Effect of process parameters on arc behavior and weld formation in weaving gas tungsten arc welding

1. Introduction

As an important welding method between black and non-ferrous metals, gas tungsten arc welding (GTAW) uses the arc between the non-melting electrode and the base metal (BM) as the heat source to melt the welding wire or BM for weld forming [1–2]. However, the divergent arc in GTAW makes the heat input uneven, which results in a shallow molten pool, and the weld strength and toughness are lower than that of the BM [3–5]. In addition, because of heat input and other reasons, which result in large residual stress and plastic deformation in the weld and surrounding areas, are caused by the highly localised transient heat and strongly nonlinear temperature fields in both heating and cooling processes. The weld may have a large shape error between layers [6], and the poor forming of some difficult welding materials is not satisfactory for the traditional GTAW welding process.

To resolve these problems, many scholars have done a lot of research work. Clark et al. [7] studied the rules of weld forming and welding speed, wire diameter, and wire feeding speed. They found that width accuracy can be improved by increasing welding speed layer by layer to reduce heat input. Ouyang et al. [8] optimized the surfacing forming effect by changing the welding heat input, and concluded that the arc length, plate preheating temperature, and temperature between the stacks affect the dimensional accuracy and surface quality of the component. There are also many new welding techniques to control the weld morphology, of which weave arc, realized by controlling the left and right periodic weave of the welding gun, was considered to be the most suitable for weld control of surfacing forming [9–10]. In addition, the weaving GTAW (W-GTAW) welding gun has been applied to reduce the occurrence of magnetic bias blowing for the welding process of some difficult welding materials [11]. The arc of the W-GTAW proved to be able to homogenize welding heat input, control molten metal flow and prevent undercut, slag inclusion, poor fusion and other weld defects [12]. The arc welding with weaving has been used widely to obtain better weld quality by avoiding the lack of side wall fusion and improve the weld efficiency by obtaining the wide weld bead. However, the influence mechanism of this process parameters, especially in the stability of welding process and weld formation, is not clear. This also makes the relevant applied research has not been widely reported.

In this work, the effect of process parameters in W-GTAW on arc behavior, welding stability and weld formation was systematically studied to deeply tap the application potential of W-GTAW technology, including the welding current, the height from the tip of the tungsten electrode to the BM, the weave speed, the amplitude and the left and right stage time of the welding gun. This instantaneous arc shape was observed for different processes using high-speed cameras. The welding electrical signal and weld formation were obtained, the influence mechanism of the W-GTAW process parameters on the welding stability and weld formation was also analyzed, respectively.

2. Materials And Methodology

The BM was 7A52 aluminum alloy, and the welding wire was 1.2-mm ER5356 welding wire. The main components are shown in Table 1. The BM dimensions were 150 mm × 100 mm × 4 mm, and the welded joint sample (20 mm × 8 mm) was from the middle part of the weld of the plate. Alternating current GTAW wire filler surfacing was carried out. The diameter of the tungsten electrode was 2.4 mm, the shielding gas flow was 15 L/min, the welding speed was 2.25 mm/s, the wire feeding speed was 33 mm/s, the welding weave device and its path is shown in Fig. 1. The welding parameters in different cases are shown in Table 2.

Table 1

Main chemical composition of 7A52 aluminum alloy and ER5356 welding wire (wt.%)
Elements	Zn	Mg	Cu	Mn	Cr	Ti	Zr	Fe	Si	Al
7A52 aluminum alloy	4.0 ~ 4.8	2.0 ~ 2.8	0.05 ~ 0.2	0.2 ~ 0.5	0.15 ~ 0.25	0.05 ~ 0.18	0.05 ~ 0.15	≤ 0.3	≤ 0.25	Bal.
ER5356 welding wire	≤ 0.1	4.5 ~ 5.5	≤ 0.1	0.05 ~ 0.2	0.05 ~ 0.2	0.06 ~ 0.2	/	≤ 0.4	≤ 0.25	/

Table 2

Welding process parameters in different cases
Cases	Welding currents (A)	Tungsten electrode heights (mm)	Weave angles (°)	Weave speeds (10^− 1rad/s)	Weave stopping times on the left and right sides (s)
a	135	4	1.7	0.30	0.15
b	135	4	1.5	0.30	0.15
c	135	4	1.9	0.30	0.15
d	135	4	1.7	0.20	0.15
e	135	4	1.7	0.40	0.15
f	135	4	1.7	0.30	0.10
g	135	4	1.7	0.30	0.20
h	130	4	1.7	0.30	0.15
i	140	4	1.7	0.30	0.15
j	135	3	1.7	0.30	0.15
k	135	5	1.7	0.30	0.15
l	135	4	0	0	0

During welding, the welding plate of 7A52 aluminum alloy was fixed on the operating table, and only the operating table moved by setting a constant welding speed, while the welding gun weaves left and right above the workpiece according to the set weave parameters, the central axis of the weave of the welding gun was perpendicular to the upper surface of the workpiece along the welding direction. A WSME-500 welding power supplied with a LD-10-500 wire feeding machine was used during the experiments. The instantaneous arc shape was continuously collected by a high-speed camera equipped with NPX-GS6500UM dimmer, and the sampling frequency of the high-speed camera was set to a constant value of 1000 fps in all experiments. A CHV-50VD voltage sensor and a CHB-500S current sensor were respectively applied to collect voltage signals and current signals during welding, while the acquisition frequency of electric signal data acquisition system was 10kHz. After welding, a 20 mm × 8 mm metallographic sample along the vertical direction of the weld was cut by a wire electrical discharge machine, then inlaid with a cold mounting material composed of acrylic powder and rapid inlay curing agent; 600#, 800#, and 1000# sandpaper was used to grind these joint samples; and 1-m diamond polishing agent was used to polish the weld until the cross-section was bright with no obvious scratches. Corrosion operations were required before microscopic observation of weld cross-sections. Keller reagent (HF 1%: HCl 1.5%: HNO₃ 2.5%: H₂O 95%, volume ratio/%) was used to corrode the sample for ~ 30 s before macroscopic observation and photographing with a stereo microscope.

3 Experimental Results And Observations

3.1 Influence of process parameters on arc behavior

Figure 2 exhibit traditional GTAW arc morphology, because the traditional GTAW arc without weave parameters had no arc offset and periodic change. Figure 3 shows the W-GTAW arc morphology in different cases for different process parameters, compared with the traditional bell arc, the shape of the swing arc was conical. For each case, the second and sixth photographs represent the starting and stopping moment of weaving arc. The interval between the first and second photos and the sixth and seventh photos corresponds to the weave stop time on the left and right sides. A comparison of Fig. 3 (a) and (b) shows that when the weave angle increased, the period of the weaving arc was prolonged from 1330 ms to 1440 ms, and the left and right stopping times of the two arcs fluctuated up and down by 150 ms. Calculations showed that the difference in weave speed was small. An increase in weave angle lengthened the motion path of the weaving arc. With no change in weave speed, the period increased. From the calculated time interval in Fig. 3(c), the weave speed was 3.85 × 10^-2 rad/s, the weave speed in Fig. 3(a) was 2.85 × 10^-2 rad/s, and the weave angle and weave stopping times of the two groups remain unchanged. The increase in weave speed resulted in the shorter period in case e. A comparison of the arc cycle images in Fig. 3(a) and (d) showed that the main difference laid in the change in time between the two arc images on the left and the right side. From Fig. 3(d), we can see that the left and right stop times of this arc in case g were 220 and 180 ms respectively, the weave cycle had also been extended. Compared with this welding process in case a from Fig. 3(a), when the tungsten electrode height increased to 5mm in case k for Fig. 3(f), the offset of the arc increased.

In order to deeply study the influence of process parameters in W-GTAW on arc behavior, and for the convenience of representation, when the arc bottom was an ellipse, the long axis of the ellipse was chosen as the radius, the radius of the lower bottom surface of the cone-shaped arc (R_b) was measured and its average value weaving periods was calculated (see Fig. 4, the T in abscissa is the period of the weaving arc). According to the weaving arc path, when the welding gun moved from the midpoint to both sides, the arc radius decreased gradually with an increase in weave angle. The R_b changed visibly for the arc with the smallest weave angle. The radius reached 3.89 mm at the midpoint of the left-right weave path, whereas the radius was only 3.37 mm and 3.30 mm on the left and right sides respectively. The arc with a larger weave speed showed the least obvious change in radius of the bottom of the cone-shaped arc. Its value ranged between 4.00 and 4.05 mm, with the largest average value in the group. Figure 3 shows that this R_b was sensitive to this height change. When the tungsten electrode height from the electrode tip to upper surface of workpiece was reduced to 3mm in case j, the average radius was only 2.75mm. For this reason, it may be that when the tungsten electrode height was increased, this low arc energy density was conducive to maintaining the stability of total arc energy [13], which tended to compress the arc by reducing the height and radius of arc to maintain energy balance. Similarly, the increase of weave speed would also increase the bottom radius of the arc as a whole, but this radius change was not as significant as the increase in the height of tungsten electrode.

3.2 Influence of process parameters on welding stability

In order to study this effect of the process parameters in W-GTAW on welding stability, it is necessary to collect the data of the electrical signals during welding process because they contain some information that can be used to evaluate welding arc and weld formation. Figure 5 shows the welding electrical signal data diagrams in different cases with different welding parameters. In Fig. 5(a), T represents the current fluctuation cycle; P_p represents the positive peaks; V_p represents the positive valley; P_n represents the negative peaks; V_n represents the negative valleys; VR represents the fluctuation value of the arc voltage. It can be seen from Fig. 5(b) that the current and voltage fluctuated periodically, and their cycle could be affected by these welding parameters. When the weave speed was increased, the current cycle was greatly shortened by about 0.26S. Also the welding current increased, the current fluctuation area increased, and the peaks and valleys of each cycle of the current fluctuated to some extent during the weaving process. After removal of extreme values, each current showed large differences in the negative peaks and valleys. Comparative measurement showed that the overall range of case a was small, positive peaks and troughs only have 3.0A and 6.0A floats, respectively, whereas the change between negative peaks and valleys values were 3.9A and 6.4A respectively. The current was unstable for increasing the tungsten electrode height, with the float between negative peaks and valleys 13.4A and 9.6A respectively. Figure 5(c) exhibit that the negative half-axis voltage in case b, c, e and f was chaotic with an obvious jumping phenomenon, which shows that an increase in weave angle, weave speed, and tungsten electrode height destabilized the arc voltage. Here, this maximum voltage fluctuation was obtained between –36.9 V and 18.8 V when the tungsten electrode height varied. Overall, it can be seen that the larger the braiding speed or the braiding angle, the larger the peak fluctuation range of welding current and voltage, and the severe current fluctuation indicates that the stability of the welding process was poor, and the possible spatter might increase, which cause the welding seam to suffer. The mechanical properties of the forming might also have some influence.

The process parameters affect the stability of the electrical signal obtained in welding, and the stability of the electrical signal represents the stability of the arc. This has important implications for weld formation. Welding stability can be reflected by the U-I images. Figure 6 shows the U–I image for different cases. The current changed at the middle of the dynamic operating point. The moving trajectory was more concentrated, and the working points on both sides were noisy. A condition to measure the stability of the welding was to assess whether the electrical signal working points were dense. Remove the image with no weave parameter added, the working point distribution in Fig. 6(a) was more concentrated than that for other parameters, and increasing weave angle, weave speed, weave left and right stopping time, increasing current, and tungsten electrode height affect the welding stability to varying degrees. For Fig. 6(c), the effect of increasing the weave speed was the most obvious.

3.3 Influence of process parameters on weld formation

The change in process parameters affects the weld formation, for example, the changes of welding current, tungsten electrode height, weave angles, weave speed and weave stop time on the left and right sides can affect the weld formation, including weld penetration and width, as shown in Fig. 7. An analysis of the weld formation and a cross-sectional comparison of the images for different welding

parameters show that the W-GTAW weld bead was S-shape. It was the result of the action of the weaving arc and rarely suffered from undercut defects. Since the weaving arc can slow down the solidification process of the metal at the edge of the molten pool and eliminate the obstruction of metal expansion in the middle of the molten pool, it is beneficial to improve the spreadability of the weld surface for the S-shape weld obtained in weaving GTAW welding [22]. An increase in weave angle, weave speed, and weave stop time increased the weld width and the number of crescent shapes of the weld, and widened each crescent for the same weld bead. An increase in welding current yielded poor weld formation, while increased the tungsten electrode height weaken the influence of the weave arc on the weld formation.

In order to better evaluate the weld forming quality obtained under different process parameters, we calculated the weld forming coefficient, as shown in Fig. 8. The weld forming coefficient refers to the ratio of the weld width (B) to the weld depth (H), namely Φ =B/H [15]. If the weld is narrow and deep, the forming coefficient of weld Φ was small. In the center of the weld, more impurities accumulated due to regional segregation, so a smaller forming coefficient of weld would increase the stress concentration and hot crack tendency of the weld [16-17]. In our 11 sets of experiments, the reduction of the height of the tungsten electrode reduced the swing width, resulted in a more concentrated heat input, an increase in the penetration depth, and a decrease in the fusion width, so, the minimum weld coefficient with a tungsten electrode height of 3mm was only 1.81. Whereas the joint with a weave angle of 1.9° had the largest forming coefficient of 3.11, this is due to the increase of the swing angle, which increases the lateral swing path, which in turn improves the ductility of the weld surface. The weld forming factor also increases as a result

4. Analysis And Discussion

Through the above experiments, it was found that besides changed the current parameters, the weave speed had the most obvious influence on the weld appearance. To understand the relationship deeply between the weaving arc behavior and weld formation, this paper selected the weave speed control group for analysis. Figure 9 shows the weld bead profile and its cross-sectional images at different weave speeds. A smaller weave speed results in a more obvious S-shape weld. A higher weave speed results in a greater weld forming coefficient. The forming coefficient of welds that correspond to weave speeds of 0.020, 0.030, and 0.040 rad/s were 1.86, 2.61, and 2.82, but the stability of the electrical signal with a weave speed of 0.040 rad/s was worst in those three experiments. To explore the reasons for the difference in weld formation, the arc image when welding guns in cases a, d, and e were woven to the far right were considered (see Fig. 10). The arc was composed of the cathode, arc column, and anode areas. The weave parameters were relative to the cathode region and the upper part of the arc column region had little influence, but the influence on anode zone and the lower part of the arc column area was more obvious. The arc radius increased, and the change in bottom radius affects the static pressure of the arc, which is a manifestation of the arc transforming heat energy into mechanical energy, and impacts weld penetration and forming. The calculation of the arc force of the arc pressure distribution can be expressed by Equation (1) [18] through integration. Figure 11 provides a schematic diagram of the arc static pressure and weaving arc changes.

In Equation (1), the dependence of arc pressure on the current has been explained by Lin and Eagar

as follows [19]: where p_t is the arc pressure (N/m²), because the p_t is a stagnation pressure of the plasma jet remaining on the anode surface, it can be expressed by Equation (2):

An analysis of Equation (5) shows that the arc pressure p_t is determined by the constant term N, current, and current density. The current density of the arc is concentrated mostly at the upper end of the arc column area, and almost no current flows through the lower end of the arc column area [22]. The change in weave speed has little effect on the upper end of the arc column, and the current and the constant term N are fixed values, and thus the arc pressure is constant for the three weave speeds. An analysis of Equation (1) shows that the arc force F_t was determined by arc pressure and the radius of the bottom surface of the arc column, however, the arc pressure was constant. Therefore, the greater the weave speed, the greater the radius of the lower bottom surface of the conical arc column, and the greater the arc force generated. When the welding gun swung to the far right in the corresponding cross-sectional image, the arc force was correlated positively with the weave speed, therefore, the arc with high weave speed had larger arc force on the weld edge, which caused the molten pool with a higher weave speed to be a flat and inverted trapezoid, whereas the molten pool with a lower weave speed was an inverted triangle.

5. Conclusions

The W-GTAW welding of 7A52 aluminum alloy was carried out with single pass surfacing of the flat plate as the research object, and the effect of welding current, tungsten electrode height from the electrode tip to the upper surface of workpiece, weave angle, weave speed, and weave stop time on the arc shape, welding stability, and weld formation during welding were systematically studied. The main research conclusions were as follows:

1) For the W-GTAW process, the smaller the weave angle of the arc, the more obvious the change of the arc bottom diameter, the maximum variation of this radius was close to 1.2 mm in this experiment. Increasing the weave speed could reduce the radius change, with only 0.10 mm drift. An increase in left and right weave stopping time extended the weave period, and did not change the arc shape. Increase tungsten electrode height raise the overall height of the arc and the arc voltage, when the tungsten electrode height increased to 5mm，the offset of the arc increased obviously.

2) W-GTAW arc can cause the periodic fluctuation of electrical signal. An increase in tungsten electrode height had the greatest impact on the peaks and valleys of the welding current. The maximum peak difference was 13.4 A. A change in weave angle, speed, and tungsten electrode height affected the stability of arc voltage. An increase in tungsten electrode height increased the voltage range, and the voltage ranged between –36.9 and 18.8 V. In the U–I diagram, an increase in weave speed and tungsten electrode tip height decreased the stability.

3) Compared with the weld of traditional GTAW welding, the weld bead morphology for the W-GTAW welding is S-shape. An increase in weave angle expanded the weld width. Increasing the weave speed and the weave stop time on the left and right sides would reduce the maximum penetration and the shape of molten pool is trapezoidal. The increase of welding current and tungsten electrode height was not always conducive to weld formation to a certain extent.

4) The minimum weld forming coefficient of 1.81 could be obtained by increasing this welding current, a smaller forming coefficient of weld would increase the stress concentration and hot crack tendency of the weld, which was not conducive to weld formation. Increasing the weave angle and weave speed would increase the weld forming coefficient. The maximum weld forming coefficient of 3.11 corresponded to the maximum weave angle, which might be beneficial to reduce the stress concentration and thermal cracking tendency of the weld.

References

Gang Z, Yu S, Ming Z et al (2022) Effect of electric parameters on weld pool dynamic behavior in GTAW. J MANUF PROCESS TECH 77:369–379. https://doi.org/10.1016/j.jmapro.2022.03.028
Apa B, Is B, Sbb C et al (2020) Assessment of Impact Strength of Welds Produced by Cold Wire and Hot Wire Gas Tungsten Arc Welding (GTAW) Processes. MATER TODAY 24:983–994. https://doi.org/10.1016/j.matpr.2020.04.411
Rodríguez-Hernández T, Cruz-Hernandez VL, Garcia MA et al (2020) First assessment on the microstructure and mechanical properties of gtaw-gmaw hybrid welding of 6061-t6 AA. J MANUF PROCESS 59:658–667. https://doi.org/10.1016/j.jmapro.2020.09.069
Kumar A, Sundarrajan S (2009) Effect of welding parameters on mechanical properties and optimization of pulsed TIG welding of Al-Mg-Si alloy. INT J ADV MANUF TECH 42(1–2):118–125. https://doi.org/10.1007/s00170-008-1572-8
Chen J et al (2016) Microstructure and mechanical properties of aluminum alloy 7A52 thick plates welded by robotic double-sided coaxial GTAW process. MODEL SIMUL MATER SC 673:8–15. https://doi.org/10.1016/j.msea.2016.07.011
Wei B, Jia C, Wu W et al (2021) Stirring effect of the rotating arc on the molten pool during non-axisymmetric tungsten NG-GTAW. J MATER PROCESS TECH 285. https://doi.org/ 10.1016/j.jmatprotec.2020.116769
Clark D, Bache MR, Whittaker MT et al (2008) Shaped metal deposition of a nickel alloy for aero engine applications. J MATER PROCESS TECH 203(1–3):439–448. https://doi.org/ 10.1016/j.jmatprotec.2007.10.051
Ouyang JH, Wang H, Kovacevic R et al (2002) Rapid prototyping of 5356- aluminum alloy based on variable polarity gas tungsten arc welding: Process control and microstructure. MATER MANUF PROCESS 17(1):103–124. https://doi.org/10.1081/AMP-120002801
Zhu CX, Cheon J, Tang XH et al (2018) Effect of weave arc on molten pool behaviors in narrow-gap GMAW of 5083 Al-alloy. J MATER PROCESS TECH 259:243–258. https://doi.org/10.1016/j.jmatprotec.2018.04.026
Xu G, Wang J, Li P et al (2018) Numerical analysis of heat transfer and fluid flow in weave arc narrow gap GMA welding. J MATER PROCESS TECH 252:260–269. https://doi.org/ 10.1016/j.jmatprotec.2017.09.030
Li FX, Sun QJ, Jin P et al (2021) Wetting behavior of melt and its effect on lack of fusion in arc oscillating NG-GTAW. J MATER PROCESS TECH 296. https://doi.org/10.1016/j.jmatprotec.2021.117176
Zhu J, Wang J, Su N et al (2017) An infrared visual sensing detection approach for weave arc narrow gap weld deviation. J MATER PROCESS TECH 243:258–268. https://doi.org/ 10.1016/j.jmatprotec.2016.12.029
Antonello MG, Bracarense AQ, Scheuer CJ et al (2021) Effect of electromagnetic arc constriction applied in GTAW-based wire arc additive manufacturing on walls' geometry and microstructure. J MATER PROCESS TECH 71:156–167. https://doi.org/10.1016/j.jmapro.2021.09.015
Huang J, Chen H, He J et al (2019) Narrow gap applications of swing TIG-MIG hybrid weldings. J MATER PROCESS TECH 271:609–614. https://doi.org/10.1016/j.jmatprotec.2019.04.043
Sachin M, Saurabh C, Manjit S et al (2021) Analyzing the response of submerged arc welding process parameters on Form factor and dilution. MATER TODAY 40. https://doi.org/10.1016/j.matpr.2021.09.143
Ottersbock MJ, Leitner M, Stoschka M (2021) Characterisation of actual weld geometry and stress concentration of butt welds exhibiting local undercuts. ENG STRUCT 240. https://doi.org/10.1016/j.engstruct.2021.112266
Kim HS, Chen J, Lee HB et al (2021) Effect of heat treatment on grain boundary carbides and primary water stress corrosion cracking resistance of Alloy 182 weld. CORROS SCI 191(6). https://doi.org/10.1016/j.corsci.2021.109730
Oh DS, Kim YS, Cho SM (2005) Derivation of current density distribution by arc pressure measurement in GTA welding. SCI TECHNOL WELD JOI 10(4):442–446. https://doi.org/10.1179/174329305X44116
Lin ML, Eagar TW (1986) Pressures produced by gas tungsten arcs. METALL MATER TRANS B 17(3):601–607. https://doi.org/10.1007/BF02670227
Lancaster JF (1986) The physics of fusion-welding.1. the electric-arc in welding. Phys Weld (Second Edition) 134(5):233–254. https://doi.org/10.1049/ip-b.1987.0040
Maecker H (1955) Plasmastrmungen in Lichtbgen infolge eigenmagnetischer Kompression. Z für Physik 141(1):198–216. https://doi.org/10.1007/BF01327300
Benakis M (2020) Current mode effects on weld bead geometry and heat affected zone in pulsed wire arc additive manufacturing of Ti-6-4 and Inconel 718. J MANUF PROCESS 60:61–74. https://doi.org/ 10.1016/j.jmapro.2020.10.018