Two series of static full-scale lateral pile cap tests were conducted on pile caps with different aspect ratios, with full width (homogeneous) and limited width backfill conditions involving loose sand and dense gravel. The limited width backfills were constructed by placing a relatively narrow zone (3 to 6 ft (0.91 to 1.83 m)) of higher density gravel material adjacent to the cap with loose sand beyond the gravel zone. Test results indicated that large increases in lateral passive resistance could be expected for limited width backfills. The main focus of this study is to assess the contribution of plane strain stress effects and 3D geometric end effects to the total passive resistance mobilized by limited width backfills, using soil and pile cap properties associated with the field tests. For this purpose, the finite element program, PLAXIS 2D was used to investigate the static plane strain passive behavior of the full-scale tests. To validate the procedure, numerical results were calibrated against analytical results obtained from PYCAP and ABUTMENT. The analytical models were additionally validated by comparison with measured ultimate passive resistances. The calibrated model was then used to simulate the passive behavior of limited width gravel backfills. Parametric studies were also executed to evaluate the influence of a range of selected design parameters, related to the pile cap geometry and backfill soil type, on the passive resistance of limited width backfills. Numerical results indicated that significant increases in passive resistance could be expected for long abutment walls where end effects are less pronounced and the geometry is closer to a plane strain condition. Comparisons between measured and numerical results indicated that using the Brinch-Hansen 3D correction factor, R3D, as a multiplier to the plane strain resistances, will provide a conservative estimate of the actual 3D passive response of a pile cap with a limited width backfill. Based on results obtained from the parametric studies, a design method was developed for predicting the ultimate passive resistance of limited width backfills, for both plane strain and 3D geometries.

Compacted fill and rammed aggregate piers (RAPs) were separately installed adjacent to a 9-ft by 9-ft by 2.5-ft driven pile foundation founded in soft clay. The compacted fill used to laterally reinforce an area of 11 ft by 5 ft by 6 ft deep adjacent to the pile cap was clean concrete sand. The thirty-inch diameter RAPs were installed in three staggered rows to a depth of 12.5 ft below the ground surface adjacent to the pile cap to test the increase in lateral resistance afforded by their installation. The foundation was laterally loaded and load, displacement, and strain readings were recorded. The results of this testing were compared with similar tests performed with virgin soil conditions. The total lateral capacity of the pile foundation increased by 5 percent or14 kips due to compacted fill placement against the face of the pile cap. The passive force acting only on the pile cap decreased from 54 kips in the virgin case to 30 kips after installation of the compacted fill, a decrease of about 45 percent. The total lateral capacity of the pile foundation that was retrofit with RAPs was increased by 18 percent or 52 kips as compared to an identical pile cap in virgin clay. The passive force acting on the pile cap at 1.5 inches of pile cap displacement was determined to be approximately 50 kips, showing a slight decrease in passive resistance as compared to the tests performed on virgin soil. Both reinforcement techniques reduced pile head rotation and the bending moments in the shallow portions of the piles.