See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/281761455

Installation Effects of Controlled Modulus Column Ground Improvement Piles

on Surrounding Soil

Article in Journal of Geotechnical and Geoenvironmental Engineering · July 2015

DOI: 10.1061/(ASCE)GT.1943-5606.0001384

CITATIONS READS
32 5,541

5 authors, including:

Muhannad T. Suleiman Lusu ni

Lehigh University Lehigh University
137 PUBLICATIONS 2,801 CITATIONS 10 PUBLICATIONS 257 CITATIONS

SEE PROFILE SEE PROFILE

Caleb Davis Hai (Thomas) Lin

Lehigh University Louisiana State University
1 PUBLICATION 32 CITATIONS 37 PUBLICATIONS 1,011 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Hai (Thomas) Lin on 18 June 2016.

The user has requested enhancement of the downloaded file.

 Installation Effects of Controlled Modulus Column Ground
Improvement Piles on Surrounding Soil
Muhannad T. Suleiman, A.M.ASCE 1; Lusu Ni, S.M.ASCE 2; Caleb Davis, A.M.ASCE 3;
Hai Lin, S.M.ASCE 4; and Suguang Xiao, S.M.ASCE 5
Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

Abstract: The installation of foundations and ground improvement systems alters soil stresses affecting their soil-structure interaction and
behavior under vertical loading. The installation effects have been investigated for several foundation types, especially for driven piles;
however, experimental full-scale investigations of drilled displacement piles are very limited. This paper focuses on investigating the
short-term installation effects of controlled modulus columns (CMCs) using an instrumented full-scale field test unit. The soil was instru-
mented using push-in pressure sensors (PS) and shape acceleration arrays (SAAs) to monitor the evolution of soil horizontal stresses, pore
water pressures, and lateral displacements during installation and vertical load test. These sensors were installed at approximately 1D, 2D, 3D,
and 4D from the outside surface of the CMC shaft, where D is the diameter of the CMC. The measurements presented in this paper clearly
show that the soil experienced an increase of horizontal stresses and lateral movement throughout the CMC installation. During mandrel
advancement, the horizontal stresses gradually increase until the mandrel reaches the location (depth) of the pressure sensors, which was
followed by a horizontal stress decrease. The measurements presented in this paper indicate that the zone affected by the CMC installation
extend to 2 to 3D from the outer surface of the CMC shaft. DOI: 10.1061/(ASCE)GT.1943-5606.0001384. © 2015 American Society of Civil
Engineers.
Author keywords: Ground improvement; Controlled modulus columns; Installation; Vertical load.

Introduction Randolph et al. 1979). Ng et al. (2013b) reported that pile setup
may result in increasing vertical pile capacity by more than 100%
A variety of construction methods have been used to install piles within 30 days. Granular columns (or aggregate piers) serve as an-
and ground improvement foundation systems. Depending on the other example where different construction methods (e.g., vibro-
construction method and foundation type, piles and ground im- composer, vibro-compaction, vibro-replacement, impact, and
provement systems can be classified as nondisplacement, partial- ramming compaction) result in different responses and properties
displacement, and full-displacement foundations. Changes of soil (Aboshi et al. 1979; Mitchell 1981; Barksdale and Bachus 1983;
properties (i.e., strength and stiffness) and stresses which occur dur- Bergado et al. 1994; Moseley and Kirsch 2004; White and
ing the installation of foundation systems affect their soil-structure Suleiman 2004; Suleiman et al. 2014). For example, Chen et al.
interaction and response when subjected to axial loading. For ex- (2009), who analytically evaluated the effects of ramming during
ample, many researchers and practitioners have recognized the aggregate pier installation, reported a 44% reduction of the pier top
increase of driven piles capacity with time (pile setup) for both settlement during vertical loading due to ramming.
coarse-grained (e.g., Chow et al. 1998) and fine-grained materials For drilled displacement piles, a combination of vertical force
(e.g., Ghandeharioon et al. 2010; Karlsrud and Haugen 1985; and torque are used during construction. The equipment used to
Karlsrud and Mahan 2010; Karlsrud 2012; Ng et al. 2013a, b; install drilled displacement piles consists of a helical, partial-flight
auger, soil displacement body, and a tip attached to the bottom of
1
Associate Professor, Dept. of Civil and Environmental Engineering, the drilling tool, if needed (Brown 2005; Prezzi and Basu 2005).
Lehigh Univ., 326 STEPs Building, 1 W Packer Ave., Bethlehem, PA The advantages of drilled displacement piles include easy construc-
18015 (corresponding author). E-mail: [email protected] tion, no or minimal spoil, and high capacity due to lateral displace-
2
Former Graduate Research Assistant, Dept. of Civil and Environmental ment of the surrounding soil (Brown 2005; Prezzi and Basu 2005).
Engineering, Lehigh Univ., 320 STEPs Building, 1 W Packer Ave., Controlled modulus columns (CMCs) are installed using a
Bethlehem, PA 18015. E-mail: [email protected] continuous flight auger fitted with a displacement component
3
Former Undergraduate Student, Dept. of Civil and Environmental
Engineering, Lehigh Univ., 1 W Packer Ave., Bethlehem, PA 18015.
with the same diameter. The flight of the lower part of the auger
E-mail: [email protected] has opposite direction to the flight of the top part of the auger.
4
Graduate Research Assistant, Dept. of Civil and Environmental During penetration (mandrel advancement), the hollow stem auger,
Engineering, Lehigh Univ., 320 STEPs Building, 1 W Packer Ave., which is fitted with a plug added at the bottom to prevent the soil
Bethlehem, PA 18015. E-mail: [email protected] from moving inside the hollow stem auger, cuts the soil and the
5
Graduate Research Assistant, Dept. of Civil and Environmental opposite flight and the displacement component prevents the soil
Engineering, Lehigh Univ., 320 STEPs Building, 1 W Packer Ave., from moving upward to the ground surface and push the soil lat-
Bethlehem, PA 18015. E-mail: [email protected] erally into the surrounding ground (Masse et al. 2004). During
Note. This manuscript was submitted on July 13, 2014; approved on June
10, 2015; published online on July 10, 2015. Discussion period open until
withdrawal (mandrel retrieval), a lean sand mix mortar is pumped
December 10, 2015; separate discussions must be submitted for individual under moderate pressure (35–345 kPa), which releases the plug
papers. This paper is part of the Journal of Geotechnical and Geoenviron- added at the bottom of the hollow stem auger and fills the created
mental Engineering, © ASCE, ISSN 1090-0241/04015059(10)/$25.00. cavity, resulting in a concrete column. The mandrel is retrieved

© ASCE 04015059-1 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059

 while rotating in the same direction used during advancement. As (0.35% axial strain=min). The measured deviator stress-axial strain
will be discussed next, pushing the soil laterally during the con- curves are presented in Fig. 2(a). The initial modulus of the soil (Ei )
struction of CMCs is expected to affect surrounding soils; however, as a function of confining pressure (σ3 ) was evaluated using the
these effects for CMC installation have not been investigated. power function suggested by Janbu (1963) [Eq. (1)] and the calcu-
Construction methods, which are used for installing displace- lated values of k and n were 297.5 and 0.78, respectively. In Eq. (1),
ment deep foundations and ground improvement systems change Pa is the atmospheric pressure, k is the modulus number and n is
soil stresses affecting their soil-structure interaction and response the modulus exponent
when subjected to axial loading (Handy 2001; Basu et al. 2011;
Lundberg et al. 2013). Different approaches have been used to Ei ¼ kPa ðσ3 =Pa Þn ð1Þ
evaluate these effects including cavity expansion analysis, numeri-
Fig. 2(b) shows the stress invariant (p 0 − q) diagram for the CU
cal modeling methods, cone penetration tests before and after in-
triaixial tests. Based on these results, the calculated value of the
stallation, and shear wave velocity measurements. None of these
slope of the critical state line (M) equals to 1.13. The critical state
approaches, however, directly and simultaneously measured the friction angle of the soil equals to 28°. Using the average compres-
Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

variation of soil stresses and displacement (movement) during pile sion and recompression indexes (Cc and Cr ), the values of the criti-
installation. Therefore, Lundberg et al. (2013) reported that there is cal state compression index (λ) and recompression index (κ) were
a lack of direct simultaneous measurements of soil stresses and calculated as 0.14 and 0.02, respectively. Fig. 1 also illustrate the
movement surrounding displacement piles in general. For CMCs, depth of the pressure sensors and the locations of the tips (bottom)
the effects of installation on surrounding soil have not been evalu- of the SAAs installed around the CMC test unit.
ated before, which is a knowledge gap that is partially addressed in
this paper.
This paper focuses on experimentally investigating the effects of Instrumentation
CMC installation on surrounding soil. A full-scale CMC test unit Push-in pressure sensors are vibrating wire earth pressure sensors
was installed in very soft low plastic sandy silt, and the surrounding fitted with a piezometer allowing for measuring total horizontal
soil was instrumented with push-in pressure sensors (PS) and shape stresses and pore water pressure in the soil. In addition to the cal-
acceleration arrays (SAAs) located at ∼1D, 2D, 3D, and 4D (where ibration provided by the manufacturer, the research team performed
D stands for diameter) from the outside surface of the CMC shaft. a calibration check on the total pressure sensors and the difference
In addition to monitoring the soil horizontal stresses, pore water between the applied pressure and the measured pressure was less
pressure, and soil lateral displacements during installation, the ver- than 0.25%. More information about the push-in pressure sensors
tical load-displacement response and the load transfer along the and their working principles are discussed by Geokon (2013). Four
CMC test unit were measured during the vertical load test and will push-in pressure sensors and four shape acceleration arrays
be reported in this paper. were installed in the soil before constructing the CMC test unit
[Figs. 3(a and b)]. The pressure sensors were installed at a depth
of 2.1 m below the ground surface (measurements made at 1.95 m)
Site Conditions and Material Properties at radial distances of 0.305, 0.610, 0.914, 1.22 m from the targeted
shaft surface of the CMC test unit (i.e., ~1D, 2D, 3D, and 4D where
The test site used in this investigation is an elementary school con- D is the pile diameter of 315 mm). To install the pressure sensors,
struction site. The subsurface investigation at the test site illustrates initially a borehole was drilled using a solid stem auger at the
that the soil profile consists of four soil layers (Fig. 1). The top location of PS4 to a depth of 1.05 m. At this depth, a Shelby tube
0.25 m of the profile consists of an aggregate layer followed by was pushed to a depth of ∼1.68 m. The vibrating wire push-in
approximately 0.95 m of loose silty sand (classified as SM using pressure sensor was then connected to a drill rod and lowered into
the Unified Soil Classification System), which is underlain by very the open borehole. The pressure sensor was then pushed into the
soft gray sandy silt (classified as ML) extending 4.8 m to a depth of soil at the bottom of the borehole until the tip of the pressure sensor
approximately 6 m. This layer is underlain by a very soft clay layer was at a depth of ∼2.1 m (depth of measurements at 1.95 m). This
with traces of silt and rock fragments (classified as CL) that extends installation procedure was repeated for pressure sensors 3, 2, and 1,
to 7.6 m below the ground surface. The standard penetration test respectively.
(SPT) N-value ranged from 0 to 5 between the ground surface Advanced soil deformation sensors were used to measure soil
and ∼6.5 m. At 7.62 m, the SPT N-value was greater than 50 deformation (or movement). The SAA sensor consists of several
(i.e., refusal). At the time of subsurface investigation, the water micromachined electromechanical sensors installed in a flexible
table at the test site was located at 1.6 m below the ground surface. and watertight casing, which enables gravity-based shape calcula-
The basic soil properties of these layers are summarized in Fig. 1. tion along sensorized area. The shape of the array is expressed in a
Laboratory consolidation tests performed on Shelby tube series of positions in x, y, and z, and the orientations are expressed
(76 mm outside diameter) samples obtained between depths of as a set of unit orientation vectors. The SAA is flexible and bends
3.1 and 3.7 m indicated that the time of 90% consolidation ranged freely in 2 degrees of freedom. To allow for measuring the lateral
from 0.423 to 4.884 min and the average coefficient of consolida- displacement along the SAAs, reference points at the top of each
tion (Cv ) was 0.017 m2 =h. The compression and recompression SAA (above the ground surface) were utilized and the movement of
indexes (Cc and Cr ) ranged from 0.148 to 0.475, and 0.015 to these reference points was measured using displacement sensors.
0.083, respectively, with an average Cc ¼ 0.33 and average For the SAA lengths used in the experimental setup (range from
Cr ¼ 0.04. The consolidation tests were performed on samples 1.5 to 4.7 m), the accuracy of the SAA measurements is smaller
with a diameter of 64 mm and a height of 25 mm. In addition, than 1 mm (Measurand 2013). SAAs were also installed at radial
Shelby tube soil samples were collected at depths between 1.05 distances of ∼1D, 2D, 3D, and 4D from the outer surface of the
and 1.68 m to conduct a series of consolidated undrained (CU) shaft of the CMC test unit [Figs. 3(a and b)]. The SAAs, which
triaxial tests with sample diameter of 71 mm and length of have different total and segment lengths, were prepared for instal-
142 mm. The CU tests were performed at three confining pressures lation in accordance with the procedure suggested by the manufac-
(15, 25, and 50 kPa) with a loading rate of 0.5 mm=min turer (Measurand 2013). Each SAA, fully assembled in capped

© ASCE 04015059-2 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059

Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

Fig. 1. Average soil profile at the test site showing soil properties and the locations of instrumentation

schedule 40 PVC conduit, was lowered into an open borehole. strain gauges were attached to a No. 6 reinforcement rebar that was
Bentonite grout was then poured between the PVC casing and sur- installed at the center of the CMC test unit directly following the
rounding soil to provide sound contact with the surrounding soil. mandrel retrieval. The strain gauges were located at 0.33, 1.45, 4.0,
SAA3 and SAA4 were custom-designed with smaller diameter and and 6.72 m below the soil surface.
shorter segment lengths to fit the laboratory test applications of the
Soil-Structure Interaction Facility at Lehigh University. Hence, the
installation procedure suggested by the manufacturer for field ap- Construction
plication was modified and the PVC conduit was filled with sand to The CMC test unit was constructed one day after installing the
try to fill the gap between the SAA and the inside wall of the PVC pressure sensors and shape acceleration arrays. This procedure
pipe. Not having a perfect fit within the PVC pipe may have af- was followed to allow for the stabilization of stress and pore water
fected the measured lateral displacements for these two SAAs. pressure readings and thereby providing more accurate measure-
After installation, it was found that SAA4 did not function properly. ments as recommended by Suleiman et al. (2010). The CMC test
The CMC test unit has an embedded length of 7.77 m and a unit, 315 mm in diameter, was installed using an Enteco 6040
diameter of 315 mm. To measure the load transfer along the shaft continuous flight auger fitted with a displacement auger (Fig. 4).
of the CMC test unit, vibrating wire strain gauges were used. The Initially, the mandrel (auger and displacement assembly with a

© ASCE 04015059-3 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059

 250
Confining pressure, 50 The reaction piles [Fig. 3(a)] were installed in the following order:
225 σ3(kPa) R1, R3, R2, and R4. Each reaction pile, 395 mm in diameter, had a
200 length of 7.77 m and a 48 mm-diameter rebar installed at the center
of the pile. A beam was mounted to the rebars of the four reaction
Deviator stress (kPa)

175
piles to form the reaction frame. The construction equipment was
150 25 very close to the reaction piles during construction and moved dur-
125
ing their construction, which makes the use of pressure and dis-
placement measurements during the construction of reaction piles
100 15 to evaluate effect of installation unreliable.
75
σ3
Ei = kPa ( )n
50 Pa Loading Sequence
k = 297.5
25
n = 0.78 The load was applied using a hydraulic jack and measured with a
Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

0 1,330 kN capacity load cell. Four displacement transducers were

0 1 2 3 4 5 6 7 8 9 10 11 12 13 connected to the top surface of the test unit to measure the vertical
(a) Axial strain (%) displacement during the load test. The vertical load test was
performed in general accordance with ASTM D1143 (ASTM
250
q = 1.13 p 2009). A maximum applied load of 1.5 times the design load
f f
225 was initially targeted (following the European standard practice).
M=1.13
200 However, to obtain the ultimate vertical load capacity of the CMC
φcs = arcsin(3M/(6+M))=28o
test unit, the load was increased beyond the maximum targeted
175 )
SL load. A load increment of 45 kN was used during the load test.
(C
150
ne Unloading was performed at 445, 667 kN, and at the end of the
3

Li
q=σ σ

125 te test. The data of all sensors were recorded during the static load
ta
1

a lS test. The data of the soil pressure sensors were collected only for
100 itic 14 minutes after the vertical load was reduced to 0 kN at the end
Cr
75 of the test.
50

25
Results
0
0 25 50 75 100 125 150 175 200 225 250
p =(σ +2σ )/2 Load-Displacement Response
(b) 1 3
Fig. 6 shows the vertical load-displacement response of the CMC
Fig. 2. Sandy silt soil properties: (a) stress-strain of soil samples from test unit. This figure shows that the maximum load applied during
CU triaxial tests; (b) p 0 − qf critical state line diagram the test was ∼980 kN. At the design load (430 kN used by the foun-
dation designers of the school project), the displacement of the top
of the CMC test unit was 5 mm. At 150% of the design load, the
removable plug at the bottom) advanced to a depth of approxi- vertical displacement was approximately 8 mm. The capacity of
mately 2.0 m and intentionally stopped (pausing or holding stage) the CMC test unit was estimated using two different methods
to monitor the dissipation of pore water pressure when the mandrel [Davisson (1975) and 10% relative settlement (Franke 1989)].
passed the depth (location) of the pressure sensors [Fig. 5(a)]. The estimated nominal capacity ranged from ∼890 to 916 kN.
Advancement of the mandrel was resumed until it reached the tar- The commonly used Davisson Criteria provided an estimated
geted tip depth of 7.62 m. When the mandrel reached the targeted nominal capacity of ∼890 kN, which is more than twice the design
depth, a Schwing SP 2000 (Schwing America, Minnesota) rig load.
pumped cement grout through the mandrel into the space created
as the mandrel retrieved to the ground surface. The real-time con-
tinuous monitoring of the installation process was recorded using Load Transfer
the computerized quality-control system installed on the rig. Fig. 5 Using the strain gauge measurements and an elastic modulus of
summarizes the vertical pressure, torque, and speed of drilling and 26 MPa for the CMC grout material (measured using concrete sam-
extraction during installation. The small torque and fast penetration ples collected during construction), the load transfer along the
of the mandrel indicate a very soft soil between the depth of ap- CMC test unit was calculated for all loading stages, and a selection
proximately 2 and 7.5 m. In addition, the quality-control system of which is summarized in Fig. 7. The selected loading stages re-
installed on the construction equipment showed that the profile present the initial (linear or elastic) stage (222 kN), the design load
of the installed column was fully intact. (445 kN), 150% of the design load (667 kN), and ∼200% of the
Once the mandrel ascended to the ground surface, a steel casing design load (890 kN). The rate of load transfer indicates that the
610 mm in length was inserted at the top of the CMC unit to form a unit friction in the ML soil layer (i.e., the slope of the curve be-
housing for the grout, leaving the top of the casing approximately tween depths of 1.45 and 4.0 m) ranged from 0 to 25.2 kN=m
50 mm above the ground surface. Cement grout was poured and at different loading stages. It should be noted that the strain gauges
leveled with the top of the steel casing. Then, a 7.54 m-long No.6 located near the tip of the test unit may have been damaged during
rebar, with eight strain gauges attached (at four depths) was pushed installation. However, if the load transfer line between depths
into the fresh grout. of 1.45 and 4.0 m was extended to the CMC pile tip, the tip resis-
On the following day, the four CMC reaction piles (R1, R2, R3, tance at the nominal capacity estimated using Davisson criteria
R4) were installed using the same methodology outlined above. (∼890 kN) would be ∼800 kN.

© ASCE 04015059-4 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059

 Reaction Pi le 4 Reaction Pi le 2

Push-in Pressure Sensor (PS)

Shape Acceleration Array (SAA)

CMC Test Pile

1.83 m
SAA2 PS2

SAA1 PS1
SAA4 PS4
SAA3
PS3
Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

Reaction Pile 3 4.88 m Reaction Pile 1

Note: not to scale
(a)

(b)

Fig. 3. (a) plan view of the test setup and instrumentation; (b) side view of instrumentation

Soil Horizontal Stress 12.6 kPa at 2D, 3D, and 4D, respectively. Fig. 9 also shows that
Fig. 8(a) summarizes the variation of soil horizontal stress measure- the horizontal stresses were increased by 2.82 kPa at 1D, 8.03 kPa
ments during the installation of the CMC test unit, the installation at 2D, 7.11 kPa at 3D and 5.69 kPa at 4D at the end of CMC test
of the reaction piles, and the vertical load test. This figure clearly unit installation. In general, the data shows a decreasing trend of
shows that changes in the soil horizontal stresses continue to occur measured pressures as the distance from the CMC test unit in-
throughout the installation process. Fig. 8(b) shows the changes of creased. During the vertical load test, a very small pressure change
horizontal stress during the installation of the CMC test unit. It was was recorded.
observed that the soil horizontal stress increased when the mandrel Fig. 10 shows a summary of measured change of horizontal
passed the depth (location) of the pressure sensors during the down- stress for all sensors during the installation of the CMC units
ward mandrel advancement and during retrieval. This trend will be and the load test. Given the short time of the CMC test unit instal-
discussed further later in the paper. As a function of distance, Fig. 9 lation (mandrel advancement, waiting period, casting and mandrel
summarizes the measured horizontal stresses at the main stages of retrieval) relative to the waiting period between the end of instal-
the construction process. This figure also illustrates that the maxi- lation and the beginning of static load test and the time needed to
mum pressure was recorded when the mandrel passed the location perform the load test, and to clarify the variation of stresses in dif-
of the pressure sensors. At 1D, the recorded horizontal stress was ferent construction stages, the term stage was used on the x-axis of
108 kPa when the mandrel passed the location of the pressure Fig. 10. Fig. 10 illustrates that the advancement of the mandrel
sensors with an initial reading of 22.8 kPa (stress increased by caused a gradual buildup of horizontal stresses reaching a maxi-
85.2 kPa). This measured stress increase was 26.1, 45.5, and mum when the mandrel passed the location of the pressure sensors.

© ASCE 04015059-5 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059

 Applied Load (kN)
0 200 400 600 800 1000 1200
0

10

Vertical Displacement (mm)

20

30

40

50

60
Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

70

80

Fig. 6. Measured vertical load-displacement response of the CMC test

unit

Fig. 4. Installation of CMC test unit showing the penetration of the

auger into the soil reported by Lehane and White (2005) and Basu et al. (2011). It
is also important to note that the horizontal stresses increase again
as the mandrel moved upward (mandrel retrieval), reaching a local
This trend is consistent with the trend reported by Lee et al. (2004) maximum as the mandrel passed the location of the pressure sen-
who investigated the effects of installation of sand compaction piles sors. The horizontal stress then decreases as the mandrel moved
in soft clay. After passing the location of the pressure sensors, the upward to the ground surface. The increase of the horizontal stress
horizontal stresses decreased, which is consistent with the trend during the installation process will consolidate the soil laterally,

10

8 Pause at CMC tip

Withdrawal to location

7
Penetration to CMC tip

of pressure sensors

6
Depth (m)

3 Pause at location Pause at location

of sensors of sensors
2
Penetration to location Withdrawal to
1
of pressure sensors ground surface
0
0 250 500 750 1000 1250 1500 1750 2000
(a) Time (s)

0 0 0

2 2 2

4 4 4
Depth (m)

Depth (m)
Depth (m)

6 6 6

8 8 8
Penetration
Withdrawal

10 10 10
0 8000 16000 24000 0 20 40 60 80 100 120 140 160 0 2 4 6 8 10
(b) Vertical Pressure (kPa) (c) Torque (kN.m) (d) Penetration and Withdrawal Speed (m/min)

Fig. 5. Summary of installation equipment real-time monitoring using computerized quality-control system

© ASCE 04015059-6 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059

 Force (kN) Horizontal Distance (D: pile diameter)
0 200 400 600 800 1000 1200 0D 1D 2D 3D 4D
0 120
Initial
SM 110 Mandrel Passing Sensors
Mandrel at tip
100 CMC Test Unit End of CMC Installation

Soil Horizontal Stress (kPa)

Depth below Soil Surface (m)

ML Begining of Reaction Installation

2 90
End of Reaction Installation
80 Vertical Load Test at 890 kN
Vertical Load Test at 980 kN
70 Vertical Load Test at 0 kN
60 Vertical Load Test at 0 kN 14 mins
4
50
Applied Load (kN)
40
Extrapolated 222
445 30
6
667
Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

20
890
? Strain gauge not functioning
10
CL 0
8
0 200 400 600 800 1000 1200 1400
Fig. 7. Load transfer along the length of the CMC test unit during the Horizontal Distance (mm)
vertical load test showing the soil layers at the site
Fig. 9. Measured soil horizontal stress with distance during CMC
installation and vertical load testing
which is confirmed by the pore water pressure measurements and
the lateral soil movement during construction. The lateral consoli-
a time-dependent capacity similar to driven piles setup; an effect
dation and horizontal soil displacement during construction are ex-
that has not been investigated for drilled displacement piles and
pected to improve the soil properties and increase the vertical load
will be evaluated for CMCs by the research team in the future.
capacity of the foundation systems (Handy 2001) and may result in
After reducing the load to 0 kN at the end of the load test, the
data of the soil pressure sensors were collected for 14 min. Fig. 10
120 shows that the soil horizontal stresses were increased by approx-
Installation of reaction piles

Distance from CMC Shaft imately 0 kPa at 1D, 6.82 kPa at 2D, 2.55 kPa at 3D and 7.03 kPa
PS1 at 1D
CMC installation

100 PS2 at 2D at 4D. These final soil horizontal stress change represent locked-in
Soil Horizontal Stress (kPa)

PS3 at 3D
PS4 at 4D
horizontal stress.
80 The measured changes of soil horizontal stresses were compared
with the cavity expansion solutions presented by Yu (2000). Using
980 kN
PS1 PS2 PS3 PS4
Vertical load test

60 Pile
1D 1D 1D 1D

40 0 kN

20

0
11/26/12 11/28/12 11/30/12 12/2/12 12/4/12 12/6/12 12/8/12
(a) Time (date) 120
Mandrel Advancement Mandrel Retrieval Installing Vertical Load
Reaction Test
Change of Soil Horizontal Stress (kPa)

120
Passing sensors
Withdraw to below sensors and hold
Mandrel pass sensors

End of holding

Piles
Passing sensors during withdraw

Finish installation
Reaching tip

100 1D
110
Change of Soil Horizontal Stress (kPa)

Distance from CMC Shaft 2D

100 PS1 at 1D 3D PS1 PS2 PS3 PS4
90 PS2 at 2D 4D Pile
PS3 at 3D 80 1D 1D 1D 1D
80 PS4 at 4D
70 Passing sensors

60 60
PS1 PS2 PS3 PS4
50
Pile 980 kN
40 1D 1D 1D 1D
40 890 kN
30 Mandrel at tip
0 kN
20
10 20
0
-10 0
-20 0 2 4 6 8 10 12 14 16 18 20 22 24 26
14:20 14:40 15:00 15:20 15:40 16:00 16:20 16:40 Stage
(b) Time (hr:min)
Fig. 10. Measured change of soil horizontal stress during installation
Fig. 8. (a) Measured soil horizontal stress during CMC installation and and testing; the x-axis represents different stages including mandrel
vertical load testing; (b) the change of soil horizontal stress during advancement, waiting period, casting and mandrel retrieval, and verti-
CMC test unit installation cal load testing

© ASCE 04015059-7 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059

 the initial modulus at confining pressure of 25 kPa [Figs. 1 and 2(a)], Pore Water Pressure
the calculated changes of horizontal stress at the pressure sensors
Fig. 11 shows a summary of the measured change of pore water
located at 1D and 3D (PS1 and PS3) were 73.0 and 49.5 kPa, re- pressure during the installation of the CMC units and the load test.
spectively, which represent approximately 17% and 8% difference (Pore pressure measurements of PS1 did not function properly after
from the measured change in soil horizontal stress at the same loca- installation.) This figure clearly illustrates that the maximum pore
tions (85.2 and 45.5 kPa). At the location of PS2 and PS4, the differ- pressure was also observed when the mandrel passed the location of
ences were 56 and 70%. Further detailed analyses, including both the pressure sensors. This is similar to the data reported by Lee et al.
cavity expansion and finite element analysis, are currently being (2004) investigating the effects of installation of sand compaction
conducted by the research team, which will be presented in a future piles in soft clay. After passing the location of the pressure sensors,
journal paper focusing on utilizing several analytical methods. the pore pressure decreased. The pore pressure increased again

SAA1 Lateral Position (mm)

Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

-30 -20 -10 0 10 20 30 40 50 60 70 80 90 100

0
SM

2 Depth of
pressure sensors ML
16
Mandrel Advancement Mandrel Retrieval Installing Vertical Load
Reaction Test
Change of Pore Water Pressure (kPa)

14 2D Piles
3D

Depth (m)
4D PS1 PS2 PS3 PS4
12
Passing sensors Pile 4
1D 1D 1D 1D
10 Passing sensors
890 kN

8 SAA1 2 3 4
Pile
1D 1D 1D 1D
6 Mandrel at tip 6
Initial CL
4 Reaching the CMC tip
980 kN 0 kN End of CMC installation
2 Beginning of Reaction Pile Installation
End of Reaction Pile Construction
Vertical Load Test at 890 kN
0 8
0 2 4 6 8 10 12 14 16 18 20 22 24 26 (a)
Stage
SAA2 Lateral Position (mm)
Fig. 11. Measured change of pore water pressure during installation -30 -20 -10 0 10 20 30 40 50 60 70 80 90 100
and testing; the x-axis represents different stages including mandrel 0
advancement, waiting period, casting and mandrel retrieval, and SM
vertical load testing

2 Depth of
pressure sensors ML
10

9 PS2
Mandrel pass sensors
Change of Pore Water Pressure (kPa)

PS3
8
Depth (m)

PS4
End of holding

7 4

6
PS2 PS3 PS4
5 Pile
2D 1D 1D SAA1 2 3 4
4 Pile
1D 1D 1D 1D
6
3
Initial CL
Reaching the CMC tip
2
End of CMC installation
Beginning of reaction piles construction
1 End of Reaction Pile Installation
Vertical Load Test at 890 kN
0 8
14:53 14:58 15:03 (b)
Time (hr:min)
Fig. 13. Lateral location of the SAAs at the end of major stages:
Fig. 12. Dissipation of the pore pressure during mandrel holding stage (a) SAA 1; (b) SAA 2

© ASCE 04015059-8 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059

 during mandrel retrieval. During mandrel advancement, the maxi- however, has been commonly modeled using uniform displace-
mum increase of pore water pressure was 9.4 kPa at 2D, 4.8 at 3D, ments (uniform cavity expansion) that produce a similar vertical
and 0.6 kPa at 4D. During the installation of the CMC test unit, the load-displacement pile response (e.g., Chen et al. 2009; Dijkstra
mandrel was stopped (pausing or holding stage) at the location of et al. 2011).
the pressure sensors to measure the dissipation of pore pressure. Fig. 14 summarizes the SAA1 and SAA2 movement at the
As shown in Fig. 12, the pore pressure dissipated quickly (within depth of the pressure sensors during different stages of the con-
10 min). This quick decrease of pore pressure is consistent with the struction process (mandrel advancement, mandrel retrieval, instal-
fast consolidation times for the sandy silt soil reported as part lation of reaction piles, and vertical loading). As expected, SAA1
of the site conditions section. Furthermore, the quick decrease shows larger movement than SAA2. The general trend of SAA1
of pore pressure may be attributed to the thixotropic hardening and SAA2 shows the largest soil lateral displacement during man-
mechanism of soft soils where pore pressure decrease quickly after drel advancement. During mandrel retrieval, the increase of soil
remolding (Mitchell and Soga 2005). displacement was very small. Minimal change of soil lateral
displacement was observed when comparing the beginning and
Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

Soil Displacement end of reaction piles installation. At the end of the construction
of the test unit, SAA1 and SAA2 at the location of the pressure
Fig. 13 summarizes the positions of SAA1 and SAA2 at six stages. sensors experienced a lateral movement of 13 mm and 8 mm,
These stages are: (1) initial; (2) mandrel reaching the tip location respectively.
during advancement; (3) end of CMC installation; (4) beginning of
reaction piles installation; (5) end of reaction piles installation; and
(6) near the end of the vertical load test (at load of 890 kN). It was Summary and Conclusions
observed that in general the soil is moving laterally outward (away)
from the CMC test unit at SAA1 and SAA2 (1D and 2D). At SAA3 A full-scale field vertical load test was conducted on a controlled
(3D), the soil experienced minimal movement indicating that the modulus column to investigate the short-term effects of installation
effects of installation extended to 2 to 3D beyond the CMC shaft on surrounding soil. This paper presents the first direct simultane-
surface. This measured zone of soil affected by the CMC pile ous measurements of soil stresses and movement surrounding
installation (∼2 to 3D) is consistent with the analytical results re- CMCs. The soil profile at the test site mainly consists of very soft
ported by Dijkstra et al. (2011) and experimental results reported sandy silt with SPT N-value smaller than 2. The soil was instru-
by Lundberg et al. (2013), Vesic (1977), Salgado (2014), and mented with four push-in pressure sensors and four shape acceler-
Suleiman et al. (2014) for different installation and penetration ation arrays located at ∼1D, 2D, 3D, and 4D from the test unit. The
methods (not including CMC installation). Fig. 13 also indicates soil stresses and displacements were monitored during the con-
that the lateral displacement at a specific radial distance is not struction of the CMC test unit, the installation of reaction piles
uniform with depth. The effect of displacement pile installation, and the vertical load test. The vertical load test results show that
the ultimate vertical load capacity of the CMC test unit was
980 kN. Using the measured load-displacement response, the pile
capacity was estimated using two different methods, which show
that the pile capacity ranged from ∼890 to 916 kN. The results pre-
sented in this paper clearly show that the soil horizontal stresses
vary during different stages of the CMC installation. During man-
drel advancement, the horizontal stresses gradually increase to a
maximum value when the mandrel reached the location (depth)
of the pressure sensors, followed by a horizontal stress decrease.
The measured pressures increase again during mandrel retrieval.
25 Pore water pressure shows a similar trend. The pressure sensor
Mandrel Advancement Mandrel Retrieval Installing Vertical Load
Reaction Test measurements also indicate small locked-in horizontal stresses after
SAA1 2 3 4
20 Pile
1D 1D 1D 1D
Piles the end of the vertical load test (less than 8 kPa). SAA measure-
Passing sensors
Mandrel at tip ments show that the soil moved radially outward during the test
Lateral Movement (mm)

15 unit installation. The measurements reported in this paper indicate

890 kN
that the zone affected by the CMC installation extended 2 to 3D
Passing sensors
10 from the shaft surface of the CMC. These results are consistent with
the data reported in the literature related to the effects of installation
5 of jacked piles and sand compaction piles. The increase of stress
and lateral soil movement during installation are expected to
0 change the CMC load capacity with time, which will be evaluated
as part of a future study. The research team is currently evaluating
-5 SAA1
the effects of CMC installation using detailed finite element
SAA2 simulations.
-10
0 2 4 6 8 10 12 14 16 18 20 22 24
Stage
Acknowledgments
Fig. 14. Change of radial soil displacement at the depth of the pressure This research was partially sponsored by Menard. The authors
sensors for SAA1 and SAA2; the x-axis represents different stages
would like to acknowledge the help of Carl Bowman, instrumenta-
including mandrel advancement, waiting period, casting and mandrel
tion specialist at Lehigh University’s Advanced Technology for
retrieval, and vertical load testing
Large Structural Systems Engineering Research Center.

© ASCE 04015059-9 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059

 References Karlsrud, K., and Mahan, A. (2010). “Evidence of long term ageing effects
on axial capacity of piles in soft clay.” Art of Foundation Engineering
Aboshi, H., Ichimoto, E., Harada, K., and Emoki, M. (1979). “The Practice, GeoFlorida 2010, ASCE, Reston, VA, 325–342.
compozer: A method to improve characteristics of soft clays by inclu- Kulhawy, F. H., and Mayne, P. W. (1990). Manual on estimating soil prop-
sion of large diameter sand columns.” Int. Conf. on Soil Reinforcement, erties for foundation design, Electric Power Research Institute, Palo
Reinforced Earth and Other Techniques, International Geosynthetics Alto, CA.
Society, FL, 211–216. Lee, F. H., Juneja, A., and Tan, T. S. (2004). “Stress and pore pressure
ASTM. (2009). “Standard test methods for deep foundations under static changes due to sand compaction pile installation in soft clay.” Géotech-
axial compressive load.” D1143, West Conshohocken, PA. nique, 54(1), 1–16.
Barksdale, R. D., and Bachus, R. C. (1983). “Design and construction of Lehane, B. M., and White, D. J. (2005). “Lateral stress changes and shaft
stone columns.” Rep. No. 1, FHWA/RD 83/026, Federal Highway friction for model displacement piles in sand.” Can. Geotech. J., 42(4),
Administration. 1039–1052.
Basu, P., Loukidis, D., Prezzi, M., and Salgado, R. (2011). “Analysis of Lundberg, A. B., Dijkstra, J., and Tol, A. F. V. (2013). “Displacement pile
shaft resistance of jacked piles in sands.” Int. J. Numer. Anal. Methods installation effects in sand.” Proc., Int. Conf. on Installation Effects in
Geomech., 35(35), 1605–1635. Geotechnical Engineering, CRC Press, Baca Raton, FL.
Downloaded from ascelibrary.org by LEHIGH UNIVERSITY on 06/18/16. Copyright ASCE. For personal use only; all rights reserved.

Bergado, D. T., Chai, J. C., Alfaro, M. C., and Balasubramaniam, A. S. Masse, F., Brockbank, B., and Pearlman, S. (2004). “CMC (controlled
(1994). Improvement techniques of soft ground in subsiding and modulus columns): Potential application to canadian soils with a new
lowland environment, A.A. Balkema, Rotterdam/Brookfield, VT. trend in ground improvement.” Geo-Quebec 2004: The 57th Canadian
Bowles, J. E. (1996). Foundation analysis and design, 5th Ed., Geotechnical Conf., Canadian Geotechnical Society, BC, Canada.
McGraw-Hill, New York. Measurand. (2013). “Manual of ShapeAccelArray.” Fresericton, NB,
Brown, D. A. (2005). “Practical consideration in the selection and use of Canada 〈http://www.measurandgeotechnical.com〉.
continuous flight auger and drilled displacement piles.” Geotechnical Mitchell, J. K. (1981). “State-of-the-art report, session 12.” 10th Int. Conf.
Special Publication No. 129, ASCE, 251–261. on Soil Mechanics and Foundation Engineering, International Society
Chen, J. F., Jan, J., Oztoprak, S., and Yang, X. M. (2009). “Behavior of Soil Mechanics and Geotechnical Engineering (ISSMGE), London,
of single rammed aggregate piers considering installation effects.” 506–565.
Comput. Geotech., 36(7), 1191–1199. Mitchell, J. K., and Soga, K. (2005). Fundamentals of soil behavior, 3rd
Chow, F. C., Jardine, R. J., Brucy, F., and Nauroy, J. F. (1998). “Effects of Ed., Wiley, NJ.
time on capacity of pipe piles in dense marine sand.” J. Geotech. Geo- Moseley, M., and Kirsch, K. (2004). Ground improvement, Spon Press, NY.
environ. Eng., 10.1061/(ASCE)1090-0241(1998)124:3(254), 254–264. Ng, K. W., Roling, M., AbdelSalam, S. S., Suleiman, M. T., and Sritharan,
Davisson, M. T. (1975). “Pile load capacity.” Proc., ASCE Conf. Design S. (2013a). “Pile setup in cohesive soil: An experimental investigation.”
Construction and Performance of Deep Foundations, ASCE, Reston, VA. J. Geotech. Geoenviron. Eng., 10.1061/(ASCE)GT.1943-5606.0000751,
Dijkstra, J., Broere, W., and Heeres, O. M. (2011). “Numerical simulation 199–209.
of pile installation.” Comput. Geotech., 38(5), 612–622. Ng, K. W., Suleiman, M. T., and Sritharan, S. (2013b). “Pile setup in
Franke, E. (1989). “Co-report to discussion, session 13: Large-diameter cohesive soil: Analytical quantifications and design recommendations.”
piles.” 12th Int. Conf. of Soil Mechanics and Foundation Engineering, J. Geotech. Geoenviron. Eng., 10.1061/(ASCE)GT.1943-5606.0000753,
Taylor & Francis, London. 210–222.
Geokon. (2013). “Instruction manual of VW earth pressure cells, NH, U.S.” Prezzi, M., and Basu, P. (2005). “Overview of construction and design
〈http://www.geokon.com/Manuals〉. of auger cast-in-place and drilled displacement piles.” Proc., 30th
Ghandeharioon, A., Indraratna, B., and Rujikiatkamjorn, C. (2010). Annual Conf. on Deep Foundations, Deep Foundations Institute (DFI),
“Analysis of soil disturbance associated with mandrel-driven prefabri- Hawthorne, NJ, 497–512.
cated vertical drains using an elliptical cavity expansion theory.” Int. J. Randolph, M. F., Carter, J. P., and Wroth, C. P. (1979). “Driven piles in
Geomech., 10.1061/(ASCE)GM.1943-5622.0000027, 53–64. clay—The effects of installation and subsequent consolidation.”
Handy, R. L. (2001). “Does lateral stress really influence settlement.” Geotechnique, 29(4), 361–393.
J. Geotech. Geoenviron. Eng., 10.1061/(ASCE)1090-0241(2001)127: Salgado, R. (2014). “Experimental research on cone penetration resis-
7(623), 623–626. tance.” Keynote Lecture GeoCongress 2014, ASCE, Reston, VA.
Hatanaka, M., and Uchida, A. (1996). “Empirical correlation between Suleiman, M. T., Ni, L., and Raich, A. (2014). “Development of pervious
penetration resistance and effective friction of sandy soil.” Soils Found., concrete pile ground improvement alternative and behavior under ver-
36(4), 1–9. tical loading.” J. Geotech. Geoenviron. Eng., 10.1061/(ASCE)GT
Janbu, N. (1963). “Soil compressibility as determined by oedometer and .1943-5606.0001135, 13.
triaxial tests.” Proc., European Conf. on Soil Mechanics and Founda- Suleiman, M. T., Stevens, L., Jahren, C. T., Ceylan, H., and Conway, W. M.
tion Engineering, Deutsche Gesellschaft für Erd-und Grundbau e.V, (2010). “Identification of practices, design, construction, and repair
Essen, Germany, 19–25. using trenchless technology.” Final Research Rep., Institute of
Karlsrud, K. (2012). “Prediction of load-displacement behavior and capac- Transportation.
ity of axially loaded piles in clay based on analyses and interpretation Vesic, A. S. (1977). “Design of pile foundations.” National Cooperative
of pile load test results.” Ph.D. thesis, Dept. of Civil and Transport Highway Research Program, Transportation Research Board,
Engineering, Norwegian Univ. of Science and Technology, Trondheim, Washington, DC.
Norway. White, D. J., and Suleiman, M. T. (2004). “Design of short aggregate piers
Karlsrud, K., and Haugen, T. (1985). “Axial static capacity of steel model to support highway embankments.” Transportation Research Record
piles in over consolidated clay.” Proc., 11th Int. Conf. on Soil Mechanics 1868, Transportation Research Board, Washington, DC, 103–112.
and Foundation Engineering, International Society of Soil Mechanics Yu, H. S. (2000). Cavity expansion methods in geomechanics, Kluwer
and Geotechnical Engineering (ISSMGE), London. Academic Publishers, Dordrecht, Netherlands.

© ASCE 04015059-10 J. Geotech. Geoenviron. Eng.

View publication stats J. Geotech. Geoenviron. Eng., 2016, 142(1): 04015059