There are at least five different methods for interpreting the results of field pressuremeter tests. Each of these methods yields different values of undrained shear strength when applied to the same pressuremeter test. A laboratory pressuremeter model was designed to simulate the pressuremeter test. Sedimented kaolinite samples were used for the laboratory model tests. These samples were consolidated to different values of the maximum vertical pressure and with overconsolidation ratios ranging from one to eight. For comparison purposes, two sets of $K_o$‐consolidated undrained triaxial tests, compression and extension, were carried out in the same range of the maximum vertical pressure and overconsolidation ratio. The range of variation of the undrained shear strength values computed for the same test by the different methods is very wide and increases with an increase in the overconsolidation ratio. A new method of interpreting the field pressuremeter test results is developed and presented. In this method, the effect of the vertical stress and the excess pore pressure are included for the determination of the undrained shear strength. The results of the new method coincide with that of the triaxial test with a maximum variation of $\pm 5\%\text{.}$

A test apparatus to measure the load transfer along the shaft of a model pile inserted in a specimen of clay is described. This apparatus allows independent control of the boundary stresses of a cylindrical clay specimen in the vertical and horizontal directions. The top and lateral surfaces of the clay specimen can deform freely and the load transfer zone on the pile is separated from the rigid base. The load transfer along the shafts of smooth and rough surface steel model piles in a kaolinitic clay under varying vertical and horizontal effective stress combinations was measured. The pile installation technique used was designed to minimize soil disturbance so that the failure criterion for pile shaft friction could be investigated. Normally consolidated and overconsolidated conditions, both in the vertical and horizontal directions, were considered in tests using eleven clay specimens with three different thicknesses and steel model piles of three different diameters. The results show that the angle of pile‐clay friction is independent of the vertical consolidation pressure in the clay, the overconsolidation ratio (both in the vertical and horizontal directions), and the length of the pile‐soil contact. The influence of pile diameter is not clearly established. The surface roughness of the pile clearly affects the pile‐clay friction angle, which is approximately equal for smooth surface piles to the Hvorslev true friction angle and for rough piles to the effective friction angle of the clay. There is no reasonable correlation with the undrained strength. The results support the effective stress approach.

The shaft creep response of a model pile inserted in a specimen of clay is studied under constant pile loads at varying percentages of the ultimate shaft resistance. The top and lateral surfaces of the clay specimen could deform freely. The specimens were consolidated under varying vertical and horizontal effective stress combinations. The results of the tests using eight normally consolidated clay specimens with three different thicknesses and steel model piles of three different diameters and of smooth and rough surfaces show that the total displacement of a friction pile consists of a relatively small immediate displacement followed by a significant time‐dependent shaft creep. The immediate settlement was mostly elastic and recoverable upon reloading. Consolidation of the clay due to load transfer from the axially loaded pile was negligible, as negligible volume change was recorded during constant pile load tests. The time‐dependent displacement of the pile is attributed primarily to the slip between the pile and the soil. The rate of the slip displacement is found to continuously decrease with time. A parabolic equation is found to represent the creep response best. The creep response as characterized by the parameters of this relationship is dependent on the load ratio (ratio of the axial load to the ultimate shaft resistance) and the pile diameter for a given pile‐soil system; however, it is independent of surface roughness and the state of soil stresses. The horizontal effective stresses in the clay determines the shaft resistance and, therefore, the load ratio for a given axial load.

A large‐scale group of steel pipe piles and an isolated single pile were subjected to two‐way cyclic lateral loading. The tests were conducted in a submerged firm to dense sand that was placed and compacted around the piles. All of the piles were extensively instrumented so that the variation in soil resistance within the group could be determined. The response of the piles in the group was also compared with the response of the isolated single pile. The loss of efficiency of the piles in the group was related principally to “shadowing” (i.e., the loss of soil resistance of piles in the trailing rows). Piles in the leading row supported a large proportion of the group load and behaved similarly to the isolated single pile. Two‐way cyclic loading had little effect on the distribution of load to the piles in the group, but tended to densify the sand around both the single pile and the group piles.

The paper summarizes the results of analyses of soil reaction stresses on offshore gravity platforms. Three aspects are considered herein: (1) Stress distribution on individual foundation skirts; (2) horizontal load sharing between skirts and platform base; and (3) contact stresses on platform base. Results of analyses illustrate the effects of skirt geometry, skirt flexibility, and soil parameters on the reaction stresses against the skirts and the platform base. Comparative analyses were done with elastic linear and nonlinear soil formulations, and with plane strain, axisymmetric, and three‐dimensional models. Guidelines for design are proposed for the evaluation of the stress distribution with depth for short and long skirts and the evaluation of contact stresses on the platform base. The three‐dimensional finite element analyses showed that a plane strain model through the center of the platform gives a good approximation of the normal and shear stresses against the skirts predicted by a three‐dimensional calculation. Based on two‐dimensional analyses with a nonlinear soil model, it is believed that the assumptions of skirt flexibility/rigidity are much more significant for the soil stress distribution on the skirts than the assumptions of a linear or nonlinear soil model.

The influence of anchor geometry, embedment, and soil density on the uplift capacity of one‐meter prototype anchors is investigated by subjecting 25‐mm models to an acceleration of 40 g in the Liverpool centrifuge. Uplift resistances expressed as dimensionless breakout facctors increase significantly with anchor embedment and soil density but reduce with increased aspect ratio. Failure displacements also increase with embedment but reduce with increased soil density and aspect ratio. The influence of anchor geometry is relatively insensitive to anchor size but increases with both embedment ratio and soil density. In general, the design approaches considered underestimate the observed capacity of one‐meter strip anchors in dense sand, although predictions using the Meyerhof and Adams and Rowe and Davis theories are acceptable. However, both these approaches appear overoptimistic for anchors in looser sand. In this case, Vesic's theory gives the closest agreement, while the formula of Majer yields overly conservative designs. Pull‐out capacities for isolated anchors may be obtained from the strip values in combination with the empirical shape factors reported in this research.

Extensive jointing is a fabric feature that is common to many finegrained soils with high clay content. Both naturally occurring and recompacted soil deposits have been known to exhibit extensively jointed fabrics. Jointing has been shown to dramatically influence the nature of the flow of fluids through jointed soils. Field permeability tests have demonstrated permeabilities of several orders of magnitude higher than laboratory tests of intact samples of the same material. The purpose of this study is to investigate the influence of joints on permeability in a low‐permeability matrix in which the joints are not necessarily continuously connected. This was accomplished by performing permeability tests on oriented samples of remolded fine‐grained till that had been isotropically consolidated and by comparing these results with finite element models. The soil was obtained from the Lake Michigan shoreline in the state of Wisconsin. This soil has been found to be extensively jointed, perhaps due to post‐glacial stress relief. A spherical sample of this soil was consolidated and tested at various orientations to obtain an intact matrix permeability. The specimen was then artificially jointed through the incorporation of a porous plate within the specimen. Additional permeability tests were performed at various orientations in order to investigate the influence of joint orientation on apparent permeability. The results indicated that the permeability in the direction parallel to the joint was significantly greater than the permeability perpendicular to the joint. The apparent permeability decreased rapidly as the applied gradient became more perpendicular to the joint, approaching that of the intact matrix. Similar results were obtained from a finite element model of the jointed spherical specimen.

The response of a buried pipeline to seismic waves is controlled by differential movements along its length. The axial movement of the pipeline is restrained by friction forces at the soil-pipe interface. Tests performed by others on a full scale pipe indicate that the relationship between the soil resistant force and relative displacement (i.e., between the soil and the pipe) is linear at small displacements. This soil resistant force reaches a ``plateau'' when shippage of the pipe with respect to the soil occurs. This suggests that the axial soil-pipe interaction can be modeled by discrete spring-sliders distributed along the pipeline. The initial stiffness of these spring-sliders is the subject of this technical note. First, previous studies evaluating this initial soil axial stiffness are reviewed. Then, results from the field tests are used to backcalculate the value of the initial soil stiffness. On the basis of the results obtained, a range of values is proposed to estimate the initial soil axial stiffness.

It is a common belief that negative incremental pore pressures occur during undrained triaxial compression loading of clays with overconsolidation ratios greater than approximately 4. While this behavior is characteristic of isotropically consolidated clays, a review of data from anisotropically consolidated tests on natural clays suggests that incremental pore pressures actually either remain positive or become only slightly negative at very high overconsolidation ratios. This phenomenon may be attributed in part to the location of the yield surface, which extends above the normally consolidated effective stress failure envelope, which is reflected by a Mohr-Coulomb cohesion intercept on the order of 4 ± 2% of the maximum preconsolidation pressure.