Effect of Installation Method on the Ratio of Tensile to Compressive Unit Shaft Resistance of Piles in Dense Sand

It is generally accepted that the unit shaft resistance of piles in sands is lower for tensile loading than for compressive loading. So far, very little attention has been paid to the influence of the installation method on the tensile-to-compressive unit shaft resistance ratio. The objective of this paper is to analyze the effect of installation method on the ratio of unit shaft resistance in tension to that in compression for a model pile installed in a dense silica sand. To study installation effects, load tests on model piles were carried out in a half-cylindrical calibration chamber with digital image correlation (DIC) capability. The model piles were either jacked in one stroke using a constant rate of advance, or pre-installed into the dense sand samples. Digital images of the model pile and sand were taken during the load test and processed using DIC to obtain the displacement field in the sand. It was found that unit shaft resistance was essentially the same in tension and compression for the pre-installed pile, however for the jacked pile, the unit shaft resistance in tension was approximately half of that in compression. DIC analysis revealed that: (1) during the load test, more displacement occurs in the soil domain for the pre-installed pile compared to the jacked pile, (2) for the jacked pile, the magnitude of displacement in the soil domain is greater in tension than in compression and (3) for the pre-installed pile, the magnitudes of displacement in tension and compression are similar. INTRODUCTION Pile foundations can be categorized into two main types depending on the method of installation: displacement piles (e.g., jacked or driven piles) and non-displacement piles (e.g., bored piles or drilled shafts) (Basu et al. 2010; Salgado 2008). The response of these piles to loading can be studied in the laboratory by means of calibration chamber experiments on model piles that are either jacked into the samples, representing displacement piles (Jardine et al. 2005; Lee et al. 2011; Tovar-Valencia et al. 2018), or pre-installed in the sand sample, representing non-displacement piles(Galvis-Castro et al. 2019a; Le Kouby et al. 2013; Tehrani et al. 2016)