Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
K.C. Chang
1
, L.L. Chung
2
, B.J. Lee
3
, Y.F. Li
4
, K.C. Tsai
1
, J.S. Hwang
5
, S.J. Hwang
5
Abstract
This paper presents the progress of a NCREEs research program on seismic retrofit of
existing RC bridge columns during the last three years. More than sixty large-scale specimens
were designed and constructed to simulate the worst scenario of the construction practice in
Taiwan prior to 1987. Twenty-four of the test columns were used as the benchmark for
comparison with other specimens retrofitted or repaired using the carbon fiber reinforced
plastics (CFRP) jacketing, steel jacketing, and RC jacketing. Experimental results showed that,
in general, the retrofit methods used in the U.S. and Japan are effective also effective for the
existing RC bridge columns in Taiwan. However, due to the poor concrete compressive
strength and the lap splicing at the plastic hinge zone, the retrofit measures developed for
better ductility and shear strength are not always effective. More research effort is necessary
to develop effective methods for RC columns lap-spliced at the plastic hinge zones.
1
Prof., Dept. of Civil Engr, National Taiwan University, Taipei, Taiwan.
2
Research Fellow, National Center for Research on Earthquake Engineering
3
Prof., Dept. of Civil Engr., Feng Chia University, Taichung 100, Taiwan.
4
Associate Prof., Dept. of Civil Engr., National Taipei University of Technology, Taipei, Taiwan.
5
Prof., Dept. of Construction Engr., National Taiwan University of Science and Technology, Taipei, Taiwan.
Introduction
A 4-years coordinated research effort on seismic retrofit of existing RC bridge columns
has been established at the National Center for Research on Earthquake Engineering (NCREE)
since 1998. Major objectives of this program are to develop effective seismic retrofit methods
of existing bridges in Taiwan due to (1) inadequate design strength, (2) inadequate
confinement at potential plastic hinge region, (3) inadequate shear strength due to large lateral
steel spacing, and (4) lap-splicing in the plastic hinge zone, etc., identified as some the most
severe weaknesses of the existing RC bridge columns for seismic hazard.. Observations of the
bridge damage during the 1999 Chi-Chi earthquake indicate that many existing bridges in
Taiwan are indeed vulnerable to major earthquakes and the coordinated research program
conducted at NCREE is necessary and urgent. This coordinated research program includes a
master plan administrated by NCREE and seven coordinated projects handled by the
investigators from six universities and research institutions. Results of this research program
will provide a domestic test database for seismic bridge engineering applications and to
provide seismic retrofit guidelines for highway officials in Taiwan. Many retrofit techniques
that have been extensively studied and widely implemented in the United States, Japan and
New Zealand are studied in this joint research.
Experimental Program
More than 60 large scale specimens were tested during the last three years, including 24
bench-mark specimens that were designed to represent typical pre- and after 1987 bridge
columns in Taiwan. Cross sectional dimensions of the rectangular columns and circular
columns are 600mm by 750 mm and 760 mm diameter, respectively, roughly 2/5 scale of the
prototype columns. The worst details that may be expected in the existing bridges are
assumed in the specimens, such as the double U-shaped transverse reinforcements with large
spacing, and the lap-spliced of main reinforcements at the plastic hinge zone. Retrofit
techniques used in the specimens include steel jacketing, FRP wrapping, and RC jacketing. In
addition, seismic performance of column-foundation connections, beam-column connections
as well as the wall type piers are also studied. Details of the test specimens are listed in Tables
1-3.
Summary of results
Seismic Retrofit of Rectangular RC Bridge Columns Using FRP Jacket
Test specimens
Flexural failure mode specimens
These specimens represent the benchmark and the CFRP wrapped, named as BMR2,
BMR3, FR1, and FR2 respectively. Specimen FR1 is retrofitted with 4 layers of FRP (0.55 )
along the whole height.
For specimen FR2, it is retrofitted based on the ductility requirement of 6. This specimen
is retrofitted in the plastic hinged zone with 8 layers of FRP (1.1 ), and the other areas are
retrofitted with 2 layers of FRP (0.275 ).
Lap-spliced failure mode specimens
These specimens represent the benchmark, CFRP wrapped, and combined steel plate and
CFRP, named as BMRL100, FRL100, SFRL100, respectively. For specimen FRL100, one
third of the column height (1100 mm) is retrofitted with 8 layers of CFRP (1.1mm), and the
other areas are with 4 layers. For specimen SFRL100, steel plates are attached to each column
face before wrapping the CFRP. Combining the steel plate, the cross section became a little
close to oval-shaped. The curvature of the shape is advantageous for FRP to produce inward
confinement stress to prevent bond slip failure.
Shear failure mode specimens
These specimens represent the benchmark and the CFRP wrapped, named as BMRS and
FRS, respectively. In order to observe the short column effect, the column height is reduced.
For specimen FRS, it is retrofitted with 4 layers of FRP (0.55 ) along the whole height.
Experimental Results and Discussion
Fig. 1-1 to 1-9 shows the lateral force and displacement relationships of the specimens.
The sequence from top to bottom is BMR2, BMR3, FR1, FR2, BMRL100, FRL100,
SFRL100, BMRS, and FRS. It is shown that both displacement ductility and energy
dissipation in BMRL100 and BMRS are quite poor. Compared to specimen BMRL100, it can
be seen that specimen SFRL100 performs very well. Not only it gets the 7.24 times of the
dissipation energy of the specimen BMRL100, but also enhances 6.03 times of the
displacement ductility, a value close to the flexural failure mode. This figure demonstrates
clearly that this retrofit method has a good potential in seismic retrofit of rectangular RC
columns lap-spliced at the plastic hinge zones.
Conclusions
1 Flexural failure mode
Seismic Retrofit Stud of RC Bridge Columns
1. Test results show that failure of the flexural type specimen under larger axial load will
result in speeding up the degradation of strength and energy dissipation capacity.
2. Standard hoop arrangements can gain better confinement than the double-U shaped
alternation arrangement used in many existing bridges.
3. The retrofit efficiency of force-based design and displacement-based design is nearly
the same. The displacement ductility levels of 7 can be reached.
2 Shear failure mode
1. Brittle shear failure occurs due to insufficient transverse reinforcement spacing.
2. Retrofitted by wrapping FRP shows great performance in improving shear strengths,
and transfers the failure mode to flexural-shear type.
3 Lap spliced failure mode
1. Without enough confinement stress, bond slip occurred between the lap-spliced
longitudinal reinforcements and resulted in brittle failure.
2. Applying CFRP directly cant provide enough confinement stress to increase frictional
force between the lap spliced longitudinal reinforcements.
3. A new method by attaching steel plates before wrapping FRP shows great potential in
increasing the confinement stress and energy dissipate capacity for rectangular RC members.
The strength, ductility, and energy dissipation capacity are also greatly improved. Further
study is necessary to better understand the mechanism and to determine the critical
parameters for retrofit design applications.
Reference
K.C. Chang, and F.S. Chung (2000), Seismic Shear and Lap-Spliced Retrofit of Rectangular
Bridge Columns using FRP, National Center for Research on Earthquake Engineering,
Technical Report.
K.C. Chang, and H.F. Chang (2000), Seismic Flexural Retrofit of Rectangular Bridge
Columns using FRP, National Center for Research on Earthquake Engineering,
Technical Report.
K.C. Chang, K.Y. Liu, F.S. Chung, and S.B. Chang (2000), Seismic Retrofit study of RC
rectangular bridge column lap spliced at plastic hinge zone , Proceeding of the thirteen
KKNN symposium on Civil Engineering, December 7-8,2000, Taipei, Taiwan.
-250 -200 -150 -100 -50 0 50 100 150 200 250
Displacement (mm)
-400
-200
0
200
400
L
a
te
r
a
l F
o
r
c
e
(
k
N
)
-1
0
1
P
/P
y
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7
Drift Ratio (%)
0 1 2 3 4 5 6
BMR2
Py = 240 kN
Dy = 20.58 mm
Pmax = 303.37 kN
Dmax = 65 mm
Pu = 242.70 kN
Du = 130 mm
-250 -200 -150 -100 -50 0 50 100 150 200 250
Displacement (mm)
-400
-200
0
200
400
L
a
te
r
a
l F
o
r
c
e
(
k
N
)
-1
0
1
P
/P
y
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7
Drift Ratio (%)
0 1 2 3 4
BMR3
Py = 251.25 kN
Dy = 32.72 mm
Pmax = 260 kN
Dmax = 48.75 mm
Pu = 208.0 kN
Du = 130.0 mm
Fig 1-1 Hysteresis curve of specimen BMR2 Fig1-2. Hysteretic curve of specimen BMR3
S sm c tro t St d o C Br d Col mns
-250 -200 -150 -100 -50 0 50 100 150 200 250
Displacement (mm)
-400
-200
0
200
400
L
a
te
r
a
l F
o
r
c
e
(
k
N
)
-1
0
1
P
/P
y
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7
Drift Ratio (%)
0 2 4 6 8
FR1
Py = 245.0 kN
Dy = 15.3 mm
Pmax = 357.77 kN
Dmax = 97.5 mm
Pu = 286.22 kN
Du = 130.0 mm
-250 -200 -150 -100 -50 0 50 100 150 200 250
Displacement (mm)
-400
-200
0
200
400
L
a
te
r
a
l F
o
r
c
e
(
k
N
)
-1
0
1
P
/P
y
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7
Drift Ratio (%)
0 1 2 3 4 5 6 7
FR2
Py = 245.0 kN
Dy = 18.04 mm
Pmax = 297.96 kN
Dmax = 65 mm
Pu = 238.37 kN
Du = 130.0 mm
Fig1-3. Hysteretic curve of specimen FR1 Fig1-4. Hysteretic curve of specimen FR2
-250 -200 -150 -100 -50 0 50 100 150 200 250
Displacement (mm)
-600
-400
-200
0
200
400
600
L
a
te
r
a
l F
o
r
c
e
(
k
N
)
-2
-1
0
1
2
P
/P
y
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7
Drift Ratio (%)
01
BMRL100
Py = 275.0 kN
Dy = 14.21 mm
Pmax = 328.0 kN
Dmax = 24.3 mm
Pu = 262.4 kN
Du = 21.93 mm
-250 -200 -150 -100 -50 0 50 100 150 200 250
Displacement (mm)
-600
-400
-200
0
200
400
600
L
a
te
r
a
l F
o
r
c
e
(
k
N
)
-1
0
1
P
/P
y
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7
Drift Ratio (%)
0 1 2
FRL100
Py = 384.0 kN
Dy = 32.46 mm
Pmax = 415 kN
Dmax = 48.83 mm
Pu = 332.0 kN
Du = 65.94 mm
Fig1-5. Hysteretic curve of specimen BMRL100 Fig1-6. Hysteretic curve of specimen FRL100
-250 -200 -150 -100 -50 0 50 100 150 200 250
Displacement (mm)
-600
-400
-200
0
200
400
600
L
a
te
r
a
l F
o
r
c
e
(
k
N
)
-1
0
1
P
/P
y
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7
Drift Ratio (%)
SFRL100
Py = 400.0 kN
Dy = 21.5 mm
Pmax = 512.75 kN
Dmax = 97.39 mm
Pu = 410.2 kN
Du = 129.75 mm
0 1 2 3 4 5 6
-150 -100 -50 0 50 100 150
Displacement (mm)
-1200
-1000
-800
-600
-400
-200
0
200
400
600
800
1000
1200
L
a
te
r
a
l F
o
r
c
e
(
k
N
)
-1
0
1
P
/P
y
-8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8
Drift Ratio (%)
01
BMRS
Py = 657.0 kN
Dy = 17.20 mm
Pmax = 681.0 kN
Dmax = 26.78 mm
Pu = 544.8 kN
Du = 26.0 mm
Fig1-7. Hysteretic curve of specimen SFRL100 Fig1-8. Hysteretic curve of specimen BMRS
-150 -100 -50 0 50 100 150
Displacement (mm)
-1200
-1000
-800
-600
-400
-200
0
200
400
600
800
1000
1200
L
a
te
r
a
l F
o
r
c
e
(
k
N
)
-1
0
1
P
/P
y
-8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8
Drift Ratio (%)
0 1 2 3 4 5
FRS
Py = 745.0 kN
Dy = 12.85 mm
Pmax = 941.0 kN
Dmax = 53.0 mm
Pu = 752.0 kN
Du = 74 mm
Fig1-9. Hysteretic curve of specimen FRS
Steel Jacketing Retrofit Design
Lateral confinement deficient specimens
The steel jacketing details for the retrofitted specimens were shown in Fig. 2-1. The SR1
and SR3 were retrofitted using the octagonal steel jacket, while the SR2 and SR4 were
retrofitted with the elliptical steel jacket. For the SR2, the dimensions of the ellipse were
computed in order to minimize the overall cross-sectional area. The 3-mm thick A36 grade
steel elliptical jacket for the SR2, SR4 was designed considering the equivalent radius (Sun et
al. 1993) and a target ductility capacity of 6 following the procedures recommended in the
reference (Priestley et al. 1996). For the purposes of comparison, the same 3-mm thick steel
plate was adopted in the fabrication of the octagonal jacket for the SR1. Specimen SR3 has a
S ismic R tr fit Stud f RC Brid C lum s
smaller octagonal cross sectional area, but a thicker steel jacket (6mm) than SR1. The
thickness of the steel jacket for the SR3 is computed from the static equilibrium assuming the
same confinement pressure to be developed in Specimens SR1. Noted that the dimensions of
the elliptical jacket for SR4 is slightly different from those of the SR2 in order to ensure the
rods for applying the vertical load not to impinge on the retrofitted column.
Lap-splice deficient specimens
For the retrofitted specimens, SRL1 was retrofitted with a 6mm thick octagonal steel jacket
while SR2 was retrofitted with a 3mm thick elliptical steel jacket. Fig. 2-1 Details of the steel
jacketing.
Shear-deficient specimens
The SRS1 was retrofitted with an octagonal steel jacket, and the SRS2 was retrofitted with
a rectangular steel jacket. The 3mm thick steel plate was adopted in both Specimens SRS1
and SRS2. Fig. 2-1 Details of the steel jacketing.
Response of the retrofitted specimens
Test results given in Fig.2-2 and Fig.2-3 confirm that the seismic performance of
rectangular RC bridge columns can be significantly and equally enhanced by properly
designed elliptical or octagonal steel jacket following the procedures noted above. Bridge
columns retrofitted with the octagonal or the elliptical steel jacket exhibit stable lateral
force-displacement hysteretic response, possess excellent displacement ductility and energy
dissipation capacities.
For the SRS2 specimen retrofitted with a rectangular steel jacket, outward bulging of the
steel jacket was observed at a drift angle of 0.03 radian. In the mean time, significant strength
degradation occurred due to the loss of the lateral confinement. Test results confirmed that the
rectangular steel jacketing is ineffective in providing lateral confinement.
Conclusions
1. Test results confirmed that the seismic performance of the rectangular RC bridge
columns can be significantly and equally enhanced by properly constructed elliptical or
octagonal steel jacket.
2. It is found that the cost of the octagonal jacketing is 15% lower than that for the
elliptical jacketing in the case studied of SR1 and SR2 specimens.
3. Using a thicker jacket can reduce the cross-sectional area of an octagonal steel jacketing
rectangular RC bridge column. The thickness of the steel jacket can be determined from the
static equilibrium assuming the same confinement pressure is to be developed for different
thickness of the steel jacket.
4. Rectangular steel jacketing can effectively prevent a shear-deficient column from shear
failure; however, it is not effective in improving the flexural ductility.
5. Octagonal steel jacketing scheme is cost-effectively and can provide lateral confinement
and the shear strength to mitigate seismic failures of rectangular RC bridge columns due to a
lack of lateral confinement, improper lap-splice or inadequate shear capacity.
6. A smaller cross-sectional area and better seismic performance than the elliptical steel
jacketing scheme have been achieved from the octagonal steel jacketing.
Reference
S ismic R tr fit Stud f RC Brid C lum s
Sun, Z.L., Seible, F. and Priestley, M.J.N. (1993), Flexural Retrofit of Rectangular Reinforced
Concrete Bridge Columns by Steel Jacketing, Structural System Research Project, Report
No. SSRP-93/01, Department of Applied Mechanics and Engineering Science, U.C. San
Diego.
Tsai, K.C. and Lin, M.L. (2000), Steel Jacketing for Seismic Retrofit of RC Rectangular
Columns, National Center for Research on Earthquake Engineering, Technical Report
43.5cm
54cm
SR2
Steel Plate t=3 mm , A36
43.5cm
54cm
Steel Plate t=3 mm , A36
SR1
Steel Plate t=6 mm , A36
60cm
42cm
SR3
50cm
37cm
Steel Plate t=3 mm , A36
60cm
SR4
42cm
(a) lateral confinement deficient specimen
SRL1
39cm
48cm
Steel Plate t= 6 mm ,A36
46cm
56cm
Steel Plate t= 3 mm ,A36
SRL2
(b) lap-splice specimen
Steel Plate t= 3 mm ,A36
48cm
39cm
SRS1
SRS2
Steel Plate t= 3 mm ,A36
33 cm
40cm
(c) shear-deficient specimen
Fig. 2-1 Details of the steel jacketing
-500
-250
0
250
500
-8 -6 -4 -2 0 2 4 6 8
Drift Rati o(%radian)
F
o
r
c
e
(
k
N
)
SR-1
Vmax=456kN
-500
-250
0
250
500
-8 -6 -4 -2 0 2 4 6 8
Drift Ratio(%radian)
F
o
r
c
e
(
k
N
)
BMR-3
Vmax=290kN
-500
-250
0
250
500
-8 -6 -4 -2 0 2 4 6 8
Drift Ratio(%radian)
F
o
r
c
e
(
k
N
)
SR-2
Vmax=400kN
-500
-250
0
250
500
-8 -6 -4 -2 0 2 4 6 8
Drift Rati o(%radian)
F
o
r
c
e
(
k
N
)
SR-3
Vmax=418kN
-500
-250
0
250
500
8 6 4 2 0 2 4 6 8
F
o
r
c
e
(
k
N
)
SR-4
Vmax=423kN
-500
-250
0
250
500
8 6 4 2 0 2 4 6 8
F
o
r
c
e
(
k
N
)
BMR-3
SR-1
SR-2
SR-3
SR-4
P=0.15f'cAg
=1400kN
(a) lateral confinement deficient specimen
-800
-400
0
400
800
-8 -6 -4 -2 0 2 4 6 8
Drift Ratio(%)
F
o
r
c
e
(
k
N
)
BMRL100
P=0.15f'cAg
=1400kN
Vmax=- 368kN
-800
-400
0
400
800
-8 -6 -4 -2 0 2 4 6 8
Drift Ratio(%)
F
o
r
c
e
(
k
N
)
SRL1
P=0.15f'cAg
=1400kN
Vmax= - 622kN
-800
-400
0
400
800
-8 -6 -4 -2 0 2 4 6 8
Drift Ratio(%)
F
o
r
c
e
(
k
N
)
SRL2
P=0.15f'cAg
=1400kN
Vmax=602kN
-800
-400
0
400
800
-8 -6 -4 -2 0 2 4 6 8
Drift Ratio(%)
F
o
r
c
e
(
k
N
)
BMRL100
SRL1
SRL2
(b) lap-splice specimen
-1200
-600
0
600
1200
-8 -6 -4 -2 0 2 4 6 8
Drift Ratio(%)
F
o
r
c
e
(
k
N
)
BMRS
SRS1
SRS2
FRS
-1200
-600
0
600
1200
-8 -6 -4 -2 0 2 4 6 8
Dri ft Ratio(%)
F
o
r
c
e
(
k
N
)
SRS2
P=0.15f'cAg
=1400kN
Vmax= 982kN
-1200
-600
0
600
1200
-8 -6 -4 -2 0 2 4 6 8
Dri ft Ratio(%)
F
o
r
c
e
(
k
N
)
SRS1
P=0.15f'cAg
=1400kN
Vmax= 1086kN
-1200
-600
0
600
1200
-8 -6 -4 -2 0 2 4 6 8
Dri ft Ratio(%)
F
o
r
c
e
(
k
N
)
BMRS
P=0.15f'cAg
=1400kN
Vmax=- 722kN
(c) shear-deficient specimen
Fig. 2-2 Cyclic load-deflection relationships
Se sm c etro t St d o C Br dge Co m s
(a)
0
50
100
150
200
250
0 1 2 3 4 5 6
Cyclic Drift Ratio(% radian)
E
n
e
r
g
y
D
i
s
s
i
p
a
t
i
o
n
(
k
N
-
m
)
BMR3
SR1
SR3
SR2
SR4
Each Two Cycles
0
50
100
150
200
250
300
0 1 2 3 4 5 6
Cyclic Drift Ratio(% radian)
E
n
e
r
g
y
D
i
s
s
i
p
a
t
i
o
n
(
k
N
-
m
)
BMR100
SRL1
SRL2
Each Two Cycles
0
50
100
150
200
250
0 1 2 3 4 5 6
Cyclic Drift Ratio(% radian)
E
n
e
r
g
y
D
i
s
s
i
p
a
t
i
o
n
(
k
N
-
m
)
BMRS
SRS1
SRS2
Each Two Cycles
Fig. 2-3 Energy dissipation curves
Seismic Retrofit of Circular RC Bridge Columns Using Steel Jacket
Experimental Program and Results
In the test program (Hwang and Hseih 1999, Hwang and Kuo 2000 ), four 0.4-scaled
down test specimens are designed according to the details of existing circular RC bridge
columns. The retrofitted specimens corresponding to the weaknesses of the four existing
column details are also designed (Hwang et al 1998) and tested for validation purpose. The
targeted weaknesses of the existing bridge details include (1) the lack of volumetric ratio of
lateral hoop; (2) the early termination of 50% main reinforcement at the mid-height of the
column, together with a much larger pitch of the lateral hoop at the upper region of the
column; (3) the lap-splice failure at the plastic hinge zone; and (4) the shear failure of short
columns. The tests results are summarized in Fig. 3-1 to Fig. 3-4. From the figure it is
concluded that the retrofit using steel jacketing is effective in enhancing the seismic resistance
of the existing circular RC bridge columns in Taiwan.
References
Kuo, S.S. (1998) Seismic evaluation and retrofit of existing highway bridges with circular
single column bents. M.S. Thesis, Dept. of Construction Engineering, National Taiwan
University of Science and Technology.
Hwang, J.S. and Hseih, Y.M. (1999) Seismic retrofit of RC bridge columns suing steel
jacket. Report No. 99016, National Center for Research on Earthquake Engineering of
Taiwan.
Hwang, J.S. and Kuo, M.Y. (2000) Seismic setrofit of existing RC bridge columns shear
strength and lap splice retrofit Report No. National Center for Research on Earthquake
Engineering of Taiwan.
Hwang, J.S, et. al. (1998) Seismic retrofit manual for highway bridges, Research Report,
Department of Transportation and Communication, Taiwan