PSI - Issue 17

Jürgen Bär et al. / Procedia Structural Integrity 17 (2019) 308–315 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

312

5

temperature increase of 0.4 K can be observed whereas in the 50 th cycle the temperature nearly no temperature increase is resulting within the loading cycle.

0.8

100 150 200 250 300 350 400 450 500

0.0 0.2 0.4 0.6 0.8 1.0 1.2 temperature change D T [K] R = -1

400

T meas

T diss

stress

R = 0

T meas

T diss

stress

0.6

300

0.4

200

-0.4 temperature change D T [K] -0.2 0.0 0.2

100

0

-1.2 -1.0 -0.8 -0.6 -0.4 -0.2

-100

stress [MPa]

stress [MPa]

-200

-0.6

-300

0 50

cycle 1

cycle 50

cycle 1

cycle 50

-0.8

-400

200.5

201.0 249.0

249.5

250.0

200.5

201.0

249.5

250.0

time [s]

time [s]

Fig. 4. Resulting temperature changes during cyclic loading of AA7475 T761 with the stress ratios R=0 and R=-1.

In case of the aluminum alloy 7475 T761 loaded under pure tension with a maximum stress of 425 MPa and under fully reversed loading conditions with 375 MPa no significant differences in the run of the temperature changes are visible (Figure 4). It should be noted at that at a stress ratio of R = -1 a higher loading was not possible because the specimen buckled in compression. The run of the total temperature change and of T diss are comparable with those measured on copper under pure tension loading. The total temperature changes are higher compared to copper which can be attributed to a higher thermoelastic effect caused by the higher loading level and a higher thermoelastic constant. The span of T diss is with about 0.18 K for R = -1 a bit higher than under pure tension loading with about 0.12 K. For the cycles presented in figures 3 and 4 a DFT according to equation (2) was performed. With the results of the DFT sine waves for the E- and D-Mode were calculated to enable a direct comparison with the results shown above. Figure 5 shows the corresponding results for copper. In the first cycle a distinct difference between the E-Mode and the calculated thermoelastic effect is visible. The amplitude obtained from the DFT is significant higher compared to that of the calculation according to equation (1). In the 50 th cycle still a slight difference can be seen. In addition to the differences in the amplitude, the result of the DFT also shows a phase shift in the thermoelastic effect. Again this difference is more pronounced in the first cycle. In case of the D-Mode and T diss also distinct differences in the curves are observable. Both waves exhibit the same frequency and nearly no phase shift between the curves is visible. Even if the strong increase in the first cycle is disregarded, the curves differ considerably in their shape. The rising edge in the run of T diss is significantly steeper, whereas the falling edge shows an exponential decline whereby clear differences between the two curves are visible. Especially in the first cycle the amplitudes of the two curves are also different. When again the steep increase to the first maximum of T diss is neglected, in the first cycle the span of T diss is found to be 0.35 K and for the D-Mode 0.5 K. In the 50 th cycle this difference is considerable smaller and both spans are below 0.3 K. 3.2. Real temperature changes compared with the Lock-In results