2018 Volume 59 Issue 9 Pages 1433-1439
The effects of elapsed time following gas metal arc welding (GMAW) on the static strength of lap fillet-welded joints were investigated using a static tensile test. It was observed that the static joint strength did not exhibit a time dependency for cases where the fracture was located in each base metal or the softened heat affected zone (HAZ). In addition, the static joint strength was not observed to have a time dependency for cases where a wire with low-hardness weld metal was used and where the fracture was located in the weld metal. However, the static joint strength was low immediately after welding, and increased over time when a wire with high-hardness weld metal was used. It was found that diffusible hydrogen that entered the weld metal during arc welding and was emitted over time was the cause of the time dependency of joint strength. High-hardness weld metal was sensitive to diffusible hydrogen; therefore, time dependency was observed only when wires with high-hardness weld metal were used and the fracture positions were located in the weld metal during tensile tests. In addition, a correlation between the storage temperature following welding and the static strength or diffusible hydrogen content of welded parts was found: the higher the storage temperature, the earlier the joint strength increases, and the earlier the diffusible hydrogen decreases.
Fig. 9 Effects of storage temperature on joint strength (a) and nominal diffusible hydrogen content of welds (b).
In the automobile industry, it is important to make car bodies lighter and stronger in order to reduce CO2 emissions and enhance passenger safety. For these reasons, high-strength steel sheets are being increasingly applied to car bodies.1) Steel sheets of 440 MPa and 590 MPa grade tensile strength were mainly applied to the frames of car bodies in the 1990s and the 2000s, respectively. In the 2010s, advanced high-strength steel (AHSS) sheets of 1180 MPa grade were used for B pillars.2) Resistance spot welding is principally employed for assembling car bodies; however, gas metal arc welding (GMAW) is employed for parts of the chassis, and in areas of the bodies consisting of closed sections where it is difficult to perform spot welding. Most arc-welded joints applied to car bodies are lap fillet welded joints. The strength of these joints is frequently evaluated by static tensile tests. In applying high-strength steel sheets to automobiles, it is necessary to know not only the strength of the spot-welded joints but also that of the arc-welded joints precisely.
In car bodies, high-strength welding wires are not necessarily used, even when high-strength steel sheets have been used. In many cases, this is because weld metal frequently have a high strength of 780 MPa grade or higher, even when welding wires with a nominal tensile strength of 490 MPa grade are used, because the arc-welded parts of steel sheets used for automobiles have a fast cooling rate at the time of welding. In the case of car bodies, high-strength wires are often unnecessary because the many parts formed by welding are between a high-strength steel sheet and a lower-strength one, rather than welding between high-strength steel sheets. However, it is expected that high-strength welding wires corresponding to the high strength of steel sheets will be needed, depending on the part, because the significant expansion of the application of recent high-strength steel sheets over 1180 MPa grade may mean that lower-strength welding wires are unable to meet the required performance of automotive parts.
In evaluating the strength of arc-welded joints for steel sheets, the time from when arc welding is performed until the joint strength is measured is not specified. Therefore, the time until the joint strength is evaluated seems to depend on the convenience of the testing process. Moreover, an instance in which bending cracks decrease the time after welding elapses has been reported3) in the bending tests of arc-welded parts by shielded metal arc welding. Another instance in which the time dependency of the peel strength was identified in the resistance spot welded joints of 780 MPa grade transformation induced plasticity (TRIP) steel sheets was reported.4) However, there have been no reports on how the elapsed time following welding affects the static strength of arc-welded joints in high-strength steel sheets. Nevertheless, these instances suggest that the elapsed time following welding can affect joint strength. This can also occur in the case of the static strength in lap fillet welded joints by arc welding. Accordingly, in this study, the effects of the elapsed time following arc welding on the static strength of lap fillet welded joints are investigated using high-strength steel sheets with a tensile strength between 590 and 1180 MPa grade. Also examined are two types of welding wires (490 MPa grade and 950 MPa grade) with respect to the nominal tensile strength grade of the weld metal.
Table 1 lists specimens of three types of high-strength steel sheets for automobiles: 590 MPa grade, 980 MPa grade, and 1180 MPa grade in tensile strength, with a thickness of 1.6 mm. As welding wires, low-strength solid Wire A of 490 MPa grade and high-strength solid Wire B of 950 MPa grade were used. The chemical composition of these wires and mechanical properties evaluated based on JIS Z 3312 are listed in Table 2. Lap fillet welding under pulsed MAG welding was performed under the conditions listed in Table 3.
Immediately after welding, the test pieces for tensile tests shown in Fig. 1 were sampled and retained at 20°C. Tensile tests were conducted 0.5 to 72 h after welding. The fracture strength of the joints was determined by dividing the fracture load at a tension speed of 0.167 mm/s by the cross-sectional area of the steel sheet. In addition, diffusible hydrogen was measured from a welded specimen retained at the same temperature and for the same time as a piece for the tensile test. It is preferable to cut a test piece for hydrogen measurement from the weld metal itself; however, this was difficult, so the test piece was cut out from a region of “Zone A,” including the weld metal and heat-affected zone. This is shown in Fig. 2. The hydrogen content was measured using Thermal Desorption Spectroscopy (TDS) in terms of the test pieces from Zone A. Diffusible hydrogen was defined as the hydrogen emitted until the temperature reached 400°C at a temperature-increasing speed of 0.167°C/s according to TDS. Total amount of diffusible hydrogen was divided by the total mass of the test piece in the region of Zone A so that the nominal concentration of the diffusible hydrogen was determined.
Specimen for evaluation of static joint strength.
Specimen for evaluation of diffusible hydrogen content.
Arc welding was performed using three types of high-strength steel sheets and two types of welding wires. The results of the investigated relationship between the elapsed time after welding and the static strength of the joints, using the test pieces retained at 20°C after welding, are shown in Fig. 3(a) and (b). Mark 1 in Fig. 3(a) is of a joint by a steel sheet with a tensile strength of 590 MPa that was welded with low-strength Wire A. The static strength of this joint had a constant value irrespective of the elapsed time after welding. In addition, in all of the tensile tests, the fracture was located in the base metal irrespective of the elapsed time following arc welding. Mark 2 is of a joint by a steel sheet with a tensile strength of 980 MPa that was welded with low-strength Wire A. The static strength of this joint had a constant value irrespective of the elapsed time. Furthermore, in all of the tensile tests, the fracture was located in the weld metal irrespective of the elapsed time following arc welding. Mark 3 is of a joint by a steel sheet of 980 MPa that was welded with high-strength Wire B. The static strength of this joint had a constant value irrespective of the elapsed time following arc welding. In addition, in all of the tensile tests, the fracture was located in the softened part of the HAZ of the upper plate around 2.5 mm away from the weld metal, irrespective of the elapsed time. In the case of Mark 3, the use of high-strength Wire B was assumed to have increased the strength of the weld metal, so no fracture occurred in the weld metal (like that which occurred in the case of Mark 2), but a fracture did occur in the softened HAZ. However, in both cases, no time dependency was identified in the static strength.
Relationship between elapsed time following arc welding and static joint strength.
Mark 4 in Fig. 3(b) is a joint of the steel sheet of 1180 MPa grade that was welded with low-strength Wire A. The static strength of this joint had a constant value irrespective of the elapsed time following arc welding. In addition, in all tensile tests, a fracture occurred in the weld metal irrespective of the elapsed time after welding, as in the case of Mark 2, where low-strength Wire A was used. On the contrary, a joint of Mark 5 with a steel sheet of 1180 MPa grade welded with high-strength Wire B indicated behavior different from that of above-mentioned joints. In other words, while the joint strength immediately after welding is low in the joint of Mark 5, it is likely to increase as time elapses.
In the joint of Mark 5, 48 h or longer following welding was necessary to saturate the static strength. Moreover, the fracture in the tensile test was located in the weld metal, irrespective of the elapsed time after welding. The time dependency of the static strength is identified in the cases (1180 steel sheet + Wire B) where a fracture is located in the weld metal in tensile tests. By contrast, no time dependency is identified in the cases (980 steel sheet + Wire B) where a fracture is located in a softened HAZ. It is considered that this phenomenon is affected by the fracture positions of the joints.
The results above show that no time dependency of static strength was identified in the case where a fracture in static tensile tests was located in the base metal or in the softened HAZ, or in the case where low-strength welding wires were used even when a fracture was located in the weld metal. On the other hand, a time dependency was identified in the static strength of arc-welded joints in a case where a fracture was located in the weld metal in a static tensile test even when a high-strength welding wire was used.
3.2 Effects of types of welding wires on hardness distribution in arc-welded joints of 1180 MPa grade steel sheetIn order to clarify the cause of the identified time dependency in the static strength of high-strength Wire B used for welding 1180 MPa grade steel sheets, the hardness distribution of the joints shown in Fig. 3(b) were examined between the cases of Mark 4 (low-strength Wire A) and Mark 5 (high-strength Wire B). The measurement results of the hardness distribution in the arc-welded parts are shown in Fig. 4. It was confirmed that the weld metal had a mean Vickers hardness of 315 Hv with Wire A and 425 Hv with Wire B. Thus, high-strength Wire B was harder. Note that all wires have harder weld metal than that of the nominal strength grade of wires in the catalogue. It is considered that one-pass welding of low-heat input in Wire A for the use of 490 MPa grade allowed for the cooling rate to rise so that the weld metal was hardened up to 315 Hv (around 1000 MPa with respect to the tensile strength). Wire B for the use of 950 MPa grade was hardened up to 425 Hv (around 1380 MPa with respect to the tensile strength) for the same reason.
Hardness profile of arc welds of 1180 MPa steel.
Figure 5 shows the observed results of the macro section and fracture in a fractured part after a static tensile test of the joints with Mark 5 using high-strength Wire B. As previously mentioned, the fractures in all tensile tests were located in the weld metal for both Wires A and B, but a small change in the fractured pass was identified depending on the elapsed time in the case of Wire B alone. In other words, a phenomenon of uneven cracking in the weld metal was identified 0.5 h after welding, and cracking in the weld metal in the vicinity of the weld interface was identified 72 h after welding. In addition, according to the observed results of the fracture surfaces, a fracture surface was divided into two regions 0.5 h after welding: a quasi-cleavage fracture surface on the side of the overlapped region where the fracture initiated, and a dimple fracture surface on the outside region were identified. However, 72 h afterward, the dimple fracture surface covered the entire fracture surface. This means that for Wire B, while the fracture position was in the weld metal, the fracture surface changed over time after welding.
Fracture mode and fracture surface of arc-welded joints (5: 1180 MPa steel - Wire B).
It was assumed that diffusible hydrogen that enters the weld metal affects the changes in joint strength over time following arc welding because the fracture surface in a static tensile test exhibited a quasi-cleavage fracture surface 0.5 h after welding. Accordingly, the diffusible hydrogen content in Zone A in Fig. 2 was measured at various times after welding. An example of hydrogen deposition curves with Wire B, measured by TDS, is shown in Fig. 6. It turns out that the rich hydrogen content immediately after welding decreases as time elapses. In addition, the change over time in the diffusible hydrogen content of Wires A and B is shown in Fig. 7. The diffusible hydrogen from both Wires A and B immediately after welding was rich, and its amount decreased as time elapsed. However, there was little difference between Wires A and B.
Examples of diffusible hydrogen content measured by TDS.
Relationship between elapsed time following arc welding and nominal diffusible hydrogen content of 1180 MPa steel welds.
From the observed results of the fracture surfaces of the welds and the measured results of the diffusible hydrogen content, as stated above, the fact that the joint strength for Wire B is low immediately after welding and rises as time elapses is assumed to be caused by the fact that the diffusible hydrogen entering the weld metal during arc welding reduces the joint strength, and the diffusible hydrogen is emitted as time elapses. Although the diffusible hydrogen content of both wires is not much different, Wire A has no time dependence on the static strength.
3.4 Effects of hardness in weld metal on hydrogen sensitivity of weld metalIn Section 3.3, the fact that a difference in the time dependency of the joint strength was identified, even if there was no significant difference in the diffusible hydrogen content between low-strength Wire A and high-strength Wire B, was surmised to be caused by the fact that high-strength Wire B resulted in a hard weld metal so that the joint strength was affected despite the low hydrogen concentration. Therefore, it was expected that the joint strength decreased if the diffusible hydrogen content increased, even in low-strength Wire A. For the purpose of verification, the static strength and hydrogen content at 0.5 h after welding were evaluated using arc-welded test pieces with small amounts of hydrogen gas added to conventional shielding gas (Ar + 20%CO2). The plotted results for the test pieces, including those welded under conventional shielding gas, as shown in Fig. 3(b) and Fig. 7, are shown in Fig. 8.
Relationship between nominal diffusible hydrogen content of welds and static joint strength with different weld metal hardness.
It emerged that the joint strength is a constant value for both low-strength Wire A and high-strength Wire B when the diffusible hydrogen content is low. However, this declines as the diffusible hydrogen content increases. In addition, the limit of the diffusible hydrogen content that causes the static strength to decrease is low when the weld metal is hard. The limit of diffusible hydrogen content ranged between 1.1% and 1.8% for the hardness of weld metal of 315 Hv according to Wire A, and ranged between 0.1% and 0.3% for 425 Hv according to Wire B. As shown in Fig. 6, the fact that the time dependence of the joint strength was not observed in low-strength Wire A but was observed in high-strength Wire B was caused by the fact that Wire B had a lower limit of diffusible hydrogen content than Wire A.
3.5 Effects of storage temperature on time dependency in static strength of arc-welded jointsThe above findings reveal that the time dependency of the joint strength can be identified when a fracture is located in the weld metal in a static tensile test, even when a high-strength welding wire that hardens weld metal is used. However, it takes a long time to appropriately evaluate the joint strength. The diffusion of hydrogen in steel has a significant dependence on temperature.5) In the welding of steel plates, a preheating or post-heating is employed6–8) to prevent cold cracking to diffuse the diffusible hydrogen that entered the weld metal during welding into the air. Accordingly, we investigated how the storage temperature of test pieces after welding affected the joint strength.
The relationship between the elapsed time and joint strength when the test pieces were stored under various temperatures between 5°C and 60°C following arc welding is shown in Fig. 9(a). The trend in which the joint strength rises with time after welding was identified. Around one week (168 h) was required until the static strength became roughly constant in the case where the test pieces were stored after welding at a temperature of 5°C. The temperature became constant after approximately two days (48 h) in the case of a storage temperature of 20°C, and after less than one day at 60°C. In other words, it emerged that the higher the storage temperature of the test pieces, the earlier the strength increase. Note that all fracture modes are in the weld metal. The relationship between the elapsed time and the diffusible hydrogen content of each welded test piece is shown in Fig. 9(b). A welded part immediately after welding had a significant content of diffusible hydrogen, and the content decreased over time. However, it turns out that the diffusible hydrogen is unlikely to decrease if the storage temperature of the test pieces after welding is low. The higher the storage temperature, the earlier the diffusible hydrogen in a welded part decreased.
Effects of storage temperature on joint strength (a) and nominal diffusible hydrogen content of welds (b).
It emerged from these results that there is a correlation between an increase in the joint strength and a decrease in the diffusible hydrogen. According to this finding, it is necessary to use the time from welding until the strength evaluation to diffuse the hydrogen that entered during welding in order to conduct an appropriate evaluation of the static joint strength. However, it is preferable to store the test pieces after welding at an elevated temperature in order to evaluate the joint strength earlier, as it takes time to allow the hydrogen to be emitted when the atmospheric temperature is low, such as during the winter.
Below are the effects of the elapsed time following arc welding on the static strength of lap fillet welded joints. High-strength steel sheets with a tensile strength between 590 and 1180 MPa and two types of welding wires (490 MPa grade and 950 MPa grade) with respect to the strength grade of weld metal were used.
