A high-resolution and large field-of-view scanner for in-line characterization of powder bed defects during additive manufacturing

• Main Authors: Le Tan Phuc, Matteo Seita
• Format: Article
• Language: English
• Published: Elsevier 2019-02-01
• Series: Materials & Design
• Online Access: http://www.sciencedirect.com/science/article/pii/S0264127518309262

Powder bed defects are irregularities in the powder layer, which alter the energy input during the powder bed fusion process. As a result, they are directly responsible for the formation of flaws in the consolidated material, which cause quality and property variability in additive manufactured parts. Because of their small size and ubiquity across the powder bed, powder bed defects are difficult to detect and correct. In this work, we propose a new method to assess powder bed defects across the entire powder bed at the remarkable spatial resolution of ~5 μm. Our method relies on the integration of a contact image sensor taken from a flatbed document scanner to the powder re-coater module. Owing to the narrow depth-of-field of the sensor, we detect powder bed defects by identifying out-of-focus regions in the acquired scans using numerical image analysis techniques. Moreover, we show that we can assess the defects height (or depth) by quantifying the degree of “blurriness” in such regions. Our “powder bed scanner” is a rapid and cost-effective tool for in-line characterization of the powder bed quality. This technology may be instrumental to develop novel close loop strategies aimed at improving the consistency of additive manufactured parts. Keywords: Powder bed fusion, In-situ monitoring, Surface topography, In-line powder bed defect characterization, Contact image sensor, Numerical image analysis