In-situ areal inspection of powder bed for electron beam fusion system based on fringe projection profilometry

Yue Liu, Liam Blunt, Zonghua Zhang, Hussein Abdul Rahman, Feng Gao, Xiangqian Jiang

Research output: Contribution to journalArticlepeer-review

47 Citations (Scopus)


Additive manufacturing (AM) techniques provide significant advantages over conventional subtractive manufacturing techniques in terms of the wide range of part geometry that can be obtained. Powder delivery is a process that occurs thousands of times during the AM build process, consequently assessment of delivery quality would be advantageous in the process in order to provide feedback for process control. This paper presents an in-situ quantitative inspection technique for assessing the whole of the powder bed post raking, by using fringe projection profilometry. In order to increase accuracy and traceability of the inspection technique, an accepted fringe projection method, is enhanced using a novel surface fitting algorithm employed to reduce the influence of phase error and random noise during calibration. A simulation was conducted to verify the accuracy of the proposed system calibration. The proposed in-situ inspection technique has been applied in an Electron Beam Powder Bed Fusion (PBF-EB) machine, also known as Electron Beam Melting (EBM). Some examples of melting edge swelling and excessive powder delivery due to rake damage during a real part build are used to demonstrate the system capability on the actual EBM machine. A build process with presented fringe projection system is reported in this paper. Experimental results demonstrate that powder defects can be efficiently inspected and the results used as feedback information in a build process.

Original languageEnglish
Article number100940
Number of pages12
JournalAdditive Manufacturing
Early online date8 Nov 2019
Publication statusPublished - 1 Jan 2020


Dive into the research topics of 'In-situ areal inspection of powder bed for electron beam fusion system based on fringe projection profilometry'. Together they form a unique fingerprint.

Cite this