-
Hosted by
Black & Veatch Corporation
GEI Consultants, Inc.
Kleinfelder, Inc.
MWH Americas, Inc.
Parsons Water and Infrastructure Inc.
URS Corporation
21st Century Dam Design
Advances and Adaptations
31st Annual USSD Conference
San Diego, California, April 11-15, 2011
-
On the CoverArtist's rendition of San Vicente Dam after
completion of the dam raise project to increase local storage and
provide
a more flexible conveyance system for use during emergencies
such as earthquakes that could curtail the regions
imported water supplies. The existing 220-foot-high dam, owned
by the City of San Diego, will be raised by 117
feet to increase reservoir storage capacity by 152,000
acre-feet. The project will be the tallest dam raise in the
United States and tallest roller compacted concrete dam raise in
the world.
The information contained in this publication regarding
commercial projects or firms may not be used for
advertising or promotional purposes and may not be construed as
an endorsement of any product or
from by the United States Society on Dams. USSD accepts no
responsibility for the statements made
or the opinions expressed in this publication.
Copyright 2011 U.S. Society on Dams
Printed in the United States of America
Library of Congress Control Number: 2011924673
ISBN 978-1-884575-52-5
U.S. Society on Dams
1616 Seventeenth Street, #483
Denver, CO 80202
Telephone: 303-628-5430
Fax: 303-628-5431
E-mail: [email protected]
Internet: www.ussdams.org
U.S. Society on Dams
Vision
To be the nation's leading organization of professionals
dedicated to advancing the role of dams
for the benefit of society.
Mission USSD is dedicated to:
Advancing the knowledge of dam engineering, construction,
planning, operation,
performance, rehabilitation, decommissioning, maintenance,
security and safety;
Fostering dam technology for socially, environmentally and
financially sustainable water
resources systems;
Providing public awareness of the role of dams in the management
of the nation's water
resources;
Enhancing practices to meet current and future challenges on
dams; and
Representing the United States as an active member of the
International Commission on
Large Dams (ICOLD).
-
Cement Bentonite Slurry Wall Strength 653
CEMENT BENTONITE SLURRY WALL STRENGTH TUTTLE CREEK DAM SEISMIC
REMEDIATION
Amod K. Koirala, Ph.D.1
Glen M. Bellew, P.E.2 John C. Dillon, P.E.3
David L. Mathews, P.E.4
ABSTRACT Cement bentonite (CB) slurry walls were constructed as
a seismic remediation to stabilize the downstream slope of Tuttle
Creek Dam in Manhattan, Kansas. A full-scale test program was
conducted to evaluate various slurry mixes and construction
techniques prior to main construction. Grout mixes with
cement/water (c/w) ratios of 0.3, 0.4, 0.45, and 0.5 were used.
Construction equipment in the test program included long-reach and
clamshell excavators. Sampling and testing were performed on wet
grab samples and core samples. Wet grab samples were obtained from
freshly constructed slurry walls and cured in a laboratory. Core
samples were obtained from cured walls. Evaluation of the test
section results led to the selection of a c/w ratio of 0.5 and the
use of a clamshell excavator for seismic stabilization
construction. The required peak unconfined compressive strength
(UCS) of the cured wall was 300 psi based on stability and
deformation modeling. Observations showed UCS of core samples were
less than wet grab samples. UCS generally increased with specific
gravity and c/w ratio.
INTRODUCTION The Corps of Engineers-Kansas City District
conducted extensive seismic evaluations for Tuttle Creek Dam
located in Manhattan, Kansas. These evaluations concluded that an
earthquake with a magnitude of 5.7 or greater would cause
liquefaction of the foundation sand and result in large
deformations of the embankment. The maximum credible earthquake
(MCE) from the nearby Humboldt Fault Zone is a magnitude 6.6
earthquake. Fast Langrangian Analysis of Continua (FLAC) modeling
demonstrated that downstream slope and toe deformations would
exceed acceptable limits during the MCE. Limit equilibrium slope
stability analysis confirmed these findings. To reduce the risk of
deformation and slope instability during and after the MCE, cement
bentonite (CB) slurry walls were constructed in the foundation
sands through the downstream portion of the embankment. The first
phase of wall construction was a full-scale production test
section. The purpose of the Production Test was to refine materials
and methods for use in the remainder of construction (Stage One
Stabilization and Main Construction Option).
1Civil Engineer, US Army Corps of Engineers, 601 E 12th St.
Kansas City, MO 64106, [email protected] 2Geotechnical
Engineer, US Army Corps of Engineers, 601 E 12th St. Kansas City,
MO 64106 , [email protected] 3Project Manager, US Army
Corps of Engineers, 601 E 12th St. Kansas City, MO 64106,
[email protected] 4Chief, Geotechnical Branch, US Army
Corps of Engineers, 601 E 12th St. Kansas City, MO 64106,
[email protected]
-
654 21st Century Dam Design Advances and Adaptations
Additional work was completed as part of the overall dam
remediation, including construction of a buried collector system to
fill in a collector ditch at the downstream toe of the dam,
upstream riprap overlays, and emergency spillway gate
modifications. The Tuttle Creek Dam modification plan with all
construction stages is shown in Figure 1.
Figure 1. Plan view of Tuttle Creek Dam Downstream Seismic
Stabilization
SUBSURFACE CONDITIONS The soil in the alluvial foundation of
Tuttle Creek Dam consists of 8 to 27 ft of silt and low plasticity
clay underlain by sand, silty sand, and gravelly sand. The sand
deposits vary in thickness from about 25 to 60 ft and can be
separated into two distinct zones. The upper zone consists of a 15
to 20-ft-thick layer of loose fine to medium sand (SM, SP and SW)
and the lower zone consists of a 25 to 30-ft-thick layer of dense
coarse to gravelly sand that increases in grain size with depth
(SP, SW, GP and GW). Due to the alluvial nature of the foundation
deposits, multiple lenses of cohesive soil exist within the
coarse-grained layers. The upper sand zone was determined to be
potentially liquefiable during large earthquake motions. The upper
silts and clays were also expected to suffer significant strength
loss due to large strains caused by liquefaction of the underlying
sand. Bedrock consists of alternating layers of shale and
limestone. The silt and clay form a natural cohesive soil blanket
over the more-permeable sands. This natural cohesive blanket is an
important component of underseepage control. Underseepage pressures
are controlled by a line of pressure relief wells along the
downstream toe.
METHODS AND MATERIALS OF CONSTRUCTION In order to install the
walls, a working platform was constructed on the downstream slope
of the dam. The working platform was constructed of predominantly
sand with an
-
Cement Bentonite Slurry Wall Strength 655
aggregate surface. The equipment used in the production test
section to install walls consisted of a Liebherr HS855DH Crane with
a clamshell excavator and a Koehring 1466 long-reach excavator.
Figure 2 shows a plan view of a clamshell-constructed wall. Walls
constructed with the clamshell excavator were 4 feet wide and were
constructed in a series of bites. The clamshell excavator was
approximately 15 feet long. Each clamshell wall was constructed in
5 bites: three primary bites (PU, PM, PD) and two secondary bites
(SU, SD). Bites were nearly vertical, and the final wall cross
section was approximately a rectangle. A steel frame (guide wall)
was used to guide the clamshell into the wall excavation. Figure 3
shows wall construction with the long-reach excavator. Walls
constructed with the long-reach excavator were 3 feet wide and
constructed in one continuous excavation. Due to the operational
range of the long-reach excavator, walls were not rectangular in
cross section. The bottom corners of long-reach excavator
constructed walls were not square due to the reach of the
excavator. All walls were constructed with the clamshell excavator
after the production test section.
Figure 2. Plan View of Liebherr HS855DH Crane with a clamshell
excavation. Figure
from TreviIcos South (2007) The grout mix used in wall
construction was mixed at an onsite batch plant. The cement used
was Lafarge MaxChem consisting of a 50/50 mixture of Portland
cement and ground granulated blast furnace slag (slag). Cement
water ratios (by weight) of trial mixes in the test program were
0.3, 0.4, 0.45, and 0.5. Additionally, a 25/75 Portland cement to
slag cement mix-ratio was used in a small number of walls in the
test program. Bentonite was typically added at a rate of 5% by
weight of cement and was Wyo-Ben Hydrogel. The additive
Lamsperse-HS was used as a retarder and bentonite antiflocculant to
maintain workability of the mix for a minimum of 24 hours. Water
was obtained from a well screened in the foundation sands at the
downstream toe of the dam.
-
656 21st Century Dam Design Advances and Adaptations
Figure 3. Operational reach of Koehring 1466 long-reach
excavator digging CB wall. Figure from TreviIcos South (2007)
Walls were constructed by excavating and simultaneously placing
self hardening cement bentonite slurry in the trenches. By
continuously placing slurry into the excavation, the trench would
remain open during construction of the wall. Walls were oriented
transverse to the axis of the dam. Walls were typically 3 or 4 feet
wide, 45 feet long, and approximately 65 feet deep. Walls were
spaced with 10 feet of clear space between adjacent walls. The
walls extended through the upper foundation sand at least 12 feet
into the deeper coarse sand. This was done to allow for stress
transfer to stronger materials during shaking. The slurry level in
the walls was observed to drop during cure. The observed drop was a
combination of slurry permeation into the adjacent soil and slurry
bleed during curing. The walls were topped off with fresh slurry
daily to account for drop during curing. Total slurry drop was
typically 10% of wall depth.
SAMPLING AND TESTING Sampling and testing was performed at the
on-site batch plant, on the fluid slurry in the excavations via
wet-grab samples, and on the hardened slurry via core drilling. Wet
grab sampling was conducted on each wall at equally spaced depths.
Wet grab samples were cast in 3-in by 6-in cylinders. The samples
were originally stored in a 100-percent humidity curing room until
being tested for unconfined compressive strength (UCS).
-
Cement Bentonite Slurry Wall Strength 657
After it became apparent the samples were dessicating in the wet
room, all samples were stored submerged under water until UCS
testing. Coring was conducted on 92% of walls in the test section,
and approximately 40% of walls in the remainder of construction.
Initially, coring was conducted with a Geobore system
(double-barrel wireline) producing 4-in-diameter samples. Walls
were typically cored between 60 and 90 days after construction, but
some were cored as early as 28 days and as late as 200 days to
observe strength changes with time. Due to significant strength
discrepancies between wet-grab and core sample strengths, the
coring operation was changed to a triple-barrel coring device in an
effort to reduce sample disturbance and micro-fracturing. However,
the change in coring operation did not have a significant effect on
cores sample UCS. The specific gravity of most wet grab and core
samples was obtained was also. Soil content was calculated from the
difference in the specific gravity of slurry at the plant and cured
wet grab or core samples.
CB WALL PROPERTIES PRODUCTION TEST SECTION The results of
laboratory testing from the production test section were analyzed
to determine the appropriate slurry mix to be used in the remainder
of stabilization. Table 1 shows a summary of wall properties with
various c/w ratios, slag content, and construction equipment.
Thirty eight walls were constructed during the production test,
approximately 4 walls per mix and method shown in Table 2.
Table 1. Summary of cement bentonite wall properties -
production test section
Equipment Cement/Water Ratio+ Sample Type
Age (days)
Average UCS (psi)
Ave. Specific Gravity
Average Soil
Content (%)Long-Reach 0.50 Wet Grab 62 568 1.65 24Long-Reach
0.50 Core 134 387 1.50 12Long-Reach 0.45 Wet Grab 47 414 1.65
24Long-Reach 0.45 Core 161 341 1.64 26Long-Reach 0.40 Wet Grab 54
285 1.62 25Long-Reach 0.40 Core 150 284 1.63 26Long-Reach 0.30 Wet
Grab 49 130 1.58 25Long-Reach 0.30 Core 131 129 1.49 20Clamshell
0.50 Wet Grab 49 626 1.57 18Clamshell 0.50 Core 100 310 1.59
20Clamshell 0.45 Wet Grab 49 436 1.54 17Clamshell 0.45 Core 125 286
1.57 19Clamshell 0.40 Wet Grab 49 273 1.51 16Clamshell 0.40 Core
174 246 1.52 17Clamshell 0.30 Wet Grab 49 115 1.43 14Clamshell 0.30
Core 191 78 1.49 18Clamshell .40 (75% slag) Wet Grab 45 713 -
-Clamshell .40 (75% slag) Core 45 356 - -Clamshell .30 (75% slag)
Wet Grab 45 300 - -Clamshell .30 (75% slag) Core 45 239 - -
+cement content was 50/50 Portland cement to slag ratio unless
otherwise stated In general, wall strength and specific gravity
increased with increasing cement water ratio. Walls constructed
with 75% slag content exhibited significantly higher strength
-
658 21st Century Dam Design Advances and Adaptations
than walls of the same c/w ratio constructed with 50% slag
content. Wet grab UCS was higher than core UCS. On average, core
UCS was 78% of wet grab UCS for all walls constructed in the
production test. The difference in wet grab and core UCS is
attributed to sample disturbance and differences in curing
environment. Due to the discrepancy between core and wet grab UCS,
core samples were used to ensure in situ wall strength requirements
were met. Figure 4 shows core UCS for various c/w ratios and
construction equipment. Walls constructed utilizing the long-reach
excavator had slightly higher core UCS than walls constructed with
the clamshell. This is thought to be due to construction duration
and method. The long-reach excavator was slightly faster than the
clam shell at constructing walls. Walls constructed with the
clamshell generally extended into a second day which required
slurry agitation over a longer duration. The additional agitation
time may have caused some loss of early cement bonding resulting in
higher core UCS of long-reach constructed walls. Higher wet grab
UCS was not observed in long-reach constructed walls. The specific
gravity of long-reach walls was higher than clamshell walls. This
is thought to be caused by more soil being mixed in to the slurry
during construction with the long-reach excavator. Specific gravity
and strength were found to be generally directly proportional, so
the higher specific gravity (soil content) in long-reach walls is
at least partially responsible for their higher strength.
Figure 4. Core sample UCS variation with cement/water ratio and
construction technique,
50% slag walls (Figure from Axtell, Stark, Dillon (2009)) All of
the c/w mixes appeared to achieve the majority of their strength
between ages of 30 and 45 days. Figure 5 shows the variation in UCS
with time for various c/w ratio
-
Cement Bentonite Slurry Wall Strength 659
mixes. Some samples indicated a decreasing UCS with time after
the peak. This is likely caused by testing errors and minor
desiccation of samples not stored under water prior to testing.
Figure 5. Wet grab and core sample UCS variation with
cement/water ratio and age In the production test walls, cores from
the bottom of the wall generally had higher strengths from cores in
the middle or top of the wall. The UCS increase with depth is
likely the result of increased specific gravity, or soil content,
in the lower portion of the wall due to soil particle settlement
and slurry consolidation. Materials and equipment for the remainder
of construction were selected upon completion of the production
test section. A minimum 0.5 cement/water ratio with a 50/50
Portland cement to slag was required in the specifications, and the
Contractor elected to use the clamshell excavator on the remainder
of construction. The selected materials and method were proven in
the production test section as being able to provide a core peak
UCS of 300 psi, which was required in the specifications.
WALL PROPERTIES STAGE ONE AND MAIN CONSTRUCTION OPTION
The remainder of wall construction provided a large data set of
walls constructed with the same slurry mix and construction method,
as well as a smaller data set of higher c/w ratio (0.55 and 0.60)
walls constructed at the Contractors option. The higher c/w ratio
walls were constructed so the Contractor could gain information
regarding stronger mixes and to ensure the last portion of
construction would meet strength requirements and equipment could
be demobilized. The results of laboratory testing from the
remainder of
-
660 21st Century Dam Design Advances and Adaptations
wall construction, the Stage One and Main Construction Option
phases, were analyzed to determine trends in wall properties. Table
2 shows a summary of wall properties from the remainder of wall
construction. Typically the Main Construction Option (MCO)
exhibited slightly higher strengths than the Stage One
Stabilization (S1S). This is likely due to differences in the
subsurface conditions between the two reaches. The thickness of the
silty clay blanket is significantly greater in the S1S area than in
the MCO area. The aggregate qualities of the fine-grained particles
are not as good as those of coarse-grained sand particles and
likely produce lower strengths. The walls with higher c/w ratios
generally had higher wet grab and core UCS. The 0.55 c/w ratio mix
core samples are the exception, but this may be due to small sample
size or early age of core testing. As was observed in the
production test section, wet grab samples exhibited higher
strengths than core samples. Core samples exhibited between 38% and
66% of the UCS of wet grab samples for mixes used in the S1S and
MCO phases of construction.
Table 2. Summary of remainder of 0.5 c/w ratio wall
construction, Main Construction Option
Phase of Construction
Number of Walls
Constructed
c/w Ratio
Sample Type
Age (Days)
Avg. UCS (PSI)
Avg. Specific Gravity
Stage One Stabilization Wet Grab 34 506 1.59
62 0.5
Core 120 333 1.59 Wet Grab 63 658 1.52 235 0.5
Core 95 356 1.65 Wet Grab 63 851 1.57 12 0.55 Core 50 319 1.60
Wet Grab 63 1057 1.70
Main Construction Option
4 0.60 Core 67 604 1.71
Wall UCS and specific gravity with normalized depth for all 0.5
c/w mix walls constructed during the MCO are shown in Figure 6. The
middle portion of the walls exhibited a relatively constant UCS and
a slightly increasing specific gravity with depth. Research has
shown that overburden stresses are transferred to the trench sides
and the slurry does not cure under an increasing confining stress
with depth (Evans and Ryan, 2005). This explains why there is
generally no increase in strength with depth in the middle of the
wall. In the upper portion of the wall, there is an inversely
proportional relationship between strength and specific gravity -
the upper portion of the walls exhibited high strength and low
specific gravity. This is likely due to slurry drop and subsequent
top off that occurred while the wall was curing. Adding slurry
during curing increased the confining stress in this zone leading
to higher UCS. The specific gravity was lower because soil
particles in the upper portion of the wall settled toward the
bottom during cure. The lower portion of the walls exhibited higher
strength and specific gravity with depth. This is likely due to the
accumulation of settling sand particles at the bottom
-
Cement Bentonite Slurry Wall Strength 661
of the wall and consolidation of the slurry mix during curing.
The higher density and soil content of the lower wall causes higher
strengths than less-dense-lower-soil-content-slurry in the upper
and middle portions of the wall.
Figure 6. UCS and specific gravity with normalized sample depth
for 0.5 c/w mix walls
constructed during the Main Construction Option.
CONCLUSIONS A full-scale production test program at Tuttle Creek
Dam was conducted to evaluate various slurry mixes and construction
techniques for construction of cement bentonite walls to improve
seismic stability. The production test indicated slurry with a c/w
of 0.5 would produce the required peak core UCS of 300 psi at
Tuttle Creek Dam. Walls constructed with a long-reach excavator
produced walls with a higher strength than walls constructed with a
clamshell. Walls constructed with higher c/w ratios had higher
specific gravities and strengths. Wet grab samples obtained from
freshly constructed walls and cured in a laboratory exhibited
higher strengths than core samples from cured walls. Because of the
discrepancy between wet grab and core sample strength for high
strength walls, core sample strength was used to ensure in situ
wall strengths were achieved. Strength generally increased with
increasing specific gravity, with a trend deviation in the upper
portion of the wall likely due to slurry drop and subsequent
slurry
-
662 21st Century Dam Design Advances and Adaptations
top off. Construction of high strength cement bentonite walls at
Tuttle Creek Dam proved to be an effective and constructable method
to provide seismic stabilization.
ACKNOWLEDGEMENT
The authors acknowledge the support provided by the U.S Army
Corps of Engineers-Kansas City District, specifically Joe Topi and
Geoff Henggeler, and the expertise of the contractor, Treviicos
South.
REFERENCES Axtell P.J, Stark, T. D. and Dillon J.C. (2009).
Strength Difference Between Clamshell and Long reach Excavator
Constructed Cement-Bentonite Self Hardening Slurry Walls.ASCE
International Foundation Congress and Equipment Expo. pp 297-304
Evans, J. and Ryan, C. (2005) Time-Dependent Strength Behavior of
Soil-Bentonite Slurry Wall Backfill, GSP-142 Waste Contamination
and Remediation. TreviIcos South (2007). Downstream Production Test
Modification Final Report.