Welding Defects Occurrence and Their Effects on Weld Quality in Resistance Spot Welding of AHSS Steel

Abstract

An investigation was made on the weldability of DP600 steel. Influences of electrode force, welding current and welding time on welding defects occurrence were discussed. Metallographic analysis and tensile-shear tests were implemented for weld quality detections. Weld quality dependency on welding defects was studied. It was found that welding defects were easy to occur during spot welding process of DP600 steel, which could be explained by its rich chemistry. Lower electrode force, larger welding current and longer welding time all contributed to the easy occurrence of welding defects. HAZ softening caused by martensite tempering was observed under strong welding parameters. For interfacial failure, welding defects effects on weld quality should not be underestimated. As to pullout failure, the effect of expulsion on weld quality was not that significant as suggested by other researches. In the end, a method was introduced to predict peak load and maximum tensile-shear displacement for pullout failure.

1. Introduction

For resistance spot welding, the electrode force is first applied on two or more overlapped workpieces, then joule heat is generated by the passage of a large welding current in a very short period of time, base metal is finally melt and joints are formed after the cooling stage. Although many new welding procedures are introduced, such as laser beam welding and adhesive bonding, resistance spot welding remains to be the most widely used sheet metal welding process in the automotive industry owing to its low cost, high efficiency and easy to operate.

The increasingly demands for vehicle safety and energy conservation are drawing more and more attentions on non-traditional materials like advanced high strength steels (AHSSs).¹⁾ Due to the good combination of strength and ductility, dual phase steels including DP600 are more preferred by vehicle manufacturing companies. However, the poor weldability of dual phase steels has been an important issue to restrict its wide application. Welding parameters like electrode force, welding current and welding time all can have significant effects on weld quality of dual phase steels. High alloy elements level in DP600 makes expulsion and interfacial failure more frequent for spot welds. The traditional failure criterion of $D_{cr}=4\sqrt{t}$ does not work for DP600 resistance spot welding any more.²⁾ However, limited work has been done to investigate the effect of welding defects on weld quality of spot welded DP600 steels.

The presence of welding defects in weld nugget has been reported in some investigations.^3,4,5) According to these studies, shrinkage void and solidification crack were two of the main reasons for interfacial failure of spot welds under tensile-shear and fatigue tests. Voids formed by expulsion generally decreased the load bearing capacity of spot welds and interfacial failure occurred more often. Khan et al.⁶⁾ found that interfacial failure would occur for DP600 resistance spot welds although the nugget size exceeded the critical value, which was attributed to shrinkage voids and solidification cracks formation. Sun et al.^7,8) studied the microstructure of resistance spot welded magnesium alloy and suggested that a portion of welding cracks belonged to solidification cracking. An incrementally coupled finite element procedure was then developed to simulate the spot welding process of aluminum alloys. Tensile residual stress at the weld nugget was believed to promote through-thickness cracking. Marya et al.⁹⁾ showed that solidification cracks can adversely affect the overall bonded area in DP600 spot welds. Pouranvari et al.^2,10) suggested that the high tendency to shrinkage void formation during AHSS resistance spot welding was one of the main reasons for its high susceptibility to interfacial failure. An analytical model was then proposed to predict the critical nugget size:   

$$D_{cr}=\frac{4t}{Pf}\left(\frac{H_{PFL}}{H_{FZ}}\right)$$

where P referred to the porosity factor induced by shrinkage voids.

Considerable researches have been made on how to reduce welding defects occurrence. Marya et al.³⁾ investigated the weldability of dual phase steels with different thickness. They found that shrinkage voids were less pronounced in thinner DP steels. Joaquin et al.⁴⁾ and Ma et al.⁵⁾ showed that longer holding time was helpful to reduce shrinkage voids. Lang et al.¹¹⁾ studied effects of welding parameters on crack susceptibility. They tried to understand the crack formation mechanism and make improvements on weldability of magnesium alloys. Milititsky et al.¹²⁾ conducted tensile-shear tests on DP600 spot welds and found that short holding time may promote shrinkage voids and solidification cracks formation. Zhang et al.¹³⁾ presented an weld expulsion estimation method in resistance spot welding of dual phase steel through changing the electrode force.

The objective of this study is to investigate the weldability of DP600 steel. Welding defects, such as expulsion, shrinkage void and solidification crack, were detected by experiments. Influences of electrode force, welding current as well as welding time on welding defects occurrence were discussed. In the end, a preliminary analysis was made on correlation between welding defects and mechanical performance of spot welds.

2. Experimental Procedure

In this study, a 1.7 mm thick DP600 steel was used for resistance spot welding. Its chemical composition was measured and given in Table 1. Detailed mechanical property of this steel sheet is shown in Table 2. Before welding, the steel sheets were cut into dimensions of 120 × 40 mm and all the specimens surfaces were cleaned carefully by acetone. The schematic description of the overlapped steel is shown in Fig. 1.

Table 1. The chemical composition of DP600 steel, wt-%.

C	Si	Mn	P	S	Al	N
0.0790	1.0000	1.5200	0.0150	0.0049	0.0230	0.0037

Table 2. Mechanical property of DP600 steel.

Steel	Yield Strength (MPa)	Ultimate Tensile Strength (MPa)	Elongation (%)
DP600	351	633	28

Fig. 1.

Schematic illustration of the overlapped steels for resistance spot welding.

Spot welding was conducted using a 120 kVA, PLC controlled 50 Hz AC spot welding machine (1 cycle = 0.02 s). A 45⁰ truncated cone electrode made of copper alloy with a face diameter of 6 mm was used. Welding was repeated for six times at each welding condition, including five specimens prepared for static tensile-shear test and one specimen prepared for metallographic investigation. During the welding process, both the squeeze time and holding time were kept constant at 10 cycles (0.2 s). To study the effect of electrode force, welding current and welding time on joint performance, specimens were divided into three groups. In each group, only one parameter was varied and all the other parameters were kept constant. Electrode force was changed from 2.0 kN to 4.0 kN. Welding current, which referred to the peak current value, increased from 6 kA to 12 kA. Welding time was varied from 8 to 17 cycles at an interval of 3 cycles. A detailed description of the welding schedules utilized in this study is shown in Table 3.

Table 3. Welding schedules in this study.

Squeeze Time (Cycle)	10
Holding Time (Cycle)	10
Electrode Force (kN)	2.0–2.5–3.5–4.0
Welding Current (kA)	6–8–10–12
Welding Time (Cycle)	8–11–14–17

An Instron universal testing machine was used to conduct the tensile-shear tests at a constant crosshead speed of 10 mm/min. Tabs were glued to the specimen ends to reduce joints deformation. Peak load was measured at the top position of the load-displacement curve. Failure energy absorption was considered as the area under the load-displacement curve up to failure:¹⁴⁾   

$$Q=\sum _{i=1}^{N}F(i)[x(i)-x(i-1)]$$

Where F is the force, x is the displacement, the product of F and x are summed for each data set (i, N) up to the peak load. By examining the failed specimen, failure modes were identified.

For metallographic analysis, specimens were cross-sectioned, ground and polished with carbide papers and etched using a 4% nital reagent until the martensite grain boundaries can be observed under the microscope. Figure 2 shows a prepared metallographic specimen. The microstructure of weld joints was detected by an Zeiss Axiovert 200 MAT Inverted microscope. Nugget size was measured by the Carl Zeiss AxioVision Software on cross-sectioned spot welds. Microhardness testing was carried out using a HX-100 test machine (100 g load) at an interval of 0.5 mm.

Fig. 2.

A prepared metallographic specimen.

3. Results and Discussion

3.1. Microstructure and Microhardness

A typical microstructure distribution of resistance spot welded DP600 steel is shown in Fig. 3. Microstructure in the fusion zone shows directional and columnar grains growing from the fusion boundary towards the weld centerline, as depicted in Fig. 3(a). In Fig. 3(b), the fusion zone is predominantly composed of large lath martensite, which could be explained by the high cooling rate. Grain size is less coarser in the HAZ (Fig. 3(c) and left area of Fig.3(d)) when comparing to that of the fusion zone, which is due to an incompletion of austenitizing.³⁾ The base metal, as seen in Fig. 3(d), consists mainly of evenly distributed martensite and ferrite.

Fig. 3.

Microstructure variation across a DP600 spot weld.

Figure 4 shows the microhardness profiles across the weld nugget under two different welding conditions. As can be seen, the fusion zone hardness is roughly around 360 HV, owing to the lath martensite formation. Fusion zone hardness is approximately 1.5 times higher than that in the base metal. When welding time is 8 cycles, maximum hardness is found in the heat affected zone (HAZ) closing to the weld nugget. A softening region in the HAZ can be observed under a large welding time (11 cycles), which is attributed to martensite tempering. What’s more, mechanical performance of spot welds is considered to be improved by HAZ softening.¹⁵⁾

Fig. 4.

Two typical microhardness profiles across DP600 spot welds.

3.2. Welding Defects

3.2.1. Expulsion

Welding expulsion at the faying surface for a DP600 spot weld is shown in Figs. 5 and 6. Expulsion, which refers to the ejection of molten metal, is a common phenomenon during resistance spot welding process. It usually occurs at either the electrode/workpiece interface or the faying surface owing to excessive heat input. Due to the presence of expulsion, excessive electrode indentation, shrinkage voids and solidification cracks will be introduced into the weld nugget. Expulsion at the faying surface may severely affect the weld quality.¹⁵⁾

Fig. 5.

Expulsion occurrence at the faying surface.

Fig. 6.

Expulsion occurrence at the faying surface (electrode force = 3.5 kN, welding current =12 kA, welding time = 14 Cycles).

Prediction and control of expulsion are important issues in the manufacturing environment. Long welding time and large welding current are common choices for dual phase steel spot welding with consideration of its poor weldability, which make expulsion even more popular. In general, plastic annulus formed by larger plastic strain around the weld nugget is helpful to prevent expulsion.¹⁶⁾ What’s more, appropriate welding parameters, clean workpiece surface and elimination of electrode wear are all necessary to reduce expulsion. The effect of expulsion on tensile-shear failure load and failure energy absorption of spot welds are discussed in the following sections.

3.2.2. Shrinkage Void

Shrinkage voids are found under some welding conditions, as shown in Figs. 7 and 8. Low magnification microscopic images are depicted in Figs. 7(a) and 8(a). High magnification microscopic images are shown in Figs. 7(b) and 8(b). Heat input immediately reduces to zero on welding current is cut off. Temperature at the weld joint decreases rapidly under the cooling effect of water cooled electrodes. Liquid metal starts to solidify when temperature drops to the liquidus temperature. Finally, cooling rate mismatch at different locations results in the formation of shrinkage voids.

Fig. 7.

Shrinkage void at weld nugget (electrode force = 3.5 kN, welding current =12 kA, welding time = 14 Cycles).

Fig. 8.

Shrinkage void at weld nugget (electrode force = 3.5 kN, welding current =10 kA, welding time = 14 Cycles).

Effective area of weld nugget is reduced and local shear stress is increased under tensile-shear tests when shrinkage voids form at the faying surface.¹⁷⁾ Interfacial failure tendency, which means an unsatisfactory load bearing capacity, is believed to increase by shrinkage voids and the critical nugget size gets bigger. However, when pullout failure occurs, the mechanical performance will not be affected by shrinkage voids. In this case, the weld nugget and heat effected zone are nearly stress free, which was confirmed by finite element method and experimental observations.^18,19)

Shrinkage voids formation and solidification cracks are strongly dependent on carbon, alloy elements and the amount of S and P. A higher carbon equivalent results in more shrinkage voids in spot welds.⁴⁾ Shrinkage voids tend to occur during resistance spot welding of dual phase steel, owing to the inherent rich chemistry as compared to low carbon steels.³⁾ In addition, if an internal expulsion is generated during the welding process, shrinkage voids are easily formed at the weld nugget center for the reason of liquid metal loss. To reduce shrinkage voids occurrence, greater electrode force can be adopted to while controlling other welding parameters to restrict the heat input rate. As can be seen, the electrode force is relative high in Figs. 7 and 8. However, shrinkage voids are generated due to the high welding current and long welding time, leading to too much heat input and expulsion is also observed.

3.2.3. Solidification Crack

Figures 9 and 10 clearly show the solidification cracks. A rich chemistry in dual phase steels is also contributed to solidification crack occurrence. Internal solidification cracks are easy to form and often companied with shrinkage voids. Increased local shear stress and decreased weld quality are results of solidification cracks. Most of the cracks are observed to be perpendicular to the faying surface and will be explained below.

Fig. 9.

Solidification crack at the transition region from HAZ to base metal (electrode force = 3.5 kN, welding current =10 kA, welding time = 17 Cycles).

Fig. 10.

A magnified views of solidification crack at weld nugget (electrode force = 3.5 kN, welding current = 10 kA, welding time = 14 Cycles).

During the spot welding process, thermal expansion and plastic deformation occur sequentially in weld joints as a result of heat input by welding current passage. Inhomogeneous deformation results in nonuniform stress and strain fields. Under the combined effects of electrode force and heated base metal, stress in the weld nugget is mostly compressive. However, temperature decreases rapidly during the cooling stage and stress in the weld nugget translates gradually to tensile state. Contraction rate is the largest in the direction parallel to faying surface. What’s more, the martensite grain boundary direction is mostly perpendicular to the faying surface. That’s why solidification cracks is prone to occur in through-thickness direction and most of them are grain boundary cracks.

3.3. Welding Defects Occurrence

Expulsion and shrinkage void (or solidification crack) occurrence under various welding conditions are shown in Fig. 11. Resistance spot welding under the same welding parameters was repeated for five times. Expulsion was checked after the completion of tensile-shear tests. Shrinkage void occurrence was indirectly determined by the significant drop in maximum tensile-shear displacement. If no welding defects were found, the frequency is replaced by the value of 2% for a better visualization. As it is shown in Figs. 11(a)–11(c) and indicated in previous sections, expulsion and shrinkage void always occur simultaneously. A lower electrode force, larger welding current and longer welding time all make contribution to the easy occurrence of expulsion and shrinkage void.

Fig. 11.

Expulsion and shrinkage void occurrence under various welding conditions.

3.4. Effect of Expulsion on Weld Quality

Tensile-shear tests were conducted for each group with the aforementioned five spot welded specimens under the same welding condition. Both interfacial and pullout failure mode were observed during the tensile-shear tests. Partial interfacial failure was categorised as interfacial failure, depending on its failure performance. After an elimination of the welding parameters under which only interfacial failure during tensile-shear testing was taken place, a detailed analysis of the influence of expulsion on weld quality is depicted in Fig. 12. Sample number is ordered by failure mode (from interfacial failure to pullout failure) in Fig. 12. For all spot welded specimens listed in Fig. 12, only the specimen number 2–5 in Fig. 12(a) and specimen number 3–4 in Fig. 12(b) are expulsion free, all the rest experienced a slight or excessive expulsion. In addition, among the four different welding conditions, HAZ softening was not observed only when welding time is 8 cycles. Peak load, maximum displacement were recorded and the ratio between them was also plotted.

Fig. 12.

Effect of expulsion on weld quality.

Generally, plastic collapse is considered to be the failure mechanism for pullout failure. In this case, failure initiates around the weld nugget from the location where hardness is minimum.²⁰⁾ Moreover, it seems that a proportional relationship exists between peak load (F_max) and maximum displacement (L_max) under pullout failure, as can be seen in Fig. 12. This conclusion is further confirmed by an almost horizontal green line on behalf of the F_max/L_max in Fig. 12. That is to say:   

$$F_{max}\propto L_{max}$$(1)

In addition, different failure modes indicate different weld qualities at the same welding condition. Peak load and maximum displacement at interfacial failure is typically lower than those of pullout failure. However, despite the expulsion occurrence, experimental data varies slightly under the pullout failure mode, as shown in Figs. 12(a)–12(d). Thus, it can be inferred that the influence of expulsion occurrence on weld quality is not that significant for pullout failure, which is probably different from some researches.^6,21)

In addition, L_max reflects the tensile strength of the initial failure location and nugget size d is also an key factor affecting F_max. The normalized L_max/d can then be used to represent the ductility of initial failure location. Moreover, the failure location hardness H_fl is inversely proportion to its ductility. Thus,   

$$L_{max}/d=\omega H_{fl}^{-1}$$(2)

can be established. Pullout failure usually initiates from the minimum hardness position, that is to say H_fl = H_min.¹⁰⁾ ω can be calculated by the average experimental data obtained, as shown in Table 4. For occasions without HAZ softening, the average value of ω is 173.8 HV (No. A). When HAZ softening occurs, ω is approximately equal to 144.3 HV (No. B–D). Once ω is determined, L_max and F_max under pullout failure could then be predicted by Eqs. (1) and (2). For further study, a critical nugget size criterion may be constructed in combination with additional interfacial failure investigation.

Table 4. Welding parameters and average experimental data for calculating ω (Electrode force = 3.5 kN).

No. (–)	Welding current (kA)	Welding time (Cycle)	D (mm)	Lmax (mm)	Hfl (HV)	ω (HV)
A	10	8	6.42	4.65	240	173.8
B	10	14	7.08	5.57	185.1	145.6
C	12	14	7.37	6.24	170.5	144.4
D	10	17	6.71	6.42	149.2	142.8

4. Conclusions

Electrode force, welding current and welding time are main factors affecting the weld qualities of DP600 spot welds. Microstructure and mechanical performance of spot welds with welding defects were investigated. Although the test results here are limited, following conclusions can still be derived:

• Welding defects are easy to occur during resistance spot welding of DP600 steel sheets due to its rich chemistry.

• Lower electrode force, larger welding current and longer welding time all contributed to the easier occurrence of welding defects.

• HAZ softening caused by martensite tempering is observed under strong welding parameters.

• For interfacial failure, welding defects effects on weld quality should not be underestimated.

• As to pull out failure, the influence of expulsion occurrence on tensile-shear mechanical performance is not that significant.

• A parameter to is introduced to predict peak load and maximum tensile-shear displacement for pullout failure occasion.

Acknowledgements

We would like to acknowledge the financial support from the National Natural Science Foundation of China (11072083). The sponsorship of the Graduates Innovation Fund of Huazhong University of Science & Technology (HF-11-14-2013) and the Fundamental Research Funds for the Central Universities, HUST (CXY13Q047) are also appreciated.

References

• 1)  ULSAB-AVC Consortium: Technical Transfer Dispatch # 6 (Body Structure Materials), American Iron and Steel Institute, Southfield, MI, (2001).
• 2)   M.  Pouranvari,  S. P. H.  Marashi and  D. S.  Safanama: Mater. Sci. Eng. A, 528 (2011), 8344.
• 3)   M.  Marya and  X. Q.  Gayden: Weld. J., 84 (2005), 172.
• 4)   A.  Joaquin,  A. N. A.  Elliott and  C.  Jiang: Weld. J., 86 (2007), 24.
• 5)   C.  Ma,  D. L.  Chen,  S. D.  Bhole,  G.  Boudreau,  A.  Lee and  E.  Biro: Mater. Sci. Eng. A, 485 (2008), 334.
• 6)   M. I.  Khan,  M. L.  Kuntz,  P.  Su,  A.  Gerlich,  T.  North and  Y.  Zhou: Sci. Technol. Weld. Join., 12 (2007), 175.
• 7)   D. Q.  Sun,  B.  Lang,  D. X.  Sun and  J. B.  Li: Mater. Sci. Eng. A, 460 (2007), 494.
• 8)   X.  Sun and  P.  Dong: Weld. J., 79 (2000), 215.
• 9)   M.  Marya,  K.  Wang,  L. G.  Hector and  X. H.  Gayden: J. Manuf. Sci. Eng., 128 (2006), 287.
• 10)   M.  Pouranvari and  S. P. H.  Marashi: Mater. Sci. Eng. A, 528 (2011), 8337.
• 11)   B.  Lang,  D. Q.  Sun,  Z. Z.  Xuan and  X. F.  Qin: ISIJ Int., 48 (2008), 77.
• 12)   M.  Milititsky,  E.  Pakalnins,  C.  Jiang and  A. K.  Thompson: On Characteristics of DP600 Resistance Spot Welds, SAE Report 2003-01-0520, SEA International, Warrendale, PA, USA, (2003).
• 13)   Y.  Zhang,  J.  Shen and  X.  Lai: ISIJ Int., 52 (2012), No. 3, 493.
• 14)   M. I.  Khan,  M. L.  Kuntz and  Y.  Zhou: Sci. Technol. Weld. Join., 13 (2008), 294.
• 15)   H.  Zhang and  J.  Senkara: Resistance Welding: Fundamentals and Applications, 2nd ed., Taylor & Francis, Boca Raton, FL, (2011).
• 16)   Z.  Hou,  I.-S.  Kim,  Y.  Wang,  C.  Li and  C.  Chen: J. Mater. Process. Technol., 185 (2007), 160.
• 17)   X.  Sun,  E. V.  Stephens and  M. A.  Khaleel: Weld. J., 86 (2007), 18.
• 18)   X.  Deng,  W.  Chen and  G.  Shi: Finite Elem. Anal. Des., 35 (2000), 17.
• 19)   M.  Pouranvari and  S. P. H.  Marashi: Mater. Des., 31 (2010), 3647.
• 20)   Y. J.  Chao: Sci. Technol. Weld. Join., 8 (2003), 133.
• 21)   M.  Pouranvari,  A.  Abedi,  P.  Marashi and  M.  Goodarzi: Sci. Technol. Weld. Join., 13 (2008), 39.