PSI - Issue 34

D. Rigon et al. / Procedia Structural Integrity 34 (2021) 154–159 D. Rigon and G. Meneghetti / Structural Integrity Procedia 00 (2021) 000–000

155

2

α shape factor to evaluate the Mode I stress intensity factor (SIF) α R , β R , γ R , δ R R-dependent coefficients of the empirical model Eq. (1) ΔK th,LC,est,(R) estimated threshold SIF range of long cracks for a given load ratio R by means of Eq. (1) ΔK th,LC,exp,(R) experimental threshold SIF range of long cracks for a given load ratio R AM Additively Manufacturing CM Conventionally Manufactured materials EST El Haddad Smith Topper model LOF Lack of Fusion defect LEFM Linear Elastic Fracture Mechanics 1. Introduction The structural durability of AMmaterials is typically lower than CM ones due to the different microstructure, tensile residual stresses and the presence of porosities as well as Lack of Fusion (Li et al. 2016; Lewandowski and Seifi 2016; Meneghetti et al. 2017; Rigon et al. 2018; Mooney et al. 2019; Chern et al. 2019; Carneiro et al. 2019; Kan et al. 2019; Braun et al. 2021). The presence of defects is generally regarded as the main cause of lower fatigue properties thus their influence in terms of size and distribution plays an important role on the fatigue assessment of AM parts (Zerbst et al. 2021; Murakami et al. 2021). When dealing with fatigue of defect, classical approaches such as the Murakami model ((Murakami 2019), the El Haddad Smith Topper model (El Haddad et al. 1979a, b) or its extension proposed by Atzori, Lazzarin, and Meneghetti (Atzori et al. 2003, 2005) support the estimation of fatigue thresholds of materials affected by a certain defect or short crack. The correct application of the above-mentioned models require two material parameters, i.e. the fatigue limit of the defect-free material and the threshold range of the stress intensity factor for long cracks ΔK th,LC . In particular, ΔK th,LC can be determined precisely by means of dedicated experimental tests described in the relevant standards (ASTM E647 or ISO 12108:2018); however, its experimental evaluation requires time-consuming destructive experimental testing and expensive machining processes to produce the standard specimens. Estimating models from the literature are based on dislocation theory, crack tip plasticity, and environmental conditions that require the determination of physical parameters hardly obtainable during the engineering design phase. Other models are based on parameters such as tensile strength, yield strength and microstructural length resulting from standard experimental tests or micrograph examination (Taira et al. 1979; Ritchie 1979; Llorca and Sanchez-Galvez 1987; Ravichandran and Dwarakadasa 1991; Bartosiewicz et al. 1993; Herold et al. 2000; Chapetti 2011). However, these models are affected by a quite large approximation for the load ratio (R) equal to -1 and/or involve material parameters difficult to evaluate. Recently (Rigon and Meneghetti 2020) calibrated an empirical model to estimate the ΔK th,LC for R = -1 that requires only the Vickers hardness HV and a material-dependent microstructural length, l . Afterward, the authors extended the empirical model for different R values (Rigon and Meneghetti 2021). The aim of the present work focusses on calibration steps performed for fitting the empirical model on the experimental data taken from the literature relevant to AM as well as CM materials. 2. Calibration of the ΔK th,LC for different load ratios The experimental SIF thresholds ΔK th,LC,exp have been collected for the CM and AM materials and summarised in (Rigon and Meneghetti 2020, 2021), to which the reader is referred. Regarding the definition of the microstructural l parameter, the idea that it is the mean free path between (insurmountable) obstacles to slip-band transmission was followed according to Yoder (Yoder et al. 1983) for wrought metals; similarly, for AM materials, it was defined starting from the observation reported in the original papers of the interaction between fatigue cracks and microstructure in the near-threshold regime (Rigon and Meneghetti 2020), as summarized in Table 1.

Table 1. An example of a table.