PSI - Issue 41

R. Nobile et al. / Procedia Structural Integrity 41 (2022) 421–429 Riccardo Nobile et al. / Structural Integrity Procedia 00 (2019) 000 – 000

424

4

Fatigue tests were performed using a servo-hydraulic load machine INSTRON 8850 axial-torsion machine, having a load cell of ± 250 kN and controlled by a software Instron Wave Matrix. The tests were conducted in force controlled with a frequency load of 10 Hz under sinusoidal waveform in tension-tension mode (load ratio R = 0.1). Table 2 summarizes the experimental fatigue parameter of each tested specimen.

Table 2. Scheduled experimental parameters for each fatigue tests.

Number of cycles to failure (Nf)

Maximum load [kN]

Minimum load [kN]

Load amplitude [kN]

Stress amplitude [MPa]

Specimen

P1 P2 P3 P4

25 22 20 16

2.5 2.2

11.25

416.39 366.54 333.11 266.67

11875 10629 24236 94132

9.9

2

9

1.6

7.2

3. Results and discussion The thermal effects on the experimental measurement of electrical resistance were also considered to avoid the introduction of unacceptable errors. The effect of the change in resistance due to temperature effect was determined by a linear relationship (Nobile et al. (2021)) according to Eq. (1): ∆ ℎ = − 0 = ∆ (1) where the resistance temperature coefficient A = 0.002 mΩ °C -1 was evaluated interpolating the experimental data obtained during a calibration test (Fig. 4a). The specimen was heated by means of a 1 kW halogen lamp, measuring the temperature data of the three thermocouples and the corresponding electrical resistance change using the same experimental setup of the fatigue test. Figure 4b shows the increase in specimen temperature over time as measured by the three thermocouples. The trends of the resistance variation against average temperature variation measured by the thermocouples, show an approximately linear correlation between the two variables both in the heating and in the cooling phase and this last was used to determine the value of the coefficient A (Fig. 5b).

(a) (b) Fig. 4. (a) Experimental setup for determination of coefficient A; (b) average temperature variation of the three thermocouples (TC m ) vs. time.