PSI - Issue 42

3

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

994

Building direction

Fig. 1. Specimen geometry used in the low-cycle fatigue tests.

Table 1. Main mechanical properties of the tested material conditions.

Material condition:

As-built

Stress relief (300ºC) T6

Stress relief (250ºC) Stress relief (250ºC) + HIP

435 250

250 140 15.9

180 110 14.1

355 195

375 205

 UTS (MPa)  YS (MPa)

3.9

5.5

6.4

 f (%)

E (GPa)

72

69

70

72

73

550ºC for 2h, followed by aging at 175ºC for 6h); the third in a stress-relieved condition (300ºC for 2h); the fourth in a stress-relieved condition (250ºC for 2h); and the last in a stress-relieved condition followed by hot-isostatic pressure (250ºC for 2h under nitrogen gas pressure of 100 MPa). The two stress-relief temperatures selected in this research (i.e. 300ºC and 250ºC) are associated with the two exothermic peaks reported for this alloy when processed by LPBF at about 260ºC and 320ºC [14]. It has been found that the resulting microstructural peculiarities strongly affect its mechanical behaviour. In addition, the effect of HIP treatments on cyclic plastic response in this alloy has not yet been addressed. Low-cycle fatigue tests were conducted in a servo-hydraulic machine under fully-reversed conditions (R  = -1) at a constant strain rate (d  /dt = 8×10 -3 ) and room temperature. The tests were initiated in tension and were stopped when the total failure was reached for strain amplitudes (  a ) in the interval 0.2%-1.5%. The cyclic stress-strain response was acquired from a from a 12.5 mm-gauge extensometer clamped to the specimen, via two separated knife-edges, and were recorded in appropriate intervals, using a PC-based data acquisition system, with a rate of 200 samples per cycle. The main monotonic mechanical properties of the tested batches, namely the ultimate tensile strength (  UTS ), yield strength (  YS ), strain at failure (  f ) and Young’s modulus (E) are presented in Table 1. 3. Results and discussion The cyclic response of the SLM-manufactured AlSi10Mg aluminium alloy is significantly affected by the post processing treatment and by the strain amplitude. Figure 2 plots the strain amplitude against the number of cycles for the five tested conditions, i.e. as-built, T6, 300ºC stress-relief, 250ºC stress-relief, and 250ºC stress-relief and HIP. It is clear from the figure that the stress amplitude is almost constant throughout the entire test, irrespective of the material condition or the strain amplitude, which denotes limited transient cyclic effects for this alloy in the SLM manufactured state. It is also clear that the fatigue lives are relatively similar at the same strain amplitude and the maximum values, for these two strain amplitudes, were found for the stress-relief heat treatment. The as-built condition led to the maximum strain amplitudes over the tests while the T6 condition resulted in the lowest strain amplitudes. Another important outcome is that the 250ºC stress relief combined with the HIP treatment did not introduce relevant differences in the cyclic response when compared with the case subjected to 250ºC stress relief. Overall, the material exhibited a strain-softening behaviour, except for the T6 condition, which strain-hardened.