Reinforced Concrete Beams Strengthened with NSM

CFRP Rods in Shear

Firas Al-Mahmoud1, Arnaud Castel2,*, Trinh Quang Minh3 and Raoul François3
1Institut
Jean Lamour, UMR 7198, CNRS, Université de Lorraine, Nancy, France
2Centre
for Infrastructure Engineering and Safety (CIES), The UNSW, Sydney 2052, Australia
3Université de Toulouse UPS, INSA; L.M.D.C (Laboratoire Matériaux et Durabilité des Constructions), 31077 Toulouse, France

(Received: 29 January 2014; Received revised form: 18 March 2015; Accepted: 19 April 2015)

Abstract: This experimental research aims to investigate the possibility of using

Carbon Fiber-Reinforced Polymer (CFRP) rods to strengthen in shear concrete
structural members with the Near Surface Mounted reinforcement (NSM) technique.
Seven reinforced concrete beams were tested in flexure with or without traditional
shear reinforcement. Carbon-epoxy pultruded FRP (CFRP) rods with 6 mm in
diameter were used. The possibility of using a mortar as filling material with surface
pre-conditioned CFRP rod is investigated and compared to resin performance. Results
show that NSM technique is very efficient to strengthen reinforced concrete beam in
shear. Using the resin instead of mortar as the filling material still increases the
efficiency of the shear strengthening but the gain is only moderate. The experimental
results are compared with existing analytical models.

Key words: reinforced concrete, CFRP rods, near-surface mounted, beams, shear strengthening.

1. INTRODUCTION
Since the 1980s – 90s, a strengthening technique using FRP reinforcement, the NSM system has a number of
FRP laminate has been developed. FRP materials have advantages, which are: (1) NSM bars are protected by
better mechanical, physical and chemical properties the concrete cover and so are less exposed to accidental
than steel. The applications of FRP materials for impact and mechanical damage (2) the amount of site
retrofitting or strengthening existing concrete structures installation work may be reduced, as surface preparation
have been growing rapidly throughout the world. other than grooving is no longer required; (3) NSM bars
FRP bars used to strengthen RC structures are can be more easily anchored into adjacent members to
currently emerging as a new technology to increase prevent debonding failures; (4) NSM reinforcement is
their bending or shear strength. The FRP rod requires a less prone to debonding from the concrete substrate; (5)
groove to be cut in the concrete cover, a technique the aesthetic of the strengthened structure is virtually
known as Near Surface Mounted reinforcement (NSM) unchanged.
(Taljsten et al. 2003; El-Hacha and Rizkalla 2004; Numerous experimental data exists on shear
Novidis et al. 2006; De Lorenzis and Teng 2007; strengthening of concrete members with externally
Soliman et al. 2010; Al-Mahmoud et al. 2010; Kreit bonded FRP sheets, laminates and strips, whereas an
et al. 2011). NSM FRP technology could be particularly insufficient amount of data are available on shear
attractive to strengthen structures in bending areas strengthening with NSM FRP bars.
subjected to negative moment, to mechanical damage or De Lorenzis and Nanni (De Lorenzis and Nanni
environmental attack. Compared to externally bonded 2001) tested twelve large size T-beams, 6 with stirrups

*Corresponding author. E-mail address: [email protected]; Fax: +61293859747; Tel: +6129385504.

Advances in Structural Engineering Vol. 18 No. 10 2015 1563

 Reinforced Concrete Beams Strengthened with NSM CFRP Rods in Shear

and 6 without stirrups. Carbon FRP (CFRP) ribbed the beams; the contribution of the laminates for the
round rods in epoxy-filled grooves were used as NSM beam shear resistance was limited by the concrete
shear reinforcement. The test variables included bar tensile strength.
spacing and inclination angle, and anchorage of the bars All the previous studies showed that NSM FRP
in the flange. The NSM reinforcement produced a shear technique is efficient to strengthen concrete members in
strength increase which was as high as 106% in the shear. The effectiveness of this technique depends on
absence of steel stirrups, and still significant in presence the FRP bar shape: circular, square, or rectangular cross
of a limited amount of internal shear reinforcement. section, bar inclination and on the filling material used:
Barros and Dias (2003) tested strengthened beams epoxy resin, mortar.
without internal stirrups. The beams were strengthened In this paper, the performance of reinforced concrete
with NSM CFRP strips of different inclinations. Their beams strengthened in shear using NSM technique is
performance was compared to the performance of further investigated. Carbon-epoxy pultruded FRP
similar beams with equivalent amount of externally (CFRP) rod with 6mm in diameter. This work aims to
bonded FRP shear reinforcement. The strength increase enlarge the low amount of experimental data available
observed on the beams with NSW CFRP ranged from in the literature regarding shear strengthening using
22% to 77%, and was in all cases larger than the one NSM technique. The results differ from existing
obtained with externally bonded FRP. literature mainly regarding the interface condition
Nanni et al. (2004) reported the test results of a single between the CFRP rods and the filling material. Indeed,
full-scale PC girder taken from a bridge and shear- previous works deal manly with NSM technique using
strengthened with NSM CFRP rectangular smooth smooth surface FRP bars in resin. But, for some
surface bars in resin. The beam failed in flexure at a applications, when the concrete surface is exposed to
shear force close to the shear resistance predicted by De severe conditions such as high temperatures, epoxy
Lorenzis and Nanni model (2001). Rizzo and De resin cannot be used as filling material. Very few results
Lorenzis (2009a) tested a total of nine, 2.0 m long RC are available regarding using mortar as the filling
beams with a rectangular 200-mm x 210-mm cross- material because bond strength between smooth surface
section to analyze the influence on the structural rods and mortar or concrete is low. NSM shear
behavior and failure mode of selected test parameters strengthening using resin or mortar as a filling material
such as the type of NSM reinforcement (round bars and with surface pre-conditioned CFRP rods is investigated
strips), their spacing, inclination, and the mechanical in this study. One concrete mix was tested. Several
properties of the groove-filling epoxy. The experimental beams were cast with or without steel shear
results showed 22% to 44% increase in shear capacity of reinforcement (stirrups) and including different amount
RC beams strengthened with NSM reinforcement over of NSM CFRP rods. Tests were carried out up to the
the control beam in presence of a limited amount of failure load. Modifications in mechanical behaviour,
stirrups. Failure mode for all the beams strengthened cracking and failure mode of the beams were analysed.
using NSM FRP rods technique was the delamination of The experimental results were compared with existing
the concrete cover of the stirrups which is not a analytical model.
conventional shear failure mode. In another paper
(Rizzo and De Lorenzis 2009b), Rizzo and De Lorenzis 2. EXPERIMENTAL PROGRAM
(2009a) developed a model to compute the NSM bars A total of seven beams, 2 m long, rectangular (150 ∞
debonding failure load of RC beams, assuming that 200 cm) cross-section beams were cast. The first group
debonding involves each NSM bar separately. Anwarul of beams (Four Points Testing, group FPT) was tested
Islam (2009) tested four concrete beams with regular under a monotonically increasing four points load. The
steel reinforcement in flexure. The beams were tension reinforcements consisted of three 16-mm-
strengthened with CFRP rods, all placed vertically. An diameter ribbed bars running along the beam. The
increase of shear strength has been found in the range of second group (Three Points Testing, group TPT) was
17% to 25%. Dias and Barros (2010) compared the tested under a monotonically increasing three points
effectiveness of the externally bonded reinforcement load Three Points Testing (group TPT) in order to
and NSM techniques for the shear strengthening of T increase the shear load (V) for the same loading applied
cross section RC beams. The experimental results compared to group FPT. For all beams, stirrups were
showed that: inclined laminates were more effective made using 6-mm-diameter ribbed bars. The
than vertical laminates; an increase of the percentage of dimensions, reinforcement details and loading setup of
laminates produced an increase of the shear capacity of the beams are shown in Figure 1.

1564 Advances in Structural Engineering Vol. 18 No. 10 2015

Firas Al-Mahmoud, Arnaud Castel, Trinh Quang Minh and Raoul François

2.1. Material Properties shown in Figure 2. Results regarding the yield strength,
2.1.1. Concrete ultimate tensile strength and elastic modulus of steel
A 30 MPa grade concrete was selected for design purpose. bars are presented in Table 1.
The properties of the hardened concrete (compressive
strength, tensile strength and instantaneous elastic modulus) 2.1.3. CFRP rods
were measured at 28 days on concrete cylinders (diameter = The CFRP rod was made in France by the Soficar
110 mm, height = 220 mm). The specimens were removed Company. Al-Mahmoud et al. (2007) tested three
from their molds 24 hours after casting and stored for 28 specimens in axial tension to measure their mechanical
days in a confined room (T = 20 oC, RH = 60%). The tensile
strength was obtained using splitting test. 2500
The average compression strength and the elastic
modulus obtained were 35.4 MPa and 30.4 GPa 2000

Tensile stress (MPa)

respectively with a standard deviation of 1.6 MPa and
1500
0.9 GPa respectively. The tensile strength was 3.1 MPa
CFRP-Manufacturer Test
with standard deviation of 0.2 MPa. CFRP-Laboratory Test
1000
Steel bar
2.1.2. Steel reinforcing bar 500
The steel reinforcing bars were composed of natural
S500 half-hard steels; ordinary ribbed reinforcing steel 0
bars were used. The yield strength, ultimate tensile 0.00 0.01 0.02 0.03 0.04
strength, modulus of elasticity and the hardening Strain
stiffness of the steel bars were measured. The typical Figure 2. Experimentally based stress–strain curves of steel
stress-strain relationship for steel reinforcing bar is reinforcement bar and CFRP rods

P
P/2 P/2
B Stirrups Φ 6 /200
A
FPT1 TPT1

100 450 B 1350 100

100 600 600 A 600 100
Zone A Zone B

Stirrups Φ 6 /200
FPT2 TPT2
Mortar

100
NSM FRP rod
FPT3 TPT3

Resin Resin

160 160
100

2φ6
TPT4
200
160

Stirrups φ 6 /200 Mortar

160 160
3 φ 16
120
150
Sec A-A and B-B
All dimensions are in mm
Figure 1. Beam details and setup for tested beams

Advances in Structural Engineering Vol. 18 No. 10 2015 1565

 Reinforced Concrete Beams Strengthened with NSM CFRP Rods in Shear

characteristics (Figure 2). The brittle failure started with dust using air-brushing pressure to remove debris and fine
splitting and ended with rupture of the rod by tensile particles so as to ensure proper bonding between the
rupture of the fibers. Table 1 shows the strengths and the filling materiel and the substrate concrete. The groove
elastic modulus obtained; the manufacturer’s mechanical was half filled and the CFRP rod was then placed in it and
characteristics of the CFRP rods are also presented. To lightly pressed. This forced the filling material to flow
modify the surface of the initially smooth rods in order to around the CFRP rod. More filling material was applied to
enhance the bond with the filling material and particularly fill the groove and the surface was leveled. The beams
with the mortar, a surface sanding treatment was applied. were tested after one week to ensure the filling material
The smooth CFRP rods were coated with 0.2/0.3 mm-size strength. Each beam was instrumented with a numerical
sand by sprinkling it on to a thin layer of freshly applied displacement transducer to measure the mid-span
epoxy resin (Al-Mahmoud et al. 2007). deflection. Figure 1 shows both shear reinforcement

2.1.4. Filling material

Two filling materials were studied: An epoxy resin and Concrete lug
a mortar. Table 2 shows the mechanical properties
measured at 7 days. b

b
2.2. Beam Strengthening
Installation of the NSM CFRP rods was performed by (a) Concrete is sawn
cutting grooves into the side surfaces of the beams
over their full depth. The grooves had square cross-
section with a nominal size of b = 12 and 15 mm
Residual Root
(Figure 3) for CFRP rods embedded in resin and
mortar respectively and, corresponding to twice and (b) Lug is removed
twice and half the actual diameter of the rod
respectively (Al-Mahmoud et al. 2011). Filling material
A special concrete saw with a diamond blade was used
to cut the grooves. The remaining concrete lugs formed 6 mm CFRP Rod
by sawing the concrete surface were then removed using
(c) Rod is embedded in the groove
a hammer and hand chisel so that the lower groove
surface became rough. The grooves were cleaned of any Figure 3. Installation of the NSM CFRP rod

Table 1. Mechanical characteristics of Steel bars and CFRP rods

Hardening
Yield Ultimate Elastic stiffness
strength strength Ultimate modulus post-
Material (MPa) (MPa) stain (MPa) yielding
Steel Bars
(Kreit et al. 2011) 600 655 0.035 210 000 18 500 MPa
CFRP rods
Manufacturer tTest — 2 300 0.016 150 000 —
Laboratory test
(Al-Mahmoud et al. 2007) — 1 875 0.013 145 900 —

Table 2. Mechanical properties of filling materials, SD: Standard Deviation

Compressive Tensile Elastic

strength/SD strength/SD modulus/SD
Material (N/mm2) (N/mm2) 103 (N/mm2)
Filling material Epoxy resin (7 days) 83/2.2 29.5/0.8 4.94/0.2
Mortar (7 days) 74.6/1.3 6.2/0.3 31.4/1.2

1566 Advances in Structural Engineering Vol. 18 No. 10 2015

Firas Al-Mahmoud, Arnaud Castel, Trinh Quang Minh and Raoul François

(stirrups) and shear strengthening setups for all beams. In Table 3. Details of the shear strengthened
group FPT, FPT1 beam was not strengthened and not beams tested
reinforced in shear in the shear zones (one beam without Number of CFRP
any shear reinforcement). This configuration has been CFRP rods Filling
used in previous studies (Dias and Barros 2010; Usama Beam rods spacing material
and Huda 2013; Amir et al. 2014; Jacinto and Gonzalo FPT3 3 150 resin
2014) in order to quantify the contribution of the concrete TPT2 1 – mortar
and the stirrups in the total shear strength separately (Lima TPT3 1 – resin
and Barros 2011). Stirrups were placed in the constant TPT4 3 150 mortar
moment zone only with 200 mm spacing. This first beam
allows assessing experimentally the plain concrete shear
strength. FPT2 beam was not strengthened neither but was of 45° with respect to the longitudinal axis of the beams
reinforced in the shear zones using stirrups with 200 mm which is the optimal orientation regarding shear strength
spacing. This second beam is the control reinforced (Dias and Barros 2010). Details on strengthened beam are
concrete beam. FPT1 and FPT2 allow assessing presented in Table 3. Those configurations have been
experimentally both plain concrete and stirrups tested by other researchers (Dong et al. 2013; Daniel et al.
contribution to shear load resistance. FPT3beam was 2014). The purpose was to study the effectiveness of
strengthened in shear zones with three CFRP rods NSM-FRP rods in substitution of stirrups.
embedded in the resin and with 160 mm spacing.
In group TPT, TPT1 beam was not strengthened but 3. TEST RESULTS
reinforced with stirrups in the shear zones with 200 mm Table 4 shows the experimental results for all the beams. In
spacing. It is the control beam of the TPT group. TPT2, Table 4, the two last columns report the increase in
TPT3 and TPT4 beams were shear strengthened in zone A percentage in shear force at failure compared to the beam
(Figure 1) with one CFRP rod embedded in the mortar, with plain concrete only (FPT1) and to the beam reinforced
one CFRP rod embedded in the resin and three CFRP rods with stirrups (TPT1) respectively. The concrete contribution
embedded in the mortar respectively. All the strengthened to the shear strength was assessed experimentally using
beams in the group TPT were without stirrups in zone A FPT1 results and was equal to 41.8 kN. The stirrups
where the applied shear load is three times higher than in contribution to the shear strength was assessed
zone B. CFRP rods were always placed with an inclination experimentally using the TPT group control beam (TPT1)

Table 4. Analysis of experimental results

Increase
Increase versus
versus plain stirrups
Ultimate Shear force concrete shear
Shear load Experimental at failure (V) beam reinforced
Beam reinforcement (kN) failure mode (kN) beam % beam %

FPT1 concrete 83.7 shear 41.8 — —

Group FPT2 concrete+ steel 144.4 crusing of 72.2 72.6 —

(FPT) compressed
concrete + shear 72.2

FPT3 concrete+ CFRP 142.4 crushing of 71.2 70.2 —

compressed concrete

TPT1 concrete+ steel 95.1 shear 71.3 70.5 —

TPT2 concrete+ CFRP 128.0 shear 96 129.4 34.6

(failure of bond)
Group
(TPT) TPT3 concrete+ CFRP 136.6 crushing of 102.4 144.9 43.6
compressed concrete

TPT4 concrete+ CFRP 135.4 crushing of 101.6 142.7 42.5

compressed concrete

Advances in Structural Engineering Vol. 18 No. 10 2015 1567

 Reinforced Concrete Beams Strengthened with NSM CFRP Rods in Shear

and was 29.5 kN (71.3 – 41.8 kN). FPT2, the control beam strength is around 144.9 % and 43.6 % compared to
for group FPT, gives similar result. NSM CFRP rods FPT1 (plain concrete only) and TPT1 (reinforced with
contribution to the shear capacity was assessed as the stirrups) respectively. Thus, the resin still performs
difference between the ultimate shear force of the shear- better than the mortar using the same surface pre-
strengthened beam and those of FPT1 or TPT1 beams. conditioned rods but the gain is only moderate.
Beams FPT3 and TPT4 strengthened with three
3.1. Failure Modes and Shear Strength Increase CFRP rods (without stirrups) showed 70.2% and
Regarding the strengthened beams, only TPT2 beam 142.7% increase in ultimate shear force respectively but
failed by shear failure; all the other strengthened beams failed due to concrete crushing. As the failure was
failed by crushing of compressed concrete. This means controlled by the bending strength, shear strengthening
that the full shear capacity of the strengthened beams could not be assessed experimentally for those beams
FPT3, TPT3 and TPT4 was not obtained. Failure of but will be analysed further on in the paper using
TPT2 beam, strengthened with one surface pre- existing models.
conditioned CFRP rod in mortar, occurred due to the All experimental results agree with the ones reported in
bond failure at the interface mortar-substrate concrete. the literature and show the efficiency of the NSM
The increase in shear strength using one NSM CFRP rod technique to strengthen reinforced concrete beams in shear.
embedded in mortar is 129.4 % in comparison with Very few results are available in literature regarding
FPT1 (plain concrete only) and 34.6 % compared to using mortar as the filling material for shear
TPT1 (reinforced with stirrups). This range of shear strengthening because bond strength between smooth
strength increase is similar to the ones reported in the surface rods and mortar or concrete is low
literature. Beam TPT3, strengthened with one CFRP rod (Al-Mahmoud et al. 2007). Indeed, previous study
in resin, failed by crushing of compressed concrete but (Al-Mahmoud et al. 2007) showed that using smooth
observation of concrete cracking (Figures 4 and 5) surface CFRP rod in mortar would lead to an early
showed that the shear failure was almost reached. failure of the strengthening at the interface between the
Comparison with existing model will confirm this mortar and the CFRP rod. In this study, the experimental
assumption later on in the paper. The gain in shear results show that appropriate surface pre-conditioning

FPT1 TPT1

FPT2 TPT2

1emb

FPT3 TPT3

1emb1 1emb2 1emb

1emb1

TPT4

1emb2

Figure 4. Details of the failure zones of the tested beams.

1568 Advances in Structural Engineering Vol. 18 No. 10 2015

Firas Al-Mahmoud, Arnaud Castel, Trinh Quang Minh and Raoul François

FPT1 TPT1
similar to TPT1 except that concrete crushing occurred
simultaneously with shear failure.
For TPT2 beam, at about 80 kN, diagonal shear
cracks initiated at the center of shear span, at about mid-
height of the beam. These cracks did not lead to the
mortar failure because of its higher tensile strength
compared to the concrete. As the load increased, more
FPT2 TPT2
shear cracks formed within the shear span. At 100 kN, a
main shear crack led to the mortar failure separating the
mortar in the groove in two parts. Failure of TPT2
occurred by debonding of the interface mortar-substrate
concrete, leading to concrete palling along the main
shear crack (Figure 5). TPT3 beam exhibit the same
cracking sequence as TPT2 beam except that the main
shear crack led to the resin failure at about 84 kN
FPT3 TPT3
loading. TPT3 beam failed due to concrete crushing.
The difference between the TPT2 and TPT3 ultimate
loads and failure modes is due to the higher bond
strength between the epoxy resin and the substrate
concrete (Al-Mahmoud et al. 2009).
FPT3 and TPT4 beams (strengthened with three
TPT4 CFRP rods embedded in resin and mortar respectively)
showed similar cracking patterns as the ones observed
on the beams strengthened with one CFRP rod only.
Beams FPT3 and TPT4 failed due to concrete crushing.

3.3. Load-Deflection Curves

Typical mid-span deflection versus applied load (P)
Figure 5. Failure mode of the beams tested curves are shown in Figure 6 for both FTP and TPT
groups. For moderate loading value before or after
cracking, all beams show a similar behaviour. At
of the CFRP rod can enhance the bond between CFRP ultimate stage, in addition to the gain in bearing
rods and mortar. As a result, the performance of the capacity due to the strengthening (Table 4), NSM
shear strengthening using NSM CFRP smooth surface CFRP technique can lead to a slight increase in beams
rods embedded in resin (beam TPT3) is similar to the ultimate deflection.
one observed using surface pre-conditioned CFRP rod
embedded in mortar (beam TPT2) 4. EXISTING MODELS FOR FRP SHEAR
CONTRIBUTION COMPUTATION
3.2. Cracking Progression Existing codes/recommendations are based on the
Figures 4 and 5 show the cracking maps and photos of assumption that the total shear strength is the sum of the
the beams at failure respectively. For FPT1 beam (not contributions of concrete (Vc), steel (Vs), and FRP (VFRP):
strengthened and no shear reinforcement), diagonal
shear cracks with about 400 orientation initiated at the Vtot = Vc + Vs + VFRP (1)
center of both shear spans, at about mid-height of the
beam when the load reached about 82 kN. With the load This simple addition of shear components
increase, more bending and shear cracks formed within considering them as independent is one of the
the shear spans. A main shear crack widened and assumptions that have been questioned by a number of
propagated toward the load point application. This main scientists (Carlo and Mira 2014).
shear crack and the secondary horizontal cracks merged Chen et al. (2010) concluded that the maximum shear
at the level of the tensile reinforcement exposing the contributions of steel stirrups and FRP may not be
steel bars and leading to the beam failure (Figure 4). reached simultaneously, so that their combined
TPT1 (not strengthened), showed a conventional shear contribution may be less than the sum of the respective
failure mode. FPT2 (not strengthened) failure mode was peak values of VFRP and Vs. For accurate evaluation of

Advances in Structural Engineering Vol. 18 No. 10 2015 1569

 Reinforced Concrete Beams Strengthened with NSM CFRP Rods in Shear

Group FPT whereas Vc, Vs and Vf,exp are concrete, steel and FRP
160 contributions to the global shear strength respectively.
FPT2
FPT3 Formulations for the shear resistance of external FRP
140
sheets or rods bonded to beams (VFRP) are often
120
described differently by various researches and design
Load (P) (kN)

100 FPT1 guidelines. Most are in fact based on the same equation
80 that incorporates beam geometry, FRP geometry, and
FRP effective strength or strain.
60
According to ACI Committee, the basic equation for
40 the unfactored shear resistance attributed to FRP can be
20 expressed in the following form (ACI 318M-08 2008):
0
0 10 20 30 40 VFRP = ffeρuf bwdf (cotθ + cotβf)sinβf (5)
Mid-span deflection (mm)
where ρuf is the FRP reinforcement ratio, bw is the beam
Group TPT web width, ffe is the FRP effective stress in the principal
160 fiber direction, and θ is the principal compressive stress
140 TPT3 TPT4 inclination. The FRP effective stress is defined as the
effective strain multiplied by the modulus of elasticity Ef.
120
βf is the angle of the FRP principal tensile fibers,
Load (P) (kN)

100 TPT2 measured with respect to the member longitudinal axis. df

80 is the effective depth of the FRP, the definition and
60 calculation of which varies by code, and is sometimes
40
TPT1 presented as the effective height, hfe.
The design shear contribution due to continuous fiber
20
sheets VFRP is calculated according to (JSCE 1997) by
0 either method (1) or method (2) in the following clauses.
0 5 10 15 20 25
Mid-span deflection (mm)
Method (1) in which the coefficient expressing
the shear reinforcing efficiency of the continuous fiber
Figure 6. Load versus mid-pan deflection sheet is used to evaluate the ultimate mean stress of
the sheet and to determine the shear contribution of the
sheet:
the shear resistance they recommended the determination
 cos (∅)   z
VFRP = K  A f f fud  sin∅ + S   
of the maximum value of the combined contribution of
(6)
steel stirrups and FRP strips, since the simultaneous use   f γ
b
of these maximum values seems to be an unconservative
approach in the evaluation of shear strength.
K : Shear reinforcing efficiency of continuous fiber
The assessment procedure developed by (Lima and
sheets (Eqn7)
Barros 2011) implies the comparison between
experimental values and analytical ones, considering the
K= 1.68 – 0.67R, however 0.4 ≤ k ≤ 0.8 (7)
following quantities:
ref
VR,exp = Vc + Vs (2) 2 1
1  ffud  3  1 3
R = ( ρ f Ef ) 4  Ef   f 
cd
str
VR,exp = Vc + Vs + Vf,exp (3)
, however 0.5 ≤ R ≤ 2 (8)
ρf = Af /(bw sf)
str ref
Vf,exp = VR,exp – VR,exp (4)
Af is the total cross-sectional area of continuous fiber
ref
VR,exp is the experimental shear strength of the sheets in space sf , sf are the spacing of continuous fiber
str
unstrengthened reference control beam, VR,exp is the sheet, ffud is the design tensile strength of continuous
experimental shear strength of the strengthened beam, fiber sheet (N/mm2), bw is the beam web width, Ef is the

1570 Advances in Structural Engineering Vol. 18 No. 10 2015

Firas Al-Mahmoud, Arnaud Castel, Trinh Quang Minh and Raoul François

modulus of elasticity of continuous fiber sheet Chajes et al. (1995) proposed that the shear strength
(kN/mm2), ρf is the angle formed by continuous fiber contribution by FRP bars (VFRP ) can be calculated by
sheet about the member axis, γb is the member factor using the following equation:
(generally may be set to 1.25), z is the lever arm length
(generally may be set to d/1.15). A f E f ε ef d
VFRP =
Method (2) in which the stress distribution of the s (sinα + cosα) (10)
continuous fiber sheets is evaluated based on the bond
constitutive law to determine the shear contribution of where d is the distance from extreme compression fiber
the sheet. This method uses numerical calculation based to center of longitudinal tension reinforcement
on the following hypothesis to evaluate the stress (effective depth) (mm), s is the spacing of shear
distribution of the continuous fiber sheet in upgraded reinforcement (mm), Af is the area of FRP bars in shear
members for determining the shear contribution of the on both sides of the beam (mm2), Ef is the elastic
sheet. (i) Shear crack forms a 35° angle about the modulus of FRP rods (MPa), and εef is the effective
member axis. (ii) Member deformation after shear crack strain in FRP bars, α is the orientation of FRP rods.
has occurred is expressed by a rigid body rotation model For vertical NSM FRP rods, Anwarul Islam (2009)
with the end of a shear crack as the center of rotation. suggested that the ratio effective strain to ultimate strain
(iii) The progress of sheet peeling that traverses the of CFRP rods was close to one-third (Eqn 11),
shear crack is evaluated through stress analysis
assuming that the concrete is a rigid body, the sheet is an A f fyf d
elastic body, and there is a linear relationship between VFRP =
3s (11)
the relative displacement and bond stress between the
sheet and the concrete (the bond constitutive law). (iv) where fyf is the tensile strength of FRP bars. But, the
The strain of concrete in compression sections is
value of the effective strains at failure depends strongly
expressed as a function of the angle of rotation of the
on the bond strength between the FRP rods and the
members for which rigid body rotation is assumed. The
filling material. Thus, Eqn 11 is valid for particular
member factor used for this method is generally 1.25.
experimental conditions and still needs further studies to
According to fib design guideline (Triantafillou
be generalized.
et al. 2001), the external FRP reinforcement may be
De Lorenzis and Nanni ( 2001) proposed a simplified
treated in analogy to the internal steel (accepting that
approach accounting for both FRP spacing and
the FRP carries only normal stresses in the principal
orientation with respect to the horizontal direction. This
FRP material direction), assuming that at the ultimate
formulation can capture any possible bond failure: FRP
limit state in shear (concrete diagonal tension) the FRP
rod-substrate concrete and substrate concrete-filling
develops an effective strain in the principal material
material. A constant shear stress τf at the rod-filling
direction, εf,e (note: this is not the principal tensile
material interface is assumed at failure in all the FRP
strain, which may be assumed perpendicular to the
rods crossed by the shear cracks. The FRP shear
crack). The effective strain εf,e is, in general, less than
contribution VFRP can be calculated using Eqn 12.
the tensile failure strain, fu. The FRP contribution to
shear capacity, VFRP, can be written in the following: VFRP = 2lemb;tot pτf sin α (12)
VFRP = 0.9 εfd,e Efu ρf bwd (cotθ+cotα) sinα (9) Where lemb;tot is the total embedment lengths (sum of
where εfd,e is the design value of effective FRP strain, minimum embedment lengths of the bars crossed by the
bw : minimum width of cross section over the effective shear cracks):
depth, d is the effective depth of cross section, ρf is FRP
n
reinforcement ratio equal to 2tf sinθ/bw for
continuously bonded shear reinforcement of thickness tf
1emb,tot. = ∑ 1emb,i (13)
i =1
(bw : minimum width of the concrete cross section over
the effective depth), or (2tf/bw)(bf/sf) for FRP lemb,I is the minimum embedment length (i.e. the
reinforcement in the form of strips or sheets of width bf minimum between the two lengths ls,iand lg,i in which
at a spacing Sf, Efu are elastic modulus of FRP in the the rod is subdivided by the shear crack, Figures 4 and
principal fiber orientation, θ is the angle of diagonal 7), p is the perimeter along which the bond stress acts
crack with respect to the member axis, assumed equal to (failure interface), α is the angle of the FRP bars to the
45°, α is the angle between principal fiber orientation horizontal axis. Eqn 12 is valid providing that lemb,I is
and longitudinal axis of member. smaller than the development length of the bars.

Advances in Structural Engineering Vol. 18 No. 10 2015 1571

 Reinforced Concrete Beams Strengthened with NSM CFRP Rods in Shear

The model proposed by De Lorenzis and Nanni concrete in the case of a NSM CFRP rod embedded in
(2001) is used to analyze the experimental results mortar. According to a previous study (Al-Mahmoud et
presented is this paper. The model accounting for local al. 2010; Al-Mahmoud et al. 2011; Al-Mahmoud et al.
bond–slip behavior is not considered, as the minimum 2012), the values of τf can be considered equal to 5.95
embedment length lemb,i does not exceed the value: MPa and 18.3 MPa for the mortar and the resin
respectively. The values of P used to calculate VFRP are
H 45 mm for the resin and 18.85 mm for the mortar.
X= (14)
2 sin α Table 5 shows a comparison between the experimental
results and the models for VFRP. The experimental value
where H is the beam cross-section height (for H= 200 of VFRP was calculated as the difference between the
mm and α=45°, x=140 mm). As in the experimental shear force at failure (V) and the concrete shear
conditions of the study, lemb,i is equal to or smaller than contribution (VFTP1) (Lima and Barros 2011).
140 mm the distribution of shear stress along the rod- Table 5 shows that, for all the beams, all the design
filling material interface at failure can be considered codes overestimate the VFRP value as they do not
uniform. The average shear stress τf is used to calculate consider any bond failure or failure in bending due to
VFRP (Eqns 11 and 12). Failure occurred at the interface concrete crushing. For the beam TPT2 strengthened
resin-CFRP rod in the case of a NSM CFRP rod with one NSM CFRP rod embedded in mortar, the
embedded in resin and at the interface mortar-substrate
method proposed by De Lorenzis and Nanni (2001)
predict accurately the failure load. For TPT3 beam, the
visual observations during the test showed that this
beam was closed to fail in shear. This is assumption is
i
g,
1

FRP rod (i)

supported by the computed value of VFRP at shear
failure (65.9 kN) which is very close to the one
i
s,

measured experimentally (60.6 kN). Finally, for both

1

Shear crack
beams strengthened using 3 rods, theoretical results are
1emb,i =1s,i 1s,i <_1g,i 107.3 kN and 71.9 kN for the resin and the mortar
respectively. Concrete crushing occurred way before to
Figure 7. Minimum embedment length of i FRP bars reach those values.

Table 5. Comparison between the experimental results of VFRP and the predicted values by
design guidelines.

Computed Computed Computed Computed

VFRP VFRP VFRP VFRP
Number Exp at shear at shear at shear at shear
of NSM Exp VFRP at failure failure failure failure
Filling CFRP failure failure (kN) (kN) (kN) lemb;tot kN)
Beam material rods mode (kN) by Eqn 1 by Eqn 2 by Eqn 5 (mm) by Eqn 8

Fib
ACI 318M-08 JSCE Bulletin De Lorenzis and
2008 1997 2001 Nanni 2001
FPT3 resin 3 Crushing of 29.4 534 372 476 130+90 107.3
compressed
concrete
TPT2 mortar 1 Shear 54.2 267 62 238 135 51.1
TPT3 resin 1 Crushing of 60.6 267 62 238 135 65.9
compressed
concrete
TPT4 mortar 3 Crushing of 59.8 534 372 476 90+100 71.9
compressed
concrete

1572 Advances in Structural Engineering Vol. 18 No. 10 2015

Firas Al-Mahmoud, Arnaud Castel, Trinh Quang Minh and Raoul François

5. CONCLUSIONS Al-Mahmoud, F., Castel, A. and François, R. (2012). “Failure modes

The following conclusions can be drawn from this study: and failure mechanisms of RC members strengthened by NSM
The NSM technique using CFRP rods is very CFRP composites - analysis of pull-out failure mode”,
effective in strengthening reinforced concrete Composites Part B, Vol. 43, pp. 1893–1901.
beams in shear. When using only one surface Al-Mahmoud, F., Castel, A., François, R. and Tourneur, C. (2010).
pre-conditioned CFRP rod, the increase in shear “RC beams strengthened with NSM CFRP rods and modeling of
strength compared to beams reinforced with peeling-off failure”, Composite Structures, Vol. 92,
stirrups only was about 43.6% when using an pp. 1920–1930.
epoxy resin as filling material and about 34.6% Al-Mahmoud, F., Castel, A., François, R. and Tourneur, C. (2011).
when using a mortar. But, as already reported by “Anchorage and tension-stiffening effect between Near-surface-
other authors, shear capacity of the beams raises mounted carbon fiber reinforcement polymer rods and concrete”,
a lot by increasing the number of NSM CFRP Cement and Concrete Composite, Vol. 33, pp. 346–352.
rods in the shear zone. Al-Mahmoud, F., Castel, A., François, R. and Tourneur, C. (2007).
Results show that epoxy resin still performs “Effect of surface pre-conditioning on bond of carbon fiber
better than the mortar but the gain is only reinforced polymer rods to concrete”, Cement and Concrete
moderate. Interface between mortar and surface Composite, Vol. 29, No. 9, pp. 677–689.
pre-conditioned CFRP rod performs very well as Al-Mahmoud, F., Castel, A., François, R. and Tourneur, C. (2009).
the failure occurred at the interface between the “Strengthening of RC members with near-surface mounted CFRP
mortar and the substrate concrete. Increasing the rods”, Composite Structures, Vol. 91, pp. 138–147.
surface roughness of the groove in the substrate Amir, H.G., Amir, H.A., Sadegh, K. and Mostafa, G. (2014)
concrete could be considered in order to “Formulation of shear strength of slender RC beams using gene
improve the strengthening performance. expression programming, part I: Without shear reinforcement”,
For some applications, when the concrete Automation in Construction, Vol. 42, pp. 112–121.
surface is exposed to severe conditions such as Anwarul Islam, A.K.M. (2009). “Effective methods of using CFRP
high temperatures, epoxy resin cannot be used a
bars in shear strengthening of concrete girders”, Engineering
filling material. This work shows that, providing
Structures, Vol. 31, pp. 709–714.
an appropriate surface conditioning of the CFRP
Barros, J.A.O. and Dias, S.J.E. (2003). “Shear strengthening of
rods, a mortar can be used as well with much
reinforced concrete beams with laminate strips of CFRP”,
less vulnerability to high temperatures. More
Proceedings of the International Conference Composites in
generally, the cost of an epoxy resin is much
Constructions CCC2003, pp. 289–294.
higher than the cost of any mortar thus mortar is
Carlo, P. and Mira, V. (2014). “Assessment of design procedures for
an attractive alternative to epoxy resin.
the use of externally bonded FRP composites in shear
Experimental results were compared to some
strengthening of reinforced concrete beams”, Composites: Part
existing models for FRP shear strength
B, Vol. 45, pp. 727–741.
computation (ACI Committee, JSCE and Fib
Bulletin). The design codes overestimate the Chajes, M.J., Januszka, T.F., Mertz, D.R., Thomson, T.A. and Finch,
VFRP value as they do not consider any bond W.W. (1995). “Shear strengthening of reinforced concrete beams
failure or failure in bending due to concrete using externally applied composite fabrics”, ACI Structural
crushing. Journal, Vol. 92, No. 3, pp. 295–303.
De Lorenzis and Nanni (2001) model intends to Chen, G.M., Teng, J.G., Chen, J.F. and Rosenboom, O.A. (2010)
calculate the failure load of strengthened “Interaction between steel stirrups and shear-strengthening FRP
structures due to NSM FRP debonding. De strips in RC beams”, Journal of Composites for Construction,
Lorenzis and Nanni model provided very good ASCE, Vol. 14, No. 5, pp. 498–509.
predictions when applied to the beams that Daniel, B., Khaled, S. and Martin, N. (2014) “Strengthening of shear
failed due to both rod-filling material debonding critical RC beams with various FRP systems”, Construction and
or filling material-substrate concrete debonding. Building Materials, Vol. 66, pp. 634–644.
The model seems suitable for NSM CFRP shear De Lorenzis, L. and Nanni, A. (2001). “Shear strengthening of
strengthening design. reinforced concrete beams with NSM fiber-reinforced polymer
rods”, ACI Structural Journal, Vol. 98, No. 1, pp. 60–68.
REFERENCES De Lorenzis, L. and Teng, J.G. (2007). “Near-surface mounted FRP
ACI 318M-08 (2008). Building Code Requirements for Structural reinforcement: An emerging technique for strengthening
Concrete and Commentary, American Concrete Institute, USA. structures”, Composites: Part B, Vol. 38, pp. 119–143.

Advances in Structural Engineering Vol. 18 No. 10 2015 1573

 Reinforced Concrete Beams Strengthened with NSM CFRP Rods in Shear

Dias, S.J.E. and Barros J.A.O. (2010). “Performance of reinforced Nanni, A, Di Ludovico, M. and Parretti, R. (2004). “Shear
concrete T beams strengthened in shear with NSM CFRP strengthening of a PC bridge with NSM CFRP rectangular
laminates”, Engineering Structures, Vol. 32, pp. 373–384. bars”, Advances in Structural Engineering, Vol. 7, No. 4,
El-Hacha, R. and Rizkalla, S. (2004). “Near-surface-mounted fiber- pp. 97–109.
reinforced polymer reinforcements for flexural strengthening of
Novidis, D., Pantazopoulou, S.J. and Tentolouris, E. (2006).
concrete”, ACI Structural, Vol. 101, No. 5.
“Experimental study of bond of NSM-FRP reinforcement”,
Triantafillou, T., Matthys, S. and Audenaert, K. (2001). Externally
Bonded FRP Reinforcement for RC Structures, Technical Report, Construction & Building Materials, pp.1–11.
fib Bulletin No. 14, The International Federation for Structural Rizzo, A. and De Lorenzis, L. (2009a). “Behavior and capacity of
Concrete, Lausanne, Switzerland. RC beams strengthened in shear with NSM FRP reinforcement”,
Jacinto, R.C. and Gonzalo, R. (2014). “Bond and size effects on the Construction and Building Materials, Vol. 23, pp.1555–1567.
shear capacity of RC beams without stirrups”, Engineering Rizzo, A. and De Lorenzis, L. (2009b). “Modeling of debonding
Structures, Vol. 66, pp. 45–56. failure for RC beams strengthened in shear with NSM FRP
Dong, J.F., Wang, Q.Y. and Guan, Z.W. (2013). “Structural reinforcement”, Construction and Building Materials, Vol. 23,
behaviour of RC beams with external flexural and flexural–shear pp.1568–1577.
strengthening by FRP sheets”, Composites: Part B, Vol. 44, pp.
Soliman, S.M., El-Salakawy, E. and Benmokrane, B. (2010).
604–612.
“Flexural behaviour of concrete beams strengthened with near
JSCE. (1997). Recommendation for Design and Construction of
Concrete Structures using Continuous Fiber Reinforcing surface mounted FRP bars”, Canadian Journal of Civil
Materials, Concrete Engineering Series 23, Japan Society of Civil Engineering, Vol. 37, No. 10, pp. 1371–1382.
Engineers, Tokyo, Japan. Taljsten, B., Carolin, A. and Nordin, H. (2003). “Concrete
Kreit, A., Al-Mahmoud, F. Castel, A. and François, R. (2011). structures strengthened with near surface mounted reinforcement
“Repairing corroded RC beam with near-surface mounted CFRP of CFRP”, Advances in Structural Engineering, Vol. 6, No. 3,
rods”, Materials & Structures, Vol. 44, No. 7, pp. 1205–1217. pp. 201–213.
Lima, J. and Barros, J. (2011). “Reliability analysis of shear Usama, E. and Huda, S. (2013). “Hybrid shear strengthening system
strengthening externally bonded FRP models”, Structures & for reinforced concrete beams: An experimental study”,
Buildings, Vol. 164, No. 1, pp 43–56. Engineering Structures, Vol. 49, pp. 421–433.

1574 Advances in Structural Engineering Vol. 18 No. 10 2015