Nonlinear Response of Piles with Rigid Embedment: Lessons Learned from a Reduced-Scale Experiment-Paper

Paper

Pdf Summary

The study reports a reduced-scale experiment conducted at UC Irvine to refine the setup for planned tests on rock-socketed piles and investigate their nonlinear lateral response. The 10-inch-diameter pile was embedded in 52 inches of medium-dense sand over a 40-inch simulated rock socket. The pile was stronger than the 4-ksi socket concrete, allowing researchers to examine whether weaker rock would fail near the soil-rock boundary. The specimen was subjected to reverse-cyclic lateral loading at the exposed pile head and instrumented to measure strain and displacement.<br /><br />In both push and pull directions, yielding began at about 1.3 inches of pile-head displacement, at loads of approximately 3.6 and 3.9 kips, respectively. The maximum push load was 4.2 kips at 7.5 inches of displacement. After yielding, the pile continued to deflect with little further increase in load. Soil densification created a settlement cone around the pile, and excavation revealed substantial horizontal cracking, especially on the pile’s bending sides. The largest crack indicated a plastic hinge about 2.6 pile diameters below the ground surface.<br /><br />Despite the socket concrete being weaker than the pile, no cracking occurred at the pile-rock interface. This contrasts with common analytical expectations that pile shear failure may occur within the socket. The experiment also provided practical lessons for test setup, boundary-condition control, instrumentation, and comparison of measured behavior with analytical predictions.

Keywords

rock-socketed piles

nonlinear lateral response

reverse-cyclic loading

pile-head displacement

yielding

medium-dense sand

simulated rock socket

plastic hinge

pile-rock interface

strain instrumentation