Significant advances in the development of a general effective stress method for the prediction of the axial capacity of driven piles in clay are reported. These advances are in the modeling and prediction of stress changes due to pile installation and due to reconsolidation after installation. An improved methodology results. Preliminary design charts (8 charts) are presented.
An initial development of a general effective stress method for the prediction of the axial capacity of driven piles in clay was presented by Esrig et al (1977). Their methodology is based on modeling, in sequence, the major events in the life of a pile (pile driving, reconsolidation after. driving and axial loading). Such an approach provides insight to the factors controlling pile capacity and should permit better estimates to be made of pile capacity than are currently possible.
Two major uncertainties in the methodology were apparent in 1977: stress changes associated with pile driving were predicted from an analysis of a cylindrical cavity expanded under conditions of plane strain in an elastic-plastic medium; and stress changes, and the stress' path followed, during reconsolidation of the soil surrounding the driven pile were estimated from the results of analysis of a simple, mechanistic model of soil reconsolidation.
Present analyses of the stress changes due to pile driving have continued to utilize the model of a cylindrical cavity expanded under plane strain conditions to simulate pile driving. However, detailed consideration has been given to the use of more realistic stress-strain behavior of soil (Kirby and Esrig, 1979) and/or to a plasticity approach to soil behavior (Wroth et al. 1979) The more refined studies have shown that, with minor modification, the simple elastic-plastic analysis is adequate.
Two independent, and philosophically different, types of analyses of reconsolidation around a pile have replaced the mechanistic model. A closed form, elastic (Biot-type) analysis of stress changes and of the stress path followed during reconsolidation was first generated by C.P. Wroth (1976). This analysis was generalized to permit consideration of materials with radially symmetric but variable elastic module by Leifer et al (1979). The generalized elastic analysis forms the basis of the methodology presented herein. Subsequently, the elastic analysis was extended into the time domain by Randolph, et al (1979). Miller et al (1978) were concerned about modeling soil as an elastic material during reconsolidation and produced a numerical solution for reconsolidation which incorporates the concepts of plasticity and makes use of a model of soil behavior suggested by Roscoe and Burland (1968). This work has been extended by Randolph et al (1979 a and b).
The differences between the results of the two available models are described subsequently and the level of current uncertainty about the stress state after reconsolidation is indicated. Measurements reported by Cooke (1978) are cited and suggest that both analytical models are imperfect. A methodology for predicting axial capacity that is simpler in form than that presented by Esrig et al (1977) is presented herein together with the results of a verification study.
