PSI - Issue 33

Sabrina Vantadori et al. / Procedia Structural Integrity 33 (2021) 773–780 Author name / Structural Integrity Procedia 00 (2019) 000–000

777

5

For such tests, the mechanical parameters needed to apply the present criterion are:

1 197 af ,    MPa,

1 178 af ,    MPa and 689 u   MPa. As far as finite life fatigue tests are concerned, Tovo et al. (2014) performed a wide experimental campaign on a commercial DCI, identified as EN-GJS-400-18. The material had a ferritic matrix and spheroidal graphite nodules (black circles in Fig. 1(b)), with an average nodule size equal to ~50μm. The specimens, with a cylindrical shape and a diameter equal to 12.5mm along the gauge length, were cut from a large crossbar of a hydraulic press for ceramic tiles. Such specimens were subjected to both constant amplitude uniaxial and biaxial cyclic loadings under high/medium-cycle fatigue regime. In uniaxial fatigue tests, the following fatigue ratios were adopted: 3 1 0 0 5 R ; ; ; .    for tensile loading and 1 0 R ;   for torsional loading. Biaxial fatigue tests were performed under both fully reversed tensile and torsional loading with a phase shift  equal to 0° or 90°. As a fracture criterion was considered the complete separation of the failed specimens. Run-out tests were stopped at a number of loading cycles equal to 3ꞏ10 6 cycles. The details of the experimental loading conditions are reported in Vantadori et al. (2020b). For finite life fatigue tests, the mechanical parameters required to apply the present criterion are: 1 173 af ,    MPa, 1 167 af ,    MPa, 12 82 m .  , 10 10 * m .  , 6 0 0 2 10 * N N    cycles and 689 u   MPa. 3.2. Results and discussions According to the fatigue criterion here employed, the fatigue endurance condition (i.e. 1 eq,a af ,     ) is graphically represented by an ellipse in the a C against eq ,a N diagram, as is reported in Fig. 2 for infinite life fatigue tests and in Fig. 3 for finite ones. The safe domain is located inside the ellipse, whereas fatigue failure occurs if the point associated with a given experimental test lies out of the elliptical domain. The full and empty symbols reported in such figures are related to experimental failures and run-outs, respectively; moreover, blue empty symbols refer to experimental tests showing both failures and run-outs. The dashed lines represent the error band equal to ±20%.

200

200 (b)

(a)

Tension Torsion

 =90°

Failure Run-out

Failure Run-out

150

150

50 SHEAR STRESS AMPLITUDE, C a [MPa] 100

100

50

0 50 100 150 200 EQUIVALENT NORMAL STRESS AMPLITUDE, N eq,a [MPa] 0

0 50 100 150 200 EQUIVALENT NORMAL STRESS AMPLITUDE, N eq,a [MPa] 0

Fig. 2. Fatigue strength evaluation for infinite life fatigue tests under: (a) uniaxial and (b) biaxial loading.

The analysis of the results indicates that: -

For infinite life fatigue tests under uniaxial loading (Fig. 2(a)), the present criterion is in general able to correctly estimate failure or run-out conditions, in agreement with the experimental observations. Moreover, even when experimental failures are not theoretically predicted, such a criterion provides a condition of incipient failure, being the points very close to the elliptical domain;