PSI - Issue 38

Christophe Grosjean et al. / Procedia Structural Integrity 38 (2022) 94–108 C.Grosjean and al. / Structural Integrity Procedia 00 (2021) 000 – 000

105

12

Table 7. Correlation of fatigue calculation with test results (2)

Residual stress measurements taken into account

Pmin [bar]

Pmax [bar]

Possible crack initiation area

Rz [µm] in area

Nb cycles from fatigue calculation

Test reference

Experimental nb cycles to failure

No

4621

A B C D E F G A B C D E F G

90

Yes Yes

9 238 841

3

52 974 (Fig. 14) 205 837 (Fig. 15) 357 520 (Fig. 14)

2 6

386 393 174 579

60

No No

65 650

250

7

333 370

Yes

No failure

No

690 616 (Fig. 15)

60

Yes Yes Yes Yes Yes No

No failure No failure No failure

60 60

54 544

90

No failure

3 4

1 953 537 2 475 592

3

1 834 940 (Fig. 16)

60

No No

2 365 990 3 327 000 No failure 6 500 560 No failure No failure No failure

30 495

250

Yes

No

60

Yes Yes Yes

60 60

Critical areas found during the test all appear in the calculation as possible failure areas. The results are very sensitive to roughness and residual stress. The areas E, F and G are less critical than the other areas for a given loading. No specimens were broken in these areas. Despite to be the most critical area (without residual stress) , the failure doesn’t occur often in the A area. For all the other areas, the most likely hypothesis regarding the variation of the estimated life cycles is the residual stresses.

Fig. 13. Failure area A under 90-900 bars load