The size effect theory was proposed by Frossard and co-workers for evaluating the shear strength of rockfill material. However, this theory has not been validated in a general stress path, in which the intermediate principal stress may be different from the minor principal stress. A series of true triaxial compression tests on rockfill material A (RFM-A, a small-sized particle) and rockfill material B (RFM-B, a larger-sized particle) were carried out to validate the size effect theory in the general stress state. It was found that the predictions by the size effect theory, based on the material constants from the strength data of RFM-A, were in good agreement with the strength data of RFM-B.

Low-strain testing of piles is routinely performed by striking the pile head with a small hand-held hammer, usually aiming for the pile centre, and measuring the vertical velocity response by means of a receiver placed also on the pile head. In practice, however, it is impossible for the operator to perfectly align the point of impact with the pile centre, as existing solutions for the interpretation of the velocity measurements assume. In the general case where the impact load is eccentric, the wave field is no longer axisymmetric with respect to the pile axis and waves resulting from the hammer impact can propagate along the vertical, radial and circumferential directions. This is taken into consideration in the analytical solution presented in this note, which provides the non-axisymmetric dynamic response of a pile to an impact load applied at an arbitrary position on the pile head. Arithmetic examples are used to depict the three-dimensional non-axisymmetric vibration characteristics of the soil–pile system, with emphasis on the effect of the eccentricity of the impact load on the optimal position of the receiver, the identification of which is essential for obtaining meaningful results from low-strain integrity tests.

Bolton's stress–dilatancy equation is mainly used for clean sands. It needs to be investigated whether this equation could be used for sand with non-plastic fines or not. A series of drained triaxial compress tests were conducted for the sand with non-plastic fines. It was observed from these tests that the peak-state and critical-state friction angles increased with increasing the fines content. The peak-state friction angle and maximum dilatancy angle increased with decreasing the confining pressure. It was also found that the relationship between the excess friction angle (i.e. the difference between the peak-state and critical-state friction angles) and maximum dilatancy angle was unchanged with the non-plastic fines content. This has illustrated that the Bolton's stress–dilatancy equation proposed for clean sands could still be applied to sands with non-plastic fines.

In this note the concept of discretising the pile into virtual parts is introduced in order to derive an analytical solution that describes three-dimensional propagation of waves in piles during low-strain integrity tests, considering both vertical and radial pile displacements. The effect of transverse wave propagation on the results is quantified by comparing the proposed solution against a previously published method in which only vertical displacement was considered, results of dynamic numerical analyses, as well as the classical one-dimensional solution. Using the obtained waveforms, an attempt is made to identify and describe the mechanisms of three-dimensional propagation of longitudinal waves in piles, and how these may affect the interpretation of low-strain integrity tests.

A series of large-scale triaxial compression tests were performed on sandy soil confined by a circular geocell in this study. The effects of the relative density of the infill sand, the stiffness and strength of the geocell, and the aspect ratio (height-to-diameter ratio) of the specimen on the overall behavior of the geocell-soil composite were investigated. Furthermore, the measured confining pressure increment and apparent cohesion induced by the geocell membrane confinement were compared to those predicted by existing equations. The study results indicated that the end effects were very small in this experimental study. The effect of geocell stiffness on the stress-strain behavior of geocell-soil composites is closely related to the relative density of the soil. The measured confining pressure increments and the apparent cohesion are generally in good agreement with those predicted by Bathurst and Karpurapu's equations. The effect of the axial strain at failure on the apparent cohesion is discussed, and a method is proposed to determine the axial strain at failure for medium dense sand and dense infill sand on the basis of the hyperbolic nonlinear elastic model.