PSI - Issue 76

Christina Mamagkinidou et al. / Procedia Structural Integrity 76 (2026) 82–88

87

In Fig. 5, the distances of the crack-initiating inclusions from the surface of the steel sheet, as well as the inclusion sizes in terms of √ , are correlated with the residual stress profile in rolling direction. This is the relevant residual stress profile, since the loading direction during fatigue testing was parallel to the rolling direction of the material. The analysis reveals that the majority of crack-initiating inclusions are located within a narrow region around the middle of the sheet thickness. Notably, this region corresponds well with the zone exhibiting distinctive tensile residual stresses. It is further apparent that the residual stress profile obtained by incremental correction (green diamond symbols) provides the best correlation with the observed crack-initiation sites. As discussed above and illustrated in Fig. 1(b), performing a fracture-mechanics evaluation without mean-stress correction yields highly non-conservative predictions of the fatigue strength in the VHCF regime. The findings depicted in Fig. 5 further highlight the need for an appropriate mean-stress correction that considers the presence of locally varying residual stresses. Future work will, therefore, focus on quantifying the mean-stress sensitivity of the investigated material and correlating the local residual stress at each crack-initiation location with the applied stress amplitude.

Fig. 5. Correlation of the distribution of crack-initiation location and the residual stress profiles in loading (equiv. to rolling) direction.

4. Summary and conclusions In the present study, very high cycle fatigue test results obtained with a precipitation-hardened martensitic stainless steel sheet were evaluated with respect to the influence of localised residual stresses. Residual stress profiles were measured using X-ray diffraction, and the obtained values were corrected considering the incremental material layer removal during the measurement. Furthermore, fracture surfaces were analysed to identify the location of crack initiating defects. It could be demonstrated that VHCF failure originated solely in areas with high tensile residual stresses. This emphasises the importance of an appropriate correction of residual stress profiles measured by incremental material layer removal for the accurate prediction of the fatigue strength.