Abstract The micro plasma arc welding process is associated with different physical phenomena simultaneously. This results in complexities to comprehend the actual mechanism involved during the process. Therefore, a robust numerical model that can compute the weld pool shape, temperature distribution, and thermal history needs to be addressed. Unlike, other arc welding processes, the micro plasma arc welding process utilizes thin sheets of thickness between 0.5 and 2 mm. However, joining thin sheets using a high-density arc welding process quickens the welding defects such as burn-through, thermal stresses, and welding-induced distortions. The incorporation of a surface heat source model for computational modeling of the high energy density welding process impedes heat transfer analysis. In that respect, researchers have developed numerous volumetric heat source models to examine the welding process holistically. Although, selecting volumetric heat source models for miniature welding is a significant task. The present work emphasis developing a rigorous yet efficient model to evaluate weld pool shape, temperature distribution, and thermal history of plasma arc welded Ti6Al4V sheets. The computational modeling is performed using a commercially available COMSOL Multiphysics 5.4 package with a finite element approach. Two different prominent thermal models, namely, Parabolic Gaussian and Conical power energy distribution models are used. A comparative analysis is carried out to determine the most suitable heat source model for evaluating temperature distribution, peak temperature, and thermal history. The analysis is done by juxtaposing the simulated half-cross-section weld macrographs with the published experimental results from independent literature. The numerical results showed that the proximity of top bead width magnitude was obtained using the Parabolic Gaussian heat source model for low heat input magnitude of 47.52 and high heat input magnitude of 65.47 J·mm−1, respectively. In terms of percentage error, the maximum top bead width percentage error for the Parabolic heat source model is 13.26%. However, the maximum top bead width percentage error for the Conical heat source model is 18.36%. Likewise, the maximum bottom bead width percentage error for the Parabolic heat source model and the Conical heat source model is 12.3 and 25.8%, respectively. Overall, it was observed that the Parabolic heat source model produces the least deviating outcomes when compared with the Conical distribution. It was assessed that the Parabolic Gaussian heat source model can be a viable heat source model for numerically evaluating micro-plasma arc welded Ti6Al4V alloy of thin sheets.
Read full abstract