PSI - Issue 57

Matthias Hecht et al. / Procedia Structural Integrity 57 (2024) 581–588 Matthias Hecht et al. / Structural Integrity Procedia 00 (2019) 000 – 000 photograph of the test rig is shown on the left and a schematic diagram of the force application is shown in the middle. The vertical cylinder generates a peel tensile load ̅ a,Ax via a ball joint. The horizontal cylinder applies a moment via a force ̅ a,M to the bowl specimen with a lever arm. The measurement of the acting forces on the hydraulic cylinders for control and measurement data acquisition is carried out via a force transducer in each load channel. To exclude temperature rising too, the test frequency is set at 7 Hz [13]. 4

584

0

330

Fig. 1 (a) Illustration of the test rig for the experimental investigations of the bowl specimens, with marking of the measuring points for displacement; (b): schematic representation of the application of loading forces; (c) sectional view of the bowl specimen To measure the stiffness behavior and the opening of cracks, the displacement on the bowl top was measured with triangulation laser orthogonal to surface. The three measurement points L 1 to L 3 are shown on Fig. 1 and were located on the positions where the crack initiation was expected by means of FE simulation. 2.3. Load sequence under variable amplitude The load sequence utilized for the tests involving variable amplitudes follows a Gaussian spectrum with a sequence length of L S = 50,000 cycles and was also used in the investigations to butt-bonded hollow cylinders [16]. This sequence is applied to both the vertical and horizontal load channels, as depicted in Fig. 2.

Fig. 2 (a) Normalized force amplitude; (b) normalized mean force; (c) R -ratio over the cycles in the load sequence

To ensure comparability with previous projects, the same sequence length utilized in those projects is adopted. This allows for approximately three repetitions of the sequence on the high load levels. Nevertheless, no discernible influence is observed since no non-linearity was observed at the Gassner lines. For the testing, a constant minimum stress value ̅ min was selected resulting in an ̅ -ratio of 0.01. Since all amplitudes start at the same minimum stress value, the R -ratio depends on the stress amplitudes and subsequently varies within the range of 0.01 ≤ ≤ 0.4, Fig. 2 (c).