PSI - Issue 18

V.N. Shlyannikov et al. / Procedia Structural Integrity 18 (2019) 322–329

327

6

Author name / Structural Integrity Procedia 00 (2019) 000–000

Compensation hole

a)

b)

Fig. 4. Imitation model I for uniaxial tension (a) and Imitation model II with compensation hole for biaxial loading (b).

forces along two axes of P x =19 kN and P y =38 kN with stress biaxiality ratio  =  rr /   =0.5 and the failure criterion was the condition when the growing crack reaches the compensation hole. All tests are carried out with sinusoidal loading form with load control. Load control was estimated to be better than ±1%. The crack growth was monitored using an optical microscope.

a)

b)

c)

d)

Fig. 5. Test equipment for imitation model I under uniaxial tension (a, b) and imitation model II under biaxial loading (c, d).

Two different stress ratio values (0.1 and 0.5) are applied several times to each model in order to fix the current crack tip position: during each test, benchmarks are produced for each model by increasing the applied stress ratio from 0.1 to 0.5 at a constant value of the maximum cyclic nominal stress, while the surface crack length is approximately increased by 1 mm. As was shown by Shlyannikov et al. (2015, 2018), Yarullin and Ishtyryakov (2016) the benchmark loading does not induce load history effects or overload retardation. The typical fracture surface marks are shown in Figures 6b, 7b for imitation model I and II, respectively. As expected the crack path obtained experimentally on imitation model I (Fig. 6a) do not replicate the real compressor disks failure shown in Fig. 1c. Such differences could be attributed to non-modeled effects like tangential (or hoop) stresses caused by centrifugal loads. It is clear that the imitation model I is more suitable for the experimental determination of durability at the stage of the formation of macro defects.