DFI JOURNAL · Vol. 4, No. 2 December 2010 DFI JOURNAL

Vol. 4, No. 2 December 2010

DFI JOURNALThe Journal of the Deep Foundations Institute

PAPERS:Evaluating Excavation Support Systems to Protect Adjacent Structures (The 2010 Michael W. O’Neill Lecture) – Richard J. Finno [3]

Innovative Construction Techniques used at North Shore Construction Project – Kanchan K. Sen, Aaron Evans [20]

Large Scale Lateral Testing of Pile Foundations (Young Professor Paper Competition 2010) – Anne Lemnitzer [31]

Inelastic Response of Extended Pile Shafts in Laterally Spreading Ground during Earthquakes (Student Paper Competition 2010) – Arash Khosravifar, Ross W. Boulanger [41]

Thermal Integrity Profi ling of Drilled Shafts – Gray Mullins [54]

TECHNICAL NOTELoad Testing and Interpretation of Instrumented Augered Cast-in-Place Piles – Timothy C. Siegel [65]

Deep Foundations Institute is the Industry Association of Individuals and Organizations Dedicated to Quality and Economy in the Design and Construction of Deep Foundations.

DFI JOURNAL Vol. 4 No. 2 December 2010 [1]

From the Editors and Publisher 2011 DFI Board of TrusteesPresident:James A. MorrisonKiewit Engineering Co.Omaha, NE USA

Vice President:Patrick BerminghamBermingham Foundation SolutionsHamilton, ON Canada

Secretary:John R. WolosickHayward Baker Inc.Alpharetta, GA USA

Treasurer:Robert B. BittnerBittner-Shen Consulting Engineers, Inc.Portland, OR USA

Immediate Past President:Rudolph P. FrizziLangan Engineering & Environmental ServicesElmwood Park, NJ USA

Other Trustees:David BorgerSkyline Steel LLCParsippany, NJ USA

Maurice BottiauFranki Foundations Group BelgiumSaintes, Belgium

Dan BrownDan Brown and Associates, PLLCSequatchie, TN USA

Gianfranco Di CiccoGDConsulting LLCLake Worth, FL USA

Bernard H. HertleinAECOM Technical Services Inc.Vernon Hills, IL USA

Matthew JanesIsherwood AssociatesBurnaby, BC Canada

James O. JohnsonCondon-Johnson & Associates, Inc.Oakland, CA USA

Douglas KellerRichard Goettle, Inc.Cincinnati, OH USA

Samuel J. KosaMonotube Pile CorporationCanton, OH USA

Kirk A. McIntoshMACTEC Engineering & Consulting, Inc.Jacksonville, FL USA

Raymond J. PolettoMueser Rutledge Consulting EngineersNew York, NY USA

Arturo L. Ressi di CerviaKiewit Infrastructure GroupWoodcliff Lake, NJ USA

Michael H. WysockeyThatcher Engineering Corp.Chicago, IL USA

Journal PublisherManuel A. Fine, B.A.Sc, P.Eng

Journal EditorsAli Porbaha, Ph.D., P.E. California State University Sacramento, CA, USADan A. Brown, Ph.D. Dan Brown and Associates, Sequatchie, TN, USAZia Zafir, Ph.D., P.E. Kleinfelder Sacramento, CA, USA

Associate EditorsLance A. Roberts, Ph.D., P.E.South Dakota School of Mines and TechnologyRapid City, SD USAThomas Weaver, Ph.D., P.E.Nuclear Regulatory CommissionRockville, MD USA

Published By Deep Foundations Institute

Copyright © 2010 Deep Foundations Institute.

AII rights reserved. Written permission must be

obtained from DFI to reprint journal contents, in

whole or in part.

Contact

DFI Headquarters

326 Lafayette Avenue

Hawthorne, NJ 07506

dfi hq@dfi .org

www.dfi .org

DFI, its directors and offi cers, and journal editors

assume no responsibility for the statements

expressed by the journal’s authors. International

Standard Serial Number (ISSN): 1937-5247

Mission/Scope The Journal of the Deep Foundations Institute publishes practice-oriented, high quality papers related to the broad area of “Deep Foundations Engineering”. Papers are welcome on topics of interest to the geo-professional community related to, all systems designed and constructed for the support of heavy structures and excavations, but not limited to, different piling systems, drilled shafts, ground modification geosystems, soil nailing and anchors. Authors are also encouraged to submit papers on new and emerging topics related to innovative construction technologies, marine foundations, innovative retaining systems, cutoff wall systems, and seismic retrofit. Case histories, state of the practice reviews, and innovative applications are particularly welcomed and encouraged.

DFI JOURNAL

The DFI Journal has reached the stage where many geotechnical practitioners and contractors consider this publication as their first choice among the Journals available in which to have their papers published. Another source of papers for the Journal has developed from authors who submit papers for publication in the Proceedings of the DFI Annual Meeting or other DFI Conferences and find that the review committee considers the paper to be of a nature that it is better published in the DFI Journal. DFI’s Young Professor and Student paper competitions have also come to be a source of papers for the Journal. We hope that next time you decide to prepare a deep foundations related paper for publication, the DFI Journal will be your journal of choice.

This issue of the DFI Journal has not been completed on schedule, once again due to extended peer review delays, involving several cycles of revision and resubmission for some of the papers before they are fully accepted. This peer review process contributes substantially to the quality of the papers published and a good product is worth the wait in our view. We look forward to the day when we have so many papers in various stages of completion that we can defer those that are lagging for publication in subsequent editions, rather than delay the current edition’s completion date.

In the last edition we announced that we are open to publishing valid discussions on papers published in past issues of the DFI Journal. No correspondence in the way of discussion has resulted. We again invite our readers, including reviewers with dissenting thoughts, to take advantage of this feature.

The editors are pleased that the Journal is now available for free on-line to DFI members, which will increase the widespread dissemination of published papers. We hope that authors will be even more encouraged to submit and maintain the tradition of high-quality practice-oriented geotechnical papers for publication. The journal will continue to focus on subjects that are directly relevant to deep foundations practice and thus will provide a bridge between academia doing applied research, practicing professionals, and the construction industry.

Comments, suggestions, and submissions are welcome and may be submitted to [email protected]

[2] DFI JOURNAL Vol. 4 No. 2 December 2010


Evaluating Excavation Support Systems to Protect Adjacent Structures (The 2010 Michael W. O’Neill Lecture)Richard J. Finno, Department of Civil and Environmental Engineering, Northwestern University,

Evanston, IL, USA; r-fi [email protected]

ABSTRACTThis paper presents an overview of methods that can be used to predict damage to buildings as a result of excavation-induced ground movements and describes an adaptive management approach for predicting, monitoring, and controlling excavation-induced ground movements. Successful updating of performance predictions depends equally on reasonable numerical simulations of performance, the type of monitoring data used as observations, and the optimization techniques used to minimize the difference between predictions and observed performance. This paper summarizes each of these factors and emphasizes their inter-dependence. Case studies are presented to illustrate the capabilities of this approach. Examples are given to show how optimized parameters based upon data obtained at early stages of excavation can be used to predict performance at latter stages, and how they can be applied to other excavations in similar geologic conditions. .

INTRODUCTIONDamage to buildings adjacent to excavations can be a major design consideration when constructing facilities in congested urban areas. As new buildings are constructed, the excavations required for basements affect nearby existing buildings, especially those founded on shallow foundations. Excavation support system design must prevent any damage to adjacent structures or balance the cost of a stiffer support system or underpinning with the cost of repairing damage to the affected structures. In any case, it is necessary to predict the ground movements that will induce damage to a structure. Practically speaking, a designer is attempting to limit/prevent damage to either the architectural details of a building, which occurs prior to structural damage, or to load bearing walls.

To evaluate damage potential in affected buildings, one must first predict the magnitude and distribution of ground movements caused by the excavation. This may be done using empirical or finite element methods, depending on the importance of the building, budget considerations, and design phase of the investigation. After locating the affected building in relation to the expected ground movements, one then evaluates the impact of these movements on the building. The main two sources of uncertainty in this analysis are the structural evaluation of a neighboring building and the movement prediction. This

paper summarizes damage evaluation methods and describes an adaptive management approach for predicting, monitoring and controlling ground movements. This approach can be thought of as an “automated” observational approach (Peck 1969). This methodology is a useful design tool in that decisions regarding trigger levels and responses can be evaluated thoroughly during design.

CRITERIA TO EVALUATE EXCAVATION-INDUCED DAMAGE Selected criteria that are applicable to evaluate excavation-induced damage are summarized in Table 1, wherein the relevant parameter and its limiting value are shown. Note that the parameter used to relate structural movements at the foundation level to damage is not the same for all methods.

The following terms are related to the limiting parameters in Table 1, and are illustrated in Fig, 1. Differential settlement between two points, i and j, is δ

ij. The distance between two points i

and j is ℓij. Distortion between two points, i and

j, is defined as δij/ℓ

ij. A concave-up deformation

is commonly called “sagging,” while a concave-down deformation is termed “hogging.” An inflection point separates two modes of deformation. The length of a particular mode of deformation is L, and can be bounded by either the ends of a building or inflection points of the settlement profile. The average slope, m, of a specific mode of deformation is


defined as δkl/L

kl, where the subscripts k and

l are boundaries of the mode of deformation. This slope differs from the distortion, δ

ij/ℓ

ij,

which is the slope between two adjacent points. The relative settlement of each mode, ∆, is the maximum deviation from the average slope of a particular deformation mode. The deflection ratio, ∆/L, is the ratio of the relative settlement to the length of the deflected part. Rigid body rotation of the building, ω, is the tilt of the building and causes no stresses or strains in the building. Angular distortion, β

ij, is the

difference between distortion, δij/ℓ

ij, and rigid

body rotation, ω.

[FIG. 1] Quantities used to defi ne limiting parameters for damage criteria

The critical tensile strain, εcrit

., is that strain at which cracking becomes evident. Tensile strains, ε

t, can be caused by bending, ε

b,

diagonal tension due to shear, εd , or horizontal

extension, εh, caused by lateral movement in the

soil mass below the footings. Critical strains that cause failure in common building materials vary widely as a function of material and mode of deformation (Boone 1996).

Burland and Wroth (1975) modeled a building as a deep isotropic beam to relate strains in the building to the imposed deformations, as illustrated in Fig. 2. They suggested that for the sagging type deformations shown in the figure, the neutral axis is located at the middle of the beam. For hogging type deformations, they assumed the foundation and soil provide significant restraint to deformations, effectively moving the neutral axis to its bottom. They presented equations for limiting ∆/L in terms of maximum bending strain and maximum diagonal tensile strain for a linear elastic beam subjected to a point load with the neutral axis at either the center or bottom of the beam. They assumed the beam had a Poisson’s ratio, ν, of 0.3, implying a Young’s modulus/shear modulus ratio, E/G, of 2.6. All buildings are not adequately represented by this E/G ratio. They postulated that for buildings with significant tensile restraint, or very flexible in shear (i.e. frame buildings), an E/G ratio of 12.5 would be appropriate. For buildings that have little or no tensile restraint (i.e. traditional masonry buildings), they recommended that the E/G ratio should be 0.5.

Boscardin and Cording (1989) extended the deep beam model to include horizontal extension strains, ε

h, caused by lateral ground

movements. They presented a chart relating β and ε

h to levels of damage in buildings with

brick, load-bearing walls and an L/H ratio of 1. They further assumed the building was subjected to hogging deformation with the neutral axis at the bottom. Similar to Burland

ReferenceType of method

Limiting parameter

Applicability

Burland and Wroth (1975)

Deep beam model of building

∆/(L εcrit

)Load bearing wall (E/G = 2.6),framed structures (E/G = 12.5), and masonry building (E/G = 0.5) with no lateral strain

Boscardin and Cording

(1989)

Extended deep beam

modelβ, ε

h

L/H = 1 and assumption horizontal ground and building strains are equal

Son and Cording (2005)

Semi-empirical

Average strain

Masonry structures; need relative soil/structure stiffness; use average strain in distorting part of structure

Finno et al (2005)

Laminate beam model

∆ /(L εcrit

)Load bearing walls, framed structures, masonry buildings, need bending and shear stiffness of components of walls and floors

Boone (1996)Detailed

analysis of structure

crack width

general procedure that considers bending and shear stiffness of building sections, distribution of ground movements, slip between foundation and grade and building configuration

[TABLE 1] Selected damage criteria for excavation-induced damage to buildings


and Wroth (1975), the building was idealized as a linear elastic beam with υ equal to 0.3. Direct transfer of horizontal ground strain to the structure is assumed in this approach, which may or may not be reasonable depending on the structure. For example, modern frame structures with floors that act as diaphragms do not move laterally with the ground (e.g. Geddes 1977, 1991; Finno et al. 2002) when subjected to typical excavation-induced deformations.

Son and Cording (2005) extended the Boscardin and Cording approach in a semi-empirical manner. Resulting criteria are applicable to masonry buildings. They proposed use of a damage criterion based on the average state of strain within the distorting portion of a building. Their revised criterion is independent of E/G, L/H and the position of the neutral axis of the wall. They explicitly considered the shear stiffness of

the walls on the distortions imposed by the ground settlements. They used results of model tests and numerical simulations as well as case studies of building damage to calibrate the model. They noted that cracking in masonry walls significantly reduced effective wall stiffness. There is considerable overlap in categories of damage as a function of their parameters.

Finno et al. (2005) extended the Burland and Wroth (1975) equations to allow explicit input of E/G and location of the neutral axis, resulting in equations that relate limiting ∆/L to bending strain at the top, ε

b(top), and bottom

of a beam, εb(bottom)

, and the maximum diagonal tensile strain, ε

d(average). Fig. 3 shows the effects

of different E/G ratios on the conditions required for initial cracking. The kink in a curve represents the limit between shear critical and bending critical geometries of a beam. These results show that the limiting deflection ratio that causes cracks varies over wide limits, implying that structural details of a building must be considered when establishing criteria. However, it is difficult to select the beam characteristic parameter E/G and the neutral axis location when developing a deep beam model for many structures, i.e., multi-story structures.

To provide a more realistic model of a structure and yet maintain relative simplicity, Finno et al (2005) proposed a laminate beam model to represent the response of a building to imposed deformations. Burland and Wroth (1975) modeled a building as a rectangular beam with unit thickness, which implicitly assumes a constant value of I/A

v for the building. In the laminate beam

approach, the parameter EI/GAv accounts for

variations in bending and shear stiffness of a structure. This reflects the fact that bending is inversely proportional to the bending stiffness, EI, where I is the moment of inertia of the beam, whereas deformation due to shear is inversely proportional to the shear modulus times the area contributing to shear resistance, GA

v. The laminate beam model

assumes that the floors offer restraint to bending deformations, and the walls, whether load bearing or infill between columns, offer restraint to shear deformations. These parameters can be explicitly considered for each floor and wall system in a multi-story building. See Finno et al (2005) for more details.

[FIG. 2] Deep beam idealization of building (after Burland and Wroth 1975)


Boone (1996) presented a more detailed approach to evaluate building damage due to differential ground movement caused by adjacent construction. This method considers structure geometry and design, strain superposition and critical strains of building materials. Load bearing walls are modeled as uniformly-loaded, simple-supported beams. Damage to frame buildings is assumed to occur from differential vertical movements of columns, depending on the column’s tilt and degree of fixity. Damage to infill walls is presumed to occur as a result of the deformed shape of the surrounding beams and columns. If a structure is subjected to horizontal extension, then these strains are superposed on the ones caused by bending and shear.

None of the models developed consider the strains that occur when the building settles when it is first constructed. Conceptually, one could estimate these “residual” strains and superpose them upon those arising during excavation. However, defining how much settlement would have occurred prior to attaching in-fill walls to a structural frame during and after the original building construction is a difficult task. The movements that impact these architectural portions of the structure are less than the total settlements. Furthermore, the author is unaware of any performance data that includes movements caused by both original construction and adjacent excavation. In any case, this aspect of response warrants further study.

The variability of the magnitude of movements that cause damage to the architectural details, as illustrated in Fig. 3, suggests that either a conservative approach or a detailed structural analysis of an affected building is warranted when establishing allowable movements for an excavation. If possible, the owners of the

affected buildings should be kept informed of the planned operations. Because under certain circumstances it becomes very expensive to construct a stiff enough system to prevent all damage, the optimal solution may be one where a repair cost for inevitable minor architectural damage is included in the bid package, after, of course, securing the cooperation of the building’s owner.

ADAPTIVE MANAGEMENT APPROACH Once limiting movement criteria have been established, an adaptive management approach can be employed to predict, monitor and control ground movements during excavation. This approach is summarized in Fig. 4. The left hand column represents calculations made during the design and updating phases, and includes finite element computations. The center column is the optimization needed to update predictions based on the measurements. Inclinometer, optical survey and strain gage data have been used as observations against which the predictions are compared. Ideally, this process works automatically, all data collected in the field is transferred in real time to a host computer where it can be processed into format compatible with the numerical analyses. After data are collected at early stages of an excavation, updated parameters form the basis of a new simulation to predict responses at later, and presumably more critical, stages of excavations.

Successful use of this approach depends equally on reasonable numerical simulations of performance, the type of monitoring data used as observations, and the optimization techniques used to minimize the difference between predictions and observed performance. This section summarizes each of these factors and emphasizes their inter-dependence.

Neutral Axis at Center of Beam (λ = 0.5)

0

0.5

1

1.5

2

2.5

0 2 4 6 8L/H

Δ/(L

εcr

it)

E/G = 12.5

E/G = 25

E/G = 2.6

E/G = 0.5

Shear Critical

Bending Critical

[FIG. 3] Effect of E/G on critical tensile strain (from Finno et al. 2005)

[FIG. 4] Adaptive management approach


Applications of these techniques from case studies are presented to illustrate the capabilities of this approach.

Numerical simulations

While supported excavations commonly are simulated numerically by modeling stages of excavation and support installation, it is necessary to simulate all aspects of the construction process that affect the stress conditions around the excavation to obtain an accurate prediction of behavior. This may involve simulating previous construction activities at the site, installation of the supporting wall and any deep foundation elements, as well as the removal of cross-lot supports or detensioning of tiedback ground anchors. Furthermore, issues of time effects caused by hydrodynamic effects or material responses may be important.

Finno and Tu (2006) summarized the effects of a number of key numerical assumptions on the computed performance of supported excavations. The manner in which the excavation is simulated including the removal of soil elements in a finite elements mesh should satisfy the principle of superposition as described by Ghaboussi and Pecknold (1985). Other key assumptions include selecting appropriate drainage conditions during excavation (Clough and Mana 1976; O’Rourke and O’Donnell 1997; Whittle et al. 1993), starting with appropriate initial effective stresses that include the effects of past construction activities at a site (Calvello and Finno 2003), and accurately defining the initial ground water conditions for a site (e.g. Finno et al. 1988). Many times the effects of the installation of shoring are ignored in a finite element simulation as the wall is “wished-into-place” with no change in the stress conditions in the ground or any attendant ground movements. However, there is abundant information (e.g. Clough et al. 1989; O’Rourke and Clough 1990; Finno et al. 1988; Sabatini 1991; Koutsoftas et al. 2000) that shows ground movements may arise during installation of the wall, and, if ignored, these may have a significant impact on the accuracy of the computed responses, particularly in cases where the resulting ground deformations are relatively small. One also must take care when representing the bracing system in a model. In typical plane strain simulations, application of prestress for

cross-lot braces and installation of tiedback ground anchors can present problems under certain circumstances (e.g. Finno and Tu 2006).

Even with properly defined initial conditions, challenges remain. Excavation and support installation normally occur under conditions that are three dimensional. If one is making a computation assuming plane strain conditions, then one must judiciously select a data set so that planar conditions are applicable to a set of inclinometer data. If one is using an adaptive management approach wherein data is collected and compared with numerical predictions in almost real time, then it is clear that a 3D analysis would be required as a result of the uneven excavated surface and timing of the anchor prestressing operations.

Furthermore, when a sufficiently extensive horizontal excavated surface is identified, 3-dimensional effects may still arise from the higher geometrically-induced stiffness at the corners of an excavation. Based on settlement observations in a number of case studies, these boundary conditions lead to smaller ground movements near the corners and larger ground movements towards the middle of the excavation wall (e.g., Roboski and Finno 2007). Another, and less recognized, consequence of the corner stiffening effects is the maximum movement near the center of an excavation wall may not correspond to that found from a conventional plane strain simulation of the excavation, i.e., 3-dimensional (3-D) and plane strain simulations of the excavation do not yield the same movement at the center portion of the excavation, even if the movements at the center are perpendicular to the wall. This effect can be only quantified on the basis of numerical analysis. The plane strain ratio, PSR, is defined as the maximum movement at the center of an excavation wall computed by 3-D analyses divided by that computed by a plane strain simulation. As shown in Fig. 5, a key indicator is the L/H

e ratio, where L is

the dimension of the excavation where the movement occurs, and H

e is the excavation

depth. When L/He is greater than 6, the

PSR is equal to 1 and results of plane strain simulations yield the same displacements in the center of an excavation as those computed by a 3-D simulation. When L/H

e is less than 6,

the displacement computed from the results of a plane strain analysis will be larger than that from a 3-D analysis.


[FIG. 5] Effects of geometry on 3-D movements of excavations

When conducting an inverse analysis of an excavation with a plane strain simulation, the effects of this corner stiffening is that an optimized stiffness parameter will be larger than it really is because of the lack of the corner stiffening in the plane strain analysis. This effect becomes greater as an excavation is deepened because the L/H

e value decreases as the excavated

grade is lowered. This trend was observed in the optimized parameters for the deeper strata at the excavation for the Chicago-State subway renovation (Finno and Calvello 2005).

Soil Constitutive BehaviorWhen one undertakes a numerical simulation of a deep supported excavation, one of the key decisions made early in the process is the selection of the material constitutive models representing the various soil formations at the site. If the results form the basis of a prediction that will be updated based on field performance data, then the types of field data that form the basis of the comparison will impact the applicability of a particular model. Possibilities include lateral movements based on inclinometers, vertical movements at various depths and distances from an excavation wall and/or forces in structural support elements. When used for a case where control of ground movements is a key design consideration, the constitutive model must be able to reproduce the soil response at appropriate strain levels to the imposed loadings.

It is useful to recognize that soil is an incrementally nonlinear material, i.e., its stiffness depends on loading direction and strain level. Soils are neither linear elastic nor elasto-plastic, but exhibit complex behavior characterized by zones of high constant stiffness at very small strains, followed by decreasing stiffness with increasing strain. This behavior under static loading initially was

realized through back-analysis of foundation and excavation movements in the United Kingdom (Burland, 1989). The recognition of zones of high initial stiffness under typical field conditions was followed by efforts to measure this ubiquitous behavior in the laboratory for various types of soil (Jardine et al, 1984; Clayton and Heymann 2001; Santagata et al. 2005; Callisto and Calebresi 1998, Holman 2005, Cho and Finno 2010). Furthermore, the stiffness depends on the direction of loading as measured from the most recently applied stress path, or recent stress history.

To illustrate small strain nonlinearity and recent stress history effects on shear stiffness for Chicago clays, secant shear modulus from drained constant mean normal stress, CMS, and constant mean normal stress extension, CMSE, stress paths are plotted versus shear strains in Fig. 6. These specimens with an OCR of 1.7 were obtained from block samples cut from an excavation in Evanston, IL (Blackburn and Finno 2007). Of the four test results shown on the figure, shearing commenced in two of the tests after consolidating to the in situ vertical effective stress (noted as K

0 tests in the label).

In the other two tests, shearing began after first loading to the in situ vertical effective stress and then unloading via a reduced triaxial extension, RTE, path (noted as U tests). In all cases, the secant shear modulus at 0.1% strain, the smallest strain reliably measured in conventional triaxial equipment, was about 4 to 8 times less than that measured at 0.002% strain, the smallest value obtained with the internal instrumentation used in these experiments. Complete details and results of the testing program are presented by Cho (2007).

Recent stress history effects are shown in Fig. 6. The angles noted next to the stress paths are calculated as the absolute value of the angle change from the previous stress path (θ = 0°). The results of the two “K

0” probe

tests showed dependence on the angle change, with the CMSE path (unloading) exhibiting a stiffer response than that of the CME (loading type). For the “U” probe tests, with a recent stress history representative of a site where an old building with a basement was demolished before excavation, the opposite directional dependency is observed. The stiffness of loading path (U-CMS) is much greater than those of unloading path (U-CMSE). Interestingly, shear moduli magnitudes in the loading path (K

0-CMS) of the “K

0” probes and


the unloading path (U-CMSE) of the “post-unloading” probes with similar values of θ are quite alike, even though the current stress path direction is exactly the opposite. Considering the change in θ, as shown in the inset of Fig, 6, the stiffer shear moduli occur at the stress path corresponding to nearly complete stress reversals, U-CMS (θ =160°) and K

0-CMSE (θ

=147°). Although the data are limited, they show the effects of recent stress history on the shear stiffness. Also, little difference was noted at strains larger than 0.1%, as reported by Atkinson et al (1990). Thus it appears that recent stress history effects are significant for these clays – G

sec at about 0.002% strain varies

by a factor of 2.

Burland (1989) suggested that working strain levels in soil around well-designed tunnels and foundations are on the order of 0.1 %. If one uses data collected with conventional triaxial equipment to discern the soil responses in many practical situations, it is not possible to accurately incorporate site-specific small strain non-linearity into a constitutive model based on conventionally-derived laboratory data. There are a number of models reported in literature wherein the variation of small strain nonlinearity can be represented, for example, a three-surface kinematic model developed for stiff London clay (Stallebrass and Taylor 1997), MIT-E3 (Whittle and Kavvadas 1994), and hypoplasticity models (e.g. Viggiani and Tamagnini 1999). These models require either detailed experimental results or experience with the model in a given geology to derive parameters. More work is needed to relate these actual soil responses to conventionally-obtained field and laboratory data to incorporate these responses into practice.

0.001 0.01 0.1 1Shear Strain, εsh (%)

0

10

20

30

40

50

60

70

80

90

Seca

nt S

hear

Mod

ulus

, Gse

c (M

Pa)

K0-CMS (θ=35o)K0-CMSE (θ=145o)U-CMS (θ=160o)U-CMSE (θ=20o)

K0 pathθ

RTE

CMS

CMSEθ

q

p'

[FIG. 6] Recent stress history effects on secant shear modulus: Chicago glacial clay (Cho and Finno 2010)

For most current practical applications, one uses simpler, elasto-plastic models contained in material libraries in commercial codes. For these models, a key decision is the selection of the elastic parameters representative of the secant values that correspond to the predominant strain levels in the soil mass. An example of the strain levels behind a wall for an excavation with a maximum lateral wall movement of 29 mm (1.14 in) is shown in Fig. 7. These strain levels were computed based on the results of displacement-controlled simulations where the lateral wall movements and surface settlements were incrementally applied to the boundaries of a finite element mesh. The patterns of movements were typical of excavations through clays, and were based on those observed at an excavation made through Chicago clays (Finno and Blackburn 2005). Because the simulations were displacement-controlled, the computed strains do not depend on the assumed constitutive behavior.

As can be seen in Fig. 7, the maximum shear strains correspond to about 0.35% for 29 mm (1.14 in) maximum wall lateral movement, and represent good control of ground movements in these soft soils. Although not shown, shear strains as high as 0.7% occurred when 57 mm (2.24 in) of maximum wall movement developed. These strain levels can be accurately measured in conventional triaxial testing, and thus if one can obtain specimens of sufficiently high quality, then secant moduli corresponding to these strain levels can be determined via conventional laboratory testing. Because the maximum horizontal wall displacement can be thought of as a summation of the horizontal strains behind a wall, the maximum wall movements can be accurately calculated with a selection of parameters that correspond to these expected strain levels. The fact that small strain non-linearity is not explicitly considered will not have a large impact on the computed horizontal wall displacements because they are dominated by the larger strains in the soil mass. Consequently, these computed movements would be compatible with those measured by an inclinometer located close to the wall.

However, if one needs to have an accurate representation of the distribution of ground movements with distance from the wall, then this approach of selecting strain-level appropriate elastic parameters will not work. The small strain non-linearity must be explicitly considered to find the extent of the


settlement because the strains in the area of interest vary from the maximum value to zero. As a consequence, many cases reported in literature indicate computed wall movements agree reasonably well with observed values, but the results from the same computations do not accurately reflect the distribution of settlements. Good agreement at distances away from a wall can be obtained only if the small strain non-linearity and dilation responses, if appropriate, of the soil are adequately represented in the constitutive model.

[FIG. 7] Shear strain levels behind excavation (contours in %)

Self-updating Models

Self-updating models can be of two types, one wherein the constitutive responses are assumed and key parameters of the model are updated using inverse techniques based on selected field observations, and the other wherein the field observations are used to define the constitutive response using artificial neural nets (Hashash et al. 2006). Herein, an inverse technique based on a gradient method (e.g., Ou and Tang 1994; Ledesma et al., 1996; Calvello and Finno 2004) is applied. The method employs local parameter identification of a specific constitutive law. The gradient method described herein uses UCODE (Poeter and Hill, 1998), a computer code designed to allow inverse modeling

posed as a parameter estimation problem. Macros were written in a Windows environment to couple UCODE with PLAXIS, a commercial finite element code. Alternatively, the approach can be implemented by using the optimization routines in the MATLAB toolbox.

Fig. 8 shows a flowchart of a parameter optimization algorithm appropriate for a gradient method. With the results of a finite element prediction in hand, the computed results are compared with field observations in terms of weighted least-squares objective function, S(b):

( ) ( ) ( )' 'T TS b y y b y y b e eω ω⎡ ⎤ ⎡ ⎤= − − =⎣ ⎦ ⎣ ⎦ (1)

where b is a vector containing values of the parameters to be estimated; y is the vector of the observations being matched by the regression; y'(b) is the vector of the computed values which correspond to observations; ω is the weight matrix wherein the weight of every observation is taken as the inverse of its error variance; and e is the vector of residuals. This function represents a quantitative measure of the accuracy of the predictions.

A sensitivity matrix, X, is then computed using a forward difference approximation based on the changes in the computed solution due to slight perturbations of the estimated parameter values. This step requires multiple runs of the finite element code, one for each parameter that is to be estimated. The values of the parameters that result in a best fit between the computed and observed values are found using a modified Gauss-Newton method:

[FIG. 8] Flow chart for gradient method


(2)

1r r r rb d bρ+ = + (3)

where dr is the vector used to update the

parameter estimates b; r is the parameter estimation iteration number; X

r is the sensitivity

matrix (Xij=∂y

i/∂b

j) evaluated at parameter

estimate br; C is a diagonal scaling matrix

with elements cjj equal to 1/√(XTω X)

jj; I is the

identity matrix; mr is the Marquardt parameter

used to improve regression performance; and d

r is a damping parameter, computed as the

change in consecutive estimates of a parameter normalized by its initial value, but is restricted to values less than 0.5.

The updated model is considered optimized if either of two convergence criteria is met: (i) the maximum parameter change of a given iteration is less than a user-defined percentage of the value of the parameter at the previous iteration; (ii) the objective function, S(b), changes less than a user-defined amount for three consecutive iterations. After the model is optimized, the final set of input parameters is used to run the finite element model one last time and produce the “updated” prediction of future performance. See Rechea (2006) for details concerning the convergence criteria as applied to excavations.

Inverse analysis algorithms allow the simultaneous calibration of multiple input parameters. However, identifying the important parameters to include in the inverse analysis can be problematic, and it is not possible to use a regression analysis to estimate every input parameter of a given excavation simulation. The relative importance of the input parameters being simultaneously estimated can be defined using the parameter statistics, composite scaled sensitivity, ccs

j, and correlation coefficient,

cor(i,j) (Hill 1998). The value of cssj indicates

the total amount of information provided by the observations for the estimation of parameter j, and is defined as:

1/ 22

1/ 2

1

'NDi

j j iij j

b

ycss b ND

bω

=

⎡ ⎤⎛ ⎞⎛ ⎞∂⎢ ⎥= ⎜ ⎟⎜ ⎟⎢ ⎥⎜ ⎟⎜ ⎟∂⎝ ⎠⎝ ⎠⎢ ⎥⎣ ⎦∑

(4)

where y′i is the ith computed value, b

j is

the jth estimated parameter, ∂yi/∂b

j is the

sensitivity of the ith computed value with respect to the jth parameter, ω

jj is the weight

of the ith observation, and ND is the number of observations.

The values of the matrix cor (i,j) indicate the correlation between the ith and jth parameters, and are defined as:

1/ 2 1/ 2

cov( , )( , )

var( ) var( )

i jcor i j

i j=

(5)

where cov(i,j) equal the off-diagonal elements of the variance-covariance matrix V(b′)=s2(XTωX)-1, and var(i) and var(j) refer to the diagonal elements of V(b′). Parameters with correlation coefficients equal to 1 or -1 are perfectly correlated and cannot be optimized at the same time.

The number and type of input parameters that one can expect to estimate simultaneously depend on a number of factors, including the soil models used, the stress conditions of the simulated system, available observations, and numerical implementation issues. Examples of this procedure to select parameters for optimization are presented by Calvello and Finno (2004) and Finno and Calvello (2005).

Monitoring

The assumptions inherent in any prediction limit the types of data that can be used as a basis of updating performance predictions. Consequently, one must carefully select the types of data, location of the measuring points, and the excavation conditions when applying an inverse technique. Inclinometer data based on measurements close to a support wall are the most useful when typical elasto-plastic constitutive models are assumed to represent soil behavior, as is the case when employing commercial finite element codes, for reasons discussed in the last section. These data can be supplemented by ground surface settlements when using a constitutive model that accounts for small strain nonlinearities and dilation (Hashash and Whittle, 1996; Finno and Tu 2006). Furthermore, other types of measurements, such as forces in internal braces and pore water pressures, conceptually can be used in conjunction with displacement measurements to make the computed results more sensitive to parameters selected for optimization (Rechea 2006). However, if the bracing forces are used in the analyses, then either they must be corrected for the effects of temperature or the numerical simulation must explicitly include the temperature induced


The finite element software PLAXIS was used to compute the plane strain response of the soil around these excavations. The hardening-soil model (H-S) (Schanz et al. 1999) was assumed to represent soil responses for these examples. Parameters from other constitutive models have been optimized as well (e.g., Calvello and Finno 2002).

The effective stress H-S model is formulated within the framework of elasto-plasticity. Plastic strains are calculated assuming multi-surface yield criteria. Isotropic hardening is assumed for both shear and volumetric strains. The flow rule is non-associative for frictional shear hardening and associative for the volumetric cap. Six basic H-S input parameters define the constitutive soil responses, the friction angle, φ, cohesion, c, dilation angle, ψ, the reference secant Young’s modulus at the 50% stress level, E

50ref, the reference oedometer

tangent modulus, Eoed

ref, and the exponent m which relates reference moduli to the stress level dependent moduli (E representing E

50, E

oed,

and Eur

):

m

refref

pc

cEE ⎟⎟⎠

⎞⎜⎜⎝

⎛+−=

φσφ

cot

cot '3 (6)

where pref is a reference pressure equal to 100 stress units and σ’

3 is the minor

principal effective stress. A sensitivity analysis indicated that the model’s relevant and uncorrelated parameters for the Chicago excavations presented herein are E

50ref and φ’

(Calvello and Finno 2004). Results were also sensitive to changes in values of parameter m. However, parameter m was not included in the regression because the values of the correlation coefficients between parameters m and E

50ref were very close to 1.0, indicating

that the two parameters were not likely to be simultaneously and uniquely optimized. When values of φ’ were kept constant at their initial estimates, and only the stiffness parameters, E

50ref, were optimized, the calibrations of the

simulations presented subsequently were successful. Finno and Calvello (2005) showed that shear stress levels in the soil around the excavation were much less than those corresponding to failure for the great majority of the soil. This indeed is expected for excavation support systems that are designed to restrict adjacent ground movements to acceptably small levels, and hence one would expect the stiffness parameters to have a

changes in the support system. This latter feature is not normally included in commercial finite element codes.

While these different types of data can be handled within a properly formulated inverse analysis, the timely collection and screening of the data must be successfully accomplished (Finno 2007). Furthermore, for any monitoring system to be fully automated, one must be able to track construction progress so that performance data can be correlated with the excavation activities. To correlate the numerical data with the causative actions of the excavation process, imaging technologies can be employed to provide an accurate and detailed record of construction activities. Trupp et al. (2004) and Su et al. (2006) used 3-D laser scanning to capture an accurate image of the geometry of the excavation to provide an accurate, as-built digital record of construction. Sections may be taken from these scans and imported into a finite element code to provide an accurate excavation surface for input to inverse analysis. An internet accessible weather-resistant video camera has been used on several projects to allow remote visualization of the construction process in real-time, as well as a dated, photographic record of construction (Finno and Blackburn 2005). Significant developments have been made in automated systems to continuously monitor deformations due to construction activities. These systems provide the engineer with uninterrupted data in near real time without the need to wait for manual data readings. Such systems are essential tools for making timely decisions regarding changes in construction activities and support installation to mitigate potential damage to adjacent facilities. However, real-time, completely automated updating is not yet possible, although updated parameters can be obtained within 8 hours after field data have been acquired.

CAPABILITIES OF THE ADAPTIVE MANAGEMENT METHODExamples of the gradient method applied to supported excavations are presented to illustrate (i) its ability to identify optimized parameters based on observations made during early stages of excavation so as to allow accurate predictions of performance of latter stages of an excavation, and, (ii) the applicability of optimized parameters found based on performance data of one excavation to others in the same geology.


greater effect on the simulated results than failure parameters. Furthermore, use of this model restricts one to the use of inclinometer data obtained close to a support wall because the model does not include the capability for handling small strain non-linearity.

Parameter Optimization at Early Stages of Excavation

The ability of the approach to provide optimized parameters at an early stage of excavation which leads to good predictions of subsequent performance is illustrated by the Chicago Ave. and State St. subway renovation project in Chicago (Finno et al. 2002). This project involved the excavation of 12.2 m (40 ft) of soft to medium clay within 2 m (6.6 ft) of a school supported on shallow foundations. Fig. 9 shows a section of the excavation support system and the subsurface conditions. The support system consisted of a secant pile wall with three levels of support, which included pipe struts (1st level) and tieback anchors (2nd and 3rd levels). The subsurface conditions consisted of an urban fill, mostly medium dense sand but also containing construction debris, overlying four strata associated with the advance and retreat of the Wisconsin-aged glacier. The upper three are ice margin deposits deposited underwater, and are distinguished by water content and undrained shear strength (Chung and Finno, 1992). With the exception of a clay crust in the upper layer, these deposits are lightly overconsolidated as a result of lowered

groundwater levels after deposition and/or aging. Stratigraphy is shown in terms of Chicago City Datum (CCD) elevation.

A complete record of performance of the excavation can be found in Finno et al. (2002). Fig. 10 summarizes deformation responses to excavation and support. Both lateral movements and settlements are shown, although optimization was based solely on the former. The movements that occurred as the secant pile wall extend through all compressible layers. This is important when using these observations to calibrate parameters using inverse techniques in that these movements occur at an early stage of the excavation. These observations were sufficient to optimize parameters in all layers so that movements could be “predicted” at subsequent stages of excavation.

Very little movement beyond that which occurred during wall installation were observed until the excavation was lowered below EL. -1.4 m (-4.6 ft) CCD; a maximum of 4 mm (0.17 in) additional lateral movement occurred as a result of excavating to this elevation. The secant pile wall incrementally moved toward the excavation in response to excavation-induced stress relief. When the excavation reached final grade, the maximum lateral movement was 28 mm (1.10 in). The school settled as the secant pile wall moved laterally. The maximum settlement at the school at the end of excavation also was 28 mm (1.10 in) when the excavation bottomed out.

San

dFi

llSt

iffC

lay

Sof

tC

lay

Med

ium

Cla

yM

edC

lay

Stif

f Cla

yH

ard

Cla

y

EL +4.27

EL 0.00EL -1.22

EL -4.57

EL -7.01

EL -10.67

EL -15.24

EL -18.29

(m CCD)W24x131

Waler (TYP)

610 mm Dia x 15 mm Pipe Strut

T/PILE and GradeEL +4.27

2.10 m

Frances XavierWarde School

EL -0.30

Note:All elevations given in m CCD

EL -4.27

EL -0.91

EL -14.020.91 m

Existing Tunnel

T/PILEEL +4.27

Grade EL +4.22

B/SECANTPILES

W24x55in AlternatingShafts

Tieback A

nchor

Tieback A

nchor

Tieback AnchorTieback Anchor

1 (TYP)1 (TYP)

EL -4.57

[FIG. 9] Support system for Chicago-State excavation (Finno et al. 2002)


Table 2 shows the calculation phases and the construction stages used in the finite element simulations. Note that construction of the tunnel tubes and the school adjacent to the excavation was simulated explicitly in the first 12 phases of the simulation to take into account the effects of these activities on the soil surrounding the excavation. Stages 1, 2, 3, 4 and 5 in the optimization process refer to the construction stages for which the computed results were compared to inclinometer data taken from two inclinometers on opposite sides of the excavation. Construction steps not noted as “consolidation” on Table 2 were modeled as undrained. Consolidation stages were included after the tunnel, school and wall installation calculation phases to permit excess pore water pressures to equilibrate. Details about the definition of the finite element problem, the calculation phases and the model parameters used in the simulation can be found in Calvello (2002).

[TABLE 2] FE simulation of construction

Phase Construction stepSimulation

stage

0 Initial conditions

1-45

Tunnel construction (1940)Consolidation

6-1011-12

School construction (1960)Consolidation

13 Drill secant pile wall (1999)

14 Place concrete in wall Stage 1

15 Consolidation (20 days)

16 Excavate and install struts Stage 2

17Excavate below first tieback level

18 Prestress first level of tiebacks Stage 3

19Excavate below second tieback level

20Prestress second level of tiebacks

Stage 4

21 Excavate to final grade Stage 5

Visual examination of the horizontal displacement distributions at the inclinometer locations provides the simplest way to evaluate the fit between computed and measured field response. When computations were made based on parameters derived from results of drained triaxial tests, the finite element model computed significantly larger displacements at every construction stage (Finno and Calvello 2005). The maximum computed horizontal displacements were about two times the measured ones and the computed displacement profiles result in significant and unrealistic movements in the lower clay layers. As one would expect, these results indicated that the stiffness properties for the clay layers based on conventionally-derived triaxial data from thin-walled specimens were less than field values.

Fig. 11 shows the comparison between the measured field data from both sides of the excavation and the computed horizontal displacements when parameters are optimized based on stage 1 observations. The computed and measured responses are essentially the same after stage 1, as expected. Despite the fact that the optimized set of parameters is calculated using only stage 1 observations, the magnitude and distribution of the lateral movements with depth on both sides of the excavation are captured quite well for all stages of construction. At the end of the construction (i.e. stage 5) the maximum

2 4 6 8 10 12 14 16 18 20Distance from Centerline of Wall (m)

30

25

20

15

10

5

0

Settl

emen

t (m

m)

SettlementDay 15Day 49Day 86Day 103Day 117

30 25 20 15 10 5 0

LateralDeformation (mm)

-20

-15

-10

-5

0

5

Elevation(m CCD)

Lateral DeformationsDay 15 - 1st Lat. and vert. measurement after wall installedDay 51 - Excavate below strut level (EL +2.7 m CCD)Day 81 - Excavate below 1st tieback level (EL -1.4 m CCD)Day 87 - 1st tieback level stressedDay 102 - Excavate below 2nd tieback level (EL -5.2 m CCD)Day 116 - Excavate to final grade (EL -7.9 m CCD)

Secant Pile Wall

.

Frances Xavier Warde School

(Day 105)

(Day 60 to Day 74)(Day 28)

(Day 87)

(Day 81)

(Day 116)

CL

San

d

and

Fill

Stif

fC

lay

Sof

tC

lay

Med

Cla

yM

ediu

mC

lay

Stif

f C

lay

Har

dC

lay

[FIG. 10] Lateral movements and settlements at Chicago-State excavation (Finno et al. 2002)


computed displacement exceeds the measured data by only about 15%. These results are significant in that a successful recalibration of the model at an early construction stage positively affects subsequent “predictions” of the soil behavior throughout construction.

Applicability of Optimized Parameters in Similar Geology

To show the applicability of the optimized parameters that formed the basis of the good agreement in Fig. 11 to other excavation sites in these soil deposits, the results of numerical simulations are presented in Figs. 12 and 13 based on these optimized parameters for the conditions at the Lurie (Finno and Roboski 2005) and the Ford Design Center (Blackburn and Finno 2007) excavations, respectively. The geologic origin of the most compressible material is similar for all three cases, but the sites are located as much as 15 km (9.3 mi) apart. Consequently one should expect some variability in the actual parameters at each site.

Examining the comparisons in the clay layers below EL. -5 m (-16.4 ft) CCD for the Lurie data on Fig. 12, reasonable agreement is observed at stages 5 and 6, with significant differences seen at stage 4. This is likely caused by the

fact that the H-S model used herein does not include provisions to represent the large stiffness degradation with small strains. As discussed previously, one must select moduli that represent the average strains within the soil mass, and when the movements are small, the average stiffness should be higher in a model that does not consider the small strain stiffness degradation. As noted, the agreement between computed and observed responses was good for the latter stages of excavation where the lateral movements were larger.

At the Ford Center, the numerical results shown in Fig. 13 followed similar trends as the observed data, but with larger magnitudes. The parameters used in the analysis again were based on the larger deformations that were present at the Chicago-State site, and hence resulted in larger deformations than were observed at the Ford Center. In any case, the application of the Chicago-State based optimized parameters to both the Lurie and Ford sites resulted in reasonable agreement with the observed lateral movements, within the limitations of the analyses. Application of the inverse techniques to these data resulted in improved fit with minor changes to the parameters (Rechea 2006).

Stage 4 Stage 5 Stage 1 Stage 2 Stage 3

-60

-50

-40

-30

-20

-10

0

10

0 0.5 1 1.5 2displacement (in)

Elev

atio

n (ft

CC

D)

-60

-50

-40

-30

-20

-10

0

10


-60

-50

-40

-30

-20

-10

0

10


-60

-50

-40

-30

-20

-10

0

10


-60

-50

-40

-30

-20

-10

0

10

0 0.5 1 1.5 2diplacement (in)

-60

-50

-40

-30

-20

-10

0

10

0 0.5 1 1.5 2

-60

-50

-40

-30

-20

-10

0

10

0 0.5 1 1.5 2

-60

-50

-40

-30

-20

-10

0

10

0 0.5 1 1.5 2Measured field data

Computed displacements

Observations used forregression analysis

1

2

3

45

1

2

3

4

5

EASTSIDE

WESTSIDE

[FIG. 11] Comparison of observed and computed horizontal displacements (after Finno and Calvello 2005)


CONCLUDING REMARKSThis paper presents an overview of methods that can be used to predict damage to buildings as a result of excavation-induced ground movements and describes an adaptive management approach for predicting, monitoring, and controlling excavation-induced ground movements.

Review of relatively simple methods to estimate the onset of damage to structures from excavation-induced ground movements indicates the variability of the magnitude of movements that cause damage. Consequently,

either a conservative approach or a detailed structural analysis of an affected building is warranted when establishing allowable movements for an excavation. Because at times it becomes very expensive to construct a stiff enough system to prevent all damage, the optimal solution may be one where a repair cost for inevitable minor architectural damage is included in the bid package.

Successful applications of this approach depend equally on reasonable numerical

simulations of performance, the type of monitoring data used as observations, and the inverse analysis techniques used to minimize the difference between predictions and observed performance. The key to the successful calibration of an excavation lies in defining a “well

posed” inverse analysis problem to calibrate the simulation. The parameters optimized by inverse analysis are few compared to the total number of parameters defining the behavior of the simulation. Indeed, the majority of the input parameters are estimated by conventional means and never “re-calibrated.” Yet, the optimization can be effective if a finite element simulation of the excavation adequately reproduces the stress history of the soil on site and the soil model adequately represented the behavior of the clays, at least in terms of appropriate field observations.

[FIG. 12] Computed and observed lateral movements: Lurie excavation with optimized parameters from Chicago-State excavation

E l e v a t i o n ( m C C D )

F i l l ( S M ) N 3 - 7

S a n d ( S P ) N 1 5 - 2 6

S o f t t o M e d i u m

C l a y S u 2 9 - 4 3 k P a

S t i f f C l a y S u 1 0 5 k P a

H a r d C l a y S u 3 8 3 k P a

F i n a l E x

-20

-15

-10

-5

0

5

c a v a t i o n G r a d e

( T y p i c a l )

2 0 o

1 0 o

3 0 o

-25

-20

-15

-10

-5

0

500.020.040.060.08

Movement (m)

Elev

atio

n (m

CC

D)

Stage 4Stage 5Stage 6

PLAXIS

Field Data

5.2 m

0 m-0.9 m

-4.9 m

-13.1 m

-16.8 m

-20.7 m

0 mDepth ECD

5.2 m6.1 m

10 m

18.3 m

21.9 m

25.9 m

3.7 m (ECD)

-3.8 m (ECD)

-19 m (ECD)

1.5 m

9 m

24.2 m

-5 m (ECD)

1.5 m3.7 m

-1 m

1.5 m

-9.5-11 m (ECD)

Fill/Sand/Silt

Blodgett StratumSoft Clay – Su~36 kPa

Deerfield StratumMedium Clay

Su ~ 38-58 kPa

Park Ridge StratumStiff, Silty Clay

with Gravel

HardpanHard Clay, Sand, Gravel

Clay Crust

Excavated Fill/Sand/Silt 5.2 m

0 m-0.9 m

-4.9 m

-13.1 m

-16.8 m

-20.7 m

0 mDepth ECD

5.2 m6.1 m

10 m

18.3 m

21.9 m

25.9 m

3.7 m (ECD)

-3.8 m (ECD)

-19 m (ECD)

1.5 m

9 m

24.2 m

-5 m (ECD)

1.5 m3.7 m

-1 m

1.5 m

-9.5-11 m (ECD)

Fill/Sand/Silt



Su ~ 38-58 kPa


with Gravel


Clay Crust

Excavated Fill/Sand/Silt

Fil l/Sand/Silt



Su ~ 38-58 kPa


with Gravel


Clay Crust

Excavated Fill/Sand/Silt

-25

-20

-15

-10

-5

0

5-0.025 -0.015 -0.005

Movement (m)

Elev

atio

n (m

EC

D)

IntermediateFinal

PLAXIS

Field Data

[FIG. 13] Computed and observed lateral movements at Ford excavation based on optimized parameters from Chicago-State excavation


The convergence of an inverse analysis to an “optimal solution” (i.e. best-fit between computed results and observations) does not necessarily mean that the simulation is satisfactorily calibrated. A geotechnical evaluation of the optimized parameters is always necessary to verify the reliability of the solution. For a model to be considered “reliably” calibrated both the fit between computed and observed results must be satisfactory (i.e. errors are within desired and/or accepted accuracy) and the best-fit values of the model parameters must be reasonable.

ACKNOWLEDGEMENTSThis work would not have been possible without the many contributions of former graduate students and post-doctoral scholars at Northwestern University who worked on developing the inverse analysis methods, collecting detailed field performance data, and conducting careful laboratory experiments, including Michele Calvello, Cecilia Rechea, Sebastian Bryson, Jill Roboski, Tanner Blackburn, Terry Holman, Wan Jei Cho, Greg Andrianis, Miltos Langousis, Young-Hoon Jung, Taesik Kim and Fernando Sarabia. Financial support for this work was provided by National Science Foundation grant CMS-0219123 and the Infrastructure Technology Institute (ITI) of Northwestern University. The support of Dr. Richard Fragaszy, program director at NSF, is greatly appreciated.

REFERENCESAtkinson, J. H., Richardson, D., and 1. Stallebrass, S. E. (1990). Effect of recent stress history on the stiffness of overconsolidated soil. Geotechnique, 40(4), 531-540.

Blackburn, J.T., Sylvester, K. and Finno, R.J. 2. (2005). Observed bracing responses at the Ford Design Center excavation, Proceedings, 16th International Conference on Soil Mechanics and Geotechnical Engineering,” Japan, Vol. 3, 1443-1446.

Blackburn, J.T. and Finno, R.J., (2007). 3. Three-Dimensional Responses Observed in an Internally Braced Excavation in Soft Clay, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 133 (11), 1364-1373.

Boone, S. J. (1996). Ground-movement-4. related building damage, Journal of

Geotechnical and Geoenvironmental Engineering, ASCE, 122(11), 886-896.

Boscardin, M.D. and Cording, E.J. (1989). 5. Building response to excavation-induced settlement, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 115(1), 1-21.

Burland, J. B., and Wroth, C. P. (1975). 6. Settlement of buildings and associated damage, Proceedings, Conference on Settlement of Structures, Cambridge, 611-654.

Burland, J.B. (1989). ‘Small is beautiful’ – 7. the stiffness of soils at small strains: Ninth Laurits Bjerrum Memorial Lecture. Canadian Geotechnical Journal, Vol. 26, 499-516.

Calisto, L. and Calebresi, G. (1998). 8. Mechanical behavior of a natural soft clay. Geotechnique, Vol. 48 (4), 495-513.

Calvello, M. (2002). Inverse Analysis of a 9. Supported Excavation through Chicago Glacial Clays, PhD thesis, Northwestern University, Evanston, IL.

Calvello M. and Finno R.J. (2002). Calibration 10. of soil models by inverse analysis. Proceedings, International Symposium on Numerical Models in Geomechanics, NUMOG VIII, Balkema, p. 107-116.

Calvello M. and Finno R.J. (2003). Modeling 11. excavations in urban areas: effects of past activities. Italian Geotechnical Journal, 37(4), 9-23.

Calvello M. and Finno R.J. (2004). Selecting 12. parameters to optimize in model calibration by inverse analysis. Computers and Geotechnics, Elsevier, Vol. 31, 5, 2004, 411-425.

Chung, C.-K. and Finno, R.J. (1992). 13. Influence of Depositional Processes on the Geotechnical Parameters of Chicago Glacial Clays, Engineering Geology, 32, 225-242.

Cho, W.J. (2007). Recent stress history 14. effects on compressible Chicago glacial clay, PhD thesis, Northwestern University, Evanston, IL.

Cho, W.J. and Finno, R.J. (2010). Stress-15. Strain Response of Block Samples of Compressible Chicago Glacial Clays, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 136, No. 1, 178-188.

Clayton, C.R.I., and Heymann, G. (2001). 16. Stiffness of geomaterials at very small strains. Geotechnique, Vol. 51 (3), 245-255.


Clough, G.W. and Mana, A.I. (1976). Lessons 17. learned in finite element analysis of temporary excavations. Proceedings, 2nd International Conference on Numerical Methods in Geomechanics, ASCE, Vol. I, 496-510.

Clough, G. W., Smith, E.M., and Sweeney, 18. B.P. (1989). Movement control of excavation support systems by iterative design. Current Principles and Practices, Foundation Engineering Congress, Vol. 2, ASCE, 869-884.

Finno, R.J. (2007). Use of monitoring data 19. to update performance predictions of supported excavations, theme lecture in the Proceedings, FMGM 2007, International Symposium on Field Measurements in Geomechanics, ASCE, Boston, September.

Finno, R.J., Atmatzidis, D.K., and Nerby, 20. S.M. (1988). Ground response to sheet-pile installation in clay, Proceedings, Second International Conference on Case Histories in Geotechnical Engineering, St. Louis, MO.

Finno, R.J. and Blackburn, J.T. (2005). 21. Automated monitoring of supported excavations, Proceedings, 13th Great Lakes Geotechnical and Geoenvironmental Conference, Geotechnical Applications for Transportation Infrastructure, GPP 3, ASCE, Milwaukee, WI., 1-12.

Finno R.J., Bryson L.S. and Calvello M. 22. (2002). Performance of a stiff support system in soft clay. Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 128, No. 8, p. 660-671.

Finno, R.J. and Calvello, M. (2005). 23. Supported excavations: the observational method and inverse modeling. Journal of Geotechnical and Geoenvironmental Engineering. ASCE, 131 (7).

Finno, R.J. and Roboski, J.F., (2005). 24. Three-dimensional Responses of a Tied-back Excavation through Clay, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 131, No. 3, 273-282.

Finno, R.J. and Tu, X. (2006). Selected Topics 25. in Numerical Simulation of Supported Excavations, Numerical Modeling of Construction Processes in Geotechnical Engineering for Urban Environment, International Conference of Construction Processes in Geotechnical Engineering for Urban Environment, Th. Triantafyllidis, ed., Bochum, Germany, Taylor & Francis, London, 3-20.

Finno, R.J., Voss. F.T., Jr., Rossow, E., 26. and Blackburn, J.T. (2005). Evaluating damage potential in buildings affected by excavations, Journal of Geotechnical and Geoenvironmental Engineering, ASCE,131(10), 1119-2100.

Finno, R.J., Blackburn, J.T. and Roboski, 27. J.F., (2007). Three-dimensional Effects for Supported Excavations in Clay, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 133, No. 1, January, 30-36.

Geddes, J.D. (1977). Construction in areas 28. of large ground movements. Proc., Conf. on Large Ground Movements and Structures, Halstead Press, New York, NY 623-646.

Geddes, J.D. (1991). Discussion of 29. building response to excavation-induced settlement, by Boscardin, M.D. and Cording, E.J., Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 117(8), 1276-1278.

Ghaboussi, J. and Pecknold, D.A. (1985). 30. Incremental finite element analysis of geometrically altered structures. International Journal of Numerical Methods in Engineering. Vol 20(11) 2061-2064.

Hashash, Y.M.A., and Whittle, A.J. (1996) 31. Ground movement prediction for deep excavations in soft clay, Journal of Geotechnical Engineering, Vol. 122, No. 6, 474-486.

Hashash, Y. M. A., Marulanda, C., Ghaboussi, 32. J. and Jung, S. (2006) Novel approach to integration of numerical modeling and field observations for deep excavations, Journal of Geotechnical and Geoenvironmental Engineering, Vol. 132, No. 8, 1019 - 1031.

Hill, M. (1998). Methods and guidelines for 33. effective model calibration. U.S. Geological Survey. Water-resources investigations report 98-4005.

Holman, T.P. (2005). Small strain behavior of 34. compressible Chicago glacial clay. PhD thesis, Northwestern University, Evanston, IL.

Jardine, R.J., Symes, M.J. and Burland, 35. J.B. (1984). The measurement of small strain stiffness in the triaxial apparatus. Geotechnique, Vol. 34 (3), 323-340.

Koutsoftas, D. C., Frobenius, P. Wu, C.L. 36. Meyersohn, D. and Kulesza, R. (2000). Deformations during cut-and-cover construction of Muni Metro Turnback


project, Journal of Geotechnical and Geoenvironmental Engineering, Vol. 126, No. 4, 344-359.

Langousis, M.. (2007). Automated Monitoring 37. and Inverse Analysis of a Deep Excavation in Seattle, MS thesis, Northwestern University, Evanston, IL.

Ledesma, A., Gens, A, and Alonso, E.E. 38. (1996). Estimation of parameters in geotechnical backanalysis. I – Maximum likelihood approach, Computers and Geotechnics, Vol. 18, No. 1, 1-27.

O’Rourke, T.D. and Clough, G.W. (1990). 39. Construction induced movements of insitu walls. Proceedings, Design and Performance of Earth Retaining Structures, Lambe, P.C. and Hansen L.A. (eds). ASCE, 439-470.

O’Rourke, T.D. and O’Donnell, C.J. (1997). 40. Deep rotational stability of tiedback excavations in clay, Journal of Geotechnical Engineering, ASCE, Vol. 123(6), 506-515.

Ou, C.Y., Chiou, D.C. and Wu, T.S. (1996). 41. Three-dimensional finite element analysis of deep excavations. Journal of Geotechnical Engineering, ASCE, 122(5), 473-483.

Ou, C.Y., Tang, Y.G. (1994). Soil parameter 42. determination for deep excavation analysis by optimization. Journal of the Chinese Institute of Engineering, Vol. 17(5)}, pp.671-688

Peck, R.B. (1969). Deep excavations and 43. tunneling in soft ground. Proceedings, 7th International Conference of Soil mechanics and Foundation Engineering, State-of-the-ArtVolume, 225-290.

Poeter, E.P. and Hill, M.C. (1997). Inverse 44. Methods: A Necessary Next Step in Groundwater Modeling, Ground Water, Vol. 35, no. 2, 250-260.

Poeter, E.P. and Hill, M.C. (1998). 45. Documentation of UCODE, a computer code for universal inverse modeling. U.S. Geological Survey Water-Resources investigations report 98-4080, 116 pp.

Rechea, C. (2006). Inverse analysis of 46. excavations in urban environments, PhD thesis, Northwestern University, Evanston, IL.

Roboski, J.F. and Finno, R.J., (2006). 47. Distributions of Ground Movements Parallel to Deep Excavations, Canadian Geotechnical Journal, Vol. 43 (1), 43-58.

Santagata, M., Germaine, J.T. and Ladd, C.C. 48. (2005). Factors Affecting the Initial Stiffness of Cohesive Soils. Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 131(4), 430-441.

Sabatini, P.J. (1991). Sheet-pile installation 49. effects on computed ground response for braced excavations in soft to medium clays. MS thesis, Northwestern University, Evanston, IL.

Schantz, T., Vermeer, P.A. and Bonnier, P.G. 50. 1999. Formulation and verification of the Hardening–Soil Model. R.B.J. Brinkgreve, Beyond 2000 in Computational Geotechnics. Balkema, Rotterdam, 281-290.

Stallebrass, S.E. and Taylor, R.N. (1997). 51. The development and evaluation of a constitutive model for the prediction of ground movements in overconsolidated clay. Geotechnique, Vol. 47 (2) 235-253.

Su, Y.Y., Hashash, Y.M.A. and Liu, L.Y. 52. (2006). Integration of construction as-built data with geotechnical monitoring of urban excavation, Journal of Construction Engineering and Management, Vol. 132, No. 12, 1234-1241.

Son, M. and Cording, E.J. (2005). Estimation 53. of building damage due to excavation–induced ground movements, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 131(2), 162-177.

Trupp, T., Marulanda, C., Hashash, Y.M.A., 54. Liu, L., and Ghaboussi, J., (2004). Novel methodologies for tracking construction progress of deep excavations, Geo-Trans 2004, ASCE, Los Angeles, CA.

Viggiani, G. and Tamagnini, C. (1999). 55. Hypoplasticity for modeling soil non-linearity in excavation problems. Pre-failure Deformation Characteristics of Geomaterials, Jamiolkowski, M., Lancellotta, R. and Lo Presti, D. (eds.), Balkema, Rotterdam, 581-588.

Whittle, A.J. and Kavvadas, M.J. (1994). 56. Formulation of MIT-E3 constitutive model for overconsolidated clays. Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 120(1), 173-198.

Whittle, A.J., Hashash, Y.M.A. and Whitman, 57. R.V. (1993). Analysis of deep excavation in Boston, Journal of Geotechnical Engineering, Vol. 119, No. 1, 69-90.


Innovative Construction Techniques used at North Shore Construction ProjectKanchan K. Sen, Nicholson Construction Company, Cuddy, Pennsylvania, USA

[email protected]

Aaron Evans, Nicholson Construction Company, Cuddy, Pennsylvania, USA

[email protected]

ABSTRACTThe North Shore Connector (NSC) project will extend the Allegheny County Port Authority’s Light Rail Transit System from downtown Pittsburgh below the Allegheny River to the North Shore via twin bore tunnels. Jet grouting, diaphragm walls, and soil mixing were selected as cost-effective technically viable ground improvement and support of excavation techniques.

Jet grouting was specified to provide break-out (egress) and break-in (ingress) block sections at the tunnel boring machine (TBM) launching and receiving pits. Jet grout treatment was also used adjacent to buildings with shallow foundations and below retaining walls along the tunnel alignment. A new method for measuring jet grout column diameter based on electrical resistivity was employed on the project.

New equipment was introduced in the U.S for constructing soil mix temporary support of excavation using a soil-mix technique termed Cutter Soil Mixing (CSM) due to the use of two vertical cutting wheels that have evolved from diaphragm wall trench cutter technology. The system allows for continuous water tight soil mix panel construction including steel beams.

This paper highlights some applications of jet grouting and various support of excavation systems (e.g. slurry walls, cement bentonite walls and CSM walls), including new innovative techniques (e.g. Cyljet and CSM) used under the constraints of an urban environment.

INTRODUCTION

Project:

The North Shore Connector (NSC) project in Pittsburgh, PA, will extend the Port Authority of Allegheny County’s Light Rail Transit System approximately 1.2 miles (1.9 kilometers) from the Gateway Subway Station, underneath Stanwix Street and Allegheny River, and to the north shore of the river using twin bore tunnels (see Fig. 1). Nicholson Construction Company (NCC) was subcontracted to perform ground improvement and support of excavation (SOE) work.

Scope of Work:

NCC’s main scope of work was 32,700 cubic yards {25,020 cubic meters (m3)} of Jet Grouting (JG) including 14 columns to select jet grout parameters for the Column Confirmation Test Program (CCTP); 110,000 square feet {10,220 square meters (m2)} of 31.5-inch (0.80-m) thick Cutter Soil Mixing; 37,000 square feet (3,437 m2) of 31.5-inch (0.80-m) thick Diaphragm Wall and 22,000 square feet (2,044 m2) of 31.5-inch

(0.80-m) thick Cement Bentonite Wall (CB).

GEOLOGY OF NSC PROJECT SITEThe NSC project is located in the floodplains at the confluence of the Allegheny, Monongahela and Ohio Rivers. The floodplain consists of unconsolidated deposits that overlie rock. Two separate types of valley soils are recognized; an upper alluvium deposit and lower fluvioglacial deposit.

The alluvium deposit is 0 to 15-ft. (0 to 4.60-m) thick primarily consisting of silty clay or clayey sand having low strength, high compressibility, and moderate to low permeability. Typical grain size distribution of the alluvium deposit is 20% gravel, 15% sand and 65% fines.

The fluvioglacial deposit is approximately 25 to 50-ft. (7.6 to 15.2-m) thick overlying the rock. It consists of coarse grained glacial deposits typically having high frictional strength, low compressibility and high permeability. The fluvioglacial deposit was found to be thicker and coarser on the north side of the Allegheny River with a typical grain size distribution of 57% gravel, 38% sand, and 5% fines. On the south side the deposit was found to consist


typically of 40% gravel, 50% sand, and 10% fines. The majority of the tunnel is constructed through the fluvioglacial deposit, though the tunnel extends into the underlying bedrock below the river.

In general, the ground water table exists under phreatic condition. The alluvial and especially the fluvioglacial deposits provide a relatively permeable aquifer. The normal ground water level is about elevation 710 to 714-ft. (216.5 to 217.7-m). The water level fluctuates depending on the river pool and flood elevation levels.

Geologic maps of the project area indicate that the top of rock at the tunnel alignment rises from elevation 660-ft. (201.2-m) from the receiving pit of the TBM at the south end to 680-ft. (207.3-m) at the launching pit of the TBM at the north end of the Allegheny River. The existing ground level varies from elevation 723 to 729-ft. (220.4 to 222.2-m).

The upper layer of rock consists of shale with interbedded siltstone seams to siltstone with shale foliage. A thin layer of 1-2 ft. (0.30-0.60-m) thick limestone underlies the shale. Below the limestone is a layer of calcareous clay stone. The lowermost rock unit sampled was found to be fine grained sandstone. The uniaxial compressive strength of rock and permeability of soil and rock based on the Geotechnical Baseline Report (GBR) is shown in the following Tables 1 and 2.

[TABLE 1] Strength of Rock

Rock TypeUniaxial Comp.Strength (psi)

Remarks

Shale 5,400 to 11,800Mostly Shale and Limestoneencountered

Limestone 2,400 to 11,300

Claystone 200 to 4,300

Siltstone 3,400 to 11,000

[TABLE 2] Permeability of Soil and Rock

Location TBM

Receiv-ing Pit

Stanwix St.

TBM Launch

Pit

North Side

Station

Deposit Co-efficient of Permeability (cm/sec)

Fill5x10-5 1x10-6

5x10-5 1x10-6

7.1x10-3 1x10-6

7.1x10-5 1x10-6

Alluvial1x10-3 1x10-6

1x10+ 1x10-6

2.7x10-4 1.8x10-7

2.7x10-4 1.8x10-7

Fluvio-glacial

1x10-2 1x10-4

1x10-2 1x10-4

1.4x10-3 1x10-4

1.4x10-3 1x10-4

Rock1x10-4 1x10-5

1x10-4 1x10-5

2x10-4 1x10-4

2x10-4 1x10-4

JET GROUTINGJet grouting (JG) was selected to improve ground properties (mechanical characteristics of the treated soil; most notably compressive strength of the soil while simultaneously

[FIG. 1] Nicholson’s work at North Shore Connector. Adapted from map published by Port Authority of Pittsburgh


decreasing its permeability) at the launching and receiving pit of the TBM. The technical requirement was to create a jet grout block mass encompassing both tunnels 22-ft. (6.70-m) diameter each and extending 6-ft. (1.83-m) around the tunnel. Requirements were 100% coverage of the treated soil and a minimum unconfined compressive strength (UCS) of 150-psi (1.03 MPa).

Additionally, ground treatment was specified for the east tunnel at Stanwix Street to minimize ground movement and avoid settlement of shallow foundation structures. Requirements were 75% coverage of the treated soil and minimum UCS strength of 150-psi (1.03 MPa) (tested at 28 days after collection of the grout samples). Jet grouting was also considered below retaining walls to replace pile foundations located along the tunnel alignment.

Construction Method

Based on available data and past experience, double fluid JG system using air and grout was considered for the NSC project. In the double fluid system the cement grout is injected at high pressure and is aided by a cone of compressed air, which shrouds the grout and helps to produce bigger column diameters.

Before beginning production, the Column Confirmation Test Program (CCTP) was performed to select jet grout parameters and grout mixes. The test program

was performed directly adjacent to the south shore of the Allegheny River.

[FIG. 3] Layout of JG Columns

[FIG. 2] Energy vs. diameter of JG column based on fi eld test


Initial jet grout parameters were selected based on soil type, soil consistency, bulk density, atterberg limits, grain size distribution, water content and permeability of in-situ soil. 5 to 8-ft. (1.5 to 2.4-m) diameter grout columns were tested during CCTP. On the basis of trial columns, the graph of jet grouting energy versus column diameter was drawn to finalize the jetting parameters as shown in Fig. 2.

The JG columns were based on 7-ft (2.13-m) diameter. The layout of JG Columns at the Launch Pit is shown in Fig. 3. The JG spoils were directly deposited to the barges and truck mounted dumpsters during CCTP and production respectively (Fig. 4 and 5). The following were considered to check the effectiveness of the jet grouting method.

Density and Unconfined Compressive • Strength (collecting spoil and core samples).

Diameter and continuity of jet grouted • columns (inspecting core samples at the half radius and periphery of the jet grouted column).

Physical inspection of the jet grouted hole • (as the holes were deep and below water table, open excavation was not conceived; Cyljet, a method based on electrical resistivity method, was proposed to check the diameter of jet grouted columns).

Selection of Grout

The grout mix for jetting was cement, bentonite, and water. Bulk ordinary portland cement (Type-II) with bentonite was used to reduce bleed.

Quality Control

The jet grout parameters were monitored using a Lutz system (real time measuring and graphical recording system for drilling and grouting work). The specific gravity of grout was measured at the location of the grout plant. Also, spoil samples were collected to measure the specific gravity and UCS. In addition cored samples from the JG columns were tested for strength requirement. Results of JG core samples are shown in Fig. 6.

An innovative technique known as Cyljet, developed by Europenne de Geophysique (E.D.G) of France was used to confirm the diameter of jetted columns.

0

10

20

30

40

50

60

0 200 400 600 800 1,000

UCS (psi)

Dep

th (f

t)

NCC Target Minimum: 70 psi

[FIG. 6] UCS results of JG

Cyljet Method

In the Cyljet method the potential differences generated by an induced electric current is measured and analyzed around a borehole (Frapin et. al. 2001). A multi conductor cable is inserted in the borehole, which connects regularly-spaced current injecting electrodes (A) and receiver electrodes (M) to a computer controlled selector (see Fig. 7).

[FIG. 4] Spoil Management for CCTP

[FIG. 5] Spoil Management for JG


In this method the potential difference is measured with the induced electric current around a borehole in cylinder 5 to 10-m (16 to 33 ft) in diameter. Depending on ground resistance the responses are varied. The measurements show the geo-electrical pseudo-section showing resistance of equi-potential lines. A calibration borehole is drilled and measured to know the results of the original ground. Subsequently when a measurement is performed on a freshly grouted column, the result shows the contrast in resistance between the original ground and of the soilcrete forming the column. These resulting signals are processed by a computer and column diameters are simulated. The typical result from NSC project site is shown in Fig. 8. The diameter of the columns is obtained by comparing various simulations with the actual field readings.

[FIG. 7] Cyljet measurement

Observations

The use of jet grouting was found to be an effective ground improvement system for the congested urban environment. The JG had several advantages over other ground improvement technologies:

• The ability to create large treated cylindrical elements using small drill holes.

• Small drill holes can be used to avoid utilities.

• Ability to treat specific zones.

• Different energies could be used in different soil layers.

• Flexibility of drill equipment to drill angled holes.

• Ability to control/transfer spoil directly under space constraints of urban construction.

• Cyljet proved to be an effective method for determining the size of the jet grouted columns.

Challenges

The presence of coarser than anticipated soils precluded the ability to obtain continuous core samples at low compressive strengths. The specified jet grout strength was 150 psi (1.03 MPa). However, the gravel and cobble were glacially deposited materials of very high strength. A lesson learned is that the grout strength was not sufficient to hold the materials in place during the coring operation. For this reason the grout strength had to be increased in excess of 1000 psi (6.9 MPa) in order to facilitate the coring operation. Accordingly, the required grout strength was

[FIG. 8] Cyljet test result from NSC


controlled by the method of quality control, not by the as-designed requirements.

SUPPORT OF EXCAVATION The temporary support of excavation (SOE) was cement deep soil mix (CDSM) at the launching and receiving pits for the Tunnel Boring Machine (TBM) and at the cut-and-cover section of the tunnel. NCC proposed a CSM (Cutter Soil Mixing) wall. Under low headroom areas, the temporary SOE was constructed using a cement-bentonite slurry.

The permanent support system for the new station shell at the north shore was constructed using diaphragm wall as contract specified.

CSM

The CSM wall is primarily deep soil mixing constructed by mixing cementitious binder with the natural soil in-situ. The CSM tool is based on trench cutter technology used for construction of slurry walls. Cutter wheels break up the soil matrix and mix it with cement slurry.

Construction Method

The CSM process consists of primarily two phases. Penetration of the tool with outward rotation of the drums whilst injecting a “drilling in” bentonite slurry in between the drums, followed by inward rotation of the drums and withdrawal with continued injection employing a cementitious binder suspension. Basics of the CSM method are shown in Fig. 9.

The cutting and mixing drums are mounted on compact hydraulic motors. The drums are designed to combine high penetration rates and excellent soil/cement mixing. The cutting/mixing wheels are mounted on a Kelly bar supported by a hydraulic track rig as shown in Fig. 10 and 11. The Kelly bar allows accurate positioning and verticality of the soil mixed structure. The CSM machine is equipped with an onboard real time monitoring system.

The CSM walls 31.5-inches thick (0.80-m) consisted of overlapping primary (initial panels) and secondary panels (constructed in between the constructed primary panels). The panels were overlapped to account for a verticality tolerance equivalent to 1% of the panel depth ranging from 3 to 6-ft (0.9 to 1.8-m).

[FIG. 9] Basics of CSM

[FIG. 10] Cutting/Mixing drums of CSM

[FIG. 11] CSM Rig at North Shore

A compact automated grout mixer was used for preparation of grout for CSM wall system (Fig. 12). A standard bentonite mixer and slurry tanks were used for preparing bentonite slurry. The soldier piles were installed after completion of a panel.

Selection of Grout

The specified unconfined compressive • strength of the soilcrete was minimum 70-psi (0.48 MPa) and average about 150-psi (1.03 MPa). The designed grout mix was based on 190-kg of cement per cubic meter (320 lb per cu yd) of treated soil.


Quality Control

The quality of the work is monitored using a real time monitoring system attached to the CSM Rig (see Fig. 13). It measures the verticality by a two axis (X, Y) inclinometer, the excavated depth of the panel, fluid flow (slurry injection, cementitious binder), motor hydraulic pressure, hydrostatic pressure outside the tool and rotation speed of the wheel.

[FIG. 12] Grout Plant at North Shore

[FIG. 13] Instrumentation for CSM

The bentonite slurry and grout were tested during the construction for specific gravity and Marsh Funnel Viscosity of cementitious binder, Marsh Funnel Viscosity of bentonite slurry, UCS test of fresh in-situ soil cement samples and confirmatory core samples.

Observations

The CSM system was found to be a cost effective method for support of excavation below the water table producing a stiff and watertight system. As the prevailing soil was utilized as construction material, spoil removal was minimal. Also, vibration-free and low noise of

plant and equipment was helpful for an urban construction project. The CSM method was found to treat soil uniformly although the mixing was found more effective in the fluvioglacial deposit. The exposed face of the CSM wall at the location of the Launch Pit is shown in Fig.14. The soilcrete created using CSM system reached a minimum UCS 500-psi (3.5 MPa) as shown in Fig. 15. Verticality was maintained accurately using a real time data management system.

0

10

20

30

40

50

60

0 500 1,000 1,500 2,000 2,500 3,000

UCS (psi)

Dep

th (f

t)

NCC Target Minimum: 200 psi

[FIG. 15] UCS results of CSM

Challenges

Challenges on the project included encountering cobble layers, which impacted the production rate and produced higher than expected machine wear.

Cement Bentonite Wall

Cement bentonite (CB) walls were proposed under highways I-279 and SR-65 as the CSM rig

[FIG. 14] CSM wall at the Launching Pit


could not be utilized under low headroom areas along the alignment of the CSM wall. CB wall was constructed using cement bentonite mixing method.

Construction Method

The CB walls consist of overlapping primary and secondary panels excavated using conventional slurry wall technology. The trenches were excavated using a low headroom grab with bentonite slurry as a stabilizing agent (see Fig. 16).

[FIG. 16] CB wall below I-279

Once the panels were excavated to full depth, steel beams were placed and the stabilizing agent was substituted with the designed cement-bentonite grout mix. Soil disposal system is shown in Fig. 17. The bentonite slurry consisted of a uniform mixture of bentonite and water. The designed cement bentonite mix was supplied at the bottom of the excavated panel through multiple tremie pipes utilizing grout pumps. The exposed face of CB wall below I-279 is shown in Fig. 18.

Selection of Grout

The design of the grout mix for the CB walls was based on laboratory test results and past work experiences. The specified unconfined strength of the soilcrete was minimum 70-psi (0.48 MPa) and average 150-psi (1.03 MPa).

Quality Control

In order to guide the grab during initial excavation, and to ensure the position and verticality of the CB wall, a guide wall was constructed prior to commencement of the

CB walls. During excavation, the position of the suspension cable of the grab with respect to the reference points to the guide wall was measured.

The field personnel tested the viscosity, specific gravity and sand content of the bentonite slurry during panel excavation. Samples were collected from the fresh grout mix for UCS test. Also, core samples were collected from the completed CB panels.

Observations

CB walls can be constructed for a temporary watertight SOE system for variable and permeable ground conditions for urban construction. Under restricted headroom areas this could be constructed using conventional slurry wall technology with low headroom plant and equipment.

[FIG. 17] Soil disposal system for CB wall

[FIG. 18] Exposed face of CB Wall below I-279


Diaphragm Wall

A concrete diaphragm wall was constructed as a support of excavation during temporary construction stages and also to form the structural walls for the new North Shore station. The construction sequence (typical) of the Diaphragm wall is shown in Fig. 19.

[FIG. 19] Typical construction sequence for Diaphragm wall

Construction Method

Excavation was carried out using mechanical clamshell grabs. The construction began by excavating and concreting a primary panel (initial panel). After the panel excavation was completed the forms were set at both ends of the primary panel using Coffrage avec Waterstop (CWS) endstop. The CWS system is a technique that allows construction of geometrically well defined joints equipped with waterstop products (Vanel 1992).

The CWS endstop was left in place and extracted laterally during the excavation of the adjacent panel (follow-up panel) as shown in Fig. 20. As the CWS endstop was left in place while the adjacent panel was excavated, it was used as a guide, guaranteeing the geometrical continuity of the wall. The follow up panels were constructed progressively away from the completed primary panels using one CWS endstop. Eventually the secondary panels (located in between two completed panels) were constructed.

Once the excavation was completed up to the designated depth, the bottom of the trench was cleaned by a clamshell grab. Then the contaminated bentonite slurry was pumped out and desanded. The fresh bentonite slurry was recycled until the bentonite slurry complied with the specification.

Later the reinforcement cages were lowered and the concreting was carried out using steel tremie tubes. Ready mix 4000-psi (27.6 Mpa) high slump concrete was used for diaphragm walls (see Fig. 21). The bottom of the tremie pipes was kept approximately five feet into the concrete to avoid contamination with bentonite.

Quality Control

In order to guide the grab during initial excavation, and to ensure the position and verticality of the diaphragm wall, a guide wall was constructed prior to commencement. During excavation, measurement of the position of the suspension cable of the grab with respect

[FIG. 20] Typical construction sequence for CWS Endstop

Completed Primary Panel with Endstop Excavation of Follow-up Panel

Removal of Endstop from Completed Primary Panel


to the reference points to the guide wall were measured. In addition to above, verticality can also be monitored using a real time data management system. The supervisory screen of the data management system typically used for construction of diaphragm walls is shown in Fig. 22. The data management system records depth of excavation, verticality of the trench and the excavation parameters (mud pressure, flow etc.) in real time.

Field personnel tested the properties of the bentonite slurry during the excavation and concreting phase using mud balance, marsh funnel, filter press, pH meter etc. The concrete level in the panel was measured regularly.

A concrete curve showing actual volume of concrete poured versus theoretical volume was checked for unusual concrete overruns.

Observations

The diaphragm wall can be constructed in hard, variable and highly permeable ground conditions using bentonite as a stabilizing agent. An advantage of the system was minimum noise and vibration during construction of the permanent SOE for the congested urban environment. Innovative design and improved plant and equipment has practically eliminated the physical constraints for working under low headroom and/or restricted areas.

SUMMARY AND CONCLUSIONSMultiple geotechnical techniques were used to support activities associated with the construction of a new tunnel below the Allegheny River for Light Rail Transit System. These techniques included cutter soil mixing, jet grouting and diaphragm walls and were associated with support of excavation at the launch and receiving pit, break-in and break-out blocks, and construction of a new underground station.

The jet grouting operations achieved project objectives, most notably improved the compressive strength of the soil while simultaneously decreasing its permeability. The primary challenge with the jetting

operation was to achieve low strengths for the grouted mass in an environment of very dense course soils. Successful coring was not possible until the grout strength was increased dramatically because the gravel and cobbles were “plucked” from the relatively weak grout matrix. Accordingly, the required grout strength was dictated by the quality control process, not by the needs of the project. The Cyljet method was demonstrated to be an effective tool for estimating the diameter of the jet grouted column at depth.

The cutter soil mixing method proved to be a cost effective system for excavation support below the water table producing

[FIG. 21] Diaphragm Wall Construction at NSC Project

[FIG. 22] Real time data management system for Diaphragm Wall


a stiff and water tight system. The system emulates a conventional beam and lagging wall, but the soil mix serves as impermeable lagging.

Finally, the new rail station was constructed using diaphragm wall method as permanent support of excavation. Selection of slurry wall for the temporary and permanent support of the structure simplified the construction operations in the congested urban environment.

ACKNOWLEDGMENTSThe authors like to thank Chris Hynes, Fred Tarquinio and Rick Deschamps of Nicholson Construction Company for providing information and valuable suggestions for preparation of this paper. Also, the authors thank the reviewers for their valuable comments.

REFERENCESFrappin, P. & Morey, J., 2001, Jet grouting 1. column diameter measurement using the electric cylinder method, Travaux no.775, May 2001, France.

Port Authority of Alegheny County, 2005, 2. Geotechnical Baseline Report Contract NSC-003/006, Pittsburgh, PA.

Vanel, P, 1992, Making Diaphragm Wall 3. Joints Watertight with the CWS System. American Society for Testing and Materials, Slurry Walls: Design, Construction and Quality Control, David B. Paul/Richard R. Davidson/Nicholas J. Cavalli, Editors, Special Technical Publication, ASTM STP 1192, Philadelphia, PA.


Large Scale Lateral Testing of Pile Foundations(Young Professor Paper Competition 2010)Anne Lemnitzer, Ph.D, Assistant Professor, California State University Fullerton

ABSTRACTFull-Scale cyclic field testing was performed on a 3x3 pile group consisting of nine cast-in-drilled-hole (CIDH) reinforced concrete shafts and a comparable single-shaft in order to investigate group interaction effects. Both specimens were constructed with pile diameters of d = 0.61m (2 ft), fixed head boundary conditions, and placed in stiff clayey soil at a test site near the LAX Airport in Southern California. Internal instrumentation was installed in the single pile as well as in three piles within the pile group to obtain pile curvatures. External instrumentation controlled the lateral displacement of the specimens. A combination of experimentally and analytically obtained p-y curves for the single pile was used to simulate results for piles located in the group when field data were not available. All test specimens were tested to structural failure and excavated after test completion to investigate pile damage and plastic hinge formations. Calculated group interaction factors ranged between 1.0 at very small displacement levels of <2.5 mm (0.1 in), 0.8 at the initiation of the pile-soil-pile interaction, i.e. between 4 mm - 15 mm (0.16-0.6 in), and increased towards 0.9 near structural failure (displacement >25 mm or 1 in). A literature set of p-multipliers was found to reproduce an accurate fit to the available load displacement test data from the group.

INTRODUCTION The behavior of pile foundations subject to lateral cyclic loading depends on the magnitude and history of the applied forces, the pile head boundary condition, the pile material, as well as the characteristics of the surrounding soil. For bridge foundations, a common contruction method along the West Coast of the U.S. is the use of Cast-In-Drilled-Hole Shafts (CIDH piles). Piles are typically arranged in single configuration as large diameter piles (e.g. single shafts with diameters of about 1.83 m (6 ft), as indicated in Fig. 1) or in group arrangements with multiple piles of smaller diameters (e.g. 0.61 m (2 ft) diameter piles in a 2x2, 3x3 or larger arrangement) as shown in Fig. 1.

[FIG. 1] Typical Bridge Foundation components

Depending on the row spacing, the behavior of piles located within a group is strongly influenced by pile-soil-pile interaction. In

particular, piles spaced at distances less than six pile diameters on center (Cox et al. 1971, Rollins et al. 1998) may experience a reduction of capacity due to well known shadowing effects (Brown et al. 1987). Shadowing effects describe the development of stress zones that “overshadow” the pile or pile row adjacent to the piles directly subject to lateral displacement and reduce the strength of the overall pile group under horizontal loading. Hence a pile located within a group will have less capacity than an identical pile in single configuration.

The introduction of group efficiency factors and p-multipliers are two concepts to account for this reduction. P-multipliers are expressed as a function of pile or pile row position within the group (Brown et al. 1987; Ruesta and Townsend 1997; Rollins et al. 1998, 2002, 2006). Group efficiency factors are more commonly expressed as function of lateral pile displacement (Kotthaus et al. 1994, Lemnitzer et al. 2010). Group efficiencies describe a general reduction of the overall group capacity versus the sum of the capacity of the same amount of piles in single configuration and can be mathematically expressed as:

(1)


where Pg(y0) is the lateral force applied to a

pile group that causes a lateral deflection of y

0 at the pile cap; P

sp(y

0) is the corresponding

single-pile head load; Ng is the number of piles

in group, and η is the group efficiency factor. Group efficiencies represent a simple tool to assess expected group capacities and may be particularly beneficial in early stages of foundation design projects.

While this concept uses the overall load displacement relationships of corresponding single piles and pile groups, the concept of p-multipliers suggests a capacity reduction applied at the p-y level of a single pile. P-y springs, represented mathematically through parabolic p-y curves describe the soil resistance p per unit length for specified pile depths as a function of lateral pile displacement y at the same depth. Particularly well known sets of p-y curves are published by the American Petroleum Institute (API 1993). P-multipliers are identified by comparing p-y curves of single piles vs. p-y curves obtained from piles placed in group arrangements and represent a reduction factor of the measured relationships. A number of large scale experiments under different soil conditions and different head

boundary conditions as well as pile spacing have been conducted to identify p-multipliers for a variety of settings and combinations. A complete summary is presented in Lemnitzer et al. 2010.

This paper describes an experiment designed to fill the required literature gap for reinforced concrete pile foundations in single and group arrangement with fixed head boundary conditions (i.e. rotation is restrained due to a pile cap) in clayey soil. Many results were documented for steel piles with free head boundary conditions or piles in sands. An overview of the test results for a single 0.61 m (2 ft) diameter fixed head shaft and a 0.61 m (2 ft) diameter pile group consisting of 9 piles will address three important issues with respect to previous experimental studies: (1) testing of reinforced concrete columns instead of structural steel columns or concrete filled hollow steel sections; (2) testing under reversed cyclic loading to large deformations, including inelastic behavior, and (3) deriving group efficiency factors and studying previously obtained p-multipliers. An overview of the test program followed by a discussion of the test results and consecutive analytical studies

[FIG. 2 a-e] Construction of Single Pile (a) and Group Specimen (b) and completed Specimens: single pile (c-d) and pile group (e)

(a) (b)

(c) (d) (e)


will be provided in this manuscript. A lengthy discussion of this experiment was recently published by Lemnitzer et al. (2010) in the Journal of Geotechnical and Geoenvironmental Engineering (ASCE).

TEST SETUPThe single pile test consisted of a 0.61 m (2 ft) diameter reinforced concrete shaft extending 7.62 m (25 ft) below ground with reinforcement extending above ground into a load application cap with dimensions of 1.5 m x 2.1 m x 1.8 m (5 ft x 7 ft x 6 ft). A schematic test layout is shown in Fig. 4. The pile was reinforced with 8 - 29mm∅ (#9) bars, providing a vertical reinforcement ratio of ~2.0 % and a transverse reinforcement of 13 mm∅ (#4) spirals at 114 mm (4.5 in) pitch. Maximum concrete strengths f’

c

were measured between 30.3 MPa (4.4 ksi) and 35.9 MPa (5.2 ksi) (ASTM C39) and testing of longitudinal rebar samples indicated a yield stress f

y of approximately 483 MPa (70 ksi). The

group specimen consisted of nine 0.61 m (2 ft) diameter piles spaced 1.83 m (6 ft) on center in each plane direction (i.e. 3 pile diameters). Piles extended into a cap with dimensions of 4.8 m x 5.5 m x 1.8 m (16 ft x 18 ft x 6 ft). All piles extended about 7.6 m (25 ft) below ground and were reinforced with 8 - 22mm∅ (#7) bars, resulting in a longitudinal reinforcement ratio of 1%. Transverse reinforcement was provided by 13mm∅ (#4) spirals spaced at a 100 mm (4 in) pitch. This reduction in reinforcement of the group piles versus the single pile was needed to facilitate the limited capacity of the hydraulic equipment. Maximum concrete compressive strengths (f’

c) were measured to range between

28.3 MPa (4.1 ksi) and 33.8 MPa (4.9 ksi). A gap existed underneath the pile cap for both test specimens to avoid the contribution of cap-soil friction to the overall lateral pile resistance. The reaction system for both tests consisted of a reaction block with dimensions of 7.3 m x 3.6 m x 1.8 m (24 ft x 12 ft x 6 ft) cast on top of two 1.82 m (6 ft) in diameter, 14.6 m (48 ft) deep drilled shafts. Four actuators were installed between the reaction system and the test specimen to apply lateral loads/displacements. Photographs of the single pile and pile group under construction and as completed specimens are shown in Fig. 2 a-e.

GEOTECHNICAL SITE CONDITIONSThe test site is a Caltrans owned property located at the interchange of I-405 and I-105 in

Hawthorne, California. Site exploration included seismic cone penetration testing (SCPT), rotary-wash borings with standard penetration testing, down-hole suspension logging of shear wave velocities, pressuremeter testing (PMT) and test pit excavation mapping. Laboratory testing of soil samples provided information on shear strengths and soil classifications. The soil conditions at the test site consist of deep alluvial sediments. The upper 15 m (50 ft) are mostly silty clays interspersed with relatively thin layers of silty sand. Fig. 3 shows an extract of the soil profile and the measured in-situ and laboratory shear strengths of the site.

[FIG. 3] Soil profi le and soil strengths at the test site

SPECIMEN INSTRUMENTATIONBoth test specimens were equipped with internal and external sensors consisting of strain gauges, displacement transducers (LVDTs), fiber-optic Fiber-Bragg gratings, wire potentiometers, and inclinometers. The single pile and three piles within the nine-pile group (shaded in Fig. 4) were densely instrumented within the upper part of the pile where the majority of deformations were expected and moderately instrumented along the remaining pile depth. 102 sensors were used for each instrumented pile. External sensors were installed between a reference


frame and the respective pile caps to measure lateral displacement and to monitor the cap rotation. Internal instrumentation provided strain data to derive curvature profiles for the single pile. An instrumented reinforcement cage before placement into the ground is exemplarily shown in Fig. 5. Unfortunately, data at large deformations from piles within the pile group could not be recovered due to sensor

malfunction during testing. The behavior of piles within the group was therefore simulated through a combination of analytical models and literature based relationships.

The test protocol for the single pile included restraining the cap rotation to zero using three sensors mounted between an external reference frame and the vertical face of the pile cap. A zero rotation boundary condition could not be achieved for the group specimen due to equipment and test geometry limitations. Therefore, the pile cap rotation of the group was monitored using 3 LVDTs installed between an external reference frame and the top surface of the pile cap. All data were recorded using a National Instruments data acquisition system.

TEST CONDUCTIONLateral loads were applied to the specimens using four 2 MN (450 kip) actuators installed between the reaction block and the pile caps. The hydraulic actuators were controlled by an MTS Flextest GT Controller, which was able to independently operate actuators under displacement or load control. Three cycles of lateral displacement were applied at monotonically increasing peak values up to a maximum displacement of 76 mm (3 in) and 25 mm (1 in) for the single pile and pile group, respectively. The load capacity of the four actuators was exhausted at a lateral displacement of 25 mm (1 in) during the group test. Therefore, testing was paused and three additional actuators were installed. Testing was continued under monotonic loading to a lateral displacement of 250 mm (10 in). Fig. 6 shows the four 2 MN (450 kip) actuators with the three additional actuators between the reaction block and the nine pile group specimen.

[FIG. 4] Test Layout for single pile and pile group

[FIG. 5] Sensor instrumentation inside a pile[FIG. 6] Additional hydraulic jacks next to actuators to increase hydraulic capacity


TEST RESULTS

Load – Displacement Relationships

Figs. 7 and 8 present the measured load displacement backbone curves of the single pile and the group specimen, respectively. A maximum capacity of 1.21 MN (273 kips) was reached for the single pile at a horizontal displacement of 76 mm (3 in). The group specimen achieved its maximum lateral resistance of 10.23 MN (2,300 kips) at 102 mm (4 in) lateral displacement (during the monotonic push). Generally, data points for the backbone curves are taken at the peak of the first cycle of loading. When problems with the data acquisition system were encountered, load points were derived from subsequent cycles. This was only the case for few data point during the group test. Data taken from subsequent cycles are shown in Fig. 8 as light grey, individual data points. The break of the data at a displacement level of around 30 mm (1.2 in) in the load displacement curve of the pile group represents the point of exhausted actuator

capacity during testing. Continuation of testing with additional actuators under monotonic loading is labeled in Fig 8.

Pile Cap Rotation

The single pile cap rotation was controlled to zero throughout the entire experiment. Due to limitations associated with the actuator capacity, rotation was not restrained for the pile group but was monitored and recorded using the external LVDT sensors installed at the top surface of the pile cap of the group specimen. Two sensors were located along the center of the cap and one sensor was located at a specified distance in the push direction to describe a triangular plane. Fig. 9 shows the measured rotation of the pile cap versus top horizontal displacement of the group. Rotations increased linearly with increasing lateral displacements but were generally insignificant (i.e. max. measured rotation was 0.25 degrees at point of complete structural failure).

[FIG. 9] Rotation of pile group versus lateral displacement

In order to derive group efficiency factors from single pile and pile group test results, both specimens need to have identical boundary conditions. Since rotation was controlled to zero for the single pile, but allowed for the pile group, sub-studies were conducted to assess if the observed cap rotation influences the top load displacement relationship of the group specimen. This study was conducted by Khalili-Tehrani (2009) using the finite element program ABAQUS. A model of the pile group was developed and two test scenarios were simulated: (1) the model pile group was subjected to lateral displacements without rotation, i.e. under perfectly fixed conditions and (2) the group was subjected to lateral displacements while simultaneously applying the

[FIG. 7] Single pile experimental load displacement relationship

[FIG. 8] Pile group experimental load displacement relationship


measured field rotation in the computer model. Load-Displacement relationships were obtained for both scenarios and compared. Results indicated that cap rotation at the levels observed in the tests does not influence the group load-deflection response and can be neglected. Obtained load displacement relationships were almost identical and are presented in more detail in Khalili-Tehrani (2009) and Lemnitzer et al. (2010). Therefore, results obtained experimentally and analytically as described later can be directly used to adequately estimate the group efficiency of the test specimen.

Post Test Excavations

Post-test excavations of the group specimen were performed to investigate the pile damage and monitor crack layouts (Figs. 10 and 11). Extensive cracks with crack widths up to 50 mm (2 in) were observed between ground line and a depth of 0.9 m (3 ft), describing the range of pile depth with plastic hinge formations. At depths lower than 1.8 m (6 ft), no cracks were observed. Several piles experienced concrete spalling as shown in Fig. 10a. Severe damage was observed in front-row piles. Only moderate cracks were found in piles located in the middle and third row. The residual lateral displacement of the pile group was measured to be about 120 mm (5 in) which may be observed in the photograph of Fig. 11.

[FIG. 10 a & b] Pile Damage in the piles of the group specimen

[FIG. 11] Excavated group specimen with permanent lateral displacement

DISCUSSION AND ANALYTICAL STUDIESPre-test blind predictions using various analytical methods were performed for the fixed head single pile and are presented in Rha (2006), Khalili-Therani (2009) and Lemnitzer et al. (2010). Modeling studies after test completion were predominantly executed using Frame Lab, a computer code developed by Taciroglu et al. (2006), which works in a similar fashion to the commercially available program LPile. Single pile curvature profiles were developed from experimental strain data obtained through various sets of internal instrumentation, e.g. strain gauges, LVDTs and Fiber Optic Sensors. Sample curvature profiles for the fixed head pile at a lateral top displacement of 19 mm and 25 mm (0.75 and 1.0 in) are shown in Fig. 12. Data clearly show a double curvature bending of the pile and indicate plastic hinge locations of the pile at the interface of the pile and the cap as well as at a depth of around 0.6 – 1.8m (2 – 6 ft).

Due to large scatter within some of the curvature data, the traditional process of double integration and double differentiation to obtain p-y curves was not suitable. An alternative procedure was sought by taking advantage of the capabilities of FrameLab. The FrameLab program uses mathematical descriptions of traditional p-y curves and can be calibrated with test specific properties (e.g. soil properties, structural properties). A model of the single pile was created in FrameLab and program parameters were then adjusted using insitu characteristics. An optimization process was performed until the model accurately calculates the overall single pile


load displacement relationship measured in the field. This procedure also ensured that experimentally obtained curvature profiles were matched. No major specimen properties were modified during this optimization process. Once an excellent fit was obtained, p-y curves for the single pile were extracted. A detailed description of this process can be found in Khalili-Tehrani 2009. Fig. 13 presents three sets of load displacement relationships for comparison: (1) experimental single fixed head pile data shown with diamond symbols, (2) a single pile load displacement curve reproduced with the Frame Lab model shown as continuous line and labeled as “calibrated p-y model 2%” and (3) a load deflection relationship for a single pile with a reduced reinforcement ratio of 1%. Fig. 13 indicates an accurate fit of the initial stiffness and overall capacity between the 2% reinforcement single pile experimental and analytical relationships and validates the accuracy of the calibrated analytical model. The calibrated model applied with a smaller reinforcement ratio of 1% represents the reinforcement condition as present for a pile within the group. This calculated load displacement curve for a pile with a 1% longitudinal reinforcement ratio is shown as dashed line in Fig. 13 and labeled as “calibrated

p-y model 1%”. The modeled 1% reinforcement single pile and the insitu group specimen with 1% reinforcement form the basis for the calculation of efficiencies.

[FIG. 13] Test Data with model results for the single pile load displacement relationship

Group efficiency factors

Fig. 14 presents a summary of load displacement relationships needed to evaluate group efficiencies. In order to calculate group efficiencies only responses for a 1% longitudinal reinforcement ratio are considered. Fig. 14 shows the simulated response of a 1% single fixed head pile, a load displacement response corresponding to 9 times the 1% single pile response as well as the experimental response of the nine pile group specimen. Based on extensive literature studies, a set of cyclic load displacement data was developed to represent the test condition if cyclic loading beyond 35 mm (1.4 in) horizontal displacement would have continued. Simulated cyclic responses are labeled “Data adjusted for cyclic degradation” in Fig. 14. Using equation (1), group efficiencies can be calculated at each desired displacement level. Hereby cyclic degradation was considered for displacement levels larger than 35 mm (1.4 in).

Group efficiencies are plotted versus lateral displacements in Fig. 15. The group efficiency is unity at very small displacements (i.e. < 2.5 mm or 0.1 in) and falls to approximately 0.8 at displacements between 4 mm < ∆ < 15 mm (between 0.16 in < ∆ < 0.6 in). It gradually rises towards 0.9 at ∆ > 25 mm (1 in) (where ∆ describes the horizontal displacement).

[FIG. 12] Sample Single Pile Curvature Profi le for a lateral top displacement of 1.9 cm and 2.5 cm. Contributions to this Fig. by Ahlberg (2008)


[FIG. 15] Group effi ciencies vs. lateral displacement

P-Multipliers

Valuable p-y relationships were obtained for the single fixed head pile and are presented in Kahlili-Therani (2009) and Lemnitzer et al. (2010). No p-y curves were obtained for the group specimen due to selective sensor malfunction during testing, hence, p-multipliers (also called f

m factor) cannot be directly derived

from the test results. Extensive literature studies were conducted to identify large scale experiments with similar boundary conditions. Priority was given to structural boundary conditions such as head fixity and pile material (i.e. reinforced concrete). P-multipliers were identified and implemented in the Frame Lab model to reproduce possible results for the pile group specimen. Fig. 16 shows a comparison of

load displacement relationships obtained using the p-multipliers from Rollins et. al. (2006), a set of modified Rollins et al. p-multipliers, and Huang et al. (2001) p-multipliers along with the specimen test data.

[FIG. 16] Comparison of load-displacement relationships obtained with various p-multipliers and test data

Rollins’ multipliers were obtained from large scale tests of concrete filled steel piles in various group arrangements under free head boundary conditions. Recommended multipliers for a 3x3 pile group with pile spacing of 3 pile diameters range between 0.8, 0.6 and 0.4 for the leading row, middle row and trailing row, respectively (Rollins et al. 2006). This particular set of p-multipliers underestimated load capacities for the author’s test starting at relatively small displacements. The corresponding curve is presented in Fig. 16 by the dashed line. By keeping the ratio between Rollin’s suggested p-multipliers for each row constant, f

m factors (p-multipliers)

were increased to investigate if the capacity can be accurately estimated with a modified set of Rollins’ values. An increase of these multipliers to 1.0, 0.725 and 0.522 provides a better fit, but still underestimates capacities at larger displacements. This fit is labeled as “Rollins with modified f

m factors” in Fig. 16.

Huang’s multipliers of 0.93, 0.7 and 0.74 for leading, middle and trailing rows respectively, yield very accurate results across the entire displacement range. The Huang test was conducted on a 3x3, bored, reinforced concrete fixed head pile group in sand with pile spacing of 3 pile diameters and is the most accurate set of original p-multipliers found in literature to match the experimental test data. This may be predominantly attributed to the similitude in pile material (i.e. reinforced concrete) and head boundary condition (i.e. fixed head

[FIG. 14] Measured and expected load displacement relationships


boundaries). Two dominant differences in pile behavior of steel free head piles and reinforced concrete fixed head piles consist of (1) the double curvature vs. single curvature bending behavior of fixed head versus free head piles, respectively and (2) the much more non-linear bending behavior associated with the use of reinforced concrete material rather than steel. Both attributes in bending behavior were present in the tested specimens.

SUMMARY AND CONCLUSIONA reinforced concrete single pile and a group of 9 piles spaced 3d on center were tested to complete structural failure. Both specimens consisted of piles with a diameter d of 0.61 m (2 ft) located in similar soil conditions. Measured results included load displacement relationship, rotation and pile curvature. Post test excavation provided insight into crack patterns and locations. Group efficiencies were calculated using test results and analytical models and were found to range between unity and 0.8. P-multipliers found in literature were applied to an analytical model that was calibrated for this specific test. P-multipliers derived by Huang et al. (2002) properly estimate the field results and provide a good set of load factors for reinforced concrete fixed head piles which may be used in design.

ACKNOWLEDGMENTSFinancial support for this research was provided by the California Department of Transportation under Research Contract No. 59A0247, which is gratefully acknowledged. This extensive project would not be possible without the collaboration of many faculty and staff on an experimental and analytical level. Special contributors include Payman Khalili-Therani and Eric Ahlberg, who contributed a significant amount of analytical and experimental results of the single pile test. Further contributors include Ertugrul Taciroglu, John W. Wallace and Jonathan P. Stewart. Acknowledgement also belongs to Anoosh Shamsabadi and Craig Whitten of Caltrans. George Cooke of GB Cooke is recognized for his assistance with construction and contract administration. Project research support also was provided by the NEES@UCLA Equipment Site as an approved shared-use project through funding from NSFinc und National Science Foundation Award CMMI-0402490. Special thanks are being expressed to the NEES@UCLA research staff:

Robert Nigbor, Steve Kang, Steve Keowen and Alberto Salamanca for their technical support and assistance in test management during specimen preparation, testing and data analysis. Special acknowledgement from the author belongs to the Educational Trust of the Deep Foundation Institute on rewarding this work with the “Young Professor Paper Competition Award 2010”. Thank you very much.

REFERENCESAhlberg, E. (2008), Interaction between soil 1. and full scale drilled shafts under cyclic lateral loads. Ph.D. Dissertation, Department. of Civil & Environmental Engineering, Univ. of California, Los Angeles, CA.

American Petroleum Institute (API), (1993). 2. Recommended Practice for Planning, Designing and Constructing fixed offshore platforms. API recommended practice 2A (RP 2A)

Brown, D.A., Reese, L.C., and O’Neill, M.W. 3. (1987). Behavior of a large scale pile group subjected to cyclic lateral loading. Journal of Geotechnical Engineering., 113(11), 1326–1343.

Cox, W.R., Reese, L.C. and Grubbs, B.R., 4. (1971). Field testing of laterally loaded piles in sand. Offshore Technica; Conference, Houston, Texas, pp.459-472.

Huang, A.B., Hsueh, C.K., O’Neill, M.W., 5. Chern, S., Chen, C., (2001). Effects of construction on laterally loaded pile groups. Journal of Geotechnical & Geoenvironmental Engineering, ASCE, 127 (5), 385-397.

Khalili-Tehrani, P. (2009). Analysis and 6. Modeling of Soil Structure Interaction in Bridge Support Structures. Ph.D. Dissertation, Department of Civil & Environmental Engineering, Univ. of California, Los Angeles, CA.

Kotthaus, M., Grundhoff, T., and Jessberger, 7. H. L. (1994). Single piles and pile rows subjected to static and dynamic lateral load. Proceeduings, Centrifuge 94, C. F. Leung et al., eds., Balkema, Rotterdam, The Netherlands, 497-5-2.

Lemnitzer, A., Khalili-Tehrani, P., Ahlberg, 8. E.R., Rha, C., Taciroglu, E., Wallace, J.W., Stewart, J.P. (2010). Nonlinear efficiency factors for bored pile group under lateral loading Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol.136 (12)


Rha, C. (2006). Analytical studies of full-scale 9. reinforced concrete shaft/column subject to cyclic lateral loads, Ph.D. dissertation, University of California, Los Angeles

Rollins, K.M., Peterson, K.T. and Weaver, T.J. 10. (1998). Lateral load behavior of full-scale pile group in clay. Journal of Geotechnical and Geoenvirnmental Engineering, ASCE, 124 (6), 468-478

Rollins, K.M. and Sparks, A. (2002). Lateral 11. resistance of full-scale pile cap with gravel backfill, Journal of Geotechnical and Geoenvirnmental Engineering, ASCE, 128 (9), 711-723

Rollins, K.M., Olsen, R.J., Egbert, J.J., Jensen, 12. D.H., Olsen, K.G. and Garrett, B.H. (2006a). Pile spacing effects on lateral pile group behavior: Load tests, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 132 (10), 1262-1271.

Rollins, K.M., Olsen, R.J., Egbert, J.J., Jensen, 13. D.H., Olsen, K.G. and Garrett, B.H. (2006b). Pile spacing effects on lateral pile group behavior: Analysis, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 132 (10), 1272-1283.

Ruesta, P.F. and Townsend, F.C. (1997). 14. Evaluation of laterally loaded pile group at Roosevelt Bridge, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 123 (12), 1153-1161.

Stewart, J.P., Taciroglu, E., Wallace, J.W., 15. Ahlberg, E.R., Lemnitzer, A., Rha, C., Khalili-Tehrani, P., Keowen, S., Nigbor, R., Salamanca, A. (2007). Full scale cyclic large deflection testing of foundation support systems for highway bridges. Part I: Drilled shaft foundations, Report No. UCLA SGEL-01, University of California, Los Angeles.

Taciroglu, E., Rha, C. and Wallace, J.W. 16. (2006). A robust macroelement model for soil-pile interaction under cyclic loads, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 132 (10), 1304-1314.


Inelastic Response of Extended Pile Shafts in Laterally Spreading Ground during Earthquakes (Student Paper Competition 2010)Arash Khosravifar, Graduate Student Researcher, University of California at Davis; Davis, CA, USA

[email protected]

Ross W. Boulanger, Professor, University of California at Davis, Davis, CA, USA

ABSTRACTThe seismic design of extended pile shafts for the combined effects of dynamic shaking and liquefaction-induced lateral spreading is investigated using nonlinear dynamic finite element analyses (NDA). Results of NDA parameter studies are used to illustrate how inertia and lateral spreading loads combine during shaking. The NDA results are used to evaluate equivalent static analysis (ESA) methods. Implications for design practice are discussed.

INTRODUCTION

Statement of Problem

Experiences from past earthquakes have shown that lateral spreading associated with liquefaction of cohesionless soils can be a cause of severe damage to bridge foundations (e.g., JGS, 1996). The mechanisms of interaction between inertial loads and lateral spreading loads for deep foundations have been investigated in numerical modeling studies, physical modeling studies, and field case history studies (e.g., Boulanger and Tokimatsu 2006). Loads from lateral spreading can become particularly significant if the spreading ground includes a relatively thick and/or strong crust layer (Fig. 1).

Large diameter, reinforced concrete (RC), extended pile shafts (or cast-in-drilled-hole piles; Fig. 2) can be an effective foundation choice in areas subject to liquefaction-induced lateral spreading hazards. Design methods for extended pile shafts affected by liquefaction are not well developed. Issues include how to combine inertia loads with lateral spreading loads for estimating the nonlinear response of the pile shafts.

The purpose of this paper is to describe the results of a numerical study on how lateral spreading and inertial loads combine to affect the performance of extended pile shafts. Results of nonlinear dynamic finite element analyses (NDA) for a range of loading conditions are used to investigate how inertial loads and lateral spreading loads combine during earthquake shaking. The results are then compared to results of a current equivalent static analysis (ESA) method (Caltrans, 2008), with the relatively poor agreement illustrating

the limitations of methods that do not combine the two loads. Improved guidance for ESA methods is currently being developed by the authors. Implications of the NDA results for design practice are briefly discussed.

[FIG. 1] Inertia and lateral spreading loads on a bridge foundation

[FIG. 2] Examples of Type 1, reinforced concrete, extended pile shafts (Caltrans, 2006)


NONLINEAR DYNAMIC FE ANALYSIS (NDA)

FE Model

A two-dimensional finite element (FE) model was created in the OpenSees FE platform. The model included 1) soil elements to model a far-field soil column, 2) nonlinear structural elements to model the reinforced concrete, Type 1, extended pile shaft, and 3) soil interface springs that connect the far-field soil elements to the pile elements (Fig. 3). The details of the model used for the analyses presented herein are described below.

The soil profile consists of a clay crust layer with varying undrained shear strengths (s

u of

20, 40, and 80 kPa (or 2.9, 5.8, and 11.6 psi)), overlying a loose liquefiable sand layer with an average SPT (N

1)60

value of 5 (DR≈33%), overlying

a dense sand layer with an average (N1)60

of 35 (D

R≈87%). The soil elements are constrained to

produce pure shear behavior in a plane strain condition. The out-of-plane thickness was chosen big enough to simulate the far field behavior without being influenced by the pile kinematics.

The pile shaft is a 2-m (6.6 ft) diameter, reinforced concrete section which extends 20 m (66 ft) deep in the ground and 10 m (33 ft) (5 m and 15 m (16 and 49 ft)) in the parametric study) above the ground surface. It is a Type 1 shaft with similar plastic moment capacities above and below the ground surface (Fig. 2). The pile to deck connection is modeled as a free connection in the baseline analysis and also as a fixed connection as one of the varying parameters in the parametric study. The superstructure mass is modeled as a lumped mass on top of the shaft. P-∆ effects were included in the model.

The pile elements are connected to the soil elements using horizontal (p-y), vertical (t-z), and tip bearing (q-z) interface springs. The importance of using soil interface springs is to allow for relative displacements between the far-field laterally spreading ground and the pile.

The model was assembled in the following sequence. Soil elements were first subjected to constant gravity force to produce hydrostatic pore water pressure, initial effective stress, and K

o conditions. The gravity force on the soil

elements was applied at an angle to simulate sloping ground conditions. Then the pile was attached to the soil elements using soil springs

followed by applying the superstructure dead load and the shaft’s self weight. The analysis proceeded by shaking the model with uniform excitation (equivalent to a rigid base motion). The input motions were applied to the bottom nodes of the soil elements.

Dynamic analyses were conducted for three different cases.

With lateral spreading and superstructure • inertia; this base case combines the two loads in a complete solution.

With lateral spreading and without super-• structure inertia; this case reflects only the lateral spreading force demand on the pile. The mass of the superstructure is set to zero.

Without liquefaction and with super-• structure inertia; this case reflects the superstructure inertial force demand on the pile in the absence of liquefaction. In this case the sand is assigned a drained condition to preclude pore pressure generation and liquefaction.

...

every 0.5 m...

PyLiq

TzLiq QzSimple

PyLiq

TzLiq

PyLiq

TzLiq

PySimple

TzSimple

CG

10m

5m

3m

12m

2m RC shaft

=0.1

2 m

Shaft section

[FIG. 3] Schematic of the FE model

These three load cases provide an approximate assessment of the relative contributions of kinematic and inertial loading to the overall demands on the structure. This assessment is only approximate because in nonlinear problems, such as studied herein, superposition of solutions does not apply. Therefore, the relative contributions of inertial and kinematic loads to the overall demands cannot be analytically decoupled. Furthermore, the magnitude of the superstructure inertia in the absence of


liquefaction is not the same as the magnitude of the inertia with liquefaction, because the triggering of liquefaction affects the complete dynamic response of the system. Nonetheless, the first two cases provide a measure of the relative effects of lateral spreading and inertial loads in the presence of liquefaction, and the third case provides a measure of the inertial loads and demands that would develop in the absence of liquefaction (which is one of the loading conditions considered in design).

Modeling in two-dimensions instead of three-dimensions has the advantage of greatly reducing the computational time, simplifying the interpretation of responses, and enabling a broader parametric study in a reasonable time frame. The inclusion of soil springs between the pile shaft and the two-dimensional soil mesh allows for large relative displacements between the soil and the pile, and thus approximates the three-dimension effects of soil being able to deform around the pile shaft. A three-dimensional model would need to explicitly simulate slip and gap formation between the pile and the soil. It is expected that the general pattern of responses obtained in 2D and 3D analyses would be similar if load transfer across the soil springs in the 2D model is similar to that in the 3D model. The current study varies the soil and soil spring properties, such that the results are expected to provide a reasonable assessment of how inertial and lateral spreading loads combine. Future analyses using 3D models will be needed to confirm that the general conclusions are the same as for the 2D models.

The components of the 2D numerical modeling approach used herein have been validated against various experiments by a number of researchers. The numerical modeling approach used for the soils and soil springs were validated against results of centrifuge model tests involving elastic structures supported on single piles and pile groups in level and inclined soil profiles that liquefied during strong shaking (e.g., Boulanger et al., 1999; Chang et al., 2006; and Chang, 2007). The modeling approach used for representing inelastic pile shafts have been validated against component tests by a number of other researchers (e.g., Hutchinson et al., 2004). Despite these previous validation efforts, it would be valuable to perform future centrifuge experiments that include inelastic structural responses in combination with the effects of liquefaction and lateral spreading.

Soil Model

The soil column is modeled in two-dimensional domain using 9-4 quadrilateral u-p elements which couple soil skeleton displacement (u) and pore water pressure (p).

Two multiple-yield-surface constitutive models for soil were used (Elgamal et al., 2002): Pressure-Dependent-Multi-Yield02 (PDMY02) material for sand layers and Pressure-Independent-Multi-

Yield (PIMY) material for clay layers.

The yield-criteria for the sand model are described by a number of cone shape surfaces as shown in Fig. 4. The model has a non-associative flow rule and develops volumetric dilation and contraction due to shear deformation. The hardening rule is purely deviatoric kinematic. The stress-path and stress-strain response of the model, as illustrated in Fig. 5, approximates dilation and contraction of granular material and the subsequent pore water pressure generation in cyclic loading very well, without the locking problem that often occurs in constitutive models.

[FIG. 4] Multiple yield surface model for sand (Elgamal et al., 2002)

[FIG. 5] Illustration of sand model response: (a) stress path, (b) stress-strain (Elgamal et al., 2002)


The model parameters were calibrated to produce the cyclic resistance ratio (CRR) corresponding to the SPT-based liquefaction correlation of Idriss and Boulanger (2008) at the selected (N

1)60

values for the idealized soil profile. The CSR versus number of uniform loading cycles to cause 3% shear strain in cyclic undrained direct simple shear (DSS) loading is shown in Fig. 6 for SPT (N

1)60

values of 5, 15, and 25; the present analyses are for (N

1)60

=5 in the loose sand layer, while the other calibrations are being used in additional studies. The calibrated model was also checked to ensure reasonable shear strain accumulation at each cycle after liquefaction has triggered.

The model parameters used for the baseline case are summarized in Table 1. The meaning of each parameter and how it affects the model

response is given in Yang et al. (2003; 2008) and is not repeated here for brevity.

1 10 100

Number of uniform cycles

0

0.2

0.4

0.6

Shea

r str

ess

ratio

(τ/σ

′ vo) σ′vo = 100 KPa, Ko = 0.5

3% shear strain(N1)60 = 25

(N1)60 = 15

(N1)60 = 5

[FIG. 6] Simulated relationship between CRR and number of equivalent uniform loading cycles in undrained cyclic DSS loading using PDMY02 model

[TABLE 1] Soil model parameters

Model parametersSand

(N1)60

=5

Sand

(N1)60

=35

Clay

Su = 40 kPa

Material type PDMY02 PDMY02 PIMY

Relative density,DR* 33% 87%

Density,ρ 1.94 ton/m3 2.06 ton/m3 1.6 ton/m3

Reference pressure,p’r

100 kPa 100 kPa 100 kPa

Shear wave velocity,Vs1

* 141 m/s 210 m/s

Shear modulus,Gmax,1

* 38.3 MPa 91.3 MPa 28 MPa

Octahedral shear modulus,Gmax,1,oct

46.9 MPa 111.9 MPa 28 MPa

Maximum shear strain,ϒmax,r

0.1 0.1 0.1

Bulk modulus,Br

125.1 MPa 298.3 MPa 140 MPa

Pressure dependent coeff.,d 0.5 0.5 0.0

Friction angle,ϕDSS

* 30° 45° 0.0°

Octahedral friction angle, ϕoct

25.4° 42.2° 0.0°

Phase transformation angle, ϕPT

20° 32.2°

Contarction coeff.,c1

0.06 0.001


5.0 0.5


0.2 0.0

Dilation coeff.,d1

0.15 0.4

Dilation coeff.,d2

3.0 3.0

Dilation coeff.,d3

0.0 0.0

liq1

1.0 1.0

liq2

0.0 0.0

Number of Yield Surfaces,NYS 20 20 20

Cohesion,c 0.1 kPa 0.1 kPa 34.6 kPa

*These parameters are used for guiding model calibration, but are not input parameters for the soil model. In particular, the friction angles, shear strengths, and shear moduli must be converted from the stress–strain space used in conventional practice to the octahedral stress-strain space used in the constitutive model.


Reinforced Concrete PileThe reinforced concrete pile is modeled using flexibility-based nonlinear beam-column elements. These elements use a distributed plasticity formulation which allows the formation of a plastic hinge at any position.

The pile cross-section is discretized into a fiber section as illustrated in Fig. 7. The fiber sections include the confined concrete (core), unconfined concrete (cover), and the longitudinal steel bars.

The pile section used herein is a 2 m (6.6 ft) diameter section. The reinforced concrete pile is modeled using flexibility-based non-linear beam-column elements. These elements used a distributed plasticity formulation which allows the formation of a plastic hinge at any position. The expected compressive strength (f’

c) of 44.8 MPa (6,500 psi) was used to define

confined and unconfined concrete stress-strain behavior based on Mander et al. (1988). Longitudinal steel bars were modeled with yield strength of 475 MPa (69 ksi), elastic modulus of 200 GPa (29 x 106 psi), and 1% strain hardening ratio. The longitudinal steel area ratio varied from 1% to 3%.

2 m

Steel rebars63 #11

Unconfined concreteConfined concrete

[FIG. 7] Fiber section used for the pile shaft

The moment-curvature responses of a 2 m (6.6 ft) diameter section with various longitudinal steel area ratios and various axial dead load ratios are shown in Fig. 8. Different damage stages are identified on the figure to assess the progress of plastic hinge formation. The time when the first longitudinal steel bar yields is considered as the yielding point of the shaft. Capacity of the shaft is defined as the points corresponding to confined concrete compressive crushing or longitudinal steel bar snapping, whichever happens first.

The local curvature computed by the FE model is affected by local softening and hence is mesh-dependent. The FE computed rotation in the plastic hinge was shown to be mesh-independent, and thus this quantity was recorded and divided by the equivalent plastic hinge length proposed

by Chai and Hutchinson (2002) to arrive at the estimated curvature in the plastic hinge zone.

0 0.02 0.04 0.06 0.08

Curvature (rad/m)

0

10000

20000

30000

40000

50000

Mom

ent (

KN

-m)

First rebar yieldingUnconfined concrete spallingConfined concrete crushingRebar snapping

P/(f ′cAg)=0.05, ρs=0.01

P/(f ′cAg)=0.10, ρs=0.01

P/(f ′cAg)=0.05, ρs=0.02

P/(f ′cAg)=0.10, ρs=0.02

P/(f ′cAg)=0.05, ρs=0.03

P/(f ′cAg)=0.10, ρs=0.03

[FIG. 8] Moment-curvature behavior for the pile section

Soil SpringsThree types of soil interface springs were used to connect the far-field soil elements to the pile elements: horizontal (p-y), vertical (t-z), and tip bearing (q-z) springs. The soil springs are made from elastic, plastic, gap and dashpot components which make them a good choice for modeling lateral spreading where large relative displacements between soil and pile are expected to occur. The p-y and t-z springs were represented with PYliq1 and TZliq1 material models which update their capacity and stiffness based on the adjacent soil elements’ mean effective stress. These soil springs simulate the transient loss of stiffness and strength due to excess pore pressure generation in the far-field soil. A typical behavior of a PYliq spring is shown in Fig. 9.

The soil springs parameters (p-y, t-z, and q-z) were selected based on API (1987) recommendations. The stiffness of p-y elements at larger depths were modified using the relationship in Boulanger et al. (1999). The soil spring parameters in the base line analysis were as follows. In the clay layer with S

u=40 kPa

(5.8 psi), pult

=130 to 209 kPa (19 to 30 psi) and y

50=0.05 m (2.0 in). In the liquefying layer with

(N1)60

=5, pult

=165 to 495 kPa (24 to 72 psi) and y

50=0.01 m (0.4 in), and in the dense sand layer


with (N1)60

=35, pult

=538 to 10570 kPa (78 to 1533 psi) and y

50=0.01 to 0.03 m (0.4 to 1.2 in).

All soil springs had drag coefficient Cd=0.3.

-20 -10 0 10 20y /y50

Liquefied

-20 -10 0 10 20y /y50

-1

-0.5

0

0.5

1

p /p

ult

Not liquefied

(a) (b)

[FIG.9] p-y element behavior: (a) without liquefaction in the far-fi eld, and (b) with liquefaction in far-fi eld soil

Ground Motions

Forty acceleration time series were used as input to the dynamic analyses. These records were selected by Professor J. Baker (personal communication) so their median response spectra matched the median and log standard deviation values predicted by Boore and Atkinson’s (2008) ground motion prediction model for an earthquake having the following characteristics:

Magnitude = 7.5•

Source to site distance = 10 km (6.2 miles)•

Site shear wave velocity (V• S30

) = 760 m/s (2,500 ft/s)

Earthquake mechanism = strike slip•

The acceleration response spectra for the 40 input motions and the median spectra are shown on Fig. 10.

0.01 0.1 1 10

Period (s)

0.01

0.1

1

10

PSa

- Bas

e (g

)

[Fig. 10] Acceleration response spectra (5% damped) for the 40 input motions

Ground surface displacements were substantially greater with liquefaction than without liquefaction, as expected. For the non-liquified cases, the median values for the peak ground surface displacements during shaking varied from about 0.6 m to about 3 m (2 ft to 10 ft) for base motion PGA’s ranging from 0.1 to 1 g, respectively. For the [nIG.liquefied cases, the median ground surface displacements varied from less than about 0.01 m to about 0.3 m (0.4 to 12.0 in) for base motion PGA’s ranging from 0.1 to 1 g, respectively.

Example of Dynamic Response

An example of the dynamic response of an extended pile shaft is shown in Fig. 11. For this example, the crust’s undrained shear strength was 40 kPa (5.8 psi) and the shaft had a longitudinal steel area ratio of 2% and carried an axial dead load of 0.10f’

cA

g. The input motion

was recorded at the Gilroy Array #1 station with a peak ground acceleration of 0.43 g during the 1989 Loma Prieta earthquake.

The time at which the peak deck displacement occurs is marked by the red line on the time histories shown in Fig. 11. At this time, the relative displacement between the pile and far-field soil is 0.08 m (3 in) (Fig. 11a), 60% of the maximum passive earth pressure from the crust has been mobilized (Fig. 11b), the superstructure inertia force is about 58% of its peak value (which occurred slightly earlier in shaking; Fig. 11c), and the pile bending moment was near its peak value (Fig. 11d). The full passive crust load developed slightly later in shaking, whereas the ground surface displacement continued to increase throughout shaking toward its final value of about 1 m (3.3 ft). The transient dips in the pore water pressure ratio (r

u) in the middle of the loose

sand layer develop as the sand incrementally dilates under each cycle of applied shear stresses (Fig. 11e); at the time of peak pile head displacement, the r

u has transiently dipped

as the crust neared its greatest down-slope movement during that acceleration cycle. At that time, the liquefied layer is exerting down-slope forces on the upper part of the shaft and upslope forces on the lower part of the shaft, such that the net force from the liquefied layer is negligible compared to the crust force.

Acceleration time histories for the input motion, ground surface, and superstructure are also shown in Figs. 11f to 11h, with their corresponding linear-elastic acceleration


0 5 10 15 20 25 30

Time (sec)

-4

-2

0

2

4

Bas

eac

cel.

(m/s

2 )

-4

-2

0

2

4

Gro

und

surf

ace

acce

l. (m

/s2 )

-4

-2

0

2

4

Supe

rstr

uctu

re

acce

l. (m

/s2 )

-0.4

0

0.4

0.8

1.2

r u

in li

quef

ied

laye

r

-2000

-1000

0

1000

2000

Supe

rstr

uctu

re

iner

tia (K

N)

-1000

0

1000

2000

Cru

st lo

ad(K

N)

-0.4

0

0.4

0.8

1.2

Dis

plac

emen

t (m

) Ground surfaceSuperstructurePile at ground surface

(a)

(b)

(c)

(e)

(f)

(g)

(h)

-40000

-20000

0

20000

40000

Ben

ding

mom

ent

at P

H (K

N-m

)

(d)

0.08 m

[FIG. 11] Example dynamic response of an extended pile shaft


response spectra (ARS) shown in Fig. 12. The ARS for the corresponding non-liquefied case are also shown on Fig. 12 for comparison. The occurrence of liquefaction decreased the peak ground acceleration (PGA) by 7% and reduced spectral accelerations at periods less than about 1 second, but increased the ground surface’s spectral accelerations by about 15% for periods of about 3 seconds. The superstructure’s natural period (or period of strongest amplification) was about 2.5 seconds in the absence of liquefaction, and this increased to about 3.0 seconds with the occurrence of liquefaction.

Profiles of pile bending moments, shear forces, and displacements at different times are shown in Fig. 13. A plastic hinge formed in the pile shaft at a depth of 9.5 m (31 ft), which corresponds to just below the bottom of the liquefied sand layer as expected.

Results of Parametric Study

The results of a set of parametric NDA analyses are presented to illustrate the combined effects of inertia and lateral spreading. The parametric analysis included various reinforced concrete parameters, crust strengths, soil conditions, pile heights, pile to deck connections, and different scaling methods on the ground motions (Table 2). All NDA analyses were conducted for all 40 ground motions and repeated for the three cases described previously.

The baseline analysis was performed on a 2 m (6.6 ft) diameter reinforced concrete shaft with longitudinal steel area ratio of 2% carrying axial dead load of 0.10f’

cA

g. The soil conditions in the

baseline analysis consisted of 5 m (16.4 ft) clay crust with undrained shear strength of 40 kPa (5.8 psi), overlying 3 m (10 ft) loose sand layer with (N

1)60

blow count of 5, overlying 12 m (39 ft) dense sand layer with (N

1)60

blow count of 35 at the bottom. The ground slope in the baseline analysis was 10% and the pile was extended 10 m (33 ft) above the ground surface while freely connected to the deck. The excitations for the baseline analysis were 40 non-scaled ground motions.

[TABLE 2] Parametric Study

Varying parameter Parameter value

Longitudinal steel area ratio 1%, 2%, and 3%

Axial dead load ratio 0.05 and 0.10 f’cA

g

Crust undrained shear strength (Su) 20, 40, and 80 kPa

Layer thickness 7 m clay crust overlying 1 m liquefying sand layer

5 m clay crust overlying 3 m liquefying sand layer 3 m clay crust overlying 5 m liquefying sand layer

Ground slope 5%, 10%, and 20%

Pile height above ground 5, 10, and 15 m

Pile to deck connection Free and fixed

Input motions (40 ground motions) Unscaled

PGAs scaled to 0.6 g

PGAs doubled

Scaled to Sa(T=3s) = 0.10g

Scaled to Sa(T=3s) = 0.15g

[FIG. 12] Acceleration response spectra (5% damped) for the input motion, superstructure motion, and ground surface motion

0.01 0.1 1 10

Period (s)

0.01

0.1

1

10

PSa

(g)

Input motionGround surface (Nonliquefied)Ground surface (Liquefied)Superstructure (Nonliquefied)Superstructure (Liquefied)


The maximum superstructure (deck) displacements for the three cases are summarized and compared in Fig. 14. The horizontal axis in each graph is the maximum deck displacement for the full solution with lateral spreading and superstructure inertia (the base case). The y-axes in Figs. 14a and 14b are the maximum deck displacement considering lateral spreading alone or inertia alone, respectively. Note that the magnitude of the superstructure inertia in the absence of liquefaction (y axis in Fig. 14b) is not the same as the magnitude of the inertia that occurs in combination with lateral spreading (x axes for Fig. 14a-d) because the triggering of liquefaction affects the dynamic response of the system. The NDA results plot below the 1:1 line in Figs. 14a and 14b, indicating that either case underestimates the displacement demands for the full solutions. The y-axis in Fig. 14c is the maximum of the deck displacements computed using either inertia alone or lateral spreading alone; Again the NDA results plot below the 1:1 line indicating that enveloping of results for inertia alone or lateral spreading alone are not sufficient for estimating the full demand.

The y-axis in Fig. 14d is the sum of the demands from the inertia alone and lateral spreading alone cases; The NDA results now plot closer to the 1:1 line indicating that both the inertial and lateral spreading components are important to the performance of the extended pile shafts. Similar observations were drawn from comparisons of the local curvature ductility in

the plastic hinges for the three different loading cases. Note that these analyses included pile shafts that develop a range of local ductility in their plastic hinges, indicating that the above conclusions are valid for both elastic and inelastic responses of extended pile shafts.

EQUIVALENT STATIC ANALYSIS (ESA)

Caltrans Guideline

Caltrans (2006; and 2008) describes the

use of equivalent static analyses (ESA) for ordinary bridges with the effects of liquefaction and lateral spreading. ESA includes estimating the superstructure displacement demand from the design ARS and using pushover analyses for evaluating performance at different levels of displacement demand. Caltrans (2008) describes three steps in designing a Type 1, extended pile shaft.

Non-liquefaction case: Perform a push-over 1. analysis without liquefaction and check the local and global demands and capacities. (Fig. 15a). Use regular p-y and t-z springs.

Liquefaction case: Perform a push-over 2. analysis with liquefaction but no lateral spreading (Fig. 15b). Soften the p-y and t-z springs for the effects of liquefaction. The superstructure displacement demand is assumed equal to that used in the non-liquefaction case, although it is not explicitly stated in the guidelines.

Lateral Spreading case: Use softened p-y 3. springs. Apply 100% of the passive earth pressure from the crust (or lateral spreading force, LSF) and determine the maximum moment along the pile shaft. Design so the maximum moment from LSF alone (M

LS)

is less than 20% of the plastic moment capacity (M

P) of the shaft (Fig. 15c).

Comparison of Caltrans ESA with NDA

The comparison of Caltrans ESA results and NDA results is illustrated for one set of

CG

Dense sand(N1)60 = 35

Loose sand(N1)60 = 5

Clay crustSU = 40 KPa

-40000 0 40000

Moment (KN-m)

maxmin

Ground surface

-5 m

-8 m

-20 m

+10 m-4000 0 4000

Shear (KN)

at max deck dispat max inertiaat max crust load

-0.4 0 0.4 0.8 1.2

Disp (m)

pile max/minsoil max/min

[FIG. 13] Profi le of bending moments, shear forces and displacements in the pile shaft at different times


Tz Qz

p-y

Crust

Loose sand

Dense sand

p-y

p-y

p-y

V

Tz Qz

p-y

Crust

Loose sand

Dense sand

p-y

p-y

p-y

V

P-multipliers

Moment

Tz Qz

p-y

CG

Crust

Loose sand

Dense sand

p-y

p-y

p-y

LSF

P-multipliers MLS< 20% Mp

(a) (b) (c)

[FIG 15] Illustration of Caltrans (2008) ESA method

0.01 0.1 1 10

Combination of inertia and lateral spreading

0.01

0.1

1

10

(a) L

ater

al s

prea

ding

alo

ne

Max Δdeck (m)

"Collapse" cases plotted at Δ=6 m

0.01 0.1 1 10


0.01

0.1

1

10

(c) M

ax o

f ine

rtia

alo

ne a

ndla

tera

l spr

eadi

ng a

lone

Base ZPA (g)0 to 0.20.2 to 0.60.6 to 3.4

Max Δdeck (m)

0.01 0.1 1 10


0.01

0.1

1

10

(b) I

nert

ia a

lone

Max Δdeck (m)

0.01 0.1 1 10


0.01

0.1

1

10

(d) S

umm

atio

n of

iner

tia a

lone

and

late

ral s

prea

ding

alo

ne

Max Δdeck (m)

[FIG. 14] Comparison of maximum deck displacements for different loading cases [cases of numerical “collapse” are plotted at max deck displacements of 6 m]


parametric analyses. This set of analyses used an undrained shear strength of 40 kPa (5.8 psi) for the crust and a longitudinal steel area ratio of 2% and axial dead load of 0.10f’

cA

g for the

shaft. Global deck displacement demands and local curvature demands at plastic hinges were estimated for the liquefaction case based on Caltrans (2008) ESA guidelines. The soil springs in the liquefying layer were softened using a p-multiplier of 0.1 based on the recommendations of Brandenberg et al. (2007). Parametric analyses using p-multipliers of 0.01 to 0.2 had an insignificant effect on the overall results of the ESA and NDA analyses for the range of conditions covered herein. For the ESA method, the median normalized ARS shape for the 40 motions recorded at ground surface was used for estimating displacement demands for different levels of ground surface PGA. Consequently, the ESA results shown in Figs. 16a and 16b show a smooth increase of demand versus ground surface PGA, whereas the NDA results in these same figures exhibit scatter due to variable ground motion characteristics.

The results in Figs. 16a and 16b show that the ESA for the liquefaction case (step 2) significantly under-predict the demands imposed during dynamic shaking with liquefaction (i.e., the NDA results). The large majority of the NDA results would still indicate acceptable performance, however, in that the deck displacements rarely exceed the Caltrans

0 0.2 0.4 0.6 0.8 1

Ground surface ZPA (g)

0.01

0.1

1

Max

dec

k di

sp (m

)

P/(f ′cAg)=0.10, ρs=0.02, Crust Su=40 KPaNonlinear Dynamic AnalysisCaltrans ESA using Median Response Spectra

PD criterion limit (0.4m)

(a)

0 0.2 0.4 0.6 0.8 1

Ground surface ZPA (g)

0.0001

0.001

0.01

0.1

Max

cur

vatu

re in

the

plas

tic h

inge

(rad

/m)

Collapse limit (curv 0.04)Allowable ductility 3.5 (curv 0.0088)Yielding limit (curv 0.0025)

(b)

[FIG. 16] Comparison of Caltrans (2008) ESA method with NDA results for the liquefaction case

criterion that the moment due to P -∆ be less than 20% of M

p (red line in Fig. 16a) and the

local curvature demands rarely exceed the allowable local curvature ductility of 3.5 (green line in Fig. 16b).

CG

Dense sand(N1)60 = 35

Loose sand(N1)60 = 5

Clay crustSU = 40 KPa

0 10000 20000

Moment (KN-m)

Ground surface

-5 m

-8 m

-20 m

+10 m

MLS = 11688 KN-m>20% Mp = 6400 KN-m

Moment due to LS alone

LSF

[FIG. 17] The third step in Caltrans (2008) ESA method for lateral spreading load alone

In contrast, the results of the third step in the Caltrans (2008) ESA guidance would indicate that the pile shaft is unacceptable in all these cases because it fails to meet the criterion that the shaft bending moment due to the lateral spreading force alone be less than 20% of the section’s plastic moment capacity (Fig. 17).


The fact that the extended pile shaft met the deck displacement and local curvature ductility criteria suggests that this third criterion is not consistent with the expected performance of the extended pile shaft for the set of conditions covered by this example.

Additional ESA-NDA comparisons are in progress and being used to develop improved ESA guidelines for combining lateral spreading and inertial loads.

CONCLUSIONResults of parametric studies using nonlinear dynamic finite element analyses show that coupling of inertial and lateral spreading demands during earthquake shaking can be the governing load case for extended pile shafts and that this is not enveloped by analyzing the inertial and lateral spreading load cases separately.

Comparisons of the Caltrans (2008) ESA method with the NDA results for extended pile shafts under a limited set of conditions showed that:

The Caltrans ESA method underestimates • the peak displacement and ductility demands that develop in extended pile shafts that are affected by liquefaction and lateral spreading during earthquakes; and

The criterion that the bending moment due • to lateral spreading forces alone (M

LS) be less

than 20% of the section’s plastic moment capacity (M

P) did not correlate well with

either good or poor performance, and was often overly conservative.

Additional parametric analyses for extended pile shaft systems that cover a broader range of structural and soil conditions are in progress. The results of those analyses are being used in development of improved ESA procedures.

ACKNOWLEDGMENTFunding was provided by the Pacific Earthquake Engineering Research (PEER) Center. The authors also appreciate the assistance of Professor Scott J. Brandenberg with components of the OpenSees models and Professor Jack Baker for providing the ground motion time series.

REFERENCESAmerican Petroleum Institute (API), 1987. 1. Recommended Practice for Planning, Designing and Construction Fixed Offshore Platforms. API Recommended Practice 2A (RP-2A), Washington, D.C., 17th edition.

Boore, D., and Atkinson, G., 2008. Ground-2. motion prediction equations for the average horizontal component of PGA, PGV, and 5%-damped PSA at spectral periods between 0.01 s and 10.0 s. Earthquake Engineering Research Institute, Earthquake Spectra, 24(1), pp 99-138.

Boulanger, R. W., Curras, C. J., Kutter, B. L., 3. Wilson, D. W., and Abghari, A., 1999. Seismic soil-pile-structure interaction experiments and analyses. Journal of Geotechnical and Geoenvironmental. Engineering, ASCE, 126(9), 750-759

Boulanger, R. W., and Tokimatsu, K., 2006. 4. Geotechnical Special Publication No. 145: Seismic Performance and Simulation of Pile Foundations in Liquefied and Laterally Spreading Ground. ASCE Press, Reston, VA, 321 p.

Brandenberg, S. J., Boulanger, R. W., Kutter, 5. B. L., and Chang, D., 2007. Static Pushover Analyses of Pile Groups in Liquefied and Laterally Spreading Ground in Centrifuge Tests. Journal of Geotechnical and Geoenvir. Engineering, ASCE, 133(9), 1055-1066

Caltrans, 2006. Seismic Design Criteria. 6. California Department of Transportation.

Caltrans, 2008. Soil Liquefaction and Lateral 7. Spreading Analysis Guidelines, Memos to Designers MTD 20-15, July. California Department of Transportation.

Chai, Y.H. and Hutchinson, T.C., 2002. 8. Flexural strength and ductility characteristics of reinforced concrete bridge piles - experimental investigation. Journal of Structural Engineering, ASCE, 128(5), 595-602.

Chang, D., 2007. Seismic Performance of 9. Pile-Supported-Structures in Liquefied and Laterally Spreading Ground. PhD thesis, Dept. of Civil and Env. Engineering, University of California, Davis.

Chang, D., Boulanger, R. W., Brandenberg, 10. S. J., and Kutter, B. L., 2006. Dynamic Analysis of Soil-Pile-Structure Interaction in Laterally Spreading Ground during Earthquake Shaking. Geotechnical Special Publication No. 145, ASCE Press, Reston, VA, pp. 218-229

Elgamal, A., Yang, Z., and Parra, E., 2002. 11. Computational modeling of cyclic mobility and post-liquefaction site response. Journal of Soil Dynamics and Earthquake Engineering, Vol. 22, 259-271.


Hutchinson, T. C., Chai, Y. H., Boulanger, R. 12. W., and Idriss, I. M., 2004. Inelastic Seismic Response of Extended Pile-Shaft-Supported Bridge Structures. Earthquake Spectra, Earthquake Engineering Research Institute, 20(4), 1057-1080.

Idriss, I. M., and Boulanger, R. W., 2008. Soil 13. liquefaction during earthquakes, Monograph MNO-12, Earthquake Engineering Research Institute, Oakland, CA, 261 pp.

Japanese Geotechnical Society (JGS), 1996. 14. Special Issue on Geotechnical Aspects of the January 17, 1995, Hyogoken-Nambu Earthquake, Soils and Foundations, Tokyo.

Mander, J. B., Priestley, M. J. N., and Park, 15. R., 1988. Observed stress-strain behavior of confined concrete. Journal of Structural Engineering, ASCE, 114(8), 1827-1849.

OpenSees. Open System for Earthquake 16. Engineering Simulation. http://opensees.berkeley.edu. Pacific Earthquake Engineering Research Center, University of California, Berkeley.

Yang, Z., Elgamal, A., and Parra, E., 2003. 17. Computational Model for Cyclic Mobility and Associated Shear Deformation. Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 129(12), 1119-1127.

Yang, Z., Lu, J., and Elgamal, A., 2008. 18. OpenSees Soil Models and Solid-Fluid Fully Coupled Elements: User’s Manual. Department of Structural Engineering, University of California, San Diego.


Thermal Integrity Profi ling of Drilled ShaftsGray Mullins, Ph.D., P.E., Professor, Department of Civil and Environmental Engineering, University of

South Florida, Tampa, Florida, USA; [email protected]

ABSTRACTThe construction of drilled shafts as well as other cast in place foundation alternatives relies heavily on good practices from the contractor, engineer, and inspector in order to produce a quality foundation element. As many of the installation methods involve blind concreting processes, it is difficult to be certain of an intact concrete mass of the intended dimensions. A new integrity test has been developed that provides additional insight into the integrity of drilled shaft concrete. By measuring the temperature throughout the shaft via standard access tubes, the measured temperature profile can be compared to the normal signature associated with a shaft of the specified size. This article provides an overview of the new method development, test procedure, analysis, and results.

INTRODUCTIONDrilled shafts are large-diameter cast-in-place concrete structures that can develop enormous axial and lateral capacity and consequently are the foundation of choice for many large bridges subject to extreme event loads such as vessel collisions. Drilled shafts are constructed throughout the world often using the slurry method as a means to stabilize the excavation. This means that both excavation and concreting are blind processes which increase the chances of unwittingly producing defects in the shaft.

State-of-the-art methods for evaluating the integrity of drilled shaft concrete vary where no one approach is effective in providing a full picture of the actual state of the concrete (Hertlein, 2001). Some methods are better at evaluating the core of the shaft, whereas others can best detect problems more proximal to logging/access tubes. Ideally, but not practical, a combination of all current test technologies could perhaps identify most forms of anomalies. As this is often not cost-effective, it is therefore desirable to explore other technologies that could be extended to integrity testing that may provide a more comprehensive assessment. One such concept makes use of the heat of hydration of curing concrete and temperature measurements within the shaft to assess whether or not anomalies have been formed. This type of integrity evaluation is referred to herein as Thermal Integrity Profiling or TIP.

Prior to the implementation of the new approach, heat of hydration was only viewed as an undesired side effect which has been long recognized for its potentially harmful consequences. Of the numerous case studies,

the most famous is perhaps the Hoover Dam project constructed during the depression from 1932 to 1935 where over 4 million cubic meters (5.2 million cu yd) of concrete were used. At that time it was understood that staged construction and internal cooling systems would be required to help control elevated temperatures. Therein, the primary concern was concrete cracking from differential temperature and the associated tensile stresses. Without these considerations, temperature dissipation was estimated to take over 100 years and temperature-induced cracking would have severely compromised its structural integrity and its ability to prevent leakage (US Dept of Interior, 2004).

With regard to drilled shafts, these foundation elements have been routinely constructed without considering mass concrete effects and the possible long-term implications of the concrete integrity. Such considerations address the high internal temperatures that can be generated during the concrete hydration/curing phase which can be detrimental to the shaft durability and/or integrity in two ways: (1) short-term differential temperature-induced stresses that crack the concrete and (2) long-term degradation via delayed ettringite formation (Whitfield, 2006).

Understanding the parameters that affect the temperature rise in curing concrete has a two-fold benefit to the concrete and drilled shaft industry: (1) the ability to better predict the occurrence of mass concrete conditions in all concrete structures, and (2) the use of temperature generation and its diffusion to the surrounding environment to predict normal drilled shaft internal temperature distributions. This paper focuses on the latter although the


predictive computations and field measurement methods serve both needs.

BACKGROUNDVarious physical, chemical, and molecular principles are combined in the concept of thermal integrity profiling of drilled shafts that address heat production in the concrete, diffusion of the heat into the soil, and the resulting temperature signature produced by a properly shaped drilled shaft (Mullins et al., 2004, 2005, 2007, and 2009; Kranc and Mullins, 2007). At various stages of the curing process these principles have more prominent effects; heat production tends to dominate the resulting temperature in the early stages whereas the surrounding dissipation process controls later on.

Heat Production. The quantity of heat and rate of heat production are directly linked to the concrete mix design and the chemical constituents of the cementitious materials. These materials are generally comprised of cement and flyash or slag. Each material produces heat when hydrating, the total magnitude of which is dependent on the cementitious fraction p (by weight) with respect

to total cementitious material. The total heat, H

u, and the rate of production can be determined

from equations (1) – (5) where H is in units of kJ/kg (Schindler and Folliard, 2005).

FAFAslagcemcemu phppHH ++= 461 (1)

Where the energy per kilogram of slag is directly given to be 461 kJ/kg (198 BTU/lb), the cement and flyash energy production can be determined using equations (2) and (3), respectively.

MgO

FreeCaOSOAFC

ACSCSCcem

pppp

pppH

8501186624420

866260500

34

323

++++++=

(2)

FACaOFA ph 1800= (3)

Both equations (2) and (3) require precise knowledge of the chemical composition of the cement and flyash in the form of the weight fraction of the various chemical compounds, p

i. These are usually available

from the concrete supplier and flyash source (municipal power plant).

[TABLE 1] Effect of slag and fl yash in shaft mixes on energy and duration (Eqns 1-7).

ConcreteConstituents

WSDOT 4000P(Flyash)

WSDOT 4000P(Slag)

FDOT Class IV 4000 (Flyash)

FDOT Class IV 4000 (Slag)

Cement, kg (%) 276.7 (85%) 272.2 (77%) 226.8 (66%) 122.5 (39.7%)

MgO, % 0.83 1 0.7 0.9

C2S, % 13 14 10 9

C3A, % 7.1 5 7 7

C3S, % 58 60 62 63

SO3, % 2.8 2.7 2.9 2.9

C4AF, % 11.2 10 12 11.3

Blaine, m2/kg 387 411 391 386

Flyash, kg (%) 49.9 (15%) - 114.8 (34%) -

SO3, % 1 - 1.8 -

CaO, % 15.1 - 5.2 -

Slag, kg (%) - 81.3 (23) - 186.0 (60.3%)

w/cm 0.37 0.41 0.52 0.41

Energy (kJ/kg) 76.2 87.7 57.5 53.8

α 0.753 0.769 0.921 0.881

β 0.630 0.699 0.699 0.435

τ (hrs) 19.4 26.3 17.4 54.5


Schindler and Folliard (2005) further provided means to compute rate of heat production whereby curve fitting algorithms were applied to extensive laboratory studies again based on the weight fraction of the various cementitious constituents. The degree of hydration at time, t

e, can be determined using equation (4).

−=β

τααe

ue tt exp)( (4)

When α equals 1.0 all hydration energy has been developed from equation (1). The parameters α

u,

β, and τ are determined again by cementitious constituent fractions, p

i, shown in equations (5)

– (7), respectively, as well as the water cement ratio, w/cm.

0.13.05.0/194.0

/031.1 ≤+++

= SLAGFAu ppcmwcmwα

(5)

)647.0exp(

4.181558.0

535.0146.0227.0

3

33

SLAGSO

ACSC

pp

Blainepp

−⋅⋅

⋅⋅= −β

(6)

)5.9187.2exp(

78.66758.0

804.0154.0401.0

3

33

CaOFA

FASLAGSO

ACSC

pppp

Blainepp

−

−

−−−

⋅⋅+⋅⋅

⋅⋅⋅=τ (7)

For typical shaft mixes with moderate flyash percentages (15%) τ usually is around 18-24 meaning that all energy has been expended in roughly 18 - 24 hours. High slag content mixes (e.g. 60% replacement) usually take upwards of 50 hours. Mixes with no flyash or slag are usually expended in about 15 hours. Table 1 shows the effect of using flyash or slag on approved shaft mixes from both Washington and Florida DOTs.

Heat Diffusion. Just as important as the energy production is the mechanism by which the heat is dissipated into the surrounding environment. Although the thermal integrity approach can be applied to all concrete structural elements, it is most commonly used for drilled shafts wherein the surrounding environment is largely dominated by a soil structure or geo-material.

Heat flow in soils involves simultaneous mechanisms of conduction, convection, and radiation of which conduction overwhelmingly dominates the heat transport. Conductive heat flow in soils is analogous to fluid or electrical systems. The thermal conductivity, λ, is defined

as the heat flow passing through a unit area, A, given a unit temperature gradient, ΔT/L, equation (8).

LTAq

/Δ⋅=λ (8)

This value can be estimated by the geometric mean of the thermal conductivity of the individual matrix components: solids, water, and air. Thermal conductivity of soil minerals range from 2 to 8 W/m-C for clay to quartz, respectively. Although dependent on temperature and relative humidity, water is roughly 0.5 W/m-C and air, 0.03 W/m-C. For a saturated soil, the thermal conductivity can be determined using equation (9) where n represents the volumetric fraction of water (Johansen, 1975; Duarte et al., 2006).

nw

nssat λλλ )1( −= (9)

Likewise, the thermal conductivity of the solids, λs, is related to the fraction of quartz or sand, q, in the soil and is determined using equation (10). The subscript “o” denotes other soil minerals.

)1( qo

qqs

−= λλλ (10)

Not surprisingly, there is a strong correlation between thermal conductivity and mechanical properties as close contact / dense packing of the soil particles aides in transmitting heat by means of thermo-elastic waves. Farouki (1966) provided translation of this concept from Debye (1914) wherein heat flow through non-metallic crystalline solids occurs when warmer atoms vibrate more intensely than adjacent cooler atoms which in turn propagate waves by way of atom to atom contact at a characteristic speed. As a result, the thermal conductivity can be related to the compression wave velocity for a given material. The strength of the bonds between atoms affects this speed which is also dependent on the heat capacity of the material.

The heat capacity of the soil can be determined based on the volumetric fraction of solids, water, and air wherein the heat capacity of each component is defined as the heat required to raise the temperature of a unit volume of material one degree C. The heat capacity is actually the product of the mass specific heat, c, and the dry density of the soil, ρ. Farouki (1981) and Duarte et al. (2006) define the specific heat of a volume of soil by introducing Xi as the volumetric fraction of each component,


equation (11) can be used to determine the effective specific heat of the soil matrix where C

S, C

W, and C

A represent the heat capacity of the

solids, water, and air, respectively.

AAWWSS CXCXCXC ++= (11)

In essence, two almost conflicting parameters affect heat dissipation into the surrounding soils: the ability to conduct heat (λ) and the reluctance of the soil to be heated (C). The more dense the material the better it conducts while also requiring more energy to warm. This combines into an additional parameter, the diffusivity (k) which is defined as the ratio of the thermal conductivity to the heat capacity, equation (12).

ck

⋅=

ρλ

(12)

For the prediction of normal internal shaft temperature, the thermal conductivity, heat capacity, and the resultant diffusivity can be determined from boring logs whereby the soil type and blow count are used to estimate mineral content and density (Pauly, 2010).

Finally, the temperature diffusion is characterized by the partial differential equation (13) where the change in temperature, T, with respect to time, t, is proportional to the product of the diffusivity, k, and the second derivative of temperature with respect to distance in three spatial directions x, y, and z.

∂∂+

∂∂+

∂∂=

∂∂

2

2

2

2

2

2

zu

yu

xuk

tT

(13)

When a heat source, Q, is added (like concrete hydration energy) the following equation (14) governs wherein the product of the heat capacity, ρC, and the change in temperature, T, with respect to time, t, are proportional to the sum of the heat added, Q, and the divergence of the product of the conductivity, λ, and temperature gradient.

)( TQtTC ∇⋅∇+=

∂∂ λρ

(14)

This overview of heat production and dissipation provides an insight into the workings of three-dimensional finite difference algorithms that can be used to predict the temperature within the shaft at various thermal integrity testing times (Johnson and Mullins, 2007; Mullins et al., 2009). This is then coupled

with shaft geometry to provide the most beneficial timeframe for performing thermal integrity profiles of the curing shaft concrete. To that end, it is important to note that these mechanics are theoretically sound and provide the reproducibility for reliable thermal integrity assessment.

CURRENT INTEGRITY TESTING PRACTICESMost state transportation departments have adopted the Federal Highway Administration guideline for including access tubes in the reinforcing cage of drilled shafts (O’Neill and Reese, 1999). Therein, recommended tube materials, diameters, and plurality have been outlined to provide sufficient access to the shaft cross-section for non-destructive evaluation. Although originally intended for applications involving cross-hole sonic logging, CSL, or gamma-gamma logging, GGL (also called gamma density logging, GDL), this tube installation standard also provides access to measure the internal temperature of the shaft. Both CSL and GGL have a limited detection zone within the shaft cross-section: CSL is generally used to make determinations of the concrete quality directly between the tubes (inside the reinforcing cage) based on arrival times and the resultant wave speed; GGL measures the concrete density within a 76 to 114mm (3 - 4.5in) radius from the centerline of the access tube based on measured gamma counts/s (Caltrans 2005 and 2010). This leaves areas of the shaft untested. Fig. 1 shows the percentage of the cross sectional area actually tested by GGL and CSL as a function of shaft diameter based on an assumed 150 mm (6 in) cover (FDOT, 2010). The two images represent

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

3.3 4.3 5.3 6.3 7.3 8.3 9.3

Shaft Diam. (ft)

Test

ing

Cov

erag

e

1 1.5 2 2.5 3

Shaft Diam. (m)

GGL Shaft Tested GGL Cover Tested CSL Shaft Tested CSL Cover Tested

GGL

CSL

[FIG. 1] Tested area of shaft cross section from GGL and CSL (150 mm or 6 in cover).


graphically the coverage when applied to a 0.9m (3ft) diameter shaft. When less cover is permitted a larger fraction of the core concrete can be assumed when using CSL testing.

Both structurally and geotechnically, the outermost concrete of the shaft provides the most benefit. The contribution to the bending capacity from the core concrete is negligible when compared to that of the outer regions where the moment of inertia is proportional to the square of the distance from the centroid to the contributing concrete area (I = ΣA

ix

i2).

In fact, recent studies have assessed the feasibility of casting shafts with a full length central void to remove the unneeded core concrete (Johnson and Mullins, 2007; Mullins et al., 2009). These sources calculate little reduction in bending capacity but recognize reductions in axial capacity (structural) roughly proportional to the fraction of the removed concrete cross section. The focus there was to reduce the peak internal temperature and the associated mass concrete conditions.

The concrete cover that forms the bond between the shaft reinforcement and the bearing strata can be considered the most important yet is only partially tested by GGL and not routinely tested by CSL without single hole methods. The thermal method of assessing shaft integrity, presented herein, is not limited to these shortcomings and is equally sensitive to anomalies both inside and outside the reinforcing cage.

METHOD DEVELOPMENTIn the wake of cone penetrometer development in the late 1970’s and early 1980’s, cone penetrometers were being outfitted with various sensors (e.g. pore pressure, resistivity, cameras, etc). At that time, faculty researchers at the University of South Florida gave serious consideration to taking soil temperature measurements around freshly cast shafts using the cone as the means to gain access to these regions. Two hurdles seemed insurmountable: (1) the time to achieve thermal equilibrium between a cone-based temperature sensor and the soil (without creating thermal disturbances) was too long to be practical and (2) the inability to penetrate rock or stiff soil commonly the target bearing strata. Additionally, the cost of throw-away embedded instrumentation (e.g. thermocouples or similar) in the reinforcing cage was at that time exorbitant. However, as instrumented load tests came into

favor of many designers, so did embedded inclinometer casings which opened the door to measurements from reusable down-hole devices capable of monitoring inclination, lateral acceleration, axial strain, density, wave speed, and temperature.

The first full scale versions of thermal integrity profilers used inclinometer wheel bodies with much larger infrared sensors than those used today. By the turn of the 21st century, several versions of the equipment had evolved progressively smaller to provide access in smaller diameter tubes staying abreast with the trend toward smaller CSL devices. Smaller access tubes reduce cage congestion and aid in providing better concrete flow through the cage openings. Today’s probe is 32 mm (1.25 in) in diameter and 150mm (6 in) long for use in tubes as small as 38mm (1.5 in) inner diameter (Fig. 2).

THERMAL INTEGRITY PROFILINGThermal integrity profiling uses the measured temperature generated in curing concrete to assess the quality of cast in place concrete foundations (i.e. drilled shafts or ACIP piles). The necessary information is obtained by lowering a thermal probe equipped with four horizontally-directed, infrared thermocouples (radially oriented at 0, 90, 180 and 270 degrees) into access tubes and measuring the tube wall temperature in all directions over the entire length of shaft. Throw-away embedded devices can also perform the same function given adequate quantities are used to provide sufficient coverage.

[FIG. 2] TIP probe equipped with four infrared thermocouples.

In general, the absence of intact / competent concrete is registered by relatively cool regions (necks or inclusions); the presence of additional / extra concrete is registered by relatively warm regions (over-pour bulging into soft soil strata or voids). Anomalies both inside and outside the reinforcing cage not only disrupt the normal temperature signature for the nearest access tube, but also the entire shaft; anomalies (inclusions, necks, bulges, etc.) are detected by


more distant tubes but with progressively less effect. Fig. 3 shows a thermal integrity profile being performed whereby the depth of the probe is tracked by a digital encoder wheel over which the lead wire is passed.

[FIG. 3] Thermal integrity profi ler used to assess shaft concrete quality.

The internal temperature distribution across a normal cylindrical shaft is roughly bell-shaped with the effect of temperature reaching well into the surrounding soil (Fig. 4). The magnitude of the peak temperature is dependent on the concrete mix design, shaft diameter, thermal properties of the soil, and the time of hydration. However, at any time within the hydration period (and roughly the same time thereafter), a distinct, usable temperature profile exists for the given conditions. Although the magnitude of the temperature varies with time, the features (shape) of the profile do not.

50

70

90

110

130

150

170

190

-10 -8 -6 -4 -2 0 2 4 6 8 10Position (ft)

Tem

pera

ture

(F)

10

20

30

40

50

60

70

80

90

-3 -2 -1 0 1 2 3

Position (m)

Tem

pera

ture

(C)

Cage Diameter at Tubes

Excavation Diameter

Highest Measured Temp

Lowest Measured Temp

Depth 12.2m (40ft)

Lateral Temp Distribution (model)

Normal Centered Cage

Temp

Tube 6

Tube 1

[FIG. 4] Modeled temperature distribution across a 3.3m (10ft) diameter shaft at a given depth.

CAGE ALIGNMENTThe temperature measurements from each tube are sensitive to cage eccentricity as well as

the surrounding cover (effective diameter) as a function of time. Based on the temperature distribution shown in Fig. 4, the temperature in all tubes should be the same when the cage is centered. A cage slightly closer to one side of the excavation will exhibit cooler temperatures from tubes closest to the soil walls and warmer temperatures from tubes closer to the center of the shaft. Cages are often slightly off center for various reasons including: oversized excavation or casing, missing or broken spacers, bent cage, etc. Therefore a perfectly formed cylindrical shaft can exhibit higher and lower temperatures from tubes on opposite sides of the cage when the cage is not centered. By comparing both the highest tube temperature measurement and the lowest from the opposite side of the cage to the average at a given depth, cage offset can be differentiated from unwanted changes in cross section. Further, by dividing the change in temperature (from the average) by the slope of the linear portion of the modeled temperature / radius curve (Fig. 4), the magnitude of cage offset can be determined as well as the remaining concrete cover. Fig. 5 shows the results of TIP scans, for which the Fig. 4 results were modeled, showing opposite side tubes warmer or cooler than the average dependent on the amount of offset.

The data shown in Fig. 5 was collected from a 3.3m (10ft) diameter shaft (10 access tubes) constructed in Tacoma, Washington as part of the I-5 / SR16, Nalley Valley Project. By simple inspection, features of the as-built shaft geometry become recognizable. For instance, the water table was at 9.8m (32ft) and caused some sloughing before slurry was fully introduced which is seen in all tubes as being slightly warmer (bulge). The upper 4.5m (15ft) of measurements represent the access tube stick up above the top of shaft which is not of interest. The top and bottom of shaft show the normal effect of both radial and longitudinal temperature dissipation which extends a distance roughly 1 diameter down and up from the respective boundaries. At mid shaft elevations, dissipation is purely radial. Additionally, a sense of the cage alignment over the length of the shaft is obtained by comparing opposite side tubes and the change in temperature relative to the average. The amount of cage offset can be predicted as noted by Fig. 4.

The data for all tubes of the same shaft shown in Fig. 5 can be displayed for a single elevation on a radial temperature scale where warmer

Depth Encoded

Wheel Assembly

Data Acquisition

Computer


tubes are plotted closest to the graph center (Fig. 6). The local temperature axes for each tube are oriented in the direction away from the center based on tube spacing and the corresponding angles. This shows that the cage is slightly north to northwest of the excavation center at that depth; a cooler measurement indicates closer proximity to the shaft edge.

SHAFT SHAPEConcreting logs (i.e. yield plots) are a key mechanism for identifying atypical conditions. This information is collected by measuring the rise in the fluid concrete level between trucks using a weighted measuring tape. The volume of concrete from each truck and the associated rise in concrete level are compared to the theoretical volumes as a first level of post construction review / inspection and are often used to decide whether or not to perform integrity testing. When converted to the effective diameter from each truck a basic shape of the shaft can be estimated. For smaller, one or two-truck pours, no definition or shape can be defined. However, as the temperature distribution near the cage is strongly linear, the average tube temperature plotted versus depth reflects the as-built shape of the shaft. As a result, a refined rendering of the shaft can be prepared regardless of the number of trucks.

The data shown in Fig. 7 was collected from a 2.1m (7ft) diameter shaft (7 access tubes) constructed in Lake Worth, Florida as part of a Florida Turnpike widening / exit enhancement project. This shows the average temperature from all seven tubes and the concrete yield information converted to diameter as well as the planned / theoretical diameter. Note the first and last truck have not been corrected for the estimated volume required to fill the tremie and to over pour the shaft, respectively. Regardless, the diameter calculated for the

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80 100 120 140 160 180

Temperature (F)

Dep

th (f

t)0

5

10

15

20

27 37 47 57 67 77

Temperature (C)

Dep

th (m

)

T1

T2

T3

T4

T5

T6

T7

T8

T9

T10

Avg

Pier 6 Shaft B

Tube 1 61C (142F)

Tube 6 71C (160F)

Avg (bold) 66C (150F)

125F52C

150F66C

175F79C

T1

T2

T3

T4

T10

T9

T8

T7

T6T5

N

[FIG. 5] Thermal integrity data from a 3.3m (10ft) diameter shaft showing effect of cage offset on measured temperature.

[FIG. 6] TIP data displayed on radial temperature scale from a depth of 12.2m (40ft).

0

10

20

30

40

50

60

70

80

90

100

80 100 120 140

Temperature

Dep

th (f

t)

0

5

10

15

20

25

30

0 2 4 6 8 10Shaft Diameter

Dep

th (m

)

Average

Grnd Surf

TOS

WT

BOC

Loose Sand

TOR

BOS

EffectiveDiam.

TheoreticalDiam.

No Correction forOver-pouring Concrete

No Correction forTremie filling / volume

0 3 (m)1 2

(ft)

30 60 (ºC)40 50

(ºF)

[FIG. 7] Average TIP measurements from all tubes compared with the effective diameter from construction / concreting logs.


other seventeen trucks closely correlates to the measured average temperature at those depths. In this case, a large amount of additional concrete was used due to flowing sands above the top of rock (TOR). Other construction log information is also superimposed for additional understanding of the effects on measured temperature. This includes the bottom of the temporary 2.3m (7.5ft) diameter surface casing (BOC), top and bottom of shaft (TOS and BOS), water table (WT), top of loose sand layer which continued down to top of rock (TOR) and the ground surface elevation.

The correlation between radial position and temperature (Fig. 4) in the region around the reinforcing cage coupled with the similarly strong correlation between the average tube temperatures and the as-built shaft diameter (Fig. 7) provides compelling evidence that thermal integrity profiles provide a reliable indication of the overall presence of heat producing shaft concrete. Each tube temperature profile when converted to radius can be plotted radially similar to Fig. 6 but for all depths and used to produce a 3-D rendering of the as-built shaft as shown in Fig. 8.

Just as the presence of excess concrete (higher temperatures) and proximity of the access tubes to the excavation wall (closer is cooler)

affect the measured temperature, the absence of concrete is similarly telling. Interestingly, most shafts tested exhibit over-pour features rather than necks or inclusions; however, when encountered, the lack of an intact concrete volume is also detected.

A study conducted for the Florida Department of Transportation in 2005 demonstrated the effects of cave-ins or necks on the measured temperature. Therein, a 1.2m (4ft) diameter, 7.6 m (25ft) long shaft was cast with two levels of bagged natural cuttings tied to the outside of the 0.9m (3ft) diameter reinforcing cage at depths approximate 1/3 from the top and bottom. The cross sectional loss at both levels was roughly 10 percent of the total area and was about 0.45m (1.5ft) long. At the upper level the bags were split and lumped at two locations across the shaft from each other; at the lower level all the bags were grouped together. Fig. 9 shows the results of the thermal integrity profiles taken 15 hrs after concreting and the cross section of the two anomaly levels.

The reinforcing cage was outfitted with both steel and PVC access tubes (3 each). For convenience, tubes 1, 3, and 5 (PVC) were left dry dedicated for thermal scans while tubes 2, 4, and 6 (steel) remained flooded for CSL. Regretably, this did not provide for the normal

plurality of tubes but was intended to facilitate a series of thermal scans run on 3 hr intervals.

At the upper level one group of bags was directly beside tube 3, the other was close to tube 5, and neither was adjacent to tube 1. Qualitatively, the proximity of the anomaly to the tubes is shown both by the sharpness of the change in temperature with respect to depth as well as the magnitude of change in temperature. Tube 1 shows the least temperature change

0

10

20

30

40

50

60

70

80

90

100

70 90 110 130 150

Temperature (deg F)

Dep

th (f

t)

T1_2

T2_2

T3_2

T4_3

T5_1

T6_1

T7_1

Average

GrndSurfTOS

WT

BOC

LooseSandTOR

BOS

Florida TurnpikeMethod Shaft

[FIG. 8] TIP data (left) converted to 3-D shape (right).


but the broadest disturbance. At the lower level, only tube 1 was in close proximity which showed both the sharp change in profile as well as a change in temperature with magnitude similar to tube 3 above.

Variation in temperatures between tubes again indicates poor cage alignment where at the top of shaft tube 1 starts farthest from the edge (warmest), tube 3 closest (coolest) and tube 5 very near the average (normal cover). Moving down the shaft the cover increases or decreases proportional to the measured temperature where the average represents a centered cage. The dashed lines provide a reference for a straight cage that is slightly sloped; deviations from the lines show necks or bulges. From this simplified review, when a neck in one tube corresponds to bulge on the other side, it implies the cage is deviating from straight and the cross section is not varying. For this shaft, the CSL results showed no indication of flaws but those tests were only performed using 3 and not 4 tubes.

Results of the study were used to establish thermal probe requirements, testing procedures, and preliminary analysis methods. These recommendations have been incorporated into the devices and softwares now used to perform

these tests. Full details of the study can be found elsewhere (Mullins and Kranc, 2007).

CONCLUSIONS

Over the last 30 years, the trend toward higher quality assurance in constructed drilled shafts has moved from monitoring only concrete quantities to refined slurry properties and post-construction, non-destructive testing. Where practical, the use of multiple test methods can provide more information and better assessment of shaft acceptability. Therein, no one method does it all. The thermal integrity approach provides an overall perspective of the shaft based on the presence or absence of intact heat producing concrete. The shape, cage placement, cover and concrete health are all addressed.

In the interest of space, aspects of TIP data analysis have been

only briefly discussed, but several levels of analysis can be performed. These begin with a qualitative review of the temperature measurements which can identify top and bottom of shaft elevations, cage alignment, and gross section changes. When construction and concreting logs are included correlations between diameter and temperature can be established which verify the final location of the poured concrete volume. The majority of TIP results do not require modeling to be interpretted; rather, an understanding of the normal temperature profiles and features is necessary. However, results of numerical modeling can be directly compared to field measurements using the recent advancements in hydration energy predictions for modern concrete constituents. To this end, signal matching model results to field measurements can be used to determine the extent and magnitude of anomalous regions. Such comparisons additionally serve to verify the proper hydration process.

As with other test methods thermal integrity profiles identify a normal baseline temperature; GGL and CSL identify a normal baseline gamma count or arrival time, respectively. From these measurements physical parameters are

0

5

10

15

20

25

110 120 130 140 150 160

Temperature (F)

Dep

th (f

t)0

1

2

3

4

5

6

7

43 48 53 58 63 68 73

Temperature (C)

Dep

th (m

)

Tube 1Tube 3Tube 5Avg

1

35

1

35

Cross-Section ATop Anomaly

Cross-Section BBottom Anomaly

Steel Logging Tubes

PVC Logging

Known Anomaly

[FIG. 9] Thermal integrity profi les from 1.2m (4ft) shaft cast with known anomalies.


estimated (density, GGL; compression wave velocity, CSL). TIP measurements verify the presence of curing cementitious materials from which a volume of intact concrete is estimated. Consequently, predictions of normal density, velocity, or temperature can be made prior to or after testing as a measuring stick of normalcy but in reality local variations from the shaft norm are more reasonable and practical. This is often the mode of evaluation for thermal testing as well.

LIMITATIONSThermal integrity profiling requires temperature generation from hydrating materials to provide distinction between cementitious and non-cementitious materials. Testing should be performed while these materials are warm enough to establish a usable temperature gradient which ranges from 2 to 10 days depending on shaft diameter (roughly proportional to shaft diameter in feet, respectively).

When thermal modeling is used as the comparative basis for shaft acceptance, verification of mill certifications from the concrete supplier (constituent fractions) may be necessary as the most common method used by industry to establish constituent percentages are not exact tests. As a result, field validation of model predicted time versus temperature relationships can be performed by simple shaft temperature monitoring using small inexpensive thermocouple data collectors. Thermal integrity profiling using multiple embedded sensors can provide data for both purposes.

Thermal integrity profiling can be performed in both PVC and steel access tubes. However, if tubes are filled with water during construction, the water must be expelled prior to testing, stored, and returned after testing if CSL tests are to be conducted. If CSL tests are not planned, water is not necessary during construction as TIP results are not sensitive to debonding and the water is not used.

DISCLOSUREThis technology was developed by the University of South Florida, for which the author is an active researcher and faculty member. As is customary with such developments, the principal investigators were named as the inventors, and the university has licensed this technology to an outside firm, FGE, LLC, who

has in turn teamed with PDI to manufacture the equipment. The author serves as a part-time consultant for FGE in keeping with guidelines set forth by the University Collective Bargaining Agreement.

ACKNOWLEDGMENTSThe author would like to thank both the Florida Department of Transportation and Washington State Department of Transportation for their support in developing this technology and continued use of thermal integrity profiling. Likewise, the founders of Foundation & Geotechnical Engineering, LLC are gratefully recognized. Finally, the years of dedication from Rudy, Ltd, are wholeheartedly appreciated.

REFERENCESCaltrans, 2005. Method of ascertaining the 1. homogeneity of concrete in cast-in-drilled-hole (CIDH) piles using the gamma-gamma test method. California Department of Transportation Specifications, California Test 233.

Caltrans, 2010. Gamma-gamma logging 2. (GGL). http://www.dot.ca.gov/hq/esc/geotech/ft/gamma.htm

Debye, P., 1914. Vorträge über die kinetische 3. theorie der materie und der elektrizität (trans. Discussion of Kinetic Theory of Matter and Electricity), gehalten in Göttingen auf einladung der Kommission der Wolfskehlstiftung, B.G. Teuber Publisher, Liepzig and Berlin.

Duarte, A., Campos, T., Araruna, J., and 4. Filho, P., 2006. Thermal properties of unsaturated soils. Unsaturated Soils, GSP, ASCE, pp. 1707-1718.

FDOT, 2010. Standard specifications for 5. road and bridge construction. Florida Department of Transportation, ftp://ftp.dot.state.fl.us/LTS/CO/Specifications.

Farouki, O., 1966. Physical properties of 6. granular materials with reference to thermal resistivity. Highway Research Record 128, National Research Council, Washington, DC, pp 25-44.

Farouki, Omar T., 1981. Thermal properties 7. of soils. CRREL Monograph 81-1, Army Corps of Engineers Cold Regions Research and Engineering Laboratory, Hanover, New Hampshire.


Hertlein, B., 2001. Are our client’s 8. expectations realistic? Geo-Strata, Geo-Institute of the American Society of Civil Engineers, January, p.11.

Johansen, O., 1975. Thermal conductivity 9. of soils and rocks. Proceedings of the Sixth International Congress of the Foundation Francaise d’Etudes Nordigues, Vol. 2, pp.407-420.

Johnson, K. and Mullins, G., 2007. Concrete 10. temperature control via voiding drilled shafts. Contemporary Issues in Deep Foundations, ASCE Geo Institute, GSP No.158, Vol. I, pp. 1-12.

Kranc, S.C. and Mullins, G., 2007. Inverse 11. method for the detection of voids in drilled shaft concrete piles from longitudinal temperature scans. Inverse Problems Design and Optimization Symposium, Miami, FL, April 16-18, 2007.

Mullins, A. G. and Kranc, S. C., 2004. Method 12. for testing the integrity of concrete shafts. US Patent 6,783,273.

Mullins, G. and Ashmawy, A., 2005. Factors 13. affecting anomaly formation in drilled shafts. Final Report, FDOT Project BC353-19, March.

Mullins, G. and Kranc, S., 2007. Thermal 14. integrity testing of drilled shafts. Final Report, FDOT Project BD544-20, May.

Mullins, G., Winters, D., and Johnson, K., 15. 2009. Attenuating mass concrete effects in drilled shafts. Final Report, FDOT Project BD544-39, September, 148 pp.

O’Neill, M.W. and Reese, L. C., 1999. 16. Drilled shafts: construction procedures and design methods. U.S. Department of Transportation, Publication No. FHWA-IFF-99-025, ADSC-TL 4, Volume II.

Pauly, N., 2010. Thermal conductivity of 17. soils from the analysis of boring logs. Master’s Thesis, University of South Florida Department of Civil and Environmental Engineering, December.

Schindler, A. and Folliard, K., 2005. Heat 18. of hydrations models for cementitious materials. ACI Materials Journal, Vol. 102, No.1, pp. 24-33.

U.S. Department of the Interior, 2004. 19. Story of Hoover Dam; concrete. Bureau of Reclamation, http://www.usbr.gov/lc/hooverdam/History/essays/concrete.html.

Whitfield, T., 2006. “Effect of C3S content 20. on expansion due to ettringite formation. Master’s Thesis, University of South Florida Department of Civil and Environmental Engineering, June.


TECHNICAL NOTE Load Testing and Interpretation of Instrumented Augered Cast-in-Place PilesTimothy C. Siegel, P.E., D.GE, Principal Engineer, Dan Brown and Associates, Knoxville, TN, USA;

[email protected]

ABSTRACTGiven that piles are composed of different materials and installed using a variety of methods, ASTM D 1143/D 1143M-07 Standard Test Methods for Deep Foundations Under Static Axial Compression is justifiably general so that it has as many applications as reasonably possible. There are aspects of the test setup, the test procedures, and data interpretation for instrumented piles (some of which are specific to augered cast-in-place piles) that are either not discussed in detail in this ASTM standard or are not addressed at all. Specifically, the use of variable time hold times can obscure the shape of the conventional load-deflection curve, as well as, influence the interpretation of the ultimate pile capacity. Unload-reload cycles induce additional non-uniform internal axial loads within the pile which complicate interpretation of the strain gage data and increase the potential for errors. Attempts to maintain a constant top load during the hold period can lead to operational difficulties in taking measurements and are technically unnecessary. Residual load has been shown to develop in cast-in-place piles and can significantly influence the axial load distribution as interpreted from strain gage data. The stiffness of cast-in-place piles has been observed to vary with the measured strain, and a constant modulus, as provided by correlation with the unconfined compressive strength of the grout, may lead to a significantly different interpretation of the axial load distribution.

INTRODUCTIONTest procedures for conventional top-loaded axial compression testing of a single pile are presented in ASTM D 1143/D 1143M-07 Standard Test Methods for Deep Foundations under Static Axial Compression (ASTM International, 2007). Given that piles are composed of different materials and installed using a variety of methods, the referenced ASTM standard is general so that it is applicable to as many situations as reasonably possible. It is understandable that there are aspects of the test setup, the test procedures, and data interpretation for instrumented piles (some of which are specific to augered cast-in-place piles or ACIP piles) which are either not addressed in detail in the ASTM standard or are not discussed at all. The purpose of load testing is not limited to verification of the preliminary design, but is also to collect site-specific information on the pile geotechnical resistance on which to justify optimization of the pile length. Particularly in the latter case, it is useful to separate the resistance to the total applied top load into the two components of (1) the shaft resistance and (2) the toe resistance. An approximation of the shaft and toe resistances can be accomplished using a graphical construction (Bloomquist et al. 2007),

but they can be more directly computed by the collection and subsequent interpretation of strain gage data along the length of the pile. While a correct interpretation of the load-deflection response and the strain data can lead to a greater understanding of the pile behavior, misunderstandings can lead to confusion and even a misplaced loss of confidence in ACIP piles. Such misunderstandings do occur, due in part, to the higher level of complexity involved with load testing ACIP piles. Given that the foundation design depends on accurate data and interpretation, it becomes imperative that each step in the instrumentation setup, data collection, and interpretation be executed in a rational manner that is consistent with the physics of pile and soil behavior.

PILE-SOIL MODELA representative conceptual pile-soil model is necessary for the successful planning, execution and interpretation of load tests on ACIP piles. In that regard, it is understood that the internal pile strain (i.e., changes in the strain gage reading from the factory) is a result of the following four mechanisms: (1) disturbance due to shipping, handling, or installation, (2) internal stresses due to grout set and curing; (3) residual load as the pile-soil


system progresses toward stress equilibrium and strain compatibility even in the absence of a top load, and (4) transfer of the applied top load to the soil surrounding the pile (as either shaft resistance or toe resistance).

When a top load is applied, the pile undergoes compression that begins at the top and progresses to the depth where the entire applied top load is fully distributed into the soil. The pile also shortens under the compressive load. The pile tends to move downward relative to the surrounding soil and this mobilizes positive shaft resistance (i.e., the soil tends to force the pile upward). Depending on the magnitude of the movement of the pile toe, the soil immediately beneath the pile may also push upward on the pile. If the top load were removed, then the pile will tend to expand upward and the direction of the shaft resistance will reverse along a portion of the upper pile. That is, pile expansion will be resisted by negative shaft friction (also called negative skin friction or negative shaft resistance). A net effect of such an unload-reload cycle is the development of additional internal compressive loads. There was a time when the total downward pile movement minus the upward pile movement upon unloading (known as rebound) was believed to be a meaningful representation of the pile settlement. This has been shown to be erroneous given the better understanding of pile behavior and specifically the development of shaft resistance.

The data from a load test is most often used to estimate the ultimate geotechnical resistance of pile and can be used to estimate the distribution of the shaft and toe resistances. The design load may be compared to the interpreted geotechnical resistances to determine the factor-of-safety against geotechnical pile failure for the controlling load case. The duration of the applied top load is essentially always too short to conclude that the distribution of the geotechnical resistances represents a state of equilibrium. The single pile or pile group settlement should be computed using a method consistent with the unified design of piled foundations (Felllenius, 2004) which explicitly considers the relationship between pile load, soil compressibility, residual load and settlement. Similar to the accepted design methodology for shallow foundations, pile foundations should be designed based on the capacity often confirmed by load tests and the settlement based on field and laboratory data

and an appropriate analytical model (Meyerhoff, 1976; Fellenius, 2004).

PILE INSTRUMENTATIONStrain Gages. The writer is most familiar with the vibrating wire type strain gage where the measurement is the frequency of vibration of a tensioned steel wire whose frequency of vibration is proportional to the mass and length of, and strain in, the vibrating wire. The gage is attached to (or embedded inside) a section of reinforcing steel at the manufacturer’s laboratory and the entire assembly is embedded in the grout during installation. The following suggestions are made regarding the installation of the strain gages: Record the strain gage serial numbers and confirm that the ends of the wires are correctly marked with the serial number.

1. Use two gages at each location for redundancy. It has been suggested that the two gages at each location be placed at points opposite one another within the pile cross-section to allow adjustments for eccentricity. The drawback is that if one of the gages fails prior to or during testing, then the data from a single gage that is offset significantly from the pile center may result in greater error. There may also be limitations in the placement of gages associated with the reinforcing steel in the test pile.

2. Plan the strain gage wire lengths well in advance of the pile installation keeping in mind that it is much easier to cut (shorten) a long cable than it is to successfully splice a short cable. Wires that are too short are problematic while wires that are very long become unwieldy during placement of the reinforcing steel. Cable splices should be avoided wherever possible.

3. The sister bar should be securely fastened to the center reinforcing bar either with wire ties or welding so that its position does not slip during installation. The strain gage wires should be secured firmly along the bar while avoiding pinching. The strain gage wires may be bundled near the top of the center bar but there needs to be some exposed center bar (or an extension to the center bar) to allow for placing the lifting strap during installation. The writer’s experience is that the center bar should be placed in the fluid grout first and then the shorter reinforcing cage is placed so that the center bar is threaded through the


middle of the cage. The placement of the center bar separately from the reinforcing cage appears to reduce the potential for uncontrolled bending of the sister bars and stretching of the wires. The combined weights of the center bar and reinforcing cage have been observed to increase the potential bending during lifting and subsequently to increase the potential for stretching of the wires as well.

4. Telltales. A telltale can be used to directly measure pile movements at certain points along its length. While a telltale (or multiple telltales) can be used at any location along the pile length, a telltale at the pile toe can be used to collect information on which to evaluate the relationship between mobilized toe resistance (established using the strain gage data) and toe movement.

To allow the use of a telltale, a small diameter (12 to 19 mm or 0.5 to 0.75 in) metal pipe is typically attached to the center bar to be embedded in the grout. A permanent cap should be on the bottom end of the pipe and a removable cap should be on the top end. The tell tale rod should be very stiff with negligible bend. A butt welded reinforcing steel bar with a diameter slightly smaller than the pipe inner diameter can be very effective. The steel bar should make a right angle where it exits the top of the pile. Ideally, the connection at the right angle should be welded rather than created by a bend. Bends are often more flexible which is undesirable. The telltale should extend beyond the perimeter of the pile.

PILE LOADINGExperience has shown that loading the pile in equal time intervals in load increments consistent with the quick test generally provides a smooth load-deflection curve (Fellenius, 1980). Varying the time interval is undesirable and unnecessary. It may be initially perceived that a longer time interval would provide an estimation of pile settlement, but the settlement is more appropriately considered by an analytical procedure that more accurately represents the pile under design conditions. The time interval for each load increment should allow sufficient time to make at least two sets of readings for all of the dial gages, strain gages, jack pressure, and load cell. Automated load testing devices that allow nearly continuous readings are even better. Make sure that the batteries in all readout boxes and dataloggers are fully charged prior to

testing. In some cases, battery failure in a device can result in loss of data. Fully charged spare or replacement batteries should be kept on hand.

All readings of pile movement should be collected directly from the pile. Experience indicates that pile movements that reference plates, the jack, or surfaces other than the pile itself can incur error in the readings. An example is where the plate on the top of the pile exhibits some warping either prior to or during loading. During loading, the plate can experience measurable bend due to either flattening or warping even further.

The target load should initially be applied and then there will likely be some observed reduction in the applied load as the pile moves downward. The load will be transferred downward along the length of the pile and distributed as the soil resistance is mobilized. The pile-soil system will progress in time toward stress equilibrium and strain compatibility. Considering that two months or more may be required to reach a balance between stress and strain (Siegel and McGillivray, 2009), it is not practical to wait until the pile-soil system is in equilibrium within the duration of the load test.

Attempts to maintain a “constant” top load should be avoided. A practical reason is that the readings from the dial gages, strain gages and telltales will continue to change with each attempt to re-establish the target load. A technical reason is that once the pile moves approximately 2.5 to 5 mm (0.1 to 0.2 inches) then the shaft resistance along the upper portion of pile will begin to decrease after its peak value. Each attempt to re-establish the target load will result in further decrease in shaft resistance in the upper pile due to remolding at the shaft-soil interface which may be offset by the mobilization of additional shaft resistance in the lower pile and/or toe resistance. The best practice is to record the applied load that exists at the end of the time interval.

The pile should be incrementally loaded until one of the following occurs: (1) the applied load equals the limit of the test system, (2) the pile experiences structural failure, or (3) the pile is no longer able to resist the target load – i.e., geotechnical failure. Cycling of the load should be avoided as it induces an unrecoverable internal compressive load that varies non-uniformly through the pile length. During unloading, the internal compressive


load develops as lower portion of pile tends to force the pile upward while negative skin friction develops along the upper portion of pile. The distribution and direction of the shaft resistance will vary with depth which will lead to difficulties in the interpretation of the strain gage and telltale data.

Minimal significance should be applied to the unload portion of the load test. During and after removal of the applied top load, the distribution of the internal compressive load is unknown. It is also certain that the pile movement and the stress conditions will continue to change with time. As discussed earlier, the actual pile settlement should not be determined from the load-deflection behavior during load testing.

If the pile appears to have experienced a structural failure, then observations during unload can help as a confirmation. When a pile structurally fails then the readings in the lower strain gages can become inconsistent and the telltale acts as if it is wedged in the pile. During reloading, a structural damaged pile may resist a load that corresponds to the intact portion of the pile.

DATA INTERPRETATIONThe two primary parts of the data interpretation are (1) selection of the ultimate pile resistance, and (2) determining the distribution of the shaft resistance. There are an abundance of proposed methods to interpret the ultimate pile resistance. Appropriate methods are calibrated for the typical ACIP pile conditions and identify the limit state or the ultimate pile resistance. The ultimate pile resistance should not make any consideration for the pile movement under design loads. Such is not the purpose of the load test nor do the conditions of the load test sufficiently represent the behavior of the pile under the actual conditions. Ideally, the ultimate pile resistance will be defined by the shape of the load-deflection curve and specifically the point at which the slope of the curve becomes very steep. One such failure interpretation is the Brinch Hansen 90% criterion (Brinch Hansen, 1963) that defines failure where one-half of the top deflection at the failure load occurs as a result of the application of 90% of the failure load.

Determination of the distribution of the shaft resistance using the strain gage data has been described by Fellenius (2001). In the initial

step, the top load is divided by the product of the cross-sectional area and the measured strain (actually the difference between the zero load strain and the strain under the top load) near the top of the pile to calculate the secant or tangent moduli. It is also feasible to determine a direct correlation between strain and load. In any case, the uppermost strain gage should be sufficiently close to the ground surface so that the assumption of negligible shaft resistance is valid. The modulus of ACIP piles varies with strain and time (including curing conditions). Typically, the effect of time is ignored and the stress-strain response of the uppermost gage(s) is used to define the moduli in terms of strain only. Data from several projects were used to prepare Fig. 1 which illustrates the variation of secant modulus with strain where the nominal pile diameter (B) was between 406 and 610 mm (16 to 24 inches).

As shown in Fig. 1, the secant modulus decreases with strain. There appears to be some variation in the strain-modulus relationship for different load tests which emphasizes the importance for site-specific determination of the modulus for strain gage interpretation. For reference, the shaded zone illustrates the expected range of values that may be derived from correlations with the unconfined compressive strength testing of grout samples. This suggests that the use of such correlations will tend to under predict the modulus and lead to a significantly different interpretation. The writer’s experience confirms the interpretation based on such correlations tends to over predict the shaft resistance in the upper portion of the pile and under predict the resistance in the lower portion of the pile.

0

10000

20000

30000

40000

50000

60000

0 200 400 600 800 1000 1200 1400

Microstrain

Sec

ant M

odul

us (M

Pa)

B = 406 mm (16 in)

B = 406 mm (16 in)

B = 406 mm (16 in)

B = 457 mm (18 in)

B = 457 mm (18 in)

B = 610 mm (24 in)

Shaded zone is the expected range of values derived from correlations with the unconfined compressive strength testing of grout samples.

[FIG. 1] Variation in Secant Moduli with Strain for ACIP Piles


Using the strain dependent modulus, the internal compressive load is computed for each strain gage location. The difference in the internal compressive load between strain gage locations is the mobilized geotechnical resistance plus any residual load. Long term monitoring of ACIP piles indicates that the residual load developed within the typical wait time for load testing (approximately 7 to 10 days) is relatively small but that it can become significant afterwards (Siegel and McGillivray, 2009). If there is an extending wait period between pile installation and load testing, then it will become necessary to correct the load distribution for the effects of residual load development. As there is no established analytical method for determining the development of residual load over time, the writer proposes interpolation based on the results of long term observation. The limited available data indicates that the residual load begins developing approximately 5 days after pile installation and reaches near equilibrium approximately 58 days after installation. The equilibrium stress conditions are defined where the negative skin friction developed along the upper portion of pile will be resisted by the shaft resistance in the lower portion of pile and the mobilized toe resistance in accordance with the unified design of piled foundations.

EXAMPLESEffect of Residual Load. The writer considers the interpreted load distribution of a 760 mm (30 inch) diameter bored pile as presented by Reese et al. (1976) as an example of the effect of residual load on the interpretation of data from an instrumented cast-in-place pile (Fellenius and Altaee, 1996). This pile was installed in high plasticity clay and low plasticity silt where the water table was approximately 4.6 m (15 ft) below the ground surface. The strain data was collected using embedded Mustran strain cells and converted to axial compressive load. The following three quantities are presented in Fig. 2 versus depth: (1) the internal compressive load as originally interpreted by Reese et al. (1976) from instrumentation with the assumption of no residual load; (2) the internal compressive load computed using the β method and a constant β value of 0.4. and; (3) the interpreted residual load. The interpreted residual load is the difference between (1) and (2).

For comparison, Fig. 3 shows the typical shape of the residual load distribution proposed in several studies (Hanna and Tan, 1973; Fellenius and Altaee, 1995; Siegel and McGillivray, 2009). The writer believes there is sufficient similarity between the typical shape of residual load distribution (Fig. 3) and the interpreted residual load (Fig 2) to conclude that significant error was introduced by ignoring the presence of residual load.

0

1

2

3

4

5

6

7

8

0 500 1000 1500

Dep

th (m

eter

s)

Data from Reese et al., 1976

Interpreted Compressive Load for Beta = 0.4

Interpreted Residual Load

[FIG. 2] Interpreted Internal Compressive Load in Pile from Reese et al. (1978)

Load and resistance

Dep

th

[FIG. 3] Typical Shape of Residual Load Distribution

Effect of Modulus. The writer considers the interpreted load distribution of a 356 mm (14 inch) diameter ACIP pile as presented by Beck and Harrison (2009) as an example of the effect of the modulus selection in the interpretation of data from an instrumented cast-in-place pile. The 26.5 m (87 ft) long test ACIP pile was installed into (from the ground surface): (1) approximately 5 m (16 ft) of existing fill, (2) 10 m (33 ft) of soft to medium stiff clay, and (3)


loose to medium dense sand. Fig 4 compares the reported load distribution (based on an interpretation using a constant modulus) and a modified load distribution interpretations based on the non-linear modulus shown in Fig 1. It was assumed, for comparison purposes, that a constant modulus of 24.1 MPa (3.5 x 106 psi) was used by Beck and Harrison considering that the modulus was assigned based on the results of unconfined compression tests (Beck, 2009). As illustrated in Fig 4, the non-linear modulus results in the interpretation of higher internal compressive loads at smaller strains and the distribution shows a significantly greater portion of the top load being transfer to the clay and sand below the upper fill during load testing.

0

5

10

15

20

25

30

0 500 1000 1500 2000

Interpreted Internal Compressive Load (kN)

Dep

th (m

eter

s) Data from Beck and Harrison, 2009

Modified using a Nonlinear Modulus

[FIG. 4] Interpreted Internal Compressive Load Distribution in Pile from Beck and Harrison (2009)

CONCLUSIONSConsidering that the procedures in ASTM D 1143/D 1143M-07 Standard Test Methods for Deep Foundations Under Static Axial Compression are general so that it may be applied to a wide range of pile types, there are aspects of the test setup, the test procedures, and data interpretation for instrumented piles (some of which are specific to ACIP piles) that are either not discussed in detail in this ASTM standard or are not addressed at all. The following conclusions highlight several aspects of load testing not explicitly addressed in the referenced ASTM standard with an emphasis on their application to ACIP piles.

The use of variable time hold times can 1. obscure the shape of the conventional load-deflection curve, as well as, influence the interpretation of the ultimate pile capacity.

It is strongly preferred to maintain a constant hold time for each load increment. Unless the pile is experiencing a plunging failure, the dial gages typically stabilize during the hold time; however, some movement of the jack ram and decrease in the applied test load are expected. Attempts to maintain a “constant” top load are unnecessary and should be avoided for practical and technical reasons.

Unload-reload cycles should be avoided 2. as it induces an unrecoverable internal compressive load that varies non-uniformly through the pile length. During unloading, the internal compressive load develops as lower portion of pile tends to force the pile upward while negative skin friction develops along the upper portion of pile. The distribution and direction of the shaft resistance will vary with depth which will lead to difficulties in the interpretation of the strain gage and telltale data.

Residual load has been shown to develop 3. in cast-in-place piles and can significantly influence the axial load distribution as interpreted from strain gage data. Long term monitoring of ACIP piles indicates that the residual load developed within the typical wait time for load testing (approximately 7 to 10 days) is relatively small but that it can become significant afterwards (Siegel and McGillivray, 2009). If there is an extending wait period between pile installation and load testing, then it will become necessary to correct the load distribution for the effects of residual load development.

The stiffness of cast-in-place piles has 4. been observed to vary with the measured strain, and use of a constant modulus, as provided by correlation with the unconfined compressive strength of the grout, may lead to a significantly different interpretation of the axial load distribution.

REFERENCES

ASTM International (2007). ASTM D1143/1. D1143M-07 Standard Test Methods for Deep Foundations under Static Axial Compressive Load, West Conshohocken, PA, 15 pp.

Beck, W.K. and Harrison, P.J. (2009) Load 2. tests on small diameter augered cast-in-place piles through fill ASCE, Contemporary Topics in Deep Foundations, GSP No. 185,


Edited by M. Islander, D.F. Laefer and M.H. Hussein, Orlando, Florida, 430-437.

Beck, W.K. (2009) Personal communication.3.

Bloomquist, D., McVay, M., and Hu, Z. 4. (2007) Final Report – Updating Florida Department of Transportation’s (FDOT) Pile/Shaft Design Procedures Based on CPT & DTP Data, Department of Civil and Coastal Engineering, University of Florida, Gainesville, Florida, 199 pp.

Brinch Hansen, J. (1963) Discussion, 5. Hyperbolic stress-strain response. Cohesive soils, ASCE, Journal of the Soil Mechanics and Foundations Division, 89(SM4), 241-242.

Fellenius, B.H. and Altaee, A.A. (1995) 6. Critical depth: how it came into being and why it does not exist, Proceedings, Institution of Civil Engineers, Geotechnical Engineering, 113(4), 107-111.

Fellenius, B.H. and Altaee, A.A. (1996) 7. Discussion Critical depth: how it came into being and why it does not exist, Proceedings, Institution of Civil Engineers, Geotechnical Engineering, 119, 244-245.

Fellenius, B.H. (1980) The analysis of 8. results from routine pile load tests, Ground Engineering, 13(6), 19-31.

Fellenius, B.H. (2001) From strain 9. measurements to load in an instrumented pile, Geotechnical News Magazine, 19(1), 35-38.

Fellenius, B.H. (2004) Unified design of piled 10. foundations with emphasis on settlement, ASCE, Current Practice and Future Trends in Deep Foundations, GSP No. 125, Edited by J.A. DiMaggio and M.H. Hussein, Los Angeles, California, 253-275.

Hanna, T.H. and Tan, R.H.S. (1973) The 11. behavior of long piles under compressive loads in sand, Canadian Geotechnical Journal, 10(3), 311-340.

Meyerhoff, G.G. (1976) Bearing capacity 12. and settlement of pile foundations, ASCE, Journal of the Geotechnical Engineering Division, 102(3), 197-228.

Reese, L.C., Touma, F.T., and O’Neill, M.W. 13. (1976) Behavior of drilled piers under axial loading, ASCE, Journal of the Geotechnical Engineering Division, 102(5), 493-510.

Siegel, T.C. and McGillivray, A. (2009) 14. Interpreted residual load in an augered cast-in-place pile, Proceedings, 34th Annual Conference on Deep Foundations, Deep Foundations Institute, 173-182.


Underwriters: Gold

Underwriters: Silver

Geotechnical and Geostructural Engineering Dam, Tunnel and Underground Engineering Construction and Owner Services Geosciences, Environmental and Sustainable Geotechnics

West Chester, Pennsylvania / 610-696-6066 / schnabel-eng.com

Providing specialty services with solutions for a changing world


DFI Journal Paper Review Process

The peer review process for documents considered for publication in the DFI Journal is still evolving. The following is a description of the current process, however, the publication is still in its infancy and the review process is still in a state of flux. DFI reserves the right to alter the procedures as necessary.

Paper SubmittalPapers may be submitted at any time. Authors wishing to submit their papers for consideration of publication in the DFI Journal are invited to access www.dfi-journal.org. The website will ask for a login or, for new submitters, will ask for creation of an account. Once logged in the author must upload a full paper in MS Word format as well as any ancillary files such as figures, photos and other graphics which are included in the paper. The paper is then converted to a PDF file which the author must approve before the paper will be released to the publisher and journal editors for viewing. The journal editors preliminarily review the paper for relevancy to the Journal mission.

Paper Review The journal editors assign those papers deemed to be worthy of consideration for Journal publication to the appropriate editorial board member, which currently consists of DFI technical committee chairmen and other industry leaders, so that appropriate reviewers for the paper topic can be obtained. Reviewers are chosen based on their knowledge, areas of expertise, and qualifications to act as a reviewer on the particular subject matter of the paper in question. At least three reviewers will be assigned to each paper.

After the reviewers are selected, they are provided with instructions and a password for entry into the website where they can view the paper PDF and submit their evaluation. The criteria on which they base their review fall under two areas: technical content and quality of paper presentation. The criteria for technical content include relevancy, originality, appropriate references to support statements, significance of results and exclusion of personal opinion and commercialism. The criteria for paper presentation include quality of figures, quality of English language, paper organization and completeness. The reviewers enter their evaluation by responding to a number of questions rating the paper as well as entry of comments to authors. They are also required to make a recommendation to the journal editors of: accept as is; accept with mandatory changes; or reject. The author is advised by automatic email of the posting of reviews and he/she can access the reviews and respond and/or modify the paper to satisfy comments by the reviewers. A second round review can then take place if necessary, ultimately leading to second round reviewer recommendations. The publisher and editors, acting as a final review committee, make the decision, based on the reviewers’ recommendations, as to acceptance of the paper for publication in the next issue of the journal or in a subsequent issue.

Throughout the process, automatic emails are sent out to reviewers when papers are ready for their review and to the authors to keep them aware of the progress of their paper.

Paper Finalization Upon acceptance, the final paper submission by the author and all graphic files are downloaded by the publisher for processing and formatting for publication. The publisher is provided with proofs by the production house and these are edited to ensure acceptable layout, the absence of typos, clarity of figures, etc. In most cases the author(s) are provided with a final PDF for their review and approval.


2011 DFI Journal Subscriptions

Electronic Subscriptions for the two-issue volume (May & November 2011) are provided to all DFI members at no cost – the electronic version is included as a benefit of annual membership dues.

DFI Members may order print versions of the DFI Journal. Non-Members can join DFI to receive the electronic version and print version at member rates OR can purchase a subscription only of either electronic, print or both.

Build up your Journal Library! Past volumes are available for purchase while supplies last.

Volume 5 (May & Nov 2011) Subscription Rates Electronic

Print Only (USA)

Print Only (Outside USA)

Electronic & Print(USA)

Electronic & Print(Outside USA)

DFI MEMBER Free with membership

$125 $175 N/A N/A

NON-MEMBER $150 $300 $350 $405 $455

Join DFI $95 & receive electronic subscription plus other member benefits

Past Volumes Single Issue Rates Electronic

Print Only (USA)

Print Only (Outside USA)

Past 2-issue Volume Rates

Subscription Rates

Volume 1 (Nov 2007) MEMBER & NON-MEMBER

FREEonline

$45 $70 Vol. 3 (May/Nov 2009) Same as Vol .5 above

Volume 2 (Nov 2008) MEMBER & NON-MEMBER

$45 $45 $70 Vol. 4 (May/Nov 2010) Same as Vol. 5 above

Last: ________________________________________ First:__________________________________

Jr.,III,etc:_______________ P.E.,Ph.D.,Etc._________________ Mr./Ms/etc: _________________________

Organization: ___________________________________________ Position:_________________________

Address: ________________________________________________________________________________

City: __________________________________ State/Prov: __________ Zip/Postal Code:________________

Country:__________________________ Tel: ______________________ Fax: ________________________

E-mail Address:___________________________________________________________________________

***Circle Your Selections in Rate Charts Above*** Total $: _____________

Check Enclosed:*Payments from outside the US are requested to pay by credit card or Bank Draft on a USA Bank.

Visa/MasterCard/AMEX/Discover/Diner’s Club (Circle One)

Card Number:_______________________________________________________________

Exp. Date: ______/______ Signature: _____________________________________________

Order online at www.dfi.org/dfijournal.asp


DFI Journal Call for Papers

The Deep Foundations Institute compiles and publishes a Journal of practical and technically rigorous papers each year and has plans to reduce the interval between editions, aiming ultimately to publish on a quarterly schedule. The DFI Journal content is subject to quality technical review, and must meet a standard in quality on practical subjects dealing with case studies, deep foundations history, design, construction, testing, innovations and research in the field.

Each journal consists of at least five documents collected from technical papers that are invited or selected from papers submitted by international industry members based on this call. Papers presented at the DFI Annual Conference and Specialty Seminars may be included if expanded to the Journal review and standard.

The editors are herein sending out a call for original papers for consideration of inclusion in the upcoming journals. Full draft papers up to 15 pages in length are to be submitted to: http://www.dfi-journal.org for review. Authors will be required to create a login account and will be notified via email on the status of their submission.

Papers are solicited on the following topics: Case studies involving foundation systems with technical data support Historical evolution of deep foundations Relationship between use of design, construction and equipment Quality control, quality assurance and non-destructive testing Innovation in all aspects of deep foundations and earth retention Practice-oriented research

The Journal Editorial Board will review submitted papers and contact authors who have been chosen for publication in one of the journals with full reviews and edits in order to complete their final paper. Authors of papers accepted for publication will be required to sign a copyright license agreement.

Deep Foundations Institute was incorporated in 1976 in the State of New Jersey as a non-profit educational activity. DFI is a technical association of firms and individuals in the field of designing and constructing deep foundations and excavations. DFI covers the gamut of deep foundation construction and earth retention systems.

Although the bulk of the membership is in North America, the Institute is worldwide.

DFI’s strengths are:

• Communication of information concerning the state-of-the-art and state of the practice of deep foundation technologies

• Offering networking opportunities for our members

• Offering opportunities for members to improve the industry through publications produced by volunteer committees

• Offering educational conferences, seminars and workshops in the industry

The core strength of DFI is the broad spectrum of its membership. All disciplines participate on an equal footing, be they contractors, engineers, owners, academicians, equipment manufacturers and distributors or materials manufacturers and suppliers. All types of foundation systems are represented, whether installed by driving, drilling or other means. This diversity and openness without bias provides a forum for the free exchange of knowledge and a platform for the development of new technology and opportunity.

DFI is:

• An international network of heavy construction professionals dedicated to quality and economy in foundation design and construction

• A forum open to all construction professionals across disciplines and borders.

• A technological association devoted to gathering, storing and disseminating practical information

• A resource for identifying and locating the specialists and sources of expertise.

• An initiator and participant in research

DFI Sustaining MembersAECOM USA INC.AMERICAN EQUIPMENT & FABRICATING CORP.ANDERSON DRILLINGAPE/J&M

BAUER - PILECO INC.BEN C. GERWICK INC.BERKEL & COMPANY CONTRACTORS INC.BRASFOND FUNDAÇÕES ESPECIAIS S/ABRAYMAN CONSTRUCTION CORPORATIONCAJUN DEEP FOUNDATIONS LLCCASE FOUNDATION COMPANYCIPORT S.A.DEAN CONSTRUCTION CO. LTD.DEWITT CONSTRUCTION INC.DOSDOURIAN ENTERPRISES INC.FOUNDATION CONSTRUCTORS INC.FOUNDATION SUPPORTWORKS INC.GEOKON INC.HAYWARD BAKER INC.HJ FOUNDATION COMPANYINTERCOASTAL FOUNDATIONS AND SHORINGKIEWIT CONSTRUCTION GROUP INC.KLEINFELDERL.G. BARCUS & SONS INC.LANGAN ENGINEERING AND ENVIRONMENTAL SERVICESMACTEC ENGINEERING & CONSULTING INC.MCKINNEY DRILLING COMPANYMENARDMORETRENCHMUESER RUTLEDGE CONSULTING ENGINEERSNICHOLSON CONSTRUCTION COMPANYNORTH AMERICAN CONSTRUCTION GROUPO.C.I. DIVISION / GLOBAL DRILLING SUPPLIERS INC.PND ENGINEERS INC.RICHARD GOETTLE INC.SAS STRESSTEEL INC.SCHNABEL FOUNDATION COMPANYTEI ROCK DRILLS INC.THATCHER ENGINEERING CORPORATIONURBAN FOUNDATION/ENGINEERING LLCWILLIAM F. LOFTUS ASSOCIATES FOUNDATION ENGINEERSWURSTER ENGINEERING & CONSTRUCTION INC.

Deep Foundations Institute Sustaining Members Are Corporate Members Of DFI Who Have Voluntarily Granted Funding To The Institute For Expanded Support Of The Industry. The fund is managed by the DFI Educational Trust.

DFI JOURNALThe Journal of the Deep Foundations Institute

Deep Foundations Institute326 Lafayette AvenueHawthorne, New Jersey 07506 USATel: 973-423-4030Fax: 973-423-4031www.dfi .org

International Standard Serial Number (ISSN): 1937-5247

DFI JOURNAL · Vol. 4, No. 2 December 2010 DFI JOURNAL

Documents