In-process quality assurance of SS316L powder bed fusion process

• Main Author: Lim, Sean Zi Yang
• Other Authors: Zhou Wei
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Manufacturing::Quality control; Engineering::Mechanical engineering
• Online Access: https://hdl.handle.net/10356/149435
• Institution: Nanyang Technological University

Since its introduction 30 years ago, additive manufacturing has seen widespread application in numerous fields and industries. AM allows direct and rapid production of parts, reducing costs. There are many types of AM technologies available, this report will focus on selective laser melting (SLM). Despite the many benefits that AM brings, there will inevitably be defects generated in the process. There is thus an ever-perennial need for the continuous study of the optimisation of print parameters to achieve samples of good build quality. Much work has been done by other researchers into the optimisation of these parameters for achieving a good build quality. This report will focus on improving the build speed of the print job while maintaining good build quality as well as studying the in-process monitoring system. The latter of which to identify corresponding photodiode threshold values to predict part build quality and performance. Parts which have been predetermined to have poor build quality can be scrapped immediately without the need for further quality assurance testing. This will save valuable time and resources in the production phase. All samples were manufactured using stainless steel 316L powder. The layer thickness was standardised at 70 μm. A total of 6 batches of samples were printed. The first 5 batches were density cubes of dimension 10 x 10 x 7 mm [XYZ]. Batch 6 was a print of tensile coupons. The results showed that build rate could be increased by up to 20% without any significant effect on part bulk density. However, any further increase in build rate would result in a significant drop in the part’s bulk density. This study also identified a photodiode value threshold range of between 0.9-1.1 for good part performance and build quality. Photodiode values of less than 0.9, saw a steep drop in samples’ bulk density and ultimate tensile strength. However, the drop was not as significant in photodiode values above 1.1. Despite the relatively small threshold range proposed, more work can be done to further investigate if it is possible to further widen the range to allow for more leeway. However, it is also worth noting that target build quality is dependent on user’s application and needs. Hence, the photodiode values do not necessarily need to fall within the threshold range. More in-depth studies will allow for more reliable and accurate prediction of part performance across a wider range of parameters, materials, and AM machines.