PSI - Issue 38

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

404

4

2.3. Overload procedure For each material, three series of fatigue tests were carried out under cyclic tension-compression, at a stress ratio of = −1 , using the notched specimens shown in Fig. 1. The first test series was performed at constant amplitude loading (CAL), without overloads (OL), to obtain basic S-N curves. Subsequently, two test series with different overload levels, referred to as OL 1 and OL 2, were conducted according to the following procedure: – First, 5 overloads were applied as tension-compression cycles, with the last OL half-cycle corresponding to a compressive load. – Afterwards, the specimen was loaded by regular CAL cycles until failure or runout at 10 6 cycles. The low number of overloads (5) was selected to represent rare events in component’s service life which may occur due to abnormal service conditions, malfunctions or operating errors, and do not significantly contribute to the fatigue damage sum, but rather lead to localized plastic deformations at potential crack initiation positions. The overload levels were specified as follows. OL 1 was defined by 75 % of the specimen static strength analytically estimated according to FKM (2020). The latter document considers the OL 1 level as such that does not reduce the fatigue strength. Another and more severe overload level, named as OL 2, was set to 75 % of the specimen static strength determined experimentally by testing three notched samples of each material. Table 3 provides a summary of the specimen geometric parameters, the respective stress concentration factors (SCF), the static strength estimates according to the OL 1 and OL 2 definitions, and the resulting stress amplitudes at overloads. Note that the stress values specified in the table refer to the notch stress obtained by linear-elastic calculations.

Table 3. Static specimen strength and overload levels (notch stress values estimated by linear-elastic calculations).

Geometry

Static strength in MPa

Overload level in MPa

Net cross- section in mm²

Notch radius in mm

SCF --

According to FKM (2020)

Experimental from tensile tests

Material

OL 1

OL 2

42CrMoS4

12.56 50.27

1.0 1.0

1.75 2.22

1919

3114 1159

1440

2340

EN AW-6082

688

520

870

2.4. S-N curves In the fatigue tests, two characteristic material properties were derived: i) S-N curves in the range of about 10 4 to 10 6 load cycles, ii) estimates of the fatigue strength at 10 6 load cycles, denoted by a,10 6 . In the first case, the specimens were tested until fracture and, thus, a normal regression analysis was possible to determine the mean S-N curves. In the second case, the staircase test procedure according to Dixon and Mood (1948) was applied. Thereby, the specimens were tested up to a maximum number of = 10 6 load cycles and either failed at < 10 6 (rupture) or did not fail (runout). The fatigue strength was then obtained as the weighted mean value of the stress amplitudes in the respective tests, as suggested in Hück (1983). An example of how the staircase tests were performed and evaluated is given in Fig. 3. The results of all fatigue tests, S-N curves and fatigue strength values a,10 6 , without overloads, at OL 1 and OL 2 are summarised in Table 4 and Fig. 4.

Table 4. Fatigue strength of notched specimens from experimental results. S-N curve - = B ⋅ ( a / B ) − Type a,OL in MPa B B in MPa a,10 6 in MPa w/o OL -- -- 10 6 379 4.9 473 Overload (5 load cycles)

Material

42CrMoS4

OL 1 OL 2

1440 2340

-1 -1

10 6 10 6 10 6 10 6 10 6

511 402 141 163 118

6.7 4.7 4.6 5.2 4.0

588 540 172 173 133

w/o OL

--

--

EN AW-6082

OL 1 OL 2

520 870

-1 -1