INFLUENCE OF PILE GEOMETRY ON THE DYNAMIC RESPONSE OF UNDER-REAMED PILES

Abstract

Foundations form one of the most important components of a building. If the soil in the top layers does not have an adequate bearing capacity then the building loads need to be transferred to a lower strata using deep foundations i.e. piles etc. If the top soil is expansive in nature (black cotton soil) and a hard strata is not available then under-reamed piles are used. Under-reamed piles are bored cast-in-situ concrete piles having one or more bulbs formed by enlarging the pile stein using a suitable cutting tool. These piles are suitable in expansive soils, which are often subjected to considerable ground movements due to seasonal moisture content variations. They are also used in filled-up deposits and loose or soft strata. Under-reamed piles are also used in foundations for transmission towers etc. In this study, in-situ tests were conducted on a set of under-reamed piles in a silty sand deposit. Three 200mm diameter piles were cast upto a depth of 5 metres; the first pile was a normal pile with no under-reaming, the second pile had one bulb at 430mm from the bottom and the third pile had two bulbs, one at 430mm from the bottom and the second at 1430mm from the bottom. The pile cap was 700mm x 700mm x 300mm (height) and had four foundation bolts cast into it so that a mechanical oscillator-motor assembly could be mounted centrally on it. The oscillator-motor assembly was used to generate purely sinusoidal dynamic forces, either in the vertical or in the horizontal direction. Different force levels were achieved by changing the eccentricity settings within the oscillator and the motor operating frequencies, which were altered using a speed control unit. Accelerations were measured using suitably placed accelerometers on the pile cap top for vertical vibrations and along the height of the pile cap for horizontal vibrations. Forcing frequency versus displacement amplitude plots were obtained separately for the vertical and horizontal vibration modes and from these the resonant frequencies were computed. It was observed that (i) the natural frequencies decreased with an increase in the excitation force level (ii) for the same excitation level, the natural frequencies as well as the maximum amplitude of vibration are almost the same irrespective of the pile geometry i.e. number of bulbs in the pile, indicating that the fidl pile length is not being excited. An attempt was also made to compute the soil-pile stiffness and damping ratios for both vertical as well as horizontal modes of vibration, for both pre-resonance and post-resonance conditions. Under vertical vibration both stiffness and damping ratio values were different for the pre-resonance and post-resonance conditions. For horizontal excitation the values for stiffness and damping ratio are almost similar for the pre-resonace and post-resonance conditions.