THE ANALYSIS AND DESIGN OF THE FLEXIBILITY OF PAVEMENTS

Abstract

This paper deals with the application of fracture mechanics to the problem of fatigue cracking and failure of flexible pavements. It describes experiments conducted on four-foot diameter bituminous slabs supported on an elastic foundation in order to verify Paris' crack propagation law. The crack pattern and the actual crack lengths were obtained by X-ray photography and the techniques for analyzing the rate of crack propagation for multiple cracks are presented. Excellent agreement has been found with Paris' law dc/dN=AK4, where dc/dN is the rate of crack propagation, K is the stress-intensity-factor, which measures the general load transmission to the vicinity of the crack tips, in accordance with the load, geometrical and boundary conditions, and A is shown to be a materials constant. On the other hand comparisons made on the basis of the bending stress at the interface of the slab or beam and foundation showed discrepancies. This is not surprising since the stress-intensity-factor is a much more powerful parameter than the bending stress, being able to take into account much more fully the boundary and geometrical conditions in accordance with the load and the amount of cracking in the pavement. The criteria for fatigue failure of pavements is defined as the time for the stress-intensity-factor at the tip of the largest crack to reach its critical value, since rapid crack propagation would then occur, or the time for the total area of cracking to exceed ten per cent of the area of the pavement surface, whichever occurs first. Finally, assuming knowledge of the load-time history of random loading, and the variation of the material constants (such as the Young's modulus of each layer, the crack propagation constant, A, and the critical-stress-intensity factor, K1C, of the bituminous material) with temperature, rate of loading, age and moisture content, etc., a method of analysis and design is proposed to guard against flexibility failures in pavements. Further simplications are then introduced to illustrate the effect of using thicker bituminous layers to replace granular material that contribute the same strength according to the AASHO method of design, and to obtain the load equivalency factors for various magnitudes and configurations of wheel loads. /Author/