Construction and Building Materials 66 (2014) 634–644

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Strengthening of shear critical RC beams with various FRP systems

Daniel Baggio 1, Khaled Soudki 2, Martin Noël ⇑
Department of Civil and Environmental Engineering, University of Waterloo, 200 University Ave. West, Waterloo, ON, Canada N2L 3G1

h i g h l i g h t s

 Shear strengthening methods including CFRP wraps, GFRP wraps and FRCM wraps are compared.
 Effectiveness of commercially produced FRP anchors to prevent debonding are evaluated.
 Guidelines for installation procedure for FRP anchors are provided.
 Experimental results are compared to analytical predictions using current code equations.

a r t i c l e i n f o a b s t r a c t

Article history: The use of fiber reinforced polymer (FRP) sheets to shear strengthen reinforced concrete (RC) beams has
Received 18 January 2014 been studied and proven repeatedly. However, numerous strengthening configurations and materials can
Received in revised form 28 May 2014 be combined to maximize the increase in strength and repair. This study investigates the effectiveness of
Accepted 30 May 2014
using commercially manufactured carbon FRP (CFRP), glass FRP (GFRP) and fiber reinforced cementitious
Available online 28 June 2014
matrix (FRCM) sheets to increase the shear capacity of RC shear critical beams. Nine RC shear deficient
slender beams were tested. The test variables included the use of three different FRPs (CFRP, GFRP,
Keywords:
and FRCM), and the presence and type of FRP anchors (CFRP or GFRP). The experimental results revealed
Concrete
Repair
that applying FRP sheets increased the overall shear capacity, full depth u-wrapped FRP sheets perform
Strengthening better over companion partial depth u-wrapped FRP sheets, the use of FRP anchors further improved the
FRP shear capacity and ductility of failure, FRP strengthening could change the mode of failure from a shear to
Composite flexural failure and FRP debonding was delayed with the presence of FRP anchors. A comparison of the
Carbon fiber experimental load capacities with predicted values indicates that they correlate well with predictions
Glass fiber using the current Canadian design codes.
Fiber reinforced cementitious matrix Ó 2014 Elsevier Ltd. All rights reserved.
Shear

1. Introduction The recent advancement of fiber-reinforced polymer (FRP) com-

Many reinforced concrete bridges and structures around the
world are currently in need of repair or complete replacement as
they approach the end of their service life. Increases in traffic vol-
ume, traffic loads, and corrosion-induced deterioration are necessi-
tating significant expenditures to strengthen and rehabilitate
existing structures [1]. Of the 163,000 single span concrete bridges
in the United States, 23% are considered structurally deficient or
functionally obsolete [2].
⇑ Corresponding author. Present address: Queen’s University, 58 University Ave.,
Kingston, ON K7L 3N6, Canada. Tel.: +1 519 588 3124.
E-mail addresses: dbaggio11@hotmail.com (D. Baggio), martin.noel@queensu.ca
(M. Noël).
1
Present address: Transportation & Bridge Engineering MMM Group, Mississauga,
ON L5K 2P8, Canada.
2
Address: University of Waterloo, 200 University Ave., Waterloo, ON N2L 3G1,
Canada.
posites as a repair and strengthening material for reinforced con-
crete (RC) beams, slabs and columns in structural engineering
applications has increased over the past 20 years [3–9]. The non-
corrosive properties and high strength-to-weight ratio of FRP
materials make them a very viable repair material and can result
in longer service life of structures [1]. The advancement of new
materials and technologies has led researchers to investigate vari-
ous designs and materials to increase the strength of shear critical
RC beams [10–14]. Various materials, configurations, wrapping
techniques and mechanical anchors have been explored to increase
the capacity of existing RC members and postpone or delay the
debonding process in externally bonded FRP members [15–24].
FRP anchors were included in this study to further increase the
shear capacity of FRP and FRCM configurations. FRP anchors are
of particular interest because they have the same material proper-
ties as the FRP sheets and can be installed simultaneously with the
sheets using the same adhesives [25,25].

http://dx.doi.org/10.1016/j.conbuildmat.2014.05.097
0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644 635

This research study investigates the effectiveness of using vari- 2.3. FRP system
ous configurations and types of commercially manufactured FRP
The FRP strengthening system consisted of FRP sheets or FRCM grid. The FRP
sheets and FRCM grid to strengthen shear critical RC beams. This system was installed as intermittent u-wraps around the cross-section along the
objective of this research project aims to determine the structural beam length with orientation of fibers in the transverse direction. The wraps were
behavior of shear-critical beams strengthened with externally 100 mm or 200 mm wide, spaced at 200 mm or 275 mm center to center and
bonded FRP sheets and FRP anchors used to delay or eliminate extended the full-depth or partial depth (50 mm below the top surface) of the
cross-section. FRP anchors were used in half of the strengthened beams with one
FRP debonding. To the best of the authors’ knowledge, this study
FRP anchor on each u-wrap for a total of eight anchors per beam. Fig. 2 illustrates
is one of only a few studies which have used commercially manu- the FRP and grid strengthening schemes.
factured FRP anchors as the anchoring mechanism to secure exter- Three different FRP strengthening materials (CFRP, GFRP and FRCM) and two
nally bonded FRP sheets [24,26]. types of FRP anchors (CFRP and GFRP) provided by SikaÒ Canada were applied to
strengthen the beam. The mechanical properties of the FRP materials are given in
Table 2.
2. Experimental program
2.3.1. CFRP sheets
2.1. Test matrix Two beams were strengthened with SikawrapÒ Hex-230C CFRP sheets (Fig. 3a)
with the FRP sheets applied transversely as u-wraps over the full depth of the beam.
A total of nine shear deficient reinforced concrete slender beams were con- The FRP sheets were cut to a length of 850 mm and to a width of 200 mm. Testing
structed and tested. Table 1 outlines the test matrix. The program comprised of the two CFRP repaired beams revealed that the shear strengthening provided by
one control (unstrengthened) beam and eight FRP strengthened beams. Two beams 200 mm wide CFRP sheets spaced every 275 mm (center to center) exceeded the
were strengthened with CFRP sheets with and without anchors, two beams were expected design shear strength and ultimately exhibited a flexural failure mode.
strengthened with FRCM grids with and without anchors and four beams were To avoid flexural failure in the remaining beams, the beams strengthened with
either fully or partially wrapped with GFRP sheets with and without anchors. GFRP sheets used a reduced sheet width of 100 mm at a spacing of 200 mm (center
The test variables include: the type of FRP (CFRP, GFRP and FRCM), the presence to center) to ensure a shear failure mode after strengthening.
and type of FRP anchor (CFRP or GFRP), and u-wrapping schemes (full depth vs. par-
tial depth).
2.3.2. FRCM
Two beams were strengthened using u-wrapped fiber reinforced cement matrix
2.2. Test specimens (FRCM) sheets transversely applied over the full depth of the beam. The FRCM used
was SikawrapÒ 350G Grid. Fig. 3b shows a beam strengthened with u-wrap FRCM.
The beams measured 150 mm wide by 350 mm deep by 2440 mm long. The The FRCM grid was cut to a length of 850 mm and a width of 200 mm. SikaÒ
bottom flexural steel reinforcement consisted of two 30 M bundled bars. The top MonoTopÒ 623 was applied to bond the FRCM to the concrete surface. The mortar
compression reinforcement consisted of two 15 M bars. The shear reinforcement used is a one component polymer modified overhead and vertical repair and retro-
consisted of 6 mm smooth bars spaced at 180 mm center to center. The bottom con- fitting mortar with integral corrosion inhibitor. The mortar was applied to the con-
crete cover was 40 mm while the top and side covers were 30 mm. The beam was crete substrate as a 3 mm thick scrub coat. The FRCM grid was then applied on top
designed according to the Canadian design code CSA A23.3-04. A schematic of the of the scrub coat and pressed into the mortar. Once the grid was applied a second
specimen geometry and reinforcement details are shown in Fig. 1. lift (layer) of the mortar was placed and finished covering the FRCM grid.
The concrete used to construct the beams was batched with Portland cement
with a maximum coarse aggregate size of 19 mm. The average compressive 2.3.3. GFRP sheets
strength of the concrete obtained after 28-days was 50.7 ± 1 MPa and the average Four beams were strengthened using SikawrapÒ Hex-430G GFRP sheets
strength on day of testing was 50.1 ± 1 MPa. The longitudinal reinforcements con- u-wrapped (100 mm wide at a spacing of 200 mm o/c). Two beams had no anchors
sisted of Grade 400 reinforcing steel bars with a yield strength of 475 MPa and and two beams had FRP anchors (either CFRP or GFRP). Three of the four beams had
the stirrups had a yield strength of 384 MPa as reported by the manufacturer. GFRP sheets installed to a partial depth; 50 mm below the top of the beam to
Table 1 initiate debonding of the GFRP sheet. Fig. 3c shows a partial depth beam strength-
Experimental test matrix. ened with GFRP u-wraps. The GFRP sheets were cut to 100 mm width and 750 mm
length.
Specimen description Strengthening method Anchors For all beams the locations of the FRP sheet on the beam cross-section were
marked and the beams were prepared for installation of the FRP sheets as per the
Beam 1 – Control None No
manufacturer’s instructions and ACI 440.2R-08 requirements.
Beam 2 – CFRP – No anchors 230C CFRP sheets No
Beam 3 – CFRP – Anchors Yes
2.4. FRP anchor installation
Beam 4 – FRCM – No anchors Grid system: No
Beam 5 – FRCM – Anchors Fiber reinforced cement Yes To the best of the authors’ knowledge, currently no standard FRP anchor instal-
matrix (FRCM) lation procedure exists. The FRP anchor installation procedure used in this study
consisted of the following steps:
Beam 6 – GFRP – No anchors 430G GFRP sheets No
Beam 7 – PD-GFRP – No anchors No  The anchor holes were located and drilled at mid-width in the FRP sheet. The
Beam 8 – PD-GFRP – C anchors 430G GFRP sheets (partial depth) Yes vertical location of the holes varied: 55 mm from the top of the beam for the
Beam 9 – PD-GFRP – G anchors Yes sheets which extended the full depth of the beam and 90 mm from the top of
the beam for the beams with sheets that extended to partial depth.
Total 9 Beams

30
2-15M

350 30

40 2-30M
900 mm 400 mm 900 mm

150
2440 mm

Fig. 1. Beam geometry and steel reinforcement details (units in mm).

636 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644

(a) Beam with full depth FRP U-wrap

(b) – Beam with paral depth FRP U-wrap

Fig. 2. FRP repair scheme.

located in the shear region. To measure the strain in the concrete a 60 mm strain
Table 2
gauge was installed on the top surface between the two loading points. Strain
FRP and FRCM material properties.
gauges were also applied externally to the FRP sheets at multiple depths. The layout
Material Thickness (mm) Elastic modulus Elongation at of the strain gauges on the internal steel and concrete is shown in Fig. 6.
(GPa) rupture (%) One linear variable differential transducer (LVDT) with a 25 mm range was used
to measure the mid-span deflection of the beam.
CFRP – 230C 0.381 67 1.33
GFRP – 430G 0.508 26 2.21
FRCM – 350G 1.17 75 2.80 2.6. Test setup and procedure
CFRP – Anchor C 10 (diameter) 215 0.74
GFRP – Anchor G 10 (diameter) 70 >4.0 All beams were tested in four point bending using a closed-loop hydraulic MTS
actuator with a 500 kN capacity in a MTS 322 test frame. The beams were simply
supported with a clear span of 2200 mm with 400 mm spacing between the two
 Once the holes were drilled the FRP sheets were installed and finishing nails were loading points and a shear span of 900 mm. The supports and loading points con-
inserted through the weave of the FRP to mark the location of each pre-drilled sisted of a pin and roller connection. The test setup is shown Fig. 7.
hole. The FRP sheets were splayed so the predrilled hole opening was visible. The load was applied in displacement control at a rate of 0.3 mm/min. A
 After 24 h, the nails were removed and the holes were re-opened by drilling National Instruments data acquisition system recorded all readings from the instru-
through the gap in the hardened FRP sheet. mentation (strain gauges and LVDTs). Cracks and their development were recorded
 FRP anchors were then installed and fanned out to secure the u-wrapped FRP and monitored throughout the test.
sheets.

A diagram of the FRP anchor installation process is shown in Fig. 4 and a sche- 3. Test results and discussion
matic showing the FRP anchor locations is shown in Fig. 5.

3.1. Observed behavior

2.5. Instrumentation

Electrical resistance strain gauges were used to measure the strain in the steel
Three modes of failure were observed: shear failure, shear fail-
reinforcement. Three 5 mm gauges were used internally in each beam, one was ure with debonding of the FRP sheet and flexural failure with
attached to the longitudinal rebar and two gauges were installed on two stirrups crushing of the concrete (Fig. 8). Debonding and peeling of the
 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644 637

(a) CFRP - Full depth

(b) FRCM - Full depth

(c) GFRP – Paral depth

Fig. 3. u-Wrap installation.

FRP occurred in four beams (Beam 6 – GFRP – No anchors, Beam 7 –
PD-GFRP – No anchors, Beam 8 – PD-CFRP anchors, Beam 9 – PD-
GFRP anchors) of the eight strengthened beams. However, for both
beams with FRP anchors (Beams 8 and 9), the debonding was
halted by the presence of the anchors and the beams ultimately
failed by shear compression (concrete crushing at the shear crack
tip). A summary of the test results, ultimate load, deflection at ulti-
mate load, percent increase over the control and mode of failure for
all beams is provided in Table 3. Fig. 9 shows the load vs. deflection
curves for all beams.
3.1.1. Control beam
The control beam failed in shear by diagonal tension as
expected as seen in Fig. 8a. The beam reached an ultimate load
of 223 kN with a corresponding midspan deflection of 8.80 mm
(Table 3).
3.1.2. CFRP strengthened beams
Two beams were strengthened with CFRP sheets (200 mm wide
at 275 mm spacing) with and without CFRP anchors (Fig. 9a). The
failure mode observed for both beams was flexural failure with
crushing of the concrete. Flexural cracks were initiated at the beam
soffit at the loading points. As the load was increased the flexural
cracks began to propagate vertically toward the top of the beam.
Simultaneously, hairline shear cracks began to appear in the shear
span between the CFRP sheets. These cracks did not widen due to
the presence of the CFRP u-wraps. As the flexural cracks propa-
gated, the longitudinal tension steel bars and the compression steel
bars yielded, evidenced both by a change in stiffness and strain
gauge readings. Complete failure of the beam occurred with crush-
ing of the concrete at the top of the beam. Cracking and popping of
the CFRP sheets and epoxy was heard throughout loading up to
failure. Debonding of the CFRP sheets did not occur for either of
the CFRP-strengthened beams.
The increase in strength for Beam 2 – CFRP – No anchors and
Beam 3 – CFRP–CFRP anchors was 67% and 75%, respectively, over
the control beam. The benefits of using CFRP anchors in the second
beam could not be fully assessed because both beams failed in flex-
ure. The difference between the anchored and non-anchored
beams was an 8% increase in failure load and is not large enough
to be attributed to the presence of anchors.
These results are similar to the results by Jayaprakash et al.
which reported that increasing the amount of internal and external
FRP shear reinforcement may not proportionally increase the shear
capacity but can change the mode of failure from shear to flexure
[28].
638 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644

(a) FRP anchor detail

(b) FRP anchor impregnaon (c) FRP anchor installaon rod (d) FRP anchor fan detail
Fig. 4. FRP anchor installation.

50mm 50mm 100mm

90mm 55mm 55mm

75mm 75mm 75mm

fan fan fan

Wfrp=100 mm Wfrp=100 mm Wfrp=200 mm

Fig. 5. FRP anchor scheme.

Fig. 6. Strain gauge layout details.

 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644 639

Fig. 7. Test setup.

Fig. 8. Failure modes: Top to bottom: shear, debonding and flexural failure.
640 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644

Table 3
Summary of test results.

Beam description Ultimate load (kN) Deflection at ultimate load (mm) Percent increase over control (%) Failure mode
Beam 1 – Control 223 8.80 – Shear
Beam 2 – CFRP – No anchors 373 11.4 67.5 Flexure
Beam 3 – CFRP – Anchors 390 16.9 75.1 Flexure
Beam 4 – FRCM – No anchors 294 12.0 32.0 Shear
Beam 5 – FRCM – Anchors 300 10.7 34.7 Shear
Beam 6 – GFRP – No anchors 334 13.7 50.1 Shear – debonding
Beam 7 – PD-GFRP – No anchors 305 12.0 36.8 Shear – debonding
Beam 8 – PD-GFRP – C anchors 310 14.2 39.2 Shear compression
Beam 9 – PD-GFRP – G anchors 339 13.7 52.2 Shear compression

3.1.3. FRCM strengthened beams
Two beams were repaired with FRCM sheets (200 mm wide at a
spacing of 275 mm c/c) with and without CFRP anchors. The mode
of failure for both beams was diagonal tension shear failure
(Fig. 9b). As the load increased, diagonal tension shear cracks in
the shear span began to develop, which propagated and widened
until they reached the support and loading point. It was observed
that in the case of the FRCM strengthened beam with CFRP
anchors, the shear crack seemed to take the path of least resistance
travelling either above the anchor or below the fan.
The increase in strength for Beam 4 – FRCM – No anchors and
Beam 5 – FRCM–CFRP anchors was 32% and 34% over the control
beam (Table 3). The increase in strength provided by the CFRP
anchors in Beam 5 was only 3% over Beam 4. The relatively small
increase in strength compared to beams with full-depth FRP sheets
is attributed to the properties of the grid itself, the lower strength
of the cementitious mortar compared to epoxy, as well as the
reduced bond strength with the fiber grid.
3.1.4. GFRP strengthened beams
Four beams were strengthened with GFRP sheets (100 mm wide
at a spacing of 200 mm c/c). One beam had GFRP sheets running
the full depth of the beam and three beams had GFRP sheets with
a partial depth installation (50 mm below the top of the beam).
3.1.4.1. Beam with full depth GFRP sheets. The failure mode of the
beam with full depth sheets and no anchors (Beam 6) was debond-
ing of the GFRP sheet and simultaneous shear diagonal tension

Fig. 9. Load deflection curves of beams strengthened with FRP and FRCM sheets with and without anchors.
 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644
failure. Diagonal shear cracks began to appear in the mid-region of
the shear span between the support and loading point. Cracking
and popping was heard from the GFRP sheets as the load increased.
Past a load of 280 kN (83% of the ultimate load), debonding at the
end of the GFRP sheets from the concrete surface began to occur
(Fig. 8b). Subsequently, propagation and widening of the diagonal
shear cracks occurred until the shear cracks reached the loading
point and the support and failure occurred at a load of 334 kN
(Fig. 9c). The increase in strength for Beam 6 – GFRP – No anchors
was 50% over the control beam. This was 17% less than the strength
increase provided by CFRP sheets (Beam 2 – CFRP – No anchors)
and can be attributed to the differences in material properties
and amount of strengthening provided.
3.1.4.2. Beams with partial depth GFRP sheets. The failure mode
observed for the beam strengthened with partial depth GFRP
sheets without anchors (Beam 7) was debonding of the GFRP sheet
and diagonal tension shear failure. Diagonal shear cracks began to
appear in the mid-region of the shear span between the support
and loading point on both sides of the beam. As the load increased
cracking and popping could be heard from the GFRP sheets. When
the load reached approximately 260 kN (85% of the ultimate load),
debonding at the top of the sheets began. Simultaneously, propaga-
tion and widening of the diagonal shear cracks occurred, and as the
loading increased the diagonal shear crack reached the loading
point and failure occurred with a sudden drop in the load (Fig. 9d).
The failure mode observed for the two beams repaired with par-
tial depth GFRP sheets and FRP anchors was compression failure by
crushing of the concrete near the loading point. As the load was
increased, diagonal shear cracks began to develop in the shear span
and propagated towards the support and loading point taking the
path of least resistance. The trajectory of the shear crack was close
to the loading point. The crack propagated around the anchors of
two separate sheets towards the top 50 mm section of the beam
bypassing the GFRP at the loading point (Fig. 10).
At a load of approximately 255 kN (82% of the ultimate load),
the GFRP sheets began to debond starting from the top of the sheet
above the FRP anchor. However, the debonding process was halted
due to the presence of the FRP anchors. The beams failed when the
crack reached the loading point and crushing of the concrete in the
compression zone at the loading point occurred. Anchors in the
two beams prevented complete debonding of the FRP sheet from
concrete.
Diagonal tension shear crack
from support to loading point
Fig. 10. Diagonal shear crack pattern in beams with partial depth GFRP sheets and FRP anchors.
641
In comparison to the control beam, the four GFRP strengthened
beams (Beam 6 – GFRP – No anchors, Beam 7 – PD-GFRP – No
anchors, Beam 8 – PD-CFRP anchors, Beam 9 – PD-GFRP anchors)
had a 50%, 36%, 39% and 52% increase in capacity, respectively.
The strength of Beam 6 – GFRP – No anchors (with full depth
installation) was 13.3% higher than Beam 7 – PD-GFRP – No anchors
(with partial depth). Thus the full depth configuration which pro-
vided an additional 50 mm (14%) of bonded length increased the
shear capacity proportionally. This finding confirms the findings of
other researchers that providing additional bond length or securing
the sheet will increase the shear capacity [29]. In cases where the
bonded length is limited (partial depth installation), providing
anchors is a viable option to achieve the same capacity as a design
with full depth installation. This is evidenced by comparing the
results of Beam 6 (full depth) to Beam 9 (partial depth with FRP
anchors) which exhibited similar behavior and load capacities. It
should be noted that adding FRP anchors does not necessarily
increase the shear capacity of a member. The effectiveness of FRP
anchors is dependent on many factors including the amount and
type of the strengthening material that has been applied.
3.2. Strain behavior
The strains in the tension steel bars for the beams that failed in
shear were less than the yield strain of 2400 le except for beam
PD-GFRP-NA which exhibited a maximum strain of 2500 le. For
the two CFRP strengthened beams that failed in flexure, the strain
in the tension steel reinforcement exceeded the yield strain (Fig. 11a).
Two beams exhibited concrete strains which were close to the
design crushing strain value of 0.0035, namely Beams 2 and 3
which were strengthened with CFRP sheets. The strains in the con-
crete of the remaining six beams were all well below the concrete
crushing strain (Fig. 11b).
Six beams had strain readings greater than 2000 le in the second
stirrup suggesting that the stirrup has yielded (Fig. 11c). These
beams were: the control beam, beams with FRCM sheets with and
without anchors, beams with CFRP sheets with and without anchors
and beams with partial depth GFRP sheets without anchors. Initially
the shear strain was resisted by the concrete but once a diagonal
crack formed, the strain was transferred to the stirrup. With FRP
strengthening, the specimen was able to carry more load as part of
the force was carried by the FRP sheet/grid. The variation between
strain readings in the beams is also affected by the location of the
gauge with respect to the shear crack.
FRP Anchor
642 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644
Measured strains in the external FRP sheets are also highly influ-
enced by the proximity of the strain gauges to the nearest diagonal
crack. The highest failure load and lowest FRP strain was observed
for the GFRP strengthened beam with GFRP anchors (Fig. 11d). How-
ever, higher FRP strains were expected for Beams 8 and 9 which con-
tain FRP anchors since the presence of FRP anchors prevents
debonding from occurring which in turn allows the FRP sheets to
continue to contribute to load resistance. The higher strain readings
recorded for Beam 7 can be attributed to the location of the FRP
strain gauge with respect to the diagonal tension shear crack since
the crack crossed directly under the location of the FRP strain gauge
causing a very high localized reading; a similar effect occurred with
Beams 2 and 3. It is recommended that future tests include addi-
tional strain gauges bonded to multiple locations on the FRP sheets
to capture the local strain effects induced by diagonal cracking.
3.3. Effect of shear strengthening
Fig. 12 shows the effect of strengthening on the ultimate load
and maximum deflection. CFRP strengthening resulted in a 67%
Fig. 11. Maximum strain at ultimate load.
increase in strength and a 30% increase in maximum deflection
compared with the control beam. This can be attributed to the high
strength and stiffness of the CFRP sheets compared with the GFRP
and FRCM systems. The GFRP strengthened beam with 100 mm
wide sheets resulted in a 50% increase in strength and a 55%
increase in maximum deflection while the GFRP partial depth
strengthened beam displayed a 36% increase in strength and a
41% increase in maximum deflection (Fig. 12). The FRCM strength-
ened beam exhibited a 32% increase in strength and a 36% increase
in maximum deflection. The use of anchors had only a small effect
on the behavior of the full-depth strengthened beams, although
they were effective in preventing debonding of the FRP sheets for
beams with a partial-depth wrapping scheme.
4. Prediction of failure load
The shear capacity (V) of the tested beams were predicted
using equations given in the Canadian concrete design code
CSA A23.3-04 [30] and ISIS-M04 design manual [31]. In the
Canadian concrete design code the shear capacity is calculated by
 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644 643

adding the shear contributions from concrete (Vc), steel (Vs) and Table 4
FRP (Vfrp) which are shown in Eqs. (1)–(3). Prediction vs. experimental failure load.

qffiffiffiffi Specimen Ultimate Predicted load Pexp/Ppred

V c ¼ uc ab fc0 bw dv ð1Þ load P (kN) P (kN)
Control 223 214 1.03
CFRP – No anchors 373 Shear = 430 1.02
us Av fy dv coth Flexure (P) = 365
Vs ¼ ð2Þ
s CFRP with CFRP anchors 390 Shear = 430 1.06
Flexure (P) = 365
FRCM – No anchors 294 –b –b
ufrp Efrp efrp Afrp dfrp ðcoth þ cotBÞsinb FRCM with CFRP anchors 300 –b –b
V frp ¼ ð3Þ
sfrp GFRP – No anchors 334 316 1.05
PD-GFRP – No anchors 305 296 1.03
The shear contributions from concrete and steel (Vc + Vs) were PD-GFRP with CFRP anchors 310 386a .803
calculated based on the simplified method in CSA A23.3-04 and PD-GFRP with GFRP anchors 339 386a .878
(Vfrp) was predicted using equations in the ISIS-M04 manual. The a
Assuming that FRP anchors will cause FRP sheets to rupture.
flexural capacity (M) of the strengthened beams was predicted b
FRCM material data not available to calculate shear contribution.
using a strain compatibility approach. The predicted load (P) was
calculated as the smaller of the load based on shear capacity
(P = 2V) or flexural capacity (P = 2  (M/a) where a = shear span) of currently available code equations to predict the shear strength
the beam. contribution of FRCM.
The predicted failure loads vs. the measured failure loads are The first two beams tested (Beams 2 and 3) were both strength-
given in Table 4. It is clear that the predictions for five beams (con- ened with CFRP sheets. Initial predictions based on the ISIS design
trol beam, three FRP strengthened beams without anchors, one manual of these two CFRP strengthened beams did not suggest that
CFRP strengthened beam with anchors) correlate very well with the shear capacity would be larger than the flexure capacity
measured values with an average ratio of 1.03. No predictions were because the effective FRP strain (efrpe) using the manufacturer’s
made for the two beams strengthened with FRCM as there are no data was calculated incorrectly. All subsequent designs and tests
were conducted to ensure shear failure occurred and the correct
effective FRP strain was used to predict the ultimate load at failure.
The predictions for the two partial depth GFRP strengthened
beams with CFRP and GFRP anchors were un-conservative. The
predictions were higher than the experimental values because it
was assumed that the GFRP sheet would rupture with the presence
of anchors. This failure mode did not occur during testing as the
GFRP sheets experienced partial debonding. The predicted load
assuming FRP rupture is 386 kN while the predicted load assuming
debonding of FRP is 296 kN. The GFRP strengthened beams with
CFRP and GFRP anchors failed at 310 kN and 339 kN, which are
19% and 12% lower than the predicted load if FRP rupture were
to occur. Based on these preliminary results with commercially
manufactured FRP anchors it seems likely that achieving FRP rup-
ture is not practically achievable for many applications. It is recom-
mended that further testing be conducted such that an
amplification factor for the effective FRP strain for anchored sheets
can be developed so that one can successfully predict the contribu-
(a) Ulmate load tion of the FRP sheets to shear capacity. In addition, further
research and testing should be done on beams with similar
expected failure loads which have been strengthened with CFRP
and GFRP to obtain an accurate comparison between carbon and
glass FRP strengthening.

5. Conclusions

The shear performance of eight reinforced concrete beams

strengthened with FRP & FRCM sheets and FRP anchors and one
control specimen were investigated. The current exploratory study
highlighted the effectiveness of various strengthening schemes and
the use of FRP anchors while highlighting areas for further
research. Conclusions drawn from the test results are as follows:

 All beams exhibited a similar flexural stiffness during testing

regardless of the strengthening material, properties and layout
(depth or width).
 Beams strengthened with FRCM showed a 31% and 34% increase
(b) Ulmate deflecon in shear capacity with and without anchors over the control
specimen, respectively, and did not result in a change in the
Fig. 12. Strengthening material performance comparison. mode of failure.
644 D. Baggio et al. / Construction and Building Materials 66 (2014) 634–644
 Strengthening using full depth GFRP sheets and no anchors
resulted in a 50% increase in shear capacity over the control
with debonding occurring before diagonal tension shear failure.
 Beams strengthened with partial depth GFRP sheets with and
without anchors showed a 52% and 36% increase in shear capac-
ity, respectively, over the control beam and the presence of the
FRP anchors was effective in halting the debonding of the GFRP
sheets.
 Beams strengthened with CFRP sheets with and without
anchors showed a 67% and 75% increase in shear capacity over
the control beam, respectively, with both beams exhibiting a
ductile flexural failure mode.
 CFRP strengthening (with 200 mm wide sheets spaced every
275 mm) changed the mode of failure from a brittle shear fail-
ure to a flexure failure by yielding of the longitudinal steel
rebar.
 The use of GFRP anchors provided an additional 13% increase in
strength over the control compared to the equivalent GFRP-
strengthened beam with CFRP anchors.
 Although the presence of FRP anchors halted the debonding
process, the anchored FRP sheets did not reach their ultimate
rupture strain.
 Predictions using equations from CSA A23.3-04 and ISIS-M04
design codes correlated well with measured strength values
with the exception of two beams with partial depth FRP sheets
and FRP anchors. This result is due to the strengthened beams
failing prior to reaching the ultimate rupture strain of the FRP
sheets. Further research is recommended to determine an
appropriate effective strain value for anchored FRP sheets. Fur-
ther research is also needed to quantify the contribution of
FRCM grids to shear capacity.
 When the available bonded length is limited, the installation of
FRP anchors is a viable option to prevent a brittle shear failure
mode due to FRP debonding.
Acknowledgement
Research funding from the Natural Sciences and Engineering
Research Council of Canada is gratefully acknowledged.
References
[1] Noel M, Soudki K. Evaluation of FRP posttensioned slab bridge strips using
AASHTO-LRFD bridge design specifications; 2011.
[2] Mabsout M, Tarhini K, Jabakhanji R, Awwad E. Wheel load distribution in
simply supported concrete slab bridges. J Bridge Eng 2004;9(2).
[3] Cao SY, Chen JF, Teng JG, Hao Z, Chen J. Debonding in RC beams shear
strengthened with complete FRP wraps. J Comp Constr 2005;9(5).
[4] Triantafillou TC, Antonopoulos CP. Design of concrete flexural members
strengthened in shear with FRP. J Compos Constr 2000:198–205.
[5] ACI. Report on fiber-reinforced polymer (FRP) reinforcement for concrete
structures. Farmington Hills (MI): American Concrete Institute (ACI); 2007.
[6] Bank LC. Composites in construction: structural design with FRP
materials. Hoboken (NJ): John Wiley & Sons; 2006.
[7] Khalifa A, Gold W, Nanni A, Abdel Aziz M. Contribution of externally bonded
FRP to shear capacity of flexural members. ASCE, J Compos Constr
1998;2(4):195–203.
[8] Chaallal O, Nollet M, Perraton D. Shear strengthening of RC beams by
externally bonded side CFRP strips. J Compos Constr 1998;2(2):111–3.
[9] Triantafillou TC. Shear strengthening of reinforced concrete beams using
epoxy-bonded FRP composites. ACI Struct J 1998;95(2):107–15.
[10] Taljsten B, Elfgren L. Strengthening of concrete beams for shear using CFRP-
materials: evaluation of different application methods. Compos Part B: Eng
2000;32(2):87–96.
[11] Sim J, Park C, Moon Dy. Characteristics of basalt fiber as a strengthening
material for concrete structures. Compos Part B: Eng 2005;36(6–7):504–12.
[12] Ascione L, Feo L. Modeling of composite/concrete interface of RC beams
strengthened with composite laminates. Compos Part B: Eng 2000;31(6–
7):535–40.
[13] Yao J, Teng JG, Chen JF. Experimental study on FRP-to-concrete bonded joints.
Compos Part B: Eng 2005;36(2):99–113.
[14] Garden HN, Hollaway LC. An experimental study of the failure modes of RC
beams strengthened with prestressed carbon composite plates. Compos Part
B: Eng 1998;29(4):411–24.
[15] Bousselham A, Chaallal O. Mechanisms of shear resistance of concrete beams
strengthened in shear with externally bonded FRP. J Comp Constr 2008;12(5).
[16] Chaallal O, Shahawy M, Hassan M. Performance of reinforced concrete T-
girders strengthened in shear with carbon fiber-reinforced polymer fabric. ACI
Struct J 2002;99(3):335–43.
[17] Chen GM, Teng JG, Chen JF. Shear strength model for FRP-strengthened RC
beams with adverse FRP-steel interaction. J Compos Constr 2012;17(1):50–66.
[18] Chen GM, Teng JG, Chen JF, Rosenboom OA. Interaction between steel stirrups
and shear-strengthening FRP strips in RC beams. J Compos Constr
2010;14(5):498–509.
[19] Chen JF, Teng JG. Shear capacity of FRP-strengthened RC beams: FRP
debonding. Constr Building Mat 2003;17:27–41.
[20] Khalifa A, Nanni A. Improving shear capacity of existing RC T-section beams
using CFRP composites. J Cem Concr Compos 2000;22:164–74.
[21] Quinn KT. Shear strength of reinforced concrete beams with carbon fiber
reinforced polymer (CFRP) and improved anchor details. Texas
(Austin): University of Texas at Austin; 2009.
[22] Kim SJ, Smith ST. Shear strength and behaviour of FRP spike anchors in cracked
concrete. Sydney, Australia: ICE Australia; 2009.
[23] Orton SL, Jirsa JO, Bayrak O. Design considerations of carbon fiber anchors. J
Comp Constr 2008;12(6):608–16.
[24] Smith ST, Kim SJ. Shear strength and behaviour of FRP spike anchors in FRP-to-
concrete joint assemblies. Winnipeg, Canada: Multi Science Publ Co Ltd; 2008.
[25] Kim SJ, Smith ST. Behaviour of handmade FRP anchors under tensile load in
un-cracked concrete. Advances Struct Eng 2009;12(6).
[26] Teng JG, Lam L, Chen JF. Shear strengthening of RC beams using FRP
composites. Progress Struct Eng Mat 2004(6).
[27] Smith S, Zhang H. Anchorage of FRP-to-concrete bonds with dry and
impregnated FRP anchors. In: Proceedings of the 6th international
conference on advanced composite materials in bridges and structures
(ACMBS-VI 2012), vol. VI; 2012.
[28] Jayaprakash J, Abdul Samad AA, Ashrabov AA, Choong KK. Experimental
investigation on shear resistance behaviour of RC precracked and non-
precracked T-beams using discrete CFRP strips. Int J Integrated Eng
2009;1(2):1–15.
[29] Deniaud C, Roger Cheng JJ. Reinforced concrete T-beams strengthened in shear
with fiber reinforced polymer sheets. J Compos Constr 2003;7(4):302–10.
[30] Cement Association of Canada. Concrete design handbook. 3rd
ed. Ottawa: CAC; 2008.
[31] ISIS-M04. FRP rehabilitation of reinforced concrete structures design manual
4. Winnipeg (Canada): ISIS; 2009.