Evaluation of Shunt Distancein Resistance Spot Welding

(1)

IN

DEGREE PROJECT TECHNOLOGY, FIRST CYCLE, 15 CREDITS

STOCKHOLM SWEDEN 2020,

Evaluation of Shunt Distance in Resistance Spot Welding

DANIEL FREDRIKSSON MILOU GÖRANSSON

KTH ROYAL INSTITUTE OF TECHNOLOGY

SCHOOL OF INDUSTRIAL ENGINEERING AND MANAGEMENT

(2)

Abstract

The purpose of this study was to generate a greater understanding of how the shunt distance affects the shunting phenomena which occurs when working with resistance spot welding (RSW). Shunting affects the quality of the weld and the goal of the study was to create guidelines for RSW to minimize its impact on the weld quality. In order to minimize waste from production and to reduce the welding time there is need for understanding of what the shunt effect depends on. To evaluate what impact the shunt distance will have on the weld, two different experiments were performed. In the first

experiment, a one-dimensional welding lobe was manufactured for various materials and the current range was compared over three different shunt distances. The second experiment consisted of welding with a constant current on different shunt distances to investigate how this would impact the second weld performed in the welding sequence. A range of different steel grades and sheet thicknesses was used in the experiments to further explore what impact different material properties will have on the shunt effect. The result showed that the shunt distance has little impact on the acceptable weld-current range, with minor deviations. However, the shunt distance will affect the size of the shunted weld, which decreases as the shunt distance is reduced. Overall, the data collected in this study is not expansive enough to make guidelines that could be implemented in today´s industry. The phenomena of shunting require more data to fully be understood.

Keywords

Resistance spot welding, RSW, Shunt effect, Shunting, Shunt distance, One-dimension

welding lobe.

(3)

Sammanfattning

Syftet med denna studie var att skapa en större förståelse för hur shuntavståndet påverkar det

shunt-fenomen som uppstår när man arbetar med punktsvetsning (RSW). Shuntning påverkar svetsens kvalitet och målet var att skapa riktlinjer för arbete med RSW för att minimera

shunt-effektens påverkan på svetsen. I syfte att kunna minimera spill i produktionen samt att minska svetstiden krävs det förståelse av vad shunt effekten beror på. För att utvärdera vilken påverkan shuntavståndet kommer att ha på svetsen utfördes två olika experiment. I det första experimentet skapades en endimensionell svetslob för varje material och det genererade strömintervallet jämfördes över tre olika shuntavstånd. Det andra experimentet bestod av svetsning med en lika stor ström vid olika shuntavstånd för att undersöka hur detta skulle påverka den andra svetsen. En rad olika stålsorter och plåttjocklekar användes för att ytterligare se vilken påverkan dessa faktorer kommer att ha på shunt-effekten. Resultatet visade att shuntavståndet hade liten inverkan på det acceptabla

svetsströmintervallet, med mindre avvikelser. Emellertid påverkar shuntavståndet storleken på den shuntade svetsen genom att storleken minskar när shuntavståndet minskas. Sammantaget räcker inte de uppgifter som samlats in i denna studie för att skapa riktlinjer som skulle kunna implementeras i dagens industri, utan det krävs mer data för att fullständigt kartlägga shunt-effekten.

Nyckelord

Punktsvetsning, RSW, Shunt effekt, Shuntning, Shuntavstånd, Endimensionell svetslob.

(4)

Table of content

1. Introduction ... 1

1.1 Purpose and objectives of the project ... 2

1.2 Boundaries of the study ... 3

1.3 Ethical and Environmental impacts ... 3

2. Resistance spot welding ... 4

2.1 Resistance spot welding parameters ... 5

2.1.1 Weld current ... 5

2.1.2 Weld time ... 5

2.1.3 Electrode force ... 6

2.1.4 Other welding parameters ... 6

3.1 Previous investigations ... 7

4. Experimental method ... 9

4.1 Experimental equipment ... 9

4.2 Materials ... 10

4.3 One-dimension welding lobe ... 11

4.3.1 Welding process parameters ... 12

4.4 Welding with set current ... 12

4.4.1 Welding process parameters ... 12

5. Results ... 13

5.1 One-dimension welding lobe ... 13

5.1.1 DP600 ... 13

5.1.2 Boron steel ... 15

5.1.3 Forta H1000 Stainless 1.5 mm ... 16

5.1.4 Forta H1000 Stainless 2.0 mm ... 18

5.1.5 Boron blasted steel ... 19

5.1.6 Compilation of results ... 21

(5)

5.2.1 Boron blasted steel ... 21

5.2.1 DP800 steel ... 22

5.2.2 Compilation of results ... 22

6. Discussion ... 23

6.1 One-dimension welding lobe ... 23

6.1.1 Sources of error ... 24

6.2 Welding with set current ... 24

6.2.1 Sources of error ... 25

7. Concluding remarks ... 26

8. Further study ... 27

9. Acknowledgements ... 28

10. References ... 29

(6)

1

1. Introduction

With an increasing development towards more lightweight vehicles, the material combination used in the automotive industry are changing from mostly only using low carbon steel to using a mix of low carbon and high strength steel. This change of material usage creates new demands regarding the joining method in the production. A well-adapted joining method can be used to assemble the material in an efficient and reliable manner. There are many different types of welding techniques in use today, though the most extensively used method in the automotive industry is resistance spot welding (RSW).

The stage in automotive manufacturing where most of the welding takes place is when creating the body-in-white (BIW), a sheet metal welded structure, shown in Figure 1. A typical steel body-in-white contains around 5000 spot welds which creates high demands concerning the choice of welding technique. [1]

Figure 1. Body-in-white, an early state in the manufacturing process of a car. [2]

The reasons for the extensive use of resistance spot welding is due to the low cost, short weld time, and the high automation capabilities which eliminates human interference. Eliminating human interference in turn minimizes errors and provides the method with a wide range of optimization parameters. Resistance spot welding is for the most part performed by robots and the weld is

completely internal which removes the need for further treatment. RSW is a welding technique where the process works by applying an electric current through two or more sheets of metal that are clamped together by a pair of electrodes. As the current is passing through the sheets, the resistance of the circuit cause generation of heat energy. The heat energy (Q) is a function of the current (I), resistance (R) and time (t) as defined by Joule’s law, seen Equation 1. [1]

𝑄 = 𝐼²𝑅𝑡 (Eq. 1)

The generation of heat energy, mainly caused by the resistance, melts the welding area between the metal sheets. When the current is stopped and no further heat energy is generated, the temperature will decrease, and the metals will solidify as merged. The method is greatly dependent on the electrical and thermal conductivity properties of the materials in use. [3]

To achieve a repeatable weld result and a short weld time there is a need for greater understanding of how the distance between the welds, also called the shunt distance, affects the result. When working with RSW on a metal piece, welds are continuously produced within the same region and therefore a shunt effect occurs. [4]

(7)

2

The shunt effect is described as a parallel path where the current will flow through instead of the intended welding point, see Figure 2. This results in that an insufficient amount of heat generation, resulting in an increased weld time, decreased size of nugget* and a possibly failed weld. As the industry strives for shorter production times, reduction of unnecessary quality testing, and a more reliable welding result it is not efficient to use a trial and error-based approach to solve these problems. [4]

*A nugget is the molten metal between the sheets that solidifies into the welding joint. [4]

Figure 2. Illustration of current going through the previous performed weld via the shunt path, also referred to as shunting.

To achieve these goals and optimize production, there is a need for guidelines of the RSW process for the automotive industry production, to minimize the effect of shunting and thus create a more

consistent result of the weld. To obtain an efficient production, the effects of shunting must be evaluated on a larger scale to understand what it depends on and how it can be accounted for.

1.1 Purpose and objectives of the project

The project aimed to create guidelines for optimizing the weld quality for RSW. This was achieved by evaluating the RSW process, specifically by an investigation of the shunt effect and the shunt distance.

The project aimed to answer three research questions, which are stated below.

● How will the shunt distance affect the quality of the resistance spot weld?

● How do different material properties and process parameters affect shunting and the optimized shunt distance?

● How will the previous weld in a welding sequence impact the approved current range, when the shunt distance is changed?

(8)

3

1.2 Boundaries of the study

The project investigated how to optimize the shunt distance to obtain a more reliable weld quality.

This was performed by experiments where the current and the weld spacing varied. This means that this study did not focus on optimizing other welding parameters such as the electrode force or welding time. In addition, the impact of material properties was discussed only partially, when it is deemed relevant to answer the research questions. By limiting the project, the likelihood to concretely understand the impact of the shunt effect increases.

1.3 Ethical and Environmental impacts

The focus of this study lies in understanding and minimizing the effects of shunting in resistance spot welding thus making the question of ethical impacts not relevant for this report. The fact that RSW is mostly used in the automotive industry means that the welding technique is a subject of ethical impacts but due to the focus of this study being on the pure process and not the overall industry this will not be further analysed in this report.

Greater knowledge of the shunt effect can help with the environmental impact in that industries would reduce the number of discarded welds, and thereby minimize wasted production. This would lead to a higher efficiency and drive the overall costs down. Furthermore, reducing the effect of shunting, a lower current may be used in production, meaning that less energy will be used, resulting in a more sustainable industry. This type of knowledge will be useful when developing or improving the production further, with reduced number of samplings needed when changing material combinations.

The knowledge of how to account for the shunt effect would mean that less process optimization would be needed for each change in material combination, thus reducing unnecessary material waste and increasing the production rate.

(9)

4

2. Resistance spot welding

Resistance spot welding is a form of liquid phase welding, the materials are locally heated to a liquid state. By applying pressure, a fusion bonding will occur and by maintaining the pressure as the liquid fusion cools, a joint is formed. [1] The spot weld will form due to resistance heating, generated by the total electrical resistance (including contact and bulk resistivities*) between the welded sheets. [3] The timeline for the RSW process can be seen in Figure 3, where two sheets are joined by firstly clamping them together using an electrode pair. The current and resistance in the material generate heat, which in turn melts the welding area between the sheets. When the current is shut off, and the material cools down, the molten metal solidifies into a nugget. The electrodes keep the sheets clamped together throughout the welding process to ensure a consistent joining and controlling the resistance in the circuit; by applying a high pressure a lower resistance will be achieved and vice versa. [1]

*Resistivity being a material property that quantifies how strongly the material resists or conducts electric current. [5]

Figure 3. Illustration of the RSW process timeline.

The final properties such as strength, for a spot weld are dependent on the diameter of the nugget, formed during welding. [1] To verify the quality of the weld, the nugget size is determined; this can be made from a visual examination in combination with producing data visualization, such as a lobe- and current range curve. Using one dimensional-welding curves can help determine the process stability, with a bigger lobe- and current range curve showing a more stable process with higher quality. The relationship between nugget size and weld time can be seen in the current range curve and the lobe curve shows the process window of the current for producing acceptable nuggets. [6] The minimum acceptable size of the nugget can vary depending on the requirements of the industry in question but is approximately between 5√𝑡 and 4√𝑡 , where t is the sheet thickness in millimetre. [1] The maximum nugget size is defined on the curves where the current reaches its maximum value before expulsion occurs. Expulsion is the phenomenon when drops of molten metal eject from the sheets due to formation of a too large nugget. This occurs under excessive heating or insufficient electrode force.

Some expulsion during the process in production is considered acceptable. However, welds that do expulse are never approved and due to the risk of decreased strength this is often compensated by producing additional welds. [6]

(10)

5

There are different welding parameters that regulate the size of the nugget, including the weld current, welding time and the applied force of the electrodes. Other impacting factors are surface coatings, the contact area of the sheets and the design of the electrode tips. The setting of these parameters varies for different types of materials and for each combination there are a range of different quantities of the parameters required to produce acceptable welds. [1]

2.1 Resistance spot welding parameters

As previously mentioned, there are several different types of welding parameters that impact the weld quality. The three process parameters with the greatest impact and that are the easiest to control during the welding are; current, weld time and electrode force.

2.1.1 Weld current

The current in resistance spot welding is a very important variable. If the current is too low, the formed nugget will decrease in size and may not reach the required size or lack the required strength.

On the other hand, if the current applied is too high, expulsion could occur which may result in an unacceptable nugget quality and decreased size. The amount of current that is used varies for different materials and applications. For steel, the optimized weld current should approximately

be 8-10 kA. [1]

In addition to the weld current, a pre-pulse of high current can be used before the actual welding process begins. When using a pre-pulse, a better contact between the metal sheets can be achieved;

this is due to the removal of a coating layer or dirt and improving the contact area in cases of uneven thickness. Pre-pulse can also improve the process if several sheets are being welded

simultaneously. [7]

2.1.2 Weld time

The duration of the welding process includes different intervals; squeeze time, weld time, hold time and off time. Firstly, the squeeze time represents the duration needed to reach the desired electrode force. Secondly, the current flows duration is called weld time. Thirdly, the duration while the weld solidifies but the force is still applied is called hold time. Finally, the off time is the time needed for the electrodes to move to the next weld in the welding sequence. These intervals can have different durations to optimize the production, but the process can also be simplified by ignoring time intervals without material impact on the total process duration. [6]

It is important to obtain a short weld time while succeeding to produce a nugget of the required size.

The industry strives for a high production rate, which makes seeking shorter weld times of great importance. However, a too short weld time may not generate enough heat to form an approved weld and a too long weld time may cause an unwanted expulsion. The weld time varies for different materials, but for steel a typical weld time is approximately 200 ms. [6]

(11)

6

2.1.3 Electrode force

The electrode force is one central parameter due to the importance of contact resistance between the welded sheets; this is due to the need of satisfactory heat generation. A too high force will not generate a high enough heat to melt the material and form an acceptable nugget size due to a decrease of the electrical resistance of the contact area. A high force can also cause excessive deformation.

On the other hand, a too low electrode force may cause internal porosity and cracking of the nugget after the welding. A low force may also create gaps between the sheets; this in turn can lead to increased expulsions due to the generation of additional heat at certain points. [8]

A varying electrode force can be used during the process, but in the automotive industry the force is usually set to a fixed value. Typically, the electrode force for automotive is approximately

3-6 kN. Using a higher electrode force will increase the amount of wear on the equipment which could result in giving it a shorter lifetime. [8]

2.1.4 Other welding parameters

As one can imagine there are several other welding parameters that have an influence on the welding quality, not least depending on which material that is being welded. For example, the electrical resistivity, thermal conductivity, and heat capacity of the material that are being joined are all vital variables. The surface condition of the material is also one parameter, for example welding

of materials with a zinc-layer requires longer weld time or higher current compared to uncoated materials. [1] Geometric factors such as sheet thickness and fitting are also of importance. Other than material properties, the resistance spot welding machine has its own parameters such as cooling water flow, current type, electrode material, electrode degradation, and other design variables that will affect the welding. [6]

(12)

7

3. Shunt effect

When using resistance spot welding for multi-spot welding, a phenomenon called the shunt effect will occur. The shunt effect implies that the electrical current intended to flow through the welded spot instead passes through previously welded spots. Even though the pieces of material that are welded are connected in other areas than the welded spot, the shunt effect does not occur as much there. This is due to an oxide layer and lack of electrode pressure in these areas. The amount of current that flows where it is not intended mainly depends on the previous welds, with respect to the number, size and spacing between them. The spacing between the welds in a welding sequence are called the shunt distance. [9] Figure 4 shows the shunt distance between the first weld (the shunt weld) and the second weld (the shunted weld) in a welding sequence.

Figure 4. Illustration of the shunt distance in a welded sample.

The difference in distribution of current to the welded spots is one consequence of the shunt effect which causes inconsistencies in weld quality. Another consequence of the shunt effect is that due to the differences in distribution of current, the distribution of heat will also vary between welds. This causes a reduction in the diameter of the next formed nugget, which will affect the mechanical properties of the weld. [9]

The shunt effect is complex; a multitude of other factors, such as welding current, welding time, electrode force and geometry of the component all affect the mechanism. The material properties of the welded material such as bulk resistance and surface conditions will also influence the result. [9]

3.1 Previous investigations

Although the impact of the shunt effect in RSW is complex and not completely understood, previously experimental investigations have been made.

One of the first experimental studies of the shunt effect was performed in 1948 by Hard et al. and the study developed a method for measuring the shunt path. The study also showed that when performing a multi-spot-welding sequence, the tensional and shearing strength of the welds in said sequence decreases a significant amount compared to when welding a single spot. [10]

(13)

8

A two-part study including an experimental and theoretical investigation has been done by Wang et al.

The experimental study used different types of mild steel and included the effect of various factors such as material, surface condition and welding schedule. A mild steel is a steel with a low carbon content, making the material malleable and ductile. The conclusions of the study are that the distance between the welds in a welding sequence is the most important factor when examining the shunt effect and increasing this distance will have the biggest impact to avoid it. The surface condition of the material is also one highly important factor regarding shunting, and it was found that highly resistive surface conditions or low electrode force increase the shunt effect. Thereby, having high electrode force usually helps to avoid shunting. However, together with a narrow weld spacing, a high electrode force instead promotes shunting. This is due to that the low force in combination with the narrow spacing creates a lower electrical resistance in the shunting path, meaning that more current will flow there. Other conclusions from the study are that a smaller weld distance can be used when working with thinner materials and materials with large bulk resistivity. [11] The theoretical investigation of the shunt effect analysed the influences of process parameters to obtain an analytical model on the minimum weld spacing. The model developed agreed with the experimental results. The theoretical study has several conclusions, including that a wide minimum weld spacing is needed when welding sheets with larger thickness, the shunt effect is affected by the size of the weld and that a large distance between the welds is required for materials with high yield strength. The study also

emphasizes that if a small weld distance is needed the impact of the welding time, contact resistance and electrode force can reduce the distance. [12]

All in all, both the experimental and theoretical study by Wang et al. resulted in the conclusion that the shunt effect is a complicated phenomenon that depends on all the different parameters, making it difficult to single out the influence of a specific factor affecting the effect. [11,12]

(14)

9

4. Experimental method

The experiments performed in this study were divided into two parts. The first part includes construction of one-dimensional welding lobes to analyse the process window with respect to the current for producing approved sizes of the nugget. The second experiment investigated the impact of the shunt distance, more specifically by welding several welds at a set current while varying the spacing between the welds.

4.1 Experimental equipment

The experiments in this study were performed in Stockholm at SWERIM AB, a metal research institute. The equipment employed during the experiments are presented in Figure 5. Figure 5A.

shows the resistance spot welding machine from Matuschek and Figure 5B. shows the machine that was used to split open the joined sheets. The last-mentioned machine is operated manually. The magnitude of the applied pressure used to split open the sheets is irrelevant for the experiments in this study; therefore, the data was not collected. The specifications of the resistance spot welding machine are listed in Table 1 and the machine was fitted with the electrodes listed in Table 2. During the experiment, currents between 3 kA and 12 kA, as well as forces of 4 kN to 5.5 kN were used.

A. B.

Figure 5. Machines used during the experiments, A. RSW equipment from Matuschek. B. Machine used to split open samples.

(15)

10

Table 1. Specifications for Matuschek RSW equipment.

Table 2. Electrode specifications used during experiments.

4.2 Materials

A range of different materials were used for the experiments to obtain a sufficient amount of data to analyse, the used materials can be found in Table 3. However, all materials are different types of steel.

The experiments used a range of grade, coatings, and thickness of the steel sheets to further investigate how the shunt distance was affected by these variations. When performing the experiments, the two sheets which were welded together consisted of the same material. The different materials were supplied by the joining department of SWERIM AB and consisted of pieces cut in a size of

approximately 125x38 mm. In the first experiment the top five materials in Table 3 were used and the second experiment only used the bottom two.

Table 3. Materials used in experiments.

Max electrode force: [kN] 8

Short circuit current: [kA] 38

Current type: [AC, MFDC] MFDC

Water cooling per electrode: [l/min] 4

Weld control unit PC-based Matuscheck Servo studio

Transformer Expert 222 kVA (4diod) 50 turn ratio

Inverter Matuscheck Servo SPATZ M800LL

Welding gun Matuscheck Servo gun C-type

Specifications

Electrodes Iso Material Geometry

1 5821 Cu-alloy Cap B (20/8)

2 5821 Cu-alloy Cap B (20/8)

Material Coating Thickness [mm]

DP600 GI50/50 1.5

Boron Steel Z140 1.2

Forta H1000 S/S - 1.5

Forta H1000 S/S - 2.0

Boron steel blasted Zn 1.5

DP800 ZE 1.8

(16)

11

4.3 One-dimension welding lobe

The aim for this part of the experiments was to produce a one-dimension welding lobe to evaluate the impact of the shunt distance on a spot weld.

The experiment started with producing a first weld called the shunt weld which in turn was placed on the workpiece. The workpiece consisted of two sheets of the same material. Thereafter, a second weld was produced on the workpiece, which is the one that later was examined. The second weld was placed at a set shunt distance using indications on the bracket holding the sample, to compare the effects of different shunt distances. The weld current was varied during the experiment to find the maximum and minimum current for each material and shunt distance. The minimum current was defined as the lowest current to produce a minimum acceptable nugget size. The maximum current was defined as the highest used current without expulsion occurring. For each current setting, three samples were prepared with a shunt weld. If an expulsion would occur in only one of the three samples when producing the second weld, two more samples would be made in order to increase the accuracy and determine if the current setting would be classified as “failed” or not. A current setting would be deemed as “failed” if there would occur expulsion in more than one sample. For each material there was a set minimum nugget size and to measure the size of the examined welds the sheets were pulled apart. Figure 6 shows a sample that has been welded with a 40mm shunt distance and then been separated; the shunted weld is the one on the top right. With the usage of a digital sliding calliper the average nugget size was measured with the accuracy of 0.1 mm. Using an Excel- sheet a one-dimensional welding lobe was produced for each material at a set shunt distance and then displayed the highest and lowest current possible for creating an acceptable weld. For each material three one-dimensional welding lobes were performed at a shunt distance of 20, 30 and 40 mm, respectively.

Figure 6. Separated Boron blasted steel sample, where the two welds on the workpiece is shown. The shunted weld is the one on the top right.

(17)

12

4.3.1 Welding process parameters

For each material, a different setting of the welding parameters was used to obtain an acceptable nugget that would be measurable when the workpiece was taken apart. For this to be possible, a number of parameter adjustments were required: electrode force, pre-pulse and welding time needed to be tuned for the material undergoing testing. The different welding parameters used for the one- dimension welding lobes are listed in Table 4.

Table 4. Welding parameters for each material.

4.4 Welding with set current

The second part of the experiments in this study was performed to further investigate the effect of the shunt distance on the weld quality. The configuration used in this experiment, simulates a real-world scenario where the welds are placed in the same region with identical parameters.

Two materials were used, the Boron blasted steel and the DP800 steel, whose specifics can be seen in Table 3. After choosing a welding current that would always perform an acceptable weld, samples were prepared with an initial shunt weld, as in the first part of the experiments in this study. The chosen current was based on the result from the one-dimensional welding lobes that was produced in the initial experiment. The second weld was then placed at a set shunt distance, as in the initial experiment, and was welded with the same current as the shunt weld. This second weld was the one that was examined. The shunt distance varied between 20-50 mm and 10-15 samples were made for each distance. The samples were then split open and the average nugget diameter were measured with a digital sliding clipper for the shunted weld.

4.4.1 Welding process parameters

The process parameters used for welding with a set current at a set distance are listed in Table 5. for each material, respectively. For both materials, a pre-weld was employed to obtain an acceptable weld and a measurable nugget.

Table 5. Parameters used for welding with a set current.

Material Squeeze time [ms] Pre-pulse time [ms] Pre-pulse current [kA] Force [kN] Weld time [ms] Current [kA] Hold time [ms]

DP600 100 - - 4 360 7 180

Boron steel 200 500 4.4 4.4 600 6 200

1.5 Forta H1000 500 - - 5.5 250 6 200

2.0 Forta H1000 500 - - 5.5 250 6 200

Boron steel blasted 500 300 6.5 4.7 400 6 200

Material Squeeze time [ms] Pre-pulse time [ms] Pre-pulse current [kA] Force [kN] Weld time [ms] Current [kA] Hold time [ms]

Boron steel blasted 200 300 6.5 4.7 400 8.0 200

DP800 steel 200 300 4.4 5.0 400 7.0 200

(18)

13

5. Results

This chapter shows the results from the experimental parts of the study. The results include one-dimensional welding lobes for each material and a comparison of the nugget size produced at different shunt distances at a set current.

5.1 One-dimension welding lobe

In this section the produced one-dimension welding lobes are illustrated in Figure 7-21. The different lobes for the varying materials and shunt distances were produced using the same method. In the lobes, the welding current (x-axis) is plotted against the nugget size (y-axis). The measured nugget sizes are represented as dots in the diagrams and have different colours depending on whether they are approved or not. The different symbols are shown in the diagrams and to clarify, “Sprut” refers to when an expulsion have occurred and “Limits 1” is the process window for the acceptable current.

5.1.1 DP600

The minimum acceptable size of the nugget for the DP600 material was set to 6 mm. This limitation gives the following lobes for shunt distances between 20-40 mm, shown in Figure 7-9. For 20 mm distance the minimum current is 8 kA and the maximum current is 11.5 kA. For 30 mm distance the minimum current is 8 kA and the maximum current is 9.5 kA. Lastly, for 40 mm distance the minimum current is 7.7 kA and the maximum current is 11.7 kA.

Figure 7. Welding lobe for DP600 steel with 20 mm shunt distance.

4 5 6 7 8 9 10

6,0 8,0 10,0 12,0

Weld Plug Diameter (mm) Cross tensile

Welding Current (kA) Weldlobe

Nugget OK Nugget failed Sprut Nugget Min Limits I

(19)

14

5 6 7 8

6,0 7,0 8,0 9,0 10,0 11,0

5 6 7 8 9 10

6,0 8,0 10,0 12,0 14,0

(20)

15

5.1.2 Boron steel

The minimum acceptable size of the nugget for the Baron steel was set to 4.5 mm. Figure 10-12 shows the produced lobes for shunt distances between 20-40 mm. For 20 mm distance the minimum current is 8 kA and the maximum current is 11.5 kA. For 30 mm distance the minimum current is 8 kA and the maximum current is 9.5 kA. For 40 mm distance the minimum current is 7.7 kA and the maximum current is 11.7 kA.

Figure 10. Welding lobe for Boron steel with 20 mm shunt distance.

3 4 5 6 7

4,0 5,0 6,0 7,0 8,0

3 4 5 6 7

4,0 5,0 6,0 7,0 8,0

(21)

16

5.1.3 Forta H1000 Stainless 1.5 mm

The minimum acceptable size of the nugget for the Forta H1000 Stainless steel, with the sheet thickness of 1.5 mm was set to 6 mm. The minimum and the maximum current for all shunt distances are the same, as seen Figure 13-15. The minimum current is 5.5 kA and the maximum current is 6 kA.

Figure 13. Welding lobe for Forta H1000 with 20 mm shunt distance.

3 4 5 6 7

4,5 5,0 5,5 6,0 6,5 7,0 7,5

4 5 6 7 8

4,0 5,0 6,0 7,0

(22)

17

4 5 6 7 8

4,5 5,0 5,5 6,0 6,5 7,0

4 5 6 7 8

4,5 5,0 5,5 6,0 6,5 7,0

(23)

18

5.1.4 Forta H1000 Stainless 2.0 mm

The minimum acceptable size of the nugget for the Forta H1000 Stainless steel, with the sheet thickness of 2.0 mm was set to 6 mm. As for the Forta H1000 stainless 1.5 mm the current interval does not vary much for the different shunt distances, see Figure 16-18. For the 20 mm distance and the 30 mm distance the minimum current is 5 kA and the maximum current is 6.5 kA. For 40 mm distance the minimum current is 5 kA and the maximum current is 6 kA.

Figure 16. Welding lobe for 2.0 mm Forta H1000 with 20 mm shunt distance.

Figure 17. Welding lobe for 2.0mm Forta H1000 with 30 mm shunt distance.

5 6 7 8 9

4,0 5,0 6,0 7,0 8,0

5 6 7 8

4,0 5,0 6,0 7,0 8,0

(24)

19

Figure 18. Welding lobe for 2.0mm Forta H1000 with 40 mm shunt distance.

5.1.5 Boron blasted steel

The minimum acceptable size of the nugget for the Boron blasted steel was set to 4.5 mm. The lobes related to this material (Figure 19-21) differs from the rest of the lobes in this chapter. As the welding current decreases the curve in the diagram flattens out. The minimum current for 20 mm distance is 5 kA, for 30 mm distance it is 4 kA and for 40 mm distance it is 3 kA. However, considering the appearance of the curve this limit needs further analysation. For 20 mm distance the maximum current is 9 kA, for 30 mm distance it is 9.7 kA and for 40 mm distance it is 10 kA.

Figure 19. Welding lobe for Boron blasted steel with 20mm shunt distance.

5 6 7 8

4,0 4,5 5,0 5,5 6,0 6,5 7,0

3 4 5 6 7 8 9

3,0 5,0 7,0 9,0 11,0

(25)

20

3 4 5 6 7 8 9

3,0 5,0 7,0 9,0 11,0

3 4 5 6 7 8 9

2,0 4,0 6,0 8,0 10,0 12,0

(26)

21

5.1.6 Compilation of results

To get an overview of the results from the experiment creating one-dimensional welding lobes a compilation of the results regarding the current range at different shunt distances for each material with associated minimum nugget size can be seen in Table 6.

Table 6. Compilation of current range for each material at three different shunt distances.

5.2 Welding with set current

In this section the results from the experiments when welding with a set current are presented. The results are illustrated in two graphs where the measured nugget size (diameter) is displayed as a dot for each shunt distance. A line is drawn between the dots to indicate possible relationships. The calculated standard deviation for each distance is illustrated as a black vertical line.

5.2.1 Boron blasted steel

In Figure 22 the result from the experiment with the Boron blasted steel is displayed. As one can see in the graph, the line does not indicate a linear relationship. The nugget size increases for each step-up in shunt distance up to 40 mm. After that, the measured nugget size at 50 mm shunt distance decreases and is lower than both the value for 30 mm and 40 mm shunt distance. The results have a relatively large but consistent standard deviation.

Figure 22. Result when welding with Boron blasted steel. Nugget size for each shunt distance, displayed with standard deviation.

Material Min. nugget size [mm] 20 mm 30 mm 40 mm

DP600 6.0 8.0 - 11.5 8.0 - 9.5 7.7 - 11.7

Boron steel 4.5 5.5 - 7.0 5.5 - 7.0 5.7 - 6.7

1.5 Forta H1000 6.0 5.5 - 6.0 5.5 - 6.0 5.5 - 6.0

2.0 Forta H1000 6.0 5.0 - 6.5 5.0 - 6.5 5.0 - 6.0

Boron steel blasted 4.5 5.0 - 9.0 4.0 - 9.7 3.0 - 10.0

Current range [kA] for shunt distance:

6,10 6,15 6,20 6,25 6,30 6,35 6,40 6,45

20 30 40 50

Nugget size [mm]

Shunt distance [mm]

(27)

22

5.2.1 DP800 steel

Figure 23 shows the result from the experiment with DP800 steel. The graph indicates that there may be a linear relationship between the nugget size and the shunt distance when welding with this

material. The nugget size increases as the shunt distance increases. The results have a varying standard deviation. The largest standard derivation is for the welds performed at 30 mm shunt distance and the smallest standard derivation is for the welds performed at 40 mm shunt distance.

Figure 23. Result when welding with DP800 steel. Nugget size for each shunt distance, displayed with standard deviation.

5.2.2 Compilation of results

The results from the experiment when welding at a set current are shown in Table 7. The table shows the compilation of the results which includes the nugget size displayed with standard derivation for each shunt distance, for each material.

Table 7. Compilation of results where the nugget size for each shunt distance and material are presented with standard deviation.

4,40 4,50 4,60 4,70 4,80 4,90 5,00 5,10 5,20

20 30 40 50

Nugget size [mm]

Shunt distance [mm]

Material 20 mm 30 mm 40 mm 50 mm

Boron steel blasted 6.2 ∓ 0.08 6.3 ∓ 0.1 6.3 ∓ 0.1 6.3 ∓ 0.07

DP800 4.6 ∓ 0.17 4.7 ∓ 0.27 4.9 ∓ 0.16 4.9 ∓ 0.2

Nugget size [mm] for shunt distance:

(28)

23

6. Discussion

The main goal for the project was to create guidelines for optimizing the weld quality when using resistance spot welding. The study is based on the theory that the quality of a weld performed by a RSW process is dependent on the shunt distance. The aim of the project was to create guidelines for the process by investigating spot welds performed on different materials at different shunt distances.

This was made by creating welding lobes; finally generating process windows for producing acceptable nuggets. In addition, the study aims to examine differences in the size of the nugget

between the investigated welds at different shunt distances. This was made by investigating spot welds which were welded at a set current. In the study the manual measurements of the nugget sizes were performed by a digital calliper and the welding process was performed by a resistance spot welding equipment from Matuschek.

6.1 One-dimension welding lobe

When looking at the results from the experiments the plots from each shunt distance look very similar and the current interval behaves likewise. There are some exceptions where there is a larger difference in the current interval which can be seen in the experiments with the DP600 material. There is a substantial difference when changing the shunt distance from 40 mm to 30 mm as the current interval went from 4 kA to 1.5 kA as can be seen in Figure 9 and Figure 8. But when changing the shunt distance from 30 mm to 20 mm the current interval went from 1.5 kA to 3.5 kA, as seen in Figure 7.

One reason for this behaviour could be that when performing the experiments at 30 mm shunt

distance, a factor that could not be regulated affected the result, e.g. dirty sheets. If this is the case, the lack of removal of the layer on the sheets would result in expulsion at an earlier state than expected. If the sheets are in different conditions, it may be problematic comparing the results of the experiments.

Overall observing Figure 7-9, the slope for all three shunt distances are similar; this implies that the material behaves in a similar way when welding with various shunt distances which indicates that the current range should be consistent.

In the experiment with the blasted Boron steel, a pre-pulse was used to achieve a nugget that would be measurable and have the demanded size. The usage of the pre-pulse resulted in that there were not any welds that did not pass the minimum size requirement, seen in Figure 19-21. Illustrated in the plots, the nugget size is linearly dependent of the current, but at 6 kA for 40mm, 30mm and 7 kA for 20mm shunt distance the curves flatten out. This is assumed to be the minimum value when using the set parameters for the pre-pulse. A too high current will generate a large amount of heat; therefore, the nugget size cannot fail to reach its minimum. This results in that there are some differences to the minimum current when changing the shunt distance, and that the overall welding lobe for the material is smaller for 20mm than for 40mm by approximately 2 kA.

When looking at the results from the experiments performed on the two Forta H1000 steels and Boron steel, there were only minor deviations between the different distances. Because of the small variations it is difficult to draw any conclusions on how the shunt distance affects the size of the welding lobe.

For the Forta H1000 stainless steel (2.0 mm) there were some difficulties in splitting open the workpiece at the higher end of the current range which could explain the shape in Figure 16 and Figure 17.

(29)

24

The Forta H1000 stainless steel (1.5 mm) is the material that is the least affected by the changes in shunt distance. This could be due to that the Forta H1000 (1.5 mm) had a very clean surface with almost no contaminations, which could help ensure a good contact area between the sheets. The good surface condition together with high electrode force could, as stated in the literature study, decrease the impact of shunting. This could explain why the current range is unchanged over the different shunt distances.

6.1.1 Sources of error

In some of the experiments there was a limitation in the number of available samples. Due to this not as many experiments as desired could be performed. This will influence the accuracy of the results and is a source of error.

The measurement of the nugget size was done manually by hand with a digital sliding clipper. To be able to measure the nugget the joined sheets had to be separated by the machine in Figure 5B. If the material was soft and ductile, as for the Forta H1000 stainless steel, difficulties in splitting the sheets were found. Due to the material properties, instead of splitting open the joined sheets the sample was pressed together. The inability to split open the joined sheets caused that in some cases it was not possible to measure the nugget size, or the measurement was more approximately. However, during the measurements the most important values are the minimum and the maximum which defines the process window. Meaning that the accuracy of all the individual nugget sizes does not matter as much.

The process of developing the one-dimension welding lobes are not completely reliable and has its own sources of error. Nevertheless, for the applications that uses the lobes the result is accurate enough. However, with introduction of new materials and material combinations the accuracy is not high enough, which is the reason why new methods for manufacturing the lobes are under

investigation.

6.2 Welding with set current

The results from welding at a set current with two different materials, the blasted Boron steel and the DP800 steel, show very different results which were not expected. Figure 23 shows that the nugget size decreases linearly with a decreasing shunt distance for the DP800 steel. For this material, the nugget size was reduced by approximately 0.1 mm for each 10 mm reduction in shunt distance. The experiment with the Boron blasted steel which result is shown in Figure 22 does not show the same behaviour due to its nugget size increasing when going from 50 mm to 40 mm shunt distance. Apart from that, the nugget size decreases when decreasing the shunt distance.

The unexpected results from these two experiments can be an effect of the different surface conditions on the two different materials. Both material sheets used in the experiments have a zinc treatment but the method for coating the sheets are different. The blasted Boron steel has a normal zinc coating which has been applied using hot dip galvanizing and the DP800 steel has been electrogalvanized.

This difference in coating method could be one explanation to why the result differs between the two materials when using a shunting distance of 40 mm to 50 mm. The different alloy content in the two materials could also explain the different behaviour of the two materials. A higher alloy content usually implies a higher resistance which will generate more heat and form lager joints.

(30)

25

Another reason why the blasted Boron steel does not have a linear reduction of the nugget size can be the impact of the electrode force. As seen in the literature study a high electrode force decreases the shunt effect up to a certain distance between the welds. A narrow weld spacing together with a high electrode force promotes the shunting. The electrode force in the welding experiment with the blasted steel is relatively high. Therefore, when producing welds at a small shunt distance the nugget size may decrease to a greater extent. This can be an explanation to why the nugget size at the shunt distance of 20 mm deviate. However, this does not explain the deviation of the value at the shunt distance of 50 mm.

A third reason why the nugget size decreases when welding at 50mm, could be that the second weld is placed closer to the edge of the sample. The samples were cut in approximately a size of 125x38 mm, meaning that when increasing the weld spacing the weld was placed closer and closer to the edge. This placement could have an effect on the growth of the nugget, but due to the samples already been cut there was no possibility to further increase the distance and investigate how the nugget size would change was simply not possible.

The low value at 50 mm for the blasted Boron steel could have been explained by an expulsion. The expulsion ejects material from the sheets meaning that the nugget size decreases significant, as in this case. Although, no signs of expulsion were observed in this experiment.

It seems logical that the second weld would have a smaller nugget size compared to the first shunt weld when the shunt distance is decreased due to that a higher amount of current would go through the shunt weld. With the results from the experiments it is clear that the shunt effect is not simply

controlled by the shunt distance. The deviation could be an effect from the different alloying content in the materials, with higher amount of alloying elements usually resulting in a higher resistivity.

6.2.1 Sources of error

An important factor when discussing what impact the shunt distance have on the result is the accuracy of the experiments. Figure 22-23 shows a quite wide range of the nugget size, considering the standard derivation. The low accuracy may be due to that when the nugget was measured the average size was calculated from two measured values using a digital sliding calliper. Some of the nuggets were not completely round, meaning that the measuring may differ. This means that when analysing the results in Figure 22 the reduction could be linear if the extremes are used, even though it seems unlikely.

The manual measurements of the nugget size were made was with the accuracy of 0.1 mm. This magnitude of the accuracy is typically used in the applications of the automotive industry. However, Figure 22-23 plot the result in a greater accuracy, regarding the y-axis. Resulting that in Figure 22 the nugget size produced at the shunt distances of 30, 40 and 50 mm may be false due to that the nugget size is the same for all distances with the accuracy of 0.1 mm. In addition, Figure 22 and Figure 23 have different scales on the y-axis, this can be misleading if not being noticed.

(31)

26

7. Concluding remarks

The aim of this project was to create guidelines for optimizing the weld quality when working with RSW, more precisely how the shunt distance affects the quality of the welds performed in the same region. The results from this study shows that when creating a one-dimensional welding lobe for a material the shunt distance does not have a substantial effect on the results. When looking at a specific material such as the DP600 steel there was a substantial decrease in the current range, which dropped 37.5% when reducing the shunt distance with 25%. This result was extraordinarily large and was not seen when further decreasing the shunt distance or with other materials. This implies that the current range achieved for one shunt distance is the same that would be achieved at another shunt distance.

However, it is worth mentioning that from both this study and from previously made studies, one can see that the shunt effect will occur and affect cases where multiple spot welds are performed.

Regarding the Boron blasted steel and Forta H1000 there are difficulties in making any conclusions due to the usage of a high pre-pulse and not being able to split open the samples because of the ductile material properties.

The result also shows that when performing welds with a set current at different shunt distances there will occur a decrease in nugget size when the distance is decreased. The decrease of nugget size was approximately 0.1 mm for each 10 mm decrease of shunt distance for the DP800 steel. This amplifies that the shunt effect occurrence causes the electrical current which is intended to flow through the second weld spot instead flows through the previously welded shunt weld. Thus, it is neither harder nor easier to weld at a certain shunt distance, but it is important to be aware of which parameters that should be used at a specific distance in order to maintain the same weld quality. The result concerning the decreasing nugget size is highly dependent on material properties, coatings and thickness which makes creating a guideline based on this data difficult. Even if such a guideline could be written, it would be unreliable when switching between various materials.

(32)

27

8. Further study

In order to create guidelines for working with RSW, considering the impact of the shunt effect, that could be implemented in the automotive industry there is need for more experimental work. This is needed to fully verify the impact that the shunt effect has on the weld quality and what parameters that the effect depends on.

To achieve a result with sufficient accuracy, future work could investigate a wider range of different materials with various properties. Furthermore, it would be interesting to investigate the impact of the shunt effect when joining two or more sheets of different materials. Further studies could also

investigate what will happen when there are two or more spot welds in the welding sequence, what impact the shape of the samples will have, and the effect shunting will have on welds placed in a different configuration than one after another. This would be beneficial due to the better resemblance to the real-world production where the workpieces often consist of different materials and shapes.

Other process parameters than these in this study could also be investigated by doing a greater amount of experiments where some parameters are fixed, and some vary to determine their impact on the welds. The knowledge of what parameters the shunt effect depends on is needed to create a guideline that could be implemented in today´s industry and be used to reduce the environmental impact of the automotive industry production.

(33)

28

9. Acknowledgements

SWERIM AB is thanked for making this study possible and allowing us to use the company’s

premises including equipment and samples to conduct the experimental parts in the project. We would like to extend our greatest gratitude to our supervisors, David Löveborn, Research leader at the Joining department, at SWERIM and Tova Jarnerud from the unit of Process at the department of Material Science and Engineering at KTH Royal Institute of Technology. Whose support and expertise in the subject have helped us complete this study. Their assistance and remarks during the writing of this report have also been very helpful. In addition, one last thank you is directed to Emil Stålnacke who was our initial contact at SWERIM and who therefore made it possible for us to perform this study at SWERIM.