PSI - Issue 17

Zizhen Zhao et al. / Procedia Structural Integrity 17 (2019) 555–561 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

558

4

3. Ratcheting-fatigue and creep-ratcheting-fatigue behaviors

3.1. Stress-strain hysteresis loops Typical stress-strain hysteresis loops in RF and CRF under σ m =100 MPa and σ m =300 MPa are compared in Fig. 3. The non-closure of hysteresis loop in the first cycle is remarkable, but its openness is significantly reduced with continuous cycling. Strain after loading in the first cycle increases when stress holdings are incorporated. The inelastic strain accumulates in the direction of mean stress, but the growth rate is much slower in RF test, as hysteresis loops of 500 and 1000 cycle almost overlap. Peak stress holding stimulates the growth of inelastic strain, while peak/valley holding makes it even larger. When 10 s holding period is applied at either peak stress or valley stress, remarkable creep strains are created, which widens the hysteresis loops. 3.2. Fatigue life The variation of fatigue life with stress holding periods is shown in Fig. 4. Trend lines are added for each stress condition for clarity, and the stress holding period for RF tests is taken as 0.1 s to fit the logarithmic coordinate. The increase of either mean stress or stress amplitude would reduce fatigue life in RF tests, and fatigue life drops exponentially when hold period is involved. The reduction in fatigue life becomes more significant when long holding period is applied at high stress level, as for the 30 s and 60 s holding cases under σ m =125 MPa and σ a =300 MPa. However, when the sum of holding periods is equal, the corresponding fatigue lives are quite close regardless of hold directions, as in cases when t h = 10s and t h = c h =5s.

Fig. 3 Comparison of stress-strain hysteresis loops in RF and CRF tests. Fig. 4 The influence of stress holding periods on fatigue life.

3.3. Creep and fatigue damage

The time fraction rule was used in the evaluation of fatigue and creep damage in CRF tests as follows, CRF f RF D N N = (1) CRF c R N t D t   = (2) where D f is the fatigue damage, D c is the creep damage, N CRF is the number of cycles to failure in CRF tests, N RF is the number of cycles to failure in RF tests, Δt is the holding period in a cycle, t R corresponds to the creep rupture time under holding stress. t R was determined from short term creep tests of the steel, where a power-law relationship was found between creep stress and rupture time, with the coefficient and exponent being 453.5 MPa and -0.04