Search Documents

By Narong Thasnanipan

Performance of diameter 1.5 m toe-grouted bored pile founded at depth 44 m in sand-gravel formation of Chiang Mai City is presented in this paper. Toe grouting was applied by using tube-a-manchette method. Effectiveness of toe grouting was observed from the standard penetration tests conducted immediately below pile toe through grouting pipes at two stages, prior to grouting and three weeks after completion of grouting. Improvement of pile end resistance by toe-grouting is demonstrated with a series of back analyses carried out by load-transfer method employing hyperbolic transfer function. Construction experience of large diameter deep-seated bored piles in multi-layered subsoil and practical aspects of toe-grouting are also briefly discussed.

Jan 1, 2002

By Rick Deschamps

A reoccurring problem with support of excavations in urban areas involves the placement of external support components like anchors or nails in right-a-ways, subway and utility setbacks, in areas occupied by adjacent structures, and within the property of non cooperative neighbors. Such a situation occurred for a project in Arlington, Virginia, which ultimately resulted in the preliminary design of a large footing supported cantilever retaining wall that was extremely costly and impractical to build. An innovative design alternate consisting of a series of vertical concrete elements (VCEs) supported by post tensioned tiedowns was proposed to the General Contractor and accepted by the owner. The VCEs were designed to serve as the main structural elements for both the temporary support of excavation system and the permanent wall. The VCE?s were constructed using single bite diaphragm wall panels oriented perpendicular to the excavation face including a rebar cage and tremie concrete. Threaded inserts were placed within the rebar cage to attach the cast in place exterior wall. Overall stability and lateral movement of the VCEs were controlled by high capacity vertical anchors embedded within the VCE. A detailed instrumentation program was installed to monitor the movements of the VCEs with relation to the excavation depth. The paper present the project case history by discussing the design concepts of the VCE, the construction approaches used to install them, the construction challenges encountered, and the performance of the unique system.

Jan 1, 2008

By Dario Libano, Franz-Werner Gerressen

"Diaphragm walls are known as underground structural elements commonly used for retention systems and permanent foundation walls or elements. Of course they can also be used as deep groundwater barriers. This paper will focus on the construction work in an urban area like Sao Paulo. The Line 5 of Metro Sao Paulo is one of four network expansion projects undertaken at the same time. The projects will cover various areas of the city. The Line 5 on São Paulo’s south-side will be extended 7.1mi (11.5km) to serve 770,000 riders per day, stretching a total of 12.4mi (19.9km) along 17 stations. The extension line will facilitate access to the city’s central section and the commercial and hospital districts located along the line. The Brooklin Station, subject of this paper, had to be carried out in a special five shaft shape design, shaft diameter 131ft (40m) each, without using ground anchors. A focus will be taken on the special requirements of the installation of the arrows as well as on the specialties of installation for the shaft shaped layout. Therefore, it was required to install a 3.3ft (1,000mm) diaphragm wall using a specific layout especially in the connecting sections of the shafts, the so called arrows, to provide a bearing area for the struts. Maximum wall depth was about 121ft (37m); the total volume was about 20,060 cubic yard (15,337m³). The paper will describe the construction method and the sequence of activities required for the construction of this diaphragm wall using a combination of hydraulic grab and trench cutter system.INTRODUCTIONThe Line 5 Extension, in the city of Sao Paulo/Brazil, is currently the most important investment in public transport by Sao Paulo state’s government. It will connect the south part of the city to the city center, crossing through four other major METRO lines and stations, all currently in operation. The extension includes a new parking yard for the trains and 11 new stations. The Brooklin Station is one of the new stations and has a unique geometry of five secant shafts. The station should open and operate in 2016.The original design included constructing five secant shafts stabilized by two plastic slurry walls using conventional grab technology. One of them was to contain the contamination plumes in the neighborhood and the other for temporary support to enable the excavation of the shafts. The final retaining wall design was reviewed and changed to a concrete structural diaphragm wall. The reason of design change was as well the reduction of construction time of the shafts by half to meet the schedule as to avoid dewatering effects in a geological region characterized by Quaternary sediments and environmental contamination plumes. It was the first time that a trench cutter (hydromill) was used in such a situation in Brazil. The cutter increased the quality of the joints by overcutting, and decreased the probability of having seepage into the shafts. The cutter created a seepage barrier and avoided movement of contamination."

Jan 1, 2017

By Joel L. Volterra

Since 1910, Mueser Rutledge Consulting Engineers (MRCE) has regularly provided professional engineering services including inspection of pile load tests. Historically, such services included on-site inspection of the driving of test piles and subsequent static load tests by qualified Geotechnical Engineers. Load testing procedures are specified by local or State Building Codes, locally the 1968 Building Code of the City of New York (1968 BCCNY). Between 1968 and 2008, the 1968 BCCNY required a minimum 48-hour hold period for standard loads per pile type, an extended 96-hour hold period for higher capacity piles, and a settlement criterion that the pile maintain a test load equal to two times the design load with less than 0.012 inches of settlement over the last 48 hours of the hold period. Effective July 2008, the new 2008 BCCNY shortened the hold period and relaxed the settlement criteria to that which is more similar to other governing codes and standards such as the International Building Code (IBC 2006) and the American Society for Testing and Materials (ASTM). MRCE implemented an automated load test monitoring system to reduce costs and to alleviate or minimize overnight and off-hour staffing for load test readings. Initially, a series of linear variable differential transformers (LVDTs) and data loggers were included in the Contractor?s load test setup to electronically record pile head settlement and where applicable, settlement of telltales installed to varying depths in the pile. A pressure gage was subsequently added in the hydraulic line, then a load cell, strain gages, and remote access using modems. The ease of the setup and modular adaptive nature of this system have proven it to be a useful tool and improved our ability to better serve our clients and manage our pile load testing services.

Jan 1, 2008

By Haijian Fan

Currently, deep foundation designs for the service limit state are deterministic in nature. The settlements and lateral deflections of deep foundations are calculated by means of numerical techniques whereby the foundation-soil interactions are characterized by discrete load transfer curves such as t-z or p-y curves. The noted deficiency of the deterministic approach is that the uncertainties arising from various sources, such as soil properties and model errors are not considered. To address this deficiency, a performance-based reliability design methodology is developed using Monte Carlo statistical methods. The performance based design criteria are defined in terms of the specified allowable displacement. The input to the numerical solution techniques, including the soil strength parameters, unit weight of the soil are relevant to the generation of the load transfer curves are treated as random field. The well established random field generator algorithm, local averaging subdivision, is used to generate random samples of input, where the mean values, variances, the probability distribution, and correlation structures are required statistical parameters for these random variables. The objective of the performance based reliability design is to determine the optimized foundation dimensions so that the probability of failure is less than the target probability of failure. The probability of failure by the Monte Carlo approach is simply the ratio of the number of unsatisfactory performance to the sample size. A numerical example for laterally loaded drilled shaft is given to illustrate the application of the methodology. The incorporation of the correlation length is shown to be critical for properly accounting for the site soil spatial variability.

Jan 1, 2017

By Alexander Filotti, Richard Kalson

"1. INTRODUCTION- WHAT IS P3 OR PPP ?The Public Private Partnership (PPP or 3P) is a project delivery scheme based on the government providing land and/or a concession agreement, and on a private entity providing the design, construction, financing, operation and maintenance of the project for an extensive duration of time (up to thirty years). At the end of the operation-maintenance period, the project is turned over to the public entity, which fully owns the project at all times. While there are many PPP delivery schemes used around the world, for the present paper, we are going to analyze the impact of the Design-Build-Finance-Operate-Maintain (DBFOM) scheme on the deep foundation specialty subcontractor. There are couple of reasons that the PPP delivery system started to be used and implemented not only by developed nations (e.g. United Kingdom, USA etc.), but also by the International Development Bank for delivering major infrastructure projects in developing nations. From the point of view of government agencies, some of the reasons to use PPP are the following:??The primary use of private capital in an era where borrowing and/or supplemental taxation represents too much of a political risk. Budget deficits and taxation limits are extremely hot topics in every election (national or local) and the level of education of the electorate in this area is making it more and more difficult for politicians to use borrowed funds and additional taxation finance sources for public projects. At the same time, we see an increase in the social expenditures of the government in parallel with a needed increase in expenditures on infrastructure due to aging and neglect.??Bundling the construction risks with the operational and maintenance risks naturally pushes the private entity to increase the level of innovation and communications in the primary phase ( the construction phase) in order to develop future process based upon the innovation and communications utilized in the operation and maintenance phase of the project. This is true if contractual incentives are properly structured because “all revolves around incentives. In a world of ‘incomplete’ contracts, where it is difficult to foresee and contract about uncertain future events, it is important to get the incentive structure right” (Grimsey and Lewis, 2004).??The transfer of very expensive risks such as differing site conditions, schedule, design, construction, maintenance and operation costs from the public authority to the private entity.??The timespan of development and construction of a PPP project is usually shorter than the traditional procurement process (e.g. “ the I-595 P3 provided capacity improvements fifteen years sooner than a conventional plan would have offered. In Maryland, an innovative expansion of the Seagirt Marine Terminal in the Port of Baltimore was completed two years ahead of schedule due to the P3 model. As a result, the Port of Baltimore is only the second port on the US East Coast to have unrestricted access to both a fully functioning 50-foot channel and 50-foot berth.”)"

Jan 1, 2017

By Gerald Verbeek, Erick Baziw

Dynamic soil analysis (DSA) in geotechnical practice is intimately related to the capability of measuring the necessary soil properties of shear modulus and absorption. In DSA the methods of analysis are based upon the stress-deformation response of soils due to imposed shear strains. Typically the soil profile is numerically modeled as linear visco-elastic system. This technique is referred to as the Equivalent Linear (EL) method, and it requires the specification of the input parameters of low-strain (<10-5) shear modulus, modulus reduction, and the equivalent viscous damping ratio, which is directly related to the soil absorption value. Downhole seismic testing (DST) has proven to be a very accurate tool for estimating low strain shear modulus values as long as source wave raypaths are taken into account, especially for near-surface investigations. DST has also been utilized to estimate low-strain interval absorption values. In these cases standard frequency domain absorption estimation algorithms (such as the spectral ratio technique) are utilized where it is assumed that the source waves are traveling along the same travel path, which is obviously not the case for near surface DST investigations. This paper outlines a unique DST absorption estimation methodology, which is carried out in the time domain and takes into account source wave travel paths. 1. INTRODUCTION There is considerable interest in methods of geotechnical in-situ engineering that provide accurate estimates of the shear wave velocity and the associated absorption values in the ground, since these parameters form the core of mathematical theorems to describe the elasticity/plasticity of soils and they are used to predict the soil response (settlement, liquefaction or failure) to imposed loads (whether from foundations, heavy equipment, earthquakes or explosions).

Jan 1, 2019

By Peter Iovino

Unfortunately, I don&apos;t know the unit cost per cubic yard of treatment. I reflected on that yesterday and I don&apos;t think that we considered the use of micropiles. They could have been an option. The challenge may have been picking up some of the lateral load but I think the micropiles could have been installed on a batter and perhaps extended to rock. We could have considered them, but I do not believe we explored micropiles as an option for this project.

Feb 10, 2003

By Masaki Kitazume

"Abstract The wet type and dry type of deep mixing methods have been developed and widely applied for on-land and marine constructions in Japan. The quality assurance (QA) of the methods has been usually carried out by unconfined compression test on field stabilized soil sampled by core boring in Japan. In some other countries, stabilized soil excavated by the wet grab sampling instead of core sampling is often subjected to unconfined compression test for quality assurance. There are several merits and demerits in each method, but few studies have been carried out to compare the stabilized soil strength of these two types of specimens to discuss their applicability to practical construction. The author has conducted a series of field and laboratory tests to compare the stabilized soil strength and discuss the applicability of wet grab sampling to QA for wet and dry types of deep mixing method.In this study, the wet grab sampling was carried out at two road embankment construction sites where the wet type and dry type of deep mixing were applied. The cement stabilized soil was excavated at the three depths by a wet grab sampler soon after the DM column construction. The test result is compared with the other test data in the previous study to discuss the applicability of the web grab sampling method for QA of cement stabilized soil.IntroductionThe Deep Mixing Method (DMM), an in-situ soil stabilization technique using cement and/or lime as a binder, is often employed to improve soft soils (Kitazume and Terashi, 2013). A variety of factors and their interactions affect the eventual strength of a deep-mixed column, while some of which are controllable (e.g. binder type and dose) by engineers while the others are not (e.g. soil types, ground temperature, etc.; Babasaki et al., 1997). In addition to laboratory mix tests, which in essence are a parametric study to estimate the influence of these factors on a particular soil’s strength, the quality assurance (QA) practice in Japan usually involves unconfined compression test on sampling cores from the stabilized columns after a given curing period. It has been recognized, however, that the ‘field strength’, or the strength of the retrieved core, quf is on average lower than the ‘laboratory strength’, qul, evaluated by laboratory mix test. Also allowing for the inevitable variability, existing guidelines (for example, PWRC, 2004 and CDIT, 2008) recommend the design quf /qul values of, for example, 0.5-1.0 for the CDM (Cement Deep Mixing; a wet-type technique, in which binder is added as slurry) and 0.2-0.33 for the DJM (Dry Jet Mixing; a dry-type technique, in which binder is added as powder)."

Jan 1, 2014

By Mike Muchard

STATNAMIC load testing augmented with embedded instrumentation has been successfully used to obtain structural and geotechnical design information on large diameter spun cast concrete cylinder piles on many large marine construction projects. The paper focuses on one particularly interesting large marine project involving a new bridge to St. George Island, Florida where instrumentation and STATNAMIC load testing was used in conjunction with static load testing on high capacity piles. The information presented includes embedded instrumentation measurements taken during pile fabrication (post-tensioning), during pile driving, during static load testing, and STATNAMIC load testing. As much of this data is the first of its kind on concrete cylinder piles, it is exciting to present the foundation industry with this additional insight as to the behavior of this pile type and the accuracy of the methods used to test them.

Jan 1, 2005

? Hollow Stem Auger ? Drill Bits ? Leads/Torque Arm ? Hydraulic Gearbox ? Hydraulic Power Unit ? Grout Pump ? Crawler Crane

Jan 1, 2003

By Mark R. Svinkin

Regulations are needed to keep construction vibrations in tolerable limits. The existing regulations were developed for low-rise residential structures, but they are incorrectly used for different super and underground structures. The existing regulations cannot prevent most of harmful vibration effects on sensitive devices, people and structures from construction and industrial vibration sources. It is important to determine the areas of application of the existing regulations and bring attention to the necessity of measurement of the structural responses for assessment of vibration effects on structures. For multi-story residential, commercial and industrial buildings, the frequency-independent safe limit of 51 mm/s (2 in./s) can be chosen for the peak particle velocity of structural, not ground, vibrations. Specifications for construction operations with dynamic sources have to consider specifics of each construction site. Preventive measures can be used to mitigate harmful vibration effects on various structures and settlements in sand and clay soils.

Jan 1, 2006

By Yazen Khasawneh, Osvaldo P. M. Vitali, John E. Agee, Mohammad Nassim

A 3D FEM modeling of a ground improvement solution for an ice rink slab with stringent differential settlement limits is presented. Due to the complexity of the geotechnical conditions and strict deformation requirements, a structural slab supported by micropiles was initially proposed. However, this solution was costly and alternative solutions using ground improvement techniques were investigated. The proposed slab is supported directly on rock at one end and on approximately 45-ft deep cohesive soils on the other end. The numerical simulations show that differential settlement would be larger than allowable, if the ice rink slab is supported directly on the existing subgrade. The addition of the ground improvement scheme with compaction grouting demonstrated that the anticipated differential settlements are reduced significantly and would become less than the allowable magnitude. This study demonstrates that the use of numerical simulations to study complex loading and geological conditions will help to optimize designs and may result on significant cost savings.

Sep 8, 2021

By Mohamed A. Mansour, Ahmed M. Soliman, M. H. El-Naggar

"Grouted helical piles allow significant increase in pile capacity with low additional cost over that of conventional helical piles. They can be particularly advantageous in sites where installation disturbance can reduce or even eliminate the pile shaft resistance. For helical piles with slender shafts, a grout column offers increased buckling resistance under compression loading, and may reduce the potential settlement required to mobilize the bearing component of their plates. This paper investigates the performance of grouted helical piles installed in loose sand. Grouted helical piles were constructed and tested in a controlled laboratory environment. Helical piles with hollow pipe shaft and single helix were grouted during installation. The grout was injected into the soil under high pressure through two nozzles placed above the piles helices. The load test results showed significant improvement in the performance and load carrying capacity of the helical pile due to the formation of a cylindrical grout column along the entire shaft leading to a higher shaft resistance contribution to the total pile capacity. The results suggest that grouted helical piles with slender hollow shaft could be used to support large structural loads. INTRODUCTIONHelical piles are widely used as a deep foundation system for different engineering applications. They represent an attractive foundation option because of their cost effectiveness, ease of installation even in limited access areas, low levels of noise and vibrations, and their ability to be installed through ground water without casing. In addition, they allow immediate loading upon installation. Helical piles are installed into the ground by applying a mechanical torque to the pile head. The installation torque monitoring provides a kind of quality control during the construction of helical piles. It could be used in predicting the ultimate pile capacity using several capacity-torque correlations (Livneh and Elnaggar 2008; and Hoyt and Clemence 1989).However, helical piles installation can cause soil disturbance within the annular zone affected by the pile helices. This disturbance can introduce fissures within the pile cylindrical failure surface (Vesic 1971; Mooney el al. 1985; and Zhang 1999). Thus, the shear strength of the soil decreases along the affected zone. Bagheri and El Naggar (2013) evaluated the installation effects of helical piles in sand. They reported that as the helical blade proceeds into the ground, sand within the cylindrical path circumscribed by the helices experiences torsional and vertical shearing accompanied by sand displacement and stress changes. This process is more pronounced in the case of multi-helices and large diameter anchors. Bagheri and El Naggar (2016) evaluated the installation effects of helical piles in structured clay. They indicated that the behavior of the helical pile and its capacity are significantly affected by the degree of soil de-structuring induced by the penetration of the pile shaft and helices. In addition to the installation disturbance, helical piles have other drawbacks including; high potential settlement required to fully mobilize the bearing component of their plates, low buckling resistance of their slender shafts, and the incomplete rigidity of the coupling joints along the pile shaft that allow some lateral movement under compression loading (Vickars and Clemence 2000)."

Jan 1, 2016

By Julie Shang, Tim Evans, Michael McDonald, Yu Guo, Bingfeng Xue

Effective chemical grouting represents a technical and economic solution to many challenges related to soil stabilization and blocking the flow of water. Chemical grouts are facing increasing demands in durability, environmental impact and value. To better meet these challenges, a novel and cost effective process was developed to produce a sodium silicate that increases the ratio (i.e. silica to alkali) beyond what has been used in traditional silicate-based grouts. The move to a lower sodium form of silicate was anticipated to have two desirable performance enhancements; a reduction in leachable sodium and improvement in the durability of the silicate grout. An in-depth study was performed at the University of Western Ontario Geotechnical Research Centre to quantify improvement in durability as well as other performance properties such as compressive strength. The study began by first modelling the gelation time and reaction kinetics of the high ratio sodium silicate over a range of silicate concentrations using two different setting agents as well as with/without a hardener. As expected, reducing the level of sodium produced a much more water resistant grout. These same formulations also showed a four order magnitude reduction in permeability. Somewhat surprising, test results showed the higher ratio silicate having much higher compressive strength. At the time of this abstract, the high ratio silicate was moving to the trial stage for geotechnical applications while the petroleum industry was further along with several successful field trials for water blocking as well as cement remediation. This paper will review the chemistry, testing and features of this next generation form of sodium silicate. Introduction Sodium silicate based grouts have been used for decades for soil stabilization and water control. Over the years, several silicate-based systems have been tailored for specific application conditions to maximize performance and reduce cost. Past research has also focused on the selection of setting agents and hardeners for silicate grouts. A constant in the development effort has been the grade of sodium silicate. The vast majority of chemical grouts have been formulated using a ~3.2 ratio sodium silicate. As a very brief overview on the meaning of “ratio” and why it is so important to performance properties. The ratio describes the proportion of sand to alkali or SiO2:Na2O used to make different grades of sodium silicate. Comparing a 2.2 ratio to a 3.2 ratio, it is easier to polymerize and set the 3.2 ratio. Upon setting and forming a silica gel, the system has higher compressive strength values and improved durability with a reduced level of syneresis (i.e. shrinkage). This leads to the question if 3.2 is better than 2.2 why not go to 4.2 or higher?

Jan 1, 2018

By Rajesh A. Vaidya

A large scale piling work is undertaken in India for an important fast track industrial project. Thousands of solid square precast concrete piles are driven to depths of around 50m. Soils encountered are of alluvial origin with very soft clays varying in thickness from 10 to 18m at the top followed by medium dense to dense and occasionally very dense sand, stiff clay and finally dense to very dense sand layer as the founding layer. Recent site grading with around 4 metres thick hydraulically filled river sand necessitated driving the piles to the bottom sand layer to avoid problems on account of negative skin friction and to achieve optimum pile capacity. A process approach suggested by Bell et al (2002) to promote better understanding of the pile behaviour, pile capacity, and parameters assumed at design stage for static capacity and for wave equation analysis is followed with an aim to bring in effective economy in terms of pile length and pile installation speed. Soil investigation work is also undertaken in stages to cut down pile wastage. A judicious combination of PDA and static pile load tests is followed. Pile Driving Analyser (PDA) tests at end of driving and also at various stages of recovery of pile capacity are carried out to understand the pore pressure dissipation phenomenon. The paper attempts to present some of the valuable experiences that would go to establish precast pile as a very effective alternative to driven cast-in-situ piles and bored piles which are very common in this part of the world despite their limitations in terms of economy, speed and certain geotechnical aspects.

Jan 1, 2007

By Joanna Bzówka

Nowadays, there are some new requirements for the supermarket floors. The floors are loaded by high point- and moving-loads. In spite of this fact, it is necessary to ensure the stiffness of the floor and the stiffness of its support. When pile foundations are designed in the case of weak soil, the piles are recommended under the foundations of the supermarket and also under the floor layers. The example of foundation of a typical supermarket located on long­standing municipal dumping ground is presented. Indirect foundation - CFA piles, 0.5 m in diameter and 4.5 to 20.0 m long, was de­signed. Fifteen trial loading tests were carried out.

Jan 1, 2006

By Osamu Taki

This paper presents the system, the strength of column, the design concept, and the applications with four case histories from the past project as the general aspect of Geo-Column (trade mark of single soil-cement column by SCC Technology, Inc.). In the system, it shows the installation process and drilling/mixing tool with the mixing process. In the design concept, it shows how to design Geo-Column against the vertical and horizontal load. Finally, it shows the applications with four case histories, (1) building mat foundation, (2) isolated foundation on the pedestrian bridge, (3) bridge spread footing and slope erosion control wall in the channel, and (4) cut-off wall and shoring wall in the underpass roadway to the railroad.

Jan 1, 2000

By Dino Kartofilis, Rick Deschamps, Giovanni (John) Bonita

"Anew condominium complex in Arlington, Va., presented some formidable constraints. One was the requirement for a permanent retaining wall to support a 48.5-ft (15-m) cut and an existing 16- story building on top of the slope. Another was the adjacent property owners’ refusal to allow tieback anchors to encroach on their property. The General Contractor (GC) proposed an innovative design incorporating vertical concrete load-bearing elements (LBEs) supported by post-tensioned tiedowns. The owner accepted and the project was successfully completed in 2007.The original designer had proposed a system that included a temporary, raked soldier pile and lagging wall and a permanent cantilever retaining wall with a large piled footing. The system was potentially costly and its constructability questionable.The GC then asked for a design-build alternative to reduce costs and enhance the schedule. The first idea was a diaphragm wall with vertically anchored T-sections. However, configuring a crane to excavate perpendicular sections using grab equipment seemed impossible. There were also utilities approximately 25 ft (9 m) from the existing building, two tiered retaining walls down the slope and a large swimming pool at the base. However, the GC was committed to providing a stable bench 20 ft (6 m) wide along the wall alignment to support the crane.The design evolved to single bite panels using diaphragm wall technology, commonly called LBEs, or barrettes, when they carry vertical loads. For this project, 44 LBEs were horizontally spaced at 9 ft (2.75 m) on center, oriented perpendicular to the wall, extending into bedrock and anchored vertically on the upslope side to resist overturning moments. They were essentially large soldier piles that provided both temporary and permanent support of excavation.An unusual formal agreement was established with the GC to complete the work based on pricing, and prior to completion of final design. This arrangement allowed for an exchange of ideas that increased project efficiencies and lowered costs — without the fear that design concepts would be “shopped” to competitors."

Jan 1, 2009

By Song-yu Liu, Chao Yan, Yong-feng Deng

"In order to investigate the stability of embankment on soft subgrade improved by T-shaped deep mixed column composite foundation, the internal forces and failure modes of columns were analyzed on the shape of critical slip surface and the displacement fields of composite foundation through strength reduction method of three dimensional numerical simulation. Then comparison between T-shaped and conventional deep mixed column composite foundation were conducted. The simulation results indicate that the stability analysis was overestimated by limit equilibrium methods. Piles within the slope shoulder provide great resistance slide effect through the vertical bearing capacity, while the piles nearby slope shoulder and outside slope shoulder providing great moment to resistant slip. The embankment stability of T-shaped deep mixed column composite foundation is higher than that of conventional deep mixed column composite foundation while cost less cement and construction time.INTRODUCTIONThe soft soil has many unfavorable project properties such as high compressibility, low strength, and high water component. Embankment built on the soft soil will face subsidence, horizontal displacement and stability problems. Cement-soil deep mixed column composite foundation is more commonly used as a method for construction of embankment on soft ground [1-2]. For the existing problems of China's current cement-soil pile (wet jet mixing pile), including strength uneven, effective treatment with shallow depth and high cost, a new technique, T-shaped bidirectional deep mixed column, is introduced [3]. An automatic expanding mixing blade was developed by Liu et al. shown in Figure. 1, to install TDM columns [4].T-shaped deep mixed (TDM) column combines bi-directional deep mixing technique and T-shaped deep mixing technique: ?Using concentric biaxial drill pipe, opposite rotational direction of the blade on the inner drill pipe and the outer drill pipe. By squeezing action of the blade on the outer drill pipe, forward and reverse rotation of the blades bidirectional mixing at the same time, blocking the way to take the grout, the uniformity of cement mixing is improved. ?Scalable stirring blades, which can be stretched for two different radius at any depth below ground, expand cross-section of the upper pile (enlarged head), forming expanding cross-section within the upper part of a certain depth which is similar to the shape of nail. As a new composite foundation, T-shaped mixing pile composite foundation is widely applied. However, studies on the stability of T-shaped column treated soft subgrade are rare in the research. So the stability study on T-shaped column composite foundation under embankment for guiding the design and construction has important significance."

Jan 1, 2015