PSI - Issue 77

Bruno Pedrosa et al. / Procedia Structural Integrity 77 (2026) 649–656 Author name / Structural Integrity Procedia 00 (2026) 000–000

651

3

2. Fatigue assessment The geometry of the bolt hole adopted in this study is illustrated in Figure 1a). The hole has a diameter of 18 mm, corresponding to a detail representative of a bolted shear connection with M16 bolts. In line with standard specifications, the washers used would have an outer diameter of 30 mm. Consequently, any crack initiating at the hole would only become visible once it exceeds 6 mm in length, measured from the hole’s edge. This corresponds to a “visible” plate width of 15 mm on each side. According to EN 1993-1-8 (CEN, 2005a), the minimum required distance from the hole center to the plate edge is 2 , =1.2× 0 . Since the adopted value 2 , =30 is greater than 1.2× 0 =21 , the geometric requirements are thus fulfilled. Specimens were prepared from a plate in compliance with EN 10025-2 (BSi, 2004). The bolt hole was produced through a four-step drilling process. Initially, a pilot hole was drilled to guide the operation and ease the following step, where a drill of intermediate size was used. Boring was then applied to achieve the target diameter with precision, while reaming was performed to ensure a high quality surface finish. Fatigue tests were performed using a hydraulic testing machine (Instron 8801, 100 kN capacity), as shown in Figure 1b. The experiments were conducted under tensile load control with a stress ratio of =0.1 . Five stress levels were considered. For each level, three tests were performed, except for ∆ 5 (the lowest stress level), where only one test was carried out. A test frequency of =10 was adopted, except for ∆ 1 (the highest stress level), where = 7.5 was used. These values were chosen as a balance between machine capacity, testing duration, and potential heating effects. The fatigue life, , was defined as the number of cycles until complete specimen failure. Experimental conditions and results are summarized in Table 1. Mean stress effects were accounted for by calculating the normalised stress range, ∆ , , following the procedure described in Pedrosa et al. (Pedrosa et al., 2024).

a) b) Figure 1. Fatigue tests of S235 JR bolt hole: a) geometry (dimensions in mm); b) experimental apparatus and hydraulic testing machine. Fatigue behaviour in structural elements is complex, as it depends on several interacting parameters that are not always independent under cyclic loading. For this reason, empirical approaches are often employed to define and model fatigue response (Bowman, 1997). Experimental investigations play a key role in addressing this challenge and in providing reliable fatigue life predictions. The most common method for representing fatigue damage in mechanical components or structural details is the S–N curve, also known as the Wöhler curve. These curves describe the relationship between the applied stress amplitude and the number of cycles to failure. In practice, fatigue tests rarely allow a clear distinction between the crack initiation and propagation stages, therefore S–N curves typically represent the total fatigue life (Benden et al., 2009). An S–N curve is derived from experimental data obtained at different stress levels. A logarithmic scale is usually adopted for both axes, enabling a linear relationship between log ∆ and log for most results. This relationship is mathematically expressed by the Basquin equation (Basquin, 1910), as shown in Eq. (1), where the slope of the line is = − 1/ . ∆ = (1)