A series of consolidated undrained triaxial tests with pore pressure measurement was carried out to study the effects of applied back pressure on the shear strength characteristics of a normally consolidated clay. Four magnitudes of back pressure were employed as being representative of a common commercial test situation, the approximate in situ neutral pressure, a pressure that ensured complete sample saturation, and a back pressure much in excess of that required for saturation.The results of the work indicate that, when a low back pressure is used in conjuction with a low consolidation pressure, the soil exhibits an apparent overconsolidation, as evidenced by the normalized stress-strain characteristics, the pore pressure parameter A^^- during shear, the non-linearity of the strength envelopes, and the shapes of the stress paths. The total and effective strength envelopes for samples tested at the higher back pressures show no cohesion intercepts, while under low back pressures appreciable intercepts exist.

View full abstract

A stress-dilatancy relation of sand is proposed on the basis of the experimental fact that the principal stress axes and the principal strain increment axes rotate without complete coincidence with each other during monotonous increace of horizontal shear stress acting on a sample in a simple shear test. The inclination angle ψ of the major principal stress axis to the vertical direction is deduced from a simple relationship ; τ/σ_N=κtan ψ, which is derived from theoretical consideration (Oda and Konishi, 1974) and verified by Cole's experimental results. The value of κ is a material constant for a given sand which is dependent only on the interparticle friction angle. By using this relation, the principal stress parameters such as mean stress (σ₁+σ₃)/2 and maximum shear stress (σ₁-σ₃)/2 are easily deduced from the shear stress τ and the normal stress σ_N acting on the horizontal shear plane in a simple shear test or even in a direct shear test.

View full abstract

The evaluation method of design of embankment is proposed on the basis of cost functions which are formulated in relation to its probability of failure. Four design alternatives are investigated. That is, (1) alternative A₁ is to increase the number of laboratory test in order to get more informations on soil properties, (2) alternative A₂ is to decrease the slope gradient, (3) alternative A₃ is to increase the resisting moment with a counterweight fill and (4) alternative A₄ is to increase the shear strength by improving the poor ground with sand compaction piles.The proposed cost functions and the numerical examples show that the alternative A₄ gives the minimum total expected loss when the cost purchasing land is very expensive, but that the alternatives A₁, A₂ and A₃ tend to be selected for design when the land cost is not so expensive. Further, they show that the acceptable value of the failure probability for design should be reduced when the anticipated failure loss is considered very expensive.

View full abstract

Two methods are presented to study the one-dimensional propagation of seismic shear waves through tapered cross sections of earth dems with truncated crests. The earth dam material is assumed to be unsaturated and linear viscoelastic, or nonlinear strain-softening. The shear modulus may be considered to vary with depth or to be constant throughout the dam cross section. A closed form solution which involves Hankel functions with complex arguments is developed in case the earth dam material is viscoelastic and the dam base is subject to harmonic excitations. The method is extended to cases of random seismic vibrations by employing Fourier analysis in conjunction with a least squares criterion. Response curves are obtained for viscously damped tapered dam cross sections with truncated crests, by using the analytical solution developed. The method of characteristics is also used to provide a solution to the shear wave propagation problem. The method of characteristics exhibits versatility in handling different descriptions of dynamic material response, such as "strain-softening" material behavior laws. Three examples are presented to illustrate the application of the two methods and to demonstrate the relative simplicity and flexibility of the method of characteristics.

View full abstract

The stability of a vertical cut with a variable corner angle is analyzed by use of the upper bound theorem of the generalized theory of perfect plasticity. A three-dimensional collapse mechanism, symmetric in the vertical plane bisecting the corner opening, is assumed. Quantitative results are presented graphically and the physical meaning and general implication of these results are discuseed. Included in this generalized approach is the special case where the corner angle is equal to 180 degrees, for which a solution exists.

View full abstract