PSI - Issue 42

Pedro R. da Costa et al. / Procedia Structural Integrity 42 (2022) 1560–1566 Pedro R. da Costa/ Structural Integrity Procedia 00 (2019) 000 – 000

1563

4

2. Experimental results The experimental results are divided into two sections: the experimental measurements where the displacements and strain gauge measurements are presented and compared to the numerical results; Fatigue fracture surfaces where the obtained fracture intricacies between the three methods are studied and compared 2.1. Experimental measurements As previously mentioned, two major methods of experimental study of the three conducted ultrasonic methods were employed: vibrometer laser displacement and strain gauge measurements. Both were recorded with the same piezoelectric transducer settings to associate the obtained displacement with the induced stress at the fatigue testing region. Also, across all conducted measurements and fatigue tests, the generated temperature at the highest stress amplitude location (the fatigue testing region) was monitored and controlled by a thermal camera for specimen protection. The heat generation location provides the first correct function feedback of any ultrasonic fatigue machine. If a significant heat generation source is present other than the fatigue testing region (e.g. the specimen-horn connection) the setup is proving not to work as intended. Figure 2 shows the three methods heat generation. A clear primary heat generation source at the specimen center, its fatigue testing region, is present in all three methods.

Fig. 2. Thermographic image during ultrasonic fatigue testing in: (A) tension-compression; (B) pure torsion; (C) multiaxial Tension/Torsion. With the high-stress fatigue region proven by the thermal camera all mentioned laser and strain measurements were conducted. Incremental transducer settings were applied in short resonance sequences. The measured displacements were associated with the induced stress amplitude at the fatigue testing region. Figure 3 shows all three setups displacement to stress ratio. By comparing all three displacement-stress computed methods to all three ultrasonic fatigue machines a good agreement with a lower than 10% difference is always perceived. Only the rotational-shear stress ratio for the tension torsion had some tested specimens with a higher error. This is linked to the associated complex geometry and the consequent more complex resonance state that brings new challenges in the measurement of displacement and strain and experimental replicability. Still, the made groves on the pure torsion and tension-torsion specimens for vibrometer laser measurement of the rotational displacement proved reliable in quantifying the displacement and also to be applied in WES 1112 proposed analytical stress calculation method.