UC Davis

Additive manufacturing (AM) supports innovative engineering design by enabling the production of complex, high-quality parts on demand, but also introduces significant residual stress and distinctive grain morphologies that influence mechanical behavior. Specifically, AM presents challenges in reliability for fatigue critical applications where residual stress impacts measured crack growth rates for damage tolerant life predictions. Although the importance of these effects has been recognized, they are not well understood. Therefore, the aim of this work is to quantify and compare the influence of process-induced residual stress on measured crack growth rates of AM Type 304L austenitic stainless steel produced by directed energy deposition (DED) and laser powder bed fusion (PBF). Orientation dependence was evaluated by comparing crack growth parallel and perpendicular to the build direction. Alternating stress intensity factor (ΔK = Kmax - Kmin) tests, where K characterizes the stress field at the crack tip, were used to assess fatigue crack growth behavior in the near-threshold regime (<10-8 m/cycle). To connect process influence and fatigue performance, data were compared to an existing engineering reference material (annealed wrought Type 304/304L austenitic stainless steel). Fatigue crack growth rates in DED builds manufactured with identical process parameters were investigated. Macroscale residual stress and residual stress intensity factor (Kres) profiles of a secondary fatigue specimen were measured using the slitting method. Positive values of Kres led to higher fatigue crack growth rates compared to wrought material. While the slitting method provided an estimate of residual stress effects, an accurate means of quantifying Kres in individual (primary) specimens offers better insight into the fatigue crack growth behavior of AM material. The on-line crack compliance (OLCC) method was adopted to quantify Kres from data collected during a fatigue crack growth rate test. However, the published methodology does not clearly illustrate the process of determining Kres, so a validation study was undertaken using specimens fabricated from aluminum alloy 7050-T74. An improved approach to data analysis based the Schindler influence function was developed and applied to the fatigue test data of DED material. The Kres profiles from OLCC showed that DED specimens oriented for crack growth perpendicular to the build direction have larger values of positive Kres as compared to those with crack growth parallel. Correcting measured fatigue crack growth rates for the influence of residual stress using Kres caused data for both orientations to collapse into a single curve, indicating the primary difference in crack growth rates parallel and perpendicular to the build direction was due to tensile residual stress and that anisotropic grain morphology had a minor influence on the fatigue performance. PBF builds that were fabricated on different systems using similar process parameters were evaluated using the same fatigue crack growth testing and Kres data analysis methods. Initial residual stress measurements revealed higher values in the PBF as compared to DED. The OLCC method indicated similar values of Kres for both specimen orientations in the PBF material that were higher than those measured in either orientation of DED, leading to higher measured growth rates in the PBF material compared to DED at the same applied ΔK. Data for both orientations of PBF material demonstrated agreement prior to corrections for residual stress, consistent with the similarity of Kres. After corrections, fatigue crack growth rate data were similar in all PBF and DED specimens. Additionally, the corrected data of PBF and DED were consistent with data for wrought material that had been corrected for fatigue crack closure effects, indicating the differing amounts of residual stress was the primary contributor to the different apparent fatigue performance in the materials produced by the AM and conventional methods.