PSI - Issue 57

Grégoire Brot et al. / Procedia Structural Integrity 57 (2024) 53–60

58

6

G. Brot et al. / Structural Integrity Procedia 00 (2023) 000–000

Fig. 5. E ff ect of a tensile prestrain of ε pl ≈ 4 % prior to self-heating testing: initial heating speed S heat , i as a function of stress amplitude.

4. Discussion

The used experimental methodology could be improved in order to reach slightly higher σ a . First the radius of 5 mm in Figure 1 should be increased and the number of cycle per load steps should be decreased when tested at f load = 800 Hz. However, loading steps duration should be large compared to the duration needed to reach the set stress amplitude. Ti-6Al-4V-LPBF has a homogeneous microstructure and its behavior is almost elastic perfectly plastic at room temperature. Therefore, under high-cycle fatigue loading, plasticity appears only in the vicinity of stress risers like pores in Ti-6Al-4V-LPBF. For defect-free material like wrought Ti-6Al-4V or P 0 920 ◦ C, SH is very limited until microplasticity became significant in the matrix material. Thus, their second SH regime appears for high stress am plitudes and is observed for few loading steps. For P 0 920 ◦ C, fatigue limit estimation ( σ SH , lim = 650 MPa) is close to the fatigue resistance in the VHCF regime: 630 MPa in Brot et al. (2023). However the one for wrought Ti-6Al-4V ( σ SH , lim = 650 MPa) is much higher than fatigue resistance in the VHCF regime: 400 MPa in Morrissey and Nicholas (2005). For material grades for which fatigue crack was porosity-induced ( P 1 920 ◦ C, P 2 920 ◦ Cand P 1 650 ◦ C), fa tigue crack initiation appears at stress amplitudes lower than the one for which microplasticity becomes significant in the matrix material and only the first SH regime is observed. This is consistent with the fact that two SH regimes are observed in P 1 1020 ◦ C grade, which contains pores but large grains and for which fatigue crack initiation is microstructure-induced (Brot et al. (2023)). Microplasticity has to be significant before the appearance of long fatigue cracks in P 1 1020 ◦ C. However, for P 1 1020 ◦ Cgrade, σ SH , lim is higher than fatigue resistance in the VHCF regime: ≈ 350 MPa in Brot et al. (2023). These results are in accordance with the one of Bustos et al. (2023) for which tem perature increment was not significant for non-HIPed Ti-6Al-4V-EBM. In most cases, fatigue limit estimation using self-heating testing seems not to be appropriate for Ti-6Al-4V tested at room temperature. Prestrain seems to have a first-order e ff ect on the dissipation intensity. Munier (2012) and Mareau et al. (2012) also observed on steel an higher dissipation after a prestrain but without a strong change of σ SH , lim . Mareau et al. (2012) attribute this to the high density of dislocation after the prestrain that enhanced some anelastic mechanisms such as the oscillation of pinned disclocation. In our results, only the second regime seems to be a ff ected by the prestrain. This one should have reduce the resistance to microplasticity. Possible damage of the material during prestrain of ε pl ≈ 4 % could also have a ff ected the SH behavior, indeed elongation at break for annealed Ti-6Al-4V-LPBF is not much higher than the prestrain: 7 to 13 % (Vilaro et al. (2011), Su et al. (2021)).