Methods for Deriving p-y Curves from Instrumented Lateral Load Tests

With the development of new deep foundation systems and construction techniques, such as high capacity micro-piles, large diameter piles, or drilled shafts, existing p-y criteria that were developed previously for small diameter piles or shafts may no longer be adequate. Deriving p-y curves from lateral load tests on instrumented deep foundations is critical for further refinement or development of pertinent p-y criteria for these new applications. However, there is a lack of a consistent and well verified method for deducing the p-y curves from lateral load tests measurement data. In this paper, 4 existing methods are evaluated using measured results of 8 full-scale field tests on fully instrumented drilled shafts as well as 4 hypothetical numerical simulation results. It is found that the p-y curves deduced by the piecewise cubic polynomial curve fitting method, when input into the LPILE computer program, provides the smallest error on the prediction of deflections of a drilled shaft under the applied lateral loads. A procedure for determining an optimum strain gage spacing for instrumentation in a lateral load test to derive representative p-y curves is recommended. Lastly, a parametric study has shown that the errors of the deduced p-y curves are mainly due to inaccurate moment profiles from strain gage readings. Accurate estimate of moment-curvature relationship of a reinforced concrete shaft is therefore essential to the accuracy of the deduced p-y curves from strain data.