0350

SHEAR PERFORMANCE OF RC MEMBERS STRENGTHENED WITH

EXTERNALLY BONDED FRP WRAPS

AHMED KHALIFA1, ABDELDJELIL BELARBI2 And ANTONIO NANNI3

SUMMARY

This study presents the shear performance and the modes of failure of reinforced concrete (RC)
beams strengthened with externally bonded carbon fiber reinforced polymer (CFRP) wraps. The
experimental program consisted of testing twenty-seven, full-scale, RC beams. The variables
investigated in this research study included steel stirrups (i.e., beams with and without steel
stirrups), shear span-to depth ratio (i.e., a/d ratio 3 versus 4), CFRP amount and distribution (i.e.,
continuous wrap versus strips), bonded surface (i.e., lateral sides versus U-wrap), fiber orientation
(i.e., 900/00 fiber combination versus 900 direction), and end anchor (i.e., U-wrap with and without
end anchor). As part of the research program, the experimental study examined the effectiveness
of CFRP reinforcement in enhancing the shear capacity of RC beams in negative and positive
moment regions, and for beams with rectangular and T-cross section. The experimental results
indicated that the contribution of externally bonded CFRP to the shear capacity is significant and
dependent upon the variable investigated.

INTRODUCTION

Strengthening and rehabilitation of existing reinforced concrete (RC) structures is becoming an important issue
in situations such as demand in the increase of service load levels, repair due to degradation of a member,
design/construction defects, and response to requirements of newly developed design guidelines. Carbon fiber
reinforced polymer (CFRP) sheets continue to show great promise for use in these situations (ACI 440, 1996).
These materials are excellent for external strengthening because of their high tensile strength, light weight,
resistance to corrosion, superior durability, and cost-effective installation process.

Shear failure of RC beams, caused by their brittle nature, has been identified as the most disastrous failure
mode; it occurs with no advance warning of distress. Shear deficiency may occur due to many factors such as
insufficient shear reinforcement or reduction in steel area due to corrosion, increased service load, and
construction errors. In addition, there is an urgent need to upgrade shear resistance of older RC structures to
meet current seismic design standards in regions with high seismicity. In these situations, externally bonded
CFRP reinforcement is used to wrap the beam cross-section with fibers in the transverse direction and thus
enhances the shear resistance of the beam (Chajes et al. 1995, Taerwe et al. 1997, and Triantafillou 1998). The
strengthening is achieved by having CFRP reinforcement laid in a way to cross diagonal cracks similarly to
internal stirrups to provide the shear resistance and restrict the growth of diagonal cracks and therefore reduces
their progression into the compression

zone. Thus large uncracked concrete area is available at the head of the crack and thereby increases the part of
shear force carried by concrete. To date, only a limited number of studies has specifically addressed shear
strengthening with FRP. The parameters affecting the shear strength of the strengthened beams need to be
investigated and this study will address that research void and set the plan for further research.

1
Structural Engineering Department, Faculty of Engineering, Alexandria University, Alexandria 21544, Egypt
2
Department of Civil Engineering, University of Missouri-Rolla, Rolla, MO 65409,USA
3
Department of Civil Engineering, University of Missouri-Rolla, Rolla, MO 65409,USA
 The objectives of this study were to: (1) investigate the shear performance and mode of failure of RC beams
after strengthening with externally bonded CFRP sheets, (2) examine the effectiveness of CFRP reinforcement in
enhancing the shear capacity of RC beams in negative and positive moment regions, as well as RC beams with
rectangular and T-cross section; (3) address the variables affecting the shear capacity of strengthened beams such
as: steel stirrups, shear span-to depth ratio, CFRP amount and distribution, bonded surface, fiber orientation, and
end anchor. To fulfill these goals, twenty-seven full-scale, RC beams design to fail in shear, were constructed
and strengthened with different CFRP configurations and full study was carried out.

EXPERIMENTAL PROGRAM

Test Specimens, Test Setup, and Materials

The beam specimens tested in this experimental program were grouped into three main series designated as A, B,
and C. In Series A, twelve full-scale rectangular beam specimens were tested. The variable investigated in this
test series included steel stirrups, shear span-to-depth ratios (a/d ratios), CFRP amount and distribution. The
specimens were grouped into two main groups designated as A-SW for beams with stirrups and A-SO for beams
without stirrups in the shear span of interest. The stirrups were made from deformed steel bars with 10-mm
diameter bars, with yield stress of 350 MPa, ultimate tensile strength of 530 MPa, and modulus of elasticity of
200 GPa. Four 32-mm bars with yield stress of 460 MPa were used as longitudinal reinforcement with two
being placed at top and two at bottom face of the cross section. Each main group (i.e., Groups A-SW and A-SO)
was subdivided into two subgroups according to shear span-to-depth ratio namely: a/d = 3 and 4, and resulting in
the following four Subgroups: A-SW3, A-SW4, A-SO3, and A-SO4. All specimens of Series A were tested as
simply supported beams subjected to a four-point load. A universal testing machine with 1800 kN capacity was
used to apply a concentrated load on a steel distributed beam used to generate the two concentrated loads. A
summary of structural system, cross-section dimensions and details, shear span-to-depth ratio (a/d), steel shear
reinforcement, and CFRP strengthening configurations is listed in Table 1.

In the negative moment regions of continuous beams, shear cracks initiates from the top of the section. In this
case, the U-wrap FRP reinforcement may not be able to control the initiation of these cracks, and may have less
effectiveness to enhance shear capacity. However, most of the past research has dealt with shear strengthening
of simply supported beams (strengthening in positive moment regions) and shear strengthening in negative
moment regions has not been addressed. To fill this gap, nine full-scale, two-span-continuous rectangular beam
specimens were fabricated and tested (Series B). The variables investigated in this test series included steel
stirrups, CFRP amount and distribution, and CFRP wrapping schemes. The specimens of Series B were
subdivided into three groups designated as B-CW, B-CO, and B-CF. Each group had different longitudinal and
shear steel reinforcement ratios as shown in Table 1. The specimens were tested as continuous beams under
concentrated loads applied to the mid-point of each span. Two load cells were used to monitor total applied load
and reaction in the test span. This allowed the computation of the exact shear force in the span of interest.

In Series C, six full-scale, T-section RC beams were strengthened with different CFRP configurations and tested.
T-section beams are of great importance because they are the most commonly used in practice. Also, they
represent a more challenging case than rectangular beams due to the flange that reduces the FRP bonded length
over the web. The specimens were strengthened with different CFRP configurations. The selected parameters
were; (a) CFRP amount and distribution (i.e., continuous wrap versus strips); (b) bonded surface (i.e., lateral
sides versus U-wrap); (c) fiber orientation (i.e., 900-00 fiber combination versus 900 direction); and (d) end
anchorage (i.e., U-wrap with and without end anchor). All specimens were tested as simple beams using a four-
point loading with shear span-to-depth ratio (a/d) equals to 3. A steel distribution beam used to generate the two
concentrated loads.

The composite strengthening system that was used in this research program was provided by Master Builder
Technologies, Inc. According to the manufacturer’s information, the tensile strength of CFRP sheet is 3790
MPa, the modulus of elasticity is 228 GPa, and the design thickness is 0.165 mm (fiber only). Fabrication of the
specimens including surface preparation and CFRP installation is described elsewhere [Khalifa 1999].

2 0350
 Table 1. Summary of Test Specimens

Concrete
Shear reinforcement

Structural system and test set-up a/d strength

Specimen
No. Cross-section details
designation Steel stirrups
(dimensions in mm) ratio (f′c)
CFRP
Simply supported beams in test region
(MPa)

1 A-SW3-1 3 19.3 φ10@125mm ----

2 A-SW3-2 3 19.3 φ10@125mm Two plies (90°/0°)
3 A-SW4-1 4 19.3 φ10@125mm ----
4 A-SW4-2 610 760 310 760 610 4 19.3 φ10@125mm Two plies (90°/0°)
5 A-SO3-1 2 φ 32 3 27.5 ---- ----
305 mm
6 A-SO3-2 Series A-SW3 and A-SO3 (a/d=3) 2 φ 32 3 27.5 ---- U-wrap strips, 50 @ 125mm
150 mm
7 A-SO3-3 3 27.5 ---- U-wrap strips, 75 @ 125mm
8 A-SO3-4 405 1020 200 1020 405 3 27.5 ---- One ply continuous U-wrap
9 A-SO3-5 3 27.5 ---- Two plies (90°/0°)
10 A-SO4-1 4 27.5 ---- ----
11 A-SO4-2 4 27.5 ---- U-wrap strips, 50 @ 125mm
Series A-SW4 and A-SO4 (a/d=4)
12 A-SO4-3 4 27.5 ---- One ply continuous U-wrap

13 B-CW1 Continuous beams 3.6 27.5 φ10@125mm ----

2 φ 32
14 B-CW2 3.6 27.5 φ10@125mm Two plies (90°/0°)
305 mm
15 B-CO1 2 φ 32 3.6 20.5 ---- ---
150
16 B-CO2 460 3.6 20.5 ---- U-wrap strips, 50 @ 125mm
17 B-CO3 3.6 20.5 ---- One ply continuous U-wrap
915 915 300 915 915
18 B-CF1 3.6 50 ---- ----
460 2 φ 16
19 B-CF2 305 mm 3.6 50 ---- One ply continuous U-wrap
2 φ 16
20 B-CF3 150 3.6 50 ---- Two plies (90°/0°)
21 B-CF4 3.6 50 ---- One ply, totally wrapped

22 C-BT1 Simply supported beams 3 35 ---- ----

380
23 C-BT2 3 35 ---- One ply continuous U-wrap
2 φ 13 100
24 C-BT3 3 35 ---- Two plies (90°/0°)
305
25 C-BT4 2 φ 28 3 35 ---- U-wrap strips, 50 @ 125mm
355 1070 200 1070 355 150
26 C-BT5 3 35 ---- Two sides strips 50 @ 125
27 C-BT6 3 35 ---- U-wrap with end anchor

Strengthening Schemes

Of the twenty-seven specimens, one from each subgroup (eight specimens) was not strengthened and was
considered as a control specimen, whereas nineteen specimens were strengthened with externally bonded CFRP
sheets using different schemes.

Specimens A-SW3-2, A-SW4-2, A-SO3-5, B-CW2, B-CF3, and C-BT3 were strengthened with two CFRP plies
having perpendicular fiber directions (900/00). The first ply was attached in the form of continuous U-wrap with
the fiber direction oriented perpendicular to the longitudinal axis of the specimen (900). The second ply was
bonded on the two sides of the specimen with the fiber direction parallel to the beam axis (00). This ply (i.e., 00

3 0350
ply) was added to investigate the impact of horizontal restraint on shear strength. Specimens A-SO3-4, A-SO4-
3, B-CO3, B-CF2, and C-BT2 were strengthened with one-ply continuous U-wrap (900).
Specimen C-BT6 was strengthened with one-ply continuous U-wrap (900). The ends of the U-wrap were
anchored to the flanges on both sides of the specimen using a proprietary U-anchor system developed at
University of Missouri-Rolla (UMR). The purpose of using the end anchor was to address the problems
associated with the debonding of FRP from the concrete surface and to allow a better exploitation of the
strengthening system. A cross section showing details of the U-anchor system is given in Figure 1. The
installation procedure of the end anchor is described elsewhere [Khalifa et al. 1999 a].

Saturant
FRP rod
Flange
Paste

Web
FRP reinforcement

Figure 1. Details of the U-Anchor at the corner of flange-web

Specimen B-CF4 was totally wrapped with one-ply CFRP sheets. The sheets were attached to the four sides of
the specimen with an overlap on the topside. Even though total wrapping may not be possible in the field, this
case is representative of the upper threshold.

Specimens A-SO3-2, A-SO4-2, B-CO2, and C-BT4 were strengthened with one-ply CFRP strips in the form of
U-wrap with (900). The strip width was 50 mm with center–to-center spacing of 125 mm. Specimen A-SO3-3
was strengthened in a manner similar to that of Specimen A-SO3-2 but with strip width equal to 75 mm.
Specimen C-BT5 was strengthened with CFRP strips attached only on the two beams sides with 900 fiber
orientation. The strips width and spacing were similar to Specimen C-BT4.

TEST RESULTS AND DISCUSSIONS

The test results confirm that the strengthening technique using CFRP sheets can be used to increase significantly
shear capacity. The recorded CFRP strain of the tested specimens indicates that the failure of CFRP system
occurs at an average effective stress level below nominal strength due to stress concentrations or debonding of
CFRP from concrete surface.

Failure Modes

The failure mode of the control specimens was shear compression failure while the failure mode of the
strengthened specimens was either CFRP debonding, concrete splitting on a vertical plane, or flexural failure.
Figure 2 shows examples of the observed failure modes. In each specimen failed by CFRP debonding, the final
crack pattern was approximately similar to the control specimen. The failure initiated due to debonding of the
CFRP from concrete surface with spalled concrete attached to it, followed directly by shear compression failure.
The location of the debonding area varied according to the wrapping schemes. For specimens strengthened with
U-wrap configuration, the debonding area was above the diagonal shear crack (Figure 2 (a)). In Specimen C-
BT5, strengthened with CFRP strips attached to the specimen sides only, the location of debonding area was
below the main shear crack as shown in Figure 2(b). The specimens failed by concrete splitting were beams
with rectangular cross-sections and beams strengthened with either two perpendicular plies 900/00 or one ply
continuous U-wrap. No cracks were visible on the sides or bottom of the test specimen due to the FRP
wrapping. Before the failure, a longitudinal crack formed on the top surface of the specimen. The crack
initiated close to the position of applied load and extended towards the support then the failure occurred by
concrete splitting on a vertical plane as shown in the Figure 2 (c). The shear capacity of the strengthened
specimens of Group B-CF and Specimen C-BT6 was higher than their flexural capacity. For those specimens,
the failure was controlled by flexural (Figure 2 (d)).

4 0350
 (a) Specimen A-SO3-2 (Debonding of CFRP (b) Specimen C-BT5 (Debonding of CFRP below
over shear crack) shear crack)

(c) Specimen A-SO3-5 (concrete splitting) (d) Specimen B-CF4 (flexural failure)

Figure 1. Examples of Failure Modes of some Test Specimens

Shear Force-Deflection Curves

The shear force versus the mid-span deflection curves of the tested beams are shown in Figure 3. For all test
series, the strengthened specimens showed a higher failure load compared to the control specimens. In Subgroup
A-SW4, the mid-span deflection of the strengthened Specimen A-SW4-2 at ultimate was about 2.5 times the
deflection of the control Specimen A-SW4-1. In Subgroup A-SO3 and Group B-CO (Figures 3 c & f), the
strengthened specimens had a more brittle behavior than the control specimen. In Group B-CF (Figure 3 (g)),
the Specimen B-CF4, strengthened with totally wrapped CFRP sheets showed a large plateau and a notable
increment in ductility. In Series C (Figure 3 (h)), the Specimen C-BT6, strengthened with U-wrap continuous
CFRP sheets with end anchor, showed significant increase in the shear capacity compared to other specimen in
the series. In addition, Figure 3 (h) indicates that Specimen C-BT6 gained more stiffness and ductility. The
additional ductility was obtained from the flexural failure mode. The mid- span deflection of Specimen C- BT6
at failure was about 3 times the deflection of Specimen C-BT2, strengthened with U-wrap continuous CFRP
sheets without end anchor, at ultimate.

Evaluation of the Test Results

The summary of the test results for all of the beam specimens are detailed in Table 2.

Series A: For the specimens tested in this series, increases in shear strength of 40 to 138% were achieved. The
test results indicated that contribution of CFRP benefits the shear capacity at a greater degree for beams without
shear reinforcement than for beams with adequate shear reinforcement. In addition, the contribution CFRP
reinforcement was influenced by the a/d ratio and appeared to increase with increasing the a/d ratio. Based
upon the test results of

5 0350
 200 200

180 180
160 160
140 140
Shear force (kN)

Shear force (kN)

120 120
100 100
80 80
60 60
40 40
A-SW3-1
20 20 A-SW4-1
A-SW3-2 A-SW4-2
0 0
0 5 10 15 20 25 0 5 10 15 20 25
Mid-span deflection (mm) Mid-span deflection (mm)
(a) Subgroup A-SW3 (b) Subgroup A-SW4
200 200

180 180
160 160
140 140

Shear force (kN)

Shear force (kN)

120 120
100 100
80 80
60 A-SO3-1 60
A-SO3-2 A-SO4-1
40 40
A-SO3-3 A-SO4-2
20 A-SO3-4 20
A-SO3-5 A-SO4-3
0 0
0 5 10 15 20 25 0 5 10 15 20 25
Mid-span deflection (mm) Mid-span deflection (mm)

(c) Subgroup A-SO3 (d) Subgroup A-SO4

250 120

100
200
Shear force (kN)
Shear force (kN)

80
150
60
100
40

50 B-CO1
B-CW1 20
B-CO2
B-CW2 B-CO3
0 0
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Mid-span deflection (mm) Mid-span deflection (mm)
(e) Group B-CW (f) Group B-CO

160 250

140
200
120
Shear force (kN)
Shear force (kN)

100 150
80
100 C-BT1
60
C-BT2
B-CF1
40 C-BT3
B-CF2 50 C-BT4
20 B-CF3 C-BT5
B-CF4 C-BT6
0 0
0 5 10 15 20 25 30 35 0 5 10 15 20 25
Mid-span deflection (mm) Mid span deflection (mm)

(g) Group B-CF (h) Series C

Figure 3. Shear Force versus Mid-span Deflection for the Tested Specimens

6 0350
Specimens A-SO3-2 and A-SO3-4, increasing the amount of CFRP may not result in a proportional increase in
the shear strength. The CFRP amount used to strengthen Specimen A-SO3-4 was 250% of that used in
Specimen A-SO3-2, which resulted in a minimal (10%) increase in shear capacity. Moreover, the results of
Subgroup A-SO3 indicated that the added 00 ply improved the shear capacity by providing horizontal restraint.

Series B: In this test series, the shear behavior and modes of failure of two-span continuous RC beams
strengthened with CFRP sheets were investigated. The test results indicated that the externally bonded
reinforcement could be used to enhance the shear capacity of the beams in positive and negative moment
regions. For the beam specimens tested in this series, increase in shear strength ranged from 22 to 135%.

Series C: In this test series, the shear performance of the T beams strengthened with CFRP sheets was
investigated. For the beam specimens tested in this series, increase in shear strength of 35 to 145% was
achieved. The test results indicated that the performance of CFRP could be improved significantly if adequate
anchorage is provided. On other hand, applying CFRP to the beam sides only is less effective than a U-wrap.
The test results of this series also indicated that there exists an optimum amount of FRP, beyond which the
strengthening effect become inefficient.

Table 2. Summary of the Test Results

No. Specimen Failure Maximum Total applied Total applied Contribution CFRP
designation mode vertical CFRP load at shear force of CFRP to strengthening
strain measured ultimate the shear effectiveness
at ultimate at ultimate capacity ratio
(mm/mm)
(kN) %
(kN)
(kN)

1 A-SW3-1 Shear ---- 253.0 126.5 0.0 ----

2 A-SW3-2 Splitting 0.0023 354.0 177.0 50.5 40
3 A-SW4-1 Shear ---- 200.0 100.0 0.0 ---
4 A-SW4-2 Splitting 0.0019 361.0 180.5 80.5 80
5 A-SO3-1 Debonding ---- 154.0 77.0 0.0 ---
6 A-SO3-2 Debonding 0.0047 262.0 131.0 54.0 70
7 A-SO3-3 Debonding 0.0052 266.0 133.5 56.5 73
8 A-SO3-4 Debonding 0.0045 289.0 144.5 67.5 87
9 A-SO3-5 Splitting 0.0043 339.0 169.5 92.5 120
10 A-SO4-1 Shear ---- 130.0 65.0 0.0 ---
11 A-SO4-2 Debonding 0.0062 255.0 127.5 62.5 96
12 A-SO4-3 Splitting 0.0043 310.0 155.0 90.0 138
13 B-CW1 Shear ---- 508.0 175.0 0.0 ----
14 B-CW2 Splitting 0.0027 623.0 214.0 39.0 22
15 B-CO1 Shear ---- 220.0 48.0 0.0 ----
16 B-CO2 Debonding 0.0047 265.0 88.0 40.0 83
17 B-CO3 Debonding 0.0037 330.0 113.0 65.0 135
18 B-CF1 Shear ---- 268.0 93.0 0.0 0.0
19 B-CF2 Flexural Not available 337.0 119.0 > 26.0 > 28
20 B-CF3 Flexural Not available 394.0 131.0 > 38.0 > 40
21 B-CF4 Flexural Not available 400.0 140.0 > 47.0 > 50
22 C-BT1 Shear ---- 180.0 90.0 0.0 0.0
23 C-BT2 Debonding 0.0045 310.0 155.0 65.0 72
24 C-BT3 Debonding 0.0044 315.0 157.5 67.5 75
25 C-BT4 Debonding 0.0100 324.0 162.5 72.0 80
26 C-BT5 Debonding Not available 243.0 121.5 31.5 35
27 C-BT6 Flexural 0.0063 442.0 221.0 > 131.0 > 145

CONCLUSIONS

The tests results described in this study indicated that the strengthening technique based on externally bonded
CFRP composites can be used to increase significantly shear capacity of RC beams, with efficiency that varies
depending on the test variables. For all beams included in this experimental program, results show that an

7 0350
increase in shear strength of 22 to 145% was achieved. Based on the experimental results, analytical
investigations, and discussions, the main conclusions are as follows:

• Externally bonded CFRP reinforcement can be used to enhance the shear capacity of RC beams in positive
and negative moment regions.
• The FRP strengthening technique is applicable and can increase the shear capacity of rectangular as well as
T beams.
• The experimental verification of the end anchor system showed its effectiveness in increasing shear
capacity.
• The contribution of CFRP benefits the shear capacity at a greater degree for beams without shear
reinforcement than for beams with adequate shear reinforcement.
• The contribution of externally CFRP reinforcement to the shear capacity was influenced by the shear span-
to- depth ratio (a/d) and appeared to increase with an increase a/d ratio.
• Increasing the amount of CFRP may not result in a proportional increase in the shear strength because the
shear strength is significantly dependent on the interfacial bond between the FRP and concrete. This means
that, if FRP debonding failure is not prevented, there is an optimum amount of FRP, beyond which the
capacity dose not increase with increasing amount of FRP.
• The presence of 00 ply may improve the shear capacity by providing horizontal restraint to diagonal shear
cracks.
• Applying CFRP to the beam sides only was less effective than a U-wrap.
• The recorded CFRP strain of the tested beams indicated that the failure of CFRP system occurs at an
average effective stress level below nominal strength due to stress concentrations or debonding of CFRP
from concrete surface.

ACKNOWLEDGEMENTS

This work was conducted with partial support from National Science Foundation (NSF) and Repair of Buildings and
Bridges with Composites (RB2C) based at the University of Missouri-Rolla. The Egyptian Cultural and Educational
Bureau provided support to the first author.

REFERENCES

ACI Committee 440, State-of-the-Art Report on Fiber Reinforced Plastic (FRP) Reinforcement for Concrete
Structures, American Concrete Institute, Detroit, Michigan, 1996, 68 pp.

Chajes, M. J., Januska, T.F., Mertz, D.R., Thomson, T.A., and Finch, W.W., “Shear Strengthening of Reinforced
Concrete Beams Using Externally Applied Composite Fabrics,” ACI Structural Journal, Vol. 92, No. 3,
May - June 1995, pp. 295-303.
Khalifa, A., Shear Performance of Reinforced Concrete Beams Strengthened with Composites, Ph.D. Thesis,
Structural Engineering Department, Alexandria University, Egypt, 1999.
Khalifa, A., Alkhrdaji, T., Nanni, A., and Lansburg, A., “Anchorage of Surface Mounted FRP Reinforcement”.
Concrete International, American Concrete Institute, 1999 a, to appear.
Taerwe, L., Khalil, H., and Matthys, S., “Behavior of RC Beams Strengthened in Shear by External CFRP
Sheets,” Non-Metallic(FRP) Reinforcement for Concrete Structures, Proceedings of the Third Symposium,
Vol. 1, Japan, Oct. 1997, pp. 483-490.
Triantafillou, T.C., “Shear Strengthening of Reinforced Concrete Beams Using Epoxy Bonded FRP
Composites,” ACI Structural Journal, Mar.-Apr. 1998, pp. 107-115.

8 0350