Identifying and classifying killer defects and failure mechanisms in additively manufactured 17-4 stainless steel

The maturation of metal additive manufacturing (AM) technologies such as laser powder bed fusion (LPBF) has led to their relatively recent integration in application areas may which seek to improve design agility, achieve light-weighting, and lower product costs. Unfortunately, for highly stressed ‘fracture critical’ components, potential industrial adopters of metal AM must reckon with characteristically defect-ridden materials if they are to incorporate processes such as LPBF into their manufacturing workflows. While recent studies which characterize the tensile, fatigue, and fracture behavior of LPBF manufactured materials and frequently present fractographs to qualitatively support their findings, these data are often only a small proportion of all tested specimens. In this work, we present an exhaustive analysis which is a 100% sampling of all specimens from a large set (n = 45) of failed fatigue specimens. These 17-4 precipitation hardened stainless steel specimens were manufactured with optimized processing parameters, heat treated, and tested under high-cycle fatigue (HCF) loading conditions. As such, these analyses are a representative sampling of defects and failures likely to occur in a material with a realistic manufacturing pedigree for industrial applications. We identify the defects which initiate fracture, i.e., killer defects, and propose a set of identifiable features which can be used to categorize their manufacturing defect origin. The most prevalent defect types are identified as well as less common but equally important defects which smaller samplings may not capture.