13th International Conference on Fracture June 16–21, 2013, Beijing, China -6- (a) Hysteresis area under 20 Hz tests (b) Hysteresis area for several tests conditions Fig.6. Evolution of stress-strain hysteresis loop’s area Figure 6(b) allows a comparison of hysteresis loop’s area value between 0.2 Hz, 2 Hz, 20 Hz and 140 Hz fatigue tests at three different stress amplitudes. One can see clearly that hysteresis loop’s area is dependent of the fatigue test as the smaller is the frequency, the higher becomes hysteresis energy. In addition, primary cyclic softening behavior is also more severe when frequency decreases. (a) Comparison Monotonic and CSSC (b) Determination of cyclic hardening exponent Fig.7. Comparison of CSSC under 2 and 20 Hz Let us pay now a particular attention on the cyclic stress-strain curve (CSSC). Figure 7(a) allows a comparison of monotonic and cyclic stress-strain curves from fatigue tests at 2, 20 and 140 Hz. One can find that cyclic stress-strain data are lower than monotonic ones. This behavior is in accordance with general cyclic softening already mentioned[4]. In addition, if we consider only plastic strain of hysteresis loops as in Fig. 7(b), the cyclic hardening exponent is equal to 0.162, 0.135 and 0.147 for 2, 20 and 140 Hz fatigue tests respectively. One can note that theses values are close to the usual estimation 0.15 for most of metallic materials[5], regardless its initial conditions. Even though CSSC from 20 and 140 Hz fatigue tests are really close one to each other, one can see that for a similar stress amplitude, total strain amplitude and particularly plastic strain amplitude decreases when the testing frequency is increasing. This fact is of course correlated with the decrease of hysteresis energy already highlighted.
RkJQdWJsaXNoZXIy MjM0NDE=