SEEPAGE FAILURE EXPERIMENTS AND THEIR ANALYSES OF LOOSE AND MEDIUM DENSE SANDS

Abstract

Failure problems of saturated soils are classified in the present study into the following four types : Type I : The failure of loose and/or normally consolidated soils under fully drained loading conditions. Type II : The same soils but under perfectly undrained conditions. Type III : The failure of dense and/or overconsolidated soils under fully drained conditions. Type IV : The same soils but under perfectly undrained conditions. The soil-water coupling limiting equilibrium analysis on the basis of the critical state concept is shown possible to draw a distinction between type I, II and IV problems, and the limitations of the analysis procedure happen in solving the type problem in the above. This is demonstrated experimentally, in which seepage failure experiments in laboratory with the use of a saturated silty sand are employed to make distinctions clearly between the four types of failure problems. The definite distinctions both in failure load and failure mode appeared in the four types of experiments are summarized as follows : (1) the smallest failure load in loose sand in the undrained condition, (2) the largest failure load with the largest failure region in dense sand under undrained loading, (3) the observation of no global deformation before failure within the soil under the fully drained condition that suggests the development of very localized shear deformation in the sand. Three out of four experiments are described well by the limit analysis computation mentioned above as far as the failure load and the shape and the size of the failure region after boiling failure are concerned, and thus the type III problem is identified to be the problem that should be solved by any other means in near future.