PSI - Issue 17

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

311

4

f S = 1000 Hz. Due to the low testing frequency of f L = 1 Hz no perfect adiabatic conditions are achieved, but with the high sample rate of 1000 Hz a good temporal resolution of the temperature change within a loading cycle was achieved. The size of each recorded frame was 160 x 128 pixel. Additionally, the force signal of the testing machine was transferred as an analog signal to the camera and stored with each recorded frame. The evaluation according to equation (2) was performed with a self-developed Matlab program. Due to the small rigid body movement no motion compensation algorithm was applied. 3. Results To enable a calculation of the temperature change T diss due to dissipative effects out of the measured temperature signal T meas the temperature change caused by the thermoelastic effect has to be subtracted. Therefore, in the first loading steps, which were performed under pure elastic loading, the thermoelastic constant Kc was determined. With this constant and the applied force amplitude the temperature change due to the thermoelastic effect was calculated and subtracted from the measured temperature signal according to equation (3): ( ) = ( ) − ∙ ∙ ∙ ⁡(2 ∙ ) (3) 3.1. Temperature change within a loading cycle

1.2

T meas

T diss

T meas

T diss

stress

R = -1

stress

R = 0

200

0.2

1.0

200

0.8

0.0 Temperature change D T [K] 0.2 0.4 0.6

100

0.1

-0.1 temperature change D T [K] 0.0

150

0

100

stress [MPa]

stress [MPa]

-100

50

-0.2

-200

-0.2

cycle 50

cycle 1

0

cycle 1

cycle 50

-0.4

200.5

201.0

249.5

250.0

200.5

201.0

249.5

250.0

time [s]

time [s]

Fig. 3. Resulting temperature changes during cyclic loading of copper with the stress ratios R = 0 and R = -1.

Figure 3 shows the results of the cyclic loading of copper with a maximum load of 225 MPa at stress ratios of R = 0 and R = -1. The images show the first (cycle 1) and the last cycle (cycle 50) of the loading block, respectively. For R = 0 a temperature change of about 0.35 K was measured. There is nearly no visible difference between the first and the 50 th loading cycle. The temperature changes due to dissipative effects are very small. The value of T diss is beginning to raise when the loading reaches about 180 MPa and after unloading to the mean stress a decrease until the end of the cycle takes place. A dissipation of energy takes only place near the maximum stress, the rest of the cycle is dominated by cooling effects. Within the 50 cycles no change of the behavior is observable. The loading with R = -1 reveals a totally different behavior. The maximum span in the temperature change is about 1.3 K in the first and nearly 1 K in the last cycle of the loading block. The temperature change due to dissipative effects is characterized by a cooling in the first part of the cycle. After the load reached about 180 MPa a steep increase of T diss until the load reaches the maximum is visible. During unloading a decrease of T diss takes place followed by another steep increase when the load reaches -180 MPa in compression. The increase lasts until the stress reaches the minimum and the rest of the cycle is determined by a decrease of T diss due to cooling. In the first cycle a total