PSI - Issue 5

Robert Hannemann et al. / Procedia Structural Integrity 5 (2017) 861–868 Hannemann et al./ Structural Integrity Procedia 00 (2017) 000 – 000

866

6

4. Investigations on overlapping bending an torsional loading

For overlapping bending and torsional loading a special test rig was developed (Köster and Sander (2016)), Figure 7. With this test rig the investigation of the influence of pure cyclic bending and torsion as well as their superposition in-phase and out-of-phase on the fatigue crack growth is possible. Furthermore, the stress ratio and the phase shift angle of the mixed mode loading are variable.

4.1. Principle of operation of the bending-torsion test rig

For the application of a constant bending moment distribution in the area surrounding the crack, the specimen (6) is supported by four barrel roller bearings with separable bearing housings (5). While the outer two bearings are fixed, the inner ones are connected with a bridge (9), at which a hydraulic 100 kN cylinder is attached to apply the load.

Fig. 7. Developed bending-torsion test rig with components: spherical roller bearings (1), lever (2), torsion shaft (3), Omniflex®-coupling (4), barrel roller bearing (5), specimen (6), constructed adapters (7), half shell adapter (8) and bridge (9)

With regard to the application of the torsional moment additional torsion shafts (3) are mounted on both specimen ends. Furthermore, both shafts are equipped with a lever (2) connected with two hydraulic 50 kN cylinders to apply the torsional loading. Moreover, the torsion shafts are each supported by a fixed and a floating roller bearing (1) to prevent transferring a superimposed bending. To transfer the torsion moment as well as to compensate the misalignment occurred due to bending, the custom-made so called Omniflex®-couplings (4) are used connecting the torsion shafts with the specimen. For a guaranteed minimum of additional operating inertia forces the coupling elements are partly supported by half shell adapters made of polyetheretherketone (8). Furthermore, several frame carriers are connected in compensation of the height of the hydraulic cylinders as well as for the construction of a stiff framework, at which the bearing components are fixed using specially de-signed adapters (7). The whole test rig is mounted on an air-cushioned, 3.5 m wide and 5 m long clamping plate. In order to validate the experimental data, a finite element model of the specimen under test rig conditions was created. Therefore, results for the SIF could be achieved for a crack with a = 1 mm and a / c = 0.8 under a load cycle of combined bending and torsion loading. In Figure 8 the SIF for mode I, II and III versus the normalized position on the crack front s for a phase shift of 0° and 90° are exemplarily given. Under the chosen relation between bending and torsional moment, Mode I represents the significant part in both diagrams and shows a nearly constant distribution, which can be returned on the natural crack geometry at a / c = 0.8. As expected, the distribution of K II as a result of the torsion is zero in the middle of the crack front and maximum with 4.2. Numerical results without crack growing