Session: MF-01-01 The Noel O'Dowd Memorial Symposium on Fracture: Application of Fracture Mechanics in Failure Assessment

Paper Number: 155776

155776 - A Heuristic Approach to a Defect-Based Fatigue Life Estimation via Optical Microscopy and Fatigue Crack Growth Rate Data 

Abstract: 

Despite the progress additive manufacturing has made into delivering near-net shape components with minimal defects, the presence of gaseous pores, lack-of-fusion and keyhole pores, and material inclusions present a pressing challenge to the prediction of a component’s fatigue life. Depending on the defect location, its influence in fatigue life ranges from negligible to overwhelming, increasing the scatter associated with fatigue even further, posing difficulties during the design phase of components.

Interesting work has been performed to improve fatigue life estimation through the use of computerised tomography (CT) to map out existing defects, detecting the killing defect based on the defect volume and surrounding stress field, and applying a fatigue crack growth law. While this approach is valuable, its application is dependent on the size of the sample; furthermore, use of CT scanning equipment is expensive.

Hence, the chosen approach in this work used optical microscopy imaging along different cross-sections of the same component to parametrise statistical distributions for both the defect size and location. This information is then coupled with the stress field the part will be subjected to, and taking advantage of a Monte Carlo methodology, the most probable size, location and associated stress intensity factor is estimated. Additionally, completing the analysis with experimental fatigue crack growth rate data of the same material, and varying the stress field to several levels leads to a virtual SN curve, which can be validating through experimental data.

This approach is validated in this work, with optical images from additively manufactured 18Ni300 Maraging specimens used in both compact tension fatigue crack propagation and bending SN curves from the same material and stress ratio (R=0.1).

Presenting Author: Jorge Gil INEGI

Presenting Author Biography: Jorge Gil is a PhD student working on fatigue characterisation of additively manufactured metals. He is based in Porto, Portugal, and is pursuing his PhD at Faculdade de Engenharia da Universidade do Porto.

Authors:

Jorge Gil INEGI
Beatriz Silva IST
Maria Vaz IST
Ana Reis FEUP
Abílio De Jesus FEUP