PSI - Issue 54

365 5

Nikolai Kashaev et al. / Procedia Structural Integrity 54 (2024) 361–368 Kashaev et al. / Structural Integrity Procedia 00 (2023) 000 – 000

d d = [ 1− 1− Δ ] (1− Δ ℎ Δ ) .

(1) In Eq. 1 N is the number of loading cycles and the parameters C , n , and p are material constants that depend on the environment and the applied conditions. The variables f , R F ,  K th , and  K are the crack closure function, load ratio, threshold of the stress intensity factor range, and the stress intensity factor range, respectively. The values for estimation of the crack closure function f were taken from the study of Biswal et al., 2019 and the detailed procedure for the calculation is described by Newman, 1984. Stress intensity factor range ∆ K was calculated by Murakami’s equation (Murakami, 2002) as shown in Eq. (2): ∆ = ∆ √ √ , (2) where ∆ σ is the applied stress range, √area is the square root of the projected area of the defect, and parameter C M is equal to 0.5 for internal defects and 0.65 for surface defects. To account for short crack growth from defects, the approach already proven in the case of laser-welded Ti-6Al 4V joints (Fomin et al., 2018a) was followed in which the El Haddad equation (Eq. 2, El Haddad et al., 1979) was integrated into the NASGRO equation (Eq. 1): Δ ℎ =Δ ℎ, √ Δ + 1 Δ + 2 . (3) In Eq. 3 the parameter  K th, LC is the threshold of the stress intensity factor range determined in the case of the growth of long fatigue crack. The parameter  a is the fatigue crack extension and a 1 and a 2 are fit parameters. In the present work, the same procedure described in the work of Fomin et al., 2018a was applied to determine the parameters a 1 and a 2 . The derived values of a 1 and a 2 are 1.2 μm and 2.43 μm respectively. The fracture mechanics framework was applied to calculate the number of cycles until failure N f by integrating the Eq. 1: =∫ 1 [ 1− 1− Δ ] (1− Δ Δ ℎ ) d , (4) where the initial crack length a i for both internal and surface crack initiation cases was determined using fracture area analysis, in which the area of the defect, area was measured. For internal defects a i was set as = √ / and for surface defects a i was set as = √2 × / (Murakami ’s approach, Murakami, 2002). The values of the crack length at failure were set as specimen thickness (3 mm) and the distance of central of the defect to the nearest surface, in the case of surface and internal crack initiations, respectively. 4. Results and discussion 4.1. Fatigue behavior The fatigue test results are shown in Fig. 5(a). In the range of finite fatigue life between 10 4 and 10 6 loading cycles a typical fatigue behavior can be observed, whereas close to the fatigue limit (number of cycles of 10 7 ) a high scatter is present. Thus, run-outs were detected in the range of maximum stress levels of cyclic loading between 300 MPa and 450 MPa. The values for fatigue limit are lower than those reported for the plasma arc-based WAAM-fabricated Ti-6Al-4V structures in as-build conditions (500 MPa - 600 MPa, Syed et al., 2021a) and thus are in the typical range of those for the cast material (Léopold et al., 2014). The reason for the lower fatigue limit values in the present study could be due to the fact that there were not only single small pores in the WAAM structures as it was the case in the study of Syed et al., 2021a, but also some irregular lack-of-fusion defects were present. According to these authors, a detrimental effect of selected oscillating deposition strategy is not expected in current samples, since the effect of using oscillation or parallel passes was only observed in fatigue samples tested along the horizontal orientation.