Joint Special Interest Group meeting between euspen and ASPE Advancing Precision in Additive Manufacturing

Liam Blunt,Yue Liu,Zonghua Zhang,Chris Smith,David Knight,Feng Gao,Andrew Townsend,Xiangqian Jiang

2019-01-01

Abstract:Additive manufacturing (AM) of metal powder fusion techniques by electron beam fusion offers significant advantages compared with traditional subtractive manufacturing techniques. AM technologies have continued to develop over recent years and have seen uptake in particular for medical implants and aerospace parts. Unfortunately despite it clear advantages AM suffers from several significant limitations which need to be addressed before wider take up of the technology is achieved. Powder delivery is a process that occurs many thousands of times during build cycles, consequently it is important to qualify that the powder is delivered without encountering problems such as ridges and furrows resulting from rake damage or rake vibration and areas of low dosage. Like wise during the build process it is important to observe out of plane defects on parts such as thermal swelling. It is additionally useful to observe in-plane information such as the boundary of the melted powder and unmelted powder, which effectively constitutes the layer by layer part edge during the build. This paper describes the deployment of an in process powder/part surface measurement system on a commercial E-Beam AM machine. The inspection system is based on phase measurement profilometry. A temporal synchronisation technique is employed for data capture with and acquisition time of 2 seconds for each inspection. A novel calibration method based on a surface fitting algorithm is investigated and employed to reduce the influence of phase error and random noise. Examples of the development of part thermal swelling monitoring during the build and eventual rake damage are given. Experimental results demonstrate that powder quality and part geometry defects can be efficiently inspected during the build process and future work continues to investigate fusing CAD and measurement data to enhance the data usefulness.