Defect detection by multi-axis infrared process monitoring of laser beam directed energy deposition

Abstract

Abstract Laser beam directed energy deposition (DED-LB) is an attractive additive manufacturing technique to produce versatile and complex 3D structures on demand, apply a cladding, or repair local defects. However, the quality of manufactured parts is difficult to assess by inspection prior to completion, and parts must be extensively inspected post-production to ensure conformance. Consequently, critical defects occurring during the build go undetected. In this work, a new monitoring system combining three infrared cameras along different optical axes capable of monitoring melt pool geometry and vertical displacement throughout deposition is reported. By combining multiple sensor data, an automated algorithm is developed which is capable of identifying the formation of structural features and defects. An intersecting, thin-walled geometry is used to demonstrate the capability of the system to detect process-induced porosity in samples with narrow intersection angles, which is validated using micro-CT observations. The recorded results indicate the root cause of this process-induced porosity at the intersection, and it is shown that advanced toolpath planning can eliminate such defects. The presented methodology demonstrates the value of multi-axis monitoring for identifying both defects and structural features, providing an advancement towards automated detection and alert systems in DED-LB.