PSI - Issue 7

Kentaro Wada et al. / Procedia Structural Integrity 7 (2017) 391–398 K. Wada et al./ Structural Integrity Procedia 00 (2017) 000–000

395

5

There were no non-propagating cracks at the fatigue limits of R = −5, − 3 and − 1 ( cf. Fig. 3 (a)). In contrast, a non propagating crack was detected at the R of − 10, with a stress amplitude, σ a , of 900 MPa, and a total crack length, 2 a , of 535 μ m, as exhibited in Fig. 3 (b). The Δ K th and K max th calculated at each stress ratio are plotted in Fig. 4. The Δ K th was seen to increase with a concomitant decrease in the stress ratio. It is noteworthy to mention that the K max th bottomed out at an R of − 5, while the K max th at the R of − 10 turned out to be about twice as high as K max th at the R of − 5. Fig. 5 presents the FCG rate as a function of 2 a . At the stress level close to the fatigue limit, FCG initially decelerated at a certain crack length, subsequently accelerating with an increase in crack length. The occurrence of such phenomena indicates the development of crack closure during the FCG process, even in such a high-strength steel.

2 a = 535 μm

B

A

B

A

100 μm

100 μm

A

A

B

Non-propagating crack

B

No crack

50 μm

50 μm

σ a = 351 MPa σ max = 117 MPa N = 1.0 × 10 7

σ a = 900 MPa σ max = 164MPa N = 3.0 × 10 7

(a) R = − 5

(b) R = − 10, 2 a = 535 μm

Fig. 3: Optical micrographs of the notch after fatigue testing at the fatigue limit. A non-propagating crack of 2 a = 535 μm was found at R = −10, but was not found at R = −5 .

Fig. 4: Threshold stress-intensity factor range and maximum stress-intensity factor for crack propagation with respect to the stress ratio.