Fatigue Crack Paths 2003

the stress intensity factor Δ K was calculated as a function of crack length, Δ K= ΔK(a).

Finally, the results of fatigue test were plotted as the dependences v = v(ΔK). The results

are shown in Fig. 1. In the case of the tube (Fig. 1a) it can be seen that a special heat

treatment (1000°C / 2 h + oil quench + 600°C / 160 min) to produce a fine Cr-Fe-(C)

precipitate in the alloy has no influence on fatigue failure – the fatigue crack growth rate

is the same for the both type of specimens. Fromthe comparison of both graphs in Fig.1

it follows that material of the plate is more resistant against fatigue failure then material

of the tube. Fatigue crack growth rate in the plate is lower than in the tube (e.g. for Δ K=

30 MPa.m1/2, vtube ≈10⋅vplate). The threshold value of Δ K for the tube is lower than for

the plate (ΔKth)tube ≈ 14 MPa.m1/2 and (ΔKth)plate ≈ 18 MPa.m1/2.

b) fatigue crack propagation (a = 2 m m )

a) area of crack initiation (a = 0.5 m m )

c) fatigue crack propagation (a = 4 m m )

d) area of final static rupture

Figure 2. Fracture micromorphology at different crack lengths.