Interpretation of sand state from cone penetration resistance

Abstract

The cone penetration test (CPT) is a widely used test for determining in situ state (or density) of cohesionless soils. The reliability of the estimated in situ state parameter, ψ, is constrained, however, because intrinsic soil properties affect the CPT response (often called a ‘compressibility’ effect). Here, the limiting cavity expansion pressure is used as an analogue for CPT tip resistance, qc, to determine the effect of intrinsic soil properties on the qc versus ψ relation. The limiting cavity pressure is computed by a large strain finite element code using a critical state-based soil model. This cavity expansion idealisation is then scaled to reference calibration chamber data with a universal shape function. The approach leads to a practically simple and theoretically sound method in which results from a few triaxial tests allow calculation of a sand-specific CPT qc–ψ calibration. The method is tested against published reference studies on nine sands and recovers ψ with a precision in void ratio of Δe = ± 0·04 at approximately 80% confidence. This accuracy for soils with elasticity measured by bender elements is considerably greater, Δe = ± 0·04 being obtained at approximately 95% confidence.