Field Measurements of Passive Pressures Behind An Integral Abutment Bridge

FMGM 2007: Seventh International Symposium on Field Measurements in Geomechanics © 2007 ASCE
Field Measurements of Passive Pressures Behind an Integral

Abutment Bridge
Sophia Hassiotis1 and Kai Xiong2

1
Associate Professor, Department of Civil, Environmental and Ocean Engineering,
Stevens Institute of Technology, Hoboken, N.J., USA 07030; PH (201)216-8231; email:
Downloaded from ascelibrary.org by Universidad De Guadalajara on 12/15/23. Copyright ASCE. For personal use only; all rights reserved.
Sophia.hassiotis@stevens.edu
2
Graduate Student, Department of Civil, Environmental and Ocean Engineering, Stevens
Institute of Technology, Hoboken, N.J., USA 07030
Abstract
A bridge that is cast continuously with its abutments is known as an integral abutment
bridge. A short stub-abutment supported on a pile-bent creates a flexible foundation that
transfers the temperature and traffic-induced horizontal loading and eliminates the need
for expansion bearings. Integral abutments are gaining wide acceptance by many states;
however, a nationally accepted design methodology does not exist for their design and
construction. Instead, each highway department depends on the experience of its
engineers to develop adequate design criteria. By studying the response of such systems
to horizontal loading, we may be able to extend the learned concepts to different
structures and conditions.
To study integral bridges, we instrumented a medium-length (300 ft) highway bridge
in Trenton, New Jersey and have been observing the reaction of the bridge and its
foundation to seasonal and daily thermal loadings. Design parameters of interest include
the flexural behavior of the piles that support the abutment, the soil-pressure distribution
behind the abutment, and the displacement and rotation of the superstructure during
thermal loading. As such, we have gathered data every two hours for the past four years
on displacement, rotation, strains, and pressures on the bridge and its foundation. In this
paper, we will be sharing some of the data and will discuss relevant conclusions of our
studies. In general, we have been witnessing an excellent correlation between
temperature and displacement, expected behavior from the horizontally loaded piles, and
a steady build-up of soil pressure behind the abutment. A short study of passive pressure
build-up is also summarized.
Introduction
Integral abutment bridges are becoming widely accepted for new construction of
short to medium length highway bridges of limited skew. Although they offer an
economic alternative to the use of bearings, integral abutments present their own unique
challenges. One area of concern is the development of passive pressures behind the
abutments due to the cyclic loading of the soil during thermal movement of the
superstructure. The challenge of the present work is to present to the engineering
community an estimate of the passive pressures behind the abutment, and some
supporting data.
Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

FMGM 2007
Conventionally, Coulomb or Rankine theories have been used for the design of the
integral abutments. Displacement dependent earth pressure theories are also being
investigated to arrive at a closer estimate of the earth pressures generated during the
cyclic loading of the abutments. Clough and Duncan (1991) report that the required
displacement for full passive pressure development is /H of 0.04 for loose sand and 0.01
for dense sand. Here is the horizontal displacement at the top of a rigid wall due to
rotation or translation, and H is the height of the wall.
For the design of integral abutments, the maximum displacement at the top of the
abutment due to thermal variations is calculated by assuming that the total bridge
displacement is divided equally between each abutment:
=1/2 L T (11)
where L is the length of the continuous bridge; is the coefficient of thermal expansion;
and T is the difference between the temperature during construction and the maximum
(or minimum) temperature expected at the site (Figure 1).
Figure 1. Definition of wall-top displacement.

In the studies herein, the displacement is assumed zero every winter. Any
displacement towards the soil mass starts with increasing temperatures in spring time and
reach a maximum displacement in the summer months.
A recent report by the National Cooperative Highway Research Program (NCHRP)
(1991) shows the relationship between wall movement and passive pressures, as
reproduced in Figure 2.
Few previous studies have been conducted to record the effect of cyclic loading due
to daily and seasonal thermal variations on earth pressures behind integral abutments. It
was not until recently that experimental work and modeling concentrated on defining the
mechanics of the soil behind an integral abutment. Researchers have developed
experiments to shed light to the mechanism responsible for the complexity of the lateral
earth pressure behind the cyclically moving integral abutment using either full-scale
experiments or laboratory set-ups.
In the design of integral abutments, Massachusetts uses a magnitude of lateral earth
pressure that depends on the movement of the abutment. The value is assumed to be
somewhere between at-rest conditions and passive pressures. Used with the Mass
Highway’s standard compacted gravel borrow, a pressure coefficient Kh is proposed as
(Mass. 1999):
K h = 0.43 + 5.7[1 e 190 ( / H ) ] (12)

FMGM 2007
where, is displacement of the abutment and H is the height of the abutment. The
pressure varies linearly at the back of the abutment. The work is based on large-scale
testing.
Figure 2. Earth pressure coeff. vs relative wall displacement (NCHRP, 1991)
The British Code BA42 proposes an upper limit of the passive coefficient, Kh, as
Kh = ( ) 0.4 K p (13)
0.05 H
where a maximum value for Kp ,derived after cyclic loading of Leighton Buzzard sand, is
taken to be 12.5. The minimum Kh is constrained to Kp/3. The pressure distribution in
this code is linear down to mid-depth and constant to the bottom.
England et al. (2000) proposed the same pressure distribution with BA42 and a new
equation for the passive pressure coefficient as:
Kh = Ko + ( ) 0.6 K p (14)
0.03H
where the constraint for the minimum Kh is removed. The work is based on testing of a
model rigid wall that was used to study the effects of thermal loading on integral
abutments. The model simulated an abutment, which was free to rotate about its base as a
response to cyclic loading of the top.
Full-Scale Testing of the Scotch-Road Bridge.
The Scotch Road Integral Abutment Bridge crosses over I-95 in Trenton, New Jersey.
It is a 300 ft-long, two-span continuous bridge with a 15o skew angle. The steel
superstructure is made of 10 high performance steel girders and is supported on
reinforced concrete integral abutments each of which rests on 19 HP 14x102 steel piles.
The piles were cast in concrete at a depth of 13 feet from the bottom of the abutments.
The abutments are 3 ft wide and 11 ft deep. The soil behind the abutment is porous
compacted fill, whose gradation limits are specified by AASHTO T27. The bridge was

FMGM 2007
constructed in two stages. Half of the lanes were in place by October 2002 (Stage I) and
the bridge was completed by October 2003 (Stage II).
The abutment and foundation piles supporting the Scotch Road Bridge were
instrumented. A cross-sectional view of the instrumentation is shown in Figure 3. Data
from four soil pressure cells will be discussed herein. They are located at the elevations
of 58m (190ft) and 56.5m (185ft). In the transverse direction, the cells are placed at two
positions referred herein as Pile 3 and Pile 9 position. These, not shown in the figure,
cover the quarter points of the bridge completed in Stage I. Two inclinometers were
placed on the abutment to measure the rotations at the superstructure/abutment

connection. Two gages were specifically constructed to obtain the longitudinal
displacements of the superstructure. These gages were placed at the point where the
approach slab rests on the sleeper slab.
Figure 3. Side-view of instrumentation
The soil pressure cells, shown in Figure 4, were manufactured by Sensing Systems
(2002) and calibrated in the laboratory before being installed on site. They are
constructed with a thin, flat plate, 4-in in diameter and 5/8-in in thickness, which houses
several strain gages. A 3-in diameter pressure-sensing element was utilized to provide
the soil pressure area sample. The pressure cells, that were installed approximately four
and nine feet below the surface, are the subjects of the present report.
Integral abutments are affected by both maximum and minimum daily, and seasonal
temperature variations. The temperature increases gradually during the day, which
results in the expansion of the bridge superstructure, and hence, the passive movement of
the abutment. The cycle is reversed at night, when the temperature decreases and the
active movement of the abutment takes place. The pressure correlates with the

FMGM 2007
displacement during the daily cycles. However over the year intervals, we do not observe
an exact correlation of displacement and pressure due to the fact that the soil is not an
elastic medium.
Figure 4. Soil pressure gage (Sensing Systems, 2002)
Figure 5 shows the longitudinal displacement at the top of the abutment. Figure 6
shows the rotation measured on the abutment at the stringer base, and Figure 7 shows the
pressure data behind the abutment. In these figures, the beginning of each month is
marked in the abscissa with the month number.
The data span a period from April 2003 to June 2006. In general, the displacement
increases with temperature from January, reaching its highest value in July, and starts
decreasing to its lowest value in January. The same is true for the rotations which
correlate well to the displacement readings.
In contrast to the displacement, the pressure seems to increase suddenly at the very
beginning of temperature increases in February, reaches its highest value of the year
around April and decreases to an average value by July. As the displacement reverses
direction in July, the pressure starts to decrease. We can deduce from this that the soil
mass reaches a very dense state at the end of the seasonal active cycle of the winter
months, and it behaves as dense sand in shear as the temperatures start to climb and the
abutment starts pushing on the soil. As a shear band forms, the resistance to displacement
decreases. As a result, we see a lower pressure at the higher displacements.
The built up of pressure from year to year is a result of the changes in soil density and
soil fabric due to the daily cyclic motion. According England et al. (2000), the fabric of
granular soil changes during the unloading stage, which results in accumulation of plastic
deformation.
Data Analysis and Comparison with Literature
Some of the data recorded at the Scotch Road Bridge are plotted together with the
work of other researchers in Figure 8 to Figure 11. To compare the data recorded in the
Scotch Road Bridge with the literature, a pressure coefficient, Kh, is calculated using a
constant soil density of 125pcf. Any changes in density during the cyclic disturbance of
the soil are reflected in the change of pressure, thus the change of Kh. In addition, a new

FMGM 2007
at-rest position is assigned to the abutment every winter. Passive displacements are
calculated every year as the displacement of the bridge in reference to its winter position.
Figure 5. Longitudinal displacement at the sleeper slab.
Figure 6. Rotation of the abutment at the stinger connection.

FMGM 2007
The data shows that in the first two years, the pressures are low, more in line of what one
would expect for loose sands. By the fourth year, the coefficient is closer to that
predicted by the NCHRP (1991) for dense sands.
For small displacements ( /H less than 0.003), the coefficient increases with
displacement and can be predicted by the Kh values proposed by NCHRP (1991) for
dense sands. In an integral abutment bridge, this condition is met during the winter
months when passive displacements are small, and the soil has achieved a dense state due
to the densification produced by the daily cycles. At larger displacements, the Kh factor in
the integral abutment bridge does not increase with the increasing displacement. Rather,
it seems to undergo a small decrease. This condition is met during the summer months
when shear resistance of the dense soil mass has been overcome. As a result, during the
maximum passive displacements, the Kh factor is smaller than recommended by the
NCHRP work. More extensive analysis is found in Xiong (2007).
Figure 7 Average pressures on abutment wall, 2003-2006.
The data collected from the sensors located at Pile 3 and Pile 9 were used to produce
the maximum Kh factor for each year, shown in Table 1.
The change of the maximum coefficient of earth pressure during the first four years of
the bridge lifetime is shown in Figure 12. For the first three years, the Kh values (and
pressure) increase rapidly. The limited data presented herein indicate that these values
reach a plateau by the fourth year.
The question of interest is what is the maximum pressure that we should be using to
design the abutment. The Scotch Road Bridge was built assuming a Rankine passive

FMGM 2007
pressure of a soil with an angle of internal friction of 30o, and a density of 110pcf. The
pressure that has been measured (up to 30 psi) exceeds the design pressure (15 psi at the
point of measurement). But the data and design values are closer if we can assume that
the angle of internal friction increased to 40o due to the densification, and use a
maximum density for the soil. To show, the data has been plotted along the depth of the
abutment in Figure 13, and is compared to classical theories. For the calculations, the
height of the abutment is 11 ft, the soil density is taken as 125pcf, and the angle of
internal friction 40o. The wall is assumed frictionless, and the soil perfectly drained.
Table 1- Maximum Kh factor for every year.

Kh value
Year Pile 3 Pile 9
Average
El 56.5 m El 58.0 m El 56.5 m El 58.0 m
First 1.6 2.6 1.5 2.3 2.0
Second 2.3 3.3 3.0 3.7 3.0
Third 4.5 4.1 4.2 4.0 4.2
Fourth 4.2 5.5 4.0 3.5 4.3
Figure 8. Kh vs /H at El 56.5 of pile 9 for the first year.

FMGM 2007
Figure 9. Kh vs /H at El 56.5 of pile 9 for the second year.
Figure 10. Kh vs /H at El 56.5 of pile 9 for the third year.

FMGM 2007
Figure 11. Kh vs /H at El 56.5 of pile 9 for the fourth year.
Figure 12. Development of Kh factor over four years.

FMGM 2007
In Figure 14, the data is plotted against the more recent theories that predict passive
pressure as a function of abutment rotation, for /H of 0.006. The coefficients for these
methods are tabulated in Table 2. The British code (BA42), and England et al. (2000)
underestimate the pressures at the base of the abutment. This is mainly due to the
assumption of pure rotation, whilst the pile-supported abutment of the Scotch Road
Bridge experiences mainly translation. The NCHRP proposed Kh factor seems to predict
the pressure very well.
Table 2- Horizontal Pressure Coefficients Dependent on wall-top displacement

Kp Kp Kh Kh
Mass NCHRPfor BA 42 England et al.
/H
Higway1 dense Code2 (2000)2
1
sand Kp=12.5 Kp=12.5
0.001 1.42 2.01 4.17 2.02
0.002 2.23 3.48 4.17 2.86
0.004 3.46 5.04 4.55 4.13
0.006 4.31 5.57 5.35 5.16
0.007 4.62 5.72 5.69 5.62
1
Triangular pressure distribution along wall
2
Triangular pressure down to H/2; constant thereafter.
Figure 13. Comparison of data with classical passive pressures 2002-2006.
Summary and Conclusions

The full-scale testing of the Scotch Road integral abutment and other tests reported in
the literature demonstrate that integral abutments experience both translation and small

FMGM 2007
rotations induced by daily and seasonal temperature variations. The soil behind the
abutment undergoes pressure build-up due to densification attributed to the daily and
seasonal cycles for the first three years. Such changes seem to slow down by the fourth
year. The NCHRP coefficient for passive pressure provides a conservative design value.
Figure 14. Data vs. empirical passive pressures behind integral abutments;
/H=0.006; (2002-2006).
References
Clough, G. W. and Duncan, J. M. (1991). “Earth Pressures, Chapter in Foundation Engineering

Handbook.” 2nd edition, edited by Hsai-Yang Fang, van Nostrand Reinhold, New York, NY, 223-
235.
England G. L., Tsang, N. C. M., Bush, D.I. (2000). “Integral Bridges: A Fundamental Approach
to the Time-Temperature Loading Problem.” London: Telford.
Hassiotis, S. (2003) “The Scotch-Road Bridge Project”, 5th Annual NJDOT Showcase, Rutgers
University, October 24.
Massachusetts Bridge Manual, Part I, December 1999
NCHRP(1991). “Manuals for the design of bridge foundations” R.M Barker, J.M. Duncan, K.B.
Rojiani, P.S.K. Ooi, C.K. Tan, and S.G. Kim, eds. TRR343, TRB, Washington, D.C.
Sensing Systems and Associates (2002). New Bedford, Mass.
Xiong, K. (2007) Deformation of Cohesioness Fill due to Cyclic Loading. MS Thesis. Stevens
Institute of Technology, Hoboken, N.J. 07030

FMGM 2007

Field Measurements of Passive Pressures Behind An Integral Abutment Bridge

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Field Measurements of Passive Pressures Behind An Integral Abutment Bridge

Uploaded by

Copyright:

Available Formats

FMGM 2007: Seventh International Symposium on Field Measurements in Geomechanics © 2007 ASCE

Field Measurements of Passive Pressures Behind an Integral

Sophia Hassiotis1 and Kai Xiong2

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Figure 1. Definition of wall-top displacement.

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Figure 2. Earth pressure coeff. vs relative wall displacement (NCHRP, 1991)

Full-Scale Testing of the Scotch-Road Bridge.

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

placed on the abutment to measure the rotations at the superstructure/abutment

Figure 3. Side-view of instrumentation

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Figure 4. Soil pressure gage (Sensing Systems, 2002)

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Figure 5. Longitudinal displacement at the sleeper slab.

Figure 6. Rotation of the abutment at the stinger connection.

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Figure 7 Average pressures on abutment wall, 2003-2006.

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Table 1- Maximum Kh factor for every year.

Figure 8. Kh vs /H at El 56.5 of pile 9 for the first year.

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Figure 9. Kh vs /H at El 56.5 of pile 9 for the second year.

Figure 10. Kh vs /H at El 56.5 of pile 9 for the third year.

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Figure 11. Kh vs /H at El 56.5 of pile 9 for the fourth year.

Figure 12. Development of Kh factor over four years.

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Table 2- Horizontal Pressure Coefficients Dependent on wall-top displacement

Figure 13. Comparison of data with classical passive pressures 2002-2006.

Summary and Conclusions

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

Clough, G. W. and Duncan, J. M. (1991). “Earth Pressures, Chapter in Foundation Engineering

Massachusetts Bridge Manual, Part I, December 1999

Sensing Systems and Associates (2002). New Bedford, Mass.

Copyright ASCE 2007 Field Measurements in Geomechanics (FMGM 2007)

You might also like