PSI - Issue 38

100 Christophe Grosjean et al. / Procedia Structural Integrity 38 (2022) 94–108 C.Grosjean and al. / Structural Integrity Procedia 00 (2021) 000 – 000 7 This means that in the present configuration, the improvement of the internal geometry with an S-shape has strongly reduced the pressure drops, even though the roughness in the L-PBF part (Ra ≈ 5,5 µm in the channel) was significantly higher than in the machined part. a b

Fig. 6. (a) Hydraulic bloc in the pressure-drop test bench. (b) Comparison of the pressure drop results for AM part and machined part.

7. Fatigue tests All tests were performed at room temperature, with ISO VG46 oil, and at a frequency of 8 Hz in sine form (see Fig. 7(a)). The initial specification was to achieve 2 million cycles with a pressure range (Pmax - Pmin) of 465 bar which was on average achieved for tests 3, 4 and 5 (see Table 4). The other pressures tested are severe configurations compared to the usual use of hydraulic blocks. Each initiation site was located (see Fig. 7(b)) and fractography with a scanning electron microscope was performed. No clear defects from the process, such as lack of fusion, were observed.

Table 4. Results of fatigue tests.

Pmin [bar]

Pmax [bar]

Rz [µm]in failure area

Nb cycles

Observation

Reference

90  m 90  m

120

1200

13 330 21 000 386 393

Not broken but issues with seals. Test continued on the same part between 90 and 900 bars

1

90 65 30 30 30 90 65 65

900 650 495 495 495 900 650 650

2 3 4

3  m 3  m 3  m

1 953 537 2 475 592 5 300 000

- -

Not broken part. Test continued on the same part between 90 and 900 bars

5

65 000

Not broken part but issues with seals (hypothesis of part deformation)

6 7

60  m

174 579 333 370

250  m