Search

The interpretation of cone penetration test (CPT) data is important for the in-situ characterisation of soils. Interpretation of CPT data remains a predominately empirical process due to the lack of a rigorous model that can relate soil properties to penetrometer readings. Interpretation is especially difficult in layered soils, where penetrometer response can be affected by several horizons of soil with different properties. This paper aims to provide some insight into the mechanisms of soil displacement that occur as a penetrometer is pushed into layered soils. Data is presented from centrifuge modelling of probe penetration in layered soils in an axisymmetric container where soil deformation patterns around the probe can be measured. Results obtained from uniform soil tests are also presented to illustrate the effects of soil density and stress level (i.e. centrifuge acceleration). A large influence zone is found to relate to the higher penetration resistance obtained in a denser soil. Differing soil displacement patterns at low and high stresses are related to the tendency of the soil to dilate, with the well-known consequence of a non-linear increase of penetration resistance with stress level. Layered soil tests show a clear difference of soil deformation patterns compared to uniform tests, especially for vertical displacements. The peak value of vertical displacement of the soil occurs at dense-over-loose interfaces, while a local minimum occurs at loose-over-dense interfaces. Parameters are proposed to quantitatively evaluate the layered effects on soil deformations and a deformation mechanism is described for penetration in layered soils based on the transition of displacement profiles.