This paper attempts to propose an appropriate technique for precisely estimating in-soil shearing resistance of geogrids based on the test results obtained using two types of direct box shear apparatus. A series of drained direct shear tests changing the configurations of soil and geogrid, roughness of substrate material on which the geogrid is glued, and the distance between the surface of geogrid and the shear plane was performed. Based on the test results, the authors recommend the followings as the appropriate testing method : 1) an apparatus in which normal stress can be applied over a flexible boundary and be measured at the opposite side of the loading system should be used ; 2) the test specimen of the geogrid glued on the substrate should be placed under the sand layer and the surface of the substrate should be installed in a suitable position taking into account potential shear surface ; 3) a suitable substrate material should be used taking into account the mode of interaction between soil and geogrid in situ.

View full abstract

The behavior of the self-weight consolidation of suspension has been studied by the Oxford group (Been and Sills, 1981). However the deformation behavior of slurry-like soil upon additional load is not well understood. Due to the high demand for land in coastal areas and surrounding city areas, reclamation or remediation of land on ultra soft soil like waste ponds and mine tailing ponds has become necessary. Materials in such waste ponds and mine tailing ponds are extremely soft, and high water content which is usually above the liquid limit. In addition to this, these materials are in an underconsolidated condition and are still undergoing self-weight consolidation. A laboratory study with a large diameter consolidation cell equipped with pore water pressure transducers was carried out to find out the deformation behavior of slurry-like soil under a one dimensional vertical load. It was noted that before commencing Terzaghi's consolidation process, the slurry was deformed solely due to the expelling of water from the slurry. This deformation was as high as 19% strain and no pore water pressure dissipation was detected during this period. The behavior in the laboratory results was confirmed by a case study carried out with a trail embankment on slurry-like soil contained in the containment bund.

View full abstract

The response of skirted foundations to combined vertical (V), moment (M) and horizontal (H) undrained loading is investigated using finite element and upper bound plasticity analysis. The study has focused in particular on the effect of the foundation embedment depth on the size and shape of the combined V-M-H yield locus. The yield locus was found to be of similar shape to that of a surface foundation but the size of the yield locus was increased. Normalisation of the yield locus by the single load bearing capacities (V₀, M₀ and H₀) allowed the yield locus to be generalised for varying soil profiles and footing embedment depths. Upper bound plasticity analysis was carried out to calculate V₀, M₀ and H₀ for a range of footing embedment depths and undrained soil shear strength profiles and design charts are presented for their determination.

View full abstract

On the basis of the observed experimental behavior of axially loaded friction piles driven in soft clays, a soil-pile interface model is proposed using the basic rheological units : springs, dashpots and sliding elements. All the mass of soil affected by the presence of the pile is concentrated in a ring of infinitesimal width. Therefore, all the phenomena regarding the soil-pile stress distribution and displacements will take place in this surface. The proposed rheological model is introduced in a general boundary elements formulation for axially loaded piles. In this formulation, the pile is discretized in a certain number of elements in order to consider a variety of phenomena such as : the deformation of the pile, the increment of radial and axial stresses in the soil mass due to load transfer and the stiffness of the different layers it passes across. Parameters of the rheological model can be obtained from a creep triaxial test and a direct shear test. In case the pile cuts through different layers, it is necessary to compute the parameters for each layer. The proposed model is able to simulate the load-displacement behavior of axially loaded floating piles subject to monotonic as well as cyclic loading. It is also possible to determine the load and displacement at the pile tip and shear stress distribution along the pile shaft. Results of the proposed model have been compared with a very well documented series of pile tests carried out in Mexico City. Tested piles were squared 30 cm on the side and were driven 10 m in a quite homogeneous clay layer. From these comparisons it can be concluded that the model is able to adequately reproduce the most important aspects of the load-displacement behavior of friction piles driven in clay deposits.

View full abstract

The characteristics, composition, and spatial variability of naturally occurring cemented sand deposits in arid lands were examined, by extensive field and laboratory tests at two sites in Kuwait. Field tests included penetration tests, pressuremeter, and plate loading tests. Laboratory tests included basic properties, unconfined compression, triaxial tests, consolidation tests, chemical, and mineral composition. The results indicated very competent non-homogene-ous deposits with variable degrees of cementation, or with cemented and uncemented layers or bands. In general, the strength and the stiffness increased with depth below the ground surface. The disturbance of these soils, and the crushing of the cementation bonds, lead to increased compressibility and decreased strength due to the loss of the cohesion component of strength.

View full abstract

The turnability characteristics of a flexible tracked vehicle under traction on a loose sandy flat surface are investigated theoretically and experimentally for both the inner and the outer track while being driven. Based on soil mechanics relationships between the surface and the flexible track belt, the turnabilities of the inner and the outer track are considered looking at the surface shear resistance as a function of lateral and longitudinal slip velocity and the amount of slippage. Using terrain-track system constants, the relationships of the amount of depression, the thrust, the compaction resistance of the inner and the outer track, the effective tractive effort, the turning moment of a given tracked vehicle, and the turning radius are predicted for several kinds of steering ratio. From the results, it is determined that the vehicle speed decreases with the increment of turning radius along with the increment of resultant effective tractive effort, and the slip ratio of the outer track is always larger than that of the inner track. The amounts of depression of the front idler of the inner and the outer track decrease with slip ratio while those of the rear sprocket increase. The turning resistance moment of the inner track is always larger than that of the outer track for each surface shear resistance.

View full abstract

In order to establish a physical model leading to stable (uniform) and unstable (non-uniform fingering) flows in unsaturated soils, two-dimensional tests on continuous rainfall infiltration were carried out by changing not only the initial water content (air dry, 0.5%, 1.0%) but also the rainfall intensity (15, 30, 180 mm/h). The physical mechanism of fingering flow was discussed on the basis of precise suction measurements inside and outside the fingers. The results show that wetting fronts developed in unsaturated sandy soils can be classified into three types ; 1) fingering flow, 2) wavy front, and 3) plane front. Two types of fingering flow can be further distinguished by the difference of the swelling velocity ; i. e., low-swell finger and high-swell finger flows. In each of the low-swell and high-swell fingers, a core and a swelling zone are developed while a finger is growing. Based on suction measurements inside them, it is found that the water condition in the finger core changes from wetting to drying processes while the water condition in the swelling zone remains in the wetting process. This hysteretic behavior in the core can be explained by the moisture distribution changing along a finger from tip to tail. Fingers swell when water moves from a finger core to surrounding dry sand. Such a swelling process is mainly controlled by the hydraulic conductivity for the wetting process at a finger tail.

View full abstract

This study aims at numerically simulating the bearing capacity characteristics of strip footing on sand, and consequently explaining the scale effects observed in a series of plane strain model tests carried out on a particular type of sand (Toyoura sand). The model tests were performed using different sizes of footing, with the largest width being 50 cm, under normal gravity and in a centrifuge. A constitutive model developed for Toyoura sand based on the results from an extensive series of plane strain compression tests is used. A number of factors which affect the strength and deformation of sand are taken into account, including ; 1) confining pressure ; 2) anisotropy ; 3) non-linear strainhardening and strain-softening ; 4) dilatancy ; and 5) strain localization into a shear band (s). This material model is coupled with an isotropically hardening, non-associated, elasto-plastic material description. A widely used numberical technique, FEM, is applied to solve the non-linear equations. A parametric study is performed to evaluate the effects of using different assumptions for the material model. The simulation of bearing capacity compares well with the physical model test results. It is shown that the isotropic perfectly plastic modelling of soil property, assumed in most of the classical bearing capacity theories, is an overly simplified approximation to be used in FEM analysis of this issue. It is explained that the scale effect consists of the pressure level effect and the particle size effect.

View full abstract

The stress-strain relationships for soils and soft rocks have been studied in triaxial and plane-strain compression tests for a wide range of strain levels by Tatsuoka and Shibuya (1991), who demonstrated that the hyperbolic model does not satisfactorily predict the overall stress-strain behaviours of various geotechnical engineering materials. A new simple non-linear model is proposed by the Authors which requires three materials parameters, all of which are easily determined from laboratory tests and some of which are from in-situ shear wave velocity measurements. Calculated performance is in excellent agreement with precise laboratory measurements of stress-strain behaviour over a wide range of strains. This new model is used to study the problem of an axisymmetric excavation. It is demonstrated that the problem of stress redistribution is critically dependent on the assumed stress-strain behaviour, particularly at small strains.

View full abstract

A round robin test to determine the cyclic undrained triaxial strength of a silty clay called Fujinomori clay was performed with the participation of 26 laboratories in Japan. One laboratory performed the cyclic simple shear test under constant volume and the data was used for the comparison with triaxial strength. The clay was distributed in the form of air-dried powder and preconsolidation was carried out from the slurry state in each laboratory under a prescribed specification. The clay specimens thus formed were subjected to cyclic undrained triaxial tests and the cyclic strengths thus obtained were compared. Static undrained triaxial tests were also performed at each laboratory. It was found that the scattering in the cyclic undrained strength data is quite small if the testing procedure and conditions are strictly specified. A very good agreement was obtained for the cyclic undrained strength data normalized by the corresponding static undrained strength. Lastly the cyclic undrained strength is compared with the data reported by other researchers.

View full abstract

In soil dynamics, the most common parameter for measuring soil damping is the equivalent viscous damping ratio, which is derived from the theory of a single-degree-of-freedom system with viscous dashpot. The results of a comprehensive literature review on the equivalent viscous damping ratio at small cyclic shear strain amplitudes, γ_c, between 0.0001% and 0.01% are presented and discussed. On the basis of the published data from different types of laboratory tests two important trends are established. The first is a significant increase of the scatter of the data points as the plasticity index of the soil, PI, decreases. This indicates that damping in low-plasticity clays and sands is more sensitive to factors such as confining stress, overconsolidation ratio and cyclic loading waveform than the damping in high-plasticity clays. The second trend is an increase of the damping ratio with PI. This trend with PI is opposite from the trend at γ_c>0.01% established earlier, where damping ratio generally decreases with PI. An explanation is provided that the phenomena responsible for the increase of the damping ratio with PI at small γ_c are viscous creep and relaxation of the soil.

View full abstract

Sorption behavior of simazine in whole soils and four selected soil particle size fractions of two Hiroshima soils, a loam and a decomposed granitic (DG) soil, was measured during two-step sorption experiments. In the first sorption step, adsorption kinetics were followed for 36 hours in initially simazine-free soil samples. A rapid adsorption was observed for all soil particle size fractions after 10 minutes. Between 10 minutes and 36 hours adsorption slowly progressed in all loam size fractions while adsorption stopped in all DG size fractions. In the second sorption step, 70% of the solution in the soil samples was substituted with simazine-free solution and sorption behavior was followed for an additional 36 hours. A rapid desorption took place in all soil samples during the first 10 minutes, followed by a slow adsorption in the loam size fractions and no apparent adsorption or desorption in the DG size fractions. The amount of simazine rapidly adsorbed (step I) and desorbed (step II) was best correlated with the soil specific surface area for the low-organic DG size fractions and with soil organic matter content for the loam size fractions. For both sorption steps I and II, the adsorption kinetics in the loam size fractions were well described by a modified Freundlich type equation. Adsorption kinetics during step II were dependent on both the soil organic matter content and the total amount of simazine adsorbed during step I, likely due to diffusion-controlled sorption of simazine on the soil organic matter.

View full abstract