PSI - Issue 38

Kimiya Hemmesi et al. / Procedia Structural Integrity 38 (2022) 401–410 Author name / Structural Integrity Procedia 00 (2021) 000 – 000

409

9

Alternatively, the fatigue strength can be estimated using the non-linear analysis method of FKM (2019). It requires stresses and strains at a particular location in the component to be calculated by means of elastic-plastic FEA or be estimated using an appropriate engineering approach. The resulting stress and strain amplitudes, and , and the mean stress, , are combined to calculate the fatigue damage parameter RAM = { √( + ) at ( + ) ≥ 0 0 at ( + ) < 0 . (4) In Eq. (4), RAM represents a modified parameter according to Smith et al. (1970) including a mean stress correction factor k . The stress amplitude and the mean stress are defined in terms of the equivalent Mises stress with an algebraic sign corresponding to the sign of the hydrostatic stress. The mean stress correction factor is calculated as = { (2 + ) at ≥ 0 3 (2 + 3 ) at < 0 (5) with the parameter given by Eq. (2). Considering the fatigue strength values determined in CAL tests without overloads (Table 4) and the residual stresses calculated by means of FEA (Fig. 6), further numerical analyses were performed to estimate the fatigue strength in presence of overloads. Thereby, the stress amplitude was iteratively varied until the resulting RAM value reached the level for the specimens with no OL, at = 10 6 . The fatigue strength values estimated in such a way together with the corresponding RAM values are reported in Table 6. Fig. 7 compares the fatigue strength for the two materials determined from specimen tests with and with no overloads versus respective analytical estimates, using both measured (FKM, 2020) and numerically calculated residual stresses (FKM, 2020 and FKM, 2019). Both FKM Guidelines predict decreasing fatigue strength with increasing overload level, being in contradiction to the experimental findings demonstrating a non-monotonic effect of the overload level on the fatigue strength. The latter increases at OL 1, whereas OL 2 results in decreasing fatigue strength. Nonetheless, the experimental data suggest that limiting the maximum amplitude to 75 % of the calculated static strength does not reduce the fatigue life and, thus, confirm the validity of the respective limiting criterion currently employed in FKM (2020).

800

240

fatigue tests FKM (2020) - RS as measured FKM (2020) - RS from FEA FKM(2019) - RS from FEA

42CrMoS4

fatigue tests FKM (2020) - RS as measured FKM (2020) - RS from FEA FKM(2019) - RS from FEA

EN AW-6082 T6

700

200

600

160

500

400

120

300

80

200

fatigue strength in MPa

fatigue strength in MPa

40

100

0

0

no OL

OL 1

OL 2

no OL

OL 1

OL 2

Fig. 7. Estimates of the fatigue strength after overloads.

7. Discussion The experimental results demonstrate a complex effect of overloads on the fatigue strength of the two materials considered. Comparing to the test results at CAL with no overloads, a moderate overload level (OL 1) increases fatigue strength. However, this beneficial effect cannot be explained by means of stress analyses alone, as tensile residual stresses are induced by the overload cycles. The enhancement of fatigue strength in this case can likely be attributed to two different mechanisms. One of them is the material strain hardening by which an increase in the flow stress and