PSI - Issue 54

Jürgen Bär et al. / Procedia Structural Integrity 54 (2024) 188–195 Author name / Structural Integrity Procedia 00 (2019) 000–000

190

3

The fatigue experiments were conducted on a servo hydraulic testing machine with a loading amplitude of F=30 kN with a frequency of 20 Hz under fully reversed loading conditions (R=-1). In fixed intervals of 10,000 or 15,000 cycles overloads with an amplitude of 60 kN were introduced to mark the current crack geometry on the fracture surface. The measurement of the three potentials was undertaken using amplifiers of the control electronics (DOLI EDC 580V). For each cycle the maximum and minimum value of the force, the stroke and the three potentials were stored simultaneously using the software Test&Motion. For the evaluation only the maximum values of the measured potentials were used. In order to force crack initiation at defined positions secondary notches with a depth of about 150 µm at positions of 0°, 15°, 30° and 60° relative to potential 1 as shown in figure 2a were machined into the specimens using a laser engraving system.

a

b

Fig. 2. (a) positions of the secondary laser notches; (b) Illustration of the measurement of the crack depth a on the fracture surface.

To obtain crack depth values the calibration of the potential measurements, the crack depth a was measured from the specimen surface to the respective overload mark as shown in Fig. 2b. The corresponding potentials U i were determined from the recorded data by averaging the potentials U i of the last 10 cycles before the overload. The measured potentials were normalized by dividing the current potential values U i by the respective initial potentials U 0,i defined as the mean value of the potential of the first 20 cycles at the beginning of the experiment. � � � � � � , � � 1,2,3 (1) 3. Results and Discussion 3.1. Results of the potential measurement Figure 3a shows a fracture surface of an experiment with a specimen with a laser notch at 0°. To increase visibility, the markings on the fracture surface have been highlighted in color. The cracks propagate from the secondary notch into the specimen in the form of half ellipses. With increasing crack depths a slight asymmetry appears in the crack shape. In Figure 3b the three relative potentials as a function of the cycle number are plotted. The colored lines correspond to Figure 3a and mark the overload cycles. At the beginning of the experiment, all three relative potentials show a congruent course with an increase of the relative potential up to about 20,000 cycles. This increase can be contributed to a warming of the specimen due to dissipative effects in the material. A first small deviation of potential 1 from the other two potentials is visible after about 40.000 cycles indicating the formation of a crack near the position of