F4-Vitry 1 120 Failed (N ¼ 7:70x106 cycles) Wheel seat at about 10 mm from the T-transition
F4-Vitry 2 108 Run-out (N ¼ 107 cycles)
F4-Vitry 3 120 Run-out (N ¼ 107 cycles)
F4-Vitry 4 132 Failed (N ¼ 4:98 × 106 cycles) Wheel seat at about 10 mm from the T-transition
F4-Vitry 5 120 Failed (N ¼ 2:10 × 106 cycles) Wheel seat at about 10 mm from the T-transition
F4-Minden 1 132 Failed (N ¼ 107 cycles) Wheel seat at about 4 mm from the T-transition
F4-Minden 2 126 Failed (N ¼ 107 cycles) Wheel seat at about 15 mm from the T-transition
F4-Minden 3 120 Run-Out (N ¼ 107 cycles)
F4-Minden 4 132 Failed (N ¼ 107 cycles) Wheel seat at about 7 mm from the T-transition
36 S. Foletti et al. / International Journal of Fatigue 86 (2016) 34–43
2. The proposed method
The procedure is based on a finite element (FE) analysis so as to obtain the stress path under the press fit seat. The stress tensor is used as input in a multiaxial high cycle fatigue (HCF) criterion to determine an equivalent stress:
req ¼ f .HCF criterion; .rij .; X; Y . ð1Þ
as a function f of the selected multiaxial HCF criterion and the stress tensor .rij . at a specific location in the press fit seat identified by
the coordinates X and Y.
The so obtained equivalent stress is checked against the fatigue limit of the material expressed as a function of defect size through the El-Haddad correction, [18]:
sffiffiffiffiffiffiffiffiffiffipffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
where c is the crack depth of the inclined crack length along the critical plane predicted by the multiaxial fatigue criteria (Fig. 5b).
2.1. Multiaxial high cycle fatigue criteria
Two different multiaxial high cycle fatigue criteria (the Dang Van criterion and the Liu–Mahadevan criterion) have been used in the proposed procedure. In literature, applicability of the selected criteria have been discussed several times for the analyt- ical solution of rolling contact and fretting type of fatigue failures [9,21,22].