PSI - Issue 53

Andrea Zanichelli et al. / Procedia Structural Integrity 53 (2024) 3–11 Author name / Structural Integrity Procedia 00 (2019) 000–000

7

5

Finally, the fatigue assessment is performed through Equation (1), by employing the stress components acting on the critical plane at the verification point, P cr , which is located on the critical plane at a distance from the hot-spot equal to twice the average grain size of the material, 2d . 3. Experimental campaign The proposed methodology described in Section 2 is validated by considering an experimental campaign taken from the literature, carried out by Wang et al. (2021) on additively manufactured AISI 316L stainless steel subjected to cyclic loading. All the specimens were additively manufactured by means of the selective laser melting technology and then subjected to a post-manufacturing heat treatment in order to significantly reduce both the microstructure anisotropy and inhomogeneity. Finally, each specimen was machined from the AM annealed rods by means of a conventional lathe, in order to guarantee a high level of surface finishing. The mechanical properties after the above manufacturing process were experimentally determined, that is: elastic modulus, E = 191 GPa; Poisson's coefficient, 0.3 ν = ; yield stress, y σ = 380 MPa; and ultimate tensile strength, u σ = 642 MPa. The fatigue properties were also evaluated: the fully-reversed normal and shear stress fatigue strengths were found to be , 1 af σ − = 249 MPa and , 1 af τ − = 216 MPa, respectively, in correspondence of a number of loading cycles to failure 6 0 2 10 N = ⋅ , and the slope of the S-N curves under fully-reversed normal and shear stress were m =-0.065 and m* =-0.031, respectively. Moreover, as far as the material microstructure is concerned, the average grain size, d , can be computed from the values reported by Im et al. (2019), by exploiting the procedure reported in the work by Vantadori et al. (2022a). In the present case, a value of d =48 mm has been obtained. Four different specimen types were manufactured and subjected to fatigue testing, that is: plain specimens, specimens with a sharp notch (root radius r =0.07 mm), specimens with a circular notch (root radius r =2 mm), and specimens with a blunt notch (root radius r =5 mm). For each specimen type, pure axial, pure torsional and biaxial (that is, combined axial-torsional) fatigue tests were performed under force/moment control. In the case of biaxial fatigue loading, both in-phase (phase shift angle 0 β = ° between axial and torsional loading) and out-of-phase (phase shift angle 90 β = ° ) configurations were tested. The above tests were characterised by a fatigue loading ratio, R , equal to either 0 or -1, and different stress levels were considered. 4. Results and discussion This Section deals with the results determined by applying the proposed methodology (described in Section 2) to both plain and notched specimens (described in Section 3) subjected to cyclic tests. More precisely, the results related to the plain specimens are discussed in sub-Section 4.1, whereas sub-Section 4.2 is devoted to those related to the notched specimens. Note that the present analytical methodology was recently proposed by the present authors (Vantadori and Zanichelli (2021), Vantadori et al. (2022b), Vantadori et al. (2022c), Zanichelli and Vantadori (2021)) for traditional metallic components subjected to fretting fatigue, and it is here extended to the case of AM components. In this regard, some remarks are needed: (a) the examined additively manufactured AISI 316L stainless steel can be treated as homogeneous and isotropic due to the post-manufacturing heat treatment process and, consequently, the Carpinteri et al. criterion (Carpinteri et al. (2015), Vantadori et al. (2020), Vantadori et al. (2021a)), that is, a stress-based fatigue criterion originally devised to perform the fatigue assessment of traditional metallic materials, can be employed; (b) the tensile residual stresses generated during the AM fabrication process are not taken into account in the present analysis since they are completely relaxed due to the above heat treatment; (c) the influence of the surface roughness on the fatigue behaviour of the specimens analysed is here neglected, since a high level of surface finishing was ensured by using a conventional lathe.