See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/257812347

Performance of AASHTO-type bridge model prestressed with carbon fiber-

reinforced polymer reinforcement

Article  in  Aci Structural Journal · May 2013

CITATIONS READS

30 6,404

4 authors, including:

Nabil F Grace Vasant Matsagar

Lawrence Technological University Indian Institute of Technology Delhi
87 PUBLICATIONS   1,465 CITATIONS    216 PUBLICATIONS   2,527 CITATIONS   

SEE PROFILE SEE PROFILE

Chenglin Wu
Missouri University of Science and Technology
59 PUBLICATIONS   487 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Mathematics Applications in Renewable Energy Resources Engineering and Sustainable Systems Engineering View project

ANNs (Artificial Neural Networks) View project

All content following this page was uploaded by Vasant Matsagar on 17 September 2014.

The user has requested enhancement of the downloaded file.

 ACI STRUCTURAL JOURNAL
Title no. 110-S39
Performance of AASHTO-Type Bridge Model Prestressed
with Carbon Fiber-Reinforced Polymer Reinforcement
by Nabil Grace, Kenichi Ushijima, Vasant Matsagar, and Chenglin Wu
Carbon fiber-reinforced polymer (CFRP) composite material has
been widely studied and applied in bridge engineering as an alter-
native solution to the corrosion-related problems posed by steel
reinforcement. Nevertheless, adoption of CFRP reinforcement
to replace conventional steel reinforcement in highway bridges
has not been fully realized yet in the field. Therefore, large-scale
experimental investigations on bridges with CFRP reinforcement
are essential to encourage its widespread application in highway
bridges. This paper presents an experimental investigation conducted
on a one-third-scale AASHTO-type bridge model prestressed with
carbon fiber composite cable (CFCC) strands. The bridge model
was designed, constructed, instrumented, and tested to thoroughly
investigate its flexural behavior, strain response, and ultimate load
failure. A separate one-third-scale single AASHTO-type I-beam
was also constructed and tested to study its flexural and shear
behavior as a control beam. In general, both the control beam and
the bridge model experienced compression-controlled failure as
anticipated. Significant cracking and deflection were experienced
prior to failure. The ultimate strength of the control beam and the
bridge model were in close agreement with the values estimated
using the Unified Design Approach.
Keywords: AASHTO; carbon fiber composite cable; carbon fiber-reinforced
polymer; ductility; fiber-reinforced polymer; flexure; prestress; reinforcement.
INTRODUCTION
Corrosion of steel strands and reinforcement is one of the
major reasons the structural integrity of prestressed concrete
bridges is compromised before the bridges reach their full
life span. The viable solution to eliminate the corrosion-
related problems associated with conventional prestressed
and reinforced concrete bridges is the application of fiber-
reinforced polymer (FRP) materials. The high strength-to-
weight ratio, superior fatigue resistance, ease of handling,
low thermal expansion, and low relaxations are some of
the advantages of the FRP materials over conventional
steel reinforcement. These excellent characteristics made
carbon fiber-reinforced polymers (CFRPs) a potential future
construction material in the bridge construction industry.
The various types of I-beam cross sections specified in
AASHTO1 have been extensively used in the recent construc-
tion of prestressed concrete bridges.2 Features such as a
simpler cross section, higher flexural capacities, and reduced
manufacturing costs make the AASHTO beam a modest
choice for constructing long-span bridges. In addition, FRP
materials can help sustain a longer life span with minimal
maintenance costs. To adopt the innovative CFRP materials
in the bridge construction industry, thorough investigations
are essential; however, an extensive review of the literature
reveals that limited research is available on the application of
CFRP materials in AASHTO-type beams.
In 1997, Fam et al.3 tested five I-beams prestressed
and reinforced with CFRP materials with a span of 9.5 m
(30.5 ft) and one conventionally reinforced beam prestressed
ACI Structural Journal/May-June 2013
TECHNICAL PAPER
and reinforced with steel with the same geometry and span.
In this study, it states that the flexural behavior of the beams
prestressed with CFRP strands exhibited stiffness similar to
the beam prestressed with steel strands. Also, the variation
of the web reinforcement ratio did not significantly influ-
ence the flexural behavior of the beams, and the failures
were controlled by bending capacity. Further, the shape of
the stirrups did affect the shear failure of the beam.
To further investigate the load-deflection relationship of
concrete beams fully prestressed with CFRP, Abdelrahman
and Rizkalla4 conducted an experimental study that included
four prestressed concrete T-beams with a total length of
6.3 m (21 ft) and a depth of 330 mm (13 in.). The results of
the study show that the load-deflection relationship of these
beams was bilinear up to failure, and negligible residual
deformations were experienced during the tests.
To provide guidelines in the design and construction of the
Bridge Street Bridge—the first concrete bridge prestressed
with CFRP materials in the United States—Grace et
al.5 conducted a full-scale test on a CFRP prestressed double-
tee beam. It was concluded that the anticipated prestressing
levels were maintained during the test. The strain-compat-
ibility-based Unified Design Approach was proposed by
Grace and Singh,6 which was validated by experimental
results conducted on a double-tee beam bridge model
reinforced and prestressed using carbon fiber composite cable
(CFCC) strands. Furthermore, a compression-controlled
failure mode was recommended as the design failure mode
for CFRP prestressed concrete beams.7 This recommenda-
tion was based on the better ductility characteristics of over-
reinforced sections. Also, an ultimate concrete compressive
strain of 0.0025 was reported, as experienced in the experi-
mental work.
In typical designs of concrete highway bridges prestressed
with conventional reinforcement, tension-controlled failure
mode dominates, as higher ductility is exhibited by yielding
of the steel reinforcement. CFRP, however, is a brittle mate-
rial exhibiting a linear stress-strain relationship up to rupture.
Because a compression-controlled failure mode provides
better ductility than a tension-controlled failure mode in
terms of extensive deflections, ACI 440.1R-068 recom-
mends a compression-controlled design failure mode for the
concrete bridges reinforced and prestressed with the CFRP
ACI Structural Journal, V. 110, No. 3, May-June 2013.
MS No. S-2011-204.R1 received July 11, 2011, and reviewed under Institute
publication policies. Copyright © 2013, American Concrete Institute. All rights
reserved, including the making of copies unless permission is obtained from the
copyright proprietors. Pertinent discussion including author’s closure, if any, will be
published in the March-April 2014 ACI Structural Journal if the discussion is received
by November 1, 2013.
1
ACI member Nabil Grace is the Dean of the College of Engineering, University
Distinguished Professor, and Director of the Center for Innovative Materials Research
(CIMR) at Lawrence Technological University (LTU), Southfield, MI.
Kenichi Ushijima is a Senior Engineer at Cable Technologies North America, Inc.
He received his bachelor’s degree in engineering from Yamaguchi National University,
Yamaguchi, Japan.
ACI member Vasant Matsagar is a Visiting Professor at the Center for Innovative
Materials Research (CIMR) at LTU.
Chenglin Wu is a former Research Assistant at the Department of Civil Engineering
at LTU.
materials. In this experimental investigation, therefore, the
compression-controlled design failure mode was adopted.
RESEARCH SIGNIFICANCE
The experimental study presented in this paper explains
the design philosophy, construction techniques employed,
and flexural performance of AASHTO I-beam and bridge
model reinforced and prestressed with CFCC strands.
Experimental results of this investigation were validated
with the Unified Design Approach.6 This investigation also
compliments the ongoing research on merits gained by using
the CFRP reinforcement for the construction of highway
bridges. Further, the results presented in this paper should
allow engineers and designers to take full advantage of this
potential, emerging technology to overcome the corrosion-
related problems in the current practice of AASHTO-type
beam bridges.
CONSTRUCTION DETAILS
The control beam and bridge model was constructed,
instrumented, and tested at the Center for Innovative
Fig. 1—Cross-sectional details of I-beam.
2 ACI Structural Journal/May-June 2013
Materials Research (CIMR) at Lawrence Technological
University (LTU). The AASHTO Type IV I-beams9 were
designed as over-reinforced sections as per the flexural
design philosophy8 to be used in the experimental inves-
tigation. The one-third-scale (1:3.6) control beam consisted
of a single precast prestressed AASHTO I-beam with a span
of 12,141 mm (39 ft 10 in.) and a 64 mm (2.5 in.) thick
CFCC-reinforced composite deck slab. The cross section of
the AASHTO I-beams used in this investigation was 502 mm
(19.75 in.) deep with top and bottom flange widths of 203 mm
(8 in.) and a web thickness of 95 mm (3.75 in.), as shown in
Fig. 1. The one-third-scale (1:3.6) bridge model consisted of
five AASHTO I-beams placed at a center-to-center distance
of 502 mm (19.75 in.), joined with five equally spaced 64 mm
(2.5 in.) thick transverse diaphragms (404 mm [15.875 in.]
in depth below the deck slab soffit), and topped with a 2.5 m
(98.75 in.) wide and 64 mm (2.5 in.) thick deck slab,9 as
shown in Fig. 2. The structural integrity of the bridge model
was ensured by extending transverse reinforcement of the
diaphragms into the beams and tying protruded vertical
reinforcement of the diaphragms and protruded stirrups of
the beams to transverse reinforcement of the deck slab.10
Each precast prestressed AASHTO I-beam consisted of
three longitudinal CFCC prestressed strands and seven longi-
tudinal non-prestressed CFCC strands—all with a diameter
of 15.2 mm (0.6 in.)—as flexural reinforcement distributed
vertically along the depth of the beam. The CFCC stir-
rups, measuring 7.5 mm (0.3 in.) in diameter and spaced at
a center-to-center distance of 102 mm (4 in.), were used
as shear reinforcement and protruded in the deck slab
by 38 mm (1.5 in.). The details of the CFCC reinforcement
in the I-beam are presented in Fig. 1 and 3. Each beam was
subjected to a total prestressing force of 266.89 kN (60 kip)
equally distributed among the three longitudinal CFCC
 Fig. 2—Dimensions of bridge model.
Fig. 3—Cross-sectional details of control beam.
prestressed strands. A rectangular end block 533 mm (21 in.)
long, 203 mm (8 in.) wide, and 502 mm (19.75 in.) deep
was provided on each side of the beams to resist bursting
stresses generated during the transfer of the pretensioning
forces. Moreover, confinement in the end-block regions was
provided with rectangular stirrups spaced at a center-to-
center distance of 51 mm (2 in.), as shown in Fig. 3.
The deck slab of the bridge model was reinforced
by 20 longitudinal non-prestressed CFCC strands 15.2 mm
(0.6 in.) in diameter and 62 transverse non-prestressed CFCC
strands 7.5 mm (0.3 in.) in diameter spaced at a center-to-
center distance of 203 mm (8 in.). CFCC strands 7.5 mm
(0.3 in.) in diameter were passed through transverse holes
provided at the web of the I-beams as diaphragm reinforce-
ment for the bridge model. The mechanical properties of
ACI Structural Journal/May-June 2013
the CFCC strands and stirrups are shown in Tables 1 and
2, respectively. Because the CFCC stirrups were bent at
two ends by 90 and 180 degrees, tensile strength tests were
conducted on different portions of the CFCC stirrups to
determine the actual strength. The minimum strength was
selected as the design strength of the CFCC stirrups.
Construction of AASHTO I-beams
Upon completion of constructing the formwork for the
AASHTO I-beams, CFCC reinforcement cages were assem-
bled and placed inside the formwork. The required concrete
cover (38 mm [1.5 in.]) at the bottom of the beams was
provided by attaching 76 mm (3 in.) diameter plastic circular
chairs under the cages. The CFCC prestressing strands were
passed through the reinforcement cages at designated layers
3
Table 1—Mechanical properties of tested CFCC prestressing and non-prestressing strands
Types Transverse non-prestressing strands Longitudinal prestressing strands Longitudinal non-prestressing strands
Diameter, mm (in.) 1 x 7, 7.5 (0.3) 1 x 7, 15.2 (0.6) 1 x 7, 15.2 (0.6)
2 2
Effective area, mm (in. ) 30.97 (0.048) 115.48 (0.179) 115.48 (0.179)
Maximum, kN (kip) 95.01 (21.36) 305.99 (68.79) 305.99 (68.79)
Breaking
Minimum, kN (kip) 91.02 (20.46) 274.99 (61.82) 274.99 (61.82)
load
Average*, kN (kip) 91.99 (20.68) 295.98 (66.54) 295.98 (66.54)
Tensile strength, MPa (ksi) 2972 (431) 2558 (371) 2558 (371)
Modulus of elasticity, MPa (ksi) 164,000 (23,786) 157,000 (22,771) 157,000 (22,771)
*
Five specimens were tested.

Table 2—Mechanical properties of tested CFCC stirrups

Types CFCC rod at straight portion CFCC rod at 180-degree bent end CFCC rod at 90-degree bent end
Diameter, mm (in.) 7 (0.28) 7 (0.28) 7 (0.28)
Effective area, mm2 (in.2) 30.97 (0.048) 30.97 (0.048) 30.97 (0.048)
Maximum, kN (kip) 72.02 (16.19) 37.2 (8.36) 40.5 (9.10)
Breaking
Minimum, kN (kip) 60.00 (13.49) 30.4 (6.83) 25.6 (5.76)
load
*
Average , kN (kip) 64.99 (14.61) 34.0 (7.64) 38.6 (8.68)
Tensile strength, MPa (ksi) 2096 (304) 1098 (159) 1246 (181)
Modulus of elasticity, MPa (ksi) 145,000 (21,030) 145,000 (21,030) 145,000 (21,030)
*
Five specimens were tested.

and positioned between two bulkheads. Calibrated load cells
were mounted on the prestressing strands at the dead end
and connected to a data acquisition system to monitor and
record the level of pretensioning forces applied. In addition,
the strain gauges mounted on the prestressing strands and
the pressure gauge installed on the hydraulic jack were used
to monitor the applied pretensioning forces.
The pretensioning force was applied through a 305 mm
(12 in.) center hole hydraulic jack positioned at the live end.
A custom-made steel chair was attached to the hydraulic
jack and supported on the bulkhead to transfer the reactions
generated during the application of the pretensioning forces.
Moreover, to verify the pretensioning force applied, elon-
gations experienced in the CFCC prestressing strands were
measured during the prestressing operation by measuring
ram displacement of the hydraulic jack. Each prestressing
strand was stressed to an average jacking force of 88.96 kN
(20 kip) to achieve a total pretensioning force of 266.89 kN
(60 kip) on each beam. The applied pretensioning force of
88.96 kN (20 kip) in each CFCC prestressing strand was
approximately 30% of the average breaking load (Table 1).
After the prestressing operation, the concrete was placed
in the formwork and proper uniform compaction was
achieved using three electrical pencil vibrators. The average
28-day compressive strength of the concrete was 44.82 MPa
(6500 psi). After placing the concrete, the beams were wet-
cured by covering them with soaked burlap for 7 days. When
the concrete attained the required compressive strength,
pretensioning forces were released by saw-cutting the
CFCC prestressed strands simultaneously from both ends of
the beams.
Construction of control beam and bridge model
One of the six AASHTO I-beams was selected as the
control beam and was moved to the testing area underneath a
loading frame and was positioned on steel supports 762 mm
(30 in.) tall, measuring 356 x 1372 mm (14 x 54 in.). Elas-
tomeric bearing pads 25.4 mm (1 in.) thick were placed
between the beams and steel supports at each end to simulate
the field conditions. Upon placing the control beam under
the loading frame, a CFCC-reinforced 64 mm (2.5 in.) thick
and 502 mm (19.75 in.) wide deck slab was cast. The average
compressive strength of the concrete for the deck slab was
44.82 MPa (6500 psi). After the concrete in the deck slab
gained adequate strength, the control beam consisting of the
single AASHTO I-beam and the cast-in-place deck slab was
instrumented and tested for flexure.
The other five AASHTO I-beams were moved to another
testing area underneath a loading frame, and the beams
were positioned on steel supports to allow the construction
of the deck slab and transverse diaphragms for the bridge
model. The five beams were maintained at a center-to-center
distance of 502 mm (19.75 in.). Non-prestressing CFCC
strands 7.5 mm (0.3 in.) in diameter were passed through the
holes kept in the beams and epoxied to brace the beams in
the transverse direction. Vertical reinforcement of the trans-
verse diaphragm was attached to its transverse reinforcement
and formwork was provided around the reinforcement. Prior
to the deck slab reinforcement placement, formwork for the
deck slab was attached to the beams and diaphragm form-
work and supported on the ground as typically practiced in
shored construction. The deck slab reinforcement consisted
of 20 No. CFCC 1 × 7 φ 15.2 mm (0.6 in.) strands in the
longitudinal direction and CFCC 1 × 7 φ 7.5 mm (0.3 in.)
strands at 203 mm (8 in.) in the transverse direction. The
concrete was placed in the diaphragms and the deck slab, and
compacted properly with pencil vibrators and metallic rods.
After finishing the top surface of deck slab the bridge model
was covered with soaked burlap for 7 days. Upon hardening
of the concrete for 7 days, supports of the formwork were
removed. The average 28-day compressive strength of the

4 ACI Structural Journal/May-June 2013

 Fig. 4—Test setup for control beam.
Fig. 5—Flexural load test of control beam.
concrete in the deck slab and the transverse diaphragms was
27.58 MPa (4000 psi).
Instrumentation and test setup
Prior to the casting of the AASHTO I-beams, electrical
strain gauges were attached to the CFCC prestressing
strands to measure the strain responses. After the concrete
in the deck slab gained adequate strength, two linear motion
transducers were installed at the quarter-span and midspan,
respectively, to measure the deflections of the control beam
during the flexural load test. Two strain gauges were mounted
on the top surface of the deck slab at midspan to record the
strain response of the extreme compressive concrete fiber of
the control beam. Meanwhile, to observe the strain response
of the CFCC stirrups, five DEMEC stations (Rosette type)
were also installed at the shear-critical sections. A steel
loading frame—that is, a spreader with two loading points
spaced at a distance of 965 mm (38 in.)—was placed on the
deck slab at midspan. An actuator with a maximum loading
capacity of 889.64 kN (200 kip) was used to apply load at
the center of the spreader. A load cell with a capacity of
889.64 kN (200 kip) was connected to the actuator to record
the applied load. To prevent the possible twist that might
be caused during the loading process, custom-made steel
guides were installed around the beam at both quarter-span
points and beam ends. The test setup for the control beam is
shown in Fig. 4.
ACI Structural Journal/May-June 2013
The control beam was subjected to several loading and
unloading cycles to separate the elastic and inelastic ener-
gies. These loading cycles were 13.34, 31.14, 35.59, 53.38,
66.72, 88.96, and 111.21 kN (3, 7, 8, 12, 15, 20, 25 kip) and
ultimate load cycle. Figure 5 shows the control beam during
the flexural load test.
Different sensors were installed and used to analyze the
behavior of the bridge model according to the test program.
The test program for the bridge model consisted of the flex-
ural performance, decompression and cracking load, and
ultimate load tests.
In the flexural performance test, each beam of the bridge
model was loaded at quarter-span and midspan with a single
point load of 66.72 kN (15 kip). Linear motion transducers
were installed at both quarter-span and midspan to measure
the deflections of the beams under different loading cases.
In the decompression load test, a 1.22 m (48 in.) long
spreader was used to load the bridge model at midspan.
Linear motion transducers were installed at the midspan of
the bridge model to monitor and record the deflection. Strain
gauges were attached at the bottom surface of the center
beam at midspan to determine the decompression load.
In the ultimate load test, the bridge model was loaded
using a steel square tube with a length of 1.22 m (48 in.)
mounted at the midspan covering the center beam and its
two adjacent beams, as shown in Fig. 6. The strain gauges
installed on the CFCC strands, linear motion transducers
5
 Fig. 6—Test setup for ultimate load test of bridge model.
Fig. 7—Load-deflection behavior of control beam.
installed at midspan, and strain gauges installed on the top
surface of the deck slab at midspan were used to analyze the
behavior of the bridge model. Similar to the flexural load
test conducted on the control beam, the bridge model was
subjected to several loading and unloading cycles before
failure to separate the elastic and inelastic energies. These
cycles were conducted at 311.38, 355.86, 422.58, 444.82,
489.30, 533.79, and 578.27 kN (70, 80, 95, 100, 110, 120,
and 130 kip) and at ultimate load.
RESULTS AND DISCUSSION
The control beam failed with an ultimate load-carrying
capacity of 162.05 kN (36.43 kip) and a corresponding
deflection of 292 mm (11.5 in.). The ultimate load-carrying
capacity of the control beam is in close agreement with the
designed value of 164.58 kN (37 kip) using the Unified
Design Approach proposed by Grace and Singh6 (important
steps in the design procedure are shown in the Appendix*).
Because the control beam was over-reinforced with a
*
The Appendix is available at www.concrete.org in PDF format as an addendum to
the published paper. It is also available in hard copy from ACI headquarters for a fee
equal to the cost of reproduction plus handling at the time of the request.
6 ACI Structural Journal/May-June 2013
reinforcement ratio of 0.0046 (the balanced reinforce-
ment ratio for the section was calculated as 0.00196), a
compression-controlled flexural failure initiated by the
concrete crushing at the midspan was observed. The load-
deflection behavior of the control beam experienced a
bilinear response, as shown in Fig. 7, with cracking load of
28.91 kN (6.5 kip). The load-strain responses of the top and
bottom layers of the CFCC prestressing strands also experi-
enced a bilinear response, as shown in Fig. 8. However, the
strain gauges on the strands stopped functioning at a load
of 62.28 kN (14 kip). The CFCC prestressing strands were
further examined after the beam failure and no damage was
experienced. Meanwhile, the load-strain response of the
extreme compressive concrete fiber was also recorded, as
shown in Fig. 9. The maximum strain experienced by the
extreme compressive concrete fiber was 2000 με at failure.
The ductility of the control beam was evaluated by the
energy ratio, which was calculated using the energy-based
approach.11 Accordingly, the failure of a structure is defined
as brittle failure if the energy ratio is lower than 70%.
Therefore, the failure of the control beam was defined as a
brittle failure with an energy ratio of 46.88%, as shown in
Fig. 10. However, a sufficient warning prior to the failure
 Fig. 8—Strain experienced by CFCC prestressing strands in control beam.

Fig. 9—Compressive strain of top concrete fiber at midspan of control beam.

Fig. 10—Ductility of control beam.

ACI Structural Journal/May-June 2013 7

 Fig. 11—Cracks developed in control beam.

Fig. 12—Strain experienced by CFCC stirrup at DEMEC Station 2.

was provided by the excessive deflection and extensive
cracks of the control beam before reaching the ultimate load
of 162.05 kN (36.43 kip).
A significant number of diagonal cracks were observed in
the shear zone (between the loading points and beam ends)
of the control beam. Details of the cracks mapped at different
load levels are shown in Fig. 11. As the control beam was
designed to have a flexural failure, adequate CFCC stir-
rups were provided to avoid any premature shear failure.
Therefore, the widths of the cracks were insignificant and
the maximum width of the shear cracks was approximately
0.2 mm (0.009 in.) at a corresponding load of 111.21 kN
(25 kip). Thus, the CFCC stirrups provided in the AASHTO
I-beams in the desired bent shapes served the intended
purpose satisfactorily.
The strain responses of the CFCC stirrups at DEMEC
Station 2 were recorded and compared to the estimated
values using ACI 440.4R-047 and ACI 318-0512 approaches,
as shown in Fig. 12. It can be seen that both the shear
design approaches proposed by ACI 440.4R-047 and
ACI 318-0512 conservatively predicted the strains of the
CFCC stirrups at a load of 111.21 kN (25 kip). However, the
estimated strain responses of the CFCC stirrups using the
approach proposed by ACI 440.4R-047 were closer to the

8 ACI Structural Journal/May-June 2013

 Fig. 13—Strain experienced at bottom concrete fiber in bridge model.
actual strains experienced as compared to those estimated
using ACI 318-05.12 Furthermore, the maximum strain of the
CFCC stirrups measured at DEMEC Station 2 was 500 με,
which was less than the ultimate strain of the CFCC stirrups
(756 με). This fact signifies adequate strength of the CFCC
stirrups during the flexural load test. The shear failure load
for the control beam, predicted based on the strains experi-
enced, was 178 kN (40 kip).
As mentioned previously, the test program conducted on
the bridge model included the flexural performance test,
decompression and cracking load test, and ultimate load
test. In addition, the flexural performance test was repeated
after the decompression and cracking load test to examine
the behavior of the bridge model at the cracked stage.
The decompression and cracking load test was conducted
to determine the effective prestress and the cracking load
of the bridge model. During the test, the bridge model was
subjected to a single point load positioned on a 1.22 m
(48 in.) long spreader located on the center beam (B-3) at
midspan. The first flexural crack was observed at the bottom
of Beam B-3 at the midspan with the corresponding load of
104.53 kN (23.5 kip). The bridge model was then unloaded
to allow installing a set of four strain gauges at both sides
of the initial flexural cracks. Upon completing the installa-
tion, the bridge model was reloaded. The strain of the bottom
concrete fiber increased proportionally to the applied load
immediately after the reloading started. When the stress of the
bottom concrete fiber approached zero, however, the strain
stopped increasing and remained constant as the load applied
on the bridge model was still increasing, as shown in Fig. 13.
The decompression load determined through the load-strain
response of the bottom concrete fiber was 57.83 kN (13 kip).
The overall prestress loss was subsequently calculated as 11%
based on the decompression load, which is less than the typical
15% prestress loss, as usually reported for beams prestressed
using steel strands.13 At the end of the test, the bridge model
was loaded up to 266.89 kN (60 kip) to allow the development
of the flexural cracks. The flexural performance test was then
repeated after the bridge model was extensively cracked.
ACI Structural Journal/May-June 2013
The ultimate load test was conducted by applying
a uniformly distributed load on a 1.22 m (48 in.) long
and 152 mm (6 in.) wide steel square tube placed across
the bridge model at the midspan. After several loading
and unloading cycles, the bridge model failed at a load
of 689.47 kN (155 kip) with the corresponding deflection
of 240 mm (9.45 in.) at center beam, B-3. This ultimate load-
carrying capacity of the bridge model was in close agree-
ment with the designed6 value of 733.96 kN (165 kip). The
failure was initiated by concrete crushing at top compres-
sion fibers at the midspan of the bridge model, and the
CFCC prestressing strands were still intact after the bridge
model failed. The load-deflection behavior of the bridge
model, as shown in Fig. 14, showed bilinear response due
to the concrete cracking, similar to what was observed in
the control beam. The load-strain response of the extreme
compressive concrete fiber for the bridge model was also
recorded, as shown in Fig. 15. The maximum concrete strain
reached by the extreme fiber was 2520 με in compression.
The strain responses of the CFCC strands in the bridge
model were monitored throughout the test. The load-strain
responses of different layers of the CFCC strands and the
extreme compressive concrete fiber during the ultimate load
cycle are shown in Fig. 16. The maximum strain reached
by the bottom prestressing strands was 10,000 με, which
was 66.7% of the ultimate strain for the CFCC strands
(15,000 με). This fact suggests that perhaps a higher preten-
sioning force can be applied to increase the cracking load.
The energy ratio of the bridge model was 26.67%, which
was calculated using the energy-based approach,11 as shown
in Fig. 17. Although the ductility of the bridge model was
low in the terms of the energy ratio and the excessive deflec-
tion and extensive cracks experienced, the bridge model
provided significant warning before failure.
CONCLUSIONS AND RECOMMENDATIONS
This paper presents an experimental investigation
addressing the application of CFRP strands in prestressed
AASHTO I-beams and bridges. The ACI 440.4R-047 design
9
 Fig. 14—Load-deflection behavior of bridge model.
Fig. 15—Compressive strain experienced by extreme concrete fiber of
bridge model.
guidelines and the Unified Design Approach proposed by
Grace and Singh6 were used in the flexural and shear design
of the control beam and bridge model. The test results were
analyzed and compared with the estimated values, and
several conclusions are drawn as follows.
1. Both the control beam and the bridge model experienced
a compression-controlled failure as expected. The calcu-
lated ultimate load-carrying capacities for both the control
beam and the bridge model were 164.58 and 733.96 kN
(37 and 165 kip), respectively. These values are in close agree-
ment with the experimental values (162.05 kN [36.43 kip]
for the control beam and 689.47 kN [155 kip] for the bridge
model). This verified that the Unified Design Approach6 is
suitable in designing the AASHTO I-beams and bridges.
2. The maximum strain of the CFCC stirrups measured at
the shear-critical sections was 500 με, which was less than
the ultimate strain of 756 με. This demonstrates the excellent
performance of the CFCC stirrups in resisting the shear load
experienced by the AASHTO I-beams.
3. The calculated strain of the CFCC stirrup using
ACI 440.4R-047 approach was conservative as compared to
10
the experimental value experienced at the load of 111.21 kN
(25 kip). This fact further indicates that ACI 440.4R-047 shear
design approach can be adequately used in the design of the
AASHTO I-beams using CFCC stirrups.
4. An 11% prestress loss was calculated through the
decompression test conducted on the bridge model, which
is less than the typical 15% prestress loss, as traditionally
reported for beams prestressed using steel strands. This
further demonstrates the excellent performance of the CFCC
strands in prestressed concrete AASHTO I-beams.
5. The CFCC strands were not damaged after the failure of
the control beam and the bridge model. The maximum strain
experienced by the prestressing CFCC strands was 66.7%
of the ultimate strain. Therefore, a higher pretensioning
force may be recommended to be applied on the CFCC
prestressing strands.
6. The energy ratios of the control beam and bridge model
were 46.88% and 26.67%, respectively. These values clas-
sified the failures of both the control beam and bridge
model as brittle failure.11 However, the excessive deflec-
tions and the extensive cracks provided sufficient warning
ACI Structural Journal/May-June 2013
 Fig. 16—Strain experienced by CFCC strands at ultimate load of bridge model.
Fig. 17—Ductility of bridge model.
prior to the failure for this type of AASHTO I-beam bridge
with CFCC strands.
ACKNOWLEDGMENTS
This investigation was supported by the U.S. Department of Transpor-
tation (US-DOT) (Contract No. DTOS 59-06-G-0030) and MDOT-LTU
Center of Excellence. The support and guidance of B. Jacob, Senior Policy
Analyst, US-DOT; and L. N. Triandafilou, Senior Structural Engineer,
FHWA, are truly appreciated. Moreover, the Tokyo Rope Manufacturing
Company Limited, Japan, supplied the CFCC reinforcement.
REFERENCES
1. AASHTO, “AASHTO Load and Resistance Factor Design (LRFD)
Bridge Design Specifications,” third edition, American Association of State
Highway and Transportation Officials, Washington DC, 2004.
2. Martin, R. D.; Kang, T. H.-K.; and Pei, J.-S., “Experimental and Code
Analyses for Shear Design of AASHTO Prestressed Concrete Girders,”
PCI Journal, V. 56, No. 1, Dec. 2011, pp. 54-74.
3. Fam, A. Z.; Rizkalla, S. H.; and Tadros, G., “Behavior of CFRP for
Prestressing and Shear Reinforcements of Concrete Highway Bridges,”
ACI Structural Journal, V. 94, No. 1, Jan.-Feb. 1997, pp. 77-86.
4. Abdelrahman, A. A., and Rizkalla, S. H., “Deflection Control of
Concrete Beams Pretensioned by CFRP Reinforcements,” Journal of
Composites for Construction, ASCE, V. 3, No. 2, May 1999, pp. 55-62.
ACI Structural Journal/May-June 2013
5. Grace, N. F.; Enomoto, T.; Abdel-Sayed, G.; Yagi, K.; and Colla-
vino, L., “Experimental Study and Analysis of a Full-Scale CFRP/
CFCC Double-Tee Bridge Beam,” PCI Journal, V. 48, No. 4, July 2003,
pp. 120-139.
6. Grace, N. F., and Singh, S. B., “Design Approach for Carbon Fiber-
Reinforced Polymer Prestressed Concrete Bridge Beams,” ACI Structural
Journal, V. 100, No. 3, May-June 2003, pp. 365-376.
7. ACI Committee 440, “Prestressing Concrete Structures with FRP
Tendons (ACI 440.4R-04),” American Concrete Institute, Farmington
Hills, MI, 2004, 35 pp.
8. ACI Committee 440, “Guide for the Design and Construction of
Concrete Reinforced with FRP Bars (ACI 440.1R-06),” American Concrete
Institute, Farmington Hills, MI, 2006, 44 pp.
9. MDOT, “Bridge Design Guides,” Bureau of Highway Development,
MI, 2001, p. 6.60.01.
10. MDOT, “Bridge Design Manual,” Bureau of Highway Development,
MI, 2006, 466 pp.
11. Grace, N. F.; Soliman, A. K.; Abdel-Sayed, G.; and Saleh, K. R.,
“Behavior and Ductility of Simple and Continuous FRP Reinforced Beams,”
Journal of Composites for Construction, ASCE, V. 2, No. 4, Nov. 1998,
pp. 186-194.
12. ACI Committee 318, “Building Code Requirements for Reinforced
Concrete (ACI 318-05),” American Concrete Institute, Farmington Hills,
MI, 2005, 430 pp.
13. Nawy, E. G., Prestressed Concrete: Fundamental Approach, fourth
edition, Prentice Hall, Saddle River, NJ, 2003, 960 pp.
11
 NOTES:

12 ACI Structural Journal/May-June 2013

View publication stats