As mentioned in the foregoing section, the shadowing phenomenon affects the pile behavior within the group under the lateral loading [6]. Although many researchers have studied the ultimate lateral resistance and deflection of the pile group to a lateral loading, they are complex due to the interaction between the surrounding soil and the pile [7].
In 1962, Prakash carried out the pile group behavior under the lateral loading using aluminum pipes ( = 12.7 mm; = pile diameter) in the medium sand. Based on these tests, it was stated that the sum of pile capacities was more than that within the group when the spacing center-to-center of piles was less than 3 and 8 in the direction perpendicular and the direction to load, respectively. Meyerhof et al. [8] conducted tests in homogeneous sand on pile groups and rigid single pile under central inclined loads. The bored piles were tested by Franke [9] in the experimental tests. The results showed that the displacement of a group was more than a single pile in the same loading when the piles spacing was less than 6. Patra and Pise [10] studied the ultimate lateral resistance on six types of configurations of pile group with different embedment length-to-diameter ratios equal to 12 and 38. Their results were compared with the results of analytical methods. Based on their report, it can be stated that the isolation spacing is six times of pile diameter for .
Kim and his workers [11] investigated lateral load tests on aluminum single pile (driven and drilled) in dry sand. In addition, they considered the head conditions of the piles. The lateral loads of the preinstalled were less than those of the driven piles.
Zhang et al. [12] proposed the ultimate lateral resistance in cohesionless soils. They collected the experimental data done by other researchers on rigid piles and a simple method was developed by them to predict the ultimate lateral resistance (involving of side shear resistance and frontal soil resistance) to piles considering the shape factor. Another method was developed by Prakash and Kumar [13]. In this method, load-displacement relationship was predicted by means of considering soil nonlinearity using subgrade reaction. Erdal and Laman [14] purposed the behavior of short pile subjected to lateral loads in a two-layer sand deposit. The pile modeled had an embedded length-to-diameter ratio of 4 and fabricated from steel for all the tests. Based on their results, it can be stated that the lateral load capacity of short rigid piles in the dense sand was 5 times that in loose sand.
3. Experimental Setup
The schematic diagram of the test setup is shown in Figure 1. The model tests were performed in a rectangular soil tank with dimensions of 900 mm in length, 700 mm in width, and 65 mm in height. To consider the boundary conditions, the size of the soil tank was extended up to 8–12 ( = pile diameter) and 3-4 in the direction and perpendicular to the lateral loading, respectively [15]. In additional, to minimize the influence of box boundaries, the soil thickness was kept below the pile tip at least 6.