PSI - Issue 7

M. Tebaldini et al. / Procedia Structural Integrity 7 (2017) 521–529 M. Tebaldini et al./ Structural Integrity Procedia 00 (2017) 000–000

525

5

wider regions of the specimen where the stress is lower tha n 0.9 σ max . Following the “most stressed area” method, the identification of A was realized analysing the FEA results performed on the three sections of interest with CATIA solver. The FEA simulation of the bending fatigue test shows that the area where the stress ranges from 0. 9σ max to σ max is very limited, so also 0. 8σ max and 0. 7σ max were considered as lower limits. In Table 2 the largest defect size estimation detected for these considered area is summarized.

DEFECT SIZE COMPARISON

0 1 2 3 4 0 100 200 300 400 500 600 700 800 900 y

-2 -1

FERET[ μ m]

FERET A FERET B FERET C

Fig. 3: statistical defect population for the position A, B and C.

Table 2: Statistical parameters, larger defect size estimation and fatigue limit evaluation. LIKELIHOOD METHOD Statistical parameters Largest defect estimation Fatigue limit estimation δ ML λ ML = ( ) [µm] (19) σ w ’ (1) σ w ’’ (2) A Sample area (100%) 113.012 482.065 35.76 3.57 886 71 78 80 % 0.1458 -1.92 264 87 95 70 % 1.256 0.228 507 78 85

Sample area (100%)

26.3

3.27

364 334 615 630 420

68 84 76 77 83

74 92 83 84 90

B

173.895 631.036

80 % 70 %

0.182 0.9164

-1.703 -0.087

Sample area (100%)

3.09

C

76.907 392.776 22.08

90 %

1.428

0.356

3.2. Fatigue limit prediction Considering the material hardness values (94 HV, 95 HV and 97 HV for the position A, B and C respectively) and the maximum Feret diameter predicted by the extreme value analysis (Table 2) it is possible to predict the fatigue limit of the A356 alloy using equation (1) proposed by Noguchi (2007) and equation (2) presented by Ueno and reported by Tajiri et al. (2014).

′ = 1.43 ( + 120 × ) �√ � 1 / 6

Murakami modified: Surface defects

(1)