PSI - Issue 75

Andrew Halfpenny et al. / Procedia Structural Integrity 75 (2025) 234–244 Author name / Structural Integrity Procedia (2025) where and are the through-thickness notch stress concentration profiles for membrane and bending stresses, respectively, and is the residual stress profile. It is worth noting that this also allows beneficial peening or stress relief processes to be considered. Despite through-thickness notch stress concentration profiles can be found in the literature for common geometries, it is usually recommended to determine the notch stress concentration profiles using a detailed breakout FE model of the joint or cracked component of interest. For relatively simple geometries the calculation of membrane and bending notch stress concentration profiles is straightforward. However, when dealing with more complex geometries, fully isolating the two stress components might prove difficult. To address this issue, a procedure to obtain and profiles using the stress profiles from two arbitrary non-collinear load cases has been developed and implemented in nCode ‘ WholeLife ’ software package. 4. Application of the ‘Total - Life’ Method and correlation with tests An evaluation and validation of the ‘ Total-Life ’ method was carried out by the SAE Fatigue Design and Evaluation Committee (2018), as also described by Cordes et al. (2019). The validation activity was carried out by comparing the fatigue lives obtained by testing A36 steel welded and machined T-shape specimens with the same geometry (Fig. 8a) to the corresponding fatigue lives predicted using nCode ‘ WholeLife ’ analysis engine. In particular, a total of 17 welded and 15 machined specimens were tested under 7 different load histories, including constant amplitude loading at various R-ratios, block loads and variable amplitude loading. It is also worth mentioning that no special weld fatigue properties were used in the simulations and that the analyses for both types of specimens were exactly the same, with the only significant difference being that the residual stress profile in the welded components was measured by the SAE and applied only to the corresponding simulations. The comparison of the fatigue lives predicted using ‘WholeLife’, , to the corresponding experimental fatigue lives from tests, , for both welded and machined (Fig. 8b) specimens showed good agreement between simulations and tests with 0.5≤ / ≤2 for most of the welded and machined components. 243 10

Fig. 8. Validation of the ‘Total - Life’ method: (a) w elded (back) and machined (front) T-shape specimens and (b) comparison between predicted (‘WholeLife’) and experimental fatigue lives for welded and machined specimens .