PSI - Issue 13

Daniele Rigon et al. / Procedia Structural Integrity 13 (2018) 1638–1643 D. Rigon et al./ Structural Integrity Procedia 00 (2018) 000 – 000

1642

5

700

Bending Torsion

 an ,  a [MPa]

B+T, φ = 0 ° , Λ=1 B+T, φ = 0 ° , Λ=√3 B+T, φ = 90 ° , Λ=1 B+T, φ = 90 ° , Λ=√3 Axial Torsion Ax+T, φ = 0 ° , Λ=√3 Ax+T, φ =9 0 ° , Λ=√3

Specimen (a)

AISI 304L R=-1

Specimen (b)

100

1,E+02

1,E+03

1,E+04

1,E+05

1,E+06

N f , number of cycles to failure

Fig. 4. Fatigue test results in terms of nominal net-section stress amplitude.

100

10

Bending

Torsion

Scatter band: 10-90% survival probabilities, from Meneghetti et al.(2013)(a) k=2.11, T Q =2.04, T N =4.50 Q A,50% =0.133 MJ/(m 3 cycle) N A =2·10 6 cycles , R=-1

B+T, φ = 0 ° , Λ=1 B+T, φ = 90 ° , Λ=1

B+T, φ = 0 ° , Λ=√3 B+T, φ = 90 ° , Λ=√3

Axial

Torsional

AISI 304L R=-1

Ax+T, φ = 0 ° , Λ=√3

Ax+T, φ = 90 ° , Λ=√3

10

1

0,1 Q [MJ/(m 3 cycle)]

1 Q [MJ/(m 3 cycle)]

Bending Torsion

B+T, φ = 0 ° , Λ=1 B+T, φ = 0 ° , Λ=√3 B+T, φ = 90 ° , Λ=1 B+T, φ = 90 ° , Λ=√3 Axial Torsion Ax+T, φ = 0 ° , Λ=√3 Ax+T, φ =9 0 ° , Λ=√3

Specimen (a)

Specimen (b)

0,01

0,1

1,E+02

1,E+03

1,E+04

1,E+05

1,E+06

0

0,25

0,5

0,75

1

N f , number of cycles to failure

N/N f

Fig. 5. (a) Typical evolution of Q during individual fatigue tests; (b) fatigue test results in terms of Q, compared to the scatter band calibrated in Meneghetti et al.(2013)(a). During the fatigue tests, several test stops were performed in order to evaluate the specific heat loss. The resulting evolution of Q versus the normalized fatigue life is reported in Fig. 5(a) for individual fatigue test. Notably, after approximately one third of the total life, the Q parameter can be reasonably assumed constant. In Fig. 5(b) all the fatigue test results have been expressed in terms of specific heat loss evaluated at half the total fatigue life and compared to the existing heat energy-based scatter band calibrated previously (Meneghetti et al., (2013)(a)). Concerning the LCF regime, it can be noted that all the Q-based fatigue results of the type (a) specimens are in good agreement with the uniaxial scatter band. The same cannot be stated for the type (b) specimens, due to the limited fatigue test results.As to the HCF regime, it can be observed that the existing scatter band correlates only uniaxial, torsional and in-phase multiaxial fatigue tests with  =1. Conversely, in-phase multiaxial fatigue results with Λ = 3 as well as out-of-phase multiaxial data with Λ = 1 as well as 3 fall outside the scatter band in the non conservative side. In general, the results obtained for the thin-walled specimens (type (b)) are in agreement with those relevant to type-(a) specimens. This preliminary study shows that Q parameter seems to be influence by the biaxiality ratio Λ and by the phase shift angle  .