For metallic alloys, most of the fatigue to defects interactions are analyzed considering high cycle fatigue conditions with lifing methodology at initiation. Recent progress for in situ observation of fatigue damage makes propagation analysis more accurate. This has been successfully applied for lost foam casting Al-alloys for low cycle fatigue condition using computed tomography (CT) experiments. CT has been used for both analyzing crack initiation and propagation from natural porous defects in situ at 250 °C  and for comprehensive study of standard fatigue crack growth tests using post-mortem observations .
The proposed paper deals first with a systematic measurement of morphology and location of defects interacting with the major crack during SENT fatigue crack propagation test achieved at 250 °C. Optical in situ observation yields surface crack to microcrack interaction measurements. CT has enabled in this case to derive associated geometrical criterion for crack to defect interaction for LCF condition.
Secondly, based on experimental observations, a systematic analysis of crack to idealized spherical defect has been achieved. This study is based on FEA, conform remeshing technique to model crack propagation, coalescence of cracks and elastoviscoplastic evaluation of mechanical state. G-theta method is used to evaluate FCGR in different geometrical configurations. This point yields an emphasis of the role crack closure associated to both strain localization and defects.
Finally, an explicit computation of realistic crack growth together with 3D mesh of the SENT specimen including meshing of defects has been developed. This helps for discussing the relative impact of different features described above in the crack to defect interaction for LCF loading.
 Dezecot, et al. (2017), Acta Materialia, 123, 24-34.
 Dezecot, et al. (2019), Materials Science and Engineering: A 743: 87-97.
 Abecassis, et al. (2019), International Journal of Fatigue, 118, 209-224.
The authors thank the French National Research Agency for funding this study through the projects INDIANA (grant ANR-12-RMNP-0011) and SEMAFOR (grant ANR-14-CE07-0037), PSA Group, SAFRAN Group and the French National Association for Research and Technology.