Experimental investigations and analysis on different pile load testing procedures

The paper presents experimental results of two Osterberg’s cell load tests (OLTs) and three conventional load tests (COLTs) in the same subsoil conditions on Continuous Flight Auger (CFA) piles carefully monitored during construction stages. The instrumentation along the pile shaft in all the tests allows interesting comparisons of both global behaviour and local load transfer. Significant differences in the stiffness of the soil-pile system with the different test procedures is outlined. The main differences between the two test procedures occur at the two opposite ends of the pile, as could have been expected, while the observed behaviour in the middle part of the tested piles is close for the two models. A relatively simple FEM model has been calibrated on the basis of the OLTs results. The same model is capable of accurately matching the experimental results of the COLTs, proving that the observed differences are not due to random factors. Furthermore, the same model has been used to simulate ideal load tests. Such a reliable simulation shows that both the experimental procedures are actually responsible for significant differences in the behaviour of the soil-pile system even in the simple case of a concentrated axial load. Large differences arise in terms of the stiffness of the system with the OLTs providing by far the stiffest response. Despite being intermediate between the OLT and the ILT, the COLTs provide a response of the pile-soil system, which is on the average about two times stiffer than the Ideal test, where the force applied on top of the pile does not depend on a tangible reaction system. Care should be thus taken when considering the results of such tests in the prediction of the settlement of a piled foundation. Correction factors should be applied to the experimentally observed behaviour.