Charles C. Ladd

Stress Path Tests with Controlled Rotation of Principal Stress Directions

University College London (UCL) has developed a new flexible boundary plane strain shear apparatus for investigating stress-strain-strength anisotropy in sands. Called the Directional Shear Cell (DSC), it has the unique capability of controlling the major principal stress direction by varying the normal stress and shear stress acting on four faces of a cubical sample constrained between two rigid end plattens. The paper describes the device and its use, and presents results from a Massachusetts Institute of Technology (MIT) test program run on dense and loose Leighton Buzzard Sand to independently evaluate the device and for comparison with data obtained at UCL to investigate the effects of induced anisotropy due to prestraining. The DSC has now been used to measure inherent anisotropy in loose sand. These data are presented and MITs progress in modifying the DSC to enable K 0 consolidated-undrained testing of soft clays is described.

A New Fall Cone Device for Measuring the Undrained Strength of Very Weak Cohesive Soils

A new fall cone device capable of measuring undrained shear strength of extremely soft cohesive soils is described and evaluated. The Automated MIT Fall Cone has a unique mechanical counterweight system that enables the use of effective cone weights as low as 0.005 N. This weight is an order of magnitude lower than possible with a standard fall cone apparatus. The new device also uses a data acquisition system to record cone penetration versus time throughout the test. An extensive experimental program was performed on a remolded marine clay (Boston Blue Clay) to evaluate the new fall cone apparatus. The program includes measurements of cone velocity and penetration to study the dynamics of the new device. The cone velocities and penetration times agree with theoretical predictions that account for the inertia of the counterweight system. The experimental program includes strength measurements using various cone weights and geometries to check the validity of the strength-penetration relationships over a wide range of liquidity index values (0.75 to 5.6).

Formal Observational Approach to Staged Loading

Construction of a major industrial facility on soft clay was accomplished by staged loading to consolidate and thus increase the strength of foundation soils before the complete final load was applied. Extensive in situ and laboratory testing allowed a formal observational approach to be used, in which uncertainty analysis of the preliminary design was combined with field monitoring data to update soil parameter and site condition estimates. Traditional Bayesian methods were used to modify initial estimates and thus to update predictions of facility performance in the later stages of construction.

Time-Dependent Triaxial Relaxation Behavior of a Resedimented Clay

The paper describes the results of K0 consolidated-undrained triaxial compression relaxation tests on resedimented Boston Blue Clay (BBC) using a computer-automated triaxial apparatus. Specimens were either normally consolidated (overconsolidation ratio, OCR = 1) or had an OCR = 4, and they were sheared at different axial strain rates to axial strains (ea) from 0.1 to 15%. Each strain level was maintained until the monitored shear stress and pore pressure reached equilibrium levels. The results show that a single specimen can be used reliably for multiple relaxation tests. During relaxation, the rate of normalized shear stress decay with log time, (−Δq/σ′vm)/Δlogt, is approximately constant, regardless of ea or OCR. Changes in shear-induced pore pressure during relaxation were negligible except when specimens were rapidly sheared. Independent of OCR, end-of-relaxation equilibrium stress states for relaxation strains ea ≤ 1.5% lie on a line of obliquity defined by K = σ′h/σ′v = 0.50, compared to K0 = 0.49 for OCR = 1 BBC; whereas, for ea ≥ 2.5%, the equilibrium states lie on a steeper obliquity line defined by K = 0.40.

Effects of Cementation on the Compressibility of Pierre Shale

Natural cementation in cohesive soils can have a significant effect on the apparent maximum past pressure (σ v m ) estimated from one-dimensional consolidation tests. Relatively small amounts of calcium carbonate (CaCO 3 ),above a threshold value of about 2 percent by_weight, appear to cement the structure of a Nebraska Pierre shale, causing measured σ v m -values several times larger than the maximum past overburden stress inferred from geological history. However, CaCO 3 contents in excess of this threshold value (even up to 50 percent or more) produce little further increase in σ v m . On the other hand, leaching out relatively small amounts of CaCo 3 from a highly calcareous clay shale can increase its compressibility and produce a lower measured apparent maximum past pressure.

Automated Triaxial Testing of Soft Clays: An Upgraded Commercial System

The paper evaluates the ability of a widely used automated commercial testing system to perform Ko consolidated-undrained triaxial compression and extension (CKoUC/E) tests on normally consolidated clay. The commercial system uses control software on a microcomputer to operate three digital pressure-volume controllers that control and measure the pore water pressure and volume, the radial stress, and the axial stress and strain for a specimen within a Bishop and Wesley Hydraulic Triaxial Apparatus. Because the commercial system calculates axial stress and strain by relying on indirect measurements assuming idealized behavior of the axial loading assembly, the system was modified to directly measure axial stress and strain. The system was also enhanced to enable shear in triaxial extension and to improve certain other testing procedures. Comparison of results from CKoUC/E tests obtained by the resulting reference system with prior data on resedimented Boston Blue Clay shows good to excellent agreement in 1-D compressibility and undrained stress-strain-strength characteristics. In contrast, stresses recorded by the commercial system resulted in significant overestimates of preconsolidation pressure, undrained strength, and the effective stress envelope, primarily caused by unquantifiable behavior of the axial loading assembly. Based on these results and other experience, the paper suggests features warranting special attention when selecting a commercially available automated triaxial system in order to upgrade the experimental capabilities of a laboratory.

The Multidirectional Direct Simple Shear Apparatus

The paper describes a new simple shear testing device, the multidirectional direct simple shear (MDSS) apparatus, for testing soil specimens under conditions that simulate, at the element level, the state of stress acting within the foundation soil of an offshore Arctic gravity structure. The MDSS uses a circular specimen that is consolidated under both a vertical effective stress (σ′vc) and a horizontal shear stress (τ1). The specimen is subsequently sheared undrained by applying a second independent horizontal shear stress (τ2) at an angle ϑ relative to the horizontal consolidation shear stress τ1. Evaluation of the MDSS first compares conventional K0-consolidated undrained direct simple shear (CK0UDSS) test data (τ1 = 0) on normally consolidated Boston blue clay (BBC) with results obtained in the Geonor DSS device. The MDSS gives lower secant Youngs modulus values and on average 8% lower strengths, but produces remarkably less scatter in the test results than the Geonor DSS. Kinematic proof tests with an elastic material (rubber) confirm that the setup procedure, application of forces, and strain measurement systems in the MDSS work properly and produce repeatable results. Results from a MDSS test program on BBC wherein specimens were first normally consolidated with σ′vc and τ1 = 0.2σ′vc and then sheared undrained at ϑ varing in 30° increments from zero (shear in same direction) to 150° show dramatic differences in the response of the soil as a function of θ. The peak undrained strength varies almost twofold from θ = 0 to 120°, while the deformation behavior varies from very brittle at low θ angles to becoming ductile at higher angles. The experimental results indicate that dramatic changes in foundation response at different θ angles will be an important design issue.

Discussion on Laboratory Shear Devices

This discussion starts with a general assessment of the capabilities of several types of laboratory shear devices and their corresponding suitability for use in specialized design practice and for basic research into the constitutive stress-strain-strength relationships of soils. It then focuses in greater depth on two specific topics: use of the Geonor Direct Simple Shear (DSS) device for design practice involving soft clays, and experimental requirements needed to reliably measure stress-strain-strength behavior in cross anisotropic soils.