Home

Search

Collections

Journals

About

Contact us

My IOPscience

Energy reduction for the spot welding process in the automotive industry

This content has been downloaded from IOPscience. Please scroll down to see the full text.
2007 J. Phys.: Conf. Ser. 76 012022
(http://iopscience.iop.org/1742-6596/76/1/012022)
View the table of contents for this issue, or go to the journal homepage for more

Download details:
IP Address: 180.248.3.131
This content was downloaded on 23/03/2015 at 13:50

Please note that terms and conditions apply.

Sensors and their Applications XIV (SENSORS07)

Journal of Physics: Conference Series 76 (2007) 012022

IOP Publishing
doi:10.1088/1742-6596/76/1/012022

Energy reduction for the spot welding process in the

automotive industry
J D Cullen, N Athi, M A Al-Jader, A Shaw and A I Al-Shammaa
Liverpool John Moores University General Engineering Research Institute, RF and
Microwave Group, Byrom Street, Liverpool, L3 3AF, UK.
E-mail: [email protected]
Abstract. When performing spot welding on galvanised metals, higher welding force and
current are required than on uncoated steels. This has implications for the energy usage when
creating each spot weld, of which there are approximately 4300 in each passenger car. The
paper presented is an overview of electrode current selection and its variance over the lifetime
of the electrode tip. This also describes the proposed analysis system for the selection of
welding parameters for the spot welding process, as the electrode tip wears.

1. Introduction
Spot welding is an important welding technique that has been long established and extensively used in
industry, especially the automotive industry [1]. The main quality control tests are the destructive
chisel test and peel test, which are carried out on welds obtained from the production line of the
product [2,3]. These have been supplemented by non-destructive methods [4], where measurement of
a representative sample is undertaken, examined and evaluated and if there is substantial increase in
failure, then the whole batch of the products is rejected as faulty. Traditionally, resistance welding
process control has been based upon monitoring the voltage and current, or their derivatives, power
and resistance [5,6]. While these sensors work well under model conditions, surface contaminants
and/or metal impurities can cause under strength or under sized welds to be formed despite the voltage
and current values conforming to the ideal standards [7]. Recently, work has been in progress for
analysis of every spot weld, as simply having the right conditions for weld formation does not mean
that the desired weld has been formed. Therefore, other parameters must be examined as well in order
to give a complete picture of the quality of weld formation [8-10].
Recently however, a different concern has come to light, namely that of energy consumption within
industry, both financially and environmentally [11].
In practice, the manufacture of welds of acceptable quality depends on the definition of optimum
welding parameters and the implementation of suitable controls to ensure constant weld quality over a
production run. These methods to optimize the spot welding process can be formalised as Weld
Growth Curves and Weldability Lobes [12], which are the industrial accepted methods.
The ability to make a weld is best defined by a weldability lobe, which outlines the available
manufacturing tolerances between minimum and maximum limits. Both two and three-dimensional
weldability lobes exist, which are defined in terms of welding time, electrical current and electrode
force.

c 2007 IOP Publishing Ltd



Sensors and their Applications XIV (SENSORS07)

Journal of Physics: Conference Series 76 (2007) 012022

IOP Publishing
doi:10.1088/1742-6596/76/1/012022

The weldability lobe can only provide a snapshot of the welding current range, because as the
electrode tips wear, the weldability lobe can drift.
These two factors are controlled by the interaction between various parameters, which control the
temperature distribution in the metal parts during the welding thermal cycle. Galvanised steel typically
has narrower lobes and greater electrode wear when compared with uncoated steel. [13]
2. Experimental Setup
Figure 1 shows the spot welding machine used for the experimental trial, which is a TECNA 4621
Pedestal Welder [14]. The top arm is hinged to move down in an arc. On the bottom arm, which is
fixed, are several sensors to monitor in real time the weld as it takes place [10]. Sensors used on the
machine include current, voltage, cameras, infrared and ultrasonic sensors, as shown in figure 2.
(1) Ultrasonic
Transmitter
(2) Adjustable Cluster
Bracket
(3) Monitoring cluster
containing Camera and
Infra-red,
(4) Voltage Probe,
(5) Ultrasonic Receiver,
(6) Rogowski current
sensor

Figure 1. TECNA 4621 Spot Welding

Machine

Figure 2. The Experimental sensors setup

The sensors are placed in positions that give the most effective detection. All these sensors have their
readings captured using data acquisition boards and recorded using a personal computer. The dataacquisition program is specifically written for this project in the Pascal based Delphi environment. All
the data is captured continuously, and when a weld takes place, the data from 0.5s before a weld to
0.5s after a weld is saved for analysis.
3. Optimal Weld Selection
The process of selecting a suitable current for performing a weld is based upon the parameters within
the British standard [7]. This standard carries a guideline set of tables for the selection of welding
current and welding time, extracts of which are reproduced in tables 1 and 2. Table 1 and table 2 show
the requirement for uncoated mild steel and electroplated mild steel respectively. As can be seen, the
force and current requirements are greater for coated steel than uncoated steel because the process has
to melt the coating before the actual spot welding can take place.
Table 1: Spot welding conditions for uncoated mild steel
Material Thickness
(mm)
0.4 to 0.6
0.6 to 0.8
0.8 to 1.0
1.0 to 1.2
1.2 to 1.6

Force Setting
(kN)
0.9 to 1.1
1.2 to 1.3
1.4 to 1.5
1.6 to 1.8
1.9 to 2.1

Weld Time
(50Hz Cycles)
5 to 7
7 to 10
9 to 12
11 to 15
14 to 18

Weld Current
(kA)
4 to 6
5 to 7
6 to 8
7 to 9
8 to 11

Sensors and their Applications XIV (SENSORS07)

Journal of Physics: Conference Series 76 (2007) 012022

IOP Publishing
doi:10.1088/1742-6596/76/1/012022

Table 2: Spot welding conditions for electrolytically deposited zinc coated mild steel
Material Thickness
(mm)
0.4 to 0.6
0.6 to 0.8
0.8 to 1.0
1.0 to 1.2
1.2 to 1.6

Force Setting
(kN)
1.5 to 2.0
1.9 to 2.2
2.2 to 2.9
2.8 to 3.6
3.4 to 4.5

Weld Time
(50Hz Cycles)
6 to 7
8 to 10
9 to 12
10 to 13
11 to 15

Weld Current
(kA)
6.5 to 8.5
7.5 to 9.5
8.5 to 10.0
9.5 to 12.5
12.0 to 14.5

Looking at the tables, there is a range of parameters available for a given metal thickness. These data
have been determined by the use of weld growth curves, and weldability lobes. A weld growth curve,
as shown in figure 3, is constructed by the method of performing several welds at different power
settings, and taking the average size of the resultant nuggets, such that you have a range of input
powers covering settings that produce no welds through to power settings that cause splash. The
range of weld types performed covers the following criteria:
No weld: This occurs when there is insufficient current to melt the parent metal
Stuck weld: This strictly refers to the case where, when spot welding galvanised metal, the
coating metal having a lower melting point melts, but the parent metal does not. This results in
the metals being stuck together, but with minimal mechanical strength, see figure 3.
Undersized weld: This is where a weld is created, but upon destructive testing the nugget is
smaller than the required size, which is, according to the BS1140 standard [7], 3.5 times the
square root of the thinner parent metal (in mm). This minimum requirement can be overridden
by a particular requirement from the manufacturer, such as 4 times the square root of the
thickness.
Acceptable weld: This is the condition where the weld nugget is above the minimum size and
below any maximum size (if specified) and does not result in splash (expelled metal), see
figure 3.
Oversized weld (if specified): This is the upper size for the weld nugget. If not specified, the
acceptable range extends to the current where splash starts.
Splash weld: This is where some of the molten metal is expelled from the molten nugget,
causing the electrodes to collapse into the metal further, resulting in a thinner weld.
Nugget Size

Average Nugget Size

Splash welds

Nugget Size (mm)

6
5

Acceptable welds

4
3
2
1

Stuck welds

Undersized welds

0
3

Current (kA)

Figure 3. Weld growth curve for 0.8mm thick galvanised mild steel.
Based upon the weld growth curve, two acceptable weld points can be specified as being the minimum
and maximum currents to produce an acceptable weld, for figure 3 these are 6.2kA and 6.7kA. These
can then be plotted on a graph. If the weld growth curve is repeated for different settings for one of the
parameters, when plotted, an envelope of points is created, that represent the acceptable values for the
3

Sensors and their Applications XIV (SENSORS07)

Journal of Physics: Conference Series 76 (2007) 012022

IOP Publishing
doi:10.1088/1742-6596/76/1/012022

welding current for that parameter. Figure 3 shows the growth curve produced for 2bar gas pressure on
coated mild steel, resulting in an electrode force of 1.4kN. If this is repeated for the following
pressures, 2.5bar(1.75kN), 3bar (2.1kN) and 3.5bar (2.45kN), which covers the range specified in the
British Standard. The growth curves are shown in figure 4 (for uncoated steel) and figure 5 (for coated
steel), and the resultant weldability lobe for coated steel is shown in figure 6.
2.5 bar

3 bar

3.5 bar

2 bar
6

5
Nugget Size (mm)

Nugget Size (mm)

2 bar
6

4
3
2
1

2.5 bar

3 bar

3.5 bar

4
3
2
1

0
2

Welding Current (kA)

Welding Current (kA)

Figure 4. Weld growth curves for different Figure 5. Weld growth curves for different
electrode forces for uncoated mild steel
electrode forces for coated mild steel
4

Pressure (bar)

3.5
3
2.5
2
1.5
6.2

6.4

6.6

6.8

7.0

7.2

7.4

7.6

7.8

Welding Current (kA)

Figure 6. Resultant weld lobe for galvanised mild steel growth curves.
4. Electrode Tip Life and Weld Quality
To minimise energy usage for the spot welding process, it is desirable to perform the weld with the
minimum current required for an acceptable weld, but this leads to a problem. As the number of welds
performed using the same electrode tip increases, the tip becomes contaminated with the galvanising
coating, effectively alloying the tip. This results in more heat generated at the tip, with a slight
reduction in the heat generated in the metals. Thus, if the current is excessive, the tip profile can
deform, causing a reduction in the welding current density. This results in a need to increase the weld
current to maintain the same nugget size. Figure 7 shows the weldability lobes for electro-plated
0.8mm mild steel. For instance, the data for the gas pressure of 3bar (welding force of 2.1kN), when
the electrode is new, the minimum current that can be used is approximately 7.3kA, but as the
electrode ages, the minimum current increases, as shown in the table 3. In this case for example, after
1200 welds, the current required to create a minimum sized weld nugget also caused the weld to
splash.
In industry, there are two common methods for setting up the welding current:
Set the current at the maximum acceptable value, which results in excessive energy usage to
guarantee the nugget size, and also shortens the tip life, typically to approximately 300 welds.
Set the current to the mid point between minimum and maximum values, which gives a
slightly better energy usage than setting the current to the maximum. The current is then

Sensors and their Applications XIV (SENSORS07)

Journal of Physics: Conference Series 76 (2007) 012022

IOP Publishing
doi:10.1088/1742-6596/76/1/012022

incremented in steps, e.g. every 100 welds. This strategy increases the tip life to
approximately 450 500 welds.
However with an intelligent control algorithm it is then possible to set the current to the minimum
value and increment it after every weld to give the minimum energy usage. Figure 8 shows the output
of a curve-fitting algorithm to the data in table 3, which gives equation (1).
WeldingCurrent = 7.3 + 3.1 10 4 x 1.1 107 x 2 + 9.2 10 10 x 3
(1)
where x is the number of welds.

Operating Gas Pressure (Bar)

4
3.5
After 0 welds
After 300 Welds
After 600 welds
After 900 Welds
After 1200 welds

3
2.5
2
1.5
6

Welding Current (kA)

Figure 8. Curve fitting algorithm output for

minimum current

Figure 7. Weldability lobes for the increasing

number of welds

Table 3: Minimum current requirements as number of welds performed increases.

Number of Welds Minimum current (kA)
a
This setting produced nuggets
0
7.3
of
the minimum size, but they
300
7.45
were splashing at the same
600
7.6
time.
900
8.2
1200
9.1a
This rule based solution is acceptable where there is only one parameter to consider, namely the
current used for the welding process. However there is no flexibility in the system to unravel the
complexity of the spot welding parameters and factors. There are many factors, which affect the
process in a factory, for example, fluctuations in gas pressure which affects electrode force, material
contamination and variations in electrode cooling. An example of this can be seen in the weld lobe
(figure 6), where the minimum current required changes as the pressure changes. Also as the tip wears,
the minimum force that can be used changes as seen in figure 6. This illustrates the limitation of the
rule-based approach of attaining the minimum current and force in the spot welding process.
5. Artificial intelligence approach
To unravel the above complexity, the use of different artificial intelligence (AI) approaches are
currently being investigated, (adaptive rule systems, neural networks, genetic algorithms etc) [15] to
allow the creation of a system that can take into account the fluctuations in a system by monitoring the
process and feeding these parameters into an AI algorithm to generate a prediction of the minimum
current required and of the resultant weld nugget quality. This is illustrated in figure 9.

Sensors and their Applications XIV (SENSORS07)

Journal of Physics: Conference Series 76 (2007) 012022
Current (In)

Weld Nugget
Size

Voltage (V)
D. Res. (R)

Weld Quality

Infra Red (IR)

Force (Fn)
Ultrasonic (I)

IOP Publishing
doi:10.1088/1742-6596/76/1/012022

Artificial
Intelligence
Algorithm

Electrode Image

Tip Width
New Current
Setting (In+1)

Weld Number
Material Thickness

New Force
Setting (Fn+1)

Material Type

Figure 9. Overview of the Artificial Intelligence approach.

6. Conclusions
It has been shown that an electrode tip can be used for more than the automotive industry
recommended lifespan (approximately 500 welds) by reducing the current to the minimum value in
order to produce an acceptable weld. This has the advantage of reducing the energy consumption for
the welding process, which will have a positive impact both environmentally and financially. This
enhancement is currently undergoing further development, which will enable the process to consume
less energy by reducing the electrode force to a minimum. It is also envisaged that the artificial
Intelligence introduction will allow for a reduction in the time that a weld is performed, as well as
adapting the system to be used on a different machine, which will give an improved deployment speed
and hence reducing manufacturing downtime.
References
[1] Larson C.F 1972 Bibliography on resistance welding 1950-1971 Welding Journal 51
[2] Davies A C 1981 The Science and Practice of Welding (Cambridge, Cambridge University Press)
[3] Nita D 1997 Resistance welding standards Krupp Camford Internal Report KW002
[4] Krieger R J, Wenk S A and McMaster R C 1954 Non destructive test methods for inspection of
welded joints Welding Journal 33
[5] Livshits A G 1997 Universal quality assurance method for resistance spot welding based on
dynamic resistance Welding Research Supplement
[6] Savage W F, Nippes E F and Wassel F A 1978 Dynamic contact resistance of series spot welds
Welding Journal 57
[7] British Standards Institution 1993 BS1140:1993 Specification for resistance spot welding of
uncoated and coated low carbon steel
[8] Matsuyama K 2003 Spot welding system and method for sensing welding conditions in real time
US Patent No. 6,506,997 B2.
[9] Vogt Werkstoffeprfsysteme 2002 SPOTline inline ultrasonic testing system http://www.vogtnde.de
[10] Cullen J D, Athi N, Al-Jader M, Johnson P, Al-Shammaa A I , Shaw A and El-Rasheed A M A
2007 Multisensor fusion for on line monitoring of the quality of spot welding in automotive
industry Measurement, http://dx.doi.org/10.1016/j.measurement.2007.01.006
[11] Gopalakrishnan B, Plummer R W and Iskander W H 2003 A comparative study on energy
assessment data from manufacturing industry, Proc. of the 2003 ACEEE (American Council
for an Energy Efficient Economy) Summer Study on Energy Efficiency in Industry
[12] Williams N T and Parker J D 2004 Review of resistance spot welding of steel sheets. Part 1.
Modelling and control of nugget formation Inst. Mater
[13] Koenigsberger F and Adair J R 1965 Welding Technology 3rd Edition (Macmillan)
[14] Tecna 2003 website http://www.tecna.net/
[15] Ablewhite J D The Application of Artificial Intelligence Techniques to the Control of Spot
Welding (University of Swansea)
6