A concept of crack closure has been applied to data for fatigue crack growth in the low growth rate regime for two steels containing different inclusion populations. A model has been developed in which the crack-closure contribution is represented by a series of springs and has been used successfully to explain experimental observations. It is suggested that for a given material a ‘true’ threshold value of alternating stress intensity ∆K exists and that measured values of threshold ∆KTh are related to this via a crack-closure mechanism which is dependent on the roughness of the fracture surfaces.
