Vol 4, No.

1 August 2010

DFI JOURNAL
The Journal of the Deep Foundations Institute

PAPERS:
Drilled Displacement Piles – Current Practice and Design –
Prasenjit Basu, Monica Prezzi, Dipanjan Basu [3]

Analysis of a Deep Excavation In Calgary, Alberta –

Thomas Lardner, Matthew Janes, K.Y. Lo, Guangfeng Qu,
Silvana Micic [21]

Analyzing Drivability of Open Ended Piles in Very

Dense Sands – James A. Schneider, Ivy A. Harmon [32]

Support of Structures in Expansive Shale Using Recycled

Plastic Piles – Lance A. Roberts, Eric Brandner [45]

Peak and Post-Peak Shear Strength of Cement-Bentonite –

Paul J. Axtell,Timothy D. Stark, John C. Dillon [59]

Deep Foundations Institute is the Industry Association of

Individuals and Organizations Dedicated to Quality and
Economy in the Design and Construction of Deep Foundations.
From the Editors and Publisher
While the success of the Journal is principally due to the contributing authors, the
2010 DFI Board of
peer reviewers also make a substantial contribution. We wish to again compliment the Trustees
reviewers of papers for their astute comments and suggestions. President:
The peer review process contributes substantially to the quality of the papers published. Rudolph P. Frizzi
Langan Engineering &
It has been our practice to seek out at least three reviewers for each paper. The search Environmental Services
for knowledgeable reviewers is usually aided by utilizing DFI’s technical committees to Elmwood Park, NJ USA
nominate reviewers from the applicable committee, but there have been instances where
Vice-President:
we have stepped outside of this source to find reviewers whose areas of expertise suit James A. Morrison
the subject of the paper. Ideally, after one or two rounds of review and resubmission of Kiewit Construction Group Inc.
the paper, all the reviewers reach the point where they recommend acceptance of the Omaha, NE USA
paper “as is”. There have been instances where this does not happen and one of the Secretary:
three reviewers remains dissatisfied, while the other two consider the paper acceptable Robert B. Bittner
without further change. When this happens, the paper in question is delayed while the Bittner-Shen Consulting
Publisher and Editorial Board try to assess whether to accept the paper “as is” or mediate Engineers, Inc.
Portland, OR USA
some changes to address the remarks of the dissenting reviewer. Such was the situation
with two of the papers in this issue of the DFI Journal, causing later than scheduled Treasurer:
publication. Please forgive us for considering quality of content more important than Patrick Bermingham
Bermingham Foundation
schedule. Solutions
To date we have not formally invited discussion on Journal papers. We do now wish to Hamilton, ON Canada
advise our readers that we are open to publishing valid discussions on paper published Immediate Past President:
in past issues of the DFI Journal. This could serve as a useful outlet for any dissenting Seth L. Pearlman
reviewers to make known their views. Menard
Bridgeville, PA USA
The editors are pleased that the Journal will now be available for free on-line to DFI
members, which will increase the widespread dissemination of published papers. We hope Other Trustees:
that authors will be even more encouraged to submit and maintain the tradition of high- David Borger
Skyline Steel LLC
quality practice-oriented geotechnical papers for publication. The journal will continue Parsippany, NJ USA
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 Maurice Bottiau
Franki Geotechnics B
the construction industry. Saintes, Belgium
Comments, suggestions, and submissions are welcome and may be submitted via the DFI
Dan Brown
website at www.dfi.org Dan Brown and Associates
Sequatchie, TN USA
Bernard H. Hertlein
AECOM USA, Inc.
Vernon Hills, IL USA
Journal Publisher
Matthew Janes
Manuel A. Fine, B.A.Sc, P.Eng
Isherwood Associates
Journal Editors Burnaby, BC, Canada
Ali Porbaha, Ph.D., P.E. James O. Johnson
DFI JOURNAL California State University
Sacramento, CA, USA
Condon-Johnson & Associates, Inc.
Oakland, CA USA
Mission/Scope Dan A. Brown, Ph.D. Dan Brown and
Associates, Sequatchie, TN, USA Douglas Keller
The Journal of the Deep Foundations Institute Zia Zafir, Ph.D., P.E. Richard Goettle, Inc.
publishes practice-oriented, high quality Kleinfelder Sacramento, CA, USA Cincinnati, OH USA

papers related to the broad area of “Deep Associate Editors Samuel J. Kosa
Lance A. Roberts, Ph.D., P.E. Monotube Pile Corporation
Foundations Engineering”. Papers are welcome Canton, OH USA
South Dakota School of Mines
on topics of interest to the geo-professional and Technology
Kirk A. McIntosh
Rapid City, SD USA
community related to, all systems designed MACTEC Engineering &
Thomas Weaver, Ph.D., P.E.
and constructed for the support of heavy Consulting, Inc.
Nuclear Regulatory Commission
Jacksonville, FL USA
structures and excavations, but not limited Rockville, MD USA
Raymond J. Poletto
to, different piling systems, drilled shafts, Published By Deep Foundations Institute Mueser Rutledge Consulting
ground modification geosystems, soil nailing Copyright © 2009 Deep Foundations Institute. Engineers
AII rights reserved. Written permission must be New York, NY USA
and anchors. Authors are also encouraged obtained from DFI to reprint journal contents, in
John Wolosick
to submit papers on new and emerging whole or in part.
Hayward Baker
Contact
topics related to innovative construction DFI Headquarters Alpharetta, GA USA
technologies, marine foundations, innovative 326 Lafayette Avenue
Hawthorne, NJ 07506 Mike Wysockey
retaining systems, cutoff wall systems, and [email protected] Thatcher Foundations, Inc.
www.dfi.org Gary, IN USA
seismic retrofit. Case histories, state of the DFI, its directors and officers, and journal editors
practice reviews, and innovative applications assume no responsibility for the statements
expressed by the journal’s authors. International
are particularly welcomed and encouraged. Standard Serial Number (ISSN): 1937-5247

DFI JOURNAL Vol. 4 No. 1 August 2010 [1]

[2] DFI JOURNAL Vol. 4 No. 1 August 2010
Drilled Displacement Piles – Current Practice and Design
Prasenjit Basu, Assistant Professor, Department of Civil and Environmental Engineering, Pennsylvania
State University, University Park, PA 16802, USA, Ph: 1-863-814-4010, e-mail: [email protected]
Monica Prezzi, Associate Professor, School of Civil Engineering, Purdue University, West Lafayette,
IN 47907, USA, Ph: 1-765-494-5034, Fax: 1-765-494-0395, e-mail: [email protected]
Dipanjan Basu, Assistant Professor, Department of Civil Engineering, University of Connecticut, CT
06269-2037, USA, Ph: 1-860-486-5023, e-mail: [email protected]

ABSTRACT
Drilled displacement piles are being increasingly used as foundation elements, particularly in
projects requiring fast construction. Different types of drilled displacement (DD) piles are available
in practice. DD piles are classified according to the drilling tool design and installation method.
The capacity of a DD pile, depending on its type, is between the capacities of geometrically similar
nondisplacement and full-displacement piles installed in the same soil profile. This paper provides
an overview of the different types of DD piles and their installation techniques and describes three
design methods used in practice. It also compares DD pile capacities obtained with these design
methods for two different sites.

INTRODUCTION displacement piles. Many auger piles, which

are installed by drilling a continuous-,
A wide variety of pile types are currently
segmented- or partial-flight auger into the
available for use in geotechnical engineering
ground, fall under this category. A variety
practice. The response of these piles to loading
of auger piling equipment is available in the
varies greatly depending on the installation
market; each one is associated with a certain
or construction methods employed. On one
degree of soil displacement during installation.
end of the pile behavior spectrum are the
The commonly used terminologies used for
nondisplacement piles (e.g., bored piles or
auger piles in North America and Europe are
drilled shafts) and, on the other end, are the
presented in Fig. 1.
full-displacement piles (e.g., closed-ended pipe
piles or precast reinforced concrete piles). A special class of auger piles was created as a
Nondisplacement piles are constructed by result of advances in auger piling technology;
removing a cylinder of soil from the ground and these are commonly known as “screw piles” in
replacing the void created with concrete and Europe, and “drilled displacement” or “augered
reinforcement. Full-displacement piles, on the displacement” piles in the USA (Brown and
other hand, are driven or jacked into the ground. Drew 2000; Brown 2005; Prezzi and Basu 2005).
During the installation of full-displacement These piles have been used in Europe over the
piles, significant changes in the void ratio and last few decades and are becoming popular in
stress state of the in situ soil take place because the U.S. Drilled displacement (DD) piles are
the soil surrounding the pile shaft is displaced rotary displacement piles installed by inserting
mainly in the lateral direction and the soil below a specially designed helical auger segment
the base of the pile is preloaded. These changes into the ground with both a vertical force and
produce a stiffer load-displacement response a torque. Soil is displaced laterally within
for the displacement piles than that of the the ground (with minimal spoil generated),
nondisplacement piles, particularly in the case and the void created is filled with grout or
of sandy soils which gain additional strength concrete. The installation of DD piles produces
through densification. greater soil displacement than that produced
by continuous-flight-auger (CFA) or auger cast-
There are other types of piles (e.g., open-ended in-place (ACIP) piles (these piles are generally
pipe piles) that show behavior intermediate associated with small soil displacement). The
between nondisplacement and full-displacement axial capacity of the different types of DD
piles. These piles are often called partial- piles depends on the radial soil displacement
DFI JOURNAL Vol. 4 No. 1 August 2010 [3]
 Auger Piles OVERVIEW
OF DD PILING
European Nomenclature North-American Nomenclature
TECHNOLOGY
The development
Continuous - Screw Piles Auger Cast -In-Place
In-Place Drilled -Displacement (DD)/ of DD piling
Flight -Auger (ACIP) Augered -Displacement
(CFA) technology
evolved from
Continuous - Augercast or Partial - Full-
Flight -Auger Auger Pressure - Displacement Displacement the continuous-
(CFA) Grouted (APG) flight auger (CFA)
piling technology.
[FIG. 1] Nomenclature used for auger piles in Europe and the U.S. (modified after Prezzi and Basu 2005)
The remarkable
progress in piling
caused by their installation (the radial soil rig capabilities over the past few decades and
displacement depends on the soil type and the improvement of the auger pile drilling
state, the design of the DD pile drilling tool, and tools and installation techniques helped speed
the drilling rig technology). up the installation process and resulted in
larger lateral soil displacement. The piles
From a design point of view, full-displacement
that ensued because of these developments
piles are preferable because they are capable
were called DD piles. However, DD piles are
of carrying larger loads than partial- or
not just limited to those that are variations of
nondisplacement piles of similar geometry.
the CFA or ACIP piles. A variety of other piles
However, pile driving may cause excessive
that have significantly different installation
vibration to neighboring structures or create
(drilling) tools are also included in this
excessive noise that may be unacceptable
broad pile classification: Atlas, De Waal,
under certain conditions. Additionally, in
Fundex, Olivier, Omega, Pressure-Grouted
some soil profiles, the use of driven piles may
Displacement (PGD), and SVV piles (see
not be advisable. DD piles often offer a viable
Fig. 2). In general, the drilling tool of a DD
alternative in cases where the installation of
pile contains one or more of the following
driven full-displacement piles is not advisable.
components: a) a soil displacement body (an
The advantages of DD piles are (i) the ease and
enlarged-diameter section which facilitates
fast rate of construction with minimal vibration
lateral soil movement), b) a helical, partial-
or noise, and minimal spoil (important for
flight auger segment, and c) a specially
contaminated sites), (ii) the high load-carrying
designed sacrificial tip, which is attached to
capacity due to partial or full displacement
the bottom of the drilling tool. The shape
of the soil surrounding the pile, and (iii) the
of the displacement body varies from one
associated savings that result when they are
pile type to another; broadly, it consists of
installed in the right soil conditions.
a cylindrical body that, in some cases, also
This paper presents a review of the current contains single or multiple helices (Fig. 2). A
DD pile practice, including DD pile installation casing (mandrel) of diameter smaller than or
methods, quality control procedures, and three equal to the diameter of the pile is connected
design methods. The ultimate capacities of five to the drilling tool.
different types of DD piles are calculated with
the design methods used
in practice and compared
with those obtained
from pile load test
results reported in the
literature. Additionally,
in a separate design
example, the capacities
of a DD pile, a full-
displacement pile and a
nondisplacement pile in
Atlas DeWaal Fundex Olivier Omega PGD SVV
a residual soil profile are
compared. [FIG. 2] Drilling tools for installation of drilled displacement piles.

[4] DFI JOURNAL Vol. 4 No. 1 August 2010

Piling rigs with high torque capacities thick clay deposits, the excess pore pressure
(150 kN-m to 500 kN-m or more) (110 kip generated during installation of DD piles may
ft to 370 kip ft) that provide vertical thrust cause bleeding of fresh concrete and loss of
during the drilling process are required for pile integrity (NeSmith 2002). However, an
installation of DD piles [drilling proceeds as a assessment of the potential problems that may
result of both the rotation of the drilling tool occur during the installation of DD piles can
and the crowd (axial) force typically applied by only be made after full consideration of specific
hydraulic rams]. Once the drilling tool reaches site conditions and equipment capabilities
the desired depth, the sacrificial tip (if used) is (piling rig and drilling tool).
released from the casing or displacement body.
Concrete or grout is then placed through the INSTALLATION TECHNIQUES FOR
casing as the drilling tool and the casing are DIFFERENT DD PILES
extracted from the ground. The reinforcement
is inserted either before or after concrete Atlas Pile
placement. The drilling tool and casing can be The Atlas pile is a drilled, dual-displacement,
withdrawn from the ground with or without cast-in-place concrete pile (De Cock and Imbo
rotation (the rotation may be clockwise or 1994). Lateral displacement of soil occurs both
counter-clockwise). A nearly smooth pile during drilling and extraction of the auger (this
shaft is obtained if the casing is withdrawn is the reason why it is called a dual-displacement
with alternating 180° clockwise and counter- pile). The Atlas pile is installed using a purpose-
clockwise rotations (as in the case of the Fundex built drilling rig with a base rotary drive (Bottiau
pile). A nearly smooth shaft also results if 2006). The Atlas pile rig has two hydraulic
the drilling tool is rotated clockwise as it is rams that can work independently (one ram
withdrawn from the ground (e.g., De Waal, PGD, taking over from the other after its full stroke
and Omega piles). However, if the displacement is achieved) to allow a continuous drilling
body is rotated counter-clockwise (e.g., Atlas operation. In the case of hard soils, the two
and Olivier piles) during withdrawal, then a hydraulic rams can be used simultaneously. The
screw-shaped shaft is obtained. rig can be operated at dual rotational speeds.
Proper knowledge of the subsurface profile is This helps control the drilling tool penetration
required for selecting the most efficient pile rate in different soil types.
type for a given site. For a number of sites in A sacrificial tip (a lost pile shoe) is attached
the U.S., Siegel et al. (2007) reported an increase to a displacement body, which, in turn, is
in cone penetration test (CPT) resistance qc due attached to a steel casing or mandrel (Fig. 3).
to the installation of DD piles. The maximum The displacement body consists of a cast-
increase in qc was observed in the case of loose iron dismountable helical head with an
sand with normalized value of CPT resistance enlarged helical flange. The joint between the
qc1 less than 50 [qc1 = (qc/pa)(pa/σ′v0)0.5; where displacement body and the sacrificial tip is made
pa = atmospheric pressure = 100 kPa and σ′v0 watertight. The combined action of the torque
= in situ vertical effective stress; Siegel et al. and the vertical thrust forces the casing down
2007]. However, according to Bustamante into the ground with a continuous, clockwise,
and Gianeselli (1998), the performance of DD helical penetrating movement. After the desired
piles may be compromised because of possible depth is reached, the steel shoe is detached
difficulties encountered during installation in from the casing by rotating the casing counter-
very loose sandy soils or very soft clayey soils clockwise (thereby opening the connection
(characterized by SPT blow count N < 5 or qc < 1 between the steel shoe and the casing).
MPa or 145 psi). In the case of very dense sandy Subsequently, the steel reinforcing cage is
soils or thick alluvium layers, a drastic drop inserted into the casing, and high-slump concrete
in the penetration rate may be observed and is poured through a hopper placed on top of the
premature wear of the screw head (drilling tool) casing to cast the pile shaft. As the casing and
may result (Bustamante and Gianeselli 1998). the displacement body are extracted by a vertical
According to NeSmith (2002), the installation pulling force and counter-clockwise rotation,
of auger pressure-grouted displacement (APGD) concrete completely fills the helical bore formed
piles (which can more generally be called PGD by the upward-moving displacement screw.
piles) becomes difficult in dense sand layers This way, a screw-shaped shaft is formed. The
with qc > 14 MPa (2,030 psi). In the case of flange thickness of the screw-shaped shaft varies

DFI JOURNAL Vol. 4 No. 1 August 2010 [5]

 depending on the extraction procedure (i.e., the the casing, and concrete is placed. As the
ratio of rotational to translational speeds during concrete is placed, the casing is extracted in an
extraction) (De Cock and Imbo 1994; Geoforum oscillating upward and downward motion with
2008). After concrete placement is complete, it alternate 180° clockwise and counter-clockwise
is possible to push a supplementary reinforcing rotations. The withdrawal of the casing with
cage into the concrete. both clockwise and counter-clockwise rotations
The diameter of the displacement body (which produces a nearly smooth shaft.
is the same as the minimum diameter of the The diameter of the conical tip ranges from
pile shaft) typically ranges from 0.31 m to 0.56 0.45 m to 0.67 m (1.5 ft to 2.2 ft), while that of
m (1.00 ft to 1.84 ft), while that of the enlarged the casing ranges from 0.38 m to 0.52 m (15
helical flange ranges from 0.45 m to 0.81 m in to 20 in) (American Pile Driving Inc. 2007;
(1.48 ft to 2.66 ft) (Bustamante and Gianeselli Geoforum 2008). The length of the Fundex pile
1998; De Cock and Imbo 1994). The Atlas pile can reach up to 25-35 m (82-115 ft).
length can reach up to 22-25 m (72-82 ft). In
highly compressible soils or in soils with large Olivier Pile
cavities or voids, a thin-walled casing is often The installation of the Olivier pile is similar
attached to the screw head of the Atlas piles to that of the Atlas pile (Fig. 6). However, the
(Bustamante and Gianeselli 1998). The casing is drilling rigs used to install the Olivier piles
left in the ground with the sacrificial tip. are different from those of the Atlas piles (the
rotary drives are different; the Atlas pile rig
De Waal Pile has a bottom-type rotary drive with a fixed rate
The drilling tool used to install the De Waal pile of penetration, while the Olivier pile rig uses
consists of a sacrificial tip, a partial-flight auger a top-type rotary drive with variable rate of
and a displacement body (Fig. 4). The tool is penetration). A lost tip is attached to a partial-
attached to a casing. The partial-flight auger flight auger which, in turn, is attached to a
is closed at the bottom with the sacrificial tip. casing. The casing, which is rotated clockwise
To install the De Waal pile, the drilling tool is continuously, penetrates into the ground by
rotated clockwise to the required depth with a the action of a torque and a vertical force. At
torque and a vertical force. After reaching the the desired installation depth, the lost tip is
desired depth, concrete is placed into the casing released, and the reinforcing cage is inserted
to a level above the ground level, the sacrificial into the casing. Concrete is then placed inside
tip is released, and the tool is extracted using the casing through a funnel. The casing and the
clockwise rotation and a vertical force. Upward partial-flight auger are extracted by counter-
transport of soil during extraction is restricted clockwise rotation. Similar to the Atlas pile, the
due to the presence of reverse auger flights shaft of the Olivier pile has the shape of a screw.
above the displacement body. The concrete
level within the casing is maintained above the Omega Pile
ground level during extraction. A reinforcement In the case of the Omega pile, drilling is done
cage is typically installed after concrete by a displacement auger (with varying flange
placement. Unlike the Atlas pile, installation of diameter), which is closed at the bottom with
the De Waal pile creates a nearly smooth shaft. a sacrificial tip (Fig. 7). The flange diameter of
the auger segments increases gradually from
Fundex Pile both ends and becomes equal to the diameter
In the Fundex pile installation, a casing/tube of the central displacement body. A casing is
with a conical auger tip attached to its end is attached to the upper end of the displacement
rotated clockwise and pushed down into the soil auger. After reaching the required depth,
(Fig. 5). The joint between the casing and the concrete is injected under pressure, and the
conical tip is made watertight. As the casing sacrificial tip is released. The auger is slowly
penetrates into the ground, soil is displaced rotated clockwise and pulled up to produce a
laterally. In dense or hard layers, drilling can be nearly smooth shaft. The reinforcement cage
combined with grout injection or water jetting is then vibrated down into the fresh concrete.
through the conical tip. After the desired For some Omega piles, it is possible to place
depth is reached, the sacrificial conical tip, the reinforcement cage (or bar) into the drilling
which forms an enlarged pile base, is released. stem (casing) even before concrete is placed
The reinforcement cage is then inserted into (Bottiau et al. 1998).

[6] DFI JOURNAL Vol. 4 No. 1 August 2010

 3: Extraction of the casing with 4: Insertion of
1: Drilling with clockwise 2: Insertion of reinforcing
counter-clockwise rotation and supplementary
auger rotation and cage at end of drilling
vertical force reinforcement: completed
vertical force
Atlas pile with screw-
shaped shaft
Hydraulic
Ram

Casing

Displacement
Body

Sacrificial
Tip

[FIG. 3] Installation stages for the Atlas pile.

1: Drilling with clockwise 3: Concrete injection and 2: Extraction of casing with 4: Insertion of
auger rotation and release of sacrificial tip at clockwise rotation and reinforcing cage
vertical force the desired depth vertical force and completed De
Waal pile

Casing

Displacement
Body

Partial Auger

Sacrificial Tip

[FIG. 4] Installation stages for the DeWaal pile.

DFI JOURNAL Vol. 4 No. 1 August 2010 [7]

 1: Drilling with clockwise 2: At the desired depth, 3: Extraction of casing with 4: Completed
auger rotation and insertion of reinforcing cage, an oscillating upward and Fundex pile
vertical force release of sacrificial tip and downward motion and
placement of concrete into alternating 180° clockwise
the casing and counter-clockwise
rotations

Casing

Sacrificial
Conical Tip

[FIG. 5] Installation stages for the Fundex pile.

1: Drilling with clockwise 2: Insertion of reinforcing 3: Concrete pumping and 4: Completed Olivier pile
auger rotation and cage and release of sacrificial extraction of casing with with screw-shaped shaft
vertical force tip at the desired depth counter-clockwise rotation

Casing

Partial
Auger

Sacrificial
Tip

[FIG. 6] Installation stages for the Olivier pile.

[8] DFI JOURNAL Vol. 4 No. 1 August 2010

PGD Pile the pile base. Depending on the equipment
The PGD piling technology is a modification of available, some or all of the following quantities
the Auger Pressure-Grouted (APG) piling system can be measured or calculated during the
(Brettmann and NeSmith 2005). The APG pile installation of auger piles: the rate of auger
is a type of CFA pile which is constructed by rotation, the rate of auger penetration, the
pumping fluid grout under pressure during torque, the concrete pumping rate, and the auger
the withdrawal of the continuous-flight auger. extraction rate (Mandolini et al. 2002).
During the installation of a PGD pile (Fig. 8), the Automated monitoring systems are available
surrounding soil is displaced laterally as the for the drilled displacement pile rigs as well.
drilling tool is advanced into the ground. There These can be used to continuously monitor
are basically two types of PGD piles: 1) pressure- the depth of penetration, the vertical force, the
grouted with partial soil displacement and 2) torque, and the rate of auger/casing penetration
pressure-grouted with full soil displacement. and rotation. In the past, quality control (QC)
The PGD pile rigs are capable of producing of auger piles was performed mostly by field
both a torque and a downward crowd force, inspectors, based mainly on the industry
facilitating the drilling operations. Once the standards published by the Deep Foundations
desired depth is reached, high-strength grout is Institute (DFI) in the 1990’s (Brettmann 2003).
pumped under pressure through the drill stem Currently, automated systems are attached
and the drilling tool is withdrawn as it rotates to pile rigs throughout the world. Although
clockwise. The reinforcement cage is inserted these monitoring systems can provide valuable
into the grout column to complete the pile information on the integrity of the piles, they are
installation process. Note that the term Auger not meant to replace qualified field inspectors.
Pressure-Grouted Displacement (APGD) is also Automated QC monitoring techniques are
used in practice to refer to this type of DD pile based on measurements of either volume
(NeSmith 2002; Brettmann and NeSmith 2005). or pressure of the grout/concrete. Typical
The full-displacement PGD piles, which are automated systems measure: i) time, depth
typically installed in loose to medium dense and hydraulic pressure during drilling, and ii)
sands (corresponding to SPT blow count N < 25), time, depth, grout/concrete volume or grout/
can be 0.3-0.45 m (12-18 in) in diameter and concrete pressure during casting. Continuous,
up to 24 m (79 ft) in length (NeSmith 2002; real time graphs of relevant data are available to
Brettmann and NeSmith 2005). The diameter the operator during the installation of DD piles
of the partial-displacement PGD pile ranges (this facilitates any impromptu adjustments that
from 0.3-0.5 m (12-20 in). These piles reach up may be needed). These files can also be stored
to 17 m (79 ft) in length and are used in loose electronically for future reference (Bretmann and
to dense sands with N < 50 (NeSmith 2002; NeSmith 2005).
Brettmann and NeSmith 2005). NeSmith and NeSmith (2006a, 2006b, 2009)
described an automated data acquisition system
SVV Pile used for APGD piles and indicated that the
The SVV pile (STRABAG Vollverdrängungsbohrpfahl), recorded data can be useful in characterizing
developed by Jebens GmbH, is a large- the subsurface profile and in estimating pile
displacement DD pile (Fig. 9). The pile is capacity. This automated system measures
installed using a casing that has a segment the depth of penetration, the inclination of
with an enlarged diameter and a drill head. the mast of the drilling platform, the drilling
The SVV pile typically has a diameter of 0.44 stem rotation, the grout flow and pressure
m (18 in) and a length of up to 20 m (66 ft) (measured at the top of the drilling stem), and
(Geoforum 2008). the hydraulic pressure applied to the motor that
controls the rotation of the drilling tool. The
INSTALLATION MONITORING torque applied during drilling is calculated from
Continuous monitoring during the installation of the recorded hydraulic pressure. In addition,
auger piles is important to assure pile integrity. the drilling time is recorded through an internal
The data obtained through monitoring of the time counter in the main control unit of the
installation process also provide additional data acquisition system.
information on the subsurface condition and A specific energy or installation energy term
allow determination of the exact position of can be calculated from the variables mentioned

DFI JOURNAL Vol. 4 No. 1 August 2010 [9]

 1: Drilling with 3: Pressure injection of 4: Extraction of drilling tool 5: Insertion of
clockwise auger rotation concrete into the casing with clockwise rotation and reinforcement:
and vertical force and release of sacrificial concrete placement completed Omega pile
tip at desired depth

Displacement
Auger

Sacrificial
Conical Tip

[FIG. 7] Installation stages for the Omega pile.

1: Drilling with clockwise 2: Pressure injection of 3: Extraction of drilling tool 4: Insertion of

auger rotation and grout after reaching with clockwise rotation reinforcement:
vertical force the desired depth completed PGD pile

Drilling Drill
Tool Stem

[FIG. 8] Installation stages for the PGD pile.

[10] DFI JOURNAL Vol. 4 No. 1 August 2010

 1: Clockwise drilling 2: Insertion of reinforcing 3: Pumping concrete and extraction 4: Completed
using torque and cage after reaching the of casing with clockwise rotation SVV pile
vertical force desired depth and vertical force

Drill
Head

[FIG. 9] Installation stages for the SVV pile.

above and other machine-specific installation where the subscript i represents a given soil
parameters (Bottiau et al. 1998). The specific layer (i = 1, 2, 3, …) for which shaft capacity
energy along the depth of the pile can be is calculated; n is the total number of layers
correlated with in situ test results; it can crossed by the pile; qb and qsi are the unit base
potentially be used to interpret the effects of and shaft resistances; Ab (= π Db2/4) is the
pile installation and to help predict pile load representative pile base area; As (= π Ds) is the
capacity (De Cock and Imbo 1994). NeSmith pile shaft perimeter; Db and Ds are the nominal
(2003) also proposed that an installation diameters of the pile base and shaft respectively;
effort parameter (IE; defined as the product of and hsi is the thickness of the ith soil layer.
normalized values of torque and drilling tool According to the guidelines provided by
penetration rate) be used as an indicator of the Huybrechts and Whenham (2003), the nominal
capacity of DD piles. However, research on this shaft and base diameters depend on the drilling
topic is very limited and caution is necessary tool geometry. For the Atlas and Olivier piles, Db
when using these methodologies. and Ds are assumed to be equal to the measured
maximum diameter Df of the drilling auger screw
DESIGN METHODS blade (see Fig. 10). Bustamante and Gianeselli
(1993, 1998), however, suggested that the nominal
General Framework
diameter of the Atlas pile is equal to 0.9Df, except
The ultimate pile capacity Qult can be expressed as: for the thick-flanged Atlas piles, for which they
suggested a nominal diameter equal to Df. For
Qult = Qb,ult + QsL (1)
the Fundex pile, Db is equal to the measured
where Qb,ult and QsL are the ultimate base and maximum diameter of the conical auger tip, and
limit shaft capacities. These quantities are Ds is equal to the measured maximum diameter
calculated from: of the casing/tube (Huybrechts and Whenham
2003). For other DD piles that also have a nearly
Qb,ult = qbAb (2) smooth shaft, such as the De Waal and Omega
n piles, both Ds and Db are taken as the diameter
Q sL = A s ∑q si h si (3) of the soil displacement body (which is equal
i

DFI JOURNAL Vol. 4 No. 1 August 2010 [11]

 to the maximum diameter of the screw blade; design method, the ultimate pile load capacity
Huybrechts and Whenham 2003). Since the is based on a relative settlement of 6% (i.e., the
nominal pile diameter depends on the drilling load corresponding to a pile head settlement
tool geometry, the different coefficients proposed equal to 6% of the pile diameter). The unit base
(in different design methods) for pile capacity resistance qb is given by:
calculations reflect the way in which they were
determined, including the nominal dimensions of qb (MPa) = 0.4 qcm + wb for qcm ≤ 19 MPa (4a)
the pile. No specific guidelines are given in the or
literature on nominal diameter values for use in
the design of other types of DD piles. qb (MPa) = 0.19 Nm + wb for N m
≤ 50 (4b)

where qcm and Nm are representative values of

qc and uncorrected SPT blow count N in the
vicinity of the pile base, and wb is a constant
that depends on soil gradation and angularity.
For soils containing uniform, rounded particles
with up to 40% fines, wb = 0 and the upper
Db = Ds
Ds limit of qb is 7.2 MPa (1,044 psi). For soils
Df Df Db = Ds
Db with well-graded, angular particles having less
Atlas De Waal Fundex Olivier Omega than 10% fines, wb = 1.34 MPa (195 psi) and
the upper limit for qb is 8.62 MPa (1,250 psi).
[FIG. 10] Design dimensions for some DD piles.
Interpolation (based on percentage of fines)
is suggested to determine the values of wb for
Calculation of unit base and shaft resistances
other types of soils (NeSmith 2002). qcm and Nm
Available design methods for DD piles are are determined from the following equations
mostly based on in situ test results. The unit (Fleming and Thorburn 1983):
base and shaft resistances of piles are typically
related to the cone penetration test (CPT) tip qcm = 0.25qc0 + 0.25qc1 + 0.5qc2 (5a)
resistance qc, the standard penetration test
(SPT) blow count N and the pressuremeter test Nm = 0.25N0 + 0.25N1 + 0.5N2 (5b)
(PMT) limit pressure pl.
where qc0 and qc1 are the average and minimum
Method A cone resistances over a length of 4Db below the
pile base, respectively, and qc2 is the average
This design methodology was developed in
cone resistance over a length of 4Db above
the U.S. based on load tests performed on 28
the pile base after eliminating values greater
APGD piles (NeSmith 2002; Brettmann and
than qc1 (NeSmith 2002). N0, N1 and N2 refer
NeSmith 2005). The Geotechnical Engineering
to the corresponding uncorrected SPT values
Circular No. 8 (published by Federal Highway
(equivalent to qc0, qc1, and qc2).
Administration) recommends this method
for the calculation of the axial capacity of The unit shaft resistance for any soil layer i is
DD piles in the U.S. (Brown et al. 2007). The given by:
ultimate load (defined as the ‘interpreted failure qsi (MPa) = 0.01 qci + ws for qci ≤ 19 MPa (6a)
load’ by NeSmith 2002) was defined as the
minimum of the loads corresponding to (i) a or
pile head settlement of 25.4 mm (1 in) or (ii) a qsi (MPa) = 0.005 Ni + ws for Ni ≤ 50 (6b)
pile displacement rate of 0.057 mm/kN (0.02
inch/ton). The specified value of the pile head where ws is a constant similar to wb, qci is
settlement (i.e., 25.4 mm = 1 inch) is equal to the CPT cone resistance for soil layer i, and
about 6% of the diameter of the piles tested [pile Ni is the uncorrected SPT blow count for soil
diameters ranged from 0.36 m to 0.46 m (14- layer i. For soils containing uniform, rounded
18 in), with 80% of the piles having a diameter particles with up to 40 % fines, ws = 0 and the
equal to 0.41 m (16 in). According to NeSmith limiting value of qsi is 0.16 MPa (23 psi). For
(2002), the settlement-based criterion (pile head soils with well-graded, angular particles having
settlement equal to 25.4 mm or 1 in) controlled less than 10 % fines, ws = 0.05 MPa (7 psi) and
the determination of the ultimate load (or the the limiting value of qsi is 0.21 MPa (30 psi).
‘interpreted failure load’). Therefore, in this Interpolation of ws is suggested for intermediate

[12] DFI JOURNAL Vol. 4 No. 1 August 2010

soils. This shaft capacity calculation method between 0.7qca and 1.3qca, and for the zone
is recommended only for sandy soils, where below the pile base, an upper bound of 1.3qca is
pile installation results in soil densification. applied, and (iv) the arithmetic mean qce value is
Brettmann and NeSmith (2005) recommended calculated from the qce profile obtained in (iii).
the use in Eqs. (5b) and (6b) of energy-corrected To estimate the unit shaft resistance qsi, a
SPT blow count N60 values. design curve (Q1, Q2, Q3, Q4 or Q5) is first
selected depending on the soil type and the
Method B
guidelines given in Table 3. Fig. 11 is then used
Bustamante and Gianeselli (1993, 1998) to estimate qsi from the design curve selected.
developed a design method based on the
results of 24 load tests on Atlas piles. They [TABLE 2] Values of α and a for different in situ
defined the ultimate pile load capacity as the tests (Bustamante and Gianeselli 1998).
load corresponding to 10% relative settlement
(i.e., the load corresponding to a pile head
settlement equal to 10% of the pile diameter).
According to this method, the unit base
resistance is given by:

qb = Kα (7)
where K is a coefficient that depends on the a
The factor 1000 is used to maintain
soil type (Table 1), and α represents an average consistency between units.
of the in situ test results within an influence
zone extending from a distance a above to The qc value used to develop this method
a below the pile base (Table 2). For the SPT- was obtained from penetration tests using an
based design, the parameter α is the average M1-type mechanical cone. When an electrical
(geometric mean) of N1, N2 and N3 (see Table 2). CPT cone is used, a correction factor β was
For the PMT-based design, the parameter α is recommended:
the average (geometric mean) of pl1, pl2 and pl3 qc,mech = β qc,elec (8)
(see Table 2). To obtain α from a CPT profile,
the in situ qc profile is modified within the where qc,mech is the cone resistance measured
influence zone. This is done in four successive with a mechanical cone, and qc,elec is the cone
stages: (i) the in situ qc profile is smoothened to resistance measured with an electrical cone.
remove local irregularities within the influence The coefficient β is in the 1.4-1.7 range for
zone, (ii) an arithmetic mean qca is calculated clayey soils and is equal to 1.3 for saturated
within the influence zone, (iii) a qce profile is sands (Bustamante and Gianeselli 1993).
obtained within the influence zone by applying
[TABLE 3] Guidelines for selection of a design
bounds to the minimum and maximum
curve to estimate qs from Fig. 11 (Bustamante
resistances in the qc profile: for the zone above and Gianeselli 1998).
the pile base, the resistance values are clipped
Limit pressure Cone
Curves
Soil Type from PMT Resistance
[TABLE 1] Values of K for different soil types (MPa) (MPa) C M
(Bustamante and Gianeselli 1998).
Clay
/Clayey < 0.3 < 1.0 Q1 Q1
In situ Tests
Soil Type Silt > 0.5 > 1.5 Q3 Q2
PMT CPT SPT /Sandy ≥ 1.0 ≥ 3.0 Q4 Q2
Clay
Clay 1.6-1.8 0.55-0.65 0.9-1.2
< 0.3 < 1.0 Q1 Q1
Sand 3.6-4.2 0.50-0.75 1.8-2.1 Sand /
> 0.5 > 3.5 Q4 Q2
Gravel
≥ 1.2 > 8.0 Q5 Q2
1
Gravel ≥ 3.6 ≥ 0.5 –
< 1.2 < 4.0 Q4 Q2
Marl
1
Marl 2.0-2.6 ≥ 0.7 ≥ 1.2 ≥ 1.5 ≥ 5.0 Q5 Q2
1
Chalk ≥ 2.6 ≥ 0.6 ≥ 2.6 > 0.5 > 1.5 Q4 Q2
Chalk
≥ 1.2 > 4.5 Q5 Q2
1
Conservative values are reported due to C = Cast-in-place screw piles, M = Screw piles with
inadequacy of test results lost casing

DFI JOURNAL Vol. 4 No. 1 August 2010 [13]

 proposed by De Beer (De Beer 1971; Van Impe
1986; Van Impe et al. 1988).
According to the recent guidelines presented
in the Application of Eurocode 7 in the Belgian
practice (Application de l’Eurocode 7 en
Belgique, 2008), the factor λ = 1 for all the DD
piles considered in this paper, except for the
Fundex pile. For the Fundex pile, the value of λ
is obtained from :

⎧ 2
⎛ Db ⎞
⎪ = 1.0 ; ⎜ ⎟ ≤ 1.5
⎪ ⎝ Ds ⎠
⎪
⎪ ⎡⎛ D ⎞ 2 ⎤ ⎛D ⎞
2
(10)
λ ⎨ = 1− 0.429 ⎢⎜ b ⎟ − 1⎥ ; 1.5 <⎜ b ⎟ < 1.7
⎣⎢⎝ s ⎠ ⎦⎥
⎪ D ⎝ Ds ⎠
[FIG. 11] Values of unit shaft resistance qs as a function of pl, qc ⎪ 2
or N (Bustamante and Gianeselli 1993, 1998). ⎪ ⎛ Db ⎞
⎪= 0.7; ⎜ ⎟ ≥ 1.7
⎩ ⎝ Ds ⎠
Method C
The factor αb varies between 0.7 and 0.8 (see
In the Belgian pile design practice, the capacity Table 4). The coefficient εb = max [1.0–0.01(Db/
of DD piles is calculated using empirical dCPT–1); 0.476] for stiff, fissured tertiary clay,
expressions that were developed based mainly while, for all other soil types, εb = 1.0.
on CPT and pile load test results (Van Impe
The unit shaft resistance qsi for the ith soil layer
1986, 1988, 2004; Bauduin 2001; Holeyman
is related to the average cone resistance qci
et al. 2001; De Vos et al. 2003; Maertens
(obtained using a standard electrical cone) of
and Huybrechts 2003a). The design practice
that layer by:
for DD piles was strongly influenced by the
results of the pile load tests performed at the qsi = αsiηp*qci (11)
Sint-Katelijne-Waver and Limelette test sites
(Holeyman 2001; Maertens and Huybrechts where αsi and ηp* are empirical factors. αsi
2003b; Van Impe 2004); these load tests were depends on the method of installation in a
supported by the Belgian Building Research particular soil and the roughness of the pile
Institute (BBRI). This method is applicable shaft (see Table 4). Table 5 shows the values
to all types of DD piles. The current Belgian of ηp*, which are a function of soil type and
practice follows the guidelines developed for qci. Beyond a certain value of qci, a maximum
the implementation of Eurocode 7 (Application design value is prescribed for qsi (Table 5). Note
de l’Eurocode 7 en Belgique 2008). The ultimate that, in the shaft capacity calculations, the
unit base resistance corresponding to 10%
[TABLE 4] Values of αb and αsi for use in Eqs. (9)
relative settlement is given by:
and (11) (Application de l’Eurocode 7 en Belgique
qb = λαbεbqb,CPT (9) 2008).

where λ is a reduction factor accounting for αb αsi

what was referred to as the “soil relaxation”
Pile Types
that may take place around the shaft during Tertiary Other Tertiary Other
the drilling process due to the presence of an Clay Soils Clay Soils
enlarged base, αb is an empirical factor that
accounts for the pile installation technique and Piles cast in
soil type, εb is a scaling coefficient (accounting situ using 0.8 0.7 0.9 1.0
for fissuring of the soil) expressed as a function concrete
of the ratio of the diameter of the pile base Db
to that of the standard electrical CPT cone dCPT
(= 35.7 mm), and qb,CPT is the representative base Piles cast
using lost 0.8 0.8 0.6 0.6
resistance calculated from a CPT resistance casing
qc profile obtained according to the method

[14] DFI JOURNAL Vol. 4 No. 1 August 2010

contributions of soil layers with qci < 1MPa (145 Test site at Limelette, Belgium
psi) are neglected. The pile load test site at Limelette, Belgium,
The Belgian practice for DD piles, as described consists of a silty and sandy clay layer down to
above, relies on qc values obtained using a depth of 8.2 m (27 ft); this layer is underlain
an electrical cone. A reduction factor ω is by a clayey sand layer (Van Alboom and
suggested (see Table 6) for qc values obtained Whenham 2003). The water table at the site
from CPTs performed in tertiary clay using a is located at a depth well below the base of
mechanical cone (i.e., qc,elec = qc,mech/ω). the test piles. Five different types of DD piles
were installed and subjected to static load
[TABLE 5] Values of ηp* for use in Eq. 11 tests (SLTs). The pile geometries, (Db, Ds and
(modified after Application de l’Eurocode 7 en length Lp), as obtained from Huybrechts and
Belgique 2008). Whenham (2003), are given in Table 7. All the
Average test piles, except the Fundex piles, have the
Cone Maximum same nominal shaft and base diameters (Ds and
Soil Type
Resistance ηp* qsi (MPa) Db) for calculation of base and shaft resistances
qci (MPa)
(the nominal design diameters were selected
0.150 for following the guidelines described previously).
Clay 1-4.5 0.0333 qci > 4.5
MPa We used the average cone resistance profiles
[obtained from CPTs done with an electrical
0.100 for cone (for use in Methods A and C) and a
Silt 1-6 0.0167
qci > 6 MPa
mechanical M1-type cone (for use in Method B)],
Sandy Silt/ reproduced in Fig. 12, to calculate the ultimate
Clay pile capacities. Fig. 12 also shows the qb,CPT
0.125 for profile (calculated using De Beer’s Method)
or 1-10 0.0125
qci > 10 MPa
Clayey Silt/ for use in design method C. The calculated
Sand ultimate capacities of the test piles are given
in Table 8, which also includes the reported
1-10 0.0111 -
ultimate capacities of the piles obtained from
0.110 + the SLTs (Maertens and Huybrechts 2003a).
Sand 10-20 0.004 For piles A2, B3, B4, C1 and C2, the SLTs could
(qci – 10) not be continued up to a pile head settlement
> 20 - 0.150
of 10% of the pile diameter. For these piles,
Chin’s method of extrapolation (Chin 1970)
was used to extend the load-settlement curves;
[TABLE 6] Reduction factor ω (Application de the ultimate capacities of these piles were
l’Eurocode 7 en Belgique 2008).
Type of Mechanical Tertiary Other Types [TABLE 7] Pile geometries (Huybrechts and
Cone Clay of Soil Whenham 2003).
M1 1.3 1.0
Ds Db Lp
M2 1.3 1.0 Pile Tag Pile Type
(m) (m) (m)
M4 1.15 1.0
B3
Atlas 0.51 0.51 9.43
CAPACITY CALCULATIONS USING B4
THE DIFFERENT DESIGN METHODS A4
DeWaal 0.41 0.41 9.53
We selected the soil profiles of two well C4
documented pile load test sites to evaluate
the different methods of pile capacity A1 9.59
Fundex 0.39 0.45
calculation described above. The first test C1 9.65
site is at Limelette, Belgium; this site was used
A2 9.20
for the load test program supported by the Olivier 0.55 0.55
BBRI. The second test site is located at the C2 9.13
Georgia Institute of Technology campus; this A3
test site was used for a load test program on Omega 0.41 0.41 9.45
C3
drilled shafts.
DFI JOURNAL Vol. 4 No. 1 August 2010 [15]
 obtained from the extrapolated curves as the settlement curve. For the Atlas, Omega and
loads corresponding to a pile head settlement Fundex A1 piles, the total ultimate capacities
equal to 10% of the pile diameter (Maertens and estimated using method C were smaller (by
Huybrechts 2003a). 11%, 13% and 14%, respectively) than the SLT
As can be seen in Table 8, the smallest base capacities. For the DeWaal piles, the total
capacity estimates were obtained with Method A ultimate capacities obtained with method C
(originally developed for APGD piles). Note that were larger (by 5%) than the SLT capacities.
method A could not be used for shaft capacity
Test site at Georgia Institute of Technology
calculations at this site because the test piles
were installed mostly in clayey soils, for which We used the residual soil profile at the Georgia
method A is not applicable. Base capacity Institute of Technology test site to calculate
estimates obtained with method B, (originally and compare the capacities of a DD pile (using
developed for the Atlas pile) were larger (by design methods A, B and C), a full-displacement
26-38%) than those calculated using method pile and a nondisplacement pile. Subsurface
C. However, the shaft capacity estimates of information for this site is available from the
methods B and C were in good agreement results of in situ and laboratory tests performed
(within 3%). to characterize the test site (FHWA Technical
The total ultimate capacities calculated Report 1993). The subsurface at this site
using method B are consistently higher than consists of a silty sand (SM) layer extending
the capacities obtained from the SLTs; the down to depths ranging from 15.8 m to 19.7 m
maximum difference was obtained for the (52 to 65 ft); this silty sand layer is underlain by
DeWaal and Olivier piles (for these two pile a partially weathered rock bed. A fill layer 0.6-
types, the total ultimate capacities estimated 3.7 m (2-12 ft) thick comprised mostly of silt
using method B were larger than the SLT and sand, is present above the silty sand layer.
capacities by 17% and 20%, respectively). For The ground water table was recorded (at the
the Olivier pile A2, the total ultimate capacity time of site characterization) at depths ranging
calculated using method C is in good agreement from 16.7 m to 19.1 m (55.0 to 62.7 ft) from
(the difference is ~1%) with the ultimate the ground surface. Particle size analysis of the
capacity obtained from the extrapolated load- collected samples revealed that the site consists

[TABLE 8] Ultimate capacities of different drilled displacement piles at the Limelette test site, Belgium.

Calculated Base Calculated Shaft Calculated Total

Capacity (kN) Capacity (kN) Capacity (kN) Capacity
Pile
Pile Type Obtained from
Tag Different Design Methods SLTs (kN)
A B C B C B C

B3 a
3528
Atlas 1220 2160 1460 1648 1671 3808 3131
B4 a
3454

A4 2400
DeWaal 832 1456 1079 1351 1363 2807 2442
C4 2248

A1 1776 1312 1300 1309 3076 2621 2988

Fundex 992
C1 1808 1344 1313 1318 3121 2662 a, b
1778

A2 1440 2496 1560 1714 1757 4210 3317 a

3354
Olivier
C2 1416 2400 1488 1694 1742 4094 3230 a, b
2908

A3 2786
Omega 806 1456 1079 1333 1350 2789 2429
C3 2723
a
Values (corresponding to a pile head settlement equal to 10% of the pile diameter) were obtained from
extrapolated load-settlement curves (Maertens and Huybrechts 2003a)
b
Low ultimate capacities of C1 and C2 are attributed to the segregation of concrete and structural
rupture (Maertens and Huybrechts 2003a)
[16] DFI JOURNAL Vol. 4 No. 1 August 2010
 Cone Resistance (MPa) Cone Resistance (MPa)
0 5 10 15 20 25 0 2 4 6 8 10
0 0
Average qc (using M1 type cone)
profile for use in Method A
Average qc (electrical CPT) profile
for use in Method C 4
4 qb,CPT profile for piles A3, A4, C3, and C4
qb,CPT profile for piles A1 and C1
qb,CPT profile for piles B3 and B4
qb,CPT profile for piles A2 and C2 8

Depth (m)
Depth (m)

12
12 Average qc profile
qb,CPT profile (for DD pile)

16

16

20
[FIG. 12] Average cone resistance and qb,cpt profiles at the [FIG. 13] Average cone resistance and qb,CPT profiles at the
Limelette test site. Georgia Institute of Technology test site.

[TABLE 9] Calculated capacities for a DD pile, a full-displacement pile, and a nondisplacement pile (soil
profile of the Georgia Institute of Technology test site).

Shaft Capacity (kN) Base Capacity (kN) Total Capacity (kN)

Pile Types
Method Method Method

A B C A B C A B C
DD Pile
596 955 668 236 459 387 832 1414 1055

Schmertmann (1978) Aoki and Velloso (1975) -

Full-displacement
Pile
704 427 1131

Lopes and Laprovitera (1988) Franke (1989) -

Nondisplacement
Pile
251 151 402

of mostly uniform sand particles (median D50 = and nondisplacement piles in sand. Table 9
0.14 mm) with 33% fines. The average total unit shows the calculated capacities for all these
weight assumed in calculations was 19.2 kN/ piles. It is interesting to note that the capacity of
m3 (3,300 lb/cu yd) (FHWA Technical Report the DD pile calculated using method B is larger
1993). Fig. 13 shows an average CPT profile of than that of the full-displacement pile with the
this site; this figure also shows the qb,CPT profile same geometry. The DD pile capacity obtained
(calculated using De Beer’s Method) for use in with method A, however, lies between the
design method C. capacities calculated for the full-displacement
The DD, full-displacement, and and nondisplacement piles (this is in agreement
nondisplacement piles were assumed to be with the notion that the soil displacement
10 m (33 ft) long with nominal base and shaft produced during the installation of a DD pile
diameter equal to 0.4 m (15.75 in). We used is within the range of that of a partial- to
CPT-based methods (Aoki and Velloso 1975; that of a full-displacement pile). The DD pile
Schmertmann 1978; Lopes and Laprovitera capacity calculated using method C matches
1988; and Franke 1989) to calculate the base closely the capacity of the full-displacement
and shaft capacities of the full-displacement pile calculated with the Aoki and Velloso (1975)

DFI JOURNAL Vol. 4 No. 1 August 2010 [17]

 and Schmertmann (1978) methods. Note that test results can help improve the prediction
general conclusions can not be reached based capability and consistency of design methods
on the calculations presented in Table 9; they for DD piles. These load tests should be
provide only a site-specific comparison of extended to large pile settlements (certainly in
predicted capacities. excess of 10% of the pile diameter), the piles
should preferably be instrumented (so that, at
SUMMARY AND CONCLUSIONS a minimum, base and shaft resistances may
Drilled displacement piles are increasingly be separated) and the test sites must be well
used in geotechnical practice. The advantages characterized. Modeling of the pile installation
of these piles are that their construction is process in conjunction with well designed field
fast, economical and environmentally friendly. load tests and systematic monitoring of pile
Depending on the method of installation, DD installation is needed for meaningful advances
piles can be classified as partial-displacement in the analysis and design of DD piles.
piles, with capacities sometimes approaching
that of full-displacement piles. ACKNOWLEDGMENTS
Pile capacity calculation methods do not Dr. Irem Zeynep Yildirim assisted with the
always predict field capacities with acceptable drafting of some of the figures presented in
accuracy. One of the reasons for the difficulty this paper. The authors greatly appreciate her
in making good predictions is that the degree helpful efforts. The authors also thank the
of soil disturbance caused by pile installation reviewers for their valuable comments.
cannot be assessed properly in the field.
Different DD pile installation methods (with REFERENCES
different drilling tools) change the soil state 1. American Pile Driving, Inc.
differently, leading to different pile load- <americanpiledriving.com > as seen on
carrying capacities. Additionally, for the same March 2, 2007.
degree of soil disturbance, a screw-shaped
2. Aoki, N., and Velloso, D. A. 1975. An
shaft may develop a larger shaft capacity than
Approximate Method to Estimate the
a smooth shaft. The design methods described
Bearing Capacity of Piles. Proceedings of
in this paper were developed based on pile
the 5th Pan-American Conference of Soil
load tests performed at particular test sites.
Mechanics and Foundation Engineering,
Consequently, these methods have biases and
Buenos Aires, Vol. 1, pp. 367–376.
may not be applicable to other sites without
proper calibration. There is also the need for 3. Application de l’Eurocode 7 en Belgique :
the design methods to be more precise, going Directives pour le dimensionnement en ELU
beyond just textbook soils (sand and clay). de pieux sous charge axiale en compression,
Version de mars 2008.
In order to illustrate the capabilities of
currently available design methods, we used 4. Bauduin, C. 2001. Design procedure
these methods to estimate the capacities of according to Eurocode 7 and analysis of the
the DD piles load-tested at the Limelette test test results. Screw Piles – Installation and
site in Belgium. Additionally, we compared Design in Stiff Clay, Holeyman (ed.), Swets
the capacities of DD, full-displacement and and Zeitlinger, Lisse, pp. 275 – 303.
nondisplacement piles for a residual soil 5. Bottiau, M., 2006. Recent evolutions in deep
profile of granite. The comparisons of the foundation technologies. Proceedings of
calculated and measured pile capacities show the DFI/EFFC 10th International Conference
that improvements in the design methods are on Piling and Deep Foundations, 2006,
necessary. In particular, future improvement Amsterdam, The Netherlands.
of DD pile design methods should include (1)
6. Bottiau, M., Meyus, I. and Callens, S. 2008.
parameters that reflect the pile installation
Screw-in energy measurement for on-site
method and their impact on the state of the
control of the bearing capacity of Omega
soil around the pile; (2) interaction of the pile
piles. Proceedings of the 7th International
and soil in a way that reflects the stress-strain
Conference and Exhibition on Piling and
response of the soil; (3) limit states that must
Deep Foundations, 2008, Vienna, Austria.
be prevented. Development of a database
containing in situ test results (performed 7. Brettmann, T. and NeSmith, W., 2005.
before and after pile installation) and pile load Advances in auger pressure grouted piles:
[18] DFI JOURNAL Vol. 4 No. 1 August 2010
 design, construction and testing. Advances 17. De Cock, F. and Imbo, R., 1994. Atlas screw
in Designing and Testing Deep Foundations. pile: a vibration-free, full displacement,
Geotechnical Special Publication No. 129, cast-in-place pile. Transportation Research
ASCE, pp. 262-274. Record 1447, pp 49-62.
8. Brettmann, T., 2003. Constructibility of 18. FHWA Technical Report No. 41-30-2175.
augured cast-in-place piles. Geo-Strata, 8 – 11. 1993. Axial load-displacement behavior
9. Brown, D. A., 2005. Practical considerations of drilled shaft foundations in piedmont
in the selection and use of continuous flight residuum.
auger and drilled displacement piles. Advances 19. Fleming, W. G. K. and Thorburn, S., 1983.
in auger pressure grouted piles: design, Recent piling advances, state of the art
construction and testing. Advances in Designing report. Proceedings of the International
and Testing Deep Foundations. Geotechnical Conference on Advances in Piling and
Special Publication No. 129, ASCE, pp. 251-261. Ground Treatment for Foundations, ICE,
10. Brown, D. A., Dapp, S. D., Thompson, W. London, pp 1-16.
R. and Lazarte, C.A. 2007. Design and 20. Franke, E. 1989. Co-report to discussion,
construction of continuous flight auger session 13: large-diameter piles. 12th
piles. FHWA Geotechnical Engineering International Conference on Soil Mechanics
Circular No. 8. FHWA. and Foundation Engineering, Rio de
11. Brown, D. and Drew, C., 2000. Axial Janeiro.
capacity of augured displacement piles at 21. Geoforum <http://www.geoforum.com/info/
Auburn University, New Technological and pileinfo/class_list.asp?Method=4> as seen on
Design Developments in Deep Foundations, November 25, 2008.
Proceedings of sessions of Geo- Denver 2000, 22. Holeyman, A. E. 2001. Screw piles –
Geotechnical Special Publication No. 100, installation and design in stiff clay.
ASCE, pp. 397-403. Proceedings of the Symposium on Screw
12. Bustamante, M. and Gianeselli, L., 1993. Piles, Brussels, Belgium. Swets and Zeitlinger
Design of auger displacement piles from in- B. V., Lisse, The Netherlands.
situ tests. Deep Foundations on Bored and 23. Holeyman, A., Bauduin, C., Bottiau, M.,
Auger Piles, BAP II, Balkema, Rotterdam, pp. Debacker, P., De Cock, F. A., Dupont, E.,
21-34. Hilde, J. L., Legrand, C., Huybrechts, N.,
13. Bustamante, M. and Gianeselli, L., 1998. Mengé, P., Miller, J. P., and Simon., G. 2001.
Installation parameters and capacity Design of axially loaded piles – 1997 Belgian
of screwed piles. Deep Foundations on practice. Screw Piles – Installation and
Bored and Auger Piles, BAP III, Balkema, Design in Stiff Clay, Holeyman (ed.), Swets
Rotterdam, pp. 95-108. and Zeitlinger, Lisse, pp. 63 – 88.
14. Chin, F. V. 1970. Estimation of ultimate 24. HuyBrechts, N. and Whenham V., 2003. Pile
load of piles not carried to failure. Proc. testing campaign on the Limelette test site
2nd Southeast Asian Conference on Soil and installation techniques of screw piles.
Engineering, 81-90. Belgian Screw Pile Technology – Design and
15. De Beer, E. Méthodes de déduction de Developments, Maertens and HuyBrechts (eds.),
la capacité portante d’un pieu à partir Swets and Zeitlinger, Lisse, pp. 71 – 130.
des résultats des essais de pénétration. 25. Lopes, F. R., and Laprovitera, H. (1988). On the
Bruxelles, Journal des Travaux publics de Prediction of the Bearing Capacity of Bored
Belgique, volume 72, no 4 (p. 191-268), no 5 Piles from Dynamic Penetration Tests. Deep
(p. 321-353) & no 6 (p. 351-405), 1971-1972. Foundations on Bored and Auger Piles, W. Van
16. De Vos, M., Bauduin, C. and Maertens, J. Impe (ed.), Balkema, Rotterdam, pp. 537–540.
2003. The current draft of the application 26. Maertens, J. and Huybrechts, N., 2003a.
rules of Eurocode 7 in Belgium for the Results of the static pile load tests at
design of pile foundations. Belgian Screw the Limelette test site. Belgian Screw Pile
Pile Technology – Design and Developments, Technology – Design and Developments,
Maertens and HuyBrechts (eds.), Swets and Maertens and HuyBrechts (eds.), Swets and
Zeitlinger, Lisse, pp. 303 – 325. Zeitlinger, Lisse, pp. 167 – 214.

DFI JOURNAL Vol. 4 No. 1 August 2010 [19]

 27. Maertens, J., and Huybrechts, N. 2003b. Belgian 37. Van Alboom, G. and Whenham, V.,
screw pile technology design and recent 2003. Soil investigation campaign at
developments. Proceedings of the 2nd Symposium Limelette (Belgium): results, Belgian Screw
on Screw Piles, Brussels, Belgium. Swets and Pile Technology – Design and Recent
Zeitlinger B. V., Lisse, The Netherlands. Developments, Maertens and HuyBrechts
28. Mandolini, A., Ramodini, M., Russo, G. and (eds.), Swets and Zeitlinger, Lisse, pp. 21 – 70.
Viggiani, C., 2002. Full scale loading tests 38. Van Impe, W. F. (1986). Evaluation
on instrumented CFA piles. Proceedings of of deformation and bearing capacity
the International Deep Foundations Congress parameters of foundations, from static
2002, Geotechnical Special Publication No. CPT-results. Proceedings of the Fourth
116, Vol. 2, ASCE, pp. 1088-1097. International Geotechnical Seminar: Field
29. NeSmith, W. M., 2002. Static capacity Instrumentation and In Situ Measurements,
analysis of augured, pressure-injected NTI, Singapore, pp 51-70.
displacement piles. Proceedings of the 39. Van Impe, W. F. (1988). Considerations in
International Deep Foundations Congress the auger pile design. Proceedings of the 1st
2002, Geotechnical Special Publication No. International Geotechnical Seminar on Deep
116, Vol. 2, ASCE, pp. 1174-1186. Foundations on Bored and Auger Piles, BAP I,
30. NeSmith, W. M., 2003. Installation effort Balkema, Rotterdam, pp. 193-217.
as an indicator of screw pile capacity. 40. Van Impe, W. F. (2004). Two decades of full
Proceedings of Deep Foundations on Bored scale research on screw piles : An overview.
and Augered Piles (BAPIV), Van Impe (ed.), Published by The Laboratory of Soil
Rotterdam: Millpress, pp. 177-181. Mechanics, Ghent University, Belgium.
31. NeSmith, W. M., and NeSmith, W. M. 2006a. 41. Van Impe, W. F., De Beer, E., and Lousberg, E.
Anatomy of a data acquisition system for (1988). Prediction of the single pile bearing
drilled displacement piles. Proceedings capacity in granular soils out of CPT results.
of the GeoConGress 2006: Geotechnical Proceedings of the International Symposium
engineering in the information technology on Penetration Testing (ISOPT I), Speciality
age, ASCE, pp. 1-6. Session, Orlando, pp 1-34.
32. NeSmith, W. M., and NeSmith, W. M. 2006b.
Application of data acquired during drilled
displacement pile installation. Proceedings of the
GeoConGress 2006: Geotechnical engineering in
the information technology age, ASCE.
33. NeSmith, W. M., 2003. Advancements in
data acquisition based design for drilled
displacement piles. Contemporary Topics
in Deep Foundations: Geotechnical Special
Publication No. 185, ASCE, pp. 447-455.
34. Prezzi, M. and Basu, P. 2005. Overview of
construction and design of auger cast-in-place
and drilled displacement piles. Proceedings
of DFI’s 30th annual conference on deep
foundations, Chicago, U.S.A., pp. 497 – 512.
35. Siegel, T. C., NeSmith, W. M., NeSmith, W. M.,
and Cargill, P. E. 2007. Ground improvement
resulting from installation of drilled
displacement piles. Proceedings of DFI’s
32nd annual conference on deep foundations,
Colorado Springs, U.S.A., pp. 129-138.
36. Schmertmann, J.H. 1978. Guidelines for
Cone Penetration Test, Performance and
Design. U.S. Department of Transportation,
FHWA-TS-78-209.

[20] DFI JOURNAL Vol. 4 No. 1 August 2010

Analysis of a Deep Excavation In Calgary, Alberta
Thomas Lardner, Geotechnical Research Centre, University of Western Ontario, London, Ontario,
Canada; 519-661-2139; [email protected]
Matthew Janes, Isherwood Associates, Mississauga, Ontario, Canada
K.Y. Lo, Geotechnical Research Centre, University of Western Ontario, London, Ontario, Canada
Guangfeng Qu, Geotechnical Research Centre, University of Western Ontario, London, Ontario, Canada
Silvana Micic, Geotechnical Research Centre, University of Western Ontario, London, Ontario, Canada

ABSTRACT
Excavation for The Bow, EnCana, in downtown Calgary, occurred between May 2007 and September
2008. A monitoring programme was used to determine the behaviour of the shoring wall, soil
and bedrock during the excavation for the 21 metre (69 ft) deep excavation with a footprint of
approximately 17,000 m2 (183,000 sq ft). Results from the inclinometer readings indicated that
significant horizontal movement due to a weak rock layer and shear band effect. These modes of
movement account for up to 100 mm (4.0 in) and 45 mm (1.8 in) of lateral deflection, respectively.

Based upon the results of the monitoring programme, rock samples were collected from the weak
mudstone bedrock, with great difficulty, to determine the strength and deformation parameters.
Results from the uniaxial compression tests indicate a UCS around 830 kPa (120 psi) and elastic
modulus between 80 and 180 MPa (11 and 26 ksi). Mohr-Columb parameters were determined to be c'
= 340 kPa (49 psi) and φ' = 24°. Residual strengths were determined to be c'r = 0 and φ'r = 15°.

Using the results of the laboratory tests and field investigation records, an initial finite element
analysis was conducted. Results of the analysis show good correlation when compared with the
observed monitoring deformation, indicating accurate portrayal of rock parameters. Additional
analysis and investigation is required to determine the principal horizontal stresses in magnitude and
direction.

INTRODUCTION
The Bow office complex in downtown Calgary,
Alberta, will consist of a 54-storey office
complex with six levels of underground
parking. Installation of the shoring system and
excavation for the six storeys of parking began
in May 2007 and was completed in September,
2008. The excavation has a footprint of
approximately 17,000 m2 (183,000 sq ft), a
shored face of 13,200 m2 (142,000 sq ft), and
a depth of 20.5 metres (67.3 ft). Fig. 1 shows
the location of the site in Calgary, between 5th
and 7th Avenues at Centre Street. Neighbouring
the excavation are the Telus building to the
north, the PetroCanada Centre to the west,
and the Historic Royal Canadian Legion No.
1 and Calgary Light Rail Transit line to the
south. Given the proximity of the neighbouring [FIG. 1] Google Earth™ mapping service image with The Bow
job site outlined in red, the PetroCanada Centre (1),Telus
structures, horizontal movement during Centre (2), and Historic Royal Canadian Legion (3).
excavation was a significant concern.

DFI JOURNAL Vol. 4 No. 1 August 2010 [21]

 General site stratigraphy consisted of a thin internal angle of friction of 24°, an elastic
layer of fill overtop 6 to 7 metres (20 to 23 modulus between 80 and 170 MPa (11 and 25
ft) of well graded gravel and cobble. The ksi), and has some swelling potential. The
bedrock, part of the Porcupine Hills Formation, Geotechnical Research Centre also conducted
is composed of layered mudstone, siltstone, additional analysis of the monitoring results,
and sandstone. Large horizontal movements and compared these results to preliminary
in the bedrock had been observed during finite element modeling using parameters taken
previous deep excavations in Calgary. An from the post-construction testing.
extensive monitoring programme was therefore
incorporated into the design of the support GEOTECHNIQUE
system so that appropriate measures could AMEC Earth and Environmental performed
be taken in case large horizontal movement geotechnical investigations at the site in 2005
occurred. and 2006. The ground surface at the site is
The shoring design consisted of an anchored generally level, with elevations ranging between
secant caisson wall embedded a minimum of 1045.5 and 1046.3 metres (3,430 and 3,433 ft).
2 m (6.6 ft) into the bedrock with shotcrete The site consists of 0 to 1.0 m (0 to 3.3 ft) of fill
shoring on the rock face below. A typical over 5.5 to 7.0 m (18 to 23 ft) of well graded,
cross section is shown in Fig. 2. The design well rounded, fluvial gravel and cobble. Bedrock
incorporated a monitoring program composed was encountered between elevations 1039 and
of 12 inclinometers, 6 extensometers, and 1041 metres (3,409 to 3,415 ft), with localized
precision survey monitoring at neighbouring areas reaching 1035 to 1036 metres (3,396 to
buildings, the top of each pile, and along the 3,399 ft). Groundwater is located in the gravels
face of the shoring. The monitoring program and varies seasonally from 1 to 3.5 m (3.3 to
played an integral role in identifying the 11.5 ft) in depth above the rock surface.
magnitude and location of movement during The bedrock consists of horizontally bedded
construction. mudstone, siltstone and sandstone of the
To better understand the mechanics behind Porcupine Hills Formation of the Cretaceous or
observed movement, rock samples of the host Paleocene Age (Jackson and Wilson 1987 and
material were taken and subjected to extensive Osborn and Rejwijcz 1998). The siltstone and
testing by the Geotechnical Research Centre at sandstone form indurated, lenticular layers of
the University of Western Ontario and Golder varying thickness and extent. The mudstone
Associates. Results of the tests indicate the Formed as lacustrine deposits, the cementing
rock is weak with cohesion of 340 kPa (49 psi), agent of the mudstones and siltstones is clay,

[FIG. 2] Typical cross section.

[22] DFI JOURNAL Vol. 4 No. 1 August 2010

giving these formations a more soil-like quality weak rock zone. These types of movement
than the sandstone lenses. The mudstone are shown readily in Inclinometers 4, 5, and
varies from very weak to medium strength, 8. Inclinometers 4 and 8 show the behaviour
with point load axial tests recording strengths of the shoring wall and rock mass under the
typically ranging from 0.16 to 0.8 MPa (23 to influence of a shear band. All three show
116 psi). The siltstone and sandstone layers evidence of a weak rock layer in the upper
have medium to high strengths, with point load portion of the bedrock.
axial tests recording between 0.16 and 5.14 MPa Inclinometer 4 indicates the presence of a
(23 to 745 psi) in the siltstone, and 1.7 up to 8.1 shear band at the approximate elevation of
MPa (247 to 1,175 psi) in the sandstone. The 1021 m (3,350 ft). Fig. 4 shows the movement
mudstone is prone to rapid weathering with the observed in Inclinometer 4 over the duration of
addition of oxygen and water after excavation. the excavation. Original grade, final excavation
Based upon borehole logs, the rock mass quality elevation, and current excavation elevation at
is poor due to the Rock Quality Designation the time of reading are also shown. Movement
ranging from 0 to 95%, with an average value observed in Inclinometer 4 reached a maximum
of 45%. Within the mudstone, zones of rock of 57.5 mm (2.3 in) at the end of excavation.
showing slickensides were reported throughout Over half of this movement, 35 mm (1.4 in),
the geotechnical investigation. occurs over a 0.6 to 1.2 m (2.0 to 4.0 ft) thick
At the time of design, little was known in section of rock experiencing the shear band
regards to the bedrock properties. It was phenomena. Of particular interest is the
known that the bedrock was the cause of large elevation of the shear band. The movement
horizontal movements during excavation in occurs up to 5 metres (16 ft) below the final
Calgary, with the two leading theories being an excavation elevation.
elevated in-situ stress (ko = 2.0), or a shear band Inclinometer 5 is located on the east wall,
phenomenon. The shear band phenomenon is at the north end. Despite the proximity to
suspected of occurring when the stresses due Inclinometer 4, it does not show any sign
to excavation cause larger displacement of a
thin weak layer of rock (AMEC 2006). There
was minimal published data available on the
strength and deformation properties of the
various bedrock layers.

RESULTS OF THE MONITORING

PROGRAMME
The monitoring programme at The Bow
consisted of 12 inclinometers, 6 extensometers,
and survey targets at the top of every pile,
throughout the shotcrete wall, and on
neighbouring structures. The location of the
inclinometers and extensometers are shown in
Fig. 3. Inclinometers were drilled separately
from the piles and generally to a depth of
10 metres (33 ft) below the final excavation
elevation. This was to ensure observations
of any deep seated movement. The three
monitoring methods were used to ensure
accurate monitoring. Inclinometer readings
successfully recorded deformations of the rock
profile at the shoring face, and extensometers
confirmed that the deformation extended
beyond the soils and rock immediately adjacent
to the wall.
Results from the monitoring indicated that
horizontal movement is primarily caused by [FIG. 3] Location of inclinometers, extensometers, and rock
two methods: shear band movement and a samples
DFI JOURNAL Vol. 4 No. 1 August 2010 [23]
 [FIG. 4] Results from Inclinometer 4. [FIG. 5] Results from Inclinometer 5.

of shear band movement, and is instead shortly after excavation into the rock. The shear
representative of typical sites. Movement band had experienced up to 3 mm (0.12 in) of
during excavation is shown in Fig. 5. The movement when the excavation was still 12 m
maximum recorded movement at the end of (39 ft) above.
excavation was 46.5 mm (1.83 in). It should be The extent of deformation of the rock mass was
noted that the inclinometers show some signs measured with extensometers. Extensometer
of movement at their base, indicating that the 1 is located on the north wall to the west of
total movement could be greater than that Inclinometer 3. The extensometer consisted
recorded. of five sensors placed at 5 metre (16.4 ft)
Comparing the results of Inclinometer 5 intervals, and measured the amount of lateral
to Inclinometer 4 shows a similar curved deformation in the rock mass. Results from
protrusion occurring around elevation 1030 Extensometer 1 are shown in Fig. 7. The
metres (3,379 ft). This indicates that similar maximum horizontal extension measured was
mechanisms in rock behaviour are acting 19.6 mm (0.77 in) at the time of final excavation.
in the north-east corner, and that there is a The extensometer was installed at elevation
possible direction dependency of the shear 1035.0 m± (3,396 ft±), shortly after excavation
band phenomenon, as both inclinometers were into the rock began. By comparing the amount
affected by similar rates of excavation. of movement observed in Inclinometers 3
The movement during excavation, as observed and 4 at the time of installation, it can be seen
in Inclinometer 4, was compared directly to that the extensometer did not capture the
the rate of excavation, as shown in Fig. 6. total movement. Inclinometer records at that
This figure shows the movement along the elevation indicate that up to 20 mm (0.79 in)
shear band increasing, as the excavation level of movement at the face of shoring was not
decreases. From observation, there appears captured by the extensometers.
to be a direct correspondence between the The extent of movement beyond the shoring
rate of excavation and the rate of movement face can also be inferred by the extensometer
experienced at the shear band. Of particular records. The non-zero slope between the last
note, movement of the shear band begins two monitoring points indicates that movement

[24] DFI JOURNAL Vol. 4 No. 1 August 2010

 However, similar layers had been observed
throughout the site during excavation, and were
only occasionally correlated with shear band
movement in the inclinometer readings.
Such descriptions bear some similarity to
the bentonite layers encountered during the
excavation for the Edmonton Convention
Centre, as described in Chan and Morgenstern’s
1987 paper. The paper states that such layers,
with enough continuity, would control the
excavation stability. However, the detailed
mechanics of the movement may be different.
[FIG. 6] Movement at the shear band at Inclinometer 4 over The Geotechnical Research Centre investigated
the course of the excavation.
the occurrence of the weak rock layer by
compiling the axial and diametric point load
is occurring past the 25 metre (82 ft) reach test results and plotted them by elevation. The
of the extensometer. The results of the results are shown in Fig. 9. For most of the
extensometers are currently being studied. depth of the excavation, the axial load tests
Inclinometer 8 results are shown in Fig. 8, and indicate good scatter. However, both tests
show shear band movement around elevation show a zone that is predominantly the weak
1028 m (3,373 ft), and a weak rock zone mudstone. This zone, between elevations 1028
between 1031 and 1036 m (3,383 and 3,399 and 1033 m (3,373 to 3,389 ft) correlates to the
ft). Inclinometer 8 is useful in that it clearly larger intrusions shown in the inclinometers,
shows a weak rock zone and a shear band, including the results from 4, and 5, above.
the two modes of large horizontal movement.
Of the maximum 37.5 mm (1.48 in) into site ROCK TESTING AND RESULTS
movement, the shear band was responsible for Results from the monitoring program prompted
up to 20 mm (0.8 in) , and the weak rock for the collection of additional bedrock samples.
approximately 15 mm (0.6 in). Block samples were collected at between
elevations 1027 and 1028 m (3,370 to 3,373 ft)
Readings indicate that the shear band shown
near the east wall, just south of 6th Avenue SE,
in Inclinometer 8 was located at approximate
and north of Inclinometer 7. The rock samples
elevation 1028 m (3,373 ft), which daylighted
were taken from a thin, black, slickensided
into the excavation face. Unfortunately, the
mudstone layer, similar to that observed at the
inclinometer was damaged during construction
location of the shear band in Inclinometer 8.
of the wall, so thorough readings could not be
These samples were sent to Golder Associates for
taken over the full course of the excavation.
shear box testing, and the Geotechnical Research
Inspection of the rock face at this elevation
Centre for extensive testing of rock properties.
showed a 150 to 250 mm thick band (6 in to 10
Tests conducted by the GRC include uniaxial
in) of black mudstone, with some slickensides.
compressive tests, multi-stage direct shear tests,
hydrostatic compression tests, semi-confined
swell tests, null-swell tests, water content tests,
Atterberg limits tests, unit weight tests, calcite
content tests, and rock salinity tests. This paper
will concentrate on the results of the first three
test types. For full coverage of the tests, please
see the 2009 paper by Lo and Micic.
Extraction of the block samples proved to be
difficult. The weak, fissile nature of the rock
required extreme caution when excavating.
Attempts at producing large block samples
proved ineffective. The use of machines was
prohibitive, as the vibrations caused the rock
[FIG. 7] Readings from Extensometer 1. to disintegrate. Recovered hand samples were

DFI JOURNAL Vol. 4 No. 1 August 2010 [25]

 shaped test samples to be extracted from the
larger block samples.
The Geotechnical Research Centre conducted 5
uniaxial compression tests, and five multi-stage
direct shear tests. Results from the uniaxial
compression tests are shown in Table 1. The
combined results are plotted in Fig. 10.
Table 1 indicates a UCS ranging between 560
and 1230 kPa (81 to 178 psi), with a modulus of
elasticity at 50% of ultimate strain (E50) between
80 and 170 MPa (11.6 to 24.7 ksi). Chan and
Morgenstern (1987) estimated the clay shales to
have a modulus of 137.9 MPa (20.0 ksi), and the
bentonite layers 49 MPa (7.1 ksi). Results from
the GRC uniaxial compression tests indicate
that the weak mudstone is closer to the clay
shale observed in Edmonton.
The results shown in Fig. 10 indicate that
the intact rock has a cohesion value of
approximately 340 kPa (49 psi), with an angle of
internal friction of 24o. The residual strength
parameters drop to zero cohesion and an angle
of 15o. The scattering of results could be a
result of the inevitable disturbance of the weak
mudstone prior to testing. Triaxial compression
test results and shear test results by Golder
[FIG. 8] Results from Inclinometer 8. provide results consistent with the testing.
A hydrostatic test was
completed to determine
the bulk modulus of the
rock. The sample was
performed in a triaxial
compression cell on a
square sample of rock.
Volume changes in the
sample were recorded
during the increase in
pressure. The results
of the test are shown in
Fig. 11 and indicate two
distinct regions. The
first region shows a large
change in volume over
a very small increase in
pressure is attributed to
the closing of fissures in
the sample. The second
region is attributed to the
[FIG. 9] Point and Diametric Load test data indicating weak zones. deformation of the rock
and is used to derive the
irregularly shaped, and immediately wrapped in bulk modulus (Goodman 1980). Results show a
plastic cling-wrap and coated in wax, to ensure bulk modulus, K, of approximately 80 MPa (11.6
no loss of moisture. However, the weak nature ksi), which is consistent with the results from the
of the rock allowed for only square-prism uniaxial compression tests, assuming

[26] DFI JOURNAL Vol. 4 No. 1 August 2010

[TABLE 1] Results of uniaxial compressive tests.

Test Uniaxial Modulus (MPa)

Number Compressive
Strength (kPa) Et E50

UCT 1 720 138 122

UCT 2 1070 180 170

UCT 3 570 85 80

UCT 4 560 135 132

UCT 5 1230 168 160

[FIG. 11] Results of the hydrostatic compression test.

represents the as-built shoring wall at

Inclinometers 4 and 5. The model considered a
mesh with a width of 190 m (623 ft) and a depth
of 80 m (262 ft). The boundary conditions
consisted of a rough and rigid elements along
the bottom, and rigid, smooth elements on the
sides. The model used beam elements for the
caisson wall and anchor elements for the rock
anchors. Material properties were matched
with those shown on the shoring construction
drawings. The beams were modeled as
W460X68 (W18X45) piles. Loads were applied
to the anchors as shown on Fig. 2, and in
accordance with stressing records from the site.
Soil and rock parameters were derived from the
test results, as well as from past experience in
similar geological conditions. These parameters
are outlined in Table 2, and have been separated
into five layers: soil, upper rock, weak rock,
lower rock and the shear band.
Soil parameters were chosen based upon
experience in the area. The weak rock
[FIG. 10] Results of multi-stage direct shear tests. properties were based upon the results of the
multi-stage direct shear tests, as shown in Fig.
that the Poisson’s ratio varies from 0.15 to 0.3 9. The elastic modulus was chosen as 165 MPa
by the equations of elasticity. (24 ksi) based upon median values from the
uniaxial compression tests. The shear band was
Additional tests indicated that the average unit
chosen to have the residual values derived from
weight of the rock was 22.24 kN/m3 (3,823 lb/
the testing. The upper and lower rock layers
cu yd).
were assigned a higher friction angle value
and elastic modulus. This was based upon
POST CONSTRUCTION MODELING
the higher axial and diametric point load test
Using the parameters gained from the values, as shown in Fig. 8. The lower rock was
testing, numerical analyses were performed modeled with an elastic modulus that increased
and the results compared to the deflection by 5 MPa/m (226 psi/ft) with depth. Due to the
measurements observed over the course of the poor quality of the rock samples, the Poisson’s
excavation. ratio was not able to be measured. As such, it
The 2D finite element analysis was completed was assumed to be 0.3 for all rock layers. Unit
with PHASE2, using a cross section similar to weight was also based upon results from the
that shown in Fig. 2. This section appropriately laboratory tests.

DFI JOURNAL Vol. 4 No. 1 August 2010 [27]

 Vertical in-situ stresses were determined through
the unit weights described above. Horizontal
stresses in the soil were determined assuming a
ko = 0.8. In the rock, it was assumed that there is
a transition zone that increased from 0.145 MPa
(21 psi) at the bedrock surface to 1.5 MPa (218
psi) at the top of the weak rock zone. From that
elevation down, the stress was held constant.
For this analysis, it was assumed that the
horizontal stresses in the rock were isotropic.
Using the above parameters, the finite element
analysis was used to model the stress and
strain of the soil, rock, and shoring wall during
excavation. Shoring installation, excavation
rates, and anchor loading were modeled using
field inspection records from the north wall.
The results of the analysis, as well as the soil
and rock layering and horizontal stresses, are
shown in Fig. 12.
The results of the analysis are shown with
[FIG. 12] Results of preliminary finite element modeling.
readings from Inclinometers 3 and 4, located
along the north wall. The analysis accurately
the rockmass at the time of final excavation.
reflects the magnitudes and behaviour of
The figure, using results from the previously
the sudden shear band movement, and the
mentioned analysis, outlines the extents
bulging of the weak rock. It is noted that the
of the model, the rockmass layers, and the
inclinometer readings assumed zero movement
horizontal movement, shown in 10 mm (0.4
at the inclinometer toes, whereas the modeling
in) intervals. From the figure, the FEA results
indicates up to 6 mm (1/4 in) of movement
indicate that the effect of the shear band is to
at this elevation. The model results were
propagate movement much farther away from
then compared with the remainder of the
the excavation than is normally observed. The
inclinometer readings, as shown in Fig. 13.
results indicate that horizontal movement is
Several observations are made based upon these occurring 60 metres (197 ft) or more from the
results. Inclinometer readings show two general excavation, a distance of 3 times the depth
types of movement in the soil, either cantilever of the excavation. The results can also be
style, or as a significant top retaining effect, such compared with the readings from Extensometer
as in Inclinometers 3 and 4. The model indicated 1. Horizontal movement at 25 metres (82
movement similar to the former style. The cause ft) from the excavation face are around 10
of the discrepancy is still being investigated. mm (0.4 in) , which is similar to the amount
The extent of the movement beyond the of movement expected at the end of the
excavation face was also investigated. Fig. 14 extensometer, as previously discussed. The
shows the horizontal movements calculated in results of the extensometer readings and the

[TABLE 2] Parameters used in initial finite element modeling.

Unit Weight E c' c'r

Layer φ' φ' v
(kN/m )3
(MPa) (kPa) (kPa)

Soil 21 150 10 40 0 15 0.3

Weak Rock 22 165 340 24 0 15 0.3

Shear Band 22 165 - - 0 15 0.3

Upper Rock 22 825 340 40 0 15 0.3

Lower Rock 22 825 340 40 0 15 0.3

[28] DFI JOURNAL Vol. 4 No. 1 August 2010

[FIG 13] Cumulated inclinometer results.

Distance (m)
[FIG. 14] FEA results showing the horizontal movement in 10 mm contours.

corresponding FEA are currently undergoing (1.8 in). The weak rock was observed to cause
further study. as much as 100 mm (4 in) of deflection, as
shown in Inclinometer 12 (Fig. 13). Due to the
CONCLUSIONS monitoring results, mudstone samples were
Inclinometer monitoring during the 21 m (69 collected for laboratory testing. Despite the
ft) deep excavation for The Bow in downtown difficulty in obtaining samples from the field
Calgary, Alberta, indicated that large lateral and preparing them for the testing, several rock
deflections in the bedrock are due to a weak parameters were measured.
rock layer, and the shear band effect in a thin These include:
rock layer. Total lateral movement measured a. The uniaxial compression tests indicate
ranged between 45 mm and 150 mm (1.8 in that the elastic modulus ranges from
and 6.0 in). Movement due to the shear band 80 to 180 MPa (11.6 to 26.1 ksi) and the
along the north wall was a maximum of 45 mm uniaxial compressive strength ranges
DFI JOURNAL Vol. 4 No. 1 August 2010 [29]
 between 560 MPa and 1230 MPa (81 and 178 manifested shear band behaviour. This issue
ksi). Representative values for the elastic underlines the importance of field monitoring in
modulus and the UCS could be 160 MPa and deep excavations in soft rock.
830 MPa (23 ksi and 120 ksi), respectively.
b. Results from multi-stage direct shear tests REFERENCES
indicated that c' = 340 kPa (49 psi) and 1. AMEC 2006. Supplementary geotechnical
φ' = 24°. Residual strengths were found to investigation proposed EnCana Tower 5
be c'r = 0, and φ'r = 15°. Avenue SE and Centre Street S Calgary,
c. Results from the hydrostatic compression Alberta, AMEC Earth & Environmental,
tests showed that the bulk modulus was Calgary, Alberta.
approximately 80 MPa (11.6 ksi). This value 2. ASTM D 4318. 2005. Standard test methods
was in the range of moduli derived from the for liquid limit, plastic limit, and plasticity
uniaxial compression tests. index of soils. ASTM.
Finite element modeling was conducted based 3. Brachman, R.W.I., Moore, I.D, and L0,
upon results of the laboratory tests, field records K.Y., 1998. Analysis and performance of
of construction, past experience in similar geology, a shoring system with tie-back anchors.
and compared to the movements observed along Proceedings of the 51st Canadian
the north wall. This initial analysis indicated that: Geotechnical Conference, Edmonton, Alberta,
a. Using the above mentioned parameters and vol. 1, pp 439-446.
information, magnitude and location of 4. Brendley, G.W., 1951. X-ray identification
lateral movement along the north wall was and crystal structure of clay minerals.
modeled with sufficient accuracy. Taylor and Francis Ltd., London, UK.
b. The presence of the weak rock layer resulted 5. Chan, D. H., and Morgenstern., N.R., 1987.
in large lateral movements, and should be Analysis of progressive deformation of the
considered in future excavations. Edmonton Convention Centre excavation.
It should be noted that a limited number Canadian Geotechnical Journal, volume 24,
of laboratory tests of the mudstone were pp. 430-440.
completed, and that more are required to 6. Dreimans, A., 1962. Quantitative gasometric
accurately determine the rock parameters. determination of calcite and dolomite using
No samples were taken of the siltstone and Chittick Apparatus. Journal of Sedimentary
sandstone. Due to the weak and fragile nature Petrology, vol. 32, pp. 520-529.
of the samples, the Poisson’s ratio was not 7. Goodman, R.E., 1980. Introduction to rock
measured. This important parameter should mechanics. John Wiley & Sons, New York,
be determined, either through additional NY, USA, pp. 64 and 68.
laboratory tests, or through in-situ testing.
8. Hoek, E., Kaiser, P.K., and Bawden, W.F.,
A significant feature of the mudstone to be 2005. Support of underground excavations
recognized is its swelling characteristics (Lo et in hard rock. Taylor and Francis, New York,
al. 2009). Both laboratory tests and monitoring New York, USA, pp. 103.
results indicated that the rock expanded with
time similar to shale formations in Southwestern 9. Hudson, J.A., and Harrison, J.P., 1997.
Ontario and adjoining United States. Swelling Engineering rock mechanics – An
of the rock could result in higher than expected introduction to the principles. Elsevier
stresses on final structures, leading to undesired Science Ltd., Oxford, UK, pp. 96.
deformations and damage. 10. Jackson, E.J., and Wilson, C.W., 1987.
The mechanics behind the shear band effect Geology of the Calgary area. Canadian
are also currently not well defined. The current Society of Petroleum Geologists, Calgary, AB,
theory is that the shear band is a thin layer of Canada.
rock that undergoes a large amount of plastic 11. Lardner, T., Janes, M. and Halliwell, M. 2008.
strain due to the effects of excavation. The The Bow, EnCana excavation support system
undetermined mechanics of the shear band design and performance. Proceedings from
prevents the prediction of its manifestation. the 61st Canadian Geotechnical Conference
Layers of rock identified as hosting shear bands & 9th Joint CGS/IAH-CNC Groundwater
were found throughout the site, yet only a few Conference, Edmonton, AB, Canada, pp. 63-70.

[30] DFI JOURNAL Vol. 4 No. 1 August 2010

12. Lee Y.N. and Lo, K.Y., 1993. The swelling
mechanism of Queenston shale. Canadian
Tunnelling Journal, The Tunnelling
Association of Canada, pp. 75-97.
13. Lo, K.Y., 1989. Recent advances in
design and evaluation of performance of
underground structures in rock. Tunneling
and Underground Space Technology, vol.
4(2), pp. 171-183.
14. Lo, K.Y. and Lee, Y.N., 1990. Time-dependant
deformation behaviour of Queenston shale.
Canadian Geotechnical Journal, vol. 27, pp.
461-471.
15. Lo, K.Y., Cooke, B.H. and Dunbar, D.D., 1987.
Design of buried structures in squeezing
rock in Toronto, Canada. Canadian
Geotechnical Journal, vol. 24(2), pp. 232-241.
16. Lo, K.Y., Wai, R.S.C., Palmer, J.H.L. and
Quigley, R.M., 1978. Time-dependent
deformation of shaly rocks in Southern
Ontario. Canadian Geotechnical Journal, vol.
15, pp. 537-547.
17. Lo, K.Y., Micic, S., Lardner, T., and Janes,
M., 2009. Geotechnical properties of a
weak mudstone in downtown Calgary.
Proceedings of the 62nd Canadian
Geotechnical Conference, Halifax, Nova
Scotia, pp. 454-462.
18. Martin, R.T., 1955. Glycol retention analyses.
Soil Science Society America Proceedings, vol.
19(2), pp. 160-164.
19. Moore, D.M. and Reynolds, R.C., 1997.
X-ray diffraction and the identification and
analysis of clay minerals. Oxford University
Press, New York, USA, pp. 227-297.
20. Mohamed, A.M.O., and Antia, H.E., 1998.
Geotechnical engineering. Elsvere Science
Ltd., Amsterdam, The Netherlands, pp. 75.
21. Osborn, G. and Rajewicz, R., 1998. Urban
geology of Calgary. Geological Association of
Canada, Special Paper 42, Urban Geology of
Canadian Cities, pp. 93-115.
22. Rocscience, 2008 PHASE2 v.6.027 user’s
guide. Rocscience Inc. Toronto, Ont.
23. Wong, R.C.K., 1998. Swelling and soft
behaviour of La Biche shale. Canadian
Geotechnical Journal, vol. 35, pp. 206-221.

DFI JOURNAL Vol. 4 No. 1 August 2010 [31]

 Analyzing Drivability of Open Ended Piles in Very Dense
Sands
James A. Schneider, University of Wisconsin - Madison, Madison, WI, USA; [email protected]
Ivy A. Harmon, University of Wisconsin - Madison, Madison, WI, USA

ABSTRACT
The successful installation of long piles driven into very dense sands relies on the occurrence of the
reduction in local friction with increased pile embedment, a phenomenon known as ‘friction fatigue’.
The underlying mechanisms controlling friction fatigue are poorly understood, with some design
methods including an adjustment for the influence of pile diameter while others do not. This paper
back calculates the installation resistance of 0.356m to 2m (14in to 78in) diameter open ended piles
driven into very dense sands using wave equation analyses. Cone penetration test data are used
to link soil properties to installation resistance. The study illustrates consistent interpretation of
a variety of case histories of open ended piles driven in very dense sands using newly developed
analysis techniques and normalized parameters. Results provide information on methods for
incorporating friction fatigue into drivability studies as well as a discussion of mechanisms related to
pipe pile installation resistance in sandy soils.

INTRODUCTION height normalized by the pile diameter (e.g.,

Observations of the resistance of a pile during Lehane et al. 1993; Randolph et al. 1994; Jardine
installation provides near immediate feedback et al. 2005, Kolk et al. 2005a), and others
on assumptions related to design at a given consider it to be controlled by the number of
site. Additionally, these measurements can installation cycles (or hammer blows) (e.g.,
be used to evaluate assumptions inherent in White & Lehane 2004; White 2005).
design method formulations. Despite high Three case histories with pile diameters varying
levels of uncertainty in procedures for back from 0.356m to 2m (14in to 78in) are back
analysis of pile driving records, these data have analyzed in this paper. Analyses are focused
a significant advantage over static load tests on the performance of open ended piles in very
for calibrating design methods. Specifically, the dense sands, with soil strength and high stress
effects of pile length (L), or slenderness ratio compressibility characterized using results of
(L/D), on soil resistance can be assessed for a cone penetration tests (CPTs). A CPT measures
large range of values at a single site. the two parameters of cone tip resistance
Length or L/D effects are commonly discussed (qt) and sleeve friction (fs). Fig. 1 illustrates
as the phenomenon known as ‘friction fatigue’. terminology used in this paper when comparing
Friction fatigue is the reduction in pile shaft CPT and pile resistance.
friction in a given soil horizon that occurs as
the pile tip is driven deeper (e.g., Heerema SITES
1980). Friction fatigue has been hypothesized Three case histories of driving open ended piles
to be most severe for the case of sandy soils in very dense uncemented siliceous sands are
(as compared to clayey soils) and some studies back analyzed in this paper; (i) Pigeon Creek
imply a greater effect for piles driven into (PC), Indiana, USA; (ii) Euripides (Eur-I & Eur-
dense sands as compared to loose sands (Alm & II), The Netherlands; and (iii) Trans Tokyo Bay
Hamre 2001). Bridge (TTB), Japan. Table 1 presents pile,
There is debate on the appropriate formulation hammer, and installation information for each
for assessment of friction fatigue; some design of the case histories analyzed, with additional
methods relate the concept to height of a soil information presented by Schneider et al. (2008).
layer above the pile tip (e.g., Heerema 1980; Profiles of CPT tip resistance for each site are
Toolan et al. 1990; Alm & Hamre 2001; Claussen shown in Fig. 2. All three sites have relatively
et al. 2005), others relate the behavior to the soft/loose near surface materials that are

[32] DFI JOURNAL Vol. 4 No. 1 August 2010

underlain by very dense sands. The Tokyo Bay & Denver 1988, Svinken et al. 1994, Randolph
site includes many more occurrences of low tip 2003), and definition of ‘failure’ in static load
resistance, indicating interlayers of soft/loose tests. Time effects result from changes in total
material. Along the pile shaft, the weighted stress and pore pressures that occur due to
average apparent relative densities based on soil displacement during the pile installation
normalized cone tip resistance [qc1N=(qt/pref)/ process, among other factors.
(σ'v0/pref)0.5] range from 85% for Pigeon Creek to
100% for Euripides and Trans Tokyo Bay. The
Eur-I and Eur-II sites may be overconsolidated
and have a slightly lower actual relative
density, although the overconsolidation ratio
is suspected to have a small effect on the
correlation between τf and qt (e.g., Foray et al.
1998). Pile driving operations at the Euripides
site are shown in Fig. 3.

[FIG. 1] Illustrations of cone penetration test (CPT) and pile

nomenclature

STATIC RESISTANCE DURING

DRIVING
During pile driving the soil resistance will result
from static and dynamic components. The
dynamic components are related to increases in
resistance due to inertial and viscous rate effects
(e.g., Randolph 2000), although commonly
quantified using damping factors. The static
component of the soil resistance during driving
(SRD) is similar to the pile static axial capacity,
except the resistance often differs due to
consolidation, equalization, and ageing, or
[FIG. 2] CPT qt profiles for sites analyzed in this paper
‘time-effects’ (e.g., Tavenas & Audy 1972, Skov
DFI JOURNAL Vol. 4 No. 1 August 2010 [33]
 [TABLE 1] Characteristics of open ended piles for sites in this studya (see also Schneider et al. 2008)

ztip, Water Avg Enthrub Approx.

D, t, Avg
Site ID [Lpile] Table, Ar,eff Hammer [Rated] Total Reference
m mm IFR
m m [Ar] (kN·m) Blows
Pigeon 7 0.44 19-28.5 Paik et al.
2 0.356 32 3 0.83 ICE 42-S 225
Creek [8.24] [0.33] [57] 2003
47
[27, 0.21 68
Euripides I 0.763 36 1 0.96 IHC S-90 8400 Zuidberg &
49.4, [0.18] [90] Vergobbi
57c] 1996; Kolk
47 0.22 71 et al. 2005b
Euripides II 0.763 36 1 0.95 IHC S-90 8550
[57] [0.18] [90]
Trans 31 0.07 Menck 300 Shioi et al.
Tokyo TP 2.0 34 0 1.0 MRBS 1275
[62] [0.07] [736] 1992
Bay 5000
a
Nomenclature illustrated in Fig. 1
b
Transferred energy at end of driving [hammer rated energy in brackets].
c
Field welds were made at 27m and 49.4m (89ft and 162ft) for the first installation of the Euripides pile.
Initial pile driving (and static load testing at intermediate depths) from 22 March 1995 to 10 May 1995.

increment of driving (∆hplug) as compared to the

length of pile installed (∆hpile). For large diameter
open ended piles in relatively uniform soil
deposits the soil within the pile tends to remain
near the original ground surface during driving,
∆hplug = ∆hpile, IFR=1 (e.g., Stevens 1988). Under
slow loading conditions, such as a static load
test, the soil plug will tend to move with the pile,
∆hplug = 0, IFR=0 (e.g., Randolph et al. 1991).
Considering that the inertia of the soil plug
tends to lead to large diameter pipe piles
installing in a predominantly coring manner
(IFR=1), the unit ‘end bearing’ (qb) will occur
solely on the pile annulus (qann) and the pile
shaft friction (τf) at the time of pile driving will
occur both internally (τf,in) and externally (τf,out)
along the surface area of the pile. The total SRD
is the sum of the pile shaft resistance (Qs) and
the pile base resistance (Qb):

ܴܵ‫ ܦ‬ൌ ܳ௦ ൅ ܳ௕ (1a)

௭೟೔೛
ܴܵ‫ ܦ‬ൌ ߨ‫ ܦ‬σ ߬ ȉ ȟ‫ ݖ‬൅ ‫ݍ‬௔௡௡ ‫ܣ‬௔௡௡ (1b)
[FIG. 3] Pile driving at the Euripides site (photo courtesy ௭೟೔೛ Ǧ௅೐೘್ ௙
Fugro Engineers B.V.)
where Aann is the pile annular area [=π/4(D2-
The ability to predict SRD, as well as static Di2)], D is the pile outer diameter, Di is the pile
capacity, of an open ended pile is complicated inner diameter, z is depth, ztip is the pile tip
by mechanisms related to soil entering or being depth, ∆z is the change in depth, and Lemb is
displaced by the advancing pile. The relative the pile embedded length. It is common to take
amount of soil entering an open ended pile can the internal shaft friction simply as a fraction
be quantified using the incremental filling ratio of the external shaft friction (τf,in/τf,out). Initial
(IFR=∆hplug/∆hpile) (e.g., Paikowski et al. 1989). analyses in this paper use the assumption that
IFR is equal to the change in plug height for an the ratio of τf,in/τf,out is equal to 0.5 (e.g., lower

[34] DFI JOURNAL Vol. 4 No. 1 August 2010

bound from Stevens et al. 1982). The effect of exterior of the pile, i.e., τf,in/τf,out = 1. Since the
τf,in/τf,out on driving resistance is addressed later analyses in this paper considers τf,in/τf,out = 0.5,
in this paper. The differences between internal the external shaft friction is considered to be
and external diameter are small compared to 2/3 of the value from Equation 2, such that the
the uncertainty in estimating τf,in/τf,out, and are total pile friction during installation agrees with
not explicitly considered in this study. The the recommendations of Alm & Hamre (2001).
following modified format for SRD is used for is ଴Ǥଵଷ
௤೟ ఙᇱೡబ
paper: ߬௙ǡ௠௔௫ǡ௢௨௧ ൌ ൬ ൰ ‫ߜ݊ܽݐ‬௙ (3b)
ଵଵସ ௣ೝ೐೑
œ
ൌɎ σœ –‹’Ǧ ͳǤͷɒˆǡ‘—– ȉȟœ ൅“ƒ ƒ (1c) While the ratio of the end bearing of statically
–‹’ ‡„
loaded closed ended displacement piles to CPT
To quantify the effects of friction fatigue on tip resistance (qb/qt) is often considered to vary
shaft friction during pile installation, two from 0.6 to 1.0 (e.g., White & Bolton 2005, Xu
detailed models for soil resistance are used: et al. 2008), normalized annular end bearing
• AH-01 (Alm & Hamre 2001) during installation is estimated to range from
0.35 to 0.55, as:
• UWA-05 (Lehane et al. 2005, 2007)
௤ೌ೙೙ ௤೟ ଴Ǥଶ
The formulations for each model are ൌ ͲǤͳͷ ȉ ቀ ቁ (6)
summarized as follows, and the interested ୯౪ ఙᇱೡబ

reader is referred to the original references for Equations 2, 3b, and 4 through 6 are combined
more detail. within Equation 1c and referred to as the AH-
01 method for the remainder of this paper.
Alm & Hamre (2001) method for sands This allows for ease of comparison of results
The Alm & Hamre (2001) method was calibrated produced by applying AH-01 to those of other
primarily based on drivability studies of large design methods which explicitly separate
diameter piles in the North Sea. For sands, internal and external shaft friction.
pile diameters ranged from 2.4m to 2.7m
(8ft to 9ft), and pile penetration ranged from Modified UWA-05 method for uncemented
55m to 70m (180ft to 230ft). To minimize siliceous sands
additional uncertainty related to selection of Unlike the AH-01 method, UWA-05 was
soil parameters and modeling changes in radial developed for estimation of static axial
stress due to pile installation, both qb and τf are capacity rather than SRD. The database of
correlated to CPT qt. Shaft friction is estimated piles used in calibration of UWA-05 was more
to exponentially decay from a maximum (τf,max) representative of onshore conditions than
to a residual (τf,res) value: offshore, with a mean diameter of 0.5m (20in)
and a mean embedded length of 17m (55ft).
߬௙ ൌ ߬௙ǡ௥௘௦ ൅ ൫߬௙ǡ௠௔௫ െ ߬௙ǡ௥௘௦ ൯݁ ି௞௛ (2) The mean time between installation and load
testing of the database piles was 9 days. Some
଴Ǥଵଷ
௤೟ ఙᇱೡబ differences between shaft resistance at 9 days
߬௙ǡ௠௔௫ ൌ ൬ ൰ ‫ߜ݊ܽݐ‬௙ (3a)
଻଺ ௣ೝ೐೑ after installation and that during installation
should be expected, but these variations will be
߬௙ǡ௥௘௦ ൌ ͲǤʹ߬௙ǡ௠௔௫ (4) addressed later in this paper.

Where h is the height above the pile tip, σ'v0 Shaft friction is evaluated using the Coulomb
is the initial vertical effective stress, pref is a failure criterion:
reference stress equal to 100 kPa, and δf is the
soil-pile interface friction angle at failure. The ߬௙ǡ௢௨௧ ൌ ߪԢ௥௙ ‫ߜ݊ܽݐ‬௙ ൌ ሺߪԢ௥௖ ൅ ȟߪԢ௥ௗ ሻ‫ߜ݊ܽݐ‬௙ (7)
shape factor for the ‘rate’ of degradation (k) is
related to the normalized cone tip resistance: where σ'rf is the radial effective stress on the
external wall of the pile at failure, σ'rc is the
ଵ ௤೟ ଴Ǥହ
݇ൌ ቀ ቁ (5) radial effective stress on the external wall of
଼଴ ఙᇱೡబ the pile after pile installation and equalization,
The result of Equation 2 (through 3a, 4, and 5) and ∆σ'rd is the change in radial effective stress
is discussed by Alm & Hamre (2001) to be the during pile loading. The radial stress after
total internal and external shaft friction. It was installation and the change in radial stress are
recommended by Alm & Hamre to reduce this estimated separately, and both typically use
value by 50% and apply to both the interior and correlations to cone tip resistance. For driven
DFI JOURNAL Vol. 4 No. 1 August 2010 [35]
 piles in siliceous sands, the radial stress after ௤ೌ೙೙
installation and equalization is estimated as: ൎ ͲǤ͵ͷ (9)
௤೟
௤೟ ȉ஺ೝǡ೐೑೑ ್ ௛ ି௖ Equations 7 through 9 are applied within
ߪԢ௥௖ ൌ ቂ݉ܽ‫ ݔ‬ቀ ǡ ߥቁቃ (8a)
௔ ஽ Equation 1b and referred to as the modified
௤೟ ȉ஺ೝǡ೐೑೑ బǤయ ௛ ି଴Ǥହ UWA-05 method for the remainder of this paper.
ߪԢ௥௖ ൌ ቂ݉ܽ‫ ݔ‬ቀ ǡ ʹቁቃ (8b)
ଷଷ ஽
WAVE EQUATION ANALYSES
Where (e.g., White 2005, Lehane et al. 2005):
The program GRLWEAP (Pile Dynamics 2005)
a = parameter to account for the reduction in
was used for wave equation analyses in this
radial stress behind the pile tip = 33
paper. Standard quake values of 2.5mm (0.1in)
b = parameter to account for differences were used for both the shaft and base resistance
between open and closed ended piles = 0.3 (e.g., Roussel 1979; Stevens et al. 1982), with a
c = exponent which accounts for friction fatigue Smith shaft damping of 0.25s/m and Smith base
= 0.5 damping of 0.5s/m applied in the sandy soils.
When clay soils were encountered, the shaft
ν = parameter which provides an upper limit on
damping was increased to 0.65s/m.
h/D-c at the pile tip = 2
It is difficult to use the two previously
Ar,eff is the effective area ratio that explicitly
discussed methods for estimating shaft friction
accounts for the differences in radial stress
within the program GRLWEAP, in that the shaft
induced by open, closed, and partially plugged
friction distribution changes each time the
piles.
pile tip is advanced. Fig. 4 presents calculated
஽೔ మ shaft friction distributions for installation of
‫ܣ‬௥ǡ௘௙௙ ൌ ͳ െ ‫ ܴܨܫ‬ቀ ቁ (8c)
஽మ piles at the Pigeon Creek site (e.g., Paik et al.
The incremental filling ratio (IFR=∆hplug/∆hpile) is 2003, Table 1). It is observed that the shaft
defined previously. The use of Ar,eff in Equations friction distribution based on a pile tip depth
8a and 8b results in higher shaft friction on of 8m (26ft) will underpredict pile shaft friction
closed ended piles as compared to open ended when the pile tip is, for example, at 5m (16ft).
piles, and partially plugged piles have an This is the phenomenon of friction fatigue. To
intermediate value (e.g., White et al. 2005). The accurately assess the effects of friction fatigue
pile area ratio, Ar=1-Di2/D2, is the same as Ar,eff during drivability studies, a separate wave
when IFR = 1 (fully coring). equation analysis would need to be performed
The change in radial stress during loading is for each tip depth.
based on elastic cylindrical cavity expansion Alm & Hamre (2001) appear to get around this
theory (e.g., Lehane et al., 1993): difficulty by developing a bearing graph using
ସீ୼௬
wave equation analyses and then estimating pile
ȟߪԢ௥ௗ ൌ (8d) blowcount profile from SRD (or SRD from pile
஽
blowcount). Three parametric studies of factors
In the absence of sufficiently varied case
influencing site specific bearing graphs are
histories presenting reliable data on
summarized in Fig. 5 for an IHC S-90 hydraulic
changes in radial stress during loading, the
hammer and pile conditions for the Euripides
recommendations of Jardine et al. (2005) were
test (Zuidberg & Vergobbi 1996, Kolk et al. 2005b,
used by Lehane et al. (2005, 2007) to define
Table 1). It is assumed that the pile diameter,
G=G0 and ∆y = 2Ra = 20μm for Equation 8d. G0
wall thickness, and damping parameters (in
is usually estimated from CPT tip resistance:
these uniform soil deposits) will not change, so
ି௡
ீబ ௤೟ Τ௣ೝ೐೑ the major factors influencing the bearing graph
ൌ ‫ ீܭ‬ቈ బǤఱ ቉ ൎ ͳͺͷ ȉ ‫ݍ‬௖ଵே ି଴Ǥ଻ (8e)
௤೟ ൫ఙᇱೡబ Τ௣ೝ೐೑ ൯ are (i) embedded length of the pile; (ii) fraction
of resistance from Qb or Qs; and (iii) shape of
For assessment of annular resistance of open
the distribution of τf.
ended piles or end bearing stress of closed
ended piles, UWA-05 takes the ratio of pile base Relatively similar bearing graphs are calculated
stress to properly averaged cone tip resistance irrespective of shape of shaft friction
(qb/qt) as 0.6. For this application to annular distribution or pile length in these parametric
resistance of open ended piles during driving, studies. Conversely, the fraction of shaft
a lower value of qann/qt is used, based on friction does have a significant influence when
discussion in Alm & Hamre (2001): estimating blowcount from SRD. The use of
[36] DFI JOURNAL Vol. 4 No. 1 August 2010
bearing graphs is therefore limited since Qs/SRD
will typically increase with pile embedment in
uniform deposits, and large variation may also
be expected in layered soil deposits.

[FIG. 5] Parametric study of the shape of SRD-blowcount

curves for an IHC S-90 hammer

The shape of the shaft friction distribution has

been shown to have a minimal effect on the
[FIG. 4] Comparison of calculated friction fatigue and pseudo bearing graph, so the change in shaft capacity
average incremental shaft friction (Δτf,avg) at the Pigeon Creek site between two increments of driving can be used
DFI JOURNAL Vol. 4 No. 1 August 2010 [37]
 to estimate a pseudo average incremental shaft pile) with area ratio is illustrated in in Fig. 6,
friction (∆τf,avg). The pseudo average incremental with calculations based on White et al. (2005)
shaft friction is calculated as: and Lehane et al. (2005). Since the AH-01 model
does not explicitly account for Ar,eff, the graph
ஊொೞǡಽ ିஊொೞǡಽషభ
ȟ߬௙ǡ௔௩௚ ൌ (10) is normalized for a typical open ended pile
గ஽ȉ୼௅
with a D/t ratio of 40, such as those used to
where ΣQs,L is the cumulative shaft resistance calibrate AH-01.
at the pile tip depth, ΣQs,L-1 is the cumulative
shaft resistance at the depth of the previous
SRD calculation, and ∆L is the length of pile
driven between the two sets of calculations. Fig.
4 compares pseudo average incremental shaft
friction for the two methods. Due to the effects
of friction fatigue, ∆τf,avg is significantly less
than τf near the pile tip after continued driving
in a given layer and ∆τf,avg may even be negative,
particularly for small values of ∆L.
It is interesting that the pseudo average
incremental shaft friction in Fig. 4 appears to
reach a constant value after a ‘critical depth’
of 3m (10ft) (or 8D). Observations such as
these may have led to development of the
concepts of ‘critical depth’ and limiting shaft
friction values. It should be stressed that the
average shaft friction values will be a function
of pile geometry and soil density (among other
[FIG. 6] Comparison of effective area ratio to inferred
factors), and a unique limiting shaft friction
influence on maximum shaft friction for test piles in this study
value for a given sand density is not valid in (trend from White et al. 2005)
most situations, particularly in layered deposits.
Analysis results are presented as calculated
ANALYSIS AND DISCUSSION and measured blowcount profiles with depth.
Digitized CPT profiles were used for assessment Measured resistance is presented as blows per
of τf and qb. Due to similar zones of influence meter of penetration (bpm). Fig. 7 compares
attributed to a CPT and pile annulus, no qt calculations based on the method of AH-01 for
averaging techniques were employed for these bearing graph and ∆τf,avg studies. A similar plot
studies. The soil profile was discretized into is presented for the modified UWA-05 method
approximately 100 elements for input into in Fig. 8.
GRLWEAP. The same discretization was used for In general it can be seen from Figs. 7 and 8 that
bearing graph (BG) and drivability analyses. the AH-01 method predicts drivability well for
While the piles for this study (as summarized the Euripides and Trans Tokyo Bay piles, while
in Table 1) have a large range of diameters, the modified UWA-05 method tends to better
the wall thickness is relatively consistent at predict the drivability response for the small
about 34mm (1.34in). This results in the pile diameter, larger area ratio, Pigeon Creek pile.
area ratio increasing from approximately Similar results are obtained with the bearing
0.07 for the Trans Tokyo Bay pile to 0.33 for graph and ∆τf,avg drivability analyses for both
the smaller diameter Pigeon Creek pile. Since resistance calculation methods (AH-01 &
smaller diameter piles tend to plug more modified UWA-05).
than larger diameter piles in uniform soil Due to the distinctly different method
conditions (e.g., Lehane et al. 2005), the Ar,eff formulations for AH-01 and modified UWA-05,
of the Pigeon Creek pile is actually 0.44 and there are a few reasons that could lead to these
is estimated to have an initial radial stress similarities and differences:
(and thus shaft friction) that is potentially 60%
1. Evaluation of annular end bearing;
greater than that for the Trans Tokyo Bay pile.
The increasing trend of calculated initial radial 2. Evaluation of initial shaft friction behind the
stress (on the exterior side of a displacement pile tip (aavg/Ar,eff0.3 term from Equation 8a);

[38] DFI JOURNAL Vol. 4 No. 1 August 2010

[FIG. 7] Calculated and measured blow counts using AH-01 [FIG. 8] Calculated and measured blow counts using modified
SRD method UWA-05 axial capacity method

3. ‘Rate’ of friction fatigue. of τf is more significant than qann, and qann/qt

Factors 1 and 2 are compared in Table 2 and appears to have a small effect on these pipe
Fig. 9, and Factor 3 is evaluated using Figs. 10 pile drivability analyses. The ’a’ parameter is
and 11. Despite having a larger end bearing stress dependant within the formulation of
component (qann/qt), the AH-01 method gives AH-01 [(σ'v0/pref)0.13; Equation 3a/3b], so ‘aavg’
lower blowcount during driving. Estimation parameters are presented in Table 2. Area ratio
DFI JOURNAL Vol. 4 No. 1 August 2010 [39]
 and plugging are not explicitly included in and this ratio will decrease with height (h) or h/D
AH-01, likely since these factors have a small behind the pile tip due to friction fatigue.
effect on drivability of large diameter piles with
a constant D/t. However, these factors are of
significant importance for modeling the Pigeon
Creek pile. Fig. 9 illustrates that if the shaft
friction in AH-01 method is increased using
the area ratio correction from Fig. 6, a better
estimation of pile driving resistance is achieved.

[FIG. 9] Performance of AH-01 method modified for wall

thickness and plugging effects for open ended Pigeon Creek pile

[TABLE 2] Summary of analysis results

Pile Tip Total
Site Depth Blows
(m) [Drive.]1
AH-01
Pigeon 155
7 126 0.45
Creek [185]
7870
Euripides 47 105 0.41
[9580]
Tokyo 1595
30.5 108 0.42
Bay [1715]
Modified UWA-05
Pigeon 200
7 42 0.35
Creek [245]
20000+
Euripides 47 55 0.35
[20000+]
Tokyo 2465
30.5 74 0.35
Bay [2445]
1
Total blow counts based on bearing graph
calculations, with number in brackets from
drivability studies
The ‘rate’ of friction fatigue is compared in Figs.
10 and 11 for the two methods. The ratio of τf/qt
at a given soil horizon is compared to the ratio
of τf/qt at the pile tip. This ratio is referred to as [FIG. 10] Comparison of friction fatigue using ratio of τf/qt at
the Shear Stress Tip Ratio, or TTR. Shaft friction a given height above the pile tip to τf/qt at the pile tip (or ‘TTR’)

[40] DFI JOURNAL Vol. 4 No. 1 August 2010

 Alternatively, overprediction may result from
overestimation of τf,in/τf,out. Fig. 12 shows the
sensitivity of pile drivability studies using
modified UWA-05 method to the ratio of
internal to external shaft friction.
Calculated and measured blowcounts of the
larger diameter piles in this study, Euripides
I / II and Trans Tokyo Bay, are better correlated
when applying modified UWA-05 with τf,in/τf,out
of zero. Blow counts for the partially plugged
piles at Pigeon Creek are better correlated when
using the modified UWA-05 method and τf,in/τf,out
equal to 0.75. These trends could be interpreted
to suggest that the ratio of internal to external
shaft friction is much higher for the case of
[FIG. 11] Comparison of friction fatigue using ‘TTR’ at each of partially plugged piles as compared to the case
the three sites from this paper
of coring piles. It is unlikely that the internal
shaft friction for the Tokyo Bay and Euripides
Similar values of TTR are observed for the
piles is equal to zero and that the influence
modified UWA-05 and AH-01 for the Pigeon Creek
of partial plugging at Pigeon Creek has such a
and Euripides piles, but significant differences
significant effect on τf,in/τf,out. It is clear, however,
occur for the Trans Tokyo Bay pile. The ‘rate’ of
that the assumptions related to internal shaft
friction fatigue predicted using the AH-01 method
friction are important to assessment of driving
is more severe than that predicted by modified
behavior, and experimental studies are required
UWA-05 at TTB. This appears to occur due to
to increase the understanding of underlying
the presumed diameter effect within the friction
mechanisms controlling this resistance. It is
fatigue formulation of modified UWA-05. A larger
likely that current soil models for pile drivability
database and more detailed analyses (e.g., a series
studies involve some level of compensating
of full wave equation analyses such as CAPWAP)
errors to produce accurate predictions.
should provide insight into this potential
diameter effect.
CONCLUSIONS
RATIO OF INTERNAL TO EXTERNAL While this study examined a relatively small
SHAFT FRICTION number of case histories, the large variation in
pile size combined with relatively consistent
The ratio of internal to external shaft friction is
soil conditions at each of the different sites has
uncertain, yet has significant implications on the
led to the following conclusions:
interpretation of pipe pile drivability. Stevens
et al. (1982) recommended performing analyses • To rationally assess drivability of open
with τf,in/τf,out ranging from 0.5 to 1.0. On the ended piles in very dense sands, analyses
contrary, Dutt et al. (1995) suggest that the must incorporate the effects of friction
internal shaft friction (for normally consolidated fatigue, pile area ratio, and potential for
clays) does not contribute to driving resistance. partial plugging during driving. Each of
Alm & Hamre (2001), suggest applying shaft these factors has been hypothesized to be
friction equally on the inside and outside of the influenced by pile diameter.
pile, but recommend using half the static shaft • When using GRLWEAP for drivability
friction for this value. This logic is similar to studies that include effects of friction
applying no shaft friction to the inside of the pile fatigue, analyses may be performed using
and full shaft friction on the outside of the pile. the bearing graph option or a drivability
Figs. 7 and 9 show that the AH-01 method analysis with a shaft friction distribution
can perform well for the cases in this paper, based on ∆τf,avg (Equation 10). While similar
while Fig. 8 indicates that the modified UWA- results were achieved in this study (for <
05 method tends to overpredict SRD. This 300 bpm or 91 bpf), the drivability (∆τf,avg)
overprediction may result from overestimation analysis is preferred since GRLWEAP can
of shaft friction, since the UWA-05 method is account for effects of the changing ratio of
based on resistance at 9 days after installation. pile shaft friction to total resistance, soil

DFI JOURNAL Vol. 4 No. 1 August 2010 [41]

 type dependant damping, as well as variable well for the range of pile diameters in
hammer stroke / fuel setting / efficiency this study provided that the ratio of pile
that cannot be accounted for in bearing diameter to pile wall thickness (D/t) was
graph analyses. approximately 40. For low values of D/t
or partial plugging during installation,
application of the effective area ratio
correction (Fig. 6) was required to minimize
the unconservative bias in the original
formulation.
• The modified UWA-05 method tended to
overpredict SRD when using τf,in/τf,out of 0.5
for drivability analyses. This observation
is expected since the modified UWA-05
method was calibrated to a database of
load tests performed, on average, 9 days
after pile driving. Increase in capacity with
time after driving needs to be considered
for assessment of driving resistance as
well as assessment of static capacity based
on SRD.
While the analysis procedures discussed
in this paper have been useful to highlight
mechanisms influencing diameter effects on
drivability of piles, an expanded database and
more detailed analyses of pile driving records
would extend these insights into a broader
understanding of friction fatigue for use in
pile drivability and static capacity analyses.
Additional experimental studies evaluating the
ratio of internal to external shaft friction are
also warranted.

ACKNOWLEDGEMENTS
Fugro is thanked for access to site
characterization, pile installation, and static
load test data related to the Euripides project.
Pile Dynamics, Inc. is acknowledged for use of
the academic version of GRLWEAP during these
studies. Reviewers of this paper are thanked for
their valuable suggestions and comments.

REFERENCES
1. Alm, T. & Hamre, L. (2001). Soil model for
pile drivability predictions based on CPT
interpretation. Proceedings of the 15th
International Conference on Soil Mechanics
and Geotechnical Engineering, 2, Istanbul,
Turkey, 1297-1302.
2. Clausen, C.J.F., Aas, P.M., & Karlsrud, K.
[FIG. 12] Effect of internal shaft friction (quantified using
τf,in/τf,out) on drivability assessment using the modified UWA-05 (2005). Bearing capacity of driven piles in
method sand, the NGI approach. Proceedings of the
International Symposium on Frontiers in
• Since the AH-01method was calibrated Offshore Geomechanics, Perth, Australia,
based on pile driving records, it performed 677-682.
[42] DFI JOURNAL Vol. 4 No. 1 August 2010
3. Dutt, R.N., Doyle, E.H., Collins, J.T., & 13. Paikowski, S.G., Whitman, R.V., & Baligh,
Ganguly, P. (1995). A simple model to M.M. (1989). A new look at the phenomenon
predict soil resistance to driving for long of offshore pile plugging. Marine
piles in deepwater normally consolidated Georesources & Geotechnology, 8(3), 213–
clays. Proceedings of the 27th Annual 230.
Offshore Technology Conference, OTC 7668, 14. Pile Dynamics, Inc. (2005). GRLWEAP:
Houston, USA, 257-269. Wave Equation Analysis of Pile Driving,
4. Foray, P., Balachowski, L., & Colliat, J.L. Procedures and Models Manual, Cleveland.
(1998). Bearing capacity of model piles driven 15. Randolph, M.F. (2000). Pile-soil interaction
in dense overconsolidated sands. Canadian for dynamic and static loading. Proceedings
Geotechnical Journal, 35(2), 374-385. of the 6th International Conference on
5. Heerema, E.P. (1980). Predicting pile Application of Stress Wave Theory to Piles,
driveability: Heather as an illustration of the Sao Paulo, Appendix, 3-11.
friction fatigue theory. Ground Engineering,
16. Randolph, M.F. (2003). Science and
13(3), 15-37.
empiricism in pile foundation design.
6. Jardine, F.M., Chow, F.C., Overy, R.F., & Géotechnique, 53(10), 847-875.
Standing, J.R. (2005). ICP design methods
17. Randolph, M.F., Dolwin. J., & Beck, R. (1994).
for driven piles in sands and clays. London:
Design of driven piles in sand. Géotechnique,
Thomas Telford.
44(3), 427-448.
7. Kolk, H.J., Baaijens, A.E., & Senders, M.
18. Randolph, M.F., Leong, E.C., & Houlsby, G.T.
(2005a). Design criteria for pipe piles in
(1991). One-dimensional analysis of soil plugs
silica sands. Proceedings of the International
in pipe piles. Géotechnique, 41(4), 587-598.
Symposium on Frontiers in Offshore
Geomechanics, Perth, Australia, 711-716. 19. Roussel, H.J. (1979). Pile driving analysis of
large diameter high capacity offshore pipe
8. Kolk, H. J., Baaijens, A.E., & Vergobi, P.
piles. PhD Thesis, Tulane University, New
(2005b). Results of axial load tests on pipe
Orleans, LA.
piles in very dense sands: the EURIPIDES JIP.
Proceedings of the International Symposium 20. Schneider, J.A., Xu, X., & Lehane, B.M., (2008).
on Frontiers in Offshore Geomechanics, Database assessment of CPT based design
Perth, Australia, 661-667. methods for axial capacity of driven piles in
siliceous sands. ASCE Journal of Geotechnical
9. Lehane, B.M., Jardine, R.J., Bond, A.J., &
& Geoenvironmental Engineering, 134(9),
Frank, R. (1993). Mechanisms of shaft
1227-1244.
friction in sand from instrumented pile
tests. Journal of Geotechnical Engineering, 21. Shioi, Y., Yoshida, O., Meta, T., & Homma,
119(1), 19-35. M. (1992). Estimation of bearing capacity
of steel pipe pile by static loading test
10. Lehane, B.M., Schneider, J.A. & Xu, X., (2005).
and stress-wave theory (Trans-Tokyo Bay
The UWA-05 method for prediction of axial
Highway). Application of Stress Wave Theory
capacity of driven piles in sand. Proceedings
to Piles, Rotterdam, 325-330.
of the International Symposium on Frontiers
in Offshore Geomechanics, Perth, Australia, 22. Skov, R. & Denver, H. (1988). Time-
683-690. dependency of bearing capacity of piles.
11. Lehane, B.M., Schneider, J.A. & Xu, X., (2007). Proceedings of the 3rd International
Development of the UWA-05 design method Conference on Application of Stress Wave
for open and closed ended driven piles in Theory to Piles, Ottawa, Canada, 888-897.
siliceous sand. Contemporary Issues in Deep 23. Stevens, R.F. (1988). The effect of soil plug
Foundations, ASCE GSP 158, 1-10. on driveability in clay. Proceedings of the 3rd
12. Paik, K., Salgado, R., Lee, J., & Kim, B. (2003). International Conference on Application of
Behavior of open- and closed-ended piles Stress Wave Theory to Piles, Ottawa, Canada,
driven into sands. Journal of Geotechnical 861-868.
and Geoenvironmental Engineering, 129(4),
296-306.

DFI JOURNAL Vol. 4 No. 1 August 2010 [43]

 24. Stevens, R.F., Wiltsie, E.A., & Turton, T.H.
(1982). Evaluating pile driveability for hard
clay, very dense sand, and rock. Proceedings
of the 14th Annual Offshore Technology
Conference, Houston, USA, 465-469.
25. Svinkin, M.R., Morgano, C.M., & Morvant, M.
(1994). Pile capacity as a function of time in
clayey and sandy soils. Proceedings of the
5th International Conference on Piling and
Deep Foundations, Bruges, Belgium, 1.11.1-
1.11.8.
26. Tavenas, F.A. & Audy, R. (1972). Limitations
of the driving formulas for predicting
bearing capacities of piles in sand. Canadian
Geotechnical Journal, 9(1), 47-62.
27. Toolan, F.E., Lings, M.L., & Mirza, U.
A. (1990). An appraisal of API RP2A
recommendations for determining skin
friction of piles in sand. Proceedings of
the 22nd Annual Offshore Technology
Conference, Houston, USA, 33-42.
28. White, D.J. (2005). A general framework
for shaft resistance on displacement piles
in sand. Proceedings of the International
Symposium on Frontiers in Offshore
Geomechanics, Perth, Australia, 697-703
29. White, D.J. & Lehane, B.M. (2004). Friction
fatigue on displacement piles in sand.
Géotechnique, 54(10), 645-658.
30. White, D.J., & Bolton, M.D. (2005).
Comparing CPT and pile base resistance in
sand. Proceedings of the Institute of Civil
Engineers, Geotechnical Engineering, 158, 3-14.
31. White, D.J., Schneider, J.A., & Lehane, B.M.
(2005). The influence of effective area ratio
on shaft friction of displacement piles
in sand. Proceedings of the International
Symposium on Frontiers in Offshore
Geomechanics, Perth, Australia, 741-747.
32. Xu, X., Schneider, J.A., & Lehane, B.M.
(2008). Cone penetration test (CPT)
methods for end-bearing assessment of
open- and closed-ended driven piles in
siliceous sand, Canadian Geotechnical
Journal, 45(8), 1130-1141.
33. Zuidberg, H.M. & Vergobbi, P. (1996).
Euripides, Load tests on large driven piles in
dense silica sands, Proceedings of the 28th
Annual OTC, Houston, 193-206.

[44] DFI JOURNAL Vol. 4 No. 1 August 2010

Support of Structures in Expansive Shale Using Recycled
Plastic Piles
Lance A. Roberts, South Dakota School of Mines and Technology, Rapid City, SD, USA;
[email protected]
Eric Brandner, South Dakota School of Mines and Technology, Rapid City, SD, USA

ABSTRACT
Expansive soil formations can be found throughout the United States. When subjected to wetting,
these formations have the potential to swell and exert large uplift forces on buildings and
foundations. Lightly loaded structures, such as single family residences founded in areas of expansive
soils, can be significantly damaged due to uplift movement from swelling actions. Designing an
economical deep foundation that can resist uplift forces is critical to prevent damage to these
structures. The current solutions to control uplift due to swelling soils, such as over-excavation and
replacement of the expansive material or the use of drilled shafts can be costly. Piles made from
recycled polymer materials could provide a solution. Due to a lower coefficient of friction along the
interface of the soil-pile interface compared to traditional pile materials, solid recycled plastic piles
can allow expansive soils to move nearly independently from the pile when wetted. This results in a
much smaller magnitude of uplift force being transferred to the structure, which minimizes the risk
of significant structural damage from excessive movements.

This paper presents the results of research conducted on the use of recycled plastic piles in an
expansive shale environment. The preliminary phase of the project involved the installation of six
recycled plastic piles at a test site on the South Dakota School of Mines and Technology campus.
Two of the piles were subjected to a full-scale compression load test in order to determine ultimate
capacity. The remaining four piles were subjected to long-term monitoring of uplift movement
during the course of the project. A concrete anchor was also installed at the test site for uplift
monitoring. Data gathered during the field and laboratory testing was utilized in a non-linear soil-
structure interaction model to predict the displacement behavior and internal stresses within a plastic
pile and concrete anchor subjected to uplift forces from the swelling shale. While more research is
needed to further understand the application for recycled plastic piles, the results from this research
indicate that their use is a viable alternative for support of lightly loaded structures in expansive soil
environments.

INTRODUCTION interior walls and slabs, serviceability damage,

Expansive soils typically contain minerals such such as inoperable doors and windows, and
as bentonite. When wetted, these materials structural damage of framing systems. While
attract large quantities of water within their the damage is often caused at a slow rate,
structure, thereby expanding and creating the repair cost for these problems can be
uplift forces on structures and foundations high. Estimated costs of damages caused by
(Das 2002). Expansive soil formations are expansive clay are as high as $7 billion annually
found throughout the United States, but are (FEMA 1997).
most prevalent in the Midwest from Texas, In most instances, the effects of expansive soils
Oklahoma, and the upper Missouri Valley are mitigated by over-excavation of the material
(Das 2006). Lightly loaded structures, such as and replacement with a granular material. The
residences and other single story structures, depth of replacement will vary from locale to
are particularly vulnerable to damage caused locale, but is generally 5 to 10 ft. (1.5 to 3 m). For
by the uplift forces. Due to the weight and sites with highly expansive soils and movement
large ground coverage area of these structures, sensitive structures, a combination of over-
differential movements as small as 1 in. per excavation and support of the structure with
20 ft. (25 mm per 6 m) can cause significant a deep foundation system may be required.
damage (Meehan and Karp 1994). This can Typically, small diameter drilled shafts are used
include cosmetic damage, such as cracking of for these applications. This allows the structure
DFI JOURNAL Vol. 4 No. 1 August 2010 [45]
 to be supported independently on grade beams smooth interface when compared to other
with crushable forms to minimize potential conventional pile materials, such as steel or
uplift forces from the soil. However, Johnson concrete. A research hypothesis for this study
and Stroman (1985) suggest that the uplift is that a smooth soil-pile interface would
forces on drilled shafts installed in expansive provide a low coefficient of friction, thereby
soils can be so extreme that drilled shafts have reducing pile displacement induced by the
failed due to tensile factures along their length. surrounding expansive material. A reduction
Designing an adequate foundation system to in pile displacement would in turn lower the
accommodate and resist the uplift loads can be magnitude of uplift forces within the pile itself,
a very costly initiative, thus new and innovative along with movement transferred to a structure.
methods to address this problem are warranted. Since expansive soils can cause significant
damage to lightly loaded, single story structures
OBJECTIVE AND SCOPE or to the deep foundation elements that
The objective of this research is to examine support them, the use of recycled plastic piles
the feasibility of utilizing recycled plastic piles to carry these buildings is of particular interest.
for support of lightly-loaded structures, such Fig. 1 provides a cross-section of the recycled
as single family residences, in an expansive plastic pile used for this project and identifies
soil environment. The project described the three major components.
herein includes full-scale field and laboratory
experimentation, as well as modeling the
response of plastic piles in swelling soil
conditions. During the course of the project,
six solid recycled plastic piles reinforced with
fiberglass bars were installed at a field test
site. A static load test was performed on two
15 ft. (4.5 m) piles equipped with strain gage
and telltale instrumentation. The remaining
four piles, instrumented with telltales, were
monitored for vertical movement for the
duration of one year. Laboratory tests were
conducted to determine the characteristics
of the in-situ soil at the test site. Field and
laboratory data were utilized to develop a
non-linear, soil-structure interaction model to
predict the displacement response of a plastic
[FIG. 1] Cross-section and components of Seapile® composite
pile and drilled shaft due to variations within marine piling.
expansive soil conditions.

RECYCLED PLASTIC PILES FIELD TEST SITE

Recycled plastic piles are commonly used Full-scale field load testing and long-term
in marine applications as an alternative to uplift monitoring were conducted on several
typical wooden, steel, and concrete piles. plastic piles installed in an expansive shale
Their composition enables resistance against site. The field test site was located on the
corrosion, marine borers, and deterioration campus of the South Dakota School of Mines
from the sun (FHWA 2006). Ductile and Technology (SDSM&T) in Rapid City, South
characteristics make them ideal as fender piles Dakota. As shown in Fig. 2, the site was located
and bumpers to absorb impact forces from approximately 300 ft. (90 m) north of the corner
ships and other small marine vessels (Seaward of E. St. Andrew Street and Hawthorne Avenue.
2009). Recycled plastic piles are produced in The selection of the test site was based on
many shapes and configurations. Variations both the presence of undisturbed highly
include hollow tubes, non-reinforced solid expansive soil and the absence of developed
piles, piles mixed with reinforcing fibers, steel structures that would be affected by the
reinforced piles, and fiberglass reinforced piles. research efforts. A geotechnical investigation
Since the recycled plastic piles are produced was conducted at the site by drilling three
in an extrusion process, they have a relatively boreholes and obtaining samples of the shale
[46] DFI JOURNAL Vol. 4 No. 1 August 2010
materials. The site was comprised of Belle affixed with four stainless steel telltales. Two
Fourche shale with little to no overburden at of the telltales were attached 8 in. (200 mm)
the surface. Laboratory testing of the shale from the toe of the pile while the other two
samples indicated swell pressures over 20 ksf telltales were attached near the mid-point of the
(960 kPa), along with swell percentages of up pile. The remaining four piles, used to monitor
to 18%. Additional information regarding the long-term vertical movement, received telltale
site conditions can be found in Brandner et al. instrumentation only. One of the 15 ft. (4.5 m)
(2009). piles, designated as Pile 3, contained four
telltales in a similar configuration to that of the
load test piles. The remaining 15 ft., 20 ft., and
25 ft.(4.5 m, 6.0 m and 7.6 m) piles, designated
as Pile 4, Pile 5, and Pile 6, respectively,
contained two telltales each, with one telltale
attached near the toe of the pile and another at
the mid-point of the pile. Plastic conduit was
placed over the telltales to allow independent
Field Test Site movement from the surrounding soil. A fully
instrumented toe portion of a plastic pile that
was used in one of the static load tests is shown
in Fig. 3.

[FIG. 2] Location of field test site on SDSM&T campus (from

Bing™).

FIELD TESTING PILES

Six recycled plastic piles, termed Seapiles® and
manufactured by Seaward Engineered Plastics,
were installed at the test site for compression
load testing and long-term vertical movement
monitoring. Four of the piles had a length of
15 ft. (4.5 m), one pile had a length of 20 ft.
(6.0 m), and one pile had a length of 25 ft
[FIG. 3] Strain gages and telltales attached at the toe of a pile.
(7.6 m). All recycled plastic piles had a diameter
of 13 in. (330 mm) and were reinforced with
12 - 1.5 in. (38 mm) diameter fiberglass PILE INSTALLATION
reinforcing bars spaced equidistantly around
The recycled plastic piles were installed in
the perimeter of the pile. The reinforcing bars
September 2008. The installation began by
were 0.55 in. (14 mm) from the edge of the pile.
selecting the locations of the piles. To reduce
Two 15 ft. (4.5 m) piles were subjected to static
the influence that adjacent piles would
field load testing. The remaining four piles,
have on each other during load testing, a
along with a concrete anchor, were monitored
minimum spacing of 15 ft. (4.5 m) was utilized.
for vertical movement due to seasonal moisture
Installation of all plastic piles was similar:
fluctuations.
1. An 18 in. (457 mm) diameter dry shaft was
All of the recycled plastic piles used in the field
drilled to a depth roughly 1 ft. (300 mm)
testing program were instrumented. Two piles
shorter than the length of the pile to be
selected for static load testing, designated as
installed;
Pile 1 and Pile 2, were each instrumented with
two Geokon® vibrating wire embedment strain 2. A large diameter clamp, attached to the
gages, located approximately 8 in. (200 mm) hoist on the drill rig, was secured to the
from the toe of the pile and installed on the head of the plastic pile;
pile per the manufacturer recommendations. In 3. The drill rig hoisted the pile into a vertical
addition to the strain gages, Pile 1 and 2 were position over the shaft;

DFI JOURNAL Vol. 4 No. 1 August 2010 [47]

 4. While the pile was lowered into the shaft, as infiltrate the borehole, causing the shale to
observed in Fig. 4, telltale extensions were swell at all depths along the pile and at the toe.
added along the length of the pile, along Minimal construction equipment was required
with plastic conduit over each telltale. The for installation of the plastic piles. A Production
conduit was secured with duct tape. A Digger 6030 drill rig with hoist was utilized,
metal washer was placed behind the plastic along with a skid steer to move material on-site.
conduit for vertical alignment. Strain gage The installation time per pile varied due to the
cables were also attached to the perimeter amount of instrumentation attached to the pile.
of the pile using duct tape. The plastic pile Total installation time included drilling to the
was lowered into the shaft in such a manner appropriate depth, installing instrumentation,
to minimize the amount of loose material placing/seating the pile, and backfilling the
knocked from the side of the excavation; annulus with the low plasticity clay. Pile 1
5. Once at the bottom of the shaft, the pile was and Pile 2, which contained strain gages and
leveled and repositioned within the shaft to telltales, required approximately one hour each
ensure a relatively equal annulus around the to install as extreme caution was exercised to
perimeter of the pile; ensure no damage to the strain gauges of the
6. A low plasticity clay material was tamped telltales. Conversely, Pile 4, Pile 5, and Pile 6,
into the annulus to stabilize the pile within which were instrumented with two telltales
the hole and prevent water from flowing each, took only about thirty minutes per pile
into the annulus. to install. After installation was completed,
verticality measurements were obtained for all
plastic piles. The measurements indicated that
all piles were within the general plumbness
requirements reported in most drilled shaft
construction specifications (1/4” per 1 foot or
6 mm per 300 mm). During installation of the
plastic piles, drill cuttings from various depths
were sampled in order to determine the average
in-situ moisture content of the shale versus
depth as shown in Table 1.
A cast-in-place, concrete anchor with a diameter
of 6 in (152 mm) was placed in October
2008 for uplift monitoring and performance
comparison with the recycled plastic piles. The
anchor was installed to a depth of 15 ft (4.5 m).
A Grade 75, #14 (45 mm) Dywidag threadbar
was placed in the center of the anchor for its
entire length.

[TABLE 1] In-situ moisture content (%) of shale

during installation.

[FIG. 4] Pile being lowered into the shaft. Depth Pile

(ft)
1 2 3 4 5 6
Clay from the Spearfish Formation near Rapid
City was chosen because of its low swell 5 17.5 16.7 18.4 14.6 15.5 13.5
potential, low permeability when compacted,
and abundant local availability. The clay 10 12.7 12.5 13.1 13.0 13.4 12.8
was tested for Atterberg limits and found to
possess a LL of 30 and a PI of 9. The reuse of 15 13.3 12.8 13.1 12.7 12.5 12.8
the excavation spoils from the drilling process
was not considered since adequate compaction 20 12.3 12.6
of the shale was not feasible. Sand was also
eliminated as backfill material due to its high 25 13.2
permeability and potential to allow water to

[48] DFI JOURNAL Vol. 4 No. 1 August 2010

STATIC LOAD TESTING the embedment strain gages. All strain gages
The static load testing of Pile 1 and Pile 2 took were connected to a vibrating wire readout
place in November 2008. Four reaction anchors device through a terminal switch box. The
were installed near Pile 1 and Pile 2 to provide movements of the four telltales were monitored
adequate resistance for a reaction support and recorded at each loading increment.
frame during the load testing procedure. Vertical movements of the reaction anchors
Each anchor was drilled to a depth of 30 to were monitored using an automatic level and
35 ft. (9 to 10.5 m) and had a diameter of 6 in. level rod.
(152 mm). A Grade 75, #14 (45 mm) Dywidag Pile 2 was tested first due to its location
threadbar was inserted into each hole before at the test site. While attempting the first
being tremie grouted with cement grout. The load increment of 10 kips (44.5 kN), Pile
reaction support frame, consisting of built- 2 experienced a total settlement of nearly
up steel sections and wood cribbing, was 3 in. (75 mm). The excessive settlement was
constructed and integrated with the reaction theorized to be the result of loose drill cuttings
anchors as shown in Fig. 5. and other unconsolidated material beneath
the toe of the pile from the installation. The
application of load against Pile 2 continued
until resistance was observed via the pressure
gage attached to the hydraulic jack. Once a
10 kip (44.5 kN) load was attained, the pile
was completely unloaded and the dial gauge
instrumentation was reset. The load test then
progressed as originally planned. A load-
settlement plot of Pile 2 is shown in Fig. 6. The
hydraulic jack was “bumped” at each load to
compensate for loss in pressure and creep.
This resulted in increased vertical displacement
at each bump in the hydraulic jack pressure.
The load testing of Pile 2 provided valuable
[FIG. 5] Reaction frame setup for static load testing of plastic information for development of a load testing
piles.
procedure for Pile 1 as follows:

The static load testing generally followed 1. Apply a seating load of 10 kips on the pile
Procedure “A” of ASTM D1143. However, the to ensure that the pile was properly seated
test piles were loaded in increments of 10 kips at the toe. Maintain this load on the pile for
(44.5 kN) because the ultimate compressive at least 30 minutes;
capacity of the plastic piles was unknown. 2. Unload the pile completely and set all dial
Load was applied by a hydraulic jack with a gauge instrumentation;
calibrated pressure gage. The hydraulic jack 3. Perform the load test by applying the load in
had a capacity of 1,000 kips (4.5 MN). Vertical 10 kip (44.5 kN) increments. Do not bump
head settlement measurements of the test the hydraulic jack to maintain the load;
piles were recorded at each loading increment
using four dial gage indicators with sensitivity 4. Record head settlement, telltale, and strain
of 0.001 in. (0.025 mm). The dial gages were gage readings at each load increment after
equidistantly spaced around the head of the the pressure gauge on the hydraulic jack has
test pile and were attached to a reference frame, stabilized.
which was independent from the test pile Pile 1 was loaded until plunging was evident
and test reaction system. Four additional dial and the hydraulic jack was no longer able to
gages, with accuracy to 0.001 in., (0.025 mm) register an increase in pressure. The load-
were used to measure the movements of the settlement plot of the Pile 1 load test data
telltales during the compression test of Pile 2. is shown in Fig. 7. Pile 1 was allowed to set
A standard tape measure was used to determine overnight unloaded and was reloaded the next
telltale movement during the compression morning without the seating procedure. The
test of Pile 1. Strains within the test pile were load-settlement plot for the reloading of Pile 1
measured during each loading increment using is also shown in Fig. 7.

DFI JOURNAL Vol. 4 No. 1 August 2010 [49]

 Load (kip) field load test data. A review of the relevant
0 20 40 60 80 100 120 literature resulted in the identification of more
0
than ten different criteria (Fellenius 1990,
Salgado 2006). Although many of these criteria
0.5
were developed for specific deep foundation
1 systems, the use of varying criteria often
results in a different ultimate capacity for
S ettlem en t (in )

1.5 the deep foundation (Yang 2008). A method

that can interpret the results of the field load
2
test data without bias toward the predicted
2.5
ultimate capacity is beneficial to ensure
efficiency in the design. To that end, the field
3 load test data was analyzed using the t-z
model approach.
3.5
The t-z model approach has been widely
[FIG 6] Load-settlement curve for Pile 2. utilized for analytical and numerical modeling
of deep foundation load-settlement behavior
Load (kip) (Randolph and Wroth 1978, Scott 1981, Misra
0 20 40 60 80 100 120
0
and Chen 2004, Reese et al. 2006). In the t-z
model, axial resistance along the interface of a
0.5 deep foundation and at the toe is represented
by a spring-slider system. The spring-slider
1
system is assumed to behave in either an ideal
Settlement (in)

Original Test

1.5 elasto-plastic or hyperbolic manner, and each

spring is assigned a strength and stiffness
2
magnitude. Along the interface, the spring-
2.5 slider system represents the annular band of
Reload
soil that participates in the axial resistance, as
3
shown in Fig. 8, and thus an imperfect interface
3.5 is assumed (i.e. movement of foundation and
[FIG. 7] Load-settlement curve for Pile 1 (original load test interface soil are not equal). The foundation
and reload test). itself is assumed to behave elastically
throughout considering that the load required
The design load for a single deep foundation to reach yield of the interface is much smaller
element supporting a lightly-loaded structure is than that to yield the foundation material.
typically on the order of 25 to 35 kips (110 to
155 kN). It is observed from Figs. 6 and 7
that the plastic piles are able to adequately
support a design load of this magnitude with
a head settlement of approximately 0.25 in.
(6 mm). It is evident from Fig. 7 that the load-
unload-reload of Pile 1 caused an increase in
the stiffness of the soil at the toe, but that the
pile achieved its plunging load much sooner
compared to the original load test. The static
load tests demonstrated that adequate cleanout
of loose material from the shaft bottom, along
with seating of the pile during placement of
[FIG. 8] Schematic of foundation when describing behavior
the pile in the shaft, appears to be critical in
using t-z model.
order to ensure acceptable load-settlement and
ultimate capacity performance.
In the t-z model, the following equilibrium
equation is solved (Reese et al. 2006):
ANALYSIS OF LOAD TEST DATA
There are a number of methods to compute the d2 u
Km − K u (z ) = 0 (1)
ultimate capacity of a deep foundation using dz 2
[50] DFI JOURNAL Vol. 4 No. 1 August 2010
where, u(z) is the deep foundation deformation, deep foundation perimeter and the ultimate
Km is the deep foundation axial stiffness, and K shear strength of the soil-structure interface,
is the shear modulus of soil-structure interface τu, in drained or undrained conditions. The toe
sub-grade reaction. soil force-displacement curve can be similarly
The solution to the equilibrium equation, represented using the initial tangent tip soil
assuming non-linear behavior of the spring- stiffness, Kti, and the ultimate (asymptotic)
slider system, is presented elsewhere (Misra strength of the tip soil, qt, in drained or
and Roberts 2006). However, for completeness, undrained conditions. The soil-structure
there are a number of additional parameters interface strength and stiffness parameters
necessary to utilize the t-z model approach. are related to the deep foundation type,
Under a compression load, the deep foundation construction techniques, and the properties
will develop a toe force, Pt, proportional to the of the soil strata. The toe soil strength and
toe displacement, ut, given by: stiffness parameters are generally only related
to the deep foundation type and properties of
Pt = K t u t (2) the soil at the toe.

where, Kt is the toe soil stiffness. Based on

theories for rigid punch bearing upon elastic
half-space, the initial tangent stiffness of
the toe soil, Kti, may be related to foundation
diameter and the elastic properties of the toe
soil as follows (Misra and Roberts 2006):

0.3π D E s
K ti = (3)
( 1 − μ s2 )

where, Es is the modulus of soil sub-grade

reaction, μs is toe soil Poisson’s ratio, and D is
the diameter of the foundation element.
As an axial load is applied to the head of the
deep foundation, the springs will displace based [FIG. 9] Interface force-displacement curve for a hyperbolic
spring-slider system.
on their strength and stiffness properties and
the deep foundation element will settle. As the
applied axial load increases, yielding of the A t-z model back-computation procedure
springs begins in the uppermost elements and using load test data is described extensively
progresses to the toe. At some load, all springs in Roberts et al. (2008). The method is an
yield and the deep foundation will fail by iterative process where the load-settlement and
plunging. The ultimate capacity of the toe soil strain magnitude predicted by the t-z model
can be determined assuming a punching shear is matched to the load-settlement and strain
failure from the following equation: magnitudes of the deep foundation using the
load test data. Software has been developed at
Putip = qt Am (4) SDSM&T to perform the t-z back-computation.
The back-computation procedure using the
where, qt is the unit tip bearing resistance and t-z model and the load test data from Pile 1
Am is the cross sectional area of the foundation is shown in Figs. 10 and 11. In Fig. 10, head-
element. The use of a t-z model approach settlement measurements recorded from the
therefore results in the development of a load- field load test (points) are superimposed on
settlement curve that represents the behavior the predicted head-settlement curve from the
of the deep foundation over a wide range of t-z model (solid line). In Fig. 11, the telltale
applied axial loads. movement at the pile toe from the field load
In Fig. 9, the force-displacement behavior for a test (points) is superimposed on the predicted
nonlinear (hyperbolic) spring-slider system is toe movement from the t-z model (solid line).
shown. In Fig. 9, Kinit is the initial tangent shear By fitting the t-z model to the head-settlement
modulus of soil-structure interface, and qo is and telltale data, the strength and stiffness
the ultimate (asymptotic) strength of the soil- parameters for the interface and toe soil springs
structure interface given by the product of the were back-computed. Non-linear (hyperbolic)
DFI JOURNAL Vol. 4 No. 1 August 2010 [51]
 displacement behavior was assumed for all the pile, should allow for a reduction in uplift
springs in the back-computation process. In forces due to the expansion of the shale at the
addition, the composite elastic modulus of the site. Interestingly, approximately 30% of the
plastic piles was assumed as 450 ksi (3,100 MPa) total head settlement at any given load is due to
(Seaward 2009), the non-interaction zone length elastic shortening of the pile as the composite
was taken as 1 ft. (305 mm) due to potential elastic modulus of the plastic pile is on the
ground disturbance, and the Poisson’s ratio of order of 10% that of conventional concrete.
the shale at the pile toe was assumed as 0.30. It is important to note that in Figs. 10 and 11,
The back-computed t-z model strength and differences between the measured values from
stiffness parameters are reported in Table 2. the field load tests (points) and predicted values
from the t-z model (solid line) are realized.
[TABLE 2] Strength and stiffness parameters of
Discrepancies of this nature are often due to
the t-z model interface and toe springs.
slight errors in the field measurements. In the
Interface Springs Toe Spring back-computation process, the predicted values
Strength Stiffness Strength Stiffness
from the t-z model are within ±10% of the field
measured values.
psi ksi ksf ksf

0.75 0.5 100 2000 SOIL SWELL MODEL

The swell potential of soil varies between
Load (kip)
different geological formations (Das 2002). Key
0 20 40 60 80 100
0
factors characterizing the magnitude of swell
pressure include the in-situ void ratio, eo, and
0.5 change in the degree of saturation, ∆S, of the
Head Settlement (in)

soil. During the subsurface investigation of

1
the test site, laboratory tests were conducted
1.5 to determine the swell pressure at 5 ft.
(1.5 m) intervals along the entire depth of
2
the exploratory borings. Using the in-situ
2.5 void ratio, change in the degree of saturation,
and swell pressure data from this laboratory
3 testing, a potential swell pressure equation was
[FIG. 10] Measured head settlement versus predicted head developed to predict the response of the Belle
settlement from t-z model. Fourche shale at the test site for a range of in-
situ conditions:
Load (kip)
0 20 40 60 80 100 Sp = 0.458 Pa ΔS 0.874 eo -5.016 (5)
0

0.5
where, Sp is the swell pressure (psf) and Pa is
Tip Settlement (in)

the atmospheric pressure. The value of ∆S

1 in Eq. 5 is taken as a decimal. The equation
was developed using multiple variable linear
1.5
regression and was found to have a correlation
2 coefficient of 0.98, indicating a strong
correlation within the test data.
2.5
Using Eq. 5, the swell pressure for the shale at
3 the test site was predicted for different field
[FIG 11] Measured telltale pile toe movement versus conditions by: (1) varying eo while maintaining
predicted toe movement from t-z model. a constant value of ∆S as shown in Fig. 12, and
(2) varying ∆S while maintaining a constant
As shown in Table 2, the pile interface strength value of eo, as shown in Fig. 13. From Fig. 12,
and stiffness parameters are minimal and that it is observed that as eo increases, the swell
a majority of the pile resistance is due to the pressure of the material will decrease sharply.
spring at the pile toe. This confirms that the This is expected and caused by an increase in
combination of the pile installation method, the volume of air and water in the three phase
along with the low coefficient of friction of soil element. This in turn allows for greater

[52] DFI JOURNAL Vol. 4 No. 1 August 2010

expansion of the soil particles as the soil a tensile fracture failure of the drilled shaft
particles must expand farther to exert pressure (Johnson and Stroman 1985). Therefore, the
against others. From Fig. 13, it is observed that internal forces and uplift movement of a drilled
as ∆S increases, the swell pressure will also shaft were modeled to observe whether tensile
increase. This is also expected since the greater fracture failure could occur due to the internal
the change in the degree of saturation, the pile forces from soil expansion. In addition,
greater the amount of water that is absorbed a similar model was developed for a recycled
into the soil system and thus the greater the plastic pile with a diameter of 13 in. (330 mm)
expansion. When comparing Figs. 12 and 13, and length of 15 ft. (4.5 m). A comparison of the
the in-situ void ratio, eo, has a significantly two models allows for improved understanding
greater impact on the swell pressure over the of the behavior of these deep foundation
change in the degree of saturation, ∆S. systems in expansive soil and permits easy
10000000
parametric analyses and performance prediction
ΔS (%) studies to be conducted.
1000000
20 40 60 80 100
The deep foundation model was developed
using the t-z model method. For the drilled
shaft, τu and Kinit for the t-z springs were
100000
assumed as 20 psi (138 kPa) and 5 ksi
Sp (psf)

(34.5 MPa), respectively, based on the

10000
characteristics of the shale at the site and
using criteria given in Coduto (2001) and
1000 FHWA (1999). For the plastic pile, the strength
and stiffness values for the t-z springs were
100
assigned the back-computed values from
0 0.2 0.4 0.6 0.8 1 the field load test. A compressive load of
eo
30 kips (133 kN) was applied to each model
[FIG. 12] Swell pressure versus eo for various magnitudes of ΔS. to simulate the typical design load for a
foundation element within a lightly loaded
eo
structure. Swell pressures were computed
0.2 0.3 0.4 0.5 0.6 0.7 0.8
10000000 using Eq. 5 for various values of eo along the
foundation interface, while assuming that ∆S
1000000 was 100% at the ground surface and linearly
decreased to 0% at the depth of wetting.
100000
An upward swell force was applied at each
Sp (psf)

discretized nodal element along the length of

the foundation. The upward swell force, SF,
10000
at each element was determined as follows:
1000
SF = Sp π D Δz ß (6)

100 where, ∆z is the discretized element length and

0 20 40 60 80 100 120
β is a factor to account for the soil-pile interface
Δ S (%)
friction angle and in-situ coefficient of earth
[FIG. 13] Swell pressure versus ΔS for various magnitudes of eo. pressure (FHWA 1999). All additional variables
have been defined previously. A β value of
1.10 and 0.10 was assigned to the drilled shaft
PILE MODEL USING SWELL EQUATION and plastic pile, respectively, during the swell
The diameter and length of drilled shafts used force modeling. The β value for the plastic
in the mitigation of expansive soil will vary; pile was based on the t-z back-computation of
however, the minimum size typically used in the field load test data, while the β value for
expansive shale environments in South Dakota the drilled shaft was based on typical values
is a 24 in. (610 mm) diameter shaft with a length for intermediate geomaterials given in Coduto
of 25 ft. (7.6 m). As mentioned previously, the (2001) and FHWA (1999). The swell forces along
uplift force exerted on a drilled shaft due to the the foundation interface provided resistance
expansion of soil can be large enough to cause to the applied head load of 30 kips (133 kN)

DFI JOURNAL Vol. 4 No. 1 August 2010 [53]

 (along with the weight of the foundation), while Stress in Plastic Pile (psi)
uplift resistance from the shale acted only along 0 50 100 150 200 250
0
the soil-structure interface below the depth of
wetting. It was assumed that the toe did not
contribute any uplift resistance due to suction. 5

Lastly, a perfect interface between the shale and

clay was assumed which would allow for full

Depth (ft)
transfer of the developed swell pressure from 10
Initial Void Ratio (eo)
the expansive shale through the clay annulus to
0.60
the foundation element in order to satisfy strain
15 0.70
compatibility of the radial soil.
0.80

Using the developed deep foundation soil

expansion model described above, the internal 20

axial stresses within a drilled shaft and plastic

[FIG. 15] Internal stresses developed in plastic pile under
pile subjected to an uplift force were computed expansive shale uplift forces.
for a range of in-situ void ratios. The computed
axial stresses are shown in Figs. 14 and 15 for design load of 30 kips (133 kN). Overall, both
the drilled shaft and plastic pile, respectively. It the drilled shaft and plastic pile exhibited
is observed in Fig. 14 that as the magnitude of upward movement and the movement of the
the in-situ void ratio, eo, decreases, significant drilled shaft is always greater than the neutral
axial stresses are developed within the drilled point. As eo decreases, the overall rate of
shaft at the depth of wetting. The magnitude upward displacement between the plastic
of the axial stress is such that tensile fracturing pile and drilled shaft is nearly equal, but
of normal strength concrete is highly likely, the upward displacement of the plastic pile
thereby confirming observations by Johnson reaches infinity sooner than the drilled shaft.
and Stroman (1985) and Chapel and Nelson Based on these results, the research team is
(2000). In Fig. 15, the axial stresses developed examining various methods to increase the
within the plastic pile also increase as eo side resistance of the plastic pile near the toe
decreases. However, the axial stresses within to help control the magnitude of uplift.
the plastic pile are always compressive and
1
thus structural failure of the plastic pile due to
tensile stress is not a concern. 0.8
Head Displacement (in)

Stress in Drilled Shaft (psi) 0.6

-1600 -1400 -1200 -1000 -800 -600 -400 -200 0 200 400
0.4
0

0.2
Plastic Pile
5 Upward
Drilled Shaft
0
Downward
10 -0.2
Depth (ft)

-0.4
15

Initial Void Ratio (eo) -0.6

0 0.2 0.4 0.6 0.8 1
20 0.55
0.60
eo
0.70
25 [FIG. 16] Head displacement for drilled shaft and plastic pile
0.80 Tension Compression
under expansive shale uplift forces.
30

[FIG. 14] Internal stresses developed in drilled shaft under

expansive shale uplift forces. PILE UPLIFT MONITORING AND SITE
FLOODING
Head displacements for the drilled shaft and Pile movement due to seasonal moisture
plastic pile were also determined using the fluctuations was a crucial component for this
deep foundation soil expansion model. Fig. 16 project. To monitor uplift, weekly surveys
shows the predicted head displacement for using a conventional surveying level and
both the drilled shaft and plastic pile when rod, with an accuracy of 0.01 ft. (3 mm),
subjected to variable swell pressures under a were conducted over the course of one year.

[54] DFI JOURNAL Vol. 4 No. 1 August 2010

Surveys included pile head elevation and pile was removed from the interior of the berm area
elongation measurements from the telltales. prior to flooding. To fill the area within the
Precipitation totals during each weekly survey berm, 10,500 gallons (40,000 litres) of water
interval were also recorded. Establishing a was required.
benchmark was especially difficult at the site
Site flooding occurred in July 2009. Prior to
due to the presence of the expansive shale and
flooding, moisture samples were taken near
the potential for movement of a temporary
the outer edge of the berm using a hand soil
benchmark placed in the ground. For this
sampler. Samples were obtained at depths
reason, a fire hydrant located near the site was
ranging from 0.5 to 4 ft. (0.15 to 1.2 m) The
used and the survey datum was established
use of the hand sampler limited the depth at
from one of the bolts connecting the bonnet to
which samples could be obtained. The degree
the barrel of the hydrant.
of saturation of the samples was determined
Overall, the plastic piles and concrete anchor and found to increase from 65% at the surface
experienced minimal vertical movement during to 94% at 4 ft. (1.2 m). Elevations of the plastic
the one year monitoring period, as shown in piles and concrete anchor were determined
Fig. 17. Due to the accuracy of the surveying prior to site flooding.
equipment, it is possible that most of the
Flooding the test site took approximately two
apparent vertical movement may actually be
hours. Standing water was present at the
measurement error, as all plastic piles and the
conclusion of the flooding process with an
concrete anchor generally moved the same
average water depth of about 6 in. (150 mm) as
magnitude (upward or downward) at each
shown in Fig. 18. The overcast and cool weather
measurement as shown in Fig. 17. However,
conditions during the day limited the amount
the plastic piles did appear to settle within
of evaporation. Complete infiltration into the
the first couple of months likely due to the
shale took approximately 12 hours. The next
presence of loose drill cuttings and other
day, moisture samples were taken inside the
material beneath the toe of the pile. Pile 6
berm to a depth of 4 ft. (1.2 m). The degree of
settled the most, with a magnitude of 0.13 ft.
saturation of all samples was found to be 100%.
(40 mm). It is unclear why Pile 6 settled such a
As before, the use of a hand sampler limited the
large magnitude.
depth of the sampling.
Pile 3 Pile 4 Pile 5 Pile 6 Concrete Anchor
17.00

16.50
Elevation (ft)

16.00

15.50

15.00

Date

[FIG. 17] Observed maximum and minimum elevations of the

plastic piles and anchor.
[FIG. 18] Flooding the test site.
The small apparent magnitude of vertical
movement observed during the monitoring Pile head elevations were recorded for a
period encouraged the researchers to examine a period of one week after the site flooding.
worst case scenario of moisture variation at the However, as shown in Fig. 17 during this
test site. This was accomplished by flooding a period, very minor head movements were
portion of the site. To control the area of water recorded for the plastic piles and concrete
infiltration, an earth berm was constructed anchor. In order to confirm the site
around Piles 3, 4, 5, 6, and the concrete anchor. observations, the potential head displacement
The thicker vegetation and miscellaneous debris of the plastic piles and concrete anchor

DFI JOURNAL Vol. 4 No. 1 August 2010 [55]

 were computed using the developed deep increase in the number of damaged homes in
foundation soil expansion model described Rapid City over the last five years. Therefore,
previously, along with the magnitudes of ∆S the cost analysis assumed a one story, single
given in Table 3 for the upper 4 ft. (1.2 m) family dwelling with a basement plan area
of the shale due to the flooding activities. of 1500 ft2 (140 m2). The typical expansive
Substituting these ∆S values into Eq. 5, along soil mitigation options were examined and
with a constant value of eo = 0.60, provides included (1) over-excavation and replacement
the magnitude of swell pressure at these of the expansive shale with a granular material
depths. The swell pressures, provided in to a depth of 10 ft. (3 m) beneath a slab-on-
Table 3, were substituted in Eq. 6 to compute grade basement floor; (2) use of 24 in. (610
the swell forces within the shale zone. The mm) diameter, 25 ft. (7.6 m) long drilled shafts
swell forces were applied linearly along each supporting a structural floor/grade beam
discretized element of the t-z model within system placed on crushable forms; and (3) use
the upper 4 ft. (1.2 m) of the shale. All of 13 in. (330 mm) diameter, 15 ft. (4.6 m) long
assumptions regarding the interface behavior recycled plastic piles supporting the same
of the plastic piles and concrete anchor within structural floor/grade beam system. Based
the t-z model from the previous modeling on estimated total loads, it was assumed that
exercise were maintained. Based on this 18 plastic piles would be required, while only
analysis, the upward head displacement of 12 drilled shafts would be necessary. Using
the plastic piles and concrete anchor were information provided by local contractors,
predicted to be 0.034 in. (0.86 mm) and the use of recycled plastic piles could provide
0.047 in. (1.19 mm), respectively. These a cost savings of nearly 10% and 20% when
magnitudes are significantly below the compared to Option 1 and 2, respectively.
accuracy of the surveying equipment used The efficiency of using recycled plastic
on this project. Interestingly, if the depth of piles as deep foundation elements was also
wetting was actually 10 ft. (3 m), the upward investigated. A drilled shaft foundation
head displacement of the plastic piles and must be long enough to provide necessary
concrete anchor would be approximately resistance to the uplift forces. However, the
0.073 in. (1.85 mm) and 0.055 in (1.40 mm)., minimum length required for uplift translates
respectively, given the saturation conditions to a relatively high ultimate compressive
at the site. To that end, obtaining actual capacity when compared to the design loads.
uplift measurements was not possible without This over-design may be as high as 11 times
the use of more sensitive equipment. depending on the weathering characteristics
and unconfined compressive strength of
[TABLE 3] Degree of saturation before and after the shale, along with the construction
flooding test site. method for the drilled shaft. Recycled plastic
S (%) piles, however, can adequately support the
Depth ∆S Sp anticipated design loads with a sufficient
(ft) Pre Post (%) (psf) factor of safety against ultimate bearing
Flood Flood
failure, which leads to increased efficiency in
0.5 66.4 100 34 4843 the design.
Environmental impact was also considered.
1 64.1 100 36 5125 An over-excavation and replacement option
2 70.3 100 30 4348 would likely produce a significant amount of
soil waste material, along with the highest
3 83.4 100 17 2607 emissions from the construction equipment.
The difference in the volume of waste material
4 93.7 100 6 1120 generated from the use of drilled shafts
or recycled plastic piles would likely be
COST AND BENEFIT ANALYSIS insignificant and the difference in emissions
A cost benefit analysis was performed to is not known at this time. However, the use
determine the potential economic impact of of recycled plastic piles may be considered
using recycled plastic piles. The analysis “green engineering”, which may motivate
focused on expansive shale mitigation builders and geotechnical engineers to
techniques for residential structures due to an consider this type of foundation system.
[56] DFI JOURNAL Vol. 4 No. 1 August 2010
CONCLUSIONS soil expansion model was developed based
Expansive soil is a problematic material found on predicted swell pressures. The predicted
in the Midwest of the United States from Texas swell pressures were found to be a function
to the upper Missouri Valley. Lightly loaded of in-situ void ratio and change in the degree
structures constructed on expansive soils are of saturation of the shale. The model was
susceptible to damage due to uplift forces employed to predict the internal stresses and
caused by swelling action. This paper examines uplift behavior of a plastic pile and drilled shaft
the potential use of recycled plastic piles for under varying degrees of swell pressure. While
support of such structures. Due to a low the model indicated significant differences in
coefficient of friction along the pile interface, the tensile stresses developed within each type
plastic piles should allow the expansive soil of deep foundation system, the model predicted
to move independently of the pile without similar magnitudes of vertical movement.
causing significant uplift forces that could be A cost benefit analysis was conducted by
transferred to the structure. examining typical mitigation options for a
An expansive shale test site was chosen on the single family residence. The analysis revealed
SDSM&T campus in order to perform two field that the use of recycled plastic piles could
load tests and observe the long-term vertical result in cost savings of between 10% and 20%
movement of plastic piles. After an initial field over conventional methods. Other potential
investigation consisting of three exploratory benefits of using recycled plastic piles include
borings, the shale was established to have less soil waste, less construction emissions, and
a medium to high potential of swelling. Six better efficiency in terms of capacity.
recycled plastic Seaward Seapiles® of varying
lengths were installed at the site by drilling
ACKNOWLEDGEMENTS
an oversized hole, inserting the pile and back- This research was made possible by a
filling the annulus between the pile and hole competitive research grant from the South
with low plasticity clay. Two of the installed Dakota Board of Regents and generous
piles were subjected to static load tests. After donations of time, expertise, and financial
conducting the field load tests, a t-z model support from Hayward Baker Inc. Special
was used to back-compute the strength and thanks are also due to Seaward Engineered
stiffness parameters for the shale along the Plastics for the donation of the recycled plastic
soil-pile interface and at the toe. Vertical piles, FMG, Inc. for geotechnical exploration and
movements of the four remaining piles, along testing services, and B&B Foundation Services
with a small diameter concrete anchor, were for pile installation.
monitored with weekly surveys for the duration
of a one year period. Negligible movement of REFERENCES
both the plastic piles and concrete anchor was 1. Brandner, E., Roberts, L.A. and Surovek,
observed, even after flooding the test site with A. 2009. Investigation of recycled plastic
water, predominately due to the accuracy of piles for support of structures in expansive
the surveying equipment. The field load tests shale environments. Proceedings of the 34th
did highlight the importance of ensuring that Annual Conference on Deep Foundations,
prior to installation of the piles in the pre- Kansas City.
drilled hole, the bottom of the hole must be 2. Chapel, T.A. and Nelson, J.D. 2000.
relatively free of loose drill cuttings and other Strain measurements of concrete piers in
slough. Since some loose material is likely to expansive soils. GSP 106, Proceedings of
reside in the hole regardless of the cleanout Geo-Denver 2000, Denver, pp. 151-163.
method, subjecting the piles to a seating load,
3. Coduto, D.P. 2001. Foundation design
potentially applied using the drill rig prior to
– principles and practices. Prentice Hall,
backfilling the annulus, should ensure adequate
Upper Saddle River, 883 p.
resistance and performance of the piles during
service life. This was observed during field load 4. Das, B.M. 2002. Principles of geotechnical
testing of Pile 1, which underwent a seating load engineering. Brooks/Cole, Pacific Grove,
of 10 kips (44.5 kN) prior to full load testing. 686 p.
Using swell test data from shale samples 5. Das, B.M. 2007. Principles of foundation
collected at the project site, a deep foundation engineering. CL-Engineering, Toronto, 794 p.

DFI JOURNAL Vol. 4 No. 1 August 2010 [57]

 6. Fellenius, B.H. 1990. Guidelines for the 19. Scott, R.F. 1981. Foundation analysis. First
interpretation and analysis of the static Edition. Prentice Hall, Englewood Cliffs, NJ.
load test. Deep Foundations Institute, 20. Yang, X., Han, J., Parsons, R.L. and
Hawthorne, NJ. Henthorne, R.W. 2008. Resistance factors for
7. FEMA. 1997. Multi-hazard identification drilled shafts in weak rock based on O-Cell
and risk assessment (MHIRA). Subpart B: test data. Proceedings of the TRB Annual
Geological hazards, Chapter 11: Expansive Meeting 2008, Transportation Research
Soils. Federal Emergency Management Board, Washington, D.C.
Agency, pp. 122-128.
8. FHWA. 1999. Drilled shafts: construction
procedures and design methods, FHWA-IF-
99-025. FHWA, McLean, VA.
9. FHWA. 2006. Behavior of fiber-reinforced
polymer composite piles under vertical loads,
FHWA-HRT-04-107. FHWA, McLean, VA.
10. Johnson, L.D. and Stroman, W.R. 1985. Long-
term behavior of a drilled shaft in expansive
soil. Transportation Research Board,
Transportation Research Record No. 1032,
pp. 53-59.
11. Meehan, R.L. and Karp, L.B. 1994. California
housing damage related to expansive soils.
Journal of Performance of Constructed
Facilities, ASCE, 8(2), pp. 139-157.
12. Misra, A. and Chen, C.-H. 2004. Analytical
solutions for micropile design under
tension and compression. Geotechnical and
Geological Engineering, 22(2), pp. 199-225.
13. Randolph, M.F. and Wroth, C.P. 1978.
Analysis of deformation of vertically
loaded piles. Journal of the Geotechnical
Engineering Division, ASCE, 104(12), pp.
1465-1488.
14. Reese, L.C., Isenhower, W.M., and Wang S.-H.
2006. Analysis and design of shallow and
deep foundations. John Wiley, New Jersey,
574 p.
15. Misra, A. and Roberts, L.A. 2006.
Probabilistic analysis of drilled shaft service
limit state using ‘t-z’ method. Canadian
Geotechnical Journal, 43(12), pp. 1324-1332.
16. Roberts, L.A., Misra, A. and Levorson,
S.M. 2008. Practical method for Load and
Resistance Factor Design (LRFD) of deep
foundations at the strength and service limit
states. International Journal of Geotechnical
Engineering, 2(4), pp. 355-368.
17. Salgado, R. 2006. The engineering of
foundations. McGraw-Hill, New York, 896 p.
18. Seaward 2002. Seapile® and Seatimber®
composite marine products typical
performance characteristics.

[58] DFI JOURNAL Vol. 4 No. 1 August 2010

Peak and Post-Peak Shear Strength of Cement-Bentonite
Paul J. Axtell, P.E., Dan Brown and Associates, Overland Park, Kansas;
[email protected]
Timothy D. Stark, Ph.D., P.E., University of Illinois
John C. Dillon, P.E., U.S. Army Corps of Engineers, Kansas City District

ABSTRACT
Self-hardening cement-bentontie (c-b) slurry walls were constructed as shear walls to stabilize the
downstream slope of Tuttle Creek Dam near Manhattan, Kansas. The slope stabilization was required
to protect the existing pressure relief well system located at the downstream toe of the dam. The
wells require protection from slope deformation induced by liquefaction of the foundation sands
during or immediately after the design seismic event. The shear walls are transverse to the axis
of the dam, unreinforced, and relatively brittle members that may be exposed to relatively large
shear strains, and possible cracking, during or immediately after shaking. An extensive laboratory
investigation was conducted on recovered core samples to optimize the mix design and stabilization
scheme. Furthermore, as is the topic of this paper, a portion of the laboratory investigation was to
determine the large-strain, or post-peak, shear strength of the c-b material for use in limit-equilibrium
slope stability analyses and numerical deformation modeling to assess the magnitude of permanent
deformation caused by the design earthquake. These data may be beneficial to other projects that are
considering the use of unreinforced c-b slurry walls for seismic retrofit purposes.

INTRODUCTION at 20 km (12.5 miles) with a return period

Tuttle Creek Dam, located on the Big Blue River of about 3000 years. The peak horizontal
in the Kansas River Basin, is part of a system ground acceleration, PHGA, of the MCE is
that provides a comprehensive plan for flood 0.30g mean and 0.56g mean plus one standard
control and other functions in the Missouri deviation. The threshold liquefaction event is
River Basin. The dam was designed and a magnitude 5.7 with a return period of about
constructed by the US Army Corps of Engineers, 1700 years. The Kansas City District found
Kansas City District in the 1950’s. It is located that rehabilitation of the liquefiable foundation
about 10 km north of the city of Manhattan in sands is required to prevent an uncontrolled
eastern Kansas, as shown in Fig, 1. release of the reservoir during or after the
design ground motion.
The embankment is 2,300 m (7,550 ft) long
and about 43 m (140 ft) high. A typical
cross-section of the dam is shown in Fig. 2,
identifying the general locations of the various
embankment fill zones. The crest width is 15.2
m (50 ft) and the base width varies from about
430 to 490 m (1,400 t0 1600 ft). The top of the
dam is at elevation 353.3 m (1,159 ft) while the
original ground surface varies in elevation from
about 310 to 313 m (1,017 to 1,027 ft)across
the valley. Tuttle Creek Dam is a rolled earthfill
dam; details of the fill zones and construction [FIG. 1] General Location of Tuttle Creek Dam
of the dam can be found in Lane and Fehrman
(1960).
As part of the required seismic rehabilitation,
The main influential seismic source zones transverse shear walls were constructed
are the Nemaha Ridge uplift zone and the through the embankment and underlying
Humboldt Fault zone. The maximum credible foundation soils in the downstream slope
earthquake (MCE) is a magnitude 6.6 event and toe of the dam. Some preliminary design
DFI JOURNAL Vol. 4 No. 1 August 2010 [59]
 [FIG. 2] Typical cross-section of Tuttle Creek Dam

drawings depicting the plan and profile of relatively brittle members that will be exposed
these shear walls are shown in Figs. 3 and 4, to relatively large shear strains during or
respectively. The walls are 1.22 m (4 ft) wide, immediately after the design seismic event.
13.72 m (45 ft) long, and generally about Such loading may crack the shear walls, after
21 m (69 ft) deep. A 3.05 m (10 ft) clear- which the frictional resistance of the cracked
space generally exists between them. Design section will govern the ability of the shear walls
of the clear-spacing considered requirements to resist gravitational forces induced by the
for unimpeded seepage between the walls in slope. Large deformations at the downstream
both the pervious drain and foundation sands, toe are not acceptable because of the presence
while also considering
soil displacement between
the walls using limit
equilibrium methods. These
transverse shear walls are
self-hardening cement-
bentontie (c-b) slurry walls,
primarily excavated with
a clam-shell. Note that
slightly smaller walls were
also excavated with a long-
reach excavator early in
the project for comparison
purposes between the two
construction methods.
The c-b slurry walls are
oriented perpendicular
to the crest of the
dam, unreinforced, and [FIG. 3] Plan View of Transverse Shear Walls (units in feet, 1 m = 3.28 ft)

[FIG. 4] Profile View of Transverse Shear Walls (units in feet, 1 m = 3.28 ft)

[60] DFI JOURNAL Vol. 4 No. 1 August 2010

of a fragile pressure relief well system. This seepage control system for the dam, as are the
relief well system provides vital underseepage pressure relief wells at the downstream toe.
pressure relief during operation of the reservoir The sand deposits vary in thickness from about
and damage could lead to foundation erosion 7.6 to 18.3 m (25 to 60 ft) and can be separated
and piping. into two distinct zones. The upper zone
A laboratory investigation was conducted on consists of a 4.6 to 6.1 m (15 to 20 ft) thick
recovered samples obtained from production loose fine to medium sand (SM, SP, and SW)
walls (initially they were test walls) to and the lower zone consists of a 7.6 to 9.1 m
determine the large-strain, or post-peak, shear (25 to 30 ft) thick dense coarse to gravelly sand
strength of the hardened cement-bentonite that increases in grain-size with depth (SP,
material. Testing included isotropically SW, GP, GW). Due to the alluvial nature of the
consolidated, undrained shear (R-bar) triaxial foundation deposit, multiple lenses of cohesive
compression tests and drained direct shear soil exist within the coarse-grained layers.
tests. Testing was performed on samples The upper sand zone was determined to be
that were recovered from walls constructed potentially liquefiable during the design ground
with cement-to-water (c/w) ratios of 0.3, motion. The bedrock consists of alternating
0.4 and 0.5. Both mixes include a 5 percent layers of shale and limestone (Permian age);
bentonite component. The results of the however, the transverse shear walls do not
laboratory investigation were required for use penetrate bedrock. The c-b walls were keyed
in limit-equilibrium slope stability analyses into the dense, coarse to gravelly sand and
used to design the shear walls and numerical occasionally were founded on bedrock,
deformation modeling to assess the earthquake particularly near the left abutment.
induced permanent deformation of the dam
and foundation materials. For the majority of SAMPLING AND TESTING
the production work, unconfined compression Wet-grab samples were cast in 7.6 cm by
tests were used to validate the design. The 15.2 cm (3 in by 6 in) cylinders and stored
test results presented in this paper are for underwater until testing was performed. The
samples recovered from a test section that also grab sample was obtained from a shear wall
serves as production shear walls. The testing shortly after construction and before the slurry
presented herein was required for design. hardened. After hardening, the shear walls
were cored and the resulting core samples
SOIL PROFILE were also stored underwater until testing was
A working platform was constructed on the performed. Wall coring was conducted with
downstream slope of the dam to facilitate the Geobore system (double-barrel wireline)
construction of the shear walls. The platform producing 10 cm (4 in) diameter samples.
was constructed by: 1) removing the existing Coring was conducted about three weeks
random fill material to expose the underlying after construction of the walls. Testing was
pervious drain fill; 2) importing and placing conducted at least 70 days after construction.
sand (SP); and 3) placing approximately 60 cm Based on an independent laboratory
(2 ft) of road sub-base for a working surface. investigation of the proposed mixes, and
The only portion of the embankment that the verified by full-scale field measurements,
walls are in contact with is the pervious drain relatively minor strength increases can be
material downstream of the core, which lies expected beyond 90 days for these materials.
above the natural cohesive blanket (ML and The majority of the tests occurred within the
CL). The pervious drain material is composed 90 day time frame. A significant unconfined
of dense dredged SP soil, and is approximately compressive strength discrepancy between
4.6 m (15 ft) thick. the wet-grab and core sample strength was
The soils in the alluvial foundation of the dam observed at higher c/w ratios as described by
consist of 2.4 to 8.2 m (8 to 27 ft) of silt and Axtell, et al. (2009).
clay (natural cohesive blanket) underlain by
sand, silty sand, and gravely sand to a depth of INTERPRETATION OF TEST RESULTS
12.2 to 24.4 m (40 to 80 ft). The silt and clay Testing included isotropically consolidated,
form a natural cohesive soil blanket over the undrained shear (R-bar) triaxial compression
more permeable sands. This natural cohesive tests and drained direct shear tests. The tests
blanket is an important component of the were performed by Kleinfelder, in Topeka,
DFI JOURNAL Vol. 4 No. 1 August 2010 [61]
 Kansas. The post-peak or ultimate strength 207, and 552 kPa (10, 30 and 80 psi), which are
measured on core samples via R-bar and thought to adequately encompass the expected
direct shear tests was taken into consideration in-situ stress range. The strain rate for all
during the design of these walls for the seismic of the tests was 0.08 mm/min (.003 in/min).
retrofit because some cracking of the walls is This rate was chosen to facilitate drainage of
expected during the design ground motions. excess pore pressures generated during shear
Thus, the peak strength of the hardened shear and was estimated based on consolidation
walls would not be operational, i.e., would have test results. Each specimen was tested to the
been exceeded due to cracking of the walls. maximum axial strain practical, which was
The results of the R-bar tests are provided usually less than 20 percent in the R-bar tests.
below, along with the results from the direct A common constraint was ripping or tearing
shear tests. of the specimen membrane during shear
due to the sharp pieces of concrete from the
R-BAR SHEAR RESULTS specimens. Post-peak values were obtained at
The R-bar tests were performed on 6.6-cm the maximum axial strain measured (excluding
(2.5 in) diameter samples with heights ranging data after a membrane tear occurred). The
from 11.5 to 14.5 cm (4.5 in to 5.7 in). R- results of the tests are summarized in Table
bar tests were conducted on recovered core 1. The values of dry unit weight, moisture
samples from walls constructed by both the content, and void ratio provided in Table 1 are
clam-shell and long-reach excavation methods. average values for the three specimens tested
Trimming of the samples was achieved by re- at each location (three data points are used
coring the selected specimens to the proper to define the failure envelope). The moisture
diameter. It is unknown if the trimming contents reported in Table 1 are from portions
process had any effect on the results. In of the specimen collected after the shearing
addition, two suites of tests were conducted on phase of the test. Generally, but not always,
wet-grab samples from clam-shell constructed the moisture content of the samples decreased
walls. Three of the samples were from walls by around 5 percent during the consolidation
constructed with a c/w ratio of 0.4 and one phase of the test. Back-pressure saturation
with a c/w ratio of 0.3. The remaining six was utilized; the mean B was 0.96, with a
samples were obtained from walls constructed standard deviation of 0.05.
with a c/w ratio of 0.5. Total stress failure Deviator stress versus axial strain relationships
envelopes for peak and post-peak strength for the 15 specimens of c/w=0.5 core samples
were determined, as was the effective stress (five suites of tests, each with three points
failure envelope for post-peak strength. All associated with the three different confining
failure envelopes were determined by testing stresses) are provided in Fig. 5. The results
separate samples at confining stresses of 69, of tests conducted on c/w=0.4 walls are not

[TABLE 1] R-bar results (both total and effective stress).

Sample Dry Unit Moisture Strain Peak Post-Peak Post-Peak

Core Excavation c/w Void
Depth Weight Content at Peak c Φ c Φ c' Φ'
Hole method ratio Ratio
(m) (kN/m3) (%) (%) (kPa) (°) (kPa) (°) (kPa) (°)

VC06 Long Reach 19.8 0.3 10.36 68 1.68 1.3 276 36 255 34 0 51

9.1 0.4 9.11 69 2.03 2.4 593 0 476 0 0 46

VC08 Clam-Shell
19.8 0.4 12.09 45 1.44 2.2 310 41 407 30 0 51
VC05 9.1 0.5 10.21 60 1.71 0.7 538 29 421 22 0 45
Long Reach
C-958 19.8 0.5 12.40 42 1.22 0.8 1151 18 731 25 0 49
VC14 9.1 0.5 10.99 48 1.52 1.6 352 37 262 39 0 46
15.2 0.5 10.21 57 1.69 0.6 931 8 434 17 0 46
VC17
Clam-Shell 19.8 0.5 10.68 54 1.58 0.9 690 27 310 34 0 46
Wet 15.2 0.4 9.11 66 2.04 0.9 986 7 207 30 0 46
Grab 13.7 0.5 10.83 56 1.55 0.9 1655 12 807 27 0 50

[62] DFI JOURNAL Vol. 4 No. 1 August 2010

shown, nor are the wet-grab results. The 500
production walls are to have a c/w=0.5 so 400
the c/w = 0.4 results are not relevant and
300
data from core testing was being used for
200
the design and acceptance criteria; hence,
those were the samples used for design. The

Δu (kPa)
100
linear line super-imposed on Fig. 5 indicates 0
the stress-strain relationship modeled in the
-100
permanent deformation analyses performed
-200
using FLAC (Itasca, 2000). The post-peak,
or large-strain, strength of core samples of -300
c/w=0.5 walls exceeds that required by the -400
design. The measured initial stiffness is also 0.00 0.05 Axial Strain 0.10 0.15

somewhat greater than modeled (average initial [FIG. 6] Pore pressure change versus axial strain from R-bar
Young’s modulus equals 538 MPa (78 psi) with tests on c/w = 0.5 core samples (both long-reach and clam-shell
excavators, at all confining stresses).
a standard deviation of 148 MPa (21.5 psi)) but
the majority of the stress-strain relationships
still indicate stronger material than modeled DIRECT SHEAR RESULTS
in the FLAC analyses. Thus, the permanent
The direct shear tests were performed on 6.35-cm
deformations estimated after wall cracking
(2.5 in) diameter samples with a height of 2.54
using FLAC are probably conservative and
cm (1 in). All direct shear tests were conducted
within allowable values.
on recovered core samples from shear walls
The change in pore pressure during shear constructed by the clam-shell excavation method.
versus axial strain is shown in Fig. 6 for the Trimming of the samples was achieved by re-
15 specimens of c/w=0.5 core samples. As coring the selected specimens to the proper
expected from the relatively high void ratios diameter. It is unknown if the trimming process
measured prior to shear, all samples tended to had any effect on the results. The tests were
initially generate high positive pore pressures. performed by Kleinfelder in Topeka, Kansas.
At higher axial strains, the excess pore Three of the samples were obtained from walls
pressures became negative for all 10 specimens constructed with a c/w ratio of 0.4. The remaining
tested at the lower confining stresses (69 six samples were from walls constructed with a
and 207 kPa or 10 and 30 psi), whereas the 5 c/w ratio of 0.5. Failure envelopes for peak and
specimens at the higher confining stress (552 post-peak strength were normal stresses of 96,
kPa or 80 psi) remained positive. 192, 384, and 574 kPa (14, 28, 56 and 83 psi).
Note that the actual strain values are reported The shear displacement rate for all of these
on the x-axis in Figs. 5 and 6 (∆l/l) whereas tests is 0.005 mm/min (0.0002 in/min). This
the corresponding values in Table 1 have been rate was chosen to facilitate drainage of excess
reported as a percentage. pore pressures generated during shear based
on consolidation test results. Each specimen
3.5 was tested to a 0.64 cm (0.25 in) horizontal
displacement. Post-peak strength values were
3.0
obtained at the maximum horizontal displacement
2.5 (0.64 cm) (0.25 in), whereas the peak values were
generally observed at a horizontal displacement
σ 1-σ 3 (MPa)

2.0
of less than 0.25 cm (0.1 in). The results of these
1.5 tests are summarized in Table 2. The values of
dry unit weight, moisture content, and void ratio
1.0 provided in Table 2 are average values for the
four specimens tested at each location (four data
0.5
Stress-Strain Curve used in Deformation Model
points defining the failure envelope).
0.0 Approximately three-quarters of the 36
0.00 0.05 Axial Strain 0.10 0.15
specimens (9 tests, each with four normal
[FIG. 5] Stress-strain relationships from R-bar tests on c/w = stresses) show a slight contraction initially,
0.5 core samples (both long-reach and clam-shell excavators,
at all confining stresses) and relationship used in FLAC after which the specimens began to dilate.
deformation analyses. Initial contraction on the order of about 0.5
DFI JOURNAL Vol. 4 No. 1 August 2010 [63]
 [TABLE 2] Direct shear results (Clam-Shell Constructed Walls and Core Samples).

Sample Dry Unit Moisture Peak Post-Peak

c/w Void
Core Hole Depth Weight Content c' Φ' c' Φ'
ratio Ratio
(m) (kN/m3) (%)
(kPa) (°) (kPa) (°)
9.1 0.4 8.01 76 2.5 172 44 21 41

VC08 15.2 0.4 9.26 63 1.8 400 56 110 60

19.2 0.4 9.73 60 1.8 296 37 41 39

3.7 0.5 8.16 79 2.4 303 33 48 36

4.6 0.5 8.64 79 2.4 386 20 193 23

VC17 6.1 0.5 8.64 76 2.2 276 32 14 42

9.1 0.5 8.64 72 2.0 400 25 97 27

12.2 0.5 8.95 71 2.1 241 41 28 37

VC14 15.5 0.5 9.89 59 1.7 352 40 62 40

percent of the original sample height was shear strength, as shown in Fig. 9. This seems
common whereas dilation on the order of 0.5 apparent for the effective cohesion and friction
to 5 percent was observed with increasing angle at both peak and post-peak values.
horizontal displacement. Opposite behavior Effective Cohesion (kPa)
was observed for the remaining one-quarter of 0 50 100 150 200 250 300 350 400 450
the specimens. Unfortunately, no discernable 0

trend was apparent between volumetric change

and c/w ratio, depth, void ratio, moisture 5
content, or dry unit weight.
The two upper samples from core hole VC08 10
Depth (m)

indicate effective friction angles that are

noticeably higher than the other samples, both 15
at peak and post-peak. The exact reason for
the phenomenon is not known, but expected 20 Post-Peak Peak
Post-Peak Avg Peak Avg
to be a result of the presence larger or more +1 Stan Dev +1 Stan Dev
angular natural soil particles in the sample. -1 Stan Dev -1 Stan Dev
25
The R-bar results from VC08 do not appear to
validate or dispel this conclusion.
[FIG. 7] Direct shear effective cohesion versus depth in shear
The effective cohesion and effective friction wall for c/w=0.5 samples.
angle versus depth as determined by direct shear
testing are provided in Figs. 7 and 8, respectively. Effective Friction Angle (degrees)
19 21 23 25 27 29 31 33 35 37 39 41 43
These figures only present the results of tests
0
conducted on samples with a c/w ratio equal
to 0.5, because the production work is utilized
5
for this mix. Also included in Figs. 7 and 8 are
the mean, mean minus one standard deviation,
10
and mean plus one standard deviation for each
Depth (m)

data set. Based on these figures, there does

not appear to be a discernable trend between 15

shear strength and depth in the shear wall. The

presence of a post-peak cohesion value indicates 20 Post-Peak
Post-Peak Avg
Peak
Peak Avg
that the shear displacement imposed in the +1 Stan Dev +1 Stan Dev
-1 Stan Dev -1 Stan Dev
direct shear tests was not sufficient to reach a 25
residual strength condition.
Unlike soil, there also does not appear to be [FIG. 8] Direct shear effective friction angle versus depth in
shear wall for c/w=0.5 samples.
a distinct relationship between void ratio and

[64] DFI JOURNAL Vol. 4 No. 1 August 2010

 Effective Cohesion (kPa) unreinforced walls during or following the
0 50 100 150 200 250 300 350 400 450
2.45
design seismic event due to cracking of the
2.35
walls during shaking. Laboratory R-bar and
direct shear testing of recovered core and wet-
Void Ratio at Start of Shear

2.25

2.15
grab samples was conducted to evaluate both
2.05
peak and post-peak strength for use in the
wall design and estimate of post-earthquake
1.95
permanent deformations. The results of the
1.85
laboratory testing program are presented
1.75
and indicate that a c/w=0.5 mix that includes
1.65 Post-Peak Friction Angle Peak Friction Angle
Peak Cohesion Post-Peak Cohesion a 5 percent bentonite component will meet
1.55
19 21 23 25 27 29 31 33 35 37 39 41 43 or exceed the peak and post-peak strength
Effective Friction Angle (degrees) requirements dictated by the design. These
[FIG. 9] Direct shear strength parameters versus void ratio data may be beneficial to other projects that are
for c/w=0.5 samples (all cored samples, constructed with clam considering the use of unreinforced c-b slurry
shell excavator).
walls for seismic retrofit purposes.

ACKNOWLEDGMENTS
REPRESENTATION OF TEST RESULTS
IN ANALYSES The contents of this paper are the authors’
and do not necessarily reflect those of the
The data presented herein was used to estimate
represented entities. The authors acknowledge
a strength and modulus profile for the depth of
the support provided by the U.S. Army Corps of
a transverse shear wall to model the variation
Engineers – Kansas City District and Kleinfelder.
in strength and stiffness with depth in the
The authors are particularly appreciative of
FLAC analyses. Results from both the R-bar
the efforts by Joe Topi, Francke Walberg, Bill
and direct shear tests were considered in
Empson, and David Mathews. Finally, the
determining the strength and stiffness design
expertise of the contractor, Treviicos South, was
values. However, results from the R-bar tests
essential.
were more heavily relied upon as a result of the
forced failure plane orientation in the direct REFERENCES
shear tests, as well as questions resulting from
the somewhat limited magnitude of the direct 1. Axtell, P.J., Stark, T.D., and Dillon, J.C.
shear test displacements. Based on this data, (2009). “Strength Difference between
the following average stress-strain behavior was Clam-Shell and Long-Reach and Excavator
used in the deformation analyses: Constructed Cement-Bentonite Self-
Hardening Slurry Walls.” Contemporary
1. Peak strength (total stress): c = 655 kPa Topics in Ground Modification, Problem
(95 psi) and φ = 24°. Soils, and Geo-Support, ASCE Geotechnical
2. Post-peak strength (effective stress): φ' = 46°. Special Publication No. 187; Iskander, Laefer,
3. Young’s modulus (tangent): E = 496 MPa and Hussein, editors.
(72 ksi). 2. Itasca Consulting Group, Inc. (2000). FLAC
4. Peak strength attained at axial strain: – Fast Lagrangian Analysis of Continua,
ε = 0.8% version 4.0, Itasca Consulting Group, Inc.,
Minneapolis, MN.
5. Post-peak strength begins at axial strain:
ε = 1.6%. 3. Lane, K.S., and Fehrman, R.G. (1960).
“Tuttle Creek Dam of Rolled Shale and
This characterization may be beneficial to
Dredged Sand.” Journal Soil Mechanics and
other projects that are trying model the seismic
Foundation Division, ASCE, 86(SM6). 11-34.
performance of shear walls.

CONCLUSIONS
Cement-bentonite (c-b) self-hardening slurry
walls were constructed as a seismic retrofit of
the downstream slope of Tuttle Creek Dam.
Post-peak, or large-strain, shear strength
will likely dictate the performance of these
DFI JOURNAL Vol. 4 No. 1 August 2010 [65]
 Underwriters: Gold

Underwriters: Silver

Providing specialty services with solutions for a changing world

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

[66] DFI JOURNAL Vol. 4 No. 1 August 2010

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 Submittal
Papers 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.

DFI JOURNAL Vol. 4 No. 1 August 2010 [67]

 2010 DFI Journal Subscriptions
Electronic Subscriptions for the two-issue volume (May & November 2010) 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.

Electronic & Electronic &

Volume 4 (Aug. & Nov. 2010) Print Only Print Only Print Print
Subscription Rates Electronic (USA) (Outside USA) (USA) (Outside USA)
DFI MEMBER Free with $125 $175 N/A N/A
membership
NON-MEMBER $150 $300 $350 $405 $455

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

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

Past Volumes
Single Issue Rates Print Only Print Only
Electronic (USA) (Outside USA)
Volume 1 (Nov 2007) FREE $45 $70
MEMBER & NON-MEMBER online
Volume 2 (Nov 2008) $45 $45 $70
MEMBER & NON-MEMBER

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: _____________________________________________

Orderonlineatwww.dfi.org/dfijournal.asp

[68] DFI JOURNAL Vol. 4 No. 1 August 2010

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.

DFI JOURNAL Vol. 4 No. 1 August 2010 [69]

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: APE/J&M

ASC Geosciences Inc.
• Communication of information concerning
the state-of-the-art and state of the practice Ben C. Gerwick Inc.
of deep foundation technologies Berkel & Company Contractors Inc.
Brasfond Fundações Especiais S/A
• Offering networking opportunities for Brayman Construction Corporation
our members
Cajun Deep Foundations LLC
• Offering opportunities for members to Case Foundation Company
improve the industry through publications Ciport S.A.
produced by volunteer committees Dean Construction Co. Ltd.
• Offering educational conferences, seminars Dewitt Construction Inc.
and workshops in the industry Dosdourian Enterprises Inc.
The core strength of DFI is the broad spectrum Foundation Constructors Inc.
of its membership. All disciplines participate Geokon Inc.
on an equal footing, be they contractors, Hayward Baker Inc.
engineers, owners, academicians, equipment HJ Foundation Company
manufacturers and distributors or materials Intercoastal Foundations and Shoring
manufacturers and suppliers. All types of Kiewit Construction Group Inc.
foundation systems are represented, whether
Kleinfelder
installed by driving, drilling or other means.
L.G. Barcus & Sons Inc.
This diversity and openness without bias
provides a forum for the free exchange of Langan Engineering and Environmental Services
knowledge and a platform for the development Mactec Engineering & Consulting Inc.
of new technology and opportunity. McKinney Drilling Company
Menard
DFI is: Moretrench
• An international network of heavy Mueser Rutledge Consulting Engineers
construction professionals dedicated to Nicholson Construction Company
quality and economy in foundation design North American Construction Group
and construction North American Steel Sheet Piling Association
• A forum open to all construction O.C.I. Division / Global Drilling Suppliers Inc.
professionals across disciplines Pileco Inc.
and borders. Pnd Engineers Inc.
• A technological association devoted to Richard Goettle Inc.
gathering, storing and disseminating Sas Stressteel Inc.
practical information Schnabel Foundation Company
Tei Rock Drills Inc.
• A resource for identifying and locating the
ThatCher Engineering Corporation
specialists and sources of expertise.
Urban Foundation/Engineering LLC
• An initiator and participant in research William F. Loftus Associates Foundation Engineers.

DFI Sustaining Members Deep Foundations Institute Sustaining Members Are Corporate
Members Of DFI Who Have Voluntarily Granted Funding To The
Aecom USA Inc.
Institute For Expanded Support Of The Industry. The fund is
AGL Manufacturing Ltd. managed by the DFI Educational Trust.
American Equipment & Fabricating Corp.
Anderson Drilling
DFI JOURNAL
The Journal of the Deep Foundations Institute

International Standard Serial Number (ISSN): 1937-5247

Deep Foundations Institute

326 Lafayette Avenue
Hawthorne, New Jersey 07506 USA
Tel: 973-423-4030
Fax: 973-423-4031
www.dfi.org