Crack Paths 2012

It can be clearly seen that for the test with the longest coplanar propagation, kI,max

does not reach the ModeI threshold, even if it exceeds the Dang Van’s locus in Fig. 8.a.

(a)

ModeI threshold

ModeIII R C Fthreshold

a=480

a=117Pm N=1,000,000 cycles P m N=750 cycles

h

/K I t

m a x

K I I I

a=0 P m N=2,000,000 cycles

í0.5

í2

í1.5

í1

í0.5

0

0.5

1

KImax/KIth

01.5012

(b)

Figure 9. Analysis of O O Ptests in terms of SIF’s: a) cylindrical coordinate system

along the crack; b) kI,max-kIII,max

plots for the the out-of-phase tests onto the gear steel.

So it appears that for Vh < 0 the fatigue limit criterion is not able to catch the physics of

the phenomena, which is on the other hand very clear from Fig. 9.b. It can be seen that

there are two different limits: a ModeI limit described by 'KI,TH and a ModeIII limit

that for R C Fapproximately corresponds to 0.5 'KI,th for the gear and bearing steel.

By adopting this approach it is possible to identify the fatigue limit as the

superposition of two conditions (corresponding to the two failure mechanisms):

'kI,max<'KI,th for ModeI and 'kIII,max < 'KIII,th for co-planar propagation is then also

possible to correctly catch the dependence of fatigue limits on crack size (see Fig. 10).

25