Soil Structure Interaction and Group Mechanics of Vibrated Stone Column Foundations

The development of transparent synthetic soil has enabled non-intrusive measurement of internal soil displacement and detection of pre-failure strains using laser aided imaging and Particle Image Velocimetry. The work described in this paper applies this novel methodology to evaluate the deformation and failure behaviour of stone column foundations in reduced scale physical models. The overall accuracy and precision of the test system were optimised to 53 μm and 0-4.3μm respectively through the implementation of newly developed image enhancement and photogrammetric correction techniques. This represented a considerable improvement compared to the previous state of the art in transparent soil modeling. A total of 7 isolated column tests and 6 column strip footing tests were carried out in repeatable transparent clay beds. Visualisation of real-time internal displacement has indicated that isolated stone columns fail in an axisymmetric manner through a combination of compression and bulging with only a small amount of column base penetration, indicating a critical column length of 4d. Consequently, increasing column length from 4d to 6d and 8d lead to only moderate increases of 8.5% and 13% in bearing capacity respectively In contrast to the isolated columns, local shear failure, bulging, bending, punching and block failure were all observed for column strip foundations, depending on the geometrical configurations employed. Column length played a particularly important role in optimising the bearing capacity of column strip foundations. Increasing column length from 4d to 6d and 8d was shown to improve the load capacity of the composite foundation by 29% and 67%. It was concluded that the critical length in terms of optimising the bearing capacity of a vibro strip foundation is 8d. However, increasing foundation size rather than column length was found to be more effective in increasing the load capacity in both cases. Increasing footing size and correspondingly increasing Ar from 3 to 4 and 5 was seen to protect the upper column from bulging and increase its radial zone of influence of the column in the surrounding soil, resulting in increases in load capacity of 28% and 51% for isolated columns and 22% and 29% for strip foundations respectively. It was concluded that provision of a footing overhang of over 0.4d should prevent shear failure in vibro strip foundations. A new design approach has been proposed which accounts for the size of the footing. The performance of each isolated column foundation was predicted to within 6% of the measured load capacity for all area ratios tested, while the load capacities of the strip foundations were predicted to within 14%. Thus although the new method does not model the ‘group effect’, it offers an improvement on cavity expansion based design methods, as column structure interaction is accounted for. Overall, the nature of the configuration of vibro strip foundations and their limited size in practice limit the bearing capacity of stone columns within a strip configuration to values only slightly over that of a single column, if at all.