Finite element modeling of jacked piles in clay and loose sand

Ripon Karmaker, Bipul Hawlader, Chen Wang

In the proceedings of: GeoSt. John's 2019: 72nd Canadian Geotechnical Conference

Session: Pile Foundations

ABSTRACT: Pile jacking is one of the alternative pile installation methods to conventional dynamic installations. Jacking does not cause significant vibration or disturbance of the surrounding soil. The present study uses a Coupled Eulerian-Lagrangian (CEL) approach of finite element analysis to simulate the pile jacking process. Quasi-static penetration of a rigid pile in clay and loose sand is simulated. The simulations are performed for an undrained condition for clay and a drained condition for sand. The penetration resistance, penetration-induced radial stress, ground surface heave, and plastic shear strain resulting from pile jacking in clay and sand are compared.

RÉSUMÉ: Le levage de pieux est l'une des méthodes alternatives d'installation des pieux aux installations dynamiques conventionnelles. Le fonçage ne provoque pas de vibrations importantes ni de perturbation du sol environnant. La présente étude utilise une approche d'analyse par éléments finis Coupled Eulerian-Lagrangian (CEL) pour simuler le processus de levage de pieux. La pénétration quasi statique d'un pieu rigide en argile et en sable meuble est simulée. Les simulations sont effectuées pour une condition non drainée pour l'argile et une condition drainée pour le sable. La résistance à la pénétration, la contrainte radiale induite par la pénétration, le soulèvement de la surface du sol et la contrainte de cisaillement du plastique résultant du fonçage de pieux dans l'argile et le sable sont comparés. 1. INTRODUCTION Dynamic pile driving techniques using hammers or vibrators are commonly used to drive piles. However, the dynamic installation could cause noise, vibration, and even slope failure (e.g., of sensitive clay slopes). The jacking of the pile is an alternative approach to avoid these issues, where a pile is pushed into the soil using a jacking machine up to the jack stroke length, usually at a constant rate. A large volume of soil displaces when displacement piles (e.g., concrete, closed-ended pipe and plugged open-ended pipe piles) are penetrated into the soil. The displacement of soil could create a number of issues. If the soil has a low hydraulic conductivity (clay), significant excess pore water pressure generates, which cannot dissipate during the period of installation as commonly practiced. This type of penetration could be considered as an undrained loading case. However, for the highly permeable soils (sand), the generated excess pore water pressure would be dissipated; therefore, the drained soil behaviour governs the installations. The installation of a pile causes a large radial stress increase. At small penetrations, the soil flows outward and upwards, which causes ground surface heave. However, at greater depths, the soil primarily displaces radially; therefore, the cavity expansion theory can be used for modeling the response (Randolph et al. 1979). Using the modified Cam-clay model and assuming the process as an undrained expansion of a cylindrical cavity, Randolph et al. (1979) calculated the radial stress, excess pore water pressure generation, and subsequent consolidation. The concept of cavity expansion has also been used by other researchers (e.g., Basu et al. 2014). The strain path method (SPM) has also been used to simulate the penetration of piles. Sagaseta and Whittle (2001) modeled the ground movement caused by the installation of piles in clay. They have shown large plastic shear strains around the pile, and their developed SPM could handle such large strains. The installation of a pile involves a significantly large deformation of soil. The typical finite element (FE) modeling cannot handle such large deformations; therefore, advanced large deformation FE modeling techniques have been used to simulate this process (Qiu et al. 2011; Ko et al. 2016). Note that the penetration of a pile is similar to cone penetration on a small scale. Some studies focused on analytical and numerical modeling of cone penetration in clay and sand (Teh and Houlsby 1991; Wang et al. 2015). Also, field and small-scale laboratory tests were conducted to understand the mechanisms involved in pile installations (Deeks et al. 2005; Yang et al. 2006). The plastic shear strains generated around the pile could reduce the shear strength of the soil. Moreover, the soil displacement could cause ground heave which could be another design issue. The main objective of the present study to investigate pile installation in clay and loose sand using a large deformation FE modeling technique. 2. PROBLEM STATEMENT A solid pile of 0.4-m diameter (D) is penetrated into the soil at a constant velocity (vp) along the z-axis, as shown in Fig. 1. The bottom of the pile is modeled as a half sphere. At any instant, the depth of the pile tip (point T in the inset of Fig. 1) from the ground surface is denoted as wtip. The pile is penetrated to a maximum depth of 10D. The pile penetration in both clay and loose sand is simulated. Initially, the tip of the pile is placed slightly above the ground surface (wtip = 0.01 mm) to avoid any interaction of the pile with the soil when the soil layer is brought to the in-situ stress condition through a gravity loading step, as discussed later. The groundwater table is considered at the ground surface.

Please include this code when submitting a data update: GEO2019_436

Access this article:
Canadian Geotechnical Society members can access to this article, along with all other Canadian Geotechnical Conference proceedings, in the Member Area. Conference proceedings are also available in many libraries.

Cite this article: