Experimental Investigation of Deep Penetration in Sand

Purdue University, West Lafayette IN

Investigators

Abstract

This project examines the fundamentals of the problem of deep penetration in sand. The deep penetration problem contains virtually all the challenges that geomechanicians may face: large displacements, large rotations, large strains, strain localization, crushing, and a moving soil-metal boundary. The impact of the research therefore is dual: it will advance the understanding of a problem with applications to diverse fields, including projectile penetration into the ground, design of pile foundations and interpretation of the cone penetration test, but it will also help advance the understanding of fundamental processes that happen in granular (particulate) media, such as strain localization and particle crushing and how all of these complex processes should be handled in computational simulations. Benefits for society therefore are multiple, including the general advancement of the science of granular media, economies in foundation design, a better understanding of an application (projectile penetration) that is vital to the defense industry, the production of a database of images that can be used in education and research, and the training of researchers in advanced technology and computer simulations. The research specifically uses the powerful combination of calibration chamber testing and the digital image correlation (DIC) technique to study cone penetration in crushable sands. A unique half-cylindrical chamber with a flat Plexiglass wall that allows observation of the vertical motion of half-circular penetrometers will be used to acquire images of the penetration process under various conditions with state-of-the-art high-resolution digital cameras and microscopes. The images will be processed in two ways: using DIC to obtain the displacement and strain fields in the soil and an algorithm that produces the particle size distribution of the soil from an image of it. The penetration resistance, local stresses, and the displacement and strain field dataset may be viewed as an experimental solution to the penetration boundary-value problem. Particle size gradation will be assessed during the entire penetration process, thereby enabling the research team to consider the relationship between crushing, wherever it occurs, penetration resistance and the displacement and strain fields. The calibration chamber tests are complemented by a complete laboratory testing program to produce data that can later be used in validation of theoretical simulations and development of methods of analysis and design. The high-quality experimental data will include data from tests in a unique Hopkinson bar especially designed to test soil under uniaxial or triaxial conditions. Work will also be done to advance computational simulation capabilities of penetration problems using the material point method.