ASTM A723 Q & T steel with a yield strength and ultimate strength of 1170 and 1262 MPa respectively was evaluated for mean stress-strain effects under smooth specimen axial strain controlled low cycle fatigue conditions with strain ratios R of -2, -1,0,0.5 and 0.75. Cycles to failure ranged from 15 to lo5. Cyclic stress-strain response based upon half-life hysteresis loop peaks were similar for all R ratios. Mean stress relaxation occurred for R # -1 only when plastic strain amplitudes were present and this occurred above total strain amplitudes of 0.005. Thus, mean stress relaxation was completely dependent upon cyclic plasticity. Mean strains did not affect low cycle fatigue life unless accompanied by half-life mean stress. Tensile mean stress was detrimental and compressive mean stress was beneficial and these effects only occurred at strain ampltidues below 0.005. Three different mean stress models were used to evaluate the low cycle fatigue data and the SWT log-log linear model best represented the data. These results can be used with the local notch strain fatigue life prediction methodology. NOMENCLATURE b = fatigue strength coefficient c = fatigue ductility exponent A6 t, = -= strain amplitude 2 t,,, = mean strain tmln = minimum strain tmar = maximum strain At = strain range Ate = elastic strain range Atp = plastic strain range 6 ; = fatigue ductility coefficient E, = modulus of elasticity from the first quarter cycle E, = modulus of elasticity from tensile unloading E3 = modulus of elasticity from compressive loading HR, = Rockwell hardness C scale K = strain hardening coefficient K' = cyclic strain hardening coefficient n = strain hardening exponent n' = cyclic strain hardening exponent Nf = cycles to failure 2Nf = reversals to failure R = strain ratio (t,,JcmaX) Au = stress range ( T~ = stress amplitude urn = mean stress u; = fatigue strength coefficient (T, = ultimate tensile strength uys = monotonic yield strength uks = cyclic yield strength SWT = Smith-Watson-Topper parameter 413 414

No abstract

No abstract

The objective of this research was to obtain and compare constant and variable amplitude fatigue behavior of AZ91E-T6 cast magnesium alloy in both an air and 3.5 percent NaCl aqueous corrosive environment. An additional objective was to determine if commonly used models that describe fatigue behavior and fatigue life are applicable to this material and test environment. Fatigue tests included constant amplitude strain-controlled low cycle fatigue with strain ratio, R, equal to 0, −1 and −2, Region II constant amplitude fatigue crack growth with load ratio, R, equal to 0.05 and 0.5 and variable amplitude fatigue tests using keyhole notched specimens. In all fatigue tests, the corrosion environment was significantly detrimental relative to the air environment. Mean strains influenced fatigue life only if accompanied by significant mean stress. The Morrow and Smith, Watson, and Topper mean stress models provided both accurate and inaccurate fatigue life calculations. Likewise, variable amplitude fatigue life calculations using the local strain approach and based upon the formation ofal mm crack at the keyhole notch were both accurate and fairly inaccurate depending on the specific model used.

No abstract