PSI - Issue 53

S. Senol et al. / Procedia Structural Integrity 53 (2024) 12–28 Author name / Structural Integrity Procedia 00 (2019) 000–000

18

7

The three-point bending fatigue (3PBF) tests with a stress ratio (R) of 0.1 and a frequency of 30 Hz were conducted on an Instron E10000 machine with a 10 kN dynamic load cell. For each stress level and condition, 3 samples were tested. The tests were run at room temperature under constant amplitude axial loading till fracture or run-out, and the run-out condition was specified as 2 million cycles without failure. The 3PBF sample dimensions (Beretta et al., 2020; Ordnung et al., 2022) and the test setup are shown in Fig. 1(a) and Fig. 1(c), respectively. The applied load ( F ) levels were calculated using the Equation 1 and Equation 2 where; moment of inertia ( I ) (Equation 2) and bending span ( L ) were constant, while height ( h s ) and width ( w ) were measured for each sample. The stress ( σ ) levels were selected by the authors considering the quasi-static properties ( σ yield : 407 ± 10 MPa). The applied stress amplitudes were namely, 124, 160, 212, 265, and 338 MPa, corresponding to 30, 40, 52, 65, and 83 % of the yield strength, respectively. ܨ ൌ ఙ ೘ೌೣ ൈ ଼ ൈ ூ ௅ ൈ ௛ ೞ (1) ܫ ൌ ௪ ൈ ௛ ೞయ ଵଶ (2) The stress amplitude ( σ a = (( σ max - σ min )/2) and number of cycles to failure (N f ) were used to construct S-N curves. Finally, fractured surfaces of samples tested at the same stress level (265 MPa) were analysed by scanning electron microscopy (SEM) (FEI-Nova NanoSEM 450). 3. Results Four different surface conditions, namely, AB (as-built), R (dL-PBF processed), EDM (wire electric-discharge machined) and M (milled), are studied. The effect of different surface treatments on surface roughness, concomitant stress concentration factor, surface residual stresses, and hardness are evaluated as their local variation, specifically at the RoI, where the samples experience the highest stresses during three-point bending fatigue loading, will affect the structural integrity of the part during dynamic loading. 3.1. Effect on surface and sub-surface characteristics First of all, the average surface roughness parameters ( Ra and Rv ) and the critical stress concentration factor ( k t ) for the different surface conditions namely, AB, R, EDM, and M, are listed in Table 1. The highest surface roughness is observed in the case of AB samples. Following the dL-PBF treatment, a 73% lower surface roughness is recorded for the R samples as compared to AB samples, demonstrating the beneficial effect of the dL-PBF surface treatment. However, it should be emphasized that although the surface roughness of the dL-PBF surface-treated samples is comparable to the roughness values of the reference EDM samples, it is still slightly higher as compared to M samples, and the milled (M) surface exhibits the lowest surface roughness. The critical stress concentration factors ( k t ) of the varying surface conditions also follow the same trend (Table 1), where the highest k t is calculated in the case of AB surface condition, while it is greatly lowered (> 60%) for R, EDM, and M. Table 1. The average surface roughness parameters ( Ra and Rv ) and the critical stress concentration factor ( k t ) calculated from the surface profiles measured by tactile method for four surface conditions: as-built (AB), dL-PBF processed (R), wire electric-discharge machined (EDM), and milled (M). Surface condition AB R EDM M Ra (µm) 19.1 ± 1.6 5.1 ± 0.4 5.4 ± 0.1 2.9 ± 2.1 Rv (µm) 76.1 ± 7.6 25.1 ± 10.8 24.5 ± 1.2 16.9 ± 7 k t 3.5 ± 0.2 1.3 ± 0.1 1.1 ± 0.0 1

The maximum principal surface residual stress values ( σ I ) with a tensile nature are measured for AB (178 MPa) and R (161 MPa) samples. Considering the high sensitivity of the measurement method to the sample positioning and