Citation
Soil suction applications in Colorado

Material Information

Title:
Soil suction applications in Colorado
Creator:
Sengul, Serkan
Publication Date:
Language:
English
Physical Description:
x, 71 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Soils -- Testing -- Front Range (Colo. and Wyo.) ( lcsh )
Soil mechanics -- Front Range (Colo. and Wyo.) ( lcsh )
Soil consolidation test ( lcsh )
Soil consolidation test ( fast )
Soil mechanics ( fast )
Soils -- Testing ( fast )
United States -- Front Range ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 66-71).
General Note:
Department of Civil Engineering
Statement of Responsibility:
by Serkan Sengul.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
655269253 ( OCLC )
ocn655269253
Classification:
LD1193.E53 2010m S46 ( lcc )

Full Text
SOIL SUCTION APPLICATIONS IN COLORADO
by
Serkan Sengul
B.S., University of Colorado at Boulder, 2002
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Civil Engineering
2010


This thesis for the Master of Science
degree by
Serkan Sengul
has been approved
by
Brian Brady
ShingChun Trever Wang,


Sengul, Serkan (M.S. Civil Engineering)
Soil Suction Applications in Colorado
Thesis directed by Professor Nien-Yin Chang
ABSTRACT
The objective of this study is to provide the results of several laboratory tests
performed on samples obtained from various project sites in the Colorado
Front Range Area in order to show the effectiveness and ineffectiveness of
utilizing soil suction test results for depth of wetting and heave prediction
calculations.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed
Nj^n-Yin Chang
/o


DEDICATION
I dedicate this thesis to my parents and my sister, who gave me an appreciation
of learning and taught me the value of perseverance and resolve. I also
dedicate this to my wife, Karin, for her unfaltering support and understanding
while I was completing this thesis.


ACKNOWLEDGEMENT
My thanks goes to my advisor, Dr. Nien-Yin Chang for his guidance and
understanding of a very demanding schedule. I also wish to thank Mr. Andrew
J. Suedkamp and Mr. James B. Kowalsky of Ground Engineering Consultants,
Inc. who has supported this research, served as a mentor for this and other
endeavors, and supported my educational accomplishments over the past four
years.


TABLE OF CONTENTS
Figures ...............................................................viii
Tables....................................................................x
Chapter
1. Introduction........................................................ 1
1.1 Objective and Scope of Research.......................................1
1.2 Background Information...............................................4
2. Literature Review.....................................................6
2.1 Soil Suction Concept.................................................6
2.2 Filter Paper Method.................................................11
2.3 Chilled Mirror Dew Point Method.....................................16
2.4 Expansive Soil Classification Method (McKeen 1992)..................21
2.4.1 Suction Compressive Index (C^......................................23
2.4.2 Lateral Restraint Factor (f)......................................25
2.4.3 Reduction Factor To Account For Overburden (s).....................27
3. Presentation of Test Results.........................................29
3.1 Filter Paper Versus Chilled Mirror Dew Point........................29
3.2 Soil Suction Versus Water Content...................................40
vi


3.3 Soil Moisture and Suction Relationship.............................43
3.4 Heave Prediction Using McKeens Expansive Soil Classification
Method..................................................................53
3.5 Generating a Soil Moisture Curve in Interbedded Subgrade Material.60
4.0 Discussion and Conclusions....................................... 64
References..............................................................66
vii


LIST OF FIGURES
FIGURE
2.1 TOTAL SUCTION VERSUS RELATIVE HUMIDITY AT 25C.53
2.2 CALIBRATION CURVES FOR TWO TYPES OF FILTER PAPERS
(REPRODUCED FROM ASTM D5298-92)....................14
2.3 PHOTOGRAPH OF WP4-T INSTRUMENT.................17
3.1 SCATTER OF FILTER PAPER TEST DATA..............31
3.2 SCATTER OF WP4-T TEST DATA.....................32
3.3 SCATTER OF AVERAGE SOIL SUCTION VS. WATER
CONTENT............................................33
3.4 SAMPLE 1 REMOLDED SOIL SUCTION VS. WATER
CONTENT............................................36
3.5 SAMPLE 2 REMOLDED SOIL SUCTION VS. WATER
CONTENT............................................36
3.6 SAMPLE 1 REMOLDED SOIL SUCTION VS. WATER CONTENT
WITH TRENDLINE.....................................37
3.7 SAMPLE 2 REMOLDED SOIL SUCTION VS. WATER CONTENT
WITH TRENDLINE.....................................38
3.8 SOIL SUCTION VS. WATER CONTENT, BASED ON 138 DATA
POINTS COLLECTED IN COLORADO FRONT RANGE AREA......41
3.9 SOIL SUCTION VS. LIQUID LIMIT, BASED ON 138 DATA
POINTS COLLECTED IN COLORADO FRONT RANGE AREA......41
3.10 DEPTH VS. SOIL SUCTION........................45
3.11 WATER CONTENT VS. SOIL SUCTION................46
viii


3.12 PLASTIC LIMIT VS. AFTER SWELL MOISTURE CONTENT.47
3.13 WATER CONTENT VS. SOIL SUCTION.................48
3.14 WATER CONTENT VS. SOIL SUCTION.................49
3.15 WATER CONTENT VS. SOIL SUCTION.................50
3.16 WATER CONTENT VS. SOIL SUCTION.................51
3.17 WATER CONTENT VS. SOIL SUCTION.................52
3.18 DEPTH VS. SOIL SUCTION.........................56
3.19 WATER CONTENT VS. SOIL SUCTION.................57
3.20 SOIL MOISTURE CURVE............................58
3.21 DEPTH VS. SOIL SUCTION.........................63
IX


LIST OF TABLES
TABLE
2.1 Unit conversion chart...............................................10
2.2 Suction correlation with physical behavior of soils................11
2.3 Decagon wp4-t device range and accuracy............................18
2.4 Expansive soil classification scheme...............................21
2.5 Description of values used in equation 2.5..........................22
2.6 Overburden reduction factor values..................................28
3.1 Soil suction by filter paper vs. wp4-t potentiameter................30
3.2 Soil suction by filter paper vs. wp4-t potentiameter................35
3.3 Soil suction by filter paper vs. wp4-t potentiameter................39
3.4 Summary of laboratory test results..................................44
3.5 Summary of laboratory test results..................................54
3.6 Summary of laboratory test results..................................56
3.7 Summary of laboratory test results..................................61
3.8 Summary of additional laboratory test results.......................62


1. INTRODUCTION
1.1 OBJECTIVE AND SCOPE OF RESEARCH
Various quantitative and semi-quantitative methods are used by geotechnical
engineers in the Colorado Front Range area to estimate post-construction
heave of buildings, pavements, and other movement sensitive structures as a
step toward development of geotechnical recommendations for foundations,
floor systems, and remedial earthworks. Those typically used are based on
practical engineering experience and judgment using combinations of
measured values of soil moisture content, density, plasticity, one-dimensional
swell, and recently soil suction.
At any point in the design life of a structure, the vertical and lateral extents of
future wetting of the underlying expansive soils will exert a dominant influence
on the extent and distribution of future uplift. Professional experience and
opinions differ with regard to depths to which significant, post- construction,
soil moisture changes take place. Differing assumptions for this future depth
of wetting at a site will give rise to differing estimates of probable, post-
construction heave. Movement estimates based on much deeper than typically
estimated depths of wetting will be larger than those based on more typically
observed depths.
1


Generally, a geotechnical consulting engineer in Colorado has three to four
weeks to start and finish a conventional geotechnical study. The turn around
time is controlled by the industries demands and the competition. If the
consultant exceeds the 4-week period, his ability to satisfy the clients needs
would be significandy impaired. More than approximately sixty percent of all
projects in the Colorado Front Range area require the consulting engineer to
predict potential post-construction movements that may be caused by
expansive soils.
The probable depth of wetting estimate seems to be the mam factor that
affects the outcome of the geotechnical engineers recommendation. Depth of
wetting value is referred as an estimate because currently it is not economically
feasible to calculate the actual value with the limited budget the geotechnical
engineer has. Depth of wetting estimates historically increased in the Colorado
Front Range area from depths of approximately 5 feet (Colorado Association
of Geotechnical Engineers, 1999) to depths of approximately 20 feet
(Colorado Association of Geotechnical Engineers, 1999) between 1970s and
late 1990s. Since the late 1990s depths of wetting of 30, 40 or 70 feet or more
have been considered (Chao et. al., 2006) and have been encountered locally in
the field.
This increase was mainly due to the advances in the heave prediction
methodology. As depth of wetting estimates increased, depth of exploratory
2


borings, deep foundation systems, and over excavation and replacement with
non- or less expansive backfill also increased.
The objective of this study is to provide the results of several laboratory tests
performed on samples obtained from various project sites in the Colorado
Front Range Area in order to show the effectiveness and ineffectiveness of
utilizing soil suction test results for depth of wetting and heave prediction
calculations.
3


1.2 BACKGROUND INFORMATION
Soil suction is a quantity that can be used to characterize the effect of moisture
on volume, and it is a measure of energy or stress that holds the soil water in
the pores or a measure of the pulling stress exerted on the pore water by the
soil mass.
Soil suction sigmficandy influences the behavior of unsaturated soils. Scientists
introduced the concept of soil suction in the early 1900s, in order to describe
the soil water equilibrium for agricultural purposes. Since then more than
twenty methods for measuring soil suction have been considered, none of
which have been satisfactory in measuring soil suction for all conditions.
Lightly loaded structures constructed on expansive soils are often subjected to
severe distress subsequent to construction as a result of changes in the pore-
water pressures in the soil. The structures most commonly damaged are
roadways, airport runways, small buildings, irrigation canals, spillway
structures, and all near ground surface structures associated with infrastructure
development. Changes in the pore-water pressure can occur as a result of
variations in climate, change in the depth to water table, water uptake by
vegetation, removal of vegetation, or the excessive watering of the surface
surrounding the subject structure. The problems associated with expansive
soils have been addressed in many international and regional conferences.
4


'ITiere have been three symposiums on expansive clays (from 1957 to 1960),
seven international conferences on expansive soils (from 1965 to 1992), three
international conferences on unsaturated soils (from 1995 to 2002), and many
other regional conferences. The research literature shows that the prediction of
heave associated with the wetting of an expansive soil has received more
attention than any other problem involving unsaturated soils. Numerous
reports of expansive soil problems and related damage have been documented
in various countries (e.g. Miller and Nelson, 1992; Chen, 1988)
The worldwide interest in research on expansive soils in the last four decades
has resulted in numerous methods being proposed for the prediction of heave.
The heave prediction methods are based either on one-dimensional oedometer
test results or on direct matric suction measurements (Fredlund and Rahardjo,
1993). Although an analytical tool for the prediction of heave is extremely
important, there has been little advancement in the development of an
analytical method for solving engineering problems. There does not appear to
be a computer program, for example, that has been written and widely
accepted for multidimensional heave predictions in expansive soils.
5


2. LITERATURE REVIEW
2.1 SOIL SUCTION CONCEPT
The theoretical concept of soil suction was developed in soil physics in the
early 1900s (Buckingham, 1907; Gardner and Widtsoe, 1921; Richards, 1928;
Schofield, 1935) in relation to the soil-water-plant system. Soil suction is
commonly referred to as the free energy state of soil water (Anderson and
Edlefsen, 1943). The free energy of the soil water can be measured in terms of
the partial vapor pressure of the soil water (Richards, 1965).
The soil suction is composed of two components: matnc suction and osmotic
suction (Fredlund and Rahardjo, 1993).
The osmotic suction in a clay results from the forces exerted on water
molecules as a result of the chemical activity of the soil (Miller and Nelson,
1992). The osmotic suction is caused by the concentration of soluble salts in
the pore water, and it is pressure independent. The effect of the osmotic
suction on swell is not well known. Osmotic effect may be observed if the
concentration of soluble salts in the pore water differs from that of the
externally available water. Therefore, the osmotic suction should not
significandy affect heave (swell) if the salt concentration of the soil is not
altered significandy.
6


The matric suction in a clay results from combination of many factors. These
factors include the water absorption forces of the clay particles, geometrical
configuration of the soil, texture, capillary tension in the pore water, pore-
water pressure, and ambient atmospheric pressure. Matric suction is pressure
dependent and assumed to be related to the in situ pore water pressure.
The total suction is expressed as a positive quantity and is defined as the sum
of matrix and osmotic suction. This relationship can be formed in an equation
as follows (Bulut, Lytton, Warren, 2001):
h, = hm + hp Equation 2.1
Where ht = total suction (kPa), hm = matric suction (kPa), and hp = osmotic
suction (kPa).
Total suction can be calculated using Kelvins equation, which is derived from
the ideal gas law using the principles of thermodynamics and is given as:
V
Equation 2.2
Where, ht = total suction, R = universal gas constant, T = absolute
temperature, K =molecular volume of water, P / P0 = relative humidity, P =
partial pressure of pore water vapor, and P0 = saturation pressure of water
vapor over a flat surface of pure water at the same temperature.
7


If Equation 2.2 is evaluated at a reference temperature of 25C, the following
total suction and relative humidity relationship can be obtained:
h, -137182 In
A
Equation 2.3
Figure 2.1 shows a plot of Equation 2.3 at 25C temperature. From Figure 2.1,
it can be seen that there is nearly a linear relationship between total suction (hj
and relative humidity (P/PJ over a very small relative humidity range. It can be
concluded, in general, that in a closed system under isothermal conditions the
relative humidity may be associated with the water content of the system.
Figure 2.1 TOTAL SUCTION VERSUS RELATIVE HUMIDITY AT 25C
8


Therefore, the suction value of a soil sample can be inferred from the relative
humidity and suction relationship if the relative humidity is evaluated in some
way. In a closed system, if the water is pure enough, the partial pressure of the
water vapor at equilibrium is equal to the saturated vapor pressure at
temperature, T. However, the partial pressure of the water vapor over a partly
saturated soil will be less than the saturation vapor pressure of pure water due
to the soil matrix structure and the free ions and salts contained in the soil
water (Fredlund and Rahardjo, 1993).
In geotechnical engineering practice, suction values are generally defined in
pF units. pF is defined as the base 10 logarithm of the suction expressed
in cm of water. (Schofield, 1935) However, soil suction is also currently being
represented in log MPa unit system (Fredlund and Rahardjo 1993) (i.e., suction
in log MPa = loglO (| suction in MPa |)).
To convert between MPa and pF, first convert MPa to cm of water. The
conversion factor is -10,200 cm/MPa. Then take base 10 logarithm to get pF.
The following formula can be used for conversion:
pF=log(10,200xMPa) Equation 2.4
Table 2.1, provides a conversion chart created using Equation 2.4.
9


Table 2.1 Unit conversion chart
MPa pF
-0.1 3.01
-0.2 3.31
-0.3 3.49
-0.5 3.71
-1 4.01
-10 5.01
-20 5.31
-30 5.49
-60 5.79
-100 6.01
-200 6.31
-300 6.49
-600 6.79
-780 6.90
The relation between suction and water content tends to be a straight line
between the levels of approximately 6 pF and 3 pF, which covers most real
soils. Between 2 and 2.5 water will begin to run off as the soil reaches its field
capacity, the point at which additional water drains away rather than being
absorbed into the soil. Table 2.2 (McKeen, 1992) illustrates some familiar soil
behavior correlated with associated suction level.
10


Table 2.2 Suction correlation with physical behavior of soils
Behavior Suction Level (pF)
Saturation 0.0
Liquid Limit 1.0
Field Capacity 2-2.5
Plastic Limit 3.2 3.5
Plant Wilting Point 4.2 4.5
Tensile Strength of Water 5.3
Shrinkage Limit / Air Dry 5.5
Oven Dry 7.0
2.2 FILTER PAPER METHOD
The filter paper method is based on the water-absorptive characteristics of a
filter paper. When a filter paper is placed in a temperature-controlled
environment with a soil sample for a period of time, it will absorb or de-absorb
the moisture in the controlled environment until complete moisture
equilibrium is reached between the soil sample and the filter paper. With the
filter paper method, both total and matric suction can be measured. If the
filter paper is allowed to absorb water through vapor flow (non-contact
method), then only total suction is measured. However, if the filter paper is
allowed to absorb water through fluid flow (contact method), then only matric
suction is measured.
11


The minimum period of time for establishing the equilibrium between the
filter paper and the soil has been determined to be at least seven days. The
minimum seven-day period is required in order to reach equilibrium condition
among the vapor pressure of the pore water in the filter paper and the partial
vapor pressure of the water in the air inside the container. After seven days the
water content of the filter paper is carefully measured and the water content in
the percent dry weight is subsequendy determined. The water content of the
filter paper can be related to the soil suction value by using a calibration curve.
Idle filter paper must be calibrated, prior to placing the filter paper in the air
tight container, by either suspending it above salt solutions such as reagent
grade potassium chloride or sodium chloride of known molality in distilled
water or using the pressure membrane and ceramic plate. The filter paper used
must be ash-free quantitative Type II filter paper, such as Whatman No. 42,
Fisherbrand 9-790A, or Schleicher and Schuell No.589 White Ribbon.
After the calibration curve is generated the soil suction can be determined by
using this curve. Typical calibration curves can also be obtained from the filter
paper manufacturers.
The filter paper method was evolved in Europe in the 1920s and was first
introduced to United States in 1937 by Gardner. Since its introduction, the
filter paper method has been used and investigated by numerous researchers
(Fawcet and Collis-George 1967; McQueen and Miller 1968; Chandler and
Guierrez 1986; Houston et al. 1994), who researched numerous different
12


aspects of the filter paper method, lire researches performed used different
types of filter papers, suction measuring devices, and calibration techniques.
The majority of the procedures used in the researches prior to 1992 are
presented under the filter paper method adopted by ASTM (D5298-92) as a
standard method for soil suction measurement.
The most commonly used filter papers include Whatman No. 42 filter paper
and Schleicher & Schuell No. 589 filter paper. The Whatman No. 42 filter
paper is generally used in Europe and Australia, whereas the Schleicher &
Schuell No. 589 filter paper method is more popular in North America. The
standard method adopted by ASTM (D5298-92) recommended the best
calibration curves, which were originally presented by Greacen et al. 1987.
Figure 2.2 is a copy of the calibration curves presented in the latest ASTM
standard (D5298-94). The recommended calibration curves were obtained
mostly from the research by Fawcett and Collis-George (1967) for Whatman
No. 42 filter paper, and McQueen and Miller (1968) for Schleicher and Schuell
No. 589 filter paper.
13


Figure 2.2 CALIBRATION CURVES FOR 1WO TYPES OF FILTER
PAPERS (REPRODUCED FROM ASTM D5298-92)
14


The persistent use of filter paper for over half a century includes the following
reasons:
1. small differences in water-absorptive behavior within the same brand
of filter paper,
2. low costs of filter paper and measurement as compared to other
methods,
3. short equalization time over a wide range of suction due to small
thickness of filter paper,
4. the ability of filter paper to maintain physical integrity and chemical
stability during calibration and measurement, and
5. failure to find other suitable moisture-absorptive materials.
Filter paper method has been subject to numerous controversies, denial or
approval, mainly due to the following factors (Petty and Bryant 2002; Gardner
2002):
1. filter paper methods vulnerability to significant operator errors,
2. filter paper methods vulnerability to temperature changes in the
surrounding environment,
3. differences in the calibration curves used for same type of filter papers
that are produced during different batches, and
4. slow test result mm around time due to the minimum 7-day
equilibrium process.
The need for a more reliable and rapid procedure for determining the soil
suction has simulated the interest in chilled-mirror devices for measuring soil
15


suction using the chilled mirror dew point technique (Gee at al. 1992; Petty
and Bryant 2002, Bulut et al. 2002; Leong et al. 2003; Likos and Lu 2004).
2.3 CHILLED MIRROR DEW POINT METHOD
The measurement of the water vapor content of a gas by the dew-point
technique involves chilling a surface, usually a metallic mirror, to the
temperature at which water on the mirror surface is in equilibrium with the
water vapor pressure in the gas sample above the surface. At this temperature,
the mass of water on the surface is neither increasing (too cold a surface) nor
decreasing (too warm a surface).
In the chilled-mirror technique, a mirror is constructed from a material with
good thermal conductivity such as silver or copper, and properly plated with
an inert metal such as iridium, rubidium, nickel, or gold to prevent tarnishing
and oxidation. The mirror is chilled using a thermoelectric cooler until dew just
begins to form. A beam of light, typically from a solid-state broadband light
emitting diode, is aimed at the mirror surface and a photodetector monitors
reflected light.
As the gas sample flows over the chilled mirror, dew droplets form on the
mirror surface, and the reflected light is scattered. As the amount of reflected
light decreases, the photodetector output also decreases. This in mm controls
the thermoelectric heat pump via an analog or digital control system that
16


maintains the mirror temperature at the dew point. A precision miniature
platinum resistance thermometer properly embedded in the mirror monitors
the mirror temperature at the established dew point.
An example of a chilled-mirror device that has been commonly used by the
geotechnical industry is a device manufactured by Decacon Devices, Inc.,
Pullmann, WA, U.S.A. The device is referred as the WP4-T. A photograph of
the device is provided in Figure 2.3.
Figure 2.3 PHOTOGRAPH OF WP4-T INSTRUMENT
WP4-T features internal temperature control and uses the chilled-mirror dew
point technique to measure the water potential of a sample.
17


Water potential is defined as the potential energy of water per unit mass of
water in the system. The total water potential of a sample is the sum of four
component potentials: gravitational, matnc, osmotic, and pressure.
Gravitational potential depends on the position of the water in a gravitational
field. Matric potential depends on the adsorptive forces binding water to a
matrix. Osmotic potential depends on the concentration of dissolved
substance in the water. Pressure potential depends on the hydrostatic or
pneumatic pressure on the water. The WP4-T measures the sum of the
osmotic and matric potentials otherwise referred as the total suction in a
sample.
The WP4-T measures the total soil suction, and is therefore well suited to
classify expansive soils. The WP4-T device accuracy, based on the information
provided by Decagon Devices, Inc., Pullman, WA, U.S.A., is summarized in
Table 2.3.
Table 2.3 Decagon wp4-t device range and accuracy
Range Accuracy
Device Accuracy: 0 to -10 MPa + 0.1 MPa
(Decagon Devices, Inc.) -10 to -60 MPa + 1%
-60 to -300 MPa 5%
Chilled mirror devices, such as the WP4-T, have been used for measuring total
suction for over a decade (Gee et al., 1992; Petry and Bryant, 2002; Bulut et al.
18


2002; Leong et al., 2003; Likos and Lu, 2004; Petry and Jiang, 2007) with
various success levels.
The reviewed literature that compares the measurement of total suction using
the filter paper method versus the chilled-mirror device, WP4 or WP4-T,
concluded controversial results. Some of these literature reviewed concluded
differences and scatter between the filter paper method and the chilled-mirror
device (Petry and Bryant, 2002; Bulut et al. 2002). Whereas, some of the
literature reviewed showed very close agreements between total suction values
obtained using the two different methods (Gee et al. 1992; Likos and Lu, 2004).
However, majority of the literature reviewed, including the research performed
by Petry and Bryant (2002), evaluating the consistence and convenience of
measuring soil suction with the WP4 device compared to the filter paper
method concluded that the WP4 and WP4-T devices are more practical, less
costly in time and expense, and less likely to contain errors than the filter paper
method.
A comparison of the total suction measurements performed by chilled-mirror
and water retention curves estimated from filter paper total suction data
performed by Patrick et al. (2007) showed general agreement between the
results obtained from both methods, noticing the following sources as causes of
possible discrepancies:
19


a. errors in chilled mirror total suction measurements due to incomplete
equilibrium in the sealed test chamber of the chilled-mirror device, and
b. errors in the estimated filter paper total suction values due to natural
variations of the zero water content intercept in the log total suction
versus water content relationship.
The comparison study performed by Patrick et al. (2007) also suggested that
equilibrium times of nearly 30 minutes were required versus the 5 to 10 minutes
of equilibrium time recommended by the cited manufacturers literature for
WP4-T chilled-mirror device and that such errors can be easily avoided by using
the continuous reading mode on the WP4-T device to monitor total suction
versus time until equilibrium has been clearly reached.
A round-robin testing of Dewpoint Potentiameter versus filter paper to
determine total suction was performed by Petty and Jiang (2007) that involve
several geotechnical laboratories concluded that the WP4 chilled-mirror device
provides a slightly more conservative measure of total suction and that the
scatter of data between the laboratories is significandy less using the WP4 rather
than the filter paper method.
20


2.4 EXPANSIVE SOIL CLASSIFICATION METHOD (MCKEEN
1992)
Costly engineering mistakes are likely if expansive tendencies of clays under
structures are misjudged. Geotechnical engineers therefore need a reliable and
quick method to determine how expansive a soil is. Such a method was
proposed by McKeen (1992). The method uses the slope of a soil moisture
characteristic (relation between soil suction and water content for to classify
the soil into one of five categories. Low numbered categories are problem
soils. High numbered categories give little or no expansion when wetted and
dried. The expansive soil classification scheme adapted from McKeen (1992) is
presented in Table 2.4.
Table 2.4 Expansive soil classification scheme
Category Slope (Ah/Aw)a chb AHC(%) Expansion
I > 6 -0.227 10 Special Case
II - 6 to -10 -0.227 to -0.12 5.3 High
III -10 to -13 -0.12 to -0.04 1.8 Moderate
IV -13 to-20 -0.04 to 0 - Low
V <-20 0 - N on-Expansive
a Ah/Aw is change in pF per unit change in water content,
b Ch is the suction compression index, from McKeen (1992)
c AH is the vertical movement computed by McKeen (1992)
21


Equation 2.5 was presented by McKeen (1992) as a model for predicting
expansive soil behavior during the 7th International Conference on Expansive
soils in 1992.
AH=Ch Ah At f s Equation 2.5
The descriptions of values utilized in Equation 2.5 are presented in Table 2.5.
Table 2.5 Description of values used in equation 2.5
AH Surface Heave
ch Suction Compression Index of the Soil
Ah Suction Change the Soil Experiences, pF
At Layer Thickness Being Considered, inches
f Lateral restraint factor, dimensionless
S Reduction factor to account for overburden, dimensionless
22


2.4.1 SUCTION COMPRESSION INDEX (Ch)
Suction compression index is the slope of a data plot computed by measuring
the volume change of samples and soil suction values at different moisture
contents. The volume change of the sample is generally measured by the use of
the CLOD test.
The CLOD test basic procedure was explained by Hamburg (1985), and by
Nelson and Miller (1992). The test procedure involves coating the soil samples
with a liquid resin. The resin coating allows sample volume measurements at
different moisture contents. A commonly used resin that is suitable for
performing this test is DOW Saran F310. Once the resin dries on the soil
sample, it forms a waterproof membrane around the soil sample, so that when
the sample is dipped in water moisture content of the sample does not change.
The volume of the sample is then determined by weighing the sample while it
is submerged in a water-filled container that is placed on a very sensitive
electronic scale. The reading of the balance, adjusted for the weight of the
water-filled pan, is equal to the buoyant force on the sample. Sample volume
can then be determined by Archimedes principle, which states that the weight
of the displaced fluid is directly proportional to the volume of the displaced
fluid (if the surrounding fluid is of uniform density). Thus, among completely
submerged objects with equal masses, objects with greater volume have greater
buoyancy.
23


However, the empirical relationship presented in Equation 2.6 (McKeen 1992)
has often been used by the geotechnical engineers.
Ch (-0.02673)(Ah/Aw) -0.38704 Equation 2.6
In Equation 2.6, Ah presents the suction change the soil experiences and Aw
presents the change on water content. The relationship presented in Equation
2.6 is based on a line constructed at the 85 percentile for laboratory tests on
Shelby tube and California drive samples obtained from five different locations
in the U.S.
A laboratory investigation was conducted by Perko et al. (2000) in order to
further determination of the suction compression index. As part of the
investigation a total of 89 relatively undisturbed soil and bedrock samples
collected from 22 sites around the Denver, Colorado area was tested. The
results of the study (Perko et al. 2000) agree closely with the empirical
correlation previously presented by McKeen, also.
In order to utilize Equation 2.6 and compute Ch the Ah/Aw value is obtained
by generating a soil moisture curve.
A moisture release curve, also known as a soil moisture characteristic, relates
the water potential of a particular soil to its water content. This information is
24


important for describing water storage in soil and water availability to plants,
and for predicting water and contaminant transport in soil. A moisture
characteristic curve is obtained by measuring the soil suction and water content
on a set of soil samples. Such examples are presented in Chapter 3.
2.4.2 LATERAL RESTRAINT FACTOR (f)
The most common method used by the geotechnical engineering industry to
predict heave of expansive soils is the one-dimensional oedometer test. One-
dimensional oedometer test results are obtained under fully lateral restraint
conditions by placing the soil sample in a steel ring. However, in the field
heave is a three-dimensional process. In an effort to convert the potential
volume change determined by performing the oedometer test to the
anticipated vertical heave, some researchers have suggested a lateral restrain or
a correction factor to the applied to the swell percentages obtained from the
oedometer tests.
The potential vertical rise method, first described by McDowell (1956),
assumes that only one-third of the total volume change goes upward. Lytton
(1977) multiplied the predicted heave by a factor that varies from ineOthird for
heavily cracked soil to one for intact soil with high lateral restraint.
Furthermore, studies performed by Erol et al. (1987), Richards (1967), and
Fityus and Smith (1998) suggested that in predicting soil heave one-third of the
volume change is reflected as surface heave, the remainder will be laterally.
25


Based on comparison of the measured field heave with oedometer data,
Thompson (1997) and McOmber and Thompson (200) have found an
empirical adjustment factor of 0.70 is needed to account for the difference
between observed heave and predictions obtained from the oedometer swell
tests. McKeen (1992) applies a lateral restraint factor, f, to convert the volume
change to vertical swell using Equation 2.7, where K,, is the coefficient of earth
pressure at rest.
f = (1 + 2Kq)/3 Equation 2.7
McKeen (1992) suggested that the value of f ranges between 1 at 1 to
0.333 at Kq = 0. Thompson (1997) proposed that in using McKeens equation
for predicting soil heave, a value of f = 0.5 is to be used for highly fractured
clay and f = 0.83 for massive clay with little or no fracture.
An experimental study was performed by A1 Shamrani (2004) to evaluate the
effect of lateral restraint conditions on swelling behavior of expansive soils by
performing a series of tri-axial and one-dimensional swell tests on samples
compacted at various moisture contents to different dry' densities and restraint
under various confining pressures. The study results suggested that the
vertical swells obtained from oedometer tests were considerably higher than
the tri-axial test measurements and the ratio of the ultimate vertical swells from
tri-axial tests to those obtained from odometer tests varied between 0.30 and
0.66.
26


2.4.3 REDUCTION FACTOR TO ACCOUNT FOR
OVERBURDEN (s)
Reduction factor to account for overburden factor is indicated by letter (s) and
is a coefficient for load effect on heave. It can be calculated by using Equation
2.9, where (70 is the swell pressure that is being applied to the soil due to
overburden pressure.
s 1.0 0.01 ((7o) Equation 2.8
By assuming that the moist soil weight is 125 psf, Table 2.6 was generated to
show overburden reduction factor change with depth.
27


Table 2.6 Overburden reduction factor values
Overburden Density 125 psf
Depth (ft) Overburden Pressure (psf) Overburden Reduction Factor (s) Depth (ft) Overburden Pressure (ksf) Overburden Reduction Factor (s)
1 125 0.99 31 3875 0.61
2 250 0.98 32 4000 0.6
3 375 0.96 33 4125 0.59
4 500 0.95 34 4250 0.58
5 625 0.94 35 4375 0.56
6 750 0.93 36 4500 0.55
7 875 0.91 37 4625 0.54
8 1000 0.9 38 4750 0.53
9 1125 0.89 39 4875 0.51
10 1250 0.88 40 5000 0.5
11 1375 0.86. 41 5125 0.49
12 1500 0.85 42 5250 0.48
13 1625 0.84 43 5375 0.46
14 1750 0.83 44 5500 0.45
15 1875 0.81 45 5625 0.44
16 2000 0.8 46 5750 0.43
17 2125 0.79 47 5875 0.41
18 2250 0.78 48 6000 0.4
19 2375 0.76 49 6125 0.39
20 2500 0.75 50 6250 0.38
21 2625 0.74 51 6375 0.36
22 2750 0.73 52 6500 0.35
23 2875 0.71 53 6625 0.34
24 3000 0.7 54 6750 0.33
25 3125 0.69 55 6875 0.31
26 3250 0.68 56 7000 0.3
27 3375 0.66 57 7125 0.29
28 3500 0.65 58 7250 0.28
29 3625 0.64 59 7375 0.26
30 3750 0.63 60 7500 0.25
28


3. PRESENTATION OF TEST RESULTS
3.1 FILTER PAPER VERSUS CHILLED MIRROR DEW POINT
In the recent ASCE GeoDenver 2007 conference Round-Robin Testing of a
Dewpoint Potentiameter versus the Filter Paper to Determine Total suction
was discussed (Petty and Jiang, 2007). Findings of this study concludes that
the WP4 Dew Point Potentiameter provides a slighdy more conservative
measure of total suction and the scatter of data is significantly less using the
WP4 than using the filter paper method.
In order to further evaluate the effectiveness of the WP4-T, soil suction
measurements were conducted on twenty different soil samples collected from
twenty different project sites located in the Colorado Front Range area, using
both the WP4-T Dewpoint Potentiameter and the filter paper method.
Test holes were drilled with truck-mounted, continuous flight, power auger
rigs. Relatively undisturbed samples of the subsurface materials were taken
with 2-inch I.D. California -type liner samplers. The samplers were driven
into the substrata with blows from a 140-pound hammer falling 30 inches.
This procedure is similar to the Standard Penetration Test described by ASTM
Method D1586. Soil suction tests were performed on the same California liner
sample using the filter paper method and WP4-T. Both methods used for soil
suction measurement was performed by the same laboratory technician. The
same laboratory technician was used on these tests in effort to reduce the
29


operator error component of the filter paper test. In addition, the samples
collected were tested for standard property tests, such as natural moisture
contents, dry unit weights, grain si2e analyses, and liquid and plastic limits.
Laboratory tests were performed in general accordance with applicable ASTM
protocols. A summary of the test results is presented in Table 3.1.
Table 3.1 Soil suction by filter paper vs. wp4-t potentiameter
uses Classification Natural Moisture Content Natural Dry Density Percent Passing No. 200 Sieve liquid limit Plastic Limit Plasticity Index Soil Suction By Filter Paper Soil Suction By WP-4 Potentiameter
Test 1 Test 2 Test 1 Test 2
sc 15.8 107.1 30 31 20 ii 3.73 3.65 3.66 3.65
sc 16.7 103.8 38 38 25 13 4.05 4.12 4.00 4.00
sc 15.9 116.4 42 35 20 15 4.23 4.18 4.20 4.18
sc 13.9 119.6 48 31 18 13 4.02 3.89 4.04 4.02
sc 14.1 114 49 35 21 14 3.92 3.84 3.90 3.90
CL 19 104 50 41 24 17 3.67 3.67 3.63 3.62
CL 18.1 106.1 56 39 22 17 3.82 3.76 3.80 3.80
CL 9.9 113.7 68 36 18 18 4.68 4.65 4.62 4.63
CL 23.5 108.6 71 46 25 21 3.73 3.68 3.62 3.62
CL 27 96 92 47 29 18 3.77 3.69 3.67 3.67
CH 16.7 1129 91 50 25 25 4.0.3 4.00 4.07 4.08
CII 13.2 116.1 98 58 28 30 4.98 4.95 4.93 4.93
CH 19.3 101 86 63 19 44 4.38 4.46 4.48 4.48
CH 18.9 109.5 95 89 30 59 4.69 4.73 4.64 4.65
CH 15.8 107 95 101 29 72 4.58 4.62 4.50 4.52
MH 39 82.1 73 75 47 28 3.32 3.27 3.20 3.22
MH 34 85.1 59 76 49 27 3.15 3.17 3.12 3.12
MH 35.9 82.5 71 76 47 29 3.32 3.42 3.34 3.35
MH 35.6 95.4 57 59 37 22 3.03 3.04 3.02 3.01
MH 39.3 79.8 85 77 48 29 3.35 3.39 3.31 3.30
30


Based on the information provided in Table 3.1, the scatter charts shown in
Figures 3.1 and 3.2 were generated by scattering the two consecutive soil
suction test results obtained using the Filter Paper Method and WP4-T
Potentiameter method versus the natural soil water content.
Filter Paper Test 1
Filter Paper Test 2
Water Content (%)
Figure 3.1 SCATTER OF FILTER PAPER TEST DATA
31


5.()0 n
9
1 m
fT"
C o f t \ 1 I WP4-T Test 1 WP4-T Test 2
3 'Jj 1
O 1 ft

1 1 m m
1 1
l 0 1 2 1 4 6 1 8 2 0 22 2 Water ( 4 26 2 Content ( 8 3 y) 0 32 3 4 3 6 3 8 40
Figure 3.2 SCATTER OF WP4-T TEST DATA
As shown in Figures 3.1 and 3.2 the scatter of test data is significantly less
using the WP4 -T Potentiameter method versus the Filter Paper method, thus
majority of the points are overlapping in the Figure 3.1 scatter chart.
Therefore, based on the tests performed it can be concluded that the
repeatability of the soil suction test using the filter paper method shows a
higher variation than the WP4-T method. The cause of this variation appears
to be generally due to the vulnerability of the filter paper methods weighing
procedure to large errors. In addition, although filter paper method
equilibrium process was performed in a cooler located in the laboratory, it still
32


does get affected from temperature changes. The WP4-T controls the
temperature of the sample better than the filter paper method.
Based on the information provided in the above Table 3.1, the average values
of Test 1 and 2 for both the Filter Paper method and the WP-4 Potentiameter
method versus the natural moisture content is presented in Figure 3.3 below.
5.00 - i
2
1
EL
C? 4.20 - i | Filter Paper Tests 1 & 2 Average WP-4 lest 1 & 2 Average
u 4.UU 3 v: t
0 ^5U " 1 1 1 P t 1

1 1 I
",
i 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4 2 6 2 8 3 0 3 2 34 3 6 3 8 40
Water Content (%)
Figure 3.3 SCATTER OF AVERAGE SOIL SUCTION VS. WATER
CONTENT
33


As shown in Figure 3.3, test results obtained using the WP4-T Potentiameter
device are generally providing slightly lower soil suction results, average of 0.03
pF lower calculated over the 20 test results provided in Table 3.1, compared to
the filter paper method.
All of the results presented above were performed by conducting Filter Paper
suction tests on undisturbed samples. In order to provide a wider range of test
results additional tests were performed on two remolded specimens using both
the filter paper method and WP4-T method. Two bulk samples were collected
from different sites in the Colorado Front Range area. A standard Proctor
compaction test was performed on each sample to determine its optimum
moisture content and maximum dry density. Laboratory testing was performed
in general accordance with applicable ASTM protocol. Then each sample was
remolded to 0, +2, and +4 percent above its optimum moisture content. The
results of the laboratory test results are summarized below:
34


Table 3.2 Soil suction by filter paper vs. wp4-t potentiameter
Soil Sample Number uses Classific ation Optimum Moisture Content (%) Maximum Dry Density (pcf) Percent Passing No. 200 Sieve (%) Liquid Limit (%) Plastic Limit (%) Plasticity Index (%) Remolded Moisture Content (%) Soil Suction By Filter Paper (pF) Soil Suction By WP4-T Potentia meter (pF)
20.4 4.26 4.38
1 CL 20.4 91.9 70 45 29 16 22.4 4 4.22
24.4 3.8 4.01
18.6 4.45 4.63
2 CH 18.6 103.9 92 52 24 28 20.6 4.26 4.43
22.6 4.18 4.33
Based on the data provided in Table 3.2, below scatter charts as shown in
Figures 3.4 and 3.5 were generated by scattering the soil suction results
obtained using the Filter Paper and the WP-4 Potentiameter method versus
the remolded moisture contents for each sample.
35


Figure 3.4 SAMPLE 1 REMOLDED SOIL SUCTION VS. WATER
CONTENT
&
c
3
SI
'o
St
T3
3
2
1
4.70 -I
4.60
4.50
4.40
4.30
4.20
4.10
17.0
19.0


21.0

23.0
SAMPLE 2-
REMOLDED: Inker
Paper Method
SAMPLE 2-
REMOI.DED: WP4-T
Method
Remolded Water Content ( o)
Figure 3.5 SAMPLE 2 REMOLDED SOIL SUCTION VS. WATER
CONTENT
SAMPLE 1 -
REMOLDED: Ihlter Paper
Method
SAMPLE 1 -
REMOLDED: WP4-T
Method
Remolded Water Content (%)
36


As shown in Figures 3.4 and 3.5, the test results scatter in a linear pattern using
both methods. Moreover, the remolded soil suction test results obtained using
the filter paper method are consistendy lower than the results obtained using
the WP4-T device. When trendlines are constructed using the information
provided in Figures 3.4 and 3.5, the following information provided in Figures
3.6 and 3.7 are obtained. Please note that the remolded moisture contents are
not presented in percent value in these graphs.
Figure 3.6 SAMPLE 1 REMOLDED SOIL SUCTION VS. WATER
CONTENT WITH TRENDLINE
37


Figure 3.7 SAMPLE 2 REMOLDED SOIL SUCTION VS. WATER
CONTENT WITH TRENDLINE
The slope of the trendlines, which represents the change in soil suction per
change in water content as shown in Figures 3.6 and 3.7 are summarized in the
Table 3.3 below:
38


Table 3.3 Soil suction by filter paper vs. WP4-T potentiameter
Soil Sample No uses Classification Optimum Moisture Content (%) Maximum Diy Density (pcf) Percent Passing No. 200 Sieve (%) Liquid Limit (%) Plastic Limit (%) Plasticity Index (%)
1 CL 20.4 91.9 70 45 29 16
2 CH 18.6 103.9 92 52 24 28
Soil Sample No Remolded Moisture Content (%) Soil Suction By Filter Paper (PF) Soil Suction By WP-4 Potentiameter (pF) Slope by Filter Paper (Ah/Aw) Slope by WP4-T (Ah/Aw)
1 20.4 4.26 4.38 -11.5 -9.25
22.4 4 4.22
24.4 3.8 4.01
2 18.6 4.45 4.63 -6.75 -7.5
20.6 4.26 4.43
22.6 4.18 4.33
As shown in Figures 3.6 and 3.7 and Table 3.3, consistently lower results
obtained by the filter paper method does not indicate consistently steeper or
flatter slopes representing change in soil suction per change in moisture
content.
39


3.2 SOIL SUCTION VERSUS WATER CONTENT
A total of 138 samples collected from 50 separate sites located within
Colorado Front Range area were tested for natural water content and soil
suction. Soil suction, water content, and Atterberg limit tests were performed
on all samples in general accordance with ASTM procedures. The soil suction
tests were performed using the WP4-T device. Test results are presented in
Figures 3.8 and 3.9.
As shown in Figure 3.8 the for variety of samples with moisture contents
ranging between 0 and 59 percent, soil suction values tend to stay between the
values of 5 pF and 3.2 pF for all samples tested. Majority of the samples were
located between 4.6 pF and 3.6 pF.
40


Figure 3.8 SOIL SUCTION VS. WATER CONTENT, BASED ON 138
DATA POINTS COLLECTED IN COLORADO FRONT RANGE AREA
41


Figure 3.9 SOIL SUCTION VS. LIQUID LIMIT, BASED ON 138 DATA
POINTS COLLECTED IN COLORADO FRONT RANGE AREA
Figure 3.9 shows the same soil suction results scattered with the liquid limit
values obtained from the Atterberg test. Liquid limit values were ranging
between 10 and 105 percent where soil suction results were between 3.2 pF
and 5 pF.
42


3.3 SOIL MOISTURE AND SUCTION RELATIONSHIP
One test hole was drilled in a project site located in the Colorado Front Range
area. The test hole was drilled with a truck-mounted, continuous flight, power
auger rig using 7-inch diameter hollow stem augers. Relatively undisturbed
samples of the subsurface materials were taken with a 2-inch I.D. California
-type liner sampler. The sampler was driven into the substrata with blows
from a 140-pound hammer falling 30 inches. This procedure is similar to the
Standard Penetration Test described by ASTM Method D1586.
Laboratory testing of soil samples obtained from the subject site included
standard property tests, such as natural moisture contents, dry unit weights,
grain size analyses and liquid and plastic limits, Denver swell-consolidation and
WP4-T soil suction tests were performed on each sample, as well. Laboratory
tests were performed in general accordance with applicable ASTM protocols.
The subsurface conditions at the project site consisted of a layer of overburden
soils approximately 3 feet in thickness, underlain by very hard claystone
bedrock extending to the test hole termination depth. Groundwater was not
encountered during subsurface exploration. Test results are presented in Table
3.4.
43


Table 3.4 Summary of laboratory test results
Sample Location Natural Moisture Natural Percent Atterberg Limits
Test Moisture Content Dry Passing Liquid Plasticity
Hole Depth Content After Swell Density No. 200 Limit Index
No. (feet) (%) r/o) (P 1 4 15.8 19.2 108.5 89 90 66
1 9 21.1 24.7 107 95 101 72
1 14 20.8 25 104.5 94 98 69
1 19 19.9 24.8 103 93 116 96
1 24 17.9 25.3 106.9 93 108 84
1 29 18.9 22 109.5 95 89 59
Sample Location Suction Denver Surcharge Suction uses
Test Before Swell Pressure After Classifi-
Hole Depth Swell Test Swell cation
No. (feet) (PF) (%) (psf) (PF)
1 4 4.6 3.7 1,000 4.01 CH
1 9 4.49 8.9 1,000 4.06 CH
1 14 4.49 11.4 1,000 4 CH
1 19 4.63 16.3 1,000 4.05 CH
1 24 4.64 6.8 1,000 4.08 CH
1 29 4.63 4.8 1,000 4.05 CH
As shown in Table 3.4 all 6 samples collected from the subject site were tested
for moisture content and soil suction both before and after the Denver swell
test was performed.
Denver swell test results indicate that the soil samples tested have high swell
potentials when they are wetted against 1,000-psf surcharge. In order to
calculate the potential for heave, the depth of wetting for the project site will
44


have to be estimated. Since all soil samples were tested for soil suction at their
natural moisture contents, a depth versus soil suction profile was generated
(Figure 3.10) in order to determine the depth to constant suction at this project
site. As shown in Figure 3.10 a depth to constant suction data was obtained
between 19 and 29 feet. Where all three samples tested indicated no significant
change in the soil suction values.
Figure 3.10 DEPTH VS. SOIL SUCTION
45


Figure 3.10 shows the results of soil suction tests performed on the samples
after the completion of the swell test. All after swell suction results obtained
were between 4.00 pF and 4.10 pF. Since all samples were wetted for several
days (in select samples up to 10 days) during the swell test, it can be concluded
that the equilibrium suction value at this site is around 4.00 pF.
4.70 -I
h
4.60
4.50 -
Pi 1 A n Natural Moisture Content
vs. Before Swell Soil
0 Suction
a 4.30 -
D After Swell Moisture
CO A OA . Content vs. After Swell Soil
0 C/3 Suction
4.10 -
I
4.00
3.90 - i 1 1
0 10 20 30
Water Content (%)
Figure 3.11 WATER CONTENT VS. SOIL SUCTION
Figure 3.11 shows the scatter of before swell and after swell soil suction values
versus the water content of the sample. As expected the soil suction values
drop as water content of the samples increase.
46


Figure 3.12 PLASTIC LIMIT VS. AFTER SWELL MOISTURE
CONTENT
Figure 3.12 shows the relationship between the after swell moisture content of
the samples and their respective plastic limits. The moisture content of all
samples was within 5 percent of their plastic limit after the completion of the
swell test.
Using the data presented in Figure 3.11, a linear trendline was created. As
shown in Figure 3.13 the created trendline was forecasted to intercept the soil
suction axis, and indicated a soil suction value of 5.7 pF when water content
equals to 0.
47


Figure 3.13 WATER CONTENT VS. SOIL SUCTION
However, based on previous studies it has been shown that the oven dry
sample exhibits a soil suction value of 7.0 pF. (McKeen, 1992). Therefore, the
trendline was adjusted to intercept the soil suction axis at 7.0 pF. Figure 3.14
shows the adjusted trendline.
48


Figure 3.14 WATER CONTENT VS. SOIL SUCTION
Due to the lack of data between the maximum soil suction value obtained and
the 7.0 pF suction value assumed for the oven dr}' condition. All samples
tested and shown in Table 3.4 were air dried. Several soil water content and
soil suction tests were performed during the air drying period. The results of
the tests are presented in Figure 3.15. Please note that when samples were
oven dried the WP4-T results obtained did not appear accurate and were not
included in Figure 3.15.
49


7.00 -
6.50 -
A
6.00 -
£V
5.50 c %
0 Moisture Content vs.
o 5.00 3 Soil Suction
C/J
^ 4.50 "
C/3
4.00 - * *
3.50 -
3.00 - ! 1 1
0.0 10.0 20.0 30.0
Water Content (%)
Figure 3.15 WATER CONTENT VS. SOIL SUCTION
Using the data presented in Figure 3.15, a linear trendline was created and is
shown in Figure 3.16. The created trendline was forecasted to intercept the soil
suction axis, and indicated a soil suction value of approximately 6.7 pF when
water content equals to 0.
50


7.00
Moisture Content
vs. Soil Suction
Linear Trendline
0.0 10.0 20.0 30.0
Water Content (%)
Figure 3.16 WATER CONTENT VS. SOIL SUCTION
Using the same data presented in Figure 3.15 another trendline was
constructed by setting the soil suction intercept at 7.0 pF, as shown in Figure
3.17.
51


7.00
G-7
A
a
o
a
o
3
(/>
'o
CO
6.50
6.00
5.50
5.00
4.50
4.00
3.50
3.00
y = -0.1255x + 7
R2 = 0.9154
0.0 10.0 20.0 30.0
Moisture Content
vs. Soil Suction
----Linear Trendline
intercepting at 7.0
pF
Water Content (%)
Figure 3.17 WATER CONTENT VS. SOIL SUCTION
As shown in Figures 3.16 and 3.17 the slope of the trendlines vary with
different assumptions. However, based on the scatter of data, it appears that
the trendline shown in Figure 3.16 appears to be the most representative line
when compared to Figure 3.17. The trendline presented in Figure 3.16 appear
to be the most conservative of all trendlines presented above.
52


Based on the results of this study, it can be concluded that in order to properly
generate a soil moisture and soil suction relationship it is beneficial to obtain
additional data points by air drying the samples and performing additional soil
suction tests.
3.4 HEAVE PREDICTION USING MCKEENS EXPANSIVE SOIL
CLASSIFICATION METHOD
One test hole was drilled in a project site located in the Colorado Front Range
area. The test hole was drilled with a truck-mounted, continuous flight, power
auger rig using 7-inch diameter hollow stem augers. Relatively undisturbed
samples of the subsurface materials were taken with a 2-inch I.D. California
-type liner sampler. The sampler was driven into the substrata with blows
from a 140-pound hammer falling 30 inches. This procedure is similar to the
Standard Penetration Test described by ASTM Method D1586. Laboratory
testing of soil samples obtained from the subject site included standard
property tests, such as natural moisture contents, dry unit weights, grain size
analyses and liquid and plastic limits, Denver swell-consolidation and WP4-T
soil suction tests were performed on each sample, as well. Additional soil
suction tests and moisture content tests were performed on the samples after
the completion of the swell test, and after they were air dried. Laboratory tests
were performed in general accordance with applicable ASTM protocols. The
subsurface conditions at the project site consisted of a layer of overburden
soils approximately 6 feet in thickness, underlain by very hard claystone
53


bedrock extending to the test hole termination depths. Groundwater seepage
was encountered at a depth of 35 feet below the existing grades during
subsurface exploration. Actual groundwater table is anticipated to be much
deeper. Test results are presented in Table 3.5.
Table 3.5 Summary of Laboratory Test Results
Sample Location Natural Moisture Natural Percent Atterberg Limits
Test Moisture Content Dry Passing Liquid Plasticity
Hole Depth Content After Swell Density No. 200 Limit Index
No. (feet) CZo) r/o) (P 2 6 16 19.8 114.5 93 50 31
2 9 14.1 18.2 119.6 88 43 25
2 14 12.3 16.8 122.6 92 38 21
2 19 11.3 18.2 121.9 84 46 29
2 23 15.1 20.4 115.5 95 50 31
2 24 13.3 24.8 118.3 82 40 23
2 29 13.9 19.3 117.8 79 45 27
2 34 16.9 23.4 111.6 88 51 25
2 39 16.4 21.2 111 90 44 19
2 44 16.8 22.8 112.4 91 46 22
54


Table 3.5 Summary of Laboratory Test Results (cont.)
Sample Location Suction Denver Surcharge Suction uses
Test Before Swell Pressure After Classifi-
Hole Depth Swell Test Swell cation
No. (feet) (%) (psf) m
2 6 5.1 4.3 1,000 4.73 CH
2 9 4.13 5.8 1,000 3.83 CL
2 14 4.27 7 1,000 4.11 CL
2 19 4.65 4.9 1,000 4.21 CL
2 23 4.5 7.2 1,000 4.22 CH
2 24 4.57 .5.3 1,000 4.19 CL
2 29 4.57 5.9 1,000 4.23 CL
2 34 4.58 4.9 1,000 4.18 CH
2 39 4.6 3.9 1,000 4.25 CL
2 44 4.59 4.2 1,000 4.22 CL
55


Table 3.6 Summary of Laboratory Test Results
Sample Location Moisture Content Air Dned C/o) Soil Suction Mr Dried (PF)
Test Hole No. Depth (feet)
2 6 8.9 5.73
2 9 7.4 5.14
2 14 6.3 5.63
2 19 5.9 5.49
2 23 6.8 5.71
2 24 8.5 5.68
2 29 9.4 5.79
2 34 9.2 5.88
2 39 7.8 5.89
2 44 7.6 5.71
Figure 3.18 DEPTH VS. SOIL SUCTION
56


7.00
6.50
6.00
a
| 5.50
V
o
3
CO
5.00
Si 4.50
4.00
3.50
A^



5.0 10.0 15.0 20.0 25.0 30.0
Water Content (%)
Natural Moisture Content
vs. Before Swell Soil
Suction
After Swell Moisture
Content vs. After Swell
Soil Suction
a Air Dried Moisture
Content vs. Air Dried
Soil Suction
Figure 3.19 WATER CONTENT VS. SOIL SUCTION
57


Figure 3.20 SOIL MOISTURE CURVE
Using the data presented in Table 3.5, Table 3.6, Figure 3.18, 3.19, and 3.20
and McKeens equation the following results were obtained:
1 Using the data presented in Figure 3.18, the depth current depth of wetting
at the project site is estimated to be 24 feet below the existing grade. The
depth to groundwater seepage of 35 feet was not considered, since no
significant change in soil suction was observed and anticipated actual
groundwater table is much deeper than the explored depth.
58


2 Ah/Aw value can be calculated to be -12.16 using the slope of the trendline
shown in Figure 3.20. Please note that the value provided in Figure 3.20
has to be multiplied by 100 in order to convert it from percent water
content to water content by weight. Therefore,
Ch = (-0.02673) x -12.16 0.28704 = -0.062 Equation 3.1
Ah, value can be calculated using the maximum difference between the before
and after swell soil suction test results obtained from the top 24 feet. The test
results below 24 feet are not accounted for because they are below the
estimated depth of wetting. Therefore, based on the information provided in
Table 3.5, a Ah value of -0.44 was calculated using the data obtained from Test
Hole 2 at 19 feet.
At, is the layer thickness that is considered, which in this case is equal to the
depth of wetting at the project site. Converting the depth of wetting value
from feet to inches a At value of 312 inches was calculated.
f, is the lateral restraint factor, and is generally between 0.333 and 1. A
comparison tri-axial shear versus oedometer test was not performed to
determine this value. Based on the review of the available information in
regard to the determination of the lateral restraint factor, as previously outlined
in Section 2.4.2, a value of 0.5 was utilized for the lateral restraint factor.
59


s, is the reduction factor to allow for overburden pressure. An estimated
overburden pressure value of s=0.9 is calculated by assuming s0is equal to
1,000 psf.
Using the above calculated values potential surface is calculated as:
AH = -0.062 x -0.44 x 312 x 0.5 x 0.9 = 4.3 inches. (Please note that this value
will only be realized if wetting to the estimated depth actually occurs.)
3.5 GENERATING A SOIL MOISTURE CURVE IN
INTERBEDDED SUBGRADE MATERIAL
One test hole was drilled m a project site located in the Colorado Front Range
area. The test hole was drilled with a truck-mounted, continuous flight, power
auger rig using 7-inch diameter hollow stem augers. Relatively undisturbed
samples of the subsurface materials were taken with a 2-inch I.D. California
-type liner sampler. The sampler was driven into the substrata with blows
from a 140-pound hammer falling 30 inches. This procedure is similar to the
Standard Penetration Test described by ASTM Method D1586. Laboratory
testing of soil samples obtained from the subject site included standard
property tests, such as natural moisture contents, dry unit weights, grain size
analyses and liquid and plastic limits, Denver swell-consolidation and WP4-T
soil suction tests were performed on each sample, as well. Laboratory' tests
were performed in general accordance with applicable ASTM protocols.
60


The subsurface conditions at the project site consisted of a layer of overburden
soils approximately 5 feet in thickness, underlain by very hard and very dense
interbedded claystone, siltstone, and sandstone bedrock extending to the test
hole termination depth. Groundwater was not encountered during subsurface
exploration. Test results are presented in Tables 3.7 and 3.8.
Table 3.7 Summary of laboratory test results
Sample Location Natural Moisture Content r/o) Natural Dry Density (ttf) Percent Passing No. 200 Sieve Atterberg Limits
Test Hole No. Depth (feet) Liquid Limit (%) Plasticity' Index
3 4 25.9 94.6 63 26 95
3 9 24.7 107 95 101 72
3 14 40.2 79.1 73 62 24
3 19 17.2 110.3 85 76 56
3 24 39 82.1 73 75 28
3 29 39.4 79.9 41 71 28
3 34 34 85.1 59 76 27
3 39 30.7 91.3 35 58 15
3 44 35.9 82.5 71 76 29
3 49 22 109.5 95 89 59
3 54 25.6 95.4 57 59 22
61


Table 3.7 Summary of additional laboratory test results
Sample Location Suction Denver Surcharge uses
Test Hole Depth Before Swell Swell Test Pressure Classifi- cation
No. (feet) (PF) (%) (psf)
3 4 3.64 0 1,000 MH
3 9 4.49 3.8 1,000 CH
3 14 4.02 -1.4 1,000 MH
3 19 4.68 2.9 1,000 CH
3 24 4.2 0.6 1,000 MH
3 29 3.96 0 1,000 SM
3 34 4.12 -0.3 1,000 MH
3 39 3.64 0 1,000 SM
3 44 4.34 0.4 1,000 MH
3 49 4.63 5.8 1,000 CH
3 54 4.02 -0.5 1,000 MH
62


0
Figure 3.21 DEPTH VS. SOIL SUCTION
As shown in Figure 3.21, due to the heavily interbedded nature of the subgrade
soils, the soil suction versus depth profile cannot be used to estimate the depth
to constant suction on this project site. Engineering judgment must be used in
order to estimate the depth of wetting and predict potential heave.
63


4. DISCUSSION AND CONCLUSIONS
As the primary objective of this study, the effectiveness of measurement of soil
suction using the WP4-T Dew Point Potential Meter device manufactured by
Decagon Devices, Inc. was shown. An experimental study was provided to
compare the suction measurement obtained using WP4-T to the conventional
filter paper method. The repeatability of soil suction tests using the WP4-T
device was also shown by performing several tests using undisturbed and
remolded samples. These results were related to the conventional filter paper
method.
The purpose of the main objective was to provide a cost-effective, fast, and
reliable alternative to the filter paper method. As the secondary objective of
this research program, 138 samples obtained from 138 different sites within
the Colorado Front Range Area were tested for water content, liquid limit, and
soil suction. The relationships between these values were shown in two scatter
type graphs. The purpose of this secondary objective was to show the limited
range the soil suction values have when they are used in the commonly used
pF units. As the third objective of this research program three project sites
were selected where extensive soil suction tests were conducted on samples
along with several other commonly performed geotechnical engineering soil
tests. Innovative and cost effective methods were provided for generating soil
moisture relationship curves, estimating depth of wettings, and predicting
64


heave using McKeens 1992 method. In addition, the ineffectiveness of the soil
suction test data was briefly discussed in interbedded subsurface conditions
that are commonly encountered in the Colorado Front Range Area.
The main conclusions associated with the tests performed in this study are as
follows:
Measuring soil suction using the WP4-T device is a cost effective, reliable,
and time-saving alternative to conventional filter paper method.
The water content of soil samples tested approached to their plastic limit
after the completion of the Denver swell test.
Soil suction results obtained from several samples that are tested at varying
moisture contents provided a relatively linear water content and soil
suction relationship between 6.0 pF and 3.5 pF.
When subgrade conditions are relatively uniform, generating a soil
moisture curve to determine the depth to constant suction on a project site
can be an excellent tool to aid in depth of wetting calculations and heave
prediction.
McKeens 1992 heave prediction method can easily be used to predict
post-construction vertical movement in uniform subgrade conditions.
Performing soil suction tests is not an effective method to identify
potential post construction heave in non-uniform subgrade conditions,
such as heavily interbedded subgrade materials.
65


REFERENCES
Anderson, A.B.C., & Edlefsen, N. E. (1943). Thermodynamics of Soil
Moisture. Journal ofAgricultural Science, 15 (2), 31-298.
Buckingham, E. (1907). Studies on the movement of soil moisture. U.S.
Department of Agriculture bureau of Soils bulletin No. 38.
Bulut, R., Lytton, R. L., and Wray, W. K. (2001). Soil Suction Measurements by
Filter Paper. American Society of Civil Engineers Geotechnical Special Publication
No. 115 (pp. 243-261.).
Campbell, G.S., & Gardner, W.H. (1971). Psychrometric Measurement of Soil
Water Potential: Temperature and Bulk Density Effects. Soil Science Society of
America Journal, 35 (1), 8-12.
Chandler, R.J., & Gutierrez C.I. (1986). The filter paper method of Suction
Measurement. Geotechnique, 13, 265-268.
Chao, K.C., Overton D.D., and Nelson J.D. (2006). The Effects of Site
Conditions on the Predicted Time Rate of Heave. Proceedings of the Fourth
International Conference on Unsaturated Soils (pp. 2086-2097). Carefree, AZ:
American Society of Civil Engineers.
Chen, F.H. (1988). Foundations on Excpansive Soils. New York: American
Elseveir Science Publication.
Coduto, D.P. (2001), Foundation Design, Principles and Practices. Second Edition.
New Jersey: Prentice Hall.
66


Cokca E. (2001). Relationship between Methylene Blue Value, Initial Soil
Suction and Swell Percent of Expansive Soils. Turkish Journal of Engineering
and Environmental Science 26(2), 521-529.
Colby C.A., Houston S.A, Houston W.N, &, Walsh, K.D. (2006). Evaluation
of Changes to Soil Suction Resulting from Residential Development.
Proceedings of Unsaturated Soils 2006 (pp. 203 212). Carefree, Arizona:
American Society of Civil Engineers
Craig, R.F. (1998). Soil Mechanics. New York: E & FN Spon
Das, B.M. (2000). Fundamentals of Geotechnical Engineering. Pacific Grove, CA:
Brooks/Cole.
Department of Army. (1983). Technical Manual TM-5-818-7, Foundations in
Expansive Soils. Washington D.C.: Department of Army.
Gardner, D.N. (2002). Soil Suction for the Investigating Engineer. Proceedings of
the Texas Section of the American Society of Civil Engineers Fall 2002 Meeting.
Waco TX: American Society of Civil Engineers.
Gardner R. (1937). A Method for Measuring Capillary Tension of Soil
Moisture Over a Wide Moisture Range. Soil Science, 43, 277-283.
Gardner W., & Widtsoe J.A. (1921). The Movement of Soil Moisture. Soil
Science, 11(2), 215-224.
67


Gee, G., Campbell, M., Campbell G., & Campbell, J. (1992). Rapid
Measurement of Low Soil Potentials Using Water Activity Meter. Soil
Science Society of America Journal, 56, 1068-1070.
Goss, C.M., Parekh M.L., & Wiltshire, R.L. (2006). Geo-volution The Evolution of
Colorados Geological and Geotechnical Engineering Practice. Reston: American
Society of Civil Engineers.
Goss, C.M., Parekh, M.L., Malama, J.C., & Wiltshire, R.L. (2008). Geo-velopment
The Role of Geological and Geotechnical Engineering in New and Redevelopment
Projects. Reston: American Society of Civil Engineers.
Greacen E.L., Correll R.L., Cunningham R.B. Johns G.G., & Nicolls K.D.
(1987). Calibration. In: E L Greacen (ed.), Soil water assessment by the neutron
method. Commonwealth Scientific and Industrial Research Organization,
Australia (pp. 50-81).
Guan Y. (1996). The Measurement of Soil Suction. (Doctoral dissertation,
University of Saskatchewan, 1996)
Erzin Y. & Erol O. (2007). Swell Pressure Prediction By Suction Methods.
Engineering Geology, 92, 133-145.
Fawcett R.G., & Collis George, N. (1967). A Filter-Paper Method for
Determining Moisture Characteristics of Soil. Australian Journal of
Experimental Agriculture and Animal Husbandly, 7, 162-167.
Fredlund, D.G., & Rahardjo H. (1993). Soil Mechanics For Unsaturated Soils. New
York: John Wiley & Sons, Inc.
68


Houston S.L., Houston W.N., & Wagner A. (1994) Laboratory Filter Paper
Suction Measurements. Geotechnical Testing journal, 17(2), 185-194.
Houston S.L., & Wray W.K. (1995). Soil Suction Applications in geotechnical
Engineering Practice. New York: American Society of Civil Engineers.
Jefferson R. (1998). Liquid Limit and Temperature Sensitivity of Clays.
Engineering Geology, 49, 5-109
Leong, E.C., He, L., & Rahardjo, H. (2002). Factors Affecting the Filter Paper
Method for Total and Matric Suction Measurements. Geotechnical Testing
journal, 25(3), 1-12.
Likos W.J. & Lu N. (2002). Hysteresis of Capilary Cohesion in Unsaturated
Soils. Proceedings of the 15th ASCE Engineering Mechanics Conference. New York,
NY: American Society of Civil Engineers
Likos, W.J. & Lu, N. (2002). Filter Paper Technique For Measuring Total
Suction. Transportation Research Record: journal of the Transportation Research
Board, 1786, 120-128
Likos, W.J. & Lu, N. (2004). Unsaturated Soil Mechanics. New York: John Wiley
& Sons, Inc.
McCarthy, D.F. (1993). Essentials ofSoil Mechanics and Foundations. New Jersey:
Prentice Hall, Inc.
69


McKeen, R.G. (1992). A Model For Predicting Expansive Soil Behavior.
Proceedings of 7th International Conference on Expansive Soils (vol. 1, pp. 1 -6).
Dallas, TX: Texas Tech University Press.
McKeen, R.G., & Nielsen, J.P. (1978). Characterization of Expansive Soils For
Airport Pavement Design, Report No. DOT/FAA-RD-78-59, U.S.
Department of Transportation, Federal Aviation administration.
McQueen, I.S., & Miller R.F. (1968). Calibration and Evaluation of Wide
Range Gravimetric Method for Measuring Soil Moisture Stress. Soil Science,
10(3), 521-527.
Miller D.J. and Nelson, J.D. (1992), Expansive Soils. Problems and Practice in
Foundation and Pavement Engineering. New York: John Wiley & Sons, Inc.
Nelson, J. D., Overton, D. D., & Chao, K. C. (2003). Design of FoundaUons
for Light Structures on Expansive Soils. California Geotechnical Engineers
Association 2002-2003 Annual Conference. Carmel, CA
Noe, D.C., Jochim C.L., & Rogers W.P. (1997). A Guide to Swelling Soils for
Colorado Homebuyers and Homeowners. Colorado Geological Survey, Special
Publication 43.
Patrick, P.K., Olsen, H.W., & Higgins, J.D. (2007). Comparison of Chilled-
mirror Measurements and Filter Paper Estimates of Total Soil Suction.
Geotechnical Testing Journal, 30(5)
Perko, H. A., Thompson, R. W., &Nelson, J.D. (2000). Suction Compression
Index Based on CLOD Test Results. Geo-Denver 2000, 393-408.
70


Petty M., and Bryant J. (2002). Evaluation and Use of Decagon WP4
Dewpoint Potentiameter. Proceedings of the Texas Section of the American Society
oj Civil Engineers Tall 2002 Meeting. Waco, TX: American Society of Civil
Engineers
Petty M., & Jiang C. (2007). Round-Robin Testing of a Dewpoint
Potentiameter Versus Filter Paper to Determine Total Suction. Advances in
Measurement and Modeling of Soil Behavior (GSP 173) Part of Geo Denver 2007:
New Peaks in Geotechnics Proceedings of Sessions of Geo-Denver 2007. Denver,
CO: American Society of Civil Engineers.
Richards, L.A. (1928). The Usefulness of Capillary Potential to Soil Moisture
and Plant Investigators. Journal of Agricultural Research, 37, 719-742.
Richards, S.J. (1965). Soil Suction Measurements with Tensiometers. Methods of
Soil Analysis Agronomy (Black CA., Evans D.D., White J.E, EnsmingerEE.,
and Clark F.E. ed.), 9, 153-163.
Schofield, R.K. (1935). The pF of the water in soil. Third International Congress on
Soil Science Col. 2 (pp. 3748).
Vipulanandan C., Addison M., & Hasen M. (2001). Expansive Clay Soils and
Vegetative Influence of Shallow Foundations. Proceedings of Geo-Institute
Shallow Foundation and Soil Properties Committee Sessions at the ASCE 2001 Civil
Engineering Conference. Flouston, TX: American Society of Civil Engineers
Wray, W.K. (1979). Analysis of Stiffened Slabs-On-Ground Over Expansive
Soil. Proceedings of Fourth International Conference on Expansive Soils. Denver,
CO: American Society of Civil Engineers.
71