158
A. Cruzado et al.
Fig. 16 Experimental results and numerical predictions of the number of cycles for fatigue crack
initiation in Inconel 718 alloy at 400 ◦ C as a function of the strain range ε normalized by the
minimum cyclic strain range ε min . (a) For R ε = −1 and ASTM 8.5 grain size. (b) For R ε = 0 and
ASTM 8.5. (c) Idem as (a) for ASTM 3 grain size
The fatigue model fitted with the fine-grained LCF experiments was then used to
predict the fatigue life of the coarser microstructure, ASTM 3. The results obtained
are represented in Fig. 16c. The fatigue initiation predictions are in general very
accurate, with only a slight underestimation of life for strain ranges in the middle of
the range explored. It must be noted that this material microstructure has not been
used to fit the fatigue life initiation and the effect of grain size is a direct consequence
of the differences in the plastic response.
Figure 17 presents together the results of the simulations obtained for the two
grain sizes considered. It can be observed that the model is able to capture the grain
size effect in crack initiation observed in the experiments; for the largest values
of the cyclic strain range (where the accumulated plasticity is homogeneous), the
scatter is low, and the fatigue lives were independent of the grain size. Nevertheless,
at the lowest strain ranges, where the localization of strain in the most favorable
A. Cruzado et al.
Fig. 16 Experimental results and numerical predictions of the number of cycles for fatigue crack
initiation in Inconel 718 alloy at 400 ◦ C as a function of the strain range ε normalized by the
minimum cyclic strain range ε min . (a) For R ε = −1 and ASTM 8.5 grain size. (b) For R ε = 0 and
ASTM 8.5. (c) Idem as (a) for ASTM 3 grain size
The fatigue model fitted with the fine-grained LCF experiments was then used to
predict the fatigue life of the coarser microstructure, ASTM 3. The results obtained
are represented in Fig. 16c. The fatigue initiation predictions are in general very
accurate, with only a slight underestimation of life for strain ranges in the middle of
the range explored. It must be noted that this material microstructure has not been
used to fit the fatigue life initiation and the effect of grain size is a direct consequence
of the differences in the plastic response.
Figure 17 presents together the results of the simulations obtained for the two
grain sizes considered. It can be observed that the model is able to capture the grain
size effect in crack initiation observed in the experiments; for the largest values
of the cyclic strain range (where the accumulated plasticity is homogeneous), the
scatter is low, and the fatigue lives were independent of the grain size. Nevertheless,
at the lowest strain ranges, where the localization of strain in the most favorable