Influence of heat source arrangement on coupling characteristics of low-power pulsed laser-MAG hybrid welding

Low-power pulsed laser-MAG hybrid welding was carried out by different heat source arrangement methods (lead mode, distance between laser spot and welding wire tip (DLA)). The coupling effect and welding characteristics of hybrid heat sources under different modes were studied by using high-speed image (HSI) and real-time electrical signals. HSI observation and electronic signals showed that the lead mode and DLA significantly affected the coupling state of the heat source and welding characteristics (weld morphology, process stability, and droplet transfer). The influence of lead mode on weld forming was more significant than that of DLA. In laser-lead mode, when DLA = 1 ~ 2 mm, the laser and arc generated a perfect coupling effect, resulting in a good weld formation, less spatter, and greater penetration. In arc-lead mode, the weld formation was poor and the undercut defect was common, while the penetration was greater than that in laser-lead mode. In terms of welding process stability, the arc-lead mode was better than the laser-lead mode. When DLA = 1 ~ 2 mm, stable droplet transfer could be realized under both modes.


Introduction
As an advanced and efficient laser welding technology, laser arc hybrid welding has promising application potential in different materials such as high-strength steel and aluminum alloy and different plate thickness, especially in thick plate welding [1,2]. At present, it has shown great application prospects in the oil and gas pipeline industry [3]. However, most existing laser arc hybrid welding technologies adopt high-power CW lasers, which are expensive and have strict requirements on the operating environment [4]. Meanwhile, high-power laser arc hybrid welding is mainly laser welding, which has poor tolerance to joint gaps and requires high assembly accuracy. Low-power pulsed laser MAG arc is a hybrid welding technology that is mainly arc based and laser assisted. The pulsed laser with an average power below 1 kW is used to induce and enhance the arc, thereby significantly improving the energy density of the arc. It can maintain the characteristics of the arc welding molten pool under the condition of small heat input and has the advantages of energy saving and good process adaptability [5].
Heat source arrangement involved the relative spatial arrangement of two heat sources, that is, the lead-trail relationship and the relative distance of heat sources (D LA ), which had an obvious impact on the coupling characteristics of welding heat sources, the stability of the welding process, and weld formation. The lead-trail relationship determined the sequence of heat source action. Generally, there were two lead modes: laser-lead mode and arc-lead mode. Chen et al. [6] utilized the effect of lead mode on heat and mass transfer in laser-MIG hybrid welding of aluminum alloy. It indicated that it is easier to obtain larger penetration in arc-lead mode, and laser-lead mode is more conducive to the homogenization of weld microstructure. Zhang et al. [7] stated that in laser MAG welding, porosity and undercut defects are more likely to occur in arc-lead mode. Tang et al. [8,9] observed the hump phenomenon on the back of the weld in laser arc hybrid butt welding and found that arc-lead mode can effectively inhibit the hump formation. Bunaziv et al. [10] applied double-sided fiber laser arc hybrid welding to weld 45-mm-thick highstrength steel, and results showed that to obtain greater penetration, D LA should be larger in arc-lead mode and smaller in laser-lead mode.
The heat source coupling effect was generally realized by adjusting D LA . D LA was one of the main factors affecting the coupling behavior of laser and arc plasma. Meanwhile, it also had a great influence on welding stability. Gui et al. [11] investigated that when the D LA was small in low-power laser-TIG welding, the arc would be significantly compressed, and the preheating of the arc would greatly improve the laser absorption. Liu et al. [12] ascertained that D LA in laser TIG hybrid welding seriously affects the stability of the welding process. By accurately adjusting D LA , the behavior of the keyhole and molten pool could be controlled to obtain appropriate weld penetration. Shi et al. [13] demonstrated that undercut defects during laser-TIG wire filling welding of TA15 titanium alloy could be effectively suppressed by adjusting the D LA and shielding gas. Zhang et al. [14] found that the droplet transfer mode and the ratio of laser energy to arc energy were greatly affected by D LA in laser-MAG hybrid welding. When the D LA is 2.5-3.5 mm, the influence of the droplet on the keyhole could be ignored and the welding process was stable. According to Xu et al. [15], the effect of D LA on energy transfer behavior in keyhole during low-power pulsed laser arc hybrid welding was explored by spectral diagnosis method. It is found that precise adjustment of D LA could control the energy transfer mode in the keyhole and improve the utilization rate of the heat source. However, there are few reports on the effects of different heat source arrangements on arc stability, droplet transfer, and defect formation in low-power pulsed laser-MAG hybrid welding by real-time HSI and electrical signals.
In this paper, the effects of different heat source arrangements (lead mode and D LA ) on the heat source coupling effect and welding characteristics of low-power pulsed laser MAG hybrid welding were explained and analyzed in detail. The arc shape, droplet transfer mode, and molten pool behavior were researched by HSI and electronic signal synchronously.

Experimental methods
The experimental system includes pulsed Nd: YAG laser, Lorch S8 GMAW welding machine, FANUC Robot M-10iA, and moving platform, as shown in Fig. 1. The wavelength of the laser beam was 1064 nm, and a circular light spot with a diameter of 0.6 mm was focused on the workpiece by a convex lens with a focal length of 150 mm. The rated average power of the laser was 1 kW, and the pulse energy was controlled by the excitation current and pulse width. The laser beam was irradiated vertically to the workpiece, and the welding torch was tilted 45° from the vertical. The welding power supply adopted DC reverse connection mode and the wire extension was 12 mm. The 80% Ar and 20% CO 2 were used as shielding gases, and the gas flow was 20 L/min. The composite welding torch was installed on the FANUC robot arm to realize the free movement of the x, y, and z axes. The horizontal distance (D LA ) between the MAG welding wire tip and the laser beam axis was precisely adjusted by a screw Fig. 1 Laser-MAG hybrid welding and diagnosis system micrometer with an accuracy of 0.003 mm, and the D LA adjustment range was − 1 ~ 3 mm. The constant speed of the moving platform was set to 0.72 m/min. The specific welding parameters are shown in Table 1. Welding was carried out on the plate, and the influence of lead mode and D LA on arc morphology, droplet transfer, and molten pool flow was discussed. There were two configurations on the heat source lead modes: laser-lead and arc-lead, as illustrated in Fig. 2.
The welding base metal (BM) was Q345 steel, the workpiece size was 200 mm × 100 mm × 6 mm, and the welding wire was H08Mn2SiA with Φ 1.2 mm. Before the welding experiment, the surface of the workpiece was ground and cleaned with an acetone solution to remove surface contamination. MS50K high-speed camera and image acquisition card were adopted to collect images of the arc shape, droplet transition behavior, and molten pool flow, and the selected sampling frequency was 2000 frames per second. In addition, the narrow-band filters with the central wavelengths of 659.5 and 809.5 nm are respectively placed in front of the camera lens to filter and observe the arc plasma and molten pool behavior. When photographing the molten pool, a diode laser with a center wavelength of 808 nm and an output power of 0 ~ 40 W was used as the auxiliary lighting source. The beam was focused beneath the laser action point, forming an elliptical spot large enough to cover the entire molten pool. The collected information was transmitted to the software on the computer through the cable and displayed as images. Hannover analyzer, data acquisition card, and industrial PC were employed to simultaneously collect and save transient welding current and voltage data, with a sampling rate of 10 kHz.
The weld size (weld penetration, width, and reinforcement) was evaluated using Image Pro Plus image analysis software. According to the standard process of metallographic preparation, the welded metallographic samples were prepared and corroded in 4% nitric acid + alcohol solution for 8 s. The morphology of welded joints was observed with a scanner.

Weld feature
Figures 3 and 4 demonstrate the weld morphology and joint cross-section characteristics of laser-MAG hybrid welding in two modes: laser-lead and arc-lead, respectively. There existed obvious differences, which indicated that the coupling effect of the heat source and the thermal transfer mechanism between heat source and workpiece were different in the two modes. Compared with a single MAG welding, the penetration still increased to a great extent in both modes. Figure 3a presents the weld formation of single MAG, the surface was well formed, while existed many welding spatters. As depicted in Fig. 3b and c, at D LA = − 1 ~ 0 mm, the laser pulse acted on the tip of the welding wire, which interfered with the generation and falling off of the droplet. The large size of welding spatters occurred. At D LA = 1 ~ 2 mm, the weld surface was intact and smooth without spatter and undercut, as demonstrated in Fig. 3d and e. Due to the excellent coupling effect between laser and arc, the arc volume was compressed and the current density increased, increasing in arc stiffness [16]. The arc energy was more concentrated, the welding process was stable, and the spatter was reduced. At D LA = 3 mm, the coupling effect weakened. The induced compression effect on the arc was limited and the arc stability deteriorated and spattering occurred, as shown in Fig. 3f. The cross-section of the weld obtained in laserlead mode presented a "wine cup" shape approximately, and the keyhole generated by the pulsed laser could achieve greater penetration. At D LA = 1 mm, the heat source coupling effect was the best and the heat source utilization rate was maximized; the maximum penetration was 2.70 mm. Compared with laser-lead mode, the weld width in arclead mode was narrower, and the weld reinforcement was bigger, as indicated in Fig. 4. Welding spatter was less and serious undercut defects occurred. In Fig. 4a, there were serious undercuts at the weld toe on both sides of a single MAG weld. Owing to the arc force accelerated the backward flow of the molten pool, and the metal at the rear end of the molten pool solidified prematurely before spreading, resulting in the undercut. At D LA = − 1 mm, there were many small spatters on the weld surface. At D LA = 0 ~ 2 mm, there was almost no spatter in the weld but undercut still existed, as presented in Fig. 4b-e. The keyhole was not only to increase the penetration depth, but also to promote the flow of metal inside the molten pool. Undercut defects improved but still existed due to excessive arc force and plasma flow force. In Fig. 4f, at D LA = 3 mm, a large number of fine spatters were produced. This showed that excessive D LA led to the weakening of the coupling effect. The arc stability deteriorated and the spatter increased.
As shown in Fig. 5, D LA had a significant effect on the weld penetration, width, and reinforcement. The weld penetration in both modes first increased and then decreased It showed that the coupling effect between laser and arc was the best, and the welding energy density of the hybrid heat source increased, resulting in maximum penetration. For the arc-lead mode, the arc force accelerated the backflow of the molten pool and made the keyhole effect more obvious, and more heat was transmitted to the inside of the molten pool, thereby increasing the penetration depth more obviously. The lead mode had a great influence on the weld width, and the weld width in the laser-lead mode was larger, while D LA has little impact on the weld width, and the weld width fluctuates only in a small range. In laser-lead mode, the welding torch was tilted forward and the backflow of molten pool metal was weakened, so the weld width increased. Figure 5c exhibits the effect of D LA on the weld reinforcement in two modes. In laser-lead mode, it could catch smaller weld reinforcement. D LA had little effect on weld reinforcement, which remained unchanged. In arc-lead mode, the backflow of the molten pool was strengthened and the weld reinforcement increased. At D LA = 2 mm, the minimum weld reinforcement reached 2.75 mm. At this time, the coupling effect between the pulsed laser and arc was the best. The coupled heat source was beneficial to the backflow of molten pool metal and the two sides of the weld.

Arc shape and stability
Since the welding power supply adopted DC pulse mode, the arc shape changed periodically with the pulse current waveform. To characterize the arc shape and stability under Weld formation and joint cross-sectional morphology in arc-lead mode different modes and D LA , the arc shape changed within a certain pulse period was observed. The arc morphology at two specific moments (peak current and base current) were selected, as shown in Figs. 6 and 7. Figure 6a demonstrates the arc shape change during single MAG welding in laser-lead mode, which regularly changed in size according to the pulse period. In Fig. 6b, at D LA = − 1 ~ 0 mm, the laser acted on the tip of the welding wire, which generated droplets at the peak current moment, causing a spatter. It was not conducive to the stability of the arc shape. At D LA = 1 ~ 2 mm, the arc shape was stable, and the coupling effect between the laser and the arc was excellent. The attraction and compression of the laser on the arc could be seen at the peak and base values. When the pulsed laser beam interacted with the arc, the arc shrunk as a whole, and a conductive channel was formed between the welding wire and the keyhole. The stiffness of the arc increased, and the plume sprayed obliquely upward, which indicated the intense effect of laser plasma and arc plasma, as illustrated in Fig. 6d and e. At D LA = 3 mm, at the peak value moment, it could be seen that the laser keyhole acted on the molten pool and produced a plume that sprayed vertically upward. Due to the long distance between the laser and the arc, the coupling effect was weakened, and the shape of the arc was not compressed. At the base value moment, the arc force interfered with the stable existence of the keyhole and droplet transfer, resulting in a certain splash, as shown in Fig. 6f.
As shown in Fig. 7, the arc in arc-lead mode acted on the molten pool in a "drag-like" manner. Owing to arc force, the molten pool generated a deeper arc crater, and more heat acted on the molten pool. On the whole, the arc shape was stable and there was little spatter. In Fig. 7a, without laser action, the arc shape changed regularly. At D LA = − 1 mm, as the laser action point was too close to the arc, the expansion of the laser plasma plume intensified, resulting in the expansion and instability of the arc plasma, as shown in Fig. 7b. At D LA = 0 ~ 2 mm, the coupling effect between the laser and the arc at the peak value moment was favorable. The direction of laser plasma plume sprayed obliquely upward from the keyhole was the same, indicating that the generation and closing of the keyhole had little disturbance to the molten pool. The heat sources coupled relatively stable, and the arc shrunk significantly. At the base value moment, the laser mainly acted on the molten pool, and the coupling effect was not obvious, as demonstrated in Fig. 7c-e. At D LA = 3 mm, the laser acted on the convex part of the molten pool and the keyhole collapsed and closed prematurely, producing some splash, as illustrated in Fig. 7f.
The real-time synchronous acquisition electric signal waveforms of welding current (I) and arc voltage (U) was used to assist in the evaluation of the welding process [17], as shown in Figs. 8 and 9.
The red waveform represents the welding current and the black waveform represents the arc voltage. In MAG  Fig. 8b and c, the current waveform fluctuated greatly within 5 s, the regular fluctuation was destroyed, and the voltage dropped sharply down to a low value for many times in the voltage waveform, which meant that short circuit transition occurred. The shielding and absorption of the laser by arc plasma were enhanced, and the laser beam directly hit the droplet, resulting in spatters. Meanwhile, owing to the droplet transition path that was opposite to the flow direction of the molten pool, a disordered flow field resulted at the front of the molten pool, which was prone to splashes, as indicated in Fig. 3b and c. In Fig. 8d, the current waveform was very stable and the frequency of short-circuit transition in the voltage waveform was significantly reduced. With the increase of D LA , when D LA = 2 mm, as shown in Fig. 8e, the fluctuation of the current waveform increased again, but the frequency of short-circuit occurrence was still better than that in Fig. 8b and c. At D LA = 3 mm, the fluctuation of the current waveform became worse regularly, and the frequency of short circuit in voltage waveform increased, which meant that the welding process stability decreased, as depicted in Fig. 8f. As the laser acted on the arc edge, arc plasma and laser plasma were separated from each other, and the effect of the laser on arc-induced compression weakened, resulting in poor arc stability.
For arc-lead mode, the dynamic characteristics of signal waveforms had similar change tendencies with the increase of D LA , as shown in Fig. 9. The current waveforms under different D LA were relatively stable, and the short circuit transition frequency in the voltage waveform was very low. According to the research of [18], the arc plasma was compressed by the laser, resulting in the change of arc force acting on the droplet and the decrease of droplet transfer frequency. The droplet transfer time was prolonged, which was beneficial to improve the stability of droplet transfer.
For the welding process, the coefficients of variation (C V ) of current and voltage were not affected by welding parameters and could be employed to objectively evaluate welding process stability [19]. The small C V meant slight signal waveform fluctuation. The better the arc stability, the more stable the welding process. To further study the effect of D LA in different modes on arc stability, C V was introduced to evaluate the welding stability, as shown in Fig. 10. C V can be expressed by formula (1): where σ and μ are the standard deviation and the mean value, respectively. N and x i are total and continuous random variables.
In Fig. 10, the change trends of arc stability in both modes were similar. The C V of current and voltage first decreased and then increased with the increase of D LA , and the minimum C V appeared at D LA = 1 mm, the coupling effect was best, and the arc stability was optimum.
Overall, the C V in arc-lead mode was lower, indicating that the arc stability was better in arc-lead mode.
(1) Figure 11 demonstrates the effects of D LA on arc shape, droplet transfer, and molten pool flow in laser-lead mode. The HIS images of MAG welding, the schematic diagram of droplet transfer, and molten pool flow are depicted in Fig. 11a, g, and m; the arc shape was larger and the spatter phenomenon was relatively severe. With D LA increased, the coupling effect of laser and arc first increased and then decreased. At D LA = − 1 ~ 0 mm, the path of the laser incident to the molten pool passed through the tip of the welding wire, which interfered with the formation and shedding of the droplet. The transition stability of the droplet was poor, and the frequency of splashing was high. Meanwhile, the laser plasma and arc coupled to discharge, and the droplet was pulled toward the keyhole. Due to the plume recoil generated by the keyhole, splashing was easy to occur, as indicated in Fig. 11b, c, h, and i. At D LA = 1 ~ 2 mm, the coupling effect was favorable, and the addition of laser produced a high-density laser plasma, which could stabilize the arc and compress the arc column. The arc shape shrunk as a whole, making the arc energy more concentrated. The arc plasma deflected toward the keyhole, as depicted in Fig. 11j and k. The droplet gradually moved away from the keyhole, and the reaction force of metal vapor in the keyhole to the droplet decreased accordingly. In addition, owing to the laser no longer being directly irradiated on the droplet, the laser energy loss was less, and the energy was more concentrated. The thermal radiation became stronger, resulting in a decrease in the surface tension of the droplet and more projected transitions. It was beneficial to the welding stability and the transition was stable with less splash. In Fig. 11p and q, the arc force pushed the liquid metal to the front of the molten pool. The keyhole would affect the molten pool and generate a radial force, which acted together to form a composite fluid field. The part of molten metal flowed forward first and then flowed to the tail of the molten pool through the edges. The laser keyhole effect strengthened, and the current density of the keyhole increased. The electromagnetic convection in the molten pool enhanced, and the liquid metal in the middle of the Fig. 8 Waveforms of electrical signals in laser-lead mode molten pool flowed downward faster from the surface, which benefited the transfer of more heat to the bottom of the molten pool and increased the penetration depth. At D LA = 3 mm, the arc plasma and laser plasma were at the separation edge. The coupling effect was weak, and the arc shape was not affected by the laser and did not shift to the keyhole. The liquid molten pool expanded and squeezed the side wall of the keyhole. Meanwhile, the metal vapor collided with the expanded metal liquid under the action of laser recoil to form a splash, as shown in Fig. 11f, l, and r. Figure 12 demonstrates the arc shape, droplet transfer, and molten pool flow in arc-lead mode. Since the arc force in the horizontal direction was opposite to the welding direction, the molten metal flowed backward and became more aggravating. The weld toe was prone to premature solidification, which hindered the wetting of the weld and led to an undercut. The liquid level at the front of the molten pool concaved obviously, and more arc heat acted on the molten pool, which was beneficial to increase the penetration. The arc shape change was similar to that in laser-lead mode, as illustrated in Fig. 12b ~ f. At D LA = − 1 mm, laser Fig. 9 Waveforms of electrical signals in arc-lead mode plasma and arc plasma were very unstable. The laser beam directly irradiated the droplet, and the arc interfered with the laser keyhole, which reduced the stability of the composite plasma. Meanwhile, due to the laser-induced steam reaction, the droplet transfer was difficult and unstable. At D LA = 0 ~ 2 mm, the plasma of the laser and arc was relatively stable, and the coupling effect was excellent. The arc shape shrunk and the stiffness increased. The arc shrinkage led to the effective width of the molten pool being reduced, which was beneficial to reduce the undercut. A discharge channel was generated between the arc and the keyhole, and more heat was transferred to the molten bottom through the keyhole. The droplets regularly transited to the molten pool with little spatter. In the meantime, due to the coupling effect of laser and arc, the arc stiffness increased and the area of action decreased. In the width direction of the upper molten pool, near the center line of the molten pool, the liquid metal temperature was the highest and the surface tension was the lowest. Nevertheless, near the boundary of the molten pool, the temperature of liquid metal decreased and the surface tension increased. Therefore, driven by the increasing tension gradient on the surface of the molten pool, the liquid metal on the surface of the molten pool accelerated to flow from the center line of the molten pool to the boundary. The flow rate of liquid metal rose and the time for the weld metal to flow from the molten center to the edge was more sufficient, which ultimately made more liquid metal move to the weld toe, filling the weld toe and lightening undercuts. While at D LA = 3 mm, the droplet fell into the molten pool and was close to the keyhole, which was on the back wall of the molten pool pit. The impact force of the droplet and the backflow force of molten metal shortened the existence time of the keyhole and the keyhole was easier to close [7]. Figure 13 reveals the typical force diagram of droplets in low-power pulsed laser-MAG welding. To make the droplet fall off from the end of the welding wire, the resultant force that promoted the droplet transfer in the vertical direction must be downward. That is, the following formula is satisfied:

Force analysis on droplet transfer
where F g is the gravity of the droplet, F em is electromagnetic pinch force, F p is plasma flow force, F γ is surface tension, and F RL is laser recoil pressure.
F g is the force that promotes the droplet to escape from the end of the welding wire, which could be expressed as follows [20]: where r is the droplet radius, ρ is the droplet density, and g is the gravitational acceleration.
F em acting on the droplet could usually be decomposed into two components, radial and axial. When pulsed laser and arc  coupled, the number of fine droplets increased. Meanwhile, due to the shrinkage of the arc, the stiffness increased, the current lines in the droplets were dense, and the Lorentz force hindered the separation of droplets. The droplet transition frequency decreased. F em could be expressed by parameters such as the current size in the droplet [21]: where D w is the diameter of the welding wire, D a is the diameter of the droplet, μ 0 is vacuum permeability (4π × 10 −7 H·m −1 ), θ is the angle of arc coverage, and I is current intensity.
When the current lines diverged, the F em acted as a detachment force. When the current lines converged, it was the retaining force. It could be seen that with the change of D LA , the arc stiffness changed. When D LA = 0 ~ 2 mm, the coupling effect between laser and arc was the strongest, the arc stiffness was the largest, the arc coverage angle θ was the smallest, and the F em was the largest.
F p facilitates droplet detachment from welding wire, which had an important influence on the separation of the droplet and the movement after separation. F p could be expressed as follows [8]: where C d is the plasma flow coefficient taken as 0.45, A p is the action area of plasma flow, V f is the flow velocity of plasma flow, and ρ f is the density of plasma flow. The coupling effect between the pulsed laser and arc increased first and then decreased with the increment of D LA . When the laser was coupled with the arc, the cross-sectional of the arc column decreased, which increased the current density of the arc and the plasma flow velocity. The peak value of arc pressure increased, that is, the F p acting on the droplet increased.
F γ hindered the transfer of droplets, and it could be expressed as follows [22]: where R is the radius of the welding wire, π is the radius of the droplet, γ is the surface tension coefficient, and γ∝ 1/T, T is the temperature of the droplet. The higher the droplet temperature, the lower the surface tension. When the laser was coupled with the arc, due to the thermal radiation of laser-induced plasma, the γ of the droplet was reduced. F γ reduced, which was more conducive to the refinement of droplet and promoted the transfer of the droplet.
F RL could be expressed as follows [23]: where R h is the maximum metal vapor action radius, A p is the action area of plasma flow, v 0 is the melt velocity, and N a and k B are the Avogadro number and Boltzmann constant. T S and M a are the surface temperature and molecular weight of the metal vapor, respectively. B 0 is the vaporization constant, L v is the latent heat of vaporization, and D LA is the horizontal distance between the laser spot and the welding wire tip. It can be seen that with the increase of D LA , F RL increased first and then decreased, and when D LA = 0 mm, F RL was the largest.
Based on the analysis of force acting on the droplet, it could be seen that with the increase of D LA , the coupling effect between laser and arc first increased and then weakened, and the current lines flowing through the droplet gathered first and then dispersed. Therefore, the trend of F em and F p was the same as the change of coupling effect, while the F γ first decreased and then increased. In laser-lead mode, at D LA = 1 ~ 2 mm, the laser-induced plasma of pulsed laser increased the thermal conductivity of the arc plasma, led to the pinch effect of the arc, increased the energy density of the arc, and promoted the generation of mixed electromagnetic force. At this time, the F γ decreased, the F P increased, and the F RL was the smallest, which was conducive to droplet transfer. The projected transition was stable and perfect, more small droplets were generated, and the generation and separation of droplets were more stable and regular. The situation was similar in arc-lead mode as illustrated in Fig. 12.
In summary, D LA had a direct effect on the droplet transfer mode, while the heat source lead mode has little effect on the metal droplet transfer mode. This, in turn, affected the process stability and formation of the weld morphology. The process stability improved when D LA was in the range of 1 ~ 2 mm and the projected transition was more excellent.

Conclusions
1. The influence of heat source lead mode on weld formation was more significant than that of D LA . Better weld formation could be obtained in laser-lead mode. In laser-lead mode, at D LA = 1 ~ 2 mm, the coupling effect between the pulsed laser and arc was excellent, and the welding seam was well formed. 2. In terms of welding process stability and maximum penetration depth, the arc-lead mode was better than the laser-lead mode. 3. At D LA = 1 ~ 2 mm, stable droplet transfer could be achieved in laser-lead mode. At D LA = 0 ~ 2 mm, stable droplet transfer could be achieved in arc-lead mode.