Many important buildings and bridges are supported by foundations consisting of groups of steel pipes or piles driven into the ground. The horizontal resistance provided by these groups of piles is critical in determining whether or not the structures will survive an earthquake without significant damage. Although fairly reliable methods have been developed for predicting the horizontal resistance of single piles under slowly applied loads, there is very little information to guide engineers in the design of closely spaced pile groups, particularly under rapidly applied loads. Nevertheless, the data from these limited field tests indicate that piles in groups will undergo much more movement and higher stresses for a given load per pile than will a single isolated pile. These pile group effects are commonly accounted for in design by using reduction factors to reduce the resistance provided by the soil on the piles in the group, but there is considerable uncertainty regarding appropriate reduction factors. The proposed research study has the following objectives: (1) Evaluate the effect of pile spacing on measured group reduction factors and develop a design curve for these factors as a function of pile spacing, (2) Determine the validity of the reduction factor concept for a large (5-row) pile group and determine if the reduction factors remain constant beyond the third row, (3) Determine the effect of repeated loading and gap formation around the piles on the measured group reduction factors, (4) Examine the effect of loading rate on the resistance provide by the soil around the pile, and (5) Provide a well-documented case history for use in evaluating and calibrating computer and physical models. These objectives will be accomplished by conducting a series of horizontal load tests on a full-scale pile and pile groups. These tests will supplement pile group testing already conducted at the site with pile spacing of three feet on centers. First, horizontal load testing will be performed on a single 12-inch diameter steel pipe pile for comparison purposes. Load will be applied in 10 increments with 15 cycles per increment to simulate the repeated loading from a large earthquake. Next, horizontal load tests will be conducted on a five-row pile group spaced at four feet on centers. The same cyclic loading procedure will be applied and the load carried by each pile will be measured. The load frame will then be moved to an adjacent pile group for testing at six feet spacing. In addition to the slowly applied loads, load will be applied rapidly using a rocket sled (Statnamic device) which will simulate the speed at which earthquake loads are applied. Following data reduction, the test results will be analyzed using available computer models and appropriate pile group reduction factors will be determined as a function of pile spacing. The results of the load tests from this and previous work at the site will also be provided to several researchers in Japan for detailed analysis using sophisticated 2-D and 3-D computer models. These analyses should help quantify the increased resistance seen during rapidly applied loads relative to slowly applied loads. Potential Japanese researchers will include Prof. Matsumoto of Kanazawa University, Dr. Susumu Iai of the Japanese Port and Harbor Research Institute, and other researchers expressing interest in analyzing the results.
