Effects of Welding Factors on Liquid Metal Embrittlement Cracking during Resistance Spot Welding of Zinc-coated High-strength Steel

Abstract

During resistance spot welding of zinc-coated advanced high-strength steels (AHSS), cracks which are promoted by liquid metal embrittlement (LME) may occur. According to previous researches, welding current, welding time, electrode tip diameter and electrode misalignment angle are significant factors on LME cracking. In this study, the effects of these four factors on LME cracking were compared with an orthogonal design experiment. After welding, numbers and lengths of cracks were measured, and Crack Index (CI) of each experiment group was calculated for comparison. As the result, two types of LME cracks, i.e., Type A cracks located at the indentation of electrode tip and Type B cracks located at the indentation of electrode shoulder were observed. The order of influence degree on CI of Type A crack from high to low was welding current > electrode tip diameter > electrode misalignment angle > welding time. The order of influence degree on CI of Type B crack was electrode tip diameter > electrode misalignment angle > welding time > welding current. The result reveals that modifying welding current is the most effective way to reduce Type A cracks and optimizing the electrode tip diameter is significant for preventing Type B cracks.

1. Introduction

In recent years, for improving the vehicle safety while satisfying the need of lightweight, the use of advanced high strength steels (AHSS) have been continually developed for Body-in-White(BIW) of vehicle. Among them, the third generation of AHSS, presented by Dual Phase steel with high formability (DH), Quenching & Partitioning (Q&P), Transformation Induced Plasticity Aided Bainitic Ferrite (TBF) steel is gradually becoming the focus in the field of BIW material, due to its higher combination of outstanding mechanical property and relatively low cost.^1,2,3) At the same time, in order to protect the BIW against the impact of corrosion environment during service, AHSS is usually coated with zinc-based coating layer via hot-dip galvanized (GI), Galvannealed (GA) or Electrogalvanized (EG) process.^4,5)

Resistance spot welding (RSW) is the most commonly used welding method in BIW assembly.^6,7) However, the third generation of AHSS with zinc-based coating is facing the liquid metal embrittlement (LME) problem during RSW process because of its susceptible alloying design and microstructure.^8,9,10) During RSW process, the zinc-based coating with low melting temperature melts and penetrates into the grain boundary of the steel substrate, which will weaken the cohesion of grain boundary, then, intergranular cracks can be initiated and propagate under tensile stress induced by welding process.^11,12,13) Since LME cracks potentially affect the mechanical property of the RSW joint, resulting in the risk of strength degradation, the application of the third generation AHSS are being seriously restricted because of this problem.^14,15,16)

According to the mechanism mentioned above, temperature and tensile stress are two critical factors impacting LME cracks. Temperature and stress distributions during RSW process are directly related to welding parameters, electrode geometry and undesirable welding conditions, so that several previous researches focus on the effects of these variables on LME cracks. Among them, Kim et al.¹⁷⁾ investigated the influences of four welding parameters namely, welding current, welding time, electrode force and holding time on LME surface cracks, and found that surface cracks became severer with the increase of welding current, prolonging of welding time and decrease of electrode force. Meanwhile, surface cracks tended to be slightly decreased due to the increase of holding time. Ling et al.¹¹⁾ also studied the impacts of these four welding parameters above by an orthogonal design experiment, and came to a similar conclusion. At the same time, the analysis of this study revealed that sequence of influence degree on LME cracks from high to low was welding current > electrode force > welding time > holding time. At the same time, LME cracking has been tried to mitigate via modification of welding parameters by numbers of studies. Ashiri et al.¹⁸⁾ reduced LME cracks with a multi-pulse welding schedule, contributed to a slower increase of heat input. DiGiovanni et al.¹⁹⁾ developed a ramp down current to alleviate LME cracking, as a result of a shorter time in the critical temperature range and averting a sudden increase in tensile stresses at the weld surface. Patel et al.²⁰⁾ and Midawi et al.²¹⁾ both found that 5-pulse welding schedule could result in less LME severity compared to 2-pulse one with the same time, due to the reduction in temperature and stress. Böhne et al.²²⁾ showed that prolonging holding time could also alleviate LME cracks happening, attributed to controlling the cooling rate after welding. Choi et al.²³⁾ studied the influence of electrode force on LME cracks by experiment and simulation, indicating that the LME cracks were decreased by higher electrode force because of modified temperature and thermal stress. Meanwhile, current researches show that electrode geometry is also critical for LME cracks. Böhne et al.²⁴⁾ investigated the effect of electrode geometry on LME cracks locating at the periphery of electrode indentation. The result showed that the crack formation was stimulated by elongating welding time from 1 to 4 times, and electrode with larger tip diameter could effectively reduce this risk resulted from the excessive welding time. Murugan et al.²⁵⁾ compared the LME behaviors welded by a radius type electrode with different radius of curvature (R) and a dome radius type electrode with variable tip diameter (d), and showed that the radius type ones had a lower LME tendency than the dome radius type ones, LME tendency decreased with the increase of R and d. The mechanism was explained as that, with the increase of R and d, the contact area between the electrode and steel sheet increased and current density decreased, resulting in the reductions of temperature and tensile stress at the surface of steel sheet, which led to a lower LME tendency. DiGiovanni et al.²⁶⁾ compared the differences of LME severity among 3 types of electrodes, radius type, dome radius type and truncated type. The result demonstrated that LME cracks locating in the shoulder region of indentation became severer when welded with truncated and dome radius type electrodes. According to the simulation result, electrode collapse, a phenomenon that the electrode suddenly pressed into the steel sheet, would occur during RSW process welded with these two types of electrodes. This phenomenon would lead to a sudden local cooling and a material contraction in the shoulder region, resulting in large tensile stress on the adjacent position, and large LME cracks occurred. Furthermore, in practical industrial production, RSW process would face some unexpected welding conditions because of assembly errors from welding equipment or parts. The electrode misalignment is one of the undesirable conditions that can affect LME occurrence. Studies by Siar et al.²⁷⁾ and DiGiovanni et al.²⁸⁾ both found that the electrode misalignment could lead to more severe LME cracks, owing to the serious stress concentration caused by the bending of workpiece and asymmetric stress field.

As mentioned above, LME crack severity is significantly affected by welding parameters, electrode geometry and undesirable welding conditions such as electrode misalignment. However, there is still a lack of study about the combined impacts of these three aspects of factors on LME cracks. In this study, an orthogonal experiment with four factors and four levels was designed to investigate the influences of welding current, welding time, tip diameter of dome radius type electrode and angle of electrode misalignment, and the influence degrees of this four factors on different types of LME cracks were obtained respectively, to provide a reference for LME cracking controlling in practical industrial RSW process.

2. Experimental Method

2.1. Experimental Material

In this test, a GI coated 1000 MPa grade tensile strength third generation AHSS was selected, with a sheet thickness of 1.6 mm. The chemical composition of base material is 0.15 C, 1.50 Si, and 2.20 Mn, in mass percentage. A few of Al, Nb, and Cr elements were also added to enhance the mechanical performance, and the remained elements were Fe and other impurities. The thickness of GI coating was about 7 μm, and the mechanical properties of experimental material is shown in Table 1.

Table 1. The Mechanical properties of the as-received steel.

Material	Yeld Strength (MPa)	Ultimate Tensile Strength (MPa)	Elongation to failure (%)
third generation AHSS	700	1000	18

2.2. Resistance Spot Welding

Before welding, experimental material sheet was cut into 125 × 40 mm specimens, then the specimens were cleaned with ethanol for preparation. The spot welding experiment was conducted by a stationary direct current (DC) inverter-controlled spot welding machine, with a single-pulse welding current approach. During experiment, electrode force and holding time were fixed, as 4.5 kN and 300 ms separately, welding current and welding time were set as variables. Cu–Cr dome radius type electrode with different tip diameters were used for welding, as illustrated in Fig. 1, tip diameter was the variable, and curvature radius of the tip was a constant, as R40. In this test, several welding tests were conducted with electrode misalignment larger than 0°, as shown in Fig. 2. The angle of electrode misalignment was set as variable. A L₁₆(4⁵) orthogonal experiment was designed to evaluate the influences of four variables above on LME cracks. The factors and their levels for the orthogonal design were summarized in Table 2. The detailed welding conditions for each group were demonstrated in Table 3. Among them, the group 1 welding condition was a base condition, with the welding time, electrode size and electrode condition commonly used in industrial production, and with a highest current which doesn’t cause expulsion. For facilitating LME crack occurring, higher welding current and longer welding time were selected for this test. Furthermore, in order to ensure the reliability of the evaluation result, three samples were welded separately for each experimental group.

Fig. 1. Schematic of dome-radius type electrode.

Fig. 2. Schematic of spot welding with electrode misalignment.

Table 2. Selected factors and their levels for the orthogonal design.

Factors	Symbols	Level 1	Level 2	Level 3	Level 4
Electrode tip diameter (mm)	A	6	8	10	12
Electrode misalignment (°)	B	0	3	5	7
Welding current (kA)	C	7.8	8.6	9.4	10.1
Welding time (ms)	D	380	760	1140	1520

Table 3. Welding conditions for each group of L₁₆(4⁵) orthogonal design.

Welding Group	Factor Symbol
	A	B	C	D
1	6	0	7.8	380
2	6	3	8.6	760
3	6	5	9.4	1140
4	6	7	10.1	1520
5	8	0	8.6	1140
6	8	3	7.8	1520
7	8	5	10.1	380
8	8	7	9.4	760
9	10	0	9.4	1520
10	10	3	10.1	1140
11	10	5	7.8	760
12	10	7	8.6	380
13	12	0	10.1	760
14	12	3	9.4	380
15	12	5	8.6	1520
16	12	7	7.8	1140

2.3. LME Crack Characterization

According to previous studies,^29,30) LME cracks could be classified into three types based on locations, as shown in Fig. 3(a). Type A outer cracks are located inside the indentation region, Type B outer cracks are at or outside the shoulder of the electrode indentation region, and Type C inner cracks are found at the lap edge of the sheets around the welding. Several researches have revealed that crack location has a significant influence on strength of welding joint. Specifically, Type A cracks hardly had influence on joint strength, while Types B and C cracks may reduce it when the LME cracks are located on the same side as the load direction.^{14,15,16,29,30)} However, when the electrode misalignment is larger than 0°, the cross-sectional schematic diagram of a spot weld is different from the one without misalignment, presenting an asymmetrical shape. In order to categorize LME crack types more easily under this condition, Type A cracks are identified as the outer cracks locating inside the area of electrode tip diameter, and Type B cracks are other ones locating outside the area mentioned above, reffering to the method by Okigami et al.,³¹⁾ as shown in Fig. 3(b).

Fig. 3. (a) Schematic of LME crack classification, (b) Identified location of Type A crack with the electrode misalignment. (Online version in color.)

After RSW welding experiment, all RSW samples were cross-sectioned through the center of the welds. For samples belonging to the group with a 0° electrode misalignment, firstly, outer cracks were detected by a microscope, then cutting direction was determined as through the detected most severe crack. For others, cutting direction was parallel to the electrode misaligned direction. Then, cut RSW samples were all mounted, ground and polished for LME crack observation. LME cracks of each sample were observed by a Leica DMI5000 M optical microscope, and the magnification was 100 times. After that, the length of each crack was measured and the numbers of 3 types of cracks were calculated, using an image processing software Image J. Then, Crack Index (CI) for all 16 experimental groups was calculated by the Eq. (1):

  

$$CI=\text{max}\left\{\frac{{{\displaystyle \sum }}_{\text{i}=1}^{{\text{n}}_1}{L}_{1i}}{t},\frac{{{\displaystyle \sum }}_{\text{i}=1}^{{\text{n}}_2}{L}_{2i}}{t},\frac{{{\displaystyle \sum }}_{\text{i}=1}^{{\text{n}}_3}{L}_{3i}}{t}\right\}$$(1)

where L_1i, L_2i and L_3i are the measured lengths of crack i for three repeated samples in the same experimental group, n₁, n₂ and n₃ are the numbers of cracks for three repeated samples in the same experimental group, and t is the sheet thickness. CI represents the maximum value of the ratio of the sum of crack length to sheet thickness for each group. CI of each LME crack type was calculated separately for analysis. After crack observation, all samples were etched with 3% HNO₃+97% ethanol, and then nugget diameters were observed and measured. The average of nugget diameter (ND) for each group was calculated for analysis.

3. Experimental Results and Discussions

3.1. Results of Orthogonal Design Experiment

The results of Crack Indexes for Type A and B LME cracks (CI_A and CI_B), and NDs for each group were summarized in Table 4. Expulsion conditions were also recorded in the same Table. There were no type C cracks observed in all samples, proposing that the experimental material used didn’t cause the formation of Type C cracks under the welding conditions in this test.

Table 4. Results of orthogonal design experiment.

Experimental Group	Factor Combination	CIA	CIB	ND	Expulsion
1	A1B1C1D1	0.01	0.05	6.73	No
2	A1B2C2D2	0.34	0.37	7.35	Yes
3	A1B3C3D3	2.05	1.51	7.92	Yes
4	A1B4C4D4	0.94	1.02	7.43	Yes
5	A2B1C2D3	0.00	0.00	8.08	No
6	A2B2C1D4	0.00	0.18	7.93	No
7	A2B3C4D1	0.35	0.04	8.19	No
8	A2B4C3D2	0.53	0.16	8.55	No
9	A3B1C3D4	0.03	0.00	8.84	No
10	A3B2C4D3	0.74	0.04	9.48	No
11	A3B3C1D2	0.20	0.00	7.24	No
12	A3B4C2D1	0.44	0.00	6.37	No
13	A4B1C4D2	0.38	0.00	8.79	No
14	A4B2C3D1	0.48	0.00	8.35	No
15	A4B3C2D4	0.23	0.00	8.40	No
16	A4B4C1D3	0.00	0.00	6.43	No

Analysis of mean value is a common method used for evaluating the influences of factors on the result. k value, the index for this analysis, was calculated by the Eq. (2):

  

$$k_{Ai}=\frac{{{\displaystyle \sum }}_{j=1}^n{k}_{ij}}{n}$$(2)

where k_Ai is the mean value on level i for factor A, k_ij represents each experimental result on level i, and n is the number of results on level i. Then, R value, the difference between the maximum and minimum k_Ai values, was calculated to evaluate the effect degree of each factor on the result.

The analysis results of mean value for CI_A are summarized in Table 5, and plotted in Fig. 4. It can be seen that CI_A generally decreases with the increase of the diameter of electrode tip and the decreasing of electrode misalignment angle and welding current, the influence of the welding time is not significant. The order of influence degree on CI_A from high to low is, welding current > electrode tip diameter > electrode misalignment angle > welding time.

Table 5. Analysis result of k-value of CI_A.

	Electrode tip diameter A	Electrode misalignment B	Welding current C	Welding time D
k1	0.84	0.11	0.05	0.32
k2	0.22	0.39	0.25	0.36
k3	0.35	0.71	0.77	0.70
k4	0.28	0.48	0.60	0.30
R	0.62	0.60	0.72	0.40

Fig. 4. Effects of four factors on CI_A. (Online version in color.)

Table 6 and Fig. 5 represent the analysis results of the effects of each factor on CI_B. It reveals that CI_B obviously decreases with the increase of electrode tip diameter, and slightly increases with the increase of electrode misalignment angle, welding current and welding time. For CI_B, the order of influence degree from high to low is, electrode tip diameter > electrode misalignment angle > welding time > welding current.

Table 6. Analysis result of k-value of CI_B.

	Electrode tip diameter A	Electrode misalignment B	Welding current C	Welding time D
k1	0.74	0.01	0.06	0.02
k2	0.09	0.15	0.09	0.13
k3	0.01	0.39	0.42	0.39
k4	0.00	0.30	0.28	0.30
R	0.74	0.38	0.36	0.37

Fig. 5. Effects of four factors on CI_B. (Online version in color.)

The analysis results of ND are shown in Table 7, and Fig. 6. For the nugget diameter, ND significantly increases with the increase of welding current, and slightly increases with the increase of electrode tip diameter and welding time. Generally, it is well known that the nugget diameter decreases with a large electrode tip diamater, so this study comes to a contrary result. However, it could be seen from Table 4 that expulsion occurred when welded with the minimum tip diameter of 6 mm, and the nugget couldn’t grow up because of the expulsion phenomenon, so that in this study the increase of nugget diameter with the larger electrode tip diameter results from the suppression of expulsion with larger electrode tip. At the same time, ND decreases with the increase of electrode misalignment angle. The order of influence degree from high to low is, welding current > electrode misalignment angle > electrode tip diameter > welding time.

Table 7. Analysis result of k-value of ND.

	Electrode tip diameter A	Electrode misalignment B	Welding current C	Welding time D
k1	7.36	8.11	7.08	7.41
k2	8.19	8.28	7.55	7.98
k3	7.98	7.94	8.42	7.98
k4	7.99	7.20	8.47	8.15
R	0.83	1.08	1.39	0.74

Fig. 6. Effects of four factors on ND. (Online version in color.)

Analysis of variance was used to quantitatively evaluate the contribution of each factor to CI_A, CI_B and ND. The Eqs. (3) and (4) were used for this analysis:

  

$$S_A={\sum }_{i=1}^m{(k_{Ai}-\overline{k})}^2$$(3)

  

$$C_A=\frac{S_A}{\sum S_A}\times 100\text{\%}$$(4)

where S_A is the sum of the square variations between levels of factor A, k_Ai is the mean value on level i for factor A, $\overline{k}$ is the mean value of k_Ai for all factors on all levels, m is the number of level for factor A. C_A is the contribution of factor A, ΣS_A is the sum of S_A of all factors.

The analysis of variance results for CI_A, CI_B and ND are shown in Tables 8, 9, 10. It can be seen that the analysis of variance is similar to the results of the mean value. C_A of welding current to the CI_A is the greatest, about 38%. For CI_B, electrode diameter provides the substantial contribution, about 60%. C_A of welding current to the ND is most significant, about 50%.

Table 8. Results of the analysis of variance for CI_A.

Factor	k value of CIA on each level				Sum of squares (SA)	Contribution (CA)
	1	2	3	4
Electrode tip diameter	0.84	0.22	0.35	0.28	0.24	27.97
Electrode misalignment angle	0.11	0.39	0.71	0.48	0.19	21.74
Welding current	0.05	0.25	0.77	0.60	0.32	38.00
Welding time	0.32	0.36	0.70	0.30	0.10	12.28
					0.85	100

Table 9. Results of the analysis of variance for CI_B.

Factor	k value of CIB on each level				Sum of squares (SA)	Contribution (CA)
	1	2	3	4
Electrode tip diameter	0.74	0.09	0.01	0.00	0.38	60.00
Electrode misalignment angle	0.01	0.15	0.39	0.30	0.08	13.30
Welding current	0.06	0.09	0.42	0.28	0.09	13.58
Welding time	0.02	0.13	0.39	0.30	0.08	13.13
					0.63	100

Table 10. Results of the analysis of variance for ND.

Factor	k value of ND on each level				Sum of squares (SA)	Contribution (CA)
	1	2	3	4
Electrode tip diameter	7.36	8.19	7.98	7.99	0.39	14.03
Electrode misalignment angle	8.11	8.28	7.94	7.20	0.68	24.51
Welding current	7.08	7.55	8.42	8.47	1.39	50.13
Welding time	7.41	7.98	7.98	8.15	0.31	11.33
					2.77	100

3.2. Metallographical Analysis of LME Cracks

In this study, samples of several experimental groups were welded under the conditions with electrode misalignment. According to the results, when samples are welded with electrode misalignment, there are two kinds of typical cross-sectional morphologies, i.e., the cross-sectional image with the electrode indentation shoulders, and the cross-sectional image without the electrode indentation shoulders. Figures 7 and 8 demonstrate the cross-sectional images and crack occurrence conditions of the two kinds of morphologies. Therein, Fig. 7 shows the results of a sample with electrode indentation shoulders, from A₂B₄C₃D₂ group. For this sample, the electrode tip diameter is small, and the electrode indentation includes electrode tip regions and electrode shoulder regions. At the same time, plastic deformation regions, reported in previous study,³²⁾ also exist resulting from the electrode misalignment. In this case, Type A cracks occurring at the electrode tip regions near the plastic deformation regions, and Type B cracks locating at the deeper sides of the electrode shoulder regions can be observed. Figure 8 demonstrates the results of a sample from A₄B₂C₃D₁ group. It can be seen that when the diameter of electrode tip is on a high level, different from Fig. 7, the electrode indentation includes only electrode tip regions, without the electrode shoulder regions. Meanwhile, plastic deformation regions still exist from the image. For this situation, Type B cracks are not observed, and there are only Type A cracks, locating near the plastic deformation regions.

Fig. 7. Cross-sectional metallographical image and crack occurrence condition of sample with electrode indentation shoulder. (a) Cross-sectional image; (b) Type B crack distribution; (c) Type A crack distribution. (Online version in color.)

Fig. 8. Cross-sectional metallographic image and crack occurrence condition of sample without electrode indentation shoulder. (a) Cross-sectional image; (b) Type A crack distribution-1; (c) Type A crack distribution-2. (Online version in color.)

3.3. Mechanism of LME Cracks Under the Condition with Electrode Misalignment

Based on the former results, Type A crack is most affected by welding current, and electrode tip diameter is the most significant factor for Type B crack. For explaining the results, the mechanism are discussed.

The schematic diagram of mechanism of Type A LME crack under the condition with electrode misalignment is shown in Fig. 9, the occurrence of Type A cracks is explained by the following presumption. Under this condition, since the electrode is not perpendicular to the sheet, the parts of the sheet under the electrode would have a tendency to rotate into the direction perpendicular to the electrode. Due to the constraint of the surrounding materials and welding jig, a bending moment M will act on the sheet under the electrode. Then, under the bending moment M, a force F_f along the sheet surface in contact with the electrode will be generated. During welding process, the welded sheet softens because of resistance heating. Under the force F_f, the softened material exhibits plastic flow along the tip of electrode, and at last the plastic deformation regions form because of this phenomenon. However, during the process of plastic flow, a reaction force F_c, opposite to F_f, is also created along the sheet surface because of the constraint. Finally, if the welded material has a high LME susceptibility, Type A cracks could be induced by the tensile stress generated under the action of F_f and F_c. According to the above analysis, Type A cracks occur due to the large tensile stress caused by the plastic flow during the welding process. Gaul et al.³²⁾ demonstrated that excessive plastic deformation is resulted from electrode misalignment. According to the Figs. 7, 8 and the study by Okigami et al.,³¹⁾ Type A cracks almost occur in or near the plastic deformation regions when welded with electrode misalignment. So the presumption that the occurrence of Type A crack is mainly caused by plastic flow, is reasonable. It could be supposed that the degree of plastic flow is directly related to the heating input, so that in the case of high current, the degree of plastic flow will be more significant, leading to more serious Type A cracks. At the same time, Okigami et al.³¹⁾ showed that areas, which are in or near the plastic deformation, trend to expose LME susceptible temperature range easily due to insufficient cooling from electrodes. It could be speculated that welding with high current could aggravate this insufficient cooling due to more heat input, so that this area would suffer a longer time in the range of LME susceptible temperature in comparison to low current. Therefore, in order to decrease Type A cracks effectively, welding with a low current is significant.

Fig. 9. Schematic diagram of the mechanism of Type A LME crack with electrode misalignment. (Online version in color.)

Figure 10 demonstrates the schematic diagram of mechanism of Type B LME crack under the condition with electrode misalignment, Type B cracks phenomenon could be explained by the hypothesis below. As mentioned in the previous research by Lalachan et al.,³³⁾ the contact area between the electrode and sheet surface could be divided to the initial contact and transient contact areas. At the initial stage, the electrode tip region contacts with the sheet surface for both type of the electrodes with small and large tip diameter. At this time, because of the non-perpendicular between the electrode and the sheet surface, the electrode force F_E at this initial contact area is resolved into a force component F_N perpendicular to the sheet surface and a force component F_A along the sheet surface. Because the curvature radiuses of tip region for the both type of the electrodes are the same, so that there are no differences in F_A between the two type electrodes, as shown in Figs. 10 (a)–10(b) and 10(e)–10(f). During the RSW process, the contact area gradually expands, at some critical time, the contact area reaches the edge of the electrode tip, and the mechanical collapse reported by DiGiovanni et al.,²⁶⁾ a phenomenon that the electrode plunge into the sheet and the electrode shoulder contact with the sheet, will occur at this critical time. Based on the research by Murugan et al.,²⁵⁾ for the electrode with small tip diameter, the contact area is smaller than the one with large tip diameter during the RSW process, leading to a higher pressure on sheet surface, so that the mechanical collapse is easier to happen for the electrode with small tip diameter. On the other hand, for the electrode with large tip diameter, the contact of the electrode shoulder and the sheet could be avoided due to the large contacting area, so that the indentation of shoulder region hardly be generated. Comparing the Figs. 7 and 8, it could be seen that shoulder regions are observed when welded with a small tip electrode, but disappear when welded with a large one. This is also an evidence for the analysis above. After mechanical collapse, the electrode shoulder region contact with the sheet, and the force decomposition of F_E will change since the curvature radius of the shoulder is smaller than the tip region. As shown in Figs. 10 (c)–10(d), after this moment, the force along the surface F_A increases owing to the decreasing of the curvature radius, causing a large tensile stress at the contact surface, so that Type B cracks would be induced if the generated tensile stress exceeds the limit strength of the material. On the other hand, since mechanical collapse hardly occurs for the large tip diameter, the welded sheet will be contact with the electrode tip region during the whole RSW process, so that the force decomposition of F_E is almost unchanged and F_A hardly increases, as shown in Figs. 10(g)–10(h), thus avoiding the the occurrence of large tensile test and the Type B cracks. At the same time, for the larger tip diameter electrode, the wider range of the contact area between the electrode and sheet leads to the lower current density. This would result in a lower temperature and alleviate the possibility of Type B cracks. Therefore, according to the above analysis, it could be supposed that Type B cracks occur with small tip electrode due to large tensile stress caused by the contact between the sheet and the electrode shoulder region and high temperature. Using a electrode with large tip diameter could mitigate this kind of cracks effectively by avoiding the contact between the sheet and the electrode shoulder and controlling the temperature of the sheet surface.

Fig. 10. Schematic diagram of the mechanism of LME Type B crack with electrode misalignment. (Online version in color.)

4. Conclusions

In this study, an orthogonal experiment with four factors and four levels was designed to investigate the influences of welding current, welding time, tip diameter of dome radius type electrode, and angle of electrode misalignment on the different types of LME cracks. After welding, numbers and lengths of cracks were measured, the Crack Index of each experiment group was calculated to evaluate the LME crack.

(1) Type A cracks located at the indentation of electrode tip and Type B cracks located at the indentation of electrode shoulder region were observed in this study. The order of influence degree on CI of Type A crack from high to low was welding current > electrode tip diameter > electrode misalignment angle > welding time. The order of influence degree on CI of Type B crack was electrode tip diameter > electrode misalignment angle > welding time > welding current.

(2) In the case of welding with electrode misalignment, the cross-sectional morphologies are different between the electrodes with small and large tip diameters. When electrode tip diameter is small, the electrode indentation includes electrode tip regions, electrode shoulder regions and plastic deformation regions. Type A cracks occur at the electrode tip regions nearby the plastic deformation regions, and Type B cracks occur at the electrode shoulder regions. When electrode tip diameter is on a high level, the electrode shoulder regions disappear, but the plastic deformation regions still exist, and only Type A cracks can be observed at the electrode tip regions nearby the plastic deformation regions.

(3) In the case of welding with electrode misalignment, the occurrence of Type A cracks could be explained by the following presumption. Type A cracks occur due to the large tensile stress caused by the plastic flow during the welding process, and the degree of plastic flow is directly related to the heating input. Welding current is the main parameter affecting heating input, so that modifying welding current is the most effective way to reduce Type A cracks.

(4) In the case of welding with electrode misalignment, it could be inferred that Type B cracks occur due to the large tensile stress caused by the contact between the sheet and the electrode shoulder with a small radius of curvature. Using the electrode with large tip diameter could alleviate Type B cracks effectively by avoiding the contact between the sheet and the electrode shoulder, so that optimizing the electrode tip diameter is significant for preventing Type B cracks.

Statement for Conflict of Interest

The authors declare that they have no conflict of interest.

References

• 1)   J.  Schmitt and  T.  Iung: C R Phys, 19 (2018), 641. https://doi.org/10.1016/j.crhy.2018.11.004
• 2)   A.K.  Perka,  M.  John,  U.B.  Kuruveri and  P.L.  Menezes: Metals, 12 (2022), 1051. https://doi.org/10.3390/met12061051
• 3)   C.T.  Wu,  N.  Ma,  Y.  Guo,  W.  Hu,  K.  Takada,  H.  Okada and  K.  Saito: Adv. Eng. Softw, 123 (2018), 1. https://doi.org/10.1016/j.advengsoft.2018.05.009
• 4)   A.R.  Mardera: Prog. Mater. Sci., 45 (2000), 191. https://doi.org/10.1016/S0079-6425(98)00006-1
• 5)   D.  Fan,  P.  Kaushik and  H.  Pielet: ISIJ Int., 58 (2018), 1538. https://doi.org/10.2355/isijinternational.ISIJINT-2018-132
• 6)   Y.B.  Li,  D.L.  Li,  Z.Q.  Lin,  S.A.  David,  Z.  Feng and  W.  Tang: Sci. Technol. Weld. Joining, 21 (2016), 59. https://doi.org/10.1179/1362171815Y.0000000059
• 7)   J.W.  Kim,  S.P.  Murugan,  J.-H.  Yoo,  R.  Ashiri and  Y.-D.  Park: Sci. Technol. Weld. Joining, 25 (2020), 235. https://doi.org/10.1080/13621718.2019.1680483
• 8)   D.  Bhattacharya: Mater. Sci. Technol., 34 (2018), 1809. https://doi.org/10.1080/02670836.2018.1461595
• 9)   M.H.  Razmpoosh,  C.  DiGiovanni,  Y.N.  Zhou and  E.  Biro: Prog. Mater. Sci., 121 (2021), 100798. https://doi.org/10.1016/j.pmatsci.2021.100798
• 10)   Y.  Yu,  Y.  Zhang,  N.  Cai,  W.  Chen,  P.  Wang and  C.  Fu: ISIJ Int., 63 (2023), 1233. https://doi.org/10.2355/isijinternational.ISIJINT-2023-001
• 11)   Z.  Ling,  T.  Chen,  L.  Kong,  M.  Wang,  H.  Pan and  M.  Lei: Metall. Mater. Trans. A, 50 (2019), 5128. https://doi.org/10.1007/s11661-019-05388-6
• 12)   S.P.  Murugan,  J.B.  Jeon,  C.  Ji and  Y.  Park: Weld World, 64 (2020), 1957. https://doi.org/10.1007/s40194-020-00975-3
• 13)   H.  Kang,  L.  Cho,  C.  Lee and  B.C.  De Cooman: Metall. Mater. Trans. A, 47A (2016), 2885. https://doi.org/10.1007/s11661-016-3475-x
• 14)   C.  DiGiovanni,  E.  Biro and  Y.N.  Zhou: Sci. Technol. Weld. Join, 24 (2018), 218. https://doi.org/10.1080/13621718.2018.1518363
• 15)   C.  DiGiovanni,  X.  Han,  A.  Powell,  E.  Biro and  N.Y.  Zhou: J. Mater. Eng. Perform, 28 (2019), 2045. https://doi.org/10.1007/s11665-019-04005-2
• 16)   Y.  Ma,  Y.  Yu,  P.  Geng,  R.  Ihara,  K.  Maeda,  R.  Suzuki,  T.  Suga and  N.  Ma: Mater. Des., 210 (2021), 110075. https://doi.org/10.1016/j.matdes.2021.110075
• 17)   Y.G.  Kim,  I.J.  Kim,  J.S.  Kim,  Y.I.  Chung and  D.Y.  Choi: Mater. Trans., 55 (2014), 171. https://doi.org/10.2320/matertrans.M2013244
• 18)   R.  Ashiri,  M.  Shamanian,  H.R.  Salimijazi,  M.A.  Haque,  J.-H.  Bae,  C.-W.  Ji,  K.-G.  Chin and  Y.-D.  Park: Scr. Mater., 114 (2016), 41. https://doi.org/10.1016/j.scriptamat.2015.11.027
• 19)   C.  DiGiovanni,  S.  Bag,  C.  Mehling,  K.W.  Choi,  A.  Macwan,  E.  Biro and  N.Y.  Zhou: Weld World, 63 (2019), 1583. https://doi.org/10.1007/s40194-019-00790-5
• 20)   M.  Patel,  M.  Shojaee,  O.  Sherepenko,  A.R.H.  Midawi,  H.  Ghassemi-Armaki and  E.  Biro: J. Mater. Res. Technol, 26 (2023), 22. https://doi.org/10.1016/j.jmrt.2023.07.176
• 21)   A.R.H.  Midawi,  M.  Patel,  M.  Shojaee,  K.  Pearson,  O.  Sherepenko,  H.  Ghassemi-Armaki and  E.  Biro: Metals, 13 (2023), 491. https://doi.org/10.3390/met13030491
• 22)   C.  Böhne,  G.  Meschut,  M.  Biegler and  M.  Rethmeier: Sci. Technol. Weld. Joining, 25 (2020), 617. https://doi.org/10.14279/depositonce-10514
• 23)   D.-Y.  Choi,  A.  Sharma,  S.-H.  Uhm and  J.P.  Jung: Met. Mater. Int., 25 (2018), 219. https://doi.org/10.1007/s12540-018-0180-x
• 24)   C.  Böhne,  G.  Meschut,  M.  Biegler,  J.  Frei and  M.  Rethmeier: Sci. Technol. Weld. Join, 25 (2019), 303. https://doi.org/10.1080/13621718.2019.1693731
• 25)   S.P.  Murugan,  K.  Mahmud,  C.  Ji,  I.  Jo and  Y.  Park: Weld World, 63 (2019), 1613. https://doi.org/10.1007/s40194-019-00797-y
• 26)   C.  DiGiovanni,  L.  He,  U.  Pistek,  F.  Goodwin,  E.  Biro and  N.Y.  Zhou: J. Manuf. Process, 49 (2020), 1. https://doi.org/10.1016/j.jmapro.2019.11.015
• 27)   O.  Siar,  Y.  Benlatreche,  T.  Dupuy,  S.  Dancette and  D.  Fabrègue: Metals, 10 (2020), 1166. https://doi.org/10.3390/met10091166
• 28)   C.  DiGiovanni,  A.  Ghatei Kalashami,  F.  Goodwin,  E.  Biro and  N.Y.  Zhou: J. Mater. Process. Technol, 288 (2021), 116917. https://doi.org/10.1016/j.jmatprotec.2020.116917
• 29)   O.  Siar,  S.  Dancette,  T.  Dupuy and  D.  Fabrègue: J. Mater. Res. Technol, 15 (2021), 6678. https://doi.org/10.1016/j.jmrt.2021.11.100
• 30)   K.  Kwon,  G.  Jang,  W.  Kim,  S.  Uhm,  T.  Lee and  C. S.  Lee: J. Mater. Res. Technol, 13 (2021), 2482. https://doi.org/10.1016/j.jmrt.2021.06.041
• 31)   F.  Okigami,  N.  Avedissian,  H.  Ghassemi-Armaki,  X.  Gao and  W.  Zhang: J. Manuf. Process, 110 (2024), 173. https://doi.org/10.1016/j.jmapro.2023.12.065
• 32)   H.  Gaul,  S.  Brauser,  G.  Weber and  M.  Rethmeier: Weld World, 55 (2011), 99. https://doi.org/10.1007/BF03321547
• 33)   A.  Lalachan,  S. P.  Murugan,  W.-S.  Jin and  Y.-D.  Park: J. Mater. Process. Tech., 318 (2023), 118009. https://doi.org/10.1016/j.jmatprotec.2023.118009