Extreme value analysis of the equivalent flaw size for fatigue limit estimation in additively manufactured Ti6Al4V specimens with as-built surface condition

This work presents a preliminary investigation into the estimation of fatigue limit in Ti6Al4V specimens produced via Laser Powder Bed Fusion of metals (PBF-LB/M) with surfaces in as-built conditions, by combining fracture mechanics with surface and volumetric defect characterization. Building upon a previously proposed approach that uses Extreme Value Statistics (EVS) to estimate the deepest micro-notch of PBF-LB/M as-built surfaces, this study presents a preliminary comparison between the estimated deepest notch from optical profilometry (OP), as a function of sample size, and the actual deepest surface notch measured across the entire outer surface of the gauge section of the specimen. In addition, internal and subsurface defects were characterized by X-ray computed tomography (CT) to provide a preliminary assessment of their relevance to fatigue behavior. The results show that EVS provides reliable estimates of the maximum depth, with less than 5% error for sufficiently large sample sizes. Finally, both surface roughness and subsurface defects led to similar effective crack sizes and fatigue threshold estimates, suggesting their comparable roles in early crack initiation for the considered material.