Deep Foundations 2002: An International Perspective on Theory, Design, Construction, and Performance

This paper outlines some local and global shortcomings in the analysis and design of foundations. Among the topics discussed are design philosophy, site evaluation, analytical models, design decisions, reliability, construction, and acceptance criteria. Broad concepts are stressed, fundamental questions are posed, and suggestions are given to improve practice.

Evaluating static pile capacity by dynamic testing has become routine procedure in contemporary foundation engineering practice worldwide. Testing is typically performed during initial pile driving, and/or during restrike some time after installation. Situations limiting the accurate assessment of static pile capacity by dynamic methods include: (1) the pile capacity changes due to the time-dependent characteristics of the supporting soils and/or rock, and (2) the hammer energy is insufficient to fully mobilize all soil/rock resistance forces present during the test. This paper presents discussions on the fundamental mechanics of these two limitations and. proposes the use of superposition of dynamic test results for evaluating total pile capacity. A case history where both a significant increase in pile capacity occurred after the end of initial driving, and insufficient hammer energy limited the activation of full resistance during restrike is presented that demonstrates the applicability of the proposed approach. Dynamic test results performed with end of initial driving (EOID) and beginning of restrike (BOR) data are provided along with full scale static loading test results. Using superposition by combining the pile end bearing resistance from EOID and shaft resistance from BOR dynamic test results produced excellent correlations with the static loading test results including the ultimate pile capacity as well as the pile top load-movement relationship.

The drained elastic (small-strain) behavior is evaluated for drilled shafts in cohesionless soils. A database of load tests is evaluated using the L₁-L₂ interpretation method, in which L₁ is the elastic limit, defined at a specific load (Q_L1) and corresponding displacement (ρ_L1). Values of drained elastic soil modulus (E_d) are back-calculated from Q_L1 and ρ_L1 and then are normalized by the unit side resistance (f) and cone tip resistance (q_c). For the database, the back-calculated E_d/f and E_d/q_c ratios had coefficient of variation (COV) values on their means of approximately 70 to 90% in uplift and 90 to 100% in compression. These COVs are consistently larger than those for ρ_L1 alone. Comparisons also are given to results for drilled shafts in cohesive soils. Overall, the results demonstrate that elastic behavior can be quantified using either normalized ρ_L1 or E_d values, but large COVs are inherent for subsequent displacement predictions.

In contrast to other types of anchors, prestressed ground anchors (or tie-backs) are routinely used in clayey or cohesive soils. One of the problems that needs to be considered concerns soil creep (i.e. rheological properties). The load-deformation behavior of ground anchors has received considerable attention in recent years, but relatively little attention has been paid to the time dependency of deformation and pull out force of ground anchors embedded in creep-sensitive soils. Using the results of experimental investigations for ground anchors in clayey soils, a method of analysis is proposed to account for creep of ground anchors in tied-back structures.

This paper presents both treatment of the bearing capacity of and estimated settlements for vertically-loaded piles in an elasto-plastic foundation ground. The solution presented is based on an elasto-plastic analysis of the interaction between bored reinforced piles and a non-linear foundation ground. Both the piles and the foundation ground are analyzed using the finite element method. In the latter case, an elasto-plastic constitutive model for dry soil conditions is also considered. Based on the results of analytical solutions and numerical analyses, a generalized model for interaction analysis between elastic piles and an elasto-plastic foundation ground has been made. The significance of the proposed generalized model is reflected above all in the fact that more accurate predictions of vertically-loaded piles' bearing capacity and deformations with different ground characteristics and geometrical data are made possible, based on a limited number of non-linear analysis results. The results of some parametrical analyses concerning the most important impact factors are also presented. The results of the research are presented in the form of diagrams which, for the selected ground characteristics, including the relative stiffness of the piles and the ground, as well as actual geometrical data, enable us to make more precise simultaneous predictions of the piles' bearing capacity and displacement. The capacity of individual piles is expressed as the dependency of allowable displacements, the shear strength and deformability of the ground, the relative stiffness of the pile and the ground, and the diameter and depth of bored piles. The second part of the paper presents a comparison between the results of non-linear interaction analyses and the results of field static tests. The conclusions provide indications as to the application of the research results in geo-technical practice. The results of the research are presented in the form of diagrams, which are very simple to use in geo-technical practice.

A database is presented of case histories showing long-term set-up of driven piles in non-cohesive soil, revealing that long-term set-up can be substantial — an average of 40% per log cycle of time being observed. The main results from a comprehensive study of the mechanisms behind pile set-up are presented. In the main, the study is based on pile loading tests, undertaken at different points in time, on piles instrumented with earth pressure cells on the shaft. In addition, dynamic testing and torque testing on small-scale rods were also performed at different points in time. It was concluded that pile set-up is caused by two strongly interconnected mechanisms: creep (stress relaxation) and soil aging (increasing dilation and stiffness). A conceptual model explaining the underlying process is presented.

An analytical procedure for analysis of laterally loaded piles in a two-layered elastic soil system is proposed. The procedure uses the variational method to obtain the governing differential equation of the pile and soil system. The important foundation properties are quantified through one interaction parameter γ for the soil layers. The γ-parameter controls the stress decay in the soil away from the pile. Either free or fixed pile head conditions may be considered. The results obtained using the proposed method are compared with experimental and theoretical results obtained by other investigators.

A design phase load testing program has been recently completed for the planned Cooper River Bridge in Charleston, SC. This $500 million (+/–) bridge is subject to major lateral loadings from seismic, vessel impact, and wind forces and thus the load testing program includes lateral testing as a significant component. A total of 12 drilled shaft foundations were subject to lateral loading at the three field test locations. The 2.5 m (8.5 ft) and 2.0 m (6.5 ft) diameter test shafts provide a measure of soil response during lateral loading with the following variations: 1) shafts with and without permanent steel liners, 2) cyclic loading in both sands and soft clays, 3) lateral loading at a high rate of displacement, 4) lateral loading using the split cylinder technique with embedded O-cells, and 5) test shafts loaded in sands which are subject to blast-induced liquefaction during the test loading. The paper provides an outline of the overall testing program, comparisons of measurements of the test results with the variations cited above, and the interpretation and use of the test data to develop soil properties for design.

Micropiles are being utilized to both enhance the foundation ultimate capacity as well as reduce foundation deflection. Load-displacement relationships may be used in the design of micropiles that conform to deflection and load-capacity criteria. Consequently, we develop a mathematical model for micropile load-displacement behavior. The model explicitly considers soil-micropile interaction. Motivated by field measurements, we assume the micropile to be partially bonded, consisting of a top debond zone that, typically, extends up to the casing depths and a bottom bond zone. Thus, the micropile is assumed to transmit its load to surrounding soil through the micropile-soil interface in the bond zone only. Furthermore, to ensure model simplicity and accessibility to designers, we assume the micropile-soil interface to be elastic-perfectly plastic. Closed form expressions of micropile deformation as a function of applied load are derived. The derived expressions are used to study the effect of model parameters on micropile yield behavior. Scaling factors are identified that are found to significantly influence the micropile load-displacement behavior. Micropile strain distribution and the consequent load transfer behavior predicted by the model are discussed. The model predictions are compared with the field measured load-displacement curves.

Driven H-piles and pipe piles are frequently used as substructure units for highway bridges constructed in the Province of Alberta, Canada. While geotechnical investigations and recommendations often precede the design of the substructure units, test piles are traditionally driven at sites to determine the capacity of the piles using "so called" pile driving formulae, the efficiency of installation of the driven piles and to assess what pile lengths are required. Where geotechnical investigation and pile driving information correlate, there is generally confidence in the proposed piling scheme. Where there are differences, very often the pile driving results tend to govern except in circumstances whereby the results show exceedingly large variances in pile lengths. This was the situation at a bridge site across Battle River in the Province of Alberta. Pile driving records of this site showed piles reaching refusal at a depth of about 55 meters (180 ft), whereas the geotechnical design indicated that piles could be stopped at about 30 meters (100 ft). Pile Driving Analyzer (PDA) testing was done during the installation of production piles to ensure that geotechnically derived design capacity values were achieved and thus optimize the depth of pile driving. This paper addresses details of the site characteristics, the details of the foundation investigation and pile driving analyzer test results. The results of the PDA testing confirmed the geotechnical recommendations.

Conventionally, medium to large bored piles deep-seated in the multi-layered soils of Bangkok were constructed by rotary drilling under wet process or slurry replacement method, using betonite slurry as support fluid in boring process. As an alternative to bentonite slurry, polymer-based slurry has been increasingly used in Bangkok recently. Polymer-based slurry having environmentally friendly properties, is found to be a good solution in minimizing the problems associated with the disposal of contaminated soils as well as the site pollution. This paper presents the performance of medium to large diameter bored piles constructed with polymer-based slurry in Bangkok. The static pile load test results of polymer-based slurry bored piles are compared with the predicted values using bentonite piles parameters. The measured frictional resistance of pile shaft from instrumented pile load tests are also reported. Shaft resistance of bored piles constructed with polymer-based slurry is found to be better than that of bored piles constructed with bentonite slurry.

This case study presents results from the Pile Installation Demonstration Project (PIDP) that was conducted as part of the San Francisco-Oakland Bay Bridge (SFOBB) East Span Seismic Safety Project. The PIDP was conducted to better understand pile handling, marine construction operations, pile driving behavior, and pile setup associated with installing large-diameter, steel pipe piles into San Francisco Bay sediments. Three full-scale piles (one vertical and two on a 1:6 batter) were driven in the area between the existing Bay Bridge and the planned replacement bridge alignment. The expected ultimate axial compression capacity was on the order of 120 meganewtons. Two large hydraulic hammers (Menck MHU 500T and MHU 1700) were used. These hammers drove the three indicator piles to the specified tip elevations at acceptable blow counts (typically less than 50 blows per quarter meter) and without pile damage. Dynamic monitoring data collected during initial driving and subsequent restrikes were used as input for CAPWAP (Case Pile Wave Analysis Program) analyses. The results of those analyses were used to estimate skin friction along the pile length, and the soil-pile setup with time. As expected, soil resistances to driving were significantly less than the estimated ultimate pile capacities due to remolding of the soil and generation of excess pore pressure during continuous driving in the primarily clay soils. The procedures used to predict soil resistance to driving and blow counts were validated by the pile dynamic monitoring data and subsequent CAPWAP analyses. The estimated setup rates were also compared with results of procedures currently available in the literature for predicting the pile setup rate.

Nine bored piles were built at the National Geotechnical Experimentation Sites at Texas A&M University: five at the sand site and four at the clay site. The piles had a nominal diameter of 0.915 m and varied in length from 10.7 m to 24.1 m. Various defects were intentionally placed in the piles including necking, bulging, soft bottom, tremie displacement, mud cake, and inclusion. Five companies were invited to make class A predictions of the length of the piles and of the defects size and location. They used small strain dynamic methods including sonic echo, impulse response, impedance logging, and sonic logging. The results are an indication of how reliable these methods are for predicting length and defects.

A large load test program was performed for the US Highway 17 Replacement Bridge over the Cooper River in Charleston, South Carolina, USA. A total of 12 drilled shafts (1.8-m and 2.4-m in diameter, 30-m or 46-m long) were constructed at three different test sites along the 11+ km (7+ mile) long alignment. The bearing stratum for all shafts was the Cooper Marl, a stiff calcareous marine clay. The load test program was structured to help answer many design-related questions, one of which is the consequences of using different drilled shaft construction methods. Shafts were constructed dry, with bentonite slurry, with polymer slurry or with plain freshwater. To evaluate bottom conditions prior to concrete placement, all shafts were inspected with a Mini Shaft Inspection Device (Mini-SID). Steel casing was driven through loose sands and/or soft clays into the Cooper Marl and was pulled after concrete placement on some shafts and left in place on others. All shafts were instrumented with vibrating wire and resistance type strain gages and axial testing consisted of single-level and double-level Osterberg Cell load testing and Statnamic load testing. For all shaft construction methods, average back-calculated unit side shear values within the bearing stratum were within 10% of each other and there were no discernible construction-method-related trends in the data. However, there was a marked reduction in unit side shear values between uncased portions of the shafts and the cased portions of the shafts. This was true even when the casing was pulled. Within the range of Mini-SID results that were observed on the project (10mm<sediment/softened zone<40mm), there was no apparent correlation between unit end bearing resistance and bottom cleanliness.

The paper deals with some geotechnical problems which were encountered in the construction of deep foundations for structures that have been under construction over the last two years along the demanding, 16.8 km long Vransko-Blagovica section of the southeast-northwest motorway link, which is part of the Slovenian National Motorway Construction Programme. Along this section the motorway runs across morphologically very varied hilly ground, so that 13 viaducts, 18 bridges, and 5 tunnels, as well as many retaining and supporting structures up to 12m high, are needed. The bedrock consists mainly of weak, tectonically altered Carboniferous — Permian rocks, which are predominantly clayey schists, siltstones and sandstones. To assist in the design of this motorway section and its structures, over 13,000 metres of trial boreholes, as well as various field and laboratory tests, were executed. A general geological description of the area is presented, together with some of the strength and deformation characteristics of the typical rock types. An example of the deep, piled and caissoned foundations which were needed for the construction of a viaduct is presented, including the assembly of geotechnical data and the execution of the works. Some geotechnical problems, which were encountered during the construction of an anchored piled wall are also presented, with emphasis on the results of geotechnical monitoring obtained during and after construction. Conformity between predicted and actual in-situ geotechnical conditions was verified by means of quality assurance/quality control procedures, according to the programme specified in the National Motorway Construction Programme.

This paper presents a quantitative analysis of the interaction between soil, structure and friction piles located in a consolidating clayey soil using the Finite Element Method (FEM). The consolidation process due to the weight of the structure and to a pumping-induced decrease in pore water pressure is simulated. The stresses developed within the soil and at the interface between pile and soil are evaluated. Progressive increase of soil strength due to consolidation is taken into account. The analysis performed provides a valuable insight into the complex interaction between soil, piles and structure and suggests some important new elements to be taken into account in the design of friction piles.

The effect of retaining structures on the performance of raft foundations enhanced with deep foundation elements (typically piles), simply known as piled rafts, was examined numerically. Illustrative piled rafts were analyzed using a simplified, linear elastic, plane strain finite element model. The effect of the structure was evaluated on the displacement profile, differential displacement, raft bending moments, and pile butt load ratio. For fully piled rafts, the effect is minimal. However, for partially piled rafts, the effect can be significant. Details are given in dimensionless plots.

Analytical solutions have recently been developed for analyzing the behavior of piles in a group subjected to loading in three-dimensional space. The piles may be installed vertically or on a batter and their heads may be fixed, pinned, or elastically restrained by a rigid pile cap. The several components of the analytical method have been coded and iteration is used to accommodate the nonlinear response of the soil around each pile in a group. With availability of a rational and convenient tool, the engineer may consider several alternatives in a design, including pile spacing, batter angle, and type of pile. This paper presents an example of use of the analytical tool to optimize a design. The example is site-specific, but the demonstrated procedures are fully general.

A database was developed at the University of Florida that serves as a depository for the results of lateral load tests on single drilled shaft and driven pile foundations. Current design software for laterally loaded deep foundations (FLPIER, LPILE) utilizes the p-y method to estimate the foundation lateral load resistance. Input parameters used to develop these curves are typically obtained from insitu tests (SPT, CPT) correlations. This paper presents the structure of the database and its current 24 SPT and 6 CPT case histories. In addition, an evaluation of predictions using input soil parameters from SPT and CPT correlations vs field measured values is presented. The comparisons showed that 4 different SPT correlations for φ coupled with 3 different k-values, all produce similar results. Therefore, little difference exists between the SPT correlation combinations. Albeit, the estimated k value has a greater effect on deformation. Similar combinations of CPT correlations also show little effect among the commonly used correlations. SPT predictions are quite conservative at low load levels; and remain conservative at high load levels. Also, the scatter is high. CPT evaluations gave unconservative predictions. In addition, the scatter (standard deviation) for both methods is high.

Liquefaction of the soil is one of the major factors affecting the behaviour of piles founded on seismically active areas. Although methods are available for dynamic analysis of pile foundations, many of them are purely elastic methods or consider only the non-linearity of the surrounding soil. Here a numerical model is presented which takes into account the reduction in stiffness and strength of the soil due to pore pressure build up and subsequent soil liquefaction, in addition to the non-linearity of the soil and pile material. Soil in the far field is represented by the displacements calculated from a ground response analysis based on the effective stress method. The interaction between pile and the surrounding soil is modelled using the Mindlin's fundamental solution with appropriate extensions to account for the soil radiation damping. Interaction coefficients and ultimate lateral pressure of soil at the pile-soil interface are calculated by taking into account the reduction in effective stresses due to build-up of pore water pressures. The non-linearity of the soil at the pile-soil interface is modelled via an iterative procedure which ensures that no where along the pile-soil interface, soil pressure will not exceed the ultimate lateral pressure of the soil. Results given by the numerical model are compared with the data recorded during centrifuge tests and the 1995 Hyogoken-Nambu earthquake. It is demonstrated that the new method gives results in agreement with the recorded data despite its relative simplicity.

In order to evaluate the axial and lateral response of instrumented steel piles, static pile load tests were carried out during the design stage in Busan, Korea. In this study, the stress analysis with FLAC code and the subgrade reaction analysis with load transfer curves according to the API code were performed consistently. These results were compared with the observed behavior of both axial and lateral load tests. Through both analysis methods, it was available to validate the axial load capacity of the pile and the load transfer in each layer. Both numerical methods could predict both the axial and the lateral behavior of piles consistently in this site.

In this paper, Artificial Neural Networks (ANNs) are applied to the prediction of lateral load — deflection relationships of group piles. The results of neural networks are compared with the measured data from the model tests. A series of model tests were performed with group piles in the Nak-Dong River Sand. For the verification of applicability of BPNN (Back Propagation Neural Network) and SNN (Sequential Neural Network), a total of 64 model test results for group piles were used. Also, in this study, the structure of neural network with one input layer-two hidden layer-one output layer was used. The number of neuron for each hidden layer determined to be 30 from the results of model test and the learning rate determined to be 0.9 to optimize network learning. The results show that the prediction of BPNN can provide a good matching with model pile test results by training with more than 50% of the total data. The investigation confirmed that a SNN with feedback is more effective than a conventional ANN without feedback to simulate the lateral load-deflection relationship for group piles. It is concluded from this study that artificial neural networks based on pile-soil interaction models can be developed by properly predicting and learning algorithms based on a comprehensive data set, and that useful inferences can be made from such models.

The use of drilled shafts as temporary foundations in the renovation of Camegie Hall provides an interesting case history of how traditional geotechnical solutions can be integral to accomplishing atypical and difficult projects. In the case of Carnegie Hall, the installation of drilled shafts enabled reliable interim support of the existing structure during the subsurface removal of over 4,600 m³ (6,000 y³) of rock. The project created a new 600 plus seat, multi-use, performance space beneath the main concert space of Carnegie Hall. The site provided challenges related to limited headroom, scheduling restrictions, and concerns about the preservation of an internationally acclaimed concert hall. The structural and geotechnical work was completed without imperiling aesthetics or acoustics and without interfering with the operation of the facility.

The tip (toe) capacity of drilled shafts in sands can often be many times greater than the side shear component; however, the tip capacity is often discounted from the total shaft capacity due to the relatively large displacements required to mobilize this end bearing component. Concerns of soil disturbance at the shaft tip (i.e. insitu stress relief) and cleanliness also discourage the use of end bearing as available capacity within any reasonable service load displacement criteria. Pressure grouting the shaft tip has been successfully employed throughout the world as a method of mitigating these conditions. However, there is an apparent lack of ration design procedures and construction guidelines for its use. In cooperation with the Florida Department of Transportation, the University of South Florida is researching the effects of post-grouting on shaft capacity. This paper describes two sites at which full scale shafts have been tested: (1) five shafts, including one control shaft, tipped in a shelly sand, and (2) three shafts, including one control shaft, tipped in a silty sand. The results of these tests show that significant tip capacity improvement can be realized through the use of pressure grouting drilled shaft tips.

The driving of a casing pile with the impact of the hammer at the toe or bottom of the pile is an effective method compared to driving with the hammer at the pile top. A significant reduction of driving time and noise has been observed for this type of driving. All energy of the impact hammer is transferred to the driving point at the pile bottom. Depending on the state of the configuration, part of the energy will be used to pull the casing downward. The driving of the casing is done by tension forces, instead of compression forces as is the case with normal impact driving at the pile top. So, the energy of the impact hammer is delivered to the point of the pile where it has the highest effect. The bottom driving casing technique has been developed by IHC Hydrohammer and Bouygues Offshore. This method is presently called Pile Toe Driving System (PTDS). The TNOWAVE stress wave program has been extended with the model for the bottom driving casing technique. The applied model for the simulation is discussed and a practical application is shown and acceptable results have been obtained.

Conventional dynamic piling methods are ill-suited to the urban environment. The press-in method offers an alternative technique of pile installation, which allows pre-formed piles to be installed with minimal noise and vibration. Field measurements of noise and ground vibrations during press-in piling are presented and compared to existing recommended limits. Based on this initial database, tentative prediction curves are presented. Equipped with these tools, designers can assess the relative environmental impact of each installation method when planning piling works.

Single concrete piles subjected to free field motion and ground lateral spreading are studied using the beam on Winkler foundation method. Soil loads on the pile are modeled based on nonlinear soil springs combined with a distributed viscous dashpot placed in parallel. These springs and dashpots are modeled using the modified Bouc-Wen model, which is also used to incorporate the variation of the pile stiffness beyond cracking loads. Three numerical examples are presented to study the importance of taking the concrete cracking effect into account in concrete pile analysis. Comparison of the predicted results with available test results shows that the presented method accurately modeled the static and seismic concrete pile behavior.

The maximum scour depth around bridge piers in sand is calculated using well established formulas based on experimental model calibrations. There are no such formulas for the maximum scour depth around bridge piers in clay. In practice and by conservatism the maximum scour depth in clay is taken to be equal to the maximum scour depth in sand. However no such evidence exists and common sense tells us that clays scour much more slowly than sand. This paper presents flume test results of pier scour in clay. The piers are cylinders with diameters varying from 25 mm to 210 mm. The soils were a low plasticity porcelain clay, a medium plasticity armstone clay, a high plasticity bentonite clay and two uniform sands. The results show that the maximum depth of scour is approximately the same in the sands and in the clays. However the rate of scour is drastically different. This shows that in clays a site specific scour rate analysis is necessary while it may not be necessary in sands.

Most of the analytical methods for estimating the earthquake response of pile foundations are based on the dynamics of the structure and consideration of seismic pore water pressure's influences on soil dynamic behavior (e.g. liquefaction, softening) is difficult. The objective of this paper is to describe a dynamic coupled analysis for earthquake response of pile foundations and investigate the effects of seismic pore water pressure's generation and dissipation on dynamic pile-soil interaction. According to the character of earthquake effect, soils and pile foundations, the research model of the earthquake response of pile foundations is established. The model can consider the dynamic pile-soil-structure interaction. Based on the basic dynamic equations for porous media, the dynamic coupled analysis with the model is recommended to predict the earthquake response of pile foundations. The equivalent linear approach is used to compute the approximate nonlinear response of soil. The calculation procedure is also reported in detail. Finally, using proposed dynamic coupled analysis theory and software for earthquake response of pile foundations, a pile foundation is calculated and such results as acceleration, vibration pore water pressure and subsidence are given and analyzed.

Satisfactory methods of analysis of the raft-pile-soil interaction are available by now. The search for design criteria more rational and economic of the conventional ones is thus justified. The present paper discusses the guidelines for an optimum design. It is suggested that different criteria should be applied for "small" and "large" piled rafts. In the former case, the primary reason to add piles is that of achieving a sufficient factor of safety; in the latter the addition of piles is essentially intended to reduce mean and/or differential settlement. In both cases, it is shown that substantial savings can be achieved in respect to the present design approach, by considering the raft-soil-pile interaction.

The piled raft foundation has shown its validity in the last two decades as a very economic geotechnical foundation type, where the structural loads are carried partly by the piles and partly by the raft contact stresses. The structural serviceability requirements regarding the settlements and tiltings of buildings can be fulfilled with relatively fewer piles in comparison with a pure piled foundation. This foundation system was successfully applied in stiff as well as soft subsoil. An innovative application of the piled raft is its special adjustment to cases of foundations with large load eccentricities or very different loaded parts of buildings to avoid the need of complex settlement joints especially below ground water table. Extensive measurements of the load transfer mechanism of piled raft foundations during and after the construction were performed to verify the design concept and to prove the serviceability requirements. Calculation procedures to model the behavior of such complex three-dimensional problems have been developed since the 1970s. But some important requirements concerning the raft stiffness, the nonlinear behavior of the pile support and the slip developing along the pile shafts even under working loads were not sufficiently considered in these analyses. For these reasons an improved numerical model based on a mixed technique of the Finite Element Method (FEM) and the Boundary Element Method (BEM) was developed taking into account all above mentioned effects. The results of measurements and of the developed calculation method will be shown and compared in a short overview.

In situations where a raft foundation alone does not satisfy the design requirements, it may be possible to enhance the performance of the raft by the addition of piles. The use of a limited number of piles, strategically located, may improve both the ultimate load capacity and the settlement and differential settlement performance of the raft. This paper outlines the development of a simplified method of analysis which can provide a useful tool for preliminary design of piled raft foundations. It involves two phases: 1) The assessment of the overall foundation behavior and 2) The assessment of the behavior under individual column loads. In both cases, use is made of simplified solutions to compute foundation stiffness and capacity characteristics. The selection of design geotechnical parameters is an essential component of both design stages, and some approximations for estimating the necessary parameters are summarized. Typical applications to a case history of a piled raft and to model centrifuge tests are described, and it is found that the behavior predicted by the simplified analysis is broadly consistent with the measured behavior.

In India a number of tunnels have been constructed and are under construction in varied geological conditions. The entire Himalayan ranges are of young origin and consist of geologically active mountains induced with locked up tectonic stresses, squeezing ground conditions surcharged with water. The proposed tunnel crosses the Pir Panjal ranges in higher Himalayan ranges connecting the Lahaul & Kullu Valleys. At the tunnel location, due to high overburden, steep slopes, dense forest covered with perennial snow, it was difficult and expensive to conduct geo-technical investigations and predict accurately the geological features along the entire tunnel alignment. To avoid geological surprises during tunnel construction a comprehensive site specific geo-technical programme was formulated for obtaining design data to meet any eventuality during excavation. The paper discusses the various investigations carried out in most difficult terrain conditions for an elaborate design for the proposed 9.0 km long highway tunnel at elevation 3,000 m under overburden up to 1,900 m.

Loess is found in many places in Northeast Thailand including Khon Kaen Province. It is generally considered as a soil with good engineering parameters. However the collapse of loess was found in many construction projects due to an increase in its moisture content. This paper presents some research results which show the strong effect of moisture content on loess behavior. An increase in moisture content caused the loess internal friction angle to decrease. Furthermore, an increase in moisture content caused large settlements of footings founded on the loess. A case study is presented herein. The water supply pipe in front of a three-story building founded on spread footings in Khon Kaen University was broken. The loess supporting the foundation at the middle part of the building was soaked for many days and then the foundation severely settled. Some reinforced concrete structures were damaged. An underpinning technique was employed for the rehabilitation. So-called composite piles were installed for transferring the load from the building to a laterite layer overlain by the loess layer. The rehabilitated building became serviceable again.

This paper deals with the quality control methods and procedures, which were employed for foundations of a steel frame high-rise building, involving twenty-five (25) stories above level and two (2) below street parking levels. The main building tower is supported on fifty-two (52) drilled straight shaft caisson members bearing within sound shale formation. The installed lengths of caisson members range between approximately 160 feet (48.8 m) and 170 feet (51.8 m). Bearing diameter of the individual caissons varies between 3 feet – 8 inches (1.1 m) and 6 feet – 10 inches (2.1 m). The caissons were installed at the base of an approximately twenty-five (25) foot depth excavation below street level. The field control included proof testing and visual examination of all excavations for drilled straight shaft caissons, plumbness evaluation using laser alignment equipment, and comprehensive concrete placement control operations. Statistical analysis of concrete strengths and post-construction evaluation of the in-situ caisson concrete density and strength supplemented the inspection operations. The comprehensive inspection program furnished sufficient information to assure the design team and owner of the integrity of the foundation system.

The case history presented in this paper deals with the concrete-filled steel pipe pile installation for an Integral Abutment Bridge in Brooklyn, New York, and vibration monitoring to minimize the impacts to the adjacent residential buildings. Integral Abutment Bridges do not incorporate deck joints but instead rely on flexible piles to accommodate the bridge deck movements. The project site is underlain by medium to dense fine silty sand. Groundwater is approximately 7.6 m below the ground surface. The piles were driven by an impact hammer to a depth of 13.7 m to 18.3 m to achieve the required capacity. The residential buildings, located as close as 3 m from the pile driving activity were typically two-story masonry/timber framed structures with basements and were approximately 50 to 70 years old. Pile load tests, vibration-monitoring using three-component seismographs, monitoring of the existing cracks with crack gages, and field surveys to monitor settlements of the nearby structures were performed during the pile driving operation. Field measured Peak Particle Velocities were plotted against distance to obtain an Attenuation Graph for the specific hammer energy. This data may be applicable for other projects with similar conditions. This paper provides the summary of the results obtained from the above tests, as well as pile driving and vibration control criteria, to minimize the probability of damages to the nearby structures.

The use of a subsurface retaining wall to arrest an active landslide offers a viable alternative to remedial grading. This paper presents a case study of the application of such a wall in the densely populated suburban residential neighborhood of San Ramon, California. The slide was directly impacting several homes and the abutting wildlife preserve. Due to the spatial constraints imposed by the residences and the preserve and the potential for additional movement during construction, the use of a retaining wall consisting of tieback-reinforced drilled piers was deemed the optimal remedial choice. The wall consisted of 69 0.9-meter-diameter drilled piers reinforced with 18.3m to 21.3m W27 x 129 steel sections. During construction, additional piers were placed at the deepest location of the slide plane to assure adequate resistance. Two rows of tiebacks were placed along the wall. Tiebacks were drilled through the slide plane and grouted in place within competent bedrock. To insure that tieback resistance was developed only from behind the slide plane, a section of the tieback borehole at the estimated slide plane/competent bedrock interface was backfilled with a bentonite slurry to act as a non-structural grout. The tiebacks were tested with and without the bentonite plug to verify its effectiveness in limiting the tieback resistance to that section behind the estimated slide plane. The construction proceeded in May 1999 in a top-down manner. Following pier construction, the slope was excavated to allow for installation of the top row of tiebacks. Following completion of top tieback installation, the excavation was continued to the bottom row of tiebacks. After installation, the slope was rebuilt with engineered fill to resemble the original grade. The slope was hydroseeded, and several horizontal drainpipes were installed to assist in drainage of the slope. The entire project was completed within four months, on schedule and within budget in September 1999. The completion of the wall demonstrated that the use of a subsurface retaining wall could successfully meet engineering criteria required to arrest a landslide within the tight spatial and environmental constraints imposed by a suburban/urban environment.

Large diameter friction caissons were adopted as foundations for a 66-storey tower block in the Singapore bouldery clay formation. The caissons were designed to carry working loads primarily in shaft resistance. Parametric studies involving interaction between the caissons and caisson cap were carried out to determine the load distribution and settlement behavior for a group of eight caissons. Insitu jacking tests performed on caisson lining shells during shaft installation verified that significant skin friction resistance could be obtained in the bouldery clay. Monitored settlements during construction of the superstructure confirmed that tolerable settlements could be achieved with large diameter friction caissons. Comparison is made with the performance of caissons for two other nearby sites.

Two static load tests on H-piles were conducted in Jacksonville, Illinois to evaluate how well methods for predicting capacity based on driving behavior agree with results from static load tests. The predictive methods included dynamic formulae, WEAP analyses, PDA, and CAPWAP. The pile load test program consisted of dynamically monitoring three test piles during initial driving and at selected re-strike intervals over a three-week period. In addition, re-taps were conducted over several weeks to identify and quantify the effects of time on pile capacity. Two static load tests were conducted on one of the driven piles. Results from the two static load tests are compared with predictions made by several dynamic methods. To determine uniqueness of these load tests, results from the Jacksonville load tests are compared with the results from a database of driven piles (and accompanying static load tests) containing over 100 load tests. It can be seen for the predictive methods investigated and for the EOD and BOR conditions, results from the Jacksonville load tests fall in the middle of the data from both databases. This indicates the load tests conducted at Jacksonville yield typical results, but also provides the reader with a perspective on the range in capacity that may be predicted using these different methods and employing EOD or BOR conditions.

The aim of this paper is to review the performance of jet grouting after the Kocaeli Earthquake and to discuss the quality assurance during production. The jet grouting technique was implemented for soil improvement under several structures at the Ford Automobile Plant. The plant's production facilities have been in operation since April 2001 in Golcuk, Kocaeli, Turkey. The structures and their foundations were thoroughly checked and evaluated after the Kocaeli Earthquake on August 17, 1999. The earthquake epicenter was only 1 km away from the site. At the time of the earthquake, 90% of the foundation works were completed. An extensive testing and investigation program was adopted after the earthquake to verify the performance of the jet grouted system and to reassess the soil conditions. It was then decided to proceed with the construction in January 2000, with certain foundation design revisions.

This paper evaluates the reasons for pile sweep ("dog-legging") and collapse during pile-driving operations for the pier foundations of the Conrail Crossing Bridge on the Waterfront light-rail line in Cleveland, Ohio. The pile foundations consisted of 355-millimeter (mm) [14-inch], closed-end, steel-pipe piles that were driven and then filled with concrete. The piles tended to sweep and, in some instances, they collapsed when driven using equipment without lateral support (free leads or a pile template). For the foundation construction of the bridge abutments, which were completed under a different construction contract, similar piles were driven with equipment with lateral support (fixed-lead). This paper discusses the problems of pile sweep and collapse, their potential causes, and the relevance of using of free leads vs. fixed leads. Finally, the paper presents a hypothesis for the problem and conclusion.

The object of this publication is to alert readers about the problems and unknowns associated with timber piling. Included are 21 case histories of problems with treated wood piling dealing with installation, usage, marine effects, and environmental impacts. Eight of the case histories concern the most serious problem, the breakage of timber piling during construction (see Figure 1). Usage problems include downdrag, settlement, degradation of wood, and potential fire hazards. Problems in salt-waters include marine borer attack, Chromated Copper Arsenate (CCA) sheathing that fails in about 10-years, breaking from impact forces, and excessive spiral grain. Environmental problems involve preservatives leaching into the water. The three major building codes, BOCA, SBC, and UBC, publish creosote treated timber piling compressive design stresses of 8.6 MPa (1250 psi) for Douglas fir and 8.3 MPa (1200 psi) for southern pine. Compression parallel to grain tests on southern pine timber piling give average values as low as 15.5 MPa (1820psi). The result is an inadequate safety factor of about 1.5. To be safe, engineers should consider designing creosote treated piling with 6.2 MPa (900) psi for Douglas fir and 5.5 MPa (800 psi) for southern pine.

At an offshore site along the Red Sea coast of Saudi Arabia, marine structures were supported on tubular steel piles driven into coral and coral contaminated sands and silts. Surprises and disappointments faced during the installation of the piled foundation were, to a large measure, due to inappropriate design methodology and absence of proper soil information. The paper describes the changes introduced to the design during the execution of the project, highlights the pile test program conducted to verify the design, and compares pile design penetration with actual penetration depth. Despite corrections made to the design using site-specific load tests, the final design adopted overestimated capacities in over 50 percent of total piles driven. The unpredictable and erratic pile behavior observed, ascertains the need for more appropriate pile design and installation guidelines for driven piles in coral and coral contaminated formations.

Six cases are presented showing the benefit of dynamic testing and analysis of driven piles, comprising (1) a comparison between two types of piles and a correlation between CAPWAP determined capacity and load-movement response to that determined in a static loading test, (2) the benefit of time-dependent increase of pile capacity, (3) a study how a soil plug can be shown to replace a pile shoe, (4 & 5) the effect on pile capacity from removing soil inside the pile, (6) set-up and effect of excavation. Together the cases show that the dynamic testing can be used to investigate problems in design and construction of piles much beyond a simple capacity determination.

The results of many years of accumulated experience with deep foundations in the San Francisco Bay Area have led to gradual but significant improvements in our understanding of the behavior and key aspects of the deign and installation of pile foundations. The increased confidence from past experience has resulted in a recent trend for higher design capacities that allow more economical designs and shorter construction schedules. The paper describes several case histories that illustrate the design and construction opportunities and challenges with high capacity piles. The case histories include: (1) 356 mm and 406 mm square prestressed concrete piles driven into very dense sands at two sites; (2) Tubex piles, a special adaptation of pipe piles; and (3) composite piles consisting of soil-cement columns with steel H-pile inserts.

Settlement and subsequent cracking of a mat foundation supporting heavy power plant equipment required a guaranteed and verifiable solution. Mini-piles grouted into the underlying basalt were designed to bypass the damaged foundation. The design, load testing, and installation of the piles are described, along with the load transfer from the mat foundation to the piles.

Two indicator pile programs were conducted to support design and construction of new access bridges for the new International Terminal Complex and supporting facilities at the San Francisco International Airport. This paper is a brief synopsis of the indicator pile program for the Inbound and Outbound Ramps and Structures (IORS) and its use in confirming design axial capacities and shortening approximately 4,000 production piles. These piles were used to support both the IORS and an adjacent set of double-decked Elevated Circulation Roads (ECR). The less comprehensive ECR indicator pile program is not presented here. The IORS indicator pile program included six pile load tests (both tension and compression) conducted on 0.61-meter-diameter steel pipe piles that were driven from 20 to 28 meters into supporting soils that consisted of soft young bay mud underlain by stiffer older alluvial and marine deposits. A total of 30 load test piles and reaction piles were instrumented during driving with a Pile Driving Analyzer system. Subsequent analysis included Case Pile Wave Analysis. Program (CAPWAP) analyses and correlation with load test results. A series of pile restrikes allowed the development of pile-soil setup curves over time that were required by the owner as part of production pile driving acceptance criteria. The load tests and CAPWAP analyses showed that the axial capacity and skin friction that developed were significantly higher than the required design values. A procedure was developed to shorten the production piles that was based on the results of the test program. Pile setup curves demonstrated significant increases in skin friction and total pile capacity over time, which greatly aided pile acceptance during the construction. During production pile installation, less than about two dozen piles (less than 1%) encountered hard driving and early refusal, but they all were found to be acceptable. Smaller diameter concrete piles for nearby facilities had encountered appreciable problems with early refusal and breakage. The IORS indicator pile program results confirmed that open-end steel pipe piles were an excellent choice for this project because they provided predictable drivability and satisfactory capacity without pile damage during installation.

When designing structures that will impose a significant load over a large area, such as buildings, tanks, walls, slopes or embankments, the geotechnical engineers must address the following situations, especially when dealing with weak foundation soils: bearing capacity failures, intolerable total and differential settlements, large lateral pressures and movement, and slope instability. Geogrid-reinforced and pile-supported platforms have been successfully used for supporting earth structures. In this system, geogrids create a stiffened platform that spans weak soils and prevents deflection between piles (or columns by ground improvement techniques) or pile caps from being reflected to the surface. This paper briefly discusses the mechanisms of this system and then focuses on three case studies in which geogrid-reinforced and pile-supported platforms were used for supporting segmental walls in Brazil, embankments, and a building in the US. The first two sites have peat deposits and the foundation soils are very soft. The third site had a large cut and fill situation and the anticipated differential settlement is substantial if no ground improvement is implemented. The three projects adopted three different deep foundation techniques: vibro-concrete columns (VCC), jet-grouting columns, and cast-in-place concrete piles. In this paper, each case study provides a description of the site, the design solution, construction, and performance. The performance of each project was investigated by survey, instrumentation, and/or observation. Satisfactory performances of these projects have proved the effectiveness of this technique in reducing total and differential settlements.

This paper presents the analysis, design, implementation and construction of a Value Engineering Cost Proposal (VECP) for the support of excavation system for parts of the underground Central Artery/Tunnel in downtown Boston. The paper explores various aspects of the work, ranging from the innovative analysis techniques that enabled the VECP, to construction and contractual concerns that helped shape its final form. Topics include the constructability concerns that motivated the redesign; the negotiations between the Owner and the contractor; the conceptual and final designs, highlighting and detailing the analysis methods employed; the actual construction focusing on the flexibility of the design to adapt to field changes; and, finally, an assessment of the success of the undertaking, considering the accuracy of the design as measured by site instrumentation, constructability and cost effectiveness. Although the focus is broad, the authors hope to bring to light the realities of implementing such a VECP, while providing valuable information on the analytic tools which made it possible.

Diaphragm wall construction causes changes in the original stress state, resulting in soil movements. These deformations can be quantified using the 3-dimensional finite element model presented in this paper. The predictions made are compared with measurements made during diaphragm wall construction for the SONY excavation in Berlin. The realistic stress state in the soil which is determined results from the application of a hypoplastic constitutive law with intergranular strain. It was found that the panel dimensions and the trench excavation sequence have a decisive effect on soil movements.

Soil-Cement-Bentonite (SCB) slurry walls have been used with increasing frequency in recent years to provide barriers to the lateral flow of groundwater in situations where the strength of a normal soil-bentonite wall would be inadequate to carry foundation loads. The addition of cement to the backfill blend allows the backfill to set and form a more rigid system that can support greater overlying loads. Construction and quality control for the SCB wall is more demanding than that needed for conventional soil-bentonite slurry walls. Backfill mixing, sampling and testing of this type of wall involve more exacting procedures. Recommendations are made herein for methods to carry out pre-job design mix testing and in-field quality control testing for the most reliable results. Designing the SCB backfill is a complex issue involving conflicting actions of the various materials involved. While the SCB wall provides additional strength, permeability is one property that generally suffers in comparison to soil-bentonite slurry walls. A normal permeability specification would be a maximum of 1 x 10^–6 cm/sec. With special attention to materials and procedures, a specification of a maximum 5 x 10^–7 can be achieved. Data are presented from design mix studies and field-testing programs to illustrate the effect of increasing concentrations of the key materials in the mix design and also the impact of other factors such as time on the measured properties. Comparisons are made between soil-cement and SCB materials as used in slurry walls and other types of installations. The SCB material is normally highly variable, even when mixed under carefully controlled conditions; engineers must account for this variability in designs and when drafting specifications.

The relocation of existing retaining structures and the lengthening of an existing bridge structure are required to accommodate a widening of Interstate Route H-1 on the Island of Oahu, Hawaii. Drilled shaft retaining walls up to 7.3 m (24 feet) in height, utilizing 90 to 150 cm (3 to 5-foot diameter shafts, were used. An innovative design to lengthen the existing bridge structure utilizing a cantilever box girder abutment structure on drilled shaft foundations was used. This paper describes the design, construction and performance of Hawaii's first drilled shaft retaining wall and abutment structures. Instrumentation data including wall deflections are discussed. The performance predicted by the original design methodology is compared to the measured performance of the walls.

Extreme variations in ground conditions which categorize karstic limestone terrain inevitably create challenges during design and installation of deep foundation systems. For example, the elevation of rock head may vary greatly over short distances, while for substantial depths below rock head, one may anticipate the presence of major solution features. Such features may be entirely open, or may be partially or completely filled with products of the limestone degradation. High capacity micropiles of the type described by FHWA have, for some years now, proved to be a technically and economically viable deep foundation system in such terrain. Frequently, they have been the only plausible solution. This adaptability is a result of the large variety of small diameter borehole drilling techniques which can be exploited to ensure penetration in such conditions, and also the fact that micropiles transfer load by skin friction. The design implication is that if certain thickness of "good" rock is encountered, then a certain pile capacity can be guaranteed. There is therefore dramatically reduced risk of piles "punching through" into a solution feature as would occur in the case of an end-bearing pile. During the last several years, the authors have been involved in the design, construction and performance testing of high capacity micropiles in karst. This paper provides an overview of these experiences and provides recommendations as to most appropriate practice. Reference is made to several recent case histories.

The Dabhol Power Company (DPC) is constructing India's first Liquefied Natural Gas (LNG) Terminal on a remote strip of the western coast along the Arabian Sea about 160-km south of Mumbai. Designed to handle the largest LNG carriers, the Terminal's marine facilities includes a Jetty that extends 1750-m into open sea to reach adequate depth, a jetty head supporting the unloading arms and the control tower, four berthing dolphins, four mooring dolphins, walkway support dolphins, three navigation dolphins and one tug berth. 473 steel open-end piles (dia.762-mm) support all these structures and are driven into weathered basalt to support working compression loads up to 4000 kN. Tensile loads up to 2000 kN are supported by 610-mm diameter sockets that are bored beneath the pile tip in the underlying basalt. The subsoil consist of three subsequent layers: 1) Soft clay layer with a thickness varying between 0 and 6-m, 2) Weathered Basalt with a thickness of 1 to 5-m's, and a RQD value varying between 0 to 90%, and 3) Sound Basalt with unconfined compression strength varying between 29 and 115 MPa. A significant problem consists of defining a installation procedure for the piles, which reconcile the requirement to guarantee an adequate bearing capacity with a risk of damaging the pile tip, and the requirement of limiting the deformation of the pile tip in such way that the installation of the socket through the steel open end pile remains possible without damaging the bore hammer. An onshore test program was performed to solve this problem. To confirm the findings of the test program, Dynamic Pile Tests were carried out on offshore piles. In this way the bearing capacity of the installed piles was controlled. This paper discusses the test program which was carried out to define the installation conditions (Hammer Energy and Penetration per blow) in order to guarantee the bearing capacity in compression and tensile together. For compression piles, a penetration of 1 mm per blow is required and for tension piles 5 mm per blow for 16 m piles and 2,5 m per blow for 19 m piles is not to exceed.

This paper discusses the challenges encountered during construction of the foundation system for a large retail store in the northwest suburbs of Philadelphia, Pennsylvania. The original shallow foundation design plan had to be abandoned when a deep, historic limestone quarry, filled with miscellaneous fill material including construction debris, municipal waste and organic material, was encountered within a portion of the building footprint during excavation. Following a subsurface investigation program (utilizing SPT, CPT and rock probes) and research into feasible foundation solutions, an alternative foundation system consisting of grade beams supported on drilled mini-piles in the quarry portion of the building, and strip footings in the non-quarry portion of the building was selected. Further complicating foundation design was the highly pinnacled, sinkhole susceptible, limestone and dolomite bedrock underlying the site. Installation of the mini-piles was performed through difficult subsurface conditions — along the former quarry face, through waste material, and into limestone rock with pinnacles and voids — pushing the contractor and their equipment's capabilities. This paper highlights the site history, subsurface investigation, foundation design, and construction issues.

Pullout behavior of large-diameter steel pipe production piles (Diameter = 2,500 mm, Length = 38∼40 m), which were designed as compression piles but used as reaction (uplift) piles during a static compression load test of a pile (Diameter = 1,000 mm, Length = 40 m), were investigated. Steel pipe piles were driven 20 m into a marine deposit and weathered soil layer and then socketed by drilling 10m into underlying weathered and soft rock layers, following which the drilled sockets and pipe were filled with reinforced concrete. The steel pipe and concrete in the steel pipe zone and concrete and rebars in the socketed zone were heavily instrumented to measure strains in each zone. The pullout deformations of the reaction pile heads were measured with LVDTs. Over the course of the study, a maximum uplift deformation of 7 mm was measured in the heads of the reaction piles when loaded to 10 MN, and 1mm of residual uplift deflection was measured. In the reaction piles, about 83 % and about 12 % of the applied pullout load was transferred in the weathered rock layer and in the soft rock layer, respectively. Also, under an uplift force of 10 MN, developed shearing resistances in the weathered rock layer and soft rock layer were as large as 125.3 kPa and 61.8 kPa, respectively. Thus, the weathered rock layer, located above the less weathered soft rock, can be utilized as a resisting layer, in which substantial frictional force can be mobilized.

The replacement of the existing I-95 Fuller Warren Bridge over the St. Johns River in I Jacksonville, Florida, required the installation of more than 750 drilled shafts founded in limestone and marl for support of the new 2,304-m (7,559-ft) long bridge. This project included Osterberg load cell tests on 915, 1220, and 1830-mm (36, 48 and 72-inch) diameter drilled shafts located on land and in the river to confirm shaft capacities and to establish shaft tip elevations for construction. One of the test shafts had a single O-Cell® while the remaining three test shafts had two levels of O-Cells to determine both ultimate friction and end bearing resistance. The two shafts in the river were installed with isolation casing to eliminate shaft friction above the design scour elevation. The paper describes the design of the drilled shaft foundations and the installation of the test shafts. The paper also presents and discusses the results of the O-Cell load tests, and compares the test results with predicted design values. In addition, the paper provides a comparison of the test results with values predicted using methods described by O'Neill and Reese in the current FHWA manual on drilled shaft construction and design. The data evaluation also includes an assessment of the influence of the hydrostatic pressure of the plastic concrete on friction resistance by comparing the results of two 1,830-mm (72-inch) diameter shafts.

The replacement of the existing I-95 Fuller Warren Bridge over the St. Johns River in Jacksonville, Florida, required the installation of more than 750 drilled shafts founded in limestone and marl for support of the new 2,304-m (7,559-ft) long bridge. This project included a lateral load test performed on two instrumented 1,830-mm (72-inch) diameter drilled shafts founded in marl and located in the river in water approximately 15 m (50 ft) deep. The shafts were installed with a center-to-center spacing of 7.13 m (23.4 ft), and were loaded by jacking the shafts apart above the water level. The two test shafts were installed with isolation casing to eliminate lateral soil resistance above the design scour elevation. The paper describes the design of the drilled shaft foundations for lateral loading, including the design p-y curves developed from the site investigations. The paper also discusses the installation of the test shafts and the instrumentation used for monitoring the load test. The results of the lateral load test are presented and discussed, and compared with predicted shaft performance. Among the issues discussed is the effect of concrete cracking on the flexural rigidity of the shaft.

It is now universally acknowledged that in bored piles (drilled shafts, caissons) flaws can, and therefore will, occur. The majority of the important flaws can be detected by standard non-destructive testing methods. There are situations, however, when none of these methods is viable. One example is that of small diameter piles, penetrating through rockfill with large cavities and then socketed into rock. The Israel Ministry of Construction and Housing initiated a research project to evaluate the applicability of single hole ultrasonic testing (SHUT) to such piles. The following parameters were to be investigated: Minimum age of concrete that enables testing, the most suitable tube material, tube diameter and most advantageous probe spacing. The goal of this work was to establish the performance envelope of the method and propose a suitable specification for the test procedure. A suitable test site was selected in Ma'ale Modi'in, Israel, where the subsoil consists of weathered hard dolomite with hard clay pockets. Altogether ten piles were tested, of which eight contained known defects. All testing was performed using the CHUM system manufactured by Piletest.com Ltd. The results prove that SHUT is a very suitable means for integrity testing of small-diameter piles, and may also serve as a clean alternative to radioactive testing of large-diameter piles.

Due to their quick and simple execution, dynamic load testing is a preferred method of Quality Assurance in the United States and around the world for both driven piles and drilled shafts. The test is particularly simple for driven piles, where the loading apparatus, the pile driving hammer, is readily available. To perform the test, strain and acceleration are measured a short distance below the pile top. The strain multiplied by the pile material's elastic modulus and the pile's cross sectional area yields the pile top force. For prefabricated, driven piles the elastic modulus varies only slightly over the cross section, and two strain transducers mounted to opposite sides of the pile usually yield the true average strain. For drilled shafts, the effort is somewhat more involved because a ram must be brought to the construction site. A crane or free release system must then drop this ram from heights ranging from 0.3 to 3 m. The acceleration and strain measurements are typically acquired one or two pile diameters below the pile top. Since bored piles without permanent casing often have uncertain cross sectional and concrete material properties, it is generally best to add a pile top extension of a thin steel pipe, one to two pile diameters long and filled with good quality concrete. The sensors are then anchored directly to the concrete after removing the steel from the attachment area. Frequently four strain sensors are utilized to improve the measured average strain because of possible variations in concrete quality and strain magnitudes over the cross section (Figure 1). The steel pipe also acts as external reinforcement and prevents the concrete from spalling due to the impact of the drop weight. A recent development has combined the measurement system with the dynamic loading system, improving and simplifying the measurements. The acceleration is still measured at the pile top, but the pile top force is calculated from the product of the ram's mass and its measured deceleration. This paper describes dynamic pile tests recently conducted with both strain and ram mounted force measurements at the Federal Highway Administration's National Geotechnical Experimentation Sites in Amherst, Massachusetts and Opelika, Alabama. It compares the dynamic pile testing results to static load tests and rapid load tests (Statnamic) performed at the sites prior to the dynamic testing.

In a current research project funded by the German Research Society (DFG) the low-strain integrity method is investigated to gain improved results in terms of localisation and identification of structural defects in piles. Variations of measurement gauges and impact hammers are examined especially to cover the high frequency content of the impact vibrations. To widen the limits of application the development of analysis algorithms is a central part of the research. The up to date analysis consists solely of the visual interpretation of the velocity signal gained from the acceleration signal measured. To increase the objectivity of the results analysis algorithms are necessary. Those algorithms based on the physical behaviour of the stress wave propagation also include additional information from site investigation and pile installation. The performance of pile tests of static or dynamic nature is standardised in Germany in the Recommendations for Static and Dynamic Pile Tests of the German Society for Geotechniques, Working Group 2.1. Moreover the IGB-TUBS uses high-strain measurements during driving and as a restrike for the continuous quality assurance on several construction sites in Germany. Depending on the local conditions a flexible testing programme consisting of three process-phases was developed. Due to the critical nature of a soft boulder clay at a site in the harbour of Hamburg, Germany extensive use of dynamic pile test was made to optimise the piled foundation of a quay wall structure. Finally finite element computations were used to analyse the influence of nearby excavation on the load-settlement behaviour of the piles.

Impulse response tests were conducted on the drilled shafts at the Amherst National Geotechnical Experimentation Site (NGES) as part of a prediction symposium. The nominally 14-m-long drilled shafts were constructed with anomalies, the nature and number of which were unknown when the impulse response tests were conducted and the analyses made. The planned anomalies included necking, voids, caving and soft bottoms. The predictions were based on conventional sonic echo and mobility plots, and comparisons with numerical simulations. The conditions at the test site are summarized, and comparisons between the predicted and as-built anomalies are presented. The rationale with which the predictions were made and the limitations of such predictions are discussed. Reasonable agreement between the predicted and actual anomalies was obtained within the upper half of the shaft, but poor agreement was found in the lower half of the shaft. After the non-destructive testing was completed, axial load tests were conducted on the shafts. These load test results are discussed in light of whether these anomalies could be considered "defects," i.e., anomalies that are sufficiently large as to impact the axial load capacity of the piles.

The application of dynamic methods to driven piles has advantages in evaluation of the hammer-pile-soil system and in data acquisition during pile driving and restrikes. Therefore during the last twenty five years, dynamic methods have become an integral part of pile capacity prediction and measurement for numerous projects. Dynamic methods use good quality hardware and software, but such great tools cannot themselves solve geotechnical problems of piling without engineering judgement. This paper shows some engineering assessments of determining pile capacity by dynamic formulas, wave equation analysis and dynamic testing.

Until recently, the analysis of Statnamic test data has typically incorporated the "Unloading Point Method" to determine an equivalent static capacity. The Unloading Point (UP) method requires that the foundation move as a rigid body, thus excluding stress wave phenomenon from the analysis. If this requirement is met the foundation capacity can be determined using this simplified method. However, many foundations do not meet the UP criteria (e.g. fixed end or relatively long piles) and have proven difficult to analyze without more complex techniques. This paper presents a new analysis method that uses measured strain data as well as the standard Statnamic test data to determine foundation capacity. This new method discretizes the foundation into smaller segments that each meet the rigid body criteria of the UP method. Thereby, a more refined inertia and viscous damping evaluation can be implemented that individually determines the contributions from the various segments. This approach, termed the "Segmental Unloading Point" (SUP) method, is developed herein and then demonstrated with results from full-scale Statnamic test data.

In this paper, two CPT based interpretation methods, European and LCPC, were evaluated for providing better predictions of ultimate pile load capacities of driven piles. Fourteen driven piles supporting bridge foundations of Interstate Highway 280 in California were selected from database. These piles were used for testing the reliability of the interpretations. Ultimate loads of these piles were measured from pile load test results by using Davisson's method. The CPT data from the test sites were used with both interpretation methods to predict ultimate pile load capacities. Comparisons between measured and predicted ultimate loads showed that the LCPC method provided the best predictions. Variations in the interpretations with respect to both methodology steps are also explained.

A methodology is developed to determine resistance factors for driven piling. Several pile load test databases are used to identify simple statistical parameters necessary to describe the uncertainties associated with predicting pile capacities. These statistics are used to determine the uncertainty with which axial pile capacities are predicted, and use these estimates of uncertainty to determine resistance factors, φ, for and LRFD approach.

Results of reliability analyses of pile groups subjected to deep-seated soil displacements are presented for two case studies. The results of these analyses are reviewed and the research findings are presented. Reliability was evaluated with respect to global pile group displacements, axial pile capacity, and bending moment. Both case studies found that reliability was dominated by the variability of the stronger soil layers that supported the structures. The case studies found that no unique relationship exists between factor of safety and reliability index for the performance modes studied.

During the design of a project, the engineer has to calculate the best estimate of the project behavior. Taking the average of several estimates is a process which is sometimes used by engineers. The TEAM approach, which stands for Technical Expert Averaging Method, takes its roots in this averaging process. The TEAM often improves the precision of the estimate (scatter around the mean) and sometimes the accuracy of the estimate (proximity between predictions and measurements). The TEAM approach is described together with various levels of sophistication in the averaging technique. Examples of TEAM successes and failures at prediction events are presented, the theory behind the TEAM is discussed, and a data base of pile load tests is assembled and used to show the improvement in precision and accuracy that can be obtained with the TEAM.

This paper aims at incorporating the results of pile load tests into pile design using the Bayesian approach and assisting decision-making on the outcome of load tests. The analysis is carried out in the framework of Load and Resistance Factor Design (LRFD), since this methodology is being seriously considered to eventually replace the conventional Allowable Stress Design (ASD). The first-order, second-moment reliability method is used to develop resistance factors for pile design using statistics of the pile capacity distribution that has been updated using the Bayesian method. A criterion of minimum acceptable test outcome is proposed for assisting decision-making in load tests. To illustrate the methodology, results associated with the design method for driven piles using information from the standard penetration test (SPT) are presented. Parametric studies are performed to evaluate the effect of number of load tests and outcome of the load tests on the updated resistance factors. For the same test outcome, the Bayesian approach allows for higher resistance factors or shorter pile length than those interpreted from the ASD approach.

Simplified reliability-based design (RBD) equations are developed that are suitable for transmission line structure and similar foundations. Two formats, Load and Resistance Factor Design (LRFD) and Multiple Resistance Factor Design (MRFD), are calibrated rigorously using the First-Order Reliability Method to produce designs that achieve a known level of reliability consistently. Based on comparisons with existing designs, target reliability indices of 3.2 and 2.6 are proposed for ultimate and serviceability limit states, respectively. Illustrative drilled shaft design calculations are given using these formats.

In the late 1980s, we proposed a set of consistent criteria for interpreting the results of load tests conducted on drilled shafts under axial uplift, axial compression, and lateral/moment loading. These criteria have been used extensively (correctly and incorrectly) over the past decade. This paper re-examines the axial compression and uplift interpretation criteria (the L₁-L₂ method), provides additional guidelines and clarifications, and recommends detailed procedures for applying these criteria in a consistent manner. Since that time, these criteria also have been applied to other types of drilled foundations, including augercast piles, pressure-injected footings, and micropiles. The basic interpretation criteria appear to hold for these other foundation types as well, with only small differences. It appears that these interpretation criteria may be appropriate across the full range of drilled foundations.

Over the past several years, a seven-step process has been developed for selecting driving criteria that has proven to be a successful formula for the installation of driven piles. The process incorporates static pile capacity estimates, wave equation analyses, pile driving analyzer use, and static load testing in a rational and consistent sequence, and has resulted in reliable and cost-effective driving criteria. The process has been used successfully for relatively small projects, but most frequently on projects that include a hundred to several thousand piles. Experience with this process on many projects has resulted in a 100 percent success rate in static load testing, with measured factors of safety between 2.5 and 3.0, avoiding excessive conservatism in design and unnecessary cost.

Axial load tests were conducted on four micropiles installed from the floor of a quarry southwest of Chicago. The micropiles consisted of 178 mm diameter steel drill casing with a wall thickness of 13 mm. The piles were drilled into dolomite with sockets that varied in length from 0.3 to 2 m, were tremie filled, and then grouted with pressures of about 140 kPa. The tests were conducted on piles instrumented with strain gages to determine axial load distributions and load transfer characteristics. The non-linearity of the response of the composite pile was considered when determining the axial load distribution in the piles. Results of the tests suggest the current practice based, on structural considerations can be quite conservative, if the rock is competent. Load transfer through side friction is governed by the interface shear strength between the steel casing and the surrounding grout.

Augercast piles have become increasingly common. Engineers are attracted to this pile type as a way to reduce foundation costs. In addition, the applied design loads have increased as the equipment used by the augercast contractors has increased to both larger diameters and longer lengths, allowing augercast piles to be considered on a wider range of project sites. The performance of this pile type requires a shaft with good integrity and sufficient soil resistance. However, inspection for this pile type is difficult. The augercast pile is constructed in situ and at no time during the process can the "hole" or injected concrete along the shaft be inspected visually. By contrast, driven piles can be inspected prior to installation. Closed end steel pipes can be visually inspected after driving and prior to concreting. The integrity of driven piles is further indirectly confirmed by driving the pile to the required blow count, and in many cases dynamic pile testing is used to confirm whether or not a pile has suffered damage, even for solid section piles like H piles or square concrete piles. While pile and shaft performance can be verified by static load tests, the cost and time constraints prevent static testing for all but a few augercast piles on even the largest site. For augercast piles, current typical practice with visual inspection does inspect the pile completion process including installation of reinforcement. However, typical inspection during the critical augercast grouting phase in the United States is often limited to counting the total number of pump strokes (and computing a total volume based on that count). This paper describes automated electronic monitoring during construction to assure proper incremental grout volume versus depth for every augercast pile on site. Dynamic integrity inspection after installation and dynamic pile capacity determination for larger percentages of augercast piles on site are discussed. These alternate inspection methods increase the confidence in augercast pile foundations.

The deep foundation provisions of the International Building Code (IBC) are based principally on those of the Standard Building Code. These code provisions have not undergone substantial evolution since 1982. Although they have served the industry well through the years, there are a number of developments in the state of the practice which have subsequently left portions of these provisions outdated, inadequate, and in many cases stilling to innovation in a discipline that is traditionally one of the more progressive in the design and construction industry. This paper begins to address the shortcomings of the IBC from the perspectives of design engineers in an effort to bring about a constructive dialogue concerning the overhaul of these provisions. Comparison is made between the IBC, ASCE 20-96, and Eurocode7. The authors recommend that the revised code be structured similarly to Eurocode7.

A static load test carried out on barrette having cross sectional size of 1.5 m x 3.0 m seated at 57 m below ground level is presented in this paper. Over 50,000 kN load was applied to evaluate the performance of this fully instrumented barrette. The load applied on barrette with a reaction frame "tower" capacity of 60,000 kN was claimed to be one of the highest test loads in the region. The test results were compared with those from the instrumented load test on bored pile of diameter 1.5 m with the same length, being located 30 m away. Since bored pile and barrette were constructed by different method and the parameters such as construction time, pile shape, slurry properties etc, were considerably different, results from the load tests have provided a unique opportunity to assess the extent of difference in behavior.

Continuous flight auger piles (CFA) procedure allows the rapid installation of piles with diameters usually ranging from 0,4 to 1,0 m and a length up to 30–35 m, free from the vibration and noise of driven piles and substantially reducing the ground loosening of bored piles. CFA are becoming increasingly popular in many countries, and sometimes they are claimed to have the advantages of both driven and bored piles, without the related shortcomings. The behaviour of CFA piles, however, is actually affected by technological factors; among them the capacity of the equipment in terms of thrust and torque and the volume of concrete pumped in during auger retrieval. In the paper the results of three failure loading tests on instrumented CFA piles are presented. The analysis of the observed behaviours confirm that the CFA piles are somewhat intermediate between bored and driven piles.

The ground vibrations induced by high-speed train have drawn engineering attention in recent years. Many influences have to be considered to decide the vibration mitigation scheme. For the train-supporting viaduct structures, the foundations were found to be one of the key factors governing the amplitude of vibration level for the nearby ground area. An overview of these topics with numerical prediction analysis and practical vibration measurement which were conducted based on Japan SKS and Euro-Star high-speed train are presented in this paper.

The cruise industry is a growing market in the United States. In order to accommodate safely today's largest cruise ships and foreseeable ships of the future, such as the Worlds City, cruise terminal design should take into consideration berthing, hurricane and earthquake loading if the site is in an active seismic zone. Because a cruise terminal is a special waterfront structure different from bridges and buildings, there is no national unified code for analysis and design. UBC and AASHTO codes are generally used. However, due to the complexity of the analysis procedure involving berthing, wave impact and seismic loading, design of cruise terminals is a challenge to engineers. This paper discusses the analysis and design procedures of a typical cruise terminal. Guidelines and construction details are described.

This paper presents the results of a study to quantify the effect of defects on drilled shafts under axial loading. These results represent the efforts of the first phase of a multi-phase project with the goal of delivering a validated computation tool that will permit the direct assessment of the load carrying capacity and performance of drilled shafts that have been rejected due to the detection of defects by non-destructive evaluation (NDE) methods. Current and historical industry practices for considering defects have been investigated and commonly occurring defects have been characterized in terms of shape, size, and location. A two-dimensional model for assessing drilled shaft behavior has been developed using the commercial finite element program PLAXIS and verified by reproducing existing axial load test data of "near perfect" drilled shafts. The validated model was then used to analyze weak layer and neck-in type defects in three cohesive soil profiles. A parametric study of the size, strength, and position of the defects was been performed and their effect on the load carrying capacity of an idealized shaft is presented. The results of this analysis show that under axial loading, the effect of a defect is highly dependent upon the geotechnical capacity of a drilled shaft. A shaft in stronger soil will experience greater loads and will therefore be more affected by presence of a defect than a shaft in weaker soil. The position of a defect is also critical in determining its effect on shaft capacity Under axial loading, the top of the shaft is the critical location. Defects located at the midpoint and toe of the shaft were found to have a greatly reduced effect on shaft capacity compared to those located at the top of the shaft.

Drilled shafts are among the most popular deep foundation solutions because of their cost-effectiveness, their high load-carrying capacity, and the minimal disturbance (vibration, noise) that they cause to the surrounding environment. However, due to the nature of their construction, minor structural flaws are sometimes introduced without the knowledge of the designer. These flaws can consist of small voids, soil inclusions or drilling slurry mixing with concrete; weak concrete; misalignment of steel rebar cages; and corroded rebars. Quality assurance procedures, which include non-destructive evaluation (NDE) techniques, are implemented to ensure that constructed drilled shafts are free of such flaws (particularly voids and soil inclusions). However, these procedures are not always successful, especially for small flaws, and some small flaws unavoidably go undetected. Drilled shafts are often designed to carry lateral loads, for example, drilled shafts in bridge abutments, cantilevered or tied-back retaining walls or bridge and building piers subjected to asymmetric vertical loads or lateral loads from wind, vessel impact or seismic events. Hence, it is important that designers consider the effects of minor, undetectable flaws on the flexural resistance of drilled shafts. This paper presents the results of an experimental study that consisted of scaled laboratory tests and large-scale structural laboratory tests designed to investigate the flexural capacity of drilled shafts with such minor undetectable flaws. The study showed that a concrete void occupying 15 percent of the cross-sectional area of the drilled shaft, penetrating inside the core of the shaft (defined by the rebar cage) arid having a length along the centerline of the shaft equal to the diameter of the shaft, may reduce the flexural resistance of the shaft by about 32 percent of the resistance of an intact shaft. The ductility of the shaft due to the presence of an anomaly was also affected. Conclusions concerning loss of capacity and ductility for other flaws are also presented for pure flexural loading.

Drilled shafts are often designed to carry lateral loads, as in cantilevered or tied-back retaining walls, bridge abutments, building and bridge piers subjected to asymmetric vertical loads or lateral loads from wind, vessel impact or seismic events. The nature of construction of drilled shafts sometimes involves the introduction of structural flaws. Flaws large enough to be detected by non-destructive evaluation methods (NDE) are almost always repaired or the drilled shaft is replaced. However small flaws could remain undetectable and may adversely affect the moment capacity of drilled shafts, especially when they occur near the positions of maximum moments. It is therefore important to investigate the effect of such undetectable minor structural flaws on the flexural capacity of drilled shafts. A numerical model was selected, verified and calibrated against data obtained from an experimental study that addressed the effect of various types of various minor structural flaws on the flexural strength of drilled shafts. The adopted numerical model, after calibration, reproduced moment capacities for the tested shaft specimens with sufficient accuracy (avg. error of 2.8 %). The calibrated numerical model was then used to study the effect of a wide range of flaws on the flexural capacity of drilled shafts. Reduction factors were calculated for worst-case scenarios of single flaws as well as flaw combinations assumed to occur simultaneously at positions of maximum moments. The numerical study revealed that the most severe reduction in flexural capacity for a single flaw was about 27 percent and was associated with a void flaw that occupied 15 percent of the cross section and penetrated inside the concrete shaft core. Reductions in flexural capacities were also calculated for cases were more than one flaw occurred simultaneously.

This paper analyzes the application of a technique for improvement of loose sandy soils supporting heavily loaded structures using shallow foundation. The technique used was deep compaction through sand columns driven with Franki-type equipment, and shallow compaction by vibratory plate. Results of the compaction processes are presented and discussed to provide guidance for future projects. The discussion includes distance from the sand column, initial relative density, time delay for results verification after compaction, and depth. The analyses demonstrate the method efficiency.

Augered, pressure-injected displacement (APID) piles develop significantly higher capacities than conventional augercast (APG) piles in loose to medium dense, granular materials. During APID installation, a Ko to Kp environment is created in the materials surrounding the pile immediately prior to casting, and this condition results in increased shaft resistance in soils where displacement leads to densification. Current augercast design methodologies assume a Ko to Ka condition in the vicinity of the pile, and thus the body of knowledge associated with conventional APG piles would be expected to be a conservative base line from which to begin assessing the capacity APID pile systems. This paper offers an empirical design methodology for APID piles, based on evaluation of 28 load tests, primarily in granular materials and in a variety of geologic settings. A design process that includes modification for soil types within the coarse-grained range is presented for CPT and SPT exploratory information.

Ground improvement using stone columns was used for the foundations of an extension of a pump house. Three identical new centrifugal pumps are to be installed in such extension. Each pump weighs about 15.8 tons. The pumps will rest on one rigid footing measuring 4.5 x 12 m. In addition to the static load carrying capacity, the design requires that a dynamic modulus of at least 500 MPa should be achieved for the foundation material. It was therefore necessary to improve the ground using vibroreplacement techniques. The fieldwork prior to and after the installation of stone columns consists of Cone Penetration Testing (CPT), Standard Penetration Testing (SPT), seismic cross-hole test and full-scale plate-loading test. Results have shown that the rigidity of the plate for the full scale loading is critical for the evaluation of the efficiency of the ground improvement technique. Furthermore, the CPT, SPT have yielded similar results indicating that the stone columns have improved the foundation material substantially. In addition, the seismic data confirmed the penetration resistance data. The details of testing and findings are presented in detail in this paper.

The drained side and tip resistances of drilled shafts in cohesionless soils were evaluated from field load test case histories with axial uplift and compression loading. The side resistance was evaluated using the effective stress or beta (β) method, and the tip resistance in compression was evaluated using bearing capacity theory. For the side resistance, the results show that β values can reach 6.5 at shallow depth, but they decrease with depth. The β values in uplift and compression are essentially the same and generally vary by less than 4%. Comparing the measured and predicted (traditional) values of β, the predicted β values are low and overly conservative. For the tip resistance, the average ratio of the measured to predicted values is about 0.3 at a displacement/diameter of 4%. This ratio shows some scatter at shallow depths, but the scatter decreases with increasing depth. To mobilize the tip resistance fully, a displacement/diameter of about 10% is required. Based on these data and evaluations, design recommendations are proposed.

Piles and piers are extensively used as laterally loaded structural members, especially for offshore platforms, chemical plants and communication tower foundation structures. Engineers are often confronted with the task of determining the maximum deflections and moments of laterally loaded pile or piers. The problem is complex because it involves bending of the piles or piers along with the soil movements. Several researchers have tried various methods to estimate the laterally loaded pile response ranging from closed form solutions using a constant value of "reaction modulus" to linearly varying reaction modulus using a finite element elastic approach or finite difference, ‘p-y' type approach. Finite difference techniques or ‘p-y' methodology involves defining the soil resistance (p) along the length of the pile as a non-linear function of lateral pile deflection (y). The analysis involves generating a set of ‘p-y' curves and solving a fourth order differential equation to determine the pile response. The computer program LPILE developed by Lymon C. Reese and Shin-Tower Wang is available to perform such an analysis. The program utilizes ‘p-y' criteria developed for different kinds of soils by various researchers. The present study includes the analyses of different pile types with lateral load and moment (i.e., lateral load at an eccentricity from the ground surface) applied to the pile head at the ground surface in uniform granular soil deposits. Utilizing the LPILE program, pile head deflections and maximum bending moments were determined for various pile types and pile sizes. A total of about 720 runs were performed. The results of the computer runs were then normalized to obtain simple equations. The equations can be easily used to determine the pile-head deflection at the ground surface and the bending moment given the applied load and moment on a pile-head, soil friction angle (φ) and diameter of the pile. Findings of this study are limited to pile and pier-head deflections of about 50 mm (2 in.) at the ground surface. The equations presented herein should be utilized for preliminary design purposes only. For an accurate determination of pile behavior, LPILE program or such other routines should be used Similar study was performed for uniform clay profile and is reported elsewhere by Endley et al, Vennalaganti et al., and Vennalaganti et al. Similar research was performed by Duncan, et al. The results of that research proposes method known as "CLM" method to determine the deflection and moment of the laterally loaded piles and piers. In Duncan's CLM method, for a pile or pier applied with a combined load and moment, a series of steps (about 6 steps) are required to compute pile head deflection. These series of steps are associated with nonlinear superimposition as the independent load induced deflection and the isolated moment induced deflection cannot be simply added. The present research has eliminated these steps and a unique closed form solution is established. The required nonlinear superimposition is simplified and this can be seen from the findings of the present research.

Factors that influence axial capacity of piles in sands such as soil parameters, pile parameters, loading effects and installation effects are listed. Physical phenomena such as grain crushing, pile shaking, stress-redistribution and plugging which all influence pile capacity are also examined. To better understand the method of prediction, each parameter and the associated uncertainty is revisited. It is evident that these factors are inter-related and sometimes compensating. This discussion also draws attention to the influence that pile installation procedures can have on pile capacities arid highlights the need for field experience with various pile installation procedures for designers. In summary, while the various available methods of pile capacity prediction are deceptively simple to use, the associated uncertainty of each method cannot be ignored. The possible savings generated by an improved method are too substantial to ignore.

Glacial soil deposits usually contain inter-bedded layers of fine and coarse grained materials. Pile capacity prediction in glacial soil is difficult because a clear boundary between fine and coarse grained soils can not be established easily. Unfortunately, the conventional pile capacity prediction methods require classification of soil as either cohesionless or cohesive. The piezocone (CPTU) can provide a continuous soil profile and a better "picture" of soil stratigraphy. But the limitation of piezocone based pile capacity predictions is that reliable soil sampling and/or soil classification is needed since the pile capacity estimate can vary widely depending on the soil type. Therefore, design pile capacity is less reliable when soil sampling is not possible. The paper discusses the authors' approach for pile capacity prediction in a glacial soil profile where soil sampling was not possible.

Design and analysis methods for use in the design of groups of laterally loaded drilled shafts are reviewed. The methods in current use are primarily adapted from methods developed for driven piles, and very little data is available for validation of these approaches which is specific to drilled shaft groups. The available data for drilled shafts are reviewed within the context of drawing conclusions about the adequacy and validity of currently used design methods and their corresponding reduction parameter recommendations for drilled shafts. Statistical methods are used to identify parameters for use in p-multiplier and group reduction factor approaches, and compared with values currently used or recommended in common design standards. The potential contribution of the pile cap is found to be very significant, and calculation methods for this contribution are reviewed.

The pressure-injected footing (PIF) can develop significant uplift capacity in cohesionless soils. This high capacity is developed mainly by the unique shape of the PIF and the soil compaction effect caused by the installation process. In this paper, a rational approach is proposed for evaluating the axial uplift capacity of PIFs in cohesionless soils. This approach is based on the general effective stress model used for drilled foundations, taking into account the special construction effects that alter the in-situ soil conditions and the unique PIF geometry. Case histories of axial uplift load tests were collected and analyzed, and it was found that three distinct effects were evident: (a) compaction effect, (b) shape effect, and (c) change in stress state. These factors were quantified from the available data. The resulting analysis model that incorporates these factors is versatile and can be used to evaluate the axial uplift capacity of PIFs in a straightforward and rational manner.

Methods were evaluated for assessing axial compressive capacities of augered, cast-in-place (ACIP) piles in the Pleistocene terrace deposits of the Texas Gulf Coast and Recent to Modern alluvial soils. The study involved a combination of data base analyses and the performance of new loading tests on instrumented piles. The results in sand deposits mirrored earlier studies that indicated that common methods used for designing drilled shafts are also appropriate for ACIP piles, although the patterns of load transfer in the tested piles suggested zones of both lower load transfer (running sands below the water table) and higher load transfer (compacted surface soils) than would be expected in drilled shafts. The action of drilling and grouting resulted in net increases in lateral effective stresses in the soil around the tested piles, near the ground surface, where such measurements were made. In overconsolidated clays, slightly higher unit side resistances were inferred for ACIP piles than are predicted by common drilled shaft design methods. However, common drilled shaft design methods, notably the FHWA method, produced generally accurate capacities for mixed soil profiles.

The behavior and performance of barrette piles in Bangkok subsoil has been evaluated based on the data of test barrette piles with full instrumentation of vibrating wire strain gauges and mechanical extensometers. The skin friction and end bearing capacity of tested piles were separated according to load transfer phenomena. The skin friction factors α and β values as well as end bearing coefficient, N_q of barrette piles were compared with those of bored piles of tip level in various soil condition as well as with and without applied base grouting. The skin friction factors α and β of barrette pile were found to be more or less the same as bored piles. The N_q of barrette pile, however was much less than that of bored piles. The differences in plane strain effects on the skin friction between narrow side and wide side of barrette piles are limited.

Very often highway departments drive test piles at proposed bridge locations to refusal or to a predetermined set and utilize these records to determine the pile capacity using the Hiley and Engineering News Record pile driving formulae. This test pile driving also provides information on the depth to which the piles can be driven and on problems that may likely be encountered during production piling. The pile capacity obtained from pile driving formulae is generally used by the structural engineer to undertake the preliminary design of the bridge foundations. The use of the pile driving approach to capacity determination often works well when the ground is competent at relatively shallow depths. However, where piles cannot achieve refusal unless driven into stiff or hard ground a geotechnical evaluation of pile capacity becomes more relevant and is often relied upon for pile capacity determination. This paper describes a site where H-pile and closed end pipe piles attained refusal at a depth of 31 metres in hard clay till and where the geotechnical evaluation recommended that the pier piles be terminated at a higher elevation. To demonstrate that the geotechnical recommendations were acceptable, static load testing and Pile Driving Analyzer tests were undertaken. The detailed testing program demonstrated that the driving of piles to refusal was not necessary to achieve the desired pile capacities and that conventional static analysis provided capacities that were sufficiently reliable for design.

In designing drilled shafts with expanded bases, often referred to as belled piles, it is necessary to determine whether these expanded bases can be constructed to the desired design sizes at the depth where these piles are to be founded. In fairly homogeneous cohesive ground, this does not pose a problem and often drilled shaft test piles can confirm the adequacy of the construction of these piles. However, where the ground is variable and non-homogeneous and contains sand deposits, this is not readily determined and the dimensions of piles may have to be varied to meet the design criteria. This situation poses a construction problem, increases the project costs, and may lead to contractual problems if piles have to be redesigned on site during the installation process. This paper describes a case history of fifty one (51) drilled shaft belled piles in which the design shape of a number of the piles had to be changed on account of erratic subsoil conditions. A variety of pile shapes — belled piles with long bottom shafts, short belled piles, straight shaft piles and double belled piles were utilized during installation. These pile shapes were not predetermined but rather dictated by existing conditions encountered during installation. The paper addresses the site conditions, design requirements and changes that had to be made on site to achieve an acceptable product. The need to recognize certain aspects of the subsurface site conditions prior to full scale installation is of importance to the successful installation of these piles.

The use of drilled shafts for supporting vertical downward loads from building foundation is provided in the case histories presented herein. The paper presents the results of some evaluation and design performed at three sites. A total of three sites are discussed in this paper. The paper includes results of the subsurface investigations, results of the in-situ tests that were conducted during the design phase, and the results of the load tests that were conducted on the drilled shafts prior to the installation of production shafts. The drilled shafts were designed to develop the design capacity through shaft friction and end bearing. In-situ tests were performed at two of the sites that were underlain by a weathered rock stratum. Statnamic tests were performed at two of the sites.

A key element in the design of laterally loaded piles is the determination of the ultimate resistance that can be exerted by the soil against the pile. Several methods have been published for predicting the ultimate resistance of laterally loaded piles in cohesionless soils. However, these methods often produce significantly different ultimate resistance values. This makes it difficult for practical designers to effectively select the appropriate method when designing laterally loaded piles in cohesionless soils. By analyzing the lateral soil resistance distribution along the width of the pile and based on the test results of model rigid piles in cohesionless soils collected from the published literature, an expression is developed for the ultimate resistance (including frontal soil resistance and side shear resistance) of laterally loaded piles in cohesionless soils. Application of the developed expression to predict the lateral load capacity of laboratory and field rigid test piles in cohesionless soils gives satisfactory results.

At a recent power plant project in Taiwan, large diameter (1,200 mm), cast-in-place concrete, drilled shafts were selected to support heavy equipment and structures. These were selected based on the available types of piling in Taiwan, the subsurface soil conditions, and the need for high lateral capacity under seismic conditions. The subsurface profile consisted of 3 to 5 m of a medium-to-stiff, highly plastic clay soil underlain by a dense gravel and cobble/boulder layer in a clayey matrix, which is the bearing layer for the drilled shafts. This latter soil layer, which extends to substantial depth and is locally known as the "Tamaopu" formation, is similar in appearance to a hardpan/glacial till material. The initial design required that the drilled shafts be able to withstand repeated lateral loading of up to 206 kN with less than 12 mm of movement. The lateral drilled shaft capacity calculation determined that to achieve this required capacity, the drilled shaft would require an embedment of 5 m into the Tamaopu bearing material. During the final design phase of the project, as a result of the earthquake in Taiwan and a revision to the seismic design codes, the design lateral loading on the drilled shafts was increased by approximately 125 percent to 470 kN. To accommodate the higher lateral loading, the drilled shaft design was revised to extend the drilled shaft to a depth of 9 m into the Tamaopu material. At that time it was considered that the drilled shaft might not be able to achieve the 470 kN at less than 12 mm of movement. As a result of these concerns, a decision was made to perform pre-production testing of the drilled shafts of varying depths into the Tamaopu material to verify the design. Drilled shafts were constructed and tested with embedment depths of 5, 7, and 9 m into the Tamaopu material and then tested to a total load of 920 kN. The test data clearly showed that the actual properties of the Tamaopu material were considerably better than those assumed in preliminary design. As a result of these better-than-assumed properties, the load test results showed that even the shortest drilled shaft would provide the required lateral capacity. As a result, the drilled shaft design before installation of the production drilled shafts was reduced to a 5 m penetration into the Tamaopu material, yielding considerable cost and schedule savings to the project.

Destruction of three bridges across the Danube river in Novi Sad, in the April of 1999. — during NATO bombing, caused the traffic interruption on the international highway. Belgrade-Novi Sad-Subotica, which is a part of Europian highway E-75 and international railway line Ankara-Wien. In efforts to ease the life-problems to the population of Novi Sad, it was decided to build temporary railway-road bridge, in the shortest term that is possible. The bridge is located about 75 m upstream from the destroyed concrete arch bridge system. In spite of extremely short term, the complex geotechnical investigations and foundation analyses of the columns founded on piles were performed successfully. The bridge was built in the 8 months period of time and the building process was completed in the April of 2000.

In this paper we have shown results of research on the influence of the geotechnical environment on pile bearing capacity. Experimental research was carried out at full scale and consisted of test loading of 3 (three) piles. Field investigation for defining geotechnical models was carried out by cone penetration tests. The results and their analysis strongly suggest use of nondimensional coefficients in pile bearing capacity calculations.

This paper presents an evaluation of four methods for the design of auger cast piles in mixed soil conditions. Various failure criteria for interpreting load test results were also evaluated. These evaluations were performed by comparing predicted capacities determined by each design method with the results of 32 pile load tests. Recommendations are made for defining failure when performing load tests on auger cast piles. Modifications to the FHWA method for drilled shafts are also recommended to improve its ability to predict the ultimate capacity of auger cast piles in mixed soils.

This paper presents a simple procedure to develop stiffness matrix for bridge foundations under seismic loading conditions. The methodology is based on the results of a pilot program for the Washington State Department of Transportation in which a design manual was developed to assist the bridge engineers for seismic design of bridge foundations for typical bridges in the state. The proposed procedure consists of a simple approach, which combines the results of a dynamic soil-structure interaction finite element program SASSI with the results of SHAKE, LPILE and APILE to develop stiffness matrices of pile group/shaft foundations, which are more realistic. The results are presented in the form of design charts, which show the variation of horizontal, vertical, rocking, torsional, and cross coupling stiffness components with the deflection. The results of our analyses show that the cross coupling terms are important for single shafts but can be ignored for the pile groups. In addition, we found that the contribution of the pile caps in the lateral stiffness of pile foundation can be as high as 75 percent of the total stiffness.

The load-displacement performance and the working load of bored piles are critically affected by its construction procedure; nevertheless, current design methods do not take into account these effects. These effects include relaxation of soil due to excavation of the borehole, roughness of borehole sides, recompression of soil by fluid concrete placement, intermixing between concrete and drilling fluid, and numerous other factors. Due to these factors, conservative design parameters for soil are implemented in pile design and consequently a conservative approach to design is therefore pursued. This conservativeness, which has no theoretical bases, is different from one country to another all over the world. Well-trained personnel and expert engineers can eliminate some of the construction effects totally or partly: The paper presents case studies on full scale loading tests on bored piles, constructed at different subsoil conditions. The results confirm that a close coordination between the designer and the construction team, along with loading tests on piles, can increase the reliability in assessing the pile capacity and significantly reduce the effects of construction procedure on the performance of bored piles. Finally, the paper outlines that improper sub-surface soil investigation results in a defected bored pile.

Tests on piles—driven or bored—sometimes include instrumentation for determining the load distribution. The instrumentation can range from a telltale or two through a sophisticated array of strain gages, and from tests with the limited purpose of separating shaft and toe resistances through tests for the purpose of detailing the shaft resistance distribution along the pile. Most of the time, the measurements are analyzed from the assumption that the "zero readings", which are the readings taken at "zero" time, i.e. at the outset of the test, also have registered "zero" load. This assumption is more than a little off. It neglects the existence of locked-in loads—residual load—in the pile and is one of the sources of the myth of the so-called "critical depth". Neglect of the residual load distribution is also the main foundation behind conclusions of instrumented tests that suggest shaft resistance to be smaller when the pile is loaded in tension as opposed to when it is in loaded in compression. This paper presents examples of measured distributions of residual load and true resistance, and indicates how to determine the distribution of residual load from measurements when the residual load distribution is not measured directly.

The I-15 Reconstruction Project consisted of the complete reconstruction of nearly 17 miles of urban interstate highway through Salt Lake City, Utah and the metropolitan area south of the city. The $1.6 billion design/build project included construction of 142 new bridges, each supported on driven pile foundations. Subsurface conditions along the alignment ranged from deep, lightly over-consolidated Lake Bonneville clays to dense alluvial sands above the groundwater table. Piles derived support from shaft friction or a combination of shaft friction and end bearing, depending on the strata present. Design pile capacities were evaluated using the effective stress, or beta, method. The UNIPILE computer program was selected, as it was preferred by the Utah Department of Transportation (UDOT) and because of previous successful application in Lake Bonneville deposits. The basic design parameters are side shear (beta) and end bearing (Nt) factors. The design-build contract required full-scale load tests to assist in selecting the design parameters. However, because of the construction schedule, about half the bridges were designed before the pile load tests were completed. Therefore, initial beta and Nt parameters were selected based on geotechnical evaluation of the subsurface data, combined with past experience in Lake Bonneville deposits. Nine full-scale load tests were performed at locations representative of subsurface conditions along the alignment. Static compression, uplift and lateral load tests (not discussed in this paper) were conducted, with each test pile carried to failure in compression. Load distributions in the test piles were evaluated using internal strain measuring devices that allowed the test loads to be separated into side shear and end bearing, and the corresponding beta and Nt values to be determined. Many of the test pile locations were adjacent to structures that would be designed and constructed. Therefore, correlations were made between the measured beta and Nt values and the design pile capacities calculated/predicted before the load tests. With one exception, the ratios of measured to predicted pile capacities were within 0.94 to 1.3. The single exception was a location where the test pile was driven into dense sands above the water table, and the Nt value assigned to the dense sand layer was too great. Adjustments in the design pile capacities were made for this subsurface condition as a result of the load test program, and pile design was completed using the measured design values.

Problems associated with the use of traditional piling materials in harsh environments include concrete deterioration, steel corrosion, and marine borer attack on timber piles. It has been estimated that the U.S. spends over $1 billion annually in repair and replacement of waterfront piling systems. Such high repair and replacement costs have led several North American highway agencies and researchers to investigate the feasibility of using fiber reinforced polymer (FRP) composite piles, such as concrete-filled FRP piles. These piles, if found viable, could offer advantages such as improved durability and reduced life-cycle costs. However, composite piles have a relatively short track record of performance, and there are few readily accessible, well-documented projects involving use of FRP composite piles. FRP composite materials differ from steel and concrete in properties such as anisotropy, lower stiffness, lower surface hardness, and different surface roughness. Therefore, existing pile design approaches may not be directly applicable to FRP composite piles. The principal focus of this paper is on the skin friction characteristics of FRP composite piles against sand. This paper summarizes the results of sand-to-composite pile interface shear tests on two types of FRP composite piles. The test results are compared with those from sand-to-concrete interface tests. The influence of interface shear strength on pile axial capacity of composite piles is discussed.

It is difficult to predict accurately the capacity of driven piles in sand. In practice, this uncertainty either leads to conservative designs if the capacity is under-predicted, or worse, if the capacity is over-predicted, it can lead to severe foundation problems. One significant possible cause of the problem is that currently available semi-empirical design methods do not capture the fundamental mechanisms that govern the load capacity of driven piles in sands. As-part of a recent research project at City University centrifuge tests on a driven model pile in sands have been undertaken. A fully instrumented model pile was constructed and pushed into samples of quartz and carbonate sand thus covering a wide range of mineralogies. Triaxial tests were used to determine the fundamental patterns of behavior of the sands. This allowed a general framework to be established, in which the in situ stress:volume relationship is shown to govern the capacity of the pile. The paper examines existing design methods for piles in sand in the light of the new understanding gained from the research, highlighting their deficiencies and suggesting an improved approach.

A large design phase load test program performed for the US Highway 17 Replacement Bridge over the Cooper River in Charleston, South Carolina, USA provided an opportunity to compare drilled shaft axial design methods to measured shaft performance. A primary objective of the load test program was the determination of unit side shear values and unit end bearing values for use in final bridge design. Ten axial test shafts were instrumented with vibrating wire and resistance type strain gages and subsequently loaded with single-level and double-level Osterberg Cells and/or a Statnamic device. The bearing stratum for all shafts was the Cooper Marl, a stiff calcareous marine clay with relatively interesting behavior. Using the instrumentation data, unit end bearing resistances and unit side shear resistances for the Cooper Marl were back-calculated for each shaft. This paper compares these back-calculated values to the theoretical values that are predicted using five completely different design methods. Due to the atypical behavior of the Cooper Marl, most of the five design methods failed to accurately predict the actual unit side shear and unit end resistance values.

This paper investigates the percentage of top load P, on a top-loaded compression drilled shaft (or bored pile), that reaches its tip in end bearing, B, with top compression movements mostly in the typical working range of 3–15 mm, and at "effective" length/diameter (L/D) ratios of less than 25. The authors begin with a review of elastic theory results for predicting B/P. They then show that the results of B/P predictions from finite element studies using simulated sand gave much higher B/P ratios. Finally, they introduce comparative B/P ratio data as obtained from their review of a sample of 50 full scale, randomly selected, Osterberg load tests performed on a great variety of shaft sizes, loads and geological conditions. They included many rock sockets and 15 shafts where the. test showed disturbed bottom conditions. All this produced much scatter in B/P. A literature check of 21 instrumented, conventional top-load tests gave similar B/P results, but even more scattered. All 71 Osterberg and conventional tests gave an "upper bound" of B/P = 0.56 at an average L/D = 7. Our cited elastic solutions gave B/P = 0.03 to 0.07 at L/D = 7, which indicates that they can seriously underpredict the availability and potential importance of end bearing of real shafts in real geotechnical materials. Finite element studies suggest nonlinearity, effective stress, slip, dilatancy effects and layering as reasons for the underprediction.

The 1988 edition of the FHWA drilled shaft manual provides equations for drilled shafts in compression in clay and in sand. The equations used, proposed by Reese and O'Neill, were based on data from load tests and are empirical. The design equations appear to do well in clays and in loose to medium dense sand. Practitioners who are designing drilled shafts for soils of relatively density of 50% to 100%, often consider other design methods including the one proposed by Mayne and Harris. The 1999 edition of the FHWA drilled shaft manual added to the earlier edition an equation for designing for compression in gravelly sand or gravel and also adopted the Mayne and Harris equation for "intermediate geomaterials". The authors of this paper started working on a new design equation for cohesionless soil materials of all relative densities around 1998. The development of that equation was based in part on results of the full-scale fully instrumented tests. A nonlinear large-deformation capable finite element program, GEOT2D, was also. The design equation was published in 2000. Subsequently that design equation was advanced to include cohesive soil materials and is the unified design equation presented here. The unified equation predicts side resistance of a cylindrical drilled shaft in compression. A procedure for estimating the end-bearing capacity is also presented. The end-bearing equation is related to the side resistance near the tip of the shaft. The proposed unified equation is rational and simple to use and involves lateral earth pressure coefficients. These earth pressure coefficients depend only on the cohesion, the angle of internal resistance, and the overburden pressure. The unified equation complements the FHWA procedure and enhances the understanding of the FHWA equations. It is shown that by using specific values of cohesion and friction angle in the unified equation, one could obtain predictions similar to that of the FHWA equations for clay, for sand, and for gravelly sand or gravel.