Experimental study of the generation and propagation of acoustic emission signals in laser micro welding

Several studies have formulated methods for monitoring welding quality based on acoustic emission (AE) signals. However, most of these studies have focused only on developing a data-driven model to predict welding quality. The relationships between weld pool behavior and AE signal features, as well as the features of AE signals propagating through two-layer metal sheets, have not been investigated carefully. Because the variation of material/laser beam interaction and weld pool condition during welding can easily alter AE signal features, it is still challenging to deliver a reliable quality monitoring system to the production line. Therefore, single-layer spot welding experiments were conducted under different laser power levels in this study to examine the fundamental mechanism generating various AE signal features in different welding stages. Moreover, experiments were performed to investigate the differences in AE signal propagation through a two-layer steel sheet and a two-layer steel plate. Potential mechanisms of AE signal generation were determined for different welding stages based on the fundamental theory of AE generation, the analysis of generated AE signals, sound signals corresponding to the generated AE signals, and photographs captured using a high-speed camera. The results show that, in the first two welding stages, the frequencies of the AE signals concentrated in different ranges and the AE signals generated at the beginning of heating were unaffected by the laser power. In the third welding stage, high gas pressure accumulated inside the produced keyhole, which resulted in the random generation of bubbles. These bubbles subsequently collapsed, which resulted in the generation of a burst AE signal. Experimental results on the propagation of AE signals through two-layer steel sheets and two-layer steel plates indicated that the generated AE signals collected at different locations were affected by the contact condition between layers and signal propagation paths. This result provides crucial information when developing a reliable welding quality monitoring system with an AE sensor on the base plate in laser spot welding.