The arching phenomena observed in experimental trap door model tests

This paper presents experimental data on a well graded angular sand used to investigate the impact of particular design parameters such as sample density and sample height on the arching mechanism. Soil arching is a natural phenomenon which occurs in all granular soils as a result of stress redistribution. When the trapdoor at the base of the model was released the soil block above the trapdoor settled while the soil block above the supports did not thus shear stresses developed in the fill material between the yielding and unyielding zones. The base of the model was released at two speeds, rapid and gradual, to simulate consolidation of the soft subsoil between piles in a piled embankment and the formation of a sudden void. It was noted that the final load transfer recorded was the same for both the rapid and gradual tests, however there was an initial peak load transfer recorded for the rapid tests. The relationship between surface and base displacements recorded in the model are also discussed.