PSI - Issue 38

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

408

8

6. Consideration of overloads in the FKM Guidelines According to FKM (2020), S-N curves for notched specimens have a unique slope of = 5 , and the component fatigue strength is estimated based on the material tensile strength, either assuming its value according to the material specification sheet or using experimental measurements. The influence of the component geometry, manufacturing process, mean stress, etc., are regarded for through several factors, which only affect the value of fatigue strength, thus shifting the S-N curve along the stress axis. As overloads induce residual stresses, RS , their effect on the fatigue strength can be taken into account via the mean stress correction. For this purpose, a mean stress sensitivity factor is defined in FKM (2020) by = ∙ 10 −3 ∙ MPa + . (2) Here is the tensile strength in MPa, and are material constants equal to = 0.35, = −0.1 for steel, and = 1.0, = −0.04 for wrought aluminium. Given the fatigue strength ,10 6 for specimens with no OL, the fatigue strength for specimens of the same geometry and material, however, undergone some overloads and thus containing residual stresses, can be estimated from the relation a,10 6 ,OL = ,10 6 − ⋅ RS for 1 − −1 ≤ RS a,10 6 ≤ 1 + 1 . (3) Eq. (3) predicts decreasing fatigue strength with increasing tensile residual stresses. To assure a consistent comparison of fatigue strength estimates due to overloads according to FKM (2020) and FKM (2019), as well as to compare estimated and experimentally determined values, the subsequent calculations are performed assuming ,10 6 = 473 MPa for 42CrMoS4 and ,10 6 = 172 MPa for EN AW-6082, respectively (cf. Table 4). Those values refer to the median fatigue strength for specimens with no overloads and no residual stresses. Thereby, residual stresses due to machining are neglected, since they are rather low as compared to the residual stresses due to overloads. Considering the tensile strength according to Table 1, Eq. (2) yields = 0.329 for 42CrMoS4 and = 0.347 for EN AW-6082, respectively. Then, the fatigue strength for specimens subjected to overloads can be calculated using Eq. (3) together with residual stresses either experimentally measured (see Table 5) or derived by means of the FEA (see Fig. 6). The calculation results based on FKM (2020) and the respective residual stress values are summarised in Table 6. The latter also includes the values of the fatigue strength determined from the test results on specimens subjected to OL 1 and OL 2.

Table 6. Fatigue strength for notched specimens subjected to overloads: experimental measurements vs. estimates.

Estimated based on FKM (2020) RS as measured RS from FEA RS in MPa a,10 6 ,OL in MPa RS in MPa a,10 6 ,OL in MPa RAM in MPa a,10 6 ,OL in MPa Estimates based on FKM (2019) RS from FEA

Measured values a,10 6 ,OL in MPa

Overload type

Material

42CrMoS4

OL 1 OL 2 OL 1 OL 2

588 540 173 133

217 226

421 418 150 129

334 638

363 263 146

424 424 154 154

417 355 147

EN AW-6082

63

76

125 86 Besides residual stresses, overloads give rise to fatigue damage and, thus, may decrease the component lifetime. The amount of fatigue damage can be estimated following Miner's rule, see Palmgren (1927) or Miner (1945). Considering the five overloads at OL 1 or OL 2, the respective fatigue damage for the steel specimens is found to be = 0.003 or = 0.16 , respectively. For the aluminium alloy, the corresponding values are = 0.002 at OL 1 and = 0.03 at OL 2. Thus, only the OL 2 has a measurable influence on the calculated fatigue life, which agrees with experimental results for the alloy EN AW-6082. In all other test series, the fatigue strength increases due to overloads, which effect can obviously not be treated by the damage accumulation approach employed in FKM (2020). 346 52