Reo IKEDA

Eccentric Vertical Loading Bearing Capacity of Direct Foundation in Sandy Ground

Satoru OHTSUKA

Existing bearing capacity formulas have some problems, such as the lack of quantitative study of the size effect of footing on bearing capacity, and the fact that only simple cases of combined loads were considered. The bearing capacity equations in the specifications for road bridges are developed based on large-scale model experiments, so their applicability to complex loading schemes and soil conditions is not clear, especially for surface footings subjected to eccentric loading, for which there are fewer experiments than for centric vertical loading. 
In this study, the authors focused on numerical analysis, which can be applied to complex loading conditions, and investigated the ultimate bearing capacity of footings subjected to eccentric vertical loads in sandy soils using a two-dimensional analysis based on the rigid-plastic finite element method (RPFEM). However, the applicability of the bearing capacity equation proposed by the numerical analysis using RPFEM needs to be verified by model tests. Since the comparison with centrifugal model tests in the existing literature showed good results for the central load, but eccentric loading tests have not been studied, this study employed small model tests (eccentric vertical) and numerical analysis using RPFEM to compare the ultimate bearing capacity. Model tests and numerical analyses were conducted for sandy soils of different relative densities, and numerical analyses were conducted for Mohr-Coulomb's linear behavior model and nonlinear strength model. The nonlinear strength model can reasonably explain the size effects of footing on ultimate bearing capacity for sandy soils of different relative densities based on comparisons with centrifuge model tests with different footing widths.
In the experiment, deformation occurred the test and strongly affected bearing capacity estimation due to the embedding effect. Therefore, in cases where it is difficult to directly evaluate the bearing capacity, it was estimated from the yield stress value. The comparison in the normalized bearing capacity vs eccentricity ratio between model tests and numerical analyses using the M-C model was good and proved to match with the existing guidelines. In the case of the nonlinear strength model, the numerical analysis was somewhat unstable, however, the results were obtained to be almost consistent with the existing guideline, which is valuable and unique result.