The effect of parameter changes in CO2 laser welding on spectra

Abstract

The use of laser welding is evermore increased in metal industry. Due to its high process speed, the bottleneck of the production is usually the quality inspection of the weld. A solution to this is the use of automatic monitoring equipment. The monitoring systems are able to identify the changes in the process and weld defects generated but are not always able to identify different weld defects. The monitoring systems usually measure the intensity value of the electromagnetic spectra in quite a wide wavelength range, which may be the reason for the poor ability to identification of detects.Reliable weld defect identification would make it possible to develop a closed-loop control and adjustment of the process. Since with the current systems the weld defects can not be identified reliably enough, it is reasonable to study if there could be certain narrow areas in spectrum which can indicate a creation of a specific weld defect. Therefore, we started to study the spectral output from the welding.In this study on-line spectroscopy was used to measure the effects of parameter changes in CO2 laser welding process. The material welded was mild steel of thickness 6 mm and the joint type was butt joint. The analysis of spectra is very promising: the variations in welding parameters and groove preparation affect the spectra, changing its level and the intensity distribution of it.The use of laser welding is evermore increased in metal industry. Due to its high process speed, the bottleneck of the production is usually the quality inspection of the weld. A solution to this is the use of automatic monitoring equipment. The monitoring systems are able to identify the changes in the process and weld defects generated but are not always able to identify different weld defects. The monitoring systems usually measure the intensity value of the electromagnetic spectra in quite a wide wavelength range, which may be the reason for the poor ability to identification of detects.Reliable weld defect identification would make it possible to develop a closed-loop control and adjustment of the process. Since with the current systems the weld defects can not be identified reliably enough, it is reasonable to study if there could be certain narrow areas in spectrum which can indicate a creation of a specific weld defect. Therefore, we started to study the spectral output from the welding.In this st...