PSI - Issue 38

Michaela Zeißig et al. / Procedia Structural Integrity 38 (2022) 60–69 Zeißig, Jablonski / Structural Integrity Procedia 00 (2021) 000–000

67

8

5. Results and discussion Two different approaches describing the fatigue strength in the presence of defects have been shown. Both approaches do not account for the interaction of defects, which, however, might be present in SLM specimens. In the DSG approach, defects are incorporated via their defect size and geometry. The required data for irregular defects needs to be obtained via FE calculations. In this way, also the defect orientation with respect to loading direction can be adjusted. In Vincent, Nadot-Martin et al. (2014), it is stated that the DSG approach led to good results in the defect size range of 100 µm to 1000 µm. Therefore, the approach should be assessed regarding smaller defect sizes. As only independent defects are assessed, the most critical defect in terms of shape and size determines the fatigue limit. The Weibull distribution as applied above incorporates the defect distribution via the Weibull exponent. However, Yusuf et al. (2017) report areas of higher pore concentrations in parts made of 316L thus an entirely random distribution seems not to be the case. Adjustments regarding further local defect information may be included via the local fatigue strength. As has been shown, the Weibull approach can be used under consideration of residual stress data. Here again, FE models are a useful extension and it is shown that residual stresses significantly influence the fatigue limit. For the DSG approach a corresponding extension may be found in Nadot and Mendez (2008). Further extensions towards the incorporation of the surface roughness have been shown for the Weibull distribution and may be done for the DSG approach e.g. via surface defects as applied in Nadot et al. (2020). Moreover, the microstructure can also be included as shown in Vincent, Nadot et al. (2014).

Table 2. Comparison of required and optional input parameters for DSG and Weibull approach.

DSG Weibull Required input 3 different fatigue results and defect size Local fatigue limit and Weibull exponent Explicit defect geometry Implicit consideration of porosity Fatigue criterion Fatigue criterion Optional input Residual stresses Residual stresses Surface condition Surface condition Microstructure Microstructure

For both approaches, fatigue evaluation with respect towards a specific failure mechanism via a corresponding fatigue hypothesis is possible. Obtaining all required material data seems to be more difficult for the DSG approach. The determination of fatigue inducing defect geometries and sizes might be difficult from fracture surfaces and the material parameters for defect-free material might just not exist. However, a solution to the latter problem is described in Nadot and Mendez (2008). As discussed, both approaches allow for the incorporation of the most detrimental factors on fatigue as summarized in Tab. 2. As for now, the results shown above merely demonstrate fundamental applicability of the approaches and highlight some of the influencing parameters and sensitivities of the approaches. However, due to the lack of consistent experimental data, incorporating e.g. complete information on the building parameters, the results lack experimental validation and thus a direct comparison of the results is not possible. 6. Conclusions and outlook Two concepts regarding the assessment of fatigue limits of SLM specimens have been shown with focus on 316L. The DSG approach as well as the Weibull distribution seem very suitable for the assessment of fatigue of SLM