Self guided laser welding

Abstract

High flexibility and consistent high quality are key benefits of laser welding in production. However, high precision of welding trajectory and velocity are required; e.g. in butt welding the focal point of the laser beam with respect to the joint must be maintained with accuracy better than 50 - 100 microns, depending on the focused beam radius.To meet these requirements, seam tracking devices can be used. A sensor measures the joint position, usually 50 to 200 mm ahead of the welding position (TCP – Tool Center Point), and computes a correction vector to compensate the gap trajectory offset. However, the robot still needs to meet accuracy restrictions since the seam tracking control has to receive accurate on-line information about the TCP position from the robot. Furthermore, the time based seam tracking concept contains systematic error sources, which need to be hindered by pre-production sensor calibration.A new approach for seam tracking is introduced by an enhanced sensor concept, where both the ahead joint position and the relative velocity between the welding head and the workpiece – and therefore the relative TCP position - are measured by one camera based sensor. By carrying out the laser beam correction with a robot-independent scanner system, seam tracking can be performed completely independent to the handling. This reduces cost in robot systems due to lower accuracy and output requirements, simplifies interfacing and installation and disengages from sensor calibration necessity.High flexibility and consistent high quality are key benefits of laser welding in production. However, high precision of welding trajectory and velocity are required; e.g. in butt welding the focal point of the laser beam with respect to the joint must be maintained with accuracy better than 50 - 100 microns, depending on the focused beam radius.To meet these requirements, seam tracking devices can be used. A sensor measures the joint position, usually 50 to 200 mm ahead of the welding position (TCP – Tool Center Point), and computes a correction vector to compensate the gap trajectory offset. However, the robot still needs to meet accuracy restrictions since the seam tracking control has to receive accurate on-line information about the TCP position from the robot. Furthermore, the time based seam tracking concept contains systematic error sources, which need to be hindered by pre-production sensor calibration.A new approach for seam tracking is introduced by an enhanced sensor concept, where both the ah...