Optical detection of defects during laser metal deposition: Simulations and experiment

Abstract

Laser metal deposition is a rapidly evolving method for additive manufacturing that combines high performance and simplified production routine. Quality of production depends on instrumental design and operational parameters that require constant control during the process. In this work, feasibility of using optical spectroscopy as a control method is studied via modeling and experimentally. A simplified thermal model is developed based on the time-dependent diffusion-conduction heat equation and geometrical light collection into detection optics. Intense light emitted by a laser-heated spot moving across a sample surface is collected and processed to yield the temperature and other temperature-related parameters. In a presence of surface defects the temperature field is distorted in a specific manner that depends on a shape and size of the defect. Optical signals produced by such the distorted temperature fields are simulated and verified experimentally using a 3D metal printer and a sample with artificially carved defects. Three quantities are tested as possible metrics for process monitoring: temperature, integral intensity, and correlation coefficient. The shapes of the simulated signals qualitatively agree with the experimental signals; this allows a cautious inference that optical spectroscopy is capable of detecting a defect and, possibly, predicting its character, e.g. inner or protruding.