Module and Busbar Welding Inspection
3D machine vision inspects welding quality on EV battery modules and busbars
Related Products
3D laser displacement sensor with PC-based development environment
Combines artificial intelligence with In-Sight vision systems and VisionPro software to automate complex inspection applications.
Module Welding Inspection
At the M&P stage, an EV battery module’s housing plates are welded together. It is crucial to ensure that the welds aren’t too big and that the welding process hasn’t altered critical dimensions. To verify dimensions, a Cognex 3D laser displacement sensor travels along preselected points of the module, generating a 2D profile in x and z. During inspection, measurements are delivered as a pass/fail to confirm that the module has been assembled correctly.
Busbar Welding Inspection
During busbar welding, batteries are welded together into one battery module assembly. Connectors are welded to join the modules, connecting positive and negative ends. Busbars are connected to the connectors and modules, taking energy and moving it from one place to another. The weld slag material needs to be evenly distributed; gaps or “misses” in the weld can cause a power blowout or slow energy transfer. Welds also need to be completely even in width to achieve maximum connection efficiency. This requires a 3D inspection.
Cognex 3D laser displacement sensors inspect welding track quality without slowing down cycle time. The battery module remains stationary as the motion stage carries the DS camera to scan 12 welding tracks on the negative electrode and then the positive electrode. Cameras gauge whether the welding track width falls within tolerance and is accurately positioned relative to the center and at the right height, as well as checking that the welds are complete. Traditionally, it has been difficult for machine vision to accurately inspect welding quality due to natural process variations and the resulting slight changes in appearance between welds. Cognex Deep Learning is able to successfully detect misses and gaps in the welding material, despite naturally occurring variations and confusing surface textures, to accurately flag defects.
