PSI - Issue 38

W. Radlof et al. / Procedia Structural Integrity 38 (2022) 50–59 W. Radlof et al. / Structural Integrity Procedia 00 (2021) 000 – 000

52

3

Design

Strut size, mm

Pore size, mm

No. of cells

Porosity, %

50% porosity 60% porosity 70% porosity

1.08 0.93 0.78

1.24 1.42 1.59

6 6 6

50 60 70

a

b

BD

Fig. 1: Test samples of a 70% porosity-design for (a) bending and (b) torsion fatigue tests and their build directions (BD).

2.2. Experimental set-ups and measurement techniques Cyclic four-point bending tests were carried out using a servohydraulic testing machine with a 25 kN load cell at constant stress amplitudes with a stress ratio of R = 0.1 and a frequency of 20 Hz. The distance between the inner and outer bearing is 27 mm and between the two inner bearings 38.36 mm. Cyclic torsion tests were performed using a servohydraulic tension/torsion machine with constant stress amplitudes with a stress ratio of R = -1 and a frequency of 5 Hz. Two loading levels are selected, which are normalized to the design-dependent maximum moments determined in quasi-static bending ( , ) and torsion ( , ) tests, respectively and are listed in Table 2. The tests were stopped after the machine displacement reached 0.4 mm for bending tests or an angle of ±8 ° was obtained for torsion tests, which correlates with the total fracture of the porous structures.

Table 2. Parameter for fatigue bending and torsion tests. Loading Normalized load level

Design-dependent moments

0.15 ∙ , 0.2 ∙ , 0.12 ∙ , 0.17 ∙ ,

50% porosity

60% porosity

70% porosity

26.7 Nm 35.7 Nm 12.8 Nm 18.2 Nm

19.5 Nm 26.2 Nm 6.8 Nm 9.7 Nm

12.2 Nm 16.2 Nm 3.4 Nm 4.9 Nm

Bending

Torsion

To investigate the damage behavior of the porous structures the digital image correlation technique, the temperature field measurement as well as the potential drop method were implemented into the test rigs, as shown in Fig. 2. With the digital camera system VIC-2D HR (correlated Solutions, Irmo, USA) the deformation process of the samples was recorded on different specimen sides with two 12 megapixel (4096 × 3000 pixel) cameras. For all tests, the cameras were positioned to maximize the gauge sections in the pictures. This set-up resulted in a resolution of about 121 pixel/mm. In the bending tests, the tensile loaded side (bottom) of the porous structure was investigated with a mirror system (Fig. 2a). The temperature was measured at the sample surface with the VarioCam @ HD head (InfraTec, Dresden, Germany) with frequencies of 3 and 5 Hz for bending and torsion fatigue tests, respectively. Due to the test set-up, the distance of the thermography camera to the sample in the torsion tests was 120 mm, whereas the distance in the bending tests was only 25 mm. Furthermore, the crack propagation measuring device DCM-2 (MATELECT LTD, Harefield, United Kingdom) based on the direct current potential drop method was used for the measurement of crack initiation and propagation. Therefore, four potential probes are attached to the sample as shown in Fig. 2a and b.