Advances in Deep Foundations

The use of continuous flight auger (CFA) and drilled displacement (DD) piles in the U.S. market has increased significantly in recent years. This paper describes some of the advantages and limitations of these drilling systems compared to other deep foundation systems. Geotechnical and project considerations are described which favor the use of CFA and DD piling. In addition, conditions are discussed which may be problematic for this type of construction or which may limit the practicality and competitive advantages of CFA and DD piles.

There have been several interesting new advances in the design, construction, and testing of auger pressure grouted (APG) piles recently. The new construction techniques involve the addition of displacement piles (no spoils) and partial displacement piles (some spoils) that utilize many of the same installation procedures as traditional APG piles. These new pile types are more applicable where soft to firm clay and loose to medium dense sand soil conditions are present. The new testing and quality assurance procedures available include (1) automated pile installation monitoring equipment, (2) non-destructive integrity testing and (3) grout maturity strength testing procedures. In addition to new design methods, analytical methods are now used that can utilize the installation effort or drilling resistance measured when these piles are installed to provide a much higher level of confidence of the capacity and integrity of these pile types.

This paper examines several issues related to the behavior of augered cast-in-place (ACIP) piles under axial compression loading, including the: (a) load-displacement response, (b) method to interpret the "failure load" from the load-displacement curves, and (c) prediction of capacity using conventional geotechnical parameters. Case histories of field load tests were used herein. First, the generalized load-displacement behavior was examined, and it was found that the basic load-displacement characteristics of ACIP piles are comparable to those of other types of drilled foundations. Then the compression capacity was evaluated in a consistent manner. The results were similar to drilled shafts, but some key differences were noted. These results will be useful in evaluating overall ACIP behavior.

Increasing use of ACIP piles, as an alternative to drilled shafts, to support load-bearing structures has resulted in socketing the piles in rocks. In this study, the behavior of eight ACIP piles socketed in soft clay-shale was investigated. The diameter and length of piles varied from 457 to 610 millimeters (18 to 24 inches) and from 12 to 25 meters (40 to 83 ft) respectively. The piles were socketed in clay-shale up to 7.5 diameter of the pile and the measured maximum loads from field tests varied from 1915 to 4000 kN (215 to 450 tons). The unconfined compressive strength (q_u) of clay-shale was in the range of 1300 to 2300 kPa (188 to 333 psi). The load-displacement relationships for the ACIP piles were predicted using a non-dimensional hyperbolic relationship and the model parameters were related to the rock properties and pile dimensions. For the ACIP piles socketed in clay-shale, 50% of the ultimate load was achieved in the displacement range of 0.005 to 0.012 of the diameter. Simple relationships have been developed to predict the pile capacity and model parameters based on unconfined compressive strength (q_u) of rock and Texas Cone Pentrometer (TCP) results.

In this study, the finite element method was used to simulate the construction effects on the behavior of an Augered Cast in Place (ACIP) pile installed in the Pleistocene terrace formation of Texas Gulf Coast. A mixed soil profile, typical for the Gulf Coast, was considered. The construction effects were modeled by taking into account stress changes during drilling and grouting and by assigning modified properties (c^', φ^', E) due to remolding in the soil surrounding the pile. Due to the construction procedure, the ACIP pile is expected to affect the surrounding soil differently compared to a bored pile or a driven pile. The validity of the assumptions was assessed, by comparing numerical analysis results to full scale test data. Furthermore, the importance of the construction effects was demonstrated by comparing to the wished-in-place pile case. The various stages of the ACIP pile construction were modeled numerically by using the commercially available F.E. code PLAXIS and by adopting the hardening soil model (HS). The numerical analysis showed that load settlement behavior of an ACIP pile installed in Texas Coastal Soils gave realistic settlement results when construction effects were considered, while it significantly underestimated the settlement in the case where the pile was wished in place.

In the desert southwest of the United States (Arizona and parts of New Mexico, California, Nevada, Utah and Colorado), the soils are often dense and cemented. The design procedures for drilled shafts described by the American Association of State Highway and Transportation Officials (AASHTO) or the Federal Highway Administration (FHWA) do not incorporate the effect of density or cementation. Use of β-α methods cited in these publications often result in an over-design of the drilled shafts with associated greater costs. The FHWA publication by O'Neill and Reese suggests classifying dense or stiff soils as intermediate geomaterials (IGMs) and proposes alternative design methodology. Based on results of the recent large-scale load tests in Arizona reported in this paper (2.44 m diameter test shaft) and the results of tests in New Mexico described by Chua et al. (1.37 and 1.83 m diameter test shafts), it appears that the actual skin friction in desert southwest soils is much higher than that predicted by the equations for IGMs. This paper presents the results of the test program performed for the Arizona Department of Transportation (ADOT) that included a large scale load test on a 2.44 m diameter, 41.1m deep shaft to re-design large diameter drilled shaft foundations for the I-10/I-19 Traffic Interchange (TI) in Tucson, Arizona. The methodology implemented on this successful project may be useful for future projects in dense and variably cemented soils.

A laboratory experimental program was conducted to evaluate the performance of partially hydrolyzed polyacrylamide (PHPA) polymer drilling slurries for applications in Missouri shale. The series includes drum slake, jar slake, linear swell, and hardness testing for representative Missouri shale specimens allowed to interact with a variety of commercially available PHPA slurries. Slurries were prepared at concentrations ranging from 25% to 150% of manufacturer-recommended mixing ratios (by mass for solid polymers, by volume for liquid polymers) to examine the associated effects on performance. Indices used to quantify performance in terms of slaking, swelling, and softening inhibition are shown to be relatively insensitive to polymer manufacturer and form (granular or emulsified) but significantly sensitive to concentration. The performance indices improve with increasing slurry concentration and appear to reach optimum values at the recommended mixing ratios. Over-concentration does not enhance, and may diminish, slurry performance.

Drilled shafts have become the preferred deep foundation element in many arid and semi-arid regions such as the southwest USA because soil conditions are usually unfavorable to driven piles due to coarse gravel and cobbles. In addition, scour depths on the ephemeral river channels are quite large and there is increased confidence in the bearing layer afforded by the drilled shaft construction process. Skin friction predictive equations have been available for determining the contribution of skin friction to the drilled shaft axial load carrying capacity for a number of years. In clays and sands many load tests have been performed. These load test results have served to create and validate the equations used. In the past, only a limited number of load tests have been performed on granular materials with high gravel content. It is presumed that the skin friction factors of gravelly soils would be higher than those for pure sands, because of the increased dilatancy of gravels prior to failure. As the use of drilled shafts increases, more data from gravelly soils has become available from load tests to determine how well the current predictive equations work. This paper focuses on skin friction factors arising from drilled shafts in granular materials with a significant to high gravel content. Fifty six load tests were analyzed to determine the actual skin friction factors for drilled shafts in granular soils. These skin friction factors were compared with the various predictive equations currently employed for design purposes. The results show that the predictive equations are extremely conservative for predicting the skin friction factor in gravelly soil conditions. A new empirical model was developed by back calculating the ratio of horizontal stress to vertical stress which leads to the development of the actual skin friction in the field. This new model produces more or less unbiased estimates of the actual skin friction in the field.

The shear behaviour of rock/concrete interface is a critical factor controlling the performance of drilled shafts socketed into rock. The authors have conducted an extensive series of direct shear tests on rough sandstone/concrete interfaces to investigate the performance of bored piles socketed into sandstone. During these tests, wear of the sandstone surface was apparent. As a result of this wear, the shear strength of the interface was significantly reduced compared to rock/concrete joints of similar roughness profiles in which no wear was observed. In order to develop an understanding of the wear mechanisms, an experimental program that focused on the wear process was carried out. Factors that affect the wear mechanism have been identified and a wear model developed. This model is based on the experimental observations, analysis of the test data and principles adopted from the field of Tribology. Comparisons of predicted and measured wear show good agreement.

This paper describes the results of a study investigating the influence of soil shear strength, overburden stress, and shaft diameter on the squeeze susceptibility of an open shaft in soft clay. Both field instrumentation and finite element analysis were conducted. The inclinometer data from monitoring two shafts of different diameters showed that up to a particular depth in the same soil layer, noticeable lateral deformation was recorded; however, the magnitude was significantly greater in the larger diameter shaft than in the smaller diameter shaft. The finite element analysis showed similar results as observed in the field. Based on the data collected from field observations, in-situ tests, laboratory tests, and the finite element analysis, this study shows that there exists a relationship between the above mentioned variables and the potential for clay squeeze. A formula containing the shear strength, depth of excavation, and diameter of shaft was developed to provide an assessment of squeeze susceptibility. The results of this study can provide guidelines on the potential for shaft squeeze and indicate whether temporary steel casing or slurry will be required during caisson construction.

Five contracts from the Central Artery/Tunnel Project (CA/T) in Boston, Massachusetts were reviewed to document issues related to the design and construction of driven pile foundations. Given the soft and compressible marine clays in the Boston area, driven pile foundations were selected to support specific structures including retaining walls, abutments, roadway slabs, transition structures, and ramps. This paper presents the results of a study to assess the lessons learned from pile driving on the CA/T. This study focused on an evaluation of static and dynamic load test data and a case study of significant movements of an adjacent building during pile driving. The load test results showed that the piles have more capacity than what they were designed for. At the site of significant movement of an adjacent building, installation of wick drains and preaugering to mitigate additional movements proved ineffective. Detailed settlement, inclinometer, and piezometer data are presented.

Resistance factors were calibrated using the framework of reliability theory for the Load and Resistance Factor Design (LRFD) of driven pile's axial capacity in North Carolina utilizing pile load test data available from the North Carolina Department of Transportation (NCDOT). A total of 140 Pile Driving Analyzer (PDA) data and 35 static load test data were compiled and grouped into different design categories based on four pile types and two geologic regions. Resistance statistics were evaluated for each design category in terms of bias factors. Reliability analysis of the current NCDOT practice of pile foundation design was performed to evaluate the level of safety and to select the target reliability indices. Resistance factor calibration was performed for the three methods of static pile capacity analysis commonly used in the NCDOT: the Vesic (1977), the Nordlund (1963), and the Meyerhof (1976) methods. Two types of First Order Reliability Methods (Mean Value First Order Second Moment method and Advanced First Order Second Moment method) were used for the reliability analysis and the calibration of the resistance factors. The calibrated resistance factors varied significantly for the different pile types and geologic regions. The advanced first order second moment method resulted in larger resistance factors than the mean value first order second moment method. Recommended resistance factors are presented for the three methods of static pile capacity analysis and for seven different design categories of pile types and geologic regions.

The research performed in the last decade at Purdue University on the vertical load response and design of piles (particularly drilled shafts and pipe piles) is reviewed. Both numerical and experimental investigations were performed, and values for unit base and shaft resistance normalized with respect to cone resistance are proposed for design. The study takes into account the loading processes associated with the pile installation, and the corresponding alteration of soil properties that affect the pile response. The normalized unit resistances are provided for a relative settlement of 10%, which is widely used as a design criterion. A strong dependence of the normalized unit resistances on relative density has been observed and the suggested values account for the effect. Thus, the proposed normalized unit resistances account for the soil state and the loading processes, and can be used in design. Results with respect to IFR and PLR are provided as well, for open-ended piles.

To predict the ultimate limit state of precast driven piles, there are different empirical and analytical approaches available. A simple prediction of the resistance-settlement behavior is so far not possible. During field load tests of precast driven piles, commonly measured quantity is the total load on the pile head. Hence, model tests were performed to separate the skin friction and base resistance on the pile. Based on the statistical analysis of more than 200 static load tests on precast driven steel and concrete piles, a simple approach to predict the resistance-settlement behavior of precast driven piles was derived.

Piles driven into low permeability soils gain capacity over time. This capacity gain, often called "set-up" or "freeze", is believed to be controlled by the mechanisms of effective stress increase due to the dissipation of excess pore pressures built up during driving and stress independent phenomena such as strength increase due to thixotropic bonding. Practical efforts have been recently made to quantify time dependent pile capacity and develop a methodology of incorporating it into pile design by utilizing insitu testing. The following paper examines the use of the Multiple Deployment Model Pile (MDMP) and a full-scale instrumented pre-stressed concrete pile to examine time dependent pile capacity gain in Boston Blue Clay. Normalized relationships were used to examine the overall and segmental pile capacity gain for both the MDMP and the full scale pile and correlations are made to normalized excess pore pressure dissipation. The results of the MDMP and full scale instrumented pile compare favorably with previously compiled pile capacity gain and excess pore pressure measurements.

A pile with multiple stepped two diameters showed higher bearing capacity than a conventional cylinder pile. In this paper, model tests were conducted to examine for the bearing mechanism of this pile. The purpose of this study was as follows. (1) To measure the load bearing characteristics of model piles installed a small load cell at the end of pile. (2) To observe the behavior of soil particles around the model pile after loading by excavating the suction applied ground, and comparing with the behavior of soil particles of model piles of longitudinally half-cut shape. (3) To measure load transfer characteristics by earth pressure transducers placed in the model ground in the loading tests. The same cases were analyzed by DEM (Distinct Element Method). As a result, it can be concluded that a multiple stepped pile shows higher bearing capacity because of the load transmitted from the convex parts to the outside diagonally.

In some instances, pile foundations are subjected to lateral loading caused by lateral movement of the surrounding soil. This paper describes an analytical method for evaluating pile foundation response to lateral ground movement. A model using beam on linear and nonlinear elastic foundations has been analyzed using the finite difference method. Dimensionless plots are developed to evaluate pile behavior for different soil and geometric conditions. These plots can be used for estimating pile displacements and moments caused by lateral soil movement. Failure mechanisms are examined, and are related to relative soil-pile stiffness and soil yielding. A design example has been presented to demonstrate the use of dimensionless plots and to illustrate failure mechanisms.

Sleeving a pile is believed to be capable of reducing horizontal stresses transferred into the ground at shallow depths significantly when the sleeved pile is subjected to lateral loads such as wind loads. However, current design methods of sleeved piles are largely empirical and the influence of sleeved piles on adjacent retaining walls is not fully understood. In this paper, two series of three dimensional numerical analyses have been carried out to investigate the influence of sleeving on load transfer mechanisms from a laterally loaded pile onto an adjacent retaining wall, wall displacements, lateral stresses transferred onto the wall and the factor of safety of the wall against sliding and overturning.

Deflections of laterally loaded pile groups are strongly affected by the degree of pile head fixity. If the cap provides perfect restraint against rotation, deflections are only about 25% as large as the deflections for no rotational restraint at the tops of the piles. The degree of rotational restraint is governed by (a) the rotation of the piles with respect to the cap, and (b) the rotation of the cap itself. This paper reviews available information on pile head fixity and proposes a simple method for considering pile head fixity in analyses of pile groups. The method includes consideration of nonlinear soil behavior, pile-soil-pile interaction, the resistance to lateral loads provided by pile caps, and the degree of pile head fixity. The proposed method can be implemented through computer programs such as LPILE and COM624P, and spreadsheets based on the Characteristic Load Method.

The use of piled raft foundations has become popular in recent years as the system can act as a settlement reducer. When the raft is subjected to non-uniform loadings, piles of different diameters and lengths may be required to prevent the foundation from rotating and to reduce the differential settlement. This paper presents a finite layer method used for the analysis of piled rafts with piles of different lengths and diameters. The soil is divided into horizontal layers with different material properties and only vertical loads may be applied to the raft. Interactions between raft-soil-pile are computed. Results from the present method are shown to agree closely with those from the finite element method.

A beam-on-a-nonlinear Winkler foundation (BNWF) model in which the axial and lateral responses are coupled is developed in this paper for the analysis of piles in sand under cyclic and dynamic loading. To do this, the general equilibrium equations of an existing cyclic P-y analysis program are reformulated to account for both the axial and lateral cyclic responses. An example pile problem is then used to study the effect of the various model parameters on the lateral cyclic hysteretic response, and also the influence of both vertical and horizontal ground motion on the inertial interaction response. The results showed that the model parameters, namely: the angle of wall friction, initial confining pressure and effective vertical tangential stiffness did not have a significant influence on the lateral cyclic hysteretic response. Under horizontal and/or vertical ground motion, horizontal ground motion was seen to dominate the inertial interaction response, and significantly affected both the horizontal and vertical displacements due to gapping and second-order P-Δ effects. The results suggest that the use of the uncoupled analysis approach could lead to unconservative designs.

Piles are often installed as the foundation for large oil storage tanks constructed on soft soils. The transfer of loads between the piles and soft soils is still not well understood. A series of centrifuge model tests were conducted on oil tank supported by piles having the same size and plan layout with various pile cap sizes to investigate the load transfer mechanism of the foundation system. The model setup was instrumented such that the load carried by each pile and the settlement at the center and the corner of the tank were obtained. Based on the test results, the efficacy (percentage of applied loads carried by the piles) of the foundation system with different pile cap sizes is determined and the practical implications of the findings are elaborated.

The behavior of composite pile foundation is different from that of common pile foundation, which has not been understood fully yet. Therefore, this paper was designed to analyze the additional stress and the deformation of soil in composite pile foundation. The finite element method was applied to simulate the performance of pile foundation under vertically uniform loads. And the classical Boussinesq's solution was used to calculate the additional stress. Some cases were presented to compare the behavior of composite pile foundation and common pile foundation. The calculated results show that the biggish additional stress exits in the soil between piles and the stress along the pile is uniformly distributed. It is certain that the soil between piles shares a certain part of loads transferred from superstructure. Hence, it is suggested that the additional soil stress between piles and under piles should be considered to calculate the settlement of composite pile foundation. On the base of above suggestion a simplified method of calculating the composite pile foundation settlement is presented. The case comparisons demonstrate that the results of simplified method presented in this paper have a good agreement with the test date.

Piled raft systems have been commonly used for supporting buildings and other superstructures. In most applications, piles are installed in layered soils. It has been a challenge how to properly analyze the behavior of piled raft systems in layered soils. This paper presents a methodology for analyzing an axially loaded single pile-rigid circular raft system in layered soils based on the extended Sneddon and Muki solution for layered elastic media. Field test results for a single pile-rigid raft system at the F1 motor raceway site in Shanghai, China were used to verify the proposed method. The comparison indicates that the proposed method can reasonably predict the elastic behavior of the pile-raft system. This study also demonstrates an obvious difference in side resistance along the pile in layered soils with or without a rigid raft.

The inelastic deformations for pile foundations under lateral loading occur below grade; therefore, the overall lateral load behavior of the system is influenced by the interaction between the shaft and the surrounding soil. This complex interaction is commonly modeled using p-y curves. The standard p-y method does not explicitly take into account all the key aspects of soil-pile interaction, namely the nonlinear response of the soil, frictional interface stresses between the soil and the pile, the generation of gaps between the soil and pile, damping under dynamic loads, and the structural response of the pile. Past research efforts have been aimed at developing unidirectional p-y interaction curves comprised of an assembly of distinct sub-elements; the sub-elements are each in charge of modeling a specific component of soil-pile resistance under lateral loading. These assembled soil-pile interaction curves are then compared to a backbone curve based on the existing standard p-y curves. However the current models for p-y curves are calibrated primarily from lateral load tests and lump all components of lateral resistance into one curve. Thus the contribution of each sub-element to the overall assembled behavior of the interaction "p-y" curve becomes somewhat arbitrary. This study examines the effects of nonlinear soil resistance to relative displacements between a model pile and soil, including the effects of gapping, cyclic degradation, and interfacial slip through laboratory testing. Measurements are presented for the normal compressive soil stress on the leading face of the pile, soil/pile interface stress along the sides of the pile, and active normal soil stresses on the pile. Using current sensor technology, normal stresses in soil are measured and side-frictional interface stresses are inferred. Professor Michael O'Neill stated in his Thirty-Fourth Karl Terzaghi Lecture that "the design of deep foundations is a complex matter that should be addressed in a design context by engineers who are experienced in the observation of pile behavior, theoretical modeling, and the appropriate use of design methods" — this paper attempts to stay true to Professor O'Neill's words by applying both observations from experimental measurements and theoretical modeling to a complex topic in foundation engineering.

This paper describes the results for a numerical analysis of single piles and pile groups in clayey soils subjected to monotonous lateral loading using the ABAQUS finite element software. The investigated variables in this study include free head and embedded capped single piles, pile diameter (1.0 m, 0.5 m), pile length (7.0 m, 10.0 m), and pile groups. The 1 x 3 pile group was selected to investigate the individual pile and group lateral resistance, the distribution of the resistance among the piles, the effects of lateral stresses in front of and on the sides of the piles, and the effect of a cap on the lateral resistance of the leading pile. The soil was modeled using Cam-clay constitutive relationship and the pile was a circular concrete pile. The results show that the size of the cap influences lateral capacity of the single pile. The results also show that in pile groups, the pile-soil-pile interaction and the cap affect the resistance in the leading pile, and the p-multiplier for the leading pile of greater than 1.0 was able to be obtained.