Documente Academic
Documente Profesional
Documente Cultură
By
SANG-HO KIM
UNIVERSITY OF FLORIDA
2003
Copyright 2003
by
Sang-Ho Kim
To My Mother and My Wife
ACKNOWLEDGMENTS
I thank God, at first, with all my heart. Without his leading, I would not be here
right now. During my one year and eight months at the University of Florida, I have
learned a lot of things and met many people. This period has been the most memorable
I would like to express my appreciation for all the professors (Dr. Michael C.
McVay, Dr. Frank C. Townsend, Dr. David Bloomquist, Dr. Paul Bullock, Dr. John L.
Davidson, and Dr. Bjorn Birgisson) in this department for giving me the opportunity to
study here. In addition, I would like to extend a special thank to Professor Michael
McVay for providing me the opportunity and for assisting me throughout the pursuit of
my degree. He taught me many geotechnical concepts and how to think throughout our
I would also like to thank all of my friends, Lila, Badri, Zhihong, Landy, Evelio,
Minh, Erkan, Carlos, Thai, Scott, and Dinh, in our geotechnical field in civil engineering.
It has been a pleasure studying with them and getting to know one another. In addition, I
want to thank Sang-Min Lee, who helped me with the statistical aspects of this research,
for being a good friend and colleague. Also, I would also like to thank Ben Watson of
brothers, and my wife, who support me and pray for me every day. I especially appreciate
iv
my wife who has been a constant source of understanding and encouragement. We are
expecting our first baby next January and can not wait to see her or him, as well.
Since the beginning of my master’s course, many other people have helped me
and encouraged me. Even though I could not list all their names here, their help and
v
TABLE OF CONTENTS
page
ACKNOWLEDGMENTS ................................................................................................. iv
LIST OF TABLES............................................................................................................. ix
ABSTRACT..................................................................................................................... xiv
CHAPTER
1 INTRODUCTION ........................................................................................................1
1.1 General....................................................................................................................1
1.2 Purpose and Scope..................................................................................................3
2 SITE DESCRIPTIONS.................................................................................................5
2.1 General....................................................................................................................5
2.2 17th Street Bridge ...................................................................................................7
2.2.1 Site Description ............................................................................................7
2.2.2 General Soil Profile ......................................................................................7
2.3 Acosta Bridge .........................................................................................................8
2.3.1 Site Description ............................................................................................8
2.3.2 General Soil Profile ......................................................................................8
2.4 Apalachicola Bridge ...............................................................................................9
2.4.1 Site Description ............................................................................................9
2.4.2 General Soil Profile ......................................................................................9
2.5 Fuller Warren Bridge............................................................................................10
2.5.1 Site Description ..........................................................................................10
2.5.2 General Soil Profile ....................................................................................10
2.6 Gandy Bridge........................................................................................................11
2.6.1 Site Description ..........................................................................................11
2.6.2 General Soil Profile ....................................................................................11
2.7 Victory Bridge ......................................................................................................12
2.7.1 Site Description ..........................................................................................12
2.7.2 General Soil Profile ....................................................................................12
vi
3 LITERATURE REVIEW: OSTERBERG AND STATNAMIC TEST .....................13
3.1 General..................................................................................................................13
3.2 Osterberg Load Cell Test......................................................................................14
3.2.1 Summary of Osterberg Load Cell Test.......................................................14
3.2.1 Measurement of Osterberg Load Cell Test ................................................16
3.2.3 Advantage of the Osterberg Load Cell Test ...............................................19
3.2.4 Limitations of the Osterberg Load Cell Test..............................................22
3.3 Statnamic Load Test .............................................................................................23
3.3.1 Summary of Statnamic Load Test ..............................................................23
3.3.2 Advantages and Disadvantages of the Statnamic Test ...............................24
4.1 General..................................................................................................................27
4.2 Rock Cores Testing...............................................................................................27
4.2.1 Unconfined Compression Strength Test (qu) .............................................29
4.2.2 Tensile Strength Test (qt) ...........................................................................30
4.2.2.1 Brazilian test.....................................................................................30
4.2.2.2 Flexural test ......................................................................................33
4.2.2.3 Ring shear test ..................................................................................34
4.2.3 RQD (Rock Quality Designation) ..............................................................35
4.2.4 Percentage of Recovery (%).......................................................................36
4.2.5 Elastic Modulus Test (E) ............................................................................37
vii
6.3.4 Fuller Warren Bridge..................................................................................87
6.3.5 Gandy Bridge..............................................................................................91
6.3.6 Victory Bridge ............................................................................................94
6.3.7 Comparison Between Measured and Predicted Unit End Bearing.............96
6.4 Prediction of the Unit End Bearing using Boring of All Site ...............................97
6.4.1 Monte Carlo Simulation .............................................................................97
6.4.2 Number of Monte Carlo Simulations .........................................................98
6.4.3 Comparison Between Measured and Predicted Unit End Bearing...........101
APPENDIX
LIST OF REFERENCES.................................................................................................151
viii
LIST OF TABLES
Table page
6.1 Measured Unit End Bearing from Osterberg Load Test in Florida..........................69
6.6 Prediction of Unit End Bearing Using the Nearest Boring(17th bridge) ..................79
6.8 Prediction of Unit End Bearing Using the Nearest Boring (Acosta) .......................82
ix
6.12 Prediction of Unit End Bearing Using the Nearest Boring (Apalachicola) ...............85
6.15 Prediction of Unit End Bearing Using the Nearest Boring (Fuller Warren)..............89
6.18 Prediction of Unit End Bearing Using the Nearest Boring(Gandy) ..........................92
6.20 Prediction of Unit End Bearing Using the Nearest Boring (Victory)........................95
6.21 Measured and Predicted Unit End Bearing Using the Nearest Boring......................98
7.3 λmean, λst.d, and COV Based on Predicted Values and Measured Values ...............111
8.2 LRFD Phi Factors, Probability of Failure (Pf) and FS Based on Reliability, β
(Mean Method for All Site)....................................................................................117
x
LIST OF FIGURES
Figure page
3.1 Osterberg Setup When the Osterberg Cell is Installed Above the Tip.....................17
3.4 Osterberg Cell Load-Movement Curve for Shaft 10-2 in Victory Bridge ...............21
4.2 Bend Saw Blade to Cut the Core Sample in FDOT Transportation Lab .................28
4.3 Diamond Saw to Cut the Core Sample in FDOT Transportation Lab .....................29
4.8 Stress vs. Strain Curve for Calculating Initial Young’s Modulus............................39
4.9 Stress vs. Strain Curve for Tangent and Secant Modulus ........................................39
5.4 Correlation Chart between the qu and Ei for 17th Street Causeway Site..................47
xi
5.6 Frequency Distribution for Ei of Acosta Bridge Site ...............................................49
5.7 Correlation Chart between the qu and Ei for Acosta Bridge Site ............................49
5.11 Correlation Chart between the qu and Ei for Apalachicola Bridge Site ....................52
5.15 Correlation Chart between the qu and Ei for Fuller Warren Bridge Site...................55
5.19 Correlation Chart between the qu and Ei for Gandy Bridge Site ..............................58
5.23 Correlation Chart between the qu and Ei for Victory Bridge Site.............................61
6.3 17th Street LTSO 3 – Comparison between Predicted vs. Measured FDOT
Failure End Bearing .................................................................................................76
xii
6.4 Predicted vs. Measured End Bearing (LTSO 3 in 17th Street Causeway)................79
6.5 Predicted vs. Measured End Bearing (LTSO 4 in 17th Street Causeway)................80
6.7 Predicted vs. Measured End Bearing (46-11A in Apalachicola Bridge) .................86
6.8 Predicted vs. Measured End Bearing (62-5 in Apalachicola Bridge) ......................86
6.9 Predicted vs. Measured End Bearing (69-7 in Apalachicola Bridge) ......................87
6.10 Predicted vs. Measured End Bearing (LT-3a in Fuller Warren Bridge)....................90
6.11 Predicted vs. Measured End Bearing (LT 4 in Fuller Warren Bridge)......................90
6.12 Predicted vs. Measured End Bearing (52-4 in Gandy Bridge) ..................................93
6.13 Predicted vs. Measured End Bearing (91-4 in Gandy Bridge) ..................................93
6.17 Sample and Random Field Population for qt in Apalachicola Site. ........................100
6.18 Sample and Random Field Population for Ei in Apalachicola Site. ........................100
6.19 Sample and Random Field Population for RQD in Apalachicola Site ....................101
6.20 Comparison of Measured and Predicted End Bearing for Mean Method................103
8.2 Comparison of Measured and Predicted End Bearing (All Site Method)..............116
xiii
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
By
Sang-Ho Kim
December 2003
This research is conducted as part of a project, Static and Dynamic Field Testing of
The main purpose of this thesis is to predict unit end bearing of drilled shafts using
the O’Neill method employed as FDOT standard practice in Florida bridge sites.
Predicted unit end bearing values were compared with measured unit end bearing values
of the shafts obtained from the Osterberg Load Test. In the O’Neill method, many rock
properties should be required, and the properties of compressibility of the rock (i.e.,
Young’s Modulus, E and Unconfined compressive strength, qu) is acted for prediction of
As the way of predicting end bearing for the shafts, the nearest boring method and
mean method of all sites are used based on laboratory data of Florida limestone. These
two methods are to consider rock properties data near shaft and all site.
xiv
Based on a probability approach, Load Resistance Factor Design as approved by
AASHTO could be determined by the reliability. The same probability of failure related
LRFD factor affects the factor of safety. These LRFD factors or factor of safety can be
xv
CHAPTER 1
INTRODUCTION
1.1 General
contracted with the University of Florida to evaluate their current load testing approach
for drilled shafts. The drilled shafts in Florida are playing an important role in deep
foundation and are a favorable foundation solution in soil and /or rock where an
excavated hole will remain open and high capacities are desired.
Limestone is a common material that provides support for most of the area's
significant civil structures in Florida. Generally larger structures are supported on deep
foundations. Drilled shafts and driven piles for deep foundations are installed to combine
frictional and end bearing support in the limestone. Vertical loads are resisted by both the
system was regarded as too complex to understand and to model entirely theoretically.
regard, drilled shafts were considered to be less reliable than others because of the
uncertainty of the effects of construction on the actual service behavior, and the limited
inexpensive load test procedures. Even if there are only occasional failures, they highlight
1
2
the variables and unknowns present when working underground, particularly in water-
bearing and potentially caving soils. This results in a lower risk tolerance for a single or
double shaft supported pier compared with multiple pile supported foundations.
Currently, design methods for drilled shafts in soil or competent rock are
reasonably well founded; however, comparatively little effort has been expended to
and marls. In September 1981, FDOT and the University of Florida developed an
equation to be used as a recommended criterion for determining the skin friction of the
drilled shafts made in Florida limestone. The equation is recommending for drilled shafts
in Florida.
The strength of Florida limestone varies according to site and type of limestone.
This variability makes it difficult to predict the end bearing of drilled shaft. Generally, a
load test has been used to measure the end bearing of the shaft for deep foundation, but
prediction of the end bearing of the shaft for foundation design has not been introduced
as much as prediction of skin friction. In some cases, end bearing of the shaft is ignored
in favor of the total shaft capacity, because of exceeding service load displacement
criteria to mobilize the tip and engineer’s concept. Evident from the Osterberg tip results,
significant end bearing (>145 tsf) has been generated on drilled shafts founded in Florida
limestone, however, the end bearing of a shaft should be considered in the foundation
Based on soil/rock properties and the analysis of the properties, some predicted
methods for shafts’ ultimate or failure tip capacity of a shaft (i.e., Kulhawy, Tomlinson,
etc) have been suggested. Yet, very few methods identify tip resistance vs. tip
3
displacement. The O’Neill method, which identifies both, is the FHWA approach
employed in FB-Pier. In this thesis, the prediction of end bearing for the shafts on Florida
limestone was calculated using this O’Neill method. For the prediction of end bearing,
many rock properties of each site were used when the O’Neill method was applied.
Developed for intermediate Geomaterials, the O’Neill method is dependent on the rocks’
Using the O’Neill method, end bearing values derived from the load test of several
Florida bridge sites were compared with the values predicted from analysis of laboratory
data of Florida limestone. In predicting end bearing for the shaft, one of two approaches
may be used: 1) based on nearest boring, determine measured vs. predicted end bearing
of each shaft; or 2) based on random selection, compare the mean of measured vs.
Recently, the FDOT has implemented load resistance factor design (LRFD) for
drilled shafts. That is, the LRFD specifications as approved by AASHTO recommend the
use of load factors to account for uncertainty in the loads, and resistance factors to
account for uncertainty in the materials. Based on a probability approach, in this research,
LRFD resistance factors (φ), factor of safety (FS) for modeling tip resistance can be
The scope of work for end bearing of drilled shaft includes field studies, laboratory
data for rock properties and load test result in Florida. The various components and
analysis of the laboratory data are summarized in this report and in the Appendix. By
4
examining many rock properties, these studies were conducted to identify the limestone
at the bridge sites and evaluate and analyze their strength characteristics.
Some rock core test for this research was conducted at the FDOT transportation lab
in Gainesville and all data for rock core were obtained by design plans, Osterberg &
Statnamic load test reports, and Geotechnical report from FDOT. The data and
information for a total of 6 bridge projects (17th Causeway, Acosta, Apalachicola, Fuller
Warren, Gandy, Victory) of a total of 11 bridge projects were used for the predicted end
bearing.
It is the intent of this thesis to compare and analyze the results of the end bearing
for drilled shaft performed the Osterberg load test at the sites of projects selected with
predicted end bearing using the O’Neill method (FB-Pier method) and to suggest useful
guidelines for the drilled shaft design for Florida limestone through the LRFD factor or
• Introduce Osterberg and Statnamic axial load tests for Florida limestone
• Organize and Analyze insitu and laboratory testing (i.e. unconfined compressive
test, spilt tension test, RQD, Recovery, Young’s Modulus) on the soil and
limestone around the bridges in Florida
• Predict and analyze end bearing for drilled shaft using the O’Neill method (FB-Pier
method) and compare to measured end bearing of Osterberg load test
• Develop design guidelines of drilled shaft for Florida limestone using LRFD φ
factors or factor of safety (FS).
For the load test reductions and rock core data in Florida, the research works
Comparison between Osterberg and Statnamic Load Tests by Michael Kim (University
of Florida, 2001) and Geotechnical Predictions for the Design of Drilled Shaft in Florida
2.1 General
Much information and data (Osterberg and Statnamic load test reports,
Geotechnical reports for the sites, and Construction design plans) were used for this
whole project, and in order to meet the goals and scope of this research, some bridges’
information and data were used. The data and information for a total of 6 bridge sites
(17th Causeway, Acosta, Apalachicola, Fuller Warren, Gandy, Victory) of these 11 bridge
The location of each project is shown in Figure 2.1. Osterberg and Statnamic tests
are presented on the map. All the projects are specifically located in coastal areas of
5
6
Florida, and all the shafts are constructed on Florida limestone. A total of 27 Osterberg
load tests was performed and a total of 12 Statnamic load tests in whole project sites was
performed as shown in Figure 2.1. A description of each site along with a general soil
Victory Bridge
(Chattahoochee) Acosta Bridge
(Jacksonville)
O-cell: 5 Stat: 1
Lateral: 4 O-cell: 4 Conv: 2
Fuller Warren Bridge
Apalachicola Bridge
(Jacksonville)
(Calhoun Liberty)
O-cell: 4
O-cell: 6
Lateral: 2
Lateral: 1
Crista Bridge
Gandy Bridge (Brevard)
(Tampa)
O-cell: 3 Stat: 3 17th Causeway
Lateral: 6 (Fort Lauderdale)
O-cell: 4 Stat: 6
Hillsborough Bridge Lateral: 2
(Tampa)
O-cell: 1 Stat: 2 MacArthur Bridge
(Miami)
Conv: 1
Venetian Bridge
(Miami)
West 47 Bridge
(Miami)
Note:
Stat: Statnamic Load Test (number of test: 12)
O-cell: Osterberg Load Test (number test: 27)
Conv: Conventional Load Test (number of test 3)
This is a bascule replacement bridge for the old movable bridge on the S.E. 17th
Street Causeway over the intracoastal waterway in Fort Lauderdale, located in Broward
County. The new bascule bridge provides about 16.76 meters of clearance over the
The construction started on the west end at Station 28+73, which is approximately
127 meters west of the intersection between Eisenhower Boulevard and the S.E. 17th
Street Causeway. The end of construction was on the east at Station 41+60, which is
approximately 540 meters east of the intersection between S.E. 23rd Avenue and the S.E.
The general topography on the west end of the S.E. 17th Street project was level
outside of the area of the embankments, i.e. elevation in the range of +1.5 to +2.0 meters
(NGVD). The project alignment from the west end to the intracoastal waterway, the
the vicinity of the west abutment of the bridge. The average elevation of the ground
surface on the N.W. and S.W. frontage roads ranges from approximately +1.5 to +2.0
meters (NGVD).
Navigation Channel, the elevation of the bottom of the bay drops smoothly to elevations
as low as 4.6 meters (NGVD). From the Navigation Channel, the elevation of the bottom
8
of the bay increases smoothly until the ground surface is encountered on the east side of
the Intra-coastal Waterway. The average elevation of the ground surface on the N.E. and
S.E. Frontage Roads ranges from approximately +1.5 to +2.0 meters (NGVD). The
elevation of the project alignment on S.E. 17th Street on the east approach embankment
alignment proceeds to the east, the ground surface elevation drops to approximate
elevations between +1.5 and +2.0 meters around stations 40+50 to 41+00. At this point,
the ground surface elevation starts to increase again as the project alignment approaches
The newly erected 4-lane Acosta Bridge crosses the St. Johns River in the
downtown district of Jacksonville, FL. It replaces a 2-lane lift span bridge (completed in
1921) and carries the Automated Skyway Express (ASE), a light-rail people mover, for
The average elevation of the ground surface of the project ranges from +3.0 to
+15.0 feet (NGVD). In the shallow areas of the river crossing (i.e. less than 30’), there is
a 2’ to 10’ thick layer of sand. This thin upper sand layer is very susceptible to scour.
The upper sand layer is underlain by a layer of limestone varying in thickness from 10’ to
20' thick, which is underlain by overconsolidated sandy marl. The limestone layer is
The Florida Department of Transportation widened State Road (SR) 20, crossing
the Apalachicola River between the towns of Bristol and Blountstown in Calhoun
County, by constructing a new 2-lane bridge parallel to the existing 2-lane structure. The
existing steel truss bridge was constructed in the 1930's and was recently designated as an
bridge, and renovating the old bridge. The final bridge consists of two lanes traveling
east-west (new bridge) and two lanes traveling west-east (renovated old bridge).
Each of the structures consists of a trestle portion crossing the surrounding flood
plain as well as a high-level portion spanning the river itself. The trestle portion of the
new structure is 4,464 feet long while the approaches and main span comprise 3,890 feet,
resulting in a total structure length of 8,362 feet. The main span provides a vertical
clearance of 55 feet from the normal high water level of the river. The river is about 700
The new bridge alignment runs approximately parallel to the existing structure just
to its south. Natural ground surface elevations in the flood plain generally range from
about elevation +41 feet to +47 feet on the West Side of the river and from elevation +44
feet to +48 feet on the East Side of the river. Mud line elevations at pier locations within
the river range from about + 17 feet to + 18 feet. According to the project plans, the
mean low river water elevation is + 32. 0 feet and the normal high river water elevation is
+46.5 feet.
10
The subsurface stratigraphy consists of soft to very stiff sandy clays, sandy silts,
some clayey sands of 10 to 20 feet in thickness underlain by sands, and silty clayey sands
ranging in density from loose to dense with thickness from a few feet to a maximum of
30 feet. Beneath the sands, calcareous silts, clays, sands and gravels, with layers of inter
bedded limestone, generally extend from about elevation zero feet to about elevation -50
feet to -60 feet. The calcareous material is limestone that is weathered to varying
degrees. While the upper 10 to 15 feet of the material generally ranges from stiff to
medium dense, it appears to become very dense to hard with increasing depth. At
approximately elevation -50 feet to -60 feet, very well cemented calcareous clayey silt
with sand is encountered that extends to elevation -65 feet to -75 feet. This material is
generally underlain by very hard limestone that extends to the maximum depth of 135
The new Fuller Warren Bridge replaces the old Gilmore Street Bridge in
Jacksonville, Florida. The new bridge spans Interstate Highway 95 (I-95) across the St.
Johns River in downtown Jacksonville. The old bridge was a four-lane concrete structure
with steel, with a drawbridge bascule extending across the channel. The new concrete
high span bridge has a total of eight travel lanes and was constructed parallel to the old
The average elevation of the ground surface for this project ranges from +4.0 to
+20 feet. Overburdened soils are generally encountered from these surface elevations
11
down to the limestone formation at elevations -12 to -27 feet. Overburdened soils
generally consist of very loose to very dense fine sands with layers of clayey fine sands
and/or layers of very soft clay. A variably cemented sandy limestone formation is
encountered between elevations of -12 to -45 feet (MSL). The limestone formation is
The Gandy Bridge consists of two double lane structures across Old Tampa Bay
between Pinellas County to the west and Hillsborough County to the east in west central
Florida. The new bridge replaces the westbound structure of the existing Gandy Bridge
across Old Tampa Bay. The age, deterioration, and other factors of the old bridge
The average elevation of the ground surface of the project ranges from +0.0 to -22
feet. The surface soils consist of approximately 45 feet of fine shelly sand and silt.
Underlying the sands and silts are highly weathered limestone. The limestone is
encountered at depths varying from 58 to 65 feet below existing grade. The elevation of
the top of the limestone varies from approximately -4 feet (NGVD) to -53 feet (NGVD)
along the axis of the bridge across the bay. Four-inch rock cores were taken in selected
borings. The recovered rock samples are generally tan white shelly calcareous slightly
phosphatic limestone, which contains chert fragments. Much of the limestone has been
weathered and, due to solution processes, has pockets of silts and clays within the matrix.
12
The Victory Bridge crosses the flood plain of the Apalachicola River about one
mile west of Chattahoochee and is on U.S. 90. The Jim Woodruff dam is located
approximately 0.6 miles north (upstream) of the bridge. The original bridge was
completed shortly after the end of World War I and is supported by steel H-piles. The
bridge was recently designated as an historic structure to prevent its demolition. The new
bridge is located approximately 50 feet south of the old bridge and is supported by drilled
shafts.
The soil profile at the Victory Bridge is quite variable, ranging from silt and clay to
sand with gravel over limestone. The ground surface occurs at an elevation of +48 to +58
feet, with weathered limestone at the surface at some locations. Overburdened soils
generally consist of very loose to very dense fine sands and clayey fine sands with some
elevations of +40 to –20 feet. The limestone formation is typically 10 to 50 feet thick.
CHAPTER 3
LITERATURE REVIEW: OSTERBERG AND STATNAMIC TEST
3.1 General
A Load test is a method used to determine the amount of weight that can be carried
by a structural unit. Load tests can be performed on individual units, groups of units, or
an entire foundation. The generally accepted practice is to load test one or more
individual units of the foundation. The test results of individual piles or drilled shafts are
then used within accepted engineering boundaries for the design of groups and entire
foundations.
develop more site-specific knowledge regarding pile types and the soils into which they
are driven. Load tests can be a part of the construction contract for a particular project. In
this case, load tests are used to ascertain that a minimum capacity or minimum necessary
tip elevations can be obtained utilizing the actual pile type and equipment furnished by a
contractor for a project. Load tests are often used as a demonstrative tool for sales of
foundation products, equipment, or services. Contractors may use load tests to help
develop and prove new foundation products, be they structural units or installation
equipment. In the case of drilled shafts for load tests, strain gauges are generally installed
at equal spacing along the shaft to assess skin and tip resistances in the shaft.
It is critical that the test shaft be founded in the same formation and by the same
construction procedures as the production shafts. Generally more than one load test is
13
14
plunging of the drilled shaft, or 2) a gross settlement, uplift or lateral deflection of 1/30 of
Until recently, the only feasible way of performing a compressive load test on a
drilled shaft was the conventional method, which requires large reaction frames, but the
conventional method also has a limit capacity as the capacity of 1500 tons with
significant installation and testing time (Justason et al., 1998). Recently, two new
methods for conducting drilled shaft load testing have been developed that do not require
reaction systems. These methods have a higher capacity (about 3000 to 6000 tons) and
shorter testing time than the conventional load test. These are the Osterberg and
Statnamic testing methods. Osterberg and Statnamic tests are ordinarily less expensive
The Osterberg (O-cell) method makes it possible to separate the side shear
resistance (skin friction) from the end bearing and to separate load-deflection curves for
skin friction and end bearing, as well as determining at each increment of load how much
of the total load is in skin friction and how much is in end bearing. The O-cell is placed
on or near the bottom of a drilled shaft and after the concrete is poured and cured, an
equal upward and downward pressure is applied. From the test, the load-upward
deflection curve in side shear and the load-downward deflection curve in end bearing are
drawn. The O-cell developed by Jorj Osterberg (1989) is basically a hydraulic jack that is
cast into a shaft. Osterberg load tests have been performed in many countries.
15
Because the deflection of the shaft is small when the ultimate load is reached in
side shear compared to the deflection at which the ultimate end bearing is reached, the
majority of the load is taken in shear as the load is applied and shifts to end bearing as the
load is increased. For many cases, where the soil profile is relatively uniform throughout
the depth of the shaft, the great majority of the load, (70~80%), is taken in side shear
at the working load. If the shaft penetrates through a relatively soft soil with end bearing
of a much harder soil, the side shear still develops first, but when the working load is
reached, the end bearing may take the majority of the load. In rock sockets, the side shear
also develops faster than the end bearing as the load is increased. Based on tests in rock
sockets of different types of rock and in both strong and weak rocks, the side shear is
found in most cases to be much larger than generally assumed by the design engineer. In
relatively few cases for shafts in soil and in rock sockets, side shear is surprisingly
smaller than expected. The O-cell load-deflection curves make it possible to determine
approximately how much disturbance is on the shaft bottom and how it influences the
In the Osterberg test, end bearing provides a reaction for the skin friction, and
skin friction provides a reaction for the end bearing. This unique mechanism makes the
placement of the cell critical. If the cell is placed too high (as shown Figure 3.1), the
shaft would most likely fail in skin friction on the shaft above the O-cell. If the O-cell is
placed too deep in the shaft, the portion of shaft below the cell will likewise fail too soon.
If either occurs too soon, the information about the other is incomplete. As a
consequence, it is not easy to get both the ultimate side and tip resistances with just one
Osterberg cell. If only the ultimate tip resistance is desired, the cell should be installed at
16
the bottom of the shaft. On the other hand, if the ultimate side resistance is needed, the
Osterberg tests are typically performed in accordance with ASTM D1143 (Quick
Load Test Procedures). The loads are applied during each stage in increments of 5% of
the estimated maximum applied load. The shafts are unloaded in increments of about
Figure 3.2: used in Apalachicola Bridge and Fuller Warren Bridge), that is capable of
fully mobilizing both side and tip resistances. Nine thousand tons of combined side and
base resistances have been achieved with this arrangement. Obviously, this configuration
In the United States, electronic gages have normally been used to measure the
movement during the Osterberg test. The basic instrumentation schematic for the test on a
drilled shaft is shown in Figure 3.3. Figure 3.3 shows a hydraulic jack like device placed
on or near the bottom of a drilled shaft. After concrete is poured and cured, hydraulic
pressure is applied to the Osterberg cell which exerts an equal upward and downward
force on the shaft. The force is determined by recording the pressure and converting it
into force from a pre-determined calibration curve. The downward force is resisted at all
times by skin friction and therefore no overhead load frame with hold down piles or a
dead weight reaction is needed. The total opening, or extension, of the Osterberg cell is
lower ends of which are attached to the bottom plate of the Osterberg cell. The upward
movement of the top of the Osterberg cell is measured directly from a pair of steel
17
Dial Gages
Reference Beam
Tell-tale to
bottom cell
Pressure Source
Concrete
Hydraulic
Supply Line
Skin Friction
Osterberg Cell
(Expands)
Figure 3.1 Osterberg Setup When the Osterberg Cell is Installed Above the Tip
18
Upper O-cell
Lower O-cell
Physical Arrangement
Side Shear
Failure Upper portion side
resistance is obtained.
Locked Active
Bearing resistance
is obtained.
Active Locked
Bearing Failure
Step 1 Step 2
Open Active
Side Shear Side Shear Failure
Failure (Reverse Direction)
Active Open
Step 3 Step 4
Figure 3.2 Multi-level Osterberg Testing Setup (After Reese and O’Neill, 1999)
19
telltales which extend to the top of the Osterberg cell as shown in the first and second
strain gages of Figure 3.3 and is recorded on a data logger from which the movements
can be plotted or shown directly on the screen. These telltales also allow the
measurement of the compression of the test shaft. Subtracting the upward movement of
the top of the Osterberg cell from the total extension of the Osterberg cell determined by
the LVWDTs provides the downward movement of the bottom plate. The upward
movement of the top of the test shaft is measured with digital gages mounted on a
reference beam and set over the top of the test shaft as shown in the third and fourth
Movements will continue until either the ultimate in skin friction, the ultimate in
end bearing, or the capacity of the device is reached, whichever occurs first. When this
occurs, the test is completed. For the largest capacity size Osterberg cell, three feet
diameter cell, the maximum force which can be applied is 3,000 tons up and 3,000 tons
down. For large diameter shafts, three Osterberg cells have been used, capable of exerting
a total of 18,000 tons for upward and downward. Figure 3.4 shows the load-movement
curves for skin friction and end bearing at Victory Bridge in the Osterberg load cell test.
First, the Osterberg test is an economic load test compared to a conventional static
test. Cost savings are realized through reduced construction time and capital outlay for a
test, no top-of-pile rection equipment requirements, and less test design effort. Osterberg
tests typically cost 1/3 to 2/3 as much as conventional tests. The comparative cost reduces
Second, the Osterberg test can be used to carry out very high load capacity for a
drilled shaft. Osterberg test have been tested to equivalent conventional tests in many
20
PressureTransducer
AttachedToHydraulic
ReturnLine
1. 2.
Data
Logger Digital
Interface
Air Driven
HydraulicPump
withPressureGuage
SkinFriction SkinFriction
TelltaleCasing
ReinforcingSteel
HydraulicSupplyLine
LVWDT
Osterberg
Cell
EndBearing
Figure 3.3 Schematic of Osterberg Load Test (after Reese and O'Neil, 1999)
21
Load vs. Movement Curves for Skin Friction and End Beaaring
2.0
1.5
Skin Friction
1.0
0.5
Movement (inches)
0.0
-0.5
-1.0
-2.0
-2.5
-3.0
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Load (tons)
Figure 3.4 Osterberg Cell Load-Movement Curve for Shaft 10-2 in Victory Bridge
places in the United States. In Florida a drilled shaft has been tested with a group of
Osterberg cells that have the capacity to reach a load around 16,000 tons. Very high
Third, the Osterberg test automatically separates the skin friction and end bearing
Fourth, the Osterberg test improves safety. The test energy lies deeply buried and
Fifth, rock sockets can be usefully used in Osterberg tests. The Osterberg cell
places its large test load capability directly at the bottom of the socket, and also gives an
difficulty testing rock sockets because of limited reaction capacity and load shedding in
the soil above the rocket. Instrumentation interpretation problems often preclude any
Sixth, work area is reduced. The work area required to perform an Osterberg test,
both overhead and laterally, is much smaller than the area required by a conventional
load test.
Seventh, the Osterberg test can be performed over water or on a batter pose,
1994, researchers took a test in which five 460 mm square prestressed concrete piles were
driven to take advantage of the Osterberg test cell’s capability of sequential testing. The
purpose of this ongoing research program is to assess setup or “aging” effects on heavily
instrumented piles driven in a variety of soil types. That is to say, the Osterberg test can
be performed to carry out long term stage testing with minimal effort and equipment.
First, the Osterberg test cell must be installed prior to construction or driving in the
Second, in the Osterberg test, single cell tests generally fail by either mobilizing the
full skin friction or end bearing which limits the information on the other component. To
Third, the Osterberg cell is normally considered expendable and not recovered after
the test is completed. However, grouting the cell after completion of the test allows using
the tested drilled shaft or driven pile as a load carrying part of the foundation.
23
Fourth, the Osterberg test is not suitable for some types of piles. For example, the
The Statnamic load test has been developed to meet the construction industry's
demand for an accurate and cost-effective method of determining the load bearing
Foundation Equipment of Canada and TNO Building & Construction Research of the
Netherlands, the Statnamic test can be used on any pile type with minimum pile
Loading is perfectly axial and the relatively slow application and release of
compressive forces eliminates tensile stresses, compressing the pile and the soil as a
high capacity piles to failure in both skin and end bearing resistances simultaneously.
Statnamic devices have been constructed that are capable of applying loads of
approximately 4000 tons. The cost of a Statnamic test is usually similar to that of an
The principle of the Statnamic test is shown in Figure 3.5. Dead weights (reaction
masses) are placed upon the surface of the test shaft. Small propellants and a load cell
are placed underneath the dead weights. Solid fuel pellets in a combustion chamber
develop large pressures, which act upward against the shaft and dead weights (reaction
masses). The pressure acts against the top of the shaft, inducing a load-displacement
response that is measured with laser and load cell devices. The pre-determined load is
24
controlled by the size of the reaction mass and propellants. The duration of the applied
load is typically 120 milliseconds. Pile/shaft acceleration and velocity are typically on
the order of 1g and 1m/s respectively. Displacement is monitored directly using a laser
datum and an integrated receiver located at the center axis of the pile/shaft. In addition,
displacement may be calculated by integrating the acceleration measured at the top of the
Since there are some dynamic forces (i.e. damping and inertia), some analysis is
necessary. Currently, the Unloading Point Method (UPM) (Middendorp et al., 1992) is
the standard tool for assessing the damping inertial forces and determining the static
The Statnamic load test provides the following advantages when compared to
First, propellants are a safe and reliable way to produce a predetermined test load of
humidity. Second, the Statnamic test can be performed on a drilled shaft for which a
loading test was not originally planned, because the Statnamic requires no equipment to
be cast in the shaft. Third, the device for the Statnamic test can be reused on multiple
piles or shafts. Fourth, the Statnamic test produces the load-movement results
immediately on site. Fifth, the Statnamic test has little or no effect on the integrity of the
shaft (non-destructive). Sixth, the Statnamic test is a top-down test simulating a real load
The main disadvantage of the Statnamic test is its dynamic nature and the need to
25
Reaction
Masses
Laser sight
Load Cell / Fuel (movement)
Skin Friction
End Bearing
Load
Statnamic
Static
Movem ent
assess the dynamic forces (inertia and damping) that are developed during the test. The
dynamic forces can be computed using the Unloading Point Method (UPM) (Middendorp
et al., 1992). AFT (Applied Foundation Testing, Inc.) has recently developed the
instrumentation placed along the side of the pile/shaft. The derived static loads presented
in this thesis were calculated using the UPM method since this method was the only
4.1 General
The laboratory testing of undisturbed samples of soil and rock is an integral part of
any serious soils investigation and is used by the Geotechnical engineers to obtain the
soil or rock formations. Many properties of and information about soil and rock can be
obtained through laboratory testing and engineers will determine the extent of the work to
Laboratory tests on rock are performed on small samples of intact cores. However,
the properties of insitu rock are often determined by the presence of joints, bedding
planes, etc. It is also important that the rock cores come from the zone in which the
conjunction with knowledge of the insitu characteristics of the rock mass. Some of the
more common laboratory tests are: unconfined and confined compression tests, shear
Other test configurations are preferred for special applications and a great variety of
procedures has been investigated. Figures 4.1, 4.2, and 4.3 show rock core samples from
a boring in a field and the rock sample preparation equipment required to prepare
27
28
Figure 4.2 Bend Saw Blade to Cut the Core Sample in FDOT Transportation Lab
29
Figure 4.3 Diamond Saw to Cut the Core Sample in FDOT Transportation Lab
The unconfined compression test is one of the most basic parameters of rock
strength, and the most common determination performed for boreability predictions.
Unconfined compression (Figure 4.4) is the most frequently used strength test for rocks,
yet it is not simple to perform properly and results can vary by a factor of more than two,
as procedures are varied. The test specimen should be a rock cylinder of length-to-width
ratio in the range of 2 to 2.5 with flat, smooth, and parallel ends cut perpendicularly to the
cylinder axis. If the sample length to diameter ratio is greater or less than 2, ASTM
and by Bieniawski and Bernede (1979). Loading data and other test parameters are
recorded with a computer based data acquisition system, and the data is subsequently
reduced and analyzed with a customized spreadsheet program. Capping of the ends with
30
However, introduction of Teflon pads to reduce friction between the ends and the
loading surfaces can cause outward extrusion forces producing a premature splitting
failure, especially in the harder rocks. When mine pillars are studied, it is sometimes
preferable to machine the compression specimen from a large cylinder to achieve loading
through the rock of the upper and lower regions into the more slender central region.
However, in the standard laboratory compression test, cores obtained during site
exploration are usually trimmed and compressed between the crosshead and platen of a
testing machine. The compressive strength (qu) is expressed as the ratio of peak load P to
initial cross-sectional area A and is calculated by dividing the maximum load at failure by
P
qu = Equation 4.1
A
P = Peak load
Indirect, or Brazilian, tensile strength is measured using NX-size core samples cut
has been found that a rock core about as long as its diameter will split along the diameter
and parallel to the cylinder axis when loaded on its side in a compression machine
31
Table 4.1 Unconfined Compressive Strength (qu) for Specimens of Representative Rocks
(Goodman, 1989)
(Figure 4.5). The reason for this can be demonstrated by examining the stress inside a
disk loaded at opposite sides of a diametric plane. In such a configuration, the horizontal
stresses perpendicular to the loaded diameter are uniform and tensile with magnitude.
Brazilian tensile strength provides a measure of rock toughness, as well as strength. The
2P
σt = Equation 4.2
π × dt
P = Peak load,
d = Cylinder diameter
It is much easier to perform this type of test than to arrange the precise alignment
The "Brazilian tensile strength" is estimated from the test result by reporting the
understood that the actual cause of failure might also reflect the action of the vertical
stress along the vertical diameter in concert with the horizontal tension. The vertical
stress is non-uniform, increasing from a compressive stress of three times σt at the center
According to the Griffith theory of failure, the critical point should to be the center
where the ratio of compression to tension is 3. With a principal stress ratio of 3, failure
ought to result from the application of the tensile stress alone, without any complication
from the simultaneous compression parallel to the eventual rupture plane. In fact, the
33
Brazilian test has been found to give a tensile strength higher than that of the direct
tension test, probably due to the effect of fissures. Short fissures weaken a direct tension
specimen more severely than they weaken a splitting tension specimen. The ratio of
Brazilian to direct tensile strength has been found to vary from unity to more than ten as
the length of pre-existing fissures grows larger (Tourenq and Denis, 1970).
A Flexural test causes failure of a rock beam by bending. Like the Brazilian test,
flexural tests can also be run on rock cores lacking machined ends. Four-point flexural
loading (Figure 4.6) is when the bottom of the core is supported on points near the ends,
and the top of the core is loaded from above at the third points. This type of loading
produces a uniform moment in the central third of the specimen and gives better
reproducibility of results than three-point loading, where the upper load is central. The
34
flexural strength or "modulus of rupture" is the maximum tensile stress on the bottom of
The flexural strength is found to be two to three times as great as the direct tensile
strength. For four-point bending of cylindrical rock specimens, with loads applied at L/3
from each end and reactions at the ends, the modulus of rupture (MR) is:
where Pmax is the maximum load, L is the length between load reactions on the
The ring shear test (Figure 4.7) provides a relatively simple method to test intact rock
tests, core specimens for the ring shear test do not require perfectly square and smooth
ends. As with the triaxial test, the results permit an appreciation of the rate of increase of
strength with confining pressure. The latter is provided by the load parallel to
the axis of the core. Two sets of complex fracture surfaces form along the two planes of
If P is the peak load on the plunger, the peak shear stress (τP) is called the “shear
The RQD (rock quality designation) was proposed by Deere (1963) as an index of rock
fracturing. The RQD is the percentage of the length drilled that yields core in pieces
longer than 100 mm (or, more generally, twice the diameter of the core). Most of the
methods advanced for engineering classification of rocks use the RQD as one of the
(1974).
The RQD can be computed from each core run and plotted in a strip alongside the
Boring log. Pieces broken on new fractures during core drilling should be reassembled
for this measurement. Lengths are measured along the centerline of the core. Equation 6
25-50 Poor
50-75 Fair
75-90 Good
90-100 Excellent
index of rock fracturing. The % REC is the ratio of the length of core obtained to the
distance drilled. The % REC can be computed from each core run and plotted in a strip
alongside the Boring log. The FDOT uses the % Recovery in the calculation of skin
friction for piles and drilled shafts (Soil Handbook). Equation 7 is used to calculate the %
strength testing (Figure 4.4), and consists of measuring and recording the axial
deformation history of the sample in addition to its load history. The Rock core samples
to Initial Young Modulus’ value are loaded to 40% of the ultimate compressive strength
for three unload and reload cycles according to ASTM 469-94. The 40% limit is intended
to measure an elastic zone response and prevent any permanent deformation within the
sample. The first cycle is thought to contain seating errors and is disregarded in the
analysis. The remaining two stress and strain measurements are averaged together to
obtain the modulus of elasticity of the material. This test process requires an estimate of
the sample’s ultimate strength prior to the actual test. Figure 4.8 shows stress vs. strain
Tangent Young’s modulus, Et, is the slope of a line tangent to the sample's stress-
strain curve at 50% of its ultimate strength. The procedure for elastic modulus is that
Secant Young’s modulus, Es, is the slope of a straight line joining the origin of the
axial stress-strain curve to a point on the curve at some fixed percentage of the peak
strength. Figure 4.9 shows tangent modulus and secant modulus in a stress-strain curve.
Poisson's ratio can be determined at the same time. This involves the
measurement of the diameter deformation at the core's mid-level. These data allow
calculation of lateral strain. The loading data, recorded at the same rate, are converted to
stress by
38
p
σ = Equation 4.8
A
where
∆L ∆D
εl = , and ε r = Equation 4.9 ,4.10
L D
where
εl = Longitudinal strain
εr = Radial strain
800
700
600
400
300
200
100
Εi = σ / ε
0
0.0000 0.0020 0.0040 0.0060 0.0080 0.0100
Corrected Strain(in/in)
Figure 4.8 Stress vs. Strain Curve for calculating Initial Young’s Modulus
1,500
Slope = Es
1,000
Stress (psi)
500
Slope = Et
Figure 4.9 Stress vs. Strain Curve for Tangent and Secant Modulus
CHAPTER 5
ANALYSIS FOR LABORATORY DATA
rock strength due to the fact that sedimentary rocks are made from the mineral calcite
which came from the beds of evaporated seas and lakes and from sea animal shells. This
variability has been considered and estimated by foundation engineers. The large
The variability of the rock properties over a specific site is special information for
foundation design (i.e. skin friction, end bearing, and lateral resistance). Designers have
to consider the cost of collecting laboratory specimens as well as testing vs. spatial
variability and representative sample population sizes for each site. The means to
quantify the spatial variability and uncertainty (sample size) is found in the mathematical
material variability has grown rapidly in the last 20 years, and now forms one of the
of values (i.e. the sample, is taken from the larger set of values that compose the
population). This process is called sampling. The validity of the inferences concerning
the population is dependent on how well the sample represents the population. Samples
that systematically differ from the population are said to be biased. More technically, a
40
41
biased sample is a sample selected in such a way that all values in the population do not
have an equal chance of being selected. Samples that are not biased are random samples.
Error in the value of rock properties can arise in several ways. First, the sample
population may be small and not representative of the whole site (rock variability);
second, the sample properties have been altered or disturbed in the process of sampling
and/or transport to the laboratory (i.e. sampling errors); or finally, the laboratory tests
were not scrupulously performed according to prescribed standard methods (i.e. testing
errors). Each of these error sources may contain both bias and random error. Lumb
(1974) has discussed the identification and the contribution of each of these error types to
(i.e. over length of bridge foundation, etc.). However, sample size must be “large
general, an under-sized study can be a waste of resources for not having the capability to
produce useful results, while an over-sized one uses more resources than are necessary.
To quantify the latter, Equation 5.1 is used to estimate the sample size, N, needed to
σ
N = * (Zα + Z β ) 2 Equation 5.1
ε
σ = Standard Deviation
value of the physical quantity, based on a confidence level (i.e., 95%). The statistical
power is used to indicate the closeness with which the measurement agrees with another,
independently of any systematic error involved. Typically, the precision is set to be 20%,
which results in an “accurate” estimation with a small bias. A “precise” estimation has
both small bias and variance. Quality is proportionate to the inverse of variance. There is
nothing special about a 95% confidence level and a 20% precision level, however,
statisticians agree that this combination is “very significant” for the total population.
It has been found in this research that variability, specifically the standard
deviation, for Florida limestone becomes constant below 120 tsf for qu (unconfined
strength) and 20 tsf for qt (split tensile strength), which are the upper limits for rock
strength, that fail the rock instead of the interface for drilled shafts. The result shows a
constant standard deviation of about 30 tsf for qu and 6 tsf for qt. Using Equation 5.1, for
example, the 17th St bridge has s = 29, Zα = 1.96, Zβ = .84, a total of 23 samples are
needed, if an error, ε, of 10 tsf is acceptable. However, the values of qu <120 tsf, qt < 20
tsf, represent about 60% of the sample data population. Due to the significance of the
latter data, the sample size should be adjusted for the data that is not in the range of 120
tsf for qu and 20 tsf for qt. A simple solution is to adjust the number obtained for
From Equation 5.2, the average minimum number of samples to be taken for this
project is approximately 40 each (i.e. qu and qt). Moreover, since all the sites have
similar standard deviations, the latter number (i.e., 40 samples), should result in
homogeneous. The latter is not the case when different formations are present on a site
(e.g. 17th Street both sides of causeway) or layering is present. For such scenarios,
representative samples should be recovered from each formation or layer. A simple way
to identify layering or different formations is from the frequency strength (qu and qt)
distribution plots and the presence of multiple strong peaks over the site.
tensile strength (ASTM D-39) tests were performed on the north and south sides of the
existing bridge. Both core Recovery (%) and RQD (%) tests for a total of 119 Recoveries
and RQD were performed. The tests were performed on rock cores collected at an
approximate interval ranging from -29 to -142 ft (NGVD). A total of 16 tests for Young’s
modulus values were also performed near the site of the load tests.
The averages of qu and qt values for this site were 161.0 and 58.3 tsf. The Standard
Deviations of qu and qt values were 138.6 and 44.3 tsf. From the rock core samples of
this site, the average of Recovery was 24 % and the average of RQD values was 6 %. The
average (mean) of Young’s modulus values was 422,170 psi and the Standard Deviation
populations, the mode and median values will probably represent a better estimate of the
distribution of the data. Figures 5.1, 5.2 and 5.3, which depicts the limestone strength
and stiffness properties (i.e., qu, qt, Young’s Modulus), present the mode and median
values as well as the mean, and standard deviation of the data over the site.
17th ST Bridge
qu Frequency Distribution
100%
90%
80% qu Sample
Total Number = 56
70% Mode = 47.15 tsf
Frequency (%)
17th ST Bridge
qt Frequency Distribution
100%
90%
qt Sample
80% Total Number = 34
70% Mode = 25.3 tsf
Median = 47.5 tsf
Frequency (%)
17th ST Bridge
Ei Frequency Distribution
100%
90%
Ei Sample
80% Total Number = 16
70% Mode = 165,286 psi
Frequency (%)
Median =304,780psi
60% Mean = 422,170 psi
50% St.D = 356,360 psi
40%
30%
20%
10%
0%
0
0
00
00
00
00
00
00
00
00
00
0
00
00
00
00
00
00
00
00
00
00
00
00
00
00
50
50
50
50
50
15
25
35
45
55
65
75
85
95
10
11
12
13
Ei (psi)
Figure 5.3 Frequency Distribution for Ei of 17th Street Causeway Site
46
Generally, both of the paired items will have uncertainty associated with them. Of
course, the graphical and numerical summaries are applicable to each of the items, but it
of data.
A plot of the observed pairs, one against the other, is called a scattergram. The
2 2
_
_
_
_
R xy = ∑ xi − x * y i − y / ∑ i i
x − x * y − y Equation 5.3
= ±1; whereas if no linear relationship exists, then Rxy = 0. Note that this is a measure of
how well the data fits a straight line, data which has a nonlinear relationship will also
give a lower correlation coefficient. Figure 5.4 shows the results for the correlation
between qu and Ei. Evident from the Figure, there is a high correlation coefficient (Rxy
1,600,000
1,000,000
Ei(psi)
800,000
600,000
400,000
200,000
0
0 1,000 2,000 3,000 4,000 5,000 6,000
qu(psi)
Figure 5.4 Correlation Chart between the qu and Ei for 17th Street Causeway Site
performed on limestone samples collected in the vicinity of the bridge. No split tensile
tests were conducted for this project. Both core Recovery (%) and RQD (%) tests for a
total of 14 Recoveries and RQDs were performed and a total of 11 tests forYoung’s
modulus values were performed near the site of the load tests.
The average of the qu values for this site was 74.9 tsf. The Standard Deviation of
qu values was 82.7tsf. From the rock core samples obtained from the site, the average of
Recovery was 64.4 % and the average of RQD values was 37.1 %.
48
The average of Young’s modulus values was 480,911 psi and the Standard
Deviation of the modulus values was 841,720 psi. Figure 5.5 and Figure 5.6 show the
variability of the limestone properties (i.e., qt, Young’s Modulus) along with their mode,
Figure 5.7 shows the results for the correlation between qu and Ei. From Figure
5.8, it appears that there is high correlation coefficient (Rxy = 0.99) between qu and Ei
Acosta Bridge
qu Frequency Distribution
100%
90%
qu Sample
80% Total Number = 21
Mode = 20.8
70%
Median = 38.5 tsf
Frequency (%)
Acosta Bridge
Ei Frequency Distribution
100%
90%
Ei Sample
80% Total Number = 11
Mode = 97,997 psi
70%
Median = 99,174 psi
Frequency (%)
2,800,000
y = 536.74x - 145825
R = 0.9887
2,400,000
2,000,000
1,600,000
Ei(psi)
1,200,000
800,000
400,000
0
0 1,000 2,000 3,000 4,000 5,000
qu(psi)
Figure 5.7 Correlation Chart between the qu and Ei for Acosta Bridge Site
50
A total of 121 unconfined compressive strength (ASTM D-2938) tests and 76 split
tensile strength (ASTM D-39) tests were performed in the vicinity of the bridge.
Both core Recovery (%) and RQD (%) were obtained; 384 Recoveries and RQD
were recorded. Laboratory strength tests were performed on rock cores collected at over
the interval ranging from +30 to –40 feet (NGVD). A total of 23 tests for Young’s
The averages of qu and qt values for this site were 29.7 and 3.2 tsf. The Standard
Deviations of qu and qt values were 50.4 and 5.9 tsf, respectively. From the rock core
samples, the average Recovery was 56 % and the average of RQD values was 30 %.
From 23 rock core sample tests, the stiffness (Young’s Modulus) was determined.
The average Young’s Modulus and associated Standard Deviation was 170,735
and 177,847 psi respectively. Figure 5.8, 5.9 and 5.10 plot the frequency distribution of
the limestone’s strength and stiffness properties (i.e., qu, qt, Young’s Modulus) along
with their mode, median, mean, and standard deviation for the site.
Figure 5.11 shows the results for the correlation between qu and Ei. From Figure
5.11, it appears that there is high correlation coefficient (Rxy = 0.82) between qu and Ei
Apalachicola Bridge
qu Frequency Distribution
100%
90%
80% qu Sample
Total Number = 121
70% Mode = 5.56 tsf
Frequency (%)
17
24
31
38
45
52
59
66
73
80
87
94
0
9
10
10
11
12
12
qu (tsf)
Apalachicola Bridge
qt Frequency Distribution
100%
90%
80% qt Sample
Total Number = 76
70% Mode = 0.91 tsf
Frequency (%)
Apalachicola Bridge
Ei Frequency Distribution
100%
90%
80% Ei Sample
Total Number = 23
70% Mode = 41,218 psi
Frequency (%)
00
00
00
00
00
00
00
00
00
00
00
00
50
50
50
50
50
50
50
50
50
50
50
35
10
17
24
31
38
45
52
59
66
73
80
Ei (psi)
800,000
700,000
y = 584.35x
R = 0.8238
600,000
500,000
Ei(psi)
400,000
300,000
200,000
100,000
0
0 100 200 300 400 500 600 700 800 900
qu(psi)
Figure 5.11 Correlation Chart between the qu and Ei for Apalachicola Bridge Site
53
Testing was performed on intact core samples from both the limestone and marl
formations. In the case of the marl, both cemented and non-cemented specimens were
tested. A total of 53 Recovery (%) and RQD (%) were recorded over the site.
Laboratory testing included 51 unconfined compression strength, qu, and 22 split tension
tests, qt. A total of 33 Young’s Modulus values were recorded from the unconfined
strength tests.
The averages of qu and qt values were 74.0 and 21.0 tsf, respectively. The
Standard Deviation of qu and qt values was 74.6 and 12.8 tsf. From the rock core
samples over the site, the average of Recovery was 58 % and the average of RQD values
was 37 %. The average and standard deviation of Young’s Modulus was 362,167 and
475,836 psi from 33 rock core samples from the site. Figure 5.12, 5.13 and 5.14 display
the frequency distribution of the limestone properties (i.e., qu, qt, Young’s Modulus)
along with the mode, median, mean, and standard deviation for each property on the site.
Figure 5.15 shows the results for the correlation between qu and Ei. From Figure
5.15, it appears that there is a high correlation coefficient (Rxy = 0.98) between qu
00
0
0
00
00
00
00
00
0
00
00
00
00
00
00
00
00
00
50
50
50
50
50
50
15
25
35
45
55
65
75
85
95
10
11
12
13
14
Ei (psi)
3200000
2800000
y = 314.39x + 66578
R = 0.9803
2400000
2000000
Ei(psi)
1600000
1200000
800000
400000
0
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000
qu(psi)
Figure 5.15 Correlation Chart between the qu and Ei for Fuller Warren Bridge Site
56
strength (ASTM D-39) tests were conducted in the laboratory on the recovered cores. In
addition, the 38 unconfined compression tests had the Young’s Modulus measured (i.e.
cross-head movement was recorded). A total of 39 Recoveries (%) and RQD (%) were
The averages of qu and qt values for this site were 115.0 and 17.8 tsf, respectively.
Standard Deviations of 148.4 and 17.7 tsf were found for qu and qt on the site. From all
the rock core samples, the average of Recovery was 83.2 % and the average of RQD
values was 56.3 %. The mean and standard deviation of Young’s modulus for this
site were 547,642 and 428,656 psi, respectively. Figures 5.16 through 5.18 plot the
frequency distribution of the limestone properties (i.e., qu, qt, Young’s Modulus) along
with their mode, median, mean, and standard deviation over the site.
Figure 5.19 shows the results for the correlation between qu and Ei. From Figure
5.19, it appears that there is high correlation coefficient (Rxy = 0.84) between qu and Ei
Gandy Bridge
qu Frequency Distribution
100%
90%
qu Sample
80%
Total Number = 38
70% Mode = 44.8 tsf
Median = 46.9 tsf
Frequency (%)
40%
30%
20%
10%
0%
0
15
45
75
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
10
13
16
19
22
25
28
31
34
37
40
43
46
49
52
55
58
qu (tsf)
Gandy Bridge
qt Frequency Distribution
100%
90%
80% qt Sample
Total Number = 28
70% Mode = 8.1
Frequency (%)
Median =9.4
60%
Mean = 17.8 tsf
50% St.D = 17.7 tsf
40%
30%
20%
10%
0%
0 7.5 22.5 37.5 52.5 67.5 82.5 97.5 113 128 143 158 173 188 203
qt (tsf)
Gandy Bridge
Ei Frequency Distribution
100%
90% Ei Sample
Total Number = 28
80%
Mode = 223,506 psi
70% Median = 348485 psi
Frequency (%)
25 00
35 00
45 00
55 00
65 00
75 00
85 00
95 00
10 000
11 000
12 000
13 000
14 000
15 000
16 000
17 000
18 000
19 000
0
00
0
00
00
00
00
00
00
00
00
50
50
50
50
50
50
50
50
50
50
0
Ei (psi)
Figure 5.18 Frequency Distribution of Ei for Gandy Bridge Site
1,800,000
1,600,000
1,000,000
800,000
600,000
400,000
200,000
0
0 2,000 4,000 6,000 8,000 10,000
qu(psi)
Figure 5.19 Correlation Chart between the qu and Ei for Gandy Bridge Site
59
D-39) tests were performed on cores recovered from the site. A total of 277 Recoveries
and RQD values were recorded. The strength tests were performed on specimens
collected from elevation + 40 to –20 feet (NGVD). From the unconfined strength tests,
The averages of qu and qt values for this site were 90.3 and 31.0 tsf, respectively.
Standard Deviations of 63.2 and 21.7 tsf were found for the qu and qt values. From the
recovered rock core samples, an average Recovery of 67 % and an average RQD value of
42 % was found. The mean and Standard Deviation of the Young’s Modulus were
2,676,038 and 1,112,087 psi, respectively. Figures 5.20, 5.21 and 5.22 display the
variability of the limestone properties (i.e., qu, qt, Young’s Modulus) along with their
Figure 5.23 shows the results for the correlation between qu and Ei. From the
Figure 5.23, it appears that there is a high correlation coefficient (Rxy = 0.75) between
Victory Bridge
qu Frequency Distribution
100%
90%
qu Sample
80% Total Number = 56
Mode = 45.0 tsf
70% Median = 74.8 tsf
Frequency (%)
Victory Bridge
qt Frequency Distribution
100%
90%
80% qt Sample
Total Number = 28
70%
Mode = 37.1 tsf
Frequency (%)
Victory Bridge
Ei Frequency Distribution
100%
90% Ei Sample
80% Total Number = 24
Mode = 2,240,752 psi
70%
Median = 2,985,000 psi
Frequency (%)
45 0
75 0
10 00
13 00
16 00
19 00
22 00
25 00
28 00
31 00
34 00
37 00
40 00
43 00
46 00
49 00
0
00
0
0
00
00
00
50
50
50
50
50
50
50
50
50
50
50
50
50
50
Ei (psi)
5,000,000
y = 1078x + 1E+06
4,500,000
R = 0.7473
4,000,000
3,500,000
3,000,000
Ei(psi)
2,500,000
2,000,000
1,500,000
1,000,000
500,000
0
0 500 1,000 1,500 2,000 2,500 3,000 3,500
qu(psi)
Figure 5.23 Correlation Chart between the qu and Ei for Victory Bridge Site
62
Tables 5.1 and 5.2 summarize the results for the mean, standard deviation, and
coefficient of variation of unconfined compressive strength (qu) and tensile strength (qt).
Table 5.3 shows the correlation between the coefficient of variation of qu and qt. From
the Table 5.3, it appears that there is a high correlation between the coefficient of
variation of qu and qt (Rxy = 0.93) calculated by Equation 5.3. Figures 5.24 and 5.25 also
shows graphically the high correlation between the coefficient of variation of qu and qt.
Standard
Bridge Mean COV
Deviation
17th Street Causeway 161.0 138.6 86%
Apalachicola 30.0 50.4 168%
Fuller Warren 74.0 74.6 101%
Gandy 115.0 148.4 129%
Hillsborough 89.5 130.5 146%
Victory 90.3 21.3 70%
Standard
Bridge Mean COV
Deviation
17th Street Causeway 58.3 44.3 71%
Apalachicola 3.2 5.9 184%
Fuller Warren 21.0 12.8 61%
Gandy 17.8 17.7 99%
Hillsborough 12.0 16 133%
Victory 31.0 21.7 70%
qu qt
Pearson Correlation 1 .915(**)
qu Sig. (2-tailed) . 0.01
N 6 6
Pearson Correlation .915(**) 1
qt Sig. (2-tailed) 0.01 .
N 6 6
200%
180%
160%
140%
R2 = 0.8381
100%
80%
60%
40%
20%
0%
0% 20% 40% 60% 80% 100% 120% 140% 160% 180% 200%
qu_cov
200%
180% y = 0.7194x + 0.4257
Coefficients of Variation 160% R2 = 0.8381
140%
120%
100%
80%
60%
40%
20%
0%
a
y
dy
ST
gh
n
ol
or
re
an
ou
ic
ct
th
ar
ch
Vi
or
17
rW
ra
sb
lle
pa
ill
Fu
H
A
qu_cov qt_cov
Of interest was the correlation of % Recovery with Rock Quality Designation (i.e.
RQD), since both are determined from the recovered cores and are reported for all the
on all the RQD and % REC from boreholes, as shown in Table 5.4.
Based on the Pearson Correlation and Sig. (2-tailed) values, RQD and % Recovery
are correlated. Specifically, the analysis shows a progressive increase in RQD with
corresponding increase in % Recovery. However, there are exceptions (i.e. when the
rock is interleaved with clay instead of sand (Victory), then the % Recovery jumps
relative to RQD).
65
Recovery RQD
Pearson Correlation 1 .810(**)
Recovery Sig. (2-tailed) . 0.01
N 976 976
Pearson Correlation .810(**) 1
RQD Sig. (2-tailed) 0.01 .
N 976 976
The latter is important, since the FDOT recommends in their “Soil Handbook” to
adjust the predicted unit skin friction by the average % Recovery to account for voids in
the rock (i.e. reduced skin friction). But in the case of Victory Bridge, which has high
recoveries (i.e. clay filling the rock voids), the RQD would be a better indication of rock
The correlation of RQD or % Recovery with rock strength (i.e. qu) was also
expected, since both measure different quantities (i.e. strength) and presence of rock, and
5.5 Correlation Between Rock Strength (qu & qt) with SPT N Values
site, an attempt was made to correlate SPT N value with qu and qt values. Subsequently,
the N value at each spatial point where either a laboratory qu or qt test had been
conducted was gathered. Figures 5.26 and 5.27 show the correlation between N and qu
and N and qt, respectively. Evident from the figures, it appears that there is little if any
correlation between SPT N values and rock strength. Also, from the variability, it is
66
doubtful if SPT N values could be correlated directly to unit skin friction on the drilled
shafts.
qu vs. N
140
120
N ( Blow Counts)
100
80
60
40
20
0
0 100 200 300 400 500
qu (tsf)
qt vs. N
140
120
100
N ( Blow Counts)
80
60
40
20
0
0 10 20 30 40 50 60 70 80 90
qt (tsf)
In deep foundation, drilled shafts may develop their load capacity in side shear and
end bearing between the concrete and soil/rock. End bearing contributes to the total load
capacity of the shaft, but it is a minor participant at low deflections where the shear is
developed. The amount of drilled shaft movement required to fully mobilize the ultimate
shear transfer is usually on the order of 1/4 of an inch and typically independent of the
drilled shaft diameter. The end bearing or point resistance, however, reaches its ultimate
load capacity generally upon tip movement on the order of 5 percent of the shaft diameter
(i.e., FHWA Failure Criterion). Although end bearing resistances can produce high load
capacities for drilled shafts installed in dense granular soils or sound rock, the movement
of the shaft tip required to mobilize these end bearing capacities and may result in shaft
butt movements that could exceed the tolerable limits of the structure.
Because of the nature of Florida limestone formation, most drilled shafts are
designed primarily in skin friction. During design, the estimated contribution from the
end bearing was often neglected in the projects analyzed. Because of the movement
required to "mobilize" base resistance, there was no attempt from the designers to
maximize end bearing. However, this research had the advantage of several load test data
The Osterberg cell automatically separates the skin friction and end bearing. It was
then possible to obtain a measured end bearing. In the case of end bearing, failure is
68
69
usually defined on a settlement criteria (i.e. FDOT defines failure as settlement equal to
the elastic compression plus 1/30 of the diameter of the shaft), which may occur prior to
plunging (very large settlement). Consequently, the unit end bearing is divided into
three different values: Mobilized (settlement less than FDOT), FDOT failure, and
Maximum failure.
predicted End Bearing of drilled shaft could be calculated by the O’Neill method (FB-
geomaterials,” or geomaterials at the boundary between soil and rock, since borehole in
such geomaterials are relatively stable, geomaterials are not usually difficult to excavate,
predicted end bearing for drilled shaft in Florida limestone require determination of both
expected deflections of the top deflections and deflections of the tip of drilled shaft.
In the calculation of end bearing for prediction, the following parameters for each
- Em, Young’s modulus of rock mass of the limestone from Ei and RQD
Table 6.1 Measured Unit End Bearing from Osterberg Load Test in Florida
70
* Tip
Shaft Mobilized FDOT Maximum
Shaft Unknown Move- Failure
Bridge Length Bearing Failure Failure
Name Friction ment Status
(ft) (tsf) (tsf) (tsf)
(ft) (in)
LTSO1 119.4 5.2 0.624 Both x x x
17th Street LTSO2 142.0 9.1 1.95 Tip Failure x x x
Bridge LTSO3 100.1 11.1 1.89 Both 41.5 x x
LTSO4 77.5 2.6 3.53 Tip Failure x x 66.4
Test 1 64.2 0 4.41 Tip Failure x 61.7 90.3
Acosta Test 2 101.2 0 2.97 Tip Failure x 28 39
Bridge Test 4 113.9 0 3.2 Tip Failure x 22.4 30.2
Test 5A 87.8 0 5.577 Tip Failure x 18.5 29.4
46-11A 85.0 0 5.977 Both x 62 92
53-2 72.0 0 2.1 Both 62.5 x x
Apalachicola 57-10 84.0 0 1.7 Both 56** x x
Bridge 59-8 134.0 9 1.3 Both 57 x x
62-5 89.2 0 2.69 Both x 33.2 40
69-7 99.1 0 4.46 Both x 30.5 44
LT-1 41.0 0 0.23 Skin Failure 87 x x
Fuller Warren LT-2 27.9 0 2.56 Both x 80.8 89.5
Bridge LT-3a 120.7 0 2.94 Both x 34 34
LT-4 66.8 0 3.12 Both x 54 70
26-2 38.4 9.8 0.4 Skin Failure x x x
Gandy
52-4 54.5 4.33 2.9 Both x 139.2 x
Bridge
91-4 74.7 6.7 2.5 Both X 42.9 x
3-1 33.2 0 0.5 Both 109 x x
3-2 38.6 9.66 0.4 Skin Failure X x x
Victory
10-2 46.6 7.7 2.367 Both X 145 x
Bridge
19-1 45.0 0 0.528 Both 124.4 x x
19-2 50.7 12.14 0.4 Skin Failure X x x
Hillsborough
4-14 70.8 7.33 1.74 Both X x x
Bridge
Note :
1) *Unknown Friction : Distance from tip to the lowest strain gage.(Side skin friction is unknown)
2) ** : Ultimate unit end bearing due to plunging
3) x : Not determined
4) Shaft Length : Distance from the top to the tip of the shaft
5) Failure Status : Both means that the shaft fails in both side and end resistance
6) Mobilized : The mobilized unit end bearing when the bottom movement is 1/30 of the shaft diameter.
7) FDOT Failure : The unit end bearing when the bottom movement is 1/30 of the shaft diameter.
8) Maximum Failure : The unit end bearing when bottom movement is larger than 1/30 of the shaft diameter.
In the O’Neill method for calculating end bearing, many properties of the rock
samples are required and these values are derived from laboratory and field tests. Using
properties (qu, qt, Ei, RQD) of rock samples, some functions for the prediction were
calculated and finally total resistance (Qt) and the settlement (Wb) at the bottom of the
The following procedure is the method of predicting end bearing for drilled Shaft
based on the O’Neill method. A schematic of a typical drilled shaft foundation for
Axial Load
Drilled Shaft
Overburden
L2 Sand(Clay)
L
Qs
L1
Limestone
Tip Resistance , Qb
♦ The settlement at the bottom of the shaft (Wb) is calculated as assuming the
settlement at the top of the shaft (Wt).
72
2(Qt + Qb )L
Wb = Wt − Equation 6.1
πEc D
2
♦ Calculate total resistance, Qt of the shaft to estimate the settlement of the shaft
bottom. In order to Qt, a value of Wt is assumed, and Θf (Theta) is evaluated before
deciding which equation to use.
πD 2
Qt = πDLΘ f f s + qb , Θf ≤ n Equation 6.2
4
π D2
Qt = πDLK f f s + qb , Θf > n Equation 6.3
4
0.67
qb = ΛWt Equation 6.4
♦ Evaluate Θf(Theta), which is a function of Wt (the settlement of the top of the shaft)
and Kf, which is a function of n and Θf. Wt, which is assumed for calculating Wb (the
settlement of the shaft at the bottom) is the settlement at the top of the shaft. If Θf is
greater than n value, Kf will be used for calculating end bearing.
Em Ω
Θf = W Equation 6.5
πL Γ f s t
Kf = n+
(Θ f − n )(1 − n )
≤ 1 Equation 6.6
Θ f − 2n + 1
73
♦ Determine the “characteristic parameter” n, which is a fitting factor for the load-
settlement syntheses produced by the finite element analysis and Estimate σn, the normal
stress between the concrete and borehole wall at the time of loading. This σn is evaluated
at the time the concrete is fluid. If no other information is available, general guidance on
the selection of σn can be obtained from Figure 2, which is based on the measurements of
Bernal and Reese.
σn
In case rock socket is rough, n = Equation 6.7
qu
σ n = M .γ c .Z c Equation 6.8
γc is the unit weight of the concrete and Zc is the distance from the top of the
completed column of concrete to the point in the borehole at which σn is desired (usually
1.0
0.8
0.6
Depth = 0 m
M
Depth = 4 m
0.4
Depth = 8 m
Depth = 12 m
0.2
0.0
125 150 175 200 225
Slump (mm)
L
0.5
L 0.5 E
Γ = 0.37 − 0.15 − 1 log10 c + 0.13 Equation 6.9
D D Em
74
L
0.5
L 0.5 E
Ω = 1.14 − 0.05 − 1 log10 c − 0.44 Equation 6.10
D D Em
0.67
0.5
L 200 L − Ω L + 1
D
Λ = 0.0134 Em D D Equation 6.11
L πLΓ
+ 1
D
♦ Calculate fs, the apparent maximum average unit skin friction at infinite
displacement for limestone. For calculating fs, Dr. Michael McVay’s equation for Florida
limestone was used.
Ei, Young’s modulus of the intact rock core values, were calculated from Standard
D3148) for limestone samples. Many geotechnical reports did not include Ei value, so
here Ei values were calculated through the correlation of qu, Unconfined Compressive
Em, Young’s modulus of the rock mass values was estimated based on the
RQD value of the limestone, and the adjustment values for the RQD values were applied
to each rock. If RQD value is less than 20 percent, the correlation with RQD was used as
20 percent’s RQD (Load Transfer for Drilled Shaft in Intermediate Geomaterials, 1996).
Table 6.2 Estimation of Em/Ei based on RQD (Load Transfer for Drilled Shafts in
Intermediate Geomaterials, 1996)
75
The results of the O’Neill method for founding unit end bearing are represented in
Table 6.3 as an example in the case of LTSO 3’s shaft in the 17th Street bridge.
Table 6.3. Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (LTSO 3 in the 17th Street Causeway)
bearing of the shaft is also presented as applying FDOT failure’s criterion in Figure 6.3.
76
* AVERAGE
17th Street Predicted vs. Measured(LTSO 3) qu :30.75 (tsf)
using Nearest Boring qt : 9.6 (tsf)
150
100
75
Measured Load Test
LTSO 3
50
Predicted
Em = 59,237 (psi)
25
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Deflection(inches)
FDOT Failure Measured-LTSO3 Predicted
Figure 6.3 17th Street LTSO 3 – Comparison between Predicted vs. Measured FDOT
Failure End Bearing
6.3 Prediction of the Unit End Bearing using the Nearest Boring
6.3.1.1 Site Description and Data Analysis for the nearest boring
Based on both the Geotechnical report and the drilled shaft boring logs, it was
observed that the elevation of the top of the limestone formation varied considerably
within the site. Since the strength of the limestone was very different on each side of the
channel, it is estimated that the site was divided into soft limestone and hard limestone
areas. The soft limestone area includes LT1 shaft (ST. 35+46) and LTSO3 shaft (ST.
34+82), and the hard limestone area includes the remaining load test shafts.
The nearest borings for the Osterberg Tests of LTSO 3 (ST. 34+82) and LTSO 4
(38+04) shaft were BB4 (ST. 34+81) and BB6 (ST.38+07). However, qu value of boring
77
BB9, which was obtained from elevation -118 feet and -130 feet, was also considered as
a nearest boring for predicting tip strength of LTSO 3 shaft. Because the qu and qt values
of BB4 boring, which were obtained from an elevation of -66 feet to -95 feet, are sparsely
composed of soft limestone and were not effective for predicting hard limestone after -96
In the case of BB6 boring for predicting side and tip strength of LTSO 4, limestone
was sparsely found from an elevation of -43 to -105 feet, and under the elevation of -105
RQD (%) of the nearest boring BB9 for predicting the tip of LTSO3 shaft was 43.0
%. From the rock core samples of this nearest boring, the average of RQD values for the
nearest boring BB6 for LTSO 4 was 41.9 %. The averages of qu and qt values for BB4
were 30.7 tsf and 9.6 tsf and the average of qu and qt values for BB6 were 138.6 tsf and
14.5 tsf.
For Initial Young Modulus values of this site a new boring was performed near the
nearest boring BB 4 (St. 34+81) of the LTSO 3 (St. 34+82) shaft. The new boring in a
general plan for this site is located around 10 feet distance to the north side from the
boring BB 4. Limestone was found from a depth of -90 to -121 feet. A total of 16 rock
core samples was tested for Young Modulus values for this site.
Modulus value, Young’s Modulus value for limestone of the nearest boring was 466,377
psi for BB9 and 626,529 psi for BB6. Em, Young’s Modulus of the rock mass values for
the limestone of the nearest borings were estimated by an adjustment value based on
78
RQD (FHWA-RD-95-172, 1996). The adjustment values for RQD value 43 (%) and 41.9
(%) for Em value were 0.127 and 0.123 as shown in Table 6.2.
The mobilized unit end bearing, FDOT failure values, and maximum failure values
for this site were reported through four Osterberg load tests as shown in Table 6.1. Based
on FDOT failure criterion, predicted unit end bearing values for LTSO3 and LTSO4 were
estimated by the O’Neill method using the nearest boring data. The Result of the
spreadsheet for the decision of unit end bearing of the load-settlement is shown in Table
6.4. and 6.5. The predicted values, 50.0 tsf and 64 tsf, were presented in Table 6.6, and
the graphical representations for the predicted unit end bearing value are presented in
Table 6.4 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (LTSO 3 in 17th Street Causeway)
Table 6.5 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (LTSO 4 in 17th Street Causeway)
Table 6.6 Prediction of Unit End Bearing Using the Nearest Boring(17th bridge)
* AVERAGE
17th Street Predicted vs. Measured(LTSO 3) q u :30.75 (tsf)
using Nearest Boring q t : 9.6 (tsf)
150
100
75
Measured Load Test
LTSO 3
50
Predicted
Em = 59,237 (psi)
25
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Deflection(inches)
FDOT Failure Measured-LTSO3 Predicted
th
Figure 6.4 Predicted vs. Measured End Bearing (LTSO 3 in 17 Street Causeway)
80
* AVERAGE
17th Street Predicted vs. Measured(LTSO 4) qu :138.6 (tsf)
using Nearest Boring qt : 14.5 (tsf)
150
Ei = 626,529 psi
FDOT Failure Value
RQD = 41.9 %
125
Measured Load Test
LTSO 4
End Bearing(tsf)
100
75
Predicted
Em = 77,063 (psi)
50
25
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Deflection(inches)
FDOT Failure Measured-LTSO4 Predicted
th
Figure 6.5 Predicted vs. Measured End Bearing (LTSO 4 in 17 Street Causeway)
6.3.2.1 Site Description and Data Analysis for the nearest boring
A total of 4 Osterberg tests (Test 1, Test 2, Test 4, and Test 5A) were performed in this
site and the hard limestone was only found from an elevation of –20 feet to –47 near the
bottom of the Test 1 shaft. The sites for Test 2, Test 4, and Test 5A were composed
by medium sand, clay, and silt ranging from an elevation of -40 to -100 feet around the
bottom of the shafts. The nearest borings for the Osterberg Tests of Test 1 (ST. 134+39)
shaft was estimated as the boring WA-2 (ST. 134+40) which is composed of limestone
From the rock core samples of this nearest boring, the average of RQD values was
34.2 % and Recovery was 74 %. The average of qu and qt values for WA-2 were 85.3 tsf
For Initial Young Modulus values of this site, 11 rock core samples were taken from two
new borings on the ground near the shaft Test 1 on the river. The location of the
first boring is at a distance of 21ft to the right of the boring SB 7(St.130+34) and the
location of the second boring is the place where is located at the distance 25ft to the right
of the boring NB 1(St.150+35). Soft limestone was found in first borings from a depth of
-42 feet to -47 feet, and in second borings limestone was found from a depth of -27 feet
to -32 feet.
From these rock core samples, the tests for Initial Young Modulus values were
performed and the average of Young Modulus values was 480,911 psi. The Standard
modulus value, Young’s Modulus value for the limestone of the nearest boring was
579,042 psi. Em, Young’s Modulus of the rock mass values for the limestone of the
172, 1996). The adjustment values for RQD value 34.2 (%) were 0.097 as shown in Table
6.2.
The FDOT failure values in unit end bearing and maximum failure values for this
site were reported through four Osterberg load tests as shown in Table 6.1. Based on
FDOT failure criterion predicted unit end bearing values for Test 1 were estimated by the
O’Neill method using the nearest boring data. The Result of spreadsheet for the decision
of unit end bearing of load-settlement is shown in Table 6.7. The predicted value for the
Test 1 shaft was 40.0 tsf as shown in Table 6.8, and the graphical representation for the
Table 6.7 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (Test 1 in Acosta Bridge)
Table 6.8 Prediction of Unit End Bearing Using the Nearest Boring (Acosta)
* AVERAGE
Acosta Predicted vs. Measured(Test 1) qu :85.3 (tsf)
using Nearest Boring qt : 18.4 (tsf)
150
Ei = 579,042 psi
FDOT Failure RQD = 34.2 %
125
100
Test 1
75
Predicted
50 Em = 57,904 (psi)
25
0
0.0 1.0 2.0 3.0 4.0 5.0
Deflection(inches)
FDOT Failure Measured-Test 1 Predicted
Figure 6.6 Predicted vs. Measured End Bearing (Test 1 in Acosta Bridge)
83
A total of 6 Osterberg tests in this site were performed and FDOT failure values
were measured to 46-11A, 62-5, and 69-7 shafts. The nearest borings for the Osterberg
Tests of 46-11A (ST. 624+03) were TH-46A and TH-46B and for 62-5 (ST. 645+97)
shaft were TH-62A and TH-62B, and for 69-7 (ST. 653+41) shaft were TH-69A and
TH-69B. The weathered limestone included calcareous and cemented clayey in boring
TH-46A and TH-46B from an elevation of -13 to -42 feet. In the case of boring TH-62A
and TH-62B, the limestone is as thick as 27 feet from the elevation of -23 feet, In the case
of boring TH-69A and TH-69B, limestone was found at an elevation of -23 to -35 feet.
From the rock core tests of these nearest boring TH-46A and TH-46B, the average of
RQD values was 64.0 % and Recovery was 85 %, and for the borings TH-62A and
TH-62B, the average of RQD values was 46.7 % and Recovery was 80 %. For the boring
TH-69A and TH-69B, the average of RQD values was 47.0 % and Recovery was 70 %.
The averages of qu and qt values for the borings TH-46A and TH-46B were 12.2 and 1.0
tsf and the averages of qu and qt values for TH-62A and TH-62B were 5.3 and 0.75 tsf.
The averages of qu and qt values for TH-69A and TH-69B were 5.1 and 0.4 tsf.
Modulus value from 23 rock core samples of this site, Initial Young’s Modulus values for
the nearest borings (TH-46A and 46B, TH-62A and 62B, TH-69A and 69B) were 98,608
psi, 43,014 psi, and 41,391 psi. However, since Young’s Modulus, qu and qt values are
too small Em, Young’s Modulus of the rock mass values for the limestone of the nearest
borings in this site were estimated by an adjustment value based on Recovery value
instead of RQD (FHWA-RD-95-172, 1996). The adjustment values for the nearest
84
borings (TH-46A and 46B, TH-62A and 62B, TH-69A and 69B) of Recovery values
78.0 (%), 58.2 (%), and 54 (%) were 0.54, 0.37, and 0.26. Em, Young’s Modulus of the
rock mass values for the limestone of the nearest borings were 52,755 psi, 15915 psi, and
10,762 psi.
The mobilized unit end bearing and maximum failure values for this site were
reported through four Osterberg load tests as shown in Table 6.1. Based on FDOT failure
criterion predicted unit end bearing values for 46-11A, 62-5, and 69-7 shafts were
estimated by the O’Neill method using the nearest boring data. The Result of Spreadsheet
for decision of unit end bearing of load-settlement is shown in Tables 6.9, 6.10, and 6.11.
The predicted values, 45.5 tsf, 15.2 tsf, and 12.0 tsf, are shown in Table 6.12, and the
graphical representations for the predicted unit end bearing value are presented in Figures
Table 6.9 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (46-11A in Apalachicola Bridge)
Table 6.10 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (62-5 in Apalachicola Bridge)
Wt(in) Wb(in) qb(psi)
0.2 0.1935 41.3
0.4 0.3919 65.8
0.8 0.7897 104.6
1.2 1.1879 137.3
1.6 1.5863 166.5
2 1.9849 193.3
2.4 2.3835 218.5
2.8 2.7822 242.2
4 3.9786 307.6
Table 6.11 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (69-7 in Apalachicola Bridge)
Table 6.12 Prediction of Unit End Bearing Using the Nearest Boring (Apalachicola)
* AVERAGE
Apalachicola Predicted vs. measured(46-11A) q u : 12.15 (tsf)
using Nearest Boring q t : 1.0 (tsf)
200
Ei = 98,608.2 psi
180 RQD = 64 %
160
140 FDOT Failure Value
End Bearing(tsf)
120
100
Measured Load Test
80 46-11A
60
40
Predicted
20 Em=52,755 (psi)
0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
Deflection(inches)
Predict FDOT Failure Measured-46-11A
Figure 6.7 Predicted vs. Measured End Bearing (46-11A in Apalachicola Bridge)
* AVERAGE
Apalachicola Predicted vs. measured(62-5) q u :5.3 (tsf)
using Nearest Boring q t : 0.8 (tsf)
200
Ei = 43,014.3 psi
175 RQD = 46.8 %
150
End Bearing(tsf)
125
100
FDOT Failure
75
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Deflection(inches)
Predicted FDOT Failure Measured-62-5
Figure 6.8 Predicted vs. Measured End Bearing (62-5 in Apalachicola Bridge)
87
175
Ei = 41,391.1 psi
150 RQD = 47.1 %
FDOT Failure Value
End Bearing(tsf)
125
100
75
50
Measured Load Test Predicted
25 69-7 Em = 10,762(psi)
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Deflection(inches)
Predicted FDOT Failure Measured-69-7
Figure 6.9 Predicted vs. Measured End Bearing (69-7 in Apalachicola Bridge)
A total of 4 Osterberg tests were performed in this site and FDOT failure values
were measured to LT-2, LT-3a, and LT-4 shafts. The nearest boring for the Osterberg
Tests of LT-3a (ST. 323+81) was estimated as a boring BW-14 and the nearest boring
From the rock core tests of these nearest borings, BW-14 and BL-23, the averages
of RQD values were 31.0 % and 33.0 % and Recovery were 51 % and 49 %. The
averages of qu and qt values for the boring BW-14 were 43.0 and 9.6 tsf, and the
averages of qu and qt values for BL-23 were 88.5 and 11.4 tsf.
To obtain the correlation equation between Initial Young Modulus values and qu of
this site, a new boring was performed near the nearest boring BL 1 (St. 284+04) of LT 1
88
(St. 285+30) shaft, since LT-3a and LT-4 shafts are on the river. Limestone in this new
boring was found from a depth of –25 to –40 feet. A total of 33 rock core samples was
Initial Young’s Modulus values were calculated using 33 rock core samples of this
Modulus value. Initial Young’s Modulus values for the limestone of the nearest borings
at this site were 258,780 psi for LT-3a and 456,618 psi for LT-4. Em, Young’s Modulus
of the rock mass values for the limestone of the nearest borings at this site, were
adjustment values for the nearest boring BW-14 of RQD value 31.0 (%) was 0.093, and
the adjustment values for the nearest boring BL-23 of RQD value 33.0 (%) was 0.096.
Em, Young’s Modulus of the rock mass values for the limestone of the nearest borings
The mobilized unit end bearing, FDOT failure value, and maximum failure values
for this site were reported through four Osterberg load tests as shown in Table 6.1. Based
on FDOT failure criterion predicted unit end bearing values for LT-3a and LT-4 shafts
were estimated by the O’Neill method using the nearest boring data. The Result of
Spreadsheet for decision of unit end bearing of load-settlement is shown in Tables 6.13
and 6.14. The predicted values, 22.5 tsf and 37.0 tsf, are shown in Table 6.15, and the
graphical representations for the predicted unit end bearing value are presented in Figures
Table 6.13 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (LT-3a in Fuller Warren Bridge)
Wt(in) Wb(in) qb(psi)
0.2 0.1844 57.3
0.4 0.3728 91.2
0.8 0.7549 145.0
1.2 1.1413 190.3
1.6 1.5305 230.8
2 1.9216 268.0
2.4 2.3140 302.8
2.8 2.7074 335.7
Table 6.14 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (LT 4a in Fuller Warren Bridge)
Table 6.15 Prediction of Unit End Bearing Using the Nearest Boring (Fuller Warren)
* AVERAGE
Fuller Warren Predicted vs. measured(LT-3a) qu :43.0 (tsf)
using Nearest Boring qt : 9.6 (tsf)
150
FDOT Failure
125
Ei = 258,780 psi
RQD = 31 %
End Bearing(tsf)
100
75
Figure 6.10 Predicted vs. Measured End Bearing (LT-3a in Fuller Warren Bridge)
* AVERAGE
Fuller Warren Predicted vs. measured(LT4) q u : 88.5 (tsf)
using Nearest Boring q t : 11.4 (tsf)
150
100
50
Predicted
Em = 43,835 (psi)
25
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Deflection(inches)
FDOT Failure Measured-Lt-4 Predicted
Figure 6.11 Predicted vs. Measured End Bearing (LT 4 in Fuller Warren Bridge)
91
A total of three Osterberg tests were performed at this site and FDOT failure values
were measured to 52-4 and 91-4 shafts. The nearest boring for the Osterberg Tests of 52-
4 (ST. 93+63) was estimated as a boring SB-36, and the nearest boring for 91-4 (ST.
From the rock core tests of these nearest borings, SB-36 and SB-91, the averages of
RQD values were 78.0 % and 52.0 % and Recovery were 88 % and 70 %. The averages
of qu and qt values for the boring SB-36 were 36.6 and 13.9 tsf and the averages of qu
Initial Young’s Modulus values were calculated using 28 rock core samples of this
Modulus value. Initial Young’s Modulus values for the nearest borings at this site were
322,857 psi and 274,136 psi. Em, Young’s Modulus of the rock mass values for the
limestone of the nearest borings at this site were estimated by an adjustment value based
on RQD value (FHWA-RD-95-172, 1996). The adjustment values for the nearest boring
SB-36 of RQD value 78.0 (%) was 0.78, and the adjustment values for the nearest boring
SB-91 of RQD value 52.0 (%) was 0.21. Em, Young’s Modulus of the rock mass values
for the limestone of the nearest borings were 251,828 psi and 57,569 psi.
The mobilized unit end bearing, FDOT failure value, and maximum failure values
for this site were reported through four Osterberg load tests as shown in Table 6.1. Based
on FDOT failure criterion, predicted unit end bearing values for 52-4 and 91-4 shafts
were estimated by the O’Neill method using the nearest boring data. The Result of
92
Spreadsheet for decision of unit end bearing of load-settlement is shown in Table 6.16
and Table 6.17. The predicted values, 154 tsf and 45.0 tsf, are presented in Table 6.18,
and the graphical representations for the predicted unit end bearing value are presented in
Table 6.16 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (52-4 in Gandy Bridge)
Table 6.17 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (91-4 in Gandy Bridge)
Table 6.18 Prediction of Unit End Bearing Using the Nearest Boring(Gandy)
* Average
Gandy Predicted vs. Measured(52-4) qu :36.57 (tsf)
Using Nearest Boring qt : 13.94 (tsf)
300
Ei = 322,857 psi
FDOT failure Value RQD = 78 %
250
Predicted
Em = 251,828(psi)
End Bearing(tsf)
200
100
50
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Deflection(inches)
Predicted FDOT Failure Measured-52-4
Figure 6.12 Predicted vs. Measured End Bearing (52-4 in Gandy Bridge)
* Average
Gandy Predicted vs. Measured(91-4)
q u :4.32 (tsf)
Using Nearest Boring q t : 1.08 (tsf)
300
Ei = 274,136 psi
RQD = 52 %
250
FDOT failure Value
End Bearing(tsf)
200
150
100
Measured Load Test
91 - 4
50
Predicted
E m = 57,569(psi)
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Deflection(inches)
Predicted FDOT Failure Measured-91-4
Figure 6.13 Predicted vs. Measured End Bearing (91-4 in Gandy Bridge)
94
6.3.6.1 Site Description and Data Analysis for the nearest boring
A total of five Osterberg tests (3-1, 3-2, 10-2, 10-2, 19-1, and 19-2) were performed
at this site, and the FDOT failure value for the 10-2 shaft in this site was measured. The
nearest borings for the Osterberg Test of the 10-2 (ST. 99+31) shaft was estimated as the
boring TB-9 (ST. 99+70) which is composed of hard limestone, from an elevation of +37
feet to -15 feet. Through the boring logs for this site in the Geotech report, very hard
From the rock core samples of this nearest boring, the average of RQD values was
29.8 % and Recovery was 100 %. The averages of qu and qt values for core rock in the
For 24 rock core samples of the nearest boring, the tests for Initial Young’s
compressive strength) and Young’s Modulus value, Young’s Modulus value for the
limestone for the nearest boring was 2,880,476 psi. Em, Young’s Modulus of the rock
mass values for the limestone of the nearest borings were estimated by an adjustment
value based on RQD (FHWA-RD-95-172, 1996). The adjustment value for RQD value
The FDOT failure values in unit end bearing and maximum failure values for this
site were reported through four Osterberg load tests as shown in Table 6.1. Based on
FDOT failure criterion, predicted unit end bearing values for 10-2 was estimated by the
O’Neill method using the nearest boring data. The Result of Spreadsheet for decision of
unit end bearing of load-settlement is shown in Table 6.19. The predicted value for the
95
10-2 shaft, 159 tsf, is presented in Table 6.20, and the graphical representations for the
Table 6.19 Result of Spreadsheet for Decision of Unit End Bearing of Load-Settlement
Curve (10-2 in Victory Bridge)
Table 6.20 Prediction of Unit End Bearing Using the Nearest Boring (Victory)
300
End Bearing(tsf)
150
Measured Load Test
100 10-2
50
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Deflection(inches)
FDOT Failure Measured-10-2 Predicted
Figure 6.14 Predicted vs. Measured End Bearing (10-2 in Victory Bridge)
The principal goal of this study is to predict end bearing of shafts, and how it
compares with the measured end bearing from the load tests. Presented in Figure 6.15 is a
comparison between measured and predicted unit end bearing using the nearest boring
data of the shafts. The solid lines represent end bearing values for each shaft and the
dashed lines are measured end bearing values from the Osterberg load tests.
Based on the nearest boring data for load tests, a total of 11 of the unit end bearing values
was predicted, as shown in Table 6.21. The mean of measured end bearing values
was 65.7 tsf and the mean of predicted end bearing values was 58.6 tsf. The standard
deviation of measured end bearing values was 39.8 tsf and the standard deviation of
6.4 Prediction of the Unit End Bearing using Boring of All Site
The following is called the hit-or-miss procedure of the Monte Carlo method, or
simulation. Historically, the term “Monte Carlo” is believed to have been introduced by
John von Neumann as a code word connected with his secret work on the atomic bomb at
Los Alamos during World War II (Hammersley and Handscomb, 1964). The underlying
methodology had been known for many years, such as Buffon’s needle problem of 1777
1971). Present-day methods for generating uniform random (pseudorandom) numbers are
140
120
End Bearing (tsf)
100
80
60
40
20
0
LTSO 3 LTSO 4 46-11 62-5 69-7 LT3a LT4 52-4 91-4 Test 1 10-2V
17th ST Apalachicola Fuller Warren Gandy Acosta Victory
Bridge
Predicted End Bearing Measured End Bearing
Table 6.21 Measured and Predicted Unit End Bearing Using the Nearest Boring
The most widely used technique for generating random values of distributions
makes use of the cumulative probability distribution function (CPF), F (r) = P[x ≤ r]. By
definition, the CPF for any continuous variety is uniformly distributed over the interval
F(r) = Ru (0,1) would be a random value of the probability distribution function f(x)
whose CPF is F(r). Step 1 is the generation of the random value Ru (0,1), step 2 sets Ru
(0,1) = F(r), and step 3 determines x = r corresponding to F(r) for the particular
probability distribution.
The major application of the Monte Carlo technique is in the approximation of the
the simulation requires a high-speed computer so that a very large number of trials can be
made.
99
The SPSS (Statistical Package for Social Science) data analysis shows a lognormal
distribution is fitted for qu, qt and Ei and normal distribution for RQD. Using a Monte
Carlo simulation computer package with the given sample population and a lognormal or
normal probability density function, the best-fit field population (Figure.6.16 for qu;
Figure 6.17 for qt; Figure 6.18 for Ei; Figure 6.19 for RQD) was generated.
100%
90%
80% Sample Population
Mean = 26.9 Mean = 26.9
70%
St.D = 46.1 St.D = 46.7
Frequncy (%)
60%
50%
40%
30%
20%
10%
0%
0 7 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 119 126 133
qu (tsf)
Sample Random Field Population
Figure 6.16 Sample and Random Field Population for qu in Apalachicola Site.
100
100%
90%
80%
Sample Population
70% Mean = 3.0 Mean = 3.1
St.D = 4.2 St.D = 4.3
Frequncy (%)
60%
50%
40%
30%
20%
10%
0%
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48
qt (tsf)
Sample Random Field Population
Figure 6.17 Sample and Random Field Population for qt in Apalachicola Site.
100%
90%
80% Sample Population
Mean = 218,415 Mean = 223,107
70%
St.D = 374,392 St.D = 378,906
Frequncy (%)
60%
50%
40%
30%
20%
10%
0%
40K 120K 200K 280K 360K 440K 520K 600K 120K 760K 840K 920K
Ei (psi)
Sample Random Field Population
Figure 6.18 Sample and Random Field Population for Ei in Apalachicola Site.
101
100%
90%
80% Sample Population
Mean = 45.0 Mean = 45.1
70%
St.D = 22.8 St.D = 22.7
Frequncy (%)
60%
50%
40%
30%
20%
10%
0%
0 10 20 30 40 50 60 70 80 90 100
RQD (%)
Sample Random Field Population
Figure 6.19 Sample and Random Field Population for RQD in Apalachicola Site
The principal goal of this study is to predict end bearing of shafts, and how it
compares with the measured end bearing from the load tests. Presented in Figure 6.20 is a
comparison between measured and predicted unit end bearing for a whole site using the
To characterize the variability (standard deviation, Figure 6.20) of the end bearing
of a drilled shaft, it was recognized that it was not appropriate to use the standard
deviation of the laboratory data. Since a shaft represents a finite length/volume, it was
decided to randomly sample qu, qt, Ei, and RQD six to ten times (the number of pairs
should be a function of shaft embedment length) using the simple random selection
(SRSs) approach. Subsequently, the means of the six to ten laboratory values were
assumed to represent the average strengths, i.e. quavg, qtavg, Ei avg, and RQD avg, along a
102
drilled shaft. Next, another random six to ten samples were selected and another set of
average qu, qt, Ei, and RQD values were found. The process was repeated (500 to 1000
times) until an average qu, qt, Ei, RQD distribution and their associated standard
deviations were determined. Subsequently, the average end bearing value and its
associated standard deviation were found from the O’Neill method based on the quavg,
qtavg, Ei avg, and RQD avg, distributions. Figure 6.20 shows the mean and standard
deviation (S.D) values between predicted and measured unit end bearing.
From Figure 6.20, the bold solid lines represent the predicted end bearing average
values for each site and the thin solid lines are measured end bearing average values of
the site from the Osterberg load tests. The bold dashed lines represent the predicted end
bearing standard deviation values for each site and the thin lines are measured end
bearing standard deviation values of the site from the Osterberg load tests.
103
120
End Bearing (tsf)
90
60
30
0
17th st Apalachicola Fuller Gandy Acosta victory
Warren
Bridge
Predicted (Mean) Measured (Mean) Load test
Predicted (St. D) Measured (St. D)
Figure 6.20 Comparison of measured and predicted end bearing for Mean method
CHAPTER 7
LOAD AND RESISTANCE FACTOR DESIGN
In driven pile design, an important object is to prevent a limit state from being
reached. This object is traditionally implied by both Allowable Stress Design method (by
means of the safety factor Fs) and the Load and Resistance Factor Design method (by
means of the load factors γi and the resistance factor, φ). However, other objects that
must be considered and balanced in the overall design are function, appearance, and
economy.
The design of deep foundations has traditionally been based on ASD. For the
strength limit state in ASD, the estimated loads (or stresses) ∑ Q i are restricted as shown
below.
Rn
≥ ∑Q i Equation 7.1
Fs
Rn = Nominal resistance
where
104
105
QD = Dead load
QL = Live load
Rs = Side resistance
Rp = Tip resistance
recommend the use of load factors to account for uncertainty in the loads, and resistance
factors for the uncertainty in the material resistances. This safety criterion can be written
as
Rn = Nominal resistance
operational importance. The value of η usually ranges from 0.95 to 1.00. In this thesis,
η = 1.00 is used,
Qi = Load effect
γi = Load factor
Based on the current AASHTO recommendation, the following factors are used
If different resistance factors are used for tip and side resistance, then
106
where
Rs = Side resistance,
φs; φp= Resistance factors for side and tip resistance, respectively
• It accounts for variability in both resistance and load. (In ASD, no consideration is
given to the fact that different loads have different levels of uncertainty). For
example, the dead load can be estimated with a high degree of accuracy; therefore,
it has a lower factor (1.25) in LRFD.
• It achieves relatively uniform levels of safety based on the strength of soil and rock
for different limit states and foundation types.
• Using load and resistance factors provided in the code, no complex probability and
statistical analysis is required.
• Resistance factors vary with design methods and are not constant.
• The most rigorous method for developing and adjusting resistance factors to meet
individual situations requires the availability of statistical data and probabilistic
design algorithms.
107
adjusted by the past and future performance of foundations designed using that factor.
Second, the resistance factor φ is fitted through the factor of safety Fs and other
load parameters as
QD
γD +γ L
QL
φ= Equation 7.6
Q
Fs D + 1
QL
Rm
λRi = Equation 7.7
Rn
where
The mean, standard deviation, and coefficient of variation of the set of bias data λRi are
Mean λR =
∑λ Ri
Equation 7.8
N
∑ (λ − λR )
2
Standard deviation σR = Ri
Equation 7.9
N −1
108
σR
Coefficient of variation COV R = Equation 7.10
λR
The mean of the bias factor represents a trend between what is predicted and what
is measured.
Figure 7.1 presents the graph of the probability distribution function of g=ln(R/Q),
in which: R is the pile capacity, which is φ * Rn, and Q is the load effect, which is λQD *
When the pile capacity R is smaller than the load effect Q, then ln(R/Q) < 0. The
shaded area that ln(R/Q) < 0 is the failure region (pf). In pile foundation design, pf ranges
from 1*10-3 to 1*10-1, meaning that the probability of failure ranges from 0.1% to 10.0%.
This range is high because piles are usually grouped. The failure of one pile does not
The reliability index β is defined as the ratio between the lognormal mean, g , and
g
β= or g = β*ξg Equation 7.11
ξg
In Figure 7.1 if β is higher, then β*ξg is higher, the graph is stretched further to the
In 1972, Rosenblueth and Esteva (cited in Withiam et al., 1997) developed the
Table 7.1 Relationship between Probability of Failure and Reliability Index for
Lognormal Distribution (Rosenblueth and Esteva, 1972)
With that assumption, together with λR and COVR known from Section … and βT
QD 1 + COVQD2
+ COVQL
2
λ R (γ D +γ L)
QL 1 + COV R2
φ= Equation 7.13
(λQD
QD
QL
(
+ λQL ) exp β T ln[(1 + COV R2 )(1 + COVQD
2
+ COVQL
2
)] )
where
110
φ = Resistance factor
Table 7.2 λQD, λQL, COVQD, COVQL as recommended by AASHTO (cited in Withiam et
al., 1997)
Dead load
Factory-made 1.03 0.08
(QD)
Hansell and Viest, 1971 (cited in Withiam et al., 1997) developed the following
To estimate φ (resistance factor), the resistance’s bias factor (λR), and the
resistance’s coefficient of variation (COVR) calculated from measured average values and
predicted average values are required. The resistance’s bias factor (λR), and the
resistance’s coefficient of variation (COVR) computed from this research are shown in
Table 7.3.
Table 7.3 λmean, λst.d, and COV based on Predicted Values and Measured Values
factors to account for uncertainty in the loads, and resistance factors to account for the
uncertainty in the materials. Based on the nearest boring data for load test, a total of 11
end bearing values was predicted as shown in Table 6.21. From the 11 measured and
predicted unit end bearings from all the sites (Figure 6.15), the mean, λR, standard
deviation, σR, and coefficient of variation, COVR, were found for the FDOT/FHWA
design approach.
112
Using the computed mean, λR (1.40) and coefficient of variation, COVR (0.29), the
LRFD resistance factors, φ, were determined for different values of reliability index(β)
Using a dead load factor, γD, of 1.25, a live load factor, γL, of 1.75, and a dead to
live load ratio, QD/QL, of 2 (typically varies between 1 and 3), the Factor of Safety, Fs, in
Table 7.4 is obtained. Also note that the same probability of failure or risk associated
with the LRFD phi values apply to the ASD Factors of Safety.
7.3.2 φ factors and FS based on reliability using mean method of all sites
On the other hand, using the Monte Carlo simulation approach for all sites, the 21
measured unit end bearings and average predicted unit end bearings from all the sites are
compared (Figure 6.20). The mean, λR, standard deviation, σR, and coefficient of
Using the computed mean, λR (1.21) and coefficient of variation, COVR (0.46), the
LRFD resistance factors, φ, were determined for different values of reliability index(β)
Rock properties and the results of the Osterberg load tests for six bridge sites (17th
Street Causeway, Acosta, Apalachicola, Fuller Warren, Gandy, Victory) in this report
were collected to predict end bearing of the drilled shaft. The data contained 344
Unconfined compressive strength tests (qu), 188 Tensile strength (qt), 135 Young’s
modulus (Ei) and 11 Osterberg tests. Data on rock core properties was organized and
analyzed by each bridge, as well. From the analyzed rock properties, the variability of
rock strength of the bridge sites were identified according to ground elevation in Florida
limestone.
The results of eleven field load tests (Osterberg tests) for drilled shafts socketed in
six bridges were inspected for prediction of end bearing of the shaft. According to the
Osterberg test, the length of the shafts vary from 28 to 100 feet and the diameters of the
shafts range from 3 to 6 feet. Based on this data and rock core properties of the sites,
predicted end bearing values of the shafts were computed using the O’Neill method (FB-
Pier method) described in chapter 6. This method is commonly used as FDOT standard
practice.
In the O’Neill method, the Young Modulus of rock mass was ascertained as the
largest effective factor to determine predicted end bearing value of the drilled shaft. This
mass Modulus was obtained by the employment of the adjusted value (Table 6.2 in
Chapter 6) between RQD of the rock core and initial Young’s Modulus values. Two
approaching methods presented in chapter 6 were employed for the prediction of end
114
115
bearing. One is the nearest boring method from the shafts. The other is mean method of
all site. As shown in Figure 8.1 and 8.2, the results of predicted end bearing values were
compared with measured end bearing values from Osterberg load cell tests.
Using these predicted end bearing and measured end bearing values, LRFD
resistance factor and factor of safety based on reliability were calculated as shown in
In this research, the nearest boring data for predicted end bearing were used near
the shaft tested Osterberg load cell test within one station (100 feet) in the construction
plans. According to existing rock core data (qu, qt, Ei, RQD and REC) near the shaft at a
consider the variability of the rock for end bearing of the shafts in foundation design and
consider actual φ values or factors of safety of the shafts in Florida limestone that are
This research recommends that engineers obtain exact rock core data near the shaft
area for the drilled shaft design. In case where there is no rock information near the
conservative method.
116
140
120
End Bearing (tsf)
100
80
60
40
20
0
LTSO 3 LTSO 4 46-11 62-5 69-7 LT3a LT4 52-4 91-4 Test 1 10-2V
17th ST Apalachicola Fuller Warren Gandy Acosta Victory
Bridge
Predicted End Bearing Measured End Bearing
Figure 8.1 Comparison of Measured and Predicted End Bearing (Nearest Boring Method)
120
End Bearing (tsf)
90
60
30
0
17th st Apalachicola Fuller Gandy Acosta victory
Warren
Bridge
Predicted (Mean) Measured (Mean) Load test
Predicted (St. D) Measured (St. D)
Figure 8.2 Comparison of Measured and Predicted End Bearing (All Site Method)
117
Table 8.1 LRFD φ Factors, Probability of Failure (Pf) and FS Based on Reliability, β
(Nearest Boring Method)
Table 8.2 LRFD Phi Factors, Probability of Failure (Pf) and FS Based on Reliability, β
(Mean Method for All Site)
A.1 Summary of the Unconfined Compressive strength (qu) and Tensile Strength
(qt) Data
118
119
th
17 Street
State Project No. 86180-1522
BB-1 (32+93) S-4 (33+00)
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-65 32.2 30 22 -32 211.2 68.3
-72 27.34 67 28 -36 116.9 19.4
-85 114.2 13 7
-131 140.6 35 10
-131 64.6 35 10
A.1.2 Acosta
Acosta
State Project No. 87060-1549
WA-1 WA-2
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-35 109 47 -36 50.5 74
-37 120 74
WA-4L WA-5R
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-36 68 88 -34 52.5 100
-37 15 100
WA-6 WA-7
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 82 52 -16 30.5 30
-33 82 52
WA-8
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-40 38.5 60
Added Boring (130+34, 21' R+) Added Boring (150+35, 25' R+)
Dep.(ft) qu REC RQD qt REC RQD Dep.(ft) qu REC RQD qt REC RQD
-37 99.4 25 13 -22 53.6 100
332.1 -27 27.6
-42 274 -27 16.7 50
-42 28.8 49 38 -32 13.7
34.9
23.2
-47 20.9
122
BL-11 BL-13
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-35 92 25 25 -45 82 87 51
-35 13.9 25 25 -45 17.2 87 51
-45 68.5 30 17
-45 24.35 30 17
BL-20 BL-23
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-35 56.5 25 17 -20 95 50 47
-35 8.4 25 17 -20 9.9 50 47
-25 82 48 19
-25 12.95 48 19
BL-36 BL-37
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-20 244.5 70 49 -15 258 100
-20 54.05 70 49 -15 25.2 100
-25 104 100 -20 22.5 87 47
-20 5.6 87 47
-22 117.5 77 25
BW-1 BW-3
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 92.5 32 17 -45 28.75
-30 16.75 32 17
123
BW-5 BW-12
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-25 44.8 -35 31.85 27
-30 32.75 -40 27.4
-35 29.95
BW-14
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 43 20 42
-30 9.55 20 42
Added Boring (Hole 1: Near Boring BL-1) Added Boring (Hole 2: Near Boring BL-1)
Dep.(ft) qu REC RQD qt REC RQD Dep.(ft) qu REC RQD qt REC RQD
-20 596.2 38 38 -23 65.4 15 10
284 -28 30.8
401.6 -28 34.6 52 33
17.4 24.6
60.2 33.3
80.9 14
-25 110.5 11
-30 43.4 63 26 -33 58.5
32.2
52.9
87.7
15.8
6.6
-35 50.4
-35 87.7 73 66
85.1
77.9
20.4
29.4
12
23.6
17.9
18.3
18.2
47.8
-40 55.1
124
A.1.4 Apalachicola
Apalachicola
State Project No. 47010-3519
121+00 122+11
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-40 82.7 90 78 -20 2.8 95 58
-40 4 90 78 -20 0.8 95
-50 8.5 100 93 -30 64.2 83
-50 4.8 100 93 -30 34.2 83
-58 5.9 100 50
123+21 124+30
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-20 5.1 95 72 -25 5.8 60 30
-20 1 95 72 -25 1.1 60 30
-25 3.4 95 72 -30 6.2 80 40
-25 1.1 95 72 -30 0.8 80 40
-40 76.9 95 72 -40 10.8 100 50
-40 0.9 100 50
124+31 125+41
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-25 6.8 100 75 -15 5.5 80 26
-25 0.6 100 75 -25 4.4 100 75
-30 13.45 100 57 -25 0.7 100 75
-30 1.8 100 57 -35 4.5 80 25
-35 1.3 80 25
-50 7.4 100 60
125+42 126+52
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-15 6.7 100 64 -15 3.6 100 50
-25 11.4 100 67 -15 1.3 100 50
-25 1.9 100 67 -25 5.2 80 41
-45 7.9 100 70 -25 0.85 80 41
-45 1.3 100 70
127+60 127+64
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-15 4.1 80 21 -15 2.9 85 45
-15 0.7 80 21 -15 0.9 85 45
-30 36.4 70 45 -25 4.4 95 45
-30 3.7 70 45 -25 0.8 95 45
-35 208.7 35 25
-35 4.1 35 25
125
128+75 128+73
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-25 10.5 95 36 -25 1.2 100 67
-30 25.8 95 70 -25 1.8 100 67
-30 6.8 95 36 -30 16.1 100 60
-30 2.2 100 60
-35 4.9 100
129+84 129+88
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-25 8.5 90 40 -25 5.9 100 90
-25 0.8 90 40 -25 0.6 100 90
-45 10.5 100 60 -35 8 90 50
-45 3.5 100 60 -35 0.6 90 50
-50 66.2 65 65
132+05 133+16
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-25 5.9 80 50 -25 5.6 100 50
-25 1 80 50 -30 15.4 80 50
-35 7.4 95 73 -30 6 80 50
-35 0.6 95 73
134+26 135+36
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-20 6 100 66 -25 4.3 100 50
-20 1.2 100 66 -25 0.9 100 50
-45 102.6 80 80
-45 10.4 80 80
135+33 136+36
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-20 5.5 95 66 -30 3.55 30 0
-20 0.7 95 66 -30 1 30 0
-25 4.1 100 25 -40 66.2 30 22
-25 1.4 100 25 -40 21.9 30 22
-35 89.2 100 25 -50 95.8 67 47
-50 8.1 67 47
126
136+53 136+48
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-25 3.1 62 37 -35 7 45 7
-25 0.8 62 37 -45 69 50 18
-50 6.1 85 81 -55 11.4 93 73
-50 7.8 85 81 -55 1.1 93 73
-60 32.75 92 82 -60 13.7 78 46
-60 1.4 92 82 -60 1.5 78 46
136+34 138+64
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 3.95 80 42 -45 15.7 38 20
-30 1.45 80 42 -55 5.7 65 55
-35 15.2 80 22 -55 4.4 65 55
-40 9.65 80 -65 7.1 65 23
-40 1.2 80 -65 0.8 65 23
-50 117.5 45 28
138+84 138+82
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 2 63 47 -25 3.1 75 68
-30 0.2 63 47 -25 0.6 75 68
-55 150.55 45 42 -40 104.85 27 10
-55 21.8 45 42 -55 0.8 100 63
-65 5.6 83 72
-70 12.35 83
-70 1.3 83
138+60 141+40
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-45 37.05 37 25 -25 1 65 37
-45 7.5 37 25 -25 0.2 65 37
-60 0.2 100 82 -55 12.85 40 23
-55 4.85 40 23
-60 5.9 100 50
-60 1.2 100 50
-85 75.35 100 88
141+67 141+39
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 4.2 85 67 -30 3.2 50 45
-30 0.3 85 67 -30 0.4 50 45
-65 9.95 100 80 -45 60.65 27 13
-65 1.5 100 80 -60 9.8 100 77
-60 3.65 100 77
-70 14.3 100 85
-70 1.9 100 85
127
143+60 143+88
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-40 16.95 35 8 -52 157.4 50 0
-40 5.75 35 8 -52 15.9 50 0
-45 15.7 23 -65 6.2 93 82
-50 60.4 37 7 -65 0.9 93 82
-55 32.8 100 83 -70 4.1 100 82
-55 0.9 100 83
-60 8.7 100
-60 0.85 100
-65 9 92
144+81 144+86
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-25 5.3 100 59 -25 5.7 100
-25 0.8 100 59 -25 0.9 100
-30 3.5 100 88 -30 4.5 100
-30 0.4 100 88 -30 0.8 100
-40 5.9 90
-40 1.2 90
145+96 147+06
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 4.8 68 57 -30 4 75
-30 1 68 57 -30 1.1 75
-45 2.8 87 50 -40 2.5 75
-45 0.5 87 50 -45 174.4 30
148+15 149+21
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-25 5.7 68 41 -25 2.6 82 50
-25 0.8 68 41 -25 0.3 82 50
-35 7.7 60 41
-40 6 45 33
-40 0.2 45 33
149+30 150+34
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-45 5.9 73 50 -30 5.2 100 56
-45 0.7 73 50 -30 0.8 100 56
-40 5.4 64 34
128
151+44 152+54
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-27 6.5 62 35 -25 3.2 30 24
-27 1.1 62 35 -30 2.4 56 15
-40 9.7 53 37
153+63 153+63
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-25 3.8 82 68 -30 3.9 100
-25 0.9 82 68 -30 0.4 100
-35 6.3 58 26
-35 2.2 58 26
154+73 154+73
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 1.2 100 89 -30 2.6 50 32
-30 0.4 100 89 -35 8.2 57 13
-40 3.2 35 11
155+83 155+83
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-28 3.6 50 40 -32 5.1 60 32
-42 6.7 100 61 -32 1.2 60 32
-42 0.9 100 61 -40 1.3 28 14
156+92 157+91
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 5.5 80 67 -35 3.4 90 42
-30 0.6 80 67 -35 0.9 90 42
-35 5.8 40 40
-45 0.9 35 40
129
A.1.5 Gandy
Gandy
State Project No. 10130-1544
B-4 B-9
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-40 4.329 92 64 -29 8.2973 84
-40 48.341 92 64 -30 0.3608 84
-42 186.87 92 64 -32 2.2 84
-42 17.677 92 64
-45 46.176 92
B-15 B-21
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-30 20.202 100 -20 20.202 100
-30 3.2468 100 -25 322.51 100 50
-25 13.709 100 50
B-26 B-31
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-15 141.41 100 56 -29 36.075 100 28
-15 5.772 100 56
B-36 B-42
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-22 49.784 88 -39 38.961 82
-22 6.8543 88 -39 10.101 82
-23 9.7 88 -42 28.139 82 72
-23 5.7 88 -42 4.329 82 72
-24 33.911 88 -46 28.86 100 42
-25 23.81 88 86 -46 7.215 100 42
-26 28.86 88 42
-27 46.898 88
-27 15.873 88
-28 40.404 88
B-47 B-52
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-40 25.253 100 90 -40 75.758 90
-40 5.0505 100 90
-41 23.088 100
-41 15.152 100
-43 74.315 100
-43 9.0188 100
130
B-58 B-63
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-21 58.442 96 -45 169.55 64 29
-25 3.6075 96 58 -50 514.43 40
-50 394.3 40
B-68 B-74
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-37 46.176 40 -64 627.71 70
-64 67.1 70
B-75 B-86
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-80 18.759 32 14 -50 21.645 96 70
-80 5.4113 32 14 -51 98.846 96 70
-51 16.96 96 70
-52 167.39 96
-56 162.34 90
B-91 B-96
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
-72 4.329 70 52 -71 481.96 68
-72 1.0823 70 52 -73 41.847 68
-74 43.651 68
-75 553.39 68
131
A.1.6 Victory
Victory
State Project No. 53020-3540
TB-3 (90+38) TB-4 (92+60)
EL.(ft) qu REC RQD qt REC RQD EL.(ft) qu REC RQD qt REC RQD
35 47.4 50 11 40 52.38 21 8
30 10.67 40 8 24 18.47 74 48
30 17.6 40 8
TB-27 (131+65)
EL.(ft) qu REC RQD qt REC RQD
20 127.6 25 17
20 52.1 25 17
9 167.5 55
2 207.5 55 55
134
A.2 Summary of the Unconfined Compressive strength (qu) and Young’s Modulus
Data
Gandy Victory
Core Sample qu(psi) Ei(psi) Core Sample qu(psi) Ei(psi)
S1 670 333,333 V1 1,131 3,125,000
S2 2,590 1,133,333 V2 2,398 3,041,900
S1 30 600,000 V3 2,859 4,651,200
S1 280 200,000 V4 370 1,299,000
S1 4,470 833,333 V5 623 2,222,000
S1 1,960 909,091 V6 1,745 3,625,000
S1 500 129,630 V7 1,943 3,333,300
S1 690 500,000 V8 1,735 4,286,000
S2 470 333,333 V9 1,397 2,186,000
S3 330 166,667 V16 380 1,923,000
S4 400 214,286 V17 350 1,800,000
S5 650 400,000 V18 87 715,500
S1 540 363,636 V20 1,330 3,496,000
S2 390 166,667 V21 1,703 2,200,000
S3 400 100,000 V22 2,484 2,216,912
S1 350 200,000 V28 1,225 3,073,508
S2 320 222,222 V29 181 479,000
S3 1,030 625,000 V30 350 1,429,000
S1 810 269,231 V31 1,380 3,226,000
S1 2,350 833,333 V32 1,393 3,896,000
S2 7,130 1,578,947 V33 1,220 2,985,000
S1 640 235,294 V34 910 2,325,600
S1 8,700 1,153,846 V35 515 2,294,000
S1 300 16,129 V36 1,806 4,396,000
S2 2,320 1,000,000
S3 2,250 416,667
S1 7,670 1,000,000
S2 7,670 1,400,000
APPENDIX B
SHAFT DIMENSIONS AND ELEVATIONS
B.1 Note
Bottom of Casing: the elevation of the bottom of casing in the test shaft.
Last Strain Gauge Elevation: the elevation of the lowest stain gauge that measures the
load transfer. For Osterberg, commonly the Osterberg cell is located at the lowest
location.
Total Rock Socket: the length of the shaft from the top of rock elevation to tip elevation.
Embedded Length: the length of the shaft from the ground elevation to the tip of the
shaft.
137
138
Crowther, Carroll L. Load Testing of Deep Foundations The Planning, Design, and
Conduct of Pile Load Tests, John Wiley & Sons, New York, 1988.
Goodman, Richard E. Introduction to Rock Mechanics, 2nd Edition. John Wiley & Sons,
New York, 1989.
Hoit, M.I. and McVay, M.C. (1996). “FB-Pier User’s Manual,” Department of Civil
Engineering, University of Florida, Gainesville.
Justason, M.D., Mullins, G., Robertson, D.T., and Knight, W.F. (1998). “A Comparison
of Static and Load Tests in Sand: A Case Study of the Bayou Chico Bridge in
Pensacola, Florida,” Second International Statnamic Seminar, Canadian Embassy
of Japan, Tokyo.
Matsumoto, T., and Tsuzuki, M. (1994), “Statnamic Tests on Steel Piles Driven in a Soft
Rock,” International Conference on Design and Construction of Deep Foundations,
U.S. Federal Highway Administration, Orlando.
McVay, M.C. (1991). “Design of Socket Drilled Shaft in Limestone,” Report No.
4910450425612, University of Florida, Geotechnical Engineering Group,
Gainesville, Florida, July.
McVay, M.C. (1998). “LRFD Assessment of Static Pile Capacity from Dynamic
Methods and The determination ok Skin and Tip Resistance from PDA Signals,”
Report No. 4940450460312, University of Florida, Geotechnical Engineering
Group, Gainesville, Florida, November, pp. 34-100.
McVay, M.C, and Townsend, F.C. (1990). “Design Guidelines for Shafts Foundations,”
Report No. 4910450425612, University of Florida, Geotechnical Engineering
Group, Gainesville, Florida, July, pp. 276.
McVay, M.C., Townsend, F.C., and Williams, R.C. (1991). “Design of Socket Drilled
Shafts in Limestone,” Ninth Panamerican Conference on Soil Mechanics and
Foundation Engineering, Vina del Mar, Chile.
151
152
Middendorp, P. and Bielefeld, M.W. (1995). “Statnamic Load Testing and the Influence
of Stress Wave,” First International Statnamic Seminar, Vancouver, British
Columbia, Canada, September.
Moore, David S. The Basic Practice of Statistics, W.H. Freeman and Company, New
York, 1995.
O'Neill, M. W., and Hassan, K. M. (1994). “Drilled Shafts: Effects of Construction and
Performance and Design Criteria,” in Proceedings, International Conference on
Deep Foundations, Orlando, Florida, Federal Highway Administration, December,
pp. 137-187.
O'Neill, M. W., Townsend, F. C., Hassan, K. H., Buller, A., and Chan,'P. S. (1996).
“Load Transfer for Drilled Shafts in Intermediate Geomaterials,” Report No.
FHWA-RD-95-1 72, Federal Highway Administration, Washington, DC. January.
O’Neill, M. W., and Reese, L. C (1999). “Drilled Shafts: Construction and Procedures
and Design Methods,” FHWA-IF-99-025, Federal Highway Administration,
Washington, DC. August.
Osterberg, J. (1989). “New Device for Load Testing Driven and Drilled shafts Separate
Friction and End Bearing,” Proceedings of the International Conference on Piling
and Deep Foundations, London, pp. 421-427.
Withiam, J., Voytko, E., Barker, R., Duncan, M., Kelly, B., Musser, S. and Elias, V.
(1997). “Participant Workbook Load and Resistance Factor Design (LRFD) for
Highway Bridge Substructures,” FHWA, Washington, DC.
BIOGRAPHICAL SKETCH
Sang-Ho Kim was born at Jeonju City in Jeollabuk-Do, South Korea, in 1970.
Sang-Ho is the oldest of three sons in a Christian family. He attended elementary, middle,
also served in the military for three years during his bachelor’s study. After graduating,
he began working as a civil engineer at ByuckSan Construction Co. Ltd in Seoul, South
Civil and Coastal Engineering Department at the University of Florida. For one and one
half years, he has studied the research of the FDOT with Dr. McVay.
In August 2003, he will graduate with a Master of Science degree. In the fall of
2003, he will move somewhere and begin his career in the profession of civil engineering.
153