Some Basic Problems Of O-CELL Pile Test

As a new pile test technique, O-cell test of pile has been applied in many projects at home and abroad. But some theory problems of O-cell test technique have not been summaried systematically and solved. It is a fundamental problem about O-cell pile test how to determine integer pile vertical behavior based on the O-cell test results of piles. Also, this problem is hot and hard in researches of O-cell pile test. The solution of this problem deal with several basic research of O-cell pile test problems: In order to determine the resistance parameters of pile side and tip, how to analysis and apply results of O-cell pile test; How much the interaction between upper pile and lower pile in O-cell test and how to study the interaction between upper pile and lower pile; For upper pile, load acts on pile bottom pushing upper pile up in O-cell pile test, but load acts on pile top pressing down in conventional pile test, loading mode how to influence on upper pile side resistance, and how the influence to be considered; How to utilize results of O-cell pile test to determine the load-settlement curve of integer pile reasonably.In order to study above problems, the simulation analysis models of O-cell pile test and conventional pile test are set up in this paper. The models consist of pile, soil and pile-soil interface. FEM is used to analysis pile-soil system behavior. Pile is regarded as linear elastic. Duncan-Chang model is adopted to consider nonlinear behavior of soils. Nonthick Goodman element is adopted, and the nonlinear relationship between the shear stress and the tangential relative displacement on pile-soil interface is expressed with the hyperbolic curve. Top and bottom of load box are treated as load boundary in O-cell pile-soil system.The approach fitting the load-displacement curves and the axial force curves of upper pile and lower pile obtained from the O-cell test of pile is presented. The principle and detail adjusting parameters in process of fitting curves is also presented. With fitting and analysis of O-cell pile test data of a case, the approach to determine resistance parameters of pile side and tip in this paper is verified to be feasible.With the models of O-cell pile test, the upper and lower pile behavior are analyzed with numerical simulation method under the three cases: loading upper pile and lower pile simultaneously,loading lower pile individually,and loading upper pile individually . Compared stress and deformation of pile and soil in these three cases, the interaction between upper pile and lower pile is analysised, and the degree of the interaction influence on upper pile and lower pile is studied. The results show that if load box is in relatively hard soil the interaction between upper pile and lower pile in O-cell test is so little that it can be ignored. In fact, load box usually is in relatively hard soil in engineering. The results also show that site soil condition and pile length and pile diameter have such little influence on interaction between upper pile and lower pile in O-cell test as it is can be ignored.Without the pile tip resistance, the pile behavior under load of pressing down at pile top, pushing up at pile bottom and pulling up at pile top are analyzed with numerical simulation method. The influence of loading models on pile side resistance is studied same time. The comparing results of three loading models show that the additional pile lateral stress ratio ? induced by load of pulling up at pile top and pushing up at pile bottom is minus, that mean total pile lateral stress decrease; the additional pile lateral stress ratio ? induced by load of pressing down at pile top is plus, that mean total pile lateral stress increase. The influence of soil type and pile length and pile diameter on additional pile lateral stress ratio is also studied. Base on above results, it is accepted that difference of the additional pile lateral stress induced by load models is a fundamental reason of different pile side resistance under different loading model.Base on the above research about different side resistance behavior under different load model, a model in which interface lateral stress influence on interface mechanics parameters is considered. In this model the ultimate shear stressÏ„_ult and the max tangential stiffness K_Ï„,max in loading process are related with pile lateral stress ratioα. Equivalent lateral stress ratio is the key parameter of this model. The method by which equivalent lateral stress ratio is determined based on the equivalence of failing relative displacement is provided. With the model, the interface parameters under pressing down at pile top can be got from those under pushing at pile bottom load. Reverse procedure is feasible. Loading model influence on pile side resistance can be considered in this model. With data of pushing at pile bottom, the behavior of pile under press at pile top is simulated and analyzed. Result of analysis shows that the bearing capacity under loading press at pile top is higher than that under pushing at pile bottom, and the increase extent is agree with practical experience.In this paper, considering loading model influence on upper pile side resistance, two approaches determining integrate pile P-S curve based on the results of O-cell pile test are presented. The O-cell pile test results of a case are analysised. It is verified that the approaches presented in this paper are feasible. The complete integrate pile P-S curves, which includes linear part, transition part and failure part, can be provided by the methods presented in this paper. The P-S curves can be used to determine bearing capacity of integrate pile by code criterion. The curves determined by methods presented in this paper are also compared with ones determined by other methods. As the compression of upper pile does not been considered,the settlement of the curves determined by Equi-displacement method and Equi-load methods are underestimated. The P-S curve determined by simplified method and accurate method are close to according segment of P-S curve determined by method A and method B. But the conclusion need be checked further. Integrate pile P-S curves from simplified method and accurate method only include linear segment and partial transition segment. In theory, Method B is more reasonable than method A. It also can be considered the influence of soil type pile length and pile radius on integrate pile P-S curves. The bearing capacity determined by method B is 10% more than that determined by method A. The integrate pile bearing capacity is underestimated by popular formulas of determining the bearing capacity of integrate pile from O-cell test results, about 20³0%.