STRESS STATE DEPENDENCE OF CYCLIC RATCHETTING BEHAVIOR OF 2 RAIL STEELS

Cumulative inelastic deformation or ratchetting occurs during cyclic l oading in the presence of a mean stress. This problem has received con siderable recent attention. The nonlinear kinematic hardening rule ori ginally proposed by ARMSTRONG & FREDERICK (AF rule) [1966] has been wi dely used for description of the overall character of hysteresis respo nse during cyclic plasticity. However, this model generally overpredic ts cyclic strain accumulation (ratchetting) under asymmetric loading w ith mean stress (BOWER [1987]; BOWER & JOHNSON [1989]; CLEMENT & GUION NET [1985]; CHABOCHE [1989,1991]; CHABOCHE & NOUAILHAS [1989]; MCDOWEL L & LAMAR [1989]; MCDOWELL [1992]). Moreover, unloading-reloading beha vior associated with subcycle events under even uniaxial conditions (C HABOCHE [1989]) are not adequately represented by the AF rule. The sam e comments apply to other nonlinear kinematic hardening rules of equiv alent or similar nature such as bounding surface plasticity theory (cf . MCDOWELL and MOYAR [1991]). In this work, a modification of the dyna mic recovery term of the AF rule is considered as recently proposed by OHNO & WANG [1991a,1991b]. The approach is established based on certa in assumed crystalline slip system behavior and interaction between di slocation interactions at different size scales. Several important cri teria are discussed for models capable of representing stress state an d amplitude dependence of ratchetting behavior. Experimental results o btained on both a carbon rail steel and a heat-treated rail steel subj ected to various uniaxial and nonproportional loading conditions are p resented and correlated with an extension of the Ohno and Wang model, which accounts for a broader range of stress state and amplitude effec ts. An algorithm is developed to extrapolate the ratchetting rates obt ained at relatively early cycles to very large numbers of cycles for t est specimens. The method provides for dependence on the mean stress a s well as amplitude of loading. The algorithm is applied to both carbo n and heat-treated rail steels.