Abstract
The basis of argument that isotropic nonliniear deformations are made by some theoretical facts and a lot of assumptions, the soils must be treated as essentially anisotropic bodies. From this viewpoint, the stress-strain relationship is verified by an independent principal stress control test.
To improve on some of the inadequacies occurring in the past, a counterbalance is set up as a prop to cancel the mass quantity of the upper-cap. And for reducing the friction of the Bellofrom-Cylinder axis, oil-less metal and teflon-coating is inserted between the cylinder and axis that is directly connected to the plate suppling intermediate principal stress.
In order to define the two coefficients of anisotropic nonlinear deformation, a test was conducted under the condition of σ1>σ2=σ3 in which mean principal stress was maintained at 98, 196, 294 and 392 kN/m2. Five factors (a, α, b, β, n) are fixed due to the results obtained.
Regarding the results of previous test with Toyoura Sand, one matching factor was made use for fi tting the calculation to the data. However, from the results of the calculation with five factors obtained by the present elementary test with cohesive soil, it is not nessesary to use the matching factor. And the proposal equation used for the five factors can expressed the stress-strain relationship in a uniform manner.
