Real-time defect detection through lateral monitoring of secondary process emissions during ultrashort pulse laser microstructuring

Abstract

Ultrashort pulse (USP) laser machining is characterized by a large spatial precision in the micrometer range and very high processing speeds: Laser pulses with durations of only a few hundred femtoseconds are deflected over the workpiece at speeds of up to 10 m/s. Due to the tradeoff between the precision and productivity, machining processes already last up to multiple days, as is the case for example for structuring complete mold tools of dashboards. It is therefore essential to implement online defect detection and their elimination in order to increase the stability of the established processing as well as accelerate the process development. Because of the rapid changes during the machining, cost-intensive sensor integration, as well as the high requirements on spatial accuracy, online monitoring of USP laser micromachining represents a great challenge, we solve by spatially resolving optical process emissions at different wavelength ranges collected laterally. The monitoring system, that had previously been developed and had undergone initial testing, was further evaluated in this work. Analyses were carried out to investigate the potential of detecting the surface roughness prior to processing as well as its change induced by the USP laser machining. Their success is however dependent on many factors described in this work. Furthermore, successful localization of defects that emerge during processing was shown. Additionally, the possibility of online process control was demonstrated by transferring the analysis algorithms to FPGA, therefore implementing real-time defect detection and feedback.