Simultaneous to the rapid evolution of laser welding, with increased improvements concerning reliability and effectiveness, the need for quality control has grown. The development of system which monitor process quality online, and detect faulty welding results immediately after the defect occurs, has proven to be useful. Such process monitoring systems are usually based on the evaluation of the radiation emitted from the welding spot. This procedure offers a simple way to rate the welding quality, and to detect typical errors. However, it is not possible to recognize the nature of the fault with these single-sensor- setups. To obtain more specific information about the process, additional sensoring is necessary. Within this paper, an approach with IR signal acquisition is described, and correlations are presented. It is shown that the supplementary information coming from a 2D thermo-image allows the identification of typical welding defects. Subsequently, a method for an optimized and practical evaluation strategy using a 1D, line-shaped IR-acquisition is introduced.

Date Published: 24 November 1999
PDF: 10 pages
Proc. SPIE 3832, Sensors and Controls for Intelligent Machining and Manufacturing Mechatronics, (24 November 1999); doi: 10.1117/12.371170

Published in SPIE Proceedings Vol. 3832:
Sensors and Controls for Intelligent Machining and Manufacturing Mechatronics
Patrick F. Muir; George K. Knopf; Patrick F. Muir; Peter E. Orban, Editor(s)