The precise prediction of maximum load carrying capacity of bored piles is a complex problem because it is a function of a number of factors. These factors include method of boring, method of concreting, quality of concrete, expertise of the construction staff, the ground conditions etc. besides the pile geometry. The performance of pile load tests is, therefore, of paramount importance to establish the most economical design of piles especially where bored cast-in-situ piles are to be provided to support a structure.

This paper describes the experience gained from five pile load tests at Greater Thal Canal Project in the Punjab Province of Pakistan. Geotechnical investigations at the site were carried out through a number of boreholes drilled to depths varying between 60 ft and 70 ft below the existing ground level. These investigations revealed that the subsoils consist of silty sand in very loose to medium dense state up to depth of 10 ft followed by medium dense to very dense fine sand up to the maximum explored depth of 70 ft. The historic data on depth to water table expose that the ground water levels are not evenly distributed over the whole project area. The ground water table was encountered at a depth varying from 9 ft to 30 ft. Topography is also characterized by elevated sand dunes and interdunal depressional areas. Five test piles of diameter 2.5 ft and length 55 ft, but in different GWT levels were casted insitu by reverse rotary method and were then loaded by axial compression load to failure.

The load test data were analyzed using various state of the art techniques (intercept of two tangents, point of change of slope, 6 mm net settlement (AASHTO), 90 percent (Hansen 1963), 80 percent (Hansen 1963), limit value (Davisson 1972), Chin (1970). Based on a comparison of pile capacities from these methods with the theoretical values, recommendations are made on the method most applicable to estimate the pile capacity in the local conditions.