The behavior of free-head rigid drilled shafts under static and cyclic lateral and moment loading was investigated using laboratory models in relatively large test chambers. This testing program represents perhaps one of the first larger-scale laboratory test series to utilize cast-in-place concrete shafts in consolidated and prestressed cohesive soil deposits for realistic simulation of prototype drilled shafts in clays. The construction procedure incorporated the actual effects of concrete curing and soil/concrete interface roughness, and the soil-deposit preparation included the characteristic anisotropy and overconsolidation associated with natural clays. A total of 28 cylindrical shafts having diameters of 51, 89, and 175 mm (2.0, 3.5, and 6.9 in.) and depth-to-diameter ( D /B) ratios of 3–8 were constructed and tested. Many of the shafts were instrumented with total stress cells and pore-water stress transducers to permit both total and effective stress measurements during the load testing. The results of the lateral and moment load tests indicated a high degree of nonlinearity in the monotonic static load-displacement response, but it can be represented adequately by a hyperbola. This hyperbola also provides a reference backbone curve for the cyclic loading behavior.

The Masing rules, often used in modeling the behavior of soils under cyclic loading, are unreliable when there is significant cyclic degradation of the soil-strain curves. In the present paper it is shown that degradation under simple shear loading conditions may be attributed, mainly, to decrease in the mean effective stress. The influence of decreasing mean effective stress on the parameters of Iwan's series-parallel model, which obeys the Masing rules, is considered. The slip stresses of the model elements are shown to degrade proportionally to the decrease in mean effective stress, while the small strain shear modulus appears to be invariant to this change. Incorporation of slip stress dependence on mean effective stress into Iwan's model leads to the concept of a normalized backbone curve that is nondegrading, and to predicted cyclic simple shear behavior consistent with the Masing rules. The nondegrading backbone curve concept is shown to be consistent with experimental results obtained in torsional simple shear tests.

The purpose of this paper is to review available case-history data on the treatment of liquefiable soils by compaction grouting, and to provide observations that are useful in evaluating the effectiveness of future compaction-grouting applications. Observations are made on practical aspects such as the effect of construction procedures on treatment effectiveness, the types of soil effectively treated, and issues in in-situ testing for quality control. An approximate relationship is established between the mean grout take and treatment effectiveness, which illustrates that grout takes have been consistently greater than would be anticipated based on common design calculations. This apparent discrepancy is attributed to a concentration of volumetric strains near the grout mass while in-situ penetration tests are commonly performed at treatment grid centers (farthest point from grout mass). Quantifying the composite behavior of a compaction-grouted soil mass will require an improved understanding of the treatment mechanisms.

Relationships between cone penetration tip resistance and the liquefaction potential of sandy soils are presented to facilitate use of the cone penetration test (CPT) in liquefaction assessments. The relationships are based on 180 liquefaction and nonliquefaction field case histories where CPTs were performed and illustrate the importance of median grain size and fines content on liquefaction resistance. The proposed CPT-based relationships were developed to describe the field case histories where CPT data are available, and eliminate the need to rely on conversions of standard penetration test (SPT) blow counts to CPT tip resistance used by existing CPT liquefaction-potential relationships. A new conversion between CPT tip resistance and SPT blow count is also proposed using the liquefaction-potential relationships developed from CPT data and existing liquefaction-potential relationships developed from SPT data. Finally, tentative CPT-based liquefaction-potential relationships are proposed for clean and silty gravel based on 18 liquefaction and nonliquefaction case histories.

A laboratory study is presented of the liquefaction behavior of sand subjected to bidirectional monotonic and cyclic simple shear loading conditions using a new bidirectional simple shear apparatus. Tests were performed with the monotonic and cyclic components being “parallel” and “perpendicular” to provide an approximate representation of the seismic loading conditions produced within earth embankments or slopes by the transverse and longitudinal components of motion, respectively. For a static horizontal shear stress ratio of 0.2 to 0.3, for a sand at relative densities of 35%, 45%, and 55%, and for failure criteria of 3% to 7.5% single amplitude shear strain, the perpendicular (i.e., longitudinal) cyclic loading resistance ranged from 70% to 95% of the parallel (i.e., transverse) cyclic loading resistance. The lower cyclic loading resistance for perpendicular versus parallel static and cyclic loads is attributed to differences in the amount of principal stress rotation and shear stress reversal that occurs for these two loading conditions.

This paper shows how analytical solutions, proposed by the writers to predict in-situ stresses in rock masses with smooth and irregular topographies, can help in the selection of the alignment of unlined pressure tunnels near slopes and valley walls. The proposed methodology can be applied to ridges and valleys in isotropic or anisotropic rock masses subject to gravity or to combined gravitational and tectonic loading. The analytical solutions are two-dimensional and assume plane or generalized plane strain. They can be used to substitute existing design charts for pressure tunnels based on the finite-element method. It is found that the safe alignment of unlined pressure tunnels depend greatly on the extent of tensile regions in valley walls, which itself depends on parameters such as valley geometry, the degree of rock anisotropy, the orientation of the planes of rock anisotropy, and the in-situ loading conditions (gravity, gravity and tectonic).

Drilled shaft load-settlement analysis is frequently performed using load-transfer curves known as t-z curves. Many t-z curves are discussed in the literature, and are often used for approximate analyses. The peak skin friction is generally considered to be developed with deflections on the order of 6–12 mm (0.25–0.5 in.) or 0.5–1% of the shaft diameter. The results of several load tests in calcareous cemented fine-grained soils in the arid southwestern United States are reviewed. It is demonstrated that significantly higher deflections are required to develop the peak skin friction in these soils than in the soils for which most of the available curves were developed. Deflections at peak skin friction are typically 25 mm (1 in.) or more. This results in an increase in the deflection required to attain a given fraction of ultimate load. Although cemented soil deposits have high shear-strength parameters, deflections may exceed expectations for working loads.

Laboratory and field evidence on residual shear strength, density, and penetration resistance of very loose, water-deposited silty sands and sandy silts capable of liquefying and developing large shear deformations is summarized. A program of laboratory monotonic and cyclic undrained tests is conducted on reconstituted, water-deposited, layered triaxial specimens of silty sand retrieved from the Lower San Fernando Dam. The results indicate that the soil behaves contractively in shear, and both its density and steady-state strength increase rapidly with consolidation pressure. These trends are verified for field conditions by comparison with available case histories of earthquake-induced flow failure and large lateral deformation of embankments, slopes, and mildly sloping ground. The 1971 flow slide in the Lower San Fernando Dam is examined in detail. The normalized Standard Penetration Index ( N₁ ) ₆₀ used to develop engineering charts is developed from the case histories. These charts can be used to evaluate residual shear strength and to indicate if a saturated silt-sand slope or site is capable of developing flow or lateral deformations greater than 1–3 ft during earthquakes of magnitude M _W ≲ 8.

It is common for fill that has a low dry density and high moisture content to have poor engineering properties, such as high compressibility and low shear strength. For the tested diatomaceous fill, the range in dry density varied from about 1.0 to 1.2 Mg/m ³ with a moisture content of 38–53%. The low dry density and high moisture content are due to the diatoms, which are essentially outer shells of silica that can contain water. Even though diatomaceous fill has a low dry density and high moisture content, it behaves as a dense granular material below an effective stress of 50 kPa. This is due to the diatoms, which can resist shear and compression because of their interlocking and rough surface features. At a higher effective stress the diatoms can be crushed, which significantly increases the compressibility of the fill.