PSI - Issue 57

Sven Maier et al. / Procedia Structural Integrity 57 (2024) 731–742

738

8

S. Maier et al. / Structural Integrity Procedia 00 (2023) 000±000

used for excitation and 26 sensors were installed to record the acceleration of the system. The measurement setup is displayed in Figure 5. A series of six measurements were carried out for the nominal condition (variant 1). For each of the other two variants three measurements were conducted. For each measurement the plates, mounting and mass component were replaced. The modal analysis was conducted and evaluated with the commercial software Siemens Testlab 2011. The results can be found in Table 4. The mean values and C.o.V. gained from the measurements for the modal damping and eigenfrequencies are displayed for each mode and variant.

Fig. 5. Experimental modal analysis setup of test hardware with the ap plied accelerometers and shaker for excitation.

Fig. 6. Vibration fatigue test setup. The nominal variant 1 is representa tively shown.

Table 4. Results of the experimental modal analysis. Left side: Modal damping ζ , right side: eigenfrequencies f n .

Modal Damping ζ [%]

Eigenfrequency f n [Hz]

Modes

Mode 1

Mode 2

Mode 3

Mode 1

Mode 2

Mode 3

C.o.V.

C.o.V.

C.o.V.

C.o.V.

C.o.V.

C.o.V.

µ

µ

µ

µ

µ

µ

Variant 1 0.6 17.41% 0.43 13.85% 0.23 16.1% 41.02 0.77% 76.21 0.43% 83.77 0.32% Variant 2 0.52 1.37% 0.38 0.8% 0.23 3.01% 38.25 0.54% 68.97 1.14% 79.97 0.95% Variant 3 0.55 4.17% 0.33 3.57% 0.37 4.7% 36.42 0.5% 64.02 0.99% 76.13 0.85%

The damping was determined by means of the half-width method from the measured frequency response functions. It is found that modal damping, as well as its scattering, is significantly lower for the investigated system than expected from literature values. A modal damping between 0.23% and 0.6% was observed and also a decrease of damping with increasing frequencies for the eigenmodes for all variants. For the eigenfrequencies low C.o.V. values can be noted and for variant 2 with an increased mass of the component, as well as for variant 3 with thinner plates, the eigenfrequencies have shifted to lower frequencies and the system shows therefore a reduced sti ff ness.

3.2. Vibration Fatigue test

The investigation of the vibration fatigue behaviour for the analysed system was performed by an experimental test on a shaking table. Four specimens were tested for each of the three variants. The specimens were screwed to the shaking table and acceleration sensors were fastened to the mounting near the connection points ( Figure 6). The uniaxial PSD signal in z direction from Figure 3 was applied as load. The samples were visually inspected every 15 minutes for cracks that exceed 1 mm length as failure criterion. In extension to the visual inspection, a frequency response analysis by means of a sine sweep was conducted as additional parameter for failure indication. If a damage