each about 6.5 mm thick, are then wrapped under pipelines, but for convenience they are dealt with in the
tension around the inner shell, and in each case the section on power generating plant.
Fig. 19.5: Elevation and cross-section of a multilayer cylindrical section

Piping: Industrial pipings are critical items in a Pipes are available in long lengths as seamless or welded
production plant and they frequently operate under pipes. Pipings are longitudinally welded in a tube mill
high pressures, high temperatures and in corrosive from strips by using the electric resistance butt or high-
atmospheres. The efficiency, productivity and safe frequency resistance welding process, while pipes for
operation of plants depend to some extent on how pipelines are welded along their long seams in a pipe
effectively piping systems withstand the rigours of mill by the automatic submerged-arc or M I G / C 0 2
service. Serious consideration has to be given to the process. In the erection of pipings and pipelines, welding
selection of grades and sizes of materials, design, is restricted to girth joints or to joints between pipes
fabrication, erection, testing and inspection. Guidance and their attachments. Hence in the following sections,
is provided by various codes and standards applicable only girth welding techniques are described.
to w e l d e d p i p i n g systems p r e p a r e d by technical The metals used for piping are: carbon steel, wrought
societies, trade associations and standardisation bodies. iron, C-Mo steels, Cr-Mo alloy steels, cryogenic steels,
For example, the American National Standards Institute stainless steels, Al and its/ alloys, Ni and its alloys, Cu
(ANSI) has issued Code for Pressure Piping, which and its alloys and Ti and its alloys. Their weldability
covers Power Piping, Industrial Gas and Air Piping, characteristics are described in Chapter 10.
Petroleum Refinery Piping, Oil Transportation Piping, Carbon steel: Carbon steel piping is mostly welded
Refrigeration Piping, Chemical Industry Process Piping, by the manual metal-arc process using E6010 or E7018
Nuclear Power Piping, Gas Transmission and class of electrodes. For critical applications which
Distribution Piping Systems. demand full penetration welds, split or solid backing
Piping connected to boilers are covered in several rings are provided on the inside, or the well-penetrated
sections of the ASME Boiler and Pressure Vessel Code. root pass is made with the TIG process as described in
The American Petrol Institute (API) has issued a Chapter 5. This technique applies to all metals. MIG/
standard for Field Welding of Pipe Lines. ASME Guide C 0 2 process using gas mixtures of C 0 2 and argon is used
for Gas Transmission and Distribution Piping Systems is on less critical pipings, where full root penetration and
another useful publication. The American Welding fusion are not essential. In shop fabrication of thick-
Society has published the following recommended walled p i p e h a v i n g O.D. of m o r e than 200 m m ,
welding practices: automatic submerged-arc welding is used for the filling
a) Welding of Austenitic Chromium-Nickel Steel passes, after the root pass has been completed with the
Piping and Tubing, D10.4. manual metal-arc or TIG process. If backing rings are
b) Welding of Chromium-Molybdenum Steel Piping, used and the fit-up is good, the entire joint can be made
D10.8. by the SA process. Generally preheating is not necessary
c) Recommended Practices for Gas Shielded-Arc if the carbon content of the steel is below 0.30%. If the
Welding of Aluminium and Aluminium Alloy Pipe, wall thickness exceeds 19 mm, postweld heat treatment
D10.7. is usually recommended. It consists of heating to 600-
d) Welding Ferrous Materials for Nuclear Power 650°C and holding for one hour per 25 mm of wall
Piping, D10.5. thickness, with a minimum holding time of 30 min, and
e) Gas Tungsten-Arc Welding of Titanium Piping and then cooling in still air. For further details, relevant codes
Tubing, D10.6. must be consulted.
To ensure satisfactory welding of piping installation, During manufacture of boiler units, large numbers
it is first necessary to establish and qualify the welding of tube butt welds have to be made with the tubes
procedure covering base metal specifications, filler positioned at any angle from horizontal to vertical, and
metals, edge preparation and joint fit-up, pipe position, being often in positions of restricted access. Automated
welding process, process parameters, welding techniques, orbital TIG welding machines with automatic cold wire
preheat, interpass and postheat schedules, and final inspe- feed have been developed for this purpose. A typical
ction and testing. It is also necessary to qualify the welders orbital TIG welder has a weldhead, covering tube sizes
for the welding procedure adopted. Standard procedures in the 25-50 mm O.D. range and requires only 44.4 mm
for the qualification of welders and welding machine clearance between adjacent tubes. It features an integral
operators are given in relevant codes, for example, in wire-feed system, i.e. the wire-feed facility is mounted
Section IX of the ASME Boiler and Pressure Vessel Code. on the head and rotated with the electrode block. Arc-
Pipe materOials and fittings are available in standard- voltage control provides a means of maintaining a
ised specifications, sizes and with standard tolerances. constant preset distance b e t w e e n electrode and
workpiece. These facilities allow for a number of d r a w i n g u p p r o c e d u r e s , relevant codes must be

continuous orbits (i.e. multiple weld passes) to be made consulted.
around the tube joint. Such a machine can be applied Low-temperature steels: The types of steel used for
on pipings of all industrial metals. Lately welding heads various low-temperature service pipings are given in
capable of joining tubes 18.2 mm O.D. with a clearance Table 19.4. They are usually welded by the MMA
of only 16.8 mm have been produced. process. The suitable AWS classes of electrodes are
Wrought iron: Wrought iron piping has low carbon indicated in the table. Preheating is a must for Ni steels,
content (0.12% maximum). It is usually welded by the because nickel renders the steel air-hardened. Preheat
manual metal-arc process. It is advisable to use low and postheat data are given in Chapter 10 while
w e l d i n g currents a n d s p e e d s . Preheating a n d discussing their weldability.
postheating are generally not required. Austenitic stainless steels: These steels are used for
C-Mo steels: The welding processes used for these high-temperature as well as low-temperature service
steels are the same as those used for carbon steels. For down to -196°C. Root pass is deposited by the TIG
manual welding, electrodes of E7010-A1, E7016-A1 or process, while subsequent passes are completed with
E7018-A1 are used. For SA welding, the Mo alloy of the the manual metal-arc or MIG process. For critical
weld-metal is derived either from the wire or the flux. applications, as in nuclear and power plants, the inside
Preheat and postheat data are given in Chapter 10 while of the pipe is purged with inert gas or a consumable
discussing the weldability of these steels. insert ring is used. No preheat or postheat is necessary.
When used in service temperatures exceeding 425°C, Martensitic stainless steels: These are hardenable steels
C-Mo steels h a v e been k n o w n to u n d e r g o and are susceptible to cracking during welding. Preheat
graphitisation, i.e. the carbon transforms to nodules or and postheat operations are necessary. The postweld
flakes of graphite, which substantially reduce the heat treatment must immediately follow the completion
toughness of the steel. Though such unfavourable of welding without withdrawing the preheat. Welding
phenomenon can be suppressed by stress-relieving the data are given in Table 19.5. If for some reasons
welded joints at 720°C for four hours, use of C-Mo steel postheating is not possible, type 310 or 309 stainless steel
pipings for high temperature applications is being filler wire must be used.
discouraged. Ferritic stainless steels: These steels are less susceptible
Cr-Mo steels: These grades are mostly used for service to cracking during welding than the martensitic types,
in the 400-593°C temperature range. They are usually b u t they may become embrittled due to the high
welded by the manual metal-arc process, using low- temperatures attained during welding and consequent
hydrogen type low-alloy steel electrodes of matching grain growth. To remove embrittlement, the steel is
alloy contents. An orbital narrow gap TIG welding annealed for one hour between 705 and 790°C, and then
equipment of special design capable of welding pipes quenched or air-cooled. The welding data are given in Table
of 100 mm maximum wall thickness has been described 19.6.
in Chapter 5. For submerged-arc welding, it is advisable Al and its alloys: These alloys are commonly welded
to use neutral flux and alloyed wire in preference to by the TIG process and in some cases by the MIG
alloyed flux and neutral wire, because in the latter case, process. Before attempting to weld pipings, welders
the alloy balance in the weld deposit gets upset during must undergo training and gain some experience. In
multi-pass welding at high interpass temperatures. welding horizontally positioned fixed piping, the
Preheat and postheat data are given in Chapter 10, while molten metal sinks due to its high fluidity. Aluminium
discussing the weldability of these steels. However, for backing rings a n d c o n s u m a b l e insert rings are
Table 19.4: Steels and electrodes for low-temperature service

Min. temp. Type of steel AWS class
°C MMA electrode
-46 Fine-grained fully deoxidised steel E7016-1E7018-1
-60 2.25% Ni steel E8015-C1
-100 3.5% Ni steel E8015-C2
-196 9% Ni steel ENiCrFe-2
Table 19.5: Recommendations for wrought martensitic stainless steel pipes

Type of Chemical composition Preheat and Postheat
steel % Recommended interpass temperature
electrode or temperature °C
C Cr welding rod °C
12 Cr 11.5-13.5 E, ER410 320-370 705-760
E, ER310 or
R, ER309 200-320 705-760
12 Cr 0.08 max. 11.5-13.5 E, ER410 150-260 705-760
E, ER310 or
E, ER309 150-260 705-760
13 Cr over 0.15 12.0-14.0 E, ER410 or 320-370 705-760
E, ER430
E, ER310 or 200-320 705-760
E, ER 309
Table 19.6 : Recommendations for welding ferritic stainless steel pipes

Type of Chemical composition Recommended Preheat and Postweld 1cteat-
steel /o electrode or interpass temperatui re
welding rod temperature °C
C Cr Others °C 610-760
12 Cr, Al 0.08 max. 11.5- 0.10- E, ER430 Not necessary Highly recommended
14.5 0.30 Al E, ER310 or •
E, ER309 Not necessary Recommer ided
16 Cr ■0.12 max. 14.0- E, ER310 or Not necessary Recommer ided
18.0 E, ER309
27 Cr 0.20 max. 23.0- 0.25 446 150-200 Essential

27.0 max. N E, ER310 or
E, ER309 Not necessary Recommended
sometimes used to obtain good root penetration. It is important to remember that Ni and its alloys
Preheating is generally not necessary, but may be used are susceptible to embrittlement by accidental presence
with advantage when the diameter exceeds 60 mm. of lead, sulphur, phosphorus and some low-melting
Preheat temperature ranges between 280 and 300°C. metals.
Some Al alloys are u n f a v o u r a b l y affected w h e n Copper and its alloys: They are commonly welded by
p r e h e a t e d above 200°C. Hence, h i g h p r e h e a t oxyacetylene, MMA, TIG and MIG processes. It is
temperatures must be used with care. advisable to use backing rings whenever possible,
M and its alloys: These alloys are commonly used in because of the high fluidity of molten copper. Because
piping because of strength properties, good corrosion of the high heat conductivity of copper, preheating with
resistance to many acids, and easy weldability. They can a gas torch is necessary when large diameter or heavy-
also be readily welded to ferritic and austenitic steels. walled pipes are being welded. Red brass and yellow
The welding processes commonly used are: MMA, TIG brass are preferably welded by the oxyacetylene process
and MIG. Backing rings should not be used, because to minimise vaporisation of zinc. Cupronickel 30 (i.e.
they promote crevices, root cracks and corrosion. 70:30 alloy) is extensively welded and used for water
Consumable insert rings should be preferred. During pipe and condenser tubing on ships, because of its
root pass welding, the inside of piping must be purged superior resistance to sea water corrosion. The most
with inert gas, which can be helium, argon, hydrogen suitable welding processes for this alloy are MMA and
or their mixtures. TIG.
Ti and its alloys: As explained in Chapter 10, welding mm for stainless steel. Thicknesses greater than these
of these materials demands special techniques and and up to 22 mm should have their edges prepared as
specialised skill on the part of the welder. Pipes of wall at (a) in Fig. 19.6, while thicknesses greater than 22 mm
thickness 1.6 mm and below are normally welded by should have edge preparation as at (b) in the same
the TIG process without filler wires. For heavier pipes, figure.
filler metals are used. Unless the filler wire is thoroughly In critical applications where carbon and low-alloy
cleaned and handled with care, it can contaminate the steel piping, stainless steel piping and most non-ferrous
weld. Contamination also occurs if the hot end of the piping is to be TIG welded, joint preparations including
wire is withdrawn from the gas shield and exposed to consumable insert rings as shown in Fig. 19.7 are used.
atmosphere during intermittent deposition. Special care In all the cases shown, U or flat-land bevel preparations
must be taken that there is 100% root penetration allover are employed, because they help to minimise excessive
the joint. A small root defect can develop into a crack shrink.
during service and lead to serious failure. For butt joints between unequal wall thicknesses (for
Dissimilar metals: Pipings of dissimilar metals are example, between a pipe and a cast steel fitting or valve
often welded in power plants, oil refineries, nuclear body), codes recommend that a smooth taper be
plants, etc. The metals commonly involved are carbon provided on the edge of the thicker member.
steels, low-alloy steels, stainless steels and nickel and Fillet-welded joints are often used for pipe sizes 50
its alloys. Normal welding procedures can be used in mm in diameter and smaller, and for joining pipe to
these cases, because the melting points of these metals flanges, pipe to valves and pipe to socket joints. Three
are fairly close. They have been described in Chapter examples are shown in Fig. 19.8.
10. The m a i n considerations are filler metal Backing rings: Backing rings are commonly employed
compositions and preheat/postheat temperatures. For for welding carbon steel and low-alloy steel piping by
dissimilar joints involving non-ferrous alloys, the filler the MMA process in steam power plants and other
metal and w e l d i n g p r o c e d u r e m u s t be carefully applications. While split rings are sometimes used for
determined after studying the metallurgical aspects of non-critical applications, solid flat or taper-machined
the joint in question. Sometimes, it helps to butter the backing rings are preferred for critical applications.
joint edge of the metal having the higher melting point Some designs of backing rings and the manner in which
before final welding. For example, when carbon steel is they are fitted are shown in Fig. 19.9. The figure shows
to be joined to silicon-bronze, the carbon steel is buttered that the pipe-end must also be suitably machined on
with silicon-bronze weld deposit. When the metals to the inside diameter. Chemical composition of the ring
be joined have widely different melting points, brazing, is important as also the seat contact between the pipe-
brazewelding or soldering should be resorted to. end and the ring. Guidance for the correct use of backing
Joint design: As stated earlier, the usual joint to be rings is available in relevant codes. Backing rings are
welded in pipings is the circumferential butt joint. To rarely used for piping in oil refineries and chemical
weld such a joint by the MMA process, the pipe edge plants.
can be square or slightly chamfered when the wall Where the weld joint quality and especially its
thickness is below 5 mm for carbon steel, and below 3.2 corrosion resistance are important, consumable insert
1/16" ± 1/32"
Fig. 19.6: Edge preparations of pipe end for MMA welding
1/8" to 1/V 371/2o
TO 1/8"
Square butt (Flat land) V Bevel
1/4" to 3 / 4 "
A"
25°
1/8'
"3/32"l 1/16"
Flat land bevel U bevel
Over 3/4"
1/16"
Flat land bevel U bevel
Fig. 19.7: Joint fit-up using consumable insert for TIG welding
rings are placed at the root, as mentioned earlier and

illustrated in Fig. 19.7, and fused with a TIG torch, so
that a sound root weld pass results. This technique
dispenses with the addition of filler metal, which could
interfere with the welding operation and cause lack of
h V 3 2 " - 1/16"clearance penetration. The subsequent passes, if required, are then
i deposited by the TIG process using a filler wire or by

the MMA process. If instead of using an insert, the pipe-
end is suitably machined at the root and autogenously
Welded sleeve coupling
welded, cracking or porosity is likely to occur because
of the unfavourable base metal composition. Use of a
c o n s u m a b l e insert ring of p r o p e r l y balanced
composition and dimensions (a) provides the best
welding conditions even in horizontal fixed or 5G
position, (b) minimises human element and thereby
ensures weld uniformity, (c) gives the most favourable
weld contour which can resist cracking arising from
weld metal shrinkage, and (d) gives weld-metal
composition which can guarantee optimum mechanical
properties and corrosion resistance.
At this point, it is pertinent to mention that the various
Socket end welding elbow pipe welding positions are defined by standard symbols
(1G, 2G, etc.) as shown in Fig. 19.10. Among these, 5G
position is the most difficult and it calls for high welding
Fig. 19.8: Examples of fillet-welded joints
skill. For this position, it is advisable to insert the
37 1/2" 42 1/2"
37 1/2°* 2 ^
• ? M - ' ' 3 / 1 6 " nominal
< " 1 ~ 4 - 3 / 1 6 " nominal
rnsspi:
1/16"±1/32"
y\
If**-
AB Z f c * . - - ^ c OT 1/8-R miriT (Bore)(R|ng 0 D )
Break corners 3/4" ( B o r e ) ( R j n g QD) Break corners*
For wall thickness (T) 9/16" For wall thickness (T) 9/16"
to 1" inclusive and tapered to 1" inclusive and straight
internal machining. internal machining
10° ± 1°
.10° ± 1°
.Rounded .0 Rounded
37 1/2°"±2t ^ 3 7 1/2° ± 2 1/2°
* H I M | 3 / 1 6 " nominal t | =T 4 ; ^ V ' |Aj3/16 M nominal
v I ! 4*»
mm ^ u ! i / V ^ . X^y'/yl/2Z\ -1/16"±1/32
+<%30° m a x ^ ? M / l\/il.
^ • D K R i n g OD)
Break corner (Bore) 0 D )
1/8" R M T n r B ^ a t - o T n ^ ^ ^ / ^ ^
For wall thickness (T) greater than 1" For wall thickness (T) greater than 1"
and tapered internal machining and straight internal machining
Fig. 19.9: Edge preparation using flat or taper machined solid backing rings
Fiat position Horizontal Vertical Overhead

position position position
1 G aG
2G
^Q
0
Plates and axis Plates and axis

ill"., Plates horizontal
2 a n d axisvertical
of pipe horizontal of pipe vertical Plates
Roll welding j j of pipe vertical!
0) Test position
> Test position
o horizontal 2 G Horizontal fixed
o 6G
O 5G
/ 4 5°±5°
Pipe shall not

be turned or
Axis of pipe vertical rolled while welding
Fig. 19.10: Standard symbols for designating welding positions

consumable ring eccentric to the centreline of the pipe as applications are provided by manufacturers in advanced
shown in Fig. 19.11, so that it compensates for the down- countries.
ward sag of the liquid weld-metal and helps to obtain In the installation of piping systems, tees, laterals,
uniformly smooth root contour on the inside of the joint. wyes and vessel openings have to be welded, and they
Consumable insert rings of proper shapes, diameters normally involve intersection joints. Since such joints
and chemical compositions to suit various metals and are difficult to weld, standard welding fittings supplied
by manufacturers are used. These fittings possess
p*.l<SSSf
bursting strengths equivalent to those of pipes of the
same weight and they are designed to be connected by
simple butt welds. Some examples of such fittings are
3/32" 3/16"
shown in Fig. 19.12. Manufacturers also provide factory-
made nozzles, necks, outlets, tees, etc., specially designed
>
1 for welding to simplify the fabrication of piping.
^dbd
Heat treatment: Preheating, concurrent heating and
postweld heating are important steps in the welding of
pipings; and their successful performance in service
X
1/16" ► often depends upon correct heat treatment. The heat
treatment procedure includes consideration of the
Fig. 19.11: Eccentric insertion of consumable insert ring
maximum temperature to be attained, time at maximum
for 5G position pipe welding
temperature, rates of heating and cooling, and the width
90° Long
radius elbow 90° Short Elbow 180° Return bend
radius elbow
J
Tee Reducing tee Tee reducing on run
Tee with concentric reducers Lateral straight run
Concentric reducer Eccentric reducer Cap
Fig. 19.12: Examples of standard manufactured commercial welding fittings

of the heating b a n d . The u s u a l m e t h o d s of heat suitably, to prevent deformation and distortion. This is
treatment are: (a) oxyfuel, (b) electric resistance heating, accomplished in the shop by placing adjustable roller-
(c) induction heating, and (d) heating in a furnace. type supports under the parts being welded as near to
In the oxyfuel method, a simple gas torch is adequate the joint as pos-sible, allowing sufficient space for the
for small diameter p i p e s . For larger p i p e s a n d placement of the heating apparatus over the joint. In
connections, ring b u r n e r s are more effective. For field work, where the welds are made in position, chain
temperature control, temperature indicating crayons are falls or other suitable rigging secured to the building or
used. Electric resistance heating has been dealt with in other suppor-ting structures are used to accomplish the
detail in Chapter 20. In this method, surface thermo- same objectives.
meters or electrically operated pyrometers are used to Offshore pipework: A company in the Netherlands
control automatically the current flow to the heating fabricates exacting offshore pipework in their shop using
units. Thermocouples are usually attached to the metal several automatic TIG and MIG welding installations,
to be heated by induction heating. The thermocouple each having a turntable with two sets of adjustable roller
wires are then connected to control equipment, which beds. For 100 mm diameter pipes, manual TIG is used
may automatically control the time-temperature cycle for the root pass and automatic TIG with cold wire feed
and even programme the heating and cooling rates of for filling and capping passes. For 300 mm diameter
the metal. pipes, the same procedure is used for the root pass, and
During postweld heat treatment, it sometimes automatic MIG with a flux-cored wire is used for
becomes necessary to support the welded pipe sections subsequent passes. Table 19.7 shows a procedure for
Table 19.7: Procedure for offshore pipework welding

Material A333 GR6
Root pass TIG hand
Filling / capping MIG auto
Welding position 1G
Preheat temp. (°C) 100
Interpass temp. (°C) 300
Pipe dia. (in.) 4

Wall thickness (mm) 10.7
Joint preparation V2 x 30°
Root pass
TIG wire type PZ 6500
Wire dia. (mm) 2
Welding current (amp) 100
Filling I caspping
MIG wire type Flux-cored
Wire dia (mm) 1.2
Gas type Mixed gas, 80/ 20 (Argon / Co2)
Gas flow (1 / min) 10
Welding current (amp) 205 230 225 225
Welding voltage 28 28 27.5 27.5
Wire-feed speed (cm / min) 788 788 788 788
Welding speed (cm / min) 24.5 19.6 23.5 23.5
Total welding time (min) 2.0 2.5 2.1 2.1
Joint preparation and runs

C:Mn pipe steel, in which two types of flux-cored wires 7) The pipe is welded finally by the submerged-arc
can be used for the MIG passes, one for temperatures process, one run on the inside and another run on the
down to -25°C and the other for temperatures below outside. For the first pass, water-cooled backing is used.
-25°C The latter deposits a 2.5% Ni steel weld-metal 8) The finished pipe is moved on to the expander,
with Charpy-V notch value of 47 J minimum a -60°C, where it is surrounded by locked restraining dies, while
which also meets the COD test requirement. water at extreme pressure is pumped in, expanding the
Pipelines: This section deals with cross-country pipe against the enclosing dies. The expander does the
transmission pipelines which conduct natural gas or following functions:
liquid products such as crude oil. a) Pipe ends are mechanically expanded to size
Pipes of reasonably long lengths are produced in a b) Hydrostatic pressure expands the pipe to the exact
tube mill or pipe mill. They are either seamless or size of the mechanically locked dies
electric-resistance welded, or submerged-arc welded. c) Pipe is tested to code requirements
Laying of pipelines involves only circumferential d) H a m m e r s are d r o p p e d , while pipe is under
welding in the field. maximum code pressure
Seamless pipes are made from solid round billets of e) Inspector examines welds for leaks.
proper diameter and length. Surface defects of the billets Two 13 m long pipes may be butt welded to make 26
are initially removed by scarfing. The billets are heated m lengths, again using submerged-arc welding.
and pierced to make a hole in the solid billet. The so- Finally, there is end facing and bevelling.
formed pipe is passed successively through a plug- The forming is at the rate of 20 m / m i n and output is
rolling mill to elongate it and reduce the wall thickness up to 3,000 tons in eight hours.
to the desired dimension. The pipe is rounded and A typical boom welder used for the internal welding
smoothed on the inside and outside surfaces by passing of pipe by the submerged-arc process is shown in Fig.
through a reeling machine. The pipe is finally sized by 19.13. It is fitted with a television monitor. The 375 mm
passing through sizing rolls, straightened, expanded, diameter boom enables pipes of 450 mm and larger
hydrostatically tested and bevelled at the two ends. diameters and lengths up to 10 m to be welded internally.
Resistance-welded pipes are made from rolls of steel Pipes are also welded by the submerged-arc process,
strip in a tube-making machine. In this machine, the using the so-called spiral welding technique. The main
continuously fed strip is passed through forming rolls to advantage is that with a given width of plate or coil, a
form a straight O-shaped section, which is electric- wide range of pipe diameters can be fabricated.
resistance welded at the seam. The emerging pipe is In this technique, the edges of plates or coils are
tested continuously by means of a non-destructive trimmed to the required width and bevelled. They are
testing device and cut to the desired length. A coiler is then subjected to a modified three-roll bending arrange-
used if a long length of pipe is to be supplied in coil form. ment supported by internal or external cage rolls, and
The operation of producing large diameter pipes by the result is a continuous helix. The first welding pass is
the submerged-arc process is best understood by laid on the internal of the seam and then on the external
referring to the procedure followed by a firm in the diameter, 180° away. The conventional single electrode
U.S.A. or two electrodes in tandem may be used for the
This firm produces mild steel pipes up to 13m length submerged-arc process. To feed the stock continuously
and diameter between 500 and 900 mm and thicknesses into the machine, ends of plates or coils are welded only
between 6.3 and 12.7 mm in the following stages: on the inside by the submerged-arc process prior to
1) Shearing the edges to exact widths, bevelling the forming. After seam welding, the required length of pipe
edges and pre-forming the plate by an initial bending is cut off and the external cross-weld is completed.
of the edges The maximum outside diameter of seamless pipes
2) U-ing press is 650 mm. High frequency resistance seam welding is
3) O-ing in a semi-cylindrical die with another top used to produce pipes and tubes of diameters ranging
semi-cylindrical die activated by two massive hydraulic from 12.5 m m to over 1,250 m m a n d w i t h wall
rams of 6,000 tons capacity thicknesses of between a fraction of millimetre and 25
4) Tack welding and lack grinding mm. Submerged-arc welding is best suited for large
5) Cleaning the pipe in a degreasing bath diameter pipes, which can be internally and externally
6) Tab is welded at each seam end to assure proper welded. Penstock pipes of 10 m diameter and above
lead-in and cut-off of finish welds have been welded by this process.
Electrode Boom height

wire reels adjustment handwheel
SA welding head
Electrode nozzle tube
Flux hopper
Flood lamp
T.V. Monitor
W screen
Welding nozzle \ Pointer

Flux flow
T.V. Camera regulating valve
control panel Control panel
for welding head
I Operator's and roller beds
control
desk
Fig. 19.13: Diagrammatic arrangement of boom and controls for internal pipe welding equipment
Generally, pipes for the transmission of liquid prod- efficiency of pipes, and tests for the qualification of
ucts are smaller in diameter than pipes meant for natural procedures and welders.
gas. The common diameters used for gas transmission Pipeline welding: Most pipeline welding involves
are 600, 750 and 900 mm (24, 30 and 36 inch), though girth welding from external side only, because the
recently these have been increased to 1,400 or 1,500 mm. diameters are too small to permit welding from the
Transmission pipelines are usually manufactured to inside. The commonly used joint design is shown in Fig.
the API Specifications for Line Pipe. They specify, among 19.14. It is well suited for the stovepipe technique
other things, the strength levels of various steels to be described below. In special cases, the angle of bevel is
used, working stress levels and longitudinal joint increased from 30° to 37.5°.
»1.6 mm
Fig. 19.14: Standard joint preparation for pipeline welding
Internal backing rings are avoided as far as possible,

because they not only cause turbulence in the flow of
material, but also make it difficult to use devices for
internal p i p e cleaning. Moreover, the s t o v e p i p e
technique enables the welder to deposit sound weld-
metal at the root through the entire 360° in 5G position.
If welders cannot guarantee complete root fusion and
freedom from internal protrusions (icicles), the use of
backing rings is indicated.
Stovepipe technique: Stovepipe welding is the term used
when a number of pipes are laid and welded together
in 5G position one after another to form a continuous
line, and welding is carried out vertically downwards,
and not by the conventional vertical upwards method
which is time consuming and expensive.
Fig. 19.15: Stovepipe technique; positions
In this technique, welding starts at the 12 o'clock
for stripper beads
position on the pipe, and progresses vertically down
until the 6 o'clock position is reached. On completion
of one-half of the pipe, the opposite side is welded in shield of carbon monoxide and hydrogen evolved from
the same manner, thus producing an endless root run the cellulosic coating during welding.
known in the field as a stringer bead. For stovepipe w e l d i n g , the m a x i m u m current
The second run, known as the hot pass, is then put specified by the producer for the size of electrode is
into the joint. Its name comes from the fact that a high increased by approximately 10%. DC supply with elec-
current is used to deposit the run, so as to burn out any trode positive (positive polarity), is often recommended.
defects that may be present from the stringer bead. With There may be occasions, however, where scale on the
the exception of the final run, all subsequent runs after pipe causes surface porosity. In such cases, changing
the hot pass are termed filler beads. Their purpose is to the electrode polarity from positive to negative tends
bring the weld deposit to just below the level of the to reduce this problem.
pipe surface. The number of filler beads required will For deposition of the stringer bead (root run), once
depend largely on the pipe-wall thickness and the the arc has been established, the cup of the electrode
preparation. must be literally pushed into the root of the joint. No
There are times, however, when it is unnecessary to weave of the electrode is necessary, only a tight drag
deposit a filler bead all round the pipe periphery, action as welding proceeds, to ensure that the arc is
especially as the weld nears completion. In most cases allowed to burn inside the pipe. An electrode angle of
only the areas between 2 to 4 and 10 to 8 o'clock on the 60° in the direction of travel to the pipe tangent (see
joint (see Fig. 19.15) will require additional weld-metal. Fig. 19.16) must be held throughout.
These concave areas are rectified by the quick deposition This practice produces a very small root run, which
of a weld run called a stripper bead, which brings the allows for a controlled penetration bead. If one or more
concave area flush with the remaining weld-metal burn-throughs (windows) occur during the laying of the
elsewhere in the joint. To finish the pipe weld the final stringer bead, they can be quickly rectified by the
run is made, which is appropriately called the capping remelting process of the second run.
bead. Immediately following the stringer bead and while
The joint preparation and fit-up is as shown in Fig. it is still warm, the hot pass is p u t d o w n with an
19.14. Welding is done with AWS E6010 and E7010 class electrode angle held at 60° to the pipe tangent. A short
electrodes. These electrodes are chosen because the small arc must be held with a light drag, together with a
volume of stiff, thin slag coating deposited on the weld forward and backward movement of the electrode (see
bead, together with the forceful arc, facilitates rapid Fig. 19.17), in order to fuse out any undercut a n d / o r
changes of electrode angle d u r i n g vertical-down wagon tracks, caused by the stringer bead. In addition
welding on fixed pipes. to remelting the portions containing windows, the
To compensate for the thin slag coverage, extra higher current used for this run prevents the formation
protection from the atmosphere is provided by a gaseous of slag lines at the toes of the stringer bead.
Welding
direction Welding
direction
Finish
Finish
Fig. 19.16: Stovepipe technique; electrode angle during

deposition of the stringer and hot pass runs Fig. 19.18: Stovepipe technique; electrode angles for
filler and capper beads. From positions * electrode
angle changes from 90° to 130°.
Hot pass
the electrode may be found beneficial, depending on
Direction) the current setting and width and depth of the bead
of required. The angle of the electrode is held at 90° to the
welding pipe tangent, irrespective of the position on the pipe
periphery.
Finally the capping bead completes the joint, using
a medium to long arc length, with a rapid side-to-side
Stringer bead movement of the electrode tip. The angle is maintained
Fig. 19.17: Stovepipe technique; electrode manipulation at 90° to the pipe tangent, except from 4 to 6 and 8 to 6
during deposition of the hot pass o'clock positions, when the electrode angle is increased
to 130°.
For the filler bead deposition, it is necessary to alter For these sections, the tip of the electrode should be
the electrode angle from 60° to 90° to the pipe tangent. manipulated to produce a lifting and flicking action. To
However, on reaching the 4 o'clock position (8 o'clock achieve best results, the capping bead should be
on side 2 of the pipe), the electrode angle is increased restricted to the width and depth of~19Xl.6 mm. Weld
from 90° and reaches 130° at the 6 o'clock position of beads wider than this are somewhat difficult to control.
tire pipe (see Fig. 19.18). The electrode size for various passes depends on wall
From the 12 o'clock position down to 4 o'clock (8 on thickness. For depositing the stringer bead, for example,
side 2), a normal arc length with a rapid weave across 3.25 mm diameter electrode is used for wall thickness
the weld face is required, pausing momentarily at the below 6.3 mm, and 4 mm diameter for larger thicknesses.
toes. From 4 o'clock (8 o'clock) down to the 6 o'clock For first and second filler passes, 4 mm diameter
position, the electrode manipulation is changed from a electrode is commonly preferred. For third filler, stripper
weave to a lifting or vertical movement of the arc away and cover passes, 4 or 5 mm diameter electrodes are
from the deposit on to the weld pool. By adopting this used depending on wall thickness.
technique on the filler beads, flat weld faces with the It is difficult even for a normally well-experienced
absence of undercut are produced. welder to use stovepipe technique successfully, unless
For the stripper beads, a medium to long arc is he is given special training with suitable electrodes on
required to spread the weld deposit. A slight weave of actual pipe joints. Experience has shown that only about
20% of the otherwise skilled welders are capable of considerable spatter when C 0 2 is used for shielding. The
mastering the stovepipe technique. dip transfer technique using argon/ C 0 2 mixture for
The adoption of stovepipe technique in pipeline shielding is better suited for 360° welding, but the
construction demands a well-planned disposal of the shallow p e n e t r a t i o n of this process can lead to
crew, in order to ensure that welding operations take incomplete fusion. Moreover, the u p k e e p of the
place rapidly along the line. The pipes are first lined up equipment at site demands the services of properly
by the line-up crew with the help of an internal line-up trained mechanics and a regular supply of spares.
clamp. A good joint fit-up is the necessary condition for For the welding of pipes large enough to accommo-
a flawless, well penetrated stringer bead, and it is the date a M I G / C 0 2 welding head inside, fully automatic
responsibility of the line-up crew to ensure it. Two equipment has been developed. A typical piece of equip-
welders then complete the stringer bead (first pass). The ment consists of four welding heads, mounted at 90°
line-up men and these welders then move on to the next spacing, for internal welding and two welding heads
joint, while a second group of welders deposits the hot for external welding. The two top internal welding
pass (second pass). They then shift to the next joint, heads proceed simultaneously from the top of the pipe
while the third group of welders completely fill the joint. downward to make the weld. The two opposite inter-
The third group, called firing line, includes a larger nal heads then counter rotate to complete the joint. The
number of welders, since more welding is involved in external welding units are light and portable, and they
completing the joint. The stringer welders and the hot are used in conjunction with a tracking band, which is
pass welders work in groups of two or four. attached around the pipe at a fixed distance from the
Stovepipe technique is not possible with rutile type weld. The two units operate simultaneously on each side
(E6013 class) electrodes, because the relatively large of the joint, proceeding from the top of the pipe down-
volume and high fluidity of the slag render vertical ward. It is also possible to use the external units simul-
downward welding difficult. With these electrodes, taneously with the internal units.
good joints can be made by welding vertically upwards. For the internal weld which is made first, a small V-
But the technique is slow a n d results in lower groove is provided. For external welding, a V-groove
productivity. with 20° included angle is adequate to ensure complete
LH electrodes: In recent years, increasing use is made fusion. This means reduced weld-metal required to
of high-yield steels for pipeline, for example, the 5Lx60 complete the joint. The welding wire is of 0.8 mm
and 5Lx52 steels. These steels are more p r o n e to diameter and the shielding gas is 75% argon - 25% C0 2 .
h y d r o g e n - i n d u c e d cracking in the H A Z than the This argon-rich shielding reduces spatter to the
conventional mild steel. Hence the pipe ends need to minimum.
be preheated when E6010 - E7010 electrodes are used. The system m a y also i n c o r p o r a t e a pipe-end
When this is done, the stringer pass and the hot pass preparation machine, which is used ahead of the
have to be made with an increased speed of 230-300 welding operation. The internal welding machine may
mm/min. This increases the strain on the welder. be combined with a line-up clamp.
Special LH electrodes have been developed for Such systems have been used with success for
welding 5Lx52 and 5Lx60 steels using the stovepipe various onshore and offshore construction projects in
technique, without the need for preheating. With these the U.S.A., Canada and England.
electrodes, the root gap is increased to 2.5 m m to Flux-cored process: A typical system utilising this
accommodate the heavier coating and the welding process incorporates an end preparation machine and
speed is kept as low as 150 m m / min. The disadvantage makes all the weld passes from the outside. It uses two
of reduced speed is more than made up by the thickness welding heads, mounted 180° apart, for the root pass
of the root pass, which is twice that deposited with E6010 a n d four w e l d i n g h e a d s , spaced at 90°, for the
type. The deposition efficiency of the LH electrode being subsequent passes. The root pass is deposited over a
20% higher than the E6010 type, the joint can be copper back-up attached to a specially designed internal
completed with fewer layers and in shorter arc time. line-up clamp. All welding proceeds from the top to the
MIG/COz process: The inherent advantageous features bottom. The flux-cored w e l d i n g wire is of 2 m m
of this process could make it preferable to MMA diameter. No external gas shielding is used, which is a
welding, but there are several difficulties. The normal welcome feature for site welding. The joint consists of
spray transfer technique which is capable of giving high 58° included angle, 1.6 mm root gap and 2.5 mm root
deposition rates would give rise to burn-through and face.
Underwater pipelines: Pipelines for underwater service into the pipeline. Such an equipment can travel several
are laid in m a r s h y land, shallow w a t e r s or in kilometres t h r o u g h a pipeline, thus enabling the
considerable water depths. MMA process is commonly contractor to proceed continuously with welding
used for welding. The welders work at stations located without waiting for radiographic inspection to catch up
on barges. The pipe laying starts from the land or shore with him.
and proceeds towards deeper waters. As many as five For small diameter pipe, radiography has to be done
welding stations may operate on several barges, from outside. In this case, the source is placed on one
followed by two radiographic stations and a coating side and the film 180° opposite. At least three exposures
station. Coating is meant for corrosion protection. Large are necessary to cover the entire joint, and increased
diameter pipes are preferably concrete coated to provide exposure time per exposure is required. Hence external
corrosion resistance as well as negative buoyancy. In radiography is more time-consuming than internal
offshore construction, the completed pipe section is radiography. Other NDT methods are rarely used.
lowered gradually by means of a semi-buoyant stringer, Ultrasonics, for example, cannot perform reliably because
which holds the pipe until it has neared the sea bed. of the irregularities of the manual-arc welded stringer bead
After laying, the pipe is buried in the sea bottom. and cover pass.
Inspection and testing: For i m p o r t a n t pipeline Sometimes the completed pipeline needs to be
construction, the welding procedure as well as the pressure-tested prior to being placed in service. The
welders must be qualified. The necessary guidance is common practice is to test it hydrostatically with water
obtained from any of the following or equivalent to stress levels equal to the actual yield point of the base
standards: metal.
a) API Standard 1104, Standard for Welding Pipelines
and Related Facilities 5) Power Generating Plant
b) ASME Boiler and Pressure Vessel Code, Section In power generation, whether it is thermal, hydel or
IX nuclear, welding plays an important role at every stage
c) ANSI B 31.8. Code for Gas Transmission and of fabrication and erection. The main components of a
Distribution Piping power plant are: pressure vessels like boiler drums,
In the qualification test, a sample pipe is welded in economisers, superheaters, heat exchangers; reheaters
accordance with the procedure adopted and coupons and waterwall pipings; turbines and generators.
are removed by gas cutting. They are then subjected to Welding techniques and procedures applicable to
various tests such as tensile, nick-break, root and face- pressure vessels, tubings and pipings have been
bend tests. If these tests meet the code requirements, described in sufficient detail in the earlier sections.
the welder or procedure is taken as qualified. While the early steam generation plants were of 60
Inspection is carried out both during and after MW size, 200 and 500 MW plants are becoming more
welding. During welding, the points to be checked are : common. A 200 MW boiler involves about 50,000 welded
(i) edge cleanliness, edge preparation and joint fit-up; joints, in a d d i t i o n to the w e l d s in turbines and
(ii) physical condition of the electrodes; (iii) functioning generators. Of these, nearly 30% have to be made at
of the power source and current setting; (iv) soundness erection sites. The materials to be welded range from
and penetration of the stringer bead; (v) soundness and mild steel to sophisticated alloys like stainless steel,
quality of hot passes; and (vi) interpass cleaning. After monel and inconel. Thickness range is from 0.5 to 300
welding, the joints are subjected to visual a n d mm. Considerable amount of welding is carried out at
radiographic inspection. The latter is carried out with site using m a n u a l w e l d i n g . For shop welding,
X-rays or gamma-rays. Special radiographic equipment submerged-arc, electroslag, MIG/C0 2 , resistance and
has been designed for large diameter pipelines, which induction welding processes are used to the maximum
enables the X-ray or gamma-ray source to be propelled extent. Other interesting welding applications in a
through the pipeline on a battery-driven or engine- thermal plant are: welding of stubs to headers, automatic
driven crawler unit. The unit is provided with a welding of peg fins to tubes, spot welding on the
mechanical or radiological device to locate and stop at collecting electrodes of electrostatic precipitators,
a welded joint. Film belts are wrapped around the joint projection welding of studs to the valve bodies, welding
circumference to radiograph the entire joint in one of columns and structurals. In the manufacture of spirals
exposure. The unit is programmed for speed, exposure for HP heaters, where the quality requirements are very
time and other radiography parameters before insertion stringent, TIG process is preferred. Automatic hot-wire
TIG welding is well suited for the welding of tubes in A 283 is a normal mild steel grade, and for penstock
boilers and HP heater manufacture. application the carbon content is restricted to 0.30%
For the manufacture of transformer cases and maximum. ASTM A 442 is made with definite limits
capacitor boxes, electric resistance process can be used on Mn and Si contents and carbon is held low. Plates
with advantage. Impellers of boiler feed pump, valves, thicker than 25 mm are made to fine-grain practice and
etc., are stellited or hardfaced with other suitable alloys may be normalised if so stipulated. ASTM A 516 is
using automatic e q u i p m e n t . Large forgings a n d available in several grades with increasing strength
castings are sometimes made by joining of smaller levels a n d is m a d e to fine-grain practice in all
forgings a n d castings by the s u b m e r g e d - a r c or thicknesses. H e a v i e r plates are s u p p l i e d in the
electroslag process. For bimetallic finned tubing, normalised condition and the lower thicknesses are
induction brazing may be used. normalised if so stipulated. The A 537 Grade A and B
As mentioned earlier, for high temperature and steels are two varieties of weldable steels with the
pressure service, creep-resistant alloys such as 1 Qx/Vi mechanical p r o p e r t i e s well s u i t e d for penstock
Mo, 1>A C r / 1 Mo, type 347 and type 304 stainless steels application. Both are fine-grained steels. Grade A is
are used. For sulphur-rich oil fired boilers, it becomes normalised in all thicknesses and Grade B is quenched
necessary to use an alloy such as 12 C r / 1 M o / V type and tempered. A large number of high head and large
martensitic steel for the tubings and castings of turbines diameter pipes in the U.S.A. and elsewhere have been
and generators. Such an alloy d e m a n d s a special built with the Grade A steel. A 517 is a high-strength,
welding procedure. quenched and tempered alloy steel, which is used for
Steam and hydro-turbines demand high quality the very high heads and combinations of high heads
w e l d i n g . For example, the d i a p h r a g m s of steam and large diameters. Compared to a low-carbon steel,
turbines made of Cr-Mo-V steel have to be welded to A 517 will require much reduced thickness and will not
extremely close tolerances. For this application, a typical require costly postweld heat treatment.
automatic unit consists of two C 0 2 welding machines In the absence of recognised specifications for the
complete with synchronised welding heads and turn- d e s i g n a n d fabrication of p e n s t o c k s , design
tables. requirements are based on pressure vessel standards
Penstock pipes: Penstocks have an important place such as ASME Boiler and Pressure Vessel Code, Section
in a hydel project. They are laid on an inclination to VIII, Div. 1 or 2. The diameter of penstock is determined
convey water under pressure from a large reservoir from various factors such as fabrication cost, value of
located at a high altitude to hydraulic turbines at a power lost due to flow in the pipe, lowest annual cost
lower level to generate electric power. They may extend including depreciation and fixed charges.
in length from a few h u n d r e d metres to several In the design of penstocks, efficiencies of welded
kilometres, and the pipe diameter may vary up to a butt joints must be determined from the relevant
maximum of 15 m. Plate thickness can be as high as 60 sections of the ASME Code. Special attention must be
mm. paid to longitudinal joints, since they are subjected to
Smaller diameter and lower pressure pipelines may the full operating head. These joints must be subjected
be made from structural quality steels, such as ASTM to complete radiographic inspection to ensure that they
A 283, Grades B and C. For large diameter high pressure are free from defects. Preheat and p o s t h e a t
penstocks, pressure vessel quality steels such as ASTM requirements must also be considered. On high pressure
A 442, A 516, A 537 and A 517 are preferred. These ASTM lines using plate thicknesses over 38 mm, postweld heat
steels are described as below: treatment becomes necessary. This costly operation may
A 283 Low a n d Intermediate Tensile Strength be avoided by using a higher strength steel, which
Carbon Steel Plates of Structural Quality. permits thinner plates to be used. The advantage of
A 442 Carbon Steel Plates with Improved Transition using quenched and tempered steels has been pointed
Properties for Pressure Vessels. out in Chapter 10.
A 516 Carbon Steel Plates for Pressure Vessels for The designer must also consider the selection of
Moderate and Lower Temperature Service. appropriate grades of steel for structural steel supports,
A 537 Carbon-Manganese-Silicon Steel Plates, Heat stiffener rings and flanges for connection to gates, valves
Treated, for Pressure Vessels. or turbines. Pressure surges and impact loads due to
A 517 High-Strength Alloy Steel Plates, Quenched surge conditions must be taken care of by adjusting the
and Tempered, for Pressure Vessels. profile of the penstock suitably, and reinforcing the shell
with stiffeners at critical points along the profile. The important Indian penstock projects have been described
design of support rings and the length of span between in the following paragraphs.
supports must receive special attention. When the Kundah project: This project is located in the Nilgiri
penstock header has to feed two or more turbines or hills. Its plan is shown in Fig. 19.19. By building two
divert a part of the flow to an irrigation line, the designer dams. Avalanche and Emerald reservoirs were created
must consider the design of branch outlets or lateral in a wide catchment area. Water from the reservoirs is
connections between the header and the branch pipe. carried by a 162 m long low-pressure welded steel
Such outlets or lateral connections require special pipeline to the mouth of a 4,200 m long tunnel, which
reinforcement to compensate for the metal removed for extends to the other side of the hill. From the exit end of
hydraulic reasons. Exposed penstock must be provided the tunnel, which is at an altitude of 1,966 m the water
with expansion joints to prevent the b u i l d - u p of jumps down a height of over 360 m through three all-
temperature stresses caused by changes in atmospheric welded steel penstock lines to provide energy for the
temperature. running of turbines in power house No. 1. The exit water
To u n d e r s t a n d the techniques involved in the from the power house No. 1 pours into another reservoir
fabrication and erection of penstock lines, a few of the called the Kundah-Palam reservoir, for the completion
of which two more dams had to be built. From here also,
water is led in turn through a 196 m long low-pressure
all-welded steel pipeline, through a 3,988 m long tunnel
Emerald inside a hill, and then down through five all-welded
Avalanche reservoir steel penstock lines to drive the turbines in power house
reservoir
No.2.
Mild steel pipes for low-pressure pipelines and high-
tensile T-l steel pipes for high-pressure penstocks were
L P . pipeline imported from Canada in 7.2 m lengths. Each length
consisted of three 2.4 m long shells welded together,
and therefore contained three longitudinal welds and
two circumferential welds. These welds were completed
in the shops of Davie Shipbuilding Ltd. of Canada by
the automatic submerged-arc process. Site welding at
3 penstock lines
K u n d a h , therefore, consisted of m a k i n g only
PJH.No.1 +5335' circumferential joints by manual metal-arc welding.
Kundah-Palam Gasoline engine driven welding sets, electrodes and
reservoir certified welders for site welding were brought over
from Canada by Davie Shipbuilding Ltd., whose
representatives were in complete control of pipeline and
penstock welding at Kundah.
Welding on low-pressure mild steel pipeline was
carried out with AWS E6010 Class electrodes. Welding
on high-pressure T-l steel penstocks was carried out
with high-tensile low-hydrogen electrodes of AWS
E11015-16 class.
5 penstock lines All the electrodes b r o u g h t from Canada were
contained in hermetically sealed cans and these were
PJH. No.2 +2860' stored on high wooden racks in a dry storage room. But
since low-hydrogen electrodes are extremely susceptible
to moisture pick-up, the electrodes on use at site were
carried in specially designed electrode stabilising ovens.
The oven consisted of a simple cylindrical steel container
with insulated lining and an electrical heating element
Fig. 19.19: Plan showing penstocks, power houses and
reservoirs at Kundah Project of the resistance type. The oven heats the electrodes up
to a t e m p e r a t u r e of 150°C. The electrodes stand
vertically on a handle that can be raised up from the

top to pick up the electrode. The oven is illustrated in
Fig. 19.20. Handle to be raised up
The edges of the circumferential joints of pipes to be for picking the electrode
welded at site were already prepared before being sent
to India. The edges were bevelled so as to give a 60°
included angle and a shoulder of 3 mm at the butt joint. Insulated
As shown in Fig. 19.21, the top half of the pipe has the walls
root on the inside, so that the groove is filled up by
welding from the exterior of the pipe. The root is then
Electrodes
gouged out by Arcair carbon electrode to sound metal
from inside the pipe, in the overhead position, and the
final sealing pass is made also in the overhead position.
The bottom half of the pipe has the root on the outside,
so that the groove is filled u p by welding from the
interior of the pipe. The root is then gouged out by Arcair
carbon electrode to sound metal from outside the pipe
in the overhead position, and the final sealing pass is
also made on the outside in the overhead position. A
metre d e e p trench h a d to be d u g in the g r o u n d
underneath the pipe joint for the last two operations. Lamp
Extreme care and skill on the part of the operator
was found necessary in laying the first pass on US T-l
steel. The pass is very susceptible to porosity, which may
occur at any point and at any time. The welder must
immediately detect and gouge out the defective portion
and reweld. Such care and conscientious work on the
part of the welder can save considerable labour and
expense that may be incurred, if the defects are left over Fig. 19.20: Electrode stabilising oven
to be subsequently detected in radiographic testing.
Across AXB
v-"-v
70
\ / A 1.6 mm
1.6-3 mm
v*V
Across AYB
\a
H|— T
1.6mm
1-6-3
Fig. 19.21: Edge-preparation for circumferential joints to be welded at site

Radiographic inspection was carried out at site. A design stress of 18,300 psi (125 N/mm 2 ) is taken, while
200 kVA portable X-ray machine called Baltospot 200 for ASTM A-517 Steel (quenched and tempered) the
and manufactured by Balteau Electric Corporation of stress is 38,300 psi (262 N/mm 2 ). Therefore for a 32 mm
Connecticut, U.S.A. was used. The unit weighed only thick plate of A-285-C steel, the equivalent thickness
35 kg. Another 120 kVA unit was used later side by side. in A-517 steel would be 15.2 mm (say 15 mm). While
About 3,000 m of welded joints were radiographed on the cost of A-517 steel would be about 1.6 times, the
the project. tonnage to be imported and handled would be only
For use in places where access for heavy X-ray about 0.4 times, thus effecting a saving of 35% in the
machines was difficult and as a stand-by in case of cost of steel itself. The weld-metal requirements will be
damage to X-ray machines, a gamma-ray camera was reduced by 56%. The percentage saving in the weld-
also provided. It carried a Cesium 137 isotope. metal goes on increasing as the thickness goes on
While X-rays required an exposure time of only three increasing. There is saving in almost every operation in
to five minutes on steel thicknesses u p to 25 mm, higher tensile steel, except rate of consumables, the total
gamma-rays required an exposure time of 50 to 60 min. cost of which is also less as the saving in requirements
However, as gamma-rays are emitted by an isotope more than compensates for the higher rate.
source over 360°, filming of an entire circumferential For butt welding plates, the general practice is to
welded joint was possible in one exposure. A one-hour adopt 60° V-groove with a root gap of 3-4 mm and a
single exposure was made on a circumferential joint 11.0 shoulder of 3-4 mm. The root gap can be conveniently
mm thick and 1.42 min diameter using 12 Dupont 506 reduced to 1.5 mm or even nil in carbon steels like A -
slow films wrapped around on the exterior of the pipe 285-C, while using the automatic sumberged-arc
joint. The isotope was located exactly at the centre of process, since high welding currents are permitted to
the pipe after being pushed out of the heavy lead be used on this steel. The angle of V-groove can also be
container by means of an extension rod. The necessary reduced to 45°. This results in a saving of 30% in welding
precautions were taken by keeping people at least 10 m consumables for plate thickness of 15 mm. This saving
away from the source d u r i n g the exposure. The will increase as plate thickness increases.
radiographs were clear enough and had a sensitivity of Though it is considered impractical to reduce the V-
3% or below as shown by the impressions of the groove angle to less than 60° for manual welding, some
penetrameter. fabricators have successfully reduced the angle to 45°
This project called for considerable planning, skill in thickness of 20 m m a n d below, and obtained
and experience, even though by comparison it was a acceptable radiographic results. In one instance, instead
simple straightforward job. It involved small diameters, of bevelling both edges, only one edge was bevelled to
1.27 to 1.57 m, and thicknesses only up to 14.28 mm. 45° and the other edge was kept straight, thus effecting
Many projects involving penstocks of larger diameters some saving in the cost of edge preparation also. Use of
and of considerably higher thicknesses have been two torches mounted on the oxyacetylene cutting
successfully executed in India since then. machine reduces time and cost of such edge preparation.
Among them is the Kalinadi project involving As far as practicable, welding should be done in the
Welten-80, a Japanese quenched and tempered low-alloy d o w n h a n d position, because in this case heavier
steel. Before taking it up, it would be worthwhile to electrodes can be used, which are usually less costly per
discuss the productivity aspects of penstock fabrication ton than thinner electrodes. For example, changing over
in a general way. from 4 to 5 mm electrodes reduces the cost of labour
High strength steel plates of maximum possible plus overheads by as much as 25-30%.
width, in line with the capacity of the bending roll must It is always economical to use low-hydrogen iron
be procured to minimise the total tonnage of steel to be powder electrodes. Though the price of these electrodes
handled and the number of girth joints. The plate lengths is higher, the ultimate weld deposit is definitely cheaper,
should also be such as to reduce the n u m b e r of because of higher deposition efficiency, higher rate of
longitudinal joints in accordance with the shop handling welding, reduced overheads and less completion time.
facilities. In a typical shop, it was found that single plates Besides low-hydrogen electrodes are less prone to
up to 11 m length, suitable for shells of approximately underbead cracking. Where preheating is required, iron
3,350 mm diameter, could be conveniently handled. p o w d e r l o w - h y d r o g e n electrodes also save in
The economy of using high strength steel can be preheating costs, because of higher heat input obtained
demonstrated thus: For ASTM A-285 Grade C Steel, a with these electrodes.
Maximum possible welding in penstock fabrication be successfully overcome by adopting unequal V-

should be done by automatic welding processes. It grooves having a difference of 4-5 mm and welding the
requires some experience and skill to control the deeper side first. It will be advisable not to complete
curvature at the longitudinal joints when butt welding the first side fully, but leave about two to three layers
with the automatic or semi-automatic process. For girth and lay these depending upon the behaviour of the joint.
or circumferential seams, all welding should be done This deformation or bend at the joint is undesirable,
by automatic process to save completion time and especially in high strength steels exceeding 60 kg class,
reduce welding costs. The chances of distortion are as the joint is liable to crack even under low loading.
almost negligible in girth joints. According to experience, the limit of this bend has been
Devices like turning rolls and rotators greatly determined at 2.5° as shown in Fig. 19.22. Control of
facilitate the work by bringing the workpiece quickly this bend is very important as re-rolling can be avoided.
in the most convenient position and dispense with the When attempt is made to correct excessive bend by re-
need of helpers to a great extent. They also permit the rolling, cracks occur in the weld.
use of large sizes of electrodes by presenting the joint in For matching the abutting edges, the usual practice
the downhand position and increase operator factor. is to weld an L-shaped lug on the lower surface and
Accurate dimensioning, correct edge preparation, drive a wedge between the overhanging arm of the lug
close fit-up and proper alignment contribute to saving and the higher surface as shown at (A) in Fig. 19.23.
in requirements of weld materials and permit use of This involves unnecessary welding and wastage of lug
proper sizes of electrodes by eliminating the necessity steel. The welding sometimes causes damage to the base
of using smaller diameter electrodes to prevent burn- metal, which has to be repaired before the job can be
throughs. Badly cut joints mean wider V-groove, accepted. Besides, this operation involves cost of
requiring more weld-metal than necessary, necessitate fabricating the lugs and removing them by cutting or
building up of edge, more cost in back cleaning of the gouging afterwards.
joint, longer time of execution and increased distortion. A considerable saving can be effected in this
Overwelding and throwing away of long stub ends can o p e r a t i o n by u s i n g a jig w h i c h goes r o u n d the
increase cost of welding by as much as 50% or even circumferential joint and matching the edges by simply
more. driving the wedges at the required points. The jig, shown
The submerged-arc process can also be used for in Fig. 19.23 (B), is made in two semicircles and is easily
welding stiffener rings, in which the welding time is put on the joint. Such a jig is also suitable for alignment
reduced by 50-60% as compared to manual welding. of longitudinal joints.
Experience has shown that an automatic SA welding
installation, if used in a planned manner, can give output
equal to eight to ten manual welders.
Double bevel or X-grooves are commonly used for max. limit
heavy butt welds. The so-called equal grooves in this
type of joint are in fact unequal, as after back gouging,
the second side (back side) becomes deeper by 4-5 mm.
This occasionally presents problems in the control of Fig. 19.22: Curvature control of longitudinal joint of pipe
curvature at the longitudinal joints. This problem can
Wedge
Lug
Wedge
n Alignment
iiJ ^li
Fig 19.23: Jigs for pipe alignment
Chipping and cleaning of each completed weld layer

before commencing the next one is an essential part of
the welding job. Proper tools and more joints in hand
(about twice the number of welders in a shift) give
optimum outturn by keeping the welders busy for most
of the time. Cleaning, chipping and welding should
proceed on alternate joints, thus keeping workmen busy
at all times and resulting in maximum utilisation of Radiographic source are fixed on
labour. Treatment of back side of welding groove can (Isotope) a || joints
be greatly expedited by the combined use of arc-air Fig. 19.24: Improved radiography technique for pipes
gouging and high speed grinding wheels.
Welded joints m u s t h a v e some m i n i m u m
reinforcement to ensure that no point on the weld seam cheaper but has other advan-tages. Testing can go on in
is below the plate surfaces. Excessive reinforcement, say, the shop along with other jobs at all times, as no
above 1.5 mm for 12 mm plates, above 2.5 mm for 13- radiation hazards are involved. At erection sites, when
25 mm plates and above 3.0 mm for plates more than 25 penstocks are being concreted from the outside, the
mm thickness is not permissible. It causes wastage of concreting need not wait for the completion of testing
weld-metal and also results in head losses on the inside and repair, as would be the case with radiography,
surface of penstocks. Besides, this extra reinforcement because ultrasonic testing can be done even with
needs to be removed and this involves avoidable labour concrete on the outside. For radiography, the other side
and materials for grinding off. The ideal joint would be has to be approachable for affixing the films, while the
just flush with plate surfaces. radiation source is on the inside.
Radiographic inspection is an important step in All specifications provide for hydrostatic testing of
penstock fabrication, especially for longitudinal pipe penstocks with a view to check any hidden defects. This
joints, which are subjected to maximum stress. Weld operation incidentally also effects mechanical stress-
surface irregularities, especially low points, are built up relieving to a certain degree. While all straight sections
beforehand and some grinding on the interior surface can be tested on a hydrostatic test structure, this is not
is also done, so that radiograph contrast resulting from possible for bends and conical sections, which will
any irregularities does not mask or is confused with that require welding of bulkheads and holding of bends from
of any objectionable defect. Either an X-ray machine or opening out. It is permissible to test straight sections
a gamma radiography source is used. A common individually or in assemblies, depending upon the
procedure is to radiograph the longitudinal joints design of test structure, and then cut the sections to the
separately, covering about three to four films (about 38 required profile and weld to make bends. The final girth
cm length each) in one exposure, thus requiring about welds will have to be radiographed or ultrasonically
eight exposures for four longitudinal joints of two shells tested and this p r o c e d u r e has been accepted as
of 2 m length each. This procedure increases the hazard satisfactory. While making assemblies for hydrostatic
of radiation absorption by the radiography operator. test (if it is a must, depending upon structure design
One fabricator has successfully radiographed one girth which may be suitable only for a fixed length), it is not
joint (after joining two shells) and four longitudinal at all essential to fully weld such temporary girth joints.
joints each of 2 m length of a 4,880 mm diameter and 50 About 20-25% welding on circumferential joints has
mm thick penstock in one exposure with a gamma-ray p r o v e d more t h a n satisfactory. This results in
source, thus cutting down the radiography time and considerable savings.
labour by 40% (see Fig. 19.24). This has also reduced The conventional hydrostatic test structure shown
the cordoning off period and the disturbance caused to in Fig. 19.25 involves closing or sealing the ends with
other works in the vicinity. bulk-heads fitted with rubber seals, in which huge
The stress in the girth joint is about 50% of the stress holding down bolts and blocks are used to counteract
in the longitudinal joint when there are no expansion the force on account of test pressure. This also involves
joints or couplings. When expansion joints are provided, consider-able time and labour for fixing of blocks and
this stress Considerably reduces in the direction of the tightening of heavy bolts. A simple and quick test
flow. Ultrasonic testing can be effectively substituted structure shown in Fig. 19.26 can save all this trouble
for radiography for girth joints. This method is not only and increase the speed of testing four to five times.
Top seal
Detail-A
-4- *-4
IlL
i M m m aH3 m> 1
Water spaceii
127 mm i!
i! ftar
Top seal Detail-A

AJternative to Detail-A
B o t t o m seal
g Ufi HI
fl . D e t a i l - B Q } a Q 1 N
B o t t o m seal D e t a i l - B Alternative to D e t a i l - B
Fig. 19.25: Conventional hydrostatic test structure
Longer assemblies in accordance with handling and three days a joint (aggregate of many months for the
transportation facilities greatly reduce the cost of work entire project), thus not only slashing costs but greatly
and its completion time. It has been observed that by increasing overall productivity.
making an extra circumferential joint in the shop instead Proper care of electrodes and other consumables
of at the erection site (or field), one can save up to Rs. through dry storage facilities, redrying ovens and
200 per metric ton, i n c l u d i n g the saving in portable holding ovens not only reduces wastage but
transportation, crane handling, site alignment, site minimises welding defects, thereby lowering the cost
welding and welding consumables. The completion and adding to productivity.
time of the job is also greatly reduced at a rate of up to Preheating of joints to 95-100°C can be substituted
for stress-relieving u p to 38 mm thickness and may
eliminate the necessity of setting up a stress-relieving
Detail-A furnace for some projects. No preheating and stress-
relieving is needed for carbon steels up to 32 mm plate
thickness. Where preheating is a must, a controlled
speed of welding and use of thicker electrodes can save
Water space
75 mm
preheating costs up to 20-25%, as interpass temperature
can be maintained with reduced preheating, while using
the heat of the welding arc retained in the plate.
High pressure Welding of penstocks at site involves welding in all
pipe for annular positions. This situation can be greatly overcome (as was
water surface done at Kundah Project) by providing a V-groove in the
..i upper-half circle on the outside and in the lower-half
circumference on the inside, with transition at the
High pressure horizontal centre of the pipe. This will mean provision
pipe for Detail-A of single V-groove instead of X-groove, which becomes
rubber seal
uneconomical as thickness goes beyond 25 mm and is
not recommended beyond 32 m m plate thickness. This
will involve somewhat more weld-metal requirement;
b u t h i g h e r w e l d d e p o s i t i o n rate a n d less costly
electrodes of larger diameters and comparatively less
skilled welders will in many cases offset the cost of extra
Fig. 19.26: Simplified hydrostatic test structure
weld-metal.
In tunnels excavated in rock, there is usually high

*•—r Surge
humidity, because of seeping water. The space between
L i--:rrrc=—OChamber
the liner and the tunnel is usually 600 to 750 mm. This
restricted space with high humidity and poor ventilation
greatly reduces the welding efficiency. Visibility is redu-
ced d u e to welding fumes and steam formation,
especially when concreting is following the installation.
These prob-lems are minimised by providing ventilation
blowers and handling/hauling larger length assemblies.
In many cases, the liners are joined and welded near
the tunnel face (portal) and pushed in part of the way
in the tunnel, as each assembly/liner is jointed. In this
way, most of the field welding is done outside the
congested and moist atmosphere, at a lower cost and
with better quality
For economical productivity in penstocks, a lathe-
turned roundness and workmanship should not be Fig. 19.27: Schematic diagram of
demanded and in fact cannot be justified. Out of Kalinadi hydro-electric project
roundness of 1% between maximum and minimum
diameters and minor (acceptable type) welding defects finger bars are welded to keep the joint in correct align-
should be accepted, as joint efficiency assumed in design ment before welding. Joint preparations adopted for
takes care of these factors. welding the Welten-80 steel, both at shop and field are
Kalinadi project: This project constitutes the first stage shown in Fig. 19.28. For one-side welding of circular
of hydro power development from the river Kalinadi,
which flows westwards in the Dharwar and North
Offset Less than 2 mm
Canara districts (Karnataka, India) and finally joins the
Arabian Sea. The Kali valley has an estimated power Fit up Less than 2 mm
potential of 1.3 million kW, which will be developed in Root gap For shop 5 mm
several stages. It is proposed to install six 135 MW for field 6 mm
generators at Nagjhari power house and two 50 MW Root face 3 mm
generators at Supa power house in the first stage. The
♦ Strong back, key plate
gross head is 560 m. Water will be supplied from the Key nut
surge tank through three main pressure shafts 4.75 m Carrot wedge
diameter, branching off to penstocks of 3.5 m diameter.
The pressure shaft which has an inclination of 50° is Finger bar
lined with A-285 Gr. C (M.S.) and the branch penstocks
are lined with Welten-80 steel (ASTM 517 Gr. F). Figure
19.27 shows the schematic diagram of the project. Tab plate
Welten-80 is a Japanese quenched and tempered low- Example of welding of plate
alloy steel which is an equivalent of T-l and T-1A of Backing plate
the U.S. Steel Corporation. It can be readily cold worked, 15(300)75 i.50 m m
J
cold bent and sheared. It provides high corrosion- and 3 mm
abrasion-resistance as it contains copper and chromium.
Fabrication of Welten-80 steel is based on ASME Sec.
VIII, 1971. All the plates are accurately marked and mm
trimmed to correct size by gas cutting or by edge
planing. They are then cold rolled to a true circle, once Field joint
•1st & 2nd pass
the edges are correctly made. The gas cut edges are manual backing
ground 2 to 3 mm to remove the hardened zone. Filler pass by A.W.
The edge-prepared plates after bending are matched
Fig. 19.28: Joint preparation for pipe welding
properly with matching screws and thereafter sufficient
joints, the edges are ground to prevent the occurrence Specifications of consumables for MMA and SA
of welding defects. The following conditions ensure welding of Welten-80 steel are furnished below:
good welding :
Root faces exceeding 3 mm are ground down to 3 A) Manual electrodes
mm. A root gap up to 6 mm is permissible, if the heat Classification: E10016(L-74)
input can be well controlled. Weld composition (%) C—0.07, Mn—1.20, Si—0.48
Before the fabrication is started, procedure tests and Ni—2.2, Cr—0.20, Mo—0.40
welders' qualification tests are carried out in accordance Mech. properties: UTS 74-85 k g / m m 2
with Sec. IX ASME Code or equivalent Indian standard. YP 63 k g / m m 2 min.
In addition to the tests specified in the Code, the Eln 20% min.
following tests were carried out on Welten-80 steel as Impact (CVN) 4.8 kg.
recommended by the steel manufacturer : m (-15°C) min.
1) Window cracking test Diff. hydrogen: Less than 3 ml/100 g.
This is a restraint cracking test, commonly used to
study the occurrence of transverse cracks in multi-layer B) SA flux-wire combination
weldments, in joints made out of 80 k g / m m 2 class tensile Solid wire C—0.18, Mn—1.64, Si—0.17,
strength steels. composition(%): Ni—2.56, Cr—0.52, Mo—0.70
2) Y-groove restraint cracking test Flux composition (%): Si02—38.6, CaO—15.2,
This test is the most common and is a highly effective MnO—18.1,
method of testing root cracking in tack welds and in the A1 2 0 3 —8.6, NbO—3.8,
initial layers of main welds. CaF2—14.2
3) Circular patch test Weld composition (%):C—0.06, Mn—1.26, Si—0.30,
In the welding of grout holes in penstocks, possibility Ni—2.38, Cr—0.40
of weld cracking is considered to be relatively high. The Typ. mech property UTS 84.8 kg/mm 2
circular patch test is to be conducted to investigate values: YP 74.1 k g / m m 2
cracking in the grout hole welding, since the welding is Eln 22%
carried out under high restraint. Impact (CVN) 7.9 kg- m (-15°C)
4) Diffusible hydrogen test
This checks the amount of hydrogen in the deposited In view of the extremely difficult working condition
metal of the electrode. in the tunnel where the humidity is more than 90%, the
Type of weld Type of welding Location Root pass Fillet pass

1) Long, joint SAW Shop E10016* Nittetsu Y-80M wire
(Nittetsu L-74) and Nittetsu NF16
Flux
2) Cir. joint SAW Shop £10016* Nittetsu Y-80M wire
(Nittetsu L-74) and Nittetsu NF16
Flux
MMAW Shop E7016** E10016
(Nittetsu 16LH) (Nittetsu L-74)
3) Fillet weld MMAW Shop E7016/E7018 E7016/7018
Welten-80+Welten-80 (Nittetsu 16LH)
m.s.+Welten-80
4) Cir. joint Welten-80 MMAW Site E7016 E10016
(Nittetsu 16LH) (Nittetsu L74
5) Cir. joint m.s. + Welten -80 MMAW Site E7016 E7016
* Remove the root pass by gouging.

** Strength should be secured by gouging.
steel manufacturers r e c o m m e n d e d the following f) Before using the flux recovered from previous jobs,
manual electrodes and SA wire-flux combination to get care should be taken to see that the impurities are removed.
the best results (given in the Table below): Conditions of preheating and interpass temperature
The following conditions for drying of welding were stipulated as follows :
consumables were stipulated: A preheat and interpass temperature of 150 to 200°C
The manual electrodes after being taken out of the shall be maintained. The preheat shall be uniform and
container shall be dried at 400 ±15°C for about three continuous. Electric stripheaters or gas burners may be
hours before use. The electrodes thus dried should be used for preheating the longitudinal and circular joints
transferred on to storage ovens immediately, and in the shop. It is necessary to heat each part over a width
maintained at 120 to 175°C. Only the electrodes required of 100 mm (4 x t) or more and maintain satisfactory
for an hour's use shall be removed at any time. If an temperature control. It should be kept in mind that the
electrode is exposed to the atmosphere for more than surfaces to be welded must be absolutely moisture-free.
an hour, it shall be redried under the same conditions When the preheat temperature is too high, or welding
as before. The redr-ying can be done only once. If the is locally concentrated, the overall temperature of the
drying temperature is too high or the electrode is dried joint may become too high and adversely affect the
more than four times, the coating will embrittle and peel toughness of the weld-metal or the heat-affected zone.
off during welding. Therefore, the interpass temperature should be checked
Fluxes for SAW shall be d r y and free from and controlled within the limit. When thermostats are
contamination of dirt, mill scale or other foreign not available, freq-uent measurements of temperatures
material. In principle, the flux should be fused type. by indicators should be made. The measurements
a) As a rule, neutral-type flux should be dried at a should be taken as close to the welding line as possible.
temperature of 250°C for two hours and highly basic Local preheating may be carried out while repairing the
type of flux at a temperature of 400°C for two hours. w e l d i n g defects of m a i n joints revealed in the
For uniform drying, the flux should be spread in a thin radiographic tests. The recom-mended heat input is less
layer and dried in an agitated oven. than 45,000 J/cm in Welten-80 steel. Hence, input should
b) After drying, the flux should be placed in an oven not exceed 50,000 J/cm.
at 150°C. In order to prevent embrittlement and softening of
c) Flux should be used within 60 min of removal from the fusion line and heat-effected zone the heat input by
the oven. welding must be restricted as described below:
d) If the flux has been exposed to the atmosphere for Classification of plate thickness (mm)
more than 60 min after removal from the oven, it must 6 to 13 13 to 19 19 to 26 26 to 40
be dried again before use. 25,000 J/cm 35,000 J/cm 45,000 J/cm 48,000 J/cm
e) If the flux has been kept in an open container or max. max. max. max.
hopper for more than four hours, the upper 25 mm Welding voltages commonly range from 24 to 26 V,
should be discarded. when welding is done with proper arc length. The
Tabulated below are typical examples of MMA welding conditions:

Welding Electrode Welding Welding Heat input
position dia. mm current, amp speed, m m / m i n J/cm
Flat 3.2 100 to 120 100 to 120 12,000 to 18,000
4.0 160 to 180 100 to 180 13,300 to 27,000
5.0 200 to 230 100 to 180 17,000 to 34,500
Vertical 3.2 90 to 110 40 to 120 22,500 to 37,000
4.0 130 to 150 40 to 120 30,000 to 45,000
Horizontal 3.2 110 to 130 100 to 180 12,000 to 15,000
4.0 160 to 180 100 to 180 13,500 to 17,000
5.0 200 to 230 100 to 180 17,500 to 34,500
Overhead 3.2 90 to 110 40 to 140 12,000 to 18,700
4.0 120 to 140 40 to 140 12,500 to 17,500
welding current is DC, straight polarity. The actual The tack welding of jigs is carried out under the same
welding heat input (J / cm), can be calculated from the conditions as for regular welding (welding of the main
formula: body). The bead length is kept 80 mm minimum.
Current x Voltage In periods of high wind, the operator and joint are
Heat input (J/cm) protected to prevent the protective gas blanket of the
100 x Speed in c m / m i n welding arc from being blown away. No welding is perfo-
rmed when water is falling/dropping on the surface to
In the case of SAW also, this formula holds good. be welded, unless the work is properly protected.
Typical examples of heat input are as under:
Extreme care is taken to remove slag during each
Thickness Preheat and interpass Heat input pass and inspect the surface with the naked eye. The
mm temp. 0°C J/cm next pass is never deposited without looking for defects
such as overlap, pits, excessive undercut, etc. Any
25 150 to 200 44,200 defects that appear on the surface are removed by
38 150 to 200 45,000 g r i n d i n g or g o u g i n g before d e p o s i t i n g the next
50 150 to 200 45,000 successive bead. No arc strike is allowed on the base
In the fabrication and installation of penstocks, metal, except in the groove. If this happens, that part of
welding is performed at three different locations, the base metal will be hardened abnormally and may
namely, the fabrication shop, temporary site workshop become the starting point of crack.
and the installation site. Generally MMA and SA The welds are ground till all weld ripples and surface
welding are carried out at the shop and only MMA at irregularities are removed and the weld surface is left
the erection site. The longitudinal joints are welded smooth, but not necessarily flush with plate surface. All
either by SAW or by MMA. For circumferential welding, double welded butt joints are examined in accordance
the longitudinally welded shells are matched with with the requirements of paragraph UW51 of Sec. VIII-
sufficient finger bars to maintain alignment during 1971, ASME Unfired Pressure Vessel Code. Longitudinal
welding. SAW is permitted only if the thickness is more joints and the circumferential joints are X-rayed 100%
than 19.2 mm. If it is less than 19.2 mm, only MMA of the length of welds in the shop. Radiographs are not
process should be adopted. taken until 24 hours have elapsed after completion of
In SAW, penetration is relatively deeper and the welding. Defects disclosed by the radiographs are arc-
chemical composition of the deposited metal varies air gouged or chipped out, rewelded, re-X-rayed and
widely with dilution of the base metal. For this reason, the cycle repeated until satisfactory results are obtainejl.
the mechanical properties and especially the crack All fillet welds, joining diaphragm plates, rings, lugs,
sensitivity of the deposited metal are easily affected by etc., must have smooth transitions into the sides of plates
the cooling rate of the deposit. Therefore, in the selection with toes of the weld made before depositing the major
of the wire-flux combination, sufficient consideration bead. The surfaces are ground to merge smoothly into
is given to the type of joint, build up, method of preheat, the plate surfaces. The fillet joints are subjected to
interpass temperature, heat input, etc. magnetic particle testing. The procedure of carrying out
In manual welding, the arc length is kept as short as the above test is as per Appendix 6 of ASME Code Sec.
possible, consistent with maintaining satisfactory VIII, 1971.
operation. The straightforward stringer bead method Grout holes are provided in the pipes for grouting
is preferred for welding of Welten-80 steel. Weaving is purposes in the field after they are installed in the tunnel.
permissible if a steady arc length is maintained, but the Welding of grout plugs in the field is difficult in the case
w i d t h of the weave should not exceed thrice the of Welten-80 steel due to the difficult conditions and
diameter of the elect-rode used in the vertical position, high percentage of humidity. Hence a mild steel nozzle
and twice the diameter of the electrode used in the flat with reinforcement plate is welded in the shop as shown
position (diameter of core wire). When welding is in Fig. 19.29. All precautions are taken while welding to
performed in the vertical position, all beads progress prevent any underbead cracks. The grout holes are
from the bottom to top, except the first pass, which may welded with E7018 electrodes both at the shop and in
be uphill or downhill, and the finish or wash beads are the field for plug welding. As there is no scope to carry
r u n top to bottom using 3.2 or 4.0 m m diameter out radiographic tests, the grout holes are plugged
electrodes and either two or three stringer beads to cover temporarily and hydraulic testing is carried out to make
the uphill passes. sure that there is no leakage. The plugs are of mild steel
The t u n n e l is kept well ventilated, such that

temperature and humidity are kept under 30°C and 80%
I SQ. OR. DIA. 100 mm
1 1/2" STD. Pipe Threads
respectively. If the humidity exceeds 90% in the tunnel,
^ B a c k i n g plate the welding work is suspended.
M.S.
The field circumferential joints must be radiographed
Welten 80 100% for Welten-80 steel, but in the Kalinadi project,
pipe
due to the lack of space around the pipe at site,
radiographic tests could not be carried out. Ultrasonic
Low carbon test was the only alternative test that could be used. This
Shop weld (0.18-0.20%)
Seamless tube-M.S.
testing was carried out 24 hours after the welding was
Field weld
65 mm 1.0 x 10 mm completed. It was based on the ASME Code. The
(Use E 7018 electrodes for all welds) sensitivity was kept higher than what is specified in the
code, so that minute defects would not be missed.
Fig. 19.29: Detail of grout hole plug ASME Code specifies 2.25 MHz probes as a general
rule for testing. Rut for testing Welten-80 joints, 70° angle
probes of 5 MHz are recommended for plates of
and the same are later welded in the field after the thickness up to 40 mm. The execution and interpretation
grouting operation is completed. of ultrasonic testing requires specialised experience to
The shop assembled pieces are aligned at the site correctly judge the defects. A set of test pieces with
along the alignment. Sufficient finger bars are used on defects are prepared and radiographed. Ultrasonic tests
circumferential joints to maintain alignment during are thereafter carried out on these pieces. The results
welding. No tack welds in the joints are permitted. As are then tabulated for reference. The inspectors are
there is no access to weld the joint from outside, only trained with the help of these test pieces to acquire
inside welding is carried out with back-up bar, which is correct j u d g e m e n t . Wherever space permits,
made of mild steel having approximately 0.18% carbon radiographs are also taken and the results compared
content. Preheating and interpass temperatures are the with the ultrasonic method.
same as followed in the shop.
In the case of embedded type penstocks where the Nuclear power
welding work may be carried out in highly humid In India, as in many developed and developing
tunnels, delayed hydrogen-induced cracking of weld is countries, nuclear power is being increasingly used for
likely to occur. To prevent such weld cracks, low-carbon the generation of electricity. C o m p a r e d with
equivalent steel materials are used and preheating conventional plant, the capital cost of a nuclear plant is
temperatures, heat input and electrodes are stringently much higher, but the running costs are much lower.
controlled. The first pass is deposited with a lower The heart of a nuclear plant is the nuclear reactor. In
tensile electrode to prevent cracking. designing a reactor, the important problems to be
Heavy joints requiring multi-pass welding have a considered are: efficient extraction of heat from the core
high possibility of cracks occurring in the root pass, for conversion into electricity; containment of the fissile
which can be prevented by the application of continuous material and the resulting fission products, and the
welding. This kind of cracking is likely to occur when provision of a moderator. The fuel is usually placed in
temperature is u n d e r 150°C and w h e n h y d r o g e n sealed metal cans which prevent the release of the
contained in the weld zone is high. In continuous radioactive fission products. The cans are usually finned
welding, however, it will take a much longer time before to aid heat transfer to the coolant.
the temperature comes down to 150°C or lower. During A nuclear power plant is composed of extremely
this period, hydrogen levels are lowered and the complicated and sophisticated components. It incorpo-
possibility of crack occurrence is thus eliminated. rates several thousand welds to provide pressure-tight
In fact, welding should be done continuously as far containment of the liquids in the reactor vessel and
as practicable. As a general rule, the working condition steam raising circuits. In addition, some reactor designs
should be so set, that within the groove, continuous rely heavily on welded structural steelwork to support
welding can be done from the first pass to one-half of all the ancillary piping and equipment.
the plate thickness. In any case, two passes must be In India, important nuclear plants are located at
continuously welded without fail. Kotah, Chennai and Narora, and are known respectively
as Rajasthan Atomic Power Plant (RAPP), Madras Briefly, these diaphragm plates are subjected to large
Atomic Power Plant (MAPP) and Narora Atomic Power in-service stresses. The weld joint between the carbon
Plant (NAPP). All the three are heavy water cooled and steel and stalinless steel was critically reviewed to
moderated plants. determine delta-ferrite content at various locations, i.e.
In the case of RAPP/MAPP design, the calandria and carbon steel/weld boundary, centre and stainless steel/
the two end-shields, one at each end, are separate weld boundary. It is well known that ER 309 electrodes
components, each of which is suspended from the roof are generally used for this conventional transition weld.
by support rods. The design aims at optimum alignment However, studies revealed that at the boundary of
of the calandria a n d end-shield at the operating stainless steel and weld material, the ferrite content is
conditions. lower than the minimum specified value of 5% by ASME
In the NAPP design shown in Fig. 19.30, the calandria standards. To offset this, it was established that by giving
and the end-shields have been made integral by field an overlay with ER 308L on stainless steel edge, the
welding. The lattice tubes, 306 in number, which support problem of low ferrite content can be overcome. Hence,
the end-fittings of the coolant channel, pass through the the risk of undetected microfissures is taken care of.
end-shields and are welded at each end. End-fitting seal Details A and B in the figure give the ferrite content
bellows are provided at each end of the coolant channel with conventional method of welding carbon steel/
and are required to contain and seal the protective stainless steel and the modified welding technique
atmosphere in the annulus between the coolant tubes/ developed for the project, respectively.
end-fittings assembly and calandria tube/lattice tubes
assembly. At the calandria end, the end-shields have a c) Welding of bellows
second tube sheet (baffle plate), which is situated 25 mm End-fitting seal-bellows are provided at either end
behind the calandria side-tube sheet of the end-shield. of the coolant channel: one end of the bellow assembly
The lattice tubes are arranged on a 228.60 mm square would be welded to a collar, shrink-fitted on the end-
lattice pitch and welded to the fuelling machine side fitting and the other end to be rolled into the lattice tubes.
tube sheet at one end and to the baffle plate and the These bellows are required so as to contain and seal the
calandria side-tube sheet (tri-junction weld joint) at the protective atmosphere around each coolant channel
other end. Two support diaphragms, provided in each assembly.
end-shield, are d e s i g n e d to a c c o m m o d a t e the In bellow welding, approach and space restrictions
differential thermal expansion between the stainless were the main limitations for obtaining sound welds.
steel and carbon steel portions of the end-shield and Since manual welding is ruled out, a special welding
between the Zircaloy-2 calandria tubes and the concrete jig and fixture were designed and developed to make
walls in which the end-shields are housed. the required weld of consistent quality.
Some important and challenging welding jobs To make the weld, the motorised welding gun of the
developed for the NAPP project are described below. automatic TIG welding equipment is mounted on the
end-fitting and is firmly held on it. The automatic
a) Site welding between calandria and two end-shields sequence now takes over and the arc is struck between
For integrating the main reactor vessel (calandria) the welding torch and the job at the partition line
and the adjoining shielding components (end-shields) between the shrink-fitted ring and the bellow ring. The
by field welding, a special welding procedure involving motor of the welding gun drives it around the end-fitting
the MMA process and a well-planned block welding precisely at the required uniform speed. Initially, a
sequence was developed to control distortion. Four pulsed TIG power source was considered in order to
welders, each one working diametrically opposite his obtain consistent weld quality and greater depth-to-
partner, welded together on each seam. width ratio. However, by making suitable modifications
in the weld geometry, the desired results were achieved
b) Welding carbon steel to stainless steel with the conventional TIG power source. The welding
The end-shield diaphragm plates consist of two arrangement details are shown in Fig. 19.32.
parts, one of stainless steel (Type 304L) and the other of
carbon steel (ASTM A516 Gr. 60) as shown in Fig. 19.31. d) Tri-junction welding
These p a r t s s u p p o r t the end-shield-calandria The tri-junction joint comprises lattice tube, calandria
assembly, and in addition to supporting the dead load, side-tube sheet (CSTS) and baffle plate (BP) of each end-
are subjected to both lateral and inplane loadings. shield. The schematic diagram of the tri-junction weld
Outer
Cover supporU
lid A Solid shut off rod assy. 14

o Regulating rod (SHRC) 18
• Regulating rod (fine) 2
-► Booster rod
4- Liquid shut off rod assy. 12
+ Flux monitor 1
O.F. - Overflow pipe 1
OPR - Overpressure tube end shieldL
relief line 6 Balls HJ) Calandria
I-Moderator inlet 12 .2 tube
OT - Moderator outlot 4 Enlarged Detail , C I
Top hatch—î
V
Journal ring and
See bearing sleeve Lattice
HjtO vault)
enlarged (Calandria ^r-ri tube
Detail ' C coolant shield
side) mil C a l a n d r i a * ^
End fittim
tearSh-^^ Pressure relief
End _sh[eld body
line
End shield"
Lattice tube
Calandria-
Ion chamber Moderator -Calandria-

inlet side tube sheet
♦ Fuel Moderator inlet.- Baffle plate
channel
(306) Vault liner- .Moderator Enlarged Detaij ,A,
Carbon steel outlet
End shield
Enlarged Detail 'B'
alternative to Detail 'A' m
>
♦o
O
5
5
z
CO
Fig. 19.30: General arrangement of reactor components at NAPP

Octagonal flange
C.S.
Key diagram
Electrode 309
Section C-C
( Tube sheet
F / M side 5.S.
Overlay with
30 81 electrode
S.S,
Detail Detail B
Fig. 19.31: Welding of end-shield diaphragm plates
is shown in Fig. 19.33. The lattice tubes are seamless to be kept to a m i n i m u m , in order to maintain
austenitic stainless material of extra-low carbon type dimensional stability during final boring operations; and
ASTM A312 Gr. 304L and tube sheet and baffle plate a high rate of production is required since the number
are made out of ASTM 240 type 304L stainless steel. The of weld joints is quite large.
finished w e l d s h a v e to meet ASME Class 1 For depositing the root run, a unique arrangement
requirements. The welding has to be done in horizontal of placing consumable welding wire was successfully
(2G) position and the welding is almost a blind operation developed for getting full root fusion with a specially
in the sense that operator's view is blocked during designed auto-TIG torch. As seen in the figure, in this
welding. Welding distortion and locked-in stresses have case the insert is placed on the partition line between
945.5 225.5
Proposed welding head
End fitting-expansion
bellows assy.
Backing ring
Tig fusion site Bellow assembly
weld end piece
Shrunk fit ring
End fitting
Detail »X«
Dimensions are in mm.
Part end view of end fittings at

reactor lattice positions
Fig. 19.32: Details of bellow welding for NAPP reactor
Octagonal frame Outer shell j Calandria side

Manholi u . . , , , Tube sheet
Hand hole
Main shell
Baffle -7
plate /
Lattice tube
Detail 'A'
Key Diagram
Fig. 19.33: Joint configuration for welding CSTS baffle plate lattice tube
the calandria side-tube sheet (CSTS) and the baffle plate.

The necessity of preplacing the insert arises mainly
from introducing the required amount of ferrite in the
root of the weld, because fully austenitic welds are
likely to develop microfissures. Secondly, the insert so
placed also helps in achieving welds of consistent good
quality during the production welding. The subsequent
filling passes were deposited with the pulsed-MIG
process. The main problem faced during the filling
operation was the sticking of the feed wire. It was
observed that the wire was getting stuck due to arc Typ. weld detail for Typ. weld detail for
jumping from workpiece to the contact tip. This problem joining bulb and 'S* joining bulb to 'S' bend
was overcome by providing a ceramic sleeve on t h e bend to side plate
c o n t a c t tip a n d by spraying a silicone-base anti- Fig. 19.34: Weld details for dump port box assembly
s p a t t e r c o m p o u n d on t h e contact tip before the
deposition of each pass.
puddle difficult, besides making the operation very
s t r e n u o u s . Moreover, d u e to the large thickness
e) Welding of calandria (Reactor Vessel) difference, it is very difficult to achieve good root
In the fabrication of calandria for RAPP and MAPP, p e n e t r a t i o n on this weld. Good penetration w a s
which was made of Type 304L stainless steel, the overall achieved by developing a two-welder technique for TIG
dimensions o f 7 m x 6 m x 5 m had to be very precisely welding of the joint. Dual physical functions of a welder
controlled, bearing in mind that an austenitic stainless in TIG welding, viz. directing the arc and feeding the
steel is more distortion prone than a carbon steel due to filler wire, normally done by him with two hands, are
its higher coefficient of thermal expansion and lower separately done in this modified technique. One welder
thermal conductivity. TIG welding process with its holds the torch and guides the arc along the joint at a
following variations was found to be most suitable: proper angle and speed. The other welder, sitting on
i) Manual TIG with filler wire the opposite side, adds the filler wire to the weld puddle
ii) Manual TIG with consumable inserts and with- at a proper rate and also directs the torch bearer.
out filler wire The joint design for S-bend to side plate welding is
iii) Automatic autogenous TIG (i.e. without filler the same as that for bulb to side plate welding. However,
wire or insert). the approach for welding S-bends is very restricted. It
One of the most critical areas of welding in the cannot be welded by one hand, and the welder has to
calandria is the d u m p port box assembly. The d u m p change his position continuously while progressing
port box consists of three sub-assemblies, each along the weld (see Fig. 19.35). The welders were trained
incorporating 28 bulbs, 14 S-bends and two side plates.
The bulbs and S-bends are 10 mm thick and side plates
are 31 and 44 m m thick. Figure 19.34 shows the
configuration of welds joining bulb to side plate, bulb
to S-bend and S-bend to side plate. It is absolutely
necessary that the welds be free from defects and have
a smooth appearance without undercuts or excessive
reinforcement. All these joints are single-sided with no
access from the back side, and are done by manual TIG Right hand
process. The welders must have enough skill and for torch
familiarity with the joints to do a faultless job every time. Left hand Left hand for torch
It should be noted that accessibility for welding the for filler Right hand for filler
joints between the bulbs and side plate over its entire
length is very limited, and that compared to the
thickness of the side plate, the bulb is very thin. Limited Fig. 19.35: Welding of'S' bend
accessibility makes a welder's judgement of the weld
to control the welding torch and its movement along to carry heavy water from the main primary system
the joint with either right or left hand, while feeding feeders to the delayed neutron monitoring room. The
the filler wire with the other hand. welds in the tubing system must be of highest integrity.
Using the TIG process with consumable inserts, high Any failure in the system will result in the loss of heavy
integrity welds with full penetration were achieved in water, which cannot be tolerated from the economic and
several hundred single-sided pipe joints varying in size safety points of the operation of a nuclear power station.
from 12.7 mm NB to 100 mm NB in each calandria. In addition to the financial loss incurred by the heavy
As an alternative to the above method, an automatic water leakage, it creates a radioactive atmosphere inside
orbital tube welder was used, as shown in Fig. 19.36. the reactor building, which will further hamper any
Using various welding heads, this system is capable of rectification work. The material selected for this system
welding the diameter range of 3-550 mm, which is the is extra-low carbon austenitic stainless steel to overcome
full range of diameters encountered in calandria piping. the weld decay problems during welding.
The system consists of: An automatic tube welding equipment using the TIG
i) Orbital welding heads process was selected for this critical application. It
mainly consists of the weld head, the pendant box, the
ii) A crater filling unit
power supply and control and shielding and purging
iii) Preheating time relay gas systems. The weld head is made of two parts. The
iv) Current control units for fine current selection as head proper is interchangeable for different tube sizes
well as for low currents, down to 5 amps. and it is provided with clamps for holding the tubes.
Type 304L bulbs for MAPP calandria were welded The head is covered from atmospheric air entering the
by the submerged-arc process using ER 308L wire and weld chamber and can be opened by a cover. The head,
a specially designed agglomerated flux. The weld sketched in Fig. 19.37, contains the electrode assembly
deposit contained less t h a n 0.04% carbon a n d in the form of a slotted gear and the electrical energy is
approximately 7% delta ferrite as stipulated. Welding supplied through the slip rings. Proper diffusers are
wires of 3.15 and 4 mm diameter were used in the provided to avoid the argon gas entering the weld
current range of 400-500 amp. The welded joints met chamber directly, which otherwise will result in arc
the radiographic requirements. shifts. The head also contains a reduction gear unit to
control the output speed of the motor in the handle. The
f) Welding of small diameter tubes pendant box contains head control switches and an
At MAPP, the S.S. tubes used in the instrumentation ammeter to give the operator a visual check on the weld
system are less than 12.7 mm O.D., and they are designed programme. Fullwave silicon-controlled diode (SCR)
rectification power supply is employed for this machine.
The programme is obtained by changing automatically
the reference voltage by means of relay contacts and
preset potentiometers used as voltage dividers. Suitable
timers are employed to control the programme. The
complete programme consists of preshielding gas flow,
preheat, weld, Decay I, Decay II, taper current and
postshielding gas flow. The preheat is meant to heat the
metal to fusion temperature in the initial stage, the
absence of which otherwise will result in lack of penet-
ration. A high frequency circuit is included for arc initiat-
ion. The whole power supply can be switched off or on
from either of main unit or pendant box. The schematic
circuit of shielding gas and purging gas systems is
illustrated in Fig. 19.38. A recorder is also provided to
record the weld current versus time of each weld.
In this process, the success of achieving a good weld-
bead contour depends entirely on the internal purge gas
pressure, which is maintained manually. The internal
Fig. 19.36: Automatic orbital tube welder
pres-sure is adjusted by introducing a manometer and
Rotor
Drive gear Oiffuser
chamber
Tube clamped Electrode

hen
Type II Tube clamped here
Fig. 19.37: A simplified sketch of the inside of weld head
a tee connection in between the argon cylinder and tube time is over, the high frequency unit starts up, ionising
as shown in Fig. 19.38. The required internal pressure is the shielding gas and establishing an arc. After the
adjus-ted by throttling the pressure control valve, till initiation of the arc, the electrode will stay stationary
the required height of water column is obtained in the until the preheat time elapses. Then the electrode starts
manometer (U tube) for a given value of flow rate. Then rotating circumferentially with the tip of the electrode
the manometer and the tee connections are removed making a concentric circle over the joint with an arc gap
and the argon gas cylinder is directly connected to the of 1.5 ± 0.125 mm, making a full penetration weld. Then
tube. This procedure has to be followed for every five the welding continues with two-step reduction of
or six welds to check the internal pressure and to make current called Decay I and Decay II. By this time the
correction if required. electrode would have completed one revolution and
First the inside of the tubes are purged with argon continued to rotate approximately 160° more with
gas and the required pressure and flow rate are tapering current, to complete the weld programme as
maintained. On pressing the start switch, the shown in Fig. 19.39. The arc extinguishes automatically
preshielding gas starts flowing into the weld chamber at the end of taper current but the electrode will continue
to provide an inert atmosphere. After the pre-gas-flow to rotate and it will stop at the original starting position.
Pendant box
Recorder
zO
20 l/min
Purge gas weld head —^
pressure control i ^
velve r ^ Tube
51/min
$ 4 * Regulator-***
f*—Valve—«X
U-Tube
Purge gas Shielding gas

cylinder (Argon) cylinder (Argon)
Fig. 19.38: Schematic diagram of shielding and purging gas system

The shielding gas flow is also maintained for eight to subjected to tensile tests, 15 were bend tested and the
ten seconds after the arc extinction, to protect the hot remaining cut opened, visually inspected and subjected
weld from oxidation. Arrangements for no-weld to liquid dye penetrant examination. It was stipulated
sequence operation and circumferential inching of that the weld shall not have any root concavity, lack of
electrode are also provided. A graphical representation fusion, excess inside reinforcement and burn-through.
of the weld sequence is shown in Fig. 19.40. Even a single specimen failure would lead to the
To qualify the procedure, the machine had to produce requalification of the automatic welding procedure.
50 continuous welds of acceptable quality in each of 5G The initial problems of arc striking and arc wander
and 2G positions. Out of the 50 welds, six welds were were overcome by using 2% thoriated-tungsten
520°
Full weld current

Decay I current
Taper current
Decay II current
Full weld current

270° Decay I current 360
360° Decay II current
520° Taper current
Fig. 19.39: Change of welding current with reference to welding sequence
TIME
Taper adjusts slope

a
E
<
c
£
3
U
"3
Push
sequence m 1 Rotation
button HF initiates arc Electrode Arc is Rotation
starts rotation completed extinguished stops
Fig. 19.40: Welding sequence

electrodes of proper size and controlling the tip size and

the taper of these electrodes. Tolerances permitted for Hot sodium
weld acceptance are detailed in the following Table: Concrete shield
Control rods S3Z
The complete welded system was leak tested using
fteactor jacketl Steam
helium as tracer gas and observing the following
generator
conditions: the helium gas pressure to be maintained is
1 bar (15 psig) and the gas inside shall contain at least
50% helium. All joints shall be leak proof and when
tested, the leak rate should not exceed 7.6 x 10 3 m litre/
sec (1 x 10"5 standard cc/sec).
Fast breeder reactor: Many countries have been experi-
menting with fast breeder reactors, because they per-
mit more efficient utilisation of uranium, which in course
of time will either become scarce or rise substantially in Primary p u e | Intermediate
vessel elements heat
price. The reactor consists of a compact core, which con- exchanger 23E
tains the enriched P u 0 2 / U 0 2 fuel in stainless steel cans, Cool Sodium
surrounded by more cans containing U0 2 , which will
Fig. 19.41: Fast breeder reactor (FBR)
be partially converted to P u 0 2 by the breeding process
(see Fig. 19.41). There is no moderator. To remove the
intense heat generated in the core, liquid sodium metal feasibility of the design, performance of material vis-a-
is used as a coolant. This becomes slightly radioactive, vis neutron environment and other aspects so vital for
so instead of circulating directly to the steam genera- development of future reactors in the country. On the
tors, an intermediate sodium circuit is interposed. This p e r f o r m a n c e of this reactor will d e p e n d the
removes the dangerous possibility of radioactive sodium development of further fast reactors, and the decision
coming into contact with superheated steam, with which to switch over to fast breeder technology for the future
it would react violently, should there be a leak in the nuclear power programmes of our country.
steam generator. The fuel elements, primary sodium All over the world, austenitic stainless steel Types
pumps and intermediate heat exchangers are all sited 304 and 316 are used as principal construction materials
inside a stainless steel tank. All external connections are for sodium-cooled reactors. For the fabrication of FBTR
made through the roof, thus avoiding the need for pen- at Kalpakkarn, Type 316 (with 3% Mo) was the predom-
etrations through the tank walls or base. The coolant is inant material. For welding this steel, the important
not pressurised, and the high sodium outlet tempera- stipulations were that the weld-metal must give a
ture (560-600°C) gives a high thermal efficiency. minimum elongation value of 41% in the tensile test and
India is specially interested in fast breeder reactors, its ferrite content must be in the range of 2.5-5.0%. The
because they are meant to use Thorium-232 (converted modified 316 basic type electrodes successfully used on
from U-233) and Plutonium-239 (converted from U-238), this steel had the following composition (%) range:
and minimise our dependence on other countries for C .. 0.055-0.060 Cr .. 18.6-18.8
import of enriched Uranium-235 or natural Uranium. Mn .. 1.25-1.50 Mo .. 2-2.10
India has made a start in this direction by setting up the Si .. 0.50-0.56 S .. 0.010-0.012
Fast Breeder Test Reactor (FBTR) at Kalpakkam in Tamil Ni .. 11.8-12.0 P .. 0.010-0.017
Nadu. Research will be carried out on this reactor to The factors which helped to achieve the stipulations
gain experience in sodium technology, fuel cycle aspects, were: low heat input, rapid cooling of weld deposit with
6 mm Tube 10 mm Tube
Maximum radial misalignment between butting tube ends 0.125 mm 0.25 mm
Electrode tip deviation from the joint along tube axis ± 0.125 m m ± 0.125 m m
Maximum concavity 0.125 mm 0.125 mm
Maximum reinforcement 10% of wall 10% of wall
thickness thickness
Bead width 3.175 ± 0.5 mm 3.30 ± 0.9 mm
dry air, short arc and stringer bead technique, sequential c) Centring fixture and dial gauges were used to
welding, maintaining narrowest groove and minimum check up concentricities.
root gap, thoroughly dried electrodes, avoiding cold d) Welding was carried out in horizontal position
work, and employing trained welders. The welded for initial layers.
components included reactor vessel with double e) Welding effect on shifting of centre was observed
envelope and thermal shields, large and small rotating by means of dial gauges during initial runs and
plugs, vessel and plug supports and liquid metal seals. steps were taken to follow suitable welding se-
The fully welded reactor vessel consisted of five quences.
shells (one conical) and five flanges of diameters varying Another vital component is the steam generator.
from 3 to 0.5 m and of total length of 7 m. The Austenitic stainless steel cannot be used for the
dimensional requirements which were extremely generator, because it is highly susceptible to stress
stringent were as follows: corrosion attack in aqueous media. In the FBTR, 2.25
a) Maximum mismatch of the edges of shells and Cr-1 Mo steel has been used for the generator shell and
flanges allowed is -0.3, meaning that tolerance on the tubing. Steam is generated in the exchanger with water
ovality of the shells at the edges is -0.6 mm. flowing inside the tubes and sodium in the shell. Sodium
b) Maximum eccentricity allowed between centres enters the shell at 510°C and leaves at 300°C, while water
of flanges after assembly and welding of the reactor enters at 200°C and superheated steam leaves at 480°C
vessel is ±0.5 mm on a span of 7 m. and 125 kg/cm 2 pressure. This design called for the
Before venturing on the actual fabrication, mock-up welding of transition Joint between 2XA Cr-1 Mo steel
tests were conducted with stainless steel plates of 10 on the generator side and Type 316 steel on the sodium-
mm thickness. As each of the reactor vessel shells is system side. The joint was successfully executed by first
made of two half shells, mock-up tests were carried out buttering the 2VA Cr-1 Mo steel face with Inconel
to fabricate two cylindrical shells, enabling study on four welding consumable and then making the weld between
longitudinal and one circumferential joint. Based on the the buttered face and Type 316 steel face with Inconel
results of the mock-up tests, welding of shells of reactor welding consumable. The detailed welding and testing
vessel and other components was carried out procedure is described below:
successfully, all the time, however, analysing and
recording the results after each operation. To keep
eccentricity under control, the following measures were
Min. Width 4mm
taken:
a) Each of the rolled shells was completely rectified
to limit ovality.
b) Assembly and tack welding of shells with flanges S.S.316
was carried out in the horizontal position.
Details of tests
S. No. Test/Inspection Buttering Final welding
Procedure qualification Welder qualification Procedure Welder
1 Liquid Penetrant Inspection V V V •
2 Radiographic Inspection S •
3 Mechanical Tests
3.1 Tension Test S
3.2 Bend Test S
3.3 Impact Test S
4 Metallographic Examination
4.1 Macrographic S s S •
4.2 Micrographic S
4.3 Hardness Survey Across Weld S
5 Ultrasonic Examination S
Execution vacuum technique has been applied on stainless steel

(1) Preheat 200-250°C on 2VA Cr-1 Mo during the but- cups in the end plenums of nuclear vessels.
tering with Inconel 82/182 welding consumables Weighing around 8 tons each, and constructed from
(2) Postweld 725°C for 1 hour Type 304L stainless steel plates of over 50 mm thickness,
(3) Machine chamfer and weld to S.S. 316 with the end plenums form a vital part of the nuclear vessel
Inconel 82/182 welding consumables and can be compared to the tube sheets of a heat
exchanger. However, the design and construction is
During development work on this reactor, it was more complex and the requirements much more
observed that while manual welding with 316 Type stringent, because apart from meeting (or bettering) the
electrodes gave satisfactory results, this was not the case strict standards envisaged in ASME Sec. Ill, the fact that
when the conventional 316 Type wires were used with heavy water would be circulating through the plenums
the TIG process. Studies showed that the modified 16- has a further bearing on the quality. For this reason the
8-2 composition was better suited, provided the joints in the plenum have to be absolutely leak-proof.
compositional limits were properly set. The final Even more important was the requirement regarding
specification drawn up is shown below. The composition control of distortion, which has to be held within very
has been controlled in order to avoid the possibility of close limits. The joints between the S.S. cups and the
martensite formation. For example, the ferrite forming plenum are comparable to tube-to-tube sheet joints. The
elements Cr, Mo have been specified at the higher levels, welding job covers over 3 m in diameter. The joint
as also the austenitisers C and Mn. The minimum carbon configuration is shown in Fig. 19.42. Over 300 joints had
level has been specified from strength consideration to be welded in each plenum.
also. The maximum carbon has been kept below 0.1%,
In EB welding, the maximum clearance between the
the maximum value allowed in AWS specification for
mating components should not exceed 100 microns—
16-8-2 electrodes (no equivalent AWS specification
the less the better, preferably 30 to 60 microns, to achieve
presently exists for filler wire); this was done to avoid
weld quality. In this application, the cups were
the risk of sensitisation in the weld and HAZ. A lower
machined to close tolerances and the holes into which
limit of Mn is specified to combat against microcracking.
these cups would be fitting in the plenum were precision
Test results confirmed its suitability.
bored with the aid of a jig. To leave no stone unturned,
C .. 0.05-0.07 Cr . 15.5-17.0 selective mating of the cups and relevant holes was
Si .. 0.25-0.50 Ni . . 8.0-9.0 carried out, so as to obtain the most ideal fit required
Mn .. 1.0-2.0 Mo . 1.5-2.0 for the purpose. The cups and their locations in the holes
S .. 0.02 max. B 30 ppm max. were match-marked to avoid any possibilities of mix-
P .. 0.03 max. up. Further, the standard of cleanliness has to be much
At Bhabha Atomic Research
Electron beam welding: superior to that in conventional types of welding, to
Centre, Bombay, electron beam welding by partial ensure quality in EB welding. Hence, it was ensured
Diameters 38, 44 & 133
Note: All dimensions are in mm.
Fig. 19.42: Joint configuration for electron beam welding on stainless steel grade 304L
that the mating surfaces were thoroughly clean. In order textile chemicals and dyes, pharmaceuticals, paper pulp,
to minimise distortion, welding on the various locations foods and dairy products, in addition to the common
was carried out in a phased, staggered manner in chemicals like acids, alkalis, salts and organic
accordance with a predetermined plan based on careful c o m p o u n d s , all of which are vital to a country's
analysis. This involved shifting of the welding head to industrial development and prosperity.
locations not necessarily adjacent, and turning the job In this section we shall deal in detail with oil
completely for doing the welds on the reverse side at refineries and fertiliser plants, as they are the best
planned intervals, in the interest of balancing distortions examples of the application of welding technology in
and achieving an overall satisfactory result. The the chemical industry
mechanical system for radial motion of the welding It has been said that stainless steels and the
head was used for positioning it on the weld diameter petrochemical industry have grown together, and it is
and for its circular movement. Initial setting was done an accepted fact that many of the chemical processes
by using a suitably machined mandrel (fixed in the would not have been chemically viable without stainless
bottom of the welding head) to fit into the centre of the steel. In some cases, it is the only material that can be
cup. This was followed by fine- or micro-adjustments, used for the chemical in question. In other cases, it is
by means of levelling screws on the four a r m s the most economical material w h e n the costs of
(outriggers) attached to the welding head, till it was installation and maintenance of the plant and the
ensured through the use of a closed circuit TV that the production cost of the product as well as the cost of the
locus of the beam and the joint coincided within about initial equipment are considered.
± 0.1 to 0.2 mm on the joint. The next operation was to Stainless steels used in the chemical industry can be
confirm the correct coincidence of the beam path and b r o a d l y classified as: austenitic stainless steels,
the joint by making four tack welds separated by 90° on martensitic stainless steels and ferritic stainless steels.
the joint while operating the beam on low power. Figure These steels have been described in detail in Chapter 10.
19.43 explains the welding cycle used. Welding plays a pivotal role in the manufacture of
various equipments in the chemical and petrochemical
6) Chemical Plants industry. They can be broadly classified as follows:
Most chemical p l a n t s are subject to v a r i o u s • Pressure vessels, boilers and heat exchangers
combinations of corrosion, temperatures and pressures • Storage tanks
in service. Their efficient operation therefore depends • Industrial and transmission piping.
largely on the correct choice of alloys and welding Stainless steel vessels used in the chemical industry
methods during the design and fabrication stages. are either solid stainless steel vessels, stainless steel lined
Chemical plants produce a wide range of products like vessels, stainless steel clad vessels or vessels with S.S.
petroleum, petrochemicals, fertilisers, industrial gases, weld-metal overlays.
The stainless steel lining is applied after the carbon
or low-alloy steel vessel has been completely welded
T7/pr~TH\
U
B
»U
C
»U
Welding time
•*«
E
*
and stress-relieved, if necessary. Commonly used linings
are 2 to 3 mm thick. The linings are attached to the base
metal by resistance spot or seam welding or using plug
welds by shielded metal-arc welding. The vessel is not
recommended to be subjected to heat treatment after
the lining has been laid.
Symbol Description | The clad plate is supplied by the steel mill. The clad
A Accleratinq voltage (kV) | metals are welded such that the overall corrosion
Beam current delay time 1 resistance of the cladding metal and the mechanical
B with respect to H T kV (sec)
properties of the backing metal are not impaired. The
C Up slope time of beam current (s< backing metal is welded by shielded metal-arc welding
D Beam current (milli amperes) | or submerged-arc welding. The clad side is welded by
[ E Down slope time of current (sec shielded metal-arc welding processes. It has been
considered a sound practice to weld the backing metal
Fig. 19.43: Electron beam welding cycle for stainless first and cladding metal later. The first layer of welding
steel grade 304L
on the stainless steel side of the joint is deposited using
a filler metal rich in nickel, in order to eliminate the applications such as power piping, industrial gas and
dilution problems. air piping, petroleum refinery piping, chemical industry
The dished ends, shell and the nozzles made of process piping, nuclear power piping and many others.
carbon or low-alloy steels are overlaid with stainless The codes are laid d o w n to cover m i n i m u m
steel weld-metal, usually 4 to 6 m m in thickness. requirements of quality and safety. The codes laid down
Submerged-arc welding process using strip electrodes by the following societies /institutions are generally
or single and twin wires with oscillations is very popular followed:
for weld-metal overlays. Gas metal-arc welding process • American National Standards Institute
is also used successfully. Shielded metal-arc welding • ASME Boiler & Pressure Vessel Code
process is employed for weld-metal overlays when the • American Welding Society
work is of a smaller magnitude. Weld-metal overlays • British Standards Institution
have to be machined subsequently. They provide an • Indian Standards Institution
alternate source for integrally clad vessels required for Storage tanks in numerous types and sizes and a
the chemical process industry. variety of materials are required for the chemical and
In the fabrication of heat exchangers, tubes are petrochemical industry. They have been dealt with
welded to tube sheets using MMA, TIG and MIG earlier in this chapter. Storage tanks are designed and
welding processes, and the types of joints and welds fabricated to the requirements of one of the following
shown in Fig. 5.33. Automatic custom-built TIG welding codes and standards:
equipments are becoming increasingly popular for this a) American Petroleum Institute Codes—API-650;
application, because they produce consistent quality API-620
welds at high speeds. A typical one was described in
b) Indian Standards Institution Code—I.S. 803
Chapter 5.
Keeping in view the minimum requirements of safety c) Japan Industrial Standard
of the operating pressure vessels, various codes have Great care should be exercised in selection of the right
been formulated for different applications. For the welding process, fit-up, filler materials and the heat
chemical industry, the following codes and standards treatments wherever required, to minimise stresses and
are popular : distortion due to fabrication.
The vessels, heat exchangers and columns are
a) ASME Section I—for boilers
fabricated in the shops where adequate facilities are
b) Indian Boiler Regulations—for boilers available and the working conditions are conducive for
c) ASME Section VIII, Division I & II—for unfired good fabrication. But, when it comes to site fabrication,
pressure vessels the working conditions are far from ideal. Hence, great
d) Indian Standard 2825—for unfired pressure ves- care has to be exercised. The fabrication of heat resisting
sels steel should be isolated from the fabrication of other
materials to avoid rust formation and carbon inclusion
e) British Standard 1515—for pressure vessels due to mixing up of tools and materials. Stainless steel
f) TEMA Standards—for heat exchangers. chisels and wire brushes are to be used in case of
Stainless and heat resistant steel piping is required stainless and heat resistant steels. In case of pipe
for the chemical process industry in large quantities and welding, proper gas-flow rates should be maintained
various sizes. The piping which includes pipes, tubes, for shielding and purging to avoid oxidation. On
valves, headers and fittings such as elbows, tees, flanges completion of the welding operation, the weld joints
and reducers may be intended to transport highly should be thoroughly cleaned from contaminations,
corrosive fluids and fluids at extremely low or high either by mechanical or by chemical means. A clean
temperatures. polished surface minimises effects of corrosion.
Austenitic stainless steel p i p e s are far more Mechanical cleaning is done by power brushes, whereas
extensively used than the martensitic and ferritic chemical cleaning is done by using suitable pickling
stainless steel pipes. AISI 304, 304L, 316 and 316L are solutions.
very popular austenitic steel materials for piping. Any fabrication, whether it is a vessel or a piping
Since the piping systems are exposed to increasingly system, is complete only when it is pressure tested as
severe o p e r a t i n g c o n d i t i o n s , v a r i o u s codes a n d per the design r e q u i r e m e n t s . The most c o m m o n
s t a n d a r d s have been formulated for different methods of testing are by hydraulic and pneumatic
media. Hydrostatic testing of austenitic stainless steels define certain minimum standards to be observed. Table
should be done only with potable water. After the 19.8 provides a list of some of the more important codes,
hydrostatic testing, all water should be drained and but in the oil refining industry, the most familiar and
cleaned by blowing air. Any stagnant water remaining universally applied ones are: the ASME Code Sec. VIII
is likely to result in corrosion. It will be of interest to covering Unfired Pressure Vessels, ANSI B31.3 covering
cite an incident, where ordinary water was used for Pressure Piping and TEMA standards covering Tubular
hydrostatic test of a p i p i n g system instead of Heat Exchangers. These are commonly employed by
demineralised water recommended. No flushing and refinery owners and contractors on a contract basis and
drying by air was done after the test and it resulted in in some States in the U.S.A., the ASME Code and ANSI
severe corrosion within three months from the date of B31.3 have the force of law.
testing. All the weld joints in 5 G positions had to be cut It is important to remember that the codes do not
out whereas welds in 2 G were intact. It is essential that provide all the data necessary to design and fabricate
chloride content of water is restricted to 3 p p m , the p l a n t s . The designer m u s t decide, based on
wherever hydrostatic testing is done or the hydrostatic additional knowledge and experience, which materials
testing should be substituted by pneumatic test. to use for a specific set of conditions including corrosion.
He m u s t k n o w w h a t steps to take to overcome
Oil refinery metallurgical shortcomings of various alloys during
In the design of petroleum refining process plants, fabrication and service. He must be able to devise
the main considerations are safety, reliability and welding procedures to meet the code's minimum
economy. Welding plays an important role in ensuring standards. He must decide whether the minimum
that these three factors are maintained at a high level. standards for the quality of finished fabrications and
The first step on the road to safe, reliable design and for inspection set by the codes are adequate or whether
fabrication is provided by Codes of Practice, which additional conditions must be stipulated. All such
Table 19.8: Some codes of construction

Country Issuing organisation Source document(s)
U.S.A. American Society of Mechanical Engineers Boiler and Pressure Vessel Code, Section VIII
Great Britain British Standards Institution BS 1515-Fusion Welded Pressure Vessel for
use in the Chemical, Petroleum and Allied
Industries.
BS 5500-Unfired Fusion Welded Pressure
Vessels.
Germany Arbeitsgemeinschaft Druckbehalter A.D. Merkblatter
(published by Carl Heymans Verlag KG)
Italy Associazione Nazionale per il Controllo ANCC Code, Sections VRS, M, S&F
della Combustione
Netherlands Dienst voor Stoomwezen Regels Voor Toestellen (Rules for Pressure
Vessels).
Sweden Tryckkarlskommissionen (Swedish Swedish Pressure Vessel Code.
Pressure Vessel Commission)
Piping
U.S.A. American Society of Mechanical Engineers ANSI B31.3 Code for Pressure Piping.
American National Standards Institute
Great Britain British Standards Institution BS 3351 Piping Systems for Petroleum
Refineries and Petroleum Plants.
Tanks
U.S.A. American Petroleum Institute API Std. 620-Recommended Rules for Design
and Construction of Large Welded Low
Pressure Storage Tanks.
API 650-Welded Steel Tanks for Oil Storage.
Great Britain British Standards Institution BS 2654-Vertical Steel Welded Storage Tanks
for the Petroleum Industry.
important details left out or not clearly defined by the b) Chemical conversion processes — in which the size
codes must be covered by the Fabrication Specification, and structure of the hydrocarbon molecules are
which should be drawn up by the designer in changed to convert surplus products into dema-
consultation with the plant purchaser. The specification nded ones. The main processes in this group are:
should highlight all problem areas likely to occur in • Break-down processes such as cracking.
construction. It must impose restrictions on un- • Build-up processes such as polymerisation and
acceptable practices, and while leaving the door open alkylation.
to innovat-ion, must control it. The specification must
be careful to leave responsibility for fabrication with the • Type change processes such as dehydrogena-
fabricator and must allow him room to manoeuvre. This tion, isomerisation and reforming.
document, being a vital vehicle of communication, must c) Treating processes—in which the products of
give clear unambiguous instructions to the material separation and chemical conversion are purified
buyer, fabrication shop manager, welding engineer and to meet marketing specifications.
the inspector. d) Blending—in which products of the above pro-
The modern refinery is an extremely complex plant, cesses are combined, sometimes with additives
but its principal aspects may be described under the to produce optimum market properties.
following four headings : The refinery is made up of a variety of units to fulfil
a) Physical separation processes—such as distillation,the above-mentioned functions. The units are correctly
in which separation of hydrocarbon fractions is chosen and balanced in such a way that their
achieved without change in molecular structure. combination has the versatility to meet fluctuations in
5utphuc (11)
Gas recovery I > Propane (1)
J*l Butane (1)
Off gas
* » • fuel gas (12)
Sulphur -*► Light naphtha (2)
recovery Heavy
Naphtha naphtha
|Naphthas Reforming - * * Reformate (2) (3)
IHydrotreatingJ splitting
Crude
distillation kerosene
JHydfQtreating Kerosene (4) (5)
Gas oil Hydro- -Gas oil (6) (7) (8)
desulphuriser
--*» Propane (1)
Hydrogen sulphide
LonQ residue Gas recovery Alkylation Alkylate (2) (3)
Catalytic Gasoline Caustic *► Gasoline (2)

Vacuum cracking washing
Vacuum gas oil ■ Cycle oil (7) (8)
distillation -- Decanted oil (8)
-■'Naphthas (2) (3)

H HydrocracWing - * Kerosene (4) (5)
Gas oil (6) (7)
lube distilates Extraction

& ►J Hydrotteatinq Lube oil (9)
Short residue dew axing
Bitumen Asphalt (10)

blowing
— ■ Short residue (B)

-*» Long residue (8)
Fig. 19.44: Example of integrated flow scheme for crude oil processing: (1) LPG; (2) Gasoline: (3) Aviation spirit: (4)
Jet fuel: (5) Kerosene; (6) Diesel; (7) Heating oil: (8) Fuel oil: (9) Lubricating oil; (10) Asphalt: (11) Sulphur: (12)
Refinery gas
demand. Figure 19.44 gives an integrated refinery flow In Table 19.9, some of the more important processes
scheme, in which the relationship of the various in current refinery practice are listed. An attempt has
processes and units can be seen. been made in this table to summarise the different
Table 19.9: Oil refinery process units—summary

Process What is does Equipment Materials * Problems
Atmospheric crude Separates crude into its Desalter drum C.S. up to 250°C 1. Above 250°C
distillation component fractions feed/effluent sulphur attack
such as liquefied exchange train (steel oxidized to
petroleum gas Heater, fractionator 5Cr-V£Mo or above iron sulphide scale)
(LPG), naphtha, light overhead coolers. 250°C, C.S. 2. Uptol20°C
gas oil (LGO), + 13Cr clad acid (HC1) condensate
heavy gas oil correction in
(HGO), bitumen overhead condensers
LPG is a finished
product e.g. butane,
propane. Also some
gasoline produced
Vacuum distillation Takes the bottom As atmospheric unit, 5Cr-44Mo or above Ditto, only corrosion-
residue from the except no desalter 250°C, C.S.+ less severe because
atmospheric distil- 13Cr clad ^ S and HC1
lation tower and removed in atmospheric
separates it into distillation.
further fractions
Hydrotreater Removes by reaction Feed/effluent Shell Cr-Mo, tubes 1. Hydrogen embrittle-
with hydrogen those exchangers type ment if incorrect
parts of heavy Heater Type 321 s.s. tubes alloy used. Hence
gasoline which poison Reactor Type 347 s.s. clad C-^Mo, Cr-Mo
the catalyst in C-4£Mo or clad per Nelson Curve
reformers VACT-VMO 2. H2S attack
above 250°C
Catalytic reformer Improves the octane Feed/effluent l^Cr-V^Mo (possi- 1. Hydrogen embrittle-
of motor gasoline exchangers bly clad) ment if incorrect
produced by the 3 or 4 heaters Tubes 2V4Cr-lMo alloy used, Hence
hydrotreater by 3 or 4 reactors l^Cr-V^Mo C-ViMo, Cr-Mo
reacting the per Nelson Curve
treated gasoline 2. H2S attack
with hydrogen in above 250°C
the presence of a
catalyst, e.g. platinum
Hydrodesulphuriser Removes sulphur from Feed/effluent C-ViMo and Type Ditto, except that high
gas oil by reacting exchangers 347 s.s. temperature
with hydrogen at high Type 347 s.s. clad H2S corrosion
temperature VACr-ÂMo is more severe
Heater Type 321 or 347
s.s. tubes
Reactor C-'AMo (or iy4Cr)
Clad with Type
347 s.s. weld
overlay
Fluid catalytic Converts lubricating Heater 5Cr-M>Mo 1. Main problem is
cracking (FCCU) oil from vacuum Reactor lV4Cr-4*Mo+13Cr erosion by catalyst

distillation unit to cladding 2. External oxidation
gasoline by cracking Regenerator Shell C.S. + refractory because of high
large hydrocarbon lining. Air ring temperature
molecules to lighter 5Cr-44Mo 3. Some H2S corrosion
ones in the presence Fractionator C.S. or IViCr but almost zero
of a catalyst with 13Cr cladding. hydrogen
CO boilers VACr-ViMo
refractory lined.
Hydrocracker Does same as cat High pressure Stainless clad 1. Hydrogen embrittle-
cracker but it can feed/effluent 2 1 / 4 Cr-lMo ment because
produce gasoline exchangers of high pressure /
from heavier stocks Heater Type 321 s.s tubes temperature
such as bitumen Reactor Stainless clad 2. H2S attack
2V4Cr-l Mo above 250°C
Effluent cooler C-V2M0 and C.S.
Separator ASTM A516
(Killed C.S.)
*s.s.—stainless steel: C.S.—carbon steel.
functions of each process, the c o m b i n a t i o n s of In the crude distillation process, the hot product from
equipment involved and the primary process problems each of the strippers is used, by heat exchangers, to
to be met by the materials of construction. preheat the crude oil on its way to the heater. In the
A m o n g the processes mentioned, the p r i m a r y main distillation column, separation results in the light
proces-ses of crude distillation and vacuum distillation ends being taken off at the top for further processing in
are impor-tant, as also the three basic items of equipment the LPG unit, while the heavy residue goes for further
involved: separation in the vacuum unit. Intermediate fractions
a) Heaters to raise the temperature of the feedstock are taken off to the three stripper columns and the light
to a vapour phase. ends from each of these are fed back into the main
column. The heavy bottoms are taken either to storage,
b) Distillation vessels to separate the various frac- or in the case of the heavy gas oil stripper, for chemical
tions of the feedstock. conversion in the catalytic cracker.
c) Heat exchangers to control temperature and make Heaters: H e a t e r s consist f u n d a m e n t a l l y of a
most use of the heat generated in the refinery. continuous coil of tubing arranged in the form of a box
Distillation column: This is a vertical cylindrical vessel or vertical cylinder. Oil or gas burners supply heat to
with horizontal trays at fixed elevations throughout the the lower part of the heater, in which the coil and its
tower. Vaporised feedstock is fed into the tower at a contents are heated by radiation. The hot gases from
point close to the bottom and the various fractions of the radiant section pass upwards to a convection section
the feedstock are separated by distillation at each one and there the incoming feedstock receives its primary
of the trays. In Fig. 19.45, the principles of the distillation heating. Spent combustion gases are exhausted to
column are illustrated. Each of the trays shown contains atmosphere via a stack. The heater coil is contained
a liquid fraction at a specific temperature. As vaporised within a steel casing lined with refractory insulation.
feed passes through the liquid, the fraction that liquefies Tube metal temperatures of up to about 500°C may be
at that liquid temperature does so and remains in the reached in the radiant section and up to 300°C in the
tray. The rest of the vapour continues upwards and the convection section. Tube material must be chosen for
process of condensation proceeds. A separate fraction its creep- and oxidation-resistance. Tube supports
condenses at the liquid temperature of each tray. which are not cooled by the process fluid, must meet
The temperature at the bottom of the tower is the consi-derably higher temperatures (the fuel gas
determined by a heater or reboiler. The temperature at (800-1100°C).
the top of the tower is controlled by a condenser which Shell and tube heat exchangers: There are various
may be liquid- or air-cooled. designs of shell and tube heat exchangers but they all
-■ Gasoline
r— Liquid
Reflux
[A--jnuriLjnL
jnLjnLjjiLjnL r-*- First side
stripper return
[jj-—nununL-^
- * First side stripper
jrUTUTiL Tfl
QIjnLjnLJiLjni p**- Second side
stripper return
L^-JnunLii
[iijnLjnLjnLjrt ' Second side stripper
11HJuLR
hr
Bubble cap
pjiLjri_rLjnL|d
- Gas oil
hjr-nun!-Jlln
Hot crude oi, Jl*^ ^ G"7iJ mi
•L *^
n U<r-hH in»L
plilLJnLJILjnb
11 jjj
UHJRL
t t t tlfl
Steam Vapour '"
Residue
Fig. 19.45: Diagram of a fractionating column and bubble cap
In Fig. 19.45, the design of tray is of the bubble-cap type. Other designs in use are variously described as valve, jet
and sieve. While these types differ in detail, the principle of operation remains the same.
consist fundamentally of a cylindrical shell and tube Hydrocracking: To meet the ever-increasing demand
bundle. A drawing of the most common design in of the internal combustion engine for light fuels with
refinery service is shown in Fig. 19.46. The following good spark ignition, processes to break-down more
points should be noted: The tube bundle has a fixed heavy fractions are required. A m o n g them,
tube sheet at the channel end and the other end is left hydrocracking is the latest development. The modern
free to allow for longitudinal expansion and contraction hydrocracker exemplifies the principal metallurgical
of the tubes. In the example, the channel cover and the problems of general refinery practice. It involves all the
floating head of the tube bundle are removable to commonly employed materials from carbon steel,
facilitate cleaning of the tubes. When tube cleaning is t h r o u g h the m o l y b d e n u m containing alloys, to
unimportant, or when leaking on the tube-side cannot austenitic stainless steel. In addition, the combination
be tolerated, a U-tube design may be used. In such cases of hydrogen and hydrogen sulphide at very high
the channel cover may be ellipsoidal and welded-on. In pressure and temperatures stretches those materials to
all designs, the channel provides inlet and outlet nozzles their limits. Finally, the resultant, very thick-walled
for the tube-side fluid. The exchanger may be designed reactors and heat exchangers demand extraordinary
to pass the tube-side fluid once, or as many times as expertise of all those involved in fabrication.
required. The shell-side fluid is circulated round the In Fig. 19.47, a simplified example of a hydrocracker
tubes and a baffle arrangement deflects flow to achieve flow diagram is presented, while in Table 19.10 is
maximum heat transfer. presented the materials used for the heat exchangers.
The various styles of heat exchanger design are Note again how heat from the reactor is used in heat
contained in S t a n d a r d s of Tubular Exchanger exchangers to preheat the heavy oil and hydrogen feed
Manufacturers Association (TEMA) of U.K. before it enters the heater. The higher temperatures
Fig. 19.46: Floating head type heat exchanger. 1 Stationary head—channel; 3 Stationary head flange—channel or
bonnet; 4 Channel cover; 5 Stationary head nozzle; 6 Stationary tubesheet; 7 Tubes; 8 Shell; 10 Shell flange— stationary
head end; 11 Shell flange—rear head end; 12 Shell nozzle; 13 Shell cover flange; 14 Expansion joint; 15 Floating
tubesheet; 16 Floating head cover; 17 Floating head flange; 27 Tie rods and spacers; 28 Transverse baffles or support
plates; 29 Impingement baffle; 31 Pass partition; 32 Vent connection; 33 Drain connection; 34 Instrument connection;
35 Support saddle; 36 Lifting lug (Reproduced by kind permission of the Tubular Exchanger Manufacturers Association).
T321
1
(7\ CS
Reactor I
2 1/A|
Compressor Quench hydrogen
m -»from recycle gas
./^scrubber
oil feed from
Cr-Mo I I vacuum unit
clad T347 i £jj. j-+> Recycle gas
/
T321
i Hydrogen
partial
pressure
Pump
J c o o l e n t o gas|
scrubber
Heater about I CMo HP
V
M7_MN/m csf separator
T32' Heat exchangers
T321 C 320^C 275°C P r °duct to
fractionator or
LP separator
Compressor
Recycle
hydrogen from
gas scrubber +
make-up hydrogen
Fig. 19.47: Simplified hydrocracker flow diagram
dictate the material selection: carbon steel at low cold hydrogen is injected into the reactor to absorb the
temperatures progressing through carbon- heat generated by the exothermic reaction of the
m o l y b d e n u m a n d c h r o m e - m o l y b d e n u m alloy to cracking process.
austenitic c h r o m i u m - n i c k e l steel at very h i g h The range of steels used in the hydrocracker forms
temperatures. In the reactor itself, heavy fractions are the basis of refinery material selection. The compositions
broken down and combined with hydrogen by passing and room temperature properties of these steels are
them through a series of fixed catalyst beds. Additional given in Table 19.11.
Table 19.10: Heat exchanger construction
Channel Shell
Unit Temp, °C Materials Temp, °C Materials
A 385-360 2%Cr-Mo 295-315 2UCr-Mo
Clad T347 Clad T347
B 360-320 2V4Cr-Mo 205-295 ■ ltfCr-Mo
Clad T347
C 320-275 2V4Cr-Mo 95-205 C-Mo
CladT347
Table 19.11: Composition and properties of plate materials

C,% Mn,% Y min, UTS El min.
ASTM Grade max. Si, % Cr, % Mo, % Ni, % N/mm 2 N/mm 2 %
A285 C 0.28 0.90 — — — — 205 380-515 23
A516 60 0.23 — 0.15-0.30 — — — 220 410-550 21
A516 65 0.26 — 0.15-0.30 — — — 240 450-585 19
A516 70 0.28 — 0.15-0.30 — — — 260 480-620 17
A204 B 0.20 0.90 0.15-0.30 — — — 275 480-620 17
A387 11 0.17 0.40-0,65 0.50-0.80 1-1.5 0.45-0.65 — 240 410-585 18
A387 22 0.15 0.30-0.60 max. 0.50 2-2.5 0.90-1.10 — 205 410-585 18
A387 5 0.15 0.30-0.60 max. 0;.50 4.0-6.0 0.45-0.65 — 205 410-585 18
A240 T347 0.08 2.00 1.00 17-19 NblOxC 9-13 205 515 min. 40
to 1.10
Effect of temperature: Primarily, the choice of steel type compete with C-Mo steel at about 500°C. With an
is dependent on operating temperature. Increases in increase of t e m p e r a t u r e above that point their
temperature reduce tensile strength, and with some advantages become more pronounced.
particular corrodents, increases the corrosion rate. The
Code's allowable stress tables reflect the effect of
temperature on strength and this is shown graphically
in Fig. 19.48 for the group of steels involved. Here it
will be seen that from a strength point of view, carbon
steel (ASTM A516 Gr. 70) with a room temperature
strength equal to that of the ferritic alloys compares
favourably with all these alloys up to about 350°C.
Carbon steel is still competitive economically up to
400°C, but above that temperature, an addition of Vi%
molybdenum provides the equivalent of about 50°C
advantage in service usefulness. Compared with carbon
steel at 425°C, the C-Mo steel has the advantage of a
30% reduction in thickness and at 450°C over a 50%
saving in metal thickness is achieved.
From a code and strength point of view, therefore,
and weighing the thickness advantage against the
higher unit cost of the molybdenum alloy, the break-
point in choice between carbon steel and C-V^Mo steel
occurs at a design temperature of about 400-425°C. Still 400 500 600
considering only high t e m p e r a t u r e strength and
economics, VAQr-WtAo and 21/4Cr-lMo steels begin to Fig. 19.48: Allowable stress values for various steels
The austenitic Cr-Ni stainless steels, it will be noted, API Std. 941. As new experience is obtained the curves
have excellent tensile properties at the top end of the are revised.
temperature range, but their relatively high cost restricts • Hydrogen sulphide and sulphur: H y d r o g e n
their use to these high temperatures, when strength is sulphide and sulphur cause general corrosion, which, as
the only consideration. They come more into their own a rule, is less with increasing additions of chromium in
w h e n corrosion, and corrosion plus temperature, the steel. Aqueous hydrogen sulphide, in particular, causes
become a primary factor. stress corrosion cracking and this is of critical concern at
Effect of corrodent: Corrosion is a complex subject, welded joints. Stress corrosion cracking may occur when
which is not within the scope of this book, but we must any two of the three possible factors coincide:
consider briefly the three principal corrodents found in a) Tensile stress, as imposed by residual stresses at
refinery service. These, which may occur separately or welded joints
together, are: hydrogen, hydrogen sulphide and sulphur. b) Hard microstructures, as often occur in the heat-
• Hydrogen: H y d r o g e n , at certain coincident affected zones of ferritic steel welds and some-
pressures and temperatures, causes cracking and times occur in their weld deposits
blistering of steel. The higher the coincident pressure c) A corrodent such as wet hydrogen sulphide.
and temperature, the higher is the risk of cracking. It follows that, when we have an environment which
Susceptibility to cracking and blistering varies with steel involves wet H 2 S, precautions must be observed to
composition. Carbon steel has a high susceptibility, but minimise the risk of stress corrosion cracking at welded
increasing additions of molybdenum and chromium joints of ferritic steels. The hardness of welds and their
reduce susceptibility in the ferritic steels while the heat-affected zones must be controlled by postweld heat
austenitic stainless steels appear to be immune from the treatment, and this also reduces the other contributory
problem. G.A. Nelson recorded the incidences of factor, residual stress. Standards of the National
hydrogen attack on pressure containing equipment and Association of Corrosion Engineers (NACE) lay down
plotted failures of specific materials graphically as safety rules for the control of stress corrosion, and like
shown in Fig. 19.49. From this actual experience, it the Nelson curves, these rules have been developed on
became possible to select material for a given coincident the basis of collected service experience.
hydrogen pressure and temperature with reasonable Final material choice: It is obvious that the following
reliability. The up-to-date curves are now contained in main physical factors must be considered before the final
choice of material is made: pressure, temperature,
hydrogen, hydrogen sulphide, sulphur and economics.
Hydrogen partial pressure, kg/cm a Hence choice must be carefully weighed in each case;
50 100 150 general rules cannot be given. At the same time it is
T
possible to indicate general trends as follows:
• Vessels: For vessels, carbon steel is used wherever
possible with a suitable corrosion allowance up to its eco-
nomic temperature limit of about 400°C If hydrogen is
involved then the temperature limit will be lower. If
hydrogen sulphide or sulphur is involved and the added
corrosion allowance becomes too large, then the econ-
omics of using a stainless steel clad plate must be
considered.
Above 425°C, C-V2M0 steel is the first choice, and
again hydrogen will be the first limitation. Stainless
cladding must be added when the necessary corrosion
allowance becomes uneconomic.
As hydrogen, temperature and pressure exhaust the
500 1000 1500 2000 usefulness of C-V^VIo steel, the l a 4Cr-Mo steel is used.
Hydrogen partial pressure, psia Again, stainless steel cladding will be added as corrosive
conditions demand. Finally, at the extremes of hydrogen,
Fig. 19.49: Nelson curves—material selection for
hydrogen service pressure, temperature and corrosive conditions, 2V^Cr-
lMo clad plate is used.
• Piping: The general trend for piping is similar to b) Application of preheat and maintenance of a high
that described for vessels, except that it is often not pos- interpass temperature to arrest the cooling rate
sible to use stainless steel cladding. This means that where and diffuse hydrogen away.
general corrosion is a problem, a solid corrosion-resistant c) Use of a high arc energy to boost the weld pool
alloy steel such as 5Cr-V£Mo may be used. Again, the temperature and diffuse hydrogen away.
choice is somewhat different, i.e. economic one as
d) Application of postweld heat treatment to soften
compared with that in the case of vessels. Wall thicknesses
the hard microstructure, minimise residual ten-
are relatively smaller and so larger corrosion allowances
sile stress and diffuse out any residual hydrogen.
can be afforded. When these thicknesses become too
great, then the much better corrosion resistance of the A p a r t from the fundamental characteristic of
5Cr-M>Mo steel makes it an attractive choice, but the hardenability, lamellar tearing is a problem that occurs
costs of welding, preheat and postweld heat treatment with plate, particularly in thicker sections. There are
are high. Solid austenitic stainless steel, with its excellent several precautions that can be employed to avoid this
corrosion-resistance and high temperature strength, and they are explained in Chapter 12. It is up to the
enables relatively thin-wall designs to be used and hence designer, metallurgist and welding engineer to decide
becomes a close rival to the 5Cr-44Mo steel. In addition, on the most economic solution.
it presents lower welding costs. Other problems, which to a greater or lesser extent,
With materials selected to meet all the basic are dependent on the type of alloy, require the combined
conditions described above, it only remains for the efforts of the same specialists to resolve. These are:
designer to understand their various characteristics in solidification cracking, liquation cracking, reheat
relation to their weldability and possible failure. The cracking and temper embrittlement, and they have been
designer can then see what precautions must be taken dealt with in Chapters 9 and 10.
and written into the fabrication specification.
For the common ferritic steels used in refinery FERTILISER PLANT
construction, Table 19.12 shows their calculated carbon Here we shall describe welding applications in a
equivalents. When considering these CE values, it fertiliser plant in India designed for 1100 t o n s / d a y
should be borne in mind that a CE of 0.45% represents ammonia, 1,600 tons/day urea, 470 tons/day sulphuric
the limit above which various precautions such as acid, 165 tons/day phosphoric acid and 500 tons/day
preheat, low-hydrogen welding process and postheat DAP. The main materials of construction used in this
must be applied to increasing extents to achieve crack- plant were:(a) carbon steels, (b) low-alloy steels, and (c)
free welds. CE, as explained in Chapter 10, is therefore stainless steels.
a simple and more precise w a y of defining the a) Carbon steels: The project used various grades of
weldability of ferritic steels. carbon steel conforming to the specifications of ASTM,
The higher the carbon equivalent and the thicker the such as A106 Gr. B and C, API 5L Gr. A and B, A53 Gr A
part to be welded, the more likely it will be for a hard and B and JIS STPT 42, STPT 49, STPG 38, STPT 38 for
microstructure to form in the HAZ. With a suitable pipelines; and ASTM A515 Gr 60, BS 1501-161-Gr 28 A
a m o u n t of h y d r o g e n in the weld pool such a L T, etc., for plates. For welding these materials, E7018
microstructure will crack. Hence it becomes necessary class electrode was generally used. In the case of pipes
to apply the following precautions: of more than 19 mm wall thickness, the root pass was
a) Use of a low-hydrogen welding process to mini- deposited by the TIG process using TGS 50 filler wire,
and the subsequent passes were deposited with E7018
mise the amount of hydrogen in the weld pool.
electrodes. For pipes of wall thickness less than 10 mm,
Table 19.12: Carbon equivalents for various materials root passes were deposited with E6013/E6010 electrodes
and filler passes with E7018. This procedure successfully
ASME
met the radiography, face bend, root bend and tensile
group Type of steel ASTM Grade CE
test requirements.
PI Low-carbon steel A285 55 0.43
PI Medium-carbon steel
Lines carrying vetrocoke solution were welded with
A516 70 0.46
P3 Carbon-V^Mo A204 8 0.47 E7018 electrodes and the joints were stress-relieved at
P4 VACr-ViMo A387 11 0.71 600°C with soaking time of 0.5 hour. The line size was
P5 2V4Cr-lMo A387 22 0.95 from 51 to 406 m m NB and Such 40 with material
P5 5Cr-1/Mo A387 5 1.68 specification API 5L Gr. B-ERW. Vetrocoke solution is a
mixture of potassium carbonate and arsenic trioxide, Maintenance welding is carried out regularly to keep
which induces caustic e m b r i t t l e m e n t a n d stress the plant functioning at maximum efficiency. This
corrosion on the weldments. To avoid this, all the involves building up of worn-out shafts; impellers, cast
weldments which came in contact with this fluid were components, blade tips, shaft sleeves, etc., and welding
stress-relieved. The bends used on these lines were of of cast iron parts such as casings of bearing housings.
stainless steel, conforming to ASTM A 312 Gr 304 to Cast iron parts are welded with Ni-based manual
avoid corrosion on the bends. The combination weld electrodes, which do not require preheating. Chrome
between these elbows and pipeline created some cast iron is the material used for pipelines and pumps
difficulties. The weldment hardness value was very handling hot sulphuric acid. This material is extremely
much on the higher side when compared to parent difficult to weld and a matching electrode is difficult to
material. Also lot of chromium carbide precipitation was obtain. For shafts, shaft sleeves and other parts. Alloy
found on the side of carbon steel heat-affected zone and 20 and Duriamet 20 are used. Similarly, material CA 955
half of the weldments. The weldments were therefore is used for handling hot phosphoric acid in pump shaft
annealed at 900°C with a soaking time of five minutes and impeller. It is difficult to get matching electrodes
and the hardness value was found to be acceptable. for these materials also. Hastelloy B and C are used for
Annealing heat was given on the side of carbon steel phosphoric acid pump shafts, sleeves, agitator shaft, etc.
and one-half of the weldment on side of carbon steel These materials can be satisfactorily welded. Parts which
was also included in the heating. In service, no problems are subject to heavy wear and tear such as jaw crushers,
were encountered. granulators, scraper tips and plunger tips are built up
b) Low-alloy steels: The materials used conformed to by stelliting.
the specifications of ASTM A 335 Gr P 11, JIS STBA 11 Equipment for urea service: Of the nitrogenous
and STPA 22. For JIS material, TIG process was used for fertilisers, urea constitutes perhaps the most important,
the root pass with TGS 1 CM filler wire and E8016-B2 due to its high nitrogen content, approximately 46.6%,
type electrode for the filling passes. For ASTM material, combined with easy transportability, ease of handling
E8018-B2 electrodes were used entirely. Preheating and and comparative economy in price. In manufacturing
stress-relieving were done as per the code requirements. urea, the process stream, however, is subjected to severe
For the combination welding of P 11 and carbon steel A corrosive conditions. The selection of materials for urea
106 Gr B / C , E8018-B2 electrode w a s u s e d w i t h service is principally governed by corrosion-resistance
preheating and stress-relieving as per the alloy steel to a m m o n i u m carbamate at various temperatures.
requirement by the code. Ammonia (NH3) and carbon dioxide (C0 2 ) are reacted
c) Stainless steels: The materials in this category in a synthesis converter to form urea (NH 2 CONH 2 ).
conformed to the specifications of ASTM A 312, TP 304, The ammonium carbamate is formed as an intermediate
TP 321, TP 316, Inconel 800, JIS SUS 27 TP, SUS 32 TP, product during the reaction, as shown below:
SUS 33 TP. For all stainless steel welding, TIG process
2NH 3 + C 0 2 = NH 2 COONH 4 (ammonium carbamate)
was used for the root pass and matching manual
NH 2 COONH 4 = NH 2 CONH 2 (urea) + H 2 0 (water)
electrodes for the remaining passes. For most joints,
internal purging with argon gas was used. For some of The equipments, besides synthesis reactor, that come
the low-pressure lines, nitrogen was used for purging. into synthesis line are d e c o m p o s e r s , s t r i p p e r s ,
The resulting welded joints were found acceptable and condensers, separators, crystallisers, etc. Depending on
gave satisfactory service performance. For 304 and 304L the process adopted, the high pressure decomposers and
materials, E 308-16 and E 308 L-16 type electrodes and reactors work at temperatures in the range of 180-220°C
corresponding TIG Filler wires were used. For 316 and a n d p r e s s u r e s in the range of 240-140 k g / c m 2 .
316L materials, E 316-16 and E 316L-16 type electrodes Concentration of carbamate varies from equipment to
and corresponding filler wires were used. For 321 equipment and process to process, but in general, the
material, E 347-16 type electrodes and corresponding high pressure stages where high temperatures (up to
filler wires were used. For urea quality JIS SUS 33 TP 220°C) are encountered, can be termed as most critical
material, E 316L-16 type electrodes with guaranteed from the standpoint of corrosion. From the view of
ferrite content of 2% maximum and 0.6% minimum in ultimate economy and availability, three of the materials
the weld deposit were used. In the reformed gas boiler have found wide usage in urea service. These are SS
of ammonia plant, Incoloy 800 liner plates above the 316L (standard and modified grades), titanium, and
refractory were welded with ENiCrFe, type electrodes. zirconium.
In urea plants, reactors are generally made either Ni, Cr and Mo contents significantly alter the
with 316L stainless lining or titanium lining. In a few resistance of the steel to Huey test and to carbamates.
cases, zirconium-lined reactors and titanium strippers The investigations have established that a Ni content of
are used. Equipments like condensers, decomposers, 13% minimum is necessary to get reasonable life in
separators and crystallisers are made of 316L stainless carbamate service. Increase in the amount of Ni requires
steels in most plants, due to the excellent service record an increase in chromium content to get a fully austenitic
of the material. deposit free from tendency to cracking. Besides higher
Fabrication of SS 316L for urea service is no different Ni, Cr and Mo contents increase the range of stable
from fabrication of this material for other services. passivity of the material, thereby making the material
However, there are certain subtle differences which less active even with lesser oxidising additions. In the
characterise these fabrications and they must be taken stripper of some of the urea plants using C 0 2 stripping,
into consideration in very early stages: 25 Cr-22 Ni-Mo material has been used after persistent
a) Selection of filler metal with controlled ferrite and failure of 316L material, the failure being caused
controlled composition supposedly by the presence of lesser amount of oxygen
b) Use of proper welding technique and manipula- and extremely severe corrosive conditions prevailing in
tion for obtaining desired weld deposit of con- the equipment together with excess amount of C0 2 ,
trolled ferrite, free from micro-cracks which shifts the active-passive range to the active zone.
The corrosion resistance of some of the stainless steels
c) Type and nature of control tests. which find application in urea service are given in Table
In the selection of filler metal for 316L steel, ferrite 19.13, which clearly shows the superiority of 25-22-2
content of less than 2% in the weld deposit is specified. over standard 316L. However, suitably modified types
Such ferrite control is justified by the results obtained of 316L have given quite satisfactory service in reactors
in detailed investigations on the relations between as well as H.P. stage equipments. A typical composition
corrosion rates in Huey Test (65% boiling nitric acid test) usually preferred for the purpose is:
and the rate obtained in practice in urea reactors. It has
been concluded that there is a definite reproducibility C—0.035 maximum Ni—13-16 Cr—16.5-19.5
between Huey test corrosion rates and corrosion rates Mo—2.3-3.0 N—0.20 maximum
in carbamate. Hence this test (described in Chapter 13)
can be treated as a reliable test for evaluating materials Welding technique: The welding technique used for
for urea service. Investigations showed that increase in urea service should ensure the following: (a) controlled
ferrite content increases corrosion rate in Huey Test ferrite, (b) controlled composition, (c) freedom from
considerably. It is clear that to achieve a corrosion rate micro-cracking, (d) in case of cladding, proper deposit
lower than around 24 mils per year with 18-8-Mo characteristics after dilution in at least top 3 to 5 mm
composition, the deposit should contain the least ferrite. layer. Some of the precautions that should be taken in
This requirement can be variously specified in the form this connection are :
of either maximum limit of ferrite content, stipulation 1) Maintaining a short arc, so that transfer of the
of Cr (eq)/Ni (eq) ratio, etc. in the latter case the ratio elements through the arc is uniform and burning out of
being maintained at 1.22 maximum, corresponding to a the elements, specially chromium, is reduced to a
ferrite level of 0.1% maximum. The ratio is sometimes minimum.
relaxed to 1.29 maximum. For practical purposes the
2) Cleanliness of the welding edges/surfaces. The
ferrite contents that may be accepted in the weld and
surfaces must be absolutely free from carbonaceous and
base metals in the high pressure and other equipments
sulphur-bearing materials such as oil, grease, etc., so that
without much deleterious effects are as given below: pick up of C, S, etc., does not take place. This is neces-
a) For high pressure equipment such as reactor, sary to get adequate corrosion-resistance of the welds.
high-pressure decomposer, stripper, etc. 3) The weld should not be allowed to stay in the high
Parent plate : 0.6% maximum temperature zone for any length of time. Checks should
Weld deposit : 2.0% maximum be made to ensure that the adjacent parent metal does
b) Other equipments in the medium and low-pres- not cross 150°C at any time. The weld should not be
sure stages subjected to 500-850°C range for considerable time. This
Parent plate no restriction is the range where the material is susceptible to
Weld deposit : 2.0% maximum intercrystalline corrosion in the presence of nitric acid
Table 19.13: S.S. compositions for urea service

Brand name Assab 724L Assab 724LN Dew 4435 Sandvik Sandvik
Sonder-2 3R60 2 RE 69
Chem. Composition C £ 0.03 C0.02 C £ 0.03 C £ 0.03 C £ 0.03
(%) Si £1.0 Si 1.0 Si£ 1.0 Si 0.6 Si £ 0.40
Mn£2.0 Mn£2.0 Mn £ 2.0 Mn 1.70 Mn£1.75
P £ 0.045 P 0.035 — — P £ 0.02
S £ 0.030 S 0.020 — — S £ 0.015
Cr 17.5 Cr 25.0 Cr 17.0-19.0 Cr 17.0 Cr 25.0
Ni > 13.5 Ni 22.0 Ni 13.5-16.0 Ni 13.6 Ni 22.0
Mo 2.6 Mo 2.1 Mo 2.4-3.0 Mo 2.8
Others — N0.12 — — —
Ferrite content < 0.6 % < 0.6% < 0.6% — —
Corrosion rate
(Huey test) < 0.6 mm/year 0.18 mm/year < 0.6 mm/year < 0.6 mm/year < 0.6 mm/year
Remarks 0.25 mm/year 0.09 mm/year
can be achieved can be achieved
and urea. Unfortunately, deterioration of the material In clad constructions, generally two or more layers
on this count cannot be detected by simpler tests like of cladding metal are used for building up desired
strauss or acidified copper sulphate tests. thickness of corrosion-resistant layer. In case of manual
4) The selection of filler metal and welding technique metal-arc m e t h o d , the dilution m u s t and can be
should ensure that the weld is free from any micro- controlled to as low as 10 to 15%. Dilution of the SS layer
cracking. Fully austenitic deposits are quite susceptible with carbon and chrome-molybdenum steel can result
to this defect unless proper care is exercised. Researches in the formation of less corrosion-resistant intermediate
have shown that micro-cracking can be prevented by compositions and in some cases lead to martensite
increasing the Mn content of the weld-metal, without formation. The martensitic phases can cause lower
affecting the corrosion-resistance of the metal. Some of ductility, giving rise to cracks and lack of cohesion with
the brands of filler metal which have been already used the base metal. Stress-relief treatment can further lower
in this service have typical composition as below: the toughness of the material. Suitable highly alloyed
C — 0.05 maximum Cr — 20.5 Ni — 15.5 fillers such as 25 Cr-12 Ni-Mo, 25 Cr-20 Ni and lnconel-
Mo —3.0 Mn —7.5 N —Present 182 for buffer layer have been found to surmount these
The cracking susceptibility is also controlled by difficulties. However, even with their use a final layer
limiting the heat input and hence dwell time at high of corrosion-resistant material must be given, as the
t e m p e r a t u r e r a n g e . Extra-low P a n d S contents, dilution in the deposit is difficult to predict with much
preferably 0.020% maximum in the filler wire have reliability, and the corrosion-resistance with higher
beneficial effect. carbon content material (due to dilution of base metal)
5) In case of weld cladding where carbon or Cr-Mo may not be comparable to the undiluted deposit.
steels may be clad with SS material, some of the With clad constructions, stress-relieving treatments,
considerations given to the procedure are: (a) choosing when considered necessary, should as far as possible be
right filler metal for the intermediate and final layers, given in the intermediate stage and not in the final stage.
(b) selection of parameters for preheating and postheat This is because the material may get sensitised during
treatment operations, (c) choosing the right sequence stress-relief treatment and lose its corrosion-resistance
and stage for postheat treatment and selection of properties to some extent. If it is not practicable to stress-
treatment p a r a m e t e r s in such a w a y that neither relieve the cladding in the intermediate stage and has
brittleness nor loss of corrosion-resistance is induced, to be done after the fabrication is completed, proper time
(d) in case of flat plates such as tube-sheets, the sequence and temperature should be selected after consulting the
of runs and other techniques to prevent or minimise a p p r o p r i a t e TTS (time-temperature-sensitisation)
distortion of the component, and (e) choosing proper diagram for the deposit, before embarking on the heat
welding parameters to keep dilution under control. treatment.
Use of superalloys: Superalloys are heat-resistant blasting or grinding. If this is not done, the oxides
nickel alloys or highly alloyed ferrous metals, which are accumulate on multi-pass welds and they may become
used in the fertiliser industry for specific corrosion- thick enough to inhibit weld fusion and to produce
resistance and other severe service conditions, where laminar type oxide strings along the weld axis, which
the conventional materials fail prematurely. These alloys act as mechanical stress raisers and may cause stress
are used in some of the initial and intermediate stages corrosion cracking in service.
of processing with remarkable success. 4) Preheating: Preheating is generally avoided, but in
The superalloys commonly used are Inconel 625, some cases where job is below 15°C, it must be heated
Inconel 748, lncolloy-800, HK-40, Hastelloy B, Hastelloy to above ambient temperature. However, cast alloys
C, Hastelloy D and Hastelloy X, and some proprietory should be heated to 100-200°C, depending on the mass
alloys like R- 55. Their compositions are given below: of the casting at the joint.
5) Postweld heat treatment: No postweld heat
Inconel 625 —Ni—61%, C—0.05%, Fe—2.5%, treatment is needed to maintain or restore corrosion-
Cr—21.5%, Mo—9%, Nb—3.65% resistance, except where stress-relief is necessary. Age-
Incolloy 800 —Cr—21.7%, Ni—34%, Fe—Balance, hardening alloys are normally welded in the solution-
C—0.069%, Mn—0.75%, Si—0.52%, treated condition and age-hardened after welding.
Ti—0.36%, Nb—0.45%, Al—0.25%, 6) Filler metal: Ni-based alloys arc generally subject
S—0.009%, Cu—0.21% to porosity, if the weld-metal does not contain nitride
HK-40 —C—0.35 to 0.45%, Mn—2%, Si—1.5%, forming and deoxidising elements. So electrodes and
P—0.03%, S—0.03%, Cr—24 to filler rods meant for arc welding must contain Al, Ti, or
27%, Ni—19 to 22%, Mo—0.5% n i o b i u m or a combination thereof. Weld-metal
Hastelloy-B —Ni—62%, Mo—28%, Fe—5% deposited on Si-bearing 18% C r - 3 8 % Ni alloy is
Hastelloy-C —Ni—54%, Mo—17%, Cr—15%, susceptible to hot cracking, if the filler metal of identical
Fe—5%, W - 4 % chemistry is used. So the alloy should be welded with a
Hastelloy-D —Ni—85%, Si—10%, Cu—3% low silicon 80% Ni-20% Cr type filler, and dilution
Hastelloy-X —Cr—22%, Mo—9%, Fe—18%, should be kept minimum.
Ni—Balance 7) Precautions: Some additional precautions to be
R-55 alloy —Cr—22.5%, Ni—Balance, W—4%, taken are:
Mo—4%, Si—4%, Fe-10.5%. a) Parts which have been worked or deformed are
given a stress-relief treatment prior to welding.
GENERAL CHARACTERISTICS b) Welded parts are rapidly heated and cooled
1) Process: The most widely used processes for through the aging range to avoid the harmful effects of
welding the superalloys are TIG, MIG and MMA embrittling carbides in grain boundaries.
welding. Solid solution superalloys are easily weldable, c) Since a m a t r i x - s t r e n g t h e n e d filler metal is
but precipitation hardening superalloys require special generally used, yielding between 80 and 90% joint
precautions. efficiency, fusion welds are designed where possible in
2) Joint preparation: Since the weld-metal of high-Ni low-stress areas.
superalloy electrodes does not spread and penetrate 8) Application ofNi-base consumables for welding: These
deeply as does steel weld-metal, angle of V for butt joint are preferred because they have the following interesting
preparation of these alloys is kept wider and proper properties:
provisions are made between the root faces to ensure a) A high metallurgical tolerance towards dilution
complete penetration of the weld-metal. by various metals.
3) Cleaning: Ni-based superalloys are susceptible lo b) Good mechanical strength and chemical resistance
embrittlement by lead, S, P and low melting-point at high temperatures.
metals like bismuth. Therefore perfect cleaning of the c) Good strength and ductility at room temperature.
welding joint is essential, so that it is free from dirt, d) High toughness, even at very low temperatures.
grease, oils and the above materials. After welding of
each r u n , slag on the surface m u s t be r e m o v e d WELDING OF R-55 ALLOY
thoroughly; otherwise severe corrosion can occur, A large number of pump casings and impellers made
particularly in oxygen containing atmosphere. Oxide of R-55 alloy for service of s u l p h u r i c acid and
formed during welding should be removed by abrasive ammonium sulphate solution are corroded and eroded
every year. Defective parts are put to service successfully Electrode for Inconel 182 Incoweld
by weld surfacing by the following procedure: subsequent layers
Welding of wetted surface Preheat temp. 150°C nil
a) Faulty p o r t i o n of the casting is g r o u n d to Interpass temp. 150°C nil
approximately two-thirds of the d e p t h of casting
thickness and chamfered all round.
b) Casting is slowly heated to about 200-230°C. Besides other benefits, the lncoloy-800 has the
c) Then material is welded either in the furnace or advantage of nipping if any particular tube is to be
on the outer edge of the furnace. Proper care should be isolated in which the failure has occurred d u e to
taken so that no heat loss takes place during welding. excessive creep in the pigtails, or at the welds joining
d) Material is welded either with a cast R-55 alloy pigtails to catalyst tube and header.
or Hastelloy C rod using borax commercial flux. After welding, the solidification cracking is generally
e) Finally the material is cooled very slowly by restricted to the weld crater positions of the root runs.
covering with insulation materials. Utmost care is required during welding. The inside
Welding of non-wetted surface surface of the joint should be completely purged with
Arc welding process is used and there is no need for inert gas to protect from oxidation. After welding, the
heating the casting for this operation. weld surface should be finished by grinding, so as to
eliminate irregularities and examined by both the
WELDING OF INCOLOY-800 radiographic and dye penetrant methods.
In pigtails of naphtha reformer of a fertiliser plant,
lncoloy-800 is often used. The heater is spun cast and as WELDING OF HK-40
such requires no welding, but the critical joints between The alloy finds its main use in gas reformer tubes in
the reformer (HK-40) tubes and the pigtails have to be fertiliser industries. The tubes are s p u n cast, and
welded as well as repaired. For welding, lnconel-82 filler welding is required only at the ends or for the repairs of
metal having the following composition is used: the tubes which have failed due to excessive bowing,
uneven heat flux, corrosion at internal as well as external
Cr—20 to 22% Ni—65 to 67% Fe—3% Mn—2.5 to surfaces, and deactivation of catalyst followed by delay
3.5% in replacement. Effective non-destructive tests have
C—0.1% Si—0.5% Ti—2 to 3% Nb—2 to 3% shown that tubes have failed generally after attainment
of 4% creep and 0.5% or more of ferrite.
The base metal and the filler metal differ in the The filler metal used here also is lnconel-82 with TIG
following respects: welding, which gives welds having better stress rupture
a) lncoloy-800 is based upon 20 C r / 3 5 N i / 4 5 Fe strength.
composition, whereas lnconel-82 is based on 20 Cr/80 To sum up, the following principles are to be abided
Ni. by before welding the superalloys:
b) Incoloy is hardened and partly deoxidised by Ti a) To h a v e a t h o r o u g h k n o w l e d g e of the
and Al, which form a fine intermetallic dispersion, compositions of materials being welded
whereas lnconel-82 is hardened by Ti and Nb (and not b) To make choice of right rods and electrodes
by Al). In addition, about 3% manganese is present m c) To maintain the time at preheating temperatures
the filler metal and is required for sulphur control, as in d) To h a v e a control over w e l d e r s ' ability by
ferritic weld-metals. qualification tests.
The root run is completed by TIG welding having Welding of tower of methanol scrubber : A fertiliser
2% thoriated tungsten electrode with argon as shielding project in India had to weld the tower of methanol
gas at 90 amps, and the successive run is also completed scrubber, received from Europe in four pieces, which
with TIG. when erected would have a total weight of 320 tons and
The following welding conditions are observed: a height of 72.9 m. The tower was designed for operating
in the sub-zero temperature range of 80/ -50°C.
Reformer tube Outer header The base material was TTStE36 (equivalent to 537
side side C l l ) and its thickness ranged between 48 and 58 mm.
Electrode for lnconel-82, 2.3 lnconel-82, 2.3 The welding electrodes were of E8016-C3 classification.
first layer mm dia mm dia Its weld-metal analysis was (%):
C-0.06, Mn-0.93, Si-0.27, S-0.05, P-0.018, Ni-1.10, M o - 6,6', 5' and 7,8,8', 7' took up the vertical position. Under
0.32, and gave an average Charpy-V impact value of these positions, the entire volume was welded. After
4.3 kgf- m at -50°C welding of these positions, the vessel diameter and
In each of the joints, 100 mm wide heating coil verticality were checked to locate distortion if any. If
(nichrome wires 4 mm diameter) was kept and insulated any distortion was found, the heat input was adjusted
approximately to 1 m width, and the heating was done against the volume of weld left-out in 1,2 and 2', Y and
by electric resistance m e t h o d . The p r e h e a t i n g 3, 4, and 4', 3'. Figure 19.50 explains this clearly.
t e m p e r a t u r e r e c o m m e n d e d is 120-150°C. The At the root, the second side was grooved by grinding
preheating temperature was checked at various points till sound metal was reached, dye-checked and filled
by means of suitable temperature indicating crayons. u p with sealing passes in the same sequence as
The interpass temperature of 200°C was also controlled explained earlier, by rotating the vessel.
by suitable temperature indicating crayons. Soon after Postweld heat treatment of the welded joints was
the required preheat temperature was attained, two carried out by the internal bulkhead method (Chapter
welders started welding on opposite sides in vertical 20) which has been developed specifically for the heat
position as shown in Fig. 19.50. treatment of circumferential weld joints on large
The entire circumference of the tower shell was diameter vessels. Using this method, the area around
divided into eight equal parts of 45°. The shell was in the welded seam was converted into its own furnace as
such a position as to permit the two welders to start shown in Fig. 19.51. Two insulated bulkheads were
welding only in vertical position 1, 2 and 2', 1'. In this erected on either side of the welded joint. These
position, only 5% of the volume could be completed. comprised simple, single iron framework, on to which
The vessel was rotated (through 90°) in such a way that two layers of mineral wool insulation were attached.
3.4 and 3', 4' occupied the position 2,1 and 2'', V. In this Between the bulkheads, a number of 4-bank channel
3, 4,, 3', 4' position, 50% of the volume was completed. elements were mounted in tiers on a scrap channel
Likewise, the vessel was rotated so that the position 5, section. Each element was rated at 13.33 kW - 240 V, 57
S=48-58 mm
Thickness
180° 4 8 - 5 8 mm
i 50% of welding - 1, 2 2\V
\\ 50% of welding - 4,3,3' 4 '
III 100% of welding- 6,5,5'6'
IV 100% of welding - 7,8,8,7»
V 50% of welding - 1,2,2',V
VI 50% of welding 4 ^ y 4
Fig. 19.50: Tower T-501 methanol scrubber-welding sequence adopted

. Insulation two layers

mm Thk
50 x 6^ mm
M.S. Flat
Four Bank 4 Bank channel heaters up to

channel heater 1/3 height of the diameter
Note: (1) (2) (3) and (4) are

chrome I-a I u me I control
thermocouples 8 or 9 swg
Fig. 19.51: Postweld heat treatment of tower using Internal bulkhead' method
amps. The 4-bank channel elements were connected to chemical operations, vessels and pipings capable of
440V power supply via a heat control and distribution operating satisfactorily at cryogenic temperatures are
board. required. In the fertiliser industry, the cryogenic vessels
The control and distribution board provided the form the heart of the process system, and any failure in
control of energy output to each element by on/off them may either be catastrophic or very costly. The weld
switching method. Energy regulators provided semi- quality control is a single factor which will reduce such
automatic control of heat treatment operation under a losses and thereby finally result in reduced production
skilled supervisor. Two layers of mineral wool insulation and maintenance cost of the plant. Proper under-
were applied over the heated zone, b e t w e e n the standing of the quality control procedures to be followed
bulkheads to the exterior of the vessel. A further single in the welding of cryogenic materials will also avoid
layer was applied at either side of the heated zone to undue repairs and rejection of welded fabrication.
protect against any harmful temperature gradient. The The vessels are fabricated under ASME Boiler and
heat treatment procedure was as follows: Pressure Vessel Code, Sec. VIII, and yet they must meet
Stress-relieving temperature — 550-570°C additional stringent requirements. For example, the
(two hours soaking time) welds have to be absolutely defect-free and leak-tight
Heating rate—100°C per hour in order to stand vacuum. Mechanical properties,
While cooling, after the temperature dropped to specially impact property of the weldment and its heat-
300°C, the joint area was allowed to cool in air.
Macro-hardness measurements were made across a
213-
specimen of the welded joint. The results are shown in
218—-A 198 199 193 / -
Fig. 19.52.
170 1 7 2 ^ 9 3 20 f/161162
PLANT FOR CRYOGENIC SERVICE 17g 17Q\206197 200170 1 i : R
Cryogenics is a branch of physics concerned with M7ÂV*$?>^ 247
phenomena occurring below -100°C and d o w n to 236 '168 /193!170 243
-273°C (absolute zero). The cryogenic science is being
used in fields like gas separation, deep sea oil extraction,
229~~* 222
missile launching, electronics, biology a n d food
handling. Material TTSTE 36 Tower T 501
For the separation and extraction of gases like
Fig. 19.52: Hardness survey with vickers 10 kg load
oxygen, nitrogen and argon from air and for many other
affected zones at sub-zero temperatures is another major Temperature, °C

factor that needs consideration. Role of w e l d i n g
? § s
technology starting from selection of materials to final
testing thus becomes crucial in the manufacture of
cryogenic equipment.
Selection of materials for cryogenic vessels
necessitates certain special considerations to physical
and mechanical properties. They would include the
familiar yield and tensile properties, ductility, fatigue
limit, impact and toughness properties. Ductility, the
ability of material to deform plastically and toughness,
resistance to brittle failure under specific conditions of
stress concentrations, are important prerequisites in
metals that are considered for use at low or cryogenic
temperatures. Selection of materials for cryogenic
services is also based on various physical properties like
low heat conductivity, low co-efficient of thermal
expansion and low emissivity. All such properties can
be used to advantage in storage vessels, vacuum transfer
lines and other components of cryogenic systems. -400 -300 -200 -100 0 100 200
Temperature °F
Of all the metals useful in construction for low
temperature applications, alloy steels remain the most Fig. 19.53: Impact properties of carbon steel and alloy
popular, because they are efficient, most readily steels at cryogenic temperatures. Boiling points of
gases are shown on abscissa.
available, most versatile a n d most economical.
Austenitic stainless steel (which is non-hardenable by
heat treatment) is an excellent material over the entire wide range of temperature and superiority of austenitic
range of cryogenic applications, because of its toughness stainless steels at all temperatures (15 ft-lbs = 20 J).
and high ductility at low temperatures and because it is In addition to these steels, copper and aluminium
highly corrosion-resistant. It remains tough even at alloys are also extensively used in cryogenic plants.
liquid-hydrogen and liquid-helium temperatures and However, use of copper is restricted to the low and
its co-efficient of thermal expansion and thermal medium capacity air separation plants in columns and
conductivity are lower than those of non-ferrous trays, while aluminium alloy (4.5% Mg) is extensively
cryogenic structural materials. used in tonnage plants. As mentioned earlier, selection
Linnert has described in his book Welding Metallurgy, is based on factors like toughness, ductility and fatigue
a few remedial steps in steel-making for securing good resistance.
fracture toughness, of which heat treatment and alloying In the fabrication of cryogenic vessels, procedure
with nickel are important. Heat treatment can improve qualification tests are conducted as per ASME Pressure
fracture toughness of low-alloy steels and extend their Vessel Code, Sec. VIII and IX, before actual production
range of usefulness to about -60°C. By adding nickel in welding is allowed to take place. Mechanical tests
substantial amounts, the range of applicability can be including low-temperature impact tests are conducted
extended as follows: on the procedure test specimens. Figure 19.54 shows a
typical arrangement for removal of test specimens from
Steel composition Lowest service temperature the p r o c e d u r e test plate and test specimen
2VA% Nickel '/ _60°C configurations. It can be noted that notches for the
3%% Nickel —100°C impact test specimens are shifted from centreline of weld
5% Nickel / —130°C to fusion line and slowly to the heat-affected zone. This
9% Nickel —200° C technique is practised in Czechoslovakia and many
other East European countries. The results can be plotted
These temperature limits are based upon a minimum in a graph, which will show the lowest impact strength
Charpy impact strength of 15 ft-lbs. Figure 19.53 by at a particular point between weld and the HAZ. The
Linnert indicates the impact properties of steels over a acceptance of the impact test is based on the fact that
careful flux removal between the weld passes followed

by grinding the entire groove to bright metal, is required.
' 'f b) Joint design: For grooved joint, wider included
c 1
o 22V 2 C 221/i° angle, compared to ordinary constructional steels, is
1 *<n
-3 c,
a; I recommended. Joint angles of 60°-80°in both single and
u
1
" £
1 **** 5I double-vee groove welds are necessary.
a <D 1
H #1 d êj
2 c) Filler metal: Manual metal-arc welding, when first
^
0.25 Root rad. applied to the 9% nickel steel using electrodes of
la
fr^
\ o ml Enlarged view of 'A' matching metal composition, gave rise to weld-metal
unU ni
o Scale: - N.T.S. cracking. Besides, the weld-metal exhibited Charpy-V
1f
impact strength of 5 ft-lbs at -200°C, which was not
} adequate. Weldments made with Inconel type filler wire
13 5
1 tn
£ W ¥=\ have s h o w n a d e q u a t e fracture t o u g h n e s s at low
i & **— temperature, as can be seen from Fig. 19.55, having
Cn^PXTYP)- ^
4^
11 a tensile strength almost equal to base metal but having
I °- I U7T I yield strength up to 55 ksi compared to 60 ksi in base
E
JT- metal (ASTM A-353, Gr. A). No difficulty is encountered
S
' 1 V 1
1 i
Charpy V - N o t c h
test specimen
10
impact
in qualifying these electrodes for ASME Code
requirements and hence these are widely used. Because
IU & J k- the weld-metal deposited by these electrodes has
approximately 95% joint strength efficiency, the ASME
Arrangement for removal of
impact test specimens from allowable design stress is proportionately reduced for
weld procedure plate 9% nickel steel in welded constructions.
specimen
0
a
0
b
— -
c ~5~~ e
- - From parent | ' I Temperature °C
plate as shown
j F - L_g Lmm~ ^mr^J •5 mrr -in Fig. aboveJ -250 -200 -150 - 1 0 0 ' - 5 0
-1 1—r
50
Kange of applicability
Fig. 19.54: Weld procedure test: Charpy-V notch impact 1 8 - 8 Austenitic stainless steel
test specimen for cryogenic vessels ♦ 9% Nicke) stee'l"
b% Nickel steel
31/2% Nickel steel
the lowest impact value shall be equal to or more than
15 ft-lbs for a 10 x 10 mm 2 standard Charpy-V specimen
CD
CD 120 Q J — 21/4% Nickel-
steel
at the lowest service temperature specified.

Qualification tests for welding operators must also 100
be c o n d u c t e d before c o m m e n c e m e n t of actual
production welding. Since 100% X-ray radiography is u
80
75
required for the job, it is desirable that procedure and a
CO
p e r f o r m a n c e qualification test specimens are £

60
radiographed as per relevant ASTM standard before
sectioning plates to make test-pieces. c
The three major welding processes used with low- 40
temperature and cryogenic vessels are manual metal-
arc welding, submerged-arc welding and MIG welding. 20
U
Among the materials used for cryogenic service, 9%
Ni steel conforming to ASTM A353 needs special
consideration. The special precautions and procedure -200 -100 0 200
necessary for welding this material are: Temperature, °F
a) Cleaning: Adequate cleaning of joints in 9% Nickel Fig. 19.55: Impact properties of typical as-deposited weld
steel is required to produce X-ray quality welds. Mill metals at cryogenic temperatures. (SMAW = Manual
and flame-cutting scales should be removed from the metal-arc welding: GMAW = TIG or MIG C0 2 ; SAW =
weld area by blasting, grinding, or power brushing; Submerged arc welding).
Austenitic stainless steels are easily weldable with After completion of welding, the welded joints are
matching electrodes or filler wires, but some special radiographed. To ensure highly sensitive quality,
precautions must be taken when they involve cryogenic adequate measures are taken. Films of type-II grade are
plant. TIG welding is ideally suited. The intense, used and exposure selected, so that density of the
concentrated heat of this arc permits high welding rates processed film corresponding to the weldment area is
and the lowest heat input of any of the fusion joining around 2. Sensitivity should range between 1% and
methods. These characteristics of the tungsten arc 1.5%, using DIN/ISO type Image Quality Indicator.
minimise weld cracking and heat effects on corrosion- ASME flat type penetrameters are also used. Location
resistance. Root passes laid down by TIG welding are markers are used on the job, letters are written by Mag
always specified for X-ray quality welds in large, heavy ink, so that cross reference is always possible. In this
walled pipe or vessels. The root may be formed by direct manner, 100% radiographic examination is completed.
fusion of specially prepared edges machined in parent Interpretation is done by viewing the radiographs with
metal, or by fusing a consumable metal ring inserted a high power illuminator. The defects, if revealed, are
between the bottom ends of the pipe sections. TIG rigidly scrutinised against a predetermined standard,
welding of stainless steel is done with DC straight to determine acceptance. If not accepted, the defective
polarity, using 1% or preferably 2% thoriated tungsten location is marked and repaired by grinding and
electrodes, re welding by a qualified welder and procedure. Re-shot
Manual electrodes are widely used in stainless steel is taken and re-examination is performed. The process
welding. In comparison with stabilised electrodes, use continues as long as the defect is not completely
of extra-low carbon g r a d e (C-0.03% m a x i m u m ) removed or not removed to such a degree that it is within
electrodes, is mostly preferred. However, a real source acceptable limit.
of trouble with coated electrodes comes from moisture The vessel, now after completion of piping, is ready
pick-up due to atmospheric humidity. As received, the for tests such as pneumatic, hydraulic and halogen leak
rods are normally packed in polythene bags inside detection. In typical tests, hydraulic test pressure is 1.5
hardboard box to keep them dry; the coating materials times the design pressure and pneumatic test pressure
tend to absorb water depending on time of storage, is 1.25 times the design pressure. Halogen leak detection
temperature, humidity and the coating composition. A test helps to find out exact location of any existing
wet rod gives a sputtery arc and a porous weld deposit; through hole, howsoever minute it may be.
therefore, it is wise to predry the electrodes in a drying For hydraulic test, the vessel is completely filled with
oven at about 100°C. Three or four hours' exposure at water and pressure is raised to the required test pressure.
high humidity can raise moisture content in coating to Peening all over and around the weld seams with
about 4 to 5%, and t h u s cause poor w e l d i n g wooden hammer on the pressurised tank and thorough
performance. In vacuum insulated vessels used for visual inspection for locating any leaking water is
cryogenic services, difficulty has been experienced very performed. Two pressure gauges are used, as is the
often to achieve required degree of vacuum due to the standard method for any such test. The tank is kept
presence of porosity caused by wet electrodes. under pressure for an hour. Initial and final readings on
In MIG welding, axial-spray type metal transfer the indicating gauges must be the same; otherwise, it
gives a clean, spatter-free weld with good penetration becomes obvious that a leakage exists somewhere.
and no problems of slag removal. Submerged-arc Next, pneumatic test is undertaken, which should
welding, both semi-automatic and automatic types with be performed with greater care, since any leak may lead
DC power sources, results in high speed, better control even to fatal results during the test. Pneumatic test
over arc starting and arc length, high deposition rates consists of pressurizing the tank with air to the required
and good penetration. DC straight polarity gives highest pressure, and testing the weldments and joints with soap
deposition rates and best arc stability with least solution. Leak will be indicated by formation of bubbles.
penetration, and is suitable for thinner materials. DC Similar to hydraulic test, two gauges are used and
r^yerse polarity gives greatest penetration and best bead constancy of two readings taken at definite intervals is
shape. Current range of 700-800 Amps, is normally also checked.
used. Economically, SA welding is preferred to other Halogen leak detection test supplements the above
welding processes, mainly because of high deposition two tests. The tank is pressurised, the probe is taken
rates and superb quality. However, a problem of welding near the weldments and joints and moved all over, and
stainless steel by SA process is high heat input. leaks, if any, are detected.
7) Ships always be located so as to be easily accessible to the

The design approach to ships, which are large electrode or welding head. The welder must be able to
floating structures, has to be different from that applied easily see and reach his work to deposit a sound weld.
to land structures, such as buildings and bridges. Ships A few examples and data on good access are given
have to be as light in weight as possible, because any in Fig. 19.56.
avoidable excess of material reduces carrying capacity Where it is impracticable to avoid poor access due
and speed. to other limitations of space, interference of structurals,
Modern ships are all-welded. They are constructed etc., it may be necessary to provide access openings—
in accordance with the requirements laid down by permanent or temporary—which should be clearly
regulatory agencies like Lloyds Register of Shipping indicated in the drawings. Sometimes, lightening holes
(LRS), American Bureau of Shipping (ABS), and Bureau or other openings may serve the purpose. If fitting of a
Veritas. member can be delayed to achieve good access for
M o d e r n s h i p b u i l d i n g is considered a mass welding the assembly, then it should be shown in the
production industry. As much fabrication as possible is drawing in terms of sequence of fabrication.
completed in the shops using flow-line technique. These • Distortion: One of the major problems in ship
fabricated blocks are then assembled at the building construction is distortion of welded structures, due to
berths using high speed welding techniques, to complete the large a m o u n t of w e l d i n g that goes into each
the ship construction in a very short time. Different types weldment. All types of distortions, buckling, etc., are
of welding techniques are adopted for shop floor seen during the fabrication and assembly of the hull
fabrication and at berths. panels—though in varying degrees. The net result is that
Shipyards require large number of stiffened flat the dimensional accuracy is affected, and on the
panels and approximately 12-15% of total weld length shipside, the aesthetic values and on open decks,
in a ship is panel welding. Different techniques have distortions pose maintenance problems.
been developed and adopted in Japanese and European Here the designer has a responsibility for taking
shipyards to increase the panel production rate. Panel necessary precautions at the design stage itself, for the
production lines have been developed using double- p r e v e n t i o n or minimising of distortions, so that
sided welding and single-sided welding, and both dimensional accuracy of the structure and other features
techniques use multi-electrode submerged-arc welding are not hampered. The role of the designer lies in
process. Many of the European shipyards seem to favour considering and duly appreciating the likely causes of
double-sided butt welding technique, whereas the distortion a n d choosing material scantlings and
Japanese shipyards use mostly one-sided welding arrangements accordingly (usually the smaller the
technique. In the construction of a ship, the designer thickness of the plating, the larger is the distortion).
has the responsibility of ensuring minimum cost and Having minimum number of welded joints (by using
m i n i m u m use of steel with m a x i m u m safety. The largest and widest plates within the working capacity
designer must find answers to the question, how of the shop) and minimum weld sizes in a fabrication
efficiently can it be produced? Since welding design is helps in reducing distortion.
very much a part of structural design of ships, the C o n s i d e r a t i o n s h o u l d always be given to the
designer must also ask himself, how efficiently can it possibility of incorporating folded, pressed or forged
be welded? components and rolled sections instead of built-up
The steps commonly adopted to reduce weld-metal sections in the design. An example: introduction of a
deposition and welding costs have been explained in corrugation in a light thickness housing bulkhead to
Chapter 15. There are many other special points which avoid stiffeners, a lot of welding and the resultant
the designer must take into consideration. Some of them distortion. Groups of welds should be balanced about
are dealt with below. the new axis of the section to the extent possible.
• Access for welding: Access for welding is one very The designer should also try to avoid the joint
important aspect, which the designer must bear in mind. geometry that will increase distortion. When unstiffened
It is perhaps difficult for a designer without direct and panels are edge welded to heavier sections, buckling of
real experience to assess as to what might be a good or the panels may be induced. Longitudinal asymmetrical
bad access for welding. This often results in joints weld a r r a n g e m e n t (like a T-section) p r o d u c e s
without proper access or sometimes impossible to weld, longitudinal bowing due to offset line of shrinkage. Such
causing rectifications and renewals at site, joints should aspects are to be taken care of at design stage; otherwise
904
i
18+
16
T
144-
#t
12 +
10+
<0- No floor line
I in Restriction for (a)
8 |
6+
4
. #
#
2+ (a,b)
-I 1 L__ » '
2 4 6 8 10 12 14
d min inches
18
1
16
f Floor line
Restriction
141
121
1 in
10+
el
No floor line
restriction
6+ ^^
0 10 20 30
d max inches
Floor line
Fig. 19.56: Some examples and data on good access
distortion cannot be eliminated in spite of all other distortion, making use of more lapped joints, sniped
precautions on the shop floor. ends to stiffeners, etc.
Since it is not possible to completely eliminate Shrinkage of the panel due to welding is also to be
distortion in a large weldment, the designer should give appreciated and taken care of at the design stage by
m i n i m u m critical connections in areas of likely providing required allowances to achieve overall
dimensional accuracy. This means that the fabrication correct position. Such float should be clearly indicated
should be built oversize, by an amount sufficient to in the design.
cover the shrinkage. It is difficult to estimate the Some general rules of welding sequence derived
shrinkage allowance accurately, except by long practical from long experience h a v e become accepted in
experience. However, some shrinkage allowances that shipyards. They are based on the principle of welding
are found to be reasonably accurate are given in Table towards free ends> or in the direction of the greatest
19.14. Tucking allowances for stiffeners are given in freedom. Special care is necessary when they are to be
Table 19.15. applied at weld junctions.
For intermittently w e l d e d stiffeners: tucking The first basic rule applicable to the weld junction is
allowance is half the above values. illustrated in Fig. 19.57. The extension of this rule to plate
Sometimes, in cases w h e r e only one set of sequence is shown in Fig. 19.58. The concept can be used
dimensions is important, like the overall length of almost without limit as shown in Fig. 19.59. When used
fabrication, it is often possible to leave fitting of one extensively through the midship body, this sequence
member, such as an end plate, until the remainder of results in a marked tendency for the ship to rise from
the fabrication has been w e l d e d and most of the the keel blocks at bow and stern. This tendency may be
shrinkage has taken place. The remaining item of correct relieved by welding a complete band around the ship
dimensions, called float is then fitted and welded in and proceeding forward and aft, limiting the completed
Table 19.14: Shrinkage allowances
Butt welds
Transverse 1.6-3.2 mm for all thicknesses per weld of 60°-V
Longitudinal
Plate over 10 mm t 0.8-1.6 mm in 3 metres of weld
Plate below 10 mm t 1.6-3.2 mm in 3 metres of weld
Fillet welds
Transverse 0.8 mm per weld where leg length £ 3A plate thickness.
Longitudinal 0.8 mm per 3 metres of weld.
Table 19.15: Tucking allowances for stiffeners

For continuously welded stiffeners:
Plate thickness 12.5 mm No allowance
9.5 mm-12.5 mm 0.4 mm per each stiffener
6.5 mm-9.5 mm 0.8 mm per each stiffener
6.5 mm 1.6 mm per each stiffener
Crack here
OR here
OR here
Fig. 19.57: Sequence for welding junctions of two joints: (a) correct sequence; (b) incorrect sequence; (c) probable
result of incorrect sequence
s ( <D I
I B1 © B2
I A © J
Girder
Fig. 19.58: Proper sequence for welding plate: (1) butt

joint complete; (2) inboard seam weld complete; (3)
outboard seam weld complete
Fig. 19.60: Welding sequence for butts in line (with

under-framing)
welded as panels on the ship using the described

sequence. The seams or butt joints can be welded
complete with regard to local sequence, and all the
framing can be welded to the panel, complete within
about 300 mm of all panel edges. This permits many
Fig. 19.59: Welding sequence for bottom shell with welders to be put to work without damage to the
staggered butts (butts in line simpler)
structure. Since there are only a few tie-in welds to be
made, the overall structure shrinks very little, thus
welding by a vertical cross-section, instead of permitting reducing the tendency of the ship to rise at the bow and
the bpttom to progress ahead of the upper deck. The stern. This sequence is illustrated in Fig. 19.61.
use of subassemblies and panels facilitates this method. No doubt, minimisation or prevention of distortion
Plated structures for ships generally involve also depends on various other factors like accuracy in
attachment of frames, stiffeners, bulkheads, etc., to the manufacture, assembly procedure, welding processes
plating. The extension of the general sequence in Fig. used, welding sequence and metallurgy of the plates;
19.58 to include this framing is explained in Fig. 19.60. but good design and satisfactory joint details help a long
The idea is to permit most of the shrinkage to take place way in minimising distortion, rectification of which is a
before the plates are tied together at their edges. On costly matter.
small plates, where this method becomes less practical, The designer must be aware of failures of welded
the complete framing may be welded in steps 3 and 7 of structures in the form of lamellar tearing, brittle fracture
Fig. 19.60 instead of just the unwelded cross-overs. One
advantage of this welding sequence for plates is that
the framing may be welded independently of the butt
and seam welds, as long as the plates are not welded
together. This permits a greater number of welders to
be put to work and reduces the overall time of welding.
In all the above figures, the butts are staggered. When
butts are in line, subassemblies can be w e l d e d
completely in the shop. This permits trimming to exact
size after shrinkage has taken place. These Fig. 19.61: Welding sequence for butts in line: (1) weld
subassemblies are then treated as single plates. If seams in panels P1 and P2; (2) weld all to within 300 mm
subassemblies cannot be made on the ground because of panel edges; (3) treat panels as individual plates in
over-all sequence
of shape or other peculiarities, the sections can be
and fatigue cracking. They have been dealt with in

Chapter 12. Tapered end piece of Superficial weld in half round
In detailed designing, the designer must take half round \
precautions to reduce or prevent avoidable stress
concentrations, which may cause damage to the
structure. Abrupt changes of shape or section and sharp
corners and congregation of welds are to be avoided.
Unnecessary attachments and fittings, specially in areas
of vulnerability cause stress concentrations. D^311 of
Lloyd's Rules stipulates that the shear strake, specially
within 0.5 L amid ships should be kept free of isolated
welded fittings, scupper openings, etc. Similar is the case
in way of breaks of superstructures. This rule applies
equally to M.S. and H.T. steels, but is applied more
rigorously to H.T. steels as a precaution against both
brittle fracture and fatigue fracture. Welded joints
introduce local points of stress concentration, but these
are seriously magnified if HAZ cracking occurs, which
is more critical in case of H.T. steels. Hence, to avoid
HAZ cracks, welding should be limited to the bare
Fig. 19.62: Example of crack propagation
minimum required in critical zones.
from a minor weld
• Structural continuity: Structural continuity is very
important to avoid stress-concentration. Plating at
corners of all openings in strength decks is to be well Lack of proper appreciation of welding of joints in
rounded and free from notches. Welding in the region members like bilge keels—either fully welded type or
of the corners of openings must be carefully detailed. riveted-welded type—caused cracks leading to major
The failure of Encounter Bay design is an example of fractures of shell plating as shown in Fig. 19.63.
the effect of structural discontinuity. Equal attention must be giVen to the welding of such
Minor welds on main strength members are also very minor members as to the main members. Otherwise,
important, since they also are subjected to the same welding minor items to main structural members should
loading as the main members. Cracks emanating from be avoided as far as possible.
minor, seemingly unimportant welds extend into the • Hard spots: Where any unstiffened plating is locally
main structure and cause serious failures. Some attached to a stiffener, a region of intensified stress is
examples are: developed in the plating at the toe of the stiffener, and
A crack from the butt weld of a half round rubbing this is called a hard spot. Toes of brackets, etc., landing
piece extending into the shear strake caused serious on unstiffened panels of plating are examples of this
fracture in shell as shown in Fig. 19.62. (see Fig. 19.64). This effect of hard spot can be offset
Crack at butt in bulb plate

Cracks at the F.B. shell butt
Crack from butt, in F.B.
4*
F.B. welded to shell
Bulb plate riveted to flat bar
Fig. 19.63: Weld failure in bilge keel

Horizontal Gusset
or stiffener
Bottom longitudinal
Fig. 19.64: Hard spot and how to avoid it: (A) Cracking at toe of a bracket due to faulty design;
(B) Modified design prevents cracking
either by landing it on an adjacent stiffener or by adding limitations for these various grades have also been
a doubler plate or special stiffener as shown at (B) in specified with respect to ordinary applications and
the figure. special applications.
Chapter D-33 of Lloyds Rule gives a number of such The regulatory agencies demand that special notch-
detailed requirements relating to detail design and tough steel be used for a few fore- and aft-strakes, which
welding practice. Similar rules are also established by are strategically located around the main hull. These
other Classification Societies. These are based on strakes have taken the place of riveted crack-arrestor
experience of problems that have actually arisen in seams used in the earlier ships.
service, and are mainly intended to avoid incidence of A close study of the hull structure of any commercial
cracking either during welding or afterwards. A proper vessel reveals the following features (see Figs. 19.65 and
understanding and application of these rules helps the 19.66):
designer in developing a clear perspective of welding a) The hull consists of many individual structural
design in shipbuilding. units which are conveniently p r o d u c e d inside a
fabrication shop and later erected and welded u p
HULL CONSTRUCTION outside, in a building berth or dock. Size of these units
The most important component of a ship is the main depends on the cranage and other facilities of the
hull. Regulatory agencies have provided complete shipyard.
specifications of ordinary-strength and higher strength b) The parallel middle body, where shaping and
structural steels to be used for hull construction. ABS, forming is almost absent, forms the prominent part of
for example, specifies seven grades of normal strength the hull. This may be of the order of 60-90% of the total
steels identified as A, B, C, CS, D, E and R. Their tensile structural weight, depending upon the ship's size.
strength ranges between 41 and 50 k g / mm 2 Grade A c) One feature common for the structural units
has no limits on chemistry; grade B is semi-killed; grade located in the parallel middle body is the existence of
C is fully killed and normalised over 35 mm; CS is fully flat panels. Each such panel may contain about three
killed and normalised in all thicknesses. For grades D plates of thickness 10-40 mm (again depending upon
and E, impact requirements are specified at 0 and -10°C the ship size and location on the hull), which are butt
respectively. Higher strength steels u n d e r ABS welded together and stiffened.
specification have tensile strength range of 50-63 k g / In a modern shipyard, automatic welding is used to
mm and are graded as AH33 or 36, DH33 or 36 and the m a x i m u m extent in hull construction. This is
EH33 or 36. The first one is semi-killed, and the latter illustrated and well explained in Fig. 19.67.
two are fully killed and h a v e to meet impact In a typical large and modern shipyard in India,
requirements at -20 and -40°C respectively. Thickness several welding processes are used in combinations. For
Fig. 19.65: The hull structure of a ship
Fig. 19.66: Mid-ship section
the butt welding of hull plates, automatic submerged- when compared to the two-sided welding process
arc welding is commonly used. Two methods are used together with the turning gear, and increased welding
with this process. In the conventional method, welding speeds. As for weld quality, it is quite satisfactory and
is done from both the sides and this involves turning of would meet the requirements of the ship classification
the panel to enable the completion of welding from the societies. The weld impact strength, though slightly
second side. This can be a laborious and time-consuming inferior to that obtained in a two-sided process, is always
operation, as the panels are usually quite large, though well above the minimum requirements, as shown in
it can be streamlined by installing elaborate permanent Table 19.16.
turning fixtures. The other method, called the flux
copper backing (FCB) system, uses a much more Welding of stiffeners to flat panels : These joints involve
convenient one-sided welding technique. This technique straight and long fillet welds, which are deposited by
has been described in detail in Chapter 4. the twin head submerged-arc process, and by the MMA
The obvious advantages of the one-side welding process using heavy iron powder type electrodes of
process are the lower equipment and operational costs longer length. In some shops and especially in Japanese
General view of erection stage
Electroslag welding
Horizontal SA welding
Erection joint of deck assembly

One-side
automatic welding
SA welding
t^
r i i v N vr^
"">w
Consumable nozzle
type electroslag welding
Erection joint of bottom shell assembly
Erection joint of side shell assembly

One-side automatic welding
Semi-automatic welding Electroslag welding
One-side automatic One-side

welding
automatic)
welding
Consumable nozzle
Horizontal
type electroslag SA welding
welding
Fig. 19.67: Automatic welding processes for erection joints of ship's hull
Table 19.16: Comparison of impact strength of weld on

25 mm M.S. plate (typical values)
Tile
Process Impact strength, I | / 4 - 9 mm
kgf • m at 0°C
MMA welding using basic coated electrodes 14.00
Two sides SA process (multi-pass) 10.50
Vertical electrodes welding 7.00
Electroslag welding 5.50
One side (FCB) welding 8.00
shipyards, the gravity welding system and spring-

loaded type auto-contact system are used for depositing
these fillet welds. These methods have been described
in detail in Chapter 3. Fig. 19.68: Typical arrangement for ceramic backing
Erection welding : Erection welding means welding
of sub-assembly panels in the building dock after correct Submarines: Submarine building demands special
alignment. Even here normal submerged-arc welding welding considerations, not only because high-strength
can be used for many erection joints in the flat position, and ultra high-strength steels (mostly in quenched and
which have easy accessibility. In some cases, it may be tempered condition) are used, but also because the
worthwhile to use the electrogas or electroslag welding tolerances in deformation are very critical. Joints are
for the long, straight, vertical butt joints of the side shell. designed for 100% joint efficiency and pressure hull
Also, in certain cases, it may be advantageous to use w e l d s h a v e to be 100% r a d i o g r a p h i c quality.
short length electroslag w e l d i n g (preferably the Understandably, submarines are a highly classified unit
consumable nozzle type) for butt welding between thick in every country, and hence the published information
stiffening members. These sophisticated processes are on the subject is scarce.
being used in Japanese shipyards. Submarines are cylindrical cigar-shaped pressure
Apart from all these, the one major point which vessels, saddled by ballast tanks. In normal loading, the
remains is the necessity of welding the root side of butt submarine will be floating like a surface ship. When
joints. This has to be done after gouging or chipping ballast tanks are flooded, the s u b m a r i n e will be
out the weld root. In the case of a relatively big ship, submerged. In submerged condition, the hydrostatic
this operation may have to be done over a length of pressure in the saddle tanks will be same as that of
thousands of metres in all the three difficult positions, external pressure and hence design feature of these tanks
viz., overhead, horizontal and vertical. For some time is not complicated. The pressure hull, however, is
now, attempts have been made to develop and improve subjected to external hydrostatic pressure, and is a piece
the welding procedure by using suitable backing of most challenging structural design. The pressure hull
materials in the form of ceramic tiles, coated refractory is reinforced by equally spaced ring frames and the
pipes, sand-filled flexible fibreglass ropes, flux-asbestos w h o l e s u b m a r i n e is d i v i d e d into n u m b e r of
foils, etc. All of these are intended for an effective one- compartments by circular transverse bulkheads. The
sided welding system, which would obviate the need design challenge is to assess the collapse depth for a
for gouging out and welding from the root side. Out of combination of parameters, viz., diameter, thickness of
the various types of backing mentioned, the ceramic tile the shell, frame spacing, frame geometry and the
is very widely used, especially by the leading shipyards spacing between the main bulkheads. The redundancy
in Europe (see Fig. 19.68). Here the tiles, each of which being very high, the analysis of elastic instability is very
is about 150 mm long are set in a small channel-shaped complex. Here the designer aims for the lightest possible
metallic case, which is made to press against the bottom pressure hull to get the maximum feasible equipment
of the joint with the help of a suitable clamping a n d w e a p o n pack inside the p r e s s u r e hull. The
arrangement. Joint fit-up need not be of high accuracy. cylindrical shell alone forms almost 70% of the structural
Misalignment of plates can be of the order of 2 mm. It is weight a n d hence one sees the constant race for
possible to weld even curved plates, provided that the developing super high strength weldable steels for
radius of curvature is not too small. submarine construction.
Submarine steels are quenched and heat-treated. Till methods of welding are among the recommended
late 50s, not much details of these steels and their precautions taken in submarine construction.
welding particulars were disclosed. Again, the advent We have discussed the weldability of HY and Q&T
of the deep submersible in ocean exploration field found steels in Chapter 10. In submarine construction, the main
a common ground and some qualitative information is p r o b l e m is to use extra-low h y d r o g e n w e l d i n g
available. These steels are termed differently in different c o n s u m a b l e s to p r e v e n t H A Z cracking. Special
maritime countries: precautions have to be taken to store the electrodes in a
U.S.A.—HY steels, U.K.—Q&T steels, Japan—NS dry place, to predry them thoroughly before use, and to
steels and U.S.S.R.—AK steels. ensure that they do not pick up significant amount of
These are all low-carbon fully-killed alloy steels and moisture from the air while they are being handled on
in some cases low percentage of vanadium and niobium the shop floor.
is added to the steel to get grain refinement. From
known sources, the details of some of these steels are 8) Railways
given in Table 19.17. Welding is the principal process used in the
As mentioned earlier, the collapse depth is estimated fabrication of all railway facilities which include
at the design stage, taking cognisance of various locomotives, passenger coaches, wagons, rail tracks,
Table 19.17: Composition and mechanical properties of different types of high strength steels used for submarines
Type C% Mn% Si% P% S% Ni% Cr% Mo% V% Nb% Y.S. Notch
kg/mm2 ductility,
kgm
U.S.A.
HY-SO 0.25 1.15-1.5 0.15-0.30 0.030 0.030 0.40 — 0.45-0.60 — — 57.0 4.0at-10°C
0.7-1.0 0.002&-
HY-100 0.15-0.2 0.6-1.0 0.4-0.8 0.030 0.030 0.15-0.5 0.4-0.65 0.4-0.6 0.07-0.08 0.0068 69.23 2.0at-40°C
HY-130 0.12 0.6-0.9 0.2-0.35 0.010 0.010 4.75-5.25 0.4-0.7 0.3-0.65 0.05-0.10 — 90.0 6.8at-20°C
U.K.
QT-28 0.17 1.3 0.10 0.030 0.030 0.50 0.30 0.30 44.0 7.8at-20°C
QT-35 0.15 1.2 0.30 0.030 0.030 1.20 1.00 0.50 0.12 0.02 55.0 6.9at-40°C
Japan
NS46 0.17 1.5 0.55 0.030 0.030 0.60 0.30 0.50 0.12 — 46.0 4.8at-20°C
NS-63 0.16 1.3 0.55 0.030 0.030 1.00 0.80 0.44 0.15 V+Nb 63.0 4.0at-20°C
parameters. Submarine pressure hull due to external

pressure will also be encountering circumferential
instability problem. Fabrication of a perfectly circular
pressure hull will not be feasible in practice. Initial out JZ
Increasing diameter
of circularity, as the fabrication error (due to Q. of pressure hull
workmanship and welding distortion) will cause onset -8
of circumferential instability at a lower depth. The
Q.
rough qualitative nature of the curves of collapse
iS
depth versus out of circularity of the pressure hull is
shown in Fig. 19.69. U
The designer therefore lays d o w n the limit of Out of circularity
fabrication error. Apart from taking all the general
measures in the fabrication of true circular ring frames,
various steps have to be taken to minimise the welding Fig. 19.69: Curves to indicate variation of collapse
depth with increase in 'out of circularity' of pressure
distortion. Use of m u l t i w e l d e r s , p o i n t - t o - p o i n t
hull
d o w n h a n d and overhead w e l d i n g and step-back
platform structures and bridges, and in the repair and cast steel trucks having primary coil spring suspension
maintenance of components. on four sets of double equalisers.
The electrical transmission equipment has three main
LOCOMOTIVES groups, namely, rotating machines, control equipment
To understand the role of welding in the production and dynamic brake unit. The rotating equipment
of locomotives, we shall describe the main operations consists of traction generator and motors, auxiliary
in the Diesel Locomotive Works, Varanasi, which is generator and exciter. The basic excitation system
primarily engaged in the manufacture of BG (WDM-2) employed is Type E control, using semiconductor
and MG (YDM-4) diesel electric locomotives to meet devices. Both BG and MG locomotives are provided
the mixed operation r e q u i r e m e n t s of the Indian with dynamic braking units, in addition to the air-brake
Railways. They also turn out WD 56 heavy duty diesel system employed on locomotives.
shunting locomotives for use in steel plants. The m a n u f a c t u r e involves a large n u m b e r of
The BG locomotive is powered by Alco Model 251- separately fabricated and welded sub-assemblies, which
B, 16-cylinder Vee diesel engine rated 2,600 HP at 1,000 are finally assembled together to form a locomotive. The
RPM and the MG and WD56 locomotives by Alco Model techniques adopted for fabrication can be best illustrated
251-D, 6-cylinder In-line diesel engine rated 1,350 HP by describing the m a n u f a c t u r e of the following
at 1,100 RPM. Both models of diesel engines follow a important sub-assemblies:
four-stroke cycle, are t u r b o s u p e r c h a r g e d and a) Engine block
aftercooled and have a bore of 9" and stroke of 10 W.
b) Underframe
The chassis consists of the u n d e r f r a m e ,
superstructure, trucks and auxiliary equipment. The c) Superstructure
underframes are of welded construction, the BG design d) Aluminium pistons.
being of box type and the MG and WD56 being of a) Engine block: The engine block is the principal
conventional I-Beam centre-sill type. The locomotive is structural member of the diesel engine. As shown in
mounted on two six-wheel, three-motor, trimount type, Fig. 19.70, it is a composite weldment with heavy steel
Top deck (sides)
22 mm
Side plates
PC 3,5
Foundation r ,, _m
plate RM.~&£J^
Rib L.H.
Fig. 19.70: Sectional view of the engine block; (M = Manual metal-arc: SA= Submerged-arc)
forgings, which serve as the crankshaft housings and After these welding operations have been completed,
hence become the backbone of the structure. The decks the fabrication acquires the final shape of an engine
and the foundation plates are built round the steel block. The engine block is then stress-relieved at 620-
forgings to form the box section, which provides the 650°C in a furnace capable of accommodating two blocks
maximum rigidity. at a time. It is finally machined to very close tolerances
The steel material of the engine block has not only and as such, it is necessary that all stresses developed
to meet the requirements of weldability and physical during the fabrication stages are completely relieved
properties, but it should also be free from defects such before this final step. This would ensure a longer life in
as laminations. The plates and billets are, therefore, service without any distortion, which would normally
subjected to ultrasonic test before use. The plates have result on account of very high alternating stresses that
the thickness range of 6-80 mm and conform to IS:2062. the engine block is subjected to, during its service. The
The principal structural members of the engine block engine block is then shot blasted, hydrostatically tested
are profile-cut from plates on the automatic flame to a pressure of 75 psi and then handed over for layout
cutting machines. Splice is machined out of billet 5' x 7', and subsequent machining operations.
conforming to IS: 1875. The spline being the most highly b) Under frame-. The BG underframe is designed to
stressed item, it is made out of one piece instead of being withstand a maximum static squeeze test load of 400
fabricated. Considering the large size of the engine block tons without signs of any permanent distortion. The
(4,116 mm length, 1,292 mm width and 933 mm height), underframe load transfer members are so designed as
the welded joints must necessarily have accurate fit-up, to transfer the vertical load on to each trimount bogie,
particularly at the fillet welds. As such, the plates are in the ratio of 60% through the centre pivot and 20%
subjected to a straightening operation on a 200-ton through either of the two loading pads. The underframe
hydraulic press before use. is welded, utilising plates ranging in thickness from 6.3
The engine block fabrication employs both manual to 25 mm. The main structure of the underframe is made
and automatic welding processes. The long continuous out of ASTM-A-441 high tensile steel, having 1.5%
welds which run through the entire length are mostly vanadium, 1.5% manganese and 0.22% carbon. The
done by automatic submerged-arc welding and the short underframe is about 15.6 m long and is made in three
runs are made by the manual metal-arc process. sections—the two end sections and the centre section.
The foundation plates, forged saddles and the spline The centre section also forms the fuel tank of the
are clamped together on a cast iron fixture mounted on locomotive, having a capacity of 5,000 litres and the
roll-over rings. The saddles and foundation plates are fabrication has to be leak-proof. The manufacture
given a pre-offset with 3.2 m m shim in the centre involves special welding skills and use of fixtures and
tapering down to 0 at the ends. This offset is needed to positioners.
neutralise the distortion effects of welding. These are In the fabrication of the end sections and the centre
then tack w e l d e d with E6020 electrodes in the section, apart from manual welding, semi-automatic
downhand position. These welded joints are subjected flux-cored arc welding without gas shield is used. After
to 100% radiographic examination. Any defects detected the above three sections are fabricated separately, these
during this examination are gouged out and rewelded are welded together employing butt joints, the welds of
to give perfectly defect-free joints. This is necessary which must comply with radiographic standards. The
because the foundation plates, saddles and spline sequence of welding these joints plays an important part
together form the backbone of an engine block and the in inducing the required camber of the underframe. The
nucleus, around which the entire block is fabricated. vertical joints are welded first. The ends are clamped
The inside walls, outside walls, middle decks, top firmly on to the surface plate and the downhand bottom
decks, side plates, fuel-shelf compartments, cam-bearing plate joints are then w e l d e d . Two welders
supports and end plates are welded in sequence around simultaneously attend to both the downhand joints. The
the above nucleus by manual as well as submerged-arc camber is obtained by inducing a calculated distortion
processes. In between the above sequence of fabrication, in the structure by means of welding heat. This is
the critical joints are subjected to r a d i o g r a p h i c accomplished in two stages, i.e., on the surface plate
examination. The weldment is also taken to the planing and in the upright position, and is then checked at the
machine to maintain the overall dimensions and for final check stand. The final camber is 12.7 to 19 mm.
proper edge preparation to ensure a good quality c) Superstructure: The superstructure comprises of
welded joint. short hood, driver's cab, contactor compartment, hood
over engine and radiator compartment. All these are floor pans are, in turn, resistance-welded to the centre
light structures fabricated from 1.6 and 3.15 mm steel sill. The side sills enclose the beams, floor pans and end
sheets. Apart from manual welding, resistance spot underframes. Following structural assembly, the floor
welding is also used on the smaller components. is inverted for installation of underfloor equipment (see
d) Aluminium pistons : Amongst the various types of Fig. 19.71).
main pistons used in the p o w e r packs of diesel Side assembly: The sideframe is a semi-monocoque
locomotives, there is one type in which the body is made type structure with shear panels at the ends and is
of forged aluminium alloy with a ring carrier of forged constructed with a 31.8 mm camber in 25.5 m. It is a
aluminium with Ni-resist rings shrunk fit on it. These completely resistance welded unit with the exception
pistons get worn out in use and they are reclaimed at of the butt joints between the deadlight panels, which
DLW by rebanding. On receipt of the old pistons, the are TIG welded. The outer exposed surfaces below the
ring carriers are removed and after thorotigh cleaning deadlight panels may be of light gauge corrugation-type
and treatment, a new ring carrier is shrunk fit on the construction, resistance welded directly to the structure.
crown of the piston and then welded by MIG welding Fluted panels may be attached in the final assembly. The
technique. In the initial stages, there used to be rejections
MIG process is used to attach heavy stainless steel tie-
due to excessive porosity in the welds. These problems in members at the floor level in door areas.
were overcome by ensuring intensive cleaning and End assembly: The end frame consists of end sheets,
temperature control. angles for connection to roof and side frames and
collision post assemblies, which connect the longitudinal
RAILCOACHES roof members (purlins) and the end underframe. This
Lightweight all-welded coaches are manufactured assembly is resistance-welded.
at the Integral Coach Factory near Madras and in several Roof assembly: Zee-shaped, stretch-formed roof
other units. The main components of a coach are bogies, carlines, h a t - s h a p e d roof p u r l i n s a n d side plate
bogie frame and body of the coach. For bogie and bogie assemblies are put together and welded in a cambered
frame fabrication, manual metal-arc and submerged-arc fixture. Corrugated mat assemblies, running the entire
processes are used. For welding the side walls and roof roof length, are then welded to this frame. Resistance
which are made from 1.6 to 2.5 mm steel sheets, foil welding is employed throughout. This completed roof
butt seam welding is used, which utilises a thin foil 0.2 is inverted for installation of piping, wiring, insulation,
mm x 4 mm wide. Resistance spot welding is also used air ducts and air conditioning. Ceiling support members
for fixing stiffeners and other items. Aluminium and are also resistance-welded in place at this time.
aluminium alloy components which are also used on Final assembly: Structural general assembly is divided
the coach are welded by the TIG process. into two stations—assembly of sides and ends to the
In the U.S.A. and several western countries, stainless floor and installation of the roof. Resistance welding
steel passenger coaches are used. The grades of stainless with portable tools is the major joining method, with
steel usually employed are AISI Types 430,201 and 301. manual metal-arc welding used to plug weld the
Type 430 is a corrosion and heat-resisting 17% chromium collision post to the end underframe, using stainless steel
steel, and is used in non-structural areas. Type 201 is an electrode E-308-XX.
austenitic chromium-nickel-manganese steel and is used Interior trim, floor, seats, windows, partitions and
in the cold rolled or annealed condition. Type 301 is an hardware are added following water testing. Arc-spot
austenitic chromium-nickel steel capable of obtaining w e l d i n g has been used to weld stainless steel
high strength and ductility by moderate or severe cold corrugation floors to the floor pans when this type of
working. Both these stainless steels are used in structural structure is specified. Interior aluminium panels and
areas. Fabrication procedure for these stainless steel sub-assemblies are joined by resistance welding and TIG
coaches is described below. welding before installation in the coach.
Floor assembly: The floor assembly consists of two end Bogie welding: The following description refers to
underframe assemblies of high strength low-alloy steel, bogie welding in India. Passenger coaches, in particular
joined by manual and semi-automatic arc-welding Electrical Multiple Unit (EMU) coaches, which develop
processes. These units are plug-welded or riveted or fast acceleration and run at high speeds, d e m a n d
both to the centre sill, which is made up of stainless designs which can be translated to fabrication by
steel drawbench sections, which are resistance-welded welding alone. The bogies of the EMU coaches are the
together. The transverse stainless steel floor beams and primary structures which bear the thrashing of high-
Bottom of
^floor pans
Floor surface
Sub floor
Continuous s.s,
floor beams
Fig. 19.71: Construction details of S.S. coach body. Coach roof (upper left), centre sill cross-section (upper right),
side frame assembly (lower right), floor construction (lower left)
intensity dynamic loads. The welding technology assembly also utilises suitable jigs and fixtures for giving
a d o p t e d on the bogie frame takes care to avoid the final shape to the bogie frame. These jigs and fixtures
premature and sudden fatigue and impact loading arc very often mounted on positioners and/or rotators,
failure, thereby ensuring passenger safety with the ultimate object of attaining higher productivity
The steel used for this type of fabrication is a carbon- through ease of manipulation in downhand welding.
manganese steel with 0.22% maximum carbon, to obtain The typical welding involved in a bogie frame may
good weldability (IS:2062). The bogie is so designed that be grouped joint wise as shown in Table 19.18.
at no point the cross-section of the material exceeds 28 Generally, the flowing processes of welding are
mm. adopted in the fabrication of bogie in the following
A few quality steel cast components are also utilised locations:
in the bogie, particularly on account of their complicated a) Manual metal-arc welding—assembly of bogie
shapes. Other parts of the bogie made from premium frame
qualities of alloy steels and other metals are springs, b) Automatic submerged-arc welding—bolster sub-
bearings, axle-bogie cover, some suspension and brake assembly, side frame sub-assembly, etc.
rigging components and various pins, bushes and c) Semi-automatic submerged-arc welding—head
fasteners. stock sub-assembly
The bogie is composed of five major sub-assemblies, d) Gas-shielded metal arc welding (C0 2 )—spring
namely, (a) head stock, (b) side frames, (c) cross-bars, plank sub-assembly, tube assembly etc.
(d) bolster, and (e) spring planks. In the welding of EMU bogie frames and similar
Jigs and fixtures are used to fabricate these sub- fabrications subject to dynamic loading and fatigue,
assemblies individually, in order to facilitate notches, cracks, inclusions, unfused zones, etc., which
maintenance of dimensional tolerance. The final act as stress-raisers must be scrupulously avoided. These
Table 19.18: Weld details of a bogie frame
SI Joint details Locations
1. Side frame sub-assy.

^ ^1 [1.8
Joint in web.
2. Frame assy.
Head-stock flange
to side frame flange
Frame assy.
1 cross-bar to
T^^^^^T side frame
Bolster sub-assy.
Joint in web.
Bolster sub-assy
Joint in sleeve
Side frame sub-assy
» mm\\\\>K» -^
Head-stock sub-assy.
ESSW- 8 ~"*~
Bolster sub-assy
KTCTC AvmvvvvvwMii ■*-,

9. Spring plank sub-assy
EE
--H *~ io -►! K- )(/T
10. Cross-bar sub-assy
stress-raisers cause local stress much higher than the At the final assembly of the bogie frame, most of the
average values in the members joined. It is at such butt joints are comparatively short in length. Therefore,
points, that the fatigue failure starts. there always remains a possibility of presence of defects
at the ends of such joints. The defects may be incomplete wagon has to carry very heavy floor loads, additional
wTelding, a burnt notch or porous craters, or even end longitudinal floor stringers are placed between centre
cracks. All these defects are stress-raisers, and since they and side sills.
are located at the end of the welds, they may easily lead The following paragraphs describe the fabrication
to weld failures. To prevent the occurrence of such of wagons in the U.S.A. The centre sill is the largest and
defects, it is a common practice to use run-on and run- most important component of the underframe. There
off plates. After completion of the welding, the plates are several designs of the centre sill some .of which are
are broken off and the area is ground smooth. shown in Fig. 19.73. The most common design consists
Fillet welds are mostly deposited by SAW and MIG/ of two Z bars welded toe-to-toe. Normally, these sections
C 0 2 processes. Run-on and run-off plates are also used are clamped in a fixture, using a preset camber for
here. A suitable sequence of welding is followed to avoid distortion allowance. Large hydraulic cylinders may be
any irreparable distortion. used to obtain this precamber. Submerged-arc welding
Every bogie is subjected to in-process quality control, is done from one side only, using a flux backing or a
as well as final inspection. These are carried out in the copper backing. This section may be welded with a
following stages: single SAW head, using a large diameter electrode;
a) Inspection of weld fit-up however, where poor fit up is unavoidable, two welding
b) Visual inspection on weldments heads may be used.
c) Mechanical tests on run-on and run-off plates, Typical body bolsters (two per wagon) are a welded
taken at random box section. The top cover plate is continuous for the
d) Ultrasonic tests on typical weldments, taken at full width of the wagon, and is attached to the top of
random the centre sill with intermittent fillet welds, slot welds
e) Radiographic test on typical weld joints. The or mechanical fasteners. The ends of the top cover plate
radiographic acceptance standard is IIW blue category often lap under the side sill flanges, and are attached
Some of the highly stressed joints which require 100% with fillet welds. One end of the bottom cover plate laps
radiography have been shown in Fig. 19.72. the bottom flange of the side sill and in some designs,
n.
WAGONS
The main components of a railway wagon are bogies,
underframe, and the body. The bogies are somewhat
similar to those of a railcoach. The underframe is an all-
welded skeleton structure, usually consisting of centre
Two 'HCS' section welded toe to toe
sill, side sills, end sills, body bolsters and cross members
that connect the centre sill to the side sills. When the . Coverlfc
[►Structural
channel [
^Continuous
flat bar
Alternate welded section
\tf-beams
Two 'W1 beams welded flanqe to flange
I beam
Two
HCS
section
Two 'HCS' section VF reinforcing * W^' beam

Fig. 19.72: Highly stressed joints (X) which require
100% radiographic tests Fig. 19.73: Standard designs of centre sill in the U.S.
an access hole permits the welds inside of the box section In the U.S.A., PS-I types of Pullman wagons were
to be tied in. The other end is welded to the bottom of being mass produced in the 60's by adopting a planned
the centre sill flange. Various designs of welded bolsters flow of production. The two production lines with a total
are shown in Fig. 19.74. All the parts of the bolster are of 1,800 people, including 400 welders, could produce
loaded in a fixture, then two automatic welding heads, 54 box-type wagons of 12 m length or 45 box-type
which ride on a welding gantry, automatically weld two wagons of 15 m length per day.
seams simultaneously. After the first side is welded, the The PS-1 had a load-carrying capacity of 50 tons,
fixture is rotated and the two seams on the opposite weighed about 21 tons, and cost approximately Rs.
side are welded. A welding gantry may serve two 59,000. The wagon had four axles and two bogies. The
fixtures, one fixture being loaded, while the other one material used for the underframe and the shell was,
is being welded. besides a few rolled products, mostly plate of semi-killed
To construct the side walls of the wagon body, a steel with a maximum of 0.07%C. The thickness for the
framework consisting of a series of vertical posts, end underframe ranged between 5 and 15 mm and the walls
posts, and top and bottom chord members is made. The and roof between 2 and 4 mm. The latter wagons had a
framing members are assembled and tack welded in a cushion underframe, called Hydroframe 60, which
h o r i z o n t a l location fixture. Sheets or plates of caused almost no deflection to the goods in the wagon,
appropriate sizes and thickness are then tack welded even w h e n there w a s a s u d d e n acceleration or
between the framework members. The sheets and plates retardation of speed up to 10 miles per hour. Hand-
are then welded by the automatic submerged-arc or welding, semi-automatic submerged-arc and fully-
M I G / C 0 2 process. automatic submerged-arc machines were in use. Jigs and
Wagon ends are often made from one or two pieces rotators enabled quick positioning and d o w n h a n d
of steel, with transverse corrugations hot-pressed in. The welding. The underframe was assembled and welded
vertical edges are usually curved 90° for connection to by supporting between two Aranson Rotators.The
the side assemblies. whole underframe could be mounted, assembled,
welded and withdrawn in 20-30 min. The assembly flow
line is explained in Fig. 19.75.
RAIL WELDING
Railway tracks with welded joints possess distinct
advantages over those having fish-plated joints, as
welded joints offer better riding comfort with reduced
Sect A-A joint maintenance. It minimises considerably rail end
failure—a predominant cause of rail renewal. Welded
a
tracks are also less susceptible to sabotage.
In India, rail joints are welded in the shop by the
flash butt welding or gas-pressure welding process, and
in situ on the track by the thermit process. This thermit
Sect B-B process is simple and does not require expensive
e q u i p m e n t . The w e l d - m e t a l is p r o d u c e d by the
exothermic reaction between aluminium and iron oxide,
One formed ft. and one flat R. carried out in a refractory lined crucible. Short welded
panels made either by flash butt or by gas-pressure
welding processes are further welded insitu by the
thermit welding process to make continuous-welded
rails. In the conventional thermit welding process, the
time taken for completion of one joint is about 45 min,
/w -<T yp- of which the preheating time is about 35 min. In

Germany, a process known as SKV, using only 1.5 min
Four flat plates preheating of the mould has been developed. With this
process, the weld can be completed in less than 10 min
Fig. 19.74: Standard designs for bolsters
and rail welding on the tracks can be carried out without
Fig. 19.75: Production assembly track for PS-1 Pullman wagons
interruption, while the trains are running intermittently which they are immediately initiated electrically, either
on the track. The Indian Railways are keen to develop as a single large group, or successively in smaller groups,
and adopt this improved process. to produce the required spot welds.
In the assembly of component sheet-metal parts,
9) Automotive Industry where the physical size and shape make it prohibitive
The automotive industry relies heavily on welding to to move the workpiece relative to the welding machine
mass produce cars, trucks and other automotive (such as with completely assembled car bodies), it
products. The high production rates encountered in the becomes most practical to process these assemblies
automotive industry have led to a high degree of using portable resistance spot welding guns (see Fig.
automation and associated special equipment. In 19.76). These guns come in a variety of physical shapes
the case of automobile body components, multispot and sizes; the shape and throat area are dictated by the
welding machines make as many as 100 to 125 spot geometry of the component parts that the gun must
welds simultaneously, at production rates up to 600 accommodate. The force and current capacity are a
components per hour. Other associated components, factor of the metal thickness combinations, that are to
smaller in size and requiring fewer welds, are produced be welded. The air-operated guns are associated with
at typical rates of 800 per hour. Obviously, the use of the lower force requirements of the lighter metal gauges,
specialised equipment for either resistance welding or while the hydraulic-operated guns are applied to
automatic fusion w e l d i n g requires a high initial heavier metal, higher force applications. Current is
investment and must be justified by the volume to be s u p p l i e d by a transformer rated to s u p p l y the
produced. In cases of lower volume production, more m a g n i t u d e of current required at the duty cycle
versatile manually operated standardised equipment is intended, and transmitted to the gun by means of a low-
used. reactance water-cooled secondary cable. The guns are
Economics is an important factor in selecting the suspended on balancers and hanger devices to facilitate
metal-joining process for the product. In the case of high- gun handling and manipulation. Welding guns are
volume production, a fraction of a cent saving per part, sequenced by NEMA type 3B timing devices, which
in either consumable material or labour can add up to a meter squeeze, weld, hold and off-time increments.
considerable amount of money over a year's production. Welding rates will vary with the metal thickness to be
A typical automobile body is fabricated from sheet steel joined with typical rates being as high as 200 spots per
components or sub-assemblies, which for the most part, minute for light sheets (0.9 mm) and 60 spots per minute
are spot welded together. An automobile body may for heavy sheets (2.5 mm and above).
contain as many as 6,000 structural spot welds, of which Radiators are either brazed or resistance seam
half may be made by means of multiple spot welding welded. Exhaust and tail pipes are resistance upset seam
machines, often referred to as press welding. welded in one continuous operation. Exhaust mufflers,
Press welding is a multispot welding method, made of light gauge sheet, are rolled and stamped to
wherein any number of welding heads or welding guns shape. The side seams are usually spot or seam welded.
containing spot welding electrodes simultaneously End joints are rolled as well as seam welded. Baffles
engage and apply pressure to the workpieces, after and other interior parts are spot welded in place.
Fig. 19.76: Typical installation of an air-actuated portable resistance spot weld gun
In the manufacture of pressed steel rear axle welding or automatic submerged-arc welding in their
assembly, automatic SA welding is used to join the four assembly.
seams. The flanged bearing ends, formed separately, are Truck b o d i e s and freight-hauling trailers are
automatic arc welded on to the housing. The cover is economically fabricated by the judicious selection of
also welded to the housing. The various suspension MMA welding, M I G / C 0 2 welding and automatic SA
brackets that make up the assembly may be projection welding.
welded, or arc welded to the housing. The brackets In the manufacture of farm tractors, the use of
themselves are often welded sub-assemblies. The fuel welding has become increasingly important. The
tank is often soldered and welded by the resistance-seam processes in use include spot welding, seam welding,
welding process, producing an economical, leak-proof gas welding, MMA and M I G / C 0 2 welding.
unit.
Propeller and drive shafts are commonly made from 10) Machine Building Industry
resistance welded tubing, with the end forgings arc Welding is used with advantage in the manufacture
welded by the submerged-arc or M I G / C 0 2 process. of machinery used for various industrial purposes.
Passenger car and truck wheels utilise resistance spot Examples are: road-making machines, earth-moving
machines, steel plant machinery, machine tools and As a remarkable example of steel plant equipment,
production e q u i p m e n t , m i n i n g machinery, farm one may mention the fully welded 1,300 tons capacity
machinery, materials handling machinery, crushing mixer designed and fabricated by Heavy Machine
machinery, sugar and cement making machinery, etc. Building Plant, Ranchi, for the Bhilai Steel Project. The
A good example of earth-moving machinery is the mixer is meant for storing the molten iron from the blast
excavator, which mainly operates in quarries, iron ore furnace, until it is taken to the open hearth furnace or
mines and dam sites where the service conditions are converter for making steel. The main fabrication is the
most rigorous. All the important components of the shell assembly consisting of the shell, two tyres and two
excavator including the revolving frame, car body, ends. These heavy weldments were completed with
shovel attachments such as the dipper handle and crane M I G / C 0 2 , automatic submerged-arc and electroslag
boom are fabricated by welding. welding processes. Other examples of steel plant
Other equally interesting examples are the crawler machinery are: (a) 90 ton ladles and ladle cars fabricated
tractor, rear dumper and scraper. Examples of critical by TELCO for TISCO. The former involved welding of
welded parts in these machines are: the rear dumper 28-40 mm thick shells, while the latter called for the
body of quenched and tempered steel; lift arm assembly welding of 127 mm thick slab frames, (b) LD convertor
of the crawler loader, which involves the judicious and its trunnion ring.
combination of castings and plates. India's first 300 ton convertor has been fabricated
In the building of heavy machinery for cement, by welding at Bokaro Steel Plant. The convertor shell
paper, sugar, material handling, etc., composite welded having a diameter of 8,614 and a height of 10,770 mm
assemblies of rolled, cast and forged components help has a shell plate thickness varying from 60 to 100 mm.
to meet the complex service requirements, like dynamic The entire shell is supported on a trunnion ring,
and fatigue stresses due to cyclic or shock load, abrasion, comprising a box section made of plates of 60 mm (web)
erosion and corrosion problems which lead to heavy thickness and 100 mm (flange) thickness. The depth of
wear and tear. Examples are: heavy fabricated gear the section is 2.5 m.
wheel, rotary table for bucket wheel excavator of 2,000
tons/shift, and stainless steel centrifugal basket.
Welding Handbook, Sect. 5, 6th Ed., published by the American Welding Society.
AWS Dl.1-1983, Structural Welding Code—Steel by the American Welding Society.
IS 8031976, Code of Practice for Design, Fabrication and Erection of Vertical Mild Steel Cylindrical Welded Oil Storage Tanks.
IS 2825-1969, Code for Unfired Pressure Vessels.
ASME Boiler and Pressure Vessel Code, Sect. IX, by the American Society of Mechanical Engineers. Indian Boiler Regulations.
Pipe Welding Procedures by H. Rampaul, 1973, by Industrial Press Inc., New York.
Pipe Welding Techniques by Griffin, Roden & Briggs, 1978, by Van Nostrand Reinhold Co., New York.
Proceedings of 1980 Conference on Pipeline Welding and Inspection, by the American Welding Society.
API 1104, Standard for Welding Pipelines and Related Facilities.
Proceedings of Seminar on Fabrication of Pressure Vessels for Chemical, Petrochemical and Fertiliser Industries, 1978, by HW, Bombay
Branch.
Proceedings of Seminar on Welding for Construction and Maintenance of Power Plants, 1979, by IIW, Bombay Branch.
Proceedings of Seminar on Fabrication of Boilers and Pressure Vessels, lanuary 1973, by IIW, Trichy Branch,
Productivity in Penstock Fabrication by Awtar Singh, IWJ, luly 1973.
Welding of 80 kg/mm2 H.E. T. Welten-80 Steel for Kalinadi Project by B. Ramaswamy, IWJ, July 1979.
Application of Gas-Shielded Arc Welding and Submerged-Arc Welding for Fabrication of Nuclear Vessels, by M.L. Gehani & W.D.
Rodrigues, IIW National Seminar, 1976.
Development of Special Welding Procedures for Narora Atomic Power Station by Dr. Srinivasan, S.L. Kati, et a l , IWJ, April 1979.
Automatic Welding of Small Diameter S.S. Tubes in a Nuclear Station by S. Panchanathan, IIW, Madras Branch Seminar, August
1976.
Selection and Evolution of Specification for 16-8-2, 17-10-2 and 308 Welding Consumables for Fast Breeder Reactor Applications by
S.C. Chetal, et a l , IIW, Madras Branch Seminar, January 1979.
Fabrication of Components for Fast Breeder Reactor by S.N. Krishna & B.N. Murthy, IIW, Trichy Branch Seminar, January 1978.
Nuclear Power Generation—An Introduction, WIRR, October 1980.

Application ofEB Welding by Partial Vacuum Technique for Tube-to-Tube Sheet Joints in the Munufacture of Nuclear Vessels by S.
Challappa, R.S. Alekal, et al., IIW, Bombay Branch Seminar, December 1976.
Welding Applications in the Chemical and Petrochemical Industry by J.R. Prasher & M.S. Kamath, IIW, Bombay Branch Seminar,
December 1976.
Oil Refinery Process Plants, Met. Constr. November & December 1977.
Materials and Welding Consumables Applied to Constructions and Maintenance of Fertiliser Plant by K. Manickam, IIW, Trichy
Branch Seminar. January 1977.
A Study of Material Selection and Fabrication for Urea Service by S.R. Jana, IIW, Bombay Branch Seminar, December 1976.
Welding of Superalloys for Fertiliser Industry by R.C. Choudhary & S.C. Ghosh. IIW, Madras Branch Seminar, January 1979.
a) Scope of Welding Technology in the Manufacture of Vessels for Cryogenic Services by S.K. Burman, IWJ, May 1975.
b) Welding ofS.S. in Cryogenic Applications by Banerjee & Pradhan, IWJ, October 1979.
c) Welding Quality Control of the Fabricated Process Equipment for Low Temperature Service by N.K. Roy, et al., IIW National
Meet, November 1977.
Manufacture of Clad Vessels for Chemical and Petrochemical Process Equipment by W.D. Rodrigues, IIW, Bombay Branch Seminar,
December 1976.
Asia's Largest Nitric Acid Plant at Trombay by N. Srinivasan, IIW, Madras Branch Seminar, January 1979.
Tower Welding of Low Temperature Applications at GNFC, by N. Srinivasan, IIW National Welding Seminar, 1980.
Some Aspect of Welding in Modern Ship Construction by V.P. Nair, IWJ, October 1977.
Quality Aspects of Welding in Naval Construction by Cdr. M.K. Mukherjee, IIW National Welding Meet, November 1977,
Vizag.
Welding Productivity in Shipbuilding by Cdr. M.K. Mukherjee, Workshop on Welding in Shipbuilding by IIW & AIEI. November
1978.
Some Aspects of Quality Requirements in Welding Ships Main Hull Structure by K. Balamanohar. IIW National Welding Meet,
November 1977.
Proceedings of the Conference on Structural Design and Fabrication in Shipbuilding, 1976, by Welding Institute.
Contribution of Welding to the Indian Railways by A.K. Bhattacharya, et al., IIW National Seminar, 1977, Delhi.
Welding Processes Adopted for the Manufacture of Diesel Electric Locomotives by A. Badhwar, IWJ, October 1978.
Fabrication of 1300 ton mixer for Steel Plant by A. Khan & S.K. Roy, IWJ, October 1978.
Experiences in Heavy Fabrications in TELCO by Vogl & Jain, IWJ, October 1978.
Assembly Welding of India's First 300-ton Convertor by Anil Pandya, IIW National Welding Seminar, 1980, Calcutta.
CHAPTER 20
Preheat and
Postweld Heat
Treatment
PREHEATING is heating the joint and its adjacent ma- A) Carbon Equivalent = C + M n / 6 + (Cr + Mo + V)/
terial prior to joining. It is used in welding Carbon Steels 5 + (Cu + Ni)/15
and Low-Alloy Steels to (a) slow down the cooling rate
so that hardness of the weld-metal and the heat-affected B) Heat Input (KJ/mm) = {(Vxl)/S}x0.06
zone (HAZ) can be reduced; (b) eliminate Hydrogen- C) Combined Joint Thickness (mm) = t1 + t2 + t3 + t4,
induced cracking (HIC) in these areas and (c) improve where "t" is section thickness as per the sketch
CVN-Notch toughness. Preheating is also used in the below
welding of metals, which are high conductors of heat
such as Copper and Aluminium and their alloys, and D) Weldability Reference Number can be taken from
in the repair welding of Cast Iron parts. the table below, corresponding to Carbon Equiva-
Preheat temperature and method can be chosen lent (CE)
based on (a) Base metal chemistry, (b) Section thickness, E) With the help of graph below, using Combined
(c) Weld metal Hydrogen content, (d) Constraints on Joint Thickness and Weldability Reference Num-
the joint and (e) Ambient temperature. ber, suitable "Joint Weldability Index" Curve can
be identified.
METHOD TO DETERMINE THE PREHEAT TEM-
PERATURE F) Having identified the suitable "Joint Weldability
To determine the preheat temperature steps below Index" Curve and Heat Input (KJ/mm), with the
are to be followed, based on the material data and WPS use of following charts "Preheat Temperature"
parameters can be determined.
EH t,
i-t __-t }
, i
t
u For welds between t, and t,/t3,
ignore ti unless it is already
rto
welded to yt 3 .
11- average thicknee over 75mm
&m t2
For welds between t3 and t,,
ignore t i unless it is already
welded to L
Carbon Equvalent Weldability Carbon Equivalent Welability

Reference No. Reference No.
below 0.30 1 0.55 to below 0.60 7

<0.30 to below 0.35 2 0.60 to below 0.65 8
0.35 to below 0.40 3 0.65 to below 0.70 9
0.40 to below 0.45 4 0.70 to below 0.75 10
0.45 to below 0.50 5 0.75 to below 0.80 11
0.50 to below 0.55 6 0.80 to below 0.85 12
Joint
Weldability
Weldability Q Index
Reference
Number 7
0 10 20 30 40 50 60 70 80 90 100 110120 and over

Combined Joint thickness (mm)
A simple method of preheating a job, such as a cast iron possible to stress-relieve the job at the same time. The
cylinder head, is to construct a temporary brick furnace temperature of the job surface is measured with a
around it, using charcoal or coke mixed with dry coloured crayon such as Tempilstik. These crayons are
cowdung as a source of heat (see Fig. 20.1). The area to available as a set with fusion temperatures as desired,
be welded is kept exposed, so that the welder can com- say 300°C, 350°C, etc., indicated on each crayon. A mark
plete the welding in situ. With such a furnace it is often made with a 350°C crayon on the job will fuse as soon
PREHEAT AND POSTWELD HEAT TREATMENT 773
250
Joint
200- weldaility
O Index
CD
v_
•4—•
cc
v_
CD
CL
E
0)
to
CD
.C
CD
a!
Welding energy input kilojule/mm of deposit
"^^^r L ^ L . Joint
~ —"~-——-JS Weldaility
o \ ^ ^ ~ \ _ " - - - Index
\ ~ ^ \^ H ^^ --4
\ \ G ^^--^^^ ~-~--~-I
\ ^ \ F
\ E " \
\
^^J
D \ ^
^\^_
v \ >
\ B X \ ^
-v A \ V
\. \_
-\A \
1 2 3 4
W e l d i n g e n e r g y i n p u t k i l o j u l e / m m of d e p o s i t
as this temperature has been attained. Where such cray-

ons or any other means of measuring temperature are
not available, the temperatures are indicated by the
colour of the heated steel as shown in Table 20.1.
Heavy castings and forgings are often preheated in
the regular heat-treating and stress-relieving furnaces
for repair welding. When they are taken out for welding,
there is bound to be a temperature drop due to air
cooling. This can be minimised by covering the job,
except for the portion to be welded, with asbestos paper
Fig. 20.1: Temporary brick-furnace for repair work
or blankets. Allowance may be m a d e for the
Table 20.1: Temperatures indicated by the colour of steel when heated

Colour Approximate Temperatures*
(°C) (°F)
Brilliant-white 1,300 2,350
White 1,200 2,200
Yellow-white 1,100 2,000
Orange-yellow 1,050 1,900
Orange-red 1,000 1,800
Bright cherry-red 850 1,600
Cherry-red 750 1,400
Dull cherry-red 650 1,200
Dark-red 600 1,100
Faint-red in twilight 500 950
Faint-red in dark 400 750
Pale-blue 300 550
Light-straw 230 450
*The temperatures given are not necessarily exact equivalent; but are given to the nearest 50° for the sake of simplicity.
temperature drop, by heating the job to a temperature

about 50° more than specified before withdrawing it.
Otherwise, supplementary heating with oxyacetylene
torches may be continued until the welding has been
completed.
Circumferential and longitudinal joints of pipes may
be locally preheated by means of suitably-shaped
burners made out of 12 to 38 mm diameter pipe drilled
at intervals with 1.6 to 0.8 mm diameter holes along
one axis. A few types of self-made burners are shown in
Fig. 20.2. Fuel gas mixed with air is used to produce a
clean, hot flame along the joint. Care must be taken to
avoid sudden and severe heating of the area, as excessive
temperature differentials can cause high stresses,
resulting in distortion and cracked tack welds. At
Hinkley Point in U.K., a preheating ring with burners
using propane was employed to maintain the plate
temperature at 100 to 150°C during the circumferential
welding of shells of the steam-raising unit. The shells
were nearly 50 mm thick.
Electrical strip heaters connected to 110 or 220 V
supply line can also be used for preheating. These must
be carefully insulated at the terminals to avoid the Fig. 20.2: Simple designs of self-made burners
danger of shock to welders. Such heaters have been used
in the U.S.A. for the local preheating of castings prior to
repair by welding, and of pipe and pressure vessel joints powder. In this case, it is not necessary to switch off the
in the field. Strip heaters have also been used in preheat current during welding.
conjunction with a 55 V, 1,000 Amps. AC transformer Induction heating with specially-designed 60-cycle
for preheating joints of over 50 m m thickness. The transformers is sometimes used for preheating pipe
heaters are anchored to the plates about 100 mm from joints in the U.S.A. These transformers may be operated
and parallel to the joint. They are then covered with a manually or with automatic controls, to maintain a
channel or inverted angle packed with dry insulating desired temperature range or to follow a pre-determined
time-temperature programme. Coils may be heavy Stress-relief is a form of heat treatment in which a
flexible copper cable insulated with asbestos, buss bars w e l d e d fabrication or c o m p o n e n t is heated to a
or water-cooled coils. Usually asbestos paper is wrapped sufficiently high temperature and held there for a
around the pipe before placing the coil in position. sufficiently long time, until the locked-up stresses are
Heating current has to be turned off during welding to almost completely eliminated. Stress-relieving should
prevent interference with the arc. not be confused with annealing and normalising.
Induction heating can also be obtained by the use of Temperatures for annealing and normalising are above
a laminated core inside a pipe, or on the surface of a the critical r a n g e , from 750 to 900°C, w h e r e a s
plate adjacent to a joint. This method has been used with temperatures for stress-relief are well below the critical
movable coils preceding an automatic welding head range. Annealing and normalising do remove internal
along a pressure vessel joint. stresses, but at the high temperatures involved in these
The most modern method of preheating uses what processes, far-reaching changes take place in the grain
are termed surface combustion units. They are described structure of the metal and its tensile properties.
at the end of this chapter. Figure 20.3 shows a unit in Dimensional changes can also occur.
operation. The effect of elevated temperatures as used in stress-
The simplest, most flexible and accurate method of relief on locked-up stresses in a mild steel structure is
preheating and postweld heating is electric resistance presented graphically in Fig. 20.4. It will be seen that
heating. It is described in detail in the next section. even at 500°C, 27% of the stresses still remain. It is only
A welded fabrication contains internal stresses called when the temperature exceeds 600°C that they are
locked-up stresses, which sometimes rise to values as eliminated for all practical purposes.
high as the yield strength of the material. These stresses Are the locked-up stresses totally eliminated by
added to the normal load stresses may exceed design stress-relief? An answer to this question is provided by
stresses and cause failure. Stress-relief is expected to the manner in which stress-relief becomes effective.
eliminate locked-up stresses. According to a recognised authority, "the internal
stresses are reduced by permanent yielding caused first
by the r e d u c t i o n of the yield point at elevated
temperature, and second by creep under prolonged
loading at elevated temperature." Carbon steel at
temperature of 600-700°C has a yield strength of 1 to
4.5 tons/sq inch and thus the residual stresses relax to
the same level when the structure is soaked at the high
temperature. If the heating is not localised and if cooling
is slow and even over the section, the stresses will
remain at this low level, down to room temperature.
The yield strength of the carbon steel is thus largely
recovered.
600°C
Stress Relieving Temperature (°C)
Fig. 20.3: Preheating with surface combustion unit Fig. 20.4: Relation of stress-relief to temperature
Procedures of stress-relief are laid down in relevant

boiler and p r e s s u r e vessel codes. The complete
fabrication may be kept in a furnace and stress-relieved.
Alternatively, only the area around the weld joint may
Single
be stress-relieved, as in the case of a circumferential joint control
in a penstock pipe welded at site or a joint in pressure uniform
heatinq
piping. Sometimes a long vessel may be stress-relieved system
in two stages: first one-half and then the other half. In
this case, there should be an overlap of heat treatment
X Bottom couple
at the central portion. In addition to relieving stresses,
Top bottom control', Top bottom
stress-relief treatment causes the following beneficial Temperature not necessary | control essential
Differential °C
effects: «
a) It lowers the tensile and yield strengths of the
weld-metal by about 10%, but considerably in-
i /
30 y * Estimated
creases the ductility I / Ranged
b) It removes dissolved gases like hydrogen from 20-]

the weld-metal and HAZ
c) It softens the heat-affected zone 10-1
d) It eliminates undesirable consequences of cold
work, such as work-hardening and strain-ageing 25 75 125 175 225 275 325
Pipe diameter - mm
e) It eliminates the possibility of distortion of
welded components which may occur during the
final machining Fig. 20.5: Top and bottom temperature control in pipes
f) It reduces the possibility of failure by brittle frac-

ture in service.
In low-alloy and hardenable steels, it tempers the These disadvantages led to the development of
martensite formed in the heat-affected zone. improved ceramics and modular types of heating
elements. The first type of efficient heating elements
Electric Resistance Method manufactured were the popular finger elements. These
During the 40's, heat treatment was carried out by consist of twin 9 SWG nickel-chromium wire bent in
the wire and beads method. Lengths of 80/20 nickel- the form of fingers. Each finger is 100 mm long and
chromium resistance wire were used with insulating insulated by special sintered alumina ceramics. These
beads. These heating elements were wrapped around a elements can be made to suit various diameters of
pipe and connected to a welding power source, the pipework by calculating a 1.3 V drop across each finger.
current of the welding transformer adjusted manually, The longest finger element is the popular 62-finger
to provide the control required for the heat treatment element, mainly used for postweld heat treatment of
cycles. In India, this is the most common technique used vessels. These elements require 300 Amps., 80 OCV
in localised heat treatment. welding transformers. Being manufactured from 9 SWG
The principal disadvantages are as follows: nickel-chromium wire, the finger elements require
1) During heat treatment, the nichrome wire ex- approximately 110 Amps.. A 62-finger element rating is
pands, causing gaps between the insulating beads approximately 9 kW.
leading to hot spots. Hence, in many cases lock- Advantages
ing up further stresses. a) 98% sintered alumina ceramics of high mechani-
2) The insulating beads, being of very poor quality, cal yield strength are used. These do not break
often break during handling. This can cause short- easily while handling.
circuits with the wall of the pipe. b) The hardness of the ceramic bead is comparable
3) Top and bottom control in pipes over 250 mm with a sapphire.
diameter is mandatory (see Fig. 20.5). This can- c) It has excellent electrical insulation and heat
not be practicable using the wire and beads heat- conduction properties.
ing elements. d) Can be used for temperatures up to 1,050°C.
Disadvantages
a) A high inventory is required for maintaining dif-
ferent sizes of elements for various diameters of
piping.
b) Heat treatment on vertical vessels is very cum-
bersome by this method. Certain materials do not
allow tacking of cleats for holding finger elements
during heat treatment of the vessel.
c) For the postweld heat treatment of vessels of big
diameters and heavy wall thicknesses, a number
of welding transformers are required, e.g. for a
3,350 mm diameter vessel, 32 mm thick, conform-
ing to the BS 5500 code, 18 welding transformers
are required; or for a 3,920 mm diameter vessel, Fig. 20.6: Flexible ceramic pad
48 mm thick, conforming to the ASME, 45 weld-
ing transformers are required as power sources.
For vessels over 3 m diameter and 90 mm wall c) The elements being of modular type, require a
thickness, controlling such a large number of low inventory.
welding transformers becomes impractical for the
following reasons: d) Flexibility is an excellent plus point for postweld
heat treatment of awkward flanges, branches,
i) Human error plays a predominant role as one valves, etc.
operator has to control a large number of weld
e) Same elements can be used for preheating.
ing machines for numerous ours. These ma-
chines are, very often, not very close to each f) If, during a heat treatment cycle an element burns
other. out (a very rare case), the insulation in that area
can be removed and a new ceramic pad placed
ii) The welding machines get damaged due to
over the burnt out one. The heat treatment cycle
misuse.
can be carried out without removing all insula-
iii) It is more profitable to use welding machines tion, heaters, etc.
for welding purposes. g) The flexible ceramic pads can be used for tem-
iv) It is, very often, not possible to acquire so peratures up to 1,050°C.
many welding machines for one operation. For piping from 25 mm to 1 m diameter, only two
Taking these factors into consideration, special modu- types of flexible ceramic elements are required. For
lar flexible ceramic heaters have found wide applica- small diameter piping, 30 V elements can be used. For
tion, especially, in the piping field. piping of 200 mm diameter and above, where top and
bottom control is mandatory, standard 65 V, 3.25 kW
Flexible Ceramic Pads pads are ideal. Calculations for elements are straight-
Flexible ceramic pads consist of standard dimensions forward, e.g. for 200, 300, 400 and 500 mm diameter 2,
and electrical ratings (65 V, 50 Amps., 3.25 kW). In 3, 4 and 5 elements, respectively, are used.
section, the flexible ceramic pad element comprises Twelve of these elements can be used off a 40 kVA
interlocking sintered alumina beads, through which a constant potential heat treatment power source, such
multistrand 80/20 nickel-chromium resistance wire is as the one shown in Fig. 20.7. It has the following
threaded. These elements are extremely flexible (see Fig. advantages:
20.6). 1) Very portable.
These pads have the following advantages: 2) All controls in one place, saving the operator the
a) Beads are made out of sintered alumina ceram- inconvenience of controlling several welding ma-
ics having good mechanical properties and high chines at a time. Having a low voltage output is
hardness. important from the safety point of view.
b) Beads are good electrical insulators and excel- 3) Contactorised power source with percentage
lent conductors of heat. timer controls saves operator fatigue.
7) Auxiliary power from the power source can be

utilised for the temperature recorder and solid
state programmer or any other auxiliary tools.
Solid State Programmer
The solid state programmer shown in Fig. 20.8 has
found wide use in the field of heat treatment. The
advantages are the following:
a) While some pipes are being heat-treated, auto-
matically controlled by the solid state program-
mer, more pipes can be set up by the same opera-
tor.
b) For certain applications, the temperature range
available is very small. To prevent operational
errors, it is ideal to programme the heat treatment
cycle and carry out the operation automatically.
c) This programmer also has parallel thermocouple
connections to the temperature recorders.
d) Up to eight thermocouples can be controlled at a
time.
Thermocouple attachment technique: A typical error in
Fig. 20.7: 40 kVA heat treatment transformer
a heat treatment cycle can be incurred unknowingly.
Normally, 9 SWG thermocouples are inserted in a
ferrule, which is tacked within the hot zone of a pipe to
4) Only a three-phase, 415 V, 60 Amps, supply is re- be postweld heat-treated. Heating elements are fixed
quired. This saves connections to several weld- around the thermocouple. The thermocouple picks up
ing machines for the same purpose. the heat from the elements, as it will be influenced by
5) Twelve 65 V elements at a time can be used off the higher temperature. The temperature recorded on
one power source, hence 12 of 150 mm diameter the chart is not the temperature of the pipe. The error
pipes or six of 200 mm diameter pipes or three of can range between 15 and 60°C as shown below. When
400 mm diameter pipes, or a variety of different 21 SWG thermocouples arc spark discharged (4 mm
diameter pipes can be simultaneously heat- apart) to the surface of the weldment to bi heat-treated,
treated. the error is reduced to zero as the thermocouple directly
6) All cabling from one power source makes con- records the temperature of the heated joint.
nections very easy. Also, the use of twist-locks Temperature measurement: It is a recommended
ensures quick, positive and safe connections. practice to use potentiometric multipoint temperature
Fig. 20.8: Solid state programmer

Comparison of thermocouple attachment errors:

Method Temperature Error
°C °C
Direct attachment 700 0
Thermocouple nut uninsulated 751 plus 51
Thermocouple nut insulated 731 plus 31
Stainless steel thermocouple ferrule uninsulated 728 plus 28
Stainless steel thermocouple ferrule insulated 715 plus 15
recorders in conjunction with type K chromel-alumel site. A number of factors are considered before the
thermocouples. These recorders should be calibrated implementation of a procedure as follows:
regularly, e.g. once a month. 1) The interruption of a heated band by the pres-
Special constant-voltage devices are used during a ence of other members
heat treatment cycle to check the calibration of the 2) The proximity of other welded attachments
temperature recorder. The constant voltage unit sends 3) Internal access
a multivolt signal equivalent to a particular temperature, 4) Joint dimensions
e.g. 800°C. This signal is logged on one of the points of 5) The possible effect of expansion and contraction
the recorder. If there is any deviation from this 6) The permitted temperature differentials and tem-
temperature d u r i n g the heat treatment cycle, the perature gradient considerations.
recorder must be immediately checked for calibration
error. The decision to effect the heat treatment by internal
or external means is based on diameter and internal
Localised Stress-relief access, an internal method being preferred for sizes
It is a common practice to locally heat-treat the above 1.2 m, on economic g r o u n d s and ease of
closing seams of pressure vessels at site. These situations application. The typical arrangements for internal and
arise in the sub-assembly and also in the final stages at external techniques are given in Fig. 20.9.
CIRCUMFERENTIAL STRESS RELIEF OF PRESSURE VESSEL USING

TWIN BULKHEAD METHOD & INTERNAL HEATERS OFF MAINS
U
.WEATHER PROTECTION
^ CABLES VIA BRANCH 4-12 mm RODS
12«mO ^ O
©
EXPANSION | | ( A n ). IRON WIRE M _ J
ROLLER SUPPORTS
^ | U Jt \ \ FIXED SUPPORTS
FIXED SUP
U- -*J M THIS
THIS SIDE
SIDE
DISTRIBUTION BOARD
AND TEMP RECORDER
1 mtn WIDE MINERAL WOOL

IRON MESH WIRED MATS 50 mm THICK 2 SINGLE
TO BULKHEADS STEEL BULKHEADS LAYERS AND 1 DOUBLE LAYER
HEAT INSIDE THE

THERMOCOUPLES
OUTSIDE TOTAL
680 mm 34.3 kw 60 V 6 COUPLES FITTED
CHANNEL ELEMENTS
TOTAL 36 FOR 3 m
DIA VESSEL
TYPICAL MILD STEEL SHOWS INFLUENCE

RECOMMENDED CHANNELS TO OF ELEMENTS ON
HEIGHT FOR SUPPORT ZONES OF SHELL TO
GIVE CONTROL OF
MILD STEEL ELEMENTS
HEAT, TOP/BOTTOM
CHANNELS
'SIDE/SIDE'
Fig. 20.9: Internal and external stress-relief techniques

The most popular technique used for this type of Basically, the zone between the two bulkheads is con-
heat treatment is the twin bulkhead technique. On either verted into a furnace. The heat t r e a t m e n t is by
side of the closing weld (distance approximately 1 m convection.
apart), two bulkheads are fixed. These bulkheads consist The channel heaters, s h o w n in Fig. 20.10, are
of a lattice type of framework, made out of scrap steel powerful heaters, rated at 240 V, 60 Amps., and 13-1/3
flat bars or angles. On this lattice framework are impaled kW each. Groups of elements are connected, so that a
two layers of 50 mm thick, 110 k g / m 3 density, mineral balanced three-phase network is formed, and are
wool. Between the bulkheads, channel element type of controlled by the action of energy regulating units, in
heaters are placed at 5%, 15% and 35% heights, conjunction writh overriding temperature controllers.
respectively, from the bottom of the vessel. The outer Normally, three 4-bank channel heaters are connected
p e r i p h e r y of the vessel, b e t w e e n the b u l k h e a d in a star network and connected to a 415 V supply, via a
boundaries, is lagged with a double layer of mineral contactorised distribution board. Very accurate results
wool. On half metre of each side of the double layer, a can be obtained by this method. The advantage of these
single layer of mineral wool is applied. This is to prevent elements is that they can be used in temporary furnaces
harmful gradients away from the hot zone. built around the weldment, using weld mesh and angle
iron. The channel elements can be placed on the hearth,
or other positions, depending on the job requirements
as shown in Fig. 20.11.
High Velocity Gas Techniques

In large-sized vessels and complex structures, power
c o n s u m p t i o n by electrical resistance techniques
becomes appreciably high. Very often, such huge
quantities of power are not readily available. To cite an
example, a 150-ton load requires approximately 800 kW
power. To carry out postweld heat treatment of such
weldments in the shop or at site, the need for high
velocity gas techniques has become a must.
In a conventionally fired furnace, the stagnant air
forms a surface film around the load. This reduces the
convective heat transfer to the load. The high velocity
Fig. 20.10: Channel heaters
burner, due to its inherent characteristic of ejecting hot
Structure
Insulation
Heating elements
Temporary furnace
Fig. 20.11: Temporary furnace with channel elements

air at velocities of 130 m / s e c , circulates the hot air Maintaining the temperature is one aspect of heat
ground the weldment, hence breaking the surface film treatment, ensuring the lateral movements of the legs
barrier. This improves the heat transfer and also reduces of the sphere, during expansion due to heat treatment,
the heating u p time, m a i n t a i n i n g t e m p e r a t u r e is just as important. Special rollers or lubrite pads must
differentials to a minimum. The principle of the burner be kept between the base plates of the legs of the spheres
is shown in Fig. 20.12. and the foundation base plates.
The high velocity burners utilise gas or oil as fuels. Very often, attachments such as skirts of vessels, lugs,
Gas is most commonly used. A mixture of air and gas is economiser tubes, etc., need to be supplemented with
premixed and combustion takes place inside the burner electrical resistance elements, to maintain a balance in
chamber. The flame is also very small and does not temperatures and hence minimising gradients.
extend beyond the burner cone, hence there is no chance Temporary furnaces, for stress-relieving, using high
of direct flame impingement onto the weldment. The velocity gas burners, can be made out of angle iron and
flame stability over the entire firing range has an weld-metal framework with mineral wool insulation or
advantage of a 35:1 turn-down ratio. modular prefolded mild steel panels 3 mm thick. These
Eight hortonspheres have been postweld heat- panels are bolted to each other to form a furnace. The
treated in India by U.K. technicians (one at Mangalore inside of the walls are lined with mineral wool and faced
and seven at Baroda). Each complete sphere was lagged with ceramic fibre. These types of furnaces are better
with 75 mm thick mineral wool, having a density of 110 known as low-thermal mass furnaces. They are ideal
kg/m 3 . A single burner of 20 million BTU/hour, was for large structures, or weldments where there may be
introduced through the bottom manway of the sphere. restrictive internal and external projections, or where
The high velocity burner has a specially designed cone, part of a vessel with a large hot zone has to be heat
through which hot gases are expelled. These gases have treated. When we mean large temporary furnaces, a
a scrubbing action against the walls of the sphere. By typical example is 8 m x 11 m x 28 m long for a load of
careful manipulation of the air/gas and compressed air 210 tons, in which four high velocity burners of 2.5
ratio, temperature variations u p to ± 10°C can be million k c a l s / h o u r were used. These were p u t in
maintained. strategic positions, so that there was a tangential flow
It may not be out of place to mention that 48 of hot air.
thermocouples are used for spheres of approximately The mineral wool and ceramic wool are excellent
14 m diameter and 40 mm thickness. All thermocouples insulators, used instead of heavy refractory walls. The
must be within ± 20°C. low-thermal walls are approximately 100 mm thick
Spark ignition plug

S.S. outer case
/ | *— Fan air supply
Combustion t Pilot gas supply

chamber Flexible gas
supply lines Main gas supply
Exit nozzle ^^rr=i ir flow

Air flow
velocity J ►—— - Pilot gas flow
stream ► *
J Air flow
It-Main gas supply
f — - J A i r flow
Zone of random mixing
Fig. 20.12: High velocity gas burner

compared to the 450 mm thick refractory walls, which The rating of the burner is approximately 50,000
require heavy steel sections as reinforcements. BTU/sq. ft. hour or nearly 15 kW/sq. ft. Propane gas at
The same burners are to be used in permanent low- a pressure of 20 lbs/sq. inch, at a consumption of 20
thermal mass furnaces. Since the framework and cu.ft./hour is the most efficient gas. These burners are
insulation is_very light, the furnaces can be moved on of modular unit sizes of 305x152 mm, 610x152 mm,
rails or lifted (the top hat type) or split apart. 915x152 mm, 152x152 mm, etc. They can be connected
in series to give a longer length of preheat.
Preheating Using Surface Combustion Units The flame generates infrared energy. Due to the
The surface combustion units, such as the one shown electro-magnetic nature of the heat, it is transmitted to
in Fig. 20.3, work on the principle of gas (natural or the weldment instantaneously. The heater can be con-
propane) entering a plenum chamber and burning on trolled either manually or by means of electronic con-
the ceramic plaque protected by inconel mesh. The trols using temperature controllers, to monitor solenoid
ceramic honey-combed plaque forms a resistance to the valves which merely adjust the flame to high or low.
flow of gas and allows the front face of the ceramic The only disadvantage of these units is that, one must
plaque to glow at a temperature of approximately 800°C. be careful to prevent back radiation from overheating
Atmospheric air is entrained by the gas flow and the burner.
mixes inside the rear tube and inside the plenum The surface combustion burners are ideal for
chamber. A match or piezo-electronic igniter is used to preheating closing seams of vessels, hortonspheres, deck
ignite the burner. girders, castings, etc.

Modern Arc Welding Technology, 2nd Edition (2018)

Uploaded by

Copyright:

Available Formats

Modern Arc Welding Technology, 2nd Edition (2018)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Modern Arc Welding Technology, 2nd Edition (2018)

Uploaded by

Copyright:

Available Formats

Modern Arc Welding Technology

This Page is Intentionally Left Blank

Oxford & IBH Publishing Co. Pvt. Ltd.

Second eBook Edition: 2018

Published by Satish Kumar Jain and produced by Varun Jain for

1. Introduction To Welding And Allied Processes 1

WELDING, CUTTING, AND ALLIED PROCESSES

FUSION WELDING ELECTRIC

MANUAL SUBMERGED MIG MAG

STITCH MULTIPLE SERIES ROLLER FLAME POWDER LANCING

BUTT SEAM CARBON METAL n 1

POWER SOURCE CORE WIRE

ELECTRODE HOLDER ARC POOL

GROUND CABLE BASE METAL

TO AUTOMATIC WIRE FEED

Fig. 1.2: Diagrammatic sketch of submerged arc welding

ARC AND METAL TRANSFER

Fig. 1.7: Sketch of auto-contact welding device

electric arc between an electrode and the work. The weld

Arc Stud Welding

Arc Spot Welding CONTROL CABLE

• Capacitor Discharge Welding: In this case, the stud Electrogas Welding

MOLTEN WELD METAL

SOLIDIFYING WELD METAL

Fig. 1.12: Diagrammatic sketch of electrogas welding

OXYGEN TORCH 8 MIXER

Secondary stage: part of aluminium by weight is locally heated with a

SLAG SEAL ASBESTOS SEAL

WAX DRAIN — j g ? S PREHEATING GATE

HIGH VOLTAGE RECEPTACLE

ELECTRON BEAM GUN

COLUMN VENT VALVE

OPTICAL VIEWING SYSTEM

Fig. 1.15: Schematic diagram of electron beam welding

Laser Welding about 0.05%, by weight, of chromium oxide. The green

WELDING LASER (SCHEMATIC) BASIC LASER SYSTEM

Fig. 1.18: Schematic diagram of laser welding

CLAMPS JOINT TO BE WELDED

WELDING FLEXIBLE LEAD

FORCE Multiple Spot Welding

\\V^MWti&SSq z Series Spot Welding

SERIES WELDING DIRECT WELDING IN SERIES

Fig. 1.22: Sketches showing four methods of multiple spot welding

SIDE PRESSURE ROLLS

TUBE BEING WELDED

Fig. 1.25: Diagrammatic sketch of butt seam welding

BEFORE AFTER WELDING

DRIVE SPRING COMPONENT

Fig. 1.28: Diagrammatic sketch of percussion welding:

assembly of the steel plate and the cladding material such

Gas Pressure Welding

Cold welding is generally done on ductile metals like

of the interface. The butt method is used for joining

• id ; ' issi - * isi ; ■ cs^— 2 coii

THRUST CYLINDER Tool

START Welding pin with

THRUST APPLIED Backing bar

CLAMPING FORCE APPLIED - - ^ j - 1

Fig. 1.35: Diagrammatic sketch of ultrasonic welding

Furnace Brazing THERMAL CUTTING