PSI - Issue 18

Andrzej Katunin et al. / Procedia Structural Integrity 18 (2019) 20–27 Author name / Structural Integrity Procedia 00 (2019) 000–000

24

5

clamp in order to prevent transferring vibrations to the sensor and its connection with the specimen. The force sensor and accelerometer were connected through a conditioning module to the multi-channel data acquisition card, which was connected to a PC and controlled by an own-developed application for this purpose. The used application allows controlling of the excitation signal parameters through the analogue output of multi-channel signal acquisition module and controls the shaker amplifier. Temperature measurements were carried out using the infrared camera. In order to perform surface cooling of specimens during fatigue testing two-stage side channel air blower was used. The airstream was directed to a specimen using a nozzle 3D-printed from PLA mounted at the end of an air hose. During performing tests the specimens were loaded in the fully reversed bending mode with an initial force of 110 N, corresponding to the initial stress of 52.8 MPa, and resulting the initial displacements of 2.21 mm, and with a constant loading frequency of 30 Hz. In order to investigate an influence of air cooling flow rate on the fatigue process, and, in particular, on appearance of the self-heating effect two distances of the nozzle from a tested specimen were assumed: 0.8 m and 2.4 m. These distances corresponded to the airflow rates of 5.5 m/s and 1.5 m/s, respectively, with an air temperature equalled to the ambient temperature of 27°C. The flow rate was measured by the thermoanemometer. In order to obtain the reference to the air-cooled specimens, additional tests without cooling were performed. The resulting temperature evolution curves obtained during testing with no air cooling as well as with air cooling with various cooling rates are presented in Fig. 4. As it can be noticed, the experimentally obtained self-heating temperature history curves coincide with the phenomenological model presented in section 2. For the cases when air cooling was present the self-heating temperature stabilization is observed, and this interval of a stabilized self heating temperature is longer for the higher airflow rate. These history curves correspond to the curve no. 1 in Fig 2. In the conditions of lack of air cooling the constant self-heating temperature growth is observed in the second phase of the self-heating temperature curve, which corresponds with the curve no. 2 in Fig. 2.

Fig. 4. Self-heating temperature history curves for the cases with and without air cooling.

One can also observe that the presence of air cooling, and its rate, influencing on the extension of the residual life of tested specimens, and this extension is significant (comparing two extreme cases the difference in total life is almost doubled in the case of air cooling). Due to heat removal from the specimens surfaces by application the air cooling the degradation and final failure is governed by mechanical fatigue process. In order to evaluate the observed changes the AE structural responses were analyzed. The total counts of hit-cascade envelopes were selected as a measure of a structural degradation (see Fig. 5). From the presented curves one can observe that AE activity is lower during the presence of the air cooling, which confirms the conclusions made based on the temperature history evolution curves. These observations coincide with the results obtained by Kahirdeh et al. (2013), where the authors stated that for cooled specimens the damage rate was less in comparison with specimens loaded without cooling.