Sensors for narrow-gap welding

Abstract

Narrow-gap welding is an important technique for increasing productivity in the manufacture of thick-walled components. The nature of the process demands an automated approach and requires precise control to ensure consistently high weld quality. The work described in the paper gives the design, construction and development of appropriate sensors and sensing techniques for the control of the narrow-gap TIG welding (GTAW) process. Sensing systems based on arc-voltage sensing and direct vision sensing were developed and successfully tested on an experimental welding system. The arc-voltage sensing included the design of a digital technique for calculating the size of the voltage drop as the welding torch approaches a sidewall. The vision-based work involved the development of a very fast access framestore and highly optimised techniques for extracting the position of contrast features within a live image. The relative merits of arc-based and vision-based techniques are discussed.