No abstract

No abstract

Fatigue crack growth under constant and random loading conditions was investigated for a metastable austenitic-bainitic steel in comparison with a ferritic chromium steel at very low crack growth rates. Experimentally determined random crack growth was compared with linear Miner calculations on the basis of constant amplitude results. It was found that the measured crack growth rates in transforming material are a factor of 10 lower than the calculated values, whereas the ditference is only a factor of 2 for the ferritic steel. The reason for the pronounced crack growth retardation in the metastable alloy is transformation of part of the austenitic phase into martensite in the stress field of the crack tip, accompanied by a volume increase and, consequently, residual compressive stresses. Rare high load cycles in the random sequence increase the closure level, which then leads to pronounced retardation of fatigue crack growth for the numerous successive low amplitude cycles. NOMENCLATURE a = crack length Aa, = crack increment in a random loading test C(n) = constant for approximation of fatigue crack growth curve using exponent n D(n) = mean error of fit and measurement of constant amplitude fatigue crack growth data E = Young's modulus f, (a/W) = correction function for calculation of stress intensity amplitude h(x = deviation of fracture surface asperities from an ideally flat surface K , = closure stress intensity factor K,, = maximum value of the stress intensity factor K , = root mean square of maximum stress intensity factors of a random sequence n = exponent for approximation of constant loading fatigue crack growth curve N o = number of load cycles applied in a random loading test R, = standard deviation of fracture surface asperities from an ideally flat surface W = specimen width u, , = vibration amplitude at one specimen end a = number of data obtained under constant amplitude loading conditions

No abstract

No abstract