Motorsport electrification raised the need to manufacture prototypes of battery packs. The manufacturing process requires welding many cylindrical cells to reach power and energy requirements. Cylindrical cells are made of steel and for this application, they are joined together using two layers of copper busbars. Dissimilar metals and significantly different melting temperatures make forming a joint difficult. Micro GTAW spot welding can improve the weldability of such battery packs and the joint quality due to the characteristics inherited from regular GTAW. In this work, tungsten electrode geometry was changed and sharpened at different angles to enhance the weld profile. Two types of joints were welded, copper-to-copper and copper-to-steel, which are the main connections of the battery tabs and the battery cells forming a pack. The study also compared pulsed and constant current waveforms and showed that the application of the pulsed one can eliminate pores in the copper-to-copper weld, enhancing its mechanical resistance.
The battery pack welding was simulated using two electrolytic copper (with 0.4% Oxygen) sheets with 0.5 mm each, joined to a carbon steel battery cell (with 0.03% C, 0.003% Si, 0.23% Mn, 0.011% P, 0.008% S). The battery pack is formed by two joints, one is carried out to join the first copper sheet with the cell, and the second one joins the first copper layer with a second one with the same thickness. A second sheet has the purpose of reducing the electrical resistance and the battery heating under operation.
The micro TIG welds were carried out using a power source of Sunstone model 250i. This equipment can apply up to 250 Joules of energy and the power can be set from 1 kW to 20 kW. A 1.0 mm pure tungsten electrode was applied and sharpened with 3 different tips to evaluate its effect over the spot weld. The electrode sharpening angle was changed from 15░, 30░ and different truncation diameters. The characteristic waveform of this equipment has a trapezoidal shape and a configuration that enables it to work with a constant slope down the ramp or a pulsed one. Both waveforms were studied, and the results were evaluated on the macrography of the joints. The shielding gas was pure Argon for all conditions. The copper samples were prepared and etched according to the ASTM E407, which indicates a solution of 1 g FeCl3, 10 ml HCL, and 100 ml water. The etchant was swabbed over the sample for 10s.
Tests were carried out to verify the effect of the electrode’s tip on the spot-welding geometry and molten pool behavior for both conditions, the dissimilar joint copper-to-steel, and the similar copper-to-copper. The electrode’s tip was sharpened at three different angles. The first two angles, 30║ and 15║, are largely applied to regular GTAW welding, although the third one, very truncated, is not so common. After the truncation, the flat diameter of the tips was 0.65 mm, 0.55 mm, and 1,10 mm, respectively.
The molten spot of the copper-to-copper was more spread and irregular than copper-to-steel for low truncated tips. The best parameters set were found after a preliminary run with several tests using the trapezoidal waveform with constant current and linear slope down. The best welding parameters were 11 kW for copper-to-steel and 20 kW for copper-to-copper. As copper has higher thermal conductivity compared to steel, the Micro-TIG weld of copper-to-copper is more difficult, once the heat transferred by conduction away from the spot zone needs to be supplied, and the arc’s power increased.
The current density theory can explain the behavior verified in the molten spots. Furthermore, for the very sharp tips, the current density is more concentrated in the center of the spot, promoting a high pressure on that area and facilitating the blow of the material out of the molten spot. Such behavior is less dependent on the sharpening angle because the same blow phenomenon was observed on the welds with 15║ and 30║.
The current distribution changes depending on the electrode’s tip geometry, mostly on the truncation diameter. The more the flat surface the more the spread of the plasma electromagnetic force over the spot surface and the lower the pressure of the arc for the same current level. Increasing the power to supply extra heat input to avoid a lack of fusion of copper will end up only removing the molten material beneath the arc. A cross-section analysis of the copper-to-steel spot weld showed that the material remained within the spot weld even for very sharp tips. In the copper-to-copper welding using the very truncated electrode, the blow-out phenomenon was eliminated; however, pores were presented within the molten zone.
Thus, using the very truncated electrode only part of the problem welding electrolytic copper-to-copper was solved due to the presence of pores. Several different trials were carried out to change the gas flow rate, however, lowering or increasing the gas just enhanced the pores in the molten spot. As the material has a small amount of oxygen in its chemical composition (0.04%) it can increase susceptibility to such pores.
To eliminate the pores, the second hypothesis was to apply the pulsed waveform. The pulsed current can promote a vibration in the molten material facilitating gas bubbles to be expelled out of the molten pool. In the macrograph analysis, the pores were eliminated for both joints with the pulsed current.
The electrode’s tip strongly affected the copper-to-copper weld profile while the copper-to-steel was not significantly changed.
A very truncated electrode can enable the welding using high energy due to its characteristic of reducing the current density and the arc’s pressure. Welding electrolytic copper is susceptible to pores due to the amount of oxygen in the base material and changing the gas flow rate does not eliminate them. The best results in terms of reducing pores were achieved using a pulsed current waveform.