PSI - Issue 75

Martin Edgren et al. / Procedia Structural Integrity 75 (2025) 555–563 Martin Edgren et Al. / Structural Integrity Procedia (2025)

559 5

3. FEA and fatigue life estimation FEA is deployed to evaluate the strain range drop at the notch by estimating the shape and depth of an emanating crack. The solid model was discretized using tetrahedral elements (SOLID187 in ANSYS). The quadratic formulation of these elements is well suited for irregular meshes, such as those representing crack geometries in the stiffness model. A symmetry boundary condition was applied along the mid-plane of the specimen (Figure 4, left side of the modelled half-specimen), while the applied load was introduced at the opposite end (Figure 4, right side of the modelled half-specimen) to simulate the stress distribution. Each design point was analysed in two stages: first, using the full model, followed by a sub model with refined meshing in the region of interest. The mesh size of the sub model was set to 0.25 mm to ensure sufficient resolution around critical features. Based on the results of the DoE, the deployed stiffness model, and input from strain gauge distance and offset measurements, the stop criterion for strain range drop was set to 5%. The crack depth based on FEA at 5% strain range drop across the different specimens are evaluated to be in the range of 0.4-0.6mm, see Figure 4.

Figure 4 a) Workflow FEA – evaluating mean strain over modelled strain gauge resistor strip (gauge area), b) Strain range drop evaluated for different strain gauge positions and crack depths The FAT class for Structural Detail No 525 is used to calculate the fatigue life based on the nominal stress method. Furthermore, fatigue life is calculated using the FAT class for HFMI treated details according IIW recommendations [13]. The evaluation is performed at the 50% failure probability FAT curve. Table 3 is showing results from the calculations with the different fatigue life calculations. Table 3. Calculated life based on nominal stress approach. FAT class Fatigue Life (N) Nominal, FAT50, 50% failure probability 103000 Nominal HFMI, FAT90, 50% failure probability 269000 4. Experimental procedure The experimental procedure consists of two main phases: pre-fatigue preparation and fatigue testing to failure. Initially, specimens are machined with predefined mild notches to promote crack initiation at a predefined position. Prior to fatigue testing, the specimen surfaces are treated via mild sandblasting to enhance the accuracy of subsequent laser scanning. This scanning process captures the geometry and position of the notches, establishing a baseline for notch shape and depth. Strain gauges are installed adjacent to the notches to monitor crack initiation and growth. While no standardized distance exists for placing strain gauges relative to the crack when using the strain range drop method, prior studies have shown a consistent correlation between crack depth and strain range drop, provided the gauges are positioned at least 1 mm from the weld toe [20]. In this study, strain gauges are placed 2mm from the notch. Specimens are then subjected to fatigue testing under a maximum load of 200kN, with a load ratio (R) of 0.1 and a frequency of 19 Hz. The testing system is programmed to automatically stop when a 5% drop in strain range is