CRACK PROPAGATION IN SOIL-CEMENT BASES SUBJECTED TO REPEATED WHEEL LOADS

Fatigue in the soil cement has been shown to be a controlling factor in the process of selecting thickness for pavements containing cement-stabilized layers. If the fatigue criteria obtained from laboratory tests do not include consideration of the number of repetitions associated with crack propagation, conservative pavement thicknesses might result. This paper shows how laboratory fatigue data on soil-cement specimens subjected to biaxial states of stress are used to estimate how many additional repetitions will produce crack propagation in the field. The test results are interpreted by using a fatigue model based on Griffith's failure criteria and a finite-element idealization of pavement systems containing a soil-cement layer. The analyses include the influences of (a) age, maximum stress level, and interface boundary condition between base and subgrade on the rate of crack propagation through a soil-cement base; (b) maximum stress level in the base on the number of repetitions needed to propagate the crack to the surface of the base; and (c) interface friction and the ability of the subgrade to accommodate tension stresses on fatigue fracture in the soil-cement base. The examples given show that permitting crack propagation to the surface rather than crack initiation will give considerably thinner design base courses. /Authors/