Surface Roughness Repeatability Analysis for Multi-Build Overhang Parts in Powder Bed Fusion Additive Manufacturing

Abstract Additive manufacturing (AM) has been evolving from rapid prototyping towards applications in high-volume production and mass manufacturing. However, achieving consistent quality and meeting industry standards remains a challenge that hinders AM industrialization. While most traditional manufacturing processes have established process control for part quality management, the complete understanding of part quality repeatability across builds, machines and processes is still missing. Hence, quality-related features such as surface roughness cannot be consistently reproduced for mass production. This study initiates the process of investigating laser powder bed fusion part repeatability in terms of surface roughness, based on the measurements conducted at the National Institute of Standards and Technology (NIST). The research compares the roughness of different surfaces of a single AM part, identical parts from a single build, and parts fabricated from two separate builds. The part has designed surfaces varying from up-skin surfaces with different conditions to down-skin surfaces of overhangs. The roughness measurements based on XCT data reveal that average roughness, Ra, of different types of surfaces is distinguishable from each other. It’s also observed that the roughness of up-skin surfaces consistently exhibits lower values compared to down-skin surfaces. However, inherent variations between builds pose challenges to achieving full repeatability. Analysis indicates that the roughness of all four parts from the first build is lower than that of the 16 parts from the second build with the same process settings and feedstock material. Achieving repeatability becomes increasingly difficult when process, machine or material variations between builds are present.