NUMERICAL ANALYSIS OF THE OSTERBERG CELL LOADING TEST

摘要

Pile loading tests are carried out with two main objectives: 1) evaluating the ultimate resistance; 2) checking the behaviour of the pile under live load. In the first case and particularly for large diameter bored piles the cost of a loading test may be very high. In the late '80s, a new method to perform static loading tests was proposed. It is based on the use of a sacrificial hydraulic jack placed at the pile toe. The device, internally pressurized, expanding applies an upward force on the pile shaft and an equal but opposite downward force on the soil beneath the pile toe. This device is well known as Osier-berg's load cell, shortly O-cell. The obtained results are typically summarized in two separate load-movement curves: one for the shaft side shear resistance and another for the toe resistance. The two curves are used to build what is generally named the "equivalent head load-movement curve" relying upon two hypotheses: 1. the pile is rigid; 2. the load movement relationship for the shaft is not affected by the direction of movement. The former assumption can be easily removed, while it is not possible to remove the latter. On the validity of this latter hypothesis, comparisons between conventional head loading tests and toe loading tests could be very helpful. Unfortunately in the literature there are only a few case histories arid most of them are also poorly documented. In order to explore limitations and potentialities of the new testing method, a finite element model has been set up. Results of simulations of conventional head loading tests and of O-cell tests are analyzed, discussed and compared.