Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

DRILLED AND DRIVEN FOUNDATION BEHAVIOR

IN A CALCAREOUS CLAY
W. M. Camp, III, P.E., Member, Geo-Institute1
ABSTRACT: Most major structures in the Charleston, SC area of the U.S. are
supported on deep foundations bearing in the Cooper Marl. The fine grained Cooper
Marl, which typically classifies as highly plastic clay, has a calcium carbonate content
of 60 to 80 percent. Unlike some other calcareous soils, the Cooper Marl is generally
an effective foundation material for both driven and drilled foundations. In general,
the performance of driven and drilled foundations is comparable but the marls
cementation does cause some differences. The relatively open structure collapses
upon shearing which causes large excess pore pressures. As a result, the capacity of a
driven pile is highly time dependent and increases as the excess pore pressures
dissipate. Vibratory installation, which generates a large number of shear reversals, is
particularly damaging to the soil structure and negatively impacts the performance of
foundations installed in this manner. Drilled foundations often use a vibratory
installed steel casing as a construction aid and performance data consistently indicate
that the cased zone of the shaft within marl, even when the casing was extracted
during concrete placement, experiences a reduction in side resistance as compared to
an uncased excavation.
INTRODUCTION
Charleston, South Carolina, lies in the Coastal Plain of the Southeastern United
States. Locally, the greater Charleston area is appropriately known as the Low
Country. The elevation of the high-ground is typically less than 5 m and saltwater
marshes, tidal creeks, and rivers are abundant. As to be expected for such a
topography, the shallow subsurface conditions (i.e., to depths of 10 to 15 m)
frequently consist of very soft clays and loose sands. This is particularly true of much
of todays undeveloped land since the best land was developed first. As an additional
complication, the region, which experienced a major earthquake in 1886, is
considered one of the most seismically active in the eastern U.S.
The population of the Low Country has grown considerably in the last few decades
and this growth has produced a corresponding increase in commercial, industrial, and
1

Senior Engineer, S&ME, Inc., 620 Wando Park Blvd., Mount Pleasant, SC 29464

1
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

infrastructure construction projects. As a result of the difficult geotechnical

conditions, deep foundations are commonly required in the region and the vast
majority of these deep foundations are supported in a common bearing stratum, the
Cooper Marl. The site characterization and foundation testing performed in
conjunction with the numerous construction projects over the last few decades have
produced data that are valuable in understanding the behavior of this very interesting
formation. Within this paper data from various projects are combined to better
understand deep foundation performance in the area. Specifically, differences in the
behavior of drilled foundations as compared to driven foundations are examined.
GEOLOGY
The Coastal Plain in South Carolina consists of a wedge of late Cretaceous and
younger sediments that thickens to the southeast, or in a direction normal to the
present day shoreline (Heron, 1962). The marine sediments within this wedge
primarily consist of non-indurated siliciclastic materials and carbonates with varying
quantities of terrigenous matter (Horton and Zullo, 1991). Within the Charleston
area, the first formation within this wedge with a thickness, uniformity, and strength
capable of providing foundation support is the Cooper Marl. The Cooper Marl
nomenclature was created in 1848 (Tuomey, 1848) and later became the Cooper
Formation (Ward et al., 1979). In 1984, the deposit was elevated to group rank and it
is now formally known as the Cooper Group, which consists of three formations: the
Oligocene Ashley Formation, the Late Eocene Parkers Ferry Formation and the
Eocene Harleyville Formation (Weems & Lemon, 1984).
In 1848 when Tuomey named the Cooper Marl, the term marl was used to
describe any deposit containing economically important quantities of carbonate of
lime (for use as fertilizer) that was soft enough to be excavated with the ordinarily
available equipment of the day (Heron, 1962). Today, according to the Glossary of
Geology, the term marl is used to refer to a material that contains significant
quantities of both carbonates and clay. Although meeting the original definition, the
Cooper Marl clay content is too small to meet todays definition of marl.
Nevertheless, the geotechnical and construction community generally continues to
refer to all members of the Cooper Group as the Cooper Marl. Within the depths of
interest, the engineering differences between the formations are relatively minor and
the use of the older and now more generic term is an acceptable convention.
The geologic literature generally classifies the Cooper Marl as a calcarenite or
calcilutite and lists its constituents as the skeletal remains of microscopic sea
organisms (alternatively microfossils or foraminifers), quartz sand, phosphate and
clay minerals (Weems and Lemon, 1989; Weems and Lemon, 1996; Ward et al.,
1979; Gohn et al., 1977). The abundance of microfossils is readily apparent in the
scanning electron micrograph of a Cooper Marl sample shown in Figure 1.

2
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

FIG. 1. SEM Photograph of a Cooper Marl Sample

GEOTECHNICAL CHARACTERIZATION
The available Cooper Marl data, with regard to both site characterization and
foundation performance, have primarily been generated through consultants working
on various industrial and government projects (relevant reports listed in References).
Consequently, for any given project, with a few exceptions, the data are relatively
limited. Some of the deficiencies can be reduced, however, by combining the data
from multiple projects and sites. The creation of such a database must be based on
the premise that the properties and characteristics of the Cooper Marl are relatively
uniform across the Charleston area. For decades, the local geotechnical state-of-thepractice has been to treat the Cooper Marl as a relatively uniform stratum and
experience has generally justified this approach. And in fact, a comparative review of
the data from across the area, some of which are presented herein, does support the
premise of a reasonable areal uniformity.
For decades, subsurface conditions in the area were explored by standard penetration
testing (SPT) with split spoon sampling supplemented with undisturbed samples and
laboratory testing. The SPT N-value of the Cooper Marl is usually in the range of 10
to 20 and the samples are visually classified as stiff to very stiff, olive, sandy clay or
sandy silt. During the last 10 years, the area has begun to use cone penetration testing
(CPT) on a regular basis and a representative sounding is shown in Figure 2.

3
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Tip Resistance

Pore Pressure
U 2 (kPa)

q T (MPa)
10

20

0
0

1000 2000 3000 4000

0.0

2.0

4.0

Shear Wave Velocity

Vs (m/s)
6.0

10

10

10

10

12

12

12

12

14

14

14

14

16

16

16

16

18

18

18

18

20

20

20

20

22

22

22

22

24

24

24

24

26

26

26

26

28

28

28

28

30

30

30

30

Depth (m)

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

0
0

Friction Ratio
FR (%)

u2
u0

200

400

600

800

FIG. 2. Representative CPT Sounding with Top of Cooper Marl at 8 m.

The most notable feature of the CPT sounding is the large increase in the u2 pore
pressure, which identifies the top of the Cooper Marl. This is a consistent
characteristic of the Cooper Marl and is seen in all cone data from the Cooper Marl.
The tip resistance within the marl is relatively uniform, although thin stiffer lenses
are occasionally encountered as seen at depths of 19 m and 24 m in Figure 2. It
should be noted that there is a corresponding pore pressure increase at each of these
depths, which is also a typical characteristic of CPT data in the Cooper Marl.
A substantial number of deep (i.e., 30 m) CPT soundings are available from
projects throughout the Charleston area. The Cooper Marl data (i.e., data from
formations above the Cooper Marl are not included) from five of these sites are
plotted versus elevation in Figure 3. For sites with multiple soundings, the data were
first averaged together to produce a representative sounding for the site. Even though
the data were collected across an approximately 400 km2 area, the variations between
the soundings are relatively small, which indicates that the Cooper Marl may be
considered a relatively uniform deposit.

4
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Tip, q T (MPa)
10

Pore Press. U 2 (kPa)

20

2500

5000

7500

Friction Ratio, FR (% )

Sleeve, fs (kPa)
50

100 150 200

-10

-10

-10

-10

-20

-20

-20

-20

-30

-30

-30

-30

-40

-40

-40

-40

-50

-50

-50

-50

-60

-60

-60

-60

Elev (m)

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

FIG. 3. Cooper Marl CPT Data from 5 Different Sites

The large pore pressure response of the CPT within the Cooper Marl is due to
cementation, which also helps explain many other aspects of the Cooper Marl
characteristics. The data shown in Figures 4 through 6 support this conclusion. A
primary constituent (60 to 80 percent) of the Cooper Marl is calcium carbonate (Fig.
4) which is a common cementing agent. Additionally, the maximum preconsolidation
stress (Fig. 5), as measured by oedometer tests, is substantial (resulting in
overconsolidation ratios of 3 to 6) and difficult to explain by conventional
consolidation. The area has not been subjected to glaciation since the deposition of
the Cooper Marl nor is there any evidence of desiccation. And to completely account
for the level of preconsolidation, more than 50 m of soil would have had to have been
eroded from the Charleston area during the Miocene and/or Pliocene Epochs. The
geological evidence does not support such an erosional event(s). Further, the
preconsolidation stress, albeit scattered, is essentially constant with depth, which is
consistent with an apparent overconsolidation due to cementation. The shear wave
velocity (Fig. 6) is also roughly constant with depth, or stress independent, as is
typical of cemented soils (Rinaldi & Santamarina, 2003). Finally, the fabric of the
Cooper Marl is relatively open (void ratio of 1 to 2, moisture content of 40 to 60
percent) yet its shear wave velocity is in the range of 500 m/s and its undrained shear
strength (from triaxial CIUC testing) is typically around 200 kPa (and is also
generally constant with depth).

5
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

-10

-10

-15

-15

-15

-20

-20

-20

-25

-25

-30

-30

-35
-40

Vs

Vp

-25

Elevation (m)

Elevation (m)

Elevation (m)

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

Copyright ASCE 2004

-10

-35
-40

-30
-35
-40

-45

-45

-45
-50

-50

-50
-55

-55

-60

-60
0

20

40

60

80

Percent CACO3

FIG. 4. CACO3 Content

100

-55
0

500

1000

1500

Max Precon Stress (kPa)

FIG. 5. Preconsolidation Stress

-60
0

1000
2000
Velocity (m/s)

FIG. 6. Velocity Profiles

Geo-Support 2004

6
GeoSupport 2004

Camp

Particle Size Analysis

100%

1.5" 1" 3/4"1/2" 3/8" #4

#10

#20

#40 #60 #100

#200

90%
80%

Percent Passing

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

During cone penetration testing (or pile driving or any other shearing), the soil
skeleton is stressed until the cementation bonds are broken. Once broken, the
skeleton collapses and forms new particle contacts. However, the sudden collapse of
the cementation bonds in the saturated soil results in an increase in pore pressure,
hence the large pore pressure measurements during cone penetration testing.
Another notable aspect of the Cooper Marl is its clay-like behavior. Grain size data
are summarized in Figure 7. The fines content is usually in the range of 60 to 90
percent and the clay size fraction is generally in the range of 10 to 30 percent. As
shown in Figure 4, the calcium carbonate content is generally in the range of 60 to 80
percent and therefore, a significant portion of the fines content and clay size fraction
must be carbonate material rather than clay minerals. This is surprising considering
the highly plastic nature of the Cooper Marl which is illustrated by the data from
multiple projects plotted Figure 8. Although there is scatter within the data, there are
no apparent trends with respect to project location. In other words, all projects tend
to have a similar amount of scatter. In general, the material is highly plastic with
liquid limits often greater than 100 and PIs greater than 60.

70%
60%
50%
40%
30%
20%
10%
0%
100

10

0.1

0.01

0.001

Particle Size (mm)

Fig. 7. Summary of Cooper Marl Grain Size Distribution Data

The seemingly contradictory properties (high plasticity and low clay content) can be
explained in two ways. First, X-ray diffraction analyses confirm that the Cooper
Marl contains a relatively small amount of phyllosilicates; 4 to 9 percent by weight.
However, smectite is one of the primary clay minerals found in the Cooper Marl
accounting for approximately 20 to 60 percent of the total phyllosilicate content. The
highly plastic properties of smectite (synonymous with montmorillonite) are well
known. Additionally, in SEM photographs of the Cooper Marl, many of the
microfossils appear to be hollow. The interstitial water trapped in these hollow

7
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

particles will be included in laboratory moisture measurements but it will not

necessarily be free to influence the soils behavior.

120
110
100
CH or OH

80
PLASTICITY INDEX, PI

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

90

"U" Line
[PI=0.9(LL-8)]

70
"A" Line
[PI=0.73(LL-20)]

60
50
40
30
20
CL or
OL

10
CL-ML

ML or OL

0
0

10

MH or OH

20

30

40

50

60

70

80

90

100

110

120

130

140

150

160

170

LIQUID LIMIT, LL

Fig. 8. Cooper Marl Plasticity Data

Finally, the Cooper Marl has a surprisingly high effective stress friction angle, 44
degrees as measured in triaxial CIUC tests. This is a result of the rough, irregular
shapes of the microfossils and other particles, as pictured in the SEM micrograph
shown in Figure 1.
LOCAL DEEP FOUNDATION PRACTICES AND OBSERVATIONS
An overview of the local deep foundation practice, in terms of both driven piles and
drilled shafts, as well as common construction observations is an informative start to
a comparison of drilled and driven foundations in the Cooper Marl.
Driven Foundations
The most common driven pile foundation used in the Charleston area is a square
prestressed concrete pile. Sizes range from 254 mm to 610 mm. Steel H-sections
and open-end pipe piles are also used but less frequently. Pile driving into the
Cooper Marl is generally very easy. Relatively small hammers are capable of driving
piles to depths that ultimately yield a high capacity even though the final driving

8
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

8
7
6
Qbor/Qeod

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

resistance may be quite small (often less than 60 blows per meter). As seen during
cone penetration testing, excess pore pressures are large at the time of pile
installation. Pore pressure measurements made within the Cooper Marl around
driven piles have confirmed that pore pressures are suddenly and significantly
elevated (Camp et al., 1992) during pile installation. As a result of the reduction in
the effective stress, the pile capacity at the time of installation is very small but
increases as the pore pressures dissipate. As shown in Figure 9, the capacity gain
with time is substantial and the beneficial impact must be considered for an
economical foundation design. Conversely, the low capacity available soon after pile
installation must be considered when the piles are loaded shortly after installation
(e.g., top-down construction, temporary support of construction equipment).

5
4
3
2
1
0
0

10

15

20

25

Time Since Installation (days)

Fig. 9. Ratio of Capacity at the Beginning of Restrike to the Capacity at the Endof-Drive for Piles Driven into the Cooper Marl (Camp & Parmar, 1999)
Other, more detrimental effects of the driving-induced pore pressures are sometimes
seen as well. In particular, two problem scenarios occasionally develop. During
pile installation, pile bounce may occur. The term bounce in this situation refers
to an extremely high quake condition in which the pile moves down under a given
hammer blow a significant distance (e.g., 30 mm) and then rebounds an equal
amount during the hammer upstroke. The pile is essentially acting like a large elastic
spring. Although proper hammer selection can help avoid the problem, the typical
solution is to move on to the next pile and drive it until it also refuses due to bounce.
This process is continued until eventually, the initial pile is driven once again.
Usually, after 1+ hours, the pile can be driven some additional depth before bounce
will once again occur. This problem is only seen with displacement piles and in
particular with the larger displacement piles.
9

Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Load (MN)
0.0

0.5

1.0

1.5

2.0

5
Pile Head Deflection (mm)

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

Secondly, with open-end pipe piles and cylinder piles, a head of very wet soil or
water may develop within the pile. In other words, the level of the soil and/or water
within the pipe pile or cylinder pile may exceed the level of the existing ground
surface. An extreme example of this phenomenon was observed during the
installation of a 914 mm diameter concrete cylinder pile (127 mm wall thickness).
The water rose within the interior of the pile to a height of more than 9 m (Soil &
Material Engineers, 1984). The water was prevented from going higher by the
hammer helmet and the resulting pressure became so high that vent holes had to cut
near the pile top to relieve the water pressure.
In the bounce scenario, large pore pressures are generated but cannot dissipate and
in the pipe pile case, the large pore pressures are dissipated by flowing up through the
interior of the pile.
A representative load-deflection curve for a statically top-loaded driven pile is
shown in Figure 10. The shape of the curve is typical for types of driven piles
bearing in the Cooper Marl. The pile initially exhibits a relatively stiff response, then
softens considerably and eventually plunges.

10

15

20

25

30

FIG. 10. Representative Load-Deflection Response of an Axially-Loaded, 305

mm square PSC Driven Pile
Drilled Foundations
Drilled shafts are the most commonly used drilled foundation in the Charleston area.
Auger cast-in-place (ACIP) piles have occasionally been used on commercial projects
but they typically bear within the soils above the Cooper Marl or at very shallow
10
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Stage 1 - Lower O-Cell Pressurized

20
0

Movement (mm)

-20
-40
-60
-80
-100
Top of Lower O-Cell

-120

Base of Lower O-Cell

-140
0

10

12

14

16

18

Lower O-Cell Load (MN)

Stage 2 (Upper O-Cell Pressurized) and 3 (both O-Cells Pressurized)
60.0
40.0
20.0
Movement (mm)

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

depths within the Cooper Marl and relevant data on ACIP piles are therefore limited.
Most drilled shafts in the area are constructed without the use of drilling fluids.
Typically, casing is driven into the top 1 to 3 m of the Cooper Marl. This eliminates
any potential shaft wall stability problems within the very weak overburden soils and
effectively forms a seal within the marl. Over typical foundation depths, the Cooper
Marl is easily excavated with conventional augers or digging buckets. The sidewalls
remain stable and bells have been successfully constructed. Water, polymer slurries,
and mineral slurries have been used as drilling fluids as a precaution for very deep
shafts and for commercial projects where the use of casing is undesirable. As
presented in Camp et al. (2002), any influence of the type of drilling fluid (or lack
thereof) on the shaft capacity is negligible.
The Osterberg-CellTM load test method has been used on the majority of the tested
shafts and many have used two levels of O-Cells. Representative load-deflection
curves from a multi-level O-Cell test are shown in Figure 11.

0.0
-20.0
-40.0
-60.0
-80.0

Top of Upper O-Cell

Base of Upper O-Cell

-100.0
0

10

15

20

25

30

Upper O-Cell Load (MN)

FIG. 11. Representative Load-Deflection Curves from a 2.4 m diam. Drilled

Shaft loaded with 2 Levels of Osterberg CellsTM.
11
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

The axial performance of driven and drilled foundations bearing within the Cooper
Marl has been evaluated on many projects by load testing instrumented piles. The
majority of the available results are from O-Cell tests but the driven pile data and two
drilled shaft data sets are from conventional top-down load tests. From these load
tests, average unit side shear and unit end values have been back-calculated for
various foundation segment lengths. The available data have been reviewed and
summarized in Figures 12 and 13 as plots of unit end resistance versus the vertical
effective stress at the pile tip elevation and unit side shear versus the vertical effective
stress at the elevation of the midpoint of the relevant side shear segment.

Unit End Bearing (MPa)

0.0
Effective Stress at Tip Elevation
(kPa)

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

DEEP FOUNDATION PERFORMANCE

1.0

2.0

3.0

4.0

5.0

6.0

Drilled Shafts
Driven Piles

50
100
150
200
250
300
350
400
450

Fig. 12. Unit End Resistance (mobilized at a normalized tip displacement of 2%)
vs Effective Stress for Drilled and Driven Piles Bearing within the Cooper Marl.
There is obvious trend between the unit end resistance and the effective stress at the
pile tip elevation. This is true for both the driven pile and the drilled shaft data.
Since the undrained shear strength is relatively constant with depth, the end bearing
results indicate that the unit end resistance can be estimated using an effective stress
approach:
qeb = Nqvo

(1)

where the bearing capacity factor, Nq, is approximately 10.

12
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Unit Side Shear (kPa)

0

50

100

150

200

250

300

Effective Stress at Segment

Midpoint (kPa)

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

0
Driven Piles

50

Drilled Shafts

100

Drilled Shaft Casing

150
200
250
300
350

Approx range of dynamic

testing data

400

Fig. 13. Unit Side Shear Resistance vs Segment Midpoint Effective Stress for
Drilled and Driven Piles Bearing within the Cooper Marl.
As shown in Figure 13, the unit side shear for driven and drilled piles appears to
increase only slightly at higher effective stress, although there is large scatter within
the data. The driven pile data plotted in Figure 13 are from instrumented static load
tests and as a result, are relatively limited. However, unit side shear data have been
estimated from the hundreds of dynamic tests performed on driven piles in the
Charleston area and the resulting values, whether from CAPWAP analyses or
conventional static back-calculation, fall within the range of about 120 to 180 kPa.
Therefore, it appears that the driven pile unit side shear values are generally within
the range of the lower bound of the drilled shaft unit side shear values.
The third data set shown in Figure 13 is for drilled shaft casing. Of course a drilled
shaft casing is simply a driven open-end pipe pile but on some shafts, the casing was
removed after concrete placement. The drilled shaft casing data plotted in Figure 13
includes the unit side shear calculated for the segment of the shaft that was below the
top of the Cooper Marl but within the zone of the casing, regardless of whether the
casing was permanent or temporary. Additionally, unlike the other driven pile data,
the drilled shaft casings were always installed with a vibratory hammer, which is also
the reason why the unit side shear values within the cased zone are lower.
Canivan and Camp (2002) presented data from two small dynamic pile testing
programs specifically designed to evaluate the differences in capacity of open-end
pipe piles driven with impact hammers versus identical piles driven with vibratory
hammers. Both programs were performed in the Charleston area and the bearing
stratum was the Cooper Marl. As shown in Figure 14, there is a marked difference
between the capacities of the piles installed with the impact hammer as compared to
those installed with a vibratory hammer. At any given time, the piles installed with

13
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

the impact hammer had capacities that were at least 30% and, in some cases, more
than 200% higher than the companion piles installed with a vibratory hammer.
Pile A with Vibro Hammer
Pile A with Impact Hammer
Pile B with Vibro Hammer
Pile B with Impact Hammer
Pile C with Vibro Hammer
Pile C with Impact Hammer

Capacity (moblized), MN

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

7
6
5
4
3
2
1
1

10

100

Time Since End of Installation, hrs

Fig. 14. Comparison of Pile Capacity vs Time for Identical Open-End Pipe Piles
(diameters of 610 mm or 762 mm) Driven with Vibratory or Impact Hammers
into the Cooper Marl (Canivan and Camp, 2002).
The difference in capacity between the two installation methods was attributed to
differences in the level of disturbance or remolding. Specifically, under the vibratory
hammer, a pile will experience a much greater number of load-unload cycles and the
degree of shear reversal during the unload cycles would be much larger as compared
to conditions under an impact hammer loading. For a cemented material like the
Cooper Marl, the more severe loading induced by the vibratory hammer would likely
be even more detrimental (i.e., larger pore pressures, a greater degree of
destructuring).
The multi-level Osterberg-Cell testing affords an opportunity to evaluate the
consequences of cyclic loading along the shaft interface and several tests have been
performed on shafts bearing within the Cooper Marl to evaluate the effects of cyclic
loading with and without shear stress reversal. Specifically, three shafts constructed
for the Cooper River Bridge Load-Test program were subjected to supplemental load
sequences after the completion of the specified testing. Details of the load test
program can be found in Camp, Brown and Mayne, 2002; Camp, Mayne, and Brown,
2002; and Brown and Camp, 2002. The supplemental testing consisted of pushing
the socket (i.e., the portion of the shaft located between the upper and lower level
of O-cells) up and down between the two levels of O-cells.
For one shaft, the load sequence consisted of three load-unload cycles following the
completion of the Stage 3 loading (i.e., the final stage in a conventional two-level Ocell test in which both levels are pressurized resulting in an upward failure of the
portion of the shaft above the uppermost O-cell). The direction of shear along the
14

Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

socket interface for all three load cycles was therefore downward, which was the
same as the direction at the end of the Stage 3 loading. At the completion of the third
load-unload cycle, the shaft was loaded from the bottom and the socket was subjected
to three more load-unload cycles, but in the opposite direction.
For the other two shafts, at the completion of the Stage 3 loading, the shaft was
loaded from the bottom, subjecting the socket to a load in the opposite direction of
the Stage 3 load. At the completion of the bottom loading, the shaft was again loaded
from the top (i.e., from the upper level of O-cells), thereby subjecting the socket to
another shear reversal. This process was repeated two more times.
Additionally, at the start of the construction of the new Cooper River Bridge, the
designers (Parsons Brinckerhoff Quade and Douglas) had two more test shafts
constructed and loaded. Both shafts were 1.5 m in diameter and were subjected to
multi-level O-cell testing (Loadtest 2002a and Loadtest 2002b). Both of these shafts
were subjected to a second cycle of loading at each O-cell level. In particular, after
mobilizing the ultimate side shear on the segment of the shafts above the upper level
of O-cells, the segment was unloaded and loaded again. As with one of the Cooper
River Bridge Load-Test program shafts, this resulted in a second load cycle in the
same direction as the first load cycle.
The data from these five test shafts are summarized in Figure 15. For each cycle,
the load was increased until the shaft continued moving under a constant load or in
some cases, a decreasing load. This loading, normalized to the peak loading obtained
during the first load cycle, is plotted in Figure 15. The solid lines connect points that
were not subjected to shear stress reversals and the dashed lines connect those that
were.
The cyclic loading with full shear reversal is obviously much more detrimental to
the shaft capacity than cyclic loading without the shear reversal. These data support
the vibratory hammer/impact hammer hypothesis and help explain why the unit side
shear within the cased zone is so much lower. And the fact that the side shear is still
reduced even when the casing is removed indicates that the reduction is not
attributable to an interface mechanism.
CONCLUSIONS
The available in situ and laboratory testing data indicate that the Cooper Marl is
cemented. As a result of the cementation, the deposit has a relatively open structure,
as indicated by a void ratio of 1 to 2, but also has a high small strain stiffness, a
relatively large undrained shear strength, and is capable of providing significant
support for foundations. Upon shearing, the structure collapses, which produces very
large excess pore pressures. The pore pressure generation makes driven pile capacity
highly time dependent and sometimes creates installation difficulties. In general,
driven and drilled foundation performance is comparable although the unit side shear
resistances tend to be slightly higher in the drilled piles. As evidenced by differences
in the performance of impact driven and vibratory driven piles, shear stress reversal
appears to have a significant detrimental effect on the unit side shear within the
Cooper Marl. Cyclic testing of drilled shafts with Osterberg-cells confirms that the
unit side shear is greatly reduced under conditions of shear stress reversal. Even
15

Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

without shear stress reversal under working loads, the consideration of the reduction
is important since it appears within the cased zones of drilled shafts as a result of the
vibratory installation (and sometimes removal) of the steel casing.

Unit Side Shear Ratio (current fs/initial fs)

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

1.2

1.0

0.8

0.6

0.4
C1 Top Ld
C1 Bottom Ld
C2 Bottom Ld
DI-1 Top Ld
DI-1 Bottom Ld
1E Bottom Ld (upper cell)
1W Bottom Ld (upper cell)

0.2

0.0
1

Number of Cycles

FIG. 15. Summary of Cyclic Loading with (Dashed Lines) and without (Solid
Lines) Shear Stress Reversal
ACKNOWLEDGEMENTS
The research presented in this paper has been supported by the ADSC and S&ME,
Inc. and their contributions are greatly appreciated. The Case Atlantic Company
kindly provided access to project sites and test data. Loadtest, Inc. collected data
beyond their contracted scope of work and provided data reduction assistance. The
author also wishes to acknowledge the assistance of Dr. Paul Mayne and Dr. Dan
Brown.
REFERENCES
Brown, D.A. and Camp, W. M. (2002) Lateral Load Test Program for the Cooper
River Bridge, Charleston, SC, Proceedings of the International Deep
Foundations Congress 2002, Orlando, FL, ASCE GSP 116, Vol 1, 95-109.
Camp, W. M.; Brown, D.A and Mayne, P. W. (2002) Construction Method Effects
on Axial Drilled Shaft Performance, Proceedings of the International Deep
Foundations Congress 2002, Orlando, FL, ASCE GSP 116, Vol 1, 195-208.

16
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

Camp, W. M.; Mayne, P. W. and Brown, D.A. (2002) Drilled shaft axial design
values: predicted versus measured response in a calcareous clay, Proceedings of
the International Deep Foundations Congress 2002, Orlando, FL, ASCE GSP
116, Vol 2, 1518-1532.
Camp, William and Parmar, Harpal (1999). Characterization of Pile Capacity with
Time in the Cooper Marl: A Study of the Applicability of a Past Approach to
Predict Long-Term Pile Capacity Transportation Research Record, Journal of the
Transportation Research Board, No. 1663, Washington, DC, 16-24.
Camp, William M., Wright, William B., and Hussein, Mohamad (1993). "The Effect
of Overburden on Pile Capacity in a Calcareous Marl", Proceedings of the Deep
Foundations Institute (DFI) 18th Annual Member's Conference held in Pittsburgh,
Pennsylvania.
Canivan, Gregory and Camp, William. Three Case Histories Comparing Impact and
Vibratory Driven Pile Resistances In Marl, Deep Foundations Institute 27th
Annual Conference The Time Factor in Design and Construction of Deep
Foundations, San Diego, CA. 51-65.
Gohn, G. S., Higgins, B. B., Smith, C. C., and Owens, J. P. (1977).
Lithostratigraphy of the Deep Corehole (Clubhouse Crossroads Corehole 1) Near
Charleston, South Carolina Studies Related to the Charleston, South Carolina,
Earthquake of 1886 A Preliminary Report, Rankin, D. W. editor, U. S.
Geological Survey Professional Paper 1028-E, p 59-70.
Heron, S. D., Jr., Limestone Resources of the Coastal Plain of South Carolina,
(1962), Bulletin No. 28, Division of Geology, State Development Board,
Columbia, SC 128 pp.
Horton, J. Wright, Jr. and Zullo, Victor A., (1991), An Introduction to the Geology
of the Carolinas, The Geology of the Carolinas: Carolina Geological Society 50th
Anniversary Volume, edited by J. Wright Horton, Jr. and Victor A. Zullo. pp 1-10,
University of Tennessee Press, Knoxville, TN
Law Engineering, Inc. (1991) Report of Test Shaft Program Isle of Palms
Connector, Report submitted to The LPA Group, Inc., Columbia, SC, January.
Loadtest, Inc., (2002a), Report on Drilled Shaft Load Testing (Osterberg Method)
Test Shaft 1W Cooper River Bridge, Report submitted to Case Atlantic
Company, Clearwater, FL, March.
Loadtest, Inc., (2002b), Report on Drilled Shaft Load Testing (Osterberg Method)
Test Shaft 1E Cooper River Bridge, Report submitted to Case Atlantic
Company, Clearwater, FL, March.
Parsons Brinckerhoff Quade & Douglas (1993), Report on Preliminary Geotechnical
Investigation for U.S. 17 Bridges over Cooper River, Charleston, South Carolina,
report submitted to the South Carolina DOT, October.
Parsons Brinckerhoff Quade & Douglas (1999), Supplemental Geotechnical
Investigation Data Report for U.S. 17 Cooper River Bridges Charleston, South
Carolina, report submitted to the South Carolina DOT, March.
Rinaldi, Victor A., and Santamarina, J. Carlos, (2003), Cemented Soils: Behavior
and Conceptual Framework, in press.

17
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004

Downloaded from ascelibrary.org by dewi wijaya on 11/02/15. Copyright ASCE. For personal use only; all rights reserved.

S&ME, Inc. (2000a), Phase II Geotechnical Data Summary Report Cooper River
Bridge Replacement Project, Charleston, South Carolina, Report submitted to
Sverdrup Civil, Inc., Charleston, SC
S&ME, Inc. (2000b), Final Report of Geotechnical Exploration Mount Pleasant
Interchange Cooper River Bridge Project, Charleston, South Carolina, Report
submitted to Sverdrup Civil, Inc., New York, NY
S&ME, Inc., (1993), Report of Preliminary Geotechnical Exploration Mark Clark
Expressway
(I-526) - U.S. Highway 17 to Folly Road Charleston, South
Carolina, report submitted to Kimley-Horn Associates, Raleigh, NC.
S&ME, Inc., (1999) Report of Final Geotechnical Exploration Maybank Highway
Replacement Bridge over the Stono River, Charleston, South Carolina, report
submitted to HDR Associates, November.
S&ME, Inc., (2001) Report of Geotechnical Exploration Bridge Foundations and
Embankments: Ashley Phosphate/I-26 Interchange Improvements, Report
submitted to Earth Tech, Inc., Raleigh, NC, August.
S&ME, Inc. (2003) Report of Geotechnical Exploration Bridge Foundations:
Aviation Avenue & Remount Road/I-26 Interchange Improvements, Report
submitted to Arcadis G&M, Raleigh, NC, October.
Soil & Material Engineers, (1984) Report on Test Pile Program North Charleston
Test Site, I-526 Bridges, report submitted to HNTB, Atlanta, GA.
Tuomey, Michael, (1848), Report on the geology of the South Carolina, South
Carolina Geological and Agricultural Survey Report, No. 1, 293 pp.
Ward, L. W., Blackwelder, B. W., Gohn, G. S., and Poore, R. Z., (1979),
Stratigraphic Revision of Eocene, Oligocene and Lower Miocene Formations of
South Carolina, Geologic Notes, Vol 23, No. 1., South Carolina Geological
Survey, Columbia, SC, pp 2 31
Weems, R. E., and Lemon, E. M., Jr., (1984). Geologic map of the Stallsville
quadrangle, Dorchester and Charleston Counties, South Carolina, U. S.
Geological Survey, Miscellaneous Investigations Series, Map GQ-1581, 1:24,000.
Weems, R. E., and Lemon, E. M., Jr., (1989). Geology of the Bethera, Cordesville,
Huger and Kitteredge Quadrangles, Berkley County, South Carolina, U. S.
Geological Survey, Miscellaneous Investigations Series, Map I-1854, 1:24,000.
Weems, R. E., and Lemon, E. M., Jr., (1996). Geology of the Clubhouse Crossroads
and Osborn Quadrangles, Charleston and Dorchester Counties, South Carolina,
U. S. Geological Survey, Miscellaneous Investigations Series, Map I-2491,
1:24,000.

18
Copyright ASCE 2004

GeoSupport 2004

Camp
Geo-Support 2004