The plastic work done during the shearing deformation of granular materials such as sand and glass beads depends largely on stress paths. A new parameter S_s which is defined by S_s=∫(dW ^p_s)/P dW ^p_s : increment of plastic work done due to shear P : mean principal stress is independent of stress paths and the parameter S_s can be regarded as an interesting measure of procession of the shearing deformation.

View full abstract

In designing the foundations for large-scale structures, it is important to take into account the scale effect of bearing capacity of foundation soils. In this paper the mechanism of the scale effect of bearing capacity of dense sand is investigated for shallow foundations. A series of loading tests using Toyoura sand are carried out in a centrifuge by which stresses in small-scale models can be made approximately identical to those in actual foundations. Deformations of the sand measured by a set of electronic devices and slip lines in the sand mass are observed immediately after the peak load by an X-ray apparatus. It is found out by these observations that shearing strains in the foundation sand differ very much from place to place, which leads to a conclusion that the progressive failure generally exists in dense sand and that the assumption of constant shearing strain, which has been adopted in existing bearing capacity theories, cannot be accepted except for some limited cases. Kotter's equations are solved by combining a unique relationship between the angle of shear stress and shearing strains for the sand, obtained from plane strain shear tests, with observed shearing strains in models. Resulted bearing capacities compare well with measured values for the sand in which shearing strains differ from place to place along the slip line. It is also shown that the rigidity of specimen containers gives significant influence to the magnitude of observed bearing capacities.

View full abstract

Elastic and elastic-plastic problems in soil mechanics have been frequently analysed numerically by the finite element method. Most of the problems were analysed in terms of total stress and a few in terms of effective stress and pore water pressure. The aim of this study is to obtain the accurate solutions of immediate and consolidation deformations of anisotropic and heterogeneous poro-elastic media introducing the concept of effective stress by the finite element technique.Firstly, this paper describes the principle of the finite element method in the problem of consolidation accompanying with undrained deformation and the procedure of analysis. It is especially emphasized that the interpolation functions for the pore water pressure must be quadratic in space and those for the displacement must be cubic. Secondly, the deformation and the distributions of the pore water pressure or the stresses in the isotropically homogeneous and elastic foundation layer just after the application of a uniform strip load and during the consolidation process are computed as an example. The superiority of the numerical technique is ascertained by comparing the results with those based on total stress analysis.

View full abstract

Prediction of the stability of slopes under various water conditions and for a wide range of loading and soil conditions can be made with the help of a digital computer. Reservoir slopes may reach a very critical state immediately after very rapid drawdown. At present, this condition is analyzed assuming an "instantaneous" drawdown event. This paper presents a method for obtaining the slope stability under a drawdown condition with a defined time-elevation history.The analyses are based on the simplified form of Bishop's model with circular failure surface and vertical slices. Two options are available for determining the pore pressure. In the time-dependent drawdown analysis, a differential equation equates the volume of water that flows out of the slope in a given time with the product of the velocity of flow, area of flow, and time interval. The solution of this equation yields the position of the phreatic surface with time. This information helps determine the pressures along a trial circular sliding surface, and the variation of the factor of safety with time.The factor of safety decreases rapidly with the lowering of the reservoir level, followed by a slow and nonlinear increase to the steady state value.

View full abstract

An extensive number of penetration tests were performed in a laboratory model. The results obtained were compared with field penetration tests in three sites of different characteristics with a view to obtain a general relationship between various penetration devices which are to be used in evaluating the strength of sandy soils.For laboratory tests various sizes of tanks filled with sands of various grading characteristics and of controlled densities were used. The tank used for carrying out the bulk of laboratory tests was not the largest one tried by the author. In fact it was the medium sized tank of 90cm diameter out of the trial tanks used (70, 150 and 200cm diameter). But the results obtained from preliminary investigations showed that there existed a fixed relationship between the findings in this tank and the largest tank (200cm diameter). This latter tank showed from pilot tests to be the most suitable as far as boundary effects concerned. But due to the inherent difficulty in controlling its density through the large number of tests carried out, it was not used. Results obtained from it regarding the relationship observed in controlled pilot tests were used to scale out the 90cm diameter tank results using a correction factor. This factor is deemed to cover for boundary effects in the tank.Laboratory and field results were compared in order to explore the effect of difference in structure between laboratory-filled sands and natural deposits. Besides, the effect of "aging" on laboratory results is clarified.Dynamic penetration tests included the standard spoon test as well as dynamic cones of 1.4, 2 and 3 inches diameter which were driven in a similary way as the spoon.New statistical formulas were derived between various factors which influence soundings in sandy soils.

View full abstract

Cyclic triaxial shear tests were performed on granular materials with a variety of grain sizes and uniformity coefficients to determine the effect of grain composition on the liquefaction potential of cohesionless soils. In place of the conventional measure of relative density as a density index the difference between the current and minimum void ratios defined as the volume decrease potential was used as a new density parameter to define liquefaction potential in a more meaningful way.It was found that the stress ratio required to cause liquefaction in a given number of cycles is intimately related to the volume decrease potential. To aid in using this concept, design curves are presented that may be used to estimate values of this potential from field Standard Penetration Test results.

View full abstract