PSI - Issue 7

Ana D. Brandão et al. / Procedia Structural Integrity 7 (2017) 58–66 Author name / Structural Integrity Procedia 00 (2017) 000–000

62

5

prepared with a layer thickness of 30 µm (energy input of 50 J/mm³). These imperfections act as stress concentration points, hindering the fatigue properties on the AlSi10Mg specimens (Romano et al. (2017); Tang and Pistorius (2017)). The effect of the layer thickness is more pronounced in the milled samples, as for the net shaped samples the surface finishing has the strongest effect on the fatigue properties. The number of cycles until failure was significantly lower (2 to 3 orders of magnitude) for the net shaped specimens, independently of the layer thickness used (ID 7 to 12). This high dependence of the fatigue behaviour on the surface condition is well known (Brandl et al. (2012); Aboulkhair et al. (2016); Mower and Long (2016); Greitemeier et al. (2016)). The higher roughness is often described as the root cause for this trend, as it might allow earlier crack initiation from the free surface, working as a stress concentration region. This work shows that different post-process treatments, such as jet blasting (ID 9 to 11) or vibratory polishing (ID 8) result into very similar fatigue performances for the same process parameters. However, when the contour parameter (illustrated in Fig. 1b) was excluded the fatigue behaviour is significantly improved to levels similar to the ones exhibited by the milled specimens. To the best of our knowledge the influence of this parameter in mechanical properties has never been studied. As such, further investigation is needed in order to understand the impact of this change in the scan strategy during the AM process, in the fatigue properties as well as in the microstructure of the AM samples. In addition, it is observed that the selected heat treatment (applied only to the samples produced with a layer thickness of 90 µm, ID 1 and ID 5) has a detrimental effect on the fatigue properties on the AlSi10Mg specimens, at the stress level tested. This behaviour is contradictory to what is commonly reported in literature (Brandl et al. (2012); Maskery et al. (2015); Aboulkhair et al. (2016)). In the present investigation the performed heat treatment had the objective of a stress relief which is used to avoid part deformation due to residual stresses. However, this also lead to a decrease in the static yield strength of the alloy from 244 MPa to 165 MPa in the heat treated specimens (EOS GmbH -Electro Optical Systems (2015)). As such, the test was carried out at a stress amplitude close to the yield strength, resulting in a larger plastic deformation when compared to the non-heat treated milled specimens, hence the lower number of cycles endured. In this work, the results showed that the build orientation does not have a strong effect on the fatigue properties of this AlSi10Mg alloy, tested at 120 MPa, for the manufacturing parameters used. The anisotropy effect observed in this study is not as pronounced as it is typically presented in literature, where the samples produced horizontally show an improved mechanical behaviour compared to the vertically manufactured ones (Brandl et al. (2012)). When comparing the best results obtained in this work (machined samples manufactured with 30 µm layer thickness, ID 2) with the fatigue performance of common Al 6061 wrought alloy, it is seen that the specimens of this work offer an improved fatigue behaviour. At the stress level of 120 MPa, the specimens of ID 2 withstand 2 × 10 7 cycles, against to approximately 2 × 10 6 cycles of the Al 6061 wrought alloy (Mower and Long (2016)).

Fig. 3. Fatigue resistance for all the samples tested per parameter combination described in Table 1. Testing was performed at a stress amplitude of 120 MPa.