Understanding the effects of volumetric defects on the uniaxial fatigue behavior of additively manufactured metallic materials

Abstract

Volumetric defects in additively manufactured parts cause significant fatigue scatter due to the variations in their size, shape, and location. This dissertation aimed to identify the morphological features of process-induced volumetric defects that most influence the fatigue behavior of laser powder bed fused (L-PBF) Ti-6Al-4V. Firstly, fatigue test specimens were machined from round bars fabricated using eight distinct process parameter sets and three build orientations to ensure a wide range of defect morphologies and populations. Uniaxial, constant amplitude, fully-reversed fatigue tests were conducted on these specimens. Fractography was performed to identify the defect responsible for the fatigue crack initiation and extract its morphological features. The influence of these features on fatigue behavior was examined by analyzing experimental data and finite element analysis results. While size was the most critical defect feature influencing fatigue behavior, it could not fully explain the variation in fatigue life by itself. For defects of similar size located at the same location, circular defects appeared to be more detrimental than irregularly shaped ones. A new defect size parameter, √area of the maximum inscribed circle, was introduced to account for both the size and shape of the volumetric defects, which exhibited an improved correlation with life compared to existing parameters. Fatigue cracks initiated from the surface and internal defects exhibited distinct crack growth behavior. Using the distinct material constants for surface and internal defects obtained from the Shiozawa plots, a novel fatigue life prediction model was proposed, which accounted for the distinct crack growth rates for surface and internal defects. Additionally, a probabilistic approach was utilized to determine the size of the largest possible defect for a given probability of failure. The fatigue design curve estimated with this defect size, using the proposed crack growth model, has been shown to be a more accurate estimate compared to the design curves developed using conventional approaches for fatigue qualification. Furthermore, the cross-platform transferability of the structure-property relationships, specifically defect-fatigue, was assessed for L-PBF Ti-6Al-4V. The transferability of defect-fatigue relationships established for the specimens fabricated on the EOS M290 platform was tested and validated with those fabricated on the Renishaw RenAM 500Q platform. Fatigue crack growth parameters derived from the EOS M290 specimens could be successfully applied to the fatigue life prediction of the RenAM 500Q specimens. The validity of fatigue design curves constructed using the short crack growth model with parameters from the EOS M290 platform was verified with data from the RenAM 500Q.