PSI - Issue 68

J.A. Ziman et al. / Procedia Structural Integrity 68 (2025) 1159–1165 J.A. Ziman et al. / Structural Integrity Procedia 00 (2025) 000–000

1164

6

It can be stated that the CATs carried out conventionally for validation purpose agree precisely with the calculated StressLife results. In addition to a significant increase in fatigue life at higher testing frequencies, a similar progression of the curves in the range of 10 5 and 10 6 can be observed. Furthermore, the fatigue strength estimated from the curves ( σ a from which no failure occurs up to a specified limit load cycle number of 10 7 cycles) correspond to the estimated fatigue strength from the LITs ( σ a from which the first M occurs), Dengel and Harig (1980). 3.2. Residual stress measurements

Fig. 5. (a) Degradation of surface residual stress at different fatigue life stages for frequencies f = 80 Hz and 260 Hz; (b) Cyclic deformation curves for constant amplitude tests (CAT) with frequencies f = 80 Hz and 260 Hz.

Residual stress measurements were conducted to analyse the frequency influence by correlating the residual stress relaxation with the results of cyclic deformation curves from CATs. Based on the estimated number of cycles to failure from the StressLife-curves obtained in 3.1, two load levels for the test frequencies (250 MPa for 80 Hz and 280 MPa for 260 Hz) were selected for the respective partial fatigue tests. For this reason, three fatigue stages were defined and approached (5% N f , 15% N f , 50% N f ) before the specimens were tested until failure. As a result of the material depended scattering, the actual fatigue life stages were recalculated based on the corresponding numbers of cycles to failure. Since the evaluated residual stress values σ ES in axial and tangential directions are almost identical, only the axial values for the respective fatigue stage are presented in Figure 5a. Figure 5b shows the cyclic deformation curves based on the change in temperature-corrected electrical resistance of the CATs already shown in 3.1. While the maximum of the cyclic softening at 80 Hz occurred demonstrably before 50 % N f , it is transferred to higher numbers of cycles in case of 260 Hz. This observation is also consistent with the σ ES reduction, which is more pronounced at 80 Hz compared to 260 Hz in earlier stages of fatigue life. This is caused by the higher plastic deformation at lower testing frequencies and the associated longer periods for dislocations to reorient themselves in the material, whereby the residual stresses in the material can be reduced to a greater extent. The cyclic hardening of the material at 80 Hz, which takes place over a long cycle range of around 10 5 cycles, is also reflected in the low σ ES relaxation after 50 % of N f . In contrast, the cyclic hardening of the specimens at 260 Hz takes place over a significantly reduced cycle range (approx. 3∙10 4 to 4∙10 4 cycles). In general, competing processes must be considered. On the one hand, higher test frequencies reduce the dislocation mobility, while the frequency-induced high temperatures facilitate the mobility of dislocations. Depending on the frequency, one of these two effects may dominate, resulting in different deformation behaviour for different testing frequencies.