PSI - Issue 42

R. Fernandes et al. / Procedia Structural Integrity 42 (2022) 992–999 Fernandes et al. / Structural Integrity Procedia 00 (2019) 000 – 000

998

7

100 2



0.1 Total strain energy density (MJ/m 3 ) As-built T6 300ºC stress relief 250ºC stress relief 250ºC stress relief + HIP 1 -1 10 0 10 1 1 10

 W e +

 W P



As-built (fitted)

0.01 10 -2

1.0E+02 10 2

1.0E+03 10 3

1.0E+04 10 4

1.0E+05 10 5

1.0E+06 10 6

1.0E+01 10 1

Number of reversals to failure, 2N f

Fig. 5. Total strain energy density versus life relationship for the tested cases.

satisfactorily correlate the data for all material conditions (fitted for the as-built condition). This parameter, unlike the total strain amplitude (see Figure 4) seems to be relatively unsensitive to the post-processing treatments, which can be useful from a design perspective. The comparison of the different series plotted in Figure 5 shows that the more relevant differences are observed for the lower fatigue lives, where the crack initiation mechanisms are more relevant. The microstructure peculiarities associated with the different material conditions can explain this behaviour. Overall, the stress relief treatments increased the fatigue performance, in particular the standard process. On the contrary, the T6 heat treatment seems to be detrimental to fatigue strength. 4. Conclusions This article addressed the low-cycle fatigue behaviour of AlSi10Mg aluminium alloy manufactured by laser powder bed fusion for five material states, as-built condition, T6 condition, standard stress relief, stress relief at a non-standard temperature, and stress relief at a non-standard temperature followed by hot isostatic pressure. Low-cycle fatigue tests were performed at fully-reversed conditions under strain-control mode for strain amplitudes in the range 0.2-1.5%. The following conclusions can be drawn: • The cyclic stress-strain response of the tested aluminium alloy is sensitive to the post-processing treatment. The post-processing treatments changed the shapes of the hysteresis loops at the same the strain amplitude. Overall, the stress relief treatment slightly increased the fatigue performance; • The Ramberg-Osgood model can be satisfactorily used to describe the stable stress-strain response of the AlSi10Mg aluminium alloy manufactured by LPBF. The fitted functions are close to the mid-life hysteresis loops, either for ascending or the descending branches; • The strain-life relationships were affected by the material state. However, irrespective of the post processing treatment, the total strain versus life relationships were better captured via two-part piecewise functions. The HIP treatment did not introduce significant changes in fatigue performance; • The total strain energy density versus life datapoints were strongly affected by the post-processing treatment. The more relevant differences were observed for lower lives. Nevertheless, all cases were satisfactorily fitted by a power function determined from the results of the as-built condition. Acknowledgments This research is sponsored by FEDER funds through the program COMPETE – Programa Operacional Factores de Competitividade – and by national funds through FCT – Fundação para a Ciência e a Tecnologia – under the project