Crack Paths 2012

maximumstress and the fatigue life. Figure 3 shows the results of S-N relation for the

two groups of specimens, in which the datum points also symbolize the crack initiation

modeat surface or from interior of specimen. It is seen from Table 1 and Fig. 3 that, for

TT180 specimens subjected to R B cycling, the fatigue life (Nf) is from 104 to 4u108; the

S-N data present a duplex or step-wise tendency. For TT150 specimens subjected to U L

cycling at almost the same level of the maximumstress (860~880 MPa), the results

show a very large scattering of data distribution. The difference of the fatigue life is as

large as three orders of magnitude, ranging from 105 to 108.

Figure 4 shows the measurements of both F G Aand fish-eye sizes as a function of the

applied maximumstress. It is seen from Table 1 and Fig. 4 that the F G Asizes are

within a relatively small range between 40 and 100 P m and the fish-eye sizes are

distributed in a large range between 100 and 300 Pm.

Table 1. Measurement results of crack initiation parameters for two groups of

specimens with the data of applied maximumstress and fatigue life

(2aμInmc)

Speccodiemen

maxV (MPa)

N f

2(aμFmG)A 2afisheye

(μm)

323434805692.12138593

RB-12345678

875075308508

5.60×106 92 049 7 12353 7408089 8

14367670547686.54838390

160.5

121.9

108.3

206.2

160.2

164.2

163.6

300.3

UL-123456789

868680

4.53×105 7671 1065657 6 271 82301039 7

45432109287629.1519575

465453353454347.342185746

208.1

189.6

200.8

183.9

138.1

220.4

247.1

197.6

197.4

UL-11023

88680

1.02×107 51 542439 8

21390.2958

5350972.4658

155.9

95.5

247.0

151.3

The range of stress intensity factor at the periphery of inclusion ('KInc), F G A('KFGA)

and fish-eye ('Kfisheye) is calculated by using the following equation [14]:

0 . 5 a K a r e a V S ' '

(1)

908