Abstract
Multi-layered pulse MAG welding by robots is used for large structures. The sensing technology for the detection of defects is crucial towards improving the productivity and quality of robotic welding. The authors have developed a system that detects welding defects (blowholes) by utilizing welding current and voltage waveform data. By processing the measurement data, it is possible to detect welding defects with a probability of 95% or better. The developed defect detection system has been introduced in the company's robotic welding systems.
