Although it is well established that spacing of piles within a pile group influences the lateral load resistance of that group, additional research is needed to better understand trends for large pile groups (greater than three rows) and for groups in sand. A 15-pile group in a 3x5 configuration situated in sand was laterally loaded and data were collected to derive p-multipliers. A single pile separate from the 15-pile group was loaded for comparison. Results were compared to those of a similar test in clays. The load resisted by the single pile was greater than the average load resisted by each pile in the pile group. While the loads resisted by the first row of piles (i.e. the only row deflected away from all other rows of piles) were approximately equal to that resisted by the single pile, following rows resisted increasingly less load up through the fourth row. The fifth row consistently resisted more than the fourth row. The pile group in sand resisted much higher loads than did the pile group in clay. Maximum bending moments appeared largest in first row piles. For all deflection levels, first row moments seemed slightly smaller than those measured in the single pile. Maximum bending moments for the second through fifth rows appeared consistently lower than those of the first row at the same deflection. First row moments achieved in the group in sand appeared larger than those achieved in the group in clay at the same deflections, while bending moments normalized by associated loads appeared nearly equal regardless of soil type. Group effects became more influential at higher deflections, manifest by lower stiffness per pile. The single pile test was modeled using LPILE Plus, version 4.0. Soil parameters in LPILE were adjusted until a good match between measured and computed responses was obtained. This refined soil profile was then used to model the 15-pile group in GROUP, version 4.0. User-defined p-multipliers were selected to match GROUP calculated results with actual measured results. For the first loading cycle, p-multipliers were found to be 1.0, 0.5, 0.35, 0.3, and 0.4 for the first through fifth rows, respectively. For the tenth loading, p-multipliers were found to be 1.0, 0.6, 0.4, 0.37, and 0.4 for the first through fifth rows, respectively. Design curves suggested by Rollins et al. (2005) appear appropriate for Rows 1 and 2 while curves specified by AASHTO (2000) appear appropriate for subsequent rows.



College and Department

Ira A. Fulton College of Engineering and Technology; Civil and Environmental Engineering



Date Submitted


Document Type





p-multiplier, pile group, sand, group interaction, group effects, deep foundation, full-scale lateral load test