The small-strain stiffness of a railway ballast

Abstract

The large particle sizes of railway ballast and rock fill have meant that conventional techniques used to measure the small-strain stiffness of finer geomaterials have not been adopted, with the consequence that their stiffnesses are poorly defined. In a series of tests on a UK railway ballast, simple adaptations were made to existing local strain measuring systems to account for the larger particle sizes. The study showed that the small-strain stiffnesses are different in second loading compared to virgin loading, but multiple cycles had little further effect on the stiffness. The large particle size was found rarely to have any detrimental effect on the quality of the strain measurements and the two independent measurements of axial strain taken at diametrically opposite locations were generally as consistent as for finer grained soils. As for other soils, the ‘external’ measurements of strain across the apparatus platens were of little use in determining stiffness. The presence of water did not have a significant effect on the behaviour, and this was confirmed by inter-particle loading tests on single particle contacts. Despite the use of lubricated end platens, there was a significant barrelling of the sample at large shear strains so that the internal measurement of the volumetric change diverged from the external measurement at large strains. The very small volumetric strains that occurred during isotropic loading meant that each sample could only be used to obtain one measurement of the virgin loading stiffness.