PSI - Issue 57

Rando Tungga Dewa et al. / Procedia Structural Integrity 57 (2024) 762–771 Dewa et al./ Structural Integrity Procedia 00 (2019) 000 – 000

766

5

= ( ̇)

(2)

Where ε p is strain rate, M and n are constants describing the regression fit (intercept and slope), respectively. The plastic strain values were derived from the mid-life of the cycles to failure. We assume that the mid-life is best considered as stabilised cycle for LCF testing at RT condition. From Fig. 4a, the lower strain rate test had a higher plastic strain. Furthermore, for LCF test with the influence of time-dependent factor, the time to failure compared to that in terms of cycles to failure must be considered and can be illustrated in the following formula: = ( ̇) (3) Where t r is time to failure in terms of hour, X and y are constants describing the regression fit (intercept and slope), respectively. Fig. 4b shows the time to failure with strain rates plotted in logarithmic function. Although, the cycles to failure is low for lower strain rate, the time to failure is showing a higher level. The equation generated from the time to failure function with strain rates for fitting interpolation is also provided in the figure. From the result, the time-dependent crack mechanism is found as the dominant fatigue mechanism for the alloy 617. The power law function seems to be well-fitted to the experimental data in terms of the linearity agreement, indicating that the predicted life coincides well with the measured LCF life. This can be taken into account that the time-dependent deformation can take place gradually with a slow strain rate fatigue. Accordingly, the above prediction techniques are verified with the predicted life, which is derived from following the back substitute of the experimental to the equations. Based on Eq. (1) and (3), the error accuracy has been estimated for 12.6% and 13.7%, respectively. Those results are still within factor of 1.0 for conversion and confirm the linearity which was shown in Figs. 3-4. 4. Influence of Strain Rate on LCF Properties The cyclic stress response behaviors at different strain rates of the Alloy 617 are shown in Fig. 5. In Fig. 5a, all materials show similar trend with initial hardening phase, followed by cyclic softening and rapid drop of stress or failure. This is typical cyclic stress response behavior of superalloy under LCF loading at RT. However, based on our result, a significant lower strain rate test condition decreased the strength of the material, which can be seen in the lower stress range response. Furthermore, Fig. 5b shows their cyclic stress responses in order of normalised cycle. The materials exhibit a short initial hardening up to 5% of their fatigue life. The initial hardening is occurred owing to the increasing dislocation density of fatigue slips in material reconciliation. the dislocation movement increases with increasing number of cycles as well, when the dislocations multiplication rate is the same of the dislocations annihilation rate, thus, the cyclic stress is stabilized. Accordingly, material shows significant cyclic softening gradually (almost for their entire life) until the initiation of macrocrack where the stress suddenly dropped. The cyclic softening phase of this material may be due to the annihilation of dislocations that exceeds the dislocation multiplication, thus the cyclic stress is decreased. It is also noticed at lower strain rate that the dislocation mobility decreased and therefore, resulting in a lower stress range. Form the figure, also it can be seen that the macrocrack initiation for specimen with the fastest strain rate occurred earlier. The cyclic crack propagation is much higher for a high strain rate specimen. To quantify the degree of cyclic softening during LCF, the softening ratio is defined as the ratio between the peak stress after initial hardening ( S max ) and the stress at sudden drop point ( S c ). Fig. 6 shows the degree of softening phase at different strain rates. It can be seen from the figure that the trend of degree of softening increases generally with increasing in strain rate. The cyclic softening phase has a lager influence at higher strain rates due to the fact that they have more cycles to failure and accumulated plastic deformation rather than lower strain rate specimens. This result confirms the previous finding that at higher strain rate that the dislocation mobility increased with a higher stress response and vice versa.