Production Handbook - Facilities & Maintenance

Cover photo: Borescope view of second stage compressor turbine rotor blade of
Solar Centaur gas turbine.

Location: Yibat Oman.
Inspector: R. V. Kumar
This book is the property of Shell Internationale Petroleum Maatschappij

B.V., The Hague, The Netherlands. It must be promptly returned to them at
any time they may so request. It is confidentially loaned to the holder for his
personal information and use only, and neither its existence nor its contents
shall be disclosed by the holder to any third party The holder shall take every
precaution to prevent third parties from perusing, reproducing or copying the
same either wholly or in part.
The copyright of this document is vested in Shell Internationale Petroleum

Maatschappij By., The Hague, The Netherlands. All rights reserved. Neither
the whole nor any part of this document may be reproduced, stored in any
retrievat system or transmitted in any form or by any means (electronic,
mechanical, reprographic, recording or otherwise) without the prior written
consent of the copyright owner
PREFACETOTHE PRODUCTION HANDBOOK, REVISION 1991
Objective
The objective of the Production Handbook is to contribute to efficient performance by all

Engineering, Petroleum Engineering and Operations staff, by providing quick access to
and practical guidance on their own and related disciplines' technology.
Being a comprehensive combination of condensed technical manuals, it provides a ready

source of information for reference and self-training.
It is not intended to replace detailed design manuals and state-of-the-art manuals; these
should remain the first source of reference for more experienced technical specialists.
Neither can the Production Handbook replace specialised training manuals.
Distribution
The Production Handbook should be available to all Engineering, Petroleum Engineering

and Operations staff at or above JG5, in Group E&P Operating Companies and SIPM.
These staff receivethe Handbook as a personal loan; they maytake italong when going on
transfer within the Group but must return it when leaving for other reasons. Staff of other
Functions' parentages temporarily working in E&P companies may use library copies.
The Handbook is confidential and holders should note the conditions stated opposite the
title page. Issue and recovery should be registered by company secretariats/libraries.
Reprinting and updating
The Production Handbook was first published by SIPM in 1986. It is the successor to the
Field Pocketbookversionsof 1933,1947,1952 and 1955 and the Field Handbook oil 963.
The 1986 version comprised 3000 pages in five A5 ringbinders; 6000 copies were distri-
buted.
An update of some 250 revised pages was issued in 1987 and a list of further corrections
was published in the Production Newsletter of November 1988.
A complete reprint is necessary at this time (1991). For flexibility and cost-effectiveness
this updated reprint is in nine paperback volumes, each one dedicated to a major
discipline with clear òwnership' by the SIPM-EP department concerned. These
‘custodian' departments will initiate further updates of their respective volumes as and
when necessary. Additional volumes and state-of-the-art manuals in the same format
may be added later as special supplements.
Suggestions for revising and updating the Handbook should be directed to the SIPM-EP
custodian department of the respective volume, using copies of the Specimen
Amendment Sheet at the back of each volume.
Overall editorial custodianship of the Handbook rests with SIPM-EPDI1 1.
CONTENTS LIST PRODUCTION HANDBOOK SERIES (1991)
SIPM Custodian
Volume I Production General
- Units and Conversion Factors EPO/71

- Health Safety and Environment EPO/6
- Quality Management EPO/72
- Economic Analysis EPE/1
Volume 2 Drilling and Transport EPO/51

- Drilling
- Civil Engineering for drilling locations
- Transport in Production Operations
Volume3 Petrophysical Engineering EPD/22
Volume 4 Reservoir Engineering EPD/22
Volume 5 Production Technology EPD/41

- Production Engineering
- Production Chemistry
Volume 6 Production Operations EPO/53
Volume 7 Process Engineering EPD/42

- Oil Processing
- Gas Processing
Custodian for Part I, Ch. 7, Terminals: EPD/13
Volume 8 Pipelines EPD/61
Volume 9 Facilities and Maintenance

- Running Equipment EPD/622
- Piping Systems EPD/622
- Electrical Engineering EPD/623
- instrumentation EPD!624
- Telecommunications EPD/76
- Reliability and Availability Assessment EPD/621 -EPO/54
- Corrosion Engineering EPD/63
- Inspection Techniques and Maintenance Terminology EPO/54
- Diving and Underwater Operations EPO/54
- Air Conditioning MFSH/11
PREFACE TO VOLUME 9, FACILITIES AND MAINTENANCE, REVISION 1991
The subjects covered in this Volume were formerly included in the 1 S86version
of the Production Handbook as Chapter 12, Specialist Engineering, Maintenance
and Inspection.
For a more logical grouping of material, subjects have also been incorporated
that were formerly covered under:
- Chapter 6, Production Operations: Sections 2.17-2.20 (Pumps, Compres-

sors, Gas Turbines, Gas Engines and Diesel Engines);
- Chapter 8, Oil Handling and Treating: Section 8.3 Process Control and
Instrumentation, Section 8.5 Piping Systems, Section 8.6 Pumps, Section
8.10 Air Conditioning;
- Chapter 9, Gas Handling and Treating: Section 9.7 Running Equipment;
- Chapter 10, Pipelines: Section 1.3 External Corrosion Protection, Section
3.2 Pipeline Monitoring and Control, Section 3.3 Internal Corrosion and
Corrosion Monitoring.
A more comprehensive review of the contents of the Surface Facilities'

Volumes of the Production Handbook (7. Process Engineering, 6. Pipelines, 9.
Facilities and Maintenance, and possibly others to follow) is planned for the
next update. It is intended to reflect in a more consistent manner the various
E&P engineering disciplines. The update will be aligned with the structure of
the Engineering and Maintenance training courses.
For organisational aspects of maintenance, reference should also be made to

the SIPM-E&P Maintenance Management Guide, EP 64000, of which a
revised version is currently in preperation.
VOLUME 9, FACILITIES AND MAINTENANCE, REVISION 1991
SUMMARY CONTENTS LISTING page
Chapter 1: Running Equipment

1.1 Prime Movers 1-10
1.2 Compressors 1-40
1.3 Pumps 1-113
1.4 Pulsations 1-147
1.5 Turbo-expanders 1-154
1.6 Gears and Speed Changers 1-157
1.7 Couplings and Coupling Guards 1-162
1.8 Foundations and Bedplates 1-189
1 .9 Condition Monitoring for Running Equipment 1-198
1.10 Testing of Running Equipment 1-211
1.11 References and Further Reading 1-218
For the contents of the following Chapters, reference is made to the individual
contents listings preceding each Chapter
Chapter 2: Piping Systems 2-1
Chapter 3: Electrical Engineering 3-1
Chapter 4: Instrumentation 4-1
Chapter 5: Telecommunications 5-1
Chapter 6: Reliability and Availability Assessment 6-1
Chapter 7: Corrosion and Materials Engineering 7-1
Chapter 8: Inspection Techniques and Maintenance

Terminology 8-1
Chapter 9: Diving and Underwater Operations 9-1
Chapter 10: Air Conditioning 10-1

Chapter 1
RUNNING EQUIPMENT
CONTENTS
1.1 PRIME MOVERS................................................................................................................ 1-10
1.1.1 Prime Mover Selection .................................................................................................... 1-10

1.1.1.1 General .......................................................................................................................... 1-10
1.1.1.2 Electric Motors............................................................................................................... 1-11
1.1.1.3 Gas Turbines ................................................................................................................. 1-13
1.1.1.3.1 General....................................................................................................................... 1-13

1.1.1.3.2 Air Filtration ................................................................................................................ 1-17
1.1.1.3.3 Turbine Compressor Cleaning ................................................................................... 1-21
1.1.1.3.4 Acoustic Enclosures ................................................................................................... 1-23
1.1.1.3.5 Fuel Systems.............................................................................................................. 1-23
1.1.1.3.6 Life Assessment ......................................................................................................... 1-24
1.1.1.3.7 Gas Detection in Gas Turbine Combustion Air Intakes.............................................. 1-24
1.1.1.4 Gas Engines .................................................................................................................. 1-25
1.1.1.5 Diesel Engines............................................................................................................... 1-29
1.1.1.6 Fuel Systems................................................................................................................. 1-30
1.1.1.6.1 General...................................................................................................................... 1-30
1.1.1.6.2 Gas Fuel ..................................................................................................................... 1-30
1.1.1.6.3 Distillate Liquid Fuels.................................................................................................. 1-32
1.1.1.6.4 Crude Oil and Residual Fuels .................................................................................... 1-34
1.1.1.7 Starting Systems .......................................................................................................... 1-34
1.1.1.8 Economics of Driver Selection ...................................................................................... 1-36
1.1.2 Site Rating of Drivers........................................................................................................ 1-37

1.1.2.1 Gas Turbines ................................................................................................................ 1-37
1.1.2.2 Gas and Diesel Engines ................................................................................................ 1-38
1.1.3 Gas Consumption............................................................................................................. 1-39
1.2 COMPRESSORS ............................................................................................................... 1-40
1.2.1 Types of Compressors and their Application................................................................... 1-40

1.2.1.1 Application Limitations.................................................................................................. 1-40
1.2.1.2 Centrifugal Compressors.............................................................................................. 1-42
1.2.1.2.1 General....................................................................................................................... 1-42
Confidential-Property end Copyright: SIPM, 1991 1-1

1.2.1.2.2 Limitations .................................................................................................................. 1-43
1.2.1.2.3 Anti-Surge Systems.................................................................................................... 1-43
1.2.1.2.4 Shaft Sealing Systems ............................................................................................... 1-45
1.2.1.2.5 Bearings .................................................................................................................... 1-53
1.2.1.2.6 Centrifugal Compressor Check-List .......................................................................... 1-55
1.2.1.3 Reciprocating Compressors ......................................................................................... 1-57

1.2.1.3.1 General....................................................................................................................... 1-57
1.2.1.3.2 High-speed/short-stroke versus low-speed/long-stroke reciprocating compressors . 1-60
1.2.1.3.3 Capacity Control ......................................................................................................... 1-64
1.2.1.3.4 Limitations .................................................................................................................. 1-64
1.2.1.3.5 Distance Pieces.......................................................................................................... 1-64
1.2.1.3.6 Pulsations ................................................................................................................... 1-67
1.2.1.3.7 Foundations................................................................................................................ 1-67
1.2.1.3.8 Valves ......................................................................................................................... 1-69
1.2.1.3.9 Reciprocating Compressor Check-List....................................................................... 1-70
1.2.1.4 Screw Compressors ..................................................................................................... 1-73
1.2.1.4.1 General...................................................................................................................... 1-73
1.2.1.4.2 Limitations ................................................................................................................. 1-74
1.2.1.5 Application of Compressors.......................................................................................... 1-75
1.2.1.5.1 General...................................................................................................................... 1-75
1.2.1.5.2 Separator Gas Collection ........................................................................................... 1-75
1.2.1.5.3 Gas Lift ....................................................................................................................... 1-77
1.2.1.5.4 Gas Transmission....................................................................................................... 1-77
1.2.1.5.5 Gas Injection............................................................................................................... 1-77
1.2.1.5.7 Casinghead Gas Compression .................................................................................. 1-80
1.2.1.5.8 Refrigeration ............................................................................................................... 1-80
1.2.2 Compressor Systems Control........................................................................................... 1-81

1.2.2.1 General ......................................................................................................................... 1-81
1.2.2.2 Control Instrumentation ................................................................................................ 1-81
1.2.2.2.1 Parallel Compressors (see Figure 1.2.2-1) ................................................................ 1-81
1.2.2.2.2 Series Compressors ................................................................................................... 1-83
1.2.2.3 Centrifugal Compressor Surge Control ......................................................................... 1-85
1.2.2.4 Monitoring Instrumentation ............................................................................................ 1-88
1.2.3 Compressor Sizing and Power Requirement ................................................................... 1-89

1.2.3.1 Introduction.................................................................................................................... 1-89
1.2.3.2 Initial Concepts .............................................................................................................. 1-89
1.2.3.3 Calculation Routine ....................................................................................................... 1-94
1.2.3.4 Centrifugal Compressors............................................................................................... 1-96
1-2 Confidential-Property end Copyright: SIPM, 1991

1.2.3.5 Reciprocating Compressors ....................................................................................... 1-101
1.2.3.6 Field Application .......................................................................................................... 1-112
1.3 PUMPS ............................................................................................................................. 1-113
1.3.1 Applications .................................................................................................................... 1-113
1.3.2 General .......................................................................................................................... 1-114
1.3.3 Characteristics and Selection........................................................................................ 1-114

1.3.3.1 General ....................................................................................................................... 1-114
1.3.3.2 Prototype Pumps ......................................................................................................... 1-117
1.3.3.3 Pump type selection .................................................................................................... 1-117
1.3.3.4 Centrifugal Pumps ....................................................................................................... 1-118
1.3.3.4.1 General..................................................................................................................... 1-118
1.3.3.4.2 Capacity Control ....................................................................................................... 1-118
1.3.3.4.3 Limitations ................................................................................................................ 1-120
1.3.3.4.4 Selection................................................................................................................... 1-120
1.3.3.4.5 Efficiencies and Power Requirements...................................................................... 1-121
1.3.3.4.6 Centrifugal Pump Trouble Shooting Guide.............................................................. 1-125
1.3.3.4.7 Water/viscous Fluid Equivalence ............................................................................. 1-127
1.3.3.5 Reciprocating Pumps ................................................................................................. 1-131
1.3.3.5.1 General.................................................................................................................... 1-131
1.3.3.5.2 Capacity Control ...................................................................................................... 1-132
1.3.3.5.3 Limitations ............................................................................................................... 1-132
1.3.3.5.4 Selection.................................................................................................................. 1-132
1.3.3.5.5 Efficiencies and Power Requirements..................................................................... 1-133
1.3.3.5.6 Reciprocating Pump Trouble Shooting Guide......................................................... 1-136
1.3.3.6 Screw Pumps ............................................................................................................. 1-138
1.3.3.6.1 General.................................................................................................................... 1-138
1.3.3.6.2 Capacity Control ...................................................................................................... 1-139
1.3.3.6.3 Limitations ............................................................................................................... 1-139
1.3.4 Pipework Requirements ............................................................................................... 1-140

1.3.4.1 Centrifugal Pumps ...................................................................................................... 1-140
1.3.4.2 Positive Displacement Pumps..................................................................................... 1-141
1.3.4.2.1 Reciprocating Pumps ............................................................................................... 1-141
1.3.4.2.2 Rotary Pumps........................................................................................................... 1-141
1.3.5 Net Positive Suction Head (NPSH) ................................................................................ 1-142
1.3.6 Centrifugal Pump Arrangements ................................................................................... 1-143
1.3.7 Prime Movers ............................................................................................................... 1-144
1.3.8 Seals............................................................................................................................... 1-144

1.4 PULSATIONS................................................................................................................... 1-147
1.4.1 Pulsation Dampeners for Reciprocating Compressors .................................................. 1-147
1.4.2 Pulsation Dampeners for Reciprocating Pumps........................................................... 1-148

1.4.2.1 Types of Pulsations ..................................................................................................... 1-148
1.4.2.1.1 Pulsations Below Primary Pumping Frequency ....................................................... 1-148
1.4.2.1.2 High Frequency Pressure Pulsations ....................................................................... 1-148
1.4.2.1.3 Cavitation.................................................................................................................. 1-149
1.4.2.2 Design Criteria............................................................................................................. 1-149
1.4.2.2.1 Liquid Filled Surge Vessels ..................................................................................... 1-149
1.4.2.2.2 Gas Cushion Dampeners ........................................................................................ 1-149
1.4.2.2.3 Tuned Filters - Helmholtz type ................................................................................. 1-151
1.5 TURBO-EXPANDERS....................................................................................................... 1-154
1.6 GEARS AND SPEED CHANGERS................................................................................. 1-157
1.6.1 General ............................................................................................................................ 1-157

1.6.2 Gear Selection................................................................................................................. 1-157
1.6.3 Gear Box Specification .................................................................................................... 1-157
1.6.4 Misalignment .................................................................................................................. 1-158
1.6.5 Torsional Vibrations........................................................................................................ 1-158
1.6.6 Gear Tooth Problems ..................................................................................................... 1-158
1.6.6.1 Pitting........................................................................................................................... 1-158
1.6.6.2 Spalling ........................................................................................................................ 1-159
1.6.6.3 Scoring (Scuffing) ........................................................................................................ 1-159
1.6.6.4 Galling, Plastic Flow .................................................................................................... 1-159
1.6.6.5 Tooth Breakage ......................................................................................................... 1-160
1.6.6.6 Normal Wear ............................................................................................................... 1-160
1.6.6.7 Corrosive Wear............................................................................................................ 1-160
1.6.7 Start-Up Problems .......................................................................................................... 1-160

1.6.7.1 Thrust Bearings ........................................................................................................... 1-161
1.6.7.2 Lubricant Drainage ...................................................................................................... 1-161
1.6.7.3 Quality Control Errors .................................................................................................. 1-161
1.6.8 Condition Monitoring Instrumentation............................................................................. 1-161
1.7 COUPLINGS AND COUPLING GUARDS ....................................................................... 1-162
1.7.1 The function of a coupling .............................................................................................. 1-162
1.7.2 The need for coupling flexibility ...................................................................................... 1-163

1.7.3 Coupling sizing and rating ............................................................................................... 1-165
1.7.3.1 Design Margins............................................................................................................ 1-163
1.7.3.2 Coupling selection ....................................................................................................... 1-167
1.7.4 Types and characteristics of flexible couplings .............................................................. 1-168

1.7.4.1 Designs based on non-metallic components............................................................... 1-168
1.7.4.2 All-metal flexible couplings .......................................................................................... 1-168
1.7.5 All-metal flexible membrane couplings......................................................................... 1-174

1.7.5.1 Principles of construction............................................................................................ 1-175
1.7.5.2 Torque capability .......................................................................................................... 1-177
1.7.5.3 Environmental tolerance.............................................................................................. 1-178
1.7.5.4 Service life ................................................................................................................... 1-178
1.7.5.5 Misalignment forces..................................................................................................... 1-178
1.7.5.6 Dynamic balance retention .......................................................................................... 1-178
1.7.5.7 Torsional rigidity .......................................................................................................... 1-179
1.7.5.8 Installation ................................................................................................................... 1-180
1.7.5.9 Limited end float couplings .......................................................................................... 1-180
1.7.6 Design features .............................................................................................................. 1-181
1.7.7 Guards and enclosures .................................................................................................. 1-182

1.7.7.1 Guards ......................................................................................................................... 1-182
1.7.7.2 Enclosures................................................................................................................... 1-182
1.8 FOUNDATIONS AND BEDPLATES ................................................................................. 1-189
1.8.1 Installation of Running Equipment on Concrete Foundations ...................................... 1-189

1.8.1.1 Foundation Bolts......................................................................................................... 1-189
1.8.1.2 The Chock and Block Method of Grouting .................................................................. 1-189
1.8.2 Installation of Running Equipment on Offshore Platforms ............................................. 1-194
1.9 CONDITION MONITORING FOR RUNNING EQUIPMENT............................................ 1-198
1.9.1 Objectives of Condition Monitoring................................................................................. 1-198

1.9.1.1 Plant Safety ................................................................................................................. 1-198
1.9.1.2 Equipment Availability ................................................................................................. 1-198
1.9.1.3 Operating Costs........................................................................................................... 1-198
1.9.1.4 Environmental Safeguards .......................................................................................... 1-199
1.9.2 Condition Monitoring Techniques.................................................................................. 1-199
1.9.3 Gas Turbines .................................................................................................................. 1-200

1.9.3.1 Protection Systems...................................................................................................... 1-200
1.9.3.2 Condition Monitoring Trending .................................................................................... 1-201
1.9.4 Centrifugal Compressors............................................................................................... 1-204

1.9.5 Centrifugal Pumps .......................................................................................................... 1-206

1.9.6 Gas Engines ................................................................................................................... 1-207

1.9.6.1 Protection Systems..................................................................................................... 1-207
1.9.6.2 Condition Monitoring Trending ................................................................................... 1-207
1.9.7 Reciprocating Compressors ........................................................................................... 1-207

1.9.7.1 Protection Systems..................................................................................................... 1-207
1.9.8 Reciprocating Pumps ..................................................................................................... 1-208

1.9.9 Gearboxes ...................................................................................................................... 1-209

1.10 TESTING OF RUNNING EQUIPMENT.......................................................................... 1-211
1.10.1 General ......................................................................................................................... 1-211

1.10.2 Prime Movers ............................................................................................................... 1-211
1.10.2.1 Electric Motors........................................................................................................... 1-211
1.10.2.2 Gas Turbines ............................................................................................................. 1-211
1.10.2.3 Gas and Diesel IC Engines ....................................................................................... 1-212
1.10.3 Compressors ............................................................................................................... 1-212
1.10.3.1 Centrifugal Compressors........................................................................................... 1-212
1.10.3.2 Reciprocating Compressors ..................................................................................... 1-215
1.10.3.3 Rotary Compressors.................................................................................................. 1-215
1.10.4 Pumps.......................................................................................................................... 1-215
1.10.5 Gears and Speed Changers......................................................................................... 1-216
1.10.6 Couplings...................................................................................................................... 1-216
1.10.7 Complete Unit, or ‘String’ Test ..................................................................................... 1-217
1.11 REFERENCES AND FURTHER READING................................................................... 1-218

TABLES
Table 1.1.2-1 Gas turbine derating 1-37

Table 1.1.2-2 Engine derating 1-38
Table 1.2.3-1 Typical polytropic efficiencies for different types of
machine 1-91
Table 1.2.3-2 Typical values of K 1-108
Table 1.3-1 Preferred pump types for water and crude oil
applications in EP 1-116
Table 1.3-2 Centrifugal pump trouble shooting guide 1-125
Table 1.3-3 Maximum efficiency for power pumps 1-135
Table 1.3-4 Maximum pump stroke rate 1-135
Table 1.3-5 Reciprocating pump trouble shooting guide 1-136
Table 1.7-1 Design margins 1-167
FIGURES
Figure 1.1.1-1 Cross-section of aero-derivative gas turbine 1-14

Figure 1.1.1-2 Cross-section of heavy duty gas turbine 1-15
Figure 1.1.1-3 Air filtration arrangement - ALTAIR SYSTEM 2 1-19
Figure 1.1.1-4 Air filtration arrangement - Donaldson GDX 1-19
Figure 1.1.1-5 Integral gas engine compressor 1-27
Figure 1.1.1-6 Typical separable compressor arrangement 1-28
Figure 1.1.1-7 Fuel gas system 1-31
Figure 1.1.1-8 Distillate fuel system 1-33
Figure 1.2.1-1 Graph of equipment type selection 1-41
Figure 1.2.1-2 Cross-section of centrifugal compressor 1-42
Figure 1.2.1-3 Cross-section of contact sea 1-46
Figure 1.2.1-4 Cross-section of oil film seals 1-48
Figure 1.2.1-5 Cross-section of dry gas seal 1-51
Figure 1.2.1-6 Cross-section of magnetic bearing 1-54
Figure 1.2.1-7 View of reciprocating compressor 1-58
Figure 1.2.1-8 Cross-section of reciprocating compressor 1-59
Figure 1.2.1-9 Distance piece venting and draining 1-66
Figure 1.2.1-10 Reciprocating compressor unbalanced forces 1-68
Figure 1.2.1-11 Comparison between reciprocating compressor
starting torque and induction motor torque 1-72
Figure 1.2.1-12 Screw compressor 1-73
Figure 1.2.1-13 Typical compressor control system schematic 1-76
Figure 1.2.1-14 Reciprocating compressor cylinder for injection
service 1-78

Figure 1.2.1-15 Centrifugal compressor for injection service 1-79
Figure 1.2.2-1 Typical parallel connected schematic 1-82
Figure 1.2.2-2 Series connected compressors 1-84
Figure 1.2.2-3 Surge control system 1-86
Figure 1.2.3-1 Relationship between adiabatic and polytropic
efficiencies 1-92
Figure 1.2.3-2 Flow diagram for calculation nomenclature 1-95
Figure I.2.3-3a Power absorbed by centrifugal compressor,
r = 1.2 to 2.2 1-99
Figure 1.2.3-3b Power absorbed by centrifugal compressor,
r = 2.2 to 3.6 1-100
Figure 1.2.3-4a Power absorbed by reciprocating compressor,
r = 1.2 to 2.6 1-103
Figure 1.2.3-4b Power absorbed by reciprocating compressor,
= 2.6 to 4.0 1-104
Figure 1.2.3-4c Power absorbed by reciprocating compressor,
= 4.0 to 5.4 1-105
Figure 1.2.3-5 Illustration of volumetric efficiency 1-109
Figure 1.2.3-6 Volumetric efficiency vs. pressure ratio 1-110
Figure 1.3-1 Pump range chart 1-115
Figure 1.3-2 Centrifugal pumps - specific speed vs. impeller
shape 1-122
Figure 1.3-3 Centrifugal pumps - efficiency at design point 1-123
Figure 1.3-4 Centrifugal pumps - performance correction chart 1-124
Figure 1.3-5 Performance correction chart for viscous liquids 1-129
Figure 1.3-6 Performance correction chart for viscous liquids 1-130
Figure 1.3-7 Horizontal plunger pump 1-131
Figure 1.3-8 Cross-section of a typical screw pump 1-138
Figure 1.3-9 Parallel centrifugal pumps with different
characteristics 1-143
Figure 1.3-10 Cross-section of mechanical seal 1-145
Figure 1.4.2-1 Types of pulsation dampener 1-150
Figure 1.4.2-2 Choke tube ‘C' factor 1-153
Figure 1.5-1 Cross-section of a turbo-expander/recompressor 1-155
Figure 1.5-2 Turbo-expander/recompressor process flow diagram 1-156
Figure 1.7-1 Rigid coupling 1-162
Figure 1.7-2 Basic shaft misalignment conditions 1-164
Figure 1.7-3 Compound shaft misalignment conditions 1-164
Figure 1.7-4 Service factors 1-166
Figure 1.7-5 Coupling designs based on non-metallic components 1-169
Figure 1.7-6 All-metal multiple membrane couplings 1-169
Figure 1.7-7 Metal spring (grid) type coupling 1-170

Figure 1.7-8 Gear coupling - close coupled 1-170
Figure 1.7-9 Gear couplings - spacer types 1-171
Figure 1.7-10 Renold' all steel chain coupling 1-171
Figure 1.7-11 Contoured disc coupling 1-172
Figure 1.7-12 Operating envelopes for different types of flexible
coupling 1-173
Figure 1.7-13 Multiple membrane couplings 1-176
Figure 1.7-14 Maximum angular misalignment 1-176
Figure 1.7-15 Stiffness vs. displacement 1-179
Figure 1.7-16 Typical API 671 coupling 1-183
Figure 1.7-17 Coupling enclosures - cross-section. Separation of
oil from air 1-187
Figure 1.7-18 Compensation of negative pressures 1-188
Figure 1.8-1 Foundation bolt and sleeve installation 1-190
Figure 1.8-2 Procedure for chock-and-block method of grouting 1-191
Figure 1.8-3 Illustration of chocks at the grouting stage 1-193
Figure 1.8-4 Three-point mounted baseplate design 1-195
Figure 1.8-5 Baseplate and holding-down bolt arrangement 1-197

Running Equipment
1.1 PRIME MOVERS
1.1.1 Prime Mover Selection
1.1.1.1 General
Electric motors provide relatively compact machinery packages of low com-

plexity, especially where the driven machine includes simplifying features
such as dry gas seals and magnetic bearings in compressors, and product
lubricated bearings in pumps. Electric motors are now being designed, using
VSDS control, to run at super synchronous speed so that in the near future it
should be possible to drive high speed compressors directly, eliminating the
need for gearboxes. This will further reduce the requirement for lubricating
oil on machinery packages. ln the limit both motor and driven machine can
have magnetic or product lubricated bearings to provide oil-free systems.
Further designs under development which uses a sidestream of gas to cool
the motor. This results in a machine with no seals to atmosphere.
Gas turbines and reciprocating engines require complex support systems for
lubrication, cooling and control and are available only in discrete power
sizes. Also their speeds are such that a gearbox is often needed to match the
driver with driven equipment speed requirements.
At an early stage in the development of a project it is necessary to decide

whether to design for central power generation with electric motor driven
equipment, or for direct drive of each machine by gas turbine or gas engine.
In some cases the decision will be made by prevailing circumstances, such
as an existing high voltage overhead line infrastructure. In other cases a
compressor may need to be added to an existing plant without spare gen-
eration capacity, so that direct drive becomes the preferred solution.
In many cases however the decision is not clear and will depend on unique
circumstances. For example, on an offshore oil production platform there
may be perhaps 5 major machinery drives accounting for 80% of the total
power requirements. At first sight, electric drives may appear to be a capex
intensive solution due to the necessary generation equipment, high voltage
switchgear and cabling. However the alternative is to install a large number
of gas turbines as direct drives, which are considerably more maintenance
intensive and result in lower package availabilities. A system effectiveness
study can assist in developing the optimum solution but electric power gen-
eration will, in any case, need to be available. Furthermore, distributed gas

Facilities and Maintenance
turbines can have safety implications with fuel supplies and air intakes in
several areas of the platform.
The current developments with high speed motors and simplified machinery
packages together with the higher availability of electric motor drives, result
in the concept of central power generation becoming increasingly acceptable.
1.1.1.2 Electric Motors
In comparison with combustion engines, the electric motor is a simple, robust

machine of high efficiency. Squirrel cage induction motors are the most
common type used in EP installations, even for large motors up to 20MW. The
maximum speed is a function of the supply frequency, and the number of
poles.
Except for sizes up to 200 KW the manufacturer will build the motor exactly to
the user's specification on power output, supply voltage, and frequency, and
therefore establish an ideal match between supplied and required power.
When selecting an electric motor it is necessary to consider both the driven

equipment full power requirement and the torque, especially under starting
conditions. Normally the starting torque of an induction motor is low, possibly
falling to a minimum during acceleration and then rising to a peak just below
running speed. In addition, torque drops further with a reduction in terminal
voltage. It is essential to ensure that there is sufficient break-away torque,
with margin, to start the driven equipment and accelerate it to operating
speed. To ease the starting torque requirements, the driven equipment
should be unloaded, via recirculation, or bypassed, during start-up. The
torque relationship at operating speed must also be considered to ensure
there is adequate margin between required and available torque once the
driven equipment is loaded.
There is also increasing acceptance of VSDS electric motors (especially with

motor sizes less than 2.5 MW) utilising thyristor controls. This allows the
motor to run at speeds between 5% and approximately 200% of the equi-
valent synchronous speed, providing the driven equipment with significant
turn-down capability. Efficiency and reliability of variable speed drives are
similar to those found on constant speed drives, but additional space is
required to house the power electronics, transformers and harmonic filters.

Running Equipment
Important characteristics of the variable speed drive are the torque and the
current developed when starting. Both can be selected such that at standstill,
high torque values can be developed while operating at current limitation.
This results in generally ample accelerating torques and at the same time a
much lower current consumption from the electrical system.
Most of the energy loss due to motor inefficiency is evolved as heat, and
allowance must be made for dissipation of this energy. Most installations of
large electric motors are in hazardous areas, and as the motor casing needs
to comply with area classification requirements, an external form of cooling
is necessary. In the internal motor circuit the air will be circulated by shaft
driven fans. This circuit is in turn cooled by an open circuit air system or a
water cooling arrangement. In a water cooled motor it is essential to ensure
that any leaks from the heat exchanger do not contaminate the electrical
system. Where sea water is used as the cooling medium, corrosion resistant
materials such as copper-nickel alloy or titanium must be in the heat ex-
changer fabrication.
Where open circuit air cooling is used, the heat exchanger is considerably
larger than that for a water cooling system, and in addition, allowances must
be made for large cross-section ducting to supply cooling air to motors in
enclosed modules.
It may also be unacceptable to discharge hot air into an enclosed module and
so additional ducting may also be required to discharge the air elsewhere.
Depending on the operating environment it may be necessary to provide a

supply of pressurising air to the motor enclosure to satisfy hazardous area
requirements. This supply must be taken from a safe area and may require
additional ducting and pressurising fans. Careful attention to shaft seals and
access hatches is required to ensure the integrity of pressurising system.
Anti-friction bearings are used in the majority of small motors and require
little attention apart from periodic replacement and lubrication.
Slow speed motors (6 pole and greater) below 2 MW and two pole motors
between 200 kW and 5 MW are likely to be fitted with sleeve bearings. Heat
dissipation from motor bearings of this size is not large and a relatively
simple oil flinger or self-contained pumping system is satisfactory.
Bearings for slower speed motors above 2 MW and two pole motors larger
than 5MW are likely to require a pressurised oil supply system which may be
common with the driven equipment system.

Large reciprocating compressors driven by electric motors have an uneven

torque vs angle of rotation characteristic which can induce a high level of
current pulsation in the power supply.
The system design must be verified by the relevant specialist Electrical

Engineers to ensure that pulsations remain within acceptable limits, If the
high voltage distribution system is not sufficiently robust, a flywheel with a
larger moment of inertia may be required to further dampen torque varia-
tions, or an alternative motor arrangement needs to be considered.
1.1.1.3 Gas Turbines
1.1.1.3.1 General
Gas turbines provide a compact power source, are tolerant to different fuels,
relatively easy to maintain and are reliable.
Simple cycle turbine thermal efficiencies have improved from around 16% to
over 37% during the last 20 years, and the efficiencies of over 50% are now
achievable in combined cycle operation. The development of a modern gas
turbine is extremely costly and so units are available only in discrete
sizes.
Historically, gas turbine development progressed along two distinct routes.

The ‘industrial' or ‘heavy duty' machine was developed using steam turbine
technology, and is characterised by large and heavy cast iron casings.
Maintenance is carried out on site and requires the package to be out of
service for an extended during major overhauls.
Conversely, as the name suggests, the aero-derivative turbine uses an en-

gine developed for aircraft use which has been modified for industrial ap-
plications. The gas turbine engine produces hot gases (hence the term gas
producer or gas generator) which are expanded further across a power
turbine to produce the net turbine output. The power turbine is often man-
ufactured by the package supplier, with the gas producer being provided by
an alternative manufacturer. Figure 1.1.1-1 and Figure 1.1.1-2 show exam-
ples of the two technology types.
The advantages of using an aero-derivative is that the gas generator is a

more compact unit, light in weight, and can be changed out in as little as 10
hours. The power turbine can normally be expected to operate for at least
100,000 hours without a major overhaul.

Running Equipment


Running Equipment
Aero-derivatives are however not suitable for use with non-premium fuels
such as residual and crude oil. The more compact combustion system cannot
tolerate the extended combustion pattern of the heavier liquid fuels and the
larger combustion chambers of the industrial turbine are required. However
for the majority of EP applications, gas fuel is available so the aero-derivative
is judged fully suitable.
The distinction between the two types of gas turbines is becoming increas-
ingly blurred as technology first developed for military aero-engines has
been passed on to civilian aircraft engines, aero-derivatives and on to in-
dustrial types.
Compressor design, hot section component materials, and bucket cooling
techniques are three such areas of major advances.
Also some of the smaller industrial manufacturers have developed remov-
able gas producers and modularised sections to facilitate maintenance ac-
tivities.
As turbines come in discrete sizes, it is important to be aware of ratings so

that a meaningful exercise can be carried out to optimise plant size and
turbine choice. Publications such as the GTW (Gas Turbine World) Handbook
show the selection of currently available gas turbines. A significant number
of these models are not suitable for EP service, or are considered to still be
‘prototype' units and not proven for EP environments. Always consult the
SIPM-EP Mechanical Engineering Discipline before completing a selection.

A prototype is described in DEP 31 .29.70.1 1-Gen as:
- A design of gas turbine whose leading engine has not yet achieved
98% reliability over 25,000 hours, or where the total population of
identical gas turbines has achieved less than 300,000 fired hours.
Once the process of equipment selection has started, a decision is required

as to whether to select a single or two-shaft configuration, and in an industrial
or aero-derivative form. Aero-derivative gas turbines are normally two shaft
machines with the power turbine not directly coupled to the gas generator. In
general industrial turbines are produced by manufacturers in either a single
or two shaft design. The air flow compressed by the gas turbine compressor
varies as the square of the running speed, and so the power output is sig-
nificantly affected by minor changes in running speed.
Where a variable speed drive is required, using a single shaft machine and
the gas turbine compressor speed is varied according to process require-
ments, serious throughput limitations can be experienced at reduced speed.
Therefore when a wide operating speed range is required, a two shaft should
be used. The power turbine speed is varied to meet the process require-
ments, and the compressor turbine or gas producer either runs freely to
provide the hot gas to meet load conditions, or is controlled by variable
geometry power turbine entry nozzles.
For a fixed speed drive application, such as an alternator, a single shaft

arrangement is preferred, with output a function of the quantity of fuel in-
jected into the combustion system.
The larger coupled rotating inertia of the single shaft arrangement allows
easier frequency control especially on the relatively small electrical systems
found in EP. However aero-derivative are now common in power generation
duty, especially on offshore platforms but responsive control systems are
necessary to ensure that shaft speed swings are control led during rapid load
changes.
1.1.1.3.2 Air Filtration
As a gas turbine ingests large quantities of air (for example, a GE Frame 6

compresses 9.7 x 106 standard m3/d of air) atmospheric pollutants ingested
with the air can affect the compressor performance. Pollutants with a particle
size of greater than 10 microns cause stripping of coatings and blade erosion
to both static and rotating elements. Pollutants of less than 10 microns cause
fouling by coating blades and reducing the compression efficiency. The fouling

Running Equipment
is revealed as increased exhaust temperature and reduced compressor

discharge pressure under fixed operating conditions.
In environments where the pollutants include salt (offshore, coastal and

many desert environments) care must be taken in the design of the air
filtration system in order to ensure effective salt removal. Sodium chloride is
an aggressive corrosive agent under elevated temperatures to the hot gas
path components.
In early offshore developments, it was assumed that marine spray would

provide the biggest problem, but experience has shown this is not the case as
the air intakes on platforms lie above most of the spray. However the filtration
system must be able to cope with the airborne salt pollutant. The major error
in early offshore designs was to ignore the atmospheric contaminants pro-
duced by the platform itself.
During field development the major pollutants are caused by dust from drill-
ing muds under transfer, and the products from drilling mud residue burning.
As the installation matures, the major pollution is from shot blasting activities
on the platform.
Figure 1.1.1-3 shows one air filtration arrangement for an offshore environ-
ment, comprising weather hood, a two-stage bag filter arrangement to re-
move solid contaminants and coalesce moisture, with a vane pack down-
stream to separate out the coalesced droplets. This arrangement should not
be considered definitive but does provide an effective maintainable filtration
system.
Similarly, good performance can be achieved from a ‘self-cleaning' type filter
without the filter pulsing equipment fitted.


Running Equipment
The type of filter most suitable for desert applications is dependent on the
levels of humidity encountered. For inland areas with low humidity, the self-
cleaning systems provide excellent service. Cleaning is initiated by a timer
or differential pressure switch and causes a pulse of air to be directed op-
posite to that of the normal direction of flow.
The pulse (usually 100 microseconds) causes a momentary flow reversal and
the accompanying shock wave clears accumulated dust from the filter ex-
terior. The conical/cylindrical filter arrangement has a major advantage in
that the element sealing area is small compared with the large sealing area
found on the of the square panel arrangements, and the seal integrity can be
easily confirmed. Figure 1.1.1-4 shows the layout of a current production
self-cleaning filter.
In desert locations with significant periods of high humidity, filter selection is

not easy, and is an area in which the filter manufacturers are undertaking
significant research.
Self-cleaning systems, self-cleaning systems without pulsing arrangements,
self-cleaning systems with prefilter covers, and traditional three stage sys-
tems are all currently being utilised.
In older three stage systems, blow-in or bypass doors were commonly fitted
in association with filtration systems. The current philosophy is to avoid this
arrangement and maintain turbine protection by monitoring pressure in the
clean air duct with prealarm and shut-down differential pressure switches.
Blow-in and bypass doors seldom seal satisfactorily, and therefore contin-
ually provide a source of unfiltered air and allow water ingress to the clean
air duct. In many existing installations, operators allow units to continue
operation for extended intervals with bypass doors open, often in ignorance
of the effects on turbine performance and life.
In offshore environments, an all stainless construction is now considered

standard for the filter house and inlet plenum. In less harsh environments, 5
mm mild steel with appropriate coatings is still considered satisfactory.
Provision of a weather hood is imperative for offshore facilities. The design

should ensure the prevention of wind driven rain, and for North Sea and
Arctic duty, prevent the- direct ingress of snow which can pack up during
winter storms and cause filter blockage.
In all locations, the placement of the air intake is important, It should be

unobstructed and allow good stairway access for filter changes and no hy-
drocarbon processing equipment should be located in front of the intake. It is

acceptable to place an air intake in a Zone 2 area if the pressure drop and
cost to duct to a safe area are excessive. A gas turbine will not necessarily
accelerate and disintegrate if it ingests a flammable gas; the outcome of such
an event would depend upon the actual concentration, and the engine type.
The intake should be positioned so that the prevailing wind does not carry hot
exhaust gases or waste heat from other process equipment into the inlet. Any
increase in inlet air temperature above ambient will adversely affect turbine
performance.
For offshore platforms, plume studies should be carried out, not only to
ensure there is no hot gas recirculation to the turbine inlet, but also ensure
there is no ingestion by other equipment and services. The plume effect on
the helideck environment must also be carefully studied.
Any inlet pressure drop affects turbine performance, so it is important to

optimise the inlet design in constrained locations, and review closely filter
manufacturers' performance specifications.
In the past anti-icing systems have been specified for North Sea environ-
ments. However in most locations it has been found that the anti-icing sys-
tems are not required if an effective air filtration system is fitted and most
anti-icing systems have major disadvantages. Compressor discharge air
bled from the turbine and recirculated to the compressor inlet guide vanes or
to the inlet filter housing can significantly influence turbine power output.
Recirculation of hot exhaust gases provides contamination of the inlet air
system with combustion by-products. As a result few installed anti-icing
systems are operational and most North Sea Operators specifying new
equipment now omit any reference to anti-icing systems.
However for developments in harsher offshore environments, or for Arctic
onshore locations, anti-icing systems may be required, but studies of alter-
native heating arrangements should be carried out.
1.1.1.3.3 Turbine Compressor Cleaning
A suitably designed air filtration system correctly operated will ensure no

compressor erosion occurs, but no system is totally effective in eliminating
the effects of fouling, and so periodic cleaning is required to reinstate com-
pressor performance. Compressor cleaning techniques can be summarised
as follows:
Carboblast, wheatings or rice can be introduced into the compressor with the
machine under load. This method has the effect of dislodging fouling build-up

Running Equipment
and has the advantage of being able to be carried out without affecting
machine availability. However it has significant disadvantages. Oily deposits
either as a result of leakage from bearing systems or from oil vapours con-
taminating the atmosphere, are not effectively removed by this method. Also
most turbines use compressor extraction air for cooling and sealing pur-
poses, and with repeated compressor cleans, bearing labyrinth seals be-
come choked with the solid cleaning agent allowing oil leakage, and turbine
hot section cooling passages become blocked causing component overtem-
perature. This method was considered the traditional means of compressor
cleaning, but with the advances in gas turbine design over the last twenty
years, it must now be considered a redundant method.
The accepted method of compressor cleaning is now the crank soak method.
The machine is shut down, allowed to cool and a water/detergent mixture is
sprayed into the compressor bell mouth while it is rotated at crank speed. The
mixture is allowed to soak for 15-20 minutes and is then rinsed off at the same
compressor speed using clean water. The advantage of this method is that
the cleanliness of the compressor first stages can be verified before the
machine is returned to service. With the quality of detergents now available,
a high standard of compressor wash can be achieved. The primary disad-
vantage is that the machine must be off-line for several hours to allow cool-
down prior to washing, and this can have an appreciable availability impact.
The on-line wash can improve machine availability but implementation must
be carried out under strict supervision. Some turbine manufacturers have yet
to endorse on-line washing, and others only approve a limited range of
detergents for on-line cleaning. Also the type and layout of the spray nozzles
for an on-line system is often considerably different from that used for a crank
soak. An on-line wash should be considered as a method of assisting in
maintaining a compressor in good condition and cannot replace a crank soak
system. An accurate method of turbine monitoring is required in order to
achieve the correct wash interval - if the wash is too frequent, excessive
detergent costs are incurred; if it is not frequent enough, then the on-line
clean will not restore the performance and the machine must be taken off-I me
for a crank soak wash. If a turbine is in critical service, and further improv-
ements to the air filtration system are not possible, on line washing can
provide availability improvements, depending on the turbine manufacturer, if
implemented in a structured and supervised manner.

1.1.1.3.4 Acoustic Enclosures
Gas turbines emit a sound level and spectrum which are unacceptable in a
modern environment, and so are housed in acoustic enclosures which are
designed so that the package meets current limits on emission levels. How-
ever the enclosure can have the adverse effect of making access for main-
tenance and trouble-shooting difficult if the package design is inappropriate.
With an acoustic enclosure in place, an effective compartment ventilation

system is required in order to remove heat rejected from the turbine. In
addition, area classification considerations may require a differential pres-
sure between the enclosure and its surroundings. The air quality used for
compartment ventilation should be to a similar standard as that provided for
the combustion air inlet and is often drawn from the clean air duct, or pref-
erably from a partitioned section of the filter house. If air is taken from the
clean air duct, it is important to ensure the ventilation fans are designed to
operate with a suction pressure equivalent to that found in the clean air
duct.
1.1.1.3.5 Fuel Systems
All gas turbine manufacturers issue specifications on the type of fuels which
can be tolerated by their equipment. The calorific value and Wobbe Index of
gas fuel are particularly important, as they can generally only vary by + 5%
before changes to turbine control calibration setpoints are required. The
Wobbe Index allows comparison of the volumetric energy content of different
gases at different temperatures. Gas turbines can operate on low calorific
value gases - up to 40% carbon dioxide in methane has been successfully
used as fuel gas. However no gas turbine can operate successfully with
hydrocarbon liquid in gas fuel, and there are numerous instances of major
equipment failures following the carryover of liquids into the gas fuel system.
It is essential to ensure that all liquids are removed and the gas is at least 20
degrees Celsius above the dew point at the turbine gas control valve or as
prescribed by the gas turbine manufacturer. Dewpoint control can be
achieved by either low temperature separation or fuel gas heating but when
heating, most manufacturers set inlet conditions to a maximum of 70 degrees
Celcius.
All turbines will operate on distillate fuel so long as the manufacturer's

specifications are adhered to. However when residual or crude oil is to be
used as fuel, then only the industrial type gas turbine should be considered.

Running Equipment
When using these fuels, it is necessary to ensure that the fuel is supplied at
the correct viscosity and that all solid impurities have been removed by
centrifuging and filtration. If salt is present in the fuel, it must be washed using
fresh water, and vanadium must be inhibited by dosing the fuel with a mag-
nesium salt. Both sodium and vanadium, especially in the presence of sulphur,
are extremely corrosive and serious hot section component life limitation can
occur, depending on the level of fuel contamination and turbine load unless
adequate conditioning is carried out. Where distillate fuel is used offshore, it is
almost always salt contaminated and so facilities must be provided on the
platform for fresh water washing and centrifuging.
1.1.1.3.6 Life Assessment
During equipment overhaul, component disposition is usually the responsi-

bility of the equipment manufacturer or the overhaul facility, and is dictated
by established inspection criteria for component condition. However a re-
maining creep life assessment on turbine hot section rotating components is
often not carried out. Creep life is greatly affected by the actual operating
environment but cannot be accurately forecast without carrying out destruc-
tive metallurgical tests on representative samples. This service is available
at the Shell Billiton Research Arnhem (B.R.A.) facility and it is recommended
that two samples of first stage turbine blades be forwarded to them at each
turbine major overhaul in order to make a component assessment. The blade
assessment report generally covers three areas:
1. Internal Damage Accumulation: Where less than 50% of component life

has been consumed, phase alignment is evident in the grain structure. Once
the 50% mark is passed, void formation and growth is evident. This then
allows assessment of remaining creep life.
2. Temperature Profile: Analysis of gamma prime growth can indicate the

maximum temperature experienced by the component in service.
3. Rejuvenation Recommendations: Assessment may indicate that compon-

ent life extension be possible through `hipping' (Hot Isostatic Pressing). Un-
der conditions of high temperature and pressure, void formation can be
reserved and the material returned in many cases to an as-new condi-
tion.
1.1.1.3.7 Gas Detection in Gas Turbine Combustion Air Intakes
In the EP environment gas turbines are an essential component in most

process operations. Gas turbines are high capex and opex items and due
regard should be made to the protection of these important assets; as an
example a replacement industrial RB211 gas generator costs over USD 6
million.
Gas turbines have extremely large airflows (e.g. RB211 - up to 70 m3(st)/s)
and any gas releases in their vicinity are likely to be drawn into combustion
air intakes. It is appreciated that a very large amount of gas would have to be
ingested to cause a catastrophic failure, however, any sustained ingestion of
gas through the combustion air intake will almost certainly have detrimental
effects on the long-term integrity of the equipment. Due to these very large
airflow rates, gas turbines effectively act as vacuum cleaners and it can be
argued that, irrespective of whether there is a potential to damage the engine
or not; a gas turbine air inlet is the perfect place to fit detectors for early
indications of gas releases.
Statistically, the gas turbine system is most vulnerable during the start-up
sequence. It is crucial that the correct light off fuel flow is attained for a
controlled ignition. With the small compressor air flow at light off the effect of
a relatively small amount of gas leakage into the inlet may have a significant
effect causing as a minimum a ‘heavy' light off and with the potential for an
explosion and major equipment damage.
With the incorporation of new gas detection instrumentation and careful
thought to installation, gas detectors in the combustion air inlet need not
compromise the overall system reliability, nor impose an additional main-
tenance burden. The use of beam detectors would also minimise the difficulty
in detection caused by stratification often experienced with gas releases.
However, the alarm and trip settings for the detection need to be realistic if
the protection is to be effective, e.g. 10% and 20% LFL.
The assessment of the potential for gas ingestion based on the location of a
gas turbine intake with respect to possible sources of gas such as the fuel gas
treatment skid, pipelines and equipment containing process gas, is of prime
importance. If it is concluded there is a possibility of gas entering a gas
turbine, and it cannot be guaranteed that the presence of gas will be detected
by other means, then gas detectors should be fitted to the combustion air
intake.
1.1.1.4 Gas Engines
The combustion cycle in a gas engine cylinder is very similar to that of an

automobile engine where an air/fuel mixture is drawn into the cylinder and
ignited by a spark. In sizes greater than 1.5 MW integral engines are utilised
where the engine and compressor share the same crankshaft as shown in
Figure 1.1.1-5. To gauge the importance of these units it must be noted that

Running Equipment
the greatest installed capacity of stationary engines over 2 MW is found in

integral engines, due to their extensive usage in the North American gas
transmission industry.
For sizes less than 1.5 MW the integral engine is not competitive with the
large production engine manufacturers like Caterpillar, Waukesha, and Su-
perior. These engines drive a separate compressor, as shown in Figure 1.1.1-6
and there is an extensive fleet of this type of compressor in service within EP.
The engines are often turbocharged, using the exhaust flow to boost the inlet
air pressure. In combination with aftercooling, the combustion air mass flow
rate is significantly increased and can improve output by up to 45% and fuel
efficiency by upto 8%. Turbocharging has the additional benefit of reducing
altitude derating factors. Both two-and four-stroke engines are used, with the
latter having a greater specific power output and being much more commonly
used. The former usually has fewer moving parts but is normally found used in
very large gas engines:
In order to meet increasingly stringent emission limits arising from environ-

mental concerns, manufacturers have begun to utilise ‘lean burn' technolo-
gy. This involves maintaining tight control over the air/fuel mixture, and
incorporating a prechamber arrangement to ignite an initial fuel rich mixture.
This ignition then acts as a torch to the main combustion area, providing an
even combustion process resulting in improved fuel efficiency and lower
emission levels.
As with gas turbines, gas engines come in discreet sizes resulting from the
various manufacturers' development activities. Product development is an
ongoing process but care must be taken in selection to ensure that prototype
designs are not chosen in the EP environment.
When a package is chosen using a separate engine and compressor care is

essential to match the characteristics of the driver and load. Often poor
performance and operational flexibility result when design speeds are com-
promised in order to provide a cost-effective direct drive package with no
gearbox. The minimal additional Capex required to provide a gearbox and
operate the package at design speed provides significantly enhanced oper-
ability and maintainability.
Ignition systems remain the area of most concern on gas engines. The solid
state capacitor discharge system has provided major reliability improve-
ments, but timing set-up procedures remain critical. A spark plug life of 5000
hours is now considered normal, but poor timing, overload or poor quality
fuel will drastically reduce this life.


Running Equipment

There is a concerted effort within the Group towards fitting shielded ignition
systems on gas engines. Early shielded systems provided poor reliability.
and were difficult to maintain and trouble-shoot. This produced considerable
disillusionment with the arrangement; however recent design improvements
by the shielded ignition manufacturers, have lead to improved system re-
liability and maintainability, which has resulted in improved acceptance by
Users.
Gas engines are designed primarily to operate on almost pure methane fuel.
The composite fuel gas usually used in EP environments has the effect of
reducing the octane rating of the fuel. The effects of a reduced octane rating
fuel are pre-ignition, detonation, cylinder head cracking and piston burning.
In order to avoid this, the engine manufacturer can provide an engine with
low compression cylinder heads, which results in a derate but allows satis-
factory operation on higher molecular weight gases.
Air filtration is just as important as in the case of a gas turbine if premature

turbocharger and cylinder wear are to be avoided. However the mass flow
rates are far lower than gas turbines and so filtration systems are far more
modest. The oil bath type has been found suitable for most applications.
1.1.1.5 Diesel Engines
These engines are used within the Group primarily for standby power gen-
eration and temporary facilities like remote water supply wells in desert
environments. In many applications, diesel engines burn crude oil, either
because alternatives are unavailable, as in a remote oil pipeline booster
station, or because this represents a sound economic alternative.
For a diesel engine, quality control of the fuel supply is very important,
especially if run on crude oil. When burning crude oil it is necessary to wash,
centrifuge and possibly heat the fuel to meet the manufacturer's specifica-
tion. It is desirable to test the engine at the packager's facility using a re-
presentative sample of the proposed fuel to ensure that no fuel filtration or
combustion problems are encountered.
For applications where an output greater than 700 kW is required, slow speed
engines running at speeds less than 1000 RPM are preferred.

Running Equipment
1.1.1.6 Fuel Systems
1.1.1.6.1 General
Many of the problems experienced with prime movers in field applications

can be traced to poor quality fuel. Reference should be made to Manufac-
turer's literature for comprehensive coverage of the fuel purity requirements
for all types of equipment.
1.1.1.6.2 Gas Fuel
Most oil/gas separator gases, or NGL stabiliser gases do not comply with
manufacturer's specifications in that they are saturated with heavy hydro-
carbons and liquid carry-over. These heavy ends and carry-over cause burn-
er fouling and combustion problems.
Figure 1.1.1-7 shows the recommended fuel gas system. Its features are:
(a) Dual fuel sources. For start-up, the available gas is used for fuel, when
the compressors are operational the higher pressure fuel will automa-
tically take over. The fuel should be delivered to the fuel treatment sys-
tem at least 20% higher in pressure than that required by the equipment.
If the gas is water-wet, check that hydrates will not form in the fuel
system.
(b) Filter separators with removable lambswool or cotton filters elements via
a quick-opening closure, followed by a liquid removal section-vane type,
(Peco, Peerless, horizontal type). Two in parallel, each 100% duty. These
will remove particulates, pipeline corrosion products and coalesce out
heavy hydrocarbons. The volume of the condensate tanks must equal the
largest liquid slug size expected. Alternatively, use a separate slugcatch-
er upstream.
(c) Heater (optional). Use only if hydrate formation is likely upon pressure
reduction.
(d) Compression (optional). Use only if supply gas pressure is too low for fuel
purposes. Use two, each 100% duty in parallel. Aftercooling may be
required, dependent on compression ratio and flash calculations after
pressure reduction.
(e) Pressure reduction. Use two 100% self-acting regulators; one set at de-
sired pressure, the second set 10% lower pressure in case of failure of
first regulator. Never rely on manual bypass for this critical service.


Running Equipment
(f) Liquid knockout. Use one SMS type separator. The volume for liquid
holdup should equal one of the filter separators.
(g) Fuel gas superheater. Use one heater, using an appropriate fuel to su-
perheat the gas 20 oC above its dew point, or as prescribed by the gas
turbine vendor, but the fuel gas temperature should not exceed 70 0C. If
there are fuel gas compressors, use gas/gas exchangers to superheat
the fuel gas using the heat of compression.
(h) Fuel gas manifold must be insulated; fit a drain valve at its furthest end for
draining condensate before starting any engine.
(i) If dehydrated, dew point depressed gas is available, (c) may be dis-
pensed with, (g) only if the temperature on pressure reduction ap-
proaches within 20 oCof the dew point or ± 5 oC, whichever is the higher.
(I) The recommended system is suitable forall equipment. For gas engines,
the composition after the superheater will determine whether engine
derating is required. Check with the engine Vendor, if CH4 less than 90%,
and C4 greater than 1.0%, then derating will be necessary.
1.1.1.6.3 Distillate Liquid Fuels
The principal problems with these fuels is contamination by particulates and

water, especially sea water. 1 dm3 of sea water will render 10 m3 distillate
fuel unfit for gas turbine use due to salt contamination. All distillate fuel which
has been bunkered by sea water must be regarded as salt contaminated.
Furthermore, bacteria can form on the sea-water/fuel interface which plugs
fuel systems. Once established, the bacteria are very difficult to eliminate.
Figure 1.1.1-8 shows the recommended distillate fuel storage and prepara-
tion system. The features of the system are:
(a) Two storage tanks, each providing a minimum of 24 hours settling of each
fill. Tanks have conical bottoms and sight glasses to ensure no water/fuel
interface, sampling points and a filling distributor to reduce turbulence
during filling.
(b) Transfer pumps and coarse filters. Filters should be 100 µm wire wound,
self-cleaning type with magnetic elements. Twin pumps and filters.
(c) Centrifuges, primarily to remove salt water. Use potable water to seal the
centrifuges, this will also wash the salt out of the fuel. Centrifuge man-
ufacturers will advise flow rates. Cleaning may be batch or continuous.


Running Equipment
(d) Day tank, cleaned fuel storage for at least 24 hours consumption, at
maximum rate. Spillback from engines to this tank. Sampling point. If
distillate fuel is used only as a standby, this tank should be kept full.
(e) Fuel supply pumps and fine filters. Fine filters 10 µm absolute type, paper
filters are unacceptable. The filters must have magnetic elements. All
pipework downstream of filters to be stainless steel.
1.1.1.6.4 Crude Oil and Residual Fuels
It is beyond the scope of this Handbook to provide designs for these systems,
as there are too many variables. There are several reputable suppliers of
liquid fuel treatment systems, e.g. Plenty Metrol, Newbury, UK, and Petrolite
Corp., London, UK. These firms will tailor the fuel treatment system to suit the
requirements of the engine and specification of the fuel oil supply - within
limits. Basically, particulates, alkali metals and water have to be removed or
inhibited, and the viscosity corrected to meet the engine Manufacturer's
specification.
1.1.1.7 Starting Systems
There are several proven means of starting gas turbines and IC engines,
viz:
(a) Battery and Starter Motor. Used for gas turbines up to 1500 kW, and IC
engines up to 300 kW. Trouble free and reliable, only the batteries require
regular checking for electrolyte and charge. Requires special starter
motors in Hazardous Areas.
(b) AC Motors. Used for larger gas turbines. (A single shaft gas turbine of 20
MW will require a 200 kW starter.) Cannot be used for `black start' of
generator sets.
(c) Hydraulic Starters. Hydraulic fluid is stored under inert gas pressure to
drive a hydraulic starter motor. The storage cylinder is recharged by a
motor driven or manual pump after each start. Used for emergency fire
pump or generator sets in Hazardous Areas. Reliable, but the hydraulic
fluid must be kept clinically clean.
On large gas turbines, an electric motor driven pump can be used to
supply hydraulic fluid to a starter motor on the turbine if there is insuf-
ficient space in the enclosure for a large electric motor.

(d) Gas Expansion Motors. Used frequently on gas turbines and IC engines
on compressor stations, or where high pressure gas is available. Highly
reliable, but the following points have to be considered:
(i) The starter gas must not be supplied from the fuel gas header. The
drawdown of pressure during starting can stall the already running
engines. (See also Figure 1.1.1-7.)
(ii) The starter exhaust must not be connected to a Common vent system.
The variable back pressure destroys the starter shaft seals when at
rest, gas will then escape into the engine enclosure. Use a separate
exhaust for each starter, discharging 2 to 3 metres above the air
filter, upward velocity disperses the plume of gas very rapidly.
(iii) Ensure proper lubrication for the starter - this is often over-
looked.
(e) Air Starting. Compressed air stored in vessels is used in an expansion

starter, similar to a gas expansion starter. Limited to starters of small
engines due to the large volume of air required for an expansion
starter.
(f) On IC engines, air starting is preferred to gas starting. Fires can be

caused by sparks off the driving pinion.
(g) Large (over 2 MW) IC engines are started by direct injection of air into the
cylinders. This method is very effective and reliable.
(h) The following points have to be considered in sizing a starting air sys-
tem:
(i) Size the air vessels for three consecutive starts between highest and
lowest allowable air pressure.
(ii) Use at least two air compressors, capable of recharging the vessels
in 30 minutes from lowest to highest pressure. The first compressor
must start at half the difference between highest and lowest pressure
is reached (for routine topping up), the second at the minimum start-
ing air pressure.
(iii) Alternate the first-starting compressor at monthly intervals.
(iv) Do not supply maintenance or service air from the starting air
vessels.
(v) Ensure that water and oil separators are working efficiently.

Running Equipment
(vi) Put a checkvalve in the air supply to each engine, and a bursting disc
downstream of it. This prevents starting air system explosions
caused by engine backfires during starting.
(i) All gas turbine starters must be rated for continuous service, Intermittent
rating will cause starter failures when the starter is run during crank-soak
washing of the engine air compressor. Also ensure that the starter en-
ergy supply is sufficient for the engine Vendor's required crank-soak
washing cycle.
1.1.1.8 Economics of Driver Selection
Most field installations have surplus gas for which there is no commercial
exploitation which can be used as fuel. Therefore low first cost, reliable
drivers are selected without concern for fuel efficiency. The most reliable
system in this instance is to use large gas turbines to generate electric power
and use motor drives for all machines, although it might represent a greater
capital expenditure, depending on the type of installation. Of significant effect
in the achievement of early commissioning, is the selection of equipment
which can be fully works tested prior to delivery.
Where fuel has a premium value, a more careful evaluation of the simplicity
and high reliability of a gas turbine will have to be set against the higher
efficiency of the IC engine. Unless there is a use for gas turbine exhaust heat
for process heating or district heating, the IC engine will invariably offer a
more satisfactory annual return if fuel gas is valued at its sales value.

1.1.2 Site Rating of Drivers
The power delivered by an engine or turbine is dependent on the mass flow

ingested. This mass, for a given volume, is dependent on the ambient tem-
perature and pressure.
Turbocharging of reciprocating engines can partially compensate for this
effect, but gas turbines cannot be similarly compensated, and the effect is
considerably greater due to the much larger mass flow rate.
Gas turbines are normally rated at ISO conditions (15 degrees Celsius,
1013.25 mbar and 60% relative humidity), excluding inlet and exhaust duct
losses. The factors in Table 1.1.2-1 give the derating to be applied for air inlet
and exhaust losses, losses due to altitude, ambient temperature loss, and the
margin to be allowed for fouling and wear. Conversely at ambients below 15
degrees Celsius, there is an equivalent power gain. However the increase
may be limited by load gearbox ratings or speed limitations on compressor
turbine or gas producer shafts.
The losses for intake and exhaust are based on a three stage inlet system,
with intake and exhaust silencing to 85 dBA, and no waste heat- recovery
system. The fouling and wear loss is to account for compressor fouling which
will reduce the power output between water washes, and the irreversible
wear on aerofoil profiles and tip seals which takes place during operation.
This gives a satisfactory margin at the design stage to ensure that the design
output can be achieved throughout the period between overhauls.
Where there are wide extremes of ambient temperature, such as desert

locations, the plant design temperature chosen may be less than the highest
maximum temperature. To design for the absolute maximum could mean the
over design of the plant for most of the year. By studying meteorological

Running Equipment
records, the approximate number of hours per year that any chosen design
temperature will be exceeded can be calculated. The shortfall of turbine
power under these conditions can be compared with the loss of throughput,
and it may be possible to increase throughput during cooler periods to make
up the shortfall. However, it is often the case than production requirements
will be for rated throughput irrespective of ambient conditions and so the
turbine must be rated for the highest maximum temperature. If the plant is
designed for a temperature less than the absolute maximum temperature air
heat exchangers must be designed for the absolute maximum, otherwise
shut-downs will occur as a result of over-temperature whenever the design
temperature is exceeded.
1.1.2.2 Gas and Diesel Engines
There is no internationally accepted standard for rating of these engines,

although the ISO rating is achieving greater acceptance. Thus, at present,
derating for altitude and temperature must be taken from the Manufacturer's
name-plate rating. Table 1.1.2-2 applies to both diesel and gas engines.
Manufacturer's name-plate ratings customarily include intake and exhaust

losses; verify that they do before accepting any figures. Also, most engines
have a flat rating up to 150 m altitude and 30 oC ambient temperature.
Note that at high ambient temperatures (> 40ºC) and for high altitudes
(> 1000 m) gas engines become very fuel sensitive. If the C3 + content of the
fuel exceeds 1.0 mol% further derating will be necessary (see also 1.1.1.6), in
consultation with the engine Manufacturer.
Some turbocharged/intercooled engines can achieve sea level rating at any

altitude and ambient temperature. Examine Vendor's claims carefully, high
pressure turbocharging leads to a very narrow torque range unsuitable for
all compressor applications and unstable governing on alternator drivers.

In addition, derating of turbocharged aftercooled units also depends on the

temperature of the coolant supply.
Manufacturers literature should be consulted for this type of application as

there is no standardised approach among vendors.
1.1.3 Gas Consumption
See Volume 7 PROCESS ENGINEERING, Part II, Section 2.1.1-3.

Running Equipment
1.2 COMPRESSORS
1.2.1 Types of Compressors and their Application
1.2.1.1 Application Limitations
Within the EP environment, gas compressors have the following field appi Ca-
tions:
- Gas lift
- Gas transmission
- Gas injection
- Welihead separator boosting
- Casing head gas collection
- Refrigeration
To carry out recompression under these diverse applications, the following

types of compressor are used.
Centrifugal
- horizontal split
- barrel type
Positive Displacement
- Reciprocating
- low speed integral
- low speed separable
- high speed separable
- Rotary
- Screw type
Depending on application, and more especially discharge pressure and flow,

equipment selection may be limited to certain types of compressor. Figure
1.2.1-1 summarises the usual selection of equipment type within the EP en-
vironment.
Where operating conditions fall in the transition zone between equipment

types, a careful study needs to be made of operational requirements, future
field developments, and the sensitivity of anticipated operating points.


Running Equipment
1.2.1.2 Centrifugal Compressors
1.2.1.2.1 General
Centrifugal compressors are the second most widely used compressor after
reciprocating machines. A centrifugal compressor functions by increasing
the velocity of the gas as it passes through an impeller. The velocity is then
reduced in the diffuser at the same time as the temperature and pressure
increase. Very high speed compressors are available, with speeds up to
30,000 RPM, and discharge pressures of up to 700 bar are achievable.
They offer compactness, simplicity, and ease of maintenance, but require a

high level of sophistication in design and selection. A centrifugal compressor
in a well designed and operated installation should operate for no less than
five years without major maintenance intervention. Depending on the type of
driver, availabilities of up to 99% are possible. Figure 1.2.1-2 shows a cross-
section of a large horizontal split casing machine.
In general centrifugal compressors are preferred for most duties because of

their simplicity but minimum flow limitations must be recognised.

1.2.1.2.2 Limitations
Impeller stresses limit the tip speed to a maximum of approximately 300

metres/second. Keeping the impeller and diffuser widths to a practical min-
imum limit of 4 mm results in a minimum discharge flow near 300 m3/h
(actual). The 4 mm limit is chosen in order to limit the effects of fouling due to
carryover in the gas stream. Narrow impellers and diffusers mean that a
small build up of fouling product rapidly affects compressor performance,
and makes the cleaning operation difficult during overhaul activities. This
can lead to imbalance and suboptimal performance after restart.
The maximum tip speed of the first stage impeller should not exceed 85% of
the speed of sound in the suction gas. Higher velocities lead to choking and a
reduction in throughput. At pressures up to 100 bar the maximum head which
can be developed `by one impeller is approximately 30 kJ/kg and reduces
significantly above this pressure. This maximum allowable impeller head
determines the number of impellers required to meet the desired pressure
ratio. However the maximum number of impellers is restricted by rotor dy-
namic considerations, and standard manufacturers casing designs which
limit the number of stages which can be accommodated. In most cases there
exists an upper limit of 9 impellers per casing.
Traditionally a maximum allowable discharge temperature of 150 degrees

Celsius has been applied also to centrifugal compressors. This has been
done to take account of the range of upper limits allowed by manufacturers
(some are as low as 170 degrees) and-because occasionally some produced
crudes have been found to be susceptible to cracking at relatively low tem-
peratures. However many designs operate with discharge temperatures well
above, the 150 degrees' level without any adverse effects. The 150 degrees
limit should therefore be used as a basis for design, but any decision to relax
this limit -must be made in consultation with SIPM Rotating Equipment Spe-
cialists, with Process Engineering Specialists and with the compressor man-
ufacturer:
1.2.1.2.3 Anti-Surge Systems
A centrifugal compressor, unlike a pump, is operating with a compressible

fluid. At some operating point under increasing discharge pressure and
decreasing flow, a condition will be reached where the compressor impeller
is unable to maintain the pressure rise. The discharge gas then re-expands in
the direction opposite to normal flow, causing vibration, noise, and in severe
cases significant compressor damage. In order to avert this phenomenon

Running Equipment
centrifugal compressors are fitted with anti-surge control systems which

ensures there is always an acceptable minimum suction flow to the machine.
In multi-stage arrangements, each stage of the compressor requires its own
protection system.
The compressor system control arrangement is covered in Section 1.2.2. The

response of the control system is influenced significantly by the layout of the
piping system. The anti-surge line pipework should be situated physically as
close as possible to the discharge of the machine in order to minimise the
inventory of high pressure gas which the antisurge system must accept
during abnormal operating conditions.
A process simulation must be carried out on all projects involving centrifugal

compressors, during the design phase, to ensure that the intended anti-surge
system will respond satisfactorily under all transient conditions and keep the
equipment within the safe operating envelope. Considerable research has
been carried out to develop sophisticated process models, and several con-
sulting companies are now in a position to provide service to EP require-
ments.
During commissioning of a new installation, the surge control system needs

to be calibrated according to the actual field conditions. To enable this to be
easily carried out, the field design needs to include a station recycle system
capable of controlling the entire flow of one compressor train. At different
speeds throughout the compressor operating speed range, the onset of
surge is carefully established under controlled conditions, and the anti-surge
controller calibrated to ensure a minimum flow, normally 10% higher than
the surge flow, at all times. The surge controller needs to operate rapidly to
ensure that surge is not encountered during machine steady state or trans-
ient operation, but needs to be able to maintain a reasonable ability to
modulate at the minimum flow condition to limit process upsets. It is recom-
mended that the services of a representative from the surge controller sup-
plier be utilised during initial commissioning activities.
With the development of micro-based computer systems, more sophisticated

anti-surge systems are now being developed. Systems which make com-
pensation for changes in moleculair weight and suction gas temperature are
now available, together with memory of flow profiles as the machine ap-
proaches surge. This has allowed the traditional 10% surge margin to be
reduced, which provides improved compressor operating envelopes while
still maintaining adequate machine protection.

1.2.1.2.4 Shaft Sealing Systems
The compressor shaft seal is designed to prevent the process gas from
escaping along the shaft to atmosphere. In non-toxic and non-hazardous
environments, a labyrinth seal is utilised which allows a small leakage of gas
to atmosphere. In the EP environment with the hazard of allowing hydrocar-
bons escape to atmosphere, a seal arrangement is utilised to contain the
process gas within the compressor. The traditional method has used one of
two types of seal which require the injection of sealing oil at a pressure
greater than that- of the gas pressure in the sealing chamber. However the
recent developments in dry gas- seals have effectively eliminated the re-
quirement for seal oil systems from new installations. The design of seals
and seal systems is covered in API 614 and API 617.
(a) Contact Seals
Contact seals use a fixed and a rotating ring, between which a floating carbon
ring is placed. Seal oil injected at the outer periphery of the rings at a
pressure greater than gas pressure forces itself between the rings and the
carbon seal, thereby cooling the contact faces and preventing the escape of
gas. This oil is drained from the casing to a sour seal oil trap. The contact seal
has the advantage of providing a positive seal even when stationary, and so
no back up seal oil system is required. Also the seal oil flows are low and so
extensive degassing facilities are seldom required, and the seals are rela-
tively inexpensive. Seal layout is shown in Figure 1.2.1-3.
However this seal type is not suitable for high pressure duty or with a dirty
gas stream, and is prone to overheating in high speed applications.

Running Equipment

(b) Oil Film Seals
Seal oil is injected into the seal chamber and passes between a pair of rings,
which cannot rotate but are free to move in a radial direction. The seal oil at a
pressure slightly higher than gas pressure flows through the clearance be-
tween the inner ring and the shaft, preventing the escape of process gas. This
oil is drained from the sour seal oil chamber to the sour seal oil trap. A
greater quantity of seal oil flows between the outer ring and the shaft toward
the atmospheric drain chamber. A labyrinth is placed to prevent seal oil
entering the bearing chamber if the seal and lube oil systems are segregated.
Film seals are used in high pressure or particularly hazardous applications,
or where the process gas is dirty. Oil film seal layouts are shown in Figure 1.2.1-4.

Running Equipment

Without seal oil pressure an effective seal cannot be achieved, so back up

arrangements are essential during compressor shut-down and casing de-
pressurising. Also when gas pressures are at revels significantly less than
design, there is a tendency for the seal rings to overheat. This often neces-
sitates the provision of an external source of gas during start-up which
creates an artificially high reference gas pressure, to ensure adequate seal
oil flow for cooling purposes.
Both oil film and contact seals rely on a supply of seal oil controlled at a
pressure slightly above the reference gas pressure of the compressor. This
is normally achieved using an overhead seal oil tank and applying reference
gas to the top of the vessel. The required differential pressure (determined by
the compressor vendor) is established by the physical elevation of the vessel
above the centreline of the compressor, with seal oil flow being regulated to
maintain a setpoint level in the vessel. The seal oil tank is sized to ensure that
during a loss of power shut-down, adequate sealing and cooling is achieved
during rundown and depressurisation. An alternative but less suitable ar-
rangement is to rely on a differential pressure control valve to control the
seal oil systems, and provide auxiliary DC motor driven seal oil pumps for the
loss of power shut-down situation.
Whether oil film or contact seals are used, there will always be a flow of oil
which mixes with the process, gas and drains to the sour seal traps. If the flow
is very low, it may be returned to the reservoir with degassing taking place
there. However, this will lead to a gradual deterioration in flash point and
viscosity of the complete oil charge, and possibly provide an explosive en--
vironment in the air space in the reservoir, so some active method of de-
gassing is always required in the EP environment. The gas may also become
contaminated with impurities including hydrogen sulphide and so may be-
come acid and develop sludge deposits in the reservoir and machine cavi-
ties. Contamination can eventually result in the ineffective operation of tilting
pad bearings, `resulting in increased vibration levels.
Sour seal oil recovered from the traps can be actively degassed by allowing it
to reside for some time in a heated degassing tank before returning it to the
oil reservoir. This degassing may not be fully effective and further treatment
may be necessary. There is considerable EP experience with nitrogen or air
sparging and vacuum stripping although the last option if often found to be
maintenance intensive.
With the high cost of lubricating oils, recovery of sour oil is almost always
justified, rather than disposal to be bulk separators.

Running Equipment
In the situation where the vendor offers a seal oil system which is entirely
independent from the lube system, this has the effect of increasing the rotor
bearing span with the attendant adverse effect on rotor dynamics.
(c) Dry Gas Seals
Gas seals were developed over twenty years ago and until recently, only
three manufacturers made them in two basic designs. The first design, the
Raleigh pad type, has scallops in one face which traps gas in the hollow area
as the seal turns, and as pressure builds up forces the faces apart. When the
faces become too far apart, the pressure falls allowing the seal springs to
close the faces and develop a dynamic balance when the gap between the
faces is maintained at approximately 50 microns.
The second design uses spiral grooves cut in one face which face the di-
rection of rotation so that gas is forced down the groove producing a pressure
rise. The seal then operates in the same manner described for the Raleigh
pad seal except that the sealing gap stabilises between only 3 and 10 mi-
crons. User experience has shown that the second design is more stable, can
operate better under off-design conditions and results in extremely low lea-
kage rates. The only complication is that the seal is unidirectional and so
dedicated seal faces are required for each end of a compressor. A typical dry
gas seal cross-section is shown in Figure 1.2.1-5.

Running Equipment

The prime motivation for the introduction of dry gas seals instead of using the
traditional seal oil arrangement has been on the grounds of safety. Informa-
tion from both manufacturers and Users indicates that over 70% of all cen-
trifugal compressor operational faults are attributable to lube and seal oil
systems. These problems can lead to fires and leakage of toxic gas, both with
major safety and environmental implications. A seal oil system involves
significant hardware, is often difficult to troubleshoot, and provides a weight
and space impact, which may be important in an offshore environment. A dry
gas seal system for a new installation is less expensive and User experience
to date indicates that the systems are at least as reliable as oil systems, but
have a much lower risk of fire or gas leakage.
Therefore justification of dry gas seals on new installations can also be made
on the basis of capital savings, especially if carried out in conjunction with
magnetic bearings as the complete compressor lube oil system can be elim-
inated. On existing machines with seal oil systems, retrofit of dry gas seals
can be difficult to justify unless serious reliability problems exist with the seal
oil system which cannot easily be resolved, or there are possible major
consequences resulting from a gas leakage.
Dry gas seals are now considered standard for all new EP equipment pur-
chases with guidelines for incorporation outlined in the following report:
Report EP9O-0575 October 1990 (Revision 1) - Information Note - Dry Gas

Seals For Centrifugal Compressors

1.2.1.2.5 Bearings
The type of bearing utilised in a centrifugal compressor is dependent on the

duty of the equipment.
For large relatively slow speed equipment operating well below the rotor first
critical the standard liner or sleeve bearing may be utilised. The bearing may
be pressure fed or oil ring lubricated.
At higher speeds, in association with longer and lighter rotors standard liner
bearing may be machined to establish ‘dams' in operation which convert oil
film velocity into static pressure. These dams act to damp the effect of half
frequency oil whirl and rotor induced vibrations. This type of bearing is
always pressure fed.
However the majority of compressors operate above the first critical speed,
and the tilting shoe bearing is usually then standard. This bearing provides
excellent damping capability and can be fitted with bearing metal tempera-
ture detection devices in one or more of the shoes.
The most recent innovation is the introduction of the magnetic bearing which
uses magnetic forces to support the rotor and damp resonant vibrations.
Shaft position is detected by position sensors, which can also serve as vi-
bration probes. This position information is then used to vary the magnetic
forces imposed by an array of electromagnets located in the place of the
normal bearing housing. Figure 1.2.1-6 shows one layout for a magnetic
bearing arrangement. In combination with dry gas seals, this then produces
the `dry' compressor, which is rapidly gaining acceptance with compressor
Users. However caution must be used as the magnetic bearing arrangement
on a flexible rotor can create a complex rotor dynamic response. A careful
review of the rotor dynamic study by the compressor manufacturer and
Rotating Equipment engineering is required if this arrangement is being
considered.

Running Equipment

1.2.1.2.6 Centrifugal Compressor Check-List
The following points require consideration when selecting centrifugal com-

pressors for field gas compression up to 100 bar discharge pressure. For gas
injection, more stringent conditions apply (see 1.2.1.5.5).
(a) Duty Conditions
(i) Check variation of gas composition with time and production rate. Check
temperature range; summer, winter, day and night. Check that all oper-
ating conditions fall within the compressor operating envelope. On re-
frigeration compressors check that there is sufficient head and power to
condense the refrigerant at the maximum condensing temperature at
the highest ambient temperature conditions.
(ii) If there are solids in the gas stream use a suction filter separator with a
25 µm filter. Liquids up to 150 g/m3 can be tolerated provided that they
are free of dissolved solids (salts in water) and are in particle sizes less
than 100 µm.
(iii) The Head Capacity curve for a centrifugal compressor should normally
have a shape which provides the following margins between the duty
point and surge:
- Head rise to surge: 5% mm.
- Flow stability: 25% mm.
(iv) If the gas contains hydrogen sulphide, reference should be made to

NACE standard MR-01-75 sections 1 and 10.7. If corrosion rates cal-
culated for water-wet gas containing CO2 exceed 0.2 mm/a then corro-
sion resistant materials should be used; it is preferable to dehydrate
before compression.
(b) Impellers and Shaft
(i) Reject any Vendor who is unable to perform a detailed torsional and
lateral critical speed analysis for the entire rotating train. The analysis
must consider shaft seal stiffness and aerodynamic cross-coupling.
(ii) All-welded, solid machined, or cast impellers are preferred.
(c) Casing and Nozzles
(i) Check that the casing design pressure will withstand the pressure de-
veloped at maximum speed, suction pressure and molar mass.

Running Equipment
(ii) Check that the inner surfaces of the compressor casing are sufficiently
smooth to prevent seizing of the inner bundle or deterioration of the
elastomer seals.
(iii) In multi casing line-ups, the casing nearest the driver should be hori-
zontally split, to facilitate maintenance.
(iv) On all casings, down-connected nozzles improve maintenance acces-

sibility.
(d) Lubricating and Seal Oil Systems
(i) For field applications, combined lubricating and seal oil systems are
acceptable unless the process gas contains a level of H2S which cannot
be adequately removed by the sour seal oil reclaimer.
(ii) The driver and gears may be lubricated from the compressor system,
but ensure that aircraft derivative gas turbine lubricating oil and com-
pressor lubricating oil do not mix. Check also that the driver/gear man-
ufacturer designs for the oil quality and viscosity in the compressor
system.
(iii) Do not reduce the eight minutes lubricating oil reservoir residence time
called for in API Std 614.
(iv) To prevent corrosion in systems in the tropics or in intermittent use,

consider the use of all-stainless steel lubricating and seal oil sys-
tems.
(v) If the reservoirs are in the base plate, ensure that the Vendor has
allowed for thermal expansion which will affect alignment.
(vi) Attach flame trap coalescers on all breathers and vents with a separate
drain line back to the reservoir.
(e) Drive Train
(i) Use direct drive without gearboxes wherever possible, but do not com-
promise Vendor's proven operating speeds in fulfilling this require-
ment.
(ii) Provide a means of turning the drive train for maintenance.
(iii) Have the Vendor supply instruments for hot alignment checking of the
drive train.

1.2.1.3 Reciprocating Compressors
1.2.1.3.1 General
Reciprocating compressors are widely used throughout EP and are catego-

rised as positive displacement machines. This means that a quantity of gas at
suction pressure is drawn into a cylinder and compressed by a piston until
the pressure rises to discharge pressure, when a valve opens allowing the
gas to leave the cylinder at discharge conditions. Figure 1.2.1-7 shows the
typical horizontally opposed type separable compressor used for process
gas compression. High pressure ratios can be achieved by connecting a
number of cylinders in series and cooling the gas between the stages of
compression. Figure 1.2.1-8 shows a cross-section of a typical low pressure
cylinder for use up to 5 bar and a high pressure cylinder capable of handling
discharge pressures of up to 100 bar.
For a given stroke of standard frame size and running speed, the capacity is
limited by cylinder diameter. This is turn is limited by allowable working
pressure. To increase capacity, it is possible to connect several cylinders in
parallel. Therefore the reciprocating machine is very flexible in application
during the equipment selection process.

Running Equipment


Running Equipment
1.2.1.3.2 High-speed/short-stroke versus low-speed/long-stroke reciprocating compressors
Reciprocating compressors may well be the most widely used of all gas
compression equipment in E&P oil and gas field production installations.
Reciprocating machines certainly provide the widest range of sizes and
types, and are still the most efficient for most oil and gas applications when
suitably matched to the required capacity.
There are two basic groups of reciprocating compressors:
A) Moderate duty (packaged) compressors
- These units are not designed to API 618.

- They are designed for reliable operation over a reasonable service life, but
should not be installed where continuous full-load, long-term operation is
required.
Note: This does not mean that they may not operate for extended periods at
full load. It does mean that the operating cost and maintenance burden will
be greater than for heavy duty compressors.
B) Heavy duty compressors
- These units are designed to API 618.

- They are designed for reliable and safe operation for a minimum service
life of 20 years.
For E&P applications we can differentiate between two sub-groups of heavy

duty compressors:
1) Low-speed, say <400 rpm

2) High speed, say <1200 rpm
Both low-speed and high-speed machines are well worth considering for a
particular application. However neither can satisfactorily handle liquids and
dirt entrained in the gas which will destroy lubrication and cause excessive
wear.
In any operational comparison of gas compression equipment, the prevailing
local service conditions can be more detrimental to availability than any
other aspect, regardless of the degree of reliability of the detailed compres-
sor design. If these service conditions are not within certain acceptable
limits, any compressor installation, whether employing high-speed or low-
speed machines, is bound to fail.
The majority of E&P reciprocating compressors currently in operation are
low-speed units. Low-speed compressors have been the industry standard
and yet operating costs, and in particular long-term maintenance costs, have

not been fully evaluated, basically because of insufficient operating history

and manpower constraints.
While there is little argument over the critical first cost advantage of high-
speed units, many engineers are uncertain about the overall economy, par-
ticularly since the Group's early experience indicated that maintenance
costs were higher.
Acceptable service conditions, together with good operation and mainten-

ance practices, are important factors in achieving successful results. If the
service conditions are not right, then in practice the reputation (good name)
of any manufacturer is liable to be tarnished.
It should be recognised that with modern reciprocating compressors, both
high- and low-speed type, provided that they are designed and specified to
API 618, recent experience has demonstrated a high degree of availability
and unit reliability.
The current experience of most users of high-speed compressors indicates
that operating costs are not necessarily higher in comparison with those of
low-speed machines, provided that they are both heavy duty designs. It
should be realized that the basis for comparison, however, is frequently
inequitable.
Regardless of any (`believed') deficiencies of the high-speed reciprocating
compressors, their initial low investment cost, for the same installed capac-
ity, will overcome any higher operating costs, if any, that can be reasonably
anticipated.
In lower capacity applications, high-speed units have a good economic po-
tential which could result in low-cost field installations.
Historically low-speed reciprocating compressors were installed directly on

concrete foundation blocks and supplied-in a number of separable parts. The
units were built together in the field at the job location. Nowadays they come
as skid-mounted and completely packaged units. The compressor manufac-
turer has full unit responsibility.
Since the introduction of the modern high-speed reciprocating machines, the
compressor driver and the majority of required auxiliary equipment are
mounted on a common structural steel skid to form a complete, compact
compressor plant.
From a technical point of view there does not seem to be a lot of difference.
Both high- and low-speed units should be designed in accordance with the
same basic criteria. This implies that stresses and resulting reliability and
wear are similar.
Probably the most important aspects in the mechanical design are the cyclic

Running Equipment
stresses because these determine the reliability of the compressor. Mate-

rials used and actual stress levels are the governing criteria.
Therefore similar allowable stress levels must be used for all components of
specific component groups, such as crankshafts, frames, cylinders, connect-
ing rods, threaded fasteners and perhaps most important the piston rods.
Safety factors must be calculated for each component on the basis of a
statistical failure probability during an infinite life, which makes these design
margins independent of the rotational speed of the compressor crank-
shafts.
If the above design requirements are met, then in view of the similarity in
design it is considered that the following advantages and disadvantages of
high-speed versus low-speed machines are the only important criteria in
determining optimal suitability for a given application.
Advantages of High-Speed vs Low-Speed:
* Main driver
- A fast running (up to 1200 rpm) gas engine or electric motor is of smaller
size and weight and first costs are low.
- A gas engine has inherently variable speed, hence built-in variable ca-
pacity for the compressor.
- The compressor can be directly coupled, no speed-reducing gear being

necessary and thus saving installation space.
- Standard prime movers of any reputable and competitive manufacturer

can be selected.
* Compressor
- Because of the limited average piston velocity, which is near the same
value as for low-speed reciprocating compressors, high-speed recips are
not significantly smaller.
- The higher rotational speed and related shorter stroke result in shorter
connecting rods and cylinder lengths. Hence better compactness, allowing
for a smaller cylinder head-to-head width with reduced installation space,
weight and foundation costs.
- Due to compactness and smaller width the compressor is suitable for

single-skid installation together with the driver.
Disadvantages of High-Speed vs Low-Speed:
* Application
- Due to the higher speeds, reciprocating weights have to be minimised in

order to limit the inertia forces. This results in reduced maximum cylinder

diameters which in turn limits the piston displacement for a given number
of cranks.
- For higher capacities therefore more units, or a single low-speed unit, are
required.
- In conjunction with a maximum number of 4 (or 6) cranks per frame, a

reduced overall capability with respect to capacity, power input and rod
loading is available.
- Valve life is adversely effected if service conditions are not optimum.
The latter, it is believed, is the only real practical difference between high-
speed and low-speed machines. However, since the introduction and suc-
cessful operation of compressor valves which are more tolerant to liquids
and dirt in the gas stream, there is less reason for the long-standing reser-
vations against the use of high-speed reciprocating compressors.
In Summary (High-Speed)
- The high-speed reciprocating compressor has good economic potential in

many E&P applications.
- Improved unit compactness, suitable for easy relocation of unit.
- High power-to-weight ratio.
- Good running balance inherent in the balance-opposed compressor de-

sign.
Provided that service-conditions are reasonable, and scrubbers are correctly

sized and designed with appropriate internals, the above results in an ìdeal'
compressor for low-cost production field installation.

Running Equipment
1.2.1.3.3 Capacity Control
After selecting the sizes and numbers of cylinders for a particular duty, there
is usually a requirement to have the capability of varying the throughput of
the compressor in normal operation. The speed can be varied throughout the
operating range of the driver, if variable speed, and the volumetric efficiency
can be changed by opening or closing fixed pocket clearances. Further turn-
down can be achieved by use of a station recycle control valve but this option
is not attractive if overall efficiency levels are important. Some designs allow
for the use of valve depressor or plug-type unloaders, but these are not
recommended for EP use as they increase significantly the complexity of the
machine and impose maintenance penalties.
A reciprocating compressor has a constant torque independent of speed, and

so if variable speed is used for capacity control the driver must be capable of
providing sufficient torque throughout the speed range.
With a relatively large number of moving parts, reciprocating compressors

are inherently maintenance intensive with the result that availabilities are
seldom better than 90-94% and maintenance costs are high. It is important to
include this data into any system effectiveness study to establish sparing
requirements for essential equipment. However it should be noted that with
developments in non-metallic materials for valve plates, rider rings, and
pressure packings, and an awareness by manufacturers that operators ex-
pect improvements, means that gradual availability improvements will con-
tinue to be made.
1.2.1.3.5 Distance Pieces
The functions of the distance piece for a reciprocating compressor in hydro-

carbon duty can be described as follows:
- to prevent hydrocarbon carryover from the cylinder to the frame end.

The presence of hydrocarbons in the frame end will reduce lube oil vis-
cosity and create a potentially explosive environment.
- to allow disposal of leakage gas from the pressure packing to a flare or

vent system.
- to allow introduction of buffer gas and safe disposal of pressure packing

leakage in sour applications. This is especially important when equip-
ment is located in a closed environment.
- to allow monitoring of pressure packing performance.

For EP applications, the `type C' distance piece arrangement as outlined in

API 618 is recommended for both sweet and sour gas applications. This
utilises the two long compartment layout with provision for external gas
buffering if required. Figure 1.2.1-9 shows the layout of the `type C' arrange-
ment. The two separate compartments are required to ensure segregation of
the cylinder and the frame end and to provide adequate venting and purging
arrangements. The precise layout is dependent on the individual application,
but the three possible arrangements for gases other than air or nitrogen are
clearly laid out in the relevant technical specification - DEP 31.29.40.31-Gen.

Running Equipment

1.2.1.3.6 Pulsations
The reciprocating compressor delivers a pulsating flow of gas. If the pipe-

work and structure should be in a resonant condition with the pulsations of
this flow, severe vibration as a result of piping or acoustic resonance will
occur. The pulsations can be smoothed at source by providing pulsation
bottles at the compressor cylinder suction and discharge nozzles. When
determining the size of the pulsation bottles during the design phase, the
piping design around the compressor is subjected to a computer simulation
study. The simulation allows the size of the pulsation bottle to be varied in
order to bring the pulsation magnitude to acceptable limits. Pulsation bottles
are designed to be devoid of all internals. It is always advisable to have the
manufacturer's simulation checked by an independent third party, to ensure
that the vendor is proposing adequately sized pulsation bottles. The simu-
lation should include a study of the effects of changes in pressure and mo-
lecular weight to ensure that any changes in process conditions do not lead to
a reduction in damping. Such studies cannot guarantee a pulsation free
operation and provision must always be made during commissioning to test
actual pulsation levels and allow for the modification to pulsation bottles or
pipework clamping and supports. In addition refer to section 1.4.
1.2.1.3.7 Foundations
The conversion of the rotational crankshaft motion to the reciprocating piston

movement produces unbalanced forces and couples. These oscillating
forces and couples occur in both horizontal and vertical planes and occur at
running speed and twice running speed in the directions shown in Figure
1.2.1-10. Counterweights on the crankshaft can reduce these forces but gen-
erally only shift them from a horizontal to a vertical plane of action. The worst
unbalance is in the case of a two cylinder compressor, while a six cylinder
compressor is in perfect balance. Foundations must be designed to accept
the resultant forces developed. While this is relatively simple for an onshore
development, in offshore applications the unbalanced forces and couples
must be carefully reviewed as they may excite platform resonant frequen-
cies. Rotating counterweights running at twice crankshaft speed can be used
to balance residual secondary forces and couples but this significantly in-
creases the mechanical complexity. Computer simulations can be used to
evaluate the effect of unbalance forces on platform structures. In addition
refer to section 1.8.

Running Equipment

1.2.1.3.8 Valves
Careful attention must be made when selecting reciprocating compressor

valves, as these contribute more than any other single factor to poor machine
availablility in service and often account for more than 50% of machine down
time. Valve performance can be simulated under the anticipated process
conditions to ensure the optimum combination of plate mass and spring rate
combined with the appropriate damping plates, if these are required.
The primary cause of valve failure is as a result of carryover of liquid or solid

particles in the gas stream. Effective suction and interstage scrubbers are
essential if satisfactory valve life is to be achieved.
Proprietary valve manufacturers are normally used within EP, but with the
continued development of non-metallic valve components a wider range of
options is now available. In applications where liquid carryover cannot be
avoided, non-metallic valve plates have proved extremely effective in pro-
viding satisfactory valve life.
Discharge temperature measurement from each stage should be monitored,

as an increase in temperature can be a useful indication of impending valve
failure.
The maximum allowable discharge temperature of 150 degrees Celcius fol-

lows the recommendations of API 618 and is intended, among other reasons,
to ensure that adequate lubrication can be maintained on the cylinder wall. In
cases where non-metallic components are used, lower limits on discharge
temperatures may be imposed by the valve manufacturer.
High-speed compressors up to 1000 RPM may be used in non-critical duty

when discharge pressures are below 80 bar, so long as an average piston
speed of 4 metres per second is not exceeded. In all other cases the max-
imum recommended compressor speed is 375 RPM with a maximum aver-
age piston speed of 4 metres per second.

Running Equipment
1.2.1.3.9 Reciprocating Compressor Check-List
The following subjects provide a check-list for the selection of field recip-
rocating compressors with discharge pressures up to 100 bar. For gas in-
jecfion, see 1.2.1.5.5.
(a) Duty Conditions
(i) Always check the maximum variation in gas flow, composition, pres-
sure and temperatures. Plot these on a performance map to ensure that
the selected compressor and driver can meet all expected operating
conditions. It is always easier to reduce the capacity of a reciprocating
compressor than to increase it.
(ii) Ensure that under no circumstances liquid can enter the compressor.
Suction scrubbers must be big enough to catch slugs forming in station
pipework. Use vane type separators or filter-separators with renewable
elements.
If a filter separator is to be employed, it may also require protection
from slugs by means of an upstream slug catcher.
(iii) With H2S and CO2 in water-wet gas it is preferable to dehydrate before
compression, but is frequently uneconomic. Corrosion-resistant mater-
ials must be used; for H2S they must comply with NACE specification
MR-01-75.
(b) Cylinders, Piston Rods and Distance Pieces
(I) Use lined cylinders.
(ii) Use top suction/bottom discharge cylinders.
(iii) Non-lubricated cylinders should not be used on field compressors.
(iv) Piston, rods should be in chrome alloy steel. Where NACE material is
specified the piston rod must still be surface hardened in the packing
area.
(v) The piston should preferably be shorter than the stroke length to ensure
adequate cylinder-wall cooling.
(vi) Never use tail-rod cylinders.
(c) Piston Rod Packing
(i) Check that sufficient access is provided to get at the packings. This is
often obstructed by process or cooling piping.

(ii) Check that the packing vents are of sufficient size. For a leak of 0.1% of
throughput, the vent system back-pressure must be less than 2 bar.
Packing vents from the cylinders on one compressor may be mani-
folded together. However the vents from several compressors should
never be manifolded together.
(iii) Don't use non-lubricated packings on field gas compressors.
(iv) Cooled packings are required for lubricated PTFE packings at pres-
sures above 50 bar, all packings above 85 bar. If closed circuit oil or de-
mineralised and inhibited water cooling systems are used, 10 µm fil-
tration is required.
(d) Frame Lubrication
(i) If electric power is available provide two electrically driven lubricating

oil pumps - a main and a 100 percent unit standby.
(ii) Aluminium bearings and/or crosshead shoes should not normally be

used. However, if the compressor has aluminium bearings or cross-
head shoes, use 5 µm filtration. Paper filters must not be used.
(iii) Lubricating oil temperature control is sometimes not necessary. If no

temperature control is considered necessary, ensure that the correct
oil viscosity is selected for the prevailing temperature.
(iv) Ensure that, as far as possible, the driver, gear unit and compressor
frame use the same lubricant.
(e) Cylinder and Packing Lubrication
(i) If electric power is available use an electric motor-driven lubricator.
(ii) The block system of lubrication (Trabon) must not be used on pressures
exceeding 50 bar or if the lubricant has a viscosity exceeding 100 mm2/s
(cSt) at the pumping temperature. If so, use a pump-to-point lubricator.
The pressure differential on points connected to the same pump and
distributor in a block system must not exceed 20 bar.
(iii) If the lubricant has a viscosity exceeding 100 mm2/s (cSt) at atmos-
pheric temperature, heat the lubricator box. This will improve lubrica-
tor pump life.
(f) Drive Train
(i) Insist on the Vendor performing a torsional analysis on the complete

drive train including all starting and stopping modes. The Vendor shall

Running Equipment
provide all flywheels and torsionally flexible couplings to meet the

torsional and electricity supply stability requirements.
(ii) Avoid gearboxes if possible, but do not compromise driver or com-

pressor design running speeds in fulfilling this requirement.
(iii) Gas turbine drivers for reciprocating compressors are acceptable pro-
vided that the Vendor can demonstrate his capability in analysing the
torsional characteristics of the drive train. The use of Bibby ‘Form-
Gard' torque limiters will prevent drive train failures.
(iv) Electric motor drivers on compact power systems, e.g. on offshore

platforms, can create electrical instabilities. Perform a system stability
analysis under such circumstances. Figure 1.2.1-11 is a typical un-
loaded starting torque curve for a reciprocating compressor.
(v) Provide a means of barring (turning) the drive train for maintenance.
Access must be without removing guards or couplings.

1.2.1.4 Screw Compressors
1.2.1.4.1 General
This is the only other type of compressor which may be used in the EP
environment. This compressor consists of counter-rotating geared screws
which compress the gas by means of the progressive cavity technique. Fig-
ure 1.2.1-12 outlines one arrangement of this compressor. While it is simple
in concept, the screw compressor has serious limitations in process appli-
cations.
EP experience of screw compressors in hydrocarbon service has not been
satisfactory. However oil injected air compressors provide excellent service,
although not providing oil free air, and good service has been achieved with
some non-hydrocarbon refrigerant compressor applications.

Running Equipment
The dry twin screw arrangement requires four shaft seals, and while man-
ufacturers claim their seal designs are satisfactory, EP experience has
shown that only seals to API 614 are suitable, which adds significantly to
initial and operating costs. Only relatively low compression ratios are al-
lowable as differential heating of the rotors causes a change in clearances
which allows greater leakage. Injection of oil into the suction dramatically
improves efficiency, and reduces the shaft seals to one, but requires an
effective separator on the discharge. This is not suitable for hydrocarbon
processes because the oil quickly becomes contaminated.

1.2.1.5 Application of Compressors
1.2.1.5.1 General
There are few field installations today where compressors are not found.
Experience has shown that certain compressor types are best suited for
particular applications. In the past, the compressor and auxiliary selection
and package design was left entirely to the Vendor, who had been selected as
the lowest price bidder. Highly unsatisfactory compressor installations have
resulted. Adherence to the notes in this Section which are based on wide field
experience, should eliminate the worst problems of operability and reliability.
The commonest applications are described here, together with notes on the
requirements for acceptable compressor selection.
1.2.1.5.2 Separator Gas Collection
Multi-stage cascade oil/gas separation evolves gas at different pressure

levels and volumes. The separator gases are usually collected and com-
pressed to first stage separator pressure or above for utilisation or
injection.
In the design of the separator train, the pressure levels should be co-ordi-
Rated with compressor selection to achieve a satisfactory match of com-
pressor size and driver power. Also, compressor staging must be selected
such that discharge temperatures do not exceed 150 oC. An uneconomic,
inflexible compressor installation can result if separator pressures are fixed
without regard to compressor requirements.
The volume of gas usually allows the use of centrifugal compressors. The
flow rate and density can be highly variable, therefore for maximum flexibility
each compressor stage should utilise a separate variable speed driver.
Maintenance of steady separator pressure is essential if carry-over is to be
avoided. Centrifugal compressor anti-surge recycle control cannot be used
to control separator pressures, particularly during compressor start-up. Use
a back pressure controller on each separator gas offtake; Figure 1.2.1-13
illustrates a system which enables independent, stable operation of com-
pressor and separator trains.
Oil/gas separators do not remove all liquids from the gas stream. Use ele-
ment type filter-separators or vane type suction separator on each
compressor stage to catch the inevitable carry-over.

Running Equipment

1.2.1.5.3 Gas Lift
As in the case of separator gas collection, the gas density often declines as a
field ages.
Although suction and discharge pressures are reasonably steady, flow rates
can vary widely in compliance with production requirements. Many appli-
cations have sufficient volume for centrifugal compressors; use multiple
variable speed machines for satisfactory flexibility. If electric drive is con-
sidered, the need for performance flexibility must be carefully analysed and
this will generally result in the use of reciprocating compressors because of
their inherent flexibility.
The use of Variable Speed Electric Drive Systems (VSDS) may be considered
as an alternative but cost penalties still exist.
Pay close attention to discharge temperature in this application. The standard

of suction separators needs to be the same as that for separator gas collection.
1.2.1.5.4 Gas Transmission
The gas is usually dry and non-corrosive. High average pressures and low
compression ratios are typical. Either reciprocating or centrifugal compres-
sors are used, the former where high efficiency is paramount. Variable speed
drives are essential for throughput control.
Condensates rarely occur in dry gas lines, the principal contaminant is de-
bris from pigging operations. Always use cartridge type filter separators on
this duty.
1.2.1.5.5 Gas Injection
Gas in the Dense Fluid (supercritical) phase is injected into the petroleum
bearing strata for pressure maintenance or disposal. Injection pressures
often range from 200 bar upwards, gas compositions can vary widely and
contaminants in the gas stream impair reservoir injectivity.
As with other compression duties, the volume flow rate, the required flexi-
bility and turndown should all be considered in deciding upon which type of
compressors to be used. Acceptable high pressure centrifugal and recipro-
cating compressors are manufactured by very few vendors.
Always consult SIPM before selecting an injection compressor, the tech-

nology is changing and improved techniques are continually being devel-
oped.

Running Equipment
The general design criteria for reciprocating injection compressors are:
1. Slow speed, long stroke; maximum speed 375 rpm and 4 m/s average
piston speed are maxima.
2. Piston length must be less than the stroke.
3. No more than 5 piston rings, with the rider rings in the centre of the
piston.
4. Provide some means of relieving excess pressure within the cylinder.
5. Use non-mineral cylinder lubricants.
Figure 1.2.1-14 illustrates a successful gas injection compressor cylinder.
The general design criteria for centrifugal compressors are:
1. Maximum six (6) impellers in one casing, with a head rise not exceeding
18 kJ/kg.
2. A rotor stability log decrement greater than 0.3 is usually advisable.
3. Anti-whip labyrinth seals.
4. All-welded or investment cast impellers, with a tip width greater than

4 mm.

Figure 1.2.1-15 illustrates a successful centrifugal gas injection compressor

for pressures up to 700 bar.
The behaviour of mixed hydrocarbon gases in the dense fluid phase is dif-
ficult to predict, It is essential to have a detailed and accurate gas analysis
with satisfactory characterisation of the C7+ components. It is strongly re-
commended that the specialist advice of SIPM is sought before any com-
pressors are selected with a discharge pressure greater than 100 bar.
1.2.1.5.6 Wellhead Separator and Tank Gas Boosters
Usually these are low pressure applications, and handle small flows of high
molar mass gas. The gas is always saturated with hydrocarbons and water.

Running Equipment
Single and two stage centrifugal motor driven compressors with variable
inlet guide vanes are used. The variable IGVs give good capacity control. The
anti-surge recycle cannot be returned to the compressor suction as there is
usually a lot of condensation in the aftercooler. The recycle gas must be
either flared or taken back to the flow tank.
For reciprocating compressors, use cylinders with high built-in clearances

and a design volumetric efficiency less than 60%. This will make the com-
pressors largely self-regulating.
1.2.1.5.7 Casinghead Gas Compression
On oil wells with beam pumping it is sometimes required to collect the gas
accumulating in the well annulus. Small volumes of gas are involved, careful
control is required to prevent pulling a vacuum on the wellhead. If wellheads
are grouped a central compression plant can be used with a low pressure
gathering system. Reciprocating compressors with high clearance cylinders
are self-regulating and are preferred for this service.
Remote wellheads can be fitted with an ‘Ecopress' single cylinder compres-

sor driven off the walking beam. Ensure that the driver has sufficient power to
drive the combination, the Ecopress will take up to 30 kW. The compressors
are supplied ‘non-lube' as standard; experience shows that the cylinder and
trunnions require continuous lubrication.
1.2.1.5.8 Refrigeration
The type of compressor will depend upon the type and size of plant and the
refrigerant used. In the field, hydrocarbon refrigerants are preferred to Freon
on the grounds of availability and environmental concerns. However, it
should be determined whether the `readily available' propane is indeed
process grade.
Also, the type of refrigerant has to be acceptable to the compressor Vendor.

Packaged chiller units will have the Vendor's standard compressor.
The principal problem is to prevent refrigerant loss. The seals should be
buffered at higher than suction pressure and the trap gas returned to suction.
In hot desert environments, ensure that the condensers are large enough to
condense the refrigerant at the highest ambient temperatures, and that the
compressor driver has sufficient power for the higher condensing pressures
required.

1.2.2 Compressor Systems Control
1.2.2.1 General
Production installations have become more complex whilst at the same time
there has been a drive to achieve lower levels of manning. To maintain a
satisfactory degree of production control and maintain reliability and safety,
there is greater emphasis on instrumentation. This intrumentation is to fulfil
several functions, viz:
1. To achieve optimum production rates with the minimum of running equip-

ment.
2. To operate production installations from a remote control centre, the in-

stallations being manned on an às needed' basis.
3. To monitor the health of the running equipment so as to determine the

need for maintenance intervention in order to avoid component deterio-
ration developing to the point of failure.
Instrumentation thus falls into three categories, for protection, control, and
monitoring purposes. There is considerable overlap between these three
categories.
1.2.2.2 Control Instrumentation
This section will discuss the requirements for throughput control of multiple
compressor installations. It will be assumed that the compressors are sup-
plied with the logic controls for the start-up and shut-down sequences. The
logic described herein will be integrated with the compressor unit control
panels as well as the station control panel.
1.2.2.2.1 Parallel Compressors (see Figure 1.2.2-1)
Station suction and discharge pressures are required for control, station and
unit flow for monitoring only. Note that centrifugal compressor surge control
is entirely separate and plays no part in compressor throughput control.
Normally the driver speed is controlled by suction pressure, with high dis-
charge pressure override. Thus if the downstream system cannot accept the
compressor throughput, PRC-2 will assume compressor control. If the up-
stream pressure rises as a result, PRC-1 will allow gas to flare via PCV-1. This
system is more sensitive and stable than trying to control on compressor flow
and allowing pressures to settle out.
The common speed signal to each train is biased in the train control panel to
balance the flows, measured on Fl-1, 2 and 3. The compressors are unlikely to

Running Equipment
run at the same speed due to variations in manufacturing tolerances, fouling,

etc. The speed signal will accelerate and decelerate the compressors in
unison as load changes. For constant speed compressors load variations can
be achieved by suction throttling using the PRC output to regulate the throttle
valve. The capacity range is less than with variable speed control and is less
efficient.
The Station Logic Controller is an optional extra to achieve automatic start

and stop of compressors upon load changes. It allows unattended operation
of the station. It operates by matching the number of running compressors
and their speed to previously programmed set points determined for station
and train flow, station suction and discharge pressures and temperatures. On
reciprocating compressors the SLC can set the cylinder clearance pockets as
part of the program. Reputable station control system manufacturers are
able to build reliable SLCs, but they are expensive.

PCV-3 is the station recycle, and is used only to prevent the suction header
pressure falling too low. This can occur when there is only one compressor
running at minimum speed and the process flow is less than the compressor
throughput. PC-3 and PCV-3 will prevent the station shutting down on low
suction pressure trip.
1.2.2.2.2 Series Compressors
This applies to centrifugal compressors only (see Figure 1.2.2-2). There are
two usual applications:
1. Pipeline boosting, all compressors identical
2. Reservoir depletion, all compressors may not be identical.
The method of control of both cases is similar. All compressors will not run at
the same speed or pressure ratio, but all will have the same mass flow. Each
will contribute to the overall station pressure rise.
The Station Logic Controller will be programmed with the performance char-
acteristics of each compressor and the system characteristic. For a given
throughput entered by the system operator, and the prevailing station suction
and discharge pressures, the SLC will select the number of compressors
required and set the speeds for optimum efficiency. The station is not con-
trolled on suction and discharge pressures alone.
ESD of any of the running compressors can surge the remaining compres-
sors. When one unit shuts down, the remaining units must be unloaded by
opening their anti-surge recycle valves and run down to minimum speed. As
the suction pressure rises, and the discharge pressure falls, the remaining
units can be reloaded.
Series compressors are ideal for applications where gas reservoir pressures
can fluctuate widely with production rate or age, or on pipeline booster
application. It offers greater flexibility of operation than parallel compressors
in these applications.
The 'hybrid' arrangement with parallel trains of series connected centrifugal

compressors is very difficult to control, and it is not recommended.

Running Equipment

1.2.2.3 Centrifugal Compressor Surge Control
The object of a surge control system is to ensure that the flow into a cen-
trifugal compressor is sufficient to maintain stability at the required pressure
differential across the compressor. Figure 1.2.2-3 shows a typical arrange-
ment of a surge control system. It must be stressed that the surge control
system must be used only for surge control and should never be confused
with a plant or process control system. Detailed guidance on surge control
systems for centrifugal compressors can be obtained from SIPM. There is
also a tendency to overcomplicate surge control systems; adherence to the
principles set out here will fulfil surge control without unnecessary compli-
cation.
Surge control systems are characterised by their quick response. The con-
trols should be adjacent to the anti-surge recycle valve (ASRV) in a lockable
enclosure. Externally mounted indicators show the anti-surge parameter,
internal set point and controller output. Any adjustments can be made only
inside the panel.
The ASRV should be a globe type, cage-guided plug valve with linear action.
The actuator should be a yoke-mounted pneumatic (air to close, spring open).
The valve must open on signal failure to unload the compressor, and achieve
full opening in 2 seconds.
- The principle of the system shown is to generate a `surge parameter' signal

which is connected to the input of FIC. One typical surge parameter [Q/(P x
dP)] yields a family of parabolic control lines on a compressor performance
map which approximate to the compressor surge line when the transmitters
are correctly calibrated. The surge parameter line at approximately 10%
greater flow than surge flow will provide surge protection for all operating
conditions without further adjustment. If variations in gas composition are
anticipated, additional pressure and temperature tappings on the compres-
sor suction and discharge may be required.
The FIC must be a precision controller of robust design meeting these

criteria:
1. Proportional plus fast reset action
2. External reset feedback connection
3. ‘Batch' volume tank
4. Reset restrictor bypass valve.
The Foxboro model 58 and the Moore model 50 meet these criteria. The cutoff
relay between the output and the external reset feedback connection is set to
limit the reset chamber to 1.03 bar. This limits controller output pressure,

Running Equipment

preventing ‘reset wind-up'. By moving the control line to the right, this action
makes the controller follow the compressor operating point and increasing
the safety margin against rapid pressure fluctuations driving the compressor
into surge. The surge control cabinet should have an indicating control sta
tion (Moore 524 or equal) to allow monitoring of the surge parameter, setpoint
and output. These indicators also facilitate calibration and commissioning.
Solenoid valves SV-1 and SV-2 allow the ASRV to be actuated during purging
and start-up. SV-2 holds the ASRV open until the start-up logic calls for
activation of the surge control system. SV-1 will de-energise upon ESD al-
lowing rapid opening of the ASRV.
For surge control, absolute accuracy of the flow signal is not essential, re-
peatability is. Flange tap orifices are recommended. Some manufacturers
calibrate the compressor suction for flow measurement; this has proven to be
accurate and effective. PT and DPT must be downstream of the suction
strainer.
The ASRV should pass the required flow under all process conditions. To
match the rapid dynamics of compressor surge, the ASRV and its actuator
require setting. The quick-exhaust is sized to allow ASRV opening in less
than 2 seconds following signal failure. The restrictor in the booster relay air
supply is adjusted to allow closing of the ASRV in 20 to 60 seconds upon
application of full controller output to the booster relay. The response time is
determined during commissioning of the anti-surge system.
The recycle line to the ASRV must not be pocketed, and must slope upwards
all the way to the valve. Place the valve as close to the high pressure end as
possible to reduce the inventory of high pressure gas the valve needs to
handle upon opening the ASRV. The check valve positioned as near as
possible to the compressor discharge nozzle will also reduce the high pres-
sure gas inventory.
A frequent nuisance on low pressure heavy gas compressors is density

degradation upon continuous recycling. This is caused by heavy hydrocar-
bons condensing after expansion through the anti-surge recycle valve and
being removed in the suction scrubber. The drop in density prevents the
compressor from achieving the required pressure differential. Using a pre-
cooler instead of an after-cooler, will assist in maintaining a constant quality
gas to the compressor.
It is considered essential in multi-stage compression that each stage has its

own anti-surge recycle system. It is false economy to attempt having a control
system covering several stages, control is either unstable or a large section

Running Equipment
of the useful operating envelope is cut off as the surge control system has to
take the worst performance of the stages in series. Equally, it is impossible to
attempt to control compressors in parallel with one surge control system.
Apart from being virtually impossible to operate, neither compressor is ef-
fectively protected and there is a certainty of damage.
1.2.2.4 Monitoring Instrumentation
There is a close overlap between the requirements for control and for mon-
itoring instrumentation. For the latter requirements, refer to 1.9.

1.2.3 Compressor Sizing and Power Requirement
1.2.3.1 Introduction
The purpose of this Section is to enable the user to carry out a preliminary
sizing of compressors for hydrocarbon gases, and driver requirements for
estimating purposes. It also provides a means of checking the performance
of operating compressors, and a means of determining the performance of
compressors under off-design conditions, or for a different application. It is
beyond the scope of this Handbook to provide derivation of formulae and
charts, or for detailed sizing of a compressor, the user may consult the
references to this Section for more rigorous selection procedures. In all
cases, selections should be confirmed by SIPM or a reputable Vendor.
1.2.3.2 Initial Concepts
1. All compressors involve the following variables in their operation:
- Actual intake volume qs, m3/s

This is the volume aspirated into a reciprocating compressor cylinder
during the suction stroke, or drawn into the inlet of centrifugal compressor
impeller.
To calculate actual inlet volume:
(a) Given suction flow qs in m3(st)/s

Running Equipment
- Actual discharge volume qd, m3/s

This is the volume discharged by a reciprocating or centrifugal compressor.
To calculate actual discharge volume, substitute qs Ps Ts and zs, with qd Pd
Td and zd in equations (1) and (2) above where
qd = actual outlet volume, m3/s

pd = absolute discharge pressure, bar
Td = absolute discharge temperature, K
zd = compressibility factor at discharge
Note: For practical purposes zs and zd are considered to be 1.0 for preliminary
calculations.
- Pressure ratio r = pd/ps

This is the ratio of the absolute discharge and suction pressures.
- Gas density ρ , kg/m3

The greater the density, for a given actual intake volume, the greater the
mass flow and hence the power absorbed will be greater.
- Ratio of specific heats k = cp/cv

This determines the pressure rise and temperature rise in the
compressor.
2. Efficiency.
Of principal concern are:
- Adiabatic efficiency η a
This assumes that work done in compressing the gas is reversible, i.e.
there is no heat loss or gain, and on re-expansion to the original pressure,
volume and temperature will be the same as original. For practical pur-
poses similar to Isentropic efficiency.
- Polytropic efficiency η p
This recognises that heat is lost (radiation, conduction) or gained (friction)

during an actual compression process.
Both adiabatic and polytropic efficiency represent the difference in theoret-

ical energy required in compressing a gas and the actual energy required.
The adiabatic efficiency is more representative of the power absorbed, but
the polytropic efficiency gives a closer estimate of the temperature rise, It is
industry practice to use polytropic efficiency in all centrifugal compressor
calculations.

Some typical polytropic efficiencies η p are given in Table 1.2.3-1.
The relationship between adiabatic and polytropic efficiency is:
n-1 k-1
 =  (3)
n kη p
where:
k = ratio of specific heats cp/cv or adiabatic exponent for gas
n = polytropic exponent for gas
Use Figure 1.2.3-1 to determine relationship between η a and η p if adiabatic

conditions are required.
- mechanical efficiency η m
This includes windage, bearing and seal losses in a centrifugal compres-

sor; sliding friction and valve losses in a reciprocating compressor.
- volumetric efficiency ηv
This is the ratio between the swept volume of a piston moving in a cylinder
and the actual volume inhaled at intake conditions (see 1.2.2.3).
3. Head Hp or Ha
Unlike a pump which will lift a column of liquid a finite, measurable height,
`head' is less tangible in a compressor, and is a measure of the energy added
to the gas by the compressor. It may be polytropic, or adiabatic (see 2.
above). It is totally incorrect to quote compressor head in units of length,
metres, feet, etc., as it is an energy term, i.e.
Hp = kJ/kg

Running Equipment

Running Equipment
4. Gas Properties, M, k, z, pc, Tc
These can be determined from the gas composition. In most cases, computer
routines are used for calculation of molar mass (M), ratio of specific heats (k),
critical pressure (pc), critical temperature (Tc), and compressibility factor (z).
If the composition is given use the GPSA Engineering Data Book (Section 4)
method for calculation of the other properties required. These are essential
to obtain satisfactory accuracy in any compression calculations.
If the critical pressure pc and critical temperature Tc are known, values of z

can be obtained from the GPSA Engineering Data Book (Section 16, Fig 16-3).
The pseudo reduced pressure Pr and pseudo reduced temperature Tr can be
obtained from the following equations.
Tc pc
Tr =  , pr = 
T P
where
Tc = absolute critical temperature
pc = absolute critical pressure
T = absolute temperature at which the gas exists
p = absolute pressure at which the gas exists

Running Equipment
1.2.3.3 Calculation Routine
See Figure 1.2.3-2 for flow diagram, which also explains the symbols
used.
1. Given : gas flow, m3(st)/s or kg/s

: suction pressure ps, bar
: suction temperature Ts, K
: discharge pressure pd, bar
: gas properties M, R, pc, Tc
: interstage recooled gas temperature T s2, Ts3, etc., K
2. Interstage Pressure Drop.
Interstage pressures may be fixed by oil/gas separator or other process

pressures. However, compressor stage pressure ratios must not permit the
maximum allowable temperature to be exceeded. Thus each oil/gas sepa-
rator stage may require more than one compressor stage, with intercooling,
or adjustment of the separator pressures to achieve a better economy of
compressor staging.
Always allow 2% pressure drop (with a minimum of 0.2 bar) for suction
separators and aftercoolers, and 3.5% pressure drop (with a minimum of 0.35
bar) between compressor stages for intercoolers and separators. For esti-
mating purposes, the interstage pressure drop is always added to the dis-
charge pressure of the previous stage.
3. Check Approximate Final Discharge Volume qd :
Use equations (1) or (2), substituting discharge conditions and assuming that
the temperature is the maximum allowable of 150 oC (423 K) (see 1.2.1). Note
that for this rough calculation zd can be assumed as 1.0, unless the discharge
pressure exceeds 65 bar when it must be calculated. Ask the question:
Is qd > 0.08 m3/s ?
If yes - proceed to 1.2.3.4 (centrifugal compressors)
If no - proceed to 1.2.3.5 (reciprocating compressors) as volume is insuf-

ficient for centrifugal compressors.
Note: This is not the only test, as reciprocating compressors are more flexible and
more efficient than centrifugals, but are usually more expensive. Operating economics
may decide in favour of a reciprocating compressor although the flow will be sufficient
for a centrifugal (see also 1.2.1).


Running Equipment
1. Calculate the polytropic η for the compression process, using Equation (3),
and a suitable value of ηp from Table 1.2.3-1.
2. Calculate suction pressure and discharge pressure:

7. Calculate first stage discharge pressure:
pd1 = (ps1 x rQ) x 1.035 [bar] (11)

or pd1 = (ps1 x rQ) + 0.35 [bar], (whichever is the greatest)
Note: The factor 1.035 is the interstage pressure drop allowance, subject to a min-
imum of 0.35 bar.
8. And first stage pressure ratio:
r1 = pd1/ps1 (12)
9. Thus first stage discharge temperature:
10. Power requirement, first stage:

Using charts, Figures 1.2.3-3a and b, for r1 and k, read the specific power
absorbed kW/{m3 (st)/s).K}.
Calculate the stage power requirement:
W1 = (specific power) x qs(st) x Ts1 [kW] (14)
Note: The power absorbed charts (Figures 1.2.3-3a, b) include a variable efficiency
factor, decreasing inversely with pressure ratio, and a variable bearing and seal loss
factor. This W1 is the total power absorbed by the stage, no further losses need be
added.
11. Calculate the stage polytropic head:
Note: Centrifugal compressor ‘head' is an energy term and cannot be compared with
the measurable head of liquid raised by a centrifugal pump. It is thus totally incorrect to
quote centrifugal compressor head in units of length (feet, meters, etc.). To do so will
result in errors in calculations.

Running Equipment


Running Equipment

20. To obtain size and running speed of compressor: Consult a reputable

Vendor's standard centrifugal compressor catalogue or other selected
references at the end of this Section, taking account of the limitations
given in 1.2.1.
21. With the power requirement and size of the compressor it will be possible
to prepare a layout of the installation, and estimation of the size, weight
and cost of the installation.
1. Calculate suction pressure and discharge pressure:
ps1 = ps / 1.02 [bar] or ps1 = ps - 0.2 [bar] (whichever is the smallest)

(21)
pdN = pd x 1.035 [bar] or pdN = pd + 0.35 [bar] (whichever is the grea-

test)
(22)
where N is the Nth or last stage. (There are an aftercooler and a discharge
scrubber after the last stage.)
2. Calculate overall compressor pressure ratio:
ro = pdN/ps1 (23)
3. Calculate maximum stage ratio r5:
This is calculated using the highest of either the suction or intercooled gas
temperature and maximum allowable discharge temperature of 150 oC
(423 K), (k = ratio of specific heats cp/cv or adiabatic exponent for gas),
4. Calculate required number of stages Q:
Q = log ro/log rs (26)

Running Equipment
Round up to the next whole number.
5. Calculate theoretical stage pressure ratio:
1/Q
rQ = (ro) (27)
6. Calculate first stage discharge pressure:
pd1 = (ps1 x rQ) x 1.035 [bar] (28)

or pd1 = (ps1 x rQ) + 0.35 [bar], (whichever is the greatest)
Note: The factor 1.035 is the interstage pressure drop allowance, subject to a min-
imum of 0.35 bar.
7. And first stage pressure ratio:
r1 = pd1/ps1 (29)
8. Thus first stage discharge temperature:
9. Power Requirement, first stage:
Using charts, Figures 1.2.3-4 a to c, for r1 and k read the specific power
absorbed k W/ {(m3(st)/s)ּ K}.
Calculate the stage power requirement:
W1 = (specific power) x qs(st) x Ts1 [kW] (31)
Note: The `power absorbed' charts unlike those in the GPSA data book give the shaft
power required, and are calculated with decreasing efficiency with increasing r as
cylinder and valve losses increase, and have been verified in practice. Note also that
the value of kW read from the charts has only to be multiplied by the absolute suction
temperature Ts, not the ratio of Ts to standard temperature, as in the GPSA
charts.
10. Calculate suction and discharge actual volume flow rate:
First calculate zs1 and zd1 using 1.2.3.2 (Step 4)
Then use Equation (1) or (2) substituting the appropriate values of P, T and
z.
Use these volume flow rates to check suction and discharge piping size.


Running Equipment


Running Equipment
11. For second stage suction pressure:
ps2 = pd1/1 .035 [bar] or ps2 = Pd1 - 0.35 [bar] (whichever is the smallest)
(32)
12. Second stage suction temperature will be the recooled gas temperature,
modified by any sidestreams, etc.
13. Second stage discharge pressure:
pd2 = (pd1 x rQ) x 1.035 [bar] (33)
14. Proceed now as first stage Steps 6 to 10 to obtain 2nd stage power and
suction and discharge volume flow rates. Repeat procedure for all
stages.
15. For total shaft power:
W= 1.05 x (W1 + W2 + W3……....W(Q)) [kW] (34)
The 5% margin is an allowance against inaccuracies in data, etc.
16. For driver power:
If direct drive, use W [kW]

If through gearbox, speed reducer, etc. (to allow for transmission losses):
Driver power = 1.04 W [kW] (35)
17. To obtain size of compressor:
Consult a reputable Vendor's standard reciprocating compressor catalogue,

taking into account the limitations described in 1.2.1. For the power absorbed
and the number of stages, select a suitable frame size.
Note: The crankcase and crank mechanism is always standard for a frame size; only
the number and diameter of cylinders per stage is variable.

18. To calculate cylinder size:
Given (a) Inlet volume flow rate from Step 10

(b) Piston stroke L, m
(c) Piston rod diameter d, m
(d) Allowable piston rod load F, N
(e) Cylinder maximum allowable working pressure, bar
(f) Running Speed ω, rev/min.
(b), (c), (d), (e) and (f) from Vendor's catalogue.
Calculate the required cylinder swept volume using the formula:
Vs
ηv =  (36)
Vpd
where ηv = volumetric efficiency, %

Vs = actual inhaled volume per piston stroke, m3
Vpd = cylinder swept volume per piston stroke, m3
Vs
Thus Vpd =  (37)
ηv
Notes
(I) Volumetric efficiency is the ratio of the volume inhaled in a cylinder at suction
conditions to the swept volume of the cylinder (see Figure 1.2.3-5). It has nothing to
do with power absorbed or compressor mechanical efficiency. Volumetric effi-
ciency (VE) will vary according to pressure ratio valve losses, piston ring leakage.
See Figure 1.2.3-6 to select required VE vs. r. Applicable to most commonly en-
countered hydrocarbon gases, and gives a cylinder selection with satisfactory
flexibility and economy.
3
(ii) All volumes in the calculations are given in m /s instead of the actual displacement
per piston stroke. This is because Vendor's standard tables of cylinder displace-
ment are calculated at the design running speed of the compressor. The methods -
given here will therefore yield compatible results.
To calculate ηv from cylinder data, use the formula:
1/k
ηv = K – r - NcL (r -1) [%] (38)
VcL
where NcL =  x 100 (see also Figure 1.2.3-5)
Vpd
and VcL = cylinder clearance volume, m3

K is a factor to allow for valve losses, piston leakage, etc. (see Table 1.2.3-2).

Running Equipment
19. Calculate cylinder diameter:
Most Vendors have tables of cylinder swept volume vs. cylinder diameter and
running speed. If not available, use the formula (for double-acting cylin-
ders):
π ω
qpd =  (2D2 - d2) x L x  [m3/s] (39)
4 60
where:
D = cylinder diameter, m
d = piston rod diameter, m
L = piston stroke, m
ω =running speed, rpm
qpd = cylinder displacement, m3/s
qpd
Thus D = [ 0.5 x (  + d2)] 0.5 [m] (40)
0.0131 x L x ω
Useful tip:
This equation is cumbersome to use, and will be repeated for each stage of
compression. As the piston rod diameter and therefore its swept volume, is
constant for each cylinder, it is useful to precalculate its swept volume and
add it to the required cylinder swept volume before calculating the cylinder
diameter.
Thus:
Piston rod pd = 0.0131 x d2 x L x ω [m3/s] (41)

qpd + piston rod pd

D = [  ] 0.5 [m] (42)
0.0261 x L x ω
Similarly, given a cylinder diameter, the qpd is calculated by subtracting the

piston rod displacement from the total cylinder volume.
qpd = (0.0131 x D2 x L x ω) - (piston rod pd) [m3/s] (43)

Running Equipment
Note: Manufactures tables of cylinder diameters do not include liners. A cylinder will
have to be selected which is at least 30 mm larger than the 0 you have calculated. This
allows for the cylinder casting to be selected for the diameter you have calculated.
Check that the cylinder has sufficient margin of working pressure above Pd.
20. Check piston rod loads (thrust), compression and tension:
π
Compression =  x [(pd x D2) - ps(D2 - d2)] x 10-2 [MN] (44)
4
Note: As ps and pd are in bar, the multiplier 10-2 must be used to express the force in
mega Newtons.
π
Tension =  x [(ps x D2) - pd(D2 - d2)] x 10-2 [MN] (45)
4

Notes:

(i) Compare these values with the allowable load. See Step 18.
If calculated load exceeds 0.85 x allowable load (per Shell DEP Standards), then
cylinder diameter selected is too large for the required pressure ratio.
Recalculate using either (a) more stages with reduced r
or (b) smaller diameter cylinders operating in parallel.
(ii) If the value of Equation (45) is negative, then there is no piston rod thrust reversal
and lubrication of the cross-head pin will be inadequate. The solution is beyond the
scope of these notes, advice should be sought from SIPM, as special cylinders are
required.
21. Proceed to second and subsequent stages and repeat Steps 18,19 and 20
for each stage to select cylinder diameter. Remember that all cylinders
must have the same stroke length and piston rod diameter.
22. Having found the cylinder sizes, the pulsation bottle sizes can be esti-
mated using 1.4. All data will now be available for preparing a layout of
the installation, estimation of size, weight and cost of the installation. The
above techniques are useful for calculation of the performance of a
reciprocating compressor under off-design conditions (see 1.2.3.6).

Running Equipment
1.2.3.6 Field Application
1. The techniques described in 1.2.3.4 and 1.2.3.5 may be used to calculate

the performance of compressors under off-design conditions, and to verify
the efficiency of compressors running in the field. In addition the following
affinity laws for centrifugal compressors can be applied:
qs2 ω2 (46)
 = 
qs1 ω1
Hp2 (ω2)2 (47)

 = 
Hp1 (ω1)2
W2 (ω2)3 (48)
 = 
W1 (ω1)3
W2 M2 (49)
 = 
W1 M1
W2 w2 (49)
 = 
W1 w1
Note Equations (49) and (50) apply also to reciprocating compressors.
where:
Suffixes 1 and 2 refer to old and new conditions respectively

qs = inlet volume, m3/s
ω = running speed, rpm
Hp = polytropic head, kJ/kg
W = power absorbed, kW
M = molar mass, kg/kmol
w = mass flow, kg/s
2. Most calculations for field verification of compressor performance are

available in programmable calculator programs. There are a number of
SIPM developed programs for HP67 and HP41C calculators, and Solar
Turbines Inc. offers several TI-59 programs. Almost all programs require
accurate gas data, thus for verification of field performance it is essen-
tial to have a gas composition analysis performed at the same time as
gathering pressure, temperature, etc. data.

1.3 PUMPS
1.3.1 Applications
Within E&P pumps have a wide application.
Pumps are employed at the very core of EP business:

- crude oil
- water
- condensate handling
Pumps are also employed for ancillary and utility applications, for exam-
ple:
- chemical injection
- fuel transfer
- lube oil circulation
- cooling medium circulation
- fire water
- domestic sewerage
- drinking water
There are also pumps employed for special applications, such as downhole
electric submersible pumps (ESPs), downhole progressive cavity pumps
(PCPs) and beam pumps.
In order to handle the diverse products and conditions encountered, a variety

of pumps types are used.
Centrifugal
- volute
- diffuser
Positive Displacement
- Reciprocating
- piston
- plunger
- Rotary
- twin screw
- single screw (progressive cavity)
- gear
- vane
- lobe

Running Equipment
1.3.2 General
The ancillary, utility and special application pumps are not specifically cov-
ered in this section, however it is essential that the same basic procedures as
provided in this section for the requisitioning and evaluation of pumps are
also applied to these units. In particular the selection of the minimum re-
quirement of component materials should be adhered to.
The theoretical aspects of pumps are adequately covered in text books and
various notes available to individuals within the Shell organisations. No
attempt is made to cover the same ground in this section.
This section does not supersede the various SIPM Design and Engineering
Practice (DEP) publications, but is intended to highlight specific information
which is likely to be useful on a routine basis.
1.3.3 Characteristics and Selection
1.3.3.1 General
Depending on the application, and especially the flow and differential head,
equipment selection may be limited to certain types of pump. The pump
range chart (figure 1.3-1) indicates the capacity and head limitations of the
various pump types which are likely to be employed. Table 1.3-1 lists the
pump types which should be used in typical EP applications, with the order of
preference and approximate maximum casing pressure capabilities
shown.

Figure 1.3-1 Pump Range Chart
As with compressors, when conditions fall in a transition zone, careful study

is required of the degree of certainty of data provided, future field develop-
ments, and the sensitivity of the anticipated operating points.
If at all possible, the centrifugal pump is always the pump of choice in the EP
environment.
Every effort should also be made to minimise the variety of makes and types
of pumps, drivers, and auxiliary equipment selected for a given installation,
or group of installations. This rationalisation should be balanced with the
need to select the best and most cost-effective pump for the specified oper-
ating conditions.

Running Equipment

As indicated, centrifugal pumps are shown as the preferred first choice. They
should be selected where possible, but not in the following circum-
stances:
- When handling highly viscous fluids (gross viscosity >800 cSt contin-
uous).
- When handling fluids containing free gases (1 to 2%).
- Where the duty involves high head at relatively low capacity.
- For metering applications.
1.3.3.2 Prototype Pumps
The selection of equipment for EP oil-field applications generally should be

well within the range of manufacturers' proven experience and should not
involve the application of novel or unproven designs or components, without
due consideration of the risks and consequences involved.
Any intent to use unproven or prototype equipment for a particular service

should be discussed with SIPM-EP mechanical equipment engineers during
the front-end phase of the project, and subsequently during the detailed
design phase.
1.3.3.3 Pump type selection
The pump type is determined by the basic process operating conditions:
- Capacity
- Differential head/pressure
- Viscosity range at minimum and maximum temperature
- Net Positive Suction Head Available (NPSHA)
- Maximum allowable speed
In addition, any other important operating conditions that can affect the pump
and materials selection need to be identified, e.g.:
- Number of installed units and any requirement to operate in parallel or

series.
- Handling fluids of varying densities or operating temperatures.
- Requirement to operate at minimum continuous thermal or stable flow.
- Stringent start-up conditions.
- Flow control requirements.
- Presence of solids and/or other contaminates in the pumped fluid
(e.g. traces of sand, water cut percentage, hydrogen sulphide and carbon -
dioxide content).
- Presence of free gases in the pumped fluid.

Running Equipment
Once the various operating conditions have been established, the design
duty can be calculated as the first stage in the selection and specification
procedure.
1.3.3.4 Centrifugal Pumps
1.3.3.4.1 General
The fundamental design basis for a centrifugal pump is whereby velocity is

imparted to the fluid by the motion of the rotating element, which is then
converted to a dynamic head. This ‘kinematic' machine therefore indirectly
produces a head so that fluid transfer can be achieved.
As already stated, centrifugal pumps are considered to be pumps of first

choice in EP, due to their low operating costs and ease of maintenance.
However, their flexibility in service is sometimes limited, so care must be
exercised during specification and procurement.
The simplest capacity control is pump on/off, but this is seldom acceptable,
especially in larger multistage applications where the rotating assembly
represents a significant moment.
Speed control provides good capacity control, and is one reason for using
gas turbine driven pumps in pipeline booster applications. In applications
where the driver requirement is less than 2.5 MW, VSDS are gaining wider
acceptance for good capacity control.
Discharge throttling is the most common form of capacity control used in EP.
This produces a modified system resistance curve, and forces the pump to
operate further up the performance curve. However this control means is
inefficient in that the pump is required to produce a head which is then
partially dissipated over a control valve.
Adjustable inlet guide vanes or diffuser vanes can be used to control through-
put, but these means of control are seldom used in the EP environment.
Within the mechanical limitations of the pump design, the diameter of the
pump impeller or impellers may be changed to modify the pump perfor-
mance. Close co-ordination is required with the pump vendor if impeller
changes are contemplated.

The following Affinity Laws can be applied:
The Affinity Laws:
where:
Q = flow
H = head
N = speed
D = impeller diameter
P = power
ρ = relative density (specific gravity)
The laws are only to be applied to centrifugal (radial flow) pumps. The re-
lationship holds only for small changes in impeller diameter. It can be seen
that changes of the speed ratio or the diameter ratio will have the same effect
on capacity, head and power.

Running Equipment
The centrifugal pump has difficulties performing in low flow, high head ap-
plications. Under these conditions, positive displacement pumps are a more
suitable choice.
To limit out of operating range conditions during start-up, the centifugal pump
should be started against a closed discharge valve. Similarly, if a pump is
allowed to operate with flows significantly higher than design, then cavitation
may develop. This can be avoided by providing maximum flow protection
arrangements.
A centrifugal pump should not be allowed to operate continuously against a

closed discharge because of the effects of heat build up in the fluid and pump
casing. If capacity control requirements demand the possibility of little or no
system flow, then provision must be made for a minimum flow loop.
1.3.3.4.4 Selection
Based on minimum wear, maintenance and downtime coupled with maxi-

mum availability and reliability, preference should be given to centrifugal
pumps whenever possible. Figure 1.3-1 also shows the exceptions to the use
of centrifugal pumps.
An important criterion is the specific speed, Ns.
Ns = NQ0.5 H-0.75
N = speed, rpm
Q = capacity, m3/s
H =head, m
Typical shapes of impellers for different Ns values are shown in Figure 1.3-2.
If Q, H and N are known the specific speed can be calculated and the shape of
the impeller identified. By using more than one stage, e.g. x stages, the Ns for
each individual impeller will be x0.75 larger than for a single stage pump and
hence an impeller shape with a higher efficiency can be used.
Whenever possible a single stage vertical in-line pump should be selected

with the following limitations:
Differential head < 160 m liquid
Capacity < 0.11 m3/s
Temperature between -30 oC and +200 oC
Experience has shown that maintenance costs are low for this type of
pump.

When the process data fall outside the range of the vertical in-line pump a
choice should be made from the following pump types in descending order of
preference:
1. Single stage centrifugal pumps

- Horizontal single suction
- Horizontal double suction.
2. Two-stage centrifugal pumps of overhung impeller type (bearing bracket

on one side).
3. Multi-stage centrifugal pumps

- Barrel type, with horizontally split inner casing
- Barrel type, with vertically split inner casing
- Horizontally split casing
- Vertically split casing, including ring section pumps
- Vertical multi-stage bowl pumps.
1.3.3.4.5 Efficiencies and Power Requirements
The efficiency varies with the output and is maximum at the pump design
point, falling with increased as well as reduced output. The maximum effi-
ciency varies considerably with different types of centrifugal pumps. In gen-
eral a low head (H), high capacity (Q) pump will have a high efficiency,
whereas a high head, low capacity pump will have a low efficiency:
An approximate prediction of the maximum efficiency at the design point
for a single stage centrifugal pump, pumping a liquid with a viscosity of
1 mm2/s can be made using Figure 1.3-3. For liquids with-a viscosity greater
than 1 mm2/s the correction for capacity, head and efficiency are shown in
Figure 1.3-4. For multistage pumps the head per stage must be taken.
The power requirements of a centrifugal pump can be summarised as fol-

lows:
Power consumption P (W)
P = QρgH
η
where:
Q = capacity, m3/s
H = actual head, m liquid
p = density of liquid, kg/m3
η = total efficiency of pump

Running Equipment


Running Equipment

1.3.3.4.6 Centrifugal Pump Trouble Shooting Guide
A trouble shooting guide for centrifugal pumps is shown in Table 1.3-2

Running Equipment

1.3.3.4.7 Water/viscous Fluid Equivalence
A. Procedure to find equivalent water duty for a centrifugal pump required to

handle a viscous fluid.
Figures 71 and 72 of the Hydraulic Institute's Standard for Centrifugal Pumps

are used to determine the viscosity correction factors (see Figures 1.3-5 and
1.3-6).
1) Enter the chart at the bottom with the required viscous fluid capacity (Qvisc)
and proceed upwards to the required viscous head per stage (Hvisc).
2) Project horizontally to the appropriate fluid viscosity line.
3) Project vertically upwards to find correction factors CQ, CH and Cη

(CE).
The following equations are used for determining the equivalent water per-
formance:
Qvisc
Qw = 
CQ
Hvisc
Hw = 
CH
Nvisc = NW x Cη
where:
QNW = Water capacity at- best efficiency point (BEP)

QW = Water capacity
HW = Water head
NW = Water efficiency
CQ = Capacity correction factor
CH = Head correction factor (use curve for 1.0 x QNW)
Cη (CE) = Efficiency correction factor

Running Equipment
B. Procedure to determine the pump performance on a viscous fluid when

performance on water is known.
1) Locate the water capacity at BEP (QNW) from the efficiency curve.
2) Determine 0.6 x QNW, 0.8 x QNW, 1 x QNW and 1.2 x QNW.
3) Enter the chart at the bottom at QNW, project vertically upwards to locate
the head per stage (HW).
4) Project horizontally to required viscosity and then vertically upwards to

the correction curves.
5) Read CQ, CH and Cη (CE) for all four capacities (CH at 0.6/0.8/1.0/1.2
QNW).
The following equations are used for determining the viscous performance
curve:
Qvisc = QW x CQ
Hvisc = HW x CH
Nvisc = NW x Cη
4 points for viscosity curve can be calculated, e.g.:
Qvisc = 0.6 QW x CQ
Hvisc = (HW @ 0.6 Qnw) x CH (@0.6 QNW)
Nvisc = NW (@0.6 QNW) x Cη


Running Equipment
Figure 1.3-6 Performance correction chart for viscous liquids


1.3.3.5 Reciprocating Pumps
1.3.3.5.1 General
The reciprocating pump is the second most common type of pump used in EP,
with the majority being of the plunger pump design. Figure 1.3-7 shows basic
details of one type of plunger pump. As a positive displacement machine, it is
essentially a constant flow device, and this must be taken into consideration
during the selection process. Pump performance will be limited by the driver
power input or by discharge pressure relief valves.
Figure 1.3-7 Horizontal plunger pump

(Source: Reciprocating Pumps, T.L. Henshaw)
Running Equipment

Capacity control can be achieved by variable speed drive, recycling to the

suction reservoir, or if non-uniform flow is acceptable, using on/off control.
The majority of positive displacement pumps are used for high head, low flow
applications and for viscous liquid duty. Particular attention needs to be paid
to ‘cold start' conditions for sizing the driver, defining pipework and pump
case pressure capacity, and the limiting affect on available NPSH and fluid
inlet velocity.
A positive displacement pump always produces a fluctuating discharge pres-

sure which may vary between 5 - 20% of the mean discharge pressure.
During the design phase, a full system dynamic study must be carried out to
assess the need for and sizing of pulsation dampeners. If the pump is required
to operate over a significant discharge pressure range, multiple dampeners
may be required, as well as dampeners on the suction header.
1.3.3.5.4 Selection
(a) Piston pump
A piston pump consists of two main parts, the power end and the fluid end.
The fluid end consists of two or more single or double acting pistons mounted
in individual or mono-block cylinders. The piston with piston rings runs in a
renewable metal liner with the piston rod passing through a packing. Parts
heavily subjected to wear while pumping abrasive liquids like drilling mud
and sandy crude oil are valves, piston rod and rings and cylinder liner. A
maximum flow of about 0.135 m3/s and pressures of 27.5 MPa have been
achieved on mud pump applications. Suitable piston pump designs should
prevent liquid ingress into the power end in the event of piston ring and
packing failure. Excessive velocities through valves should be avoided. Dur-
ing operation, liners should be rotated regularly to prevent eccentric wear.
(b) Plunger pump
Generally the power end is similar to that of piston pumps. The difference lies
in the fluid end, where the plunger runs through the packing like a piston rod
(see Figure 1.3-7). The stuffing box can be fitted with a flushing arrangement
to prevent abrasive solids from lodging in the packing where they could score

the plungers and destroy the packing. Plungers tend to be heavy and their
weight causes eccentric wear to the plunger packing.
There is also a possibility of scoring between the stuffing box bushings and
the plungers. These problems are overcome by using pumps with plunger
diameters less than 0.102 m (4 inch) or pumps with vertical plungers. A ver-
tical plunger pump design with the valves at a distance from the fluid cylinder
has the advantages of facilitating maintenance and allows the use of a great-
er variety of valves.
Pump capacities of around 0.056 m3/s are common. Maximum pressures

developed are in the region of 68 MPa. Plunger pumps are preferable to
piston pumps when liquid containing very abrasive material is pumped or
when the discharge pressure is over 14 MPa.
(c) Direct-acting pumps
Instead of having a crankshaft arrangement at the power end, steam, air, gas
or hydraulic oil can be used to directly operate the piston. During 80-90% of
the stroke the velocity of the piston will be constant. By adjusting the power-
end cylinder valves, a small flow overlap can be achieved.
1.3.3.5.5 Efficiencies and Power Requirements
λ includes losses due to, leaks in packing, piston and valves; delayed valve
closing, air or vapour in the liquid and design or manufacturing errors.

Running Equipment
λ = 0.98 to 0.96 for large pumps with good maintenance.

λ = 0.98 to 0.93 for middle sized pumps.
If the volumetric efficiency falls below ≈ 0.92 the valves and the packing
should be checked. API specifies max. λ = 0.95 to be used by sizing the fluid
end.
Some maximum total efficiencies for power pumps (piston and plunger
pumps) pumping water at practical pump speeds are shown in Table 1.3-3.


Running Equipment
1.3.3.5.6 Reciprocating Pump Trouble Shooting Guide
A trouble shooting guide for reciprocating pumps is shown in Table 1.3-5.
Table 1.3-5 Reciprocating pump trouble shooting guide


Running Equipment
1.3.3.6 Screw Pumps
1.3.3.6.1 General
Screw pumps operate on the principle of progressively moving a fluid be-

tween sets of counter-rotating screws. If significant losses are to be avoided,
tight clearances must be maintained between the screws and liner. Screw
pumps have been traditionally chosen in EP to pump viscous fluids, and
impart minimum shear forces on the fluid with relatively low discharge pres-
sures. Figure 1.3-8 shows an outline of a typical screw pump.

Capacity control can be achieved by the same means as that applied to

reciprocating pumps.
As the maintenance of close tolerances is imperative if performance is to be

maintained, the pumped fluid must be free of abrasive solids. As this situation
seldom exists in an EP environment, screw pumps should be avoided if at all
possible. Any proposal to utilise screw pumps must be carefully scrutinised,
and if possible alternatives are substituted, to ensure that frequent pump
overhauls, poor availabilities, and high operating costs are avoided.

Running Equipment
1.3.4 Pipework Requirements
1.3.4.1 Centrifugal Pumps
Suction
1. Suction lines should be short and without elbows, but if necessary use
45º elbows.
2. The velocity of the liquid should fall within the range of 0.1-0. 5 m/s.
3. A minimum suction head of 1.2-3m liquid at the pump suction is nec-
essary, but will increase with the viscosity of the fluid pumped (see
1.3.5).
4. Use full opening valves, avoid constricting valves.
5. Use eccentric reducers (top flat) near the pump for keeping the line level
to eliminate gas pockets.
6. A uniform slope down to the pump will eliminate gas pockets.
7. Recycle design should take the fluid back to the source, not into the
suction line.
8. Anchor the suction line firmly.
9. Never use a smaller line size than pump inlet.
10. It should be noted that unlike a reciprocating pump, a centrifugal pump
cannot handle any air or vapour unless a special self-priming device is
built in.
Discharge
1. Do not have flows into ends of tees and out through branches on multi-
pump installations.
2. Anchor the discharge lines firmly.
3. Run lines straight from the pump for at least 3 in before bending.
4. Install a full opening valve, which will allow for maintenance, and up-
stream of this valve a check valve, which will protect the pump from
harmonic pulsations from other installations on the line.
5. Recommended discharge velocities vary between 1,5-2 m/s.
Starting
The suction valve should always be fully opened. Usually a centrifugal pump
is started against a temporarily closed discharge valve. This is not necessary
with a variable speed pump. Standby pumps, which are required for auto-
start on loss of system pressure, will start against a closed check valve, which
will open at sufficient head. If the pressure drop in the system is appreciable,
the auto-starting pump can begin to take on load before the running speed is
reached. The driver and pump characteristics should then be checked to ensure
sufficient margin between them. to prevent stalling.

1.3.4.2 Positive Displacement Pumps
1.3.4.2.1 Reciprocating Pumps
Suction
1-9. As for centrifugal pump suction 1-9.

10. Most pumps require pulsation dampeners (see 1.4).
Discharge
1-4. As for centrifugal pump discharge 1-4.

5. Liquid velocities in discharge line and manifold or header should fall in the
range of 0.75-1.2 m/s, with all pumps operating.
6. Excessive pressure should be safeguarded by means of a relief valve,
discharging to suction.
7. For prevention of liquid hammer install pulsation dampener (see 1.4).
Starting
1. Never start a pump against a closed line.

2. Avoid starting against a full discharge load. Bring pump up to speed using
a recycle line; then gradually place the pump on line.
3. For high viscosity fluids particular attention has to be paid to `cold start'
conditions for sizing the driver and designing the pipework.
1.3.4.2.2 Rotary Pumps
Suction
All the requirements mentioned under centrifugal pumps are valid.
Discharge
All the requirements mentioned under reciprocating pumps are valid, with
exception of point 7.
Starting
As reciprocating pumps.

Running Equipment
1.3.5 Net Positive Suction Head (NPSH)
For pumps covering a wide range of capacities and speeds some of the most
complicated and serious field problems occur from insufficient NPSH.
NPSHA: Net positive suction head available to the pump determined by the
system upstream of the pump.
NPSHR: Net positive suction head required by the pump, determined by
pump type and the pumped liquid.
Suction lift:
NPSHA should preferably be greater than 2 in liquid and exceed NPSHR with
1 in liquid, at least (see DEP 31.29.02.30-Gen.).
Suction specific speed (S)
Suction specific speed is a parameter describing the suction capabilities and

characteristics of a given first stage impeller of a centrifugal pump.
S = N. Q0-5. NPSHR-0.75
where:
N = pump speed, rpm

Q = design capacity at best efficiency point for single suction first stage
impellers (Q/2 for double suction first stage impellers, at maximum
diameter, m3/s
Recommended value of S: less than 210

An estimate of the NPSH which must be provided in the suction line can now
be calculated from:

NPSHR = (N. Q0.5/S)1.33
By lowering the speed or selecting a double suction impeller a lower NPSHR

can be achieved, but this will also reduce the efficiency of the pump. Always
check if the pump foundation can be lowered or suction vessel raised, a
booster pump installed or if the liquid can be cooled to improve the NPSHA.
This will avoid the purchase of a more expensive pump and possibly a less
reliable installation.
1.3.6 Centrifugal Pump Arrangements
Parallel Operation
The combined characteristic of pumps operating in parallel is obtained by

adding their individual ‘Q' characteristics for a given value of ‘H'. If pumps of
widely differing characteristics operate in parallel, the unit with the lower
closed valve head must be protected against operating at unacceptably low
flows (see Figure 1.3-9). If the flow falls below Qm, pump B will operate
against a closed check valve.
Series Operation
The combined characteristic of pumps operating in series is obtained by
adding their `H' characteristic at any value of `Q'.

1.3.7 Prime Movers
Shaft speeds are normally dictated by the electric motor, which is the most
common pump drive. These motors offer 1500 rpm with four poles and 3000
rpm with two poles running at 50 Hz (at 60 Hz, 1800 rpm and 3600 rpm
respectively). A variable speed electric motor may be beneficial for a large
pump or a requirement for system control.
Other drivers:
Gas turbines - for high power requirements, areas where no electricity

is available or where flow rate and head may vary con-
siderable.
Internal combustion engines

- independent power source for back-up or for, e.g. f ire-
water pumps.
Steam turbines - especially used in refineries.
For Division 0 areas the use of hydraulic drive should be considered, allow-
ing the power unit to be installed remote from the pump.
Comparison items
- energy source and availability/continuity
- cost of providing energy source
- energy cost
- initial cost
- maintenance cost
- site/location/altitude/temperature.
Driver torque available must always exceed pump torque required.
1.3.8 Seals
Statistically, seal failure is the most frequent cause of pump failure in service,
and therefore deserves far more attention than it normally receives. Seals
can be classified in the following groups.
- labyrinth seal
- dynamic screw seal
- lip seal
- packed gland seal
- mechanical seal
The last three are more commonly used in EP duty, with the mechanical seal
being by far the most common. Figure 1.3-10 shows a typical mechanical seal
layout.


Running Equipment
A mechanical seal depends for its satisfactory operation on the maintenance

of a 10 micron lubricating film of the sealed fluid between the rotating and -
stationary seal faces, so the principal limitations on satisfactory seal per-
formance are:
- seal temperature
- sealing pressure
- seal face peripheral velocity
- presence of abrasives
- gas bubbles
- vibration or shaft misalignment
It is therefore imperative that accurate information is included on pump

requisition data sheets to ensure that a correct seal material selection can be
made for a particular application.

1.4 PULSATIONS
1.4.1 Pulsation Dampeners for Reciprocating Compressors
The discontinuous nature of the flow from reciprocating compressors gives

rise to accompanying variations in pressure. The pressure variations can
cause resonance in acoustically tuned piping; `the organ pipe' effect. Pul-
sations of excessive magnitude can give rise to mechanical vibrations which
cause damage to compressors, structures and pipe supports. Amplification
of the magnitude of these pulsations can result from the interaction of the
incident and the reflected waves. This not only causes an increased pulse but
also gives rise to harmful unbalanced forces, due to phase difference in
oscillating pressure between the ends of an individual section. To predict the
pulse magnification and attendant unbalanced forces an analog or a digital
computer must be used.
The objectives attained through analysis of pressure pulsations are to

establish:
- Pulse level at each significant location and frequency

- Unbalanced force at each significant run
- Reduction in compressor power by pulsation effects
- Assessment of the benefit of piping changes, added filtering, and
attenuation devices in achieving pulsation reduction.
Most commonly, pulsation bottles are added to the suction and discharge of
each compressor cylinder. For the calculations an analog computer is pre-
fered. Modifications can be introduced easily to the system and the resulting
response is instantly observable. Many different modifications can be tried
out in a short time to reach a satisfactory solution. Also harmonics of a higher
order can be studied. Most major reciprocating compressor vendors have
their own pulsation analog computers, while independent laboratories such
as South West Research Institute, San Antonio, Texas USA and Pulsim, Deift,
the Netherlands, are able to perform check analyses. Digital computer-pro-
grammes are improving rapidly and have been used inside the Group with
satisfactory results.

Running Equipment
1.4.2 Pulsation Dampeners for Reciprocating Pumps
Reciprocating plunger pumps are widely used in production operations, both

for transferring crude oil and for injecting brine used for secondary recovery.
An inherent disadvantage of plunger pumps is the high pressure pulsations
these pumps propagate. For many plunger pump applications, gas cushion
dampeners can be effectively used to attenuate these pulsation levels. How-
ever, these dampeners are limited to pumps operating with constant suction
and discharge pressures. For large or elaborate pumping systems, acoustic
filters are normally individually designed using an analog computer study.
Between these two limits, a wide range of pumps exist that need acoustic
pulsation filters, but for which the time and expense of an analog study is not
justified. Their design is outlined below.
1.4.2.1 Types of Pulsations
Pulsations can result in forces high enough to reduce pump life and cause
equipment and piping failures.
1.4.2.1.1 Pulsations Below Primary Pumping Frequency
Pressure fluctuations in reciprocating pumps are normally of the order of 10

to 35 percent of the static suction or discharge pressure at one or two times
the primary pumping frequency (number of pump plungers times the pump
rotational speed). However, the measurements on several pumps have
shown high incidence of pulsation levels exceeding static line pressures at
frequencies below the primary pumping frequencies. Mathematical models
of flow through the pumps have indicated these pulsations to be caused by
suction and discharge valve leakage.
If a defective (always open) suction valve exists, the pump plunger simply
forces the liquid in and out of the defective valve and no net flow occurs
through the discharge valve. A leaking discharge valve also produces low-
amplitude first and second harmonics. In either case a 50 percent reduction
of flow through at least one valve was necessary to produce significant levels
of pulsation below the primary pumping frequency.
1.4.2.1.2 High Frequency Pressure Pulsations
High frequency pressure pulsations exist in reciprocating pump suction and

discharge manifolds resulting from low level excitations amplified by acous-
tical organ pipe resonances within the pump manifolds, and between the

pump manifold and adjacent filter which acts as an acoustical pipe open
end.
A possible solution to this problem would be to install an orifice at the flange

between the pump and filter. Use of pulsation filters will not necessarily
reduce these high frequency levels in the manifolds.
1.4.2.1.3 Cavitation
Cavitation, which results in erratic high frequency and amplitude pulsation at

both the pump suction and discharge is caused by the pumped liquid vapo-
rising at the pump suction valves, usually as a result of insufficient suction
head.
Suction gas cushion type pulsation dampeners can be effectively used to

avoid cavitation.
1.4.2.2 Design Criteria
Pressure pulsations caused by the piston motion, forcing the liquid through
the spring loaded valve into the discharge manifold, result in pressure waves
which create mechanical shaking forces, as the momentum of the fluid
changes due to various flow discontinuities (restrictions, piping bends,
closed ends). The purpose of discharge pulsation filters is to impede the
energy transmitted by these waves while offering minimum resistance to the
flow.
Filters with gain factors of 0.20 to 0.30 are prefered; that is 20% to 30% of the
pulsation levels at the pump discharge exist at the filter discharge. Design
criteria are discussed below for three types of filters:
1.4.2.2.1 Liquid Filled Surge Vessels
See Figure 1.4.2-1a. Due to liquid incompressibility a large volume of liquid is

needed to absorb pressure pulsations. For pumps with a flow rate of more
than 0.004 m3/s the size of this volume becomes prohibitive.
1.4.2.2.2 Gas Cushion Dampeners
See Figure 1.4.2-1b. The high compressibility of gas provides pulsation ab-
sorption. A gas pressure slightly higher than the mean liquid pressure is
needed. Depending on whether contamination of liquid by the gas is per-
missible or not, a liquid float device or a flexible membrane (impervious to

Running Equipment

chemical action by the pumped liquid) are used. From experience with a
triplex pump the minimum volume needed is about 100 times the swept
volume of one plunger; for a quintuplex pump the minimum volume needed is
about 30 times the swept volume of one plunger, or more if possible.
1.4.2.2.3 Tuned Filters - Helmholtz type
See Figure 1.4.2-1c. These involve two liquid filled vessels, connected by a
short section of small diameter pipe called a choke tube. The size of the two
vessels and the diameter and length of the choke tube determine the res-
onance frequency of the filter. Acceptable pulsation levels for tuned filters
can normally be achieved by using a design such that the resonance fre-
quency is below 60% to 70% of the primary pumping frequency (revolutions
per second times the number of pistons). The filter can be made more ef-
fective (lower resonance frequency) by:
(a) Decreasing the choke tube pipe area

(b) Increasing the choke tube length
(c) Increasing the vessel volumes.
However, it must be remembered that (a) and (b) will affect the piping head
loss.
This kind of filter is normally designed by a trial and error method. Frequently
the design of a similar installation is updated. In designing a tuned filter the
following should be considered:
(a) Accurate values of liquid bulk modulus are needed.
(b) By properly locating the entrance and exit of the filter and the choke tube,
possible standing wave organ pipe resonances within the filter vessels
(particularly relatively long ones) can be eliminated.
(c) From experience with reciprocating compressors, the choke tube length
in most filters is limited to 2 to 2.5 in by a possibility of having low fre-
quency organ pipe resonances within the choke tube corresponding to
low frequency harmonic of the pumping frequency. For liquid filters a
maximum choke tube length of 6 in is recommended.
(d) A bell mouth entrance should be used at the entrance of the choke tube
(reduction of entrance losses by a factor of 10).
(e) A choke tube diameter of one half the pump discharge pipe size is a good
first choice, choke tube pressure drop ∆p in Pa can be calculated
from:

Running Equipment
where:
(g) The two vessels should be preferably of the same size, however, if over-
all physical size is a problem, V1 should be made as large as possible and
V2 may be as small as 40 to 50 percent of V1.
(h) The resonances created by either V1 or V2 interacting with other vessels

in the system should not be within 20% of the pump excitation
frequency.
For large pumps, i.e. over 100 kW, or discharge pressures over 50 bar, an
acoustic analog study of the pumping system is recommended. This study is
similar to those undertaken for gas compressors (see 1.4.1).


Running Equipment
1.5 TURBO-EXPANDERS
Since there is little general knowledge on gas turbo-expanders, there is a

tendency to think of them as costly, complex machines of dubious quality.
However small steam turbines, effectively vapour turbo-expanders, are ac-
cepted without question and gas turbo-expanders have been in use in field
applications since 1950. Turbo-expanders are used in the following applica-
tions in EP installations:
1. Recovery of energy from expanding gas where power costs are high
2. Recovery of liquids from gas streams
3. Dew point control of pipeline gases.
Compared with Joule-Thomson expansion, turbo-expanders are very much

more efficient as polytropic work is extracted from the gas during expansion.
In applications 2. and 3. a refrigeration plant can be reduced in size or even
eliminated by the turbo-expanders, saving cost and complexity.
Turbo-expanders generate power. This power can be absorbed by:
1. A brake, usually air or hydraulic oil

2. Recompression of the expanded gas
3. Generation of electric power.
The simplest to achieve is 2., where the turbine and the compressor are
carried on the same short, rigid, shaft, the arrangement being similar to a
vehicle exhaust turbocharger. Recompression is often required if the gas is
to be transmitted or utilised for some purpose. Electric power generation is
attractive in sizes greater than 1 MW. Figure 1.5-1 is a cross-section of a
typical turbo-expander/compressor.
A typical field gas turbo-expander system for liquid recovery or dew point
control is shown in Figure 1.5-2. The expansion ratio, and hence the tem-
perature drop, will be determined by the liquid recovery required or the dew
point specification of the product gas. Turbo-expanders with matching re-
compressors are available for a wide range of pressure and enthalpy drop.
For this duty, sizes up to about 5 MW are available, and up to 80% of the
enthalpy drop can be recovered as work. A typical turbo-expander/recoin-
pressor will have an overall efficiency in the range 50-65%.
The two principal problems with a turbo-expander are:
1. Condensing liquids and ice/hydrates formation
2. Thrust balancing.

All radial inflow turbines used for this purpose are designed to condense up
to 50% of the flow in the turbine. However, hydrates can form on the nozzles,
preventing their movement by the load controller. Hydrates can be control led
by methanol or glycol injection into the inlet gas stream at a rate up to
150 g/m3 and in droplets smaller than 100µm. If these rates do not depress
the hydrate formation sufficiently or if the inlet temperature is below 30C, it is
preferable to dehydrate the gas before expansion.
Thrust balancing, which can be affected by the deposition of ice or hydrates,

is controlled by thrust balancing controllers. These dynamic devices have a
satisfactory record of service. Turbo-expander/compressors are supplied as
an ‘encapsulated' unit with all lubrication systems at the thrust balancing
pressure; shaft seals are thus eliminated.
By means of variable inlet guide vanes, the expansion ratio and flow can be
controlled over a wide range. The resulting speed and power change results
in a corresponding change in compressor throughout, which can be matched
to the turbo-expander throughput, maintaining cold separator conditions.
These processes can be satisfactorily simulated by a process computer
program.
There are several reputable turbo-expander Manufacturers whose reference

lists provide experience in particular applications. The principal suppliers
are Cryostar, Rotoflow and Mafi-Trench.

Running Equipment

1.6 GEARS AND SPEED CHANGERS
1.6.1 General
In the petroleum industry gears are used as speed reducers or increasers

between the drivers (motors, IC engines, gas turbines) and driven rotating
and reciprocating equipment. These heavily loaded gears can develop var-
ious problems under adverse running conditions.
1.6.2 Gear Selection
Up to 7 MW the general SIPM recommendation is to use single helical, rather

than double helical gears.
1.6.3 Gear Box Specification
Special requirements for gear boxes based on the accumulated knowledge

and industry experience are prescribed by the API Std 613. It contains de-
tailed specifications and sizing formulae with regard to pitting (determining
gear shaft centre distances and gear face width) and tooth bending stresses
(determining the tooth size).
The formulae are based on conservative assumptions and do not provide for
some of the important influencing factors, such as:
- the dynamic load-factor

- contact ratio describing the distribution of load on several pairs of
teeth
- actual tooth geometry
- the technique of load carrying enhancement by hardening and grinding of
gears.
In most cases this results in oversized gears, which means that a larger (and
more expensive) frame size has to be selected. It is recommended that gear
boxes are purchased conforming to API Std 613 but sized according to AGMA
which -takes the above-mentioned factors into account.
Use AGMA 421.06 when the pinion speeds are 3600 rpm and higher or the
pitch line speeds are 25.4 in/s (5000 ft/min) and higher. Otherwise use AGMA
420.04.

Running Equipment
1.6.4 Misalignment
Misalignment is one of major contributors to gear problems. Proper align-

ment of gears to the input/output shafts is essential to assure even distri-
bution of loads along the gear tooth length. While perfect alignment is prac-
tically impossible to achieve, it is important to maintain alignment within the
Vendor's tolerances.
Incorrect cold alignment instructions are sometimes issued by the manufac-

turer based on wrong assumptions, not all contributing factors considered,
unstable supports, etc. Hot checks are the best way to ensure proper align-
ment.
1.6.5 Torsional Vibrations
The torsional vibrations must be considered for all large running equipment
trains, especially when using reciprocating machinery, as they can lead to
failures of gears. The torsional critical speeds must be well outside of the
operating speed range. Changes in torsional stiffness of the couplings and
proper sizing of flywheels are some of the methods used to tune the natural
frequency of the machinery train.
When purchasing costly new equipment it is strongly advised to have tor-

sional critical speed calculations done for the complete equipment train;
followed by a factory string test, and a final ‘as installed' check.
1.6.6 Gear Tooth Problems
The different gear tooth distress modes (for illustrations see API Std 613)
rarely occur alone, one may be a result of continued nature of another, which
may make it difficult to identify their causes. The most typical gear tooth prob-
lems are:
1.6.6.1 Pitting
Pitting exhibits itself in a form of cone-shaped craters usually appearing near

the pitch line of a gear. They are caused by high frequency stresses beyond
the endurance limit of the material. These are maximum compressive
stresses, and occur about 0.5 mm below the surface. The failure starts there
and progresses to the surface until a conical piece of metal falls out.

Pitting often takes place during the initial operation of gears. The machining
process results in microscopic high and low spots which cause overstressing
of the high spots. They can develop into:
- ‘corrective pitting' - which ends when a broader contact area is establish-

ed by fine pitting and smoothing out of the surfaces takes place in oper-
ation.
- ‘destructive pitting' - when the pits become larger and closer together,
finally becoming big enough to act as a start of a fatigue crack, then a tooth
breakage (this condition can be caused by overloads or inferior lubricating
oil quality).
Pitting can also be caused by voltage differences between two meshing

gears and electrical discharges taking place. The resulting pits have a round
bottom with melted appearance, as opposed to frequency stress pitting
where the pits are conical with sharp edges and grains visible.
1.6.6.2 Spalling
Spalling is a more severe form of pitting where the cavities are larger, deeper
and of a cleaner definition. It can appear after relatively few cycles under
excessive internal stresses.
1.6.6.3 Scoring (Scuffing)
Insufficient lubrication (inadequate oil film thickness), excessive load or poor

contact may cause the teeth surfaces to weld together and immediately tear
apart as the gears turn, It appears as patches of shiny scratches up the tooth
profile in the direction of meshing, on both of the meshing surfaces.
1.6.6.4 Galling, Plastic Flow
An overload or deterioration of oil will cause galling which is a more severe

version of scoring. In the past it used to lead into plastic flow of the metal from
the pitch line toward the tip of the tooth (a wire edge appearing at the tip), but
now this is less common because high hardness (hardened and ground)
gears are often used allowing tooth loads up to 350 N/mm2 of face width.

Running Equipment
1.6.6.5 Tooth Breakage
The high root stresses above the endurance limit of material resulting in
tooth breakage or fatigue failure can be caused by:
(a) Sudden shock loads

(b) Misalignment
(c) Tooth manufacture errors
(d) Surface or subsurface defects
(e) Incorrect hardness or heat treatment
(f) Improper grinding technique.
1.6.6.6 Normal Wear
Insufficient lubricant filtration and dusty environments contribute to normal

abrasive wear of the gear teeth - a gradual degradation of the tooth surface
without pitting or scoring that will not affect satisfactory performance within
the expected life of the gears. The normal wear is proportional to how heavily
loaded the teeth are.
Normal wear can progress into destructive wear when the extent of material
removal changes the mechanical strength, shape and surface finish of the
tooth.
1.6.6.7 Corrosive Wear
High humidity, contamination of lubricating oil with chemically active sub-

stances (acid, water), and highly active EP (extreme-pressure) ingredients in
the lubricant can result in a higher rate of wear due to corrosion or rusting.
The evidence of it will also be visible on other parts of the gear besides the
active faces.
1.6.7 Start-Up Problems
It has been demonstrated that the majority (up to 80%) of gear start-up
problems can be traced back to design and quality control errors, with the
rest caused by installation errors, maloperation and miscellaneous others.
Experience shows a high incidence of start-up problems due to:

1.6.7.1 Thrust Bearings
Thrust bearings are frequently underdesigned. Uprating of the thrust bear-

ings may be required, the load carrying capability of bearings having been
overestimated by the manufacturer. High torques experienced on machinery
trains can result in alternating direction of the thrust load between the active
and inactive faces of the thrust bearing, causing failures due to overload
shortly after start-up.
1.6.7.2 Lubricant Drainage
Lubricant drainage must be assured to allow sufficient space between ro-

tating elements and the drain sump to prevent foaming of the lubricating oil. A
continuous slope and even flow in drain lines are very important. Poor drai-
nage problems are easy to avoid during the design stage, but hard to correct
in the field.
1.6.7.3 Quality Control Errors
The most common are the tooth geometry errors, which can cause mechan-
ical vibration and noise. The poor contacts resulting from these errors will
give localised gear overloads, pitting and scoring, possibly leading to tooth
breakage.
The gears with insufficient tooth contact may go unnoticed during the factory
mechanical no-load test, without exhibiting any signs of distress. However,
few manufacturers have the facilities for load testing.
If poor contact is suspected, a gear contact check (red and blue stationary
checks, Red Dykem or copper sulphate running check may be neces-
sary).
1.6.8 Condition Monitoring Instrumentation
For information on condition monitoring instrumentation see 1.9.

Running Equipment
1.7 COUPLINGS AND COUPLING GUARDS
The purpose of this Section is to explain briefly the function and behaviour of
shaft couplings.
The following aspects will be discussed:

- The significance of flexibility in couplings and the various ways it can be
achieved.
- The basic parameters and methods used in sizing and rating couplings.
The difference between general and special purpose couplings.
- Coupling types, characteristics and specifications.
- The flexible membrane coupling.
- Coupling design features.
- Coupling guards and enclosures.
1.7.1 The function of a coupling
Running equipment typically comprises a functional machine, such as a

pump or compressor, and a prime mover, which is typically an electrical
motor, steam or gas turbine, gas or diesel engine. Many running equipment
trains also include gearboxes and through-driven machines. it is necessary
to transmit torque and power from the prime mover (the driving machine) to
the functional machine (the driven machine) and this is accomplished by
connecting or coupling the shafts of these two machines.
The simplest way of achieving this is by using a rigid coupling. This could be
in the form of flanged ends on the driving and driven machine shaft ends
connected together by a series of bolts. However, more usually rigid coupling
arrangements comprise half couplings or hubs, which are assembled to the
shaft ends using keys, press fits of shrink fits (Figure 1.7-1). Such arrangements

may be acceptable if the shaft lengths concerned are short and of relative
large diameter, and where both machines are mounted on a common
torsionally stiff base plate.
If a rigid coupling is employed, it is essential that the driving and driven shafts
are carefully and accurately aligned to ensure that cyclic bending stresses
are not introduced into the shafts. Such bending stresses will cause fatigue
failure of the shafts over a period of time. Misalignment will impose loads on
the machine journal bearings, significantly reducing their life. Also any axial
loads will be transmitted across the coupling so the machines are not inde-
pendent and the position of thrust bearings must be carefully considered.
Sulzer have successfully used solid couplings on many machinery trains
between compressors and a speed increasing gearbox, with the thrust bear-
ing located on the low-speed shaft. The thrust from the compressors is trans-
mitted through a single helical gearbox using cones on the gear and pin-
ion.
Solid couplings are also used in three-bearing reciprocating compressor

drives.
1.7.2 The need for coupling flexibility
From a practical point of view, it is very difficult to perfectly align two mating
shafts. Even if trouble is taken to achieve perfect alignment on installation,
the likelihood of the two shafts remaining perfectly aligned throughout their
service life is remote.
Misalignment between two shafts will be present, caused typically by a num-

ber of factors:
- inadvertent installation errors;

- in-service wear of bearings and machine parts;
- differential thermal expansions and distortions between machines;
- machine deflections under load;
- foundation settlement or movement of the supporting structure;
- machine deflections due to piping loads; thermal, pressure and pulsat-
ing.
These factors will cause the shafts of the driving and driven shafts to be
misaligned relative to each other. Misalignment can be classified as fol-
lows:
- axial misalignment (end float);

- parallel (lateral) misalignment;
- angular misalignment.

Running Equipment
These three fundamental types of misalignment are illustrated in Figure

1.7-2; but it is important to remember that they may all be present simulta-
neously on a given installation and will exist in more than one plane. Figure
1.7-3 demonstrates possible combinations of angular and parallel misalign-
ments.
The rigid coupling of misaligned shafts gives rise to severe loads because of
the shaft's resistance to bending. These loads are carried by the machine
bearing, shortening life and possible resulting in premature failure. Howev-
er, the effect of the forces may be felt elsewhere in the machine, giving rise to
noise, vibration and shaft damage. Shaft seal failure and leakage is often
attributable to badly aligned shafts.

It is much more common to use a flexible coupling for connecting shafts, as a

certain amount of misalignment is permissible. The forces due to angular,
parallel and axial misalignment can be rendered ineffective in one of two
ways:
1) The forces may be rendered completely ineffective by the use of some

mechanical device based on-links, joints, pivots or sliding action.
2) The forces may be constrained to give rise to strains which produce

acceptable stress levels in the flexible element of the coupling.
In either case the coupling flexible element is free (within certain limits) to
move axially, at an angle and radially. Some types of flexible coupling are
radially stiff, in which case lateral misalignment between shafts is catered for
by installing two flexible elements with a spacer in between. This feature can
be exploited to ensure there is adequate distance between machines to
permit maintenance of bearings and seals.
1.7.3 Coupling sizing and rating
Couplings are designed to transmit torque, which can equally be described

as the ratio power to speed (kW/rpm). Manufacturers often quote coupling
ratings in terms of kW per 100 rpm so that the numbers are more manage-
able.
These quoted ratings are typically for a service factor of one and must be
multiplied by a factor corresponding to the torque behaviour of both driver
and driven machines. Standardised factors are given in AGMA514.02, how-
ever, manufacturers often use values based on their own experience and
designs. Factors determined for driven and driving machines must be added
to reach the factor for coupling rating. Typical factors are shown in the table
in Figure 1.7-4. Service factors are higher for greater torque fluctuations such
that the highest factor normally used is for a reciprocating gas or diesel-
engine driving a reciprocating compressor.
Coupling rating must be based on the rated output power of the driving (not
driven) machine. This will prevent any possibility of encroaching on service
factors, or indeed overloading, the coupling in the event of a future uprating
of the driven machine. It should be noted that the juncture of shaft to coupling
hub must be designed to transmit the same torque as the coupling itself. Such
junctures are made by shrinking or keying the coupling hub to the shaft in
parallel or taper configuration.
In the oil industry coupling duties are divided into two categories, general
purpose and special purpose, which should be in compliance with API 671.

Running Equipment

Maximum transient torques are likely to be encountered during start-up or

fault conditions; e.g. electric motor short circuit torque.
Service factors and application factors should be used with caution to avoid
oversizing a coupling for a given duty. Refer to section: Coupling Selec-
tion.
In addition all couplings must be capable of operating continuously with

125% of the following misalignments occurring simultaneously: -
- maximum steady-state or transient axial displacement (whichever is the
larger);
- maximum angular misalignment;
- maximum parallel offset.
Operation at greater misalignments will lower the torque transmission ca-

pability of a coupling.
A common misconception is that API 671 applies to only high duty (energy)
couplings. In fact the specification categorises three classes of duty:
- low speed; for operation at maximum continuous speed up to 1800 rpm;

- medium speed; for operation between 1800 and 5000 rpm;
- high speeds; for operation above 5000 rpm.
The requirement to utilise an API coupling depends on the power to be

transmitted at rated speed. As a rule of thumb special purpose couplings are
typically used for applications required to operate continuously and trans-
mitting powers of 500 kW or more and/or running at speeds in excess of 3600
rpm. The API definition of special purpose is the requirement to operate
continuously for an extended period (typically 24,000 hours) in an equipment
train that is normally unspared and is critical to the continued operation of the
installation.
1.7.3.2 Coupling selection
1) Select a service factor for driver and driven machines;

2) Determine the required rating in kW per rpm
P*F
Rating = 
N
Where: P is transmitted power (kW)

F is the service factor
N is shaft speed (rpm)
Generally the rating should be greater of that calculated as follows.
- With P multiplied by the design margin (application factor) and F = 1.

- With F as selected.

Running Equipment
This avoids over-sizing couplings as can happen if a service factor is used

in addition to an application factor. If in any doubt consult a coupling
manufacturer. The optimum selection may be a compromise between the
two calculations, depending on the service factor.
3) Select a coupling using manufacturers' guides, including requirements

specific to each coupling type;
4) Check that the maximum speed, misalignment capability and transient

torque capability are within the permitted ranges;
5) Select the coupling configuration, hub design and distance between shaft
ends (DBSE) to suit the application.
If there are any ambiguities, coupling manufacturers should be consulted to

confirm a selection and configuration.
1.7.4 Types and characteristics of flexible couplings
In common with the rigid coupling, the main function of the flexible coupling is
to transmit the power and torque from a prime mover (the driving machine) to
a driven machine. In addition, the flexible coupling has the important secon-
dary function of absorbing limited amounts of shaft misalignment and re-
ducing the forces associated with such misalignments to acceptable levels.
Many different principles and materials are employed in the construction of
flexible couplings, but all fall within two main categories:
1.7.4.1 Designs based on non-metallic components
a) elastomeric sleeve or hose (tyre) couplings - Figure 1.7-Sa)

b) disc or sandwich couplings (bonded or unbonded)
c) bush and pin couplings - Figure 1.7-5b)
d) rubber block and spider couplings
e) flexible ring or link couplings
f) belt and rope couplings.
1.7.4.2 All-metal flexible couplings
a) metal multiple membrane couplings - Figure 1.7-6

b) metal spring couplings (steel-grid) - Figure 1.7-7
c) metal bush or pin couplings
d) gear couplings - Figures 1.7-8 and 1.7-9
e) chain couplings - Figure 1.7-10
g) contoured disc couplings - Figure 1.7-11.


Running Equipment


Running Equipment
An alternative way of categorising couplings is as follows:
- Those that accommodate misalignment through relative sliding, and con-

sequently require lubrication. Types 2b, 2c, 2d and 2e fall into this
group.
- Those that accommodate misalignment through the flexing of an element
and do not require lubrication. All types 1, 2a and 2g above come under this
category and include a flexible component.
All these designs of flexible coupling will transmit torque and accommodate
shaft misalignment. Thus, when selecting a flexible coupling, the final choice
will probably be the result of other benefits inherent in the coupling. Low cost
would be such a benefit, but there are other characteristics which all flexible
couplings are designed to possess to a greater or lesser extent, examples
being:
i) reduction of vibration, impact loads and the effects of intermittent load

changes;
ii) reduction of backlash effects (e.g. indexing);
iii) protection against overload torques;
iv) insensitivity to atmospheric and operating environment;
v) durability and wear resistance;
vi) ease of assembly and disassembly;
vii) bi-directional operation:
viii) maintenance of balance.
Operating envelopes for various coupling types are shown in Figure 1.7-12.
Various coupling types are used in the oil industry, such as the segmental

rubber block type for pumps driven by diesel engines and elastomeric
sleeves for small electric motor-driven centrifugal pumps; however, the ma-
jority of drives utilise either gear or membrane couplings.
Over the last 20 (or so) years the membrane has superseded the gear type,
such that Shell EP operating companies are now advised to specify it when-
ever possible.
Note: In EP installations for both centrifugal pumps and centrifugal compressor

drives gear type couplings shall not be used.
The pros and cons of both types of coupling are summarized below.
Gear type Flexible membrane

Advantages
Slow failure mode (wear) Require only partial replacement
Small size Maintenance free (until failure)
Light (kg/torque transmitted) Easy to balance and retains balance
Relatively low noise generated during operation

Running Equipment
Disadvantages
Need complete replacement when Failures are sudden
worn Large size
Need lubrication Heavy (kg/torque transmitted)
When worn loses balance Relatively noisy
Difficult to balance when dismant-
led and then rebuilt
It is generally considered that the advantages of the membrane type out-

weigh the disadvantages. Membrane failures are rare, machines and bear-
ings are generally adequate for the comparatively large/heavy couplings
and noise levels are reduced by the coupling enclosure where necessa-
ry.
The major argument against using gear couplings is the requirement to

supply oil to them, both for lubrication and heat removal. An inherent pro-
blem exists as oil must be induced to flow against the centrifugal force
generated by the spinning coupling. Water and particulate matter tends to
separate out of the oil forming a sludge which cannot readily be dis-
persed.
Additional filters (down to 0.5 microns) and even cyclones have been fitted
into the oil feed to gear couplings in an attempt to minimise sludging, how-
ever, this extra equipment all requires maintenance. Also, so-called anti-
sludge couplings have been marketed by manufacturers, however, they gen-
erally require a greater flow of oil, and the additional filtration equipment
would be fitted as a precaution.
Gear couplings which are packed with grease are available for lower energy
duties, however, it has been found in some cases that the metal soaps used in
the manufacture of greases separate out and starve the teeth of lubrica-
tion.
The possible disadvantage of the membrane coupling being heavier in com-

parison to a similar rated gear coupling is being eroded by the use of light-
weight materials for spacers and hubs, such as inconel or titanium. Under
some transient operating conditions the gear coupling can display a tenden-
cy to lock-up, generating axial forces between shaft ends which are outside
design parameters.
1.7.5 All-metal flexible membrane couplings
There are two categories of flexible membrane couplings; those with a single
contoured diaphragm and those with multiple membrane ‘packs'. Both types

are radially stiff so, two units are necessary, with a spacer in between, to
cater for the lateral misalignment of shafts.
1.7.5.1 Principles of construction
1.7.5.1.1 Multiple membrane couplings
All-metal, multiple membrane couplings utilise packs of very thin metallic

plates to transmit torque and absorb misalignment. The most common type of
coupling utilises a ring-form membrane, which is alternatively bolted to an
‘input' ring and an òutput' ring; (refer to Figures 1.7-6 and 1.7-13). Thus, the
torque is transmitted by tension in alternate links. The torque transmitted is
proportional to the cross-sectioned area of the membrane pack and the
number of links; the torque capability of the membrane pack is just the same
as if it was a one-piece ring. However, by building up the disc shaped pack
from a number of very thin laminar discs, typically 0.25 to 0.38 mm thick, the
bending strain of the pack is considerably reduced under misaligned running
conditions, the stresses in the membranes being maintained within working
limits and the forces generated being kept very small.
Maximum misalignment increases with less bolts (i.e. less links). Torque
transmission increases with more bolts (i.e. more links) - lighter/smaller
coupling. Consequently there is a compromise to be made between torque
transmission and misalignment. Refer to Figure 1.7-14.
An alternative form of construction utilises packs of thin, spoked discs to

transmit the torque. In this design, the torque is transmitted by shear of the
spokes; drive comes in through the outer ring of holes and is taken out
through the inner ring of holes. Refer to Figures 1.7-6 and 1.7-13. Misalign-
ment is catered for using the same principle as with the ring form membrane.
An advantage of this design is the ability to incorporate axial softness, ef-
fectively increasing axial misalignment capability beyond that of similarly
rated ring form couplings.
1.7.5.1.2 Contoured diaphragm couplings
This type of coupling is similar to the multiple membrane type, with a solid
diaphragm in place of the membrane packs. Refer to Figure 1.7-11.
The coupling has three major parts: two rigid hubs and a spool piece. These
parts are solidly bolted together and misalignment is accommodated through
flexing of the two discs of the spool. The spool piece is made up of three
separate parts: two discs and a spacer tube, which are electron-beam
welded together.

Running Equipment
The heart of these couplings is the flexing disc; it is manufactured from

vacuum degassed alloy steel, forged with radial grain orientation, and has a
contoured tapering profile. Other materials are available. This flexibility of
construction materials enables `minor' rotodynamic discrepancies to be cor-
rected by (for instance) altering the mass of the spacer tube.
The diaphragm is contoured to maintain uniform stress when all the forces
acting on it are at their maximum. Under normal operating conditions a
diaphragm of a coupling is subjected to uniform and cyclic stresses. The

uniform stresses are generated by torque, centrifugal forces and axial defle-
ction. Cyclic stresses are induced by the angular misalignment seen-by the
disc.
a. Torque transmission
b. Centrifugal forces
Due to centrifugal action, a section through the disc is also subjected to a

radial force C. Because the disc is solidly bolted to a rigid hub, calculations of
the centrifugal force acting on the disc are rather involved, however, it can be
shown that both the tangential and radial stresses increase rapidly with a
decrease in the radius. For a uniformly stressed disc, thickness is inversely
proportional to the square of the radius.
c. Axial deflection
This type of disc deformation is similar to the head of a drum as the rim
diameter remains constant. The disc deformation has two effects: radial
stretch (the distance between rim and hub increases) and bending. The
stresses imposed by axial deflection are larger at the hub than at the rim.
d. Angular misalignment
The stresses generated by angular misalignment are influenced by the ratio

of inner to outer diameter. It has been found that for the most commonly used
ratio of 0.45 the advantage of a contoured disc is not great. In conclusion, to
ensure uniform stress levels when all the various forces acting on the disc
are at their maximum rated values, the disc must have a tapered profile.
1.7.5.2 Torque capability
As previously mentioned, the principle function of a coupling is to transmit

torque. The torque capability of membrane couplings compares very favour-
ably with other types of flexible coupling. This is illustrated in Figure 1.7-12,
which indicates the power transmission envelope of the main types of flexible
coupling as a function of speed.

Running Equipment
1.7.5.3 Environmental tolerance
The use of stainless steel membranes makes the multiple membrane cou-
pling suitable for most industrial environments including marine and refinery
installations. Alternative membranes manufactured from Monel 400 and In-
conel 718 can be supplied to meet more arduous conditions. The remainder
of the coupling would normally be manufactured from carbon or alloy steel
and these materials, together with an appropriate surface finish, would again
be suitable for most industrial environments. Special materials, such as
titanium, stainless steel and brass can be employed to meet special require-
ments. The choice of metals available means that designs of membrane
couplings can be provided to meet most environments and to permit oper-
ation from sub-zero temperatures to + 3000C.
1.7.5.4 Service life
The metallic membrane coupling, if properly designed, installed and oper-

ated within its design limits, experiences no relative movement between
components. Hence it will not wear out and theoretically will have an infinite
life. There are reliable reports of multiple membrane couplings which are
still in service more than 20 years after commissioning. It is not unknown for
the coupling to outlast the equipment it is connecting together.
1.7.5.5 Misalignment forces
The forces generated by membrane couplings are very low, particularly the
bending moment imposed on the shaft ends by angular misalignment. Loads
generated by axial misalignment are predictable and very small for small
misalignments (refer to Figure 1.7-15). The very high axial loads which may
be generated by gear couplings under `lock-up' conditions do not occur with
membrane couplings.
1.7.5.6 Dynamic balance retention
The radial or shear stiffness of a pack of membranes is very high, and hence
no radial displacement takes place across the membrane pack under the
action of any radial load which might be imposed in service. This high radial
stiffness, together with the lack of any wearing parts, endows the membrane
coupling with good dynamic balance retention properties. The residual un-
balance of the coupling as installed does not change until the coupling is
disturbed, and is not affected by the speed of operation. By comparison, the
balance of a gear coupling deteriorates during its working life and the

(relatively) radially soft elastomeric couplings go out-of-balance as speed is

increased. There has been a number of successful applications of replacing
elastomeric couplings with membrane couplings on high-speed drives to
overcome vibration problems.
1.7.5.7 Torsional rigidity
The torsional stiffness of membrane couplings is readily calculable and very

high, such that they are said to be torsionally rigid. However, by modifying the
coupling spacer design or changing the membrane selection, it is possible to
modify the overall torsional stiffness of such couplings whilst maintaining the
torque transmission capability. This can be very useful if it is necessary to
torsionally tune a power transmission train to avoid running at a torsional
resonant frequency.
A torsionally rigid coupling is free from backlash. Whereas this feature is

important on many applications, it can be a drawback on pulsating torque

Running Equipment
applications (e.g. diesel engine drives) as the coupling has no capability to

damp out the torque fluctuation. This problem can be overcome by selecting
an over-rated coupling for the duty, however, other types of coupling (e.g.
segmental rubber block) are more common for these duties.
1.7.5.8 Installation
The low axial stiffness of multiple membrane couplings under small- axial.
misalignment conditions means that comparatively small axial forces will
compress the coupling to some degree. Refer to Figure 1.7-15. This feature is
of benefit when installing the coupling, as the transmission unit can be com-
pressed to permit spigots to engage in location recesses.
1.7.5.9 Limited end float couplings
These are necessary for electric motors with sleeve bearings to restrict the
motor rotor axial movement at times when the motor cannot centre itself. The
requirements of a limited end float coupling are:
1) To allow the motor to remain on its magnetic centre during normal run-
ning;
2) To allow some axial movement from thermal growths without significantly
displacing the rotor.
3) To restrict the motor rotor movement under abnormal or shutdown con-
ditions.
The non-linear axial stiffness characteristic of the multiple membrane cou-

pling meets these requirements as:
- It has a very low axial restoring force at low levels of axial displacement,
thus absorbing thermal growth without significantly disturbing the motor
rotor;
- The low axial forces generated by small axial displacements allow the
motor to run on its own magnetic centre;
- Large axial movements of the motor rotor are prevented under shutdown
conditions as the axial loads generated in the membranes by large axial
displacements are generally higher than the aerodynamic forces, etc.,
imposed on the motor rotor.
The radial spoke membrane coupling is generally considered more suitable

as a limited end float coupling than ring form designs. The gear coupling
when used with sleeve bearing motors has to be constrained by abutting
against shoulders in the hubs to ensure a limited movement.

1.7.6 Design features
Following are some of the major design features (extracted from API 671)
required in special purpose coupling installations, particularly high speed
drives.
`(2.1.4) All couplings shall be of the spacer type. Spacer shall be of sufficient
length to allow removal of coupling hubs and to allow maintenance of ad-
jacent bearings and seals without the removal of the shaft and without dis-
turbing equipment-alignment. Therefore, the minimum spacer length shall
correspond to a between-shaft-ends dimension of 18 inches (457 millimetres)
unless otherwise specified by the purchaser.'
`(2.3.3) When the flexible elements of a coupling in intermediate or high-

speed service are combined into factory-assembled disc packs, the coupling
spacer shall be removable without disturbing the factory assembly of the
elements (discs).'
The intention of this clause is to provide better inherent balance and relia-
bility.
`(2.3.4) Actual axial natural frequencies of the coupling shall not fall within 10
percent of any of the following:
1) Any speed within the range from the minimum allowable speed to the
maximum continuous speed;
2) Two times any speed within the above speed range;
3) Any other speed or exciting frequency specified by the purchaser.
(These restrictions may, forsome applications, preclude a practical coupling

design. In such cases, the vendor and the purchaser shall investigate suit-
able relaxation of these criteria).'
The non-linear axial stiffness characteristic of multiple membrane couplings

provides a means of detuning the system.
‘(2.5.1.1) All coupling components shall be match-marked. For couplings

requiring assembly balance or assembly check balance, match-marking
shall be performed after the balance or balance check and prior to subse-
quent disassembly of the coupling.'
It can be seen that, in principle, API 671 places great importance on ensuring
a high degree of balance with the installed coupling, both when new and
following removal and replacement. In addition to the above the specification
includes the following requirements:
- Interface of adjacent components must be centred by a spigot to recess

location, with a maximum clearance specified.
Running Equipment

- Interface bolt maximum diametral clearance specified. However, a posi-
tional accuracy is not mentioned and one manufacturer has suggested 0.06
mm. Inaccuracy in pitch location could result in the bolts exerting radial
loads on the flanges which override the spigot/recess location.
- Bolts/nuts which are removed during normal disassembly must be weigh-

balanced to a tolerance of 0.1 gram.
Other notable features of an API 671 coupling are: (see Fig. 1.7-16).
- Hubs are often reversed to reduce overhung moments;
- The outer surface is smooth to minimise windage;
- Factory assembled membrane packs.
1.7.7 Guards and enclosures
1.7.7.1 Guards
It is necessary to guard adequately all couplings for the safety of personnel.
This shall include the prevention of contact by personnel. Guards shall be

strong enough to support the weight of a man without deformation. Gaps
between guard and equipment shall be sufficiently small to prevent access.
The specification B55304 provides guidance on guard design.
For the majority of general purpose couplings a simple screen is sufficient. If

installed in a hazardous area this must be manufactured from a non-sparking
material, such as copper alloy (brass) or glass-fibre.
1.7.7.2 Enclosures
For high-speed applications the coupling guard is often an oil-tight enclo-

sure, with the purpose of trapping any oil escaping from the machinery
bearing housing oil-seals. This arrangement is obviously necessary for gear
couplings which are continuously lubricated, but is not essential with the dry
membrane coupling. Whenever possible simple screens should be installed,
possibly of perforated metal to ensure adequate heat dissipation.
A coupling enclosure is also a guard, which must be used on all couplings as

described in 1.7.7.1 above. Due to their much greater inherent rigidity en-
closures are not usually made of non-sparking materials as with general
guards. To satisfy its main purpose, a coupling enclosure must be oil-tight,
but not necessarily air-tight. Coupling enclosures should be provided with
breathers.


Running Equipment
When non-lubricated couplings were initially used on high-speed machinery,

attempts were made to replace the relatively cumbersome enclosure with an
open mesh guard. These attempts were not always successful, and it is now
recognised that enclosures cannot always be eliminated. Even if the quantity
of oil leaking past the seals is minimal, an installation might lose all of its oil
over a period of time. The resulting oil mist surrounding the machines could
become a fire hazard. Coupling enclosures are also needed for some appli-
cations because atmospheric dust tends to accumulate in the crevices of a
coupling.
1.7.7.2.1 Enclosures - Heat generation
Coupling enclosures were successfully used with oil lubricated couplings for
many years. When the first membrane couplings were installed in oil tight
enclosures, it was found that the enclosures could become very hot. The oil
flow to the gear type couplings had not only lubricated the coupling, but also
cooled it.
It was initially believed that heat was generated by the friction between the
gear coupling teeth, however, tests have shown that this friction is very small
and accounts only for a few degrees of the total temperature rise. It is now
believed that ‘windage' is the principle cause of heat generation in a coupling
enclosure.
(a) Heat generated by air shearing
The air in an enclosure is contained in an annular cylinder, of outer diameter

(OD) equal to the enclosure diameter, inner diameter (ID) equal to the cou-
pling's diameter, and a length equal to that of the enclosure. While the OD of
this annular cylinder is stationary, the ID rotates with the coupling. A short
time after start-up, the air in the enclosure also starts rotating. The velocity of
this rotation varies: with maximum velocity at the coupling and a minimum in
the vicinity of the enclosure. Hence, friction occurs between adjacent layers
of air and this friction generates heat.
(b) Heat generated by air turbulence
Typically the three elements of a coupling (two flexing elements and one
spacer) have dissimilar diameters. The spacer is smaller in diameter than
the rest of the coupling; this being particularly evident with diaphragm type
couplings. Due to the friction between the air and the face of the discs a
circular motion is generated, and this motion generates additional heat.

Turbulence is also generated by the heads of bolts and nuts connecting the
various components of the coupling. A well designed coupling incorporates
shrouds over bolt heads and nuts, particularly those which are placed at
large diameters.
1.7.7.2.2 Enclosures - Heat dissipation
(1) Radiation
The heat generated within the enclosure is normally dissipated by its outside
surface into the surrounding air. To increase this dissipation common engi-
neering sense should be used:
- increase the surface area, either by making the enclosure large, or by
adding radiating fins;
- allow air circulation over the enclosure;
- shade the enclosure from exposure to the sun or from other hot compo-
nents, such as turbine exhausts.
(2) Oil cooling
If the dry coupling has been installed as a retrofit such that the enclosure
cannot be modified, and its temperature is unacceptable high, spraying the
coupling or the inside surface of the enclosure with oil will dissipate addi-
tional heat.
(3) Air cooling
If the air heated by friction is allowed to escape from the enclosure and be
replaced by outside air, the enclosure's temperature can be significantly
reduced. Venting the air out of the enclosure could create the problem of oil
mist escaping into the atmosphere, however, this problem can be eliminated
by the proper design of enclosure vents, ports and baffles.
The following features should be considered when designing an enclosure
with air cooling:
(a) Sizing the air ports
The exhaust port should be tangential to the enclosure's outside diameter

and directed with the coupling rotation, as shown in Figure 1.7-17. This port
(or ports) can have any profile; however, a tubular port is the most convenient
to use. The larger the port, in comparison with the enclosure, the lower the
enclosures temperature. A large exhaust port, on the other hand, can create
problems in separating the oil from the air.

Running Equipment
The intake and exhaust ports should be placed so that no ‘dead zone' is left
in the enclosure. Hence, if one exhaust port is placed in the middle of the
enclosure's length, one intake port should be used at each end of the en-
closure. Conversely, if two exhaust ports are used (this is desirable for
enclosures longer than 36 inch), then one intake port should be placed half-
way between the exhaust ports.
The total intake port area should be half the total exhaust area. For example,
if one exhaust port of 3 inch diameter is used, each of the two intake ports
should be 1.5 inch in diameter.
(b) Separating the-oil from the air
The amount of oil that escapes past a seal is usually very small. However,
some compressor manufacturers prefer to eliminate the seals completely
and to recapture the bearing oil from the coupling enclosure rather than from
the compressor housing. It is therefore possible that flow rates of up to 3 gpm
are still flowing into the enclosure, even with a non-lubricated coupling. This
is hot oil and does not necessarily provide cooling.
To prevent oil from becoming a mist, many enclosures rely on the centrifugal
effect created by the rotation of the air, similar to the working of a cyclone
separator. One way to help the air rotation is to install the intake port(s)
tangentially to the enclosure, as shown in Figure 1.7-17. Another way to
prevent the formation of oil mist is to shorten the residency in the enclosure of
any oil particle. An oil tight baffle should be installed along the enclosure,
over the exit port (Figure 1.7-17). This baffle will direct the oil out of the
enclosure and prevent if from rotating more than one revolution.
The actual separation between air and oil is done outside the enclosure in a
vertical separation chamber. All ports should have breather caps, provided
with a coarse wire mesh filter.
(c) Avoiding negative pressures
When large diameter discs are mounted on shafts in the vicinity of the seals,
their rotation generates a negative pressure, which induces oil to be sucked
out from the machine bearing housings into the coupling enclosure. As cou-
plings are mounted closer and closer to the bearings (in order to reduce the
overhung moment), the problem of oil being lost into the enclosure is en-
countered more often.
To minimise the problem, an air tube can be installed close to the coupling.
This tube allows atmospheric pressure near the seal and cancels the nega-
tive pressures (Figure 1.7-18). Preferably, this air tube should be in addition

Figure 1.7-17 Coupling enclosure - cross-section. Separation of oil from air.
to the normal air inlet. If the tube does not eliminate oil loss it should be
connected to an area of positive pressure; usually over the largest diameter
rotating element.
1.7.7.2.3 Basic enclosure design features
Usually coupling enclosures are cylindrical and split on a diametrical plane.

The two half-cylinders are connected by bolted flanges. A few basic design
rules should be followed:
(a) A coupling enclosure should be free to float at both ends. If the enclosure
were bolted at one end it would be free more axially, but would be rigid
with respect to offset motions.
(b) The enclosure should be oil tight. The usual leakage path is at the ho-
rizontal flanges; gaskets and caulking seldom can stop leaks. One so-
lution is to include dams on the inside of the enclosure at the flanges to
prevent oil mist being centrifuged directly onto the joint (Figure-1.7-17)

Running Equipment
Figure 1.7-18 Compensation of negative pressures
(c) Breathers are required on enclosures. These should be protected from

direct impingment of oil mist by means of a baffle, similarly to (b) above.

1.8 FOUNDATIONS AND BEDPLATES
1.8.1 Installation of Running Equipment on Concrete Foundations
This Section provides guidance on the design and installation of foundation

bolts, mounting chocks and grouting of running equipment. It does not in-
clude the design of the foundation; this is a specialised activity depending on
the unbalanced forces and couples, the nature of the soil and the applied
external forces due to piping thrust, etc.
1.8.1.1 Foundation Bolts
Foundation bolts should be set in sleeves which are at least twice the diame-
ter of the bolt. The bolt should have a free length of 40 bolt diameters within
the sleeve. This allows sufficient stretch in the bolts when they are tightened
to the required torque so that the nuts do not slacken under vibrating load.
The method of attachment of the bolt head to the sleeve base is important,
if it should break loose, the entire bolt will have to be removed. Under no cir-
cumstances must the sleeve annulus be filled with any concrete, grouting
material, etc. It is essential to ensure that the annulus is properly plugged
with soft packing before concrete pouring commences (see Figure 1.8-1).
Foundation bolts, in their sleeves, should be supported from a rigid frame
during pouring of the concrete. Needless to say, the set-out of the bolts must
be accurately checked against the drawings as mistakes can be very difficult
to correct once the concrete has set.
1.8.1.2 The Chock and Block Method of Grouting
Continuous load bearing grouting beneath the machine, or its bedplate, is

neither necessary nor desirable. The `chock-and-block' method described
here is proven, simple and completely reliable. It should be specified in the
Machinery Purchase Specification as the machine Vendor will have to ensure
that the underside of the bedplate is prepared for the mounting shims. The
Vendor should supply the chocks, shims and foundation bolts.
Refer to Figure 1.8-2. After the concrete has cured, it is good practice to coat
the entire foundation block with an epoxy paint to prevent oil penetration of
the concrete. The area where the chocks are to be grouted should not be
painted but should be scarified to remove ‘laitence' and loosely adhering
material, a good bond between the grout and sound concrete is essential.
The chocks are now placed, aligned and accurately levelled using their
Running Equipment


Running Equipment
levelling screws, on the foundation. A dam is built around each chock to

contain the grout. A tube must be placed around the foundation bolt to ensure
that no grout whatever flows into the annulus. Use a specialised epoxy
grouting material (Ceilcote or equal) mixed strictly in accordance with the
Manufacturer's instructions. The temperature of the mixture is particularly
important. Non-shrinking hydraulic cement based grouts are useless for this
application, their properties are completely inadequate for the purpose. The
grout should be placed beneath each chock without disturbing it, making sure
that there are no voids whatever beneath the chock, and the grout is poured
to within 1 cm of the top of the chock, and allowed to cure. Figure 1.8-3 shows
a compressor foundation at this stage. Check that the foundation bolts are
still free in their sleeves. After curing the grout, remove the dam, levelling
screws and tube and finally check the alignment and level of each block. Fill
the levelling screw holes with epoxy; the foundation is now ready to receive
the machine.
The shims are first placed on the chocks. These shims are accurately ma-
chined in thickness to 25 ± 0.1 mm and are profiled to fit around the foun-
dation bolts and machine levelling screws. The machine is lowered onto the
foundation, and allowed to rest on the shims. All alignments and crankshaft
web deflections are checked, final adjustments are made using the machine
levelling screws. When alignment is complete, use feeler gauges to accu-
rately measure the gap between the bed plate and the top of the shim at
locations where the levelling screws have been adjusted. The best way of
compensating for these adjustments is to reduce the thickness of the thicker
shims on a surface grinder, each by an amount such that the machine is
perfectly aligned when standing on the shims and all levelling screws slack.
Alternatively, where a surface grinder is not available, the gaps should be
made up by packing stainless steel shim stock, profiled to the shape of the
shims and placed between the chock and the shim. Slacken all levelling
screws and tighten the foundation screws to the correct torque, and finally
recheck alignment. Always use a nut and lock-nut on the foundation bolts of
reciprocating machinery.

In the event that subsequent adjustment of the alignment is necessary, the

machine can be relevelled using the levelling screws and correcting the shim
thickness to the new conditions, It is advisable to recheck alignment after the
first 10, 100, 500 and 2500 hours running after commissioning.
To prevent oil spillage from flowing into the void beneath the bed plate, the
edges of the void may be packed with a stiff Portland cement grout. This grout
should be faired off level with the underside of the base plate to deflect
spillage away from the edge of the base plate. The grout should never over-
lap the machine base, any vibration will break it away.
It is almost inevitable that after ten years running the concrete will have
absorbed sufficient oil as to seriously impair its strength. At this stage it is
preferable to remove the machine entirely and to break down the foundation
until clean concrete has been reached. The foundation should be rebuilt,
using new foundation bolts, as described herein.

Running Equipment
1.8.2 Installation of Running Equipment on Offshore Platforms
On offshore structures, the problem is to have a structure with sufficient

rigidity to maintain machinery alignment with minimum weight. The use of a
three-point-mounted baseplate can solve this problem. It is rigid, light and
frees the module design from stiffness constraints dictated by the machinery
therein. See Figure 1.8-4 for a three-point baseplate design.
The baseplate has sufficient rigidity to maintain shaft alignments, and is

supported in a manner such that plattorm movements will not affect the
alignment of the machinery on the baseplate. This method can be used for
pumps, compressors and generator sets in all sizes.
Some consideration must be given to the baseplate design. All reputable

Vendors have programs to design stiff three-point mount baseplates, the
following points should be observed:
(a) Position the heaviest machinery on the baseplate at the end with the two
supports.
(b) The mountings must be designed to allow for longitudinal expansion in

the baseplate; the single bearing is made to slide longitudinally.
(c) The baseplate must allow lifting at one end only without distortion.
(d) Each mounting must be designed to withstand all static and dynamic
loads and for deflection of the module up to 60 in any direction.
(e) Piping to the machinery must be anchored to the baseplate, preferably

near the baseplate neutral axis. Flexibility is then allowed between the
baseplate and the module piping supports. A careful piping stress
analysis is neccessary.
(f) Module and baseplate design must be co-ordinated to ensure that the
module has strong points where required for the three-point mounts for
the baseplate.
Designs incorporating three-point supports for machine bases can make a

substantial reduction in overall platform weight as the module design does
not need to be stiff to provide a rigid machine base. It may be necessary to
use temporary shipping bracing, which is removed once the module is placed
in position.


1.8.3 Some Notes on Baseplate Design
Baseplates are commonly used for transportation of machinery as a package

to the operations site to reduce construction time. The design often considers
only the strength required for transport, erection and operation with no con-
sideration of dynamic behaviour. This is of little consequence on high speed
machinery, but can be troublesome on low speed reciprocating machinery.
The following should be considered in the baseplate design:
(a) The machine Vendor should check the baseplate vibration response at
the running speed, first and second harmonics to ensure that there are no
coinciding resonant frequencies.
(b) Ensure that the baseplate foundation bolts and the machinery holding-
down bolts are in the same vertical line (see Figure 1.8-5). Offsetting can
cause resonances.
(c) Baseplates are designed either low of `high' tuned. Low tuned means
that the resonant frequency of the baseplate is below the running fre-
quency. The baseplate is supported on flexible mounts and vibration will
transmit only a small force to the surrounding structure. Used for offshore
platform design. High tuned means that the resonant frequency is above
the running speed, implying a heavy foundation structure. The decision
for either type of support depends upon the nature of the machine, the
structure it is installed in, and the nature of the soil.
(d) Analysis of a vibration problem may reveal structural resonance. The

stiffness of a steel structure can be modified removing or adding mem-
bers, or packing voids with steel shot filled epoxy resin.
(e) Ensure that the baseplate is torsionally stiff.

Particularly with two pole motor and IC engine drives, a structural analy-
sis to determine, and if necessary adjust, the torsional natural frequen-
cies of the baseplate and the supporting machinery.
(f) Check that hot piping near a baseplate does not distort it. Lube oil tanks in
the baseplate cause distortion unless design has accounted for the thermal
growth induced by the hot oil in the tank.

Running Equipment

1.9 CONDITION MONITORING FOR RUNNING EQUIPMENT
1.9.1 Objectives of Condition Monitoring
The objectives of equipment condition monitoring are to achieve one or more

of the following:
- Plant safety improvement

- Equipment availability improvement
- Operating cost optimisation
- Environmental safeguarding.
Condition monitoring requires capital investment and a commitment to oper-

ate and maintain a monitoring system throughout the plant operating life. A
cost-benefit analysis must be part of the decision making process when
deciding whether condition monitoring is justified at all for an item of equip-
ment and the level of monitoring to be undertaken. Clear economic benefits
must be demonstrated before proceeding.
1.9.1.1 Plant Safety
If a machine represents a liability to the safety of life or plant, either directly or

as a result of the process of which it is part, then it should be monitored.
1.9.1.2 Equipment Availability
if a reduction in both scheduled and unscheduled machinery down-time in an

unspared arrangement can be achieved as a result of an increased aware-
ness of the machine health and operating conditions, then condition moni-
toring can be justified. Increased awareness allows intelligent decisions to
be made on the frequency and scope of maintenance interventions.
In an unspared environment, health monitoring often can be justified on the
basis of the value of deferred product relative to time. The cost of down-time
can therefore be easily established but it is important to determine accu-
rately the length of down-time due to a potential machine fault. This requires
reliable Mean Time To Repair (MTTR) statistics.
1.9.1.3 Operating Costs
Operating costs can be reduced if health monitoring results in machine being

removed for servicing prior to catastrophic failure. By structuring mainten-
ance interventions, unnecessary activities are avoided allowing reduced
maintenance costs - both in terms of spare parts consumption and man-
power requirements.

1.9.1.4 Environmental Safeguards
If health monitoring techniques can be demonstrated to assist in the prev-

ention-of accidental discharges then monitoring activities should be esta-
blished.
1.9.2 Condition Monitoring Techniques
Equipment Protection: In order to protect equipment from further deteriora-

tion or more serious damage, on-line protection instrumentation provides
signals for a machine trip. The machine trip system, however, may be com-
posed of several types of monitoring equipment and process control instru-
mentation, which will activate a trip individually or by multiple voting. This
machine protection is continuous (on-line) monitoring and takes place with-
out human intervention. In order to avoid unnecessary equipment unavaila-
bility, efforts should be made to be extremely selective in defining the ma-
chine trip system. A machine trip should only be activated when continued
operation could cause damage to the equipment or associated plant, give
rise to unsafe operating conditions or would jeopardise continued produc-
tion.
Machine protection systems must be considered to be the most basic form of

condition monitoring and fall outside the current efforts to optimise condition
monitoring systems.
Condition Monitoring Trending: This is usually done through a combination of

different techniques and inspections, both intrusive and non-intrusive. Intru-
sive techniques are used on an opportunistic basis when equipment is out of
service, to establish its condition. Borescoping is the best example of this
method. Non-intrusive techniques involve passing judgement on equipment
condition via indirect or external methods and are made with the equipment
operating and on-line.
A set of baseline measurements for future reference are of prime importance

during initial commissioning. Specification of the operating conditions at
which the baseline data should be taken requires careful consideration so
that trending of parameters can be made at, or close to, the same operating
conditions. Also trends develop at different rates so that experience with the
rate of change is important in order to determine the monitoring interval.
With the advances in microcomputer-based systems, the ability to accumu-

late parameter data has been greatly enhanced. On-line systems have
gained limited acceptance, but for installed equipment and the majority of

Running Equipment
new installations, hand-held data loggers used in conjunction with desk top
computers are all that can be justified. However a clear and precise need for
the accumulation of data must be established prior to the implementation of
any system, otherwise a proliferation of superfluous data can quickly develop.
Condition monitoring and accurate failure prediction is however no substi-

tute for the full resolution of the cause of an equipment fault. The majority of
faults in EP are the result of equipment operating outside its original design
parameters, while relatively few can be attributed to a deficiency in the
manufacturer's original design. Whatever the actual fault, steps must be
taken to resolve the cause and ensure that it becomes non-recurring, if
economically feasible.
In order to establish the appropriate level of condition monitoring for a spe-

cific item of equipment, the equipment classification must first be esta-
bushed. One classification description is as follows:
Vital Equipment: Equipment which, when failing in operation or when failing if

called upon to operate, can cause an unsafe condition on the installation,
jeopardise life, or cause major damage to the installation.
Essential Equipment: Equipment which, when failing in operation or when

failing if called upon to operate, will affect either the continuity of operation,
or the quality or quantity of product.
Non-Essential Equipment: Equipment that is neither vital nor essential.
Once a classification has been made, and an equipment maintenance strategy

established, the level, type and frequency of condition monitoring can be
formulated. The justification of a condition monitoring system must be rig-
orous. Clear economic advantages must be obvious before any effort is made
to proceed.
The following survey outlines the parameters which can be monitored for
each equipment type. Whether or not these are applied will depend on the
individual application and maintenance strategy.
1.9.3 Gas Turbines
1.9.3.1 Protection Systems
Gas turbines are provided with a standard alarm- and trip system, which
provide equipment protection on five key parameters:
- overspeed

- lube oil low pressure or high temperature

- high turbine exhaust temperature
- loss of flame in the combustion system
- high vibration.
1.9.3.2 Condition Monitoring Trending
The maintenance philosophy established for gas turbines should be one of

overhaul on-condition. The overhaul interval must be established for each
type of equipment at any one location, and must be optimised to limit over-
haul costs by ensuring that the majority of hot section- parts are repairable
rather than be replaced at overhaul.
(a) Non-Intrusive
Exhaust Temperature: The exhaust temperature of the gas turbine is the best
parameter for assessing the condition of the combustion and hot section
components. Damage to the turbine can occur very rapidly as a result of
operation at a firing temperature above the manufacturer's recommended
maximum, or as a result of an uneven exhaust temperature distribution.
Depending on the type of gas turbine, the temperature measured will be the
power turbine entry temperature (T5), or the turbine exhaust temperature
(T7). Most combustion or hot section problems will be manifest as deviations
- in measured exhaust temperatures. Some turbine manufacturers now pro-
vide temperature spread protection associated with the over temperature
protection, but trending can highlight combustion or hot section problems as
they begin to develop. The exhaust temperature can also be used as an
indirect parameter to indicate compressor fouling and intake filter problems.
When efficiency decreases as a result of fouling, more fuel will be required
for the same output, resulting in an increased exhaust temperature. For
example, an increase of 15 degrees Celsius exhaust temperature for similar
load conditions could indicate the need for compressor washing.
Vibration: Changes in vibration levels can be indicative of a change of bal-

ance or bearing wear. However it is important to ensure that the correct type
of vibration sensors are fitted, so that good machine protection is achieved,
but the data obtained can also be used for trending purposes. For aero-
derivative gas turbines, accelerometers should be utilised and mounted on
the casing of the gas generator and power turbine. A band pass filter between
25 Hz and 1 kHz allows at least the first and second order frequency com-
ponents (those which detect rotor unbalance and misalignment) to be covered
by the protection system. Higher frequencies covering blade pass frequencies
and gear mesh frequencies associated with any auxiliary gears are available
for trending considerations.

Running Equipment
On industrial gas turbines, manufacturers now often fit proximity probes in

the journal bearings of the turbine instead of the more traditional seismic
velocity probes mounted on the bearing caps or turbine case. The reason for
using pickup types other than accelerometers is because these devices on
heavy casings may not be sensitive enough to detect rotor or bearing pro-
blems. Measurement of casing vibration may be insensitive, particularly
when externally induced casing vibration is present which may mask the first
and second order vibration generated by the turbine. An increase of casing
vibration may only be noted when a catastrophic failure has occurred. Prox-
unity probes are selected because they give the-actual relative motion be-
tween the rotor and bearing irrespective of the casing vibration. The disad-
vantage of proximity probes is often inaccessibility in the case of failure, and
difficulties of checking calibration. The axial position of the power turbine and
or compressor turbine is sometimes monitored using proximity probes on
the thrust collar.
Bearing Temperature: On industrial turbines, tilting pad radial and thrust

bearings are sometimes fitted with thermocouples (TC) or resistive temper-
ature detectors (RTD). Trending of bearing metal temperatures can give
warning of impending problems with lube oil supply, bearing overload, or
failure, and is a more direct and thus more indicative method of monitoring
than measuring the lube oil temperature in the drain line of the bearing.
Monitoring of thrust bearing temperatures gives an indication of the thrust
load on the turbine. Changes in thrust loads can result from excessive seal
leakage which disturbs the thrust equalisation balance included in the ma-
chine design. Usually two RTDs per journal bearing are fitted allowing some
measure of redundancy. Roller bearings in aero-derivative turbines are
usually monitored using lube oil drain temperature.
Wheelspace Temperature: On industrial turbines the wheelspace tempera-

ture between the various turbine stages is often monitored, If trended, cool-
ing air supply and seal leakage problems can be detected before permissible
turbine disc temperatures are exceeded.
Performance Monitoring: Periodic turbine performance assessments should

be carried out for two reasons. The frequency and effectiveness of turbine
compressor water wash needs to be confirmed, and the natural power loss of
the turbine through its operating life needs to be established. The mainten-
ance strategy chosen may dictate that turbine overhaul be carried out when a
certain non-recoverable power loss is experienced, due to power require-
ments in service, or the optimisation of turbine overhaul costs. For a gas
turbine in power generation duty, performance monitoring is relatively
straightforward, especially if ambient variations are small, but in compressor

or pump drive applications power output is not easily established unless a

torque measuring coupling is fitted. Often derived parameters or correlations
can be used but care must be taken with accuracy and repeatability of
instrumentation.
Power output can be measured by using a shaft torque measuring device

fitted in the load coupling. This equipment is now becoming reliable in ser-
vice, and there is some accumulated EP experience. Care must be taken if
the retrofit of a measuring coupling is being contemplated as the inertia of the
torquemeter spacer and coupling flanges may affect the dynamic behaviour
of the complete train, It is recommended that a full rotor dynamic analysis be
carried out as part of any retrofit feasibility study.
Alignment: With the more widespread use of dry flexible couplings and their
greater tolerance to misalignment, monitoring of alignment need only be
carried out when known problems exists. Unless the equipment is mounted
on an offshore platform, and assuming it is dowelled in location, there should
be no shifting in service except for thermal effects. Hot alignment techniques
have been in use for a considerable time, using Essinger bars, but more
recent developments now enable hot alignment checks to be performed
using laser alignment instrumentation with an on-line monitoring capability.
This provides and accurate check on whether calculated growths have been
achieved and can assess any diurnal or seasonal influences on alignment.
Air Inlet Filters: Air inlet filters are a major source of normal operating costs
but can be the cause of turbine unreliability if the differential pressure rises
quickly to the shut-down value during, for example, short periods of high
humidity. Trending of pressure drop over the filters enables the performance
of a partially clogged filter to be assessed during periods of climatic change
so that filter change-out intervals can be optimised without impacting ma-
chine reliability.
Lube Oil Analysis: If the load is a compressor with a common lube and seal oil
system, regular viscosity and flash point tests must be carried out. In other
applications, only infrequent analysis is required, as turbine oil is not ex-
posed to combustion products.
For aero-derivative gas turbines with roller or ball bearing systems there are
three additional testing procedures available:
- Magnetic Particle
- Ferrography
- Spectographic Oil Analysis.

Running Equipment
To be fully effective, magnetic plug and sample points need to be located in

each scavenge line so that potential problems can be easily isolated. Mag-
netic Particle and ferrography tests are used for particles above 10 microns
but are obviously limited to ferromagnetic debris. Spectographic or ‘SOAP'
tests can effectively detect all metals and additives below 10 microns.
Acoustic Measurements: In some gear train arrangements, especially com-

plex high speed accessory drives, sound level readings can be more effec-
tive in providing warning of impending gearbox problems than absolute
vibration measurements. For trending to be meaningful, care must be taken
to ensure consistent reading locations and operating points.
(b) Intrusive
Borescope Inspection: More than any other single method, a borescope

inspection of a gas turbine gives the best opportunity to review the condition
of the machine short of a complete stripdown. Most manufacturers recom-
mend an inspection on an annual basis. The inspection interval does have
some flexibility, and should be programmed to coincide with a general plant
outage on an opportunity basis rather than require a separate shut-down.
Care must be taken to ensure that adequate training is given in both the
operation and maintenance of borescope equipment, and in the familiarity
with the machine internals. Turbine manufacturers are becoming more
aware of the advantages to users of good borescope facilities and new ma-
chines have a larger number of ports with easier access. Also documentation
has improved with clear reference to allowable deterioration limits in specific
areas. However considerable user familiarisation is still required before good
condition trends can be established. To maintain permanent records, a 35 mm
or Polaroid camera should be used which adds further to the inspection learning
curve. Video recordings can also be used to add a further dimension to effective
record keeping.
1.9.4 Centrifugal Compressors
The standard alarm and trip systems on a centrifugal compressor normally

consist of the following parameters:

- low seal oil level or differential pressure
- vibration
- bearing temperature.

The maintenance philosophy for a centrifugal compressor should-be to over-

haul on-condition with a target TBO (Time Between Overhaul) of at least 7
years. Special attention should be paid to compressor condition while plan-
ning associated scheduled plant outages so that compressor overhauls can
be carried out on an opportunistic maintenance basis.
(a) Non-Intrusive
Vibration: Trending of radial vibration levels can give background on bearing

condition and rotor imbalance. Theoretically, there should be no change in
rotor balance in service, but an uneven build-up or shedding of fouling pro-
ducts will provide an imbalance condition.
Performance: Periodic performance assessments should be carried out to

assist in determining the compressor condition. This should be done some
considerable time in advance of a planned shut-down as part of the planning
activities, so that there is ample time to organise the necessary resources.
Fouling of impellers and discharge volutes will reduce throughput, as will
excess leakage across labyrinth seals. Monitoring of leakage rates across
interstage labyrinths on back-to-back compressors is especially important,
as excessive leakage can lead to significant performance losses.
Of great assistance in the determination of compressor performance when

the driver is other than an electric motor, is the use of shaft torque meas-
urement, as outlined for gas turbines. The same comments on approach
apply equally well to compressors.
Sour Seal Oil Flows: Monitoring of the individual seal oil flow rates to the
traps gives a good indication of the condition of individual seals and can warn
of high sour seal flows before the sour seal handling facilities become over-
loaded. On systems using differential pressure seal oil controllers, excessive
sour seal flows can be the result of incorrect control set points.
Bearing Temperature: Radial and thrust bearings are instrumented to allow

trending of bearing metal temperatures. As with gas turbines, this can give
warning of problems with the lube supply or bearing overload. Thrust bearing
temperature monitoring is important in order to judge the condition of the
balance drum seals. Wear of the balance drum seals or a change in the gas
molecular weight may not affect the axial position, but will produce change in
thrust load which will result in a change in thrust bearing temperatures.
A load cell mounted in the thrust bearing can also be used to detect excessive
thrust load and is a direct measurement. However this measurement is

Running Equipment
usually not standard and often only used to confirm thrust load calculations
during development testing.
Alignment: The same comments made under gas turbines apply equally well
to centrifugal compressors.
Lube Oil: Even if there are separate lube and seal oil systems, viscosity and
flash point tests must be carried out at regular intervals on both systems.
1.9.5 Centrifugal Pumps
Depending on the size of the pump, the following protection systems may be
provided for the followed parameters:
- high vibration
- pump casing high temperature
- seal leakage flow high.
For single or two stage process centrifugal pumps designed to API 610 Stan-
dard with an installed spare, a 3 - 4 year run life should be achievable
without condition monitoring, and this should be reflected in the maintenance
philosophy developed. Monitoring may be required to assist in the resolution
of specific problems, but can seldom be justified on an ongoing basis.
For large multistage `between bearing' shipping and water injection pumps,
the maintenance philosophy chosen should be similar to that for centrifugal
compressors, that is overhaul on-condition with a 7 year TBO target.
(a) Non-Intrusive
Vibration: Periodic vibration monitoring can give an indication of bearing

condition and flow stability through the pump.
Performance: Performance assessment gives an indication of the condition

of hub and eye seals, and flow elements.

1.9.6 Gas Engines
Gas Engines are provided with standard alarm and trip systems which pro-
vide equipment protection on several key parameters: -
- overspeed
- high cooling water temperature.
The maintenance philosophy for a gas engine should be to overhaul on-

condition. Target TBOs will depend on duty and fuel quality, but overall
operating cost optimisation will always dictate precise intervals for any par-
ticular location.
(a) Non-Intrusive
Lube Oil Analysis: In addition to viscosity and flash point checks, TAN (Total
Acid Number) tests should to be carried out due to the inevitable contami-
nation of the oil system with combustion products. More than any other
analysis parameter, TAN will dictate the lube oil change interval. Once the
lube oil change intervals have been optimised for a given gas engine ap-
plication, only infrequent checks to confirm the service interval can usually
be justified.
Water Jacket Temperature: Trending of the jacket water temperature gives a

useful indication of the condition of the cooling water system and also the
ignition timing setpoint.
Cylinder Performance Analysis: Cylinder performance can be monitored

using proprietary engine analysers. Trending of this data provides a good
assessment of engine condition.
1.9.7 Reciprocating Compressors
Reciprocating compressors are provided with a trip system which provides

equipment protection for the following parameters:

- cylinder lubricator low pressure or level
Running Equipment
- cooling water high temperature

- high vibration.
The large number of reciprocating and rotating parts on a reciprocating

compressor, gives rise to difficulties in achieving the availability figures
commonly associated with centrifugal compressors. Efforts should always
be made to spare reciprocating equipment during the project definition stage
to ensure adequate overall system availability. In newer designs better av-
ailabilities are achievable with reciprocating compressors, with at least one
manufacturer aiming to produce a `30,000 hour' machine. An on-condition
maintenance philosophy should be specified with limited scheduled main-
tenance workscopes to check rider ring and cross-head bearing wear.
(a) Non-Intrusive
Valve Cover Temperature: Monitoring of valve cover temperatures can give

warning of imminent valve failure and can pin-point the actual faulty valve.
This saves considerable time when carrying out corrective maintenance
activities.
Cylinder Performance: Regular acquisition of PT chart and the development

of the derived PV diagram allows assessment of valve performance.
Distance Piece Pressure: Monitoring of the distance piece pressure provides

an indication of the condition of the pressure packing.
Rod Runout: Runout monitoring provides indication of rider ring and cross-
head bearing wear.
1.9.8 Reciprocating Pumps
Reciprocating pumps are provided with a trip system which provides equip-
ment protection for the following parameters:
- lube oil low pressure (if pressure-fed bearings)
- high vibration (on some machines).
The comments regarding availabilities of reciprocating compressors can

also be applied to reciprocating pumps.

Equipment should always be spared in order to achieve satisfactory system

availabilities. An on-condition maintenance philosophy should be applied
with limited scheduled interventions to check crosshead wear.
(a) Non-Intrusive.
Performance: Performance checks will provide the best indication of the

condition of the suction and discharge valves.
Packing Leakage: The condition of the packing can best be monitored by the
amount of leakage.
1.9.9 Gearboxes
Depending on the application, a gearbox may include a range of protection

systems, varying from no specific protection (e.g. a gas turbine accessory
gearbox) to a full -protection arrangement. The requirements for any paticular
gearbox can only be decided on a case by case basis, taking into account
normal packager arrangements, operating experience, and consequences of
failure.
- vibration
- bearing temperature
- thrust position.
The lube oil system protection is normally catered for elsewhere in the
package.
Casing mounted accelerometers are sometimes used for gearbox vibration

protection. This can be considered the preferred choice when effective fil-
tering is included. The monitored frequency range for protection purposes
must be limited to the first and second order of shaft operating speeds, to
eliminate any spurious inputs from other frequencies.
The maintenance philosophy for a gearbox should be to overhaul on-condi-

tion with a target TBO of 100,00 operating hours. Attention should be paid to
gearbox condition so that overhauls can be carried out on an opportunity
basis when associated equipment is undergoing major overhaul or inspection.

Running Equipment
(a) Non-Intrusive.
Vibration: Trending of bearing radial vibration levels can give background on

bearing condition.
Trending of vibration levels at gear mesh frequencies can give an indication

of tooth condition and meshing.
Bearing Temperature: Monitoring radial and thrust bearing temperatures

can give warning of problems with the lube oil supply or bearing over-
load.
Alignment: The same comments made under gas turbines apply equally well
to gearboxes.
Lube Oil Analysis: Magnetic chip detectors mounted in the gearbox lube oil
drain can give an early indication of any tooth distress and provide infor-
mation on abnormal wear rates.
(b) Intrusive:
Tooth Contact Checks: Tooth contact checks need usually only be made on
site during initial installation. However if gearbox problems are experienced
in service, changes in tooth contact over extended time intervals can be a
useful condition monitoring technique.

1.10 TESTING OF RUNNING EQUIPMENT
1.10.1 General
Testing is intended to reveal the fitness of the equipment for the purpose, to
discover any defects in design, manufacture or assembly, and most impor-
tant, that the entire package will work as an entity. Testing cannot be left until
the equipment has been delivered to the Job Site; it must be completed prior
to delivery of the equipment package.
It is not intended herein to cover the scope of testing to meet statutory or

regulatory requirements, e.g. metallurgical or hydrostatic tests, but to cover
the specific tests on finished or semi-finished equipment.
At the specification stage all testing requirements must be clearly specified
and the Vendors must include testing in their bid prices and deliveries. To
add testing requirements as a post-contract negotiation is to invite refusal or
disincentive prices and delivery penalties. Testing is costly; the need for it is
revealed by declining standards of workmanship, demanding greater atten-
tion by the purchaser to inspection and rejection of inferior work.
Many projects have been seriously delayed and have not functioned satis-
factorily due to faults which would have been revealed by sufficient works
testing. The object of this Section is to provide the reader with sufficient
information to fultil a satisfactory testing programme.
1.10.2 Prime Movers
1.10.2.1 Electric Motors
These shall be tested in accordance with DEP 33.66.05.31-Gen., for electrical

and mechanical soundness. During the bid analysis stage, verify that the
Vendor has the electric power capability to run a large, HV motor. The works
power supply frequency may preclude such a test if the motor is designed for
a different frequency.
These shall be tested in accordance with DEP 31.29.70.31-Gen., API Std 616,
and ASME PTC-22. Essentially the rotors are balanced, and the gas turbine is
run in the works against a dynamometer. The performance is checked, and
the governing system calibrated, It is essential to conduct this test with the

Running Equipment
contract engine control system. During this test, a full set of pressure, tem-
perature and vibration spectra data is taken as a ‘base-line' for future on-line
condition monitoring. For an aero-derivative gas turbine, the works test
should be performed using a shop hot gas generator as running starts and
hours of the contract hot gas generator on the test stand will be included in
the Vendor's guarantee period. The contract hot gas generator will have
been fully tested, as above, in the Manufacturer's works.
If the contract fuel has inert constituents in excess of 25%, or a lower heating
value (LHV) less than 22 MJ/m3 (st), the Manufacturer shall conduct perform-
ance trials in the works (before commencing manufacture) with his engine
burning a gas mixture similar to the contract fuel gas to demonstrate satisfac-
tory starting, load acceptance and load rejection. The test shall include a run
at full load of not less than four hours, after which the hot gas path shall be
inspected. An identical test shall be performed for crude oil or residual oil
firing.
1.10.2.3 Gas and Diesel IC Engines
These shall be tested in accordance with API Std 618 or API Std 7B-11C. Tests
shall include starting, stopping, and fuel consumption against a shop dynamom-
eter. As for gas turbines, if the contract fuel is ‘unusual', the test shall be done
using a sample of the proposed fuel. Base-line data such as firing pressure cur-
ves, governor valve calibration and vibration spectra shall be recorded during
the test.
1.10.3 Compressors
Tests are performed in accordance with DEP 31.29.40.10-Gen., API Std 614
and API Std 617. These will include balance checks on the rotor which must
be done in stages as the impellers are added. Final balancing is done in two
planes only. If the rotordynamic stability analysis gave a logarithmic decre-
ment less than 2.5, the final balance of the rotor should be done at up to 110%
of running speed in a vacuum balancing pit. In its casing, the rotor will be run
up to 110% of maximum continuous speed as part of the mechanical run test.
During either run up or more usually during run down the first critical speed
should be established, If it proves to be impossible to determine the first
critical speed, a further attempt should be made after adding an unbalance
weight to the coupling so as to cause excitation at the first critical speed. It is
essential to establish the first critical speed.

Aero thermal performance tests should be performed in accordance with the

ASME Power Test Code 10. This Code prescribes three classes of test:
Class 1: test performed on the specified gas, at the speed, inlet pressure and
temperature for which the compressor is intended to operate.
Class 2: performance under site conditions are predicted from tests con-
ducted at different operating conditions and with other gases.
Class 3: as Class 2, but uses different computational methods.
Class 1 testing is generally recommended only for high pressure (discharge

pressure over 100 bar) compressors and compressors for offshore installa-
tions. For other applications a Class 2 or 3 test is normally adequate.
For each constant speed line on the Head-Capacity graph for which the
performance is to be verified, six points are required for complete curve
definition:
1) and 2) to bracket at the duty point

3) at or near the surge flow
4) and 5) equally spaced between (2) and (3)
6) at a flow of 15% greater than the design point flow.
The design point performance shall have tolerances on head, flow and power
in accordance with API Std 617 (Nov. 1979). Curve shape (surge margin)
guarantees should be sought from the Vendor, in line with 1.2.1.2.6
above.
The general rule is to perform a PTC-1D test on all centrifugal compressors

purchased. If a purchase order is for two or more identical compressors, then
only one need be subjected to a performance test, the remainder only having
mechanical running tests. Similarly, in the case of a repeat order for a com-
pressor identical to one already in operation, if that unit is operating satis-
factorily and the new one will operate on the same gas, a performance test is
not required.
The mechanical running tests should be in accordance with API Std. 617 as
modified by DEP 31.29.40.30-Gen. This test may be performed using a shop
driver or it may be performed as a Complete Unit (or `String') test using
contract driver, gear unit, oil console, control panel, etc. as appropriate.
Refer to 1.10.7 below.
When a spare rotor is purchased as part of a compressor order, it should

have a mechanical run test exactly as the initial fit rotor. It need not have a
performance test.

Running Equipment
Tests are performed in accordance with DEP 31.29.40.31-Gen. and API Std
618. Nominally the compressor is checked for bearing clearances, piston rod
runout, and piston ring and packing ring clearances, It is run at no load
(suction valves removed) for six hours. Vibration levels are checked against
ISO 2372-1974. Pistons are pulled at the conclusion of the run and the con-
dition of the cylinder liner, piston rings, packings, cross-head and-bearings is
examined. As reciprocating compressor failures are usually time-dependent,
there is no purpose in performing a running test at full gas pressure.
1.10.3.3 Rotary Compressors
The testing requirements are as for reciprocating compressors, except that

screw compressors should be run on gas -up to design discharge pressure to
check noise levels.
1.10.4 Pumps
The proper testing of pumps and their ancillaries is intended to confirm the
overall fitness for purpose of the equipment purchased. It is recommended
that all such testing is carried out in the vendor's works prior to shipment. Site
testing as a means of reducing the delivery schedule should be avoided
wherever possible.
Testing requirements should be clearly specified by the requisitioning en-

gineer on the enquiry data/requisition sheets, and clarified in the requisition
specification, if appropriate. The relevant API standard and DEP specification
requirements for testing should be applied in conjunction with the following
guidelines for the various pump types.
The project electrical and instrument engineers should be consulted to en-

sure that all additional testing requirements for electrical and instrumenta-
tion equipment are properly identified and to obtain specific input to the test
procedures.
Note: For small pump units less than 250 kW, it is considered acceptable for a
third-party inspector to witness the tests. However, for larger units, and for smaller
non-standard units, it is recommended that the responsible project/requisitioning
engineer, or an appointed representative (design contractor) engineer, should also be
present during all tests not indicated as ‘by inspector'.

In addition to the range of standard inspection and testing requirements

covered by the appropriate API and DEP, the following minimum programme
of witnessed tests should be specified:
- Witnessed hydrotest all pumps (by inspector).

- Witnessed performance/mechanical run/function tests of the first of a se-
ries of identical pumps.
- Witnessed mechanical run/function test of balance of a series of identical
pumps (by inspector).
- Witnessed NPSH test of the first of a series of identical pumps, where the
margin between NPSHA and NPSHR is less than 2m.
Note: In addition to the recording of all condition monitoring readouts the requisi-
tioning engineer should ensure that the seal leakage rate is checked/monitored.
1.10.5 Gears and Speed Changers
These tests shall be in accordance with API Std 613 or API Std 677.
1.10.6 Couplings
These tests shall be in accordance with API Std 671.

Running Equipment
1.10.7 Complete Unit, or ‘String’ Test
This test shall comprise the connecting together of all equipment in the
driving train, i.e. driver, gears and driven equipment on the contract base-
plate. All contract auxiliary equipment, including control panels, lubricating
and seal oil systems, but excluding the contract air filter and exhaust silencer
on a gas turbine, shall be included in the test. The entire starting, stopping,
load acceptance and rejection tests shall be performed, and not less than
three malfunction shut-downs shall be tested, one of which shall be over-
speed.
Provided that a centrifugal compressor has been tested as in 1.10.3.1 above,

the complete unit test may be performed at no load, under vacuum. Alternator
sets shall be tested at full load, including load acceptance and rejection tests
in accordance with DEP 33.65.11.31. Reciprocating compressor sets may be
run at no load. The test shall include a non-stop six hour run.
During the test, pressures, temperatures, vibration spectra shall be recorded
using the contract instrumentation for base-line diagnostic data.
The test assembly shall be inspected for accessibility and alignment of

equipment, safety devices and maintainability. A fixed fire extinguishant
system shall be proved by triggering a fire detector head and the release of a
bottle of inert gas, e.g. CO2 (see also HSE in Vol. 1).
Any faults revealed by the above test shall be rectified before the package
leaves the shop, and any similar packages shall be identically treated. De-
pending upon the extent of faults and repairs, the right should be reserved to
test all identical packages. Any modifications resulting from the tests must be
fully recorded.
If the package is to be assembled into an offshore module, the test should be

done in the module as the scope of the test can be broadened to include all
module equipment. In any event, all functional tests must be repeated once
the module is assembled, and before loadout offshore.
A land-based package should be completely trial assembled before ship-

ping. Clearances, accessibility and completeness must be verified, and cor-
rected where inadequate, as for an offshore package.
The selection of Vendors should be restricted only to those willing and cap-
able of performing tests as described herein.

1.11 REFERENCES AND FURTHER READING
An extensive list of references and further reading for Running Equipment,

including relevant manufacturers' catalogues, is in preparation in SIPM-
EPD/62 and will be included in the next update of this Volume. This list will
also include up-to-date indices of published proceedings of the major insti-
tutions such as Turbo-machinery Symposia of Texas A&M University,
l.Mech.E. Fluid Machinery Symposia, BHRA, ASME and others.
Pending the completion of this list, users are recommended to consult SIPM-
EPD/62 for references and further reading regarding Running Equipment,
insofar as not already included in the preceding material.

Piping Systems
Chapter 2
PIPING SYSTEMS
CONTENTS
2.1 Piping Codes and Standards ..........................................................................................................................2-4
2.2 Scope of piping codes....................................................................................................................................2-4
2.2.1 B31.3 Chemical Plant and Petroleum Refinery Piping .............................................................................2-4

2.2.2 Exclusions to B31.3.....................................................................................................................................2-4
2.2.3 Packaged Units ...........................................................................................................................................2-5
2.2.4 Additional Design Documents.....................................................................................................................2-5
2.3 Stages of Development of a Project from Process and Instrument Diagrams to Requisitions......................2-5
2.3.1 Process and Instrument Diagrams (P and IDs) ..........................................................................................2-5

2.3.2 Arrangement Studies ..................................................................................................................................2-5
2.3.3 Models.........................................................................................................................................................2-6
2.3.4 Isometric Piping Drawings ..........................................................................................................................2-6
2.3.5 Detailed Layout Drawings ...........................................................................................................................2-7
2.3.6 Additional Design Considerations ...............................................................................................................2-7
2.3.7 Flexibility in Piping.......................................................................................................................................2-8
2.3.8 Stress Analysis............................................................................................................................................2-8
2.3.9 Material Take-oft and Requisitioning .........................................................................................................2-8
2.3.10 Computer Aided Design.............................................................................................................................2-8
2.4 Line sizing .......................................................................................................................................................2-9
2.4.1 General Considerations ...............................................................................................................................2-9

2.4.2 Size Limitations ..........................................................................................................................................2-10
2.4.3 Tentative Line Sizing..................................................................................................................................2-10
2.4.4 Final Line Sizing.........................................................................................................................................2-11
2.5 Specification of Piping Systems................................................................................................................2-11
2.5.1 Material Selection of Piping .....................................................................................................................2-11

2.5.2 Specification and Selection of Piping.......................................................................................................2-11
2.5.3 Piping Classes .........................................................................................................................................2-12
2.5.4 Material and Equipment Standards and Code (MESC) ...........................................................................2-12
2.5.5 Identification System for Piping Classes..................................................................................................2-13
2.5.6 Description of Piping Classes ..................................................................................................................2-13
Confidential-Property and Copyright: SIPM, 1991 2-1

2.6 Pressure Design of Piping ............................................................................................................................2-14
2.6.1 Design Limits..............................................................................................................................................2-14

2.6.2 Pressure Design of Straight Pipe...............................................................................................................2-14
2.6.3 Pressure Design of Branch Connections ...................................................................................................2-15
2.6.4 Pressure Design of Components ...............................................................................................................2-16
2.7 Particulars of Components........................................................................................................................2-16
2.7.1 Fittings......................................................................................................................................................2-16
2.7.2 Joints........................................................................................................................................................2-17
2.7.3 Valves ......................................................................................................................................................2-17
2.8 Piping Systems Operating in Sour Environments........................................................................................2-18
2.9 References....................................................................................................................................................2-18
2-2 Confidential-Property and Copyright: SIPM, 1991

Piping Systems
TABLES
Table 2-1 ANSI/ASME Code for Pressure Piping, B31 2-19

Table 2-2 ANSI/ASME and API Piping Standards 2-19
Table 2-3 Velocities for tentative line sizing (for guidance only) 2-19
Table 2-4 Nominal pipe size vs wall thickness 2-20
Table 2-5 Flow rates in m3/h for liquid and gas lines 2-22
Table 2-6 Allowable internal design pressure bar(g) for seamless
pipe, 1.0 mm corrosion allowance. Pipe APl 5L-B/ASTM
A106-B/ASTM A 333-6. 2-24
Table 2-7 Allowable internal design pressure bar(g) for seamless
pipe, 3.0 mm corrosion allowance. Pipe API 5L-B/ASTM
A 106-B/ASTM A 333-6. 2-25
Table 2-8 Allowable internal design pressure bar(g) for welded
pipe, 1.0 mm corrosion allowance. Pipe API 5L-B. 2-26
Table 2-9 Allowable internal design pressure bar(g) for welded
pipe, 3.0 mm corrosion allowance. Pipe API-5L-B. 2-27
Table 2-10 Allowable internal pressures for unreinforced 90
degree pipe to pipe connections for seamless carbon
steel 2-28
Table 2-11 Allowable internal pressures for unreinforced 90
degree pipe to pipe connections for welded carbon
steel 2-30
Table 2-12 Rating ceiling values 2-31
FIGURES
Figure 2-1 Scope of ANSI B31.3 piping - diagram 2-32

Figure 2-2 Tentative sizing for liquid lines 2-33
Figure 2-3 Line sizing for steam lines within process plant areas 2-35
Figure 2.4 Pressure drop in steam lines not covered in Figure 2-3 2-37
Figure 2-5 Pressure drop in carbon steel water lines
(water at 20ºC) 2-38
Figure 2-6 Pressure/temperature rating to ANSI B16.5/B16.34
(carbon steel materials) 2-39

2. PIPING SYSTEMS
2.1 Piping Codes and Standards
The most commonly used piping codes and standards are those produced
under the sponsorship of the American National Standard Institute (ANSI)
and the American Society of Mechanical Engineers (ASME). The ANSI/ASME
Code for pressure piping, B31, consists of a number of individually published
sections (also called codes) and is shown in Table 2-1. Selection of material
to resist deterioration in service is not covered in the code, therefore con-
sultation with a corrosion and/or materials engineer for specialist advice is
strongly recommended.
National Standards on dimension, rating and design for the large number of
piping components have been developed by the industry. The standards are
incorporated in the Code, which greatly reduces the design cost of piping.
The most important standards are listed in Table 2-2.
2.2 Scope of piping codes
2.2.1 B31.3 Chemical Plant and Petroleum Refinery Piping
This Code prescribes the requirements for the materials, design, fabrication,
assembly, erection, examination, inspection and testing of piping systems
subject to pressure or vacuum. It applies to piping systems for all fluids,
including raw, intermediate and finished chemicals, oil and other petroleum
products, gas, steam, air, water, refrigerants and fluidised solids.
This Code covers all piping within the property limits of facilities engaged in
the processing or handling of chemical, petroleum, or related products.
A diagrammatic illustration of the scope of this Code is shown in Figure

2-1.
2.2.2 Exclusions to B31.3
a) Piping systems designed for internal pressure at or above zero but less
than 15 psi (103 kPa), provided the fluid handled is non-flammable, non-
toxic, and not damaging to human tissue as defined in the Code and its
design temperature is from –20ºF (-29ºC) through 366ºF (186ºC).

Piping Systems
b) Power boilers in accordance with ASME Boiler and Pressure Vessel Code
Section I and boiler external piping which is required to conform to
831.1.
c) Tubes, tube headers, crossovers, and manifolds of fired heaters, which
are internal to the heater enclosure.
d) Pressure vessels, heat exchangers, pumps, compressors, and other fluid
handling or processing equipment, including internal piping and connec-
tions for external piping.
e) Piping located on company property which has been set aside for pipe-
lines conforming to B31.4, B31.8 or B31.11 or applicable governmental
regulations.
f) Plumbing, sanitary sewers, and storm sewers.
g) Fire protection systems constructed in compliance with insurance under-
writers' or other recognised fire protection engineering standards.
2.2.3 Packaged Units
Piping which interconnects individual pieces or stages of equipment within a

packaged equipment assembly shall be in accordance with B31.3, except that
packaged refrigeration piping may conform to either B31.3 or B31.5.
2.2.4 Additional Design Documents
Relevant Shell standards (DEPs) and further information on recommended

practices are listed under ‘References' at the end of this Chapter.
2.3 Stages of Development of a Project from Process and Instrument Diagrams to Requisitions
2.3.1 Process and Instrument Diagrams (P and IDs)
P and IDs give all the essential details on the piping such as: line size, line
number, piping class. Lines are numbered sequentially for each service.
Line sizes are determined to suit the flow rate and condition of the fluid (see
2.4). The piping class is selected from tables which have been assembled for
the project (see 2.5).
2.3.2 Arrangement Studies
The intention of arrangement studies is to develop a reasonably accurate

framework for the final equipment and piping arrangement with the least

expenditure of engineering manhours. The following basic information should

appear on arrangement studies:
a) The location and orientation of all static and running equipment nozzles,
including sight and light glasses, manholes and instrument connections.
b) The location of major valves, control sets, orifice assemblies, piping spe-
Cialities and equipment and all instruments.
c) Main or subsidiary pipe tracks, instrument and cable ducts.
d) Single runs of specialist piping such as alloy and long delivery materials.
The following related problems have to be considered during the piping

arrangement studies:
a) Process requirements.
b) Transmission of excessive loads and transfer of vibration with respect to
connected equipment.
c) Excessive stress in piping systems due to thermal expansion and con-
traction.
d) Operational accessibility.
e) Maintenance and replacement.
f) Economy.
Arrangement studies are important for an adequate plant design. Difficulties

may be recognised at an early stage and corrected, avoiding rework of the
detailed engineering design or modifications to installed plant.
2.3.3 Models
Piping models are of great assistance in the design layout stage where piping
is of a very high density and are very useful in respect of access, maintenance
and training operators. Their scale is normally 1 in 33.3.
As with arrangements studies, models may highlight potential operation and

maintenance problems.
For further information reference should be made to DEP 30.10.05.11 ‘Model

construction for processing units'.
2.3.4 Isometric Piping Drawings
The isometric drawings are produced from the piping arrangement (or direct
from the model) and are used for the purpose of fabrication. The drawings

Piping Systems
identify all in-Iine components and jointing materials required for fabrication
and assembly.
If required the isometric arrangement is spooled (sectioned) into size to meet

transportation/shipping limitations. These are maximum sizes which may be
reduced as dictated by any process (protective coating, etc.) or installation
requirements.
2.3.5 Detailed Layout Drawings
Once the arrangement studies have been completed the detailed layout
drawings are prepared. These drawings show the final relevant details of all
equipment, vessels, piping, supports, etc. and to a reasonably large scale,
say 1 in 50, and are mainly intended to provide information for the proper
installation of the piping systems.
2.3.6 Additional Design Considerations
A number of additional recommendations and statements that are to be

considered during the detailed design are listed below (Ref. 1).
a) Piping with instrument connections shall be laid so as to make these

connections easily accessible.
b) Test connections shall be flanged and closed by means of a blind flange
and suitable gasket.
c) Fuel connections at furnaces should be made by rigid pipe. Flexible hoses
shall not be used.
d) The inlet line to a relief valve shall be self-draining into the process
equipment.
e) The outlet line from a relief valve shall be self-draining into a vessel
connected to the flare header.
f) Domestic water and drinking water shall in principle not be used for
purposes other than the name indicates. If connection to process equip-
ment cannot be avoided then a reliable back-flow prevention such as a
break-tank shall be used.
g) For carbon steel and ferritic alloy steel piping the minimum corrosion
allowance shall be not less than 1 mm.
h) Threaded joints shall be used only when the piping class allows it.
Threaded joints shall not be used where corrosion or erosion may occur
and where flammable liquid/gases or toxic products are transported.
I) The use of socket weld fittings is never permitted in process piping. For
utility piping they may be used.
j) Permanent blanks and spare type blinds shall have the pressure/temper-
ature rating of the connecting piping, except for temporary spade-type
blinds which are used for pressure testing only.

2.3.7 Flexibility in Piping
To avoid undue stresses in piping, runs shall be provided with inherent

flexibility and properly designed and located supports, guides and anchors.
Flexibility may be provided by the inherent geometry of the pipe run (change
of direction and elevation, bends, loops and off-sets) or by decreased stiff-
ness by incorporating expansion joints, which are less attractive but often
needed due to limited space or to cope with settling of equipment.
2.3.8 Stress Analysis
Stress analysis should be performed on all critical lines, which are listed
below:
a) Hot lines at 315ºC (600ºF) and above.
b) Cold lines below 0ºC.
c) 150 mm (6 in) nominal pipe size and above
d) Connections to strain sensitive equipment.
e) All piping associated with rotating machinery.
f) All piping connecting wells with platforms.
g) All piping associated with platform bridges.
For lines in severe cyclic service, fatigue analysis of line sections or com-
ponents may be required.
2.3.9 Material Take-oft and Requisitioning
During the early stages of the design, a preliminary bulk piping material
take-off can made, based on flow schemes and equipment layout, for cost
estimating and tendering. As the work progresses (production of isometrics
with related material lists), the material requirements become more accu-
rate and the material requisitions are updated as required.
Bulk piping material should be ordered from accepted vendors as listed by

Group Materials of SIPM in their Procurement Coordination Publications.
2.3.10 Computer Aided Design
Computer aided design (CAD) has evolved from a simple 2D drafting system
for schematics to a powerful engineering design tool for complex installations

Piping Systems
and is more and more used by major engineering contractors. A fully integr-
ated 3D model is developed with computer graphics and when finalised,
isometrics, arrangement drawings and material listings are automatically
generated by the system.
I
The use of CAD benefits the overall quality of the design, allows a high level
of design control and provides a pay-off in the fabrication and construction
phase.
2.4 Line sizing
2.4.1 General Considerations
Proper line sizing is of great economic importance because the cost of pro-
cess piping is typically 10 to 20% of total plant investment. Line size is
dependent on required inlet and outlet conditions, namely the allowable
pressure drop, quantity of liquid flowing and the physical properties of the
liquid. There may also be a maximum allowable velocity to be taken into
account, for reasons of corrosion and erosion or excess noise and vibra-
tion.
For larger systems it may be advantageous to include also economics when

sizing lines, by considering the combined annualised costs for piping, run-
ning equipment and power consumption.
Tables and graphics used for line sizing are generally based on new pipe.
Usually, corrosion, erosion and solids deposition may foul-up the line in time
and would increase the hydraulic roughness and friction factor. Also one
should realise that e.g. a 5% decrease of the internal diameter of the pipe will
increase the pressure drop by more than 25% for the same flow.
Pump suction pressures should not be less than the vapour pressure of the
liquid at temperature. Two-phase flows need special attention in line sizing
and layout. Pressure drop may be unpredictable.
Line pressure drop for controlled flow should be consistent with the selected
control valve (specified by instrument engineering).

2.4.2 Size Limitations
Outside diameter and wall thickness (schedule number) of welded and seam-
less steel pipe is standardised and shown in Table 2-4 which should be used
with the following limitations:
a) Nominal pipe size less than DN15 shall not be used.
b) Nominal pipe size in pipe tracks and pipe bridges shall not be less than
DN40.
c) Nominal pipe sizes DN30, DN55, DN95, DN125 and DN550 should not be
used.
d) Schedule number 60S and 140S should not be used.
Note:
Nominal means that a numerical identification of dimension or rating is used as

designation, not as an exact measurement.
DN is Diameter Nominal (refers to mm).

NPS is Nominal Pipe Size (refers to inch).
PN is Pressure Number (refers to bar).
NP is Nominal Pressure (refers to psi).
2.4.3 Tentative Line Sizing
For tentative line sizing, velocity ranges for various services are given in
Table 2-3.
Approximate line sizes for liquid lines can be obtained from Figure 2-2. This
gives also the line sizes of the discharge and suction of liquid lines for a given
flow rate. Velocities may also be obtained from this chart.
Table 2-5 enables a provisional line size to be found for either liquid or gas
lines. For example, if a flow of 1200 m3/h is required at a velocity of 4.50 m/s,
then a DN 300 (NPS 12) pipe which has a wall thickness of Schedule 30 is
needed.
Figures 2-3 and 2-4 provide charts for the sizing of steam lines. Figure 2-5 is a
chart for determining pressure drop in carbon steel water lines.

Piping Systems
2.4.4 Final Line Sizing
Line sizes obtained by the above methods are satisfactory for the process
flow schemes stage, but when the engineering design stage has been
reached it is necessary to finalise the line size for the engineering flow
scheme and to check the allowable pressure drop or required pumping
head.
Pressure drop calculations may be carried out either by employing the gen-
eral equation for pressure drop (Ref. 1), or with the aid of a proven computer
program.
2.5 Specification of Piping Systems
2.5.1 Material Selection of Piping
Selection of the material of construction is based on the process conditions

such as the nature and composition of the medium, pressure and tempera-
ture, and the environment of the installation.
The selection will have to be made by a materials/corrosion engineer in

close consultation with the process designer or process engineer. For a
number of broadly specified services, suitable materials have been listed in
the Index to Service Experience (Ref. 2). However, for final definition of the
material, a materials/corrosion engineer should be consulted. See also
chapter 7 ‘Corrosion Engineering'.
2.5.2 Specification and Selection of Piping
The specification of pipe and components (fittings, flanges, valves, gaskets,

etc.) and the selection of type of component (e.g. globe valve, gate valve,
etc.), take into account the material selection, the operational requirement,
the design requirements resulting from pressure and temperature and the
adopted design principles (as governed by the applicable codes and stan-
dards). The integrity of individual components when combined in a piping
system needs to be checked to ensure that not only the individual compon-
enents are fit to purpose but also the piping system as a whole.

2.5.3 Piping Classes
Because of the wide range of suitable materials and the extensive variety
in available components, general (standard) piping specifications are devel-
oped, engineered and maintained by SIPM, with input from materials/cor-
rosion engineering, mechanical engineering, procurement and user's ex-
perience and recommendations in Operating Companies. During this pro-
cess due account is given to standardisation by limiting the variety in se-
lected materials and type of components, while observing the availability and
economics of the selection. The piping specifications are grouped in piping
classes (Refs. 3 and 4), covering the services generally dealt with by Oper-
ating Companies. Each piping class forms an assembly of fully specified
piping components that under certain conditions and within defined limits
can be used together. Currently some 40,000 individually coded components
(defining material, type, size and rating) are included in the classes. For a
project the applicable piping classes shall be selected from this set of gen-
eral piping classes.
Every effort should be made to reduce the number of pipe classes for a
project to a minimum. This lessens the risk of inadvertently using the wrong
material, e.g. ordinary grade of carbon steel where fine grain has been
specified. Introducing a class which has not been used previously at a lo-
cation will involve the stocking of additional spare pipe, valves, jointing
materials, fittings, etc.
Deviation from the general piping classes or the introduction of new piping
classes should follow the procedures as given in Ref. 5.
2.5.4 Material and Equipment Standards and Code (MESC)
The material listed in the piping classes is coded, referring to the detailed
specifications (buying descriptions) in the MESC (Ref. 8). The complete
MESC catalogue is also available on CD-ROM.
For bulk piping materials the following MESC groups are applicable:
Group 74: Pipe

Group 76: Fittings, Flanges, Clamp Connectors
Group 77: Valves (formerly covered in Group 75)
Group 81: Bolting Materials
Group 85: Gaskets and Packing Materials

Piping Systems
2.5.5 Identification System for Piping Classes
In the Index to Service Experience (Ref. 2), selected piping classes are iden-
tified by means of a four-digit figure (e.g. 1010) of which the first digit points to
the general group of selected materials, e.g.
1. Carbon steel 5. Copper and copper alloys

2. Low and intermediate alloy steel 6. Nickel and nickel base alloys
3. Stainless steel 7. Non-metallic materials
4. Aluminium and alum. base alloys 8. Carbon steel with lining
The three remaining digits have been more or less randomly chosen, al-
though they sometimes, in combination with the material-related digit, are
associated with specific services (e.g. 1260 for steam services).
Preceding this four-digit figure is placed a one- or two-digit prefix that indi-
cates the ANS class rating, e.g.
1 for ANS class 150 9 for ANS class 900

3 for ANS class 300 15 for ANS class 1500
6 for ANS class 600 25 for ANS class 2500.
In summary a piping class is defined in two steps:
step 1 Define service-related figure by means of the Index to Service Ex-

perience, e.g. 1260 for steam.
step 2 Define pressure/temperature-related prefix, e.g. 3 for ANS 300.
The full definition of the pipe class is then 31260.
2.5.6 Description of Piping Classes
Piping classes provide, in a structured way, the following information:
a) Design limits, indicating the relation between design temperature and

allowable pressure.
b) Notes, giving additional design requirements and possible options.
c) Branch table, giving construction details for 90 degree branch connec-

tions.
d) Table of schedules, giving the selected pipe wall thickness (schedule) for
each nominal pipe diameter.

e) Short material descriptions for piping components, including gasket and

valve trim material.
f) Ranges of components used, coded in accordance with and fully described

in the MESC.
g) Ranges of bolt sets for various types of flanged joints.
h) Assemblies for vent, drain and instrument connections.
2.6 Pressure Design of Piping
2.6.1 Design Limits
The design limits for the allowable internal pressure related to the design
temperature are usually set by the pressure-temperature ratings of the
flanges (see 2.6.4).
Occasionally design limits are dictated by other considerations such as pres-

sure limitations due to bolting material requirements in a sour environment
or the temperature limitations on soft seat material in ball valves.
The interconnecting pipe shall be at least compatible with the flange rating,
however, for large-bore piping, the wall thickness of the pipe may be the
limiting factor. For small-bore piping (DN40 or less), generally a heavier
schedule than required for internal pressure is used to increase the struc-
tural stability.
Unreinforced branch connections welded to the run pipe may considerably

reduce the allowable pressure (see 2.6.3).
2.6.2 Pressure Design of Straight Pipe
For carbon steel pipe, both seamless and welded, tables 2-6 to 2-9 give the
allowable internal design pressure in bar(g). They are calculated in accord-
ance with ANSI B31.3 with the factor values S, E and corrosion allowance as
shown in the Tables, using the formula below.
2tSE
p = 
D - 2tY
(1)

Piping Systems
where:
p = allowable internal design pressure, bar(g)

D = outside diameter of pipe, mm
S = allowable stress, bar
E = longitudinal weld joint factor
t = design thickness for internal design pressure, mm
Y = coefficient of value 0.4 for ferritic and austenitic steels within
the temperature range considered.
The wall thickness available to resist internal design pressure is calculated

from the equation:
t = tm - ca (2)
where:
tm = nominal wall thickness (tn) minus manufacturer's tolerance

ca = sum of mechanical, corrosion and erosion allowances
Note:
Equation (1) is valid only if t is less than D/6.
For the sake of simplicity, mechanical and erosion allowances are ignored
and only the corrosion allowance is considered in the determination of t. The
value of tm is found by subtracting the manufacturer's tolerance, as specified
in the relevant ASTM specification, from the nominal wall thickness (tn) as
specified in ANSI B36.1OM and B36.19M.
2.6.3 Pressure Design of Branch Connections
A branch connection welded to a run pipe weakens this pipe by the opening
required for the branch. Unless the branch fitting used has sufficient inherent
reinforcement (i.e. oversized wall thickness), a reinforcement ring is required
to restore the original strength of the run pipe.
For a 90-degree branch connection, the material taken out by the opening has
to be replaced (by the same amount) as reinforcement, within a limited zone
around the connection. Lateral branches require up to 100% more reinforce-
ment and are not recommended. When the branch size approaches the run
size, standard butt welding fittings (tees, reducing tees) should be used.

Maximum allowable internal pressures for unreinforced 90-degree pipe to

pipe connections are listed in Tables 2-10 and 2-11. Pressures have been
calculated for the usual header and branch size (schedule) and corrosion
allowance combinations for carbon steel. These tables can be used as an
aid to define the allowable internal pressure related to the service temperature
when such branch connections are used in the piping system.
2.6.4 Pressure Design of Components
Standardised pressure components, such as flanges, fittings and valves, with

established pressure-temperature ratings may be used, provided the design
pressure and temperature are within these ratings. Use of components with-
out pressure-temperature ratings or beyond the limits of established values
must be designed in accordance with the Code requirements.
The standard flanges, fittings and valves have been rated according to ANSI
B16.5 (see 2.5.5). The ANS Class ratings are nominal service pressure rat-
ings. For a range of usually applied materials for forging, casting and plate,
pressure-temperature ratings have been established and listed in ANSI
B16.5. The rating ceiling values derived from the ANSI tables are shown in
Table 2-12.
For carbon steel, pressure-temperature ratings for ANSI pipe flanges,

flanged fittings and valves are shown in Fig. 2-6.
2.7 Particulars of Components
2.7.1 Fittings
Bends, elbows, tees, reducers, etc. may be cast, forged or hot or cold formed
from pipe. Cold bending of pipe is normally restricted to 5 times the nominal
pipe diameter for the centre line radius of the bend. End connections of
reducers may be concentric or eccentric. The latter permit complete drain-
age of branched horizontal piping systems or permit liquid flow in horizontal
pipe to sweep the pipe free of gas or vapour. Tees instead of branch con-
nections (also when the ratio of branch size to run size is small) may be
preferred in view of the better controllability of the welding.

Piping Systems
2.7.2 Joints
The most widely used joint in piping systems is the butt-welding joint. Careful
alignment and skilled labour is required. Depending on inspection level, joint
strength may be taken equal to the original pipe. Socket weld joints are not
favoured because the crevice between pipe and socket may promote cor-
rosion.
Threaded pipe joints are normally restricted to 2 inch and below and only
acceptable for mild, non-process services. Preassembly of piping using
threaded connection is severely limited.
For pipe connections to equipment, valves, etc., the flanged joint is the most
used. Flanges are attached to pipe by various types of joints, such as
screwed, slip-on welding, socket welding, lap joint or buttwelding to a weld-
ing neck of which the latter is usually applied. A number of flange facings are
available of which the raised face is normally used. The usual flange face
finish is standardised on an average roughness of Ra3.2-6.3 µm (‘smooth
finish’), suitable for flat gaskets and spiral wound gaskets.
To obtain compact pipework for the higher pressure ratings, clamp-type

connectors are used. Disassembly of these joints is only possible in the axial
direction of the pipe.
2.7.3 Valves
Valves are the most important piping accessories and offered in a wide
variety of types and designs. For economy and interchangeability the number
of different varieties of valves should be limited.
Valve bodies may be cast, forged or machined from bar stock (smaller sizes).
Their design should prevent distortion or misalignment of the seating sur-
faces, due to pressure, temperature changes and strain from connecting
piping. The sealing surfaces should be of such materials and design that the
valve will keep its shut-off characteristics over a reasonable service period.
Various alloys are available for valve parts such as seats, discs and stems
(the valve trim), which must retain smooth finish for successful operation.
Seat material should be resistant to corrosion by the fluid handled, resistant

to erosion by suspended solids in the fluid and prevent galling (seizure at
point of contact) by difference in material or hardness or both. Often, seat
surfaces are treated or plated to obtain the required properties. All valve trim

materials have coefficients of thermal expansion which exceed those of cast

or forged carbon steel by 24 to 45% and tend to cause distortion of seats and
discs.
Valves for processing plants are generally specified in accordance to British

Standards, such as:
BS 1414 Steel Wedge Gate Valves
BS 5351 Steel Ball Valves
BS 5352 Steel Wedge Gate, Globe and Check Valves (DN50 and smaller).
Valve types or ranges not covered by the British Standards are specified in
accordance with API, ANSI or other appropriate standards (see also table
2-2). For further information on valves with respect to selection, application,
failure modes and operational aspects see Ref. 1.
2.8 Piping Systems Operating in Sour Environments
For this subject reference is made to the chapter ‘Corrosion Engineering'

elsewhere in this volume.
2.9 References
1. Piping - General Requirements DEP 31.38.01.11

2. Piping Classes Part 1 - Basis of Design DEP 31.38.01.10
3. Piping Classes Part 2 - Typical MF DEP 31.38.01.12
4. Piping Classes Part 3 - Exploration and Production DEP 31.38.01.15
5. Specification for Piping Systems DEP 31.38.01.21
6. Compilation of Bill of Material for Piping Isometrics DEP 31.38.01.31
8. MESC - Material Groups 74-75-76-77-81-85 Material and Equipment Stan-
dards and Code - Sheets and Specifications for Bulk Piping Materials

Piping Systems


Piping Systems


Piping Systems


Piping Systems

Note: Calculations made for this table have the following factor values:
CA = 1.0 mm
E = 0.85
S = as per table

Piping Systems
Note: Calculations made for this table have the following values:
CA = 3.0 mm
E = 0.85
S = as per table


Piping Systems


Piping Systems


Piping Systems


Piping Systems


Piping Systems


Piping Systems

Chapter 3
ELECTRICAL ENGINEERING
CONTENTS
3.1 Introduction ..................................................................................................................................... 3-3

3.2 Operational Safety and Reliability................................................................................................... 3-3
3.3 Protection Against Explosion and Fire Hazards ............................................................................. 3-5
3.3.1 General ....................................................................................................................................... 3-5

3.3.2 Selection and Installation ............................................................................................................ 3-5
3.4 Electrical System Design ............................................................................................................. 3-7
3.4.1 Preface...................................................................................................................................... 3-7

3.4.2 DEP Clause 4.5.3 - Distribution Systems ................................................................................. 3-7
3.4.3 DEP Clause 4.2.2 - Load Assessment and Electricity Consumption........................................ 3-8
3.4.4 DEP Clause 4.3.3 - Deviations and Variations in Supply Wave Forms.................................... 3-8
3.4.5 DEP Clause 4.7.2 – Generators ............................................................................................... 3-8
3.4.6 DEP Clauses 4.7.7.2/3/4 - Variable Speed Drives/Synchronous Motors/Asynchronous Cage-
Induction Motors.................................................................................................................................... 3-8
3.4.7 DEP Clause 5.6 - Electric Drives .............................................................................................. 3-9
3.4.8 DEP Clause 5.6.2 - Variable Speed Drives .............................................................................. 3-9
3.4.9 DEP Clause 5.8/6.3 - Cables, Wires and Accessories/Cabling and Wiring............................ 3-10
3.4.10 DEP Clause 5.10/6.5 - Lighting Equipment/Lighting Installations .......................................... 3-10
3.4.11 DEP Clause 5- Power and Convenience Outlets ................................................................... 3-11
3.4.12 DEP Clause 5.10.5 - Portable Lamps and Tools .................................................................... 3-12
3.4.13 DEP Clause 6.6 - Earthing and Bonding ................................................................................ 3-12
3.4.14 Navigation Aids Obstruction Warning Lights .......................................................................... 3-12
3.5 Electrical System Studies and Calculations................................................................................ 3-15

3.6 Installation Practices ................................................................................................................... 3-16
3.6.1 Introduction ............................................................................................................................... 3-16

3.6.2 General ..................................................................................................................................... 3-16
3.6.3 Generator Room ....................................................................................................................... 3-17
3.6.4 Switchgear Rooms .................................................................................................................... 3-17

3.6.5 Transformer Enclosure.............................................................................................................. 3-18

3.6.6 Battery Room ............................................................................................................................ 3-19
3.6.7 Cables ....................................................................................................................................... 3-19
3.6.8 Lighting...................................................................................................................................... 3-22
3.6.9 Earthing ..................................................................................................................................... 3-23
3.7 Maintenance................................................................................................................................ 3-24
3.7.1 Design/Engineering/Construction Phase .................................................................................. 3-24

3.7.2 Operational Phase .................................................................................................................... 3-25
3.8 Electrical Safe Working Practices ................................................................................................ 3-27

3.9 Additional Information ................................................................................................................. 3-28
3.9.1 Selection and Marking of Explosion Proof Equipment........................................................... 3-28

3.9.2 Guideline for Illumination Levels ............................................................................................. 3-30
3.9.3 Protection of Equipment........................................................................................................... 3-31
3.10 Relevant DEPs, Standards and Other Publications.................................................................. 3-32
3.10.1 DEPs*) .................................................................................................................................... 3-32

3.10.2 Group Standard Forms (DEP 00.00.10.05-Gen.) .................................................................... 3-33
3.10.3 Group Standard Drawings (DEP 00.00.06.06-Gen.) .......................................................... 3-34
3.10.3 Group Standard Drawings (DEP 00.00.06.06-Gen.) (continued) ........................................ 3-35
3.10.4 Other Relevant Standards and Publications ........................................................................... 3-36
TABLES
Table 3.9-1 Explosion protection 3-28

Table 3.9-2 Type marking of equipment (IEC 79) 3-29
Table 3.9-3 Suggested illumination levels at different locations 3-30
Table 3.9-4 Enclosures-degree of protection 3-31

Electrical Engineering
3 ELECTRICAL ENGINEERING
3.1 Introduction
It is a characteristic of most EP projects that the size and function tend to

change during the design/engineering phase and also during the productive
period of an oil/gas field.
Together with the fact that many production locations are remote from man-
ufacturers, often in harsh environments and subject to varying local regu-
lations, this emphasises the main requirements of the electrical system for a
cost-effective operation, i.e., reliability, maintainability and flexibility.
The foregoing illustrates that, generally, rigid rules cannot be given. Accord-
ingly this Chapter should be used as a guideline to Shell practice for the main
electrical engineering aspects of an EP project, leaving the details to the
local project engineering departments. It is divided hierarchically into nine
sections, dealing respectively with the basic principles of operational safety
and reliability, protection against explosion and fire hazards, electrical sys-
tem design, requirements for studies, installation practices, maintenance,
electrical safe working practices, additional information and relevant DEPs
and standards.
Frequent reference is made to DEPs and other publications and a list related
to electrical engineering is given in Section 3.10.
3.2 Operational Safety and Reliability
The design of the electrical installation shall be based on the provision of a

safe and reliable supply of electricity at all times. Safe conditions shall be
ensured under all operating conditions, including start-up and shut-down of
plant and equipment, and throughout the intervening shut-down periods.
Special attention shall be paid to provisional and temporary installations

required for the erection of permanent installations to ensure compliance
with basic rules for good working practice and safety and to cover against the
increased hazards which are present in temporary installations.
The design of electrical systems and equipment shall ensure that all oper-
ating and maintenance activities can be performed safely and conveniently

and shall permit periods of continuous operating periods of at least 4 years.

To fulfil the requirements above, provisions may be required for alternative
supply sources and supply routes, spare/standby capacity, automatic res-
tarting schemes, etc., details of which are given in Section 3.4.
Furthermore, standardisation of materials and equipment shall be aimed for
as far as compatible with rational design. Equipment, which will become
obsolete in the near future shall not be installed.

3.3 Protection Against Explosion and Fire Hazards
3.3.1 General
To permit the proper selection of electrical apparatus for use in areas where
flammable gas or vapour risks may arise, an area classification drawing(s)
shall be prepared, on the basis of the Institute of Petroleum Model Code of
Safe Practice - Part 15.
For an explosion to occur an explosive atmosphere and a means of ignition

must co-exist.
When electrical equipment is to be installed in areas where dangerous con-

centrations and quantities of flammable gases, vapours, mists, ignitable
fibres or dusts may be present in the atmosphere, protective measures shall
be applied to reduce the likelihood of explosion due to ignition by arcs,
sparks or hot surfaces produced either in normal operation or under spec-
ified fault conditions.
Electrical equipment should therefore as far as possible be located in non-

hazardous areas and where this is not practical, suitable equipment shall be
installed in the least onerous hazardous area. Control rooms and switch-
houses containing electrical equipment items, many of them normally spark-
ing, should therefore be situated in non-hazardous areas.
Note: Area classification shall be decided in a meeting between appropriately qual-

ified staff representing the following disciplines:
- Process technology - Engineering
- Design - Safety
3.3.2 Selection and Installation
Where electrical equipment has to be installed in hazardous areas, equip-

ment with a type of protection suitable for the relevant zones shall be se-
lected and specified in accordance with the Model Code of Safe Practice -
Part 15 of the Institute of Petroleum (see also Appendix 1 of DEP 33.64.10.10-
Gen).
The construction of electrical equipment shall preferably comply with the

requirements of the relevant parts of IEC publication 79 or CENELEC doc-
ument EN/50.014-50.020 (9).

Various types of protection are available, the following information should be

used to determine correct selection:
- Equipment complying with other internationally recognised standards may

be acceptable, refer Appendix 1 of DEP 33.64.10.10-Gen.
- For Zone 1 type ‘e’ protection should be used where possible, rather than
type ‘d' protection. Ex(d) equipment is in general more expensive than its
Ex(e) counterpart and its integrity for explosion proofness is heavily main-
tenance-quality dependent. The type of protection ‘e' is specially attractive
for the construction of lighting fixtures, terminals jointboxes and low vol-
tage squirrel cage induction motors.
- Ex(p) equipment requires clean and safe air supply and therefore requires
auxiliary equipment in the form of aircompressors, filters, ducts and con-
trol/alarm equipment for air pressure and flow.
- When installing Ex(p) type of equipment due attention shall be paid to the
design of ducting, filters, air supply source and safety devices (control/
alarm).
- Where rooms, containing industrial type equipment, need to be located in a

hazardous area, such rooms shall be pressurised from a clean source of
air supply which shall maintain an overpressure of 0.5 mbar. Additional
precautions are required to maintain a safe environment, refer to l.P.
Electrical Safety Code.
- For Zone 2 non-sparking equipment should of be used, either good indus-

trial non-sparking (Appendix 1 of DEP 33.64.10.10-Gen.), or type of protec-
tion ‘n’ material.
For lighting fixtures in Zone 2, however, type of protection ‘e' shall be
used.
For liquefaction gas plants type ‘n’ equipment shall be used on the jetties in
non-hazardous areas, with the exception of lighting fixtures, which shall be
type ‘e' in accordance with the above.
Type ‘n' restricted breathing equipment shall not be utilised, as its integrity
depends on the quality of maintenance. In non-hazardous areas type ‘n’
restricted breathing may be utilised, e.g. for floodlights.
Note: The use of Zone 1 equipment in Zone 2 areas may be considered necessary for
unmanned platforms or installations where leaks may persist for some time before
they are dealt with.

3.4 Electrical System Design
3.4.1 Preface
For electrical system design reference is made to DEP 33.64.10.10-Gen. "El-

ectrical Engineering Guidelines".
It is equally applicable to MF and EP installations, both onshore and off-
shore.
In the subsequent clauses of this Section deviations from the above DEP,
which are applicable for EP environments, will be stated. For ease of refer-
ence the applicable DEP clause number and title will be repeated. The de-
viation may be in the form of an addition, deletion, modification or a re-
mark.
Section 3.4 must therefore be seen as an EP addendum to DEP 33.64.10.10-
Gen. - 1992 edition.
When designing an electrical supply and distribution system, the following

alternatives for the electricity supply shall be considered:
- Direct drives, i.e., equipment driven by gas engines or gas turbines
- Own generation
- Outside supply (public supply system)
- Any combination of the above possibilities.
Selection of the optimum power supply arrangement is based on a number of

factors, for example:
- Initial capital outlay
- Weight and space requirements, in particular for offshore installations
- Maintenance costs, i.e. maintainability, manpower (quantity and quality)
and spare parts
- Running costs, i.e. efficiencies of equipment, luboil consumption, etc.
- Operational cost, i.e. operability and required manpower (quantity and
quality)
- Fuel, i.e. availability, type and policy.
A good insight into the merits of the various alternatives may be obtained via
a system effectiveness study, when taking into account the impact of the
electrical system selected on the whole of the process.
3.4.2 DEP Clause 4.5.3 - Distribution Systems
For large production areas (i.e. onshore) ring type distribution systems as
opposed to radial type systems are an acceptable economic solution.

3.4.3 DEP Clause 4.2.2 - Load Assessment and Electricity Consumption
In addition to the preparation of Utility Data sheets for the total installation, for
complex and for large installations, consideration may be given to preparing
load summaries for the various modes of operation, such as:
- Emergency mode
- Life support mode
- Production and export
- Production and gas reinjection
- Production and water injection
- Non-industrial loads
3.4.4 DEP Clause 4.3.3 - Deviations and Variations in Supply Wave Forms
Due consideration shall be given to the influence of reciprocating loads

(pumps, compressors) and/or reciprocating prime-motors on the supply
wave form. In particular on systems with an appreciable load of a recipro-
cating nature, a study shall be carried out to determine for the various modes
of operation, the percentage and frequency of the voltage fluctuations (see
also 3.5).
3.4.5 DEP Clause 4.7.2 – Generators
Generators over 1 MVA shall be equipped with

- Negative sequence current protection
- Diode failure detection/alarm by determining change in harmonic content
of field current
- Overload alarm.
See also condition monitoring of major rotating equipment elsewhere in this
Volume.
3.4.6 DEP Clauses 4.7.7.2/3/4 - Variable Speed Drives/Synchronous Motors/Asynchronous Cage-

Induction Motors
Equipment over 3-4 MW shall be equipped with a restrictive start relay and
stall-protection.
Motors located in Zone 1 and 2 areas shall have residual current earth fault
protection.
See also condition monitoring of major rotating equipment elsewhere in this
Volume.

3.4.7 DEP Clause 5.6 - Electric Drives
In view of failures experienced with HV drives on offshore installations an

improved specification as compared to the present DEP 33.66.05.31-Gen. is
required.
The offshore situation differs in many instances from the onshore situation
such as:
- The higher number or starts per hour offshore as a result of integrated fuel
systems and processes
- For space-saving offshore 2-pole drives are used rather than the more
reliable 4-pole drives
- Vibration of offshore structures
- The smaller power systems offshore, subject to larger and longer voltage
and frequency dips
- Cooling by sea water.
To cater for the offshore regime an addendum to DEP 33.66.05.31-Gen. has
been prepared to reflect offshore experience and to arrive at a design more
suitable for the conditions as outlined above. Refer UENA/41/620 ‘Specifica-
tion for HV electric motors for North Sea installations'.
3.4.8 DEP Clause 5.6.2 - Variable Speed Drives
Due regard shall be given to the use of variable speed drives.
Application of variable speed drives has following advantages:

- Easier control of process
- Because of ability to vary speed, less instrumentation may be required
(e.g. control valves)
- Energy conservation
- No excessive kVAr demand on power systems during motor start (Starting
current can be as low as nominal current)
- No excessive stresses on stator winding and rotor during motor start,
therefore higher reliability may effectively be achieved.
These advantages should be weighed against the disadvantages related to

variable speed drives:
- Higher capital cost (up to 2 times)
- 10 to 20% larger motor rating due to harmonics generated by the frequency
converter
- Measures may be required to reduce harmonic currents and resulting
supply wave form distortion in the power system
- Space required for converter cubicle.

3.4.9 DEP Clause 5.8/6.3 - Cables, Wires and Accessories/Cabling and Wiring
Selection and installation methods of cables for special applications such as

downhole pumping and subsea operations should be reviewed and agreed
upon in close consultation with the manufacturers.
3.4.10 DEP Clause 5.10/6.5 - Lighting Equipment/Lighting Installations
Design of the lighting and small power system shall fulfil the following func-
tions:
1. It shall provide the required level of illumination of the whole of the in-
stallation and power a structure's obstruction lights.
2. It shall provide an adequate number of small power socket outlets to meet
the needs of construction and maintenance activities.
3. It shall provide suitable fixed and portable emergency lighting sys-
tems.
(a) General
(i) In general, fluorescent luminaires, or floodlights shall be used where
practicable for large open areas.
(ii) The lighting installation shall be designed to obviate stroboscopic ef-
fects.
(iii) Lighting shall be divided into three categories:
- General lighting
- Emergency lighting
- Escape lighting for offshore installations.
(iv) Emergency lighting shall account for about 25% of the total ‘General
Lighting' and shall receive its supply from the standby generator bus bar
section. Consideration may be given to the use of fittings equipped with
batteries and self contained chargers.
(v) All lighting circuits shall be switched via double pole switches. Fuses
shall be provided on each circuit for short circuit protection and for
isolation purposes.
(vi) Special lighting shall be given consideration in areas beyond the util-
isation of normal lighting, e.g. gauge lights.
(vii) Portable emergency lighting units shall be available at all locations
where operating personnel will be present for 24 hours per day. In
addition ‘electrical' rooms shall have such units mounted next to each
door.
Each unit shall comprise a fixed wall-mounted trickle charger and ac-
cumulator powered handlamp suitable for use in a hazardous area.

Each unit shall be kept on trickle charge when not in use and shall
receive its power direct from a distribution board supplied from the
standby generator.
(b) Offshore Installations

In addition to the above statements the following requirements apply:
(i) Part of the emergency lighting system shall be dedicated to the platform
escape lighting system. Escape lighting shall be located in all areas and
shall be arranged to allow the safe movement of personnel about the
platform and to illuminate the escape routes, ladders, walkways, muster
points and direction signs to muster points. Platform status lights shall
be provided to indicate the state of platform security (approximately
60 W).
(ii) Escape lighting shall be achieved from fixed increased safety Ex(e)
fluorescent luminaires, similar to those used for general lighting but with
an electronic emergency power supply unit housed with the fitting. Dur-
ing a power failure, the output shall be maintained for about 90 minutes
with one tube in operation.
(iii) Portable emergency lighting units shall be provided at the exit doors of
all non-hazardous area modules including emergency accommodation.
The unit is identical to the one specified in (a) (vii) above.
(iv) Specific consideration shall be given to use of floodlight luminaires
which illuminate the sea beneath the lifeboats, vent stack structure, etc.
(c) Illumination Levels

These shall be in accordance with DEP 33.64.10.10-Gen. (Appendix 4) or local
or national regulation, whichever is the most stringent. When calculating
illumination levels a maintenance factor of 75% shall be used in production/
industrial areas and 85% for clean areas.
Section 3.9.2 gives a guideline for required levels of illumination at different
locations.
3.4.11 DEP Clause 5- Power and Convenience Outlets
Particularly for offshore installations these outlets shall be kept to the min-
imum and there should preferably be none provided in hazardous areas.
Small power outlets shall be limited to single phase only and controlled by
double pole switches. These shall be protected against short circuit and earth
leakage protection (30 mA earth leakage sensitivity for socket outlet supply-
ing portable equipment).
3-phase power outlets shall have short circuit and earth leakage (500 mA)
protection.

3.4.12 DEP Clause 5.10.5 - Portable Lamps and Tools
For portable equipment reference shall also, be made to DEP 70.08.10.13-

Gen., part 4, which gives specific information as to construction details and
recommended apparatus.
3.4.13 DEP Clause 6.6 - Earthing and Bonding
(a) General
All exposed metal of the electrical installation other than current carrying
parts, shall be earthed.
The resistance to earth shall be as low as practicable, such that in the case of
earth faults the electrical protective devices shall prevent shock hazard po-
tentials in any part of the equipment which is not part of the electrical cir-
cuits.
(b) Equipment and Structures

The metallic enclosures of non-electrical equipment, vessels, tanks, pipe-
lines, structures, etc. shall be earthed. In the case of offshore installations
these enclosures shall be connected to the steel deck structure via bonding
connections or via the equipment mounting/fixing bolts.
3.4.14 Navigation Aids Obstruction Warning Lights
Navigation aids shall be provided on all offshore structures and in the case of
an existing structure or an existing multi-platform complex, the designer
shall be responsible for taking into account the existing Navigation aids
equipment.
(i) General - Marking Lights Characteristics

All lights shall be exhibited from 15 minutes before sunset until sunrise and at
all times when the meteorological visibility, in any direction, is two sea miles
or less.
All lights shall exhibit the MORSE letter ‘U' about every 15 seconds. The
composition of the Morse character shall conform to the following:
- The duration of each dot, darkness period between dots and darkness
period between the dot and the dash, shall be the same
- The duration of the dash shall be three times the duration of the dot
- The eclipse between successive Morse letters shall be 8 seconds mini-
mum and 12 seconds maximum.

All lights of the same type have to operate simultaneously.
All lights shall be mounted with their beam axis directed in order to be neither
elevated above the horizontal, nor depressed below the horizon. The total
beam width in the vertical plane shall be not less than 2.5 degrees at the point
where, on the curve of intensity distribution, the intensity is 10% of the
maximum value.
The control equipment for triggering the lights, shall consist of a solid state
timer and will exhibit the following characteristics in every 15 second period:
- 1 sec on, 1 sec off

- 1 sec on, 1 sec off
- 3 secs on, 8 secs off.
(ii) Main/Secondary Lights
These lights shall normally be mounted between 12 metres (minimum) and

30 metres (maximum) above MHWS (Mean High Water Spring) and shall be
arranged so that an unobstructed flashing white light, with a nominal range of
15 sea miles is visible in any direction.
The intensity of each light, after all losses have been taken into account,
including lantern glazing, etc., must be 14 000 candela minimum.
In exceptional cases, where the height of the platform (module) is such that
the main lights cannot be mounted below 30 metres, they may be mounted at
a height about MHSW of 35 metres maximum. In such cases the beam axis
shall be directed at the horizon.
All structures shall be equipped with a secondary lighting system, due to

come into operation automatically on failure of the main lights. The secondary
system shall be capable of operation for at least 96 hours, supplied from a
power source independent of the one from which the main lights are oper-
ated. The secondary lights shall have the same characteristics as the main
lights and be mounted as for 12.1.4.15(i). However the nominal range for the
white light is reduced to 10 sea miles. An alarm signal shall be fitted in the
radio room in order to give a warning, should the main lights system fail.
(iii) Subsidiary Lights
Red lights (subsidiary) should all be positioned so as to mark the horizontal

extremities of the structure, except when and where these extremities are
already marked by main lights. Each subsidiary light must provide the maximum
arc of illumination possible from its position, and have a normal range
of three sea miles.

The intensity of each red light, after all losses have been taken into account,
including red filters and lantern glazing, etc., must be 15 candelas minimum.
Each light shall be capable of operating independently for 96 hours.
(iv) Main/Secondary Fog Signals
The fog signals, by means of fog horns, are to sound the Morse letter ‘U' as
follows:
- Blast 0.75 seconds

- Silence 1.00 second
- Silence 1.00 second
- Silent 24.00 seconds
____________
Total 30.00 seconds
The fog signal emitters shall be installed between 12 metres (minimum) and
35 metres (maximum) above MHWS and shall be designed to have a usual
range of at least two sea miles, in any direction.
The fog signal shall be sounded whenever the meteorological visibility in any
direction is less than two sea miles.
All structures shall be equipped with a secondary fog signal system with a
usual range of at least one half sea mile in any direction. The secondary
system which shall be independent of the main system, shall come into
operation in case of total failure of the main system, or if partial failure of the
main system results in the usual signal falling below one half sea mile in any
direction.
The main and secondary fog signals shall be each capable of operating for at
least 96 hours from a power source independent of the main supply. An alarm
signal is to be installed in the radio room to give a warning, should the main
system fail.
(v) Remarks
Nominal range of lights shall be computed in accordance with Appendix II of

the IALA publication ‘Recommendation for the Notation of Luminous Intensity
and Range of Lights’.
The Usual Range of the fog signal is to be computed in accordance with the
IALA report ‘The Definition and Method of Calculation of the Nominal Range
and Usual Range of a Fog Signal'.
Nothing in the marking explained in this Section shall effect the requirements
for all ‘vessels' to comply with the relevant rules of the ‘International Regulations
for Prevention of Collisions at Sea'.

3.5 Electrical System Studies and Calculations
Depending on the complexity of the electrical system the following studies

and calculations may be required:
- Load flow studies

- Short circuit calculations
- Motor starting studies
- Voltage wave form analysis, in particular for installations with thyristor
controlled equipment or reciprocating pumps, compressors or prime-
movers
- System stability studies for various disturbances, e.g., shortcircuit conditions,

generator outages, starting of large motors
- Protection studies
- Lighting calculations.
The studies and calculations should be carried out at the commencement of

the project, using preliminary data. As the project proceeds and equipment is
works tested, the previous studies are to be reassessed and repeated, if
necessary with works test values.
The outcome of the studies provides such information as:

- Required switchgear fault rating
- System voltage and frequency fluctuations due to transient conditions or
oscillatory torques
- Requirements for special protective devices
- Maximum simultaneous motor start a/o minimum generation capacity for
successful large motor start
- Setting of protective relays
- Requirements for open-bus section operation a/o two out of three inter-
locking etc.
Consideration should be given to the full use of computer-aided techniques to

complete the majority of the above studies successfully. Many computer
programmes nowadays exist which combine into one programme the capa-
bility of performing load flow analysis, short circuit calculations, dynamic
stability and waveform analysis.
Finally the system effectiveness studies should be mentioned.

These studies concern the process stream in its entirety and yield results as
to the required availability figures for each piece of equipment or system
installed in order to satisfy a desired production target for given capacities of
the equipment or system installed.
These studies provide a powerful tool for comparing the economies of direct-
driven vs. indirect-driven (electric motor driven) equipment. For the electrical

engineer they give information on the required availability of the electrical

supply system, whether self-owned or public system, and also establish
the minimum requirement for redundancy, i.e. standby and spare capacity.
3.6 Installation Practices
3.6.1 Introduction
For onshore installations basic guidelines are given in DEP 33.64.10.10-Gen.

In addition, locally developed standards will apply, as well as regulations
issued by the government of the country concerned.
In the case of offshore installations additional requirements are often necessary.

In many cases these are already laid down in government regulations
and as such their implementation is mandatory.
Further applicable instructions and practices can be found in the Fire and
Safety Philosophy manuals issued by various Operating Companies.
Finally, attention is drawn to the lEE publication ‘Recommendations for El-

ectrical and Electronic Equipment for Mobile and Fixed Offshore Installations’.
The following Section contains recommendations for installation practices

for various types of electrical equipment and installations.
3.6.2 General
1. Rooms in which electrical apparatus is placed shall be suitably con-

structed and ventilated to deal with the heat dissipated by the apparatus
under full load conditions.
2. Because of the potential dangers introduced by electrical apparatus and

of the difficulty in maintaining its safe characteristics in service, such
apparatus should not as far as practicable be installed where flammable
gases or vapours are liable to accumulate.
3. Electrical equipment shall, as far as practicable, be so placed that it is not

exposed to risk of mechanical damage of excessive vibration. If equipment
must be installed in locations subject to vibration, consideration
should be given to mounting it on anti-vibration mountings and metal
enclosures provided with flexible arrangements for earthing.
4. Electrical equipment shall be so selected, and located or protected that it

is unaffected by sea-air, water, steam, oil or oil fumes, exhaust gases,

etc. to which it is likely to be exposed. If pipes must be run adjacent to

electrical equipment there should be no joints within the immediate vicinity
of the equipment.
5. Electrical apparatus should be so installed that sufficient space is available

for maintenance.
3.6.3 Generator Room
1. Generators shall be installed in non-hazardous areas only.

2. Generator package ventilation and generator room ventilation shall be an
independent system. Fans for the ventilation system shall be suitable for
Zone I areas but duplication is not required.
3. Fire and gas protection systems shall be provided inside the generator/
turbine enclosure and generator room.
4. Additional isolation facilities for fuel and cooling water lines shall be
provided outside the generation areas.
5. Lifting facilities shall be provided for the heaviest unit.
6. Only generator/turbine control and ancilliary equipment shall be permit-
ted to be permanently installed installation in generator rooms.
7. Provision must be made for the withdrawal of the largest unit.
8. Where emergency diesel generators are located, the walls and floors
shall have A60 rating.
9. An emergency stop switch capable of stopping all generators shall be
provided outside the generator room.
10. The emergency generator shall have a black start capability and be able
to function independently of the main ventilation system operating.
11. An earthing bar shall be provided for system earthing connection.
12. Accoustic cladding shall be used to limit noise to 90 dB.
3.6.4 Switchgear Rooms
(Generally applicable also to rectifier and inverter rooms)
1. Switchgear rooms shall be located in non-hazardous areas.

2. If switchgear rooms are located in Zone 2 areas, the requirements of
DEP 33.64.10.10-Gen. Clause 6.7.1 shall be met.
3. Service lines, i.e. fuel, water and air lines, are not permitted in switch-
gear room.
4. The switchgear room shall be designed as dust-tight and water-tight as
reasonably practical and shall be provided with at least two means of
exit. Both doors shall have ‘panic bolts' on the inside for emergency exit;
one door shall be fitted with a spring lock; at least one door shall be so
designed that the largest piece of equipment can pass through it.

5. Platforms at front and rear of switchboards should have insulating non-

slip surfaces.
6. An unobstructed passage-way not less than 1 m wide should be provided

in front of the switchboard, with switchgear in a fully withdrawn position.
7. The space at the rear of any switchboard, should be ample to permit

maintenance, and in general not less than 0.6 m clearance, except that
the width may be reduced to 0.5 m where stiffeners and frames are
located.
8. The locks used on doors shall be of a special series, different from locks
used for non-electrical premises.
9. The safety (pad) locks used to lock off switches of enclosures, chambers
or cells containing exposed live parts shall be of a different series to door
locks.
10. Where necessary, switchgear rooms shall be provided with HVAC.
11. The switchgear room shall be so dimensioned and switchgear so located,

as to allow easy operation and maintenance. Sufficient room for
future extension of switchgear by at least two panels at each end without
change to the building shall be included in the building design.
12. An overhead lifting facility shall be provided for lifting the heaviest piece
of equipment.
13. Fire and gas protection system shall be provided for offshore installations.
14. Earthing bars for system earthing and equipment bonding shall be provided.
15. Switchgear rooms containing switchboards which are supplying vital

and emergency services shall have A60 rating walls and floors.
16. An emergency switch shall be provided outside the switchgear room

capable of cutting off all main power inside the switchgear room for
offshore installations.
3.6.5 Transformer Enclosure
1. A detachable fence around the transformer shall be provided. The en-

trance gate shall have locking facilities.
2. The enclosure shall have a water-tight floor and shall be bunded to hold
the total volume of transformer cooling fluid. Complete drain facilities
shall be provided.
3. Gas and smoke protection shall be provided.
4. Ventilation shall be provided.

3.6.6 Battery Room
1. The battery room shall be located next to the rectifier and inverter
room.
2. No service lines shall be permitted.
3. No cables permitted other than cables connecting rectifier or inverter
unit and local lighting.
4. Lighting fittings shall be of type certified for use in hydrogen atmosphere.
5. Switches controlling lighting fittings shall be located outside.
6. Walls and floors shall be A60 rated.
7. The ventilation system shall be independent of any other system.
8. The battery room should not form a means of access to any other compartment.
9. The battery room shall be gas tight and pressure inside shall be negative
relative to adjacent room.
10. A gas detection system shall be provided.
3.6.7 Cables
1. All cables shall be so constructed, installed and protected as to prevent

danger.
2. Cables shall be run in cable racks, trays or cable ladder systems.
3. Cables shall be laid in the following manner:
- HV and LV single cores in trefoil formation with 150mm space between
trefoils
- HV multicores in one layer, touching
- LV multicores in three layers maximum, touching, with 25% spare
capacity.
4. The cable racking shall be designed to ensure correct segregation between
instrumentation and power cables.
No DEP guidelines exist as yet applicable to offshore installations. The

segregation dimensions given below are therefore indicative only:
Electric/Instrument Power Min. Separation from Non-Power In-

Cables strument and Telecom. Cables
254 V AC 45 mm
440 V AC 600 mm
6600 V AC 1200 mm
5. Supports for horizontal cable racks shall be spaced according to the type
and width of rack and weight of cables run on it. The maximum spacing of
support shall not exceed 3 metres.

Supports for vertical cable racks shall be spaced at approximately 1

metre intervals. Vertical cable rack risers shall not be installed in front of
or over pipe risers.
Cables shall be clamped or cleated to the cable racks, with approved

type cleats along the full length of run at intervals not exceeding those
shown below.
Overall Diameter and Maximum Spacing of Cleats (mm)

Type of Cable
Horizontal Run Vertical Run
HV Multicore 2000 600
LV Multicore cables 1500 600
HV and LV Single core cables 1500 600
LV cables 35 mm in diameter and below shall be secured along their full

length by non-nylon cable ties, PVC coated stainless steel or polypro-
pylene depending on application, at intervals not exceeding 1 metre.
Vertical runs to be clipped at 600 mm intervals with PVC stainless steel

clips. Cables are not required to be individually tied, but shall be tied in
groups. Single core cables laid in trefoil formation shall be secured to
cable racks by stainless steel cleats which are designed to withstand the
forces applicable during a short circuit.
The spacing of cleats on horizontal runs may be applied to runs at an

angle of up to 60 degrees from the horizontal, and spacings for vertical
runs applied to runs at an angle of up to 30 degrees from the vertical.
Cables shall not be fixed to pipes or structural members other than via
racks etc. or structural steel specially designed i.e. control station columns.
6. Flameproof/weatherproof cable glands approved for use in a Zone 1

area, shall be used for all LV and HV cable terminations.
Barrier type glands, e.g. Hawke I.C.G., shall be used for direct entry type
equipment containing sparking contacts, and instrument connections,
e.g. pressure switches or similar equipment where failure of the sensing
element results in high pressure inside the termination box or conduct
entries.
7. The armour of single core cables shall be earthed at one end only.
8. Each cable shall be identified at its point of origin and termination, at both
sides of a cable transit, at every 15 m throughout its length and at each
change of direction.
Identification markers shall be fitted to both the cables and the cores.
9. In general all bulkheads and deck penetrations shall be fitted with

Hawke/MCT cable transits, including all accessories as recommended

by the manufacturer.
This arrangement shall particularly apply to:
- Blast walls
- Fire walls
- Walls between safe and hazardous areas
- Through walls, roofs and floors to the open air.
All transits shall be individually and uniquely numbered and the cables
passing through a transit recorded.
Spare ways in transit frames shall be fitted with blank filling blocks.
10. Multicore cables shall have preference over single core cables, except
for generator and transformer secondary cables where high current ratings
are required. Cables shall be selected in accordance with the requirement
of services to which cables serve. Selection guidance is given
in the MESC Data Sheet for cables.
11. Where there is a large concentration of cables, consideration shall be

given to the provision of heat and smoke detection in order to minimise
cable damage due to fire.
12. Where cables pass above access ways adequate protection for personnel
shall be provided.
13. Consideration should be given to a tag numbering system for cables

which identifies cables by location, discipline, function and voltage level
throughout a company's installations.
This is particularly advantageous for large, complex installations spread
over various locations. When thoroughly adhered to it facilitates engineering,
maintenance and trouble shooting.
The following serves as an example of how such a system may be applied

and which items to include:
(i) Service Code Identifies Area of Installation of

Load
(ii) Discipline Code E = Electrical

I = Instrument
T = Telecommunications
(iii) Function Code P = Power

C = Control
D = Data
L = Lighting
T = Telephone
G = Earthing
F = Fire and Gas

R = Radio Frequency
S = Status/Alarm
H = Thermocouple/RTDs
A = Analog
N = Pneumatic
(iv) Unique Numbering Code 33 kV cables 1000 - 1099

11 kV cables 1100-1199
3.3 kV cables 1200-1299
600 V cables 2000-2499
240 V cables 2500-2999
110 V cables 3000-3999 and
below
All DC cables 4000 - etc. and
telecom. cables.
Examples:
- A 600 V class electrical power cable serving the imaginary OMEGA-3
Compression Platform would be identified as OMCP-3-EP-2000
- An instrument status/alarm cable serving the imaginary BETA 6 Control

Room would be identified as BECR-6-lS-3000.
3.6.8 Lighting
1. Luminaires shall be located so that they are accessible for maintenance

and lamp-changing without the use of ladders of other equipment.
2. Installation of luminaires shall follow standard drawings, taking into ac-

count the effects of vibration, wind-force and mechanical damage.
3. Emergency lighting fittings shall be marked red for easy identification.
4. All floodlights to be mounted on stirrups. The flexible cable for connec-

tions to floodlights shall be of adequate length to permit the floodlights to
be swivelled on its mounting through 360°.
5. Floodlights (narrow beam) shall be used for illuminating tall structures

such as vent stack, flare, etc.

3.6.9 Earthing
1. The structural steel of the modules and other structures shall act as the
main earth.
2. Each module shall be provided with earth connections in at least two

positions for each of the following type of earth connections of system
neutral and bonding of HV equipment.
3. Bonding of cable rack system shall be considered only where bolted

connections and fittings do not provide adequate earth continuity.
4. For a non-metallic enclosure, an additional brass gland plate shall be

provided to ensure earth continuity of cable.
5. Where cables are terminated by glands, in equipment having clearance

holes, earthing shall be obtained by providing ‘star' washers between
the gland and the face of the gland plate.
6. The use of tapped entries into sheet metal enclosures shall not be permitted.
7. Bonding of computer, telecommunication and instrumentation drains

shall be in accordance with DEP 33.64.10.10-Gen and shall be considered
in relation to vendor documentation.
8. Protective earth conductors shall be single core stranded annealed cop-

per conductors, PVC insulated, coloured green and yellow and shall be of
minimum cross sectional area related to service cable size as follows:
(see also DEP 33.64.10.10-Gen, clause 6.6 (Appendix 11)
Service Cable Minimum Bonding Conductor Size

Size, mm2 required, mm2
Up to 50 25
70 35
95 50
120 70
150 70
185 95
240 120
9. Protective earth conductors are required for the following:
HV and LV Generators
Transformers
HV and LV Switchboards
Motors (250 kW and above)
Rectifiers and Inverter Units
Reactors

3.7 Maintenance
With the increasing sophistication and complexity of equipment, require-

ments for high availability and statutory requirements on production facilities,
an efficient Maintenance organisation is a prerequisite for a cost-
effective operation.
With regard to the technical aspects of electrical plant maintenance, no

general rules have been issued so far. Nevertheless, there are some general
considerations that should be taken into account and which are to be applied
during the various phases of a project, i.e. design/engineering/construction
phase on the one hand; and the operational phase on the other.
3.7.1 Design/Engineering/Construction Phase
The objective of achieving cost-effective maintenance is dependent upon

sound design, engineering and construction philosophies and practices. The
main ones are listed below:
- Install standard and well-proven (approved) equipment. This will result in

reduced numbers of spare parts, less work-errors and reduced training
requirements.
- The reliability of the system should be adequate, i.e. the amount of spare/
standby equipment installed should be based on sound engineering prac-
tice and system effectiveness studies for the whole of the project.
- The electrical system should be flexible such that revamps and extensions
do not result in interruptions and/or major shut-downs. This usually means
the provision of spare capacity in equipment at the time of start-up.
- The installation of electrical equipment should be easily accessible for

(in-situ) maintenance, repair or trouble shooting, leading to a reduction in
equipment down-time.
- Design-engineers should receive and apply feedback from operations and

maintenance staff regarding equipment and systems proposed. Both oper-
ations and maintenance staff can make useful contributions and give ad-
vice on the behaviour and characteristics of equipment based on their field
experience.
- Prior to starting commissioning, the engineering department shall ensure

that the maintenance and operating departments are in possession of all
as-built drawings, vendor manuals and vendor drawings.

As the commissioning proceeds preprinted check-lists shall be completed

for each system or major piece of equipment commissioned. Each check-
list shall be signed by representatives of the departments involved, i.e.,
engineering, operations and maintenance.
Electrical maintenance personnel should, as far as is practically possibly, be

involved in commissioning activities. This becomes a necessity where it
concerns energizing systems which are connected to, or have a direct impact
on already existing operational systems. Useful information on commissioning
procedures and examples of check-lists can be found in DEP 63.10.08.11-
Gen.: ‘Field Inspection of Electrical Installations and Equipment'.
3.7.2 Operational Phase
From an electrical point of view the operational phase of a project covers two
aspects, namely the operation of the electrical system (generation, switching
in the distribution system) and the maintenance of both the electrical distri-
bution system and the electrical equipment of the actual production facilities,
including (for certain Opcos) the related infrastructure such as houses, clubs,
street-lighting, etc. To maintain a given plant cost-effectively the following
points should be evaluated and conditions satisfied:
- For static electrical equipment the applicable maintenance policy is nor-

mally a compromise between breakdown and a preventive (time-based)
maintenance policy, with the emphasis, however, towards preventive
maintenance. For electric motors driving key-equipment and generators, a
policy of condition monitoring can be considered. In any case the design-
out maintenance philosophy should be considered and evaluated for its
cost-effectiveness at an early stage of the design. The main aim should
always be towards optimum maintenance effort and costs consistent with
the desired system availability and safe working practices.
- Job routines for maintaining equipment should be available. They should

be specific for each type of maintenance-activity to be performed. Prefer-
ably the check-list format should be adopted along with standard applic-
able codes for failures and corrective actions taken. Records based on
such a system, lend themselves easily to a computerised database. Com-
bined with a computerised work order system, retrieval of historical data,
performance and maintenance cost per unit of equipment are easy and
quick to access.
- The number of maintenance staff required should be derived from the

maintenance plan. This plan lists all planned maintenance actions in both
frequency and hours per year. The total number of annual man-hours

required should be increased by a percentage to cover unscheduled ac-

tivities established by previous years' experience.
It is important that the records of hours/activity and the frequency of ac-

tivities are regularly updated to reflect the actual situation in order to
permit regular reviews in the interests of optimising the maintenance cost
and effort.
The ratio supervisors/foremen vs. labour/electricians/technicians is diffi-

cult to establish as this would depend on local circumstances such as
logistics, whether plant offshore or onshore and quality of staff.
- Due regard should be given to the qualifications and training of staff.

Formal training courses should be available if outside-company training
facilities are non-existent or inadequate. This should be augmented by
dedicated equipment training and on-the-job training by line/field trainers.
- Workshop facilities for repair overhaul of equipment, including test equip-

ment and tools should be available on site. Useful information on the layout
of a workshop and recommended test equipment and tools can be found in
DEP 70.08.10.13-Gen. ‘Equipment and Tools for Maintenance and Inspection,
Part 4'.
- The Engineering department is normally responsible for defining and or-

dering the spare parts for the first year of operation. Thereafter it is the
responsibility of the Maintenance and Materials department. This should
be defined early in the project.
Consideration should be given to the methods and restrictions governing

the procurement of spare parts in the future. Government bodies could
restrict importations, also local suppliers might be favoured.
SPIR (Spare Parts and Interchangeability Record) sheets should be pre-

pared well in advance. These will assist the responsible parties in ordering
the initial and future spares in order that adequate and optimum levels are
obtained consistent with economic policy related to plant availability versus
cost of spare parts stock. If these SPIR's can not be prepared
in-house, the preparation should form part of the design contract.

3.8 Electrical Safe Working Practices
Safe working practices for electrical systems and equipment are of extreme
importance as they are a definite requirement to safeguard staff working on
or near electrical equipment and to ensure continuity of supply, where applicable.
Every Operating Company shall have a set of safety rules including a permit
to the work system, with which electrical staff should be conversant in detail.
The safety rules should at least cover the following topics:
- Definitions of terms and expression
- Responsibilities for the operation of power systems (switching and isolat-
ing/earthing)
- Switching procedures
- Isolating/earthing procedures
- Work on electrical equipment
- Permit to work procedures.
Guidance for the establishment of safety rules is given in ‘Recommendations

for Electrical Safety', issued by the Shell Safety Committee.
It should be noted that local circumstances may require enhancement of this

document as a result of additional responsibilities of the electrical mainten-
ance department, work carried out by contractors or conditions imposed by
the local governmental electrical authorities.
The safety rules should be issued as a company-wide procedure approved at

management level. The requirements set out should be strictly adhered to,
e.g. rigid adherence to the operational procedures, introduction and appli-
cation of the permits specified in the permit to work procedure and the
nomination of suitable staff as competent and authorised persons.
The appointment of authorised persons is the responsibility of the Company's
senior electrical engineer.
The purpose of the permit to work systems is to ensure safe working practices
but this document can also serve as evidence in the case of mishaps
which might have legal implications.

3.9 Additional Information
3.9.1 Selection and Marking of Explosion Proof Equipment
(See Tables 3.9-1 and 2)
Table 3.9-1 Explosion protection

Table 3.9-2 Type marking of equipment (IEC 79)

3.9.2 Guideline for Illumination Levels
Table 3.9-3 provides additional information on illumination levels referred to

in 3.4.12(c).
Table 3.9-3 Suggested illumination levels at different locations

3.9.3 Protection of Equipment
Table 3.9-4 Enclosures-degree of protection

3.10 Relevant DEPs, Standards and Other Publications
3.10.1 DEPs*)
- Symbols and identification -Electrical DEP 33.10.03.10-Gen.

- Electrical Engineering Guidlines DEP 33.64.10.10-Gen.
- Synchronous AC generators 1250 kVA and DEP 33.65.11.31-Gen.
above
Power transformers (Amendments/supple- DEP 33.65.40.31-Gen.
ments to IEC 76 and IEC 726)
- DC Supply unit DEP 33.65.50.31-Gen.
- Static AC uninterruptible power supply DEP 33.65.50.32-Gen.
unit (Static AC UPS unit)
- Engineering guidelines variable speed DEP 33.66.05.10-Gen.
drive systems (technique and applica-
tions)
- Electrical variable speed drive systems DEP 33.66.05.33-Gen.
- Electric motors - Cage Induction and DEP 33.66.05.31-Gen.
synchronous type
- Low-voltage switchgear and controlgear DEP 33.67.01.31-Gen.
assemblies (Amendments/supplements to
IEC 439-1)
- High-voltage switchgear and controlgear DEP 33.67.51.31-Gen.
assemblies (Amendments/supplements to
IEC 289)
- Field inspection of electrical installations DEP 63.10.08.11-Gen.
and equipment
- Equipment and tools for maintenance and DEP 70.08.10.13-Gen.
inspection, Part 4- Electrical workshop -
test equipment and tools
- Symbols and identification system instru- DEP 32.10.03.10-Gen.
mentation
- System cabling (instrumentation) DEP 32.37.20.31-Gen.
- Instrumentation installation procedures DEP 62.37.10.12-Gen.
- The division of responsibilities between DEP 32.00.00.10-Gen.
instrument and electrical engineering dis-
ciplines
- Spare parts for initial and normal opera- DEP 70.10.90.11-Gen.
tion
- The use of SI units DEP 00.00.20.10-Gen.
*) Design and Engineering Practice Group Publications

- Noise Control DEP 31.10.00.31-Gen.

- Standard forms DEP 00.00.10.05-Gen.
- Requisitioning DEP 30.10.01.10-Gen.
- Index to standard drawings Groups S67, DEP 00.00.06.06-Gen.
S68, S69
3.10.2 Group Standard Forms (DEP 00.00.10.05-Gen.)
- Electrical equipment summary DEP 05.00.54.84-Gen.

sheet (5 sheets)
- Utility data - electrical DEP 05.00.10.80-Gen.
- Schedule for LV switchboard (2 sheets) DEP 33.67.01.80-Gen.
- Schedule for lighting distribution switch- DEP 33.67.01.81-Gen.
board
- Schedule for AC instrument distribution DEP 33.67.01.82-Gen.
switchboard
- Schedule for DC instrument distribution DEP 33.67.20.80-Gen.
switchboard
- Schedule for high-voltage switchgear and DEP 33.67.51.80-Gen.
control-gear assemblies (3 sheets)

3.10.3 Group Standard Drawings (DEP 00.00.06.06-Gen.)
GROUP S 67
SWITCHGEAR, ELECTRICAL INSTRUMENTS AND DIAGRAMS

3.10.3 Group Standard Drawings (DEP 00.00.06.06-Gen.) (continued)
GROUP S 68
EARTHING, CABLES, WIRES AND ACCESSORIES

3.10.4 Other Relevant Standards and Publications
- Recommendations for Electrical Safety, Shell Safety Committee.
- The IP Model Code of Safe Practice Part 15; Area Classification Code for
Petroleum Installations.
- The IP Model Code of Safe Practice Part 8; Drilling and Production in

Marine Areas.
- API RP 500 B Recommended Practice for Classification of Areas for

Electrical Installations at Drilling Rigs and Production
Facilities on Land and on Marine Fixed and Mobile Platforms.
- API RP 500 C Classification of Areas for Electrical Installations at Pe-

troleum and Gas Pipeline Transportation Facilities.
- API RP 550 Manual on Installation of Refinery Instruments and Con-

trol Systems.
- UENA/41/620 Specification for HV electric motors for North Sea in-

stallations
- International Regulations for Prevention of Collisions at Sea.
- IALA Publication ‘Recommendation of the Notation of Luminous Intensity

and Range of Lights', App. II.
- IALA Publication ‘The Definition and Method of Calculation of the Nomi-

nal Range and Usual Range of a Fog Signal'.
- ERA 69.30 Current rating standards for distribution codes part

I to V.
- lEE publication on Recommendations for Electrical and Electronic Equip-

ment for Mobile and Fixed Offshore Installations.
- Relevant IEC recommendations, in particular the IEC 79-series.
- All statutory rules and regulations constituting the electrical code of the
country concerned.
Note: DEP 33.64.10.10-Gen. "Electrical Engineering Guidelines" contains a compre-

hensive list of applicable non-Group standards and regulations (chapter 9 references).

Chapter 4
INSTRUMENTATION
CONTENTS
4.1 Introduction......................................................................................................................................... 4-3
4.2 Principles of Process Control ............................................................................................................. 4-3
4.2.1 Measuring Process Variables.......................................................................................................... 4-3

4.2.2 Process Control Concepts.............................................................................................................. 4-4
4.2.2.1 Manual Control ............................................................................................................................. 4-4

4.2.2.2 Automatic Control ......................................................................................................................... 4-4
4.2.3 Accuracy......................................................................................................................................... 4-4

4.2.4 Control Systems .............................................................................................................................. 4-5
4.2.4.1 Self-Acting ................................................................................................................................... 4-5

4.2.4.2 Electronic or Pneumatic Control Systems.................................................................................... 4-6
4.2.5 Power Supply .................................................................................................................................. 4-6
4.2.5.1 Instrument Power Supply ............................................................................................................. 4-6

4.2.5.2 Instrument Back-Up Power Supply .............................................................................................. 4-7
4.2.6 Final Control Elements .................................................................................................................... 4-7
4.2.6.1 Regulators or Self-Acting Control Valves ..................................................................................... 4-7

4.2.6.2 Control Valves Operated by External Power................................................................................ 4-7
4.2.7 Signal Transmission Lines ............................................................................................................. 4-8

4.2.8 Area Classification.......................................................................................................................... 4-8
4.2.8.1 Flammable Atmospheres ............................................................................................................ 4-8

4.2.8.2 Hazard Determination .................................................................................................................. 4-9
4.2.8.3 Types of Protection ...................................................................................................................... 4-9
4.2.9 Control Room Aspects ................................................................................................................. 4-10
4.2.9.1 Centralised Control Rooms ........................................................................................................ 4-10

4.2.9.2 General Considerations.............................................................................................................. 4-10
4.2.9.3 Control Room Layout ................................................................................................................. 4-11
4.2.9.4 Auxiliary Room ........................................................................................................................... 4-11
4.2.9.5 Construction Aspects ................................................................................................................. 4-11

4.2.10 Instrument Workshop ................................................................................................................. 4-11
4.2.11 Standard Requisitions for Instruments ........................................................................................ 4-12
4.2.12 DEP Manuals, Standard Specifications and Standard Drawings................................................ 4-12
4.3 Emergency Shut-Down Systems..................................................................................................... 4-12
4.3.1 Function.............................................................................................................................................12
4.3.2 Type of System.............................................................................................................................. 4-13
4.3.3 Design Philosophy for Shut-Down and Depressurising Systems ................................................. 4-13
4.3.4 Shut-Down/Depressurising Actions............................................................................................... 4-13
4.4 Flow Quantity Measurement for Tax and Revenue Purposes ......................................................... 4-14
4.5 Control Philosophy ........................................................................................................................... 4-14
4.6 Systems........................................................................................................................................... 4-16
4.6.1 Process Control System................................................................................................................ 4-16

4.6.2 Data Acquisition............................................................................................................................. 4-17
4.6.3 Metering......................................................................................................................................... 4-18
4.6.4 Control Centre ............................................................................................................................... 4-18
4.6.5 Air and Electricity Supply for Instruments ..................................................................................... 4-19
4.6.6 Fire and Gas Detection ................................................................................................................. 4-19
4.6.7 Emergency Shut-Down Systems................................................................................................... 4-19
4.7 Instruments...................................................................................................................................... 4-21
4.7.1 Temperature Measurement........................................................................................................... 4-22

4.7.2 Pressure Measurement ................................................................................................................. 4-22
4.7.3 Level Measurement ....................................................................................................................... 4-22
4.7.4 Flow Measurement ........................................................................................................................ 4-23
4.7.5 Quality Measurements .................................................................................................................. 4-23
4.8 Maintenance ..................................................................................................................................... 4-24
4.9 Relevant DEPs ................................................................................................................................. 4-26
FIGURES
Figure 4-1 Block diagram of an automatic control loop 4-5

(crude oil to be heated to a given temperature)

Instrumentation
4 INSTRUMENTATION
4.1 Introduction
This Chapter describes process control, overall project control philosophy,

various instrumentation systems and the instruments themselves, and finally
the maintenance of instruments and its organisation, all within the context of
EP projects.
Other Volumes/Chapters refer to flow schemes and special measuring tech-

niques for particular applications.
Frequent reference is made to DEP publications and a complete list related to
instrumentation is given in 4.9.
4.2 Principles of Process Control
Basically, Process Control is achieved by using a number of techniques

which serve to stabilise the process(es) and operate the installation in a safe
and reliable manner resulting in an end-product to specification(s).
A Process Control system consists of a selection of instruments and instru-
mentation systems, the choice of which depends on the process fluids, the
product specification(s), the processing equipment, the type of process and
the processing plant location.
Note: Regarding engineering and design activities and other aspects of Process
Control and Instrumentation, the responsible Instrument engineer should be con-
tacted whenever possible.
4.2.1 Measuring Process Variables
Common process variables in the oil and gas industries to be measured (for
control purposes) are:
- Flow (rate and volume)
- Level
- Pressure
- Temperature.
Less common process variables to be measured are, e.g. density, pH value,

moisture content, conductivity, BS and W (Basic Sediment and Water), oil in
water content.

4.2.2 Process Control Concepts
Basically there are two ways to control a process:

(a) Manually
(b) Automatically
The related control equipment may be installed locally, centrally in a control
room, or part locally and part centrally.
4.2.2.1 Manual Control
The operator is assisted by (locally installed) process measuring instruments

only. He reads these local instruments and takes manual corrective action if
required, observes if this correction is sufficiently effective, and corrects again
until the desired process value is reached. This concept is only used for very
simple processes. The disadvantages are obvious.
4.2.2.2 Automatic Control
The measured process value is constantly compared with a desired value

(set point) by a process controller. Any deviation between the actual meas-
ured value and the desired value (set point) will be corrected automatically
by the relevant process controller. An automatic control system includes the
following elements (see Figure 4-1):
(a) A process measuring sensor
(b) An indicating or recording control unit
(c) A correcting unit or element, e.g. control valves
(d) Supervisory hardware (not shown).
The complete signal and conditioning system including the signal lines form
part of the automatic control loop. The main advantage of automatic control is
that the system works largely unattended and gives better, tighter and
smoother control. It also allows a single operator to be responsible for a larger
section of the process because his actual duties are reduced to supervising
and monitoring only, instead of physical interference with the process as is the
case with manual control. Thus the operator’s duty becomes limited to watching
the process values and ensuring that the automatic control system is functioning
properly.
4.2.3 Accuracy
The purpose of all measurement is to obtain the true value of the quantity
measured, and error is thought of as being the difference between the measured
and the true value.

Instrumentation
Figure 4-1 Block diagram of an automatic control loop (crude oil to be heated to a
given temperature)
Each measuring element or instrument has a certain degree of accuracy

(manufacturers state the maximum errors as plus/minus percentages of full
range). The required accuracy has to be determined for every individual
measuring case and it is left to the engineer to decide on this on the basis of
the purpose of the measurement, applicable regulations, contractual obligations
and third party or similar requirements. In the case of no specific requirements,
existing standard instruments are chosen from well-known manufacturers.
Recalibration may be necessary in order to maintain the accuracy of the
instrument's measuring range.
4.2.4 Control Systems
Basically there are three different types of automatic control systems:

- Self-acting (without external supply)
- Pneumatic (with external supply)
- Electronic (with external supply).
4.2.4.1 Self-Acting
Self-acting control takes its energy from the process and is normally limited

to the application of ‘self-acting control valves' (regulators) (see also

4.2.6.1).
The concept in general is limited to simple control loops due to the limitations
of these devices.
4.2.4.2 Electronic or Pneumatic Control Systems
These systems are designed for normal instrument installations. In general, it

is not possible to give a recommendation in favour of either electronic or
pneumatic instrumentation systems. Pneumatics have the advantages of lower
installed cost, safety in hazardous areas, and (usually) better availability of
maintenance personnel.
Electronics are favoured in case of requirements for computer compatibility,
long signal transmission lines, long-term maintenance aspects, very low
ambient temperatures, and the possibility of high density installation requiring
less space in cabinets and control panels.
Choosing between pneumatics and electronics is definitely not a choice of
controllers only, but requires a responsible engineer to consider the whole
system of measurement and control.
The important factors to be considered are summarised below:
- Capital expenditure (pneumatic usually cheaper)
- Maintenance cost (pneumatic are manpower intensive)
- Required complexity (electronics have [far] more possibilities)
- Signal transmission distance (electronics reach further)
- Signal compatibility (electronics easier to interface to telemetry computer
systems)
- Location (pneumatic [air] is safe by definition)
- Available construction and maintenance personnel/manufacturer’s support
- Accuracy (electronics are better)
- Development (mainly electronics).
4.2.5 Power Supply
4.2.5.1 Instrument Power Supply
For pneumatic instrumentation dry, clean and oil-free air is an essential

requirement. Use of associated gas should only be used in very exceptional
cases and for robust, locally mounted instruments only. Panel mounted
pneumatic instruments shall never be powered with gas as this will endanger
the control room area by the introduction of a flammable fluid.
For electronic instruments either 24 volt DC or 220 volt, 50 or 60 Hz AC is

Instrumentation
acceptable. Consult the relevant DEP manuals for more details.

Please note that 50 and 60 Hz instruments are not always interchangeable.
4.2.5.2 Instrument Back-Up Power Supply
Most plants require an extensive number of instruments and/or systems to be

provided with a back-up power supply to allow for the safe and orderly
shut-down in the event of a partial or total power supply failure. For pneumatic
instrumentation this will be an air vessel; for electronic instruments a rectifier/
battery/inverter system is required to provide AC and DC vital electrical supplies.
The number of instruments concerned, their power consumption, the actual

length of time this back-up power supply is to be maintained and the nature of
the supply must be considered along with the selection of the instrumentation.
1
In many cases, a back-up power supply having a capacity for /2 to 1 hour
independent operation will be adequate, though capacities for 6 to 12 hours
might be required for certain applications (e.g. electronic fire and gas detection
systems).
4.2.6 Final Control Elements
In the oil and gas industries a control valve is almost invariably installed as
the final control element influencing the actual process. These valves are
applied in a wide range of types, models, characteristics and construction
executions in order to fulfil their designated specific duties.
Basically, however, there are two major categories:
- Regulators or self-acting control valves
- Control valves operated by an instrument system.
4.2.6.1 Regulators or Self-Acting Control Valves
Generally, a regulator is a control valve with a built-in control mechanism.

The set point is provided integrally in the form of a mechanical adjustment
device. As stated before, self-acting regulators do not need an external
power source in order to modulate. This constitutes their main advantage.
4.2.6.2 Control Valves Operated by External Power
These control valves in contrast to regulators are available for almost any
process application.

Additional possibilities/characteristics are:

(a) Operation by external controller (remote or local) allowing any control
mode
(b) Remote manual control operation
(c) No fail-safe limitations
(d) Wide range of accessories can be attached (e.g. positioner, limit
switches)
(e) Interchangeability between many service conditions
(f) Available for operation on pneumatic, electric, or hydraulic power.
4.2.7 Signal Transmission Lines
Signal transmission lines accommodate:

(a) Pneumatic signals from pneumatic transmitters to their receiving in-
struments or controllers, and pneumatic signals from pneumatic con-
trollers to final control elements.
(b) Electric signals (including thermo-electric voltages from thermocouples)
from the transmitters to their receiving instruments or controllers and
from the controllers, to the final control elements.
Cables for comparable applications, such as plant communications, safe-
guarding systems or fire and gas systems can also be considered to form
part of the instrument signal cabling.
For connections between field, auxiliary room and control room, multicore
pneumatic and electric cables are preferred for economic reasons and
easy installation.
4.2.8 Area Classification
The electrical area classification system is used to achieve an adequate level

of safety when installing electrical apparatus in a flammable or potentially
hazardous atmosphere.
4.2.8.1 Flammable Atmospheres
A flammable atmosphere is usually a given condition; therefore adapting the

electrical instruments and/or their housing is normally the only way to main-
tain adequate safety.
When a flammable atmosphere is likely to be present, the following aspects
are of utmost importance:
-The type of instrument
- The internal release of flammable substances from the instrument
- The available types of certified protective housings
- The installation of the instrument
- The maintenance of the instrument.

Instrumentation
4.2.8.2 Hazard Determination
The determination of a hazardous area is described by various internation-

ally recognised Electrical Institutes (with minor variations for local and/or
national acceptances), e.g.:
- IP code part I - Electrical - issued by the British Institute of Petroleum

- International Electrical Code issued by the International Electrical Com-
mittee.
It should be realised that these Codes are primarily issued to determine the
minimum requirements for the design of electrical apparatus. They are by no
means exhaustive for all the dangers that may occur in petroleum installations.
Consult the process engineer for further details, who is also the authority to
issue/approve the area classification documents. In no case is an electrical/
instrument engineer to be held responsible for these documents.
The classification of hazardous areas consists of:
Zones (European standard) or

Divisions (American practice).
Definitions for these terms are given in the appropriate literature. Determining
the Zones or Divisions and thereby deciding on the probability of an explosive
mixture being present is usually the most complicated part of the exercise.
4.2.8.3 Types of Protection
The types of protection applicable to electrical instruments intended for use

in hazardous areas are generally based on one or more of the following
features:
(a) Keeping the level of electrical power so low that under any fault condition
there will be no spark of sufficient energy to cause an ignition (intrinsic
safety).
(b) Constructing the electrical and mechanical components such that there
will normally be no spark (non-sparking - non-incendive - increased
safety).
(c) Preventing the flammable atmosphere from reaching the electrical com-
ponents under normal conditions (immersion in oil, sand or powder filling,
encapsulation, hermetic sealing).
(d) Preventing intrusion of the flammable atmosphere into the enclosure

(pressurisation - purging).
(e) Preventing an ignition inside the enclosure from igniting the external
flammable atmosphere (explosion proof - flameproof).
For EP field operations usually (a), (b) or (e) are applied, (d) is often used for

analysers or to keep enclosed spaces (offices/control rooms) safe, (c) has no

special E P applications. The use of one of the above listed types of protection
depends mainly on the area classification of the location in which the instrument
is to be installed (assumes instrument has no internal source of release).
4.2.9 Control Room Aspects
4.2.9.1 Centralised Control Rooms
Centralised control rooms (CCRs) are to be considered for the following

reasons;
(a) Centralised control and supervision of the process
(b) Centralised protected location of control devices, systems, and auxiliary

equipment
(c) Application of ‘general purpose' classified equipment
(d) Reduction of operator manpower
(e) Easier maintenance due to concentration of equipment
(f) Centralised tie-in possibilities for computer assisted operation (CAO) or

telemetry systems.
4.2.9.2 General Considerations
In planning a control room, several factors should be taken into consideration.

Some are listed below:
(a) Controls to be laid out for convenience of operation and ease of main-
tenance.
(b) Control room to be designed for a low background noise.
(c) Equipment not essential for the operator should be located in the auxiliary
room.
(d) CCR/building to be located as centrally and conveniently as possible to

the process facilities (consideration for future expansion).
(e) CCR/building to be classified as a safe area by virtue of its location or by

adequate pressurisation.
(f) Plant fire and/or explosion risks are also to be taken into consideration.
Usually either the building has to be adapted to the risk (bunker type CCR)
or the distance has to be increased considerably in consultation with the
safety engineer (see also the related DEP manuals).

Instrumentation
4.2.9.3 Control Room Layout
The layout of the control room shall provide ample working space, arranged
for maximum efficiency, safety and accessibility.
Traffic through the CCR shall be minimised. Control room layout shall be
co-ordinated with the control panel design, duly considering ergonomic re-
quirements.
The best control room layout usually places the operator's desk at the centre
of his area of responsibilities.
4.2.9.4 Auxiliary Room
All rack or cabinet mounted instrumentation, pneumatic or electronic auxiliary

and safety systems, and other control related equipment shall be located in
the auxiliary room. This auxiliary room shall be adjacent to the CCR for signal
connections from and to the CCR mounted control panels.
The auxiliary room size is in general determined by the total volume of signal
distribution cabinets, electronic cabinets, instrument racks for supporting
instruments etc. The aim is to keep this equipment centralised for ease of
maintenance and accessibility but outside the CCR, thus avoiding mainten-
ance personnel interfering with the operators activities. General experience
shows that auxiliary rooms are almost invariably undersized. A floor area
equal to the CCR is strongly recommended even when initially only limited
floor space seems to be required.
When DCS systems are utilised, this usually results in a reduction in size of
the CCR and thus the auxiliary room becomes the larger of the two.
4.2.9.5 Construction Aspects
With respect to mounting of control panels and equipment cabinets in the

CCR and the auxiliary room, sufficient supporting shall be provided. Pres-
surised CCRs are to be provided with gas tight cable entries. CCRs and
auxiliary room preferably to be provided with ‘subfloors' (computer floors) to
allow for cable running. Various DEPs and standard drawings are available
on these subjects.
4.2.10 Instrument Workshop
Dependent on the required maintenance work, an ‘Instrument Workshop'

shall be provided and suitably sized to handle the maintenance and repair
activities for the instrumentation. The workshop shall be equipped with main-
tenance and test equipment for both mechanical and electronic equipment as
required.

Selecting testing instruments and tools for the instrument work shop should
be left to the instrument engineer. If necessary contact SIPM, The Hague
(EPD/64) for detailed specifications. See also DEP 70.08.10.13-Gen.
When electronic instruments are to be handled in tropical areas, part of the
instrument work shop is to be air-conditioned. Some administrative facilities
have to be catered for as well.
4.2.11 Standard Requisitions for Instruments
For requisitioning and ordering purposes, all common instrumentation items

can be specified and requisitioned by using/completing the relevant speci
fication and EDP forms.
These forms are identified with a form number, and can be found in
DEP 30.10.01.10-Gen.
4.2.12 DEP Manuals, Standard Specifications and Standard Drawings
Instrumentation work in Shell is standardised to a considerable degree. Use

these standards wherever possible.
The following indexes refer:

DEP 00.00.05.05-Gen. Index to DEP publications and standard specifica-
tions
DEP 00.00.06.06-Gen. Index to standard drawings
Documents listed in these indexes can be obtained from SIPM, The Hague,
(MFFN/52).
4.3 Emergency Shut-Down Systems
4.3.1 Function
Emergency shut-down systems are normally installed to protect

- personnel
- equipment
- surroundings (pollution)
when abnormal operating conditions occur.
The systems initiate protective actions and/or operate alarms.

Instrumentation
4.3.2 Type of System
Emergency shut-down systems can be manually or automatically initiated,

depending on the operating/safety philosophy established for the plant. A
manual system eliminates the problem of spurious trips by detecting systems
causing unwarranted alarms and plant shut-down but introduces the risk of
delayed action or no action at all.
4.3.3 Design Philosophy for Shut-Down and Depressurising Systems
The emergency shut-down and depressurising system (ESD and EDP) in

general should be:
- Fail-safe, reliable and simple
- Independent of normal control and relief systems
- Initiated in the event of operational upsets creating dangerous situations
which are outside the limits of the normal process trips e.g. a major gas
release into atmosphere, (threatening) fire or explosion.
The detailed philosophy is to be determined in close co-operation with the
process engineer, the safety engineer and the instrument engineer.
4.3.4 Shut-Down/Depressurising Actions
The actions to be carried out depend on the safety philosophy laid down for
the emergency shut-down and depressurising/emptying of the unit (part
plant isolation and sectionalising or total shut-down).
The usual action items for an ESD and/or EDP system are:
- Closing isolating valves, e.g. in and outlet of oil/gas treating unit
- Tripping of non-essential running equipment and fired process equipment
- Depressurising the protected plant section to flare, atmosphere or empty-

ing it into the drain system, API, waste pit or burn pit
- Closing liquid flows to or from vessels
- Keeping utility systems in operation as long as possible by cutting off

non-essential consumers.
The depressurising system design shall be made in close liaison with the
process, the safety and the instrument engineer. In many cases Government
requirements exist which have to be adhered to as a minimum.

4.4 Flow Quantity Measurement for Tax and Revenue Purposes
Flow quantity metering with respect to:
(a) Fiscal metering and
(b) Custody transfer/sales metering.
This type of measurement needs specific attention with respect to selection

of the metering device, the installation and calibration facilities for the fol
lowing reasons:
- ‘Fiscal Metering' is the basis for paying royalties and taxes on produced
oil and gas to the relevant government(s).
- ‘Custody Transfer/Sales Metering' is the basis for selling or transferring

custody of the oil, gas and products to customers.
- These measurement facilities are usually subject to Government Authority

approval (note that often one measurement facility serves both purposes).
Selection of metering devices for high accuracy is in general limited to:
- Positive Displacement meters for liquid flow
- Turbine meters
- Differential pressure (orifice) for gas flow.
Guidance should be obtained from a qualified Instrument Engineer, if necess-

ary in consultation with SIPM or SIPC (MK), regarding acceptable measuring
methods for the case concerned. An alternative method for accurate liquid
volume measuring for fiscal/sales metering or custody transfer is tank-dipping
of calibrated tanks. This matter should also be handled by a qualified instru-
ment engineer.
4.5 Control Philosophy
As EP production facilities range from integrated platforms linking unatten-

ded production stations to individual wells, it is not possible to present a
single control philosophy for all circumstances, except in a very general
manner. However it can be stated that a modern design for a production
facility should be based on minimum operator involvement and minimal
proliferation of instrument type and vendor. Whenever possible, facilities
should be provided in a control centre for the operator to:
(a) Monitor process and other equipment including fire and gas detection
and protective systems.
(b) Control, i.e. start/stop equipment or process trains and adjust set points
of control loops during normal operation.

Instrumentation
(c) Initiate separate control sequences, e.g. well tests, meter proving, emer-
gency shut-down.
(d) Record/report production and condition of wells, process and equipment
to assess performance.
The functions (a) and emergency shut-down initiation should always be cent-
ral, while in small production plants, it may be more cost-effective to have
recording and start/stop of processes and equipment locally or on local
panels. All aspects should be considered in the cost analysis, including
remote monitoring and future needs.
When the control system is to be developed around an existing facility, the

possibility of utilising the spare capacity already existing within the installed
system should first be studied. In the event of these facilities being inadequ-
ate, studies should be carried out to determine the feasibility of expanding
the existing systems to accomodate the additional requirements.
In selecting the control system and instrumentation for a ‘grass roots' pro-
duction facility, the following main aspects have to be taken into account:
- Size of the facility and complexity of the process.

Future extensions in number of wells and separation trains, possibility of
water and gas injection and future automatic data acquisition should be
borne in mind.
- Experience of operations and maintenance staff to be employed. Espe-

cially for new, remote areas.
- Local support from manufacturers.

This should include availability at the location of a service engineer as well
as timely delivery of replacement and spare parts.
- Environmental conditions.
For instrumentation in particular the climate may have far-reaching con-
sequences on the installation and its protection.
- Sociographical situation.
Locations near densely populated areas may require from the beginning,
or in the future, enhanced pollution control and data acquisition systems.
Considering these factors, a choice has to be made between local and cen-
tralised control. For the more complicated production facilities increasing
integration of the control functions will be seen in the future, aiming for fully
automatic, unmanned or unattended operation. The technical options may be
for pneumatic/hydraulic or for electric/electronic equipment. The actual
choice should take the following considerations into account:
If local control is preferred over centralised control, pneumatic control be-

comes the only practical option for EP projects as most control equipment is

located in Zones 1 or 2 and pneumatic systems are intrinsically safe.

It is entirely feasible to transmit all alarms and key process variables to the
control panel as pneumatic signals, feeding into pneumatic display/recording/
controlling instrumentation. This equipment is however of old design and is no
longer the subject of vendor development.
Pneumatic transmission, also, is limited to a distance of about 200 m and is

less accurate than electronic transmission. Moreover, when complicated
shut-down logic has to be applied, or in the case of thermocouples or re-
sistive temperature instruments or when much data has to be transmitted,
electronic (multiplexed) transfer is often the most cost-effective solution.
Some measurement and control functions must be electrical:

- Fire and gas detection
- Start/stop of electrical motors
- Electric power generation control and distribution
- Fiscalisation metering
- Quality measuring analysers
- Net oil computing devices.
For the high pressure required in the wellhead area, pneumatic/hydraulic

control is mostly applied.
4.6 Systems
The instrumentation requirements for an EP project will be reviewed by

considering the various subsystems, which are either ‘stand alone' to ensure
integrity and availability, or form part of an integrated system in order to
respond rapidly and without human interference to incoming signals from
other (sub)systems. The following systems will be reviewed:
1. Process control system

2. Data acquisition
3. Metering
4. Control centre
5. Air and electric supply for instruments
6. Fire and gas detection
7. Emergency shut-down systems.
4.6.1 Process Control System
The process control system governs the processes and flows under normal
conditions. Various parts of the process, each having its own control require-
ments, can be distinguished:

Instrumentation
- Wellheads and flow lines

- Separators and surge vessels
- Compressors
- Dehydration facilities
- Storage tanks and transfer pumps.
Some of these units are usually bought as a package. The Manufacturer's

(standard) control system shall then be in agreement with the overall control
philosophy as discussed in 4.5, with emphasis on reliability and simplicity.
Implementation of the control centre function is now achieved in the main by
a computer based distributed control system (DCS) which, as well as provid-
ing the monitoring and control features previously provided by single loop
control recording instruments, also provides additional features such as live
graphics packages, historical and live trending displays, alarm annunciators,
alarm and event reporting, special free format reports as well as hard copying
facilities. Many of the additional features were previously implemented in the
Supervisory Control and Data Acquisition System (SCADA), which now can be
reduced in size. If special supervisory features are required these are easily
implemented with data hand off from the DCS system.
4.6.2 Data Acquisition
With growing complexity of the operation of a production facility, the infor-

mation requirements increase. Also many remote, often unmanned, stations
require centralised monitoring, which in fact is a form of (continuous) data
acquisition.
The measuring equipment to be installed must reflect present and future

production data requirements.
Data acquisition is sometimes combined with supervisory control and is then

called a SCADA-system (Supervisory Control And Data Acquisition). Three
levels of information have been recognised:
- The first level (SCADA and manual inputs) provides data and information
for monitoring and control of the operations, i.e. through terminals/VDUs in
the control room, as well as data and information for further processing in
second level systems.
- The second level administers a selection of operational and technical data

and information for specific groups of users and/or disciplines (dedicated
systems, tactical level) and exchanges selected data and information with
the third level system(s).

- The third level administers a selection of data and information for long
term (corporate) planning and evaluation (strategic level).
4.6.3 Metering
Meter proving facilities should be available for custody transfer and fiscal
flow metering as well as for calibration of other flow meters.
A metering station should consist of a multiple meter run installation with at

least one spare meter run, designed in accordance with DEP 32.32.10.10
(para 16). For crude oil transfer, dedicated mechanical displacement provers
are common and can be supplied by various manufacturers. For gas transfer,
carefully designed metering stations, based on orifice plates as primary
elements, are used. A detailed description of such a gas station for fiscal
service is given in Code of Practice EA/029 of Shell U.K. Exploration and
Production.
For calibration of flow meters and measuring instruments for operational use,
no strict rules exist. Calibration shall always be carried out at the installation
and after breakdown and repair. The frequency of regular checks depends on
the conditions of service (see 4.7), and should be adapted to local experience.
In principle it should be possible to inspect and change out orifice plates and
flow meters that provide data for production administration systems without
facility shut-down. Similarly, valving and connections must be provided for
in-situ calibration of liquid flow meters performing that role.
4.6.4 Control Centre
The general conditions in the control room shall be managed to encourage

effective procedures, work patterns and personnel safety. This may be
achieved by minimising discomforts and distractions which tend to reduce
operator performance and increase the possibility of human errors.
The location of controls and displays should be carefully considered to assist

the operator in making speedy and correct decisions and also to assist the
maintenance technician in making a rapid and correct diagnosis and repair.
Therefore there must be a functional or geographical relationship in the
display of controls (automatic controller or push buttons), so that they are
immediately apparent and unambiguous to the operator, rather than group-
ing alarms, e.g. by type.
In particular for emergency information it is vital that fixed sequences are
executed automatically, leaving as much time as possible for operator de-
cisions.

Instrumentation
More details on ergonomic requirements of control room layouts can be

obtained from manufacturers. For complicated systems it is recommended to
have an ergonomic study carried out by specialists.
Special attention should be paid to the effects of installing additional equipment

in both the control and auxiliary room to ensure that the air conditioning plant is
able to cope with the extra demand.
4.6.5 Air and Electricity Supply for Instruments
The main parts of the air supply system are the compressor to compress air
to 8 bar in the buffer vessel and the drying equipment to reduce the dew point
to about 10 ºC below the lowest temperature occurring in the system.
The required quantity of air shall be based on use for pneumatically operated
instrumentation, pressurisation of enclosures in hazardous areas and dilution
for process analysers.
The use of associated gas should only be applied in exceptional cases, e.g.
small equipment on remote locations. It shall never be used for transmission
to a control room and for panel mounted pneumatic instruments.
The electricity requirements for instruments are mainly 110/240 V AC and/or

24V DC. Full protection shall be provided to guard against surge for short
periods of time. Particular attention shall be paid to the reliability of the
power supplies for DCS systems, the emergency shut-down, fire and gas
detection and protection, telemetry and telecommunications systems.
4.6.6 Fire and Gas Detection
A main feature of a system for fire and gas detection is, that it shall be
separate from any other system. It normally operates from a 24 V DC power
system, with continuous self-checking facilities and its own back-up power
supply.
The equipment is usually purchased as a separate system. For the sake of

overview it is recommended that the alarm and control panel for fire and gas
is separate from the other panels in the control centre. Further details are
given in DEP8O.47.10.11, section 10 and DEP32.30.20.11.
4.6.7 Emergency Shut-Down Systems
The purpose of an emergency shut-down system is to protect human life, the

environment and the capital investment against the consequences of undesir-
able events with a minimum amount of production stoppages.

A prime requirement of such a system is that it must be reliable. Every effort

shall be made at the design and specification stage to ensure that the total
system functions correctly and is free from spurious shut-downs. Therefore
these systems and eventually blowdown, shall be fully independent from all
other systems, except to give information to a data acquisition system.
Also facilities shall be provided to carry out periodic testing of the shut-down
systems with as few sensors as practicable in the override position at the
time of test. All emergency trips shall be active at start-up.
For major production facilities, these systems become complicated and a

reliability analysis may be required. In order to keep them manageable,
several levels of shut-down are introduced: local trip, process train shut-down,
production shut-down, plant shut-down.
Although, for small production facilities it may be cost-effective to have con-

ventional pneumatic shut-down logic, with modern power supply being very
reliable, new facilities are designed with electric relay solid state systems.
Modern ESD systems apply microprocessor based logic, including voting

networks for automatic shut-down (e.g. on 2 out of 3 systems detecting ab-
normal conditions). These systems equipped with diagnostic and self-check
ing procedures can be made as reliable as the relay solid state systems, but
they are then more expensive. Manual shut-down is initiated by actuation of
switches, located in the control centre, preferably on a separate panel and
locally at strategic points.

Instrumentation
4.7 Instruments
In selecting individual measuring instruments, the points mentioned in 4.5 on

experience of staff and support from manufacturers are equally applicable.
Moreover, one should take into account the next two conditions of service:
- The ambient conditions (‘external').

The characteristics of the environment in which the instrument should
operate. This always includes the maximum, minimum and normal pres-
sure, temperature and humidity of the surroundings and hazardous area
classification, but also specific influences should be considered such as
vibrations, probability of mechanical abuse, exposure to direct sunlight,
marine conditions, etc.
- The process conditions (‘internal').

Flow velocities, absolute pressure, temperature and density will always
affect the choice of measuring principle, construction materials and
mounting of the instruments. More specific characteristics to be consid-
ered are: presence of particulates, corrosive/erosive media, presence of
wax or contaminants in general.
In addition to these general considerations, instruments are reviewed below

for the measurement of:
1. Temperature
2. Pressure
3. Level
4. Flow, and
5. Quality.
Many recommendations on installation, mounting, inspection and testing of

instruments are available in DEPs dealing with these instruments (see com-
plete list of DEPs in 4.9).
With respect to the spares (parts) situation, in general the manufacturers'

recommendation should be followed and carefully vetted by an experienced
instrument engineer. Some guidelines are given in DEP 70.10.90.11 (see also
4.8).
In the case of a special measurement for which no acceptable instrument is

available, Central Office should be consulted. It may then be necessary to
seek the assistance of Group Research in modifying a suitable existing in-
strument or in developing a new one, which would be followed by field tests of
prototypes.

4.7.1 Temperature Measurement
The following types of temperature instruments are used:
1. Bimetallic thermometers for local indication only.

2. Filled systems for simple, local control purposes.
3. Thermocouples are used in EP mostly for high temperature applications,
e.g. gas turbines. They should be fitted into thermowells, which are part of
mechanical engineering. The instrument engineer should check on response
time in the case of fast control.
For the relationship between temperature and thermo-electric voltage, see
IEC Publication 584.1.
4. Resistance thermometers are used in EP more often than thermocouples,
because temperatures are mostly not very high and they remain stable
and accurate for many years. They should also be mounted in thermowells.
Normally a 100 ohm platinum type is used. The relationship between

temperature and platinum resistance type sensors is given in IEC Publication
751.
For more details on temperature instruments, see DEP 32.31.00.32, section 9.
4.7.2 Pressure Measurement
For local indication the Bourdon-type and bellows are mostly used.
In the case of electronic signal transfer, various manufacturers supply in-
struments based on deflection of a membrane, its motion being measured
according to different physical principles (capacitive, inductive, etc.).
For more directives on pressure measurements, see DEP 32.31.00.32, sec-
tion 8.
4.7.3 Level Measurement
The usual way of detecting, measuring and controlling liquid levels is by a

displacer or float type of level instrument. For high accuracy tank measure-
ment for custody transfer and fiscal purposes, servo-motor operated level
gauges are preferred.
For special applications such as high pressure (over ANSI class 900), very
viscous liquids, etc., differential-pressure measurement is recommended.
For simple level switching, external float type instruments shall be used.

Instrumentation
Other recommendations are found in DEP 32.31.00.32, section 7. For level

gauge glasses, see DEP 31.38.01.11, Piping - General requirements.
4.7.4 Flow Measurement
It is common EP practice to measure flow of gas and water by orifice meter

runs and to measure crude oil, condensate and oil/water emulsions by po-
sitive displacement (PD) meters, turbine meters or orifice plates. For high
accuracy liquid volume flow measurement, as occurs in custody transfer and
fiscal metering, turbine meters and PD meters are preferred. In the latter
case the turbines are used for low viscosity liquids and sometimes for large
volume flows of dry, high pressure gas, while the PD meters are more suit-
able for higher viscosities.
Straight lengths requirements (see DEP 32.37.00.33) shall be adhered to for

accurate measurements with flow meters.
Flow measurement is covered in more detail in a DEP available especially on

flow (DEP 32.32.10.10).
On flow measurements in particular, frequent new developments take place,

also in Group Research. Advice should be obtained from Central Office, who
will liaise with Research as necessary.
4.7.5 Quality Measurements
In the process industry the term quality measurement often refers to the
measurement of variables other than temperature, pressure, level or flow.
It covers the chemical analyses of components in gases or dissolved gases in

liquids, pH measurement and quantitative determination of concentrations.
Usually, chromatographic and light (including IR and UV) absorption tech-
niques are used. Of particular interest for EP are measurements of water-
in-oil/condensate/gas and oil-in-water for disposal. Nowadays mostly on-
line versions of this equipment are available. For high accuracies laboratory
analyses are recommended. However, much attention should be paid to the
correct method of sampling.
Broadly, conductivity meters, viscosity meters, density meters, hygrometers,

etc. also belong to this category of instruments. For each of the variables
measured there are various instruments in the market based on different
physical principles.
Compared to temperature and pressure measurement, the market for ana-

lytical instruments is continuously developing. This is a typical area where

micro-processors have a large impact on the automation of measurement

techniques. Also semi-conductor sensors are continuously extending the
possibilities.
Therefore, before applying a particular technique, advice should be sought

from Central Office, who continuously evaluate new instruments.
4.8 Maintenance
To-date, no general rules or DEPs have been issued on the technical aspects
of plant instrument maintenance. However, the following general considerati-
ons should be taken into account.
Firstly the design should aim for good maintainability. This has various aspects.
It means that the design should contain as few and as simple instruments as
required to fulfil the specifications. The instruments should be of a well-proven
design and the number of different types should be kept to a minimum. This
may sometimes require an overdesigned instrument, but the benefits of standa-
rdisation come from a reduced number of spares (parts), less work errors and
reduced training requirements.
The installation of instruments should provide easy accessibility for (in-situ)

maintenance, repair or trouble-shooting, leading to a reduction in equipment
down-time. In this respect, self-checking facilities of modern micro-processor
based equipment offer big advantages. Their LED-annunciators can indicate
printed-circuit-board faults as well as identify erroneous field wiring.
Secondly, one should provide for adequate expertise and good facilities to
execute the maintenance work tasks efficiently. In order to achieve this,
maintenance personnel should, apart from general instrument training, be
made familiar with the operating principles of the plant. Therefore they
should be involved in inspection and commissioning of the plant instruments.
Attention should be given to the possibility of getting manufacturers experts

to the site. A well-equipped instrument workshop should be made available
and test equipment should be in good working condition and operational
before commissioning the instruments.
In the interests of training, and possibly as a cross-checking method, it is

recommended that instrument loops and systems be tuned and calibrated by
different maintenance technicians from time to time. This will obviously re-
quire the guidance of competent supervision.

Instrumentation
Sufficient spares and spare parts should be available on site, well on time
(DEP 70.10.90.11).
The Engineering department is normally responsible for defining and order-

ing the spare parts for the first year of operation. Thereafter it is the respons-
ibility of the Maintenance and Materials department. This should be defined
early in the project.
Consideration should be given to the methods and restrictions governing the

procurement of spare parts in the future. Government bodies could restrict
importations, also local suppliers might be favoured.
SPIR (Spare Parts and Interchangeability Record) sheets should be prepared

well in advance. These will assist the responsible parties in ordering the
initial and future spares in order that adequate and optimum levels are
obtained consistent with economic policy related to plant availability versus
cost of spare parts stock. If these SPIRs can not be prepared in house, the
preparation should form part of the design contract.
Thirdly, a maintenance policy has to be agreed. Normally, this is a compromise

between a breakdown and a preventive maintenance policy, dependent on the
criticality of the instrumented system or process. The main aim should be towards
optimum maintenance effort and costs consistent with the desired system reliabi-
lity and safe working practices.
Emergency shut-down systems should be inspected and operated on a routine

basis. The instrumentation manufacturer’s instructions regarding maintenance
and calibration should be followed closest. During equipment overhauls all
instruments should be checked and calibrated before recommissioning the
equipment.
Finally, attention should be given to a good, well-disciplined administrative

support unit. It should provide:
- initiation of activities by maintenance personnel
- history of performance of all instruments
- retrieval facilities for performance analysis.
A small computer system can be of immense value in this administrative

task, provided sufficient effort is devoted to regularly updating the records.

4.9 Relevant DEPs

(a) DEPs on Instrument Engineering

Instrumentation
(b) Other DEPs related to Instrument Engineering

Telecommunications
Chapter 5
TELECOMMUNICATIONS
CONTENTS
5.1 Introduction 5-2

5.2 General Requirements 5-2
5.3 Telecommunications Systems in the Production Environment 5-3
5.4 Telecommunications Organisation 5-4
5.5 Telecommunications System Management 5-5
5.6 Radio Equipment and Radio Frequency Hazards 5-7
5.7 Summary 5-9
5.8 Abbreviations Used in Telecommunications 5-9
FIGURES
Figure 5-1 Integrated communications system for production areas 5-6

Figure 5-2 Working documents or recommended practices – the
use of telecommunications/radio equipment in
hazardous areas 5-8

Telecommunications
5 TELECOMMUNICATIONS
5.1 Introduction
The objective of any telecommunications system is to provide the resources

by which information may be transferred between two or more locations and
it is the task of the telecommunications department within the Operating
Company to provide cost-effective solutions which will satisfy the end user's
requirements for the exchange of voice, data, telex, facsimile, video, SCADA
information and messages.
It is recognised that telecommunications facilities are essential for managing

the company and for the safe, effective operation of the various facilities, and
that these services must be inherently reliable and available at all times.
The main functions performed within a telecommunications department are

the conceptual design, planning, construction and maintenance/operation of
the telephone, data transmission, mobile radio and other associated networks
and services.
5.2 General Requirements
The user's interface to the telecoms network is either the telephone instru-
ment or terminals for telex, facsimile and data. These are linked to other,
distant counterparts via a network of cables, radio links and strategically
located switches or exchanges to provide a functionally cost-effective infras-
tructure.
However, it is often the local legislation, difficult geographical conditions, or

the financial implications which limit the provision of service rather than the
technology.
There are two types of operations which telecommunications will be required

to support in the EP environment, namely the Single String Exploration Ven-
ture and the normal Production Operations.
Single String Exploration Ventures generally are small operations which

have a life of possibly a few years. Generally, there is unlikely to be a need for
extensive telecommunications facilities although that provided will be required
to support operations over a very wide geographical area, often for relatively
short periods of time.
The telecommunications requirements to support Single String Ventures are

detailed in report EP 91-0725.
5-2 Confidential—Property and Copyright:SIPM, 1991

5.3 Telecommunications Systems in the Production Environment
The facilities provided by a typical Opco telecommunications organisation

will provide services to:
- Office/base
- Residences
- Production field operations
- Exploration and well work-overs
- Transport for material and personnel incl. . - boats
- vehicles
- aircraft incl. helicopters.
Besides the normal production-related activities, other major telecommuni-

cations users are likely to include:
- Safety, fire, security & medical services
- Construction contractors
- Emergency co-ordination and control.
The services which the telecommunications organisation may be asked to

provide could include:
- Telephone and telex with connections to both the public and the Shell
private networks
- Electronic mail (E-Mail) service
- Data terminals, PC's and workstations
- Radio paging
- Rural telephone service for drilling camps
- Mobile radio systems
- Marine radio and radar for both ports and vessels
- Aviation HF/VHF (for air to ground communications)
- Radio navigation beacons to support aviations and marine operations.
Depending upon the country in which the Opco is established, these services
may be directly supplied by third parties such as the local PTT, or in the
absence of a reliable public infrastructure, constructed by the telecommunica-
tion organisation themselves. In the latter case, the networks will be constru-
cted from a wide variety of technologies which are likely to include:
- Microwave, troposcatter and/or satellite transmission systems for linking

centres of activity
- Local cable networks
- Telephone exchanges (something known as PABX's)
- Data switching equipment
- Cellular and trunked mobile radios

Telecommunications
- HF/SSB radio systems
- PAKTEL dataradio system
- VHF and UHF mobile and hand-held radio equipment.
A simplified integrated communications system for production areas is

shown in Fig. 5-1.
5.4 Telecommunications Organisation
With the increase in the importance and sophistication of computing and the
heavy reliance on telecommunications to deliver computing service to the
end user, the telecommunications and computing functions are considered
as a single Information Technology (IT) department. Different Operating
Companies will have their own organisation structure - there is no univer-
sally correct organigram. However, the following points should be borne in
mind:
- Information Planning should take account of all the information require-

ments and decide the best solution or mix of solutions: voice, facsimile,
telex, electronic mail, or computer data or a combination of all those
means.
- Project work, building the telecommunications infrastructure, has to be

integrated with major oil and gas projects, particularly on offshore plat-
forms. New offices and camps also require the timely provision of tele-
phones, local area networks, etc.
- Maintenance must be structured so that the end user has one contact for
reporting faults and there should be single point responsibility for restora-
tion and repair.
Other aspects which are important are:

- consideration of the interface to the local PTT administration
- consideration of security for both voice and data
- emergency/catastrophe planning, including links with the outside
world.
It is essential that all the work is appropriately planned, organised and con-
trolled to provide optimum service at minimum cost.
For companies in the start-up phase which are too small to justify a full-time
IT staff member, someone should be nominated as the IT focal point to deal
with the local telecommunications authority and telecommunications con-
tractors. As the operation grows, the telecommunications and computing
planning infrastructure should evolve together.

5.5 Telecommunications System Management
In common with other operational environments, a number of facilities are

essential to ensure the effective operation of the telecommunications sys-
tem.
These will include:
- Operational procedures and guidelines

- Technical facilities for network control
- Record maintenance
- System information covering the equipment, network and operation
- Spares, spare units, stock control and test gear
- Facilities such as offices, workshops and transport.
Below are outlined typical aspects of some of the more pertinent topics.
Operational Procedures and Guidelines

The provision of formal procedures and guidelines permits all personnel in
the organisation to standardise the methods and techniques which are to be
applied for the operation and maintenance of the network.
The procedures and guidelines must cover four basic areas:

- Breakdown maintenance
- Preventative maintenance
- Construction standards
- Administrative control
- Financial control
Technical Facilities for Network Control

The following technical tools are available for the management and super-
vision of telecommunications networks:
- Availability records of equipment
- Performance records of circuits and facilities
- Administration of facilities
- Administration of work
- Cost control of the telecommunications operation
- Engineering tools (for planning etc.)
Record Maintenance
As in any other function, it is essential that records be maintained. These
should include:
- Cable network administration
- Frequency allocation records
- Radio channel multiplex records

Telecommunications

- Subscriber apparatus records

- Equipment inventories
- Spares inventories, ordering and stock control
- Fault records and job tickets
All of the above functions should be computer-based for ease of manipula-

tion.
5.6 Radio Equipment and Radio Frequency Hazards
Radio transmitters emit radiation in the lower part of the electromagnetic

spectrum, normally in the range 100 kHz and 30,000 MHz. One of the attri-
butes of such energy is its ability to remotely induce Radio Frequency (RF)
currents in both animate and inanimate objects which in turn may cause
undesirable effects, even at considerable distances from the transmitter.
It is widely recognised that the detonators used in conjunction with high

explosives are particularly sensitive to radio frequency energy, and normally
radio operations are curtailed during blasting or the perforation of well cas-
ings.
Part of the control function of most modern radio systems requires that each
individual radio unit identifies itself from time to time. This involves the
transmitter ‘keying' and sending its identification number at predetermined
intervals. This will occur automatically without the need for the user to man-
ually key the transmitter, and thus if such equipments are to be used in areas,
say, where explosives are being handled, it is essential that they are powered
off at the main on-off switch. Simply leaving the radio in standby mode is insu-
fficient.
RF fields when induced in the human body result in localised heating. Even
relatively low field strengths can permanently damage the tissue of the eyes
and reproductive organs causing blindness and sterility. High RF currents
induced in the human body will cause very painful, lingering burns.
There are a number of on-going studies and investigations to determine the

safe levels of exposure to RF fields. Some working documents for various
types of operation have been produced or are currently in preparation.
Finally, it should be remembered that in common with many other pieces of

electrical equipment, there can be sufficient energy within the electronic
circuitry of hand-held radios to cause ignition of flammable atmospheres.
Only those radios which have been certified as being intrinsically safe should
be used in environments where flammable atmospheres may be present.
Figure 5-2 lists the various hazards together with a list of references.

Telecommunications
Figure 5-2 Working documents or recommended practices – the use of

telecommunications/radio equipment in hazardous areas

5.7 Summary
Normally the telecommunications organisation will provide telephone, telex

and data networks to offices, industrial areas, residences and the majority of
onshore and offshore production operations.
The ability to extend such services to mobile units such as drilling rigs,
construction barges, marine vessels and road transport is technically quite
feasible although the economic justifications and legislative hurdles may
preclude their implementation.
Satellite-based communications systems such as Inmarsat are the preferred

medium for providing service to field locations out of range of the normal
subscriber or mobile radio services, but occasionally, it may be necessary to
fall back upon HF/SSB systems although these are not now recommended
due to the complexity of operation and poor quality of the transmissions.
The management and operation of a large telecommunications system re-

quires considerable skill and dedication at all levels. Diligent maintenance of
the various record databases is essential in order that faults can be readily
identified and repaired. Performance indicators enable the Head of Tele-
coms to gauge the effectiveness of his organisation and, if correctly pre-
sented, will enable deterioration in service to be identified and corrected
before it is noticed by the users.
5.8 Abbreviations Used in Telecommunications
HF - High Frequency (3-30 MHz) - World-wide transmission

capability
VHF - Very High Frequency (30-300 MHz) - Normally line of
sight only
UHF - Ultra High Frequency (300~1000 MHz) - Normally line
of sight only
MICROWAVE - Any signal in excess of 1000 MHz
TROPOSCATTER - A technique for extending UHF and microwave signals
beyond normal line of sight ranges
PABX - Private Automatic Branch Exchange
PTT - Post Telephone Telegraph - an expression used for the
national controlling authority
IT - Information Technology
PC - Personal Computer
SSB - Single Side Band - a means of transmitting voice over a
radio link, normally used with HF transmitters

Telecommunications
PAKTEL - A proprietary system for transmitting data (normally

telex) over a radio link
RF - Radio Frequency
SCADA - Supervisory Control and Data Acquisition - a system
used to monitor and control remote unmanned plant and
facilities

Reliability and Availability Assessment
Chapter 6
RELIABILITY AND AVAILABILITY

ASSESSMENT
CONTENTS
6.1 Introduction......................................................................................................................................... 6-2

6.2 Maintainability.................................................................................................................................... 6-3
6.3 Data Requirements............................................................................................................................ 6-3
6.4 Methodology ...................................................................................................................................... 6-4
6.5 System Modelling Tools .................................................................................................................... 6-5
6.6 Definitions.......................................................................................................................................... 6-5

6 RELIABILITY AND AVAILABILITY ASSESSMENT
6.1 Introduction
In a number of engineering fields, such as aerospace engineering, military

equipment and nuclear engineering, tools have been developed to deter-
mine the performance of complicated systems. With these tools it is possible
to determine the availability of a system and to predict and understand the
effects of failures.
System availability assessment can be carried out for proposed new facilities
as soon as outline flowschemes are available during concept selection, as
well as during the operational phases of the facilities lifecycle. It takes into
account individual equipment reliability data, system design configuration,
equipment design or operating capacities and expected outages for repair
and scheduled maintenance.
Typically system availability assessment allows:

- Identification of the consequences which engineering decisions have on
the expected output from production systems. Development options hav-
ing various equipment types and configurations can be compared through-
out the intended lifecycle.
- Identification of the consequences which operating decisions have on the
expected output from production systems. Operations input can be focused
in the area of maintenance strategy, such as maintenance routines (pre-
ventive and repair), logistics and resources, which bear a considerable
impact on the ability to satisfy contractual obligations.
The reasoning behind failure tolerance of redundant systems is the ability
to make-up lost production following failure and that adequate maintaina-
bility is incorporated in the design to enable equipment or sections of the
entire system to be isolated for maintenance with respect to safety and
production requirements.
- Identification of the consequences which variations of product contract
delivery terms have on the expected output from the production facili-
ties.
- Input to economic analysis since lifecycle cost estimates can be provided
for varying equipment configurations and operating strategies in combi-
nation with changes in system supply and associated revenues expected
from operating the plant.

6.2 Maintainability
The main objective is to ensure sustained technical integrity of the produc-

tion facilities. Statutory inspections, routine testing, overhauls and work
arising from condition monitoring are typical functions in a preventive main-
tenance programme. Some flexibility can be permitted in maintenance plan-
ning to defer a preventive task to some opportunity which may occur in the
future or to advance some requirements to take advantage of a forced out-
age.
Key issue will nevertheless be to minimise maintenance task duration and

hence minimise production unavailability. Attention should be focused on:
- consider performing maintenance tasks concurrently rather than conse-
cutively;
- optimise spares policy;
- plan ahead by making use of opportunities that occur during periods when
the demand on the plant or system is lower for external reasons;
- consider opportunity for replacement maintenance rather than in-situ re-
pair. This will in general reduce outage times if the appropriate spares and
rapid change- out facilities are provided.
The availability assessment tools enable to consider the above aspects and
indicate where control can be exercised.
6.3 Data Requirements
Failure and recovery time data are required at system, subsystem and equip-
ment item level, depending on the required detail of the analysis to be carried
out. Their validity is crucial to the result of the study, although some com-
promise on data accuracy is acceptable when carrying out comparative
studies for different configurations of the same types of equipment. Com-
pensation for data inaccuracies should be made by sensitivity analysis. It is
recommended that the input of experienced local operators and mainten-
ance staff be sought when selecting data. Their intuition can be a guide to
sensitivity analysis.
Data can be retrieved from open literature (i.e. OREDA), from maintenance
records and from previous reliability studies.

The criticality of accurate data in maintenance records and the impact they
may have on the results cannot be over-emphasised. Areas which need to be
addressed in data collection are:
- specification of boundaries and conditions to which specified data relate
to;
- identification of cause and effect of recorded failure;
- consistent definition of failure and repair/recovery; for example, recovery
time comprises:
* time to detect failure
* time to isolate equipment
* time to analyse failure (diagnose)
* plan repair activities
* procure parts and materials
* mobilise resources
* repair
* testing
* time before equipment is reinstated (operable);
- any database should be easily installable and operated, should be flexible
with respect to update and expansion, should provide quality controls
(validity checking) and provide interpretation of results (consistency of
data formats for availability analysis).
6.4 Methodology
When carrying out availability assessment studies, the following method-

ology is recommended:
- Define objectives, boundary and assumptions.

- Gather P&ID's, PFD, shutdown logic, etc.
- Construct system logic network.
Just like a PFD and P&ID describe the process in terms of physical para-
meters (temperature, flow rate, pressure, etc.), a logic block diagram des-
cribes the state of the process, given the states of the individual compon-
ents or equipment comprising the system. The logic model is built up by
arranging the process equipment in series or parallel depending on their
effect on lost production.
- For each equipment item define capacity, failure rate, recovery time, shut-
down logic (equipment priority).
- For the overall system define the contract period, the supply and demand
profile and the production mode.

- Run the analysis.
- Evaluate option until objectives are met.
- Document conclusions.
- Make recommendations.
Most applications of availability analysis concentrate on the simplification
of production facilities, requiring an evaluation of the probabilistic perfor-
mance of various equipment and system configurations against contracted
quantities/requirements. It is important that in all these cases a number of
conditions need to be satisfied prior to the analysis. These conditions can,
however, be optimised as part of the study terms of reference.
- Envisaged operating strategy must be agreed and detailed and will be
assessed against required production targets with minimum facilities.
- The economic impact of deferred production should be assessed against
contractual constraints.
- The possibility to make-up from other production facilities should be raised
in early negotiations.
6.5 System Modelling Tools
Computer-based models were, and are still being, developed to assist in the
analysis of system performance calculations and allow the critical areas to
be identified and the effects of proposed solutions to be qualified. The tools
range from simply applied to sophisticated programs using a variety of pro-
babilistic techniques, analytical and (Monte Carlo) simulation techniques.
They have their application in oil and gas process systems optimisation
studies (i.e. minimum topside facilities, satellite development), gas flaring
consent and gas sales optimisation studies.
6.6 Definitions
This section presents some of the major terminology used in availability

assessment.
Availability The portion of period in an available state or the probability of being in an

available state, with available meaning in use or ready for immediate
use.
AV = [Total Time - (Sched. + B/D) Maintenance Time/Total time

= (Runtime + Standby time)/Total Time

In studies the following approximation is usually applied, which assumes

that there are two states, being up and running or under repair. It ignores
standby time and scheduled maintenance time. The availability of the equip-
ment item then reduces to :
AV = MTBF/(MTBF + MTTR) MTBF = Mean Time Between Failure

MTTR = Mean Time To Recover
Effectiveness The ratio of mean output over specified contract period and the nominated
(performance) output over the same contract period.
Failure rate The expected number of failures per calendar year or per recorded total
running time of the equipment item concerned. The reciprocal of failure rate
measured against total running time is equal to MTBF (Mean Time Between
Failure).
MTTR Mean Time To Recover. It is the total time since detection of failure to
bringing the item to an available state.
Capacity Throughput rate measured as a volumetric quantity per unit of time, e.g. in
3
thousands m /day. Note that the capacity of each equipment item can also
be expressed as a fraction of the system demand. In fact any unit can be
adopted, if it is consistently used throughout the system.
Overcapacity The installed total field capacity minus the nominated capacity (= system
demand).
Contract period Period over which the nominated volumetric quantity should be deliv-
ered.

Corrosion and Materials Engineering in Production Operations
Chapter 7
CORROSION AND MATERIALS

ENGINEERING IN PRODUCTION
OPERATIONS
CONTENTS
Preface................................................................................................................................... 7-5
7.1 Introduction ............................................................................................................................ 7-5
7.2 Mechanism of Wet Corrosion ...............................................................................................7-6
7.2.1 Anodic and Cathodic Reactions.......................................................................................... 7-6

7.2.2 Cathodic Reactions............................................................................................................ 7-8
7.2.2.1 Acid Corrosion (Charge Transfer Control) ....................................................................... 7-8
7.2.2.2 Oxygen Corrosion (Mass Transfer Control)..................................................................... 7-9
7.2.3. Concept of Potential.......................................................................................................... 7-9
7.2.4 Galvanic Series................................................................................................................ 7-11
7.2.5 Galvanic Corrosion .......................................................................................................... 7-12
7.2.6 Differential Concentration Cells ....................................................................................... 7-13
7.2.7 Passivity ............................................................................................................................ 7-14
7.3 Forms of Corrosion ............................................................................................................. 7-15
7.3.1 CO2 Corrosion................................................................................................................... 7-15

7.3.2 Erosion-Corrosion .............................................................................................................. 7-18
7.3.3 ‘Low flow' Corrosion .......................................................................................................... 7-19
7.3.4 ‘Top of line' Corrosion ...................................................................................................... 7-20
7.3.5 H2S Corrosion and Hydrogen Embrittlement ................................................................... 7-21
7.3.5.1 Sulphide Stress Corrosion Cracking.............................................................................. 7-22
7.3.5.2 Hydrogen Induced Cracking ......................................................................................... 7-23
7.3.6 Oxygen Corrosion ............................................................................................................. 7-24
7.3.7 Stress Corrosion Cracking in General ............................................................................. 7-25
7.3.8 Corrosion Fatigue ............................................................................................................. 7-26

7.4 Chemical Corrosion Inhibitors and Additives ...................................................................... 7-27
7.4.1 Types of Inhibitors........................................................................................................... 7-27

7.4.2 Adsorption Inhibitors ....................................................................................................... 7-27
7.4.2.1 Method of application..................................................................................................... 7-28
7.4.2.2 Limitations and testing ................................................................................................... 7-28
7.4.3 Oxygen Scavengers.......................................................................................................... 7-29
7.4.4 Biocides ............................................................................................................................ 7-29
7.5 Steels and Alloys ................................................................................................................ 7-30
7.5.1 Carbon-manganese Steels ............................................................................................... 7-30

7.5.1.1 Composition and Structure ............................................................................................ 7-30
7.5.1.2 Heat Treatment .............................................................................................................. 7-30
7.5.1.3 Welding .......................................................................................................................... 7-31
7.5.2 Stainless steels ................................................................................................................ 7-33
7.5.2.1 Ferritic/martensitic Stainless Steels.............................................................................. 7-33
7.5.2.2 Austenitic Stainless Steels............................................................................................. 7-34
7.5.2.3 Duplex Stainless Steels ................................................................................................. 7-35
7.6 Corrosion of Stainless Steels.............................................................................................. 7-36
7.6.1 Crevice Corrosion ............................................................................................................. 7-36

7.6.2 Pitting Corrosion ............................................................................................................. 7-37
7.6.3 Chloride Stress Corrosion Cracking ................................................................................. 7-38
7.6.4 Sensitisation...................................................................................................................... 7-38
7.7 Cathodic Protection............................................................................................................. 7-39
7.7.1 Control of External Corrosion ........................................................................................... 7-39

7.7.2 Control of Internal Corrosion............................................................................................. 7-41
7.8 Non-metallic Materials ........................................................................................................ 7-42
7.8.1 Plastics.............................................................................................................................. 7-42

7.8.1.1 Thermo plastics.............................................................................................................. 7-43
7.8.1.1.1 Properties..................................................................................................................... 7-43
7.8.1.1.2 Thermoplastic Pipes .................................................................................................... 7-43
7.8.1.1.3 Polyethylene Liners for Carbon Steel Pipe .................................................................. 7-44
7.8.1.2 Thermosets .................................................................................................................... 7-45
7.8.2 Plastics Composites (reinforced/filled/foamed) ................................................................ 7-45
7.8.2.1 Flexible Steel/Polymer Pipe........................................................................................... 7-45

7.8.2.2 Reinforced Plastics / GRP .............................................................................................7-47

7.8.2.3 Filled/Foamed Plastics................................................................................................... 7-49
7.8.2.4 Fire Protective Coatings................................................................................................. 7-50
7.8.3 Rubbers and elastomers.................................................................................................... 7-51
7.8.4. Paints and coatings........................................................................................................... 7-52
7.8.4.1 Paints ............................................................................................................................. 7-53
7.8.4.2 Pipeline External Coatings............................................................................................ 7-54
7.8.4.3 Internal Coatings for Pipelines and Downhole Tubulars................................................ 7-55
7.8.5 Cements........................................................................................................................... 7-55
7.9 Corrosion Monitoring............................................................................................................ 7-56
7.9.1 Corrosion Coupons ........................................................................................................... 7-56

7.9.2 Electrical Resistance Probes ............................................................................................7-56
7.9.3 Polarisation Resistance Probes ...................................................................................... 7-57
7.9.4 Other Techniques ............................................................................................................. 7-58
7.9.5 General Remarks.............................................................................................................. 7-58
TABLES
Table 7-1 Some commonly used reference electrodes 7-10

Table 7-2 Examples of metal-liquid combinations which give
stress corrosion cracking 7-25
Table 7-3 Name, approximate composition and typical application
of some corrosion resistant alloys 7-33
Table 7-4 Typical values of electrochemical consumption rates of
some anode materials 7-41
Table 7-1 Maximum temperatures for pipeline coatings 7-54

FIGURES
Figure 7-1 Mechanism of electrochemical corrosion 7-7

Figure 7-2 Example of the measurement of the potential of a
metal 7-10
Figure 7-3 Practical Galvanic Series for neutral soils and water,
measured with respect to a copper/coppersulphate
reference electrode 7-11
Figure 7-4 Bimetallic corrosion cell 7-12
Figure 7-5 Differential oxygen concentration cell 7-13
Figure 7-6 Under-deposit corrosion 7-13
Figure 7-7 Effect of temperature on CO2 corrosion 7-16
Figure 7-8 CO2 Corrosion Nomogram 7-17
Figure 7-9 Mechanism of `top of line' corrosion 7-20
Figure 7-10 Example of corrosion rate reduction from the
presence of glycol 7-20
Figure 7-11 Critical H2S concentrations according to NACE 7-22
Figure 7-12 Mechanism of Hydrogen Induced Cracking 7-23
Figure 7-13 Influence of a corrosive environment on fatigue 7-26
Figure 7-14 Continuous downhole inhibitor injection 7-27
Figure 7-15 Crevice corrosion at riveted joint 7-37
Figure 7-16 Mechanism of pitting corrosion in stainless steels 7-37
Figure 7-17 Impressed current cathodic protection of a buried
pipeline 7-40
Figure 7-18 Principle of resistance probe for corrosion rate
measurements 7-56
Figure 7-19 Schematic view of retriever systems for on-stream
installation and retrieval of corrosion coupons or
probes 7-57
Figure 7-20 Polarisation resistance probe for corrosion rate
measurements 7-57

Preface
‘Technical integrity of a facility is achieved when, under specified operating

conditions, there is no foreseeable risk of failure endangering safety of per-
sonnel, environment or asset value'.
Materials and corrosion engineering has an important role to play in main-

taining the technical integrity of Group facilities. The achievement of mini-
mum life cycle costs for a production system, built to achieve the required
technical integrity, is often dependent on the use of materials which can be
expected to corrode during normal operation and may well be planned to
require replacement during the life of a project.
A broad awareness amongst all staff of the potential threats that materials
and corrosion problems can pose to the business is essential if materials and
corrosion matters are to be adequately managed.
Most operating companies have materials and corrosion engineers on their
staff who can assist with the management of corrosion. Additional advice can
be obtained from the Materials/Corrosion department in SIPM-EPD, and
through them, from Group laboratories.
7.1 Introduction
Metals can react with various chemicals, often dissolved in water, to form
metal compounds. The resulting loss of metal is called corrosion. Corrosive
attack can be uniform or localized in the form of, for example, pitting or
cracking.
Most of the corrosion problems in production are associated with the pre-
sence of liquid water. `Wet' corrosion of equipment can occur due to expo-
sure to produced fluids or due to external factors such as exposure to the
atmosphere, the sea or soil.
The corrosivity of water is dictated by what is dissolved in it. The main
corrosive agents in production are carbon dioxide (CO2), hydrogen sulphide
(H2S) and oxygen (O2).
The most common material of construction for equipment used in production

is carbon steel which is relatively cheap and often suitable in terms of
strength, ductility and weldability. However, it can corrode quite easily.

In many cases the speed with which corrosion occurs can be expressed as a
corrosion rate. The corrosion rate of any construction material which can be
tolerated will be dependent on the design life foreseen in relation to the
corrosion allowance' applied. An economic assessment of the life cycle
costs of alternative materials in the same application could show that, given
the corrosion rate predicted for carbon steel, a more corrosion resistant (and
often more expensive) material might be more attractive.
The Sections which follow describe and illustrate the most important sources
of materials and corrosion concerns in an Exploration and Production envi-
ronment. In the main, the text describes the corrosion behaviour of carbon
steels though the more common alternatives amongst corrosion-resistant
alloys and non-metallic materials and their associated behaviours are also
introduced. Some of the options available to monitor and to reduce corrosion
are described and appropriate references to further reading and Group Stan-
dards are provided in the respective Sections.
7.2 Mechanism of Wet Corrosion
7.2.1 Anodic and Cathodic Reactions
The corrosion of steel releases positively charged iron ions into the water
phase to which it is exposed. In other words, a corroding metal atom leaves a
number of electrons (indicated with e) behind in the metal, e.g.:
++ -
Fe ---------> Fe + 2e (1)
metal water metal
In this way the metal would become more and more negatively charged as
corrosion proceeds, while the water would become more positively charged.
This separation of charge cannot continue indefinitely, and the reaction
quickly comes to a halt unless the excess of negative charge in the metal is
reduced, e.g.
+ -
2 H + 2e ------> H2 (2)
water metal water

Reaction 1 is called the anodic reaction (or the corrosion reaction), reaction 2
the cathodic reaction. The cathodic reaction consumes the excess electrons
left behind in the metal by the anodic reaction. Both reactions occur side by
side on the metal's surface, which can be thought of as being covered with
tiny electrical cells (Fig. 7-1), each consisting of an anode and a cathode, and
electrically shorted by the water. To this end the water has to be an elec-
trolyte i.e. it has to be electrically conductive. Electrolytes are formed when
ionic salts, such as sodium chloride, are dissolved in water.
A completely homogeneous distribution of anodes and cathodes will lead to
homogeneous loss of metal. It is possible that an area of the metal has
relatively more anodes, which has to be offset by an adjacent area having
more cathodes. This leads to uneven corrosion, referred to as localised
corrosion or pitting corrosion, which is often observed in practice.

7.2.2 Cathodic Reactions
There are two basic cathodic reactions: one follows from the presence of
acids, the other from the presence of dissolved oxygen in the water. They
differ fundamentally in the mechanism of the corrosion reaction. Acid cor-
rosion rates are controlled by the reaction rate constant of the cathodic
reaction (charge transfer), while oxygen corrosion rates are controlled by the
mass transport of dissolved oxygen to the metal's surface. Other cathodic
reactions with other chemical species are possible, but are relatively rare in
oil and gas production.
7.2.2.1 Acid Corrosion (Charge Transfer Control)
Acids in water can dissociate e.g.

+ -
HA ------ H + A (3)
(HA represents the formula of any acid)

and the cathodic reaction is then:
+ -
2H + 2e -----> 2H followed by 2 H -----> H2 (4)
(formation of hydrogen gas bubbles)
which consumes the electrons left in the metal by the anodic reaction:
++ -
Fe ----> Fe + 2e (5)
The transport of electrons through the water and the metal causes the flow of
an electrical current. The current carried by the cathodic reaction is exactly
equal to the current carried by the anodic reaction. The strength of this
current is directly related to the corrosion rate, and is called the corrosion
current. The overall ionic reaction is obtained by adding reactions 4 and
5:
+ ++
Fe ± 2H ----------> Fe + H2 (6)
The corrosion rate is governed by the rate of transfer of electrons at the metal
+
surface (charge transfer control) and not by the rate of supply of H . A higher
+ +
acidity (more H ), will increase the corrosion rate. For a given H concentrat-
ion, the resulting corrosion rate depends mainly on temperature. Corrosion
rates often increase with the velocity of the electrolyte flowing past the metal,
but with acid corrosion this dependence is very weak.

7.2.2.2 Oxygen Corrosion (Mass Transfer Control)
Oxygen (from air) dissolves in water in a concentration of about 6 weight

parts per million parts of water. It can cause appreciable corrosion rates. The
cathodic reaction can be formulated as:
- -
½ O2+ H2O+ 2e ----->2 OH (7)
In contrast with the hydrogen evolution reaction (4) the reaction rate is not
governed by the rate of electron transfer, but is controlled by the rate of
transport of O2 to the surface of the corroding steel, and is consequently very
dependent on the rate of O2 diffusion and on the flow velocity and turbulence
of the water (mass transfer control). The anodic reaction is the same as
before, and the overall reaction is, for this case:
Fe + H2O + ½ O2 ------> Fe(OH) 2 (8)
which can react further with oxygen to form a mixture of ferrous and ferric
hydroxides and oxides which is normally referred to as rust.
7.2.3. Concept of Potential
When a metal corrodes in the manner described above, electrical energy is

expended by the separation of charge in the anodic reaction. This energy can
be expressed as charge x voltage. The latter voltage is called the (electro-
chemical) potential of the anodic reaction. Similarly, the cathodic reaction
also has a potential associated with it. The difference between these two
potentials is the open voltage of the corrosion cells in Fig. 7-1. The larger it is
(if in the right direction), the faster both reactions will proceed when they are
electrically shorted by an electrolyte. The potential of the metal as a whole
will be between that of the anodic and that of the cathodic reaction. It may be
conveniently seen as a voltage between metal and electrolyte. However, it
cannot be measured directly. Only differences in electrochemical potential
can be measured, for example as a voltage between two different metals in
the same electrolyte, or between two pieces of the same metal in different
electrolytes. In the latter case the electrolytes have to make contact with each
other by some means which allows ionic transport, but prevents mixing e.g. a
porous barrier. By taking one of these metal/electrolyte combinations to be
the same every time (reference electrodes, see Table 7-1), these voltages
can be compared (see Fig. 7-2).

Table 7-1 Some commonly used reference electrodes*
Reference Electrode Cell Potential in mV

w.r.t. standard
o
H2 electrode 25 C
Saturated Calomel Hg/Hg2CI2/KCI(saturated) 242
Silver Chloride Ag/AgCI/seawater 250
Copper Sulphate Cu/CuSO4 (saturated) 316
* In this Table, the commonly used practical reference electrodes are themselves
compared to a standard laboratory reference electrode which involves a platinum
electrode covered with hydrogen bubbles immersed in a 1N solution of strong
acid.
Potentials reported for corroding systems are always with respect to the
arbitrary zero potential of a reference electrode.

7.2.4 Galvanic Series
The tendency of a metal to corrode depends on the potential of the anodic

reaction and that of the cathodic reaction. By comparing potentials of differ-
ent metals in an electrolyte with the same cathodic reaction, these tenden-
cies can be compared.
Fig. 7-3 gives an example of this for corrosion in oxygen containing water or
wet soil in the presence of oxygen. This is referred to as a practical galvanic
series. Very negative potentials indicate very reactive metals, which can
corrode at high rates e.g. magnesium. These metals are called base metals.
Less negative or positive potentials refer to more corrosion resistant metals.
They are called noble metals e.g. platinum. One metal can be nobler than
another.

7.2.5 Galvanic Corrosion
The fact that different metals (and alloys) have different potentials when
exposed to the same electrolyte implies that current will flow when they are
in metallic contact with each other (Fig. 7-4). This current will increase the
anodic reaction rate (corrosion rate) for one metal, while for the other metal
the cathodic reaction rate is increased. For example, in seawater steel has a
potential of -0.7 V, while copper is at -0.2 V, giving a voltage between them of
-0.2 - (-0.7) = 0.5 V. When these materials are in metallic contact, the cor-
rosion rate of the more base metal, steel, is enhanced. On the more noble
metal, copper, the cathodic reaction is increased, while the anodic reaction
(dissolution of copper) is decreased. The increase in corrosion current on the
steel (Eq. 1) is exactly equal to the increase of the cathodic reaction current
on the copper (predominantly oxygen reduction, Eq. 7).
For a small area of steel attached to a large area of copper, this increase
leads to a very high corrosion current per unit of area, and dramatic damage
of the steel can be the result. Especially in the presence of oxygen, the
simultaneous use of different metals can cause problems, which are referred
to as bimetallic or galvanic corrosion.
Sometimes, bimetallic corrosion is made to occur on purpose. In these cases
zinc (or aluminium or magnesium) metal is electrically connected to steel,
which reduces the corrosion rate of the steel, at the cost of increased cor-
rosion of the zinc, which is sacrificed to protect the steel. This is called
cathodic protection by means of sacrificial anodes (see also Section 7.7).

7.2.6 Differential Concentration Cells
Instead of having two different metals in the same environment, it is also

possible to have two pieces of the same metal e.g. steel, each in a different
environment. The most common example is where these two environments
are electrolytically connected but differ in oxygen concentration (differential
aeration, Fig. 7-5).
Since the two pieces of steel would be corroding at different rates, they also
would have different potentials, the higher oxygen concentration leading to a
more positive potential.
When these pieces are in metallic contact, current will flow such as to in-
crease the cathodic reaction rate on the high oxygen side, but on the low
oxygen side the corrosion current will be increased. This is particularly
important where the area of the cathode is large compared to the anode. In
practice this occurs under deposits like sand, etc. in vessels and pipelines
(Fig. 7-6): the oxygen at the steel under the deposit will be depleted by
corrosion, while the oxygen in the bulk of the electrolyte will remain constant.
The result is an increased corrosion under the deposit, while the rest of the

steel is less attacked. The corrosion is concentrated at the edge of the deposit
where the electrical resistance to corrosion current flow is least.
7.2.7 Passivity
Some metals are much less susceptible to corrosion than would be expected
from their theoretical position in a galvanic series like in Fig. 7-3. Examples
are chromium, stainless steels (i.e. chromium containing alloys) and titan-
ium. This is caused by a corrosion product film on the metal's surface (some-
times called the passive layer), which protects the metal against further
corrosion and reduces the corrosion rate by several orders of magnitude. It
can be very thin (thickness of a few molecules) and is often invisible. A metal
in this condition is said to be passive.
When the metal is in the passive state, the potential is that of the passive
layer, which is nobler than that of the metal in the active state. If the passive
layer is damaged, for whatever reason, galvanic action between active and
passive material can lead to high rates of localised corrosion.
The distinction between passivity and some high degree of protection from a
corrosion product film is not exact. For example, steel covered by rust is
normally not called passive, although the rust layer might have reduced the
corrosion rate considerably by the formation of a relatively thick physical
barrier to further attack.
Not all corrosion products form a protective layer. When a metal loses pas-
sivity , e.g. by exposure to a ‘wrong' environment, it can revert to the active
state and become very susceptible to corrosion.
For many metals specific environments exist in which they become passive.
The stainless steels are a special case since they are passive in a large
range of practical environments. They are discussed separately in Section
7.5.2 and in Section 7.6.
Further reading:
- Report EP 87-1780 Principles of Materials Engineering and Corrosion Con-
trol in E&P operations, Chapter VI, parts 2,3.
- NACE Corrosion Engineer's Reference Book, edited by R.S.Treseder, 2nd
printing, Houston 1980.
- Introductory Textbook: H.H.Uhlig, Corrosion and Corrosion Control, John
Wiley & Sons Inc.

7.3 Forms of Corrosion
In the following Sections, forms of corrosion most commonly encountered in

Exploration and Production operations are examined in more detail. The
order of presentation reflects the frequency of occurrence; the less common
types may, however, represent a greater threat to technical integrity!
7.3.1 CO2 Corrosion
CO2 is one of the most common compounds present in production operations,

especially in gas/condensate wells. It is a very common source of internal
corrosion problems.
When it is present in the gas phase, any water in contact with this gas will
dissolve CO2 to a concentration proportional to the partial pressure, pCO2, of
the CO2 in the gas (Henry's law):
pCO2 = (mole% CO2) x (gas pressure)

(9)
The solubility is also dependent on temperature.
Dissolved in water, CO2 forms carbonic acid:
H2O + CO-------> H2CO3 (10)

carbonic acid
which is a weak acid because it dissociates only a little:

+ -
H2CO3----> H + HCO3 (11)
bicarbonate ion
At room temperature, only one in a thousand molecules dissociates. The

second dissociation step is even weaker. Although it is a weak acid, carbonic
acid is highly corrosive because the cathodic part of the corrosion reaction
+
does not need H as with normal acid corrosion (see Section 7.2.2.1) but
involves direct reduction of undissociated acid:
- -
H2C03 + e -----> H + HCO3 (12)
followed by 2 H-----> H2. (13)

The overall corrosion reaction is:
Fe + H2CO3 -----> FeCO3 + H2 (14)

iron carbonate

For a given pH the corrosion rate of steel in carbonic acid exceeds that in
hydrochloric acid.
The corrosion rate caused by this reaction can be predicted by means of a
nomogram (Fig. 7-8), which gives a ‘worst case' corrosion rate as a function
of CO2 partial pressure and temperature. It represents the following relation
o
between corrosion rate Vcor (in mm/y) ,pCO2 (in bar) and temperature t (in C)
log (Vcor) = 5.8 – 1710 / 273 + t + 0.67 log(pCO2) (15)
The iron carbonate which is formed is only slightly soluble in water. It forms a
film on the steel's surface which can be easily washed away at low temper-
atures, but which at higher temperatures becomes more adherent and forms
a protective scale, which then reduces the corrosion rate (see Fig. 7-7).
The extent of this reduction is the ‘scale factor', which can be conservatively
estimated from the nomogram in Fig. 7-8, which assumes relatively turbulent
flow conditions.
The temperature where the scale survives the effects of liquid flow is called
the scaling temperature. It becomes lower for milder flow conditions, but
quantitative information about this is not available. Corrosion occurs often at
locations with more turbulent flow, leading to a localised corrosion pattern
which is very characteristic, and which is sometimes referred to as `mesa-
type' corrosion.

The exact composition of the fluid can change the basic CO2 corrosion rate
read from the nomogram. For example, if the pH is not that of CO2 and water,
a correction has to be applied; a higher pH (less acid) lowers the corrosion
rate w.r.t. the predicted value. This will occur, for example, when the elec-
trolyte contains dissolved sodium-carbonate or -bicarbonate, as is often the
case with formation water co-produced with oil. In the nomogram, a correc-
tion for the expected presence of dissolved FeCO3, which is slightly alkaline,
is already incorporated.

CO2 corrosion can occur in production tubing, in piping and separators and
also in pipelines carrying wet gas or wet oil where the water has dissolved
CO2 in it.
The corrosion rates are, in first approximation, independent of the type of
carbon steel and its condition w.r.t. heat treatment, although small amounts
of alloying elements can have an influence.
There is a notable exeption to this associated with the manufacture of the

threading on tubing. In order to accommodate the thread, the tubing ends are
heated to allow ‘upsetting' (to make them thicker). After this the pipes have to
be heat-treated to restore mechanical properties. If this heat treatment is not
done on the full length of the pipe, but only on the ends, a zone is formed
which is preferentially attacked by CO2, leading to a phenomenon called
‘ring-worm' corrosion. A full length normalising heat treatment after upset-
ting cures the problem.
Further reading:
- ARGR.90.090, C.de Waard cs, `Predictive model for CO2 corrosion engi-
neering'.
- ARGR.91;028, C.de Waard, Prediction of CO2 corrosion rates: Manual for
the use of computer spreadsheet `FLOLINE5'.
7.3.2 Erosion-Corrosion
The presence of a protective corrosion product film on the steel's surface can
appreciably reduce the corrosion rate. Although flow past the steel surface
can shear off some of this layer, under normal circumstances new film forms
faster than the rate of film removal. At very high flow velocities, however, the
film can be eroded away faster than new film can be formed. Under these
conditions, the corrosion rate can become extremely high.
In relation to CO2 corrosion, the phenomenon is mainly restricted to situa-
tions where fluid droplets are carried in a high velocity gas stream. The
impingement of these droplets on the protective iron carbonate film is very
effective in eroding this film away.
Commonly used information about a threshold velocity, below which no

erosion-corrosion occurs, is given in API's Recommended Practice 14E, for
piping. This is called the ‘erosional velocity' Verosion
Verosion = c/ √σ (16)
where σ is the average density of the fluid (gas + liquid), and c is a constant
which has the value 122 when SI units are used. Application of this formula

can be misleading. The velocity obtained from Eq. 16 coincides roughly with a
change in flow pattern (see PHB, Vol.8, Pipelines) from stratified flow to mist
flow. A mist flowpattern itself does not necessary lead to film erosion as long
as it is smooth; the flow should also be disturbed to some extent for this to
occur, for example by geometric effects (flanges!). Since it is not possible to
predict these effects, it is recommended to use 20 m/s as an estimate for
Verosion for production tubing and gas flowlines. Below this velocity, expe-
rience has shown that erosion-corrosion does normally not occur.
On copper-nickel alloys (e.g. 90 Cu 10 Ni) a much weaker film is formed when

used in seawater, which is susceptible to erosion-corrosion by flowing sea-
water at velocities exceeding 3 m/s. These materials have been applied for
piping and coolers handling seawater, but alternate alloys or non-metallic
materials are presently gaining preference.
Erosion-corrosion can also be caused by cavitation (implosion of gas bub-

bles) e.g. on pump impellors, or by wear (fretting) of metal surfaces rubbing
against each other.
Pure mechanical erosion of steel by high velocity liquid or gas does not occur
at practical velocities, but the presence of solid particles like sand can lead to
erosion damage (abrasion).
Further reading:
- API RP 14E, Design and Installation of Offshore Production Platform Piping
Systems.
7.3.3 ‘Low flow' Corrosion
For pipelines transporting wet crude oil, the water in the oil is not corrosive
as long as it is entrained in the oil, and is not the main (continuous) phase. In
this case the steel will be wetted by the oil and not by the water, and corrosion
is prevented. Entrainment of water is achieved when the flow velocity of the
liquid exceeds a critical value of about 1 m/s. This critical velocity is weakly
dependent on the type of crude. The water content should not exceed about
30%, otherwise water-wetting of the pipewall can occur, even if the water is
entrained. Also this critical water content depends on the type of crude. When
water separates as a separate phase below the oil, corrosion can occur
(depending on the composition of the water) which is sometimes referred to
as `low flow' corrosion. ‘Oversizing' of a pipeline can be one of the reasons
for this to occur.
With gas condensate liquids, water wetting and corrosion often occurs even
when the quantities of water are small and, based on the 1 m/s rule, would
normally be entrained.

7.3.4 ‘Top of line' Corrosion
In wet multi-phase (gas-liquid) pipelines with liquid flowing on the bottom

(stratified flow), water condenses from the gas in the top of the line, when it
cools with distance into the line, and accumulates on the bottom (Fig. 7-9).
When the presence of CO2 renders this water corrosive, injection of a cor-
rosion inhibitor will reduce the corrosion on the bottom, but not the corrosion
from freshly condensing water in the top. Preferential corrosion around the
12 o'clock position can be the result. A possible remedial measure is the
injection of glycol, which reduces the corrosion rate when present in the
water phase (Fig. 7-10). It reduces the corrosion both at the top of the line by
co-condensation and in the accumulated liquid at the bottom. Because rather
large concentrations are needed to control corrosion, glycol should not be
regarded as a corrosion inhibitor in the normal sense of the word. Glycol is
often used already to control the formation of hydrates in wet gas lines, and
combination with a glycol-based corrosion control scheme is an important
option for transport of wet natural gas.

7.3.5 H2S Corrosion and Hydrogen Embrittlement
H2S, whilst not as common as CO2 in production operations, is a major source

of internal corrosion and equipment failures.
In the presence of H2S conditions can be referred to as ‘sour', as opposed to
‘sweet' in the absence of H2S (see also Section 7.3.5.1 below).
When dissolved in water, H2S forms a weak acid which is often only mildly
corrosive because of the formation of iron sulphide, which often forms a
protective film on the surface of the steel:
Fe+H2S----->FeS + H2 (17)
However, FeS is electrically conductive and can act as a relatively noble

material. At imperfections or damage of the film, a corrosion cell is set up
between FeS and bare metal, resulting in enhanced corrosion (pitting) of the
bare steel. The main problem with the presence of H2S, however, is that it can
cause the steel to become brittle, leading to failure by cracking.
The basic cause of the embrittlement is in the cathodic reaction for this type of
acid corrosion:
+ -
2H2S ------>2H + 2H (18)
+ -
2H + 2e --------> 2H followed by: 2H ----> H 2 (19)
The last reaction is slow in the presence of H2S, the result being a relatively
high concentration of H atoms at the interface steel-electrolyte. These atoms
can diffuse into and through the metal. Some metals, including iron and steel,
can dissolve appreciable quantities of hydrogen in this way. When this hap-
pens, the mechanical properties of the metal change, and especially its
ductility can be reduced dramatically. This is called hydrogen embrittlement.
With increasing temperatures, the hydrogen diffusion through the steel goes
faster, leaving less hydrogen in the steel itself. The maximum susceptibility
of steel to hydrogen embrittlement problems is at room temperature. Above
80 ºC the degree of embrittlement becomes small.
Hydrogen uptake and hydrogen embrittlement are contributory factors to two

types of cracking: sulphide stress corrosion cracking (often abbreviated as
SSCC) and hydrogen induced cracking (often abbreviated as HIC).

7.3.5.1 Sulphide Stress Corrosion Cracking
SSCC can occur when a metal is stressed while in contact with H2S and
sulphides dissolved in water. Cracking is fast and can occur suddenly, which
can lead to an unacceptable release of toxic fluids. The phenomenon has
received considerable attention over the last 30 years, which has resulted in
a guidance document issued by NACE (National Association of Corrosion
Engineers in Houston, USA), referred to as MR-01-75. It is a basic code of
conduct to which most oil companies (voluntarily) adhere.
The document gives, in a graph, the concentration of H2S above which crack-
ing of steel with high hardness can be expected (Fig. 7-11) for gas systems
and for multiphase systems.
The left-hand sloping line of the graph corresponds to a partial pressure of
H2S of 0.05 psi (0.00345 bar). The same H2S concentration is often used to
distinguish between sweet and sour conditions.
For carbon steels this hardness limitation is 22 Rc (the Rockwell C hardness
scale) or 248 HV10 (Vickers hardness with 10 kg indenter weight). These
hardnesses should not be exceeded anywhere, e.g. not in the pipe nor in the
heat-affected zones of any associated welds. The result of this requirement is
a limitation in the strength of the steel which can be used in sour service. For

example, the hardness of a carbon steel tubing with an API grade higher than
L-80 (tensile strength 80.000 psi (552 MPa)) is likely to exceed 22 Rc. Harder
steels may sometimes be qualified for use after a special (expensive) qual-
ification procedure at the steel mill. For other alloys different hardness lim-
itations are applicable.
The NACE standard MR-01-75 is constantly being updated as the understand-

ing of SSCC improves. The guidance given with respect to carbon steels is
generally reliable, but for more highly alloyed materials its advice is less
dependable. Compliance with this standard does not offer protection against
other hydrogen uptake related phenomena such as HIC.
Further reading:
- MR0175-90, Standard Material Requirements, Sulphide Stress Cracking
Resistant Metallic Materials for OiIfield Equipment, National Association of
Corrosion Engineers, Houston. Texas, USA.
- Shell Specification Dr-1-2/3, C-95 casing and tubing for sour service.
7.3.5.2 Hydrogen Induced Cracking
The presence of dissolved hydrogen in a steel can lead to the development of

cracks even if no external load is present. In most steels which have been
rolled into plate to be made into vessels and pipe, non-metallic inclusions are
rolled out into thin sharp-edged platelets which can act as sites for the
accumulation of gaseous hydrogen (Fig. 7-12).

This hydrogen can reach high pressures and this induces high stresses in the
embrittled steel at the edges of the platelets, which can then be the starting
points of internal cracks. The cracks from different inclusions can then join to
form the characteristic step-wise cracking which is referred to as HIC. On the
surface of the steel blisters can be formed, while internally hydrogen-filled
voids can develop.
HIC can develop rapidly insusceptible steels. In situations where HIC may be
anticipated it is strongly recommended that ultrasonic inspection followed by
a fitness-for-purpose analysis is carried out.
The inclusions are in most cases manganese sulphide, which is a common

constituent of all steels. Remedial measures are the use of steels with extra
low sulphur concentration, together with alloying elements to control the
shape of the inclusions. There is no internationally recognised guidance
document for HIC. In practice HIC has caused the premature loss of pipelines
and of vessels.
Further reading:
- Shell Standard Specification L-3-2/3, Line Pipe for use in Oil and Gas
Operations under Sour Conditions.
7.3.6 Oxygen Corrosion
As described in Section 7.2.2.2. oxygen reduction can form the cathodic

reaction in the corrosion of steel. It is the source, for example, of most
atmospheric corrosion and of corrosion by seawater. In addition, because of
the inherent high speed of the oxygen reduction reaction, galvanic cells with
a large cathode and small anode surface area can cause much damage in
oxygen containing media. Examples are cells between steel (anode) and
millscale, between steel (anode) and rust, and between steel (cathode) and
weldbead (or heat affected zone) after welding. Removal of dissolved oxygen
from injection water to a level of 20 ppb is of great importance to reduce the
amount of rust which can enter a producing formation and also to obtain a
reasonable lifetime from any carbon steel injection equipment involved. In
practice this is very difficult, and the use of materials other than carbon steel
is sometimes considered for critical applications such as manifolds installed
on the bottom of the sea.
On the bottoms of oil tanks, water can separate which can be highly corrosive
when the tank is open to the air. Since about six times more oxygen dissolves
in oil than in water, the oil can act as a huge reservoir of oxygen to supply the
corrosion reaction. In the airspace under the tank roof the oxygen dissolved
in condensing water droplets is also readily replenished, and severe corro-
sion is the result.

In cases where the total amount of oxygen is limited, as with line fills for
pipelines, the oxygen is quickly depleted and the amount of corrosion re-
mains limited. However, when this depletion occurs locally under a deposit, a
differential aeration cell can be formed which causes ùnder-deposit' corro-
sion (see Section 7.2.6).
In handling water with low dissolved oxygen, anaerobic bacteria can become
active, and the addition of biocides may be necessary to avoid the forming of
H2S and pitting corrosion (see Section 7.4.4).
The simultaneous presence of oxygen and sulphides can be especially dan-
gerous. The sulphides can be partially oxidised by the oxygen to polysulfides
(H2Sx salts, x>1), which provide a particularly effective cathodic reaction
path.
7.3.7 Stress Corrosion Cracking in General
The simultaneous application of a tensile stress and exposure to certain

environments can lead to failure of a metal by cracking at loads appreciably
lower than its tensile strength level. This so-called stress corrosion cracking
can occur with a number of specific combinations of metal and environment,
which are characteristic for the metal concerned. It can be caused by high
corrosion rates on metal surfaces freshly formed as the result of shear forces
(anodic cracking), or by hydrogen embrittlement (Section 7.3.5)(cathodic
cracking). Table 7-2 lists a number of these combinations which are relevant
to production operations.
Table 7-2 Examples of metal-liquid combinations which give stress corrosion cracking

7.3.8 Corrosion Fatigue
When a metal is subjected to a cyclic stress of a certain amplitude e.g. in dry

air, it can fail by fracturing at an average stress far below the tensile strength
of the metal. However, a stress threshold exists below which the cycling does
not lead to failure (Fig. 7-13). This is called the endurance limit. If the metal is
simultaneously exposed to a corrosive environment, this threshold disap-
pears, and the number of cycles which can be survived is reduced substan-
tially. Reducing the corrosivity by addition of corrosion inhibitors or by the
application of cathodic protection can restore the endurance limit again.
Corrosion fatigue often starts from corrosion pits, which act as stress inten-
sity raisers.
Fatigue considerations are a major factor in the design of offshore drilling or
production platforms, where wave action is the cause of cyclic stresses.
Corrosion fatigue is also one of the main limiting factors for the life of sucker
rods of pumped wells.

7.4 Chemical Corrosion Inhibitors and Additives
7.4.1 Types of Inhibitors
In order to reduce the corrosion rate of carbon steel, one option is to add a
chemical corrosion inhibitor. There are various kinds of corrosion inhibitors,
depending on the mechanisms of their action. This can be interference with
the anodic (e.g. tubing corrosion inhibitors for acidizing of wells), or the
cathodic part of the corrosion reaction, or by inducing passivation (cooling
water). The most important type of corrosion inhibitor for the production of oil
and gas are the adsorption inhibitors, which form a protective organic film on
the surface of the metal.
7.4.2 Adsorption Inhibitors
These inhibitors are sometimes referred to as film-forming amines. Dis-

solved in a hydrocarbon phase, they are adsorbed on the steel and make its
surface oil (hydrocarbon) wetting and water repellent by changing the sur-
face tension between the steel and the surrounding liquids. An oil film is then
formed to act as a barrier between steel and water.
In order to provide a path for the inhibitor from the oil through a water phase
to the surface of the steel, the inhibitor is made dispersible (not soluble) in
water.

7.4.2.1 Method of application
For protection of production tubing in oil and gas wells both continuous
injection and batch inhibition can be applied. Continuous injection is nor-
mally done by filling the annulus between tubing and casing with inhibitor
solution, and pumping this into the bottom of the well (Fig. 7-14). For gas
wells, condensate is a common solvent, but if this is not available, diesel may
be used. Concentrations are designed to provide 100-200 ppmw on total
liquids in the tubing.
Batch inhibition is done by shutting in the well, and pumping inhibitor solution
into the tubing and leaving it for some 24 hours for the protective film to form,
before the well is put on stream again. The main drawback is lack of infor-
mation on how often this needs to be done.
With pipelines or flowlines, the inhibitor solution can be injected at the be-
ginning of the line, aiming at concentrations of 10-40 ppm. For flowlines,
however, the inhibitor is often carried over from the well when continuous
downhole injection is applied.
7.4.2.2 Limitations and testing
The selection of a corrosion inhibitor involves determining the following

properties:
- Corrosion inhibition efficiency. This is normally expressed as a percen-
tage e.g. 90% efficiency means a tenfold lower corrosion than the baseline
rate.
- Solubility in intended solvent, also at low temperatures (storage).
- High temperature stability and possible forming of decomposition pro-
ducts.
- Formation of solid deposits (blocking of injection nozzles).
Partition between solvent and water.
- Emulsion forming tendencies (affects oil-water separation).
- Foaming tendencies (affects gas-liquid separation).
The demand for water-dispersability of fiIm-formers is linked with the forming

of emulsions, which is one of the main disadvantages of this type of inhibitor.
Some standard tests exist to measure the inhibition efficiency (ASTM wheel
test), but these can at best serve to rank candidate inhibitors. More sophis-
ticated testing at simulated operational conditions can be carried out in
Group laboratories. Inhibition is less common in oil wells and pipelines than
with gas-condensate systems with relatively little liquid. When corrosion
becomes possible in oil systems, the quantities required to achieve sufficient
concentration are high and other corrosion prevention methods may be
needed.

7.4.3 Oxygen Scavengers
An important measure against corrosion is the removal of dissolved oxygen

by letting it react with sodium or ammonium bisulphite:
2 NaHSO3 + O2 -----> Na2S2O7 + H2O (20)

or or
2 NH4HSO3 (NH4) S2O7
With less than the stochiometric amount of oxygen, sulphates are also
formed. The sulphite is often added after the bulk of the oxygen has been
removed by stripping or vacuum pumping. Often cobalt-based catalysts are
also added. The reaction is adversely affected by the presence of H2S.
7.4.4 Biocides
Most natural waters contain sulphate reducing bacteria (SRB's), which can
become active in anaerobic conditions, e.g. after de-aeration. They can con-
vert dissolved sulphates into sulphides, e.g.:
SRB's
Na2SO4 + 2H2O + 2 C----------------------------> 2 NaHCO3 + H2S (21)
(from organic matter)
The resulting H2S can cause corrosion, SSCC and HIC (see Section 7.3.5).
When conditions are not conducive to their growth, they become dormant and
can survive very adverse conditions e.g. high temperatures, for long periods.
One method to control fouling and SRB problems is to add chlorine to the
system, either intermittently or by continuous injection. It should be remem-
bered that solutions of chlorine gas in water containing more than about 2
ppm are corrosive to steels.
An alternate method is to add bactericides (or more generally, biocides) to
sterilise SRB-infested systems. These chemicals are often related to gluta-
raldehyde, but other proprietary chemicals are also used in concentrations in
the 10-50 ppm range. Most of them are highly specific to one bacterial spe-
cies, and universally applicable biocides do not exist.
When any of the chemicals discussed in Section 7.4 need to be disposed of in

waste water, the impact on the environment needs to be carefully consi-
dered, and in many cases disposal into, for example, the sea may not be
acceptable at all!

7.5 Steels and Alloys
7.5.1 Carbon-manganese Steels
7.5.1.1 Composition and Structure
The vast bulk of steels used for engineering purposes such as offshore
structures or process equipment and piping, can collectively be described as
low carbon, or carbon-manganese steels. These are essentially alloys of iron
with small amounts of carbon (up to around 0.30%wt), manganese (up to
around 1 %wt), and some other elements such as silicon. The total amount of
alloying elements is typically less than 1.5% by weight. Measures are taken
during production of these steels to reduce the level of deleterious impuri-
ties, notably sulphur and phosphorous. Unless ordered to a more stringent
specification, these elements will typically be at levels of around 0.01% and
0.02% respectively.
Carbon-manganese steels are formulated to give the optimum combination

of strength, ductility, toughness and weldability properties, at a relatively low
cost. These properties are determined by the microstructural constituents of
the steel, which are in turn dependent on the precise chemical constituents
and production route (heat treatment) of the steel.
Carbon is the most influential of the alloying elements in these steels. At

room temperature pure iron consists of many small grains of a phase called
ferrite, which has little mechanical strength. (At much higher temperatures a
stable phase with a different structure, called austenite, is formed). The
addition of carbon, which is virtually insoluble in ferrite, leads to the forma-
tion of regions of iron carbide Fe3C, which is very hard and strengthens the
ferrite matrix considerably. Progressively more carbon leads to more Fe3C
and increased strengthening, but at the expense of reduced ductility, tough-
ness and weldability.
7.5.1.2 Heat Treatment
The combination of properties of a carbon-manganese steel can be con-

trolled to a large extent by the addition of alloying elements or by heat
treating the steel. Heat treatment is used to control the grain size of the ferrite
phase as well as the distribution and morphology of the iron carbide. Each of
these features has a significant influence on the mechanical properties of the
steel. One example of a commonly applied heat treatment is normalising,
o
whereby the steel is held at a high temperature (around 900 C) at which

austenite (in which the carbon is readily soluble) is the stable phase, then
allowed to cool slowly to room temperature. During cooling, ferrite and iron
carbide form from the austenite in a stable, fine grained structure to give a
steel with relatively low strength but good ductility and toughness.
An alternative is quenching and tempering. In this case the steel is rapidly

cooled from the austenite region by quenching in water or oil, which leads to
the formation of a phase called martensite, which is a form of ferrite in which
a relatively large amount of carbon is trapped, leading to distortion of the
crystal structure. As a consequence of its internally strained structure, mar-
tensite has high strength but is brittle and has insufficient ductility for most
applications. A subsequent tempering treatment is therefore normally car-
ned out whereby the steel is held at an intermediate temperature (around 550
o
C) to allow the formation of a more equilibrium structure of iron carbide and
ferrite containing less trapped carbon. This gives much improved ductility
whilst still retaining much of the strength.
A wide range of heat treatments can be carried out in conjunction with

various alloy additions to give specific properties such as toughness down to
o
very low temperatures (eg -50 C) or very high strength (e.g. 500 MPa yield
strength). Where special properties are present in the steel, it is often ne-
cessary to use tightly controlled welding procedures to ensure that these
properties are not lost due to the welding process.
7.5.1.3 Welding
The weldability, or ease with which a steel can be welded, is principally

determined by the chemical composition of the steel and its tendency to-
wards the formation of hard, brittle phases. Welding introduces large
amounts of heat, sufficient to melt a very localised region of the steel (tem-
o
peratures are in excess of 1600 C). A temperature gradient exists in the solid
metal around the molten weld pool which leaves a visible heat affected zone
around the solidified weld metal. Generally the solidifying weld metal and the
heat affected zone cool very quickly, in a manner which is analogous to
quenching and this can lead to the formation of martensite and other hard
phases, depending on the steel composition. The weld metal itself is usually
formulated to resist these effects but they may be troublesome in the heat
affected zone of the base metal. The welding also introduces a thermal cycle
which imposes large stresses which, in extreme cases, can be sufficient to
cause cracking if the steel has insufficient ductility to absorb the strains which
occur as the steel cools and contracts.

The tendency towards hardening as a result of the formation of martensitic

microstructures (i.e. hardenability) can be judged from a parameter called
the carbon equivalent which attempts to quantify the relative influence of
carbon and other elements in low alloy steels.
Ceq = %C + %Mn/6 + %(Cr+Mo+V)/5 + %(Ni+Cu)/15 (22)
Generally, acceptable weldability is achieved if the carbon equivalent is

below 0.45, though for some applications a lower figure may be recom-
mended.
In some cases it may be required to apply a heat treatment after welding to

reduce the hardness. This is equivalent to the tempering treatment described
above and is typically carried out by holding the welded steel at a temper-
o
ature of 550 to 600 C. This post-weld heat treatment has the dual functions of
improving the steel's ductility and toughness and of relieving much of the
residual stresses remaining after welding.
A wide range of processes exist for the welding of steels. In the oil industry,
where high quality welds are essential, processes which gain their energy
from an electric arc struck between an electrode and the workpiece, are
virtually exclusively used. Each process has its own characteristics in terms
of the amount of weld metal which can be deposited in one weld pass, the
heat input used, etc, which together determine the properties of the com-
pleted weld. Examples are:
i) Shielded Metal Arc Welding (SMAW), often referred to as manual metal

arc or stick electrode welding, where heat inputs (typically around 1 kJ/
mm .of weld) and deposition rates are relatively low. This is widely used for
the welding of pipelines and offshore structures due to its versatility.
ii) Submerged Arc Welding (SAW), which has high heat inputs (typically 5-10
kJ/mm) and high deposition rates. This is a fully mechanised process with
the feed and forward travel of the welding wire being automatically con-
trolled. It is suitable only for workshop applications such as pressure
vessel fabrication.
Because of the very large number of variables which apply (e.g. process,
heat input, travel speed, weld metal composition, etc.) it is necessary to
perform the welding in accordance with a written welding procedure spe-
cification (WPS). The WPS specifies the allowable ranges of the essential
variables and must be qualified before use by performing a series of tests on
a trial weld to produce a Procedure Qualification Record (PQR). Guidelines
for the performance of such welding procedure qualification testing are given
in industry design codes for pipelines, piping, equipment and structures (e.g.
ASME Section 9 for pressure vessels or API 1104 for pipelines).

7.5.2 Stainless steels
Small alloying additions as described in Section 7.5.1.1 can have dramatic

effects on the mechanical properties of steel but generally have a less pro-
found influence on the corrosion resistance characteristics. An effective way
of improving the corrosion resistance is to alloy the steel with chromium.
Depending on the corrosive environment, even a few percent of Cr can give a
substantial improvement and at levels above around 12%, corrosion rates
can drop to virtually zero. This ‘passivation' results from the formation of a
very thin, stable layer of chromium oxides at the steel surfaces.
A range of commonly applied stainless steels with typical applications is
given in Table 7-3.
Table 7-3 Name, approximate composition and typical application of some corrosion resistant
alloys
7.5.2.1 Ferritic/martensitic Stainless Steels
The 12-13% Cr alloys (e.g. AISI 420) are the lowest alloyed and consequently,
the least expensive in the family of stainless steels. Their structure can be
ferritic or martensitic, depending on the carbon content and they have high
strength (typical martensitic grades have yield strengths in the range 550-650
MPa) but they suffer from very poor weldability. Some low carbon, ferritic
grades can be welded, but the alloy is most widely used in unwelded com-
ponents like production tubing in moderately corrosive gas or oil wells.

An example is the successful application of 13% Cr (AISI 420) tubing and

weliheads in the Groningen gas field where they experience conditions of
0.9% CO2, bottom hole pressure of around 150 bars, low chlorides (less than
o
10 ppm) and temperatures of 105 to 70 C. Prior to the installation of 13 Cr,
corrosion rates of 0.4 mm/yr were encountered with carbon steel tubing,
despite the use of an inhibitor. Because of its high strength and hardness the
alloy has limited applicability where H2S is present, even in small quantities.
This is discussed further in Section 7.3.5.1.
7.5.2.2 Austenitic Stainless Steels
An increase in the chromium content to 15-20%, combined with the addition

of nickel in the range 8-10%, has the effect of stabilising the austenitic struc-
ture down to room temperature. This is very beneficial in terms of weldability
since it means that the steel no longer has to undergo a phase transformation
on cooling and hardenability is no longer a concern.
The austenitic alloys are non-magnetic and are characterised by excellent

ductility and toughness. Yield strengths are relatively low in most alloys (e.g.
250 MPa for AISI 304), but they work harden rapidly, which can lead to
substantial changes in mechanical and corrosion properties if they are cold
worked or deformed. Heating during welding can lead to the formation of
chromium carbides at the grain boundaries (sensitisation) with a corres-
ponding decrease in corrosion resistance due to Cr being depleted. This
tendency is overcome by using low carbon grades (e.g. AISI 316L) for con-
structions which are to be welded or heat treated. (The phenomenon of
sensitisation is discussed at more length in Section 7.6.4.)
Austenitic stainless steels have excellent resistance to selective corrosive

media and their good weldability means that they can be conveniently used in
plant piping and equipment. Molybdenum is commonly added in the range of
2-3% to improve the resistance to seawater at relatively low temperatures
(AISI 316), but generally, in oxygen-containing environments, the so-called
18Cr 8Ni types of stainless steels are susceptible to attack in the presence of
chlorides, which can lead to localised breakdown of the protective surface
film (See Section 7.6).

In oxygen-free media the resistance to high chloride levels (such as may be

encountered in formation water) at relatively high temperatures is much
superior to the ferritic or martensitic stainless steels. The low strength of
austenitic stainless steels is, however, a disadvantage in high pressure sys-
tems, so they are frequently applied in clad form for piping and pressure
vessels. A thin layer of stainless steel provides resistance to the corrosive
process media and a backing layer of high strength, carbon-manganese steel
provides the pressure retaining capacity. The addition of nickel and molyb-
denum adds considerably to the cost of stainless steels.
A number of austenitic alloys containing higher levels of alloying elements

(often referred to as ‘super-austenitics') are available for particularly corro-
sive applications, especially where relatively high levels of H2S are present.
The Sanicro 28 alloy for example, can be applied in conditions similar to
those mentioned above but with up to at least 3 bar H2S. Further examples of
higher alloyed materials are given in Table 7-3.
7.5.2.3 Duplex Stainless Steels
The range of alloys applied in exploration and production environments is

completed by duplex stainless steels. These have less nickel than the aus-
tenitic stainless steels which results in the austenite phase not being com-
pletely stabilised, leading to the formation of a microstructure with approx-
imately equal amounts of austenite and ferrite grains. The proportion of each
phase depends on the exact composition and on the cooling rate during
production or heat treatment. Tight control of the welding procedures is very
important for duplex steels to ensure they retain the desired properties.
Duplex stainless steels are generally substantially stronger than austenitic

steels (yield strengths in the range 350-550 MPa) but have limitations in low
o
temperature applications due to reduced toughness below about -50 C. They
are widely used for flowlines carrying wet produced gas containing a high
concentration of CO2 , where corrosivity is a major concern.

7.6 Corrosion of Stainless Steels
The main alloying element which makes a steel ‘stainless' is chromium.

Chromium is basically a very active metal, but it passivates very easily by
means of a protective film of chromium oxides like Cr2O3:
2Cr + 3H2O ------> Cr2O3 + 3H2 (23)
However, if the medium is very acidic, then the chromium simply dissolves as
chromium ions, again under hydrogen evolution. The film is often only a few
molecules thick and opaque, and therefore invisible. The only corrosion
which occurs is that needed for maintenance of the protective film to form
new oxide at defects on a molecular scale.
At about 12-13% Cr the surface of the alloy is almost completely covered with
chromium oxide.
The corrosion resistance of the protective film can be further enhanced by
increasing nickel and chromium, culminating in various types of nickel al-
loys. The cost of an alloy increases sharply with its nickel content. Some of
these alloys are shown in Table 7-3.
The passivity of chromium containing alloys gives a very high resistance to

corrosion in almost all environments of practical interest, with the exception
of chlorides. The presence of chloride ions leads to the initiation of defects in
the protective film, which is the cause of a number of localised corrosion
phenomena typical for stainless steels viz, crevice corrosion, pitting corro-
sion and chloride stress corrosion cracking.
7.6.1 Crevice Corrosion
In crevices, corrosion products can build up in the liquid film between metal
surfaces (Fig. 7-15), and the presence of chlorides can lead to the formation of
a concentrated metal chloride solution which can become very acidic. This in
turn can cause the loss of passivity, and subsequent high rates of corrosion in
these crevices.
In most cases of practical interest the main cathodic reaction is oxygen
reduction on the passive metal outside the crevice. Inside the crevice the
oxygen will be depleted when corrosion occurs, and both the differential
aeration cell and the active/passive galvanic cell between the metal inside
and that outside the crevice will contribute to the rate of attack.

7.6.2 Pitting Corrosion
The protective film on a passive steel acts as a noble material, which, when
connected to a more base material, can cause galvanic corrosion. When
defects occur in a passive layer as the result of the presence of chloride ions,
the galvanic cell between active and passive steel will enhance corrosion at
these defects (Fig. 7-16), and sharp deep pits can develop. The same cell
protects the area immediately around each pit, and this keeps them separate
from each other. Once initiated, repassivation of the metal inside a pit is
hindered by the acidity of the metal chlorides which can accumulate there.
The susceptibility to pitting corrosion can be significantly reduced by adding
molybdenum (Mo) to the alloy.

7.6.3 Chloride Stress Corrosion Cracking
When stainless steel is stressed while exposed to a chloride containing

solution, the metal can develop cracks and fail at loads far below the tensile
strength of the material. The mechanism involves the shearing of the pro-
tective film at areas of stress concentration e.g. a pit, and subsequent high
corrosion rates at the freshly formed metal surface. The cracks are often
highly branched and run through the austenite grains (transcrystalline crack-
ing).
o
In the presence of dissolved oxygen this happens at temperatures > 60 C,
but in the presence of H2S it may occur at lower temperatures.
7.6.4 Sensitisation
The carbon which is dissolved in austenite or which is present as iron carbide

in ferrite can react with chromium, when the metal is heated in the temper-
o
ature range of 400-850 C to form chromium carbide Cr23C6. This reaction
proceeds predominantly at the grain boundaries, where a large proportion of
the chromium in the alloy is scavenged by the carbon to form carbide. The
chromium in the carbide does not contribute to the formation of the passive
layer, and if the chromium depletion is sufficiently severe, the passive layer
breaks down at the surface over the grain boundaries. The resulting localised
corrosion at these boundaries can lead to dislodging of grains in severe
cases. In the presence of tensile stresses, intergranular cracks can develop.
Prolonged heating in the sensitisation range will tend to even out the local
dips in chromium concentration at the grain boundaries. By heating at a
sufficientIy high temperature, the chromium carbide can be decomposed
again to release the chromium (solution annealing).
Welding of stainless steels can lead to sensitisation at the heat affected zone
of the weld, which can lead to pronounced corrosion adjacent to the weld
(weld decay and `knife line' attack).
One of the remedies is to use low carbon stainless steels e.g. 316L. A second
approach involves the addition of alloying elements such as Ti or Nb. These
elements react more readily than chromium with carbon when heated in the
sensitising range. Examples are AISI 321 (Ti) and AISI 347 (Nb).

7.7 Cathodic Protection
7.7.1 Control of External Corrosion
Cathodic protection can completely stop corrosion of steel which is im-

mersed in seawater or buried in the ground. It is most often used in E&P
operations to externally protect pipelines and offshore structures, but it is
also applied to jetties, well casings, tank bottoms etc.
By connecting a base metal to the steel to be protected, a galvanic cell is

created in which this metal carries all anodic current and will corrode, while
the steel structure carries all the cathodic current and no longer supports any
anodic processes and is therefore protected from corrosion. The base metal
in this cell is sacrificed in the process, and is called the sacrificial anode. It is
not necessary that the current for the protection is provided by a galvanic
cell. It is also possible to provide the current from an external source such as
a battery, a rectifier or a solar panel. This is called cathodic protection by
impressed current. For offshore structures and offshore pipelines, sacrificial
anodes are normally used because such systems are maintenance-free over
their design life and more reliable than impressed current.
The amount of current needed to achieve protection depends on the tem-

perature of the steel and on the composition of the environment in terms of
2
acidity and dissolved oxygen, and is typically in the range of 40-80 mA/m of
bare steel.
The potential of the steel is changed in the base -more negative direction- by
the application of the protective current. The current is designed or adjusted
to depress the potential below a generally accepted threshold value of -0.850
V w.r.t. Cu/CuSO4 reference electrode. For other reference electrodes the
corresponding value can be obtained by using Table 7-1.
For the Shell Group, offshore structures, like the jackets for production plat-
forms, are made of bare (uncoated) steel, which relies completely on ca-
thodic protection by means of sacrificial aluminium anodes for corrosion
protection under water. Special aluminium alloys are needed to counteract
the tendency of aluminium to passivate in cold water or in seabottom
mud.
Offshore pipelines are coated, and sacrificial anodes are applied most com-
monly in the form of zinc bracelets e.g. on every 10th pipe joint.

Onshore buried pipelines are also coated and are normally cathodically
protected by impressed current (Fig. 7-17). The anode -referred to as anode
ground bed in this case- can consist of any metal like scrap iron, which is
buried. Longer lasting anode ground beds can be made of silicon iron rods or
graphite, surrounded by a low resistivity carbonaceous backfill medium to
reduce the electrical resistance of the bed.
The groundbed is typically situated about between 50 and 100 m from the
pipe. The protection is strongest near the anode ground bed, and attenuates
with distance. For a well coated pipeline, one anode bed may protect a
distance in the order of 40 km, but this is highly dependent on soil conductivity
and corrosivity. Impressed current systems can give rise to ‘overprotection'
which can cause deterioration of pipe and coating material through the gen-
eration of hydrogen at the metal surface.
Cathodic protection systems of different pipelines can interfere with each
other, which can lead to ‘stray current' corrosion problems.
For sacrificial anode systems, the total amount W of sacrificial material

needed can be calculated from:
W = i x (project life) x C/U (24)
where i is the total current required for protection (from surface area x
current density), C is the electrochemical consumption rate (see Table 7-4) of
the anode material, and U is a utilisation factor of about 0.9.
The total amount of material needed is distributed over the surface of the
protected steel in the form of individual anodes. Determination of the opti-
mum size and shape of these anodes requires special calculation. Individual
anodes often weigh several hundred kilograms, and for an offshore platform
thousands of these may be needed.

Table 7-4 Typical values of electrochemical consumption rates of some anode ma-
terials
Inspection of a cathodic protection system basically consists of measuring

the potential between the steel and a reference electrode which is held near
the location to be inspected. Especially for offshore pipelines, other techni-
ques are available to measure the current flowing from the sacrificial anodes
to the pipe. For onshore pipelines, contact points to the pipe are provided at
fixed distances of typically 1 km, and the potential can be measured between
the pipe and a reference electrode pressed into the (wetted) soil above it at
each of these points.
7.7.2 Control of Internal Corrosion
For systems where oxygen is the potential cause of corrosion (e.g. (sea)
water tanks) sacrificial anodes can be successfully applied, following rules
similar to those described above, though special attention must be given to
anode distribution and their limited `throwing power' in confined spaces.
Anodes are also sometimes applied in process equipment where CO2 and
H2S corrosion is to be suppressed. The design rules for such systems are less
well defined, but there is evidence that, for example, the internal protection of
tank bottoms often requires high current densities in excess of those needed
in oxygen corrosion situations. However, normal corrosion (`self corrosion')
of the anode material can be mistaken for cathodic protection activity.
Further reading:
- DEP 30.10.73.10-Gen, Cathodic Protection (June 1983, in revision).
- DDD 37.19.30.30-Gen, Design of Cathodic Protection Systems for Offshore
Structures (in revision.).
- DEP 30.10.73.31-Gen, Design of cathodic protection systems for onshore
buried pipelines (Final Draft, December 1991).

7.8 Non-metallic Materials
The non-metallic materials used in EP operations can be broadly grouped

into just a few generic categories with characteristic properties.
These include plastics, rubbers, paints, coatings, and cements. In many

cases a number of non-metallic materials are combined to form composite
materials, often with synergistic properties. In some cases non-metallic ma-
terials compete with metals and their use is then often associated with their
resistance to electrochemical corrosion, although additional features such
as light weight, ease and flexibility of fabrication, and cost-competitiveness
can be important. In other cases non-metallics are chosen because they offer
properties not readily achieved with metallics, such as low electrical and
thermal conductivity, magnetic transparency, or design / processing flexibility.
The following Sections introduce the main types of non-metallic materials

used in EP operations and their most common applications.
Further reading:
-- DEP 30.10.02.13. Non-metallic materials selection and application.
Dec. 1984.
7.8.1 Plastics
ASTM D-883 defines plastics as materials which contain, as an essential

ingredient, one or more organic polymeric substances of large molecular
weight, which are solid in their finished state and which, at some stage in
their manufacture or processing into finished articles, can be shaped by
flow.
Plastics can be divided into two main sub-groups, thermoplastics and ther-
mosets.
Further reading:
- DEP 30.10.02.13. Non-metallic materials selection and application.
Dec. 1984.
- ASTM D-883. Standard definitions of terms relating to plastics.

7.8.1.1 Thermo plastics
Thermoplastics can repeatedly be softened and hardened by heating and

cooling respectively. In the softened state they can be shaped by flow in
processes such as moulding or extrusion. Examples include PVC, polyethy-
lene, polypropylene, and polyfluorocarbons. Typical uses include piping and
fittings, and as corrosion-resistant liners for pipes, tanks, and vessels.
Jointing techniques include mechanical, welding and adhesive bonding. PVC

is often adhesively bonded using solvent-based adhesives. Polypropylene
and polyethylene are usually welded.
7.8.1.1.1 Properties
Thermoplastics' properties are usually markedly dependent on both tem-
o
perature (even below 100 C) and strain rate.
Creep is an important phenomenon to be considered in engineering with

thermoplastics. The combined effects of time, stress, and temperature must
be considered in designing with them.
Impact performance varies widely between individual thermoplastics. Tem-

perature influences the impact performance. The large strain capabilities of
most thermoplastics can impart good impact resistance.
Chemical resistance against inorganic species is generally good, but they

can be vulnerable to oxidising and organic environments (particularly aro-
matics and chlorinated hydrocarbons). The polyfluorocarbon plastics are
usually highly resistant to most chemical attack.
Ageing reactions should be considered during material selection and design.

Such mechanisms may involve reaction with the environment (e.g. oxida-
tion), or reactions of the plastic molecules without external reagents (e.g.
dehydrochlorination of PVC). Commercial polymers frequently contain ad-
ditives to inhibit degradation reactions but these often become ineffective
with time.
7.8.1.1.2 Thermoplastic Pipes

Thermoplastic pipe is widely used in the domestic water and gas distribution
industries but its use in Group operations remains relatively restricted. It is
being used more extensively by some oil companies for low-pressure ap-
plications such as water transport and some gas gathering systems.
The main materials are PVC, polyethylene, and polypropylene. Strength is
time dependent with failure occurring by creep, so pressure ratings must be

based on long-term product characterisation. Standards such as API-15LE,

based on long-term tests, have been developed in order to determine safe
long-term design values.
Further reading:
- API-15LE. API specification for polyethylene line pipe.
- DEP 31.38.70.36. Requirements for thermoplastics pipes and fittings.
- DEP 31.38.70.25. Fabrication and installation of thermoplastics piping
systems.
7.8.1.1.3 Polyethylene Liners for Carbon Steel Pipe
Polyethylene (PE) lined steel pipe aims to provide the corrosion resistance of
PE with the strength and familiarity of carbon steel. The PE liner provides the
corrosion-resistant fluid barrier whilst the steel carrier provides the
strength.
The system comprises a thick (5-10mm typically) PE pipe, the liner, fitted
tightly inside a carbon steel carrier pipe. Pressure and mechanical loads are
borne by the steel pipe. The maximum operating temperature is currently
o
considered to be about 65 C. There are a number of installation techniques
by which the liner is fitted into the steel. The two most common are:
1. An undersize PE pipe is pulled into the steel pipe and later swaged into
intimate contact with the steel, usually by pressurising with hot water.
2. An oversize/interference PE pipe is temporarily reduced in diameter by
passing through a set of rollers. Before the PE relaxes back to its original
dimensions (this can take some considerable time) it is pulled into the
steel pipe.
Lengths of up to a kilometre (sometimes more) can be lined by these tech-

niques. Lined sections are usually terminated by fusion welding PE flanges
onto the liner at each end and using flanged joints on the steel, thus giving a
continuous internal PE lining when the steel flanges are made up. Techni-
ques have been developed to enable the steel pipes to be welded without
damaging the PE liner, thus giving a continuous welded, rather than flanged,
construction.
The major concern at present relates to ensuring liner stability, particularly

on depressurisation of the line. Some liner collapses have been observed.
Vent points are often incorporated in the steel line at regular intervals to
enable venting of gases which permeate through the PE liner and accumulate
between it and the steel. Views on optimum operational procedures for vent-
ing still vary and standardised practices have yet to evolve.

Significant experience of these systems is quoted in the industry literature,

particularly in water service. Within the Group applications are currently
limited to water service although there are plans for use in hydrocarbon
service. There are currently some unresolved issues, currently being inves-
tigated, regarding the effect of hydrocarbons on the PE.
Further reading:
- EP 89-1260. The applicability of PE-liners to protect carbon steel
pipelines against corrosion.
- DEP 30.48.50.12. Application and installation of plastic-lined steel pip-
ing.
7.8.1.2 Thermosets
Thermosets are plastics which, upon curing by heat or other means, become
essentially infusible.
Thermosets fall into a number of main groups according to their chemistry.
The main groups are epoxies, polyesters, vinylesters and furane resins.
Applications for unreinforced thermosets include coatings and encapsula-
tion, but they are almost exclusively used in combination with other materials
to reinforce them or otherwise modify their properties. Examples include
glass fibre reinforced plastics, glass flake filled plastic for paints and coat-
ings, and filled plastics to form syntactic foams and polymer concretes.
7.8.2 Plastics Composites (reinforced/filled/foamed)
Composites are combinations of two or more constituent materials, which

are usually prepared to achieve properties unobtainable from the consti-
tuents alone.
7.8.2.1 Flexible Steel/Polymer Pipe
Thermoplastics play a key role in flexible steel/polymer pipes (‘flexibles' e.g.

Coflexip type). Flexibles allocate the two main roles of a pipe, containment
and strength, to different components in the pipe construction. Containment
is provided by an inner thermoplastic pipe, typically a polyamide or fluoro-
polymer. The thermoplastic pipe alone is unable to withstand the stresses
induced by pressure and mechanical loads, however, and is supported by
several sets of over-wrapped steel tendons. The typical construction is com-
pleted by an external thermoplastic sheath which protects the steel rein-
forcement from the environment

Flexibles are manufactured to length and connected to each- other if neces-

sary by means of special end fittings built into the pipe end during manu-
facture. Attention is needed to select the end fitting material and/or coating to
ensure adequate corrosion protection. Normally the end fitting terminates as
a flange, but weld-ends are also available.
Although the main materials in contact with the internal and external fluids,
the polymer layers, are not susceptible to electrochemical corrosion, several
corrosion and degradation phenomena are possible.
For wet/subsea applications (sacrificial) cathodic protection is generally pro-
vided for the end fittings and/or for the steel reinforcement should the outer
polymer sheath be locally damaged.
The steel reinforcement may also be susceptible to sulphide stress corrosion

cracking due to permeation of hydrogen sulphide through the inner polymer
liner to the steel. Such permeation can occur relatively quickly (increasingly
so with increasing pressure and temperature) and its effect can be accen-
tuated because the permeability coefficient for hydrogen sulphide is typically
a factor of about 10 higher than for methane.
The long-term stability of the sheath polymers must be considered with

regard to the fluids to which they are exposed and the operating temperature.
The essential performance requirement for the polymer is to maintain barrier
integrity and so some physical changes such as limited swelling, and limited
changes to mechanical properties (modulii, strengths) may be acceptable
but it should be realised that inner liner collapse due to swelling/annular
pressure, accentuated by modulus reduction, can occur. Degradation me-
chanisms involving embrittlement and significantly reduced ultimate strain
can also occur and could lead to leaks as the liner is strained in normal
service.
Flexibles are used in both ‘static' and ‘dynamic' applications. Dynamic ap-
plications, such as suspended risers, are characterised by significant cyclic
mechanical loads and displacements in addition to the pressure loads. Flexi-
bles are being used increasingly in static applications such as oil and gas
flowlines, often buried, where the high material cost of the flexible can be
offset by rapid and inexpensive installation. Flexibles can also be useful to
avoid alignment problems in some facilities pipework and for temporary
equipment connections.
Further reading:
Preparation of a DEP on flexibles is in the current SIPM Standards Action
Plan.

7.8.2.2 Reinforced Plastics / GRP
Reinforced plastics are composite materials comprising one or more rein-

forcement materials embedded in a plastic matrix.
A number of reinforcement material types and formats are available, al-
though long-fibre glass reinforced plastic (GRP) is most widely used in EP.
Other reinforcement fibres include aramids and carbon, which can give im-
proved mechanical properties but are considerably more expensive, whilst
highly-oriented thermoplastic fibres are also available.
In GRP the reinforcement is usually E-glass. Two other glass types are used,
S-glass for higher strength (used infrequently because of high cost) and
C-glass for increased chemical resistance.
Both thermoplastic and thermoset plastics can be used for the matrix but
thermosets currently predominate, particularly epoxies and vinylesters.
Some high performance thermoplastic systems are available but are cur-
rently too expensive to be considered as commodity products. Cheaper ther-
moplastics are likely to become more widely used in the near future.
GRP is being increasingly widely used as an engineering material for pipes,

tanks, and structural components such as gratings, handrails, panels. These
applications, and a more detailed introduction to GRP, are reviewed in EP
report 91-1450.
Fire resistant enclosures for pipeline and riser emergency shutdown valves
made from GRP sandwich materials are in service. They were selected
following full-scale test demonstrations of their ability to protect the valves
against jet-fire impingement exceeding two hours, the design requirement
for this application.
Piping has been the most common use of GRP to date. The attraction of GRP
pipe lies in its light weight, ease of installation, and resistance to electro-
chemical corrosion and chemical attack. These features can result in cost-
effective piping systems with low maintenance and high reliability and safety.
Low-pressure sewer and drainage systems are used in diameters exceeding

one metre, whilst small diameter (50-75 mm) lines are operating at over 140
bar. Most use in the past has been onshore, but a number of current offshore
projects are using GRP widely. For onshore buried firewater systems GRP
has been the recommended material for some time, but there is rapidly
growing interest in its use for offshore firewater systems. Recent work has
validated that high levels of fire resistance can be achieved whilst the

material's corrosion resistance can avoid a major source of firewater system

unreliability - deluge nozzle blockage by corrosion products.
Typically, material costs for GRP pipe components exceed those of carbon
steel but installation costs are significantly less. The resulting installed cost
of GRP is usually about the same as carbon steel. The corrosion resistance of
GRP compared to carbon steel gives improved life-cycle costs in corrosive
environments.
Pressure ratings are worthy of special mention because GRP behaves dif-
ferently to steel and manufacturer's pressure ratings have often proved op-
timistic in independent tests. GRP pipe normally fails in a time-dependent
manner and hence it is important that pressure rating methods are based on
long-term strength properties. API-15HR and API-15LR address pressure
ratings of high (static) pressure and low (cyclic) pressure components res-
pectively. A new Group DEP will be issued in 1992 which is based on API-
15HR but encompasses and extends the scope of both documents.
Further reading:
- EP 91-1450 GRP in EP/MF Operations.
- API-15HR. Specification for high pressure fiberglass line pipe.
1988.
- API-15LR. Specification for low pressure fiberglass line pipe.
1990.
- DEP 31.40.31.31. Technical specification. Requirements for glass fibre
reinforced plastic pipes and fittings. (Publication due
mid-1992)
- EP 91-1830 Technical manual for DEP 31.40.10.31 (Requirements
for glass fibre reinforced plastic pipes and fittings).
(Publication of revised version due mid-1992).
- EPD/6 Memo. 91-03 The use of GRP piping in offshore fire water systems:
the VaIhall project.
- DEP 31 .22.30.14. Design and calculation of glass-fibre reinforced
epoxy vessels and equipment. 1977.
- DEP 31 .22.30.34. Requirements for glass-fibre reinforced epoxy ves-
sels and equipment. 1978.
- DEP 31.22.30.24. Installation of glass-fibre reinforced epoxy vessels
and equipment. 1978.

7.8.2.3 Filled/Foamed Plastics
Filled and foamed plastics form another category of plastic-based composite

material, but one in which the prime aim of the ‘reinforcing' constituent is to
modify the physical rather than mechanical properties of the plastic matrix.
Typical properties tailored in this way include density, thermal conductivity,
and permeability.
Flake glass fillers are used in some paint formulations to improve barrier and
abrasion properties, whilst heavy particulate fillers are used for weight coat-
ings and polymer concretes.
Lightweight ‘fillers' are used to produce foams either for their buoyancy or
thermal insulation properties - both gas and syntactic foamed pipeline in-
sulation systems have been developed which are capable of withstanding the
mechanical loads imposed by reel-barge installation. The `fillers' may be gas
bubbles, which produce ‘gas-blown' foams, or lightweight (solid or hollow)
particles which produce ‘syntactic' foams.
Resistance to long-term creep is an important feature in both buoyancy and

sub-sea thermal insulation applications: collapse of the foam will increase its
density and thermal conductivity. Gas foamed polymers are more suscepti-
ble to creep than glass-microsphere based syntactics. Gas-foams generally
show a multi-phase creep failure mechanism : after apparently stable low
creep rates for significant periods the material can suddenly show a marked
increase in creep in which the cells (partially) collapse.
Further reading:
An EP report summarising the ‘state of the art' of (pipeline) thermal insulation
materials is under preparation. Publication is due mid-1992.

7.8.2.4 Fire Protective Coatings
Fire protective coatings are used in EP to protect personnel and equipment in

control rooms and other closed spaces, and to protect structural elements.
They achieve this by restricting the heat transmission through, and temper-
ature of, the substrate to which they are applied. This may be in order to delay
or restrict:
- loss of structural stability of the substrate,
- loss of physical integrity (burn-through),
- back-face temperature rise and heat transmission.
Several types of non-metallic material are used for fire protective coatings
including mineral and ceramic fibres, lightweight cements and epoxy-based
intumescent (swell on exposure to fire) materials. All of these either start off
with, or develop on heating, low thermal conductivities. Some also reduce
heat gain to the substrate by absorbing incident heat in chemical reactions or
dehydration within the coating, or by using low-emissivity surfaces.
In selecting a fire protective coating system, not only should thermal aspects
be considered but also features such as toxic emissions and resistance to
impact and weathering. Two types of fire condition have traditionally been
considered, characterised by different time/temperature profiles: cellulosic
(e.g. SOLAS fire curve) and hydrocarbon (e.g. MOBIL fire curve). The hy-
drocarbon curve is the more severe, having a faster temperature rise rate
and higher ultimate temperature, and has generally been used in the oil
industry. Fires arising from ignition of a high-pressure hydrocarbon leak,
known as `jet fires', are even more severe due to the high heat flux and
erosional effects they produce. Jet fire resistance is increasingly being con-
sidered in the design of hydrocarbon installations.
Further reading:
- EP 88-0387. A catalogue of fire protective coatings and allied ma-
terials for offshore platforms and related structures.

7.8.3 Rubbers and elastomers
In general for a given rubber composition the chemical resistance increases

with increasing hardness, hence ebonites generally have a higher chemical
resistance than the soft rubbers. Rubber is used mainly for sealing elements.
The main categories of rubbers are:
- Natural rubber (NR) (soft)
- Styrene-butadiene rubber (SBR)
- Chloroprene rubber (CR)
- Butyl rubber
- Nitrile rubber (NBR)
- Ethylene propylene rubbers (EPM and EPDM)
- Urethane rubber
- Chlorosulphonated polyethylene (CSM)
- Fluoro rubbers of the polymethylene type (FKM).
Chemical resistance, and to a more limited degree, thermal resistance, vary

significantly between these types of rubber. These features generally form
the main criteria for selection, although physical performance criteria are
important in some applications, e.g. resilience/hardness, and resistance to
explosive decompression in high-pressure seals.
Further reading:
- ASTM D-1566. Standard terminology relating to rubber.
- DEP 30.48.60.10. Design and installation of rubber-lined process equip-
ment piping and concrete structures.
- DEP 30.48.60.30. Requirements for rubber linings for process equipment
and linings.

7.8.4. Paints and coatings
Conventional liquid-applied coatings generally consist of up to three main

components:
i) Pigments/reinforcements - solid particles which can serve to impart co-
our and opacity, give added strength to the film, and increase chemical
and environmental resistance.
ii) Binder - that part of the system which forms the continuous phase of the
solid coating. The binder forms a matrix around pigment particles, pro-
motes adhesion to the substrate, and gives abrasion resistance and flex-
ibility to the film as well as imparting the paint's basic chemical and
environmental resistance.
iii) Solvent - a liquid component which escapes from the film, usually by
evaporation, leaving the binder and pigments to form the final solid paint
* film. The main functions of the solvent are to control the paint viscosity, to
transport the pigment and binder during application and to promote wet-
ting of the substrate.
Single-component polymeric coatings such as polyethylene, polypropylene,

and fusion bonded epoxy are also being used increasingly.
The importance of proper surface preparation prior to coating application

cannot be overstressed. Application of coatings over rust may even accel-
erate corrosion by maintaining a corrosive environment adjacent to the steel.
Preparation consists primarily of cleaning and roughening, both necessary
to obtain adequate adhesion. This is typically accomplished by blast clean-
ing.
Surface cleanliness is often measured by comparison to visual standards.

The Swedish Standard SIS 055900 (now 150-8501-1) has been widely used. It
defines a number of levels of cleanliness, e.g. ‘Sa 3' (100% contaminant
removal – ‘white metal' surface), and ‘Sa 2 ½ ' (95% contaminant removal -
‘near white metal' surface). Lesser levels, e.g. ‘Sa2' (‘commercial blast') are
defined but Sa 2½ or Sa 3 are required for adequate long-term coating per-
formance.
Surface roughness (or ‘anchor profile') is measured as ‘peak to trough'

height using wax replication tapes (e.g. Testex tape) or visual comparators.
Much better coating performance will be obtained by use of angular (not
spherical, e.g. shot) blasting media.

7.8.4.1 Paints
Painting is one of the basic methods used to prevent corrosion of a metal by

aggressive constituents in its environment. Most commonly, paints are used
for protection against atmospheric corrosion caused by oxygen in the air.
They achieve this by providing a strongly-adhering physical barrier which is
resistant to oxygen, moisture, and chemical ion permeation. Paints may also
prevent corrosion by sacrificial cathodic protection, surface film conversion,
and inhibition.
Painting also fulfills an aesthetic role, and may be used to enhance safety and
ease of maintenance by colour coding. Speciality coatings are also available
for use as linings in immersion service (e.g. potable water, aviation fuel), for
high temperature service, and in special applications such as anti-skid
floors.
Paints are normally classified according to their binder. The types most
commonly used in EP are epoxies, chlorinated rubbers, vinyls, silicones and
inorganics.
Paint is generally applied by either spray (conventional or airless), brush or

roller. The curing of paint film to form a solid barrier often involves solvent
evaporation together with a chemical reaction. Local weather conditions
(temperature, humidity, wind) will affect paint application scheduling - even
feasibility -and due consideration for seasonal conditions should be made
during project planning.
Blasting and painting operations often involve toxic materials and careful
consideration must be given to personnel protection.
Painting is usually an investment to protect an asset. Life-cycle economics

should be considered in selecting the paint system and preparation clean-
liness level. Evaluation of initial costs should reflect the fact that the costs of
preparation, personnel mobilisation, and application generally greatly ex-
ceed the cost of the paint itself.
To ensure satisfactory coating performance, thorough inspection by qualified
personnel is required during all stages of surface preparation and coating
application.
Further reading:
- DEP 40.48.00.30. Paints and paint materials.
- DEP 30.48.00.10. Painting and coating.
- DEP 30.48.00.31. Painting and coating for new construction projects.

7.8.4.2 Pipeline External Coatings
External pipeline coatings are used in conjunction with cathodic protection to

control external corrosion of subsea and buried pipelines.
The main factors influencing coating performance are moisture impermea-
bility, adhesion, cohesion, flexibility, chemical and physical stability, elec-
trical resistance, and resistance to impact, creep, weathering, and cathodic
disbondment.
Proper surface preparation prior to coating application is vital for satisfactory
long-term coating performance and the same methods and standards as for
painting (see previous Section) are used. Priming and coating should be
carried out as soon as possible after surface preparation to avoid recon-
tamination and rusting. This is particularly important in field jointing and
repairs. Special applicaton equipment is needed for some coatings neces-
sitating factory/yard application. Factory application is generally preferred
over site application since preparation and conditions can better be con-
trolled.
Materials used for pipeline coatings, and their maximum permissible work-
ing temperatures, are given in the Table below:
Table 7-5 Maximum temperatures for pipeline coatings
Note: Not yet established for PP, expected to approach 100ºC
Installation, inspection, and associated aspects of pipeline coatings are dis-

cussed in the ‘Pipelines' Section of this handbook, Vol. 8.
Further reading
- L-4-1/2/3. External polyethylene coating for line pipe.
- L-5-1/2/3. External fusion-bonded epoxy powder coating for line
pipe.

7.8.4.3 Internal Coatings for Pipelines and Downhole Tubulars
Internal coatings are sometimes applied to pipelines as ‘flow coats' to give a

smoother wall and hence reduced flow friction. Their use may provide cost
benefits by reducing the pipeline diameter and/or pumping requirements to
achieve the desired flow rate.
Internal coatings are not generally successful as corrosion protection, par-

ticularly in pursuit of a ‘zero leaks' philosophy. Such coatings generally fail in
a localised manner by, for example, blistering. Corrosion and subsequent
leakage still occurs, but at a reduced incidence rate and in a correspondingly
localised manner.
Despite their shortcomings as a corrosion prevention method, internal coat-
ings can be beneficial, particular in downhole tubulars, since the localisation
of the corrosion damage reduces the likelihood of the tubing parting down-
hole on retrieval during workover. Coating of downhole tubulars may also
enable increased re-use, after careful inspection, selection, and re-coating
as necessary.
7.8.5 Cements
Cementitious materials are used, not only in the familiar structural concrete,
but in diverse applications such as pipe linings (where recent experience has
often been unsatisfactory), weight coatings, and some fire protective coatings.
Further reading:
- DEP 30.48.30.31. Cement lining of new pipelines.
- DEP 31 .40.30.30. Concrete coating of linepipe.

7.9 Corrosion Monitoring
Various types of probes can be installed to measure corrosion rate on-stream

to obtain information for use in understanding corrosion and monitoring the
effectiveness of control measures:
7.9.1 Corrosion Coupons
These are samples of the steel concerned, which are mounted in the system
in such a way that they are exposed to the corrosive environment inside the
equipment. They are retrieved at regular intervals to be examined for weight
loss and depth of pitting corrosion. Mounting is done with the aid of a
threaded nipple, which can also be used for on-stream retrieval by means of
special tools (Fig. 7-19). Often they consist of a rectangular piece of steel
which is mounted perpendicular to the wall of the equipment. For pipelines,
special coupons which are mounted flush with the wall of the pipe are also
available.
7.9.2 Electrical Resistance Probes
These probes consist of steel sample, often in the form of a wire (Fig. 7-18) or
a strip of metal of which the electrical resistance is measured at intervals of
days or weeks. This can be done on-stream, without retrieval of the probe.
Corrosion will make the probe thinner, which can be detected as an increase
in electrical resistance. In order to compensate for the effect of temperature,
a reference wire is incorporated in the probe which is embedded in a resin to
prevent its corrosion.

7.9.3 Polarisation Resistance Probes
The application of a small current on a corroding piece of metal results in a

corresponding change in potential (Fig. 7-20). When the corrosion current
(see Section 7.2.2) is high, this perturbation will have relatively little effect,
and the change in potential will be small. On the other hand, for a system with
a low corrosion rate, the potential change will be relatively high. The quotient
of potential change and current applied is called the polarisation resistance,
which is inversely proportional to the corrosion rate.
In practice, a voltage is often applied between two electrodes, which changes
the potentials of both, but in opposite directions, and the current is measured.
In theory, the measurement can give a corrosion rate in a matter of minutes.
In practice, misleading readings are often obtained e.g. because of the pre-
sence of electrochemical reactions other than the corrosion reaction. When
H2S is present, FeS is formed which can electrically short the electrodes.

7.9.4 Other Techniques
Many other probes exist, based on e.g. hydrogen permeation, electrochem-

cal noise, electrochemical impedance, conductivity or potentials. They are,
however, less well established for routine application. Instead of a probe, a
removable test section is sometimes built in, equipped with a bypass system
which enables removal and inspection of this section without having to shut
down the operation.
7.9.5 General Remarks
All probes have in common that they only give information pertinent to the
location of the probe, and extrapolation of this to the equipment as a whole
can lead to serious errors. Low or zero corrosion rate readings are often
caused by incorrect installation or malfunctioning of a probe, and should
never be accepted at face value.
Better coverage can be obtained with tools like calipers (for tubing) and
‘intelligent pigs' for pipelines. The latter are based on the detection of metal
loss using magnetic principles or ultrasound. Rapid developments are taking
place in this area, however, at present ‘intelligent pigs' are better suited to
evaluate system integrity than corrosion rates.

Inspection Techniques and Maintenance Terminology
Chapter 8
INSPECTION TECHNIQUES AND

MAINTENANCE TERMINOLOGY
CONTENTS
8.1 Non-Destructive Methods for Evaluating Materials ............................................................................ 8-2

8.2 Equipment Identification ................................................................................................................... 8-14
8.3 Materials ........................................................................................................................................... 8-15
8.4 Maintenance Definitions and Terminology ....................................................................................... 8-16
8.5 The Job............................................................................................................................................. 8-19
8.6 Reporting and Analysis .................................................................................................................... 8-20
8.7 Units of Measurement ...................................................................................................................... 8-24
TABLES
Table 8-1 Non-destructive methods for evaluating materials 8-2
FIGURES
Figure 8.6-1 Service life and its components 8-21

8 INSPECTION TECHNIQUES AND MAINTENANCE TERMINOLOGY
8.1 Non-Destructive Methods for Evaluating Materials
Thirty methods are summarised in Table 8-1.
(continued on next page)












The following Sections 8.2-7 contain a glossary of maintenance terms.
8.2 Equipment Identification
Accessory. A part, subassembly, or assembly designed for use in conjunc-

tion with or to supplement another assembly or unit.
Component. Any self-contained part, Combination or parts, subassemblies,
or units which perform a distinctive function necessary to the operation
of a system.
Expendable. An item which, after use, cannot be economically restored to a
serviceable condition.
Interchangeable. Capable of being interchanged. An interchangeable part,
subassembly, assembly, or unit which meets or exceeds required func-
tional and structural specifications for a given application. It has the
same mounting provisions. Overall dimensions and connections may
vary provided that installation does not require rework or modifica-
tion.
Item. Any level of hardware assembly (that is, part, component, subsystem
or system).
Item, Maintenance Significant (MSI). Maintenance items that are judged to
be relatively the most significant from a safety, reliability, or economic
standpoint. (These are also known as Key or Vital items.)
Item, Structurally Significant (SSI) for offshore platforms, etc. Local areas of
primary structure which are judged to be relatively the most important
from a fatigue, corrosion vulnerability, or failure effects standpoint.
Module. A combination of assemblies, subassemblies, and parts contained
in one package, or so arranged as to be installed in one maintenance
action.
Part. One piece, or two or more pieces joined together which are not nor-
mally subject to disassembly without destruction of designed use.
Parts list. A list of the parts that make up an item of equipment or subas-
sembly.
Repairable part. An item which is repaired after usage and reused rather
than discarded because it is more economical to do so.
Part number. A unique number assigned to each part and component.
Serialised part. An item which has, in addition to a specific part number, an
individual serial number, enabling the maintenance department to iden-
tify each piece uniquely.
Spare part. An item which may be replaced during maintenance.

Spares. The individual items held (in addition to those installed already) for
the purpose of providing replacements for those removed for overhaul,
repair, or rectification.
Subassembly. Two or more parts which form a portion of an assembly or a
unit replaceable as a whole, but having a part or parts which are indi-
vidually replaceable.
Subsystem. A major functional part of a system which contributes to oper-
ational completeness of the system.
System. A combination of interrelated items arranged to perform a specific
function.
Unit. A part or any combination of parts, mounted together, normally cap-
able of independent operation.
8.3 Materials
Material cost. The cost of all materials expended in the maintenance

task.
Economic order quantity (EOQ). The quantity of an item that it is convenient
and economical to order.
Lead time. The period of time from the order request for items to their
expected availability on site.
Note: Materials ordering procedure time, manufacturer's delivery time, trans-
portation, custom's clearance, etc. should be considered when estimating a lead
time.
Storage life. The length of time an item can be stored under specified
conditions and still meet specified requirements, also called `shelf
life'.
Maintenance materials. Items required for maintenance work that cannot
or should not be reused, such as concrete/asphalt, split pins/self locking
nuts, gaskets, penetrating oil, etc.
Packing list. The-storekeeper's list of the parts and materials required for
work orders.
Reorder level. A minimum stock quantity fixed for an item to indicate when
a new order must be placed.
Service level. A probability measurement indicating the probability that a
requisition order can be immediately filled.

8.4 Maintenance Definitions and Terminology
Maintenance. Those actions required for restoring or maintaining an item

in serviceable condition, including servicing, repair, modification, over-
haul, inspection and determination of condition.
Note: The actions referred to are those associated with initiation, organisation
and implementation.
Maintenance resources. Facilities, support equipment, manpower, spares,
consumables, and funds available to maintain and support an item in its
operational environment.
Maintenance management. The organisation of maintenance within the
agreed policy of the plant or company.
Maintenance policy. A strategy within which decisions on maintenance are
taken.
Breakdown maintenance. Maintenance required to put equipment back
into production following an unexpected halt. Also known as corrective
maintenance.
Ad hoc maintenance. Maintenance which is necessary to carry out imme-
diately to avoid serious consequences. Also known as emergency main-
tenance.
Preventive maintenance. Maintenance carried out at predetermined inter-
vals, or to other prescribed criteria, and intended to reduce the likeli-
hood of an item not meeting an acceptable condition.
Planned maintenance. Maintenance organised and carried out with fore-
thought, control and the use of records to a predetermined plan.
Note: Preventive maintenance is normally planned. Breakdown maintenance
may or may not be planned.
Predictive maintenance. A technique that involves monitoring the opera-
tion of equipment so that potential defects can be identified.
Maintenance planning. Determining in advance the parts, materials, tools,
methods,manpower, and crafts required for identified maintenance jobs.
Maintenance scheduling. Allocating manpower, materials and equipment
to planned and unplanned maintenance jobs according to established
priorities and production schedules.
Maintenance concept. Generalised maintenance management ideas and
general maintenance principles which are studied and then developed
into a specific plan for application through a particular maintenance
programme.
Maintenance programme. A list allocating specific maintenance to a spe-
cific period.
Maintenance schedule. A comprehensive list of maintenance and its inci-
dence.

Maintenance manual. Collated information and advice on the maintenance

requirements of an item.
Condition-based maintenance. Preventive maintenance initiated as a re-
sult of knowledge of the condition of an item from routine or continuous
checking.
Running maintenance. Maintenance which can be carried out whilst the
item is in service.
Shut-down maintenance. Maintenance which can only be carried out when
the item is out of service.
Servicing. The replenishment of consumables needed to keep an item in
operating condition.
Note: This was sometimes used in the broader context of maintenance but this is
not now a recommended practice.
Block maintenance program. A programme which divides major structural
inspections and/or maintenance tasks into groups, or blocks, which
permit convenient, economical, and effective accomplishment.
Continuous maintenance program. A type of complete maintenance pro-
gramme which is expected to assure continuous availability of the fa-
cilities. Under this system the total maintenance effort is apportioned to
each of the various and more frequent types of maintenance. A complete
overhaul at one point in time is not a part of a continuous maintenance
plan.
Equalised maintenance program. A maintenance programme whereby
work packages are scheduled for accomplishment in such a manner that
the required maintenance manpower will remain relatively constant.
Work programme. A programme which defines a logical sequence of main-
tenance actions to be performed as events or pieces or a whole which,
when performed collectively, result in achievement of the desired main-
tenance standards.
Acceptable condition. That condition agreed for each particular usage. This
condition shall be not less than that demanded by statutory require-
ments and safety considerations (if not covered entirely by the statutory
regulations).
Breakdown. Failure resulting in the non-availability of an item.
Burn-in. The operation of an item to stabilise its characteristics, also called
`run-in'.
Calibration. The application of specifically known and accurately meas-
ured input to ensure that an item will produce specifically known output
which is accurately measured or indicated. Periodic calibration checks
and corrections (if any) should be duly recorded.
Craft. A trade or occupation that requires the use of specific manual and
technical skills.

Craftsman. A person trained in a particular craft.

Defect. Any confirmed abnormal condition of an item whether or not this
could eventually result in a failure.
Examination. A comprehensive inspection supplemented by measurement
and physical testing in order to determine the condition of an item.
Failure. The termination of the ability of an item to perform its required
function.
Failure report. A document reporting departure of an item from an accept-
able condition.
Inspection. The responsibility of audit and evaluation of the physical con-
dition of all operating plant, equipment and facilities and to keep oper-
ational and technical management fully informed of these conditions on
a continuous basis.
On stream inspection (OSI). A careful and critical scrutiny or inspection of
equipment supplemented by measurements or physical tests carried
out whilst an item remains in service and without dismantling. (Also
known as in service inspection).
Sample inspection. The monitoring and/or withdrawal of selected devices
from service to permit determination of their condition at predetermined
progressive intervals.
Visual inspection. A careful and critical scrutiny of an item carried out
without dismantling, by using the senses of sight, hearing, smell, taste
and touch.
Inspection report. The written report of an inspection or examination which
may or may not include a recommendation for actions (also known as
survey (report)).
Malfunction. The occurrence of a condition whereby the operation of an
item is outside of specified limits.
Overhaul. A comprehensive examination and restoration of an item, or a
major part thereof, to an acceptable condition.
Partial overhaul. The overhaul of a subassembly of an item with a time-
controlled overhaul to permit the longerlived item to achieve its autho-
rised overhaul life.
Rehabilitation. Extensive maintenance intended to bring plant, equipment
or buildings up to current acceptable conditions, often involving improv-
ements (colloquially ‘revamp').
Modify. To change or alter through rework and/or through the installation
or removal of an item (smaller works not as extensive as rehabilita-
tion).
Repair. To restore an item to an acceptable condition by the renewal, re-
placement or mending of decayed or damaged parts.

Restoration. Maintenance actions intended to bring back an item to its

original appearance or state.
Shut-down. Taking equipment out of production for modification or main-
tenance.
Test. To compare the response of an item with a suitable or defined ap-
plication of a stress, energy or measurement against a standard.
Troubleshooting. Locating and diagnosing malfunctions or breakdowns in
equipment by means of systematic checking or analysis.
8.5 The Job
Job instruction. A document describing the way in which the work is to be

carried out. It may define the materials, tools, time standards.
Work report. A statement recording the work done and describing the con-
dition of the equipment.
Work request. A report by operational or maintenance staff that indicates a
known or suspected malfunction or defect on an item of equipment and
requests maintenance intervention.
Remark. Those reports or log entries made by the operational staff for
information purposes only.
Permit to work. A signed document, authorising access to an item, which
shall define conditions, including safety precautions, under which work
may be carried out. This may include a document, signed on completion
of maintenance, stating that an item is safe and ready for use.
Work order. A written instruction detailing the maintenance work to be
carried out.
Removal. Any deliberate detachment of an item from plant. All removals
can be classified into one and only one of three categories: scheduled,
unscheduled, and maintenance convenience.
Removal, scheduled. A removal of an item brought about as a result of the
item's life having approached or achieved a previously defined limit,
also called `time removal'.
Removal, unscheduled. A removal of an item brought about as the result of
a known or suspected malfunction and/or defect.
Removal, maintenance convenience. A removal brought about for any rea-
son other than as a result of the item's life having approached or
achieved a previously defined limit or as a result of a known or sus-
pected malfunction and/or defect (that is, all removals other than sched-
uled and unscheduled removals), also called ‘planned removal'. The
following are types of maintenance convenience removals: access re-
moval, cannibalisation removal, engineering examination removal,
modification removal, etc.

8.6 Reporting and Analysis
Life cycle costs. The total cost of an item throughout Its life including initial,
maintenance and support costs.
Direct maintenance cost. Those maintenance labour and material costs
directly expended in performing maintenance on an item or plant. Does
not include those indirect maintenance labour and material expendi-
tures which contribute to the overall maintenance operations, servicing,
administration, record, keeping, supervision, tooling, test equipment
facilities and transportation.
Indirect maintenance cost. Those maintenance labour and material costs
not considered to be direct maintenance costs, but which contribute to
the overall maintenance programme through overhead operations, ad-
ministration, record-keeping, supervision, tooling, test equipment fad I-
ities and transportation. Also called `maintenance burden'.
Reliability. The ability of an item to perform a required function under
stated conditions for a stated period of time. This may be expressed as a
probability.
Reliability analysis. The assessment of probabilities of satisfactory perfor-
mance of an item over a given service period by means of statistical
studies.
Maintainability. A function of the rapidity and ease with which maintenance
operations can be performed to help prevent malfunctions of to correct
them if they occur.
History. Record of usages, events and actions as appropriate relating to a
particular item. (Maybe kept on cards or computerised as appro-
priate.)
Feedback. A written or verbal report of failures, defects, shortcomings and
other experience which can be used to influence design, performance
and costs.
Deterioration rate. The rate at which an item approaches a departure from
an acceptable condition.
Availability. The fraction of a specified period for which an item is capable
of performing a specified function at a specified level of perfor-
mance.
Down time. The period of time during which an item is not in a condition to
perform its intended function.
Utilisation factor. The ratio of the period of actual use to availability.
Physical asset register. A record of items, including information such as
constructional and technical details about each. This may be combined
with an inventory or history also known as equipment data base.
Life. A period of time which is related to the usability of an item.

Useful life. The length of time a population of items is expected to operate

with a constant failure rate, also called `constant period' or `failure rate'.
This excludes any infant mortality and wearout periods.
Infant mortality period. That early period, beginning at zero unit time, dur-
ing which the failure rate of a family of-items can be expected to de-
crease.
Wearout. The process of deterioration which results in an increase of the
failure rate with increasing age.
Service life. The life of-an item at which it is no longer physically or eco-
nomically feasible to repair or overhaul the item to acceptable stan-
dards (see Figure 8.6-1).
Achieved overhaul life. The life achieved by an item when an overhaul

becomes necessary.
Maximum permitted life. The time specified by an appropriate authority
after which a particular item must be removed from service, also called
‘limited life' or ‘declared life'.
Failure rate. The performance figure calculated by dividing the number of
failures by the total unit hours or cycles accumulated during the same
period. It is usually expressed as failure per 1000 unit hours or cycles.
Failure rate is the reciprocal of mean time between failure.
Time. Any appropriate unit of life measurement such as hours, cycles,
etc.
Job standard time. The time allowed for a given task, sometimes referred to
as target time.
Mean maintenance time. The statistical mean of the distribution of pre-

ventive or breakdown (or both) maintenance times. The summation of

the durations of preventive or breakdown (or both) maintenance time
during a given period divided by the total number of preventive or
breakdown maintenance actions (or both) during the same time period
s an estimate of mean preventive maintenance time (MP), mean break-
down maintenance time (MS), or mean maintenance time (M).
Mean repair time. The arithmetic mean of the times to repair a given item.
The cumulation of active repair times during a given period of time
divided by the number of repaired items during the same time
interval.
Mean task time. A representative task time, equal to the summation of task
times required to perform a specific task a number of different times
divided by the number of times performed.
Total time. The operating time an item has accumulated since new.
Turnaround time. That time needed to repair, service, or check out an item
or plant for recommitment to operational service.
Time between overhauls (TBO). The maximum time that an item is permit-
ted to operate between overhauls, also called òverhaul period'.
Time since check. The time accumulated since an item was last
checked.
Time since installation (TSI). Time accumulated since a given item was last
installed.
Time since overhaul (TSO). The time accumulated since a given item was
last overhauled.
Man-hours, direct. Man-hours of work spent directly on the equipment or
removed items.
Man-hours, maintenance. The total man-hours required to complete the
maintenance task. (Including preparatory work, shutting down through
restarting and testing.)
Man-hours per operating hour. A performance figure calculated by dividing
the direct man-hours expended to maintaining a particular process or
plant during a given period by the operating hours during that pe-
riod.
Mean time between failure (MTBF). A performance figure calculated by
dividing the total unit operating hours accrued in a period by the number
of unit failures that occurred during the same period.
Mean time between maintenance. The arithmetic mean of the time inter-
vals between maintenance actions.
Mean time between unscheduled removals (MTBUR). A performance fig-
ure calculated by dividing the total unit operating hours accrued in a
period by the number of unscheduled unit removals that occurred during
the same period.

Mean time to failure (MTTF). A performance figure calculated by dividing

the summation of times to failure for a sample of failed items by the
number of failed items in the sample. This is different from mean time
between failure because no allowance is made for experience of items
which have not failed.
Mean time to repair (MTTR). A performance figure calculated by dividing
the active repair times accrued in a period by the number of repaired
items in the same period.
Redundancy. The existence of more than one means for accomplishing a
given function. The alternative means of accomplishing the function
need not necessarily be identical.
Active redundancy. That redundancy wherein all redundant items are oper-
ating simultaneously rather than being activated when needed.
Standby redundancy. That redundancy wherein the alternative means of
performing the function is inoperative until needed and is activated upon
failure of the primary means of performing the function.
Deferability. That period of time for which an item may be inoperable or
deficient within the terms of the Minimum Equipment List (MEL) or Con-
figuration Deviation List (CDL).
Minimum equipment list (MEL). An approved list of items which may be
inoperative under specified conditions, also called `go/no-go items,'
àllowable deficiency list,' or `dispatch inoperable list'.
Configuration deviation list (CDL). Those items, such as access panels,
covers, or fencings, normally forming part of the exterior profile of the
item or plant, the absence of which does not prevent operation.
Actuarial program. A statistical technique which uses historical data to
depict the life expectancy of an item with respect to age.
Active maintenance downtime. The maintenance downtime during which
work is being done. From the time the item or plant is made available for
preventive or corrective maintenance to the time the item or plant is
back in operation at its specified performance.
Engineering change order. An order for a physical change in a component,
which is initiated by the engineering department, by the manufacturer of
the part, or by the appropriate authority to improve the component's
reliability or performance.
Out of service hours. The number of elapsed hours that an item or plant is
not available.
Removal rate. The number of removals of an item expressed in terms of a
base period: maybe per 1000 hours, per year, or as appropriate.
Scrap rate. The number of items becoming beyond the limits of economical
repair, expressed as a quantity per unit of time.

8.7 Units of Measurement
hertz (Hz) = An SI unit of frequency, e.g. vibration, of one cycle per

second
micrometre (µm) = An SI unit of measurement of one micrometre or
-6
1.0 x 10 metres
mil (mil) = An Imperial unit of measurement of one thousandth part
-3
of on inch or 1.0 x 10 inches
(See Volume 1 for details on units and conversion factors.)

Diving and Underwater Operations
Chapter 9
DIVING AND UNDERWATER

OPERATIONS
CONTENTS
9.1 Introduction......................................................................................................................................... 9-2

9.2 Selection of Method of Intervention and Planning the Underwater Work ......................................... 9-2
9.2.1 Oceanographical and Meteorological Parameters ........................................................................... 9-2

9.2.2 Geographical and Political Parameters ........................................................................................... 9-3
9.2.3 Intensity of the Diving Operations ................................................................................................... 9-3
9.2.4 Other Factors................................................................................................................................... 9-3
9.3 Identification of Tasks........................................................................................................................ 9-3
9.3.1 Drilling Support ............................................................................................................................... 9-4

9.3.2 Construction Diving ........................................................................................................................ 9-4
9.3.3 Tasks .............................................................................................................................................. 9-4
9.4 Operational Limitations....................................................................................................................... 9-6

9.5 Divers, Unmanned and Manned Submersibles.................................................................................. 9-6
9.6 Support Base...................................................................................................................................... 9-7
9.7 Support Vessels and Craft ................................................................................................................. 9-8
9.7.1 Construction and Maintenance Vessels .......................................................................................... 9-8

9.7.2 Dynamic Positioning ........................................................................................................................ 9-8
9.8 Types of Diving................................................................................................................................... 9-8
9.9 Composition of Diving Team ............................................................................................................. 9-9
9.10 Typical Diving Equipment................................................................................................................ 9-9
9.10.1 General.......................................................................................................................................... 9-9

9.10.2 Power Supplies............................................................................................................................ 9-10
9.10.3 Handling Systems ....................................................................................................................... 9-10
9.10.4 Communications.......................................................................................................................... 9-11
9.11 Regulations and Responsibilities .................................................................................................. 9-11
9.11.1 Regulations.................................................................................................................................. 9-11

9.11.2 Responsibilities............................................................................................................................ 9-12
9.12 REFERENCES AND FURTHER READING .................................................................................. 9-14

9 DIVING AND UNDERWATER OPERATIONS
9.1 Introduction
Intervention work in support of underwater operations may be carried out by

a number of methods including remotely operated vehicles (ROVs) and hy-
perbaric diving.
Diving services are always contracted (never undertaken by Shell personnel)

and where there is limited local experience, consultation with SIPM should
be undertaken at the earliest opportunity. SIPM report EP89-0777 ‘Under-
water Handbook - Diving' is the Group reference document on the subject.
For background information on the various intervention methods, see Ref-

erences and Further Reading 9.12.
It should be noted that the UK legislation on diving is used as a minimum base

case for underwater operations using divers, unless there are more stringent
local requirements.
9.2 Selection of Method of Intervention and Planning the Underwater Work
Once the decision to undertake essential subsea work has been made and a
scope of work developed, it will be necessary to assess the possible alter-
native methods for accomplishing the work from both practical and economic
standpoints. To do this, the parameters under which the work is to be per-
formed, must be defined. Careful consideration of these factors before de-
tailed planning and bar charts are drawn up will invariably lead to a more
effective underwater operation.
The success of an underwater engineering operation will often be influenced

by choice of the correct underwater tools for the job; the UEG Handbook of
Underwater Tools (Ref. 2) is recommended in this respect. The factors to be
considered are described below.
9.2.1 Oceanographical and Meteorological Parameters
The depth of water, its temperature and underwater visibility, the time of year
and prevailing weather conditions, tidal streams and currents will all influence the method
chosen. An air diving operation in 50 m of very cold

water with no visibility and strong currents will differ considerably from one
undertaken at 15 m depth in warm, clear and tranquil waters.
9.2.2 Geographical and Political Parameters
The area of the world, location of potential support bases, logistics of supply
and the import and entry restrictions of a country may influence the method.
9.2.3 Intensity of the Diving Operations
This factor is very important and may for instance lead to a situation where a
‘shallow' saturation dive may be more economical and practical than a pro-
tracted surface air diving operation. This is especially true for complex work
in the 20-50 m depth range.
9.2.4 Other Factors
In planning diving operations, care must be taken to avoid periods of inten-

sive offshore activity that are close to and therefore likely to prevent or
interfere with the work. These activities include seismic surveys, piling oper-
ations and scaffolding or construction programmes where there are likely to
be hazards from falling objects. In these situations, underwater activities are
best undertaken by ROVs, rather than divers.
Many other `local' factors may affect the efficiency of the operation and where
known, should be communicated to the diving contractor, e.g. dangerous
marine animals, heavy marine fouling or debris.
It is important to ensure that divers and ROV pilots receive a thorough brief-
ing before they start the work, in order that they clearly understand their task
and the underwater equipment and environment which they will encounter.
Part of this briefing should include study of records of previous dives at the
same locality or for a similar task/situation. Detailed and accurate records of
all dives should thus be maintained to help reduce the work-load and increase
the safety of future dives.
9.3 Identification of Tasks
The underwater work to be done will be the basis on which the means of
intervention is selected, taking into account the relevant health, safety, en-
vironmental and cost aspects.

9.3.1 Drilling Support
Points to consider include:
- The depth of water and what type of drilling rig is to be used, jack-up,
drilling ship or semi-submersible drilling rig.
- Advancing technology, including the ability to drill, install and change out
sea bed installations in very deep water, is reducing the requirement for
divers. Whether or not diving support for these operations is really nec-
essary, should be very carefully considered.
9.3.2 Construction Diving
The construction phase of the gas or oil field often requires diver intervention
and it is essential that relevant expertise should be integrated into the project
at the earliest possible time. This should preferably be during the design
phase of the structure and certainly during its building. Care taken at the
design stage to place identifying marks, work stages and rigging aids at
salient points on the structure, will result in considerable simplication of the
underwater tasks and speed up their completion. Diver or ROV operated
equipment, such as grouting connections, should be positioned in easily
accessible positions.
The depth of water in the field and possibly on the pipeline route to shore is a
significant consideration when choosing the underwater support required.
The field may require only air or bounce mixed gas diving, while the pipe-
laying support may require saturation diving. An ROV may be capable of all
or part of the work.
9.3.3 Tasks
These may include:
(a) Pipe Laying Support

Diver support in these conditions will be restricted to work on the stinger, but
in shallower depths, pipe inspection and valve operations for flooding and
de-watering the pipe may well be required.
(b) Trench Barge Support

In the event that trenching of pipe is necessary, underwater operations to
monitor the progress of the sled and check that it is working correctly may be
required. Diving work is mainly on the sea bed, so any diving system must be

capable of supporting the divers at the maximum depth of the pipe route.
Checking the sled, pipe and trench is a daily operation which can be executed
by an ROV.
(c) Platform Installation
The use of ROVs to survey and inspect the sea bed prior to the installation of
the platform and the type of vessel to be used for underwater support during
the platform installation should be carefully considered.
(d) Pipeline and Risers Tie-Ins
The type of tie-in to be used for pipelines and risers should be decided at an
early stage of the development of the field; if hyperbaric welding is specified,
the long lead time for selection, training, certification and acceptance of the
welders should be taken into account.
(e) Underwater Inspection and Maintenance
The inspection and maintenance of an installation will be greatly reduced

and essential work eased by proper planning. The access requirements to a
platform by vessels such as derrick barges and accommodation craft and
their effect on diving operations must be carefully considered. Other access
problems of diving from support ships close to the platform are:
- Deep horizontal penetration of the platform

- Acute angles of nodes
- Craft temporarily moored alongside for logistics purposes and the wires
used to secure them
- The strength of the tidal streams and currents; it may only be possible to
dive at slack water in some areas and at certain times of the month in other
areas. Strong currents will necessitate diving on the protected side of the
platform to prevent being swept underneath it.
- Construction work or crane operations immediately above the dive site.
The depth of the platform and the scope of work will determine the range of
diving equipment required. Except in very shallow water, a full range from air
to saturation diving facilities may be required, although an increasing pro-
portion of such work is carried out by ROV. Inspection and maintenance
programmes may depend on the legal requirements of the country in which
the Company is operating and the relevant Certifying Authority as well as
those of the Company itself.
(f) Repair
Repairs may be required on any part of a field or its associated pipelines. The

depth of water and the complexity of the remedial work necessary will de-
termine the type of diving equipment required. Tasks include:
-Location of the damaged area

- Visual, still photographic and video inspection of the damaged areas
- Removal of concrete cladding, cutting and removal of damaged pipe
- Measurement of length of new pipe or spool piece required
- Placement and jointing of spool piece (flanged connections, hyperbaric
welding, or other connections)
- Concrete repair on structures
- Wellhead intervention
- Mooring installations
- Storage installations.
Repairs have tended to require extensive diving operations and in only rare
instances have they been completed by ROVs.
9.4 Operational Limitations
The geography of the area will pose the main operational limitations of the
underwater operation in the development and should be carefully con-
sidered. The availability of local resources and the capability of local per-
sonnel should also be assessed.
When planning underwater operations, it should always be remembered that

generally divers can descend without significant restriction but ascent can be
very slow; therefore work required in the vertical plane will be completed
more effectively if undertaken during compression of the divers rather than
decompression.
9.5 Divers, Unmanned and Manned Submersibles
The diver plays an essential role in the development of many offshore fields
and this is expected to remain so for some years. However, the advancement
of technology is producing rapid changes in underwater activities. Some of
these changes have been forced by the steady progress into very deep water
beyond effective diver capability, i.e. deeper than 300-400 m.
Alternatives to Divers
Amongst the alternatives to divers that may be considered for underwater
intervention are:

- Remotely operated vehicles (ROVs)
- Observation submersibles
- One man, one atmosphere submersibles
- Atmospheric diving suits
- Mobile observation bells.
A small observation ROV, operating as a mobile TV camera with its own

lights, should be considered as an aid to diver safety and efficiency.
9.6 Support Base
The support base is a most important part of the pre-planning necessary

before operations start. It is quite likely that in some parts of the world the
closest established base from which the operation can be run or supported
may be hundreds of miles away from the work-site, while in other areas it
may be proposed to run a diving spread from a home base thousands of miles
away. This is unlikely to be successful if the operation is to last any length of
time, but it may be reasonable to support a short duration programme in this
manner. A construction spread requires support as near as is practical both
for logistics and management. For example, hyperbaric welding requires
plentiful supplies of gas and other back-up equipment.
Geography may well be a decisive factor in the choice of base, e.g. the most
operationally convenient base may not be acceptable politically.
When establishing the support base, some of the factors which should be
taken into consideration are:
- The type of docks or jetties available and the depth of water in the harbour.
An alongside berth with good cranage is ideal, but it may be necessary to
move stores and equipment on intermediate craft from the shore to the
supply vessels in the harbour, or even outside the harbour.
- The requirement to set up good lines of communication between the shore
base, the field, the Company headquarters and emergency services.
- The presence of other diving operations in the area. If possible, the oper-
ators should be approached to establish contacts, background information
and for mutual support, emergency procedures etc.
- A smooth arrangement must be made for the clearance of financial trans-
actions, imported goods, or the re-export of goods from the chosen base to
the work site. Delays in the delivery of diving gas, for instance, could prove
very costly.
- In remote locations it may be necessary for the Diving Contractor to pro-

vide a deck decompression chamber at the shore base. Consideration
should be given to the use of this chamber in an emergency situation and to
the provision of rapid means of transporting divers to it.
9.7 Support Vessels and Craft
The selection of the diving or ROV support vessels is usually critical to

successful underwater activities. See also Reference 1.
9.7.1 Construction and Maintenance Vessels
Vessels and craft used for construction and maintenance range from simple
barges equipped with a small crane, to purpose-built, multi-purpose semi-
submersibles (MSV), with in-built diving spreads. Selection of craft will de-
pend principally on the scope of work, weather conditions and economic
considerations.
9.7.2 Dynamic Positioning
In areas where anchoring may be undesirable or hazardous to subsea equip-

ment or where short duration diving tasks need to be undertaken in nume-
rous different locations, the use of a dynamically positioned diving support
vessel (DPDSV) may be extremely advantageous and cost effective. Selec-
tion and operation of these vessels is complicated due to stringent safety and
redundancy requirements and to the great variation of performance between
different vessels, and personnel.
Companies wishing to charter such ships for the first time should seek advice
from SIPM, but as a basic guide the vessel chosen should at least comply with
the `Guidelines for the Specification and Operation of Dynamically Positioned
Diving Support Vessels' published jointly by the UK Department of Energy
and the Norwegian Petroleum Directorate in 1983 (Ref. 3).
9.8 Types of Diving
The two main types of diving are:
- Surface-orientated, in which one or two divers leave the surface, work for
short periods at depths to a maximum of 50 metres, and then return to the
surface after carrying out essential decompression.

- Saturation diving, in which teams of 2-20 divers may live `saturated' in
compression chambers for up to 28 days. They may work up to eight hours
per day in the water, having been transported to the work site in a close
diving bell. The decompression time after saturation may be several days,
depending on the depth of dive.
9.9 Composition of Diving Team
All air diving operations should normally be conducted in the surface-sup-

ported mode, i.e. the diver supplied with air and communications via an
umbilical.
All diving below 50 m must utilise mixed gas (usually helium/oxygen) and a
closed diving bell system.
Considerable variation in the numbers making up a diving team will be found

offshore due to the different procedures, the task requirements and surface
support arrangements.
To support one surface-supported diver in the water, the minimum crew
should, however, consist of one Supervisor and five other diving team mem-
bers.
A fully dressed standby diver should always be in attendance, where a diver is

in the water. (For comfort the standby diver may not actually be wearing his
helmet or weight belt.) In bell diving, the bellman will usually act as the standby diver.
Minimum team layouts will vary according to scope of work and operation
parameters. Examples are detailed in Reference 1 -.Report EP89-0777 ‘Un-
derwater Handbook - Diving'.
9.10 Typical Diving Equipment
9.10.1 General
Reference is made to the SIPM Ùnderwater Handbook- Diving' (EP89-0777)

Appendix 17 for details of typical diving plant and equipment to support each
type of diving. Diving systems in their entirety should comply with the cer-
tification requirements of a major Classification Society (e.g. Ref. 4).
Amongst the types of tests required are the following:
- Pressure tests of vessels, e.g., deck decompression chambers, bells

etc.

- Load tests of handling systems, e.g., winches, À' frames, gantries, wire
ropes etc.
- Leak and pressure tests of pipes, valves and pipe fittings
- Pressure tests of gas cylinders.
Test certificates or copies thereof should be available for inspection on site

and must be in date.
9.10.2 Power Supplies
Electrical power may be required for the handling system, life support equipment,
hot water supply unit, LP and HP compressors etc. and may be 20 to 150
kVA from a 380 or 440 V 3-phase supply.
9.10.3 Handling Systems
Handling systems will be necessary for handling diving baskets, ROVs or

Submersible Compression Chambers (SCCs) and must have adequate re-
dundancy of equipment and power supplies. When the system is mounted on
a ship shape vessel, it must have sufficient reach to prevent the basket, bell
or anchor weight (if any) from colliding with the vessel side during rolls.
They may comprise:
- Overside systems, using an ‘A' frame or a supporting beam with a foldback
facility
- Guide wire weight and constant tension winches
- Through moonpool system using a main frame, travelling trolley, lifting
winch, cursor system (active or passive) and guide wires controlled by
constant tension winches
- A rail or tube system running down a column of a semi-submersible craft
and including various arrangements of sub-sea launch.
An important aspect of any handling system is the umbilical. This may be a

composite umbilical, with an external strength member enclosing all the
services, (breathing gas, hot water, communications, power, etc.) and pro-
duced as a single smooth circular cable, or a ‘taped' umbilical. The compo-
site type of umbilical is normally used in conjunction with a constant tension
winch with automatic spooling and slip ring arrangement and therefore does
not require man handling. A ‘taped' umbilical contains the same services but
is taped together and is normally stored in a basket and either man handled
or deployed with the aid of a capstan.

9.10.4 Communications
Communications systems allow communication between the divers and the

Supervisor on the surface, between the divers themselves and also between
the Supervisor and the OIM or vessel Captain. Helium unscrambling units are
fitted at the control console.
9.11 Regulations and Responsibilities
9.11.1 Regulations
In view of the world-wide developments in Diving Safety Regulations, SIPM

recommends that all Group diving operations are conducted in accordance
with the United Kingdom Diving Operations Regulations, except where local,
national, or state regulations are required to take precedence. Where such
local regulations are enforced, the minimum standards required by the UK
Regulations should not be lowered.
The United Kingdom Diving Operations Regulations referred to above are

primarily:
- the Diving Operations at Work Regulations 1981 (see Ref. 5).
- the Merchant Shipping (Diving Operations) Regulations 1975.
The Diving Operations at Work Regulations 1981, were prepared by the

Health and Safety Executive under the United Kingdom Health and Safety at
Work Act, 1974 and came into force on 1st July, 1981. These regulations cover
offshore installations, pipelines and proposed pipelines in UK concessionary
waters. The Merchant Shipping (Diving Operations) Regulations 1975 remain
in force, with some modifications, to cover diving from British ships in other
parts of the world outside British Waters and foreign vessels within UK
territorial waters.
There are several important aspects of the above legislation which should be
understood by Managers and Engineers, as follows:
- The Regulations require that divers must work directly for the Diving Con-
tractor and not for some third party, thereby insisting that the Diving Con-
tractor is fully responsible for the divers that he employs.
- The Diving Supervisor of a diving spread is not to dive while he is in charge

of that spread.
- Every diving company must produce a set of Diving Rules. Copies of the
rules are to be readily accessible to all members of the diving team on site.

These Rules should be made available to the Offshore Installation Man-
ager (OIM).
- The requirement to provide a means for heating the diver at depths below
50 m and to provide a means of heating the diver's breathing gas at depth
below 150 m.
The Diving Inspectorate of the Department of Energy in the UK issue diving

safety memoranda as necessary. These supplement UK Regulations and are
circulated throughout the Group by SIPM.
9.11.2 Responsibilities
Offshore Installation Managers (persons in charge of installations and facil-

ities) and Masters of Diving Support Ships and Vessels must have a clear
appreciation of their own and others' responsibilities with regard to diving
activities.
In many countries, it is accepted that the owner of the concession, offshore
installation, pipeline, or craft from which, or in connection with which, diving
is taking place, has overall responsibility for ensuring that the duties laid
down in national diving legislation for various personnel are carried out. It
should be noted that in the UK legislation, the OIM is defined as the senior
representative of the concession owner on site, with reponsibility for the
operation. Therefore, the OIM, or the Master of Diving Support Vessel, car-
ries the overall responsibility.
The Diving Supervisor has the authority to stop all operations under his
control, should he consider them unsafe. The OIM does not have the authority
to override this decision; however, the OIM does have the overall appreciation of
all activities taking place at the work-site and does have the authority
to stop them, including diving. He should not hesitate to do so, if he foresees a
possibly dangerous situation arising. It is essential that a close liaison is
maintained between the OIM and the Diving Supervisor. Daily meetings of
competent persons involved in work on site are recommended.
It is vital that the vulnerability of divers in transit from the working platform to
the water and when below the water, from ship or crane movements and from
items dropped overboard, is fully appreciated by all persons at the worksite.
Care is essential when planning the positioning of diving installations. It is

even more important to recognise the hazards that may exist when diving
systems are embarked on a platform temporarily. The Diving Supervisor may
require assistance from the OIM to meet his responsibilities in providing the
necessary facilities.

Items which may affect the safety of the diver and which should be considered
by the OIM include:
- Fishing lines and debris
- Underwater noise, explosives, pile driving etc.
- Differential pressures caused by flooding pipes etc.
- Allowances for care and maintenance on equipment to be carried out
- Fitness to work and dive, including training
- Environmental conditions
- Diving platform stability
- Dynamic positioning (DP) of platform operating divers
- Vessel movements in the work area
- Electrical safety
- Working practices
- Conduct of parallel underwater work, i.e. when more than one spread is
working on one platform
- Emergency procedures
- Communications between the Diving Supervisor and his divers in the water
and in DDCs and between the Diving Supervisor and the OIM.
The careful application of an authorised Permit to Work system is essential to

all diving activities.

DIVING AND UNDERWATER OPERATIONS
References
1. SIPM Report EP89-0777 Underwater Handbook - Diving
2. UEG Handbook of Underwater Tools. UEG, London
3. Guidelines for the Specification and Operation of Dynamically Positioned

Diving Support Vessels, 1983. Issued by UK Department of Energy and
Norwegian Petroleum Directorate
4. Rules and Regulations for the Construction and Classification of Sub-

mersibles and Underwater Systems. Lloyds Register of Shipping (Lon-
don), 1989
5. UK Department of Energy, Statutory Instrument No.399. Health and Sa-

fety, Diving Operations at Work Regulations, 1981
Further Reading
- Shilling, C. S., Werts, M. F. and Schandelmeler, N. R., The Underwater

Handbook, Guide to Physiology and Performance for the Engineer. Plenum
Press
- The Principles of Safe Diving Practice, 1984. UEG, London
- Bennett, P. B. and Elliott, D. H., (Editors), The Physiology and Medicine of

Diving and Compressed Air WorkiBailliere. Tindall (London)

Air Conditioning
Chapter 10
AIR CONDITIONING
CONTENTS
10.1 Choice of Cooling/Dehumidifying Method ................................................... 10-2

10.2 Sizing of Facility (Cooling Load) ................................................................... 10-3
10.3 Data for the Estimation of Cooling Load.................................................... 10-10
10.3.1 Human Heat Generation, qH ................................................................... 10-10

10.3.2 Transmitted Heat, qT............................................................................... 10-10
10.3.2.1 Windows Exposed to the Sun ............................................................. 10-10
10.3.2.2 Windows, Walls, Floors, Ceilings ......................................................... 10-11
10.3.3 Heat Generated by Various Equipment, qE ............................................ 10-11
10.3.4 Ventilation and Fresh Air Intake, qV ........................................................ 10-11
10.3.5 Infiltration Air, qL ...................................................................................... 10-12
10.3.6 Comfort Levels ........................................................................................ 10-12
10.4 Estimation of Power Requirement............................................................. 10-14

10.5 Choice of Equipment ................................................................................. 10-14
10.6 REFERENCES AND FURTHER READING............................................... 10-15
TABLES
Table 10-1 Air conditioning load - calculation sheet 10-13
FIGURES
Figure 10-1 Combined sketch showing the five principal classes of

refrigeration systems 10-4
Figure 10-2 Psychrometric chart for air-water mixtures at 1
atmosphere 10-5
Figure 10-3 Basic air conditioning processes 10-6

10 AIR CONDITIONING
The design of air-conditioning systems is a complex subject which will often

be subcontracted on a turnkey basis. This Section indicates some criteria for
provisional choice and sizing of systems or for evaluation of proposals, but it
is not intended to provide design guidelines. For the latter, reference should
be made to more specialist books (see Refs. 1 and 2).
10.1 Choice of Cooling/Dehumidifying Method
Nearly all systems involve a mechanical compression refrigeration cycle.

The differences are in the means of heat transfer between the cycle working
fluid and the air to be treated (see Figure 10-1). A direct system means that air is in
direct contact with the evaporator; an ‘indirect' system has secondary
circulation of brine or water. The cooling can take place at a ‘closed surface'
or with an ‘open-spray'.
A psychrometric chart (Figure 10-2) is Central to comparing treatment me-

thods. The basic air-conditioning processes are illustrated in Figure 10-3 (a.
to h.) with reference to the chart. An air-conditioning system must cope with
seasonal variations in air temperature and humidity; for this the chart pro-
vides a quick means of comparison.
The enthalpy datum of the chart for air-water mixtures takes into account the
latent heat of any water added or withdrawn.
It is recommended that when using the chart, a complete enthalpy balance
should be made over the whole air conditioning system and its components,
e.g. air humidifiers, water condensers, etc.

Air Conditioning
10.2 Sizing of Facility (Cooling Load)
The required capacity of an air-conditioning plant depends on the estimated

cooling load and the external air condition. It is essential to define the boun-
daries of the space concerned, in order to evaluate all the air-flows and
internal sources of heat. The key parameters in the sizing of a facility are
summarised as follows:
- internal/external temperature difference
- internal/external humidity difference
- total volume of ‘building'
- purpose and occupancy of rooms
- ventilation requirements (air-change rate)
- thermodynamic efficiency of chosen refrigeration cycle.
The required steady-state cooling load is the sum of internal heat generation
(human and equipment), transmitted heat (by conduction and radiation),
inward leakage of warm-air and outward flow of excess cold air:
The various contributary elements are briefly discussed in 10.3.


Air Conditioning

a. HEATING WITHOUT MOISTURE CHANGE

No moisture change in heating coil (diagram) means dew point, latent heat stay
same. Adding sensible heat means higher total heat, dry bulb, wet bulb. Process
shows on chart as A-B
b. COOLING WITHOUT MOISTURE CHANGE

It not enough cooling to cause condensation, removing sensible heat lowers dry bulb
without dew point change. With no change in latent heat, total heat goes down, wet
bulb falls.
Figure 10-3 Basic air conditioning processes

(Adapted from Power Handbook, McGraw-Hill, 1951)

Air Conditioning
c. COOLING, HUMIDIFYING BY SPRAYS

With recirculated spray water at air's wet bulb temperature, there's no heat transfer,
no change in air total heat. Some air sensible heat becomes latent, so water eva-
porates. Dry bulb drops (cooling), air picks up moisture (humidification) and dew
point rises. Dry bulb and dew point approach wet bulb.
d. COOLING, DEHUMIDIFYING BY SPRAYS

With spray water held below air's dew point by outside cooler, air gives up heat to
water. Dry bulb, wet bulb, dew point fall. Some moisture condenses out. If water
temperature stayed same, line A-B would apply, C being water temperature. But
water heats to temperature D, so line A-E shows the practical case.
Figure 10-3 (continued)

e. HEATING, HUM1DIFYING BY SPRAYS

Heater (connected like cooler of d.) keeps water temperature above air dry bulb.
Water heats air so dry bulb, wet bulb, dew point rise. Water temperature falls. Line
A-B shows air's changes, C is final water temperature, which air approaches.
f. COOLING, DEHUMIDIFYING BY COILS

With coil temperature below air initial dew point, dry bulb, wet bulb and dew point
fall, moisture condenses out. Air conditions change along line A-B. How close air ap-
proaches coil temperature C depends on design, number of sections, etc.
Figure 10-3 Basic air conditioning processes (continued)

Air Conditioning
g. CHEMICAL DRYING, COIL COOLING

Water vapour is absorbed by chemical, so dew point falls. Latent heat released by
condensation raises air dry bulb with no change in wet bulb. Cooling coils take out
sensible heat so dry bulb and wet bulb fall while dew point remains unchanged.
h. MIXING FRESH AND RETURN AIR

Fresh air A mixes with return air B to form mixture C. Where point C falls along line
A-B depends on relative amounts of air at A and B. Distance A-C (as measured on
chart) is to distance C-B as volume of air at B is to volume of air at A.
Note: American Institutes use the ‘Standard Commercial Ton of Refrigeration' to

express the capacity of a cooling installation, It equals 3,500W (288,000 Btu/day).
This unit is often used in manufacturers' literature.
Figure 10-3 (continued)

10.3 Data for the Estimation of Cooling Load
Estimation of cooling loads is inevitably uncertain, and various rules of

thumb exist. Consequently answers are neither precise nor firmly reliable.
With these provisos, the calculation sheet (Table 10-1) can be used to provide
a systematic initial estimate. This was adapted from a form prepared by the
Air Conditioning and Refrigeration Machinery Association, Inc.
10.3.1 Human Heat Generation, qH
W (per occupant)
Lying down 70- 95
Sitting 80-105
Standing 95-125
Moving slowly 175-235
Labour/Moving fast 235-470
Average will vary between 100-200W per occupant in buildings. Refer also to
ASHRAE Handbook ‘Fundamentals'.
10.3.2 Transmitted Heat, qT
Heat from the external environment is transmitted by conduction and radia-

tion. It depends on the climate and location, and on the insulation and heat
capacity of the building. For aggregate estimation, some typical values for
the overall heat transmission coefficient are given below:
10.3.2.1 Windows Exposed to the Sun

2
Insert value in W/m at ‘A' in calculation sheet (Table 10-)*)
Exposures to the sun from direction: NE E SE S

Windows:
- of clear glass (no protection) 350 560 570 335
- fully in the shade 95 160 140 95
- with venetian blinds, etc. 205 350 305 185
*) Values are equal for exposure from equivalent Western angles. Calculate heat for
largest sun exposed window area.

Air Conditioning
10.3.2.2 Windows, Walls, Floors, Ceilings

2
Insert value in W/m at ‘B' in calculation sheet (Table 10-1).
o
Outside Dry Bulb C 30 35 40 45 50
Windows, no sun 27 54 81 108 135

Wails, heavy 7 14 21 28 35
Walls, insulated 3 9 16 23 30
Partition, single, inside 16 32 48 64 80
Partition, double, ditto 10 17 24 31 38
Floor 6 13 19 26 32
Ceiling, no vent, under
1
Attic ) 35 42 51 61 72
1
Ceiling, vent, under attic ) 22 35 47 60 72
Ceiling, under flat roof 41 51 60 70 79
1) When ceiling is insulated, multiply by:

0.4 for 25 mm insulation
10.3.3 Heat Generated by Various Equipment, qE
W
Electric motors, per rated kW 1100
Gas burners 1750
Coffee makers, glass 260
Coffee machines 400
For heat load from equipment in locations with positive exhaust reduce by
50%.
10.3.4 Ventilation and Fresh Air Intake, qV
Air in air-conditioned areas in circulated, filtered and partly replaced by fresh

air.
Rates of circulation and fresh air intake ratio's vary with the functional use of
the enclosed space.
The cooling load which results is proportional to the product of both factors
(see Ventilation factor C below, which should be inserted in calculation sheet,
Table 10-1).

Type of Area Circulations Fresh Air Ventilation

per hour intake ratio factor (C)
Non hazardous 2-6 0.3 0.6- 1.8*)

Hazardous 12 1.0 12
Recreational 15 0.7 10.5
Dining 15 0.7 10.5
*) Dependent on pollution as a result of smoking, odour, etc.
10.3.5 Infiltration Air, qL
Air will infiltrate through doors and various leaks which will add to the cooling
load.
Since leakage represents a non-intentional fresh air intake, it can be com-
pensated for by reducing the intentional fresh air intake somewhat.
For an initial estimate of the cooling load, infiltration can be ignored.
10.3.6 Comfort Levels
For office and accommodation units, the following are generally acceptable
comfort levels:
o
Temperature range 20-22 C
Humidity range 35% - 65% (relative humidity)
Air movement 0.15 m/s max.
Noise, offices 55 decibels at 1 metre
Noise, plant rooms 60 decibels at 1 metre

Air Conditioning

10.4 Estimation of Power Requirement
The overall power requirement depends on the following:

- cooling load Q (allow for variations)
- thermodynamic effectiveness of chosen cycle (including isentropic effi-
ciency of the compressor)
- mechanical efficiency of equipment
- starting capacity
- contingency (up or down).
The engineer should note that there is a considerable element of discretion in
applying the above factors and hence the appropriate design will usually be a
compromise. When comparing or evaluating cooling systems, always iden-
tify the thermodynamic part. For this, the performance ratio is defined as:
β = enthalpy (heat) absorbed in evaporator (β > 1)

compressor work
Consequently, the overall power requirement (= compressor work) should

always be less than the cooling load. For Freon installations, a very rough
approximation of the power requirement, can be obtained by using an overall
factor of 0.3 to convert cooling load into power requirement, i.e. power re-
quirement = 0.3 x Cooling Load.
Note: The power requirements for condenser/cooling fans, circulation pumps, etc.
should be added to obtain full power load.
10.5 Choice of Equipment
The following should be evaluated, as applied to a particular situation:

- large (efficient) plant vs. flexible decentralised units, especially in case of
large number of window units or split units
- piped liquid vs. ducted air
- power source; starting surges
- control system
- maintenance requirements (including spares)
- leased equipment vs owned.

Air Conditioning
AIR CONDITIONING
References
1. ASHRAE Handbooks issued by the American Society of Heating, Refrig-

eration and Air Conditioning Engineers Inc., 345 East 47th Str., New York,
NY 10017, USA
2. IHVE*) Guides issued by the CIBS (the Chartered Institution of Building

Services) Delta House, 222 Balham High Road, London SW 12 9B5
Further Reading
- Carrier System Design Manual issued by Carrier Air Conditioning Com-

pany, Syracuse, New York (Obtainable from local Carrier representa-
tives)
*) IHVE = The Institution of Heating and Ventilating Engineers, 49 Cadogan Square,

London SW1 X OJB.

Production Handbook - Facilities & Maintenance

Uploaded by

Copyright:

Available Formats

Production Handbook - Facilities & Maintenance

Uploaded by

Document Information

Original Title

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

Production Handbook - Facilities & Maintenance

Uploaded by

Copyright:

Available Formats

What is the purpose of the Production Handbook?

What is the purpose of the Production Handbook?

What are the different volumes of the Production Handbook and what topics do they cover?

What are the different volumes of the Production Handbook and what topics do they cover?

Cover photo: Borescope view of second stage compressor turbine rotor blade of

Solar Centaur gas turbine.

This book is the property of Shell Internationale Petroleum Maatschappij

The copyright of this document is vested in Shell Internationale Petroleum

The objective of the Production Handbook is to contribute to efficient performance by all

Being a comprehensive combination of condensed technical manuals, it provides a ready

The Production Handbook should be available to all Engineering, Petroleum Engineering

Reprinting and updating

- Units and Conversion Factors EPO/71

Volume 2 Drilling and Transport EPO/51

Volume3 Petrophysical Engineering EPD/22

Volume 4 Reservoir Engineering EPD/22

Volume 5 Production Technology EPD/41

Volume 6 Production Operations EPO/53

Volume 7 Process Engineering EPD/42

Volume 8 Pipelines EPD/61

Volume 9 Facilities and Maintenance

- Chapter 6, Production Operations: Sections 2.17-2.20 (Pumps, Compres-

A more comprehensive review of the contents of the Surface Facilities'

For organisational aspects of maintenance, reference should also be made to

SUMMARY CONTENTS LISTING page

Chapter 1: Running Equipment

Chapter 2: Piping Systems 2-1

Chapter 3: Electrical Engineering 3-1

Chapter 4: Instrumentation 4-1

Chapter 5: Telecommunications 5-1

Chapter 6: Reliability and Availability Assessment 6-1

Chapter 7: Corrosion and Materials Engineering 7-1

Chapter 8: Inspection Techniques and Maintenance

Chapter 9: Diving and Underwater Operations 9-1

Chapter 10: Air Conditioning 10-1

1.1 PRIME MOVERS................................................................................................................ 1-10

1.1.1 Prime Mover Selection .................................................................................................... 1-10

1.1.1.3.1 General....................................................................................................................... 1-13

1.1.2 Site Rating of Drivers........................................................................................................ 1-37

1.1.3 Gas Consumption............................................................................................................. 1-39

1.2 COMPRESSORS ............................................................................................................... 1-40

1.2.1 Types of Compressors and their Application................................................................... 1-40

Confidential-Property end Copyright: SIPM, 1991 1-1

1.2.1.3 Reciprocating Compressors ......................................................................................... 1-57

1.2.2 Compressor Systems Control........................................................................................... 1-81

1.2.3 Compressor Sizing and Power Requirement ................................................................... 1-89

1-2 Confidential-Property end Copyright: SIPM, 1991

1.3 PUMPS ............................................................................................................................. 1-113

1.3.1 Applications .................................................................................................................... 1-113

1.3.2 General .......................................................................................................................... 1-114

1.3.3 Characteristics and Selection........................................................................................ 1-114

1.3.4 Pipework Requirements ............................................................................................... 1-140

1.3.5 Net Positive Suction Head (NPSH) ................................................................................ 1-142

1.3.6 Centrifugal Pump Arrangements ................................................................................... 1-143

1.3.7 Prime Movers ............................................................................................................... 1-144

1.3.8 Seals............................................................................................................................... 1-144

Confidential-Property end Copyright: SIPM, 1991 1-3

1.4.1 Pulsation Dampeners for Reciprocating Compressors .................................................. 1-147

1.4.2 Pulsation Dampeners for Reciprocating Pumps........................................................... 1-148