Behavior of Reinforced Concrete Beams Retrofitted in Flexure Usin

Portland State University
PDXScholar
Dissertations and Theses Dissertations and Theses
Spring 5-21-2015
Behavior of Reinforced Concrete Beams Retrofitted

in Flexure Using CFRP-NSM Technique
Salam Al-Obaidi
Follow this and additional works at: https://pdxscholar.library.pdx.edu/open_access_etds
Part of the Structural Engineering Commons

Let us know how access to this document benefits you.
Recommended Citation
Al-Obaidi, Salam, "Behavior of Reinforced Concrete Beams Retrofitted in Flexure Using CFRP-NSM
Technique" (2015). Dissertations and Theses. Paper 2294.
https://doi.org/10.15760/etd.2291
This Thesis is brought to you for free and open access. It has been accepted for inclusion in Dissertations and
Theses by an authorized administrator of PDXScholar. For more information, please contact [email protected].
Behavior of Reinforced Concrete Beams Retrofitted in
Flexure Using CFRP-NSM Technique
by
Salam Al-Obaidi
A thesis submitted in partial fulfillment of the

requirements for the degree of
Master of Science
in
Civil and Environmental Engineering
Thesis Committee:
Franz Rad, Chair
Peter Dusicka
Evan Kristof

2015
Abstract
A variety of retrofitting methods are used to upgrade existing structures. For example,
steel plates and Fiber Reinforced Polymer (FRP) jackets are externally bonded to
members to increase their capacity in flexure and shear. However, due to the issue of
corrosion these strengthening systems may lose their efficiency with time. FRP materials
have been used to strengthen many structural components of different shapes and types.
FRP jackets, FRP Strips, and FRP rods have commonly been used to rehabilitate existing
structural components. The many advantages of using FRP as strengthening materials
have made this material an attractive alternative: advantages such as lightweight, high
strength, and ease of setting up. Among the many applications using FRP, Near Surface
Mounted –Fiber Reinforced polymer (NSM-FRP) is a promising technique used to
strengthen concrete members. However, de-bonding issues have to be overcome to make
this technique efficient and reliable. The NSM-FRP technique consists of making a
groove along the surface of the concrete member to be retrofitted with depth less than the
cover of the member. After cleaning the groove, epoxy paste is used to fill two-thirds of
the groove’s depth. The FRP element is then mounted in the groove. Finally, the groove
is filled with epoxy and the excessive epoxy is leveled with surface of the concrete. This
technique makes the FRP material completely covered by epoxy in the cover of the
concrete. This method can be used for strengthening both the positive and negative
moment regions of girders and slabs. Groove size, paste, concrete, and rods properties are
the main variables that control the efficiency of the NSM-FRP rods. The main objective
i
of this research project is to determine the behavior of reinforced concrete beams that are
strengthened with NSM-CFRP reinforcement bars. In this research project, the bond
characteristics of NSM-CFRP reinforcement bars are first determined from pullout tests.
Then, NSM-CFRP rods are installed in reinforced concrete beams and the beams are
tested. Loads, strains, and deflections are measured and theoretical and measured
capacities are compared. Finally, the reliability and efficiency of using NSM-CFRP rods
technique in retrofitting existing structures is observed.
ii
Acknowledgment
The author would like to thank Professor Franz Rad for his guidance and advice on this
project. The author also would thank the committee members: professors Peter Dusicka
and Evan Kristof for serving the thesis defense. The author would also like to thank the
following graduate students for their support and help in the experimental work of this
project: Yasir Saeed, Karrar Al-Lami, Hayder Alkhafaji, Salih Mahmmod, Ali Salih, and
Anas Yousifani. The author extends his sincere appreciation to Mr. Tom Bennett for his
help in setting up the data acquisition system and facilitating the experimental works.
Special thanks to Robert Kosowski for his help with most of the experimental research.
Special thanks to The Higher Committee for Education Development in Iraq for funding
this project.
iii
Table of Contents
Abstract ……………………………………………………………………….……..….. i
Acknowledgment ………………………………………………………………...………iii
List of Tables ……………………………………………….…………..…..……….. . vii
Lists of Figures …………………………………………………………………...….... viii
Notations ………………………………………….……………………………..…..… xvi
Chapter 1: Introduction ................................................................................................... 1
1.1. General ................................................................................................................ 1
1.2. Near Surface Mounted Technique (NSM) ........................................................... 2
1.3. Objective .............................................................................................................. 4
Chapter 2: Literature Review.......................................................................................... 5
2.1. Introduction .......................................................................................................... 5
2.2. Bond Mechanism.................................................................................................. 5
2.3. Flexural Behavior ............................................................................................... 14
2.4. Summary of the Literature Review and Discussion ........................................... 26
Chapter 3: Experimental program ................................................................................ 30
3.1. Design guides ..................................................................................................... 30
3.2. Tension Test ....................................................................................................... 30

iv
3.2.1. Methodology ............................................................................................... 30
3.2.2. The Results.................................................................................................. 37
3.3. Pullout Tests ....................................................................................................... 38
3.4. Flexural Tests ..................................................................................................... 40
3.4.1. Materials ..................................................................................................... 41
3.4.2. Concrete ...................................................................................................... 42
3.4.3. Epoxy .......................................................................................................... 44
3.4.4. Steel............................................................................................................. 44
3.4.5. Specimens’ Analysis and Design ................................................................ 46
3.4.6. Specimens Preparation ................................................................................ 52
3.4.7. Concrete Casting and Curing ...................................................................... 53
3.4.8. Instrumentations .......................................................................................... 54
3.4.9. Preparation of the Groove ........................................................................... 56
3.4.10. Inserting the CFRP Bars. ......................................................................... 57
3.4.11. Instrumentations and Procedure .............................................................. 57
Chapter 4: The Results.................................................................................................. 59
4.1. Pullout Test ........................................................................................................ 59
4.2. Discussing the Pullout Results. .......................................................................... 65
4.3. Beam Test ........................................................................................................... 70
v
4.3.1. Set (A) Specimens....................................................................................... 71
.4.3.2 Set (B) specimens ....................................................................................... 92
4.3.3. Set A and B Results Comparison .............................................................. 108
4.3.4. Measured Crack Width ............................................................................. 112
4.3.5. Theoretical Analysis ................................................................................. 113
Chapter 5: Summery Conclusions and Recommendations ......................................... 122
5.1. Summery .......................................................................................................... 122
5.2. Conclusions ...................................................................................................... 122
5.3. Recommendations ............................................................................................ 123
References ....................................................................................................................... 124
Appendix A ..................................................................................................................... 127
Appendix B ..................................................................................................................... 130
vi
List of Tables
Table 1-1-1 Typical properties of steel and FRP materials. ............................................... 2
Table 2-1 The test results. ................................................................................................. 17
Table 2-2 Specimens details. ............................................................................................ 18
Table 2-3 Test results. ...................................................................................................... 19
Table 2-4 Specimens details and results. ......................................................................... 24
Table 3-1 The manufacturer’s data sheet .......................................................................... 31
Table 3-2 Concrete properties. .......................................................................................... 43
Table 3-3 Epoxy properties. .............................................................................................. 44
Table 3-4 Steel tensile properties. ..................................................................................... 46
Table 3-5 NSM-CFRP flexural cross section analysis for set A. ...................................... 49
Table 3-6 NSM-CFRP flexural cross section analysis for set B. ...................................... 51
Table 4-1 pullout results summery. ................................................................................. 68
Table 4-2 Set A specimens results summery. ................................................................... 71
Table 4-3 Set B results summery. ..................................................................................... 92
Table 4-4 Set A results comparison. ............................................................................... 111
Table 4-5 Set B results comparison. ............................................................................... 111
Table 4-6 Ultimate loads comparison. ............................................................................ 114
vii
List of Figures
Figure 1-1 Stress-strain curve comparison of steel and FRP bars. ..................................... 2
Figure 1-2 NSM technique.................................................................................................. 3
Figure 2-1 Hinged beam pullout test with inverted T beam section. ................................. 7
Figure 2-2 Modified pullout test scheme. ........................................................................... 8
Figure 2-3 Slip- bond stress relationship. ........................................................................... 9
Figure 2-4 Direct puulout test with spurious stress breakers. .......................................... 10
Figure 2-5 Pull out test setup and specimen dimensions. ................................................. 11
Figure 2-6 Modified pullout test with instrument. ............................................................ 12
Figure 2-7 Specimens with mechanical interloching. ....................................................... 13
Figure 2-8 Load –deflection curve. ................................................................................... 15
Figure 2-9 Load –deflection curve for retrofitted beams. ................................................. 20
Figure 2-10 Load –deflection curve for retrofitted cantilever beams with FRP bars. ...... 21
Figure 2-11 Retrofitteed beam sections details. ................................................................ 22
Figure 2-12 NSM CFRP mechanical interlocking details. ............................................... 25
Figure 3-1 Anchor details. ................................................................................................ 31
Figure 3-2 ASLAN200 CFRP rods. .................................................................................. 32
Figure 3-3 Steel pipe used to grip #2 CFRP rod. .............................................................. 33
Figure 3-4 The expansive cement. .................................................................................... 33
Figure 3-5 The expansive cement’ properties manufacturer’s data sheet. ....................... 33
Figure 3-6 Anchoring one end of #2CFRP rod. ................................................................ 34
Figure 3-7 Alignment of #2CFRP rod. ............................................................................. 34
viii
Figure 3-8 #2 CFRP specimen ready to be tested . .......................................................... 35
Figure 3-9 #2CFRP being tested. ..................................................................................... 35
Figure 3-10 Failure mode for #2CFRP rod. ...................................................................... 36
Figure 3-11 #2 CFRP Tensile Properties. ......................................................................... 37
Figure 3-12 Pull out test setup. ......................................................................................... 39
Figure 3-13 Control specimens of set A and B. ................................................................ 40
Figure 3-14 Details of set A and set B. ............................................................................. 41
Figure 3-15 Concrete mechanical tests. ............................................................................ 43
Figure 3-16 #3 steel strain stress diagram........................................................................ 45
Figure 3-17 #4 steel strain stress diagram........................................................................ 45
Figure 3-18 Tensile test of steel. ....................................................................................... 46
Figure 3-19 Strain and stress distribution (ACI-440-2R-08). .......................................... 47
Figure 3-20 Steel cage and mold fabrication. ................................................................... 52
Figure 3-21 Steel cage inside the molds with the required cover. ................................... 52
Figure 3-22 concrete casting and leveling ........................................................................ 53
Figure 3-23 Concrete casting. ........................................................................................... 53
Figure 3-24 Concrete finishing and curing. ..................................................................... 54
Figure 3-25 Strain gages preparation and posting. ........................................................... 55
Figure 3-26 Making grooves in the tension side of the strengthened beams. ................... 56
Figure 3-27 Inserting CFRP Rods..................................................................................... 57
Figure 3-28 Four-point load test setup. ............................................................................. 58
Figure 4-1 Pull out test failure without lateral grooves. .................................................. 59
ix
Figure 4-2 Load-strain diagram for #2with out lateral grooves. ....................................... 60
Figure 4-3 Lateral grooves details. ................................................................................... 61
Figure 4-4 Strain load diagram for#2 CFRP with lateral grooves. ................................... 62
Figure 4-5 Pullout-test-failure for #2 with lateral grooves. .............................................. 62
Figure 4-8 Pullout load slippage curves........................................................................... 65
Figure 4-9 #2 CFRP pullout stress distribution. .............................................................. 66
Figure 4-10 #3 CFRP pullout stress distribution. ............................................................. 67
Figure 4-11 # 4 CFRP pullout stress distribution. ............................................................ 67
Figure 4-12 Bond stress distribution (ACI 440.2R-08, p 40). .......................................... 68
Figure 4-13 Average shear stress for NSM CFRP rod. .................................................... 69
Figure 4-14 Two point load flexural test. ......................................................................... 70
Figure 4-15 Specimen AC strain and stress profile according to ACI. ............................ 72
Figure 4-16 Specimen AC load deflection-curve. ............................................................ 72
Figure 4-17 Specimen AC strain profile at different load level........................................ 73
Figure 4-18 Failure mode of specimen AC....................................................................... 74
Figure 4-19 ACI strain and stress analysis of specimen AR2. ......................................... 75
Figure 4-20 Specimen AR2 load -deflection curve. ......................................................... 76
Figure 4-21 Section strain profile for specimen AR2 at different load levels. ................ 77
Figure 4-22 Tension layer delamination (ACI440.2R-08, p 38). ..................................... 77
Figure 4-23 CFRP load-strain relationship for specimen AR2. ........................................ 78
x
Figure 4-24 Specimen AR2 failure mode. ....................................................................... 79
Figure 4-25 stress and strain analysis of specimen AR3 .................................................. 79
Figure 4-26 Load deflection curve for specimenAR3. ..................................................... 80
Figure 4-27 Failure mode of specimen AR3..................................................................... 81
Figure 4-28 Strain profile for specimen AR3 at different load level. ............................... 82
Figure 4-30 Strain and stress analysis of specimen AR4. ................................................. 83
Figure 4-31 Load deflection curve of specimen AR4. ...................................................... 84
Figure 4-32 Failure mode of specimen AR4..................................................................... 85
Figure 4-33 Strain profile of specimenAR4 at different load levels. ................................ 85
Figure 4-35 Strain and stress analysis of specimen AR4-S. ............................................. 87
Figure 4-36 Load deflection curve for specimen AR4-S. ................................................. 88
Figure 4-37 Failure mode of specimen AR4-S. ................................................................ 89
Figure 4-38 Strain profile for specimen AR4-S at different load levels. .......................... 90
Figure 4-39 CFRP load-strain relationship for specimen AR4-S. .................................... 91
Figure 4-40 ACI 318 strain and strain profile for specimen BC. ...................................... 93
Figure 4-41 Load deflection curve of specimen BC. ........................................................ 93
Figure 4-42 Failure mode of specimen BC. ...................................................................... 94
Figure 4-43 Strain profile for specimen BC at different load levels................................. 95
Figure 4-44 Strain and stress analysis of specimen BR2. ................................................. 95
Figure 4-45 Load deflection curve of specimen BC. ........................................................ 96
xi
Figure 4-46 Failure mode of specimen BR2. .................................................................... 97
Figure 4-47 Strain profile for specimen BR2 at different load levels. .............................. 97
Figure 4-48 CFRP load-strain relationship for specimen BR2. ........................................ 98
Figure 4-49 Strain and stress analysis of specimen BR3. ................................................. 99
Figure 4-50 Load deflection curve of specimen BR3. .................................................... 100
Figure 4-51 Failure mode of specimen BR3. .................................................................. 101
Figure 4-52 Strain profile for specimen BR3 at different load levels. ........................... 102
Figure 4-53 CFRP load-strain relationship for specimen BR3. ...................................... 103
Figure 4-54 Strain and stress analysis of specimen BR4. .............................................. 103
Figure 4-55 Load deflection curve of specimen BR4. .................................................... 104
Figure 4-56 Failure mode of specimen BR4. .................................................................. 105
Figure 4-57 Strain profile for specimen BR4 at different load levels. ............................ 106
Figure 4-58 CFRP load-strain relationship for specimen BR4. ...................................... 107
Figure 4-59 Set A load deflection curves. ..................................................................... 108
Figure 4-60 Set A normalized load deflection curves. ................................................... 108
Figure 4-61 Set B load deflection curves. ....................................................................... 109
Figure 4-62 Set B normalized load deflection curves. .................................................... 109
Figure 4-63 Set A normalized load crack-width curve. .................................................. 112
Figure 4-64 Set B Normalized load crack-width curves. ................................................ 113
Figure 4-65 Theoretical load deflection curves for set A. .............................................. 115
Figure 4-66 Normalized theoretical load deflection curves for set A ............................. 115
Figure 4-67 Theoretical Load deflection curves for set B. ............................................. 116
xii
Figure 4-68 Normalized load deflection curves for set B. .............................................. 116
Figure 4-69 Specimen AC theoretical and experimental load deflection comparison. .. 117
Figure 4-70 Specimen AR2 theoretical and experimental load deflection comparison. 118
Figure 4-71 Specimen AR3theoretical and experimental load deflection comparison. . 118
Figure 4-72 Specimen AR4-S theoretical and experimental load deflection curves ..... 119
Figure 4-73 Specimen AR4 theoretical and experimental load deflection comparison. 119
Figure 4-74 Specimen BC theoretical and experimental load deflection comparison.... 120
Figure 4-75 Specimen BR2 theoretical and experimental load deflection comparison.. 120
Figure 4-76 Specimen BR3Theoretical and experimental load deflection comparison. 121
Figure 4-77 Specimen BR4 Theoretical and experimental load deflection. ................... 121
xiii
Notations
NSM = Near Surface Mounted Retrofitting Technique.
CFRP = Carbon Fiber Reinforced Polymer.
GFRP = Glass Fiber Reinforced Polymer.
AFRP = Aramid Fiber Reinforced Polymer.
EFRP = External Bonded Reinforcement.
Db = Bar Diameter.
𝑓𝑦 = Steel Yielding Stress.
𝜀𝑦 = Steel Yielding Strain.
𝑓𝑓𝑢 = CFRP Ultimate Tensile Stress
𝐿𝑑𝑏 = Development Length.
𝑓𝑐′ = Concrete Compressive Strength.
𝜏𝑟−𝑒 = Shear Stress at Epoxy Rod Interface.
𝜏𝑒−𝑐 = Shear Stress at Epoxy Concrete Interface.
𝑑𝑓𝑟𝑝 = The Flexural Depth of CFRP Reinforcement.
𝜀𝑓𝑟𝑝 = CFRP Strain.
𝜀𝑠1 = Compressive Steel Strain.
xiv
𝐸𝑓𝑟𝑝 = Modulus of Elasticity for CFRP Material.
𝐴𝑓𝑟𝑝 = CFRP Cross Sectional Area.
𝐸𝑠 = Steel Modulus of Elasticity.
𝐴𝑠1 = Cross Sectional area of Compressive Steel Reinforcement.
𝜀𝑠 = Strain in the Steel Reinforcement.
𝐴𝑠 = Cross Sectional Area of Tensile Steel Reinforcement.
𝑑1 = The Depth of Compressive Steel Reinforcement.
C = The Depth of Compressive Concrete Section.
𝑑 = The Depth of Tensile Steel Reinforcement.
𝑀𝑐 = Concrete Bending Moment Capacity.
𝑀𝑠𝑡1 = Compressive Steel Bending Moment Capacity.
𝑀𝑓𝑟𝑝 = CFRP Bending Moment Capacity.
𝑀𝑠𝑡 = Tensile Steel Bending Moment Capacity.
𝛷𝑦 = Section Curvature at Steel Yielding Step.
𝛷𝑐 = Section Curvature at Concrete Crushing Step.
a,b = Dimensions of the Grooves.
xv
Chapter 1: Introduction
1.1. General
Many of the existing reinforced concrete (RC) members are exposed to degradation.
Repairing these components is costly and demanding. Plenty of strengthening members
and techniques have been carried out to repair the deteriorated elements. Steel plates and
bars are the common ones that are used to rehabilitate the concrete and masonry members
such as slabs, beams, columns, and walls. However, this technique has some drawbacks
such as corrosion, enlargement of the original repaired sections, and the difficulty of
lifting, and installation process. The most attractive alternative is the FRP materials.
They do not corrode, are lightweight, and have high level of strength compared to steel.
Nevertheless, they are expensive, are brittle, and have a low modulus of elasticity.
Generally, the FRP materials consist of fibers that are impregnated in the matrix of vinyl
ester which transvers the loads between the fibers and protects them. The fibers could be
made of glass, aramid, and Carbone. The fibers provide the composites materials with the
strength and the required stiffness. This composition makes the FRP materials strong in
the direction of the fibers and weak in the transverse direction. The CFRP’s tensile
strength is about more than four times the tensile yielding strength of the steel, and it is
much lighter. Table 1-1 and Figure 1-1 show the comparison in behavior between the
steel and different types of FRP rods. There are different texture of CFRP rods available
in the industries such as smooth surface, sand coated, ribbed, and sand coated with helical
wrap. These textures work as mechanical interlocking to provide good adhesion when
they are used as strengthening elements.
1
Table 1-1-1 Typical properties of steel and FRP materials.
Bar type Tensile strength (ksi) Modulus of elasticity (ksi)
Steel 60 29000
GFRP 100 6000
AFRP 170 12000
CFRP 315 18000

(ksi)
Stress (ksi)
Figure 1-1 Stress-strain curve comparison of steel and FRP bars.
1.2. Near Surface Mounted Technique (NSM)
According to the 2013 report card for America’s infrastructure as published by American
Society of Civil Engineers (ASCE) “one in nine of the nation’s bridges are rated as
structurally deficient”. Since many structures, especially bridges, are classified as
structurally deficient, rehabilitation of these structures is a serious concern for engineers.
A variety of retrofitting methods are in process to upgrade the existing structures. Steel
2
plates and FRP jackets are externally bonded to the retrofitted members to increase their
capacity of flexure and shear. However, due to the corrosion issue, theses strengthening
systems loss their efficiency with time. The near surface mounted technique has recently
become a promising and accepted method. It essentially makes a groove along the surface
of the retrofitted concrete with depth less than the cover of the member. After cleaning
the groove, an epoxy paste is used to fill a 2/3 of the groove’s depth. The FRP element
then is mounted in the groove. Finally, the groove is completely filled with epoxy and the
excessive epoxy is leveled with surface of the concrete. This technique makes the FRP
material completely covered by epoxy in the cover of the concrete. NSM is also
desirable for strengthening the negative regions of girders and slabs where the
mechanical and corrosions factors are severely damaging for externally bonded steel or
FRP plates. FRP rods or strips can be used to perform this technique. However, the
relatively small contacting surface of the FRP strips to the surrounding concrete surface
makes the de-bonding more feasible and controlling. As compared with the externally
bonding reinforcement (EBR), the NSM does not require any preparation for the grove
prior to the installation. CFRP material is more secured and protected in NSM than the
EBR system.
Figure 1-2 NSM technique (Taljisten et al. 2003).
3
1.3. Objective
The main objective of this research project is to investigate the behavior of the
rectangular beams that are strengthened with NSM-CFRP reinforcement bars. By
obtaining the characteristics of bonding behavior from the pullout test, the NSM-CFRP
rebars are going to be implemented in rectangular concrete beams. The efficiency of
using this method in retrofitting existing structures and bridges can be obtained by
understanding the behavior and parameter of NSM-CFRP rods technique. This research
contributes to and provides more knowledge regarding this method so that engineers can
judge and demonstrate the rehabilitation of the deteriorated structures.
4
Chapter 2: Literature Review
2.1. Introduction
FRP materials have been used to strengthen many structural components in many
different shapes and types. FRP jackets, FRP Strips, and FRP rods are the common types
that have been used to rehabilitate the existing structural components. The many
advantages of using FRP as strengthening materials have made this material an attractive
alternative. The main benefits of using FRP are that it is lightweight, high in strength, and
easy to set-up. Among the many applications of using FRP, NSM-FRP is a promising
technique of strengthening the concrete members. However, de-bonding issues have to be
overcome to make this technique more efficient and reliable. The NSM-FRP rod is
basically pressing the FRP rod into a cut that is pre made in the concrete cover of the
strengthened components. The FRP rods are attached to the concrete by epoxy paste.
Groove size, paste, concrete, rods properties are the main variables that control the
efficiency of the NSM-FRP rods.
2.2. Bond Mechanism
As stated above, the main characteristic that controls the efficiency of the NSM technique
is the bonding. Much research on the pullouts-FRP rods has been done in the last recent
years. The pullout test aims to investigate the bonding behavior of the NSM-FRP joint.
Laura De Lorenzis et al. (2002) performed an experimental test on the bond between the
NSM rods and concrete that had compressive strength of 27.6 MPa (4000psi). Bond
length, diameter of the rod, material types of FRP, surface treatment, and the size of the
5
groove were the main characteristics that were addressed in the study. Four different
bonded lengths were chosen: 6, 12, 18, and 24 times the rod diameter. Two diameters of
#3 and #4 rods for carbon and glass FRP were investigated. The CFRP rod had a two
surface configuration: deformed and sand coated. Three different sizes of groove were
tested. The test was performed on an inverted T beam where the tension face was
strengthened with the NSM technique as shown in Figure 2-1. It was observed that the
specimens with deformed rods failed by splitting the epoxy paste and cracking the
concrete surrounding the groove. However, the pullout phenomenon was dominant in the
specimens retrofitted with sand blasted rods. It was also observed that the groove size had
a significant effect on increasing the bonding strength. As the groove thickness increased,
the strength of the epoxy past increased and the failure shifted from the epoxy to the
surrounding concrete. The bond length also had a remarkable influence on the bond
behavior. By increasing the bond length, the stress was distributed over the length so that
it prevented the early de-bonding in the epoxy-rod interface. It was also reported that the
optimum groove size for #3and #4 rods is ¾ in and 1.0 in respectively.
6
Figure 2-1 Hinged beam pullout test with inverted T beam section(Laura De Lorenzis et al. 2002).
Laura De Lorenzis et al. (2002) conducted an experimental work on the configuration of
a new specimen as shown in Figure 2-2. This modified pullout test was conducted with
the purpose of eliminating any eccentricity that would occur in the conventional pullout
test. The test has investigated 36 specimens with various variables. Type of the FRP rod,
adhesive material, development length, and the groove size were the parametric study of
this research. Carbone FRP with ribbed and sand coated, as well as glass FRP, were
investigated under the pullout test. The filling adhesive materials were epoxy-based and
cement mortar. The compressive strength of the concrete was 22 MPa (3.19 ksi). The
development lengths of the rod were varied as 4Db, 12Db, and 24Db. Finally, four groove
sizes were cut in the specimens to find out the sensitivity of the groove size on the overall
bonding behavior: 1.25Db, 1.5Db, 2Db, and 2.5Db. From the test results, many failure
modes were reported: pullout at the concrete- adhesive interface; pullout at the rod-
adhesive interface; splitting of the adhesive’s cover with no concrete cracking; and
7
crushing the concrete surrounding the groove with formation of crack in the adhesive’s
cover. Curve (a) shown in Figure 2-3 represents the typical slip-stress diagram for epoxy
adhesive with the precut groove, which is usually smooth and sand coated surface. The
failure in curve (a) occurred at the concrete adhesive interface. Curve (b) represents the
typical slip-bond stress curve for a glass FRP rod when the failure is the epoxy-concrete
interface. As the groove size increased, the failure shifted from epoxy-rod interface to
concrete- epoxy interface with initiation of cracks in the surrounding concrete. It was also
reported in this study that the epoxy adhesive provides a better bonding strength than the
cement mortar adhesives due to its high shear strength. It should also be noted that the
surface configuration of the CFRP played no significant role in the bonding behaviors
since the controlling failure was at concrete-epoxy interface.
Figure 2-2 Modified pullout test scheme(Laura De Lorenzis et al. 2002).
8
Figure 2-3 Slip- bond stress relationship(Laura De Lorenzis et al. 2002).
D. Novidis et al. (2008) did a direct pullout test by using the testing system shown in
Figure 2-4. The parametric study was the development length and the groove size.
Twenty-four concrete specimens were used with compressive strength of 34.5 MPa (5000
psi). The specimens’ dimensions were 150 mm (6.0 in) squared, and 300 mm (12.0 in)
was the length of the concrete block. The various embedded lengths were 3Db, 5Db,
7.5Db, and 10Db, and the groove dimensions were either 25 mm (1.0 in) or 20 mm
(0.8in). The conclusion indicated that the adopted pullout test gave reliable results with a
manageable specimen size. Two failure modes were obtained: pullout of the rod at
concrete-epoxy interface, and pullout of the rod at epoxy-rod interface. As the groove
size increased, the strength of the joint increased. In the same manner for a given groove
size, the bonding strength increased as the development length increased up to a certain
9
length. The non-uniformly distributed stress occurred after a limit increasing of
development length, resulting in a decrease in the bonding strength.
Figure 2-4 Direct puulout test with spurious stress breakers(D. Novidis et al. 2008).
Bilotta A. et al. (2011) conducted experimental research comparing the EBR and NSM
systems. For the NSM technique, basalts, glass, and Carbone materials were used with
different geometries. All the pullout tests were performed on prisms that have dimensions
of 160x200 mm and 400 mm height as shown in Figure 2-5. The CFRP rods had 8mm
diameter with smooth surface. The groove dimensions were 1.75Db and 300 mm as
development length (37.5Db). It was reported that the pullout test of CFRP rod failed at
50 kN (11.0 kips) with de-bonding at the epoxy-concrete interface, and with the
detachment of the concrete layer as a mode of failure. It is worth noting that the concrete
compressive strength was 19 MPa (3.0 ksi) to simulate the poor existing RC components.
As a comparison between the EBR and NSM systems, it was concluded that the NSM
10
system worked more efficiently than the EBR system. For NSM system, the CFRP rod
was fully bonded to the concrete subsurface by the adhesive. Therefore, more than 50%
of the CFRP’s tensile strength could be extracted. For NSM system, the CFRP rod was
full bonded to the concrete by the adhesives. Therefore, more than 50% of the CFRP
tensile strength could be extracted by using the NSM system.
Figure 2-5 Pull out test setup and specimen dimensions(Bilotta A. et al. 2011).
Shehab M. Soliman et al. (2013) conducted an extensive bonding test on 80 specimens
with the compressive strength range between (38-44 MPa). They adopted the modified
test system that was used by De Lorezis of 340 mm x40 mm (13.38 in x13.38 in) and 500
mm (20.0 in) height as shown in Figure 2-6. The main aim of the study was to investigate
the weather effect on the bonds’ behavior. However, many bond characteristics were
investigated such as: FRP type—glass and carbon with sand coated ; embedment length:
6Db, 12Db, 18Db, 24Db, 36Db, and 48Db; adhesive type: epoxy based and cement based
adhesive; and groove size: 1.5Db and 2.0Db. Two diameters of CFRP rods were used
11
9.5mm (0.37 in.) and 12.7 mm (0.5 in.), which had a tensile strength of 1546 MPa (224
ksi) and 1250 MPa (181 ksi) respectively. The results indicated that the conditioned
specimens performed less efficiently than the reference ones. In terms of the adhesive
types, the reference specimens with the epoxy based adhesive had a consistently higher
bond strength than the specimens with the cement based adhesive. It should be
mentioned that increasing the bonding length, increased the strength of the joint.
However in the study, the rupture of the CFRP and GFRP rods was reported for the 24Db,
36Db, and 48Db, which gave full efficiency for bonding. Shear tension failure of the
surrounding concrete was the controlling mode failure for the reference specimens with
the epoxy based adhesive. However, for the conditioned specimens, the failure was
controlled by the epoxy splitting. The controlling failure mode for the reference
specimens with the cement based adhesive was the de-bonding in the concrete–adhesive
interface.
Figure 2-6 Modified pullout test with instrument (Shehab M. Soliman et al. 2013).
12
Sharaky, I. A. et al. (2013) investigated many bond parameters. The main characteristics
that were implemented and investigated by using the modified pullout test were: groove
surface, groove geometry, FRP bar type, bond length, and construction details of the
groove. Two CFRP rods were used in the test 8 mm (0.315 in) with smooth surface
texture and 9.05 mm (0.356 in) with surface texture. The compressive strength of the
concrete ranged between (35.2-42.2 MPa) (5-6.12 ksi). Many conclusions have been
drawn and addressed out of the test’s results. It was reported that the groove surface had
no effect on the bond capacity if the failure was at the bar-epoxy interface. As the bond
length increased from 40 mm (5Db) to 192 mm (24Db) for the CFRP 8mm rod, the failure
load increased from 12.75 kN (2.866 kips) to 36.59 kN (8.225 kips). Increasing the
groove dimensions from 1.5db to 2db, delayed the bar epoxy interface failure. It was also
reported that the interlocking increased the joint capacity by 14.8%. It is interesting to
mention that the transverse interlocking, which is shown in Figure 2-7 below had a
remarkable impact on the transfer load failure and stress from concrete epoxy interface to
the surrounding concrete and enhancement of the concrete epoxy interface bond.
Figure 2-7 Specimens with mechanical interloching(Sharaky, I. A. et al. 2013, p 353).
13
2.3. Flexural Behavior
Täljsten et al. (2003) conducted an experimental work on eight rectangular full-scale
beams. In this study, an experimental work was carried out to investigate the flexural
behavior of the beams that strengthened with NSM CFRR. The strengthening was with
the pre-stressed NSM CFRP and without pre-stressed NSM CFRP. The flexural
reinforcement of all specimens consisted of two16mm steel rebars, and had concrete
compressive strength of 60 MPa (9000 psi). The beams of 4 m (13.2 ft.) length and
300x200 mm (11.811x7.874 in) cross section were tested under four-point loads. The
shear span was 1300 mm reinforced with 10mm steel stirrups spaced @ 75 mm. Four
beams were tested for each series. For the first series of the four specimens, one was a
control specimen and three were strengthened with two NSM CFRP 10 mm (0.39 in)
squared rods. Epoxy bond was used to bond the beam (E4) with full development length
and E3 with a development length of 3000 mm. However,a cement grout bond was used
to reinforce beam (C3) with a development length of 3000 mm. For the second series,
pre-stressed NSM CFRP rods were used. The groove size for the epoxy retrofitting beam
was 15x15 mm (0.6x0.6 in) and 20x20 mm (0.787x0.787 in) for the cement based
adhesive beam. The reference beam had overall flexural strength of 79 kN (17.76 kips).
The cement grout strengthened beam had an overall flexural strength of 123 kN (27.652
kips) and failed by anchorage slippage. However, the overall flexural strength of the
epoxy-strengthened beam was 140 kN (31.473 kips) and failed by anchorage failure.
Finally, the full-developed length of epoxy strengthened beam failed by rupture at an
overall flexural strength of 152 kN (34.171 kips). In general, the strengthened beams
14
exhibited a significant improvement in the flexural performance over the reference beam
as shown in Figure 2-8 below.
Figure 2-8 Load –deflection curve(Täljsten et al. 2003).
Hassan, T et al. (2004) conducted an experimental and analytical study on eight simply
supported T- beams. Flexural and bond performance was investigated in this study. The
beam had a total depth of 300 mm (11.811in) and clear span of 2.5 m (8.2 ft). The
parameters of this study were the development length and adhesive materials. One of the
beams designated as Ao shown in Table 2-1 below was reinforced with only steel and
considered as a reference beam. The flexural reinforcement consisted of two No.10 and
two No15 rebars as secondary reinforcement. The others specimens were retrofitted with
9.5 mm (0.375 in) CFRP bars that had a modulus of elasticity of 111 GPa (16099.0 ksi)
and an ultimate tensile strength of 1918 MPa (278.0 ksi). All of the beams had a concrete
compressive strength of 48 MPa (6.96 ksi) and the groove size of 18 mm (0.71in) width
and 30 mm (1.18 in) depth. Different development lengths for NSM-CFRP rods were
15
used for the rest of the specimens. A1, A2, A3,and A4 were strengthened with gel epoxy
adhesive that had a modulus of elasticity of 1200 MPa (174.0 ksi), a tensile strength of
48 MPa (7.0 ksi), and a development length of 150 mm (6.0 in), 550 mm (22.0 in), 800
mm (31.5 in), and 1200 mm (4.0 ft) respectively. Whereas, A5, A6, and A7 were
retrofitted with epoxy adhesive that had a modulus of elasticity of 3000 MPa (435.0 ksi),
a tensile strength of 62 MPa (9.0 ksi), and a development length for the CFRP bars of 550
mm (22.0 in), 800 mm (31.5 in), and 1200 mm (47.0 in) respectively. The results are
shown in Table 2-1 below. It can be observed that the adhesive type did not have a
significant effect on the behavior of the bonding. It also shows that the failure modes
were mostly the de-bonding at the concrete epoxy interface. This de-bonding occurred at
the zone where the secondary reinforcing steel was terminated. The maximum stress at
the CFRP bars was 45% of the ultimate strength of the CFRP bars for the maximum
development length. This indicated that the rupture of the FRP bars did not occur before
the de-bonding failure. A new analytical relationship for bonding was proposed based on
the finite element modeling and the experimental works. By using the new proposal,
a chart of finding a development length based of the rods size was constructed and
verified with ACI code. The summary of the study indicated that the NSM FRP technique
can improve the stiffness and the flexural strength of the strengthened beam. A clear
spacing, edge distance, and development length of achieving a good tensile strength
before de-bonding were suggested. However, this length depended on many parameters
such as the dimension of the bars, concrete and adhesive properties, reinforcement
configuration, and groove width.
16
Table 2-1 The test results (Hassan, T et al. 2004,).
Tang, W. C. et al. (2006) conducted an experimental test on ten beams that were tested
under flexural to investigate the moment-deflection behavior. The parameters of this
research were the compressive strength of the concrete 58,37,and 21 MPa (8.41, 5.36,and
3.04 ksi); the types of reinforcements—steel reinforcing, sand blasted GFRP reinforcing,
and NSM sand blasted GFRP reinforcing; and the type of adhesive—epoxy past XH-130
and XH-111. All of the specimens were tested under two point symmetrical loads. The
specimens’ dimensions were 250 mm x180 mm (9.84x7.08 in) cross section and 1500
mm (4.9 ft) span length. Two different diameters of GFRP bars were used: a #3 sand
coated with a diameter of 9.5 mm (0.374 in) and an ultimate strength of 650 MPa (94.0
ksi), and a # 5 sand coated with diameter of 16 mm (0.629 in) and an ultimate tensile
strength of 512 MPa (74.0 ksi). The reinforcing steel bar was 16 mm (0.629 in) diameter
with an ultimate tensile strength of 478 MPa (69.328 ksi). The specimens’ configurations
are shown in Table 2-2 below. In Table 2-3, the results of each specimen are reported.
Different mode failure and moment-deflection behaviors were observed for different
reinforcing characteristics. The steel-reinforced concrete beams exhibited flexural failure
where the cracks were initiated near the tension face, and widely extended to the center
17
line of the beam until the failure. The GFRP reinforced concrete beams failed in a
combination of shear and concrete compression failure. For the # 5 bars, the rapture was
a remarkable failure. The steel- reinforced concrete strengthened with NSM GFRP
exhibited splitting of the epoxy simultaneously with shear failure followed by the rupture
of the NSM GFRP, particularly for the concrete with high compressive strength. As a
comparison between the steel and GFRP reinforcing, the GFRP exhibited nonlinear
behavior due to the extensive cracking. The moment capacity remained the same for
both materials. However, the GFRP attained the same moment at a high deflection rate.
The NSM GFRP reinforced beams showed a significant increasing (23% -53% over the
same corresponding to specimens without NSM technique) in the flexural stiffness and
moment capacity. The lightweight concrete performed better with the NSM GFRP. In
general, the epoxy paste type XH-130performed better than epoxy paste type XH-111.
Table 2-2 Specimens details (Tang, W. C. et al. 2006,).
18
Table 2-3 Test results (Tang, W. C. et al. 2006).
Al-Mahmoud, F.et al. (2010) carried out an experimental work on concrete beams
strengthened by two # 2 CFRP rods. In this study, seven beams were categorized into
two groups: one group was tested as conventional beams under four-point load, and the
other group was tested as a cantilever beam. For each group, different development
lengths were applied and tested. The main purpose of this study was to investigate the
flexural behavior of each group after rehabilitating with CFRP rods. Each group had the
controlling beam, which had a cross section of 150 x 280 mm (5.9 x 11.02 in) and 3m
(9.84 ft) span length reinforced with two-12 mm steel bars (0.47 in) at the tension face.
The compressive strength of the concrete for all specimens was 37.0 MPa. The retrofitted
specimens had two grooves at the tension face and two CFRP rods of 6 mm that had a
young modulus of 146 GPa (21175.5 ksi), and a tensile stress of 1875 MPa (271.945 ksi)
was placed inside the groove. Depending on the development length, two modes of
failure were observed: pullout and peeling-off failure. For the first group, two
19
development lengths were used: one with 2700 mm (106.0 in.) along the beam’s length,
and the other 2100 mm (83.0 in) ending before the supports. The beam with 2700 mm
(106.0 in.) exhibited the pullout de-bonding at the load of 53.3 kN (12.0 kips). The other
beam with 2100 mm (83.0 in) failed at 44 kN (10.0 kips) with peeling-off failure as
shown in Figure 2-9 below. The other group had three different development lengths:
2400 mm (9.44 in), 190 mm (7.5 in), and 150 mm (6.0 in). The beam with the 240 mm
(94.4 in) development length cracked at 8 kN (1.798 kips), then the steel yielded at 38 kN
(8.542 kips), after which the failure occurred at 59.5 kN (13.4 kips). The beam with the
1900 mm (75.0 in) development length cracked at 10 kN (2.25 kips), and failed at 52.0
kN (11.6 kips) by peeling off the concrete surrounding the groove. This beam maintained
a 72% flexural strength over the reference beam. Finally, the beam with the 1500 mm
(59.0 in.) development length cracked at 25 kN (5.620 kips), then at 31 kN (6.969 kips)
the steel yielded, after which the failure occurred at 36 kN (8.093kips) by peeling of the
concrete. This beam maintained a 20% flexural strength over the reference beam as
shown in Figure 2-10 below.
Figure 2-9 Load –deflection curve for retrofitted beams (Al-Mahmoud, F.et al. 2010).
20
Figure 2-10 Load –deflection curve for retrofitted cantilever beams with FRP bars(Al-Mahmoud, F.et
al. 2010).
Soliman, S. M. et al. (2010) investigated many characteristics of NSM-FRP bars. Twenty
specimens were tested to investigate their flexural behavior. The specimens were
strengthened with NSM-FRP rods to enhance their loads-deflection relationship. The
parameters of the study were the types of NSM-FRP rods, FRP bar diameter, bonded
length, and groove size. Three series were set and casted with ready mix concrete that
had a compressive strength of 40.0 MPa (6000 psi). All specimens were tested under
four point load with shear span of 800 mm. Three series were categorized based on the
steel ratio as shown in Figure 2-11 below.
21
Figure 2-11 Retrofitteed beam sections details(Soliman, S. M. et al. 2010,p 1373).
Series A consisted of twelve beams. The control beam was reinforced with 0.4% steel
reinforcement. The rest of the specimens were reinforced with 9.5 mm (0.374 in) CFRP
bar. The development lengths were 12Db, 24Db, 48Db, and 60Db with a groove size of
1.5Db and 2Db. The objective of this series was to investigate the development length and
the groove size. Series B included three specimens. The control specimens were
reinforced with 0.8% of steel. The other two beams were retrofitted with 9.5 mm (0.374
in) CFRP bars that had a development length of 24Db and 48Db with 2Db as the groove
size. The objective of this series was to investigate the moderate steel ration on the
performance of the NSM-FRP technique. Finally, series C consisted of five beams. The
control specimens were reinforced with 1.6% steel. The rest of the specimens were
additionally reinforced with 9.5 mm (0.374 in) CFRP bars that had a development length
22
of 12Db, 18Db, 24Db, and 48Db with 2Dbas the groove size. The obtained results in terms
of load-deflection curve showed the same behavior as it was observed by Al-Mahmoud,
F.et al. (2010). The general behavior had three limits: the first limit was the concrete
cracking limit where all of the specimens have the same value and behavior; the second
limit was the steel yielding limit where the steel started to yield and the FRP bars started
to involve more; and the last limit was where the FRP bars rupture occurred. From series
A, it was found that the gaining capacity was achieved by 22%, 33%, 71%, and 75% by
increasing the development length from 12Db up to 60Db. However, the gaining capacity
in series B was found to be 4%-19% only. This ration decreases dramatically in series C
down to 1%-9%. Generally, the efficiency of using NSM-FRP was increasing with the
decreasing of the steel reinforcement ratio. In this research, the optimum development
length was found to be 48Db and the optimum steel ratio 0.4%. It was discovered that
using the smaller groove size delays the de-bonding failure by increasing the distance
between the FRP rods and the steel reinforcements. CFRP and GFRP give the same load
carrying capacity at failure. However, GFRP gave more ductility due its relatively low
modulus of elasticity. A good indication of this was observed when the maximum
measured strains of FRP rods were between 75%-85% of the rupture strains depending
on the development lengths.
Micelli, F. et al (2013) investigated the flexural behavior of six reinforced concrete
rectangular beams that were strengthened with CFRP rods. The beams were 200 mm x
400 mm (7.87 in x 15.75 in) and 4.3m long. Two parameters were used: the ratio of the
internal steel and the ratio of the CFRP reinforcing bars. Series A had two 14 mm in
23
diameter internal steel bars as tensile reinforcement, whereas series B had two 18mm in
diameter internal steel bars as tensile reinforcement. Both series had a concrete
compressive strength of 15 MPa (2.17 ksi) and NSM reinforcement of one and two 8mm
(0.315 in) CFRP rods. As shown in the Table 2-4 below, the results indicated that the
NSM system sufficiently improved the overall behavior of the flexural capacity for the
strengthened beams.
Table 2-4 Specimens details and results (Micelli, F. et al 2013).
Table 2-4 also shows a comparison between the theoretical and experimental results
which are in a reasonable agreement with each other. The results also show that the high
steel ratio beams failed by concrete crushing after the steel yielding. Whereas, the low
steel ratio steel failed by CFRP rods de-bonding after the steel yielding. It was reported
that the de-bonding of the CFRP rods for (BRl-a) and BR2-a, which they have less steel
ratio, was due to the low concrete compressive strength where the crack extended rapidly
after the imitation. Therefore, de-bonding between the CFRP rods and concrete occurred.
On the other hand, BR1-b and BR2-b, which had two NSM CFRP rods, suffered from the
stress concentration at the edges and between the CFRP rods prior to the failure.
24
W.T. Jung, et al (2005) investigated the flexural behavior of the RC beams strengthened
with EBR and NSM. Eight specimens of 200 x 300 mm (7.87 in x 11.8 in) cross section
with 3.0 m long span were tested under two-point load system. All specimens had
compressive strength of 31 MPa (4.5 ksi) at twenty-eight days and tested. The main
flexural reinforcement consisted of three 10mm steel rebars and the shear reinforcements
were 10mm steel rebar spaced @100 mm (4.0 in). The control beam failed at 50 kN
(11.24 kips) by the yielding of the steel, and then by the crushing of the concrete. The
EBR beams that were strengthened with sheet and strip gave about 61.0 kN (13.71 kips)
before the CFRP sheet de-bonded. The NSM beams failed at 62.0 kN (14.0 kips) before
the NSM FRP reinforcement de-bonding, which was caused by the cracking of the
concrete surrounding the groove. Finally, the beams that were reinforced with NSM
plate and rod, in addition to the mechanical interlocking, failed at 62.0 kN (14.0 kips) and
65.0 kN (14.6 kips) respectively before the rupture of the NSM FRP reinforcement. The
mechanical interlocking was added to enhance the bonding capacity by increasing the
concrete epoxy bonding efficiency . Therefore, the concrete splitting failure was delayed.
Figure 2-12 below illustrates the mechanical interlocking details and dimensions.
Figure 2-12 NSM CFRP mechanical interlocking details(W.T. Jung, et al 2005).
25
2.4. Summary of the Literature Review and Discussion
From the literature review, it can be concluded that the bonding of NSM CFRP rod is
mainly influenced by some of the following characteristics: the groove size, the surface
texture of the FRP and the groove, the development length, and the concrete compressive
strength. These characteristics are the main parameters that affect the bonding behavior.
Much of the research focused on the load-slip behavior by conducting the pullout tests.
The load-slip relationship is crucial for modeling the behavior of NSM-CFRP interface.
Therefore, many of the studies were performed on short development lengths.
Nonetheless, Shehab.M .Soliman et al. (2013) investigated the behavior of the longer
development lengths such as 36Db and 48Db groove sizes of 1.25Db, 1.5Db, 1.75Db, 2Db,
and 2.25Db. As the groove size increased, the bond failure mode shifted from epoxy
splitting failure mode to concrete epoxy interface de-bonding failure. De Lorenzis. (2002)
suggested that the groove size of 2Db as an optimum groove size. The four possible
failure modes were: the FRP rod-epoxy interface de-bonding, or splitting of the epoxy;
the epoxy-concrete interface de-bonding; the epoxy-concrete interface de-bonding with
crushing of the concrete surrounding the groove; and the FRP rod rupture. The first and
the second failure could be avoided by increasing the development length and the groove
size respectively.
Sharaky et al. (2013) used mechanical interlocking to increase the joint capacity.
However, the reported failure mode was de-bonding at epoxy-concrete interface with
diagonal concrete cracking. Cement grout and epoxy-based paste were used as adhesive
agents. It was concluded that the epoxy-based paste performs and interacts with FRP
26
material better than cement-based paste. As long as the tensile strength of the epoxy was
greater than the tensile strength of the concrete, the difference in the performance of
different types of epoxy was negligible. It is interesting to mention that many researchers
use the modified pullout test that was invented by De Lornzis. (2002) to eliminate the
eccentricity, and the effect of bearing stress on the NSM-CFRP joint. The surface texture
of the CFRP rod does not play a significant role if the failure mode is at concrete-epoxy
interface (De Lornzis, 2002). Shehab M. Soliman et al. (2013) reported that the FRP rod
rupture in the pullout test with 36Db and 48Db development length. However, the tensile
strength of the FRP rod was 1546 MPa (224 ksi) for 9.5 mm #3 rod, and 1250 MPa
(181.3 ksi) for 12.2 mm #4 rod. These values are less the capacity of the FRP bar that is
available in the industry. The compressive strength of the concrete in most studies ranged
from between 19 MPa (2.75 ksi) and 44 MPa (6.38 ksi). This range covers most of the
existing systems that need to be retrofitted. Finally, it is interesting to mention that the
channel shape of the specimens was chosen to eliminate the bearing effect on the joint
performance (De Novidis et al., 2007).
As a flexural behavior, the NSM CFRP technique significantly improved the overall
behavior of load deflection relationship. Most of studies implemented a full development
length and a groove size of 2Db to perform the NSM CFRP systems. Most of the
specimens’ cross-section ranged between (200 mm x300 mm-180 mm x250 mm) that is
(7.87 in x11.811 in-7 in x9.84 in). This range provided enough room to accommodate
two CFRP of #3 or #5 as the maximum limits for the beam. The beam size is also easy to
carry and to maneuver in the laboratory area. The main steel reinforcement ranged from
27
0.4% to1.6%. These ratios covered the minimum and the maximum steel ration that
could be found in most of the existing beams. The minimum steel ratio simulated the
deteriorated beams so that their reinforcement was corroded or accidentally lost. On the
other hand, the medium or maximum steel ratio simulated the existing components that
needed upgrading for their flexural performance to satisfy the new applied loads. The
concrete compressive strength ranged from 15 MPa (2.17 ksi) to 60 MPa (8.7 ksi). This
range covered almost all of the existing reinforced concrete components. However, the
high compressive strength was rarely found in the old concrete beams. Since the test is
for flexural behavior, most of the beams were tested under two-point load. The two-point
load test gives a pure flexural region to be observed and monitored. Therefore, the
results will be more reliable. As it was stated previously, the NSM CFRP provides an
excellent improvement for the flexural performance of the RC beams.
Al.Mohamoud.F.et al. (2010) and Soliman S. M. et al. (2010) classified the load
deflection curve of NSM-CFRP RC beams into three stages. The first limit is the concrete
cracking limit where the retrofitted and the not retrofitted beams attend this limit at the
same time and behavior. In this limit, the steel and the CFRP bars have not engaged. The
second limit is the steel yielding limit where the steel rebar starts to yield and the FRP
rods start to engage. In this limit, the not retrofitted beams will stop picking up additional
load and behave ductile until it reaches the failure limit. Whereas, the retrofitted beams
continue picking up more load. Therefore, the NSM-CFRP beams have different stiffness
and capacity. The third limit is where the beam reaches its maximum capacity and after
this point, three possible failure scenarios occur depending on the NSM parameter. The
28
extreme scenario is the rupture of the FRP. The rupture of the FRP would occur if the
beam has low steel ratio and low FRP ratio. The second scenario is the pullout of the
FRP rod. This behavior occurs if the development length is not sufficient and the
concrete compressive strength is relatively low. The last scenario is the peeling off of the
bottom concrete layer with the NSM CFRP. This phenomenon occurs if the steel and the
FRP ratio is high, and if the CFRP has sufficient development length. W.T. Jung, et al
(2005) has introduced mechanical interlocking to overcome the premature de-bonding
and increase the overall flexural behavior by 14%. As described previously, the
mechanical interlocking increases the concrete-epoxy bonding area and delays the early
bonding, especially for concrete beams that have low compressive strength. The overall
gaining capacity in flexural for RC beams ranges from (23%-80%). This range depends
on many characteristics and parameters of NSM that described previously.
29
Chapter 3: Experimental program
3.1. Design guides
Two design guides were adopted to analyze and design the NSM-CFRP system. The
ASTM D7205/D7205M – 06 has the instructions and details for preparing and testing the
CFRP rod samples. The ultimate tensile strength, the ultimate tensile strain, and the
modulus of elasticity were obtained by using the ASTM D7205/D7205M – 06 guidelines.
The ACI 440.2R-08 “Guide for the Design and Construction of Externally Bonded FRP
Systems for Strengthening Concrete Structures” has the guidelines and philosophy of
analysis and design the NSM-FRP beams. The flexural design of the NSM-CFRP
rectangular concrete beam system was analyzed and designed using the ACI 440.2R-08.
3.2. Tension Test
3.2.1. Methodology
As described previously, the FRP materials are relatively weak in the transverse direction
compare to the longitudinal direction. This inherent property of the composite materials
causes premature failure due to the griping action in the tensile test. Therefore, ASTM
D7205/D7205M – 06 was adopted in this research to perform the tensile test properly.
The axial stress-strain behavior and the CFRP rods modulus of elasticity can be evaluated
later, and compared with the results of the manufacturer’s data sheet. The figures
provided below are adopted from the ASTM to illustrate the process of preparing the
specimens for the test. A steel tube, anchor filling materials, and the FRP samples are the
main elements of the test. Each of these elements is described in figures below.
30
Figure 3-1 Anchor details(ASTM D7205/D7205M – 06, p 8).
The CFRP material, the Aslan™ 200 series, was purchased from ASLAN FRP (Hughes
Brothers, Inc). Three CFRP rod diameters were ordered from ASLAN FRP for which the
tensile properties are shown in Table 3-1 described by the manufacturer. The CFRP rod
#2 was selected for the tensile stress.
Table 3-1 The manufacturer’s data sheet
31
Figure 3-2 ASLAN200 CFRP rods.
The specimen was cut to the desired length. The steel pipe measuring a diameter of
32mm and an inner diameter of 20 mm was ordered from Williams Form Engineering
Corp., as shown in Figure 3-3. The chosen length of the steel tube was 10” so that the
total length of the specimen was 30 in, and the available accommodating room in the
testing machine was 36 in. The ASTM suggests a filling material that provides (30-50
MPa) with an embedment length of diameter ratio of approximately 50. The
commercially available demolition agent known as BUSTAR was selected for the
grouting. It provides about 60 MPa pressure after 70 hours curing with 20oc temperature
as shown in Figure 3-5 below.
32
Figure 3-3 Steel pipe used to grip #2 CFRP rod.
Figure 3-4 The expansive cement.
Figure 3-5 The expansive cement’ properties manufacturer’s data sheet.
The slurry was mixed per the instructions on the manufacturer’s data sheet, and then
poured inside the steel tube, which had been plugged with a PVC cap from one side as
shown below. The CFRP rod then was inserted inside the steel tube. After twelve hours
33
of curing, the same process was repeated on the other side. The specimens were then left
to cure for another two days, and to wait for the expansive agent to reach the 60 MPa
pressure.
Figure 3-6 Anchoring one end of #2CFRP rod.
Figure 3-7 Alignment of #2CFRP rod.
Extra care was taken to ensure the alignment of the CFRP rod inside the tube before the
test to prevent any eccentricity from taking place.
34
Figure 3-8 #2 CFRP specimen ready to be tested .
The test (short-term static strength) was carried out in a MTS machine at the iSTAR
LAB. The strain-stress relationship was measured by laser extensometer and strain
gauges as shown in the results below. The obtained tensile stress and the modulus of
elasticity were as described by the manufacturer. However, one of the specimens
indicated higher tensile strength than manufacturer data sheet.
Figure 3-9 #2CFRP being tested.
35
It is interesting to mention that the failure happened exactly at the middle of the specimen
as shown in Figure 3-10. Prior to the rupture, a breaking of the helical fibers was heard,
after which the carbon fibers lost their composites and started to rupture.
Figure 3-10 Failure mode for #2CFRP rod.
36
3.2.2. The Results
Stress (ksi)
(a) #2 CFRP Tensile Properties.

(ks)
(b) CFRP Tensile Properties comparison.
Figure 3-11 #2 CFRP Tensile Properties.
As shown in the Figure 3-11 (b) above, the tensile stress is higher than the manufacturer
data stated for the #2 CFRP rod. However, the ultimate tensile strain is slightly lower
than the reported value. These values will result in a modulus of elasticity slightly higher
than the manufacture’s data. The overall trend of the tensile stress-strain relationship is
37
elastic until failure. This trend was reported by the manufacture, and pointed out by the
ASTM test. Therefore, these values will be used for the analysis and design of the NSM-
CFRP technique.
3.3. Pullout Tests
The modified pullout test was adopted in this study to investigate the bonding behavior of
NSM-CFRP rod. Some of the parameters of the bonding test were: the CFRP rods size,
the groove characteristic, and the development length. The objective of this test was to
obtain a pullout load of at least 50% of the rod’s capacity before de-bonding occurred.
All of the variables above were implemented in the concrete specimen blocks as
described in figure below. A grove size of 2Db was selected as suggested by the
literature. As mentioned in the literature, Shehab M. Soliman et al. (2013) reported the
FRP rod rupture in the pullout test with a 36Db and 48Db development length. However,
the tensile strength of the FRP rod was 1546 MPa (224 ksi) for 9.5mm #3 rod and 1250
MPa (181.3 ksi) for 12.2mm #4 rod. These values were less than the capacity of the FRP
bar obtained from the tensile stress of the #2CFRP rod. Therefore, 60Db was used as the
development length. Six specimens (two for each bar size) were investigated for their
NSM bonding. The specimens were pulled out the concrete block as shown in the test
setup Figure 3-12 below.
38
Steel tube holding the CFRP Bar
Hollow core ram
CFRP Bar
LVDT for loaded end slipage 2C10*20
10mmSteel plate
10mm Rubber membrane
A
concrete specimen
l/3
B
l/3 36
l/3
D
4.0 3.0 3.5 3.5 3.0 4.0

Side view Front view
20.0
2D
20.0
2D
CFRP Bar
Top view
Epoxy Paste
Figure 3-12 Pull out test setup.
39
3.4. Flexural Tests
After obtaining the bond characteristics from the pullout test, CFRP rods were inserted
into grooves that were made in the bottom surface of each of the beams. Two CFRP rods
were inserted in a groove size of 2Db. Two sets of reinforced concrete beams were
fabricated and casted. Each set contains four beams. One of them was the control beam,
and the rest were additionally reinforced with two CFRP rods of #2, #3, and #4. Each set
had a different internal steel ratio. The first set had the minimum steel ratio, which was
0.4%. This steel ratio simulated the poorly reinforced concrete beams in the real
application or the corroded reinforced concrete beams. The second set had a steel ratio of
0.7%. This series representd the existing beams that needed to be upgraded to satisfy the
new applied load. The cross section of the beams was 10 in x 6.5 in with span length of
8ft. Figure 3-14 shows set A and B cross sections.
6.5 6.5
#3@4in c-c
#3@4in c-c
10.00 10.00
2#3s 2#4s
SET (B) SET (A)
P/2 P/2
stirrups #3@4in
10.0
6.0 32.0 32.0 32.0 6.0
108.0
Figure 3-13 Control specimens of set A and B.

40
6.5 6.5
10.00 A B 10.00
2#4s 2#3s
6.5 6.5
AR2 BR2
10.00 10.00
2#4s 2#3s
2#2c 2#2c
6.5
6.5
AR3 BR3 10.00

10.00
2#3s
2#4s
2#3c
2#3c
6.5
6.5
AR4 BR4
10.00
10.00
2#3s
2#4s
2#4c
2#4c
Figure 3-14 Details of set A and set B.

3.4.1. Materials
41
The materials used in this projects were concrete, steel, CFRP, and epoxy. All the
material were tested according ASTM in iSTAR lab and in South Green House Lab at
Portland State University.
3.4.2. Concrete
A compressive strength of concrete at 4000 psi was selected for both the flexural and
pullout tests. The concrete was supplied by Miller’s Mini Mix Concrete. Cement type II
was used, and the aggregates were supplied by The Willamette River bed. The gravel and
sand gradation was as follows: 100% passing the 25 mm sieve (1.0 in.), 48.34% passing
the 12.7 mm (1/2 in.), 26.16% passing 9.5 mm (3/8in.), and 0.66% passing 4.75 mm (#4).
For the sand, the gradation was as follows: 97.81% passing 4.75mm sieve (#4), 84.67%
passing 2.00 mm (#10), 72.99% passing 0.85 mm (#20), 49.64% passing 0.425 mm
(#40), 5.11% passing 0.150 mm (#100), and 0.73% passing 0.075 mm (#200). The slump
of 4 in was achieved by reducing the added water to the mix. As shown in Figure 3-23,
the concrete compressive strength was performed on the cylinders. The average
compressive strength was (3.72 ksi). Splitting and flexure tests were also conducted
according to the ASTM to obtain the mechanical properties of plain concrete. The table
provided below describes the mechanical properties of the concrete at twenty-eight days.
42
Table 3-2 Concrete properties.
Compressive Tensile strength Modulus of rupture(ksi)

strength(ksi) (ksi)
3.72 0.342 0.55
Figure 3-15 Concrete mechanical tests.
43
3.4.3. Epoxy
The epoxy was used in this research was Sika AnchorFix®-1. It is a fast curing anchoring
adhesive, which consisted of two parts: adhesive and hardening. The mechanical
properties were provided by the manufacture as shown in the Table 3-3 below.
Table 3-3 Epoxy properties.
Epoxy Compressive Flexural Tensile Compressive Tensile Curing

Strength Strength E-Modulus time
Strength E-Modulus
MPa MPa MPa
MPa MPa
Sika 60 28 12 3500 4500 7 days
3.4.4. Steel
Steel bars # 3 and #4 were used as flexural and shear reinforcement. The tensile strength
and modulus of elasticity were obtained by testing 3 ft. specimens according to ASTM.
The prepared specimens were provided with strain gages at the middle to obtain the
tensile stress-strain diagram. The test was performed in the MTS machine in the iSTAR
laboratory. Figures 3-16 and Figure 3-17 below are the steel stress-strain diagrams that
constructed based on the collected results.
44
Figure 3-16 #3 steel strain stress diagram.
Figure 3-17 #4 steel strain stress diagram.
45
Figure 3-18 Tensile test of steel.
From graphs above the following results have been obtained.
Table 3-4 Steel tensile properties.
𝑓𝑦 (ksi) 𝜀𝑦 (strain) Modulus of elasticity(ksi)
74ksi 0.0027 27407
3.4.5. Specimens’ Analysis and Design
ACI 318-11 and ACI 440.2R-08 were adopted to perform flexural analysis and design of
rectangular reinforced concrete beams and NSM-CFRP reinforced concrete beams. ACI
440.2R-08 has some preliminary assumptions, such as full bonding between the CFRP
rod and the concrete. The section is assumed plane before and after loading. The
46
maximum service strain in concrete is 0.003. Finally, the CFRP rod has elastic tensile
properties until the rupture.
Figure 3-19 Strain and stress distribution (ACI-440-2R-08).
As shown in figure above, strain compatibility and equilibrium condition of the cross
section will be satisfied to induce the capacity of the section.
∗
𝑓𝑓𝑢 = 𝐶𝐸 𝑓𝑓𝑢 ACI 440.2R-08 (9-3) (1)
∗
𝜀𝑓𝑢 = 𝐶𝐸 𝜀𝑓𝑢 ACI 440.2R-08 (9-4) (2)
𝜀𝑓𝑒 = 𝑘𝑒 𝜀𝑓𝑢 ACI 440.2R-08 (12-5) (3)
CE is the environmental reduction factor, which is available in table 9.1in ACI 440.2R-08
for variety of cases.
ke (0.6-0.9), it is bonding coefficient that depends on many characters its such as the
internal reinforcement, member dimensions, and the surface texture of CFRP bars.
47
An assumption was made on whether the concrete crushing or CFRP rod rupture is
controlled based on the following equations. An iterative solution was guided by
assuming a value for C, the distance from the concrete block to the neutral axis. This
assumption was checked by using the equilibrium condition as explained in equation (10-
12) of ACI 440.2R-08.
𝑑𝑓 −𝑐
𝜀𝑓𝑒 = 𝜀𝑐𝑢 ( ) ACI 440.2R-08 (10-3) when concrete crushing failure governs (4)
𝑐
𝜀𝑓𝑒 = 𝜀𝑏𝑖 + 𝜀𝑓𝑑 ACI 440.2R-08 (10-3) when CFRP rod failure governs (5)
From strain compatibility, the strain at the steel and the concrete was determined from the
following relationships.
𝑑−𝑐 (6)
𝜀𝑠 = (𝜀𝑓𝑒 + 𝜀𝑏𝑖 ) (𝑑 ) ACI 440.2R-08 (10-10)
𝑓− 𝑐
𝑓𝑓𝑒 = 𝐸𝑓 𝜀𝑓𝑒 ACI 440.2R-08 (10-9) (7)
𝑓𝑠 = 𝐸𝑠 𝜀𝑠 ≤ 𝑓𝑦 ACI 440.2R-08 (10-11) (8)
𝐴𝑠 𝑓𝑠 +𝐴𝑓 𝑓𝑓𝑒 (9)

𝑐= ACI 440.2R-08 (10-12)
𝛼1 𝑓𝑐′ 𝛽1 𝑏
After the strain compatibility and equilibrium condition were satisfied, the following
equation was used to obtain the section flexural capacity.
𝛽1 𝑐 𝛽1 𝑐 (10)
𝑀𝑛 = 𝐴𝑠 𝑓𝑠 (𝑑 − ) + 𝜓𝑓 𝐴𝑓 𝑓𝑓𝑒 (𝑑𝑓 − ) ACI 440.2R-08 (10-13)
2 2
48
Ψf = is a reduction factor of 0.85 statistically suggested to account for CFRP strength
contribution.
The procedure described above was followed to analyze the specimen’s flexural
capacity.as explained in appendix A. Tables 3-5 and 3-6 below show the strain levels, the
flexural capacity, and the theoretical mode of failure for each specimen.
Table 3-5 NSM-CFRP flexural cross section analysis for set A.
6.5 0.003 0.85fc'
a/2
C
1.5 Mn=18.86 k.ft
8.25
(d-a/2)
10.00
2 # 4 s T Faiure mode is the

0.0135
steel yeilding .
6.5 0.003 0.85fc' Mn=28.86k.ft

a/2
47.34
2.4
8.25
(d-a/2)
10.00
Faiure mode is the

2 # 4 s
concrete crushing.
29.6
0.0073
2 # 2 c 17.79
0.0092
49
Table 3-4 (cont.)
6.5 0.003 0.85fc' Mn=36k.ft

a/2
57.5
2.8
8.25
(d-a/2)
10.00
Faiure mode is the

2 # 4 s 29.6
concrete crushing.
0.0052
2 # 3 c 27.55
0.007
6.5 0.003 0.85fc' Mn= 38 k.ft

a/2
68.3
3.5
8.25
(d-a/2)
10.00
Faiure mode is the

2 # 4 s 29.6
concrete crushing.
0.0039
2 # 4 c 0.005 38.743
6.5 0.003 0.85fc' Mn= 35 k.ft

a/2
59.2
2.7
8.25
(d-a/2)
10.00
Faiure mode is the

2 # 4 s 29.6
concrete crushing.
0.0065
1 # 4 c 0.0075 29.6
50
Table 3-6 NSM-CFRP flexural cross section analysis for set B.
6.5 0.003 0.85fc'
a/2
C
0.76 Mn=11.0 k.ft
8.25
(d-a/2)
10.00
2 # 3 s T
Faiure mode is the
0.029 steel yeilding .
6.5 0.003 0.85fc'

Mn=24.34 k.ft
a/2
38.8
1.9
8.25
(d-a/2)
10.00
Faiure mode is the

concrete crushing.
2 # 3 s 16.28
0.0093
2 # 2 c 0.0116 22.5
6.5 0.003 0.85fc'

Mn=31.6k.ft
a/2
52.6
2.4
8.25
(d-a/2)
10.00
Faiure mode is the

concrete crushing.
2 # 3 s 16.28
0.0073
2 # 3 c 35.75
0.00903
6.5 0.003 0.85fc'

Mn= 36 k.ft
a/2
61
3.0
8.25
(d-a/2)
10.00
Faiure mode is the

2 # 3 s
concrete crushing.
16.34
0.00525
2 # 4 c 0.0065 41.7
51
3.4.6. Specimens Preparation
The fabrication of the beams’ steel gages and the wood molds were fabricated and made
at the South Green House. Four plastic spacers were used for each beam to provide the
required cover. The inside surface of wood molds were also finished with oil to provide a
smooth surface and to easily disassemble the forms. The figures below show the steel
cages and the forms.
Figure 3-20 Steel cage and mold fabrication.
Figure 3-21 Steel cage inside the molds with the required cover.
52
3.4.7. Concrete Casting and Curing
The casting of the concrete beams was done outside of the Hoop house. Therefore, the
fresh concrete was directly poured into the wood forms. The forms were then properly
leveled and covered with plastic cover after curing. Forty plastic cylinders were prepared
and casted to find the mechanical properties of the hardened concrete. Concrete casting
and curing are showing in the figure provided below.
Figure 3-23 Concrete casting.
53
Figure 3-24 Concrete finishing and curing.
3.4.8. Instrumentations
Steel strain gages were posted on both the flexural steel rebars and the NSM CFRP rods.
Three steel strain gages were placed on steel rebars. Steel strain gages were placed at the
pure bending moment region to capture the flexural strain. For the CFRP rods, four strain
gages were distributed at 1/6th of the development length. The strain gage of “KFH-20-
120-C1-11L1M2R” was purchased from Omega Engineering, Inc. Steel and CFRP rod’s
surface preparation were done to ensure the proper bonding of the strain gage as shown in
figures below. Three concrete strain gauges were mounted on the top of each beam to
record the compression strain at concrete mid span. The concrete strain gauges were type
PL-30-11from TML. An LVDT of 100 mm (4 in) stroke was placed at the middle of the
beam to record the deflection.

54
Figure 3-25 Strain gages preparation and posting.
55
3.4.9. Preparation of the Groove
The grooves were made by a saw with a diamond blade to form two parallel cuts, then a
manual hammer with chisel was used to finalize the grooves. The groove’s size was 2Db,
that is 0.5 in x0.5 in for #2 CFRP rod, 0.75 in x0.75 in for #3 CFRP rod, and 1.0 in x1.0
in for #4 CFRP rod. The distance of the longitudinal grooves was 1.0’’from the edge, and
the spacing between the lateral grooves was 3 in as shown in figures below. The grooves
were cleaned with an air compressor to eliminate of the concrete dust before inserting the
CFRP rods.
Figure 3-26 Making grooves in the tension side of the strengthened beams.
56
3.4.10. Inserting the CFRP Bars.
In order to post the CFRP rods, a first layer of epoxy was applied to the grooves. The first
layer’s thickness was almost half of the grooves’ depth. The CFRP rod then was pressed
in the groove. After that, a second layer of adhesive was applied at the top of the CFRP
rod. The excessive adhesive was leveled and removed as shown in figures below. The
adhesive was left for at least seven days for curing at temperature of 17.0C0.
Figure 3-27 Inserting CFRP Rods.
3.4.11. Instrumentations and Procedure
The flexural tests for beams were tested in iSTAR laboratory at Portland State University.
All beams were tested under four-point loads by using a 50-kip hydraulic cylinder that
connected to a manual pump. The test setup is shown in Figure 3-28. The specimens
57
were supported on a plate that rested on a 4-in wide bearing plate. A spreader beam of
W8x35 was attached to the hydraulic ram. The spreader beam had a flexural span of 32 in
that was supported by two plates. The specimens were tested under a monotonically
increasing load until failure. The cracks width and paths were observed, and marked at
specific load stage.
Figure 3-28 Four-point load test setup.
58
Chapter 4: The Results
4.1. Pullout Test
The pullout test system shown in Figure 3-12 was adopted in this study. Three sizes of
CFRP rods were tested: #2 (0.25 in diameter), #3 (0.375 in diameter), and #4 (0.5 in.
diameter). The development length was selected as 60 times the diameter of the CFRP
rod. The bonded length was 15 in for the #2 CFRP rod, 23 in for the #3 rod, and 30 in for
the #4 rod. The first test was done for the #2 CFRP rod. In this test two concrete cuts
were made at the concrete surface, then the chisel and a manual hammer were used to
make the grooves. The failure mode was de-bonding at the concrete-epoxy interface
where the CFRP and the epoxy separated from the concrete surfaces as one block.
Epoxy-concrete interface de-bonding was the failure mode for this specimen as it seen in
Figure 4-1. From Figure 4-2, it is apparent that the failure load was about 40% of the bar
ultimate strength.
Strain1
Strain2
Strain3
Figure 4-1 Pull out test failure without lateral grooves.
59
Load (kips)
Figure 4-2 Load-strain diagram for #2with out lateral grooves.
This low bonding capacity was insufficient since it did not utilize more than 40% of the
rod capacity. Therefore, it was decided to enhance the bonding by using lateral grooves
of 0.25 inch in width and about 1.0 inch in length. The distance between the grooves was
about 3 in. The details for the lateral grooves were adopted from Sharaky, I. A. et al.
(2013). The layout of the lateral grooves are shown in Figure 4-3 below.
60
Specimens with mechanical interlocking
(Sharaky, I. A. et al. 2013).
#2&#4 lateral grooves details
Figure 4-3 Lateral grooves details.
Using the lateral grooves, the result in bonding capacity was 100% for the #2 rods, 60%
for #3 rods, and about 50% for #4 rods. It was concluded that the lateral grooves
enhanced the bonding by providing an interlocking bond between the epoxy and the
concrete. This interlocking force allowed the bonding to resist additional forces by
transferring the load to the concrete. Figures 4-4, 4-6, and 4-7 show the strain distribution
along the CFRP rods. Strain readings were recorded by strain gages that had been posted
along the CFRP rod. Strain gage no.1 was posted at the top of the bonding length. Strain
gages no.2 and no.3 were posted at third points of the embedment length. For #2 rod and
the #3 rod CFRP pullout test, the middle strain gauges were damaged and stopped
recording after certain level of load. In general, the graph shows that the strain along the
61
CFRP is maximum at the loaded end and minimum at strain gages no. 3, which is close to
free end. As the pullout load increased, the strain increased along the CFRP rod.
Load (kips)
Figure 4-4 Strain load diagram for#2 CFRP with lateral grooves.
Figure 4-5 Pullout-test-failure for #2 with lateral grooves.
62
Figure 4-5 shows # 2 and #4 pullout failure mode. It was observed that #2 failure did not
involve any damage for the surface of the concrete before it ruptured. On the other hand,
the failure of the #4 CFRP rods caused serious damage to the surface of the concrete. The
crushing of the concrete introduced the de-bonding failure. For the #3 CFRP rod, the
failure mode was exactly the same mode of failure as the #4 CFRP rod. These failure
modes reduced the bonding capacity of the #3 and the #4 CFRP rod to 60% and 50%
respectively as shown in load-strain curves below. It is apparent from the results above
that the interlocking mechanism shifted the failure from the concrete-epoxy interface to
the rupture of the CFRP rod for the size #2 rods, and concrete crushing for the #3and the
#4 CFRP rods.
Load (kips)
As Figure 4-5 shows, the failure of the #3 and the #4 CFRP rods mainly occurred at the
loaded end. This indicates that the stress at the loaded end position is significant.
63
Load (kips)
The load-slippage curve was also constructed from the LVDT reading that is explained in
Figure 3-13. Figure 4-8 shows the load-slippage relationships for #2, 3, AND 4 CFRP
rods. It was observed that the relationship between the load-slippage was reasonably
linear for the #2 and #3 rods. However, for #4 rod, the relationship is linear until about
50% of the capacity, then the curve started to behave in a nonlinear manner due to the
concrete and epoxy deformation surrounding the CFRP rod.
64
Load (kips)
Figure 4-8 Pullout load slippage curves.
4.2. Discussing the Pullout Results.
In order to understand the distribution of the pullout stress along the CFRP rod, the
figures shown below were constructed. At each load level, the strains were multiplied by
the modulus of elasticity for each rod and positioned at the same location of the strain
gauges. A, B, and C are the locations of the strain gauges. D is the free end point where
the stress was zero.
65
A
l/3
B
l/3 36
l/3
D
Front view
A
Stress (ksi)
Figure 4-9 #2 CFRP pullout stress distribution.
It was concluded that the stress variations along the CFRP rods of the stress were uniform
for all the bar sizes. Rods #2 and #4 had some of discrepancies of strain variation at the
early load stages. Nevertheless, the strain adjusted at the ultimate load when more stress
66
was transferred to the lower portion of the development length. That indicated that the
bonding was adequate to transfer the stress along the bonded length.
Figure 4-10 #3 CFRP pullout stress distribution.
C
Stress (ksi)
Figure 4-11 # 4 CFRP pullout stress distribution.
Table 4-1 shows the summery of the pullout tests. Shear stresses at the CFRP-epoxy
interface and at epoxy-concrete interface were calculated and referred to as τr-e. and τe-c
67
respectively. These stresses were calculated based on the maximum load of the pullout
and according to ACI 440.2R-08 equation shown below:
𝑃𝑢
𝜏𝑟−𝑒 = (11)
(𝛱 𝐷 𝐿𝑑𝑏 )
𝑢 𝑃
𝜏𝑒−𝑐 = ((𝑎+2𝑏)𝐿 (12)
𝑑𝑏 )
B.R =Bar Rupture. C.C = Concrete Crushing.
CFRP 𝐿𝑑𝑏 𝑓𝑐′ Pu σu 𝜏𝑟−𝑒 𝜏𝑒−𝑐 εu Slippage Failure
rod# inches ksi kips ksi ksi ksi strain (in) mode
#2 15.0 4.5 15.0 288.7 1.2 0.63 0.0155 0.39 B.R
#3 22.5 4.0 21.0 193.5 0.8 0.42 0.01 0.41 C.C
#4 30 4.5 31.0 161.14 0.67 0.35 0.008 0.34 C.C
Table 4-1 pullout results summery.
ACI 440.2R-08 assumes a linear distribution for the shear stresses along the bonded
length, with a peak near the middle of the embedment. The max stress starts at the
loaded end or the maximum bending moment region, and assuming a uniform bond
stress, the bar stress gradually decreases until it reaches the support or the free end.
Figure 4-12 shows the assumed bond distribution according to ACI440.2R-08
Figure 4-12 Bond stress distribution (ACI 440.2R-08, p 40).
68
From the force equilibrium, ACI 440.2R-08 derived the following equation to calculate
the development length. ACI 440.2R-08 reported an average shear stress from500-3000
psi based on previous studies. However, the 1000 psi was recommended as the average
shear strength for NSM.
𝑑
𝐿𝑑𝑏 = 4(𝜏𝑏 ) 𝜎𝑏 ACI 440.2R-08 (13-3) (13)
𝑏
By comparing, the bond shear stress results from the pullout results with the ones that
were presented by ACI 440. 2R-08, the obtained results were reasonably close to the
average shear stress that was recommended by ACI 440.2R-08 as shown in figure 4-13.
Recommended ACI average stress =1.0 ksi.

Shear Stress (ksi)
Figure 4-13 Average shear stress for NSM CFRP rod.
69
4.3. Beam Test
For the flexural test, nine beams were tested under a two-point load as shown
schematically in Figure 4-14 below.
Figure 4-14 Two point load flexural test.
All beams had dimensions bxh= 6.5 x10 inches. The beams were classified into two
sets. The steel reinforcement for Set A was selected to be a “ medium-low” amount to
reflect a portion of actual beams that may need to be retrofitted. The beams in set A had
a 0.7% steel ratio, which is two #4 steel rebars. The steel reinforcement for Set B was
selected to be a “low” amount to reflect a portion of actual beams that will more likely to
be retrofitted. The beams in Set B had a 0.4% steel ratio, which is two #3 steel rebar. For
each set, the beams were additionally reinforced with two CFRP rods as described in
Table 4-2 below. The load was recorded by a load cell and the mid-span deflection was
recorded by a Linear Variable Differential Transformer (LVDT). The strains in the steel
bar, concrete, and CFRP bar were also recoded. The load was applied gradually in 3-kip
intervals. The crack propagation and crack width were inspected, marked, and recorded
manually by crack width measuring device. Strains were measured in the concrete in
compression (top surface), in the steel and in CFRP bars at two locations.
70
4.3.1. Set (A) Specimens
Table 4-2 Set A specimens results summery.
Beam Reinforcement fc’(psi) Pu(Kips) ∆u(inches) Ductility Increase in Failure
label details capacity Mode
beyond AC
AC 2#4 steel bars. 4200 17.0 2.0 3.75 - Concrete
No CFRP crushing
AR2 2#4 steel bars 4200 29.0 1.5 2.5 70% Concrete
2#2 CFRP peeling off
bars
AR3 2#4steel bars 4372 23.4 1.1 1.6 38% Concrete
2#3 CFRP peeling off
bars
AR4 2#4 steel bars 4000 27.5 0.8 0 62% CFRP de-
2#4 CFRP bonding
bars
AR4-S 2#4 steel bars 4600 26.5 1.1 1.7 56% CFRP de-
1#4CFRP bars bonding
4.3.1.1. The Control Beam (AC)
The first tested beam in set A was the control beam (AC), which had two #4 steel rebars.
According to ACI318-12 (shown in Figure 4-15), the flexural capacity of this beam is
18.9 k-ft. which corresponds to a load of 14.2 kips. Figure 4-16 shows the experimental
load-deflection relationships.
71
6.5 0.003 0.85fc'
a/2
C
1.5
8.25
(d-a/2)
10.00
2 # 4 s T
0.0135
Figure 4-15 Specimen AC strain and stress profile according to ACI.
Figure 4-16 Specimen AC load deflection-curve.
The strain profile measured at the maximum bending moment region is shown in Figure
4-17. The strain profile showed that the pure bending moment section experienced a
significant strain increase when the load increased from 15.0 kips to 17.0 kips. At 17
kips, the steel and concrete strain was about 0.015 and 0.005 respectively. After this
point, concrete crushed and the capacity decreased.
72
Figure 4-17 Specimen AC strain profile at different load level.
Cracks propagated at the pure positive bending moment regions. As the load increased,
the cracks widened. The load was then increased until the concrete crashing occurred at
about 17.0 kips as shown in Figure 4-18.
73
Figure 4-18 Failure mode of specimen AC.
4.3.1.2. Beam Retrofitted with Two #2 CFRP Rods (AR2)
The second beam tested in set A was AR2. This beam had the same properties as the AC
beam, but was additionally reinforced (retrofitted) with two #2 CFRP rods at the tension
side. According to ACI 440-2R-08, the maximum capacity of this beam is 29.0 k-ft that
leads to a load of about 21.6 kips. Strain and stress profile according to ACI is shown in
Figure 4-19.
74
6.5 0.003 0.85fc'
a/2
47.34
2.4
8.25
(d-a/2)
10.00
2 # 4 s 29.6
0.0073
2 # 2 c 17.79
0.0092
Figure 4-19 ACI strain and stress analysis of specimen AR2.
From the measured load deflection curve shown in Figure 4-20, three distinct regions
were observed: the first region was where the first crack of 0.05 mm occurred at about 3
kips. That is where the beam lost some of its flexural stiffness. The second region was
where the steel and the CFRP rods behaved elastically; and the third region was where
the steel reinforcement yielded. In this stage, more of flexural stiffness was lost and more
cracks were initiated and widened at the positions of the lateral grooves. This observation
was reported by Al-Mahmoud, F.et al. (2010) where the authors categorized the moment-
deflection curve into three stages: first zone is the elastic stage, second zone is concrete
crack to steel yielding stage, and the third Zone is steel yielding to failure stage as shown
in Figure 2-9. As the load increased, the measured cracks increased. The load was then
increased until the de-bonding occurred at about 29 kips.
75
35
30
25
First zone
Load(kips)
20
15 Third zone
10
5
second zone
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Deflection(inches)
Figure 4-20 Specimen AR2 load -deflection curve.
The increased capacity was about 70% over the control beam. Before deboning occurred,
some diagonal cracking occurred at the maximum bending moment region. These cracks
were initiated at the middle and extended to the ends of the beam.
From the CFRP, steel, and concrete’s strain gauges, the strain profile at the maximum
moment section was constructed as shown in Figure 4-21. It was observed that the strain
in the section was linear up to 20 kips. After that, the strain profile behaved nonlinearly,
especially prior to the de-bonding when the strain in the CFRP material was less than the
steel bars.
76
Figure 4-21 Section strain profile for specimen AR2 at different load levels.
This phenomenon is described in ACI 440-2R-08 as the “delamination of the concrete
layer caused by the tension failure of the concrete cover” as shown in Figure 4-22 below.
Figure 4-22 Tension layer delamination (ACI440.2R-08, p 38).
77
Figure 4-23 shows the load-strain relationship at two points along the CFRP rods. The
first strain gauge was located at the beginning of the pure moment region, directly under
the point load, and the second strain gauge was located at 16 inches away from the first
guage. The strain distribution shows that the strain was larger at the pure moment region
and increased as the load increased, as expected. The maximum strain in the first and the
second strain gauge was about 70% and 40% of the ultimate strain.
P/2 P/2
stirrups #3@4in
10.0
STRAIN#1 16.0 STRAIN#2
6.0 32.0 32.0 32.0 6.0
108.0
Load (kips)
Figure 4-23 CFRP load-strain relationship for specimen AR2.
Figure 4-24 shows the failure mode for specimen AR2.
78
Figure 4-24 Specimen AR2 failure mode.
The third beam tested in set A was AR3. This had the same properties of the AC beam,
but was additionally reinforced with two #3 CFRP rods at the tension side. According to
ACI 440-2R-08, the maximum capacity of this beam is 36.0 k-ft. that is corresponded to
27.0 kips. Stress analysis of AR3 is shown in Figure 4-25.

6.5 0.003 0.85fc'
a/2
57.5
2.8
8.25
(d-a/2)
10.00
2 # 4 s 29.6
0.0052
2 # 3 c 27.55
0.007
Figure 4-25 stress and strain analysis of specimen AR3
79
In general, the same three zones were exhibited in the load-deflection curve as shown in
Figure 4-26. The first region was where the first cracking of 0.05 mm occurred at 3.44
kips. The second region was at about 18 kips. As the load increased, the selected cracks
increased as shown in Figure 4-27. The load was then increased until the de-bonding
occurred at about 24 kips, and that was where the third region of the curve was shaped.
The increased capacity was about 38% over the control beam.
Figure 4-26 Load deflection curve for specimenAR3.
Figure 4-27 shows the failure mode of specimen AR3. The de-bonding in this specimen
occurred in two stages. The first stage occurred when one of the CFRP rods de-bonded at
the middle. That is where the load dropped from 24.0 kips to 17.0 kips. The second de-
bonding occurred when the load was increased up to 19.0 kips after that the load dropped
to 10.0 kips. The expected capacity of this specimen was more than AR2. However, some
80
bonding issues in one of the CFRP rod caused the first de-bonding, and that reduced the
capacity of the specimen
First de-bonding Second de-bonding
Figure 4-27 Failure mode of specimen AR3.
From the CFRP, steel, and concrete’s strain gauges, the strain profile was constructed as
shown in Figure 4.28 below. It was observed that the strain in the section was linear up to
81
15.0 kips. After that point the strain profile behaved nonlinearly, especially prior to the
de-bonding where the strain in the CFRP was less than the steel.
Figure 4-28 Strain profile for specimen AR3 at different load level.
first strain gauge was located at the start of the pure moment region under the point load,
and the second strain gauge was located at 16 inches away from the first gauge. As
expected, the strain distribution shows that the strain was larger at the pure moment
region and increased as the load increased. The maximum strain in the first and the
second strain gauges were about 30% and 20% of the ultimate strain. These low values
indicate the deficiency of bonding.
82
Load (kips)
The fourth tested beam in set A was AR4. This had the same properties as the AC beam,
but was additionally reinforced with two #4 CFRP rods at the tension side. According to
ACI 440.2R-08, shown in Figure 4-30, the maximum capacity of this beam was 38.0 k-ft.
that resulted in a load of about 8.5 kips.
6.5 0.003 0.85fc'
a/2
68.3
3.5
8.25
(d-a/2)
10.00
2 # 4 s 29.6
0.0039
2 # 4 c 0.005 38.743
Figure 4-30 Strain and stress analysis of specimen AR4.
83
Figure 4-31 shows the load deflection curve and two distinct regions can be observed.
The first region is where the first cracking of 0.04 mm occurred at about 5 kips. That is
where the beam lost some of its flexural stiffness as shown in the load deflection curve.
The load was increased until the de-bonding occurred at about 28 kips, and that is where
the second region of the curve was exhibited by a significant drop in the load. The
increased capacity over the control beam was about 64%. However, this specimen had a
brittle behavior, which indicated the over reinforcement of the system.

Load (kips)
Figure 4-31 Load deflection curve of specimen AR4.
Before the de-bonding occurred, some diagonal cracking occurred in the shear spans.
These cracks initiated at the beginning of the shear span and extended to the ends of the
beam. The de-bonding was due to the over stress of the cover layers where the edge
concrete cracked and spalled, initiating the de-bonding as shown in Figure 4-32.
84
Figure 4-32 Failure mode of specimen AR4.
shown in Figure4-33 below. It can be observed that the strain in the maximum bending
moment section is reasonably linear up to the failure and the steel reinforcement did not
yield.
Figure 4-33 Strain profile of specimenAR4 at different load levels.
85
first strain gauge was located at the start of the pure moment region, under the point load,
and the second strain gauge was located at 16 inches away from the first one. The strain
distribution showed that the strain was larger at the pure moment region and increased as
the load increased, as expected. The maximum CFRP strain before de-bonding was
about 20% of the ultimate strain. That indicates the poor performance of the composite
action.
Load (kips)
86
4.3.1.5. Beam Retrofitted with One #4 CFRP Rod (AR4-S)
The fifth beam tested in set A was AR4-S. This specimen had the same properties as the
AC beam, but was additionally reinforced with a single #4 CFRP rod on the tension side.
The purpose of testing this specimen was to minimize the stress at the bonding area by
increasing the edge distance According to ACI 440.2R-08, the maximum capacity of this
beam is about 35 k-ft. leading to a load of about 27kips. Figure 4-35 shows the strain and
stress analysis of the specimen according to ACI 440.2R-08.
6.5 0.003 0.85fc'
a/2
59.2
2.7
8.25
(d-a/2)
10.00
2 # 4 s 29.6
0.0065
1 # 4 c 0.0075 29.6
Figure 4-35 Strain and stress analysis of specimen AR4-S.
Figure 4-36 shows the load deflection curve, with three distinct regions. As the load
increased, the selected cracks increased. The load was then increased until the de-bonding
occurred at about 26 kips.
87
Figure 4-36 Load deflection curve for specimen AR4-S.
The increased capacity was about 52% over the control beam. Before the de-bonding
occurred, some diagonal cracking occurred at the maximum bending moment region.
These cracks initiated at the middle, and did not extend to the ends of the beam. The de-
bonding was due to the over stress of the cover layer where the edge concrete cracked
and spalled, initiating the de-bonding. Figure 4-37 shows the failure mode of specimen
AR4-S.
88
Figure 4-37 Failure mode of specimen AR4-S.
shown in Figure 4-38 below. It can be observed that the strain in the concrete, steel and
CFRP bars in the section are compatible, i.e., forming reasonable straight lines up to the
failure.
89
Figure 4-38 Strain profile for specimen AR4-S at different load levels.
first strain gauge was located at the edge of pure moment region (under the concentrated
load), and the second strain gauge was located at 16 inches away from the first gauge.
The strain distribution shows that the strain was larger at the pure moment region, and
increased as the load increased, as expected. The maximum strain in the first and the
second strain gauges was about 31% and 18% of the ultimate strain. This indicated that
the maximum strain for the #4 CFRP was about 31% before the de-bonding occurred.
This value does not agree with the pullout test result where the maximum strain in
#4CFRP rod was about 50% of the ultimate strain. This difference could be because of
the flexural action and flexural cracking that raises the bond stress, the so called in-and –
out bond stress that occurs between flexural cracks.
90
Load (kips)
Figure 4-39 CFRP load-strain relationship for specimen AR4-S.
In ACI 440.2R-08, there is guidance for the distribution of the CFRP rod in tensile layer.
The edge distance should be four times the depth of the grooves and the spacing between
the bars should be at least two times the depth of the groove. For all specimens, the
minimum spacing was 2.5 inches and the edge distance was1.0 inch with the exception of
specimen AR4-S where the edge distance was 3in. The purpose of testing this specimen
was to minimize the stress at the bonding area by increasing the edge distance. However,
specimens AR2 and AR3 showed better performance in composite action than AR4-S.
91
4.3.2. Set (B) specimens
Table 4-3 Set B results summery.
Beam Reinforcement fc’(psi) Pu(Kips) ∆u(inches) Ductility Increase in Failure
label details capacity Mode
beyond BC.
BC 2#3steel bars 4560 11.6 2.5 5 - Concrete
No CFRP crushing
BR2 2#3steel bars 4348 18.0 1.8 3.5 60% CFRP
2#2 CFRP de-
bars bonding
BR3 2#3steel bars 4676 25.4 1.37 2.3 130% CFRP
2#3 CFRP de-
bars bonding
BR4 2#3 steel bars 4400 25.6 1.12 0 130% Concrete
2#4 CFRP peeling
bars off
4.3.2.1. Control Beam (BC)
The first beam tested in set B was the control beam, which had two #3 steel bars in the
tension side. As shown in figure 4-40 according to ACI318-12, the flexural capacity of
this beam is about 11 k-ft, which results in a load of 8.25 kips.
92
6.5 0.003 0.85fc'
a/2
C
0.76
8.25
(d-a/2)
10.00
2 # 3 s T
0.029
Figure 4-40 ACI 318 strain and strain profile for specimen BC.
Figure 4-41 shows the measured load-deflection relationship. The first observed cracking
was 0.2 mm at 3 kips. The load deflection curve showed that the reduced stiffness begins
at about 2 kips. That was where the beam lost a significant amount of its flexural
stiffness. It appears that steel yielding happened at about 8 kips, which is reasonably
close to the nominal capacity computed according to ACI code.
Figure 4-41 Load deflection curve of specimen BC.
93
Cracks propagated at the pure positive bending moment regions. As the load increased,
the selected measured cracks widened. The load was then increased until the concrete
crashing occurred at about 11.3 kips. The measured load capacity of the beam was about
37% higher than computed. Figure 4-42 shows the cracks pattern and the failure mode of
specimen BC.
Figure 4-42 Failure mode of specimen BC.
The strain profile measured at the maximum bending moment region is shown in Figure
4-43.The strain profile showed that the pure bending section experienced a significant
strain increase when the load was raised from 9.0 kips to 11.0 kips. At 11.0 kips, the steel
and concrete strains were about 0.02 and 0.0033 respectively. After this point, concrete
crushed and the load dropped.
94
Figure 4-43 Strain profile for specimen BC at different load levels.
4.3.2.2. Beam Retrofitted with Two #2 CFRP Rods (BR2)
The second beam tested in set B was BR2. This beam had the same properties as the AC
beam, and was additionally reinforced with two #2 CFRP rods at the tension side.
According to ACI 440-2R-08, Figure 4-44 shows the maximum flexural capacity of this
beam is about 24.3 k-ft, which leads to a load value of 18 kips.
6.5 0.003 0.85fc'
a/2
38.8
1.9
8.25
(d-a/2)
10.00
2 # 3 s 16.28
0.0093
2 # 2 c 0.0116 22.5
Figure 4-44 Strain and stress analysis of specimen BR2.
95
As shown in Figure 4-45, three distinct regions can be observed: the first region was
where the first crack of 0.05 mm occurred at about 2 kips. That was when the beam lost
some of its flexural stiffness; the second region was where the steel and the CFRP rods
behave elastically; and the third region was where the steel reinforcement yielded. In this
stage, more of the flexural stiffness was lost, and more cracks were initiated and widened
at the positions of the lateral grooves. As the load increased, the selected measured
cracks increased. The load was then increased until the de-bonding occurred at about 18
kips. which is the same value as given by ACI 440.2R-08 design formula. The increased
capacity was about 55% over the control beam. Figure 4-46 shows the specimen after
reaching failure.
Figure 4-45 Load deflection curve of specimen BC.
96
Figure 4-46 Failure mode of specimen BR2.
shown in Figure 4-47. There appears to be some incongruity between the steel and the
CFRP strains for the loads at 5 and 10 kips values. The variation of the strain profile
from the linear strain distribution seems excessive. However, the section strain
distribution at higher loads (15 and 18 kips) are as expected.
Figure 4-47 Strain profile for specimen BR2 at different load levels.
97
Figure 4-48 shows the load strain relationship for the two gauges in a CFRP rod. The first
strain gauge was located at the pure moment region under the concentrated load, and the
second strain gauge was located at 16in away from the first gauge. The load- strain
diagram shows that the strain was larger at the pure moment region, and increased more
rapidly as the load increased, as expected. It can be observed that the maximum strain in
CFRP rod located in the pure moment region was about 0.014 which is close to the
rupture strain of the CFRP rods. That means that the bonding of the CFRP rods in this
system was utilized reasonably well. The strain in the second location reached about 50%
of the ultimate strain. This is as expected since the moment at the second location was
smaller than (about half as much) the moment at the first location. The compatibility of
the steel and CFRP strain distribution at higher loads, shown in figure 4-47 indicates that
the section had good composite behavior.

Load (kips)
Figure 4-48 CFRP load-strain relationship for specimen BR2.
98
4.3.2.3. Beam Retrofitted with Two #3 CFRP Bars (BR3)
The third beam tested in set B was BR3. This specimen had the same properties as the
BC beam, and was additionally reinforced with two #3 CFRP rods at the tension side.
According to ACI 440-2R-08, Figure4-49 shows that the maximum capacity of this
beam was 31.6 k-ft. which corresponds to a load of about 23.7 kips.
6.5 0.003 0.85fc'
a/2
52.6
2.4
8.25
(d-a/2)
10.00
2 # 3 s 16.28
0.0073
2 # 3 c 35.75
0.00903
From the load deflection curve shown in Figure 4-50, three distinct regions can be
observed: the first region was when the first crack of 0.05 mm occurred at a load of
about 3 kips. That was when the beam lost some of its flexural stiffness; the second
region was where the steel and the CFRP rods behave elastically; and the third region
was where the steel reinforcement yielded. In this stage, more of flexural stiffness was
lost, and more cracks were initiated and widened at the positions of the lateral grooves.
99
Figure 4-50 Load deflection curve of specimen BR3.
As the load increased, the selected crack widths increased. The load was then increased
until the de-bonding occurred at about 25 kips. The increased capacity was about 119%
over the control beam. Before the de-bonding occurred, some diagonal cracking occurred
in the shear span. These cracks initiated at the positions of the lateral grooves, and
extended to the middle sections of the beam. The de-bonding was due to the over stress
of the beam’s tensile layer where the edge concrete cracked and spalled, initiating the de-
bonding. Figure 4-51 shows the failure mode of the specimen.
100
Initiation of the lateral grooves that caused the de-bonding.
shown in Figure 4-52. It can be observed that the strain in the section was linear up to 15
kips. At 16 kips, the steel strain gauges at the maximum bending moment region stopped
recording data. They might have been damaged during testing or had some other
technical issues in data acquisition. Therefore, the data for strain profile up to 15 kips
only are available.

101
Load (kips)
first strain gauge was located at the pure moment region, and the second strain was
located at 16 in away from the pure moment region. The strain distribution showed that
the strain was larger at the pure moment region, and increased as the load increased, as
expected. The maximum strain in the first and the second strain gauge was about 36%
and 30% of the FRP ultimate strain respectively. This shows that the variation of the
stress at the maximum moment region, and 16 inches away from the maximum moment
region, was small.
102
4.3.2.4. Beam Retrofitted with Two #4 CFRP (BR4)
The fourth beam tested in set B was BR4. This had the same properties as the BC beam,
and was additionally retrofitted with two #4 CFRP rods on the tension side. According to
ACI 440-2R-08(Figure 4-54 below), the maximum capacity of this beam was about 36 k-
ft, which corresponds to a load of 27.0 kips.
6.5 0.003 0.85fc'
a/2
61
3.0
8.25
(d-a/2)
10.00
2 # 3 s 16.34
0.00525
2 # 4 c 0.0065 41.7
103
From the load deflection curve shown in Figure 4-55, two distinct regions can be
observed. The first region was when the first cracking of 0.04 mm occurred at 5 kips.
That was when the beam lost some of its flexural stiffness as shown in load deflection
curve below. The load was then increased until the de-bonding occurred at about 26 kips,
and that was where the second region of the curve was shaped. The increased capacity
was about 121% over the control beam. However, this specimen had a brittle behavior.
That indicated the over reinforcement of the system.

Load (kips)
Figure 4-55 Load deflection curve of specimen BR4.
Before the de-bonding occurred, some diagonal cracking occurred in the shear spans.
These cracks initiated at the beginning of the shear span, and extended to the ends of the
beam. The de-bonding was due to the over stress of the cover layers where the edge
concrete cracked and fell off, initiating the de-bonding. Figure 4-56 shows the failure
mode of the specimen.

104
From the CFRP, steel, and concrete’s strain gauges, the strain profile was constructed for
the section under the concentrated load as shown in Figure 4-57 below. It can be
observed that the strain in the section was linear up to the failure, and that the steel
reinforcement did not yield.
105
Load (kips)
first strain was located at the pure moment region, and the second strain was located at
16in away from the pure moment region. The strain distribution shows that the strain was
larger at the pure moment region, and increased as the load increased. The maximum
strains in the first and the second strain gauges was about 24% and 15% of the CFRP
ultimate strain respectively. These low strain values indicate the poor performance of the
composite action due to the over reinforcement of the system.
106
Load (kips)
107
4.3.3. Set A and B Results Comparison
Figure 4-59 shows the measured load-deflection curves for beam specimens in set A.
Figure 4-59 Set A load deflection curves.
Figure 4-60 shows the normalized load-deflection curves for beam specimens in set A.
Figure 4-60 Set A normalized load deflection curves.
Figure 4-61shows the measured load-deflection curves for beam specimens in set B.
108
Figure 4-61 Set B load deflection curves.
Figure 4-62 shows the normalized load-deflection curves for beam specimens in set B.
Figure 4-62 Set B normalized load deflection curves.
109
From the data in the figures above, it is apparent that adding CFRP rods increased the
flexural capacity and the stiffness of the retrofitted beams. Another interesting fact was
that as the sizes of CFRP rod increased, the ductility decreased.
In comparing set A and set B results, it is observed that adding CFRP rods to set B
specimens works more efficiently than adding the CFRP bar sizes to set A specimens.
Soliman, S. M. et al. (2010) reported that 0.4 % is the optimum steel reinforcement ratio
to increase the capacity by adding the CFRP rods. Tables 4-4 and 4-5 shown below
reveal that the optimum reinforcement ratio for this section was within the range of 0.9-
0.6%. A reinforcement ratio larger than 0.9% would result in prematurely de-bonding
CFRP rods with sharp drop in load. On the other hand, a reinforcement ratio smaller than
0.6% would result in smaller added capacity. As the reinforcement ratio ranged
between (0.9-0.6%), the maximum capacity of the beam ranged between 18.0-29.0
kips. However, this value was obtained at a different ultimate deflection levels. For
example, AR2 had a maximum capacity of 28.7 kips and deflection of 1.5 inches. On the
other hand, BR2 had a maximum capacity of 18.0 kips at ultimate deflection of 1.76
inches. Therefore, the deflection in AR2 was compromised over the capacity, and the
capacity in BR2 was compromised over the deflection. From tables 4-4 and 4-5, we can
see that the ultimate CFRP strain in set B was larger than the corresponding ones in set A.
Therefore, he composite mechanism in set B beams was more efficient than set A beams.
110
Table 4-4 Set A results comparison.
Specimens Reinforcement Ultimate Normalized Ultimate Ultimate
No. ratio capacity Ultimate deflection(in) CFRP
(kips) Capacity strain
AC 0.73% 17.0 1.0 1.93 ----
AR2 0.92% 28.7 1.7 1.5 0.011
AR3 1.1% 23.0 1.35 1.0 0.006
AR4-S 1.12% 26.5 1.55 1.1 0.0057
AR4 1.4% 27.5 1.6 0.78 0.003
Table 4-5 Set B results comparison.
Specimens Reinforcement Ultimate Normalized Ultimate Ultimate
No. ratio capacity Ultimate deflection(in) CFRP
(kips) Capacity strain
BC 0.4% 11.0 1.0 2.5 ----
BR2 0.6% 18.0 1.6 1.76 0.014
BR3 0.82% 25.0 2.3 1.37 0.0065
BR4 1.2% 25.6 2.33 1.128 0.0044
111
4.3.4. Measured Crack Width
Figures 4-63 and 4-64 show the normalized measured load crack width curves for
specimens in set A and set b respectively.
Figure 4-63 Set A normalized load crack-width curve.
It is apparent from the Figures that adding the CFRP reinforcement limited the crack
width to less than 1.0 mm as shown in the graphs 4-63 and 4-64. For the retrofitted
beams, the cracks were distributed along the entire length of the beams specifically at the
position of the lateral grooves. However, for the control beams, the cracks were
concentrated at the region of the pure bending moment.
112
Figure 4-64 Set B Normalized load crack-width curves.
4.3.5. Theoretical Analysis
An analytical study was done and compared to the experimental results. The analysis was
based on the developing moment curvature relationship for reinforced concrete beam
section. The load deflection curve was then established from the moment curvature
relationship. A Mathcad code was programmed by Anas Yosefani, a graduate student at
Portland State University. This program allows the user to input material and section
properties. By incrementally increasing the strain in the concrete, the program uses the
equilibrium and strain compatibility to find the moment and the curvature of the section
at any strain level. Deflection was then derived from the curvature by moment - area
method. For each specimen, the load - deflection curve was established based on three
points: concrete cracking point, steel yielding point, and concrete crushing point. As
described in appendix B, the steel yielding point and the concrete crushing point were
found based on the properties of each section.
113
Table 4-6 shows ultimate loads for each specimen based on a) ACI prediction, b)
experimental values, c) theoretical value assuming𝜀𝑐𝑢 = 0.003, and d) theoretical value
assuming 𝜀𝑐𝑢 = 0.004. the ratio of ACI/ Experimental and theoretical(0.003)/
experimental are also shown.
Table 4-6 Ultimate loads comparison.
Ultimate load ( kips) Ratio
Specimen ACI Experimental Theoretical Theoretical 𝐴𝐶𝐼 0.003 𝑇ℎ𝑒𝑜.

𝐸𝑋𝑃. 𝐸𝑋𝑃.
Prediction 𝜀𝑐𝑢 𝜀𝑐𝑢
= 0.003 = 0.004
AC 14.14 17.0 15.73 16.8 0.83 0.92
AR2 21.64 29.0 26.2 30.56 0.75 0.9
AR3 27.0 23.4 32 36.15 1.2 1.4
AR4 28.5 27.5 37.4 42.2 1.0 1.4
AR4-S 26.6 26.5 30.4 34.42 1.0 1.1
BC 8.25 11.6 9.8 10.395 0.7 0.85
BR2 18.0 18.0 23.0 23.9 1.0 1.3
BR3 23.7 25.4 29.5 35.43 0.9 1.2
BR4 27.0 25.6 40.725 35.5 1.1 1.6
From table 4-6, it can be observed that ACI prediction and the experimental results are in
reasonable agreement. On the other hand, the ultimate loads according to the theoretical
analysis seemed to be generally higher than the experimental results especially for
specimens with high CFRP reinforcement ratio. Nonetheless, the bonding and epoxy
deformation were not considered in the theoretical analysis which result in the higher
114
values. Figures 4-65 through 4-68 demonstrate the theoretical performance of each
specimen, and the locations of the experimental de-bonding points. It can be concluded
from the figures that the de-bonding at set A specimens occurred between 25-30 kips. On
the other hand, set B specimens had a slightly larger range where the de-bonding
occurred between 18-24 kips.
De-bonding point
De-bonding
Figure 4-65 Theoretical load deflection curves for set A.
De-bonding point
Figure 4-66 Normalized theoretical load deflection curves for set A
115
De-bonding point
Figure 4-67 Theoretical Load deflection curves for set B.
De-bonding point
Figure 4-68 Normalized load deflection curves for set B.
116
Figures 4-69 through 4-77 below are the theoretical and experimental load- deflection
curve comparison for all specimens. In general, the theoretical load deflection curves
were comparable to the experimental results as shown in figures below. For all
retrofitted specimens, the experimental mode of failure was the de-bonding of the CFRP
rods. However, for the theoretical specimens, the mode of failure was the crushing of
concrete. Therefore, the theoretical analysis always reached a larger ultimate capacity
than the experimental results. The derivation of the theoretical curves are described in
appendix B.
Load (kips)
Figure 4-69 Specimen AC theoretical and experimental load deflection comparison.
117
Load (kips)
Figure 4-70 Specimen AR2 theoretical and experimental load deflection comparison.
Load (kips)
Figure 4-71 Specimen AR3theoretical and experimental load deflection comparison.
118
Figure 4-72 Specimen AR4-S theoretical and experimental load deflection curves.
Load (kips)
Figure 4-73 Specimen AR4 theoretical and experimental load deflection comparison.
119
Figure 4-74 Specimen BC theoretical and experimental load deflection comparison.
Figure 4-75 Specimen BR2 theoretical and experimental load deflection comparison.
120
Figure 4-76 Specimen BR3Theoretical and experimental load deflection comparison.
Load (kips)
Figure 4-77 Specimen BR4 Theoretical and experimental load deflection.
121
Chapter 5: Summery Conclusions and Recommendations
5.1. Summery
Pullout tests of the #2, #3, and #4 CFRP rods were carried out to investigate the bonding
characteristics of CFRP rods. Nine beams with different steel reinforcement ratio were
retrofitted with different sizes of CFRP rods. Each beam was tested in flexure under two-
point load. Analytical computations were carried out for each specimens and compared to
the experimental results. From the obtained results the following conclusion can be
drawn:
5.2. Conclusions
 NSM CFRP application increased the flexural capacity and the stiffness of tested
specimens.
 The failure mode of retrofitted specimen was mainly the de-bonding of the CFRP
materials and peeling of the concrete cover.
 Retrofitting the specimens that had a low steel ratio appeared to be more effective
than retrofitting specimens that had a high steel ratio.
 Bonding length of 60Db appeared to perform an excellent composite action for the
#2 CFRP rod, and a reasonable composite action for the #3 CFRP rod.
 The measured maximum bond shear stress of the pullout tests appeared to be
within the range as specified by the ACI440.2R-08.
122
 ACI 440.2R-08 reasonably predicted the flexural capacity of the NSM CFRP
beams.
 The lateral grooves played a significant role in the composite action by enhancing
the bond between the concrete and the epoxy.
 Since the bonding characteristic was not involved in the analytical calculations,
the theoretical load deflection curves were calculated based on the moment
curvature method and appeared stiffer than the experimental results.
5.3. Recommendations
Based on the experimental results, the following recommendations are made.
 Investigate the influence of the fatigue and long-term loading on beams retrofitted
with NSM-CFRP.
 Create a reliable analytical FEM model to predict the flexural performance of
beams retrofitted with NSM-CFRP.
 Further investigate the bonding behavior of CFRP rods especially size # 4.
 Investigate the performance of the NSM-CFRP applications on beams that have a
deteriorated cover.
 Investigate adding CFRP fabric to the NSM-CFRP beams to limit the de-bonding
due to concrete spalling off and delamination.
123
References
Al-Mahmoud, F., Castel, A., François, R., & Tourneur, C. (2010). RC beams
strengthened with NSM CFRP rods and modeling of peeling-off failure–Composite
structures 92.
American concrete institute (2008). ACI 440.2R -08:Guid for the Design and
Construction of Externally Bonded FRP Systems for Strengthening Concrete Structures.
ACI Manual of Concrete Practice, Farmington Hills, Michigan.
American concrete institute (2011). ACI 318 -11 : Building Code Requirements for
Structural Concrete and Commentary. ACI Manual of Concrete Practice, Farmington
Hills, Michigan.
ASCE ( American Society of Civil Engineering), 2013a, ASCE report card for America’s
infrastructure.http:// www.infrastructurereportcard.org/ (accessed on March 29,2013).
ASTM D7205 / D7205M - Standard Test Method for Tensile Properties of Fiber
Reinforced Polymer Matrix Composite Bars. American Society of Testing and Materials:
West Conshohocken, PA.2006, 12.
Bilotta, A., Ceroni, F., Di Ludovico, M., Nigro, E., Pecce, M., & Manfredi, G. (2011).
Bond efficiency of EBR and NSM FRP systems for strengthening concrete members.
Journal of Composites for Construction, 15(5), 757-772.
124
De Lorenzis, L., & Nanni, A. (2002). Bond between near-surface mounted fiber-
reinforced polymer rods and concrete in structural strengthening. ACI structural Journal,
99(2).
De Lorenzis, L., & Teng, J. G. (2007). Near-surface mounted FRP reinforcement: An
emerging technique for strengthening structures. Composites Part B: Engineering, 38(2),
119-143.
De Lorenzis, L., Rizzo, A., & La Tegola, A. (2002). A modified pull-out test for bond of
near-surface mounted FRP rods in concrete. Composites Part B: Engineering, 33(8), 589-
603.
Hassan, T. K., & Rizkalla, S. H. (2004). Bond mechanism of near-surface-mounted fiber-
reinforced polymer bars for flexural strengthening of concrete structures. ACI Structural
Journal, 101(6).
Jung, W. T., Park, Y. H., Park, J. S., Kang, J. Y., & You, Y. J. (2005). Experimental
investigation on flexural behavior of RC beams strengthened by NSM CFRP
reinforcements. ACI Special Publication, 230.
Kent, D.C., and Park, R. (1971). "Flexural members with confined concrete." Journal of
the Structural Division, Proc. of the American Society of Civil Engineers, 97(ST7), 1969-
1990
Micelli, F., & De Lorenzis, L. (2012). Near surface mounted flexural strengthening of
reinforced concrete beams with low concrete strength. Proceedings of the ICE-
Construction Materials, 166(5), 295-303.

125
Novidis, D. G., & Pantazopoulou, S. J. (2008). Bond tests of short NSM-FRP and steel
bar anchorages. Journal of Composites for Construction, 12(3), 323-333.
Parretti, R., & Nanni, A. (2004). Strengthening of RC members using near-surface
mounted FRP composites: Design overview. Advances in Structural Engineering, 7(6),
469-483.
Sharaky, I. A., Torres, L., Baena, M., & Vilanova, I. (2013). Effect of different material
and construction details on the bond behaviour of NSM FRP bars in concrete.
Construction and Building Materials, 38, 890-902.
Soliman, S. M., El-Salakawy, E., & Benmokrane, B. (2010). Bond performance of near-
surface-mounted FRP bars. Journal of Composites for Construction, 15(1), 103-111.
Soliman, S. M., El-Salakawy, E., & Benmokrane, B. (2010). Flexural behaviour of
concrete beams strengthened with near surface mounted fibre reinforced polymer bars.
Canadian Journal of Civil Engineering, 37(10), 1371-1382.
Täljsten, B., Carolin, A., & Nordin, H. (2003). Concrete structures strengthened with near
surface mounted reinforcement of CFRP. Advances in structural engineering, 6(3), 201-
213.
Tang, W. C., Balendran, R. V., Nadeem, A., & Leung, H. Y. (2006). Flexural
strengthening of reinforced lightweight polystyrene aggregate concrete beams with near-
surface mounted GFRP bars. Building and environment, 41(10), 1381-1393.
126
Appendix A
Flexural Design Example of NSM_CFRP Rods based on ACI440.2R-08
Concrete properties: 𝑓𝑐′ = 4.676 𝑘𝑠𝑖 (normal weight concrete)
Steel reinforcement properties: 𝑓𝑦 = 74.0 𝑘𝑠𝑖 𝑎𝑛𝑑𝜀𝑐 = 0.0027
∗
Carbon fire rods properties:𝐸𝑓 = 18000 𝑘𝑠𝑖, 𝜀𝑓𝑢 =0.0175
It is required to find the flexural capacity of the section shown in figure below.
6.5
d=8.25
df=9.625
10.0
2 # 3 s
2 # 3 c
The environmental reduction factor of 0.9 shall be apply based on table 9.1 in
ACI440.2R-08.
∗
𝑓𝑓𝑢 = 𝐶𝐸 𝑓𝑓𝑢 ACI 440.2R-08 (9-3)
𝑓𝑓𝑢 = 0.9(0.0175) = 0.01575 𝑘𝑠𝑖.
ACI 440.2R-08 (9-3)

127
Bond dependent coefficient ke shall be apply to limit the strain in the FRP reinforcement.
A value of 0.7 will be apply as bond coefficient factor.
𝜀𝑓𝑒 = 𝑘𝑒 𝜀𝑓𝑢 ACI 440.2R-08 (12-5)
𝜀𝑓𝑒 = 0.7(0.01575) = 0.011
ACI 440.2R-08
The section was assumed to be controlled by concrete(12-5)
crushing failure. The strain in the
FRP and the steel was calculated as following
𝑑𝑓 −𝑐
𝜀𝑓𝑒 = 𝜀𝑐𝑢 ( ) ACI 440.2R-08 (10-3)
𝑐
The value of C was found after an iterative trials. The value of 2.4 was found to satisfy
the equation of equilibrium and

ACIcompatibility conditions as explain in the following
440.2R-08 (10-3)
steps.
9.625−2.4
𝜀𝑓𝑒 = 0.003 ( ) = 0.009 < 0.01575. 𝑂. 𝑘 𝑛𝑜 𝐶𝐹𝑅𝑃 𝑟𝑜𝑑 𝑓𝑎𝑖𝑙𝑢𝑟𝑒.
2.4
𝑑−𝑐
𝜀𝑠 = (𝜀𝑓𝑒 + 𝜀𝑏𝑖 ) (𝑑 ) ACI 440.2R-08 (10-10)
𝑓− 𝑐
Since the section is not under load, the pre-existing strain was assumed zero.
8.25−2.4
ACI 440.2R-08 (10-10)
𝜀𝑓𝑒 = 0.009 (9.625−2.4) = 0.0073 > 0.0027. 𝑡ℎ𝑒 𝑠𝑡𝑒𝑒𝑙 𝑟𝑒𝑖𝑛𝑓𝑜𝑟𝑐𝑒𝑚𝑒𝑛𝑡 ℎ𝑎𝑠 𝑦𝑒𝑖𝑙𝑑𝑒𝑑 .
ACI 440.2R-08 (10-9)

𝑓𝑓𝑒 = 𝐸𝑓 𝜀𝑓𝑒
𝑓𝑓𝑒 = 18000(0.009) = 162 𝑘𝑠𝑖
ACI 440.2R-08 (10-9)

𝑓𝑠 = 𝑓𝑦 = 74 𝑘𝑠𝑖
128
𝐴𝑠 𝑓𝑠 +𝐴𝑓 𝑓𝑓𝑒
𝑐= ACI 440.2R-08 (10-12)
𝛼1 𝑓𝑐′ 𝛽1 𝑏
𝛼1 𝑎𝑛𝑑 𝛽1 𝑎𝑟𝑒 𝑡ℎ𝑒 𝑐𝑜𝑐𝑛𝑟𝑒𝑡𝑒 𝑠𝑡𝑟𝑒𝑠𝑠 𝑏𝑙𝑜𝑐𝑘 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑎𝑛𝑑 𝑐𝑎𝑛 𝑏𝑒 𝑡𝑎𝑘𝑒𝑛 𝑎𝑠 0.85
ACI 440.2R-08 (10-12)

𝑓𝑜𝑟 𝑐𝑜𝑐𝑛𝑐𝑟𝑡𝑒 𝑐𝑟𝑢𝑠ℎ𝑖𝑛𝑔 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 𝑚𝑜𝑑𝑒.
0.22(74)+0.22(162)
𝑐 = 0.85(4.67)(0.85)(6.6) = 2.36~2.4 𝑖𝑛. ACI 440.2R-08 (10-13)
𝛽1 𝑐 𝛽1 𝑐
𝑀𝑛 = 𝐴𝑠 𝑓𝑠 (𝑑 − ) + 𝜓𝑓 𝐴𝑓 𝑓𝑓𝑒 (𝑑𝑓 − )
2 2
ACI 440.2R-08 (10-13)

0.85(2.4) 0.85(2.4)
𝑀𝑛 = 0.22(74) (8.28 − ) + 0.85(0.22)(162) (9.625 − )
2 2
𝑀𝑛 = 117.7 + 260.7 = 378.4 k.in =31.53 k.ft
6.5 0.003 0.85fc'
a/2
52.6
2.4
8.25
(d-a/2)
10.00
2 # 3 s 16.28
0.0073
2 # 3 c 35.75
0.00903
129
Appendix B
Theoretical Load Deflection Relationship
The analytical program was done to verify the results. This program was based on the
developing moment curvature relationship for the reinforced concrete beam section. The
load deflection curve was then established from the moment curvature relationship,
which was done by following the procedure that described in the lecture notes of Dr.
Caesar Abi Shadid (2008). A Mathcad code was programmed by Anas Yosefani, a
graduate student at Portland State University. This program allows the user to input
material and section properties. By incrementally increasing the strain in the concrete, the
program uses the equilibrium and strain compatibility to find the flexural capacity and the
curvature of the section at any strain level.
Strain and stress properties of the materials were input as formula to account for the
nonlinear behavior. As shown in figure B-1, the steel strain diagram has three distinct
regions.
For εs ≤ 0.0027 (elastic region)
fs= Eεs (14)
For εy ≤ εs ≤0.012 (perfectly plastic region)
For 0.012 ≤ εs ≤ 0.05 (strain hardening region)
130
Regression analysis was done to find the formula that best defined the strain-hardening
trend, which turned out to be as follows
𝑓𝑠 = 22.79 ln(𝜀𝑠 ) + 175.88 (15)
The steel properties were defined for both the top (As’) and bottom reinforcement (As)
Stress (ksi)
Figure B-0-1 steel strain-stress curve.
For concrete material properties, a compression strain-stress diagram was defined by

Figure B-0-2steel strain-stress curve
some of the concrete parameters. This concrete parametric diagram was adopted from
Kent and Park (1971). Where the ascending branch of concrete compressive strain
diagram is defined by equation (16) and the descending branch is defined by equation
(17) as explained in figure(B-2).
131
2𝜀 𝜀 2
𝑓𝑐 = 𝑓𝑐′ [𝜀 𝑐 − (𝜀 𝑐 ) ] (16)
𝑐𝑜 𝑐𝑜
𝑓𝑐 = 𝑓𝑐′ [1 − 𝑍(𝜀𝑐 − 𝜀𝑐𝑜 )] (17)
0.5 3+0.002𝑓𝑐′
Where 𝜀𝑐𝑜 = 0.002 and 𝑍=𝜀 in which 𝜀50𝑢 = (𝑓𝑐′ 𝑖𝑛 𝑝𝑠𝑖)
50𝑢 −𝜀𝑐𝑜 𝑓𝑐′ −1000
For 𝑓𝑐′ = 4500 psi. 𝜀50𝑢 = 0.003. Therefore Z=350.
concrete compressive strain-stress curve

5
4.5
4
3.5
Stress (ksi)
3
2.5
2
1.5
1
0.5
0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004 0.0045
strain (in/in)
Figure B-2 Concrete strain stress curve.
For the CFRP material properties, the tensile stress-strain diagram for the #2 CFRP rod
Figure 0-3Concrete strain stress curve.
was tested, and the obtained results indicated that the tensile behavior was linear up to the
rupture of the bar. As shown in figure (B-3).
132
(ksi)
Stress (ksi)
Figure B-3 #2CFRP strain-stress curve.
Analysis of Section AC
In order to draw the load deflection diagram, the moment curvature diagram was
constructed for the section. Three points can define the moment curvature diagram: the
first point is before concrete cracking; the second point is before steel yielding, and the
third point was selected when the strain in concrete (𝜀𝑐) is 0.004.
1) The cracking moment.
6.5(10)(5) + 2.56(8.25)
𝑦′ = = 5.123 𝑖𝑛.
6.5(10) + 2.56
𝐼𝑔 103 (6.5)
𝐼𝑡𝑟𝑎𝑛𝑠𝑓𝑜𝑟𝑚𝑒𝑑 = = = 615 𝑖𝑛4
0.88 12 ∗ 0.88
𝑦𝑡 = 10 − 5.12 = 4.8 𝑖𝑛
7.5√4200
𝑓𝑟𝑢𝑝𝑡𝑢𝑟𝑒 = =0.48 ksi.
1000
133
6.5 concrete strain fc
5.11
8.25
10.00
4.88
nAs=2.56
2#4steel bars
0.486
𝑓𝑟𝑢𝑝𝑡𝑢𝑟𝑒 𝐼𝑡𝑟𝑎𝑛𝑠𝑓𝑜𝑟𝑚𝑒𝑑 0.48(615)

𝑚𝑐𝑟𝑎𝑐𝑘 = = = 5.11 𝑘. 𝑓𝑡
𝑦𝑡 5.12
𝑚𝑐𝑟𝑎𝑐𝑘 5.11 ∗ 12
𝛷𝑐𝑟𝑎𝑐𝑘 = = ≅ 3 ∗ 10−5
𝐸𝑐 𝐼𝑡𝑟𝑎𝑛𝑠𝑓𝑜𝑟𝑚𝑒𝑑 3694(615)
2) The Steel Yielding Point.
At this point, the concrete cracked and it was assumed that the strain in the
concrete was still less than 0.002.Therfore, the force in the concrete can be found
from the first equation (16).
𝜀 2𝜀 𝜀 2
𝐹𝑐 = ∫0 𝑐 𝑏 𝑓𝑐 𝑑𝑐 , 𝑓𝑐 = 𝑓𝑐′ [𝜀 𝑐 − (𝜀 𝑐 ) ]
𝑐𝑜 𝑐𝑜
134
𝜀𝑐𝑔 .
𝜀𝑐𝑔 𝜀𝑠1
.
d1
.
. 𝑑𝑐
𝜀𝑐 FC
C .
𝑑𝜀𝑐
. d
STRAIN DISTRIBUTION STRESS DISTRIBUTION
𝜀𝑠
.
𝐶
From the strain triangle, one can find that 𝑑𝑐 = 𝜀 𝑑𝜀𝑐
𝑐𝑔
𝐶𝑏 𝜀 2𝜀 𝜀 2 𝜀𝑐𝑔 1 𝜀𝑐𝑔 2
𝐹𝑐 = 𝜀 𝑓𝑐′ ∫0 𝑐𝑔 [𝜀 𝑐 − (𝜀 𝑐 ) ] 𝑑𝜀𝑐 → 𝐹𝑐 = 𝐶𝑏𝑓𝑐′ [𝜀 − 3 (𝜀 ) ]
𝑐𝑔 𝑐𝑜 𝑐𝑜 𝑐𝑜 𝑐𝑜
𝜀𝑐𝑔 𝜀𝑠𝑦 𝐶(𝜀𝑠𝑦)

From the strain compatibility, one can find that = → 𝜀𝑐𝑔 =
𝐶 𝑑−𝐶 (𝑑−𝐶)
𝐹𝑠 = 𝐸𝑠 𝜀𝑠𝑦 𝐴𝑠 → 𝐹𝑠 = 27407.4(0.0027)(0.4) = 29.6 kips.
𝜀𝑐𝑔 𝑠1𝜀
𝐹𝑠1 = 𝐸𝑠 𝜀𝑠1 𝐴𝑠1 From the strain compatibility, one can find that = 𝐶−𝑑1
𝐶
𝜀𝑐𝑔
𝜀𝑠1 = (𝐶 − 𝑑1 )
𝐶
From equilibrium, 𝐹𝑠 = 𝐹𝑐 + 𝐹𝑠1
𝜀𝑐𝑔 1 𝜀𝑐𝑔 2 𝜀𝑐𝑔

29.6 = 𝐶(6.5)(4.2) [ − ( ) ] + 27407.4(0.22) [ (𝐶 − 𝑑1 )] (18)
0.002 3 0.002 𝐶
135
𝐶(0.0027)
𝜀𝑐𝑔 = , d = 8.25 in, 𝑑1 = 1.0𝑖𝑛 apply in eq.18 above and solve for C.
(𝑑−𝐶)
C= 2.26
𝐶(0.0027) 2.26(0.0027)
Check for 𝜀𝑐𝑔 = . 𝜀𝑐𝑔 = = 0.001 ≤ 0.002 ok
(𝑑−𝐶) (8.25−2.26)
To calculate the bending moment of each element about the neutral axis, the level arm of
the concrete parabola should be integrated as shown.
𝜀 2𝜀 𝜀 2
𝑀𝑐 = ∫0 𝑐 𝑏 𝑐 𝑓𝑐 𝑑𝑐 , 𝑓𝑐 = 𝑓𝑐′ [𝜀 𝑐 − (𝜀 𝑐 ) ]
𝑐𝑜 𝑐𝑜
𝐶 𝐶
From strain triangle, one can find that 𝑑𝑐 = 𝜀 𝑑𝜀𝑐 , 𝑐 = 𝜀 𝜀𝑐
𝑐𝑔 𝑐𝑔
𝐶 2𝑏 𝜀 2𝜀 𝜀 2 𝜀𝑐𝑔 2 1 𝜀𝑐𝑔
𝑀𝑐 = 𝜀 2
𝑓𝑐′ ∫0 𝑐𝑔 𝜀𝑐 [𝜀 𝑐 − (𝜀 𝑐 ) ] 𝑑𝜀𝑐 → 𝑀𝑐 = 𝑐 2 𝑏𝑓𝑐′ 𝜀 [3 − 4 𝜀 ]
𝜀𝑐𝑔 2 1 𝜀𝑐𝑔
𝑀𝑐 = 𝑐 2 𝑏𝑓𝑐′ [ − ] (19)
𝜀𝑐𝑜 3 4 𝜀𝑐𝑜
𝑀𝑠𝑡 = 𝐹𝑠𝑡 (𝑑 − 𝐶) (20)
𝑀𝑠𝑡1 = 𝐹𝑠𝑡1 (𝐶 − 𝑑1 ) (21)
Apply the values of C, d and d1 in eqs.19,20,and 21 above.
𝑀𝑐 = 37.75 𝑘. 𝑖𝑛, 𝑀𝑠𝑡 = 174.1𝑘. 𝑖𝑛, 𝑀𝑠𝑡1 = 4.23 k.in
𝑀𝑇 = 𝑀𝑐 + 𝑀𝑠𝑡 + 𝑀𝑠𝑡1 =18.0 k.ft
𝜀𝑐 0.001
𝛷𝑦 = = = 4.425 ∗ 10−4
𝐶 2.26
136
3) Concrete Crushing Point.
At this point, the concrete has reached the limit of crushing, which was assumed as
0.004.Therfoe; the force in the concrete can be integrated based on both concrete
equations.
𝜀 𝜀𝑐𝑔
𝐹𝑐 = ∫0 𝑜 𝑏 𝑓𝑐1 𝑑𝑐 + ∫𝜀𝑜 𝑏𝑓𝑐2 𝑑𝑐
2𝜀𝑐 𝜀𝑐 2 𝐶
𝑓𝑐1 = 𝑓𝑐′ [ −( ) ] , 𝑓𝑐2 = 𝑓𝑐′ [1 − 𝑍(𝜀𝑐 − 𝜀𝑐𝑜 )] , 𝑑𝑐 = 𝑑𝜀𝑐
𝜀𝑐𝑜 𝜀𝑐𝑜 𝜀𝑐𝑔
By integrating both terms of the above equation, the result is the total force in concrete.
𝐶𝑏 2 ′ 𝜀𝑐𝑔 − 𝜀𝑜
𝐹𝑐 = [ 𝑓𝑐 𝜀𝑜 + [2𝑓𝑐′ − 𝑧(𝜀𝑐𝑔 − 𝜀𝑜 )] ( )]
𝜀𝑐𝑔 3 2
𝜀𝑐𝑔 𝜀𝑠 𝜀𝑐𝑔 𝜀𝑠1

𝐹𝑠1 = 𝐸𝑠 𝜀𝑠1 𝐴𝑠1 , 𝐹𝑠 = (22.79 ln 𝜀𝑠 + 175.88)𝐴𝑠 , = 𝑑−𝐶 , = 𝐶−𝑑1
𝐶 𝐶
From equilibrium equation → 𝐹𝑠 = 𝐹𝑐 + 𝐹𝑠1 .
(22.79 ln 𝜀𝑠 + 175.88)𝐴𝑠 = [ 𝑓𝑐 𝜀𝑜 + [2𝑓𝑐′ − 𝑧(𝜀𝑐𝑔 − 𝜀𝑜 )] ( )] + 𝐸𝑠 𝜀𝑠1 𝐴𝑠1 (22)
𝜀𝑐𝑔 3 2
0.004(𝑑−𝐶) 0.004(𝐶−𝑑1)
𝜀𝑠 = , 𝜀𝑠1 = d = 8.25 in, 𝑑1 = 1.0𝑖𝑛 𝜀𝑐𝑔 = 0.004 apply in eq. 22
𝐶 𝐶
above and solve for C.
0.004(8.25−1.29)
C= 1.29in. check for 𝜀𝑠 = = 0.0215 ≥ 0.012 𝑡ℎ𝑒 ℎ𝑎𝑟𝑑𝑖𝑛𝑖𝑛𝑔 𝑟𝑒𝑔𝑖𝑜𝑛.
1.29
137
For moment capacity, the same previous process was done by integrating the concrete
force multiplied by the lever arm to the neutral axis and the resulted equation is as shown
below.
𝑏𝐶 2 2
𝑀𝑐 = 2
[𝑓𝑐′ (6𝜀𝑐𝑔
2
− 𝜀𝑜2 ) − 2𝑧(𝜀𝑐𝑔 − 𝜀𝑜 ) 2(𝜀𝑐𝑔 − 𝜀𝑜 )] (23)
12𝜀𝑐𝑔
𝑀𝑠𝑡 = 𝐹𝑠 (𝑑 − 𝐶) (24)
𝑀𝑠𝑡1 = 𝐹𝑠1 (𝐶 − 𝑑1 ) (25)
Apply the values of C, d and d1 in eqs.23,24,and 25 above.
𝑀𝑐 = 22.364 𝑘. 𝑖𝑛, 𝑀𝑠𝑡 = 246.293 𝑘. 𝑖𝑛, 𝑀𝑠𝑡1 = 1.57 k.in
𝑀𝑇 = 𝑀𝑐 + 𝑀𝑠𝑡 + 𝑀𝑠𝑡1 =22.519 k.ft
𝜀𝑐 0.004
𝛷𝑦 = = = 3.1 ∗ 10−3
𝐶 1.29
These three values were used to establish the load deflection curve by using the moment
area method as explained below.
Point # Moment (K.ft) Curvature
1 0 0
2(cracking moment) 5.0 3 ∗ 10−5
3(yielding moment) 18.0 4.43 ∗ 10−4
4(concrete crushing moment ) 22.5 3.0 ∗ 10−3
138
25 Concrete crushing point
20
steel yielding point

Moment (k.ft)
15 Concrete crushing point
10
5 concrete cracking point
0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035
curvture(1/in)
Deflection Calculation
1) At Cracking Point.
Δ= the moment of curvature diagram about A.
𝑓𝑡 2
𝛥 = 6.42 ∗ 10−4 𝑖𝑛
δ= half of the moment of curvature diagram about the center point.
𝑓𝑡 2
𝛿 = 1.158 ∗ 10−4 𝑖𝑛
6.42∗10−4 𝑓𝑡 2
𝛥𝑐𝑟. = − 1.158 ∗ 10−4 = 2.05 ∗ 10−4 ∗ 144 = 0.03 𝑖𝑛.
2 𝑖𝑛
139
2) At Yielding Point. P/2 P/2
A
𝑓𝑡 2
𝛥 = 8.515 ∗ 10−3 𝑖𝑛 2.67ft 2.67ft 2.67ft
𝑓𝑡 2
𝛿 = 1.36 ∗ 10−3 𝑖𝑛 Cracking Moment Diagram area
𝛥𝑦𝑖. = 0.42𝑖𝑛.
Cracking Curvature Diagram area
A
yielding Moment Diagram area
yielding Curvature Diagram area
A
3) At Ultimate Point.
𝑓𝑡 2
𝛥 = 0.0415
𝑖𝑛 ultimate Moment Diagram area
𝑓𝑡 2
𝛿 = 4.86 ∗ 10−3 𝑖𝑛
𝛥𝑢𝑙. = 2.288𝑖𝑛.
ultimate Curvature Diagram area
𝛥/2 . Deflection
.
𝛿 𝛥 '
140
Analysis of Section AR2
For the specimen AR2, the analysis was basically the same steps that have been done
with specimen AC. However, the contribution of two #2 CFRP rods in the beam cover is
going to be added to the flexural analysis and the concrete compressive strength is now
4.5ksi.
1) The Cracking Moment.
Since the concrete properties of specimen AR2 is slightly different from concrete
properties of specimen AC, the cracking moment and curvature can be assumed the
same.
2) The Steel Yielding Point.
At this point, the concrete has cracked and it assumed that the strain in the concrete is still
less than 0.002.Therfoe, the force in the concrete can be find from the first equation (1).
𝜀 2𝜀 𝜀 2
𝐹𝑐 = ∫0 𝑐 𝑏 𝑓𝑐 𝑑𝑐 , 𝑓𝑐 = 𝑓𝑐′ [𝜀 𝑐 − (𝜀 𝑐 ) ]
𝑐𝑜 𝑐𝑜
𝐶
From strain triangle, one can find that 𝑑𝑐 = 𝜀 𝑑𝜀𝑐
𝑐𝑔
𝐶𝑏 𝜀 2𝜀 𝜀 2 𝜀𝑐𝑔 1 𝜀𝑐𝑔 2
𝐹𝑐 = 𝜀 𝑓𝑐′ ∫0 𝑐𝑔 [𝜀 𝑐 − (𝜀 𝑐 ) ] 𝑑𝜀𝑐 → 𝐹𝑐 = 𝐶𝑏𝑓𝑐′ [𝜀 − 3 (𝜀 ) ]
𝜀𝑐𝑔 𝜀𝑠𝑦 𝐶(𝜀𝑠𝑦)

From strain compatibility, one can find that = 𝑑−𝐶 → 𝜀𝑐𝑔 = (𝑑−𝐶)
𝐶
141
𝜀𝑐𝑔
.
.
d1
𝜀𝑠1
d
dfrp
𝜀𝑠
.
𝜀𝑓𝑟𝑝
.
𝜀𝑐𝑔 𝜀𝑓𝑟𝑝 𝐶(𝜀𝑓𝑟𝑝 )

=𝑑 → 𝜀𝑐𝑔 = (𝑑 , where 𝜀𝑓𝑟𝑝 and 𝑑𝑓𝑟𝑝 are the strain and the depth of CFRP.
𝐶 𝑓𝑟𝑝 −𝐶 𝑓𝑟𝑝 −𝐶)
𝜀𝑓𝑟𝑝
𝐹𝑠 = 𝐸𝑠 𝜀𝑠𝑦 𝐴𝑠 → 𝐹𝑠 = 27407.4(0.0027)(0.4) = 29.6 kips.
𝜀𝑐𝑔 (𝑑𝑓𝑟𝑝 −𝐶)

𝐹𝑓𝑟𝑝 = 𝐸𝑓𝑟𝑝 𝜀𝑓𝑟𝑝 𝐴𝑓𝑟𝑝 → 𝜀𝑓𝑟𝑝 = 𝑐
𝜀𝑐𝑔 𝑠1 𝜀
𝐹𝑠1 = 𝐸𝑠 𝜀𝑠1 𝐴𝑠1 From strain compatibility, one can find that = 𝐶−𝑑1
𝐶
𝜀𝑐𝑔
𝜀𝑠1 = (𝐶 − 𝑑1 )
𝐶
From equilibrium, 𝐹𝑠 + 𝐹𝑓𝑟𝑝 = 𝐹𝑐 + 𝐹𝑠1
𝜀𝑐𝑔 1 𝜀𝑐𝑔 2 𝜀𝑐𝑔

29.6 + 𝐸𝑓𝑟𝑝 𝜀𝑓𝑟𝑝 𝐴𝑓𝑟𝑝 = 𝐶(6.5)(4.2) [ − ( ) ] + 27407.4(0.22) [ (𝐶 − 𝑑1 )] (26)
0.002 3 0.002 𝐶
𝐶(0.0027)
𝜀𝑐𝑔 = (𝑑−𝐶)
, d = 8.25 in, 𝑑1 = 1.0𝑖𝑛 , 𝑑𝑓𝑟𝑝 = 9.75𝑖𝑛. 𝐴𝑓𝑟𝑝 = 0.098𝑖𝑛2, and
142
𝐸𝑓𝑟𝑝 = 19767 𝑘𝑠𝑖 apply in eq.26 above and solve for C.
C= 2.466 in.
𝐶(0.0027) 2.466(0.0027)
Check for 𝜀𝑐𝑔 = . 𝜀𝑐𝑔 = = 0.00115 ≤ 0.002 ok
(𝑑−𝐶) (8.25−2.466)
0.00115(9.75 − 2.466)
𝜀𝑓𝑟𝑝 = = 0.0034
2.466
To calculate the bending moment of each element about the neutral axis, the level arm of
the concrete parabola should be integrated as shown.
𝜀 2𝜀 𝜀 2
𝑀𝑐 = ∫0 𝑐 𝑏 𝑐 𝑓𝑐 𝑑𝑐 , 𝑓𝑐 = 𝑓𝑐′ [𝜀 𝑐 − (𝜀 𝑐 ) ]
𝑐𝑜 𝑐𝑜
𝐶 𝐶
From strain triangle, one can find that 𝑑𝑐 = 𝜀 𝑑𝜀𝑐 , 𝑐 = 𝜀 𝜀𝑐
𝑐𝑔 𝑐𝑔
𝐶 2𝑏 𝜀 2𝜀 𝜀 2 𝜀𝑐𝑔 2 1 𝜀𝑐𝑔
′
𝑀𝑐 = 𝜀 2 𝑓𝑐 ∫0
𝑐𝑔
𝜀𝑐 [𝜀 𝑐 − (𝜀 𝑐 ) ] 𝑑𝜀𝑐 → 𝑀𝑐 = 𝑐 2 𝑏𝑓𝑐′ 𝜀 [3 − 4 𝜀 ]
𝜀𝑐𝑔 2 1 𝜀𝑐𝑔
𝑀𝑐 = 𝑐 2 𝑏𝑓𝑐′ [ − ] (27)
𝜀𝑐𝑜 3 4 𝜀𝑐𝑜
𝑀𝑠𝑡 = 𝐹𝑠𝑡 (𝑑 − 𝐶) (28)
𝑀𝑠𝑡1 = 𝐹𝑠𝑡1 (𝐶 − 𝑑1 ) (29)
𝑀𝑓𝑟𝑝 = 𝐹𝑓𝑟𝑝 (𝑑𝑓𝑟𝑝 − 𝐶) (30)
Apply the values of C, d and d1 in eqs27,28,29,and 30. above.
143
𝑀𝑐 = 51.103 𝑘. 𝑖𝑛, 𝑀𝑠𝑡 = 171.03 𝑘. 𝑖𝑛, 𝑀𝑠𝑡1 = 6.04 k.in 𝑀𝑓𝑟𝑝 =47.97 k.in.
𝑀𝑇 = 𝑀𝑐 + 𝑀𝑠𝑡 + 𝑀𝑠𝑡1 + 𝑀𝑓𝑟𝑝 =23.013 k.ft
𝜀𝑐 0.00115
𝛷𝑦 = = = 4.66 ∗ 10−4
𝐶 2.466
3) Concrete Crushing Point or CFRP Rupture Point
At this point, an assumption have to be made whether the concrete will reach the strain of
0.004 or the CFRP will reach the strain of 0.016. Therefore, the assumptions are going to
be that the concrete will reach the crushing limit of 0.004 and the steel is within the
perfectly plastic region.
𝜀 𝜀𝑐𝑔
𝐹𝑐 = ∫0 𝑜 𝑏 𝑓𝑐1 𝑑𝑐 + ∫𝜀𝑜 𝑏𝑓𝑐2 𝑑𝑐
2𝜀 𝜀 2 𝐶
𝑓𝑐1 = 𝑓𝑐′ [𝜀 𝑐 − (𝜀 𝑐 ) ] , 𝑓𝑐2 = 𝑓𝑐′ [1 − 𝑍(𝜀𝑐 − 𝜀𝑐𝑜 )] , 𝑑𝑐 = 𝜀 𝑑𝜀𝑐
𝑐𝑜 𝑐𝑜 𝑐𝑔
By integrating both terms of the above equation, the result is the total force in concrete.
𝐹𝑐 = [ 𝑓𝑐 𝜀𝑜 + [2𝑓𝑐′ − 𝑧(𝜀𝑐𝑔 − 𝜀𝑜 )] ( )]
𝜀𝑐𝑔 3 2
𝐹𝑠1 = 𝐸𝑠 𝜀𝑠1 𝐴𝑠1 , 𝐹𝑠 = 29.6𝑘𝑖𝑝𝑠, 𝐹𝑓𝑟𝑝 = 𝐸𝑓𝑟𝑝 𝜀𝑓𝑟𝑝 𝐴𝑓𝑟𝑝1.
From equilibrium equation → 𝐹𝑠 + 𝐹𝑓𝑟𝑝 = 𝐹𝑐 + 𝐹𝑠1 .
𝐸𝑓𝑟𝑝 𝜀𝑓𝑟𝑝 𝐴𝑓𝑟𝑝1. + 29.6 = [ 𝑓 𝜀 + [2𝑓𝑐′ − 𝑧(𝜀𝑐𝑔 − 𝜀𝑜 )] ( )] + 𝐸𝑠 𝜀𝑠1 𝐴𝑠1 (31)
𝜀𝑐𝑔 3 𝑐 𝑜 2
144
(𝜀𝑐𝑔 )(𝑑𝑓𝑟𝑝 −𝐶) 𝜀𝑐𝑔 (𝐶−𝑑1)
𝜀𝑓𝑟𝑝 = 𝜀𝑠1 = , 𝜀𝑐𝑔 = 0.004 apply in eq.31 above and solve for C.
𝐶 𝐶
0.004(8.25−2.064)
C= 2.064in. check for 𝜀𝑠 = = 0.0119 ≤ 0.012 𝑡ℎ𝑒 𝑝𝑙𝑎𝑠𝑡𝑖𝑐 𝑟𝑒𝑔𝑖𝑜𝑛.
2.064
(𝜀𝑐𝑔 )(𝑑𝑓𝑟𝑝 −𝐶) (0.004)(9.75−2.064)

𝑐ℎ𝑒𝑐𝑘 𝑓𝑜𝑟 𝜀𝑓𝑟𝑝 = = = 0.015 ≤ 0.016 𝑂. 𝑘.
𝐶 2.064
For moment capacity, the same previous process was done by integrating the concrete
force multiplied by the lever arm to the neutral axis and the resulted equation is as shown
below.
𝑏𝐶 2 2
𝑀𝑐 = 2
[𝑓𝑐′ (6𝜀𝑐𝑔
2
− 𝜀𝑜2 ) − 2𝑧(𝜀𝑐𝑔 − 𝜀𝑜 ) 2(𝜀𝑐𝑔 − 𝜀𝑜 )] (32)
12𝜀𝑐𝑔
𝑀𝑠𝑡 = 𝐹𝑠 (𝑑 − 𝐶) (33)
𝑀𝑠𝑡1 = 𝐹𝑠1 (𝐶 − 𝑑1 ) (34)
𝑀𝑓𝑟𝑝 = 𝐹𝑓𝑟𝑝 (𝑑𝑓𝑟𝑝 − 𝐶) (35)
Apply the values of C, d and d1 in eqs.32,33,34,and35 above.
𝑀𝑐 = 56.25𝑘. 𝑖𝑛, 𝑀𝑠𝑡 = 196.19𝑘. 𝑖𝑛, 𝑀𝑠𝑡1 = 13.22k.in 𝑀𝑓𝑟𝑝 = 223.33 𝑘. 𝑖𝑛.
𝑀𝑇 = 𝑀𝑐 + 𝑀𝑠𝑡 + 𝑀𝑠𝑡1 + 𝑀𝑓𝑟𝑝 =40.751k.ft
𝜀𝑐 0.004
𝛷𝑐 = = 2.064 = 1.93 ∗ 10−3
𝐶
145
Point # Moment (K.ft) Curvature
Starting point 0 0
2( concrete cracking moment) 5.11 3 ∗ 10−5
3( steel yielding moment) 23.013 4.66 ∗ 10−4
4(concrete crushing moment ) 40.75 1.93 ∗ 10−3
The same process of calculating the deflection of AC was done to calculate the deflection
of AR2. The result are as shown below.

Load (kips)
146
Load (kips) Load (kips) Load (kips)
147
Load (kips)
The same procedure was done to calculate the theoretical load deflection curve for the
other specimens as shown in the following graphs.
148
Load (kips)
149

Behavior of Reinforced Concrete Beams Retrofitted in Flexure Usin

Uploaded by

Copyright:

Available Formats

Behavior of Reinforced Concrete Beams Retrofitted in Flexure Usin

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Behavior of Reinforced Concrete Beams Retrofitted in Flexure Usin

Uploaded by

Copyright:

Available Formats

Portland State University

Dissertations and Theses Dissertations and Theses

Behavior of Reinforced Concrete Beams Retrofitted

Follow this and additional works at: https://pdxscholar.library.pdx.edu/open_access_etds

Part of the Structural Engineering Commons

Flexure Using CFRP-NSM Technique

A thesis submitted in partial fulfillment of the

Portland State University

structural components. The many advantages of using FRP as strengthening materials

Mounted –Fiber Reinforced polymer (NSM-FRP) is a promising technique used to

strengthen concrete members. However, de-bonding issues have to be overcome to make

technique in retrofitting existing structures is observed.

List of Tables ……………………………………………….…………..…..……….. . vii

Lists of Figures …………………………………………………………………...….... viii

Notations ………………………………………….……………………………..…..… xvi

Chapter 1: Introduction ................................................................................................... 1

1.1. General ................................................................................................................ 1

1.2. Near Surface Mounted Technique (NSM) ........................................................... 2

1.3. Objective .............................................................................................................. 4

Chapter 2: Literature Review.......................................................................................... 5

2.1. Introduction .......................................................................................................... 5

2.2. Bond Mechanism.................................................................................................. 5

2.3. Flexural Behavior ............................................................................................... 14

2.4. Summary of the Literature Review and Discussion ........................................... 26

Chapter 3: Experimental program ................................................................................ 30

3.1. Design guides ..................................................................................................... 30

3.2. Tension Test ....................................................................................................... 30

3.2.2. The Results.................................................................................................. 37

3.3. Pullout Tests ....................................................................................................... 38

3.4. Flexural Tests ..................................................................................................... 40

3.4.1. Materials ..................................................................................................... 41

3.4.2. Concrete ...................................................................................................... 42

3.4.3. Epoxy .......................................................................................................... 44

3.4.5. Specimens’ Analysis and Design ................................................................ 46

3.4.6. Specimens Preparation ................................................................................ 52

3.4.7. Concrete Casting and Curing ...................................................................... 53

3.4.8. Instrumentations .......................................................................................... 54

3.4.9. Preparation of the Groove ........................................................................... 56

3.4.10. Inserting the CFRP Bars. ......................................................................... 57

3.4.11. Instrumentations and Procedure .............................................................. 57

Chapter 4: The Results.................................................................................................. 59

4.1. Pullout Test ........................................................................................................ 59

4.2. Discussing the Pullout Results. .......................................................................... 65

4.3. Beam Test ........................................................................................................... 70

.4.3.2 Set (B) specimens ....................................................................................... 92

4.3.3. Set A and B Results Comparison .............................................................. 108

4.3.4. Measured Crack Width ............................................................................. 112

4.3.5. Theoretical Analysis ................................................................................. 113

Chapter 5: Summery Conclusions and Recommendations ......................................... 122

5.1. Summery .......................................................................................................... 122

5.2. Conclusions ...................................................................................................... 122

5.3. Recommendations ............................................................................................ 123

References ....................................................................................................................... 124

Appendix A ..................................................................................................................... 127

Appendix B ..................................................................................................................... 130

Table 1-1-1 Typical properties of steel and FRP materials. ............................................... 2

Table 2-1 The test results. ................................................................................................. 17

Table 2-2 Specimens details. ............................................................................................ 18

Table 2-3 Test results. ...................................................................................................... 19

Table 2-4 Specimens details and results. ......................................................................... 24