Behaviour of R.C. Horizontally Curved Beams With Openings Strengthened by CFRP Laminates-2014

Republic of Iraq
Ministry of Higher Education and Scientific Research

Babylon University
College of Engineering
Civil Engineering Department
Behavior of R.C. Horizontally Curved Beams

with Openings Strengthened by CFRP
Laminates
A Dissertation
Submitted to the College of Engineering in the University of
Babylon in Partial Fulfillment of the Requirements for the
Degree of Doctor of Philosophy in the Civil Engineering
(Structural)
By
Sadjad Amir Hemzah

(B.Sc. Civil Engineering, 2000)
(M.Sc. Structural Engineering, 2003)
Supervised By
Prof. Dr. Ammar Yasir Ali

2014
‫ﺑﺳم ﷲ اﻟرﺣﻣن اﻟرﺣﯾم‬
‫ﻧَﺮﻓَﻊ‪ ‬ﺩ‪‬ﺭ‪‬ﺟ‪‬ﺎﺕ‪ ‬ﻣ‪‬ﻦ ﻧﱠﺸ‪‬ﺎﺀ‪‬‬
‫ﻭ‪‬ﻓَﻮﻕَ ﻛُـﻞﱢ ﺫ‪‬ﻱ ﻋ‪‬ﻠﻢٍ ﻋ‪‬ﻠ‪‬ﻴﻢ‪‬‬
‫ﺻدق ﷲ اﻟﻌﻠﻲ اﻟﻌظﯾم‬
‫ﺳﻮﺭﺓ ﻳﻮﺳﻒ ‪ /‬آﻳﺔ ‪76‬‬

Certificate
I certify that the preparation of this dissertation titled

" Behavior of R.C. Horizontally Curved Beams with
Openings Strengthened by CFRP Laminates ", is prepared by
" Sadjad Amir Hemzah ", under my supervision at the
Department of Civil Engineering in the University of Babylon in
partial fulfillment of the requirements for the degree of Doctor
of Philosophy in Civil Engineering (Structure).
Signature:
Name: Prof. Dr. Ammar Yaser Ali
Date: / / 2014
Committee Certification
We certify that we, the examining committee, have read the

dissertation titled" Behavior of R.C. Horizontally Curved Beams with
Openings Strengthened by CFRP Laminates ", which is being submitted by
" Sadjad Amir Hemzah ", and examined the student in its content and in
what is connected with it, and that in our opinion, it meets the standard
of a thesis for the degree of Doctor of Philosophy in Structural
Engineering(Structure).
Signature: Signature:
Name: Dr. Ammar Y. Ali Name: Dr. Ihsan A.S. Al-Shaarbaf
(professor) (Assist. Professor)
(Supervisor) (Member)
Name: Dr. Hayder T. Nimnim Name: Dr. Mustafa B. Dawood
(Assist. Professor) (professor)
(Member) (Member)
Name: Dr. Nameer A. Alwash Name: Dr. Hisham M. Al-Hassani
(professor) (professor)
(Member) (Chairman)
Approved by the Head of the Civil Engineering Department

Signature:
Name: Asst. Prof. Dr Abdul-Rudha I. Ahmed
(The Head of the Civil Engineering
Department)
Date: / / 2014
Approved by the Dean of the College of Engineering
Signature:
Name: Prof. Dr. Adil A. Al-Moosaway
(The Dean of the College of Engineering)
Date: / / 2013
ACKNOWLEDGMENT
In the name of ALLAH, the most compassionate the most merciful.
Praise be to ALLAH and pray and peace be on his prophet Mohammed and
his family.
First, I would like to express my appreciation and deepest gratitude
to my supervisor Prof. Dr. Ammar Y. Ali for his remarkable suggestions,
encouragement and guidance through the research; I am really indebted to
him.
Thanks to all staff of Civil Engineering Department/College of
Engineering / University of Babylon for their appreciable support. Special
thanks to Dr. Bashar Abid Hamza and my friends Majed, Mohsen, and
Bahaa for their help during this study.
Thanks also extended to the staff of the Structures Laboratory at
College of Engineering / University of Karbala for their help in using the
various facilities.
A special thank and gratitude to my family for their care, patience
and encouragement throughout the research period.
Sadjad Amir Hemzah
2014
i
Abstract
This research is devoted to investigate the behavior and performance
of reinforced concrete horizontally curved beams with and without openings,
unstrengthed and strengthened externally by CFRP laminates or internally by
steel reinforcement.
The experimental work consisted of fabrication and testing of fourteen
reinforced concrete curved beams divided into two group. The first group
included ten of semicircular beams (one without opening and nine with
openings), while the second group consisted of four full circular reinforced
concrete beams (one without opening and three with openings). The
variables considered in the experimental program included: geometry of
circular beam, location of opening through profile of beam, presence of
internal strengthening by reinforcing steel(stirrups) and external
strengthening by CFRP laminates for beam around openings. The beams
were tested under effect of point loads at top edge of each midspan .
The experimental results showed that the presence of opening at
region of maximum bending moment and shear force caused a significant
decrease in ultimate load capacity by about 35% for semi-circular beams
and 50% for full circular (ring) beams compared with control beams without
opening associated with beam shear mode failure. The use of steel
reinforcement as internal strengthening and CFRP laminates as external
confinement around openings improved the ultimate load capacity of the
semi-circular beam by a ratio of (3% to 30%) and (11% to 40%),
respectively. While for full circular (ring) beam the increasing in ultimate
load reached about 60% and 75% for internal and external strengthening,
respectively.
Also, both types of strengthening around opening (internally by steel
reinforcement or externally by CFRP laminates) enhanced crack patterns,
post cracking torsional and flexural stiffnesses.
ii
The numerical work included a three-dimensional nonlinear finite
element model using the computer program ANSYS version 12.1 suitable
for the analysis of the tested reinforced concrete horizontally curved beams
with or without openings and unstrengthened or strengthened by(CFRP
laminates or reinforcing steel) under incremental loading, as well as a
parametric study for many variables. Full bond was assumed between the
CFRP and concrete surface and between the steel reinforcement and
concrete. In general, a reasonable agreement between the finite element
solutions and experimental results has been obtained concerning estimate
load-deflection response, mode of failure, cracking and ultimate loads with
average difference about 4.1% and 16% for ultimate load and deflection
respectively.
The numerical results for the analyzed beams show that the ultimate
load decreased about 12% when the length increased by 50% and it
increased by 14% when the length of opening decreased by 50%. Moreover,
an increment in ultimate load about 41% was found for circular opening
instead of rectangular opening with the same area. Also, for the same length
of the beam, the ultimate load increased about 103%,130% and 172% when
the curvature of the curved beam (1/R) decreased (100%,200,and ∞),
respectively.
iii
List of Contents
Acknowledgments i
Abstract ii
List of Contents iv
List of Tables xi
List of Figures x
Notation xvii
Abbreviations xviii
CHAPTER ONE:INTRODUCTION
1.1 General 1
1.2 General Guidelines to Selection of the Size and Location 5

of Openings.
1.3 Reinforced Concrete Curved Beams 6
Analysis for Ultimate Strength Under Combined Torsion 7

1.4
Bending and Shear
Failure Modes Under Combined Torsion, Bending, and 10

1.5
Shear with Small Opening
1.6 Design for Torsion 17
1.6.1 Design for Beam type Failure 18
1.6.2 Design for Frame Type Failure 21
1.7 Fiber Reinforced Polymer FRP 23
1.7.1 Fiber Materials 23
1.2.2 Matrix 25
1.7.3 Properties of FRP 25
1.8 Aims of the Study. 27
1.9 Layout of Thesis 28
iv
CHAPTER TWO:LITERATURE REVIEW
2.1 Introduction 29
Experimental and Analytical Studies on Reinforced 29

2.2
Concrete Curved Beams.
Experiential and Analytical Works on Reinforced 39

2.3
Concrete Beam with Openings.
2.4 Concluding Remarks 54
CHAPTER THREE:EXPERIMENTAL WORK
3.1 General 57
3.2 Description of Specimens 57
3.2.1 Semicircular Curved Beams (Group I) 60
3.2.1.1 Semicircular Curved Beams without Opening 60
3.2.1.2 Semicircular Curved Beams with Opening 60
3.2.2 Ring Curved Beams (Group I) 60
3.2.2.1 Ring Curved Beams without Opening 60
3.2.2.2 Ring Curved Beams with Opening 60
3.3 Strengthening System 62
3.3.1 Internal Strengthening by Internal Reinforcement 62
3.3.2 External Strengthening by CFRP Straps 62
3.4 Material Properties of Tested Specimens 65
3.4.1 Concrete 65
3.4.1.1 Cement 65
3.4.1.2 Coarse Aggregate (Gravel) 66
3.4.1.3 Fine Aggregate (Sand) 66
3.4.1.4 Mixing Water 68
v
3.4.1.5 Mix Design 68
3.4.1.6 Mixing Procedure 69
3.4.2 Reinforcing Steel 69
3.4.3 Carbon Fiber (CFRP) Strengthening System 71
3.5 Mechanical Properties of Hardened Concrete 72
3.5.1 Compressive Strength (fc') 72
3.5.2 Splitting Tensile Strength (ft) 73
3.6 CFRP Installation 74
3.7 Instrument and Test Procedure 75
3.8 Loading and Support Condition 78
3.9 Steel and wood mold 79
CHAPTER FOUR:EXPERIMENTAL RESULTS AND

DISCUSSION
4.1 General 81
4.2 Semicircular Curved Concrete Beams (Group I) 82
4.2.1 Pilot and Control Semicircular Curved Beam (SCB.P) 82
Semicircular Curved Beam with Opening near External 85

4.2.2
Support (SCB.E)
4.2.2.1 Unstrengthened Semicircular Curved Beam (SCB.Eo) 85
Internally Strengthened Semicircular Curved Beam 88

4.2.2.2
(SCB.Esr)
Externally Strengthened Semicircular Curved Beam 90

4.2.2.3
(SCB.Ecfrp)
4.2.2.4 Summary of tested SCB.E Beams 93
Semicircular Curved Beam with Opening near Midspan 95

4.2.3
(SCB.M)
vi
4.2.3.1 Unstrengthened Semicircular Curved Beam (SCB.Mo) 95

4.2.3.2
(SCB.Msr)

4.2.3.3
(SCB.Mcfrp)
4.2.3.4 Summary of tested SCB.M Beams 103
Semicircular Curved Beam with Opening near Interior 105

4.2.4
Support (SCB.I)
4.2.4.1 Unstrengthened Semicircular Curved Beam (SCB.Io) 105

4.2.4.2
(SCB.Isr)

4.2.4.3
(SCB.Icfrp)
4.2.4.4 Summary of tested SCB.I Beams 113
4.2.5 Summary of Tested Semicircular Curved Beams 115
4.3 Full Circular (Ring) Concrete Beams 115
4.3.1 Control Specimen (FCB.P) 116
4.3.2 Full Circular Beam with Opening (FCB.Mo) 118
4.3.3 Internally Strength Full Circular Beam (FCB.Msr) 121
4.3.4 Externally strengthened Full Circular Beam (FCB.Mcfrp) 123
4.3.5 Summary of Tested FCB.M Beams 126
4.3.6 Summary of The Tested Curved Beams 128
CAPTER FIVE: FINITE ELEMENTS MODELING AND

FORMULATION OF NONLIEAR ANALYSIS OF TESTED CURVED
BEAMS
5.1 Introduction 129
5.2 Material Modeling 130
vii
5.2.1 Concrete Modeling 130
5.2.1.1 Uniaxial Compression Behavior for Concrete 131
5.2.1.2 Tensile Behavior of Concrete 134
5.2.1.3 Biaxial Stress Behavior of Concrete 136
5.2.1.4 Triaxial Stress Behavior of Concrete 137
5.2.1.5 Post - Cracking Model (Tension Stiffening Model) 137
5.2.1.6 Crushing Modeling 139
5.2.1.7 Shear Transfer Model 139
5.2.1.8 Cracking Modeling 139
5.2.1.9 Failure Criteria for Concrete 141
5.2.1.9.1 Determination of the Model Parameters 144
5.2.2 Reinforcement Modeling 145
5.2.3 CFRP Composite Modeling 148
5.3 Material Idealization 148
5.3.1 Element Types 148
5.3.1.1 Element SOLID65 149
5.3.1.2 Element LINK8 149
5.3.1.3 Element SOLID45 150
5.3.1.4 Element Shell41 151
5.4 Nonlinear Finite Element Analysis 151
5.4.1 Numerical Integration 152
Procedure for Solving Nonlinear Finite Element 154

5.4.2
Equations
5.4.3 Convergence Criteria 157
5.4.4 Analysis Termination Criteria 158
viii
CHAPTER SIX: FINITE ELEMENT ANALYSIS
6.1 Introduction 159
6.2 Description of Specimens in Finite Element 159
6.3 Mesh Refinement 160
6.4 Reinforcing Steel Modeling 162
6.5 Modeling of CFRP Laminates 163
6.6 Loads and Boundary Conditions Representations 164
6.7 Results of Finite Element Analysis 165
6.7.1 Load – Deformations Curves 165
6.7.2 Deflection, Cracking and Ultimate Loads 180
6.7.3 Deflection of Service Loads 180
6.7.4 Crashing and Stresses in Curved Beams. 180
6.8 Parametric Study 181
6.8.1 Effect of Curvature 182
6.8.2 Effect of Type and Size of Opening 183
6.8.3 Effect of Opening Length 186
6.8.4 Effect of Opening Height 188
6.8.5 Wrapping Schemes of CFRP (full and U- wrapping ) 190
CHAPTER SEVEN:CONCLUSIONS and RECOMMENDATIONS FOR

FUTURE WORKS
7.1 General 193
7.2 Conclusions 193
7.3 Recommendations 196
ix
References 197
Analytical Solution and Design of Semicircular A-1

Appendix A
Curved Beam - Control Beam
Design of Steel Reinforcement for strengthening of Mid A-5

Appendix B
Opening of Semicircular Curved Beams
Cracks Pattern And Stress In Concrete And CFRP A-12

Appendix C
Laminates
x
List of Tables
No. Title Page
3.1 Description of Tested Specimens 61
3.2 Chemical and Physical Test Results of the Cement 65
3.3 Grading of Coarse Aggregate 67
3.4 Fine Aggregate Properties 67
3.5 Properties of Concrete Mix 68
3.6 Specification and Test Results of Steel Reinforcing Bar Values 70
3.7 Properties of Sikadur-330 (Impregnating Resin) 71
3.8 Properties of SikaWrap Hex-230C (Carbon Fiber Fabric) 72
3.9 Concrete Compressive Strength of Specimens 73
4.1 Summary of tested SCB.E Beams 93
4.2 Summary of tested SCB.M Beams 104
4.3 Summary of tested SCB.I Beams 113
4.4 Summary of tested FCB Beams 126
5.1 Element Types 148
5.2 Sampling Points Locations and Weighting Factors 154
6.1 Theoretical and Experimental Cracking and Ultimate Loads 180
6.2 Ultimate Load for Different Types of SCB.Io with Variable 183
Radiuses
6.3 Ultimate Load Capacity for Different Opening Dimensions of 185
SCB.Io Curved Beam
6.4 Ultimate Load Capacity for Different Opening Length of 187
SCB.Io Curved Beam
6.5 Ultimate Load Capacity for Different Height of SCB.Io 189
Curved Beam
6.6 Ultimate Load Capacity of Wrapping Schemes for SCB.Io 192
and SCB.Mo Curved Beams
xi
List of Figures
No. Title Page
1.1 Applications of Curved Beams 1
1.2 Typical Layout of Pipes for High Rise Building 2
1.3 Collapses Mechanism at Large Opening 4
1.4 Guidelines for Location of Web Opening 6
1.5 Horizontally Curved Beam Loading and Forces 7
1.6 Failure surfaces for a solid beam in Modes 1, 2, and 3 8
1.7 Mode 1 failure surface for beams with a small opening 10
1.8 Geometry of the boundary of failure surface 12
1.9 Mode 2 failure surface for beams with a small opening 13
1.10 Frame-type failure of a beam with a small opening under torsion 18
1.11 Idealized free-body diagram at opening of a beam under loading 22
1.12 Stress-Strain Relationship of Fibers and Steel 24
1.13 a-Stress-Strain Curves of Fibers, FRP and Matrix 26
b-Typical Composition of FRP Material
2.1 yield surfaces for combined bending and torsion, at a section where 31
a plastic hinge was formed.
2.2 Geometry of Test Specimens by Jordaan et al (1974) 32
2.3 Geometry of Test Specimens of Badawy et al (1977) 34
2.4 Typical Shear Failure of a Beam with Small Openings 42
Containing no Shear Reinforcement
2.5 Shear Failure of a Beam at the Throat Section by Salam 43
2.6 Modes of Failure for Small Opening ,Mansur (1998). 47
2.7 Details of the tested beams by Abdulla (2003) 49
3.1 Details for Group I of Tested Curved Beams 58
3.2 Details for Group II of Tested Ring Beams 59
3.3 External Strengthening around opening with CFRP laminates 63
3.4 Details For Internally Strengthening by Steel Reinforcement 64
3.5 Cage of Steel Reinforcement 70
3.6 CFRP Laminates and Epoxy Resine 71
3.7 Compressive Strength Test and Splitting Tensile Strength Test 73
xii
3.8 Application of CFRP System on Concrete Element 75
3.9 Loading Machine Used in the Testes 76
3.10 Instruments Details 77
3.11 Details of the Typical Support Condition 78
3.12 Semicircular Molds and Tools 79
3.12 Wooden Circular Molds 80
4.1 Control Beam SCB.P 83
4.2 Mode of Failure and Cracks Pattern for Control Beam 83
4.3 Load-Midspan Deflection Curve for Control Beam SCB.P 84
4.4 Load-Midspan Twisting Angle Curve for Control Beam SCB.P 84
4.5 Semicircular Curved Beam SCB.Eo 86
4.6 Mode of Failure and Cracks Pattern for SCB.Eo Curved Beam 86
4.7 Load-Midspan Deflection Curve for Control Beam SCB.Eo 87
4.8 Load-Midspan Twisting Angle Curve for Control Beam SCB.Eo 87
4.9 Semicircular Curved Beam SCB.Esr 88
4.10 Mode of Failure and Cracks Pattern for SCB.Esr Curved Beam 89
4.11 Load-Midspan Deflection Curve for Control Beam SCB.Esr 89
4.12 Load-Midspan Twisting Angle Curve for Control Beam SCB.Esr 90
4.13 Semicircular Curved Beam SCB.Ecfrp 91
4.14 Mode of Failure and Cracks Pattern for SCB.Ecfrp Curved Beam 91
4.15 Load-Midspan Deflection Curve for Control Beam SCB.Ecfrp 92
4.16 Load-Midspan Twisting Angle Curve for Control Beam SCB.Ecfrp 92
4.17 Comparison of Load-Midspan Deflection Curves for SCB.Ep, 94
SCB.Eo, SCB.Esr and SCB.Ecfrp Curved Beams
4.18 Comparison of Twisting Angle Curves at Midspan for SCB.Ep, 94
SCB.Eo, SCB.Esr and SCB.Ecfrp Curved Beams
4.19 Semicircular Curved Beam SCB.Mo 96
4.20 Mode of Failure and Cracks Pattern for SCB.Mo Curved Beam 96
4.21 Load-Midspan Deflection Curve for Control Beam SCB.Mo 97
4.22 Load-Midspan Twisting Angle Curve for Control Beam SCB.Mo 97
4.23 Semicircular Curved Beam SCB.Msr 99
xiii
4.24 Mode of Failure and Cracks Pattern for SCB.Msr Curved Beam 99
4.25 Load-Midspan Deflection Curve for Control Beam SCB.Msr 100
4.26 Load-Midspan Twisting Angle Curve for Control Beam SCB.Msr 100
4.27 Semicircular Curved Beam SCB.Mcfrp 101
4.28 Mode of Failure and Cracks Pattern for SCB.Mcfrp Curved Beam 102
4.29 Load-Midspan Deflection Curve for Control Beam SCB.Mcfrp 102
4.30 Load-Midspan Twisting Angle Curve for Control Beam 103
SCB.Mcfrp
4.31 Comparison of Load-Midspan Deflection Curves for SCB.P, SCB. 104
Mo, SCB. Msr and SCB.Mcfrp Curved Beams
4.32 Comparison of Twisting Angle Curves at Midspan for SCB.P, 105
SCB.Mo, SCB.Msr and SCB.Mcfrp Curved Beams
4.33 Semicircular Curved Beam SCB.Io 106
4.34 Mode of Failure and Cracks Pattern for SCB.Io Curved Beam 106
4.35 Load-Midspan Deflection Curve for Control Beam SCB.Io 107
4.36 Load-Midspan Twisting Angle Curve for Control Beam SCB.Io 107
4.37 Semicircular Curved Beam SCB.Isr 108
4.38 Mode of Failure and Cracks Pattern for SCB.Isr Curved Beam 109
4.39 Load-Midspan Deflection Curve for Control Beam SCB.Isr 109
4.40 Load-Midspan Twisting Angle Curve for Control Beam SCB.Isr 110
4.41 Semicircular Curved Beam SCB.Icfrp 111
4.42 Mode of Failure and Cracks Pattern for SCB.Icfrp Curved Beam 111
4.43 Load-Midspan Deflection Curve for Control Beam SCB.Icfrp 112
4.44 Load-Midspan Twisting Angle Curve for Control Beam SCB.Icfrp 112
4.45 Comparison of Load-Midspan Deflection Curves for SCB.P, 114
SCB.Io, SCB.Isr and SCB.Icfrp Curved Beams
4.46 Comparison of Twisting Angle Curves at Midspan for SCB.P, 114
4.47 Full Circular Beam FCB.P 116
4.48 Mode of Failure and Cracks Pattern for Circular Beam FCB.P 117
4.49 Load-Midspan Deflection Curve for FCB.P Circular Beam 117
4.50 Load-Midspan Rotation Curve for FCB.P Circular Beam 118
4.51 Full Circular Beam FCB.Mo 119
4.52 Mode of Failure and Cracks Pattern for Circular Beam FCB.Mo 119
4.53 Load-Midspan Deflection Curve for FCB.Mo Circular Beam 120
xiv
4.54 Load-Midspan Rotation Curve for FCB.Mo Circular Beam 120
4.55 Full Circular Beam FCB.Msr 121
4.56 Mode of Failure and Cracks Pattern for Circular Beam FCB.Msr 122
4.57 Load-Midspan Deflection Curve for FCB.Msr Circular Beam 122
4.58 Load-Midspan Rotation Curve for FCB.Msr Circular Beam 123
4.59 Full Circular Beam FCB.Mcfrp 124
4.60 Mode of Failure and Cracks Pattern for Circular Beam FCB.Mcfrp 124
4.61 Load-Midspan Deflection Curve for FCB.Mcfrp Circular Beam 125
4.62 Load-Midspan Rotation Curve for FCB.Mcfrp Circular Beam 125
4.63 Comparison of Load-Midspan Deflection Curves for FCB.P , 127
FCB.Mo, FCB.Msr and FCB.Mcfrp Curved Beams
4.64 Comparison of Twisting Angle Curves at Midspan for FCB.P, 127
FCB.Mo, FCB.Msr and FCB.Mcfrp Curved Beams
5.1 Uniaxial Compressive Strain curve for concrete with different strength 132
5.2 Stress-Strain Curve for Concrete in Compression 134
5.3 Typical Tensile Stress-Strain Curve for Concrete 135
5.4 Biaxial State of Loading 136
5.5 Failure Surface of Concrete in 3-D Stress Space 137
5.6 Pre and Post-Cracking Behavior of Normal Strength Concrete 138
5.7 Cracking representation in discrete cracking modeling approach 140
5.8 Smeared Crack Modeling 141
5.9 Failure Surface 143
5.10 Profile of the Failure Surface as Function of Five Parameters 145
5.11 Modeling of Reinforcing Bars . 146
5.12 Models for Reinforcement in Reinforced Concrete 147
5.13 Schematic Properties of CFRP Composites 148
5.14 Geometry of Element SOLID65 149
5.15 Geometry of Element LINK8 150
5.16 Geometry of Element SOLID45 150
5.17 Geometry of Element SHELL41 151
5.18 Distribution of integration points 154
5.19 Technique for Solving the Nonlinear Equation 155
5.20 Incremental-Iterative Procedures Full Newton-Raphson procedure 157
6.1 Adopted Descriptions of Curved Beams 160
6.2 Mesh density (Cross Section and Top View) 161
6.3 Effect of Number of Elements on Load-Midspan deflection 162
6.4 Effect of Number of Elements on Load- Midspan Twisting Angle 162
6.5 Reinforcing Steel Bars Modeling 163
6.6 CFRP Laminates Arrangement of Tested Concrete Curved Beams 163
6.7 Boundary Conditions and Applied Loads Arrangements 164
6.8 Load-Midspan Deflection Curves for Control Beam 166
6.9 Figure (6.9) Load-Midspan Twisting Angle Curves for Control Beam 166
6.10 Load-Midspan Deflection Curves for SCB.Eo Beam 167
6.11 Load-Midspan Twisting Angle Curves for SCB.Eo Beam 167
6.12 Load-Midspan Deflection Curves for SCB.Esr Beam 168
xv
6.13 Load-Midspan Twisting Angle Curves for SCB.Esr Beam 168
6.14 Load-Midspan Deflection Curves for SCB.Ecfrp Beam 169
6.15 Load-Midspan Twisting Angle Curves for SCB.Ecfrp Beam 169
6.16 Load-Midspan Deflection Curves for SCB.Mo Beam 170
6.17 Load-Midspan Twisting Angle Curves for SCB.Mo Beam 170
6.18 Load-Midspan Deflection Curves for SCB.Msr Beam 171
6.19 Load-Midspan Twisting Angle Curves for SCB.Msr Beam 171
6.20 Load-Midspan Deflection Curves for SCB.Mcfrp Beam 172
6.21 Load-Midspan Twisting Angle Curves for SCB.Mcfrp Beam 172
6.22 Load-Midspan Deflection Curves for SCB.Io Beam 173
6.23 Load-Midspan Twisting Angle Curves for SCB.Io Beam 173
6.24 Load-Midspan Deflection Curves for SCB.Isr Beam 174
6.25 Load-Midspan Twisting Angle Curves for SCB.Isr Beam 174
6.26 Load-Midspan Deflection Curves for SCB.Icfrp Beam 175
6.27 Load-Midspan Twisting Angle Curves for SCB.Icfrp Beam 175
6.28 Load-Midspan Deflection Curves for FCB.P Beam 176
6.29 Load-Midspan Twisting Angle Curves for FCB.P Beam 176
6.30 Load-Midspan Deflection Curves for FCB.Mo Beam 177
6.31 Load-Midspan Twisting Angle Curves for FCB.Mo Beam 177
6.32 Load-Midspan Deflection Curves for FCB.Msr Beam 178
6.33 Load-Midspan Twisting Angle Curves for FCB.Msr Beam 178
6.34 Load-Midspan Deflection Curves for FCB.Mcfrp Beam 179
6.35 Load-Midspan Twisting Angle Curves for FCB.Mcfrp Beam 179
6.36 SCB.Io Beams With Variable Radiuses 182
6.37 Load-Midspan Deflection Curves for SCB.Io Beams With Variable Radiuses 183
6.38 Load-Midspan Deflection Curves for SCB.Io Beams With Variable Opening 184
Dimensions
6.39 Load-Midspan Twisting Angle Curves for SCB.Io Beams with Variable 185
Opening Dimensions
Length
6.41 Load-Midspan Twisting Angle Curves for SCB.Io Beams With Variable 187
Opening Length
Height
6.43 Load-Midspan Twisting Angle Curves for SCB.Io Beams with Variable 189
Opening Height
6.44 Load-Midspan Deflection Curves for SCB.Io Beams with Different Wrapping 190
Schemes
6.45 Load- Midspan Twisting Angle Curves for SCB.Io Beams with Different 191
Wrapping Schemes
6.46 Load-Midspan Deflection Curves for SCB.Mo Beams with Different 191
Wrapping Schemes
6.47 Load- Midspan Twisting Angle Curves for SCB.Mo Beams with Different 192
Wrapping Schemes
xvi
Notation
Most commonly used symbols are listed below, these and others are defined
where they appear in the research;
Symbol Description
As Area of steel.
d Effective depth.
Ec Concrete modulus of elasticity.
Es Steel modulus of elasticity.
ƒc ̀ Nominal concrete compressive strength (cylinder test).
ƒy Yield strength of steel reinforcement.
ƒs Strength of steel reinforcement
fu Ultimate strength of steel reinforcement.
fsp Splitting tensile strength of concrete.
fyv(or Fvw) yield strength of transverse steel (stirrups)
ƒt Concrete Tensile strength of splitting test.
ƒr Concrete modulus of rupture.
f1 Biaxial crushing stress of concrete under the ambient
hydrostatic stress state
f2 Uniaxial crushing stress of concrete under the ambient
hydrostatic stress state
( f cb ) Ultimate biaxial compressive strength of concrete
h/d Depth of opening/ effective depth
(I1 ) First stress invariant
(J 2 ) Second deviatoric stress invariant
n Number of stirrups.
Pᵤ Ultimate load
Pcr Cracking load
R Radius of the arch
r Radius of the curve beam
wf Width of the composite laminate
w/c Water/cement.
wc Air-dry unit weight of concrete.
X shorter dimension of a rectangular section
X1 shorter center-to-center dimension of closed rectangular
stirrup
y longer dimension of a rectangular section
y1 longer center-to-center dimension of closed rectangular
stirrup
xvii
z distance between the plastic centroids of the chord members
α angle of inclination of diagonal bars (stirrups) or web
reinforcement to longitudinal axis
ε Strain.
ε cu Ultimate strain of concrete
ε˳ Strain corresponding to the maximum compressive stress of
concrete
x,y,z Cartesian coordinates.
υ Poisson's ratio.
βc,β˳ Shear transfer coefficient for closed and opened crack.
σ Stress
σ xp Principal stress in the x – direction
σ yp Principal stress in the y – direction
σ zp Principal stress in the z – direction
εf Strain in the fiber
σa h Ambient hydrostatic pressure
σh Hydrostatic stresses
Abbreviations
Abbreviations Description
ACI American Concrete Institute
ASTM American Society for Testing and
Material
ANSYS Analysis System program
AFRP Aramid Fiber Reinforced Polymer
CFRP Carbon Fiber Reinforced Polymer
DOF Degrees Of Freedom
Exp. Experimental
Eq. Equation
FEM Finite Element Method
FE Finite Element
FRP Fiber Reinforced Polymer
GFRP Glass Fiber Reinforced Polymer
H.H Hinge – Hinge support system
H.R Hinge – Roller supports system
HM High Modulus
HS High Strength
Max. Maximum
xviii
Min. Minimum
No. Number
N-R Newton-Raphson
pp. From page … to page …
rebar Reinforcing bar
RC Reinforced Concrete
3D Three-Dimensional
L.O.I Loss on ignition
I.R Insoluble residue
L.S.F Lime saturation factor
xix
Chapter One Introduction
CHAPTER ONE
INTRODUCTION
1.1 General
Reinforced concrete horizontally curved beams are extensively used in
many fields, such as in the construction of modern highway intersections,
elevated freeways, the rounded corners of buildings, circular balconies,…etc
Figure(1-1). In the construction of modern buildings, network of pipes and
ducts, is necessary to accommodate essential services like water supply,
sewage, air-conditioning, electricity, telephone, and computer network.
Usually, these pipes and ducts are placed underneath the beam soffit and, for
aesthetic reasons, are covered by a suspended ceiling, thus creating a dead
space. Passing these ducts through transverse opening in the floor beams will
reduce the dead space and results in a more compact design as shown in
Figure (1-2). For small buildings, the saving of dead spaces may not be
significant, but for multistory buildings, any saving in story height multiplied
by the number of stories can represent a substantial saving in total height,
Figure (1-1) Applications of Curved Beams.
1
length of air-conditioning and electrical ducts, plumbing risers, walls and

partition surfaces, and overall load on the foundation (Mansur, 2006).
(a)
(b)
Figure 1.2:Typical Layout of Pipes for High Rise Building
(Mansur and Hasnat, 1979).
a-Typical layout of service ducts.
b- Alternative arrangement of service ducts
A horizontally curved beam, loaded transversely to its plane, is

subjected to torsion in addition to bending and shear. Furthermore, it is
obvious that inclusion of openings in beams alters the simple beam behavior
to a more complex one. Due to abrupt changes in sectional configuration,
opening corners are subjecting to high stress concentration that may lead to
cracking unacceptable from aesthetic and durability viewpoints. The reduced
stiffness of the beam may also give rise to excessive deflection under service
load and results in a considerable redistribution of internal forces and
moments in a continuous beam. Unless special reinforcement is provided in
2
sufficient quantity with proper detailing, the strength and serviceability of

such a beam may be seriously affected (Mansur, 2006).
In practice, the most common shapes of openings are circular and
rectangular. Circular openings are required to accommodate service pipes,
such as for plumbing, while rectangular openings provide the passage for air-
conditioning ducts that are generally rectangular in shape.
With regard to the size of openings, many researchers use the terms
“small” and “large” without any definition or clear-cut demarcation line.
Mansur and Hasnat(1979) have defined small openings as those circular,
square or nearly square in shape. Whereas, according to Somes and
Corley(1974), a circular opening may be considered as effected when its
diameter exceeds 0.25 times the depth of the web because introduction of
such openings reduces the strength of the beam. The authors however
consider that the essence of classifying an opening either small or large lie in
the structural response of the beam. When the opening is small enough to
maintain the beam-type behavior or, in other words, if the usual beam theory
applies, then the opening may be termed as small opening. In contrast, large
openings are those that prevent beam-type behavior to develop(Mansur,
2006; Weng et al, 1999).
For small openings, two different failure modes are identified. These
types of failure may be labeled as ''beam-type'' failure and ''frame-type''
failure, respectively, and required separate treatment for complete design. In
contrast, for large openings the beam-type behavior transforms into a
Vierendeel action(Figure 1-3) as the size of opening is increased. Since the
behavior of a beam depend on the size of opening, small and large openings
need separate treatment in design (Mansur, 1998).
The effects of introducing an opening on the overall response of a
beam may be summarized as follow (Mansur and Hasnat, 1979):
3
Figure (1-3) Collapses Mechanism at Large

Opening(Mansur and Hasnat,1979)
1- Introduction of an opening in the web of a beam leads to early
diagonal crack, and the load at first crack decreases with an increase
in either the length or the depth of opening.
2- Unless additional reinforcement is provided to restrict the growth of
cracks, the opening corners are liable to exhibit wide cracking.
3- When the same amount and scheme of reinforcement is used, an
increase in the opening size either by increasing the length or the
depth of opening decreases the strength as well as stiffness of the
beam. The eccentricity of opening, however, has only a marginal
effect on both strength and stiffness.
4- The chord members above and below the opening behaves in a
manner similar to the chords of a Vierendeel panel with
contraflexure points located approximately at mid-span of the
chords. Final failure occurs by the formation of a mechanism with
four hinges in the chords, one at each corner of the opening as
shown in Figure (1-3).
4
1.2 General Guidelines for Selection the Size and Location of

Openings.
The following guidelines had been proposed to facilitate the selection
of the size and location of web opening as shown in figure (1-4):
1. For T-beams, openings should preferably be positioned flush with the
flange for ease in construction. In the case of rectangular beams,
openings are commonly placed at mid-depth of the section, but they
may also be placed eccentrically with respect to depth. Care must be
exercised to provide sufficient concrete cover to the reinforcement for
the chord members above and below the opening. The compression
chord should also have sufficient concrete area to develop the ultimate
compression block in flexure and have adequate depth to provide
effective shear reinforcement.
2. Openings should not be located closer than one-half the beam depth D
to the supports to avoid the critical region for shear failure and
reinforcement congestion. Similarly, positioning of an opening closer
than 0.5D to any concentrated load should be avoided.
3. Depth of openings should be limited to 0.5 D.
4. The factors that limit the length of an opening are the stability of the
chord members, in particular the compression chord, and the
serviceability requirement of deflection. When the opening gets
bigger, it is preferable to use multiple openings providing the same
passageway instead of using a single opening.
5. When multiple openings are used, the post separating two adjacent
openings should not be less than 0.5D to insure that each opening
behaves independently.
5
Figure(1.4) Guidelines for Location of Web Opening for straight beam

(Tan and Mansur,1996)
1.3 Reinforced Concrete Curved Beams.

A horizontally curved beam may be defined as a curved girder having
an out of plane curvature, and supported at its ends by two or more supports.
The shape of the curved beam may be circular, elliptical or parabolic and
sometimes it is made up by circular arcs of several and different radii or/and
centers. It may be subjected to out of plane forces: shear, flexural and
torsional forces.
Horizontally curved beams are loaded transverse to their planes, and
are subjected to torsion, bending, and shear, Figure (1.5). Therefore, a
special feature of the analysis and design of such beams is the necessity to
include torsional effects. In a reinforced concrete structure, cracks form even
at service load. Such cracking causes local reduction in torsional and flexural
stiffness’s resulting in redistribution of internal forces. Furthermore openings
will increase cracks spatially at their corners because of the sudden change in
shear, flexural and torsional stiffness (Mansur and Tan , 1999).
6
Figure(1.5) Horizontally Curved Beam Loading and Forces
1.4 Analysis for Ultimate Strength Under Combined Torsion

Bending and Shear
The analysis of any solid beam subjected to shear, bending and torsional
moments are based on the well-known skew-bending theory for torsion in
concrete beams. In the case of a solid beam, the theory considers three basic
failure modes classified as Mode 1, Mode 2, and Mode 3, according to the
location of concrete compression zone near the top, side, and bottom of the
beam, respectively, as shown in Figure( 1-6). Aspect ratio of the beam
section, relative proportion of top and bottom longitudinal steel, and the ratio
of applied torque to bending moment with or without a combination with
transverse shear generally govern the failure modes. When a relatively small
transverse opening is introduced through the beam web, and the beam is
subjected to predominant torsion, one would expect no changes in the mode
7
of failure. This has been confirmed experimentally by several investigators

(Hasnat and Akhtaruzzaman, 1987; Mansur and Hasnat, 1979; Mansur and
Paramasivam, 1984; Mansur et al., 1983b). As a result, an analysis similar to
that of a prismatic beam is applicable to beams containing a small opening.
With this contention, the analysis is based on three failure modes classified
as Mode 1, Mode 2, and Mode 3. Since the opening represents the potential
source of weakness in a beam, the failure surface is assumed to be traversed
through the center of the opening. In developing strength equations for such
beams, each failure mode is considered separately, and the following
assumptions are made to simplify the problem (Tan and Mansur,1999):
Figure( 1-6) Failure surfaces for a solid beam in Modes 1, 2, and 3.

(Tan and Mansur,1999)
8
1) The pattern of reinforcement in the vicinity of the opening consists

of longitudinal bars above and below the opening, full-depth stirrups
close to either side of the opening, and closed stirrups at the throat
section (above and below the opening), in addition to the normal top
and bottom reinforcement in the solid section.
2) The spacing of stirrups at the solid cross section as well as at the

throat section is uniform along the length of the beam.
3) Failure occurs on a warped plane. The boundaries of the warped plane

are defined on the three sides of the beam by a spiral crack and on the
fourth side by a compression zone that joins the ends of the spiral
crack.
4) The crack defining the failure plane on three sides of the beam
consists of three straight lines spiraling around the beam at a constant
angle.
5) The concrete outside the compression zone is cracked and carries no

tension.
6) All reinforcement crossing the failure plane outside the compression

zone yields at failure.
7) Any reinforcement in the compression zone and dowel action of

reinforcement is ignored.
9
1.5 Failure Modes Under Combined Torsion, Bending, and

Shear with Small Opening
A structural member is, in general, subjected to torsion, bending,
shear, and axial forces, but when the analysis and design of a beam is
concerned, the effect of axial load is usually ignored. Combined torsion,
bending, and shear, therefore, may be considered as the general loading case
for beams, and this loading case is treated first.
• Mode 1 failure
The assumed failure surface for Mode 1 is shown in Figure(1-7) together
with the internal forces developed in steel reinforcement. The skewed
compression zone at the top makes an angle θ 1 to the normal cross section. It
originates from the tip of the tension crack, which traverses through the
opening on one of the side faces of the beam. The resultant of the forces
developed in the longitudinal bars intercepted by the tension zone is Fs, and
the force in the bottom legs of stirrups that cross the failure surface is F wh ·
Similarly, F wv and F’ wv are the forces in the vertical legs of the long and
short stirrups, respectively. Let the forces Fs and Fwh be located at Ys and
Y wh , respectively, below the center of compression and X 1 be the width of
stirrups.
Figure (1.7) Mode 1 failure surface for beams with a small

opening(Tan and Mansur,1999)
10
Take moments about axis A-A, which passes through the center of the
compression zone. The moment about A-A of the applied vectors M, T, and
V acting at the midsection of the opening (Hasnat and Akhtaruzzaman,
1987) must be equal to the moment of forces in steel reinforcement. The
resulting expression may be obtained as follows:
𝑏 2 + 𝑏ℎ
𝑇. 𝑠𝑖𝑛𝜃1 + 𝑀. 𝑐𝑜𝑠𝜃1 + 𝑉. � � 𝑠𝑖𝑛𝜃1
2𝑏 + 4ℎ
1
= 𝐹. 𝑐𝑜𝑠𝜃1. 𝑦𝑠 + 𝐹𝑤ℎ . 𝑠𝑖𝑛𝜃1. 𝑦𝑤ℎ + (𝐹𝑤
2
𝑏(𝑏 + ℎ) (1-1)
+ 𝐹 ′ 𝑤) �1 − � 𝑥 . 𝑠𝑖𝑛𝜃1
𝑥1 (𝑏 + 2ℎ) 1
The three expressions on the right-hand side of Eq. (1-1) represent the
contributions of longitudinal steel, the bottom legs of stirrups, and the
vertical legs of stirrups, respectively. However, forces in the vertical legs of
the stirrups have lever arms about A-A so small that their contributions to
the resisting moment for Mode 1 failure may be neglected. Dropping the
third expression on the right-hand side and dividing the resulting equation by
cos θ 1 , eq. (1-1) thus reduces to:
𝑇. 𝑡𝑎𝑛𝜃1 + 𝑀 + 𝑉𝜇. 𝑡𝑎𝑛𝜃1 = 𝐹𝑠. 𝑦𝑠 + 𝐹𝑤ℎ . 𝑦𝑤ℎ 𝑡𝑎𝑛𝜃1 (1-2)
where
𝑏2 +𝑏ℎ
𝜇=� �
(1-3)
2𝑏+4ℎ
Figure( 1-8) shows a developed diagram of the line PQRS (Figure 1.7)
of the failure surface, where PO is the zone of compression and QR, RS, and
SP are the cracks on the other three faces. It may be seen that the number of
stirrup legs intersected at the bottom of the beam by the crack is
approximately (x 1 tan(β1/ s)), where x 1 is the width of stirrups, s is the stirrup
11
spacing, and β1 is the inclination of the failure crack for Mode 1. Referring
to figure( 1-7), it can be shown that β1 is related to θ1 by:
𝑏.𝑡𝑎𝑛𝜃1 𝑡𝑎𝑛𝜃1
𝑡𝑎𝑛𝛽1 = = (1-4)
𝑏+2ℎ 1+2𝛼
where α = h / b, and b and h are the overall width and depth,

respectively, of the solid cross section of the beam. If A w denotes the area of
one leg of the stirrups and f yw is its yield strength, then
𝐴𝑤 .𝑓𝑦𝑤
𝐹𝑤ℎ = 𝑥1 𝑡𝑎𝑛𝛽1 (1-5)
𝑠
Figure (1-8) Geometry of the boundary of failure surface
After some modification, assumptions and derivations, the strength in

failure Mode 1, which will be denoted by T1, is obtained by the following
quadratic equation:
2𝑀𝑜1 𝐾1 1 1 1
𝑇1 = �� + − � (1 − 6)
∆ 𝐾1 (𝜓∆)2 𝜓∆
where ψ= T/ M, λ= M/ V,
12
1 𝐴𝑤. 𝑓𝑌𝑤 𝑥1 . 𝑦1
𝐾1 = � � (1 − 7)
1 + 2𝛼 𝑆 𝑀𝑜1
and
𝜇
∆= 1 + (1 − 8)
𝜓𝜆
where :
x 1 is the width of stirrups
A w is the area of one leg of the stirrups
• Mode 2 failure
The failure surface for this mode is illustrated in Figure(1-9). In this
failure mode, the compression zone is located along a lateral side of the
beam making an angle θ2 with the normal cross section. It is assumed that
the longitudinal steel is concentrated at the corners and the lever arm of the
force in this steel is x1. The equation of moments about B-B is given by
Figure( 1-9) Mode 2 failure surface for beams with a small opening
13
𝑥1
𝑇. 𝑠𝑖𝑛𝜃2 + 𝑉. = 𝐹𝑠 . 𝑐𝑜𝑠𝜃2. 𝑥1 + 𝐹 ′ 𝑤𝑣 . 𝑠𝑖𝑛𝜃2. 𝑥1
2
𝑏
+𝐹𝑤ℎ � � 𝑥 . 𝑠𝑖𝑛𝜃2 (1 − 9)
ℎ + 2𝑏 1
Similar to Mode 1 failure, the lever arm of the force F wh about B-B is
very small, and, hence, its contribution to the resisting moment may be
ignored. Also, the first term on the right-hand side of eq. (1.9) represents
approximately the lateral flexural strength, M o2 , of the beam at the opening
section. Thus, eq. (1.9) reduces to:
𝑥1
�𝑇 + 𝑉. � 𝑡𝑎𝑛𝜃2 = 𝑀𝑜2 + 𝐹′𝑤𝑣 . 𝑡𝑎𝑛𝜃2. 𝑥1 (1 − 10)
2
It may be shown that the inclination, 𝛽2, of the failure crack for Mode 2
is related to the angle θ2 by:
ℎ. 𝑡𝑎𝑛𝜃2 𝑡𝑎𝑛𝜃2
𝑡𝑎𝑛𝛽2 = = (1 − 11)
ℎ + 2𝑏 1 + 2/𝛼
and that
𝐴𝑤. 𝑓𝑦𝑤
𝐹′𝑤𝑣 = 𝑦′. 𝑡𝑎𝑛𝛽2 (1 − 12)
𝑆
in which y' is the total length of vertical legs of short stirrups at throat
section on one face of the beam.
Inserting the values of tanθ2 and F' wv from Eqs. (1.12) and
(1.11),
respectively, into Eq. (1.10), we get
𝑥1 2
�𝑇 + 𝑉 � �1 + � 𝑡𝑎𝑛𝛽2
2 𝛼
𝐴𝑤. 𝑓𝑦𝑤 2
= 𝑀𝑜2 + 𝑥1 . 𝑦′1 . �1 + � 𝑡𝑎𝑛2 𝛽2 (1 − 13)
𝑆 𝛼
Minimization of T with respect to tanβ2 yields

14
𝑥1 𝐴𝑤. 𝑓𝑦𝑤
�𝑇 + 𝑉 �=2 𝑥1 . 𝑦′1 . 𝑡𝑎𝑛𝛽2 (1 − 14)
2 𝑆
Substitution of Eq. (1.14) into Eq. (1.13)

𝑥1 2
�𝑇 + 𝑉 � �1 + � 𝑡𝑎𝑛𝛽2 = 𝑀𝑜2 (1 − 15)
2 𝛼
The strength equation for Mode 2 is obtained by eliminating tan β2
from Eqs.(1.15) and (1.14) as:
2𝑀𝑜1
𝑇2 = �𝑅 . 𝐾 (1 − 16)
1+𝛿 2 2
in which δ= x 1 .V/2T , R 2 =M o2 /M 01 , and

1 𝐴𝑤. 𝑓𝑦𝑤 𝑥1 . 𝑦1
𝐾2 = 2 � � (1 − 17)
1+ 𝑆 𝑀𝑜1
𝛼
• Mode 3 failure
The analysis for Mode 3 is very similar to that for Mode 1. In this case,
the skewed compression zone is at the bottom instead of at the top (see
figure(1-7)). The equations for Mode 1 can be used to derive the equations
for Mode 3 by turning the beam upside down and taking M = -M, V = -V
and β1 =β3. eqs. (1.7) and (1.8) thus, respectively, become:
𝑇 − 𝑉𝜇 = 2 𝑥1 . 𝑦1 . 𝑡𝑎𝑛𝛽3 (1 − 18)
𝑆
and
𝑇 − 𝑉𝜇
(1 + 2𝛼)𝑡𝑎𝑛𝛽3 − 𝑀 = 𝑀𝑜3 (1 − 19)
2
The term M o3 is the pure flexural strength in negative bending. , the
strength in Mode 3 (being denoted by T3) is obtained by as:
15
2𝑀𝑜1 . 𝐾1 1 𝑅3 1
𝑇3 = � −� + � (1 − 20)
∆′ 𝜓∆′ 𝐾1 (𝜓∆′)2
in which R3 = M o3 / M o1 , and Δ' = μ /(ψ.λ) -1
• Shear-compression mode of failure

The three modes of failure, namely Mode 1, Mode 2, and Mode 3, as
described above, may be termed as flexural type of torsional failure.
Although the failure surface is skewed and warped, it has the general
characteristics of the failure surface in pure flexure. As a result, the analysis
is very similar to that for bending. These modes of failure usually occur
when the beam contains adequate stirrups such that the main steel yields, and
the full flexural strength of the beam is reached when loaded to failure.
In the case of beams containing inadequate stirrups, the concrete
compression zone may shear through prior to yielding of the main steel and
this will precipitate failure at a load below the corresponding flexural failure
load. This type of failure may be called Shear-compression mode of failure.
A detailed analysis to predict the strength of the beam failing in this mode is
rather too complex for practical use and is hardly justified because of the
limited test data available. For beams without an opening, it was found
empirically (Collins et al., 1968) that the possibility of a shear type failure
could be checked in a single step by introducing an "equivalent shear" V eq ",
as given by
1.6
𝑉𝑒𝑞 = 𝑉𝑢 + 𝑇 (1 − 21)
𝐵 𝑢
In which Vu and Tu are the factored shear and torsion, respectively, at
the section under consideration and B width of section. The shear
compression strength of the beam can then be evaluated by means of the
16
shear strength equation for a section using Veq instead of Vu to account for
torsional effects. This procedure has been found to give results well on the
conservative side.
A similar equation has been proposed by Hasnat and Akhtaruzzaman
(1987) to evaluate the shear compression strength of a beam containing a
small opening. It is given as:
1.2
𝑉𝑒𝑞 = 𝑉𝑢 + 𝑇 (1 − 22)
𝐵 𝑢
1.6 Design for Torsion

In the preceding presentation of the skew-bending theory for torsion in
concrete beams containing a small opening, the failure surface for a
particular mode has been considered to pass through the center of the
opening and encroach the solid part of the beam. Observed in many torsion
tests (Hasnat and Akhtaruzzaman, 1987; Mansur and Hasnat, 1979; Mansur
and Paramasivam, 1984), these failure modes are basically identical to those
of a beam without an opening, and, hence, may be termed as "beam-type"
failure. A careful examination of the equations derived reveals that only the
reinforcement in the solid section outside the opening participates in carrying
the external load. When sufficient rebars are used to prevent failure to occur
in these modes, then there is a possibility that the failure may precipitate in
the members above and below the opening. This type of failure is shown in
Figure (1-10). It is similar to the "frame-type" shear failure of a beam with
small openings and will be referred as "frame-type'' failure. In this type of
failure, the entire applied actions are resisted, independent of the solid part of
the beam, by the members framing the opening, and, hence, require a
separate treatment in design
17
Figure (1-10) Frame-type failure of a beam with a small opening

under torsion
1.6.1 Design for Beam type Failure

The modes of failure considered in the skew-bending theory are all
considered as beam-type failure. These failure modes form the bases of the
torsional design provision in the Australian Code, AS 1480 (1974), in which
inclusion of the effect of torsion has essentially been reduced to usual
flexural and shear design procedures. According to the code, the cross
section is first proportioned on the basis of Mode 1 failure. Checks are then
made and, if necessary, modifications are introduced to ensure that the beam
will not fail in Modes 2 or 3, or the shear-compression mode. Steel
percentages are also limited to guard against a primary crushing failure.
Since the strength equations for different modes of failure remain basically
similar when a small opening is introduced, and they are found to agree very
well with reported test data (Hasnat and Akhtaruzzaman, 1987; Mansur and
Hasnat, 1979; Mansur and Paramasivam, 1984), the same approach with
minor modifications for the inclusion of opening may be used for designing
such beams. The design steps involved to account for the various possible
modes of failure under predominant torsion are described in sequence as
follows.
• Mode 1 failure
Equations similar to those given in the Australian Code AS 1480 (1974) can
be obtained directly from the following equations :
18
𝑇 + 𝑉𝜇 = 2 𝑥1 . 𝑦1 . 𝑡𝑎𝑛𝛽1 (1 − 23)
𝑆
and
𝑇 − 𝑉𝜇
(1 + 2𝛼)𝑡𝑎𝑛𝛽3 + 𝑀 = 𝑀𝑜1 (1 − 24)
2
According to the Code, the value of tan β1 is chosen as :

2
𝑡𝑎𝑛𝛽1 = (1 − 25)
√1 + 2𝛼
Substitution of this value into Eqs. (1.23) and (1.24) gives:

(𝑇 + 𝑉𝜇)√1 + 2𝛼 = 4 𝑥1 . 𝑦1 (1 − 26)
𝑆
(𝑇 + 𝑉𝜇)√1 + 2𝛼 + 𝑀 = 𝑀𝑜1 (1 − 27)
In a design situation, the factored bending moment, torsional moment, and

the shear force at the center of the opening are known. Thus, M = Mu, T =
Tu, and V = Vu. Designating the first term of Eq. (1.27) as M eq(1) , the
equivalent moment due to torsion and shear in Mode 1 failure, that is:
𝑀𝑒𝑞(1) = (𝑇 + 𝑉𝜇)√1 + 2𝛼 (1 − 28)
the required strength in positive bending becomes

𝑀𝑜1 = 𝑀𝑒𝑞(1) + 𝑀𝑢 (1 − 29)
The designer chooses a value of (α=h/b) if the sectional dimensions

are not given or known in advance, evaluates Meq(1), and then finds the
value of Mo1. The section and the longitudinal reinforcement must be
designed for this moment using the normal flexural design procedure. The
transverse reinforcement is obtained from Eq. (1.26). Introducing a capacity
reduction factor, Ø, the following equation is obtained:
19
𝐴𝑤 1 𝑀𝑒𝑞(1)
= � � (1 − 30)
𝑠 𝜙 4𝑥1 . 𝑦1 . 𝑓𝑦𝑤
• Mode 2 failure
In the Australian Code (1974), the design equations were derived by
assuming that tan β2 is given by the following equation:
2
𝑡𝑎𝑛𝛽2 = (1 − 31)
�1 + 2/𝛼
Substitution of Eq. (1.31) for tanβ2 in Eq. (1.15) yields:
𝑀𝑜2 = 𝑀𝑒𝑞(2) (1 − 32)
in which M eq(2) , the equivalent moment due to shear and torsion for Mode 2
failure, is given by:
𝑥1
𝑀𝑒𝑞(2) = �𝑇𝑢 + 𝑉𝑢 � (1 − 33)
2
AS 1480-197 4 suggests that if Mu < 0.5 M eq(2) , the cross section of

the beam and the area of longitudinal reinforcement should be such that the
beam can withstand an equivalent bending moment, M eq(2) , as given above in
lateral bending. Also, the area of web steel should not be less than that given
by:
𝐴𝑤 1 𝑀𝑒𝑞(1)
= � � (1 − 34)
𝑠 𝜙 4𝑥1 . 𝑦1 . 𝑓𝑦𝑤
• Mode 3 failure
Taking tan β3 = tan β1 which is given by Eq. (1.25), upon substitution,
Eq. (1.19) then reduces to
𝑀𝑜3 = 𝑀𝑒𝑞(3) − 𝑀𝑢 (1 − 35)
in which
𝑀𝑒𝑞(3) = (𝑇𝑢 + 𝑉𝑢 𝜇)√1 + 2𝛼 (1 − 36)
20
is the equivalent moment due to torsion and shear in Mode 3 failure.

Thus, if the numerical value of Mu is greater than M eq(3) , there is no
possibility of a Mode 3 failure. Physically it means that any tensile stress at
the top of the beam induced by M eq(3) is canceled by the compression due to
Mu. No top steel is, therefore, required. However, nominal steel comprising
at least one bar at each of the top two corners must be provided for
anchorage of stirrups.
If Mu< M eq(3) , there will be residual tension at the top of the
beam, and top steel should be introduced according to the usual flexural
theory to withstand a negative bending (that is, one of opposite sign to
Mu) of magnitude (M eq(3) -Mu).
• Shear-compression mode of failure

Similar to the Australian Code (1974) approach, Eq. (2.69) may be used
to preclude a shear-compression mode of failure of a beam containing a
small opening. Thus, the equivalent shear for this mode is calculated as:
1.6
𝑉𝑒𝑞 = 𝑉𝑢 + 𝑇 (1 − 37)
𝐵 𝑢
The transverse reinforcement is then designed to resist this equivalent

shear on the basis of the normal shear design provisions by assuming that the
failure plane passes through the center of opening. If the steel area, Aw, thus
determined is greater than that already found during the previous design
steps, the larger quantity should be adopted.
1.6.2 Design for Frame Type Failure

This type of failure occurs when the members above and below the
opening are not adequately reinforced for the actions being transmitted
through them. In the case of combined bending and shear, the applied shear
may be assumed to be shared by the chord members in proportion to their
21
cross-sectional areas. Similarly, the applied torque, that produces lateral

shear stresses in the chord members, may be assumed to be resisted by the
couple formed by the resultant of these stresses, as shown in Figure (1-11).
Figure (1-11) Idealized free-body diagram at opening of a beam under loading
The above assumption may be justified from the work of Mansur et al.
(1983). They have assumed that for a beam containing an opening, the
applied torque is resisted by torsion in each member and by the couple
formed by the lateral shear and shown analytically that the torsional
component becomes smaller as the length of opening is decreased and
eventually becomes negligible in comparison to the latter component (couple
formed by the lateral shear) when the opening reduces to square (or circular)
in size. Thus, referring to Figure (1-10), the lateral shear, Vz, may be
assumed to be given by
𝑇𝑢
(𝑉𝑧 )𝑡 = (𝑉𝑧 )𝑏 = (1 − 38)
𝜃𝑡 + 𝜃𝑏
With the usual mechanism for applied bending moment, the problem
of combined torsion, bending, and shear for frame-type failure thus reduces
22
to designing each chord member at the opening for shear in two directions,
as shown in Figure (1.10).
1.7 Fiber Reinforced Polymer FRP (Fiber Reinforced-

Composite)
A composite is a combination of two or more materials (reinforcing
laments, fillers, and matrix binder) with different form or composition
which, when combined into a material system, exhibit properties which are a
combination of its individual components. The matrix can be a ceramic,
metal, or polymer. Fillers may be mineral or metallic powders. Reinforcing
can be particles, fibers, rods, or bars. On the other hand, reinforced concrete
is a composite consisting of steel reinforcement, sand and gravel fillers, and
a Portland cement matrix.
1.7.1 Fiber Materials

Several materials are available for the fibers, e.g. glass, aramid and
carbon. Almost 95 percent of all applications for strengthening purposes in
civil engineering are by carbon fibers (Al-Tai, 2010). Figure(1-12),
demonstrates some typical response of uniaxial loaded fiber materials and
steel. HM and HS are abbreviations of high modulus of elasticity and high
strength, respectively. Fibers have a linear elastic behavior until failure
which is brittle.
The fibers are what make the FRP strong and there are three things
that control the mechanical properties of the FRP (Carolin, 2003):
23
Figure (1-12): Stress-Strain Relationship of Fibers and

Steel (Carolin, 2003)
1. Constituent materials
As mentioned earlier there is a wide array of different materials to be

used. What is important to remember is that the choice of fiber materials
determines, together with choice of polymer, what kind of quality, properties
and behavior the FRP finally will obtain.
2. Fiber amount
Regarding the amount of fiber used in the FRP it is easy to say that the
more fiber used the better properties will be achieved. This is somewhat true
but with too high fiber content there will be manufacturing problems. If the
fibers are tightly packed the matrix will have problems enclosing the fibers
which might deteriorate the FRP.
3. Fiber orientation
The FRP will be stiffer and stronger in the fiber direction. For
example, a rod with all the fibers as very strong in its fiber direction but in
24
perpendicular direction the FRP has not as good properties. A typical FRP
product for the construction industry has therefore an anisotropic behavior
compared to steel which is isotropic.
1.7.2 Matrix
The matrix (the polymer in the composite) is used to bind the fibers
together as shown in Figure(1.13-b), transfer the forces between the fibers
and to protect the fibers from external mechanical and environmental
damage. The shear forces created among the fibers are limited to the
properties of the matrix. The matrix is also the limited factor when applying
forces perpendicular to the fibers. It is important that the matrix has the
capability to take higher strains than the fibers as shown in Figure (1-13-a),
if not there will be cracks in the matrix before the fibers fail and fibers will
be unprotected.
1.7.3 Properties of FRP
1.Non-Corrosive
When dealing with infrastructure, corrosion is a major concern. The
high degree of deterioration can be largely attributed to the corrosive
property of metals. Composites alleviate the problem of corrosion because
they do not rust.
2.Fatigue Life
Because of the nature of the loading on bridge structures, the fatigue
life is of great concern. Cyclic loading from traffic can lead to the distress of
a structure. Most FRPs have an enhanced resistant to fatigue in a certain
loading range as compared to conventional materials.
25
Figure 1.13: (a) Stress-Strain Curves of Fibers, FRP and Matrix (Bisby, 2004)
(b) Typical Composition of FRP Material (Federico, 2001)
3.Non-Magnetic Properties
Steel reinforcement and members often inhibit the design of a
structure when the interference of electromagnetic waves is of concern.
Using FRP in place of the magnetic elements may be a solution to this
problem.
4.High Specific Strength and Stiffness

Two of the most important advantages of using FRP in civil
infrastructure applications are the high specific strength to weight ratio and
high stiffness to weight ratio. With the improvement of strength to weight
ratio over other conventional materials the designer has the ability to use
lower weights and thicknesses. With this in mind, designers are no longer
26
limited by the ratios of conventional materials thus allowing new design

concepts to be explored. By reducing the weight, installation time is reduced
which results in a large reduction in cost. Once, the size of the superstructure
is reduced and this will directly affect the required dimensions and capacity
of the foundation and substructure.
5.Brittle Material Response

FRPs do exhibit definite ductile behavior like metals as shown in
Figure (1-13). The ductile behavior is desirable in a structure to provide
ample warning to occupants prior to failure. However, FRP components can
be designed to exhibit a sequence of damage mechanisms to ensure a
relatively slow failure with extensive deformations.
1.8 Aims of the Study.

The increase in the use of CFRP composite system for strengthening,
repairing and rehabilitation of concrete structure in shear, flexure, torsion
and axial lead to use these system in strengthening the web opening in the
reinforced concrete curved beam. The present study aims to:
1- Investigate experimentally the behavior of continuous reinforced
concrete circular curved beams with and without opening.
2- Investigate experimentally the behavior of reinforced concrete circular
curved beams with openings strengthened by CFRP laminates .
3- Verify the adequacy of the design method suggested for straight
reinforced concrete beam with opening to utilize for the reinforced
concrete curved beam with openings.
4- Carry out finite element technique to analyze the nonlinear behavior of
reinforced concrete curved beams with and without openings
strengthened by CFRP laminates up to failure by using ANSYS
(version 12.1) computer program.
27
1.9 Layout of Thesis

The present study consists of seven chapters as follows:
The current chapter (chapter one) presents a general introduction about
applications and effects of the opening in the reinforced concrete beams and
their effect on behavior of R/C curved beams and properties of FRP fabrics.
Chapter two reviews the most of the previous studies carried out in the field
of the present study. In Chapter three an experimental program, detail
specimens, material properties and test procedure are presented. Chapter
four presents the review and analysis of the experimental results. Chapter
five presents the material modeling used in the finite element analysis, types
of elements used in idealization, in addition to nonlinear solution technique.
Chapter six describes the theoretical results and also a comparison with the
experimental results and parametric study of some variables. Chapter seven
presents a summary of the conclusions of the present work and
recommendations for further work.
28
Chapter Two Literature Review
CHAPTER TWO
LITERATURE REVIEW
2.1 Introduction
Horizontally curved reinforced concrete beams occur frequently as
members in buildings, bridges, and other structures. Methods for the
determination of ultimate loads for straight beams with and without opening
have been the subject of many studies and they are well established.
Ultimate loads and nonlinear behavior for reinforced concrete curved beams
have not been studied in sufficient detail
The aim of this chapter is to present a review for the available
information concerning the behavior of reinforced concrete (horizontally
curved beams and straight beams) with and without opening strengthened
and unstrengthened with CFRP laminates, especially those members which
are subjected to torsion and shear.
2.2 Experimental and Analytical Studies on Reinforced

Concrete Curved Beams.
In 1963, Chu and Thelen preformed a plastic analysis for balcony
girders of circular curved beam with a subtended angle of not more than 180
degree, with a constant cross section , fixed at both ends and subjected to
uniform load perpendicular to the plane of the curved beam. In this analysis,
as in any other plastic analysis, to determine the ultimate load the following
conditions must be satisfied: (a) static equilibrium, (b) yield condition, (c)
flow law, and (d) mechanism and compatibility conditions.
29
A yield surface for combined bending and torsion, at a section where a

plastic hinge was formed, and had been represented by:
m + t  = 1
2 2
(2.1)
Fulfillment of the flow law requirement was achieved through the
relationship:
t
tαn γ  = (2.2)
α  ⋅ m
where
m = M / M P (2.3)
t = T / TP (2.4)
α  = TP / M P (2.5)
M : applied bending moment at cross section.

T : applied torsional moment at cross section.
M P : plastic bending capacity of cross section.
TP : plastic torsional capacity of cross section.
γ  : angle between the axis of rotation at the plastic hinge and the vector
radius.
Figure (2-1) shows the yield surface for combined bending and
torsion, at a section where a plastic hinge was formed.
The general method of analysis was examined, the various
mechanisms of failure were investigated, and the results of the load capacity
of curved beams were presented in a chart to facilitate their application.
30
Figure(2-1) yield surfaces for combined bending and torsion, at a

section where a plastic hinge was formed.
In 1972, Khalifa (mentioned by Jordaan et al. 1974) modified a

plastic analysis for steel girders in the form of a circular arc subjected to
single load and uniformly distributed loads to calculate the collapse load for
reinforced concrete beams curved in plan.
In 1974, Jordaan et al., developed plastic methods of analysis for the
determination of collapse load for reinforced concrete curved beams and
subtending a central angel less than 180 degree. In this study, the plastic
analyses for the cases of a single point load besides uniformly distributed
loads were extended to take into account two concentrated loads placed at
any point on the beam. A plastic hinge was predicted at a particular cross
section of the curved beam if the internal moments at that cross section
satisfy the same yield criterion and flow law mentioned in eq. (2.1) and eq.
(2.2) respectively.
In the experimental part of this study, four curved and six straight
reinforced concrete beams were tested. All the beams had constant moment
31
and torque capacities at any section. The beams were completely fixed at the
supports.
Each curved beam had a radius of (2.21m) and a subtended angle of
(86) degree as shown in figure (2-2). According to amount of reinforcement,
the curved beams were divided into two groups. One curved beam from each
group was tested under single point load and the other curved beams were
tested under two concentrated loads.
B2 B2
6 in
A
A
R=2.21m
B1
B1 Machine Head
86ο
Figure(2-2): Geometry of Test Specimens (Jordaan,1974).
Straight beams were also divided into two groups. These two groups
had similar reinforcement to that of curved beams group. One beam was
tested under bending for each group while the other two beams from each
group were tested under pure torsion, combined bending and torsion
respectively. The purpose of these control tests (for straight beams) was to
determine the flexural and torsional strength for each type of cross section
and to check the interaction relationship given by yield criterion.
32
In this paper, the author concluded that the plastic theory used could
give satisfactory prediction of the mode of failure (the location and type of
plastic hinges), but the prediction value of ultimate load was conservative for
some modes and was slightly higher than experimental ultimate load for
other modes.
In 1977, Badawy et al. formulated two yield criteria to represent the
behavior of a reinforced concrete section under the combined action of
bending, torsion and shear, and the analysis was modified to include the
effect of shear. They tested seven straight and eight curved beams, and the
results were compared with the modified analysis. In the test on curved
beams, the plastic hinge locations and consequently the modes of failure
were recognized from the crack patterns, the deformed shape of the beam
and the measured reinforcement strains. The workers observed the plastic
hinges either torsion-shear hinges, flexural hinges, or bending-torsion-shear
hinges. The analysis of the test results and the comparison with the results
predicated by the plastic theory indicates the following conclusions:
1. The methods of plastic analysis can be applied to reinforced concrete
curved beams.
2. An analysis using the first criterion gives a good predication of the
ultimate load, mode of failure and the internal forces. Whereas an
analysis using the second criterion establishes a lower bound for the
ultimate load and the internal forces. The dimensionless equations for
these two surfaces are :
m2 +t2/(1-v2)=1 (2-6)
for the first yield criterion and
m2 +t2/(1-v)2=1 (2-7)
for the second yield criterion where :
m=M/M p , t=T/T p , v=V/V p ; M, T, and V are the bending moment,
torsion and shear respectively; and M p , T p , and V pare the
33
corresponding plastic capacities of the cross-section in pure bending,

torsion, and shear.
3. Two types of redistribution of internal forces occur in a curved beam,
one due to cracking and the other due to plastic hinge formation.
4. The effect of modes of failure is accurately predicted by the analysis.
In 1977, Badawy et al. presented an experimental study to investigate the
applicability of methods of plastic analysis to reinforced concrete
horizontally curved beams that including shear effect. Eight curved and
seven straight beams were tested. Each curved beam had a radius of (2.21 m)
and a subtended angle of (75) degree as shown in Figure (2-3). Each curved
beam was designed to be subjected to a single concentrated load. Four
curved beams were reinforced identically but were tested under different end
conditions. In the other curved beams, both the transverse reinforcement and
the end conditions were varied. The straight beams were reinforced
according to the reinforcement of the curved beams and were tested under
different combinations of bending, torsion and shear.
Figure(2-3): Geometry of Test Specimens of Badawy et al (Badawy,1977).
34
The results of the straight beams were used to determine the

intersection point of the two yield criteria (proposed in an earlier
investigation) with the bending, torsion, and shear axes and to check the
validity of yield surfaces.
This paper concluded that an analysis using the first criterion gives a
good predication of the ultimate load, mode of failure, and the internal
forces, whereas an analysis using the second criterion might establish a
lower bound for the ultimate load and the internal forces. Besides, there were
two types of redistribution of internal forces that might occur in a curved
beam, one was due to cracking and the other was due to plastic hinge
formation.
In 1978, Hsu et al. tested seven reinforced concrete horizontally
curved beams to investigate torsional and flexural moment redistribution
after cracking and to suggest an appropriate design method based on this
post-cracking behavior. Each beam had a radius of (2.74 m) and a subtended
angle of (90) degree. Each beam was designed to be fixed at both ends and
subjected to a concentrated load at mid span.
The test program for the seven specimens was as following: the first
and second beams were analyzed by the elastic theory based on uncracked
sections and the reinforcement at each section was then designed by the ACI
Building Code (1971). They were tested to check the applicability of
conventional design method. The third and fourth beams were designed for
only a portion of the maximum torsion moment calculated by the elastic
analysis based on uncracked sections. The fifth and sixth beams were
designed by the elastic method based on cracked section. The seventh beam
was provided with uniform longitudinal bars and stirrups throughout its
length to study the failure mechanism of a uniform beam. For cracked
section, the torsional rigidity was determined by the post-cracking torsional
35
rigidity of reinforced concrete member which was derived from space truss
analogy by Hsu (1972).
These tests showed that a significant redistribution of torsional and
bending moments was observed after cracking. Therefore, the conventional
design method based on "Building Code Requirements for Reinforced
Concrete (ACI 318-71)" and elastic analysis assuming an uncracked section
caused premature yielding at the supports. Besides, this paper recommended
that the design of curved beam should be based on cracked section.
In 1981, Abul Mansur and Rangan investigated three different
methods for design of reinforced concrete horizontally curved beams to
study the behavior, strength, mode of collapse, and economy of reinforcing
steel. The design methods investigated were:
1) A collapse load method proposed by Badawy et al. (1977)
2) The classic elastic method based on uncracked sections
3) A limit design method proposed by the researchers themselves.
The limit design method was based on assuming two additional
conditions to obtain the solution for a statically indeterminate system. The
first additional condition was provided by assuming an ultimate design
torsional moment equal to [ 0.33 f ' x 2 y / 3 ] in(Nmm) at critical sections

c 1 1
where x 1 and y 1 are, respectively, the smaller and the larger dimensions of a
rectangular section in (mm), and f c' is the compressive strength of concrete
in (MPa). The second additional condition was furnished by assuming the

position of an inflection point (bending moment is zero) according to elastic
analysis.
Three curved beams designed by these methods were tested. Each
beam had a radius of (2.45 m) and a subtended angle of (90) degree. The
beams were completely fixed at the supports and subjected to a vertical point
load. According to the results of this study, the authors concluded that:
36
1) Redistribution of internal forces in a reinforced concrete curved beam

occurs at two stages: one after cracking, and the other after the
formation of some of the plastic hinges, and
2) The three methods gave satisfactory designs for reinforced concrete
curved beams, but the collapse load method proposed by Badawy et
al, required significantly more steel (especially hoops) than the
conventional elastic design method and the limit design method.
Al-Temeemi (2002) studied the analysis of reinforced concrete

horizontally curved beams on elastic foundations by the finite elements
method. The material nonlinearity was taken into account for concrete and
steel. A twenty-node isoperimetric brick element with sixty degrees of
freedom is employed to model the concrete. The reinforcing bars are
modeled as axial members embedded within the brick element. Soil is
represented by normal and horizontal subgrade reactions. The normal
component is represented by Winkler, Kondner, Polynomial models, while
the horizontal component is represented by Winkler model. The finite
element and the available experimental and numerical results have shown
good agreement.
Parametric studies have been carried out to examine the influence of
some selected parameters (like radius to span-length ratio (R/L), boundary
conditions, α1 (the rate of stress release as the crack widens), α2 (the sudden
loss of stress at instant of cracking), and the type of soil on the overall
behavior of reinforced concrete curved beam on elastic foundations. From
the results obtained according the consider examples, it was found that the
ultimate load of curved beam on elastic foundation could be increased with
increasing the radius to span-length ratio (R/L)[ when (R/L) increases from 1
to 5, the ultimate load increases about 40% . The shear reinforcement bars
have a significant effect on decreasing the ultimate load, it was found that
37
the ultimate load decreases about 51 % when the shear reinforcement was
removed.
Ali A.Y. (2010) studied the analysis of reinforced concrete horizontally
curved deep beams, loaded transversely to its plane, using a three-
dimensional nonlinear finite element model in the pre and post cracking
levels and up to the ultimate load. The 20-node isoperimetric brick element
with sixty degrees of freedom was of Ansys Program employed to model the
concrete, while the reinforcing bars are modeled as axial members embedded
within the concrete brick element. Perfect bond between the concrete and the
reinforcing bars was assumed. The effects of some important numerical in
addition to material parameters had been investigated to study their influence
on the predicted load-deflection curves and the ultimate load capacity.
The numerical study was carried out to investigate the effect of the
central subtended angle, boundary conditions, amount of transverse
reinforcement, and using additional longitudinal bars (horizontal shear
reinforcement) on the behavior of reinforced concrete horizontally curved
beams with different shear length to effective depth ratios (a/d). It was found
that decreasing the central subtended angle causes an increase in the ultimate
load resisted by curved beams. Also, it was found that the effect of the
internal torsion and the amount of transverse reinforcement on the ultimate
load resisted by curved beams was decrease as (a/d) ratio decrease, while the
effect of the flexural moment and using additional longitudinal bars as a
horizontal shear reinforcement was increase when (a/d) ratio decrease.
Al-Mutairee (2013) studied the effects of non-uniform distribution of
longitudinal reinforcements on the behavior of reinforced concrete (RC)
horizontally curved beams with fixed-ends under static loads to product an
optimal strength of these beams without increasing the volume of
longitudinal reinforcement. Three dimensional nonlinear finite element
analyses done utilizing computer program called NFHCBSL, incorporate
38
20-node isoperimetric brick element used to represent the concrete elements

while reinforcing bars are idealized as axial members embedded within the
concrete elements without any relative displacement between them. The
results show that the effect of non-uniform distributions of longitudinal
reinforcement of RC horizontally curved beams with fixed-ends is effective
and can be used to improve the strength of this type of beams and its
importance increases with increasing the angle of horizontal curvature (θ).
2.3 Experimental and Analytical Works on Reinforced

Concrete Beam with Openings.
Many experimental and analytical studies on reinforced concrete beam
with openings were made by several researchers. Burton (1965) studied the
influence of embedded ducts on strength of continuous reinforced concrete T-
beams. He tested two reinforced concrete T-beams (one with ducts and the
other solid). The results showed very little difference in performance of the
two beams. He concluded that the design equations for ultimate strength
contained in ACI-1963 Building code can safely be applied to wide shallow
beams of the type used in his study with opening.
In 1967 Nasser reported tests on reinforced concrete beams with
openings and made the following assumptions:-
1. The top and bottom cross member at the opening are assumed to
behave similar to the chords of a Vierendeel panel.
2. The cross members of the openings, when they are not subjected to
transverse load, have centraflextural points at their mid spans.
3. There is a diagonal stress concentration at the corners induced by the
chords shear leading to a force whose magnitude is twice that of the
simple shear force.
39
Depending on the mentioned assumptions, the researcher tested nine

rectangular beams. His conclusions from the test results of the beams were:
a- Large opening in rectangular reinforced concrete beams behave similar to
Vierendeel panel.
b- Adequately reinforced large opening in rectangular beams do not result in
reducing the ultimate capacity of the beams, but it reduces its stiffness.
Dalal(1969), studied the behavior and design of reinforced concrete
T-beams with an opening in the web, and he proposed an empirical equation
for calculating the area of reinforcement around the opening and established
linear relationships between the ultimate shear carried by special web
reinforcement and the size of opening.
𝟏 𝑽𝒖𝟐 (𝟎.𝟏𝟖+𝟐.𝟓∝) 𝒂�
𝒅−𝟑.𝟕𝟖
• Av= × × (𝟏 − ) (2.8)
𝑸 (𝒔𝒊𝒏 𝜽 ×𝟎.𝟔ƒ𝒚) 𝟓.𝟏𝟔
Where:-
Av: area of special web reinforcement around the opening.
Q = 0.85
Vu2: Ultimate shear force at tie center of the opening.
α: Ratio of the length of the opening to the span of the beam.
a: Distance between the support and the load for the test specimen.
Ɵ: Angle of inclination of web reinforcement with the horizontal.
𝑎� : Shear-span ratio.
𝑑
Vu1= Vu (0.18+ 2.5 α) (2.9)
where:-
Vu1= Ultimate shear carried by the web reinforcement.
Vu= Total ultimate shear.
Hanson(1969), tested a series of longitudinally reinforced T-beams
representing a typical joist floor. The specimens contained square openings
and were tested to simulate the joist on either side of a continuous support.
Many parameters were used in this study, but the main ones were the size
and horizontal and vertical locations of the opening. However, as the
40
opening represents a source of weakness, the failure plane always passes

through the opening, except when the opening is very close to the support so
as to bypass the potential inclined failure plane. Figure (2-4) shows
schematically some typical shear failures of beams containing square and
circular openings.
In 1974, Somes and Corley reported a similar study of Hanson
(1969) but, in this case, the openings were circular in shape. In both cases
(square and circular opening), it was found that an opening located adjacent
to the center stub (support) produced no reduction in strength. As the
opening is moved away from the support, gradual reduction in strength
occurs until it levels off to a constant value. Test data suggest that the
vertical position of opening has no significant effect, while an increase in the
size of opening leads to an almost linear reduction in strength. However,
there appears to be a size of opening below which no reduction in shear
strength occurs. This size corresponds to about 25% of the beam depth for
square openings and 33% of the beam depth for circular openings. They
have also noted that the strength of such a longitudinally reinforced beam
may be fully restored by providing stirrups on either side of the opening.
Figure 2.4 shows schematically some typical shear failures of beams
containing square and circular openings.
To study the empirical equation for reinforcement around the opening
as suggested by Dalal(1969) on the T-beams and to determine the effect of
the variation of shear-span ratio, Jindal (1976) , tested seven rectangular
beams with opening. He conclude the followings:
• The behavior of a beam having an openings provided with special web
reinforcement is similar to the solid beams, and
• The response of the load deflection and moment rotation are similar in
the both cases.
41
• The beam with openings can be designed by providing special

reinforcement around the opening.
Figure(2.4): Typical Shear Failure of a Beam with Small Openings

Containing no Shear Reinforcement(Hanson 1969, Somes and Corley 1974)
To determine a reinforcement scheme suitable around the opening to

restore the strength of a beam to the level of a corresponding solid beam,
Salam (1977), conducted an investigation on beams with opening of
rectangular cross section tested under two symmetrical point loads. He found
that:
• A short stirrup in the members both above and below the opening are
necessary to eliminate the weakness due to the provision of opening, in
addition to the longitudinal reinforcement above and below the opening
and full depth stirrups by its sides.
• When sufficient reinforcement is provided to prevent a failure along a
diagonal crack passing through the center of the opening and traversing
the entire depth, then the failure is precipitated at the minimum section.
42
In such a case, formation of two independent diagonal cracks in the

members above and below the opening split the beam into two separate
segments. Figure (2-5) shows the sketch of the cracking pattern of the
beam that failed in this manner.
Figure (2.5): Shear Failure of a Beam at the Throat Section by Salam (1977).
In 1978 Jamal, tested eleven reinforced concrete beams with opening.

His study concentrated on the examining the influence of an opening in the
flexural zone and the shear zone, the position of the opening through the
depth of section and the influence of additional reinforcement on the
capacity. It was found that an opening in the flexural zone dose not reduce
the flexural capacity of the beam up to (h/d) ratio equal to (2/3), where h is
the depth of the opening and (d) is the effective depth of the beam. However,
the opening did reduce the stiffness of the beam and its energy absorption
capacity. When the opening was situated in the shear zone a significant
reduction in capacity resulted. Also, if it is not properly reinforced can lead
to a very premature type of failure can occur. He presented methods of
evaluating the capacity of beams with an opening in any zone and also
generalization to the case of uniformly loaded beams.
Mansur and Hasnat(1979), tested twenty-two reinforced concrete
beams with small openings under torsion. The beams were divided into three
groups according to nominal concrete strengths. In the first and the third
groups, investigations were made for a 76mm hole size, whereas in the
second group four different hole sizes of 51mm, 76mm, 102mm, and 127mm
43
in diameter were used to study the effect of opening size. All beams were
provided with both longitudinal and transverse reinforcement. Torsion load
was applied in increments. After application of each load increment the load,
rotation, and strains were recorded and cracks, if any, were marked. The
theoretical predictions with ACI Code 318-77 are found to be in close
agreement with the available test results
Mansur and Paramasivam(1984), tested ten beams, each containing
a symmetrically placed transverse circular opening in bending and torsion.
The beams were divided into two groups designated as CA and CB,
according to their cross-sectional dimensions which were 175 × 350mm and
200 × 400mm respectively. The amount and arrangement of reinforcement in
each group was kept constant. Group CA consisted of three specimens with
different size openings. These specimens were tested under pure torsion. For
group CB, opening size was held constant, but the torsion to moment ratio
was varied from pure torsion to pure bending. The load was applied in
increments. After each increment, steel strains, deflections, twists, and
maximum crack width were recorded. Torsion strength and stiffness of a
beam decreases with an increase in opening size. It was concluded that, the
presence of small amount of bending moment increases the torsion capacity
of a beam. But for substantial bending moment the increase in bending
moment leads to a drop in beam torsional capacity. This is in agreement with
the findings of earlier investigators.
Mansur et al.(1985), tested twelve beams designed by the proposed
method of the ACI Code 318-83, under a point load to study the flexural
behavior of reinforced concrete beams with large rectangular openings that
were subjected to bending and shear. The major variables were the length,
depth, eccentricity and location of openings, and the amount and
arrangement of corner reinforcement. He conclude that at a particular load
both the maximum crack width and maximum beam deflection increase with
44
an increase in opening length, opening depth, or moment shear ratio at the

center of the opening. Also, diagonal bars as corner reinforcement were
found to be more effective in crack and deflection control than vertical
stirrups. Based on test results, a suitable quantity of corner reinforcement
was recommended.
Mansur et al(1991), tested eight reinforced-concrete continuous
beams each containing a large transverse opening. The beams are rectangular
in cross section and all contain the same amount and arrangement of
longitudinal reinforcement. The number of spans, the size of opening, and its
location along the span are considered as major variables. Test results
indicate four distinctly different stages of behavior in the load-deflection
curve of a continuous beam. Final failure of the beam occurs by the
formation of a mechanism, and the two opening ends represent the most
vulnerable locations for the development of plastic hinges. Besides early
cracking, the strength and stiffness of the beam decrease with an increase
either in the length or depth of opening. Similarly, openings located in a high
moment region produce larger deflections and result in early collapse of the
beam.
To study the deflection behavior of reinforced concrete beams with
web openings, Mansur et al(1992), tested twenty-two beams (14 simply
supported, 3 two-span continuous, and 5 three-span continuous beams), each
containing a large rectangular opening. The direct stiffness method were
used analyze the beams. The beams were treated as structural members
comprising several segments, and an equivalent stiffness had been derived
for the segments traversed by the openings. A comparison of the
experimental results with the predicted deflections obtained from the direct
stiffness method had good agreement. They concluded that, the method also
gives a reasonably accurate picture of the redistribution of internal forces
and moments in continuous beams due to incorporating opening.
45
Tan et al(1996) ,studied the flexural behavior of reinforced concrete

T-beams with and without large web opening in positive and negative
moment regions. Fifteen beams were tested under a point load. The tests had
indicated the following :
• The presence of web openings leads to a decrease in the cracking and
ultimate strength as well as the post-cracking stiffness of continuous
beams.
• Performance of a beam with multiple opening is more desirable in
terms of strength and serviceability. The thickness of the post between
the adjacent openings should not less than one-half of the overall beam
depth.
Tan and Mansur(1996), suggested useful procedure for complete
analysis and design of reinforced concrete beams with large web opening.
Also they suggested the guidelines to facilitate the selection of the size and
location of web openings. Generally, the following should be considered:
• Openings should be positioned so that chords have sufficient concrete
area to develop the ultimate compression block in flexure and adequate
depth to provide effective shear reinforcement.
• They should not be deeper than one-half the beam depth
• Should be located not closer than one-half the beam depth from
supports or concentrated loads.
For structural analysis of reinforced concrete beams with openings, they
showed that in the case of a statically determinate beam, shear force and
bending moment envelopes can be obtained from statics. For continuous
beams, they suggested method can be followed, that is, the member
containing an opening is considered as a nonprismatic beam with different
cross sectional properties: those of a solid section and of the equivalent
section for opening segments. They gave the recommended design process
46
for the opening segment which is based on the observed Vierendeel behavior
of chord members at an opening.
The behavior and design of a beam containing a transverse opening
and subjected to a predominant shear are discussed by Mansur (1998).
Based on the observed structural response, some guidelines are suggested to
classify the opening as ‘large’ or ‘small’. For small openings, two different
failure modes are identified. These types of failure may be labeled as ''beam-
type'' failure and ''frame-type'' failure, respectively, and required separate
treatment for complete design. In beam-type failure, a an inclined failure
plane of angle 45°, similar to solid beam may be assumed, the plane being
traversed through the center of the opening, as shown in figure (2-6a).
Frame-Type failure occurs due to the formation of two independent diagonal
cracks, one in each of the chord members bridging the two solid-beam
segments, leads to the failure, as shown in figure (2-6b). In the proposed
method, the maximum shear allowed in the section to avoid diagonal
compression failure has been assumed to be same as that for solid beam
except for considering the net section through the opening.
Figure(2-6): Modes of Failure for Small Opening ,Mansur (1998).
To simulate drilling of holes in an existing beam either for the passage

of service ducts or for the determination of in place concrete strength, Weng
wei et al (1999) tested nine T-beams were fabricated with circular openings
through the web to simulate drilling of holes in an existing beam either for
47
the passage of service ducts or for the determination of in place concrete

strength. The major parameters considered in the study were the size and
location of openings. Test results indicated that an opening, when created
near the support region of an existing beam, leads to early diagonal cracking
and significantly reduces the strength and stiffness of the beam. They
concluded that filling an opening by nonshrink grout, as is usually done for
openings created by removing concrete cores for the determination of in
place concrete strength of an old building, is not adequate to restore the
original strength and stiffness, their study reveals that the weakness
introduced in terms of cracking, deflection, and ultimate strength by creating
an opening in existing beams can be completely eliminated by strengthening
the opening region of the beam using a suitable method, like the use of
externally bonded FRP plates as used here. In other hand they found that
after first shear cracks occurred however, the compression chord tended to
carry a larger proportion of the shear force, and this increased with an
increase in the applied load. The fraction of the applied shear carried by the
compression chord at ultimate ranges is from 65 to 90%.
Neff et al(2002), tested six full-type laboratory beams. They were
conducted in order to analyze the behavior of reinforced concrete beams
with one large rectangular opening. Thus, the shear force distribution and the
location of the contra flexure points have been examined. On the basis of the
finite element computations, a design concept for beams with one large
rectangular opening had developed, which now can be proved by means of
the test results. The results of the test series and the finite element
calculations have been compared and they have shown that the distribution
of the shear force depends on the ratio between the stiffnesses of the chords.
Abdulla et al.( 2003), tested ten reinforced concrete beams with
openings and with CFRP sheets. The experimental program included
strengthening five beams with openings, while four beams kept without
48
strengthening and the last one was solid beam used as control beam. It was
found that, when the opening with height of 0.6 from the beam depth reduce
the capacity by 75%. The application of the CFRP sheet around the opening
was greatly decreased the beam deflection, controlled the cracks around the
openings and increased the ultimate load capacity of the beam. The failure
occurred due to a combination of shear cracking of concrete and bond failure
of CFRP sheets glued to concrete surface. Figure (2-7) explain internal steel
reinforcement and types of external CFRP for strengthening.
Figure(2-7): Details of the tested beams: (a) internal steel reinforcement

and (b) types of external CFRP strengthening.(Abdulla,2003)
Mohammed(2004), used a three-dimensional nonlinear finite element

model to investigate the behavior of reinforced concrete beams with large
transverse opening under torsion. The 20-node isoparametric brick elements
have been used to model the concrete. The reinforcing bars were idealized as
axial members embedded within the concrete element and perfect bond
between the concrete and the reinforcement has been assumed to occur. In
49
general good agreement between the finite element solutions and the
experimental results has been obtained. The finite element solutions revealed
that the collapse torque and post-cracking tensional stiffness decrease with
the increase of length or depth of the opening.
Al-Kaisy(2005), investigated the behavior of reinforced concrete
beams with large transverse openings subjected to flexure by using a three-
dimensional nonlinear finite element model. Several parametric studies have
been carried out to investigate the effect of some important finite element
and material parameters. The effect of concrete compressive strength,
amount of longitudinal tensile reinforcement and opening size on the load-
deflection response has been investigated. In general good agreement
between the finite element solutions and the experimental results has been
obtained. The finite element solutions revealed that the ultimate load and
post-cracking stiffness decrease with the increase of length or depth of
opening.
Akinobu Suzuki et al(2007), tested ten reinforced concrete beams
with and without opening for the purpose of comparison. The beams were
prepared into presence or absence of web opening, method of reinforcement
of beam with web-openings. Difference between beam without shear
reinforcement, beam with shear reinforcement only, and beam with shear
reinforcement and stirrups, concrete compressive strength: Two levels of
concrete design strength (Fc)–36 N/mm2 and 60 N/mm2 and method of load
application: Difference between multi-cyclic loading and monotonic loading.
They discussed Arakawa’s formula for evaluating shear strength was applied
to the beams without opening and Hirosawa’s formula was applied to the
beams with an opening. They found that the non-reinforced specimens with
openings showed a marked decline in shear resistance compared with the
other specimens. In contrast to this, the specimens provided with shear
reinforcement and reinforcement near the openings were measured to have a
50
higher shear resistance than the calculated shear resistance (Hirosawa’s

formula).They concluded that provide an additional stirrups around the web
opening improved shear resistance
Najim(2009), tested eight reinforced concrete deep beams with cross-
section of (100x750mm) and total length (1150mm) were tested under two-
point loads. Seven specimens had two rectangular openings with dimensions
(100x200 mm), one in each center of the shear span, placed symmetrically
about the centerline of the beam. The experimental results indicated that the
use of CFRP sheet to upgrade the R.C. deep beams with web openings has
significant effect on overall behavior such as the ultimate load, crack width
and deflection. The percent of increase in the ultimate load capacity was
about (100-190) %.
In the other side, three dimensional finite element analysis was used
to investigate the performance of the R.C. member strengthened by CFRP
laminate. ANSYS computer program (version 9.0, 2004) was utilized
through this study. The comparison between the experimental and theoretical
results showed reasonable agreement and asserted the validity of the
numerical analysis and methodology developed in this study.
Bandyapadhya et al (2011), discusse the use of Glass Fiber
Reinforced Polymer (GFRP) to strengthen and rehabilitate. In this
experiment ten beams with length of 2.0 m for each one, one solid as
reference beam and other nine beams categorized as beams with
openings (strengthened and rehabilitated beams) are tested. The opening size
of these beams is increased from (100x100) mm to (100x300) mm and with
different locations. They concluded that GFRP can be used to strengthen and
rehabilitate the beams with small opening only. Also, the beams with
openings only show the maximum deflection at a point which is in between
the middle point of beam and middle point of opening instead of
maximum deflection at central point of solid beam.
51
Ali et al(2011), made an experimental study to clarify the possibilities and

drawbacks of using composite materials for the purpose of strengthening or
retrofitting concrete structures. There general goal is divided into two aims.
First one is to investigate the effect of the shape and dimensions of opening
on the behavior, and the second is to examine the effectiveness of CFRP
reinforcement in enhancing the flexural capacity of RC beams with opening
at the flexural region. Nine beams with length of 1.0 m total length and (150
by 100) mm cross section dimensions were tested. The opening position was
fixed at the mid span of the beams for all sizes and shapes. All beams were
tested as a simple beam subjected to two concentrated loads. They conclude
that the flexural strengthening of R.C. beams with CFRP is effective with
increment in deflection of about (310,300 and 360)% for beams with (L/h)
1.8, 5 and 20 respectively, but it is reduce the ultimate load of about (35,16
and 15)% respectively. On the other hand, tests have shown that the CFRP
elongation is a major factor that could affect shear crack propagation).
Al-Dolaimy(2011),tested eight continuous reinforced concrete beams
with cross-section of (150x250mm) and total length (3300mm) with two
spans (clear span is 1500 mm) were tested under two-point loads. Six beams,
each beam contains one opening, three of which have dimensions (200x100
mm) and the three other have dimensions (140x140mm). The location of
opening was in the zone of maximum moment and maximum shear.
The experimental results indicated that the use of CFRP sheet to
upgrade the R.C. continuous beams with web openings has significant effect
on overall behavior such as the ultimate load, crack width and deflection.
The percent of increase in the ultimate load capacity was about (60-106) %.
While the inclined strengthening model (strengthened continuous beam with
opening 200X100mm) shows a stiffer response compared with other
strengthened model, which gives 106% increase in ultimate load than
(control unstrengthened continuous beam with opening 200X100mm) and 13
52
% more than (strengthened continuous beam with opening 200X100mm).

The inclined stirrups with angle of 70˚ are better than the stirrups with right
angle because the stirrups with70˚ is perpendicular on the crack path. Also,
the use of CFRP as externally strengthening system decrease the crack
width, but, it cannot change the formation of inclined crack. All the tested
beams with opening failed due to splitting the beam into two pieces though
the inclined crack.
In the other hand, three dimensional finite element analysis was used
to investigate the performance of the R.C. member strengthened by CFRP
laminate. ANSYS computer program (version 9.0, 2004) was utilized
through this study. The comparison between the experimental and theoretical
results showed reasonable agreement and asserted the validity of the
numerical analysis and methodology developed in this study.
Hamza (2012), investigates the behavior and performance of
reinforced concrete arch beams with and without openings, unstrengthened
and strengthened (externally by CFRP laminates or internally by steel
reinforcement).
The experimental work consists of fabrication and testing of twelve
reinforced concrete arch beams. Eight of these beams had rectangular
openings. The experimental variables considered in the test program
included: curvature forces, location of opening through profile of arch,
present of (internal strengthening steel reinforcement or external
strengthening CFRP laminates) for opening. The beams were tested under
two points loading at top edges with (hinge-roller) supports at bottom.
The experimental results showed a significant decrease in ultimate
load capacity by about 38 % which were registered for arch beam without
confining stirrups to resist forces in radial direction (curvature). The mode of
failure is splitting failure, it occurs suddenly with no advance warning of
distress. The use of CFRP laminates as external confinement to resist
53
curvature forces which gives a little increase in ultimate load capacity by

about 2.7% , consequently, the ductility ratio decreased rapidly by about
60%, as compared with control arch. The presence of opening at zone of [
pure bending, combined of(bending, shear force and axial compression
force) and excessive compressive force ( near the support)] leads to decrease
in load capacity about 58%, 60% and 42%, respectively.
Patel and Saksena (2013), Investigate the effect of small circular
opening on the shear and flexural and ultimate strength of beams. The main
factors of the test are the changes of diameter and the position of opening. In
this investigation five beams using C20 concrete and Fy415 steel were
casted and tested. First beam was solid and was used as reference for
comparison with other beams with an opening. Second beam had opening of
110mm (0.55D) at L/8 distance, third beam had opening of 90mm (0.45D) at
L/8 distance. Beam number four and beam number five had openings as
mentioned above at L/4 distance. The tested beams have been loaded as
simple beam with two concentrated and symmetrical load. They conclude
that when the diameter of opening increased, the reduction of ultimate
strength increased and patterned of cracking as well as mode of
failure of the beam changed. They recommended usage of diagonal
reinforcement and stirrups in top and bottom of opening for increasing the
ultimate shear strength of the beam. They also concluded that the most
critical position of opening to reach the ultimate strength in beams is near the
support and also the best place for the location of opening in these
beams is in middle of a beam (flexure zone).
2.4 Concluding Remarks

The experimental and theoretical studies concerning the behavior of
concrete beams with opening can be summarized as following:
54
• Many theoretical studies about the reinforced concrete curved beams

were not complete.
• Numerous experimental and theoretical studies concerning the behavior
of concrete beams with opening have been carried out. The majority of
these studies dealt with the straight members like simply supported
beams, continuous beams and deep beams with openings. Few studies
were about the nonlinear behavior of reinforced concrete curved beams.
The overall response of a straight beam and arch beam with opening
and horizontally concrete curved beams as reported in the above
literature may be summarized as follows:
1.The presence of web opening leads to decrease in both cracking and
ultimate load, as well as post cracking stiffness of beams.
2. Unless additional reinforcement is provided to restrict the growth of
cracks, the opening corners are liable to exhibit wide cracking.
3. Performance of multiple opening is more desirable in terms of strength
and serviceability. The thickness of the post between adjacent openings
should not be less than one- half of the overall beam depth, and the post
should be adequately reinforced to avoid premature failure.
4. The increase in the opening size either by increasing the length or the
depth of opening decreases the strength as well as stiffness of the beam.
The eccentricity of opening, however, has only a marginal effect on both
strength and stiffness.
5. Specifying the type of the opening (large or small) is depending on the
type of structural response of the beam.
6. A large rectangular opening behaves similarly to Vierendeel-panel at the
opening segment. Under combined bending and shear, the chord members
bend in double curvatures with contraflexure points located approximately
at mid span of the chords. Final failure occurs by the formation of a
55
mechanism with four hinges in the chords, one at each corner of the
opening.
7. The application of the CFRP sheet around the opening generally decreased
the beam deflection, controlled the cracks around the openings and
increased the ultimate load capacity of the beam and changes the failure
mode of the beam.
It's well-known from the literature mentioned above there is no available
technical data about horizontally curved reinforced concrete beam with
openings. The present work enhance the knowledge of overall behavior of
reinforced concrete circular curved beams with and without opening,
strengthened and unstrengthened by internal reinforcing steel or external
CFRP laminates through experimental work and numerically by three
dimensional finite element method (ANSYS.12.1 package).
56
Chapter Three Experimental Work
CHAPTER THREE
EXPERIMENTAL WORK
3.1 General
Although the use of reinforced concrete curved beams leads to
several advantages in architectural plans, the fabrication of laboratory
models of such type of construction seems to be relatively difficult, costly
and time consuming process. In spite of this fact, an experimental
investigation was carried out in this study to establish the structural
behavior of fourteen specimens of reinforced concrete curved beams with
and without opening, strengthened and unstrengthened under effect of
two or four point loads.
Firstly, the experimental work will be described, which includes
the main variables; types of curved (geometry) beam, location of opening
through profile of curved beams (effect of combination of internal forces)
and, presence of internal strengthening by steel reinforcement and
external strengthening by CFRP laminates around opening. Details of
casting the specimens, strengthening by internal reinforcement and
externally with CFRP, testing procedure and measuring instruments are
also presented in this chapter. Then, standard tests are presented in this
chapter to determine the properties of concrete and steel reinforcement
used in this study.
3.2 Description of Test Specimens
The experimental program includes two groups of specimens, the
first group consists of ten semicircular continuous curved beams of two
spans simply supported at external ends and roller at middle support, with
57
and without web openings. The second group consists of four closed
circular (ring) beams of four roller supports, with and without web
openings.
Beams of the first group have an inner diameter 2000 mm and
outer diameter 2250 mm, and having cross section of dimensions 250
mm overall depth and 125 mm width, as shown in Figure (3-1). These
beams were tested under the effect of two point loads located at mid of
each span (angle 45°). Steel reinforcement (2Ø12)mm deformed bars
were provided for top and bottom for positive and negative moment
regions with clear cover of 25 mm. The closed stirrups of Ø6 mm
reinforcing bar were placed at angle 4.5o along the beam length (see
Appendix A ).
P/2 (125x50x8)mm Ø6 @ 4.5°
2Ø12
(125x50x8)mm P/2
250 mm
(125x50x8)mm
(125x50x8)
mm 2Ø12
125 mm
(125x50x8)mm
(a) (b)
(c)
1.125 m
Figure(3-1) Details for Group I of Tested Curved Beams

a) Geometry and Load , b) Cross Section and Reinforcement , c) Front View
58
Beams of second group have an inner diameter 950 mm and outer

diameter 1200 mm, and had a cross section of dimensions 250 mm
overall depth and 125 mm width as shown in Figure (3-2). These beams
were tested under the effect of four point loads located at mid of each
span of the beam (of angle 45°) at top surface. Steel reinforcement
(2Ø12)mm deformed bars were provided for top and bottom for positive
and negative moment regions with clear cover of 25 mm. The closed
stirrups of Ø6 mm reinforcing bar were placed at angle 11o along the
beam length
All curved beams of the two groups had openings with dimensions
of (100*200 mm) and each group had a control beam without opening
(solid). Table (3.1) illustrates identification, geometry and details of
reinforcement around openings.
P/4 Ø6 @ 11°
P/4 P/4
2Ø12
(125x50x8)mm
250 mm
P/4 2Ø12
125 mm
(a) (125x50x8)mm (b)
(c)
Figure(3-2) Details for Group II of Tested Ring Beams

a) Geometry and Load , b) Cross Section and Reinforcement , c) Front View
59
3.2.1 Semicircular Curved Beams (Group I)

3.2.1.1 Semicircular Curved Beams without Opening
This type consists of two curved beams. One of them represents the
pilot beam of a semicircular curved beam and the other represents a
control beam of the same type. Pilot beam is similar to control beam, and
it was used to test the devices and loading test system which will be used
for the test of other semicircular curved beams.
3.2.1.2 Semicircular Curved Beams with Opening

This type consists of nine semicircular curved concrete beams.
Three curved beams are with two opening each located at distance ‘ d/2 ‘
from each exterior support Figure (3-3-a), the other three curved beams
are with two opening each located at distance ‘ d/2 ‘ from point load
toward interior support (3-3-b), and the last three curved beams with two
opening each located at distance ‘ d/2 ‘ from both sides of interior
support (3-3-c). The first type of opening without any strengthening, the
second type of opening was strengthened with internal steel
reinforcement stirrups and the last type strengthened with CFRP
laminates as illustrated in Table 3.1.
3.2.2 Ring Curved Beams (Group 2)

3.2.2.1 Ring Curved Beams without opening
This type consists of one ring beams represents a control beam of
the second group (full circular beams). It is used as a reference beam in
which results of other ring beams will be compared with it. This ring
beam has no opening at any part of its length.
3.2.2.2 Ring Curved Beams with opening
This type consists of three full curved concrete beams (ring
beams). These full curved beams are with opening spaced at distance ‘d/2
‘ from applied load. One of them without strengthened; the other one was
strengthened internally using steel reinforcement (stirrups), while the last
60
one strengthened externally using CFRP laminates. Identifications for

each of these full curved beams and details of reinforcement around
opening are illustrated in Table 3.1.
Table 3.1 Descriptions of Tested Specimens

Specimen Location of Details of Reinforcement External CFRP Laminates
Group
No.
Designation Opening Around Opening Around Opening

SCB.P --- --- ---
SCB.Eo Near Exterior --- ---

Support
Near Exterior 3Ø6for each cord,
SCB.Esr Support 1Ø6 diagonal bar for each corner, ---
1Ø6 at each side
SCB.Ecfrp Near Exterior --- 1of 20mm width on each side
Support 3of 20mm for each cord
SCB.Mo Near Applied
--- ---
Load
Group 1
Near Applied 6Ø6for each cord,

SCB.Msr Load 2Ø6 diagonal bar for each corner, ---
2Ø6 at each side
SCB.Mcfrp Near Applied 1of 25mm width on each side
---
Load 3of 20mm for each cord
SCB.Io Near Interior
--- ---
Support
Near Interior 6Ø6for each cord,
SCB.Isr Support 2Ø6 diagonal bar for each corner, ---
2Ø6 at each side
SCB.Icfrp Near Interior ---
1of 25mm width on each side
Support 3of 20mm for each cord
FCB --- --- ---
FCB.Mo Near Applied

--- ---
Load
Group 2
Near Applied 6Ø6for each cord,

FCB.Msr Load 2Ø6 diagonal bar for each corner, --
2Ø6 at each side
FCB.Mcfrp Near Applied ---
1of 25mm width on each side
Load 3of 25mm for each cord
(*) definitions of symbols Subscript (i) denotes one of the following :
SCB = Semi-circular curved beam o = no strengthening for opening location
FCB= Full circular curved beam sr = strengthening using steel reinforcement
P = Control beam cfrp = strengthening using cfrp laminates
Ei = Exterior location of opening
Mi = middle location of opening
Ii= Interior location of opening
61
3.3 Strengthening System (schemes)

Strengthening system was chosen carefully according to crack
pattern and failure mode. The method of design adopted for strengthening
technique had been suggested by (Mansur, 1998) for straight beam. The
design specification of ACI 318-2011 and ACI Committee 440-2002 was
satisfied for steel bars reinforcement and CFRP laminates, respectively.
3.3.1 Internal Strengthening by Internal Reinforcement
Curved beam SCB.Esr was strengthened by one full depth closed
stirrups Ø6 mm on each side of the opening and three pairs of deformed
bars of Ø6 mm closed stirrups for top and bottom chords of opening and
one diagonal bars Ø6 mm for each corner of opening.
Curved beams SCB.Isr, SCB.Msr and FCB.Msr were strengthened
by two full depth closed stirrups Ø6 mm on each side of the opening and
six pairs closed stirrups of Ø6 mm for top and bottom chords of opening
and two diagonal bars Ø6 mm for each corner of opening. Layouts of
steel reinforcement scheme for all tested curved beams internally
strengthened by steel reinforcement are shown in Figure (3-3).
3.3.2 External Strengthening by CFRP Laminates

Beam SCB.Ecfrp was strengthened by two full wrap of CFRP
laminates of 0.131 mm thickness and 20 mm width at each side of the
opening and three pairs of 20 mm for both top and bottom chords of
opening . Beams SCB.Mcfrp and SCB.Icfrp were strengthened by pair of
full wrap of CFRP straps of 0.131mm thickness and 25 mm width at each
side of the opening and three pairs of 0.131mm thickness and 20mm
width for both top and bottom chords of opening. Beam FCB.Mcfrp was
strengthened by pair of full wrap of CFRP straps of 0.131mm thickness
and 25 mm width on the each side of the opening and three pairs of 20
62
mm width for both top and bottom chords of opening. Details of

externally strengthening by CFRP laminates for all tested curved beams
are shown in Figure (3-4). More details about the method of
strengthening of beams are illustrated in Appendices A&B.
-1Ø6mm full depth

stirrups at each side
of opening
-1Ø6mm at each
corner of opening
-3Ø6mm stirrups for

each cord of opening
-2Ø6mm at full
depth stirrups at
each side of opening
-2Ø6mm at each
corner of opening
-6Ø6mm stirrups at
-2Ø6mm at full
depth stirrups at
-2Ø6mm at each
corner of opening
-6Ø6mm stirrups at
-2Ø6mm at full
depth stirrups at
-2Ø6mm at each
corner of opening
-6Ø6mm stirrups at
Figure (3-3) Details For Internally Strengthening by Steel Reinforcement

Around Opening
63
(a
(b
(c
FCB.Mcfrp (d
Figure (3-4) External Strengthening Around Openings by CFRP Laminates
64
3.4 Material Properties of Tested Specimens

3.4.1 Concrete
The materials used in producing concrete are locally available
materials, which include cement, natural gravel, natural sand and water.
These materials were tested in laboratories of Engineering College of
Karbala University.
3.4.1.1 Cement
Ordinary Portland cement (Iraqi manufacturing) named Tasloga
was used throughout this investigation for casting all the specimens. The
cement was kept in air-tide plastic containers to avoid exposure to the
atmosphere. This conforms to Iraqi Standard Specification No. 5:1984 as
illustrated in Table 3.2.
Table 3.2: Chemical and Physical Test Results of the Cement

Chemical Properties Test Result Limit according to IQS
CaO % 60.21 --------------------
SiO 2 % 19.12 --------------------
Al 2 O 3 % 4.30 --------------------
Fe 2 O 3 % 2.42 -------------------
MgO % 3.57 ≤ 5%
SO 3 % 1.37 ≤2.5%ifC 3 A< 5%
≤2.8%ifC 3 A> 5%
Free Lime % 1.07
Loss on Ignition % 1.5 ≤4%
Insoluble Resdue % 0.87 ≤1.5 %
L.S.F 0.89 0.66-1.02
M.S 2.33 ----------------
M.A 1.54 ---------------
65
Table 3-2 : Continue

Test Result Limit according to IQS
Physical Properties
Setting Time, min, 120 Initial ≥45

240 final ≤600
Fineness (Blaine), 330 ≥230
m2/kg
Compressive
Strength , MPa 21.0 ≥15
3days 31.2 ≥23
7days
3.4.1.2 Coarse Aggregate (Gravel)

A maximum size of 19 mm of crushed gravel from Al-Nibaey was
used in the current study. The gravel was washed and cleaned by water
several times and left to dry in air. Table 3.4 shows the grading of
aggregate and the limits specified by the Iraqi Specification No.45/1984.
3.4.1.3 Fine Aggregate (Sand)

Natural sand from Karbala of maximum size of 4.75 mm was used
in this investigation. Before being ready to use, the sand was washed and
cleaned by water several times, later it was spread out and left to dry in
air to avoid the humidity saturation which may affect the water content
extensively. The grading test results conform to Iraqi specification No.
45/1984 specification. Table 3.5 shows the properties of fine aggregate.
66
Table 3.3 : Grading of Coarse Aggregate
Sieve %, Passing
No. size Limit of Iraqi Specification

Coarse Aggregate
(mm) No. 45/1984
1 19 99 95-100
2 14 96 -
3 10 48 30-60
4 5 1 0-10
5 2.36 0 -
SO 3 content=0.08% (specification requirements up to 0.1%)
Table 3.4:Fine Aggregate Properties

% Limit of Iraqi Specification
No. Sieve size (mm)
Passing No.45/1984,zone (2)
1 10 100 100
2 4.75 97 90-100
3 2.36 82 75-100
4 1.18 67 55-90
5 0.6 51 35-59
6 0.3 23 8-30
7 0.15 2 0-10
SO 3 content=0.29% (specification requirements up to 0.5%)
67
3.4.1.4 Mixing Water

Water used in mixing concrete should be clean and free from
injurious amounts of oils, acid, alkalis, salts, organic materials, or other
substances deleterious to concrete or reinforcement. In engineering
practice, the strength of concrete at a given age and cured in water at a
prescribed temperature is assumed to depend primarily on two factors
only, the W/C ratio and the degree of compaction. In this study, the tap
water has been used for mixing concrete and curing of all spacimens.
3.4.1.5 Mix Design

Several trial mixes were made according to the recommendations
of the ACI 211.1-97. Reference concrete mixture was designed to achieve
normal strength of concrete (about 28 MPa) at (28) days. The mixture
was (1 cement: 2.2 sand: 2.42 gravel: 0.55 water cement ratio, by
weight), and the slump was approximately (60 mm).
Mixture details are given in Table 3.5. It was found that the used
mixture produces good workability and uniform mixing of concrete
without segregation.
Table 3.5: Properties of Concrete Mix

Parameter Normal strength concrete
Water/cement ratio 0.55
Water (kg/m3) 205
Cement (kg/m3) 373
Fine aggregate (kg/m3) 623
Coarse aggregate (kg/m3) 900
68
3.4.1.6 Mixing Procedure

The mixing procedure was as follows:
(1) Before mixing, all quantities were weighted and packed in a clean
container.
(2) The coarse and fine aggregate are stocked and blended in 2/3 of the
required water for 60 seconds in a horizontal rotary mixer with 0.2
m3 capacity.
(3) The cement and rest water were added and mixed for 3 minutes.
(4) The cylindrical molds were filled with concrete into three layers;
each layer was compacted. The concrete was poured into lightly
oiled steel mold in two layers, and each layer was vibrated by
mechanical vibrator(300 rpm), 5 second for each insertion which
gave adequate compaction. The surface of the concrete was leveled
off and finished with a trowel. Then, the specimens were covered
with a nylon sheet to prevent evaporation of water.
(5) All specimens were left in the laboratory until they were removed
from their mold after 24 hours, and then burlap sacks were placed
over the beams and kept wet for (28) days. The same procedure was
followed for the concrete cylinders. After (28) days, they were taken
out of the water and left to dry for (24) hours and then tested in
accordance with the standard specifications.
3.4.2 Reinforcing Steel

Two sizes of reinforcing steel bars were used in the tested beams,
deformed bars of size ( Ø12mm) for main longitudinal reinforcement
(circumference )and deformed bars of size (Ø6 mm) for closed stirrups
and straight pieces around opening as shown in Figure (3-5). Tensile test
of steel reinforcement was carried out on at least three specimens,
69
prepared for each type of the reinforcing steel bars which were used in
the tested beam to determine their tensile properties according to ASTM
C370-2005a. The tensile test was performed in the Central Organization
for Standardization and Quality Control. The main properties are
summarized in Table 3.6 and it is in agree with ASTM A615-86.
Table 3.6: Test Results of Steel Reinforcing Bar

Nominal Measured Yield Ultimate Elongation Modulus of
Diameter Diameter Stress(*) f sy Strength f su Ratio % Elasticity(**)
(mm) (mm) (MPa) (MPa) (GPa)
6 5.72 520 615 9.5 200
12 11.67 550 625 12.5 200
(*)
Each value is an average of three specimens (each 40 cm length).
(**)
Assumed value.
Figure (3-5) Cage of Steel Reinforcement
70
3.4.3 Strengthening System by (CFRP) Laminates

Carbon fiber fabric SikaWrap Hex-230C and epoxy based
impregnating resin Sikadur-330 were used for technique of strengthening
is shown in Figure (3-6). The main properties of impregnating resin
Sikadur-330 and carbon fiber fabric SikaWrap Hex-230C are shown in
Tables 3.7 and 3.8, respectively.
Figure (3-6) CFRP Laminates and Epoxy Resine
Table 3.7 :Properties of Sikadur-330 (Impregnating Resin) (*)

Comp. a: white
Appearance
Comp. b: grey
Density 1.31 kg/l (mixed)
Mixing ratio A : B = 4 : 1 by weight
Open time 30 min (at + 35◦C)
Viscosity Pasty, not flowable
Application temperature + 15◦C to + 35◦C (ambient and substrate)
Tensile strength 30 MPa (cured 7 days at +23◦C)
Flexural E-modulus 3800 MPa (cured 7 days at +23◦C)
(*) Provided by the manufacturer
71
Table 3.8:Properties of SikaWrap Hex-230C (Carbon Fiber Fabric) (*)

Fiber type High strength carbon fibers
0◦ (unidirectional). The fabric is equipped
Fiber orientation with special weft fibers which prevent
loosening of the roving (heatset process).
Areal weight 225 g/m2
0.131 mm (based on total area of carbon
Fabric design thickness
fibers)
Tensile strength of fibers 4300 MPa
Tensile modulus of fibers 238 GPa
Elongation at break 1.8 %
Fabric length/roll ≥ 45.7 m
Fabric width 305 mm
(*) Provided by the manufacturer
3.5 Mechanical Properties of Hardened Concrete

3.5.1 Compressive Strength
During casting of each curved beams, three 150×150×150 mm
cubes and three 100×200 mm cylinders were made. After cleaning and
lubricating the molds, concrete was cast and compacted and then cured
under the same conditions. Cube compressive strength and cylinder
tensile strength were obtained by standard tests ASTM C39-2001, as
shown in Figure (3-7a). The results of each curved beam are listed as an
average in Table (3-9).
72
3.5.2 Splitting Tensile Strength

Splitting tensile strength test was carried out on plain cylindrical
concrete specimens (100 mm x 200 mm) in accordance with ASTM
C496-1996 as shown in Figure (3-7b). Results are presented in table 3.10.
(a) (b)
Figure (3-7) compressive Strength and Tensile Strength Test
Table 3.9 : Test Results of Compressive and tensile Strength
Compressive Splitting Tensile Modulus of

Beam Strength of Concrete
(MPa)(*) Strength ( fct ) Elasticity Ec
Symbol
(MPa) (MPa)(**)
( fcu ) ( fc' )
SCB.P 37.63 30.10 3.4 25786

SCB.Eo 39.38 31.50 3.5 26379
SCB.Esr 37.25 29.80 3.4 25657
SCB.Ecfrp 41.38 33.10 3.6 27040
SCB.Mo 39.25 31.40 3.5 26337
SCB.Msr 40.13 32.10 3.5 26629
SCB.Mcfrp 40.50 32.40 3.5 26753
SCB.Io 38.38 30.70 3.4 26042
SCB.Isr 37.75 30.20 3.4 25829
SCB.Icfrp 36.63 29.30 3.1 25440
73
Table 3.9 Continue

FCB.P 37.13 29.70 3.4 25614
FCB.Mo 37.50 30.00 3.4 25743
FCB.Msr 41.50 33.20 3.6 27081
FCB.Mcfrp 40.38 32.30 3.6 26711
(*)
ƒ̀c= 0.8 ƒ̀cu
(**)
(Ec=4700 fc' )
3.6 CFRP Laminates Installation

The effectiveness of strengthening or rehabilitation with externally
bonded CFRP laminates depends on the bond between the CFRP and
concrete. To insure such proper bonding the following steps were
followed:
(1) The surface of the concrete were grinded using an electrical hand
grinder to expose the aggregate and to obtain a clean sound
surface, free of all contaminants such as cement laitance, and dirt.
(2) Specimens were cleaned by washing with water and allowed to dry
prior to composite application. This procedure removed loose
particles and contaminations from the specimen's surface.
(3) The corners of the specimens were rounded (radius of
approximately 15 mm) to avoid any stress concentration in the
CFRP at corners of the beams. This stress concentration will lead
to a rupture failure of CFRP at corners before reaching their
ultimate strength.
(4) Apply the mixed resin Sikadur-330 to the prepared substrate using
a trowel or brush in a quantity of approximately (0.7 to 1.2 kg/m2),
depending on roughness of substrate.
(5) The SikaWrap Hex-230C fabric was cut by scissors to strips for
the required width and length for all the specimens.
74
(6) Place the SikaWrap Hex-230C fabric onto the resin with the plastic
roller until the resin is squeezed out between the roving.
(7) As a covering layer an additional resin layer of approximately (0.5
kg/m2) broadcast with the brush can be added, which will serve as
a bonding coat for following cementitious coatings.
(8) After allowing the laminate to cure for several days, the specimens
will be ready to test. All apparent concrete surface beams were
painted white so that crack propagation can be easily detected.
Figure (3-8) shows the procedure of application of CFRP laminates on
concrete element.
Figure 3.8: Application of CFRP System on Concrete Element
3.7 Instrument and Test Procedure

Tests were carried out using 2000 kN hydraulic testing machine
which was manufactured for the Civil Engineering Department of
Engineering College of Karbala University, as shown in Figure (3-9).
This machine was manufactured from a built up steel sections of a
thickness 30 mm with a movable braced support (two plates) of thickness
75
32 mm that could be moved up or down by a scroll fixed at the top of the

machine. These plates are fixed to the frame by two 50 mm steel rods
through holes made for this purpose as shown in Figure (3-10).
The main characteristics of the structural behavior of the beam
specimens were detected at every stage of loading during testing. A dial
gage of 0.01 mm accuracy was used at midspan of the beam and at the
outer and inner edge of the midspan section to measure the rotation at this
section as shown in Figure (3-10). The specimens were placed on the
supports of the testing machine, and then the first readings of the gages
were recorded. After that, the specimens were loaded with a constant rate
of loading. Readings of deflections was recorded at each interval of load
as well as recording the first crack load and the ultimate load
consequently.
Figure 3.9: Loading Machine Used in the Testes
76
(a)
(b)
Loading
l
Dial gauge at
the edges of
midspan
section
support
(c) support Dial gauge at the

center of
midspan section
(d)
Figure (3-10) Instruments Details

a) Semicircular Curved Beam
b) Ring Curved Beam
c) Dial Gauge
d) Dial Gauge Positions
77
3.8 Loading and Support Condition

The supporting system was hinged at exterior ends and roller at the
inner support. The hinged end consists of one smooth stainless steel shaft
of diameter 25mm welded to base plate and two shaft, of diameter 25mm
welded to steel beam (to make grove). A W-shape steel beam was fixed
weld to the supported plates to prevent relative movement and also to
provide a rigid reference beam that hold the supports. The roller support
be made of two shaft with diameter 25mm welded to the base plate and
other steel shaft of diameter 25mm free. Figure (3-11) shows the details
of support conditions.
The test was done by 2000 kN capacity hydraulic jack. All of the
semicircular beams were tested under two point loading, with the load
applied at midspan of each panel, while full circular beams were tested
under four point load applied at the midspan of each panel.
Figure (3-11) Details of the Typical Support Condition

(a) Hinge Support
(b) Roller Support
78
3.9 Steel and wood mold

Two types of molds were used in this study. The first one used for
semicircular beams which is a steel mold from plates of thicknesses
(4mm for the main frame, 3mm for stiffeners and 2mm inner cover plate)
using a CNC-plasma machine, as shown in Figure (3.12). The second
type of molds is a wooden type which was used for the circular ring
beams, a 2.44 m by 1.24m plywood blocks of thickness 10 mm were used
to form the main frame of this type of mold, while 4mm wooden fibers
plates were used to cover the inner surfaces of the molds as shown in
Figure (3.13).
a)
b) c)
Figure (3-12) Semicircular Molds and Tools

a) Steel Mold Layout
b) Steel Mild Photo
c) CNC-plasma machine
79
a)
b)
Figure (3-13) Wooden Circular Molds

a) Wood Mold Layout
b) Wood Mild Photo
80
Chapter Four Experimental Results and Discussion
CHAPTER FOUR
EXPERIMENTAL RESULTS AND DISCUSSION
4.1 General
In this chapter, the general response and observation of the tested
horizontally curved concrete beams with and without openings, and
strengthened with internal reinforcements or externally with CFRP laminates
were reported and discussed. To accomplish this objective, an experimental
program has been performed, as described in chapter three.
Two groups of circular curved beams were tested, first one includes
ten continuous semicircular curved concrete beams with inner radius of 1000
mm and with dimensions of section of 125 mm width and 250 mm height
under effect of two midspan point loads. The second group includes four full
circular beams with inner radius of 475 mm and section dimensions of 125
mm width and 250 mm height under effect of four midspan point loads. Both
of these groups were tested to study the influence of different variables that
were considered: opening location, effect of internal reinforcement by steel
reinforcements (stirrups) and effect of external strengthening by CFRP
laminates.
One of each group of the beams represents the control beam, which
has no opening and without strengthening scheme at any position. The others
have web openings with and without strengthening. Test results are
discussed in this chapter based on load-midspan deflection curves and load-
twisting angle curves to explore the influence of internal reinforcement with
stirrups and diagonal bars and external strengthening with CFRP laminates
on cracking and ultimate loads, crack pattern and failure modes.
81
4.2 Semicircular Curved Concrete Beams (Group I)

This section deals with the overall behavior of ten horizontally
circular concrete beams with and without openings, and strengthened with
internal reinforcement or with CFRP laminates.
4.2.1 Pilot and Control Semicircular Curved Beam (SCB.P)

Two specimens (SCB.P) were constructed, one as a pilot and the other
is a control specimen. The pilot beam is used to check the validity of test
processes supports and load positions, dial gages and its location, as well as
the overall test incremental load applications. The test shows that all devices
work properly and could be used for all other specimens.
The control specimen is a semicircular curved beam without opening
as shown in Figure (4.1). The beam was loaded gradually until the first crack
was observed. The first crack (flexural) appeared in the top face of
maximum negative moment (internal support) at load of about 62.3 kN. At
the same load, a torsional crack appears at both sides with an angle about
45°. As the load was increased further, several flexural and torsional cracks
initiated in the positions between interior support and points of loading,
spatially at tension zones of the beams. Torsional cracks spread rapidly and
began to propagate more than flexural crack with load increments. In the
same manner it can be noticed that width of torsional cracks increased more
than flexural cracks widths during loading. As the load reaches the value of
117.7 kN the first positive flexural crack at bottom face of midspan appears,
while torsional cracks directed to propagate at the compression zone above
the interior support. Finally, torsional failure at ultimate load about 147.2
kN occurred. Figure (4.2), show mode of failure and cracks pattern, while
Figures (4.3) and (4.4) show load deflection curve and load – angle of
twisting curves for midspan, respectively.
82
Figure (4.1) Control Beam SCB.P
a)
b)
Figure (4.2) Mode of Failure and Cracks Pattern for Control Beam at
Failure
a) Interior Support b)Under Midspan Point Load
83
Figure (4.3) Load-Midspan Deflection Curve for Control Beam SCB.P
Figure (4.4) Load-Midspan Twisting Angle Curve for Control Beam SCB.P
84
4.2.2 Semicircular Curved Beam with Opening near External

Support (SCB.E)
4.2.2.1 Unstrengthened Semicircular Curved Beam (SCB.Eo)

The curved beam SCB.Eo includes opening spaced at distance d/2 =
106.5 mm from external support and without any strengthening as shown in
Figure (4.5). The beam deformation was initially within the elastic range at
the early stages of loading. The load was increased gradually until the first
flexural crack occurred at load about 48.5 kN which was observed in the
maximum negative moment position at top face above the interior support.
The diagonal crack of opening began to appear from bottom corner of
opening that near to support at load of 55.4 kN. Torsional cracks start to
appear at load of 58.8 kN at the tension zone and near the maximum
torsional force at mid distance between load and internal support As the load
was increased further, several flexural and torsional cracks initiated in the
positions between the applied loads and internal support. Flexural cracks
moved downward and torsional cracks increased in length and numbers at an
angle of 45°.Also, diagonal cracks increased in length and width, and start
from other corners of opening. As the load was increased further, a torsional
conventional failure mode at the opening position was appeared. The curved
beam SCB.Eo failed at a load of 128.1 kN by forming two diagonal cracks
start from the skew corners of opening, then the cracks growth along top and
bottom reinforcement at an angle of 45° until brittle failure occurs of beam
type failure, as shown in Figure (4.6). The load–midspan deflection and
load–angle of twisting curves are shown Figures (4.7) and (4.8),
respectively.
85
Figure (4.5) Semicircular Curved Beam SCB.Eo
a b
Figure (4.6) Mode of Failure and Cracks Pattern for SCB.Eo Curved Beam
a) Interior Support
b) At Exterior Support
86
Figure (4.7) Load-Midspan Deflection Curve SCB.Eo Curved Beam
Figure (4.8) Load-Midspan Twisting Angle Curve SCB.Eo Curved Beam
87
4.2.2.2 Internally Strengthened Semicircular Curved Beam

(SCB.Esr)
The curved beam SCB.Esr includes opening spaced at distance
d/2=106.5 mm from exterior support, as shown in Figure (4.9), and
strengthened by three pairs of closed stirrups Ø6 mm for both top and
bottom chords of opening and one diagonal bars Ø6 mm for each corner of
opening, also one Ø6 mm stirrups at each side of the opening. In this beam,
cracks were first observed at the top face of maximum negative moment at a
load approximately 50.2 kN. These cracks appeared due to flexural stresses
of bending moment. As the load increased to 58.8 kN a torsional and skew
cracks at the corners of opening were appeared. After more loads increments
flexural and torsional cracks were propagate and widened. In spite of
forming diagonal corner cracks for curved beam SCB.Esr, the beam fails due
to torsional moment at load of 132.0 kN, as shown Figure (4.10). No
propagation of cracks at the corners of opening were recorded because of the
internal reinforcement which prevent cracks from extended, and the beam
failed in a beam type failure mode. The load – midspan deflection and load –
angle of twisting curves are shown in Figures (4.11) and (4.12), respectively.
Figure (4.9) Semicircular Curved Beam SCB.Esr
88
a) b)
Figure (4.10) Mode of Failure and

Cracks Pattern for SCB.Esr Beam
a) Inner Face of Exterior Support

b) Above Interior Support
c) Outer Face of Exterior Support
c)
Figure (4.11) Load-Midspan Deflection Curve SCB.Esr Curved Beam
89
Figure (4.12) Load-Midspan Twisting Angle Curve For SCB.Esr

Curved Beam
4.2.2.3 Externally Strengthened Semicircular Curved Beam

(SCB.Ecfrp)
The curved beam SCB.Ecfrp includes opening spaced at distance
d/2=106.5 mm from external support, as shown in Figure (4.13), and
strengthened by full wrap of CFRP laminates of 0.131mm thickness (pair of
20mm width on each side of the opening and three pairs of 20mm width
for both top and bottom chords of opening ). The beam was loaded gradually
until the first crack was observed and the deformation was initially within
the elastic range at early stages of loading. The first crack was noticed at a
load of 41.5 kN at the tension fibers of maximum negative moment above
interior support and at the down corner of the opening near the support.
Torsional cracks began to appear at the zone between the applied load and
the interior support at an angle of 45° at load of 48.5 kN, also flexure cracks
90
propagated through these increments, while corner cracks at the opening did
not propagate or increased because of the confinements of CFRP laminates.
At load of about 72.7 kN the crack at corner of opening reaches the CFRP
laminates and stop its extension through until failure occurs, while torsional
cracks increased and propagated rapidly and cause the failure of the beam. A
beam type failure mode is occurred due to torsional moment at load of 141.9
kN, as shown in Figure (4.14). The load – midspan deflection and load –
angle of twisting curves are shown in Figures (4.15) and (4.16), respectively.
Figure (4.13) Semicircular Curved Beam SCB.Ecfrp
a) b)
Figure (4.14) Mode of Failure and Cracks Pattern for SCB.Ecfrp Curved Beam
a) Outer Face of Interior Support
b) Inner Face of Interior Support
91
Figure (4.15) Load-Midspan Deflection Curve SCB.Ecfrp Curved Beam
Figure (4.16) Load-Midspan Twisting Angle Curve SCB.Ecfrp Curved Beam
92
4.2.2.4 Summary of Test Results for SCB.E Beams

Table 4.1 shows the cracking load, ultimate load, maximum rotation at
and deflection at midspan and failure mode. Also, Figure (4.17) and (4.18)
show a comparison of load- deflection and angle of twist curves at midspan
for curved beams SCB.P, SCB.Eo, SCB.Esr and SCB.Ecfrp. It can be
concluded, that the presence of openings near exterior support reduce the
ultimate load capacity (compared with control beam) by about 18%, also no
interested changes in deformations were recorded. On the other hand
strengthening of the opening by internal reinforcement or external CFRP
Laminates will increase ultimate load capacity (compared with SCB.Eo) by
about 3% and 11% respectively. Also It could be noticed that torsional
deformation of internally strengthened curved beam increased by about 25%,
while other deformations had no interested change.Results indicates that the
presence of opening in the curve beam reduced its stiffness, while the
strengthening of the opening lead to increase the post cracking flexural and
torsional stiffnesses and behavior of the curved beam.
Table 4.1 Summary of Tested SCB.E Beams
* Max. θ
Cracking Load, kN Ultimate Max. Δ
Ultimate ×10-3 Failure
Specimen Load (mm) at
Corner Flex. Tor. Load, kN (rad) at Mode
Diff. % midspan
midspan
SCB. P -- 62.3 62.3 152.3 18 44.6 20.4 Torsional
SCB. Eo 55.4 48.5 58.8 128.1 -- 44.4 18 Shear
Torsional
SCB. Esr 58.8 50.2 58.8 132.0 3 55.13 20
Shear
SCB. Ecfrp 72.7 41.5 48.5 141.9 11 44.2 17.1 Torsional
Difference = (Pu(Specimen) − Pu(SCB. Eo))/(Pu(SCB. Eo))
93
Figure (4.17) Comparison of Load-Midspan Deflection Curves for

SCB.Ep, SCB.Eo, SCB.Esr and SCB.Ecfrp Curved Beams
Figure (4.18) Comparison of Twisting Angle Curves at Midspan for

SCB.Ep, SCB.Eo, SCB.Esr and SCB.Ecfrp Curved Beams
94
4.2.3 Semicircular Curved Beam with Opening near Midspan

(SCB.M)
4.2.3.1 Unstrengthened Semicircular Curved Beam (SCB.Mo)

As previously Mentioned the SCB.Mo beam includes opening spaced
at distance d/2=106.5 mm from the applied load and without strengthening,
as shown in Figure (4.19). The first visible cracks are inclined cracks at the
corners of openings at load 34.6 kN directed toward the applied load
position. At load step of 41.5 kN another cracks appear at the opposite
corners of the opening, and no flexural or torsional cracks appear until this
stage. After increasing the load to 48.5 kN, torsional cracks began to appear
and propagated rapidly more than corners cracks, while flexural cracks
appear at load of 62.3 kN with small length and width at position of
maximum positive moment near the applied load. It is well noticed here that
there is no flexural cracks at position of tension zone which had maximum
moment overall the beam length because of the redistribution of the forces
due to existing of the opening at position of maximum positive bending
moment and torsional force , and because of the rapid propagation of cracks
at opening zone. Figure (4.20) shows the frame type failure mode and the
crack pattern of the beam at ultimate load of 83 kN. The load –deflection and
load–angle of twisting curves at midspan are shown in Figures (4.21) and
(4.22), respectively.
95
Figure (4.19) Semicircular Curved Beam SCB.Mo
a)
b)
Figure (4.20) Mode of Failure and Cracks Pattern for SCB.Mo Curved Beam
a) Outer Face of Opening Area
b) Inner Face of Opening Area
96
Figure (4.21) Load-Midspan Deflection Curve SCB.Mo Curved Beam
Figure (4.22) Load-Midspan Twisting Angle Curve SCB.Mo Curved Beam
97
4.2.3.2 Internally strengthened Semicircular Curved Beam

(SCB.Msr)
The specimen SCB.Msr semicircular curved beam includes opening
spaced at distance d/2=106.5 mm from the applied load as shown in Figure
(4.23), and strengthening by six pairs of closed stirrups Ø6 mm for both top
and bottom chords of opening and four diagonal bars Ø6 mm, two for each
corner of opening, as well as two Ø6 mm full stirrups at each side of the
opening spaced at distance 40 mm. In the beam (SCB.Esr) cracks were first
observed at the corners of the opening at load of 41.5 kN due to the stress
concentration at these positions, these cracks were inclined toward the
applied load. As the load was increased to 55.3 kN, torsional cracks start to
appear at position of maximum torsional moment between applied load and
internal support, while flexural cracks found at a load of 72.7 kN at both
positions of maximum positive (under the applied load ) and negative
moments (over the internal support). After more load increments a small
propagation of diagonal cracks at the corners of opening was noticed, while
torsional cracks at top and bottom cords propagated and widen rapidly more
than flexural and diagonal cracks. A frame type failure occurs at a load of
100.4 kN by formation of two independent diagonal cracks, one in each cord
of the opening as shown in Figure (4.24). The load –deflection and load –
angle of twisting curves at midspan are shown in Figures (4.25) and (4.26),
respectively.
98
Figure (4.23) Semicircular Curved Beam SCB.Msr
a)
b)
Figure (4.24) Mode of Failure and Cracks Pattern for SCB.Msr Curved Beam
a) At Interior Support
b) At Opening Region
99
Figure (4.25) Load-Midspan Deflection Curve SCB.Msr Curved Beam
Figure (4.26) Load-Midspan Twisting Angle Curve SCB.Msr Curved Beam
100
4.2.3.3 Externally Strengthened Semicircular Curved Beam

(SCB.Mcfrp)
The curve beam SCB.Mcfrp includes opening spaced at distance
d/2=106.5 mm from the applied load, as shown in Figure (4.27), and
strengthened by full wrap of CFRP laminates of 0.131mm thickness (pair of
40 mm width on the each side of the opening and three pairs of 25mm
width for each top and bottom chords of opening ). The beam was loaded
gradually until the first crack was observed and the deformation was initially
within the elastic range at the early stages of loading. The first crack appears
at a load of 34.6 kN at the skew corners of the opening. At a load of 41.5
flexural cracks began to appear at the section of maximum negative above
interior support at the top tension position. After more loading increments at
load of (72.6 kN) a torsional cracks at an angle of 45° began to appear and
propagate rapidly, while flexural cracks stopped until failure occurs, as well
as no significant increasing in corner cracks was noticed up to failure. The
failure mode occurred through torsional forces (beam type failure) at an
ultimate load of 116 kN, as shown in Figure (4.28). The load –deflection and
load – angle of twisting curves at midspan are shown in Figures (4.29) and
(4.30), respectively.
Figure (4.27) Semicircular Curved Beam SCB.Mcfrp
101
a) b)
c)
Figure (4.28) Mode of Failure and Cracks Pattern for SCB.Mcfrp Curved Beam
a) At Opening Region b) At Interior Support c) At 2nd Opening Region
Figure (4.29) Load-Midspan Deflection Curve SCB.Mcfrp Curved Beam

102
Figure (4.30) Load-Midspan Twisting Angle Curve SCB.Mcfrp Curved Beam
4.2.3.4 Summary of Test Results for SCB.M Beams

Table 4.2 shows the cracking load, ultimate load, maximum rotation at
and deflection at midspan. Also, Figure (4.31) and (4.32) show a comparison
of load-midspan deflection and angle of twisting curves for curved beams
SCB.P, SCB.Mo, SCB.Msr and SCB.Mcfrp.
It can be concluded that, the presence the openings near the applied
load reduced the ultimate load capacity (compared with control beam) by
about 83%, also a significant increase in twisting angle were noticed. On the
other hand, strengthening of the opening by internal reinforcement or
external CFRP laminates lead to increase in the ultimate load capacity
(compared with SCB.Mo) by about 21% and 39% respectively, also both
angle of twist and deflection at midspan for both type of strengthening
decreased by about 65% and 35% respectively. This is because of the
increasing of post cracking stiffness of the beam at the opening which could
103
be seen clearly in Figures (4.31) and (4.32). Results indicate that the
presence of opening in the curve beam will reduce its stiffness as well as
ultimate load capacity, while the strengthening of the opening will increase
the post cracking stiffness of the curved beam.
Table 4.2 Summary of Tested SCB.M Beams

Cracking Load, kN Ultimate *Ultimate Max. θ ×10-3 Max. Δ at Failure
Specimen Load (rad) at (mm)
Corner Flex. Tor. Load, kN Diff. % midspan midspan
Mode
SCB. P -- 62.3 62.3 152.3 83 44.6 20.4 Torsional

SCB. Mo 34.6 62.3 48.5 83 -- 99.1 17.2 Torsional
SCB. Msr 41.5 72.7 55.3 100.4 21 27 9.7 shear
Torsional
SCB. Mcfrp 32 41.5 72.6 116 39 35.2 12.77
and shear
Difference = (Pu(Specimen) − Pu(SCB. Mo))/(Pu(SCB. Mo))

SCB.P, SCB. Mo, SCB. Msr and SCB.Mcfrp Curved Beams
104

SCB.P, SCB.Mo, SCB.Msr and SCB.Mcfrp Curved Beams
4.2.4 Semicircular Curved Beam with Opening near Interior

Support (SCB.I)
4.2.4.1 Unstrengthened Semicircular Curve Beam (SCB.Io)

The SCB.Io curved beam includes opening spaced at distance
d/2=106.5 mm from face of internal support without any type of
strengthening, as shown in Figure (4.33). The beam deformation was
initially within the elastic ranges at the early stages of loading, and the load
was increased gradually until the first crack occurred which was observed in
the top and bottom skew corners of the opening due to concentration of
stresses at these corners at load about 34.6 kN. Torsional and flexural cracks
start to appear at load of 48.5 kN at the tension zone above interior support
and near the maximum torsional moment. As the load was increased further,
torsional cracks propagated rapidly. Flexural and corner cracks stop its
105
propagation until failure occurs, while torsional cracks increased in length

and width at an angle of 45°. As the load was increased further, a torsional
conventional failure mode at the opening position was appeared. The
ultimate load of beam SCB.Io was about 76.2 kN by forming torsional
cracks at top and bottom cords of the opening (beam type failure) as shown
in Figure (4.34). The load–deflection and load–angle of twisting curves at
midspan are shown in Figures (4.35) and (4.36), respectively.
Figure (4.33) Semicircular Curved Beam SCB.Io
a) b)
Figure (4.34) Mode of Failure and Cracks Pattern for SCB.Io Curved Beam
a) At Opening Region
b) At Interior Support
106
Figur (4.35) Load-Midspan Deflection Curve SCB.Io Curved Beam
Figure (4.36) Load-Midspan Twisting Angle Curve SCB.Io Curved Beam
107
4.2.4.2 Internally Strengthened Semicircular Curve Beam

(SCB.Isr)
The SCB.Isr curve beam includes opening spaced at distance
d/2=106.5 mm from Internal support as shown in Figure (4.37), and
strengthened by six pairs closed stirrups Ø6 mm for each top and bottom
chords of opening and two diagonal bars Ø6 mm at each corner of opening,
also two Ø6 mm stirrups at each side of the opening were used. In the beam
(SCB.Isr) cracks were first observed at the corners of the opening at a load
of approximately 41.4 kN. As the load increased to 55.9 kN a flexural crack
appears at the tension zone of maximum negative moment. After more load
increment (at load of 76.7 kN) torsional cracks appear at the location of
maximum torsional moment between applied load and internal support. After
more load increment, flexural cracks did not propagated in length or width,
while a noticeable new torsional cracks were recorded at top and bottom
cords of the opening which were the main reason of the beam failure. For
SCB.Isr beam, a frame failure mode occurs at load of 99.5 kN as shown in
Figure (4.38). The load –deflection and load –angle of twisting curves at
Figure (4.37) Semicircular Curved Beam SCB.Isr
108
a) b)
Figure (4.38) Mode of Failure and Cracks Pattern for SCB.Isr Curved Beam
a) At 1st Opening Region
b) At 2nd Opening Region
Figure (4.39) Load-Midspan Deflection Curve SCBIsr Curved Beam
109
Figure (4.40) Load-Midspan Twisting Angle Curve SCB.Isr Curved Beam
4.2.4.3 Externally Strengthened Semicircular Curve Beam

(SCB.Icfrp)
As mentioned previously, the SCB.Icfrp curved beam included
opening spaced at distance d/2=106.5 mm from the internal support as
shown in Figure (4.41) and strengthened by full wrap of CFRP laminates of
0.131mm thickness (pair of 40 mm width on each side of the opening and
three pairs of 25mm width each both top and bottom chords of opening). The
beam was loaded gradually until the first crack was observed at a load of 31
kN at the skew corners of the opening. As load increased a flexural cracks
began to appear at the section of maximum negative above interior support
and at the tension position at a load of 51.9 kN. After more loading
increments torsional cracks at an angle of 45° began to appear and propagate
at a load about 76.1kN, while no significant increase in flexural cracks up to
110
failure. It can be noticed that corner cracks propagation didn’t penetrate

CFRP laminates, but it does deeply through cross section and causing
failure. A beam failure mode occurred at load of 100.9 kN, as shown in
Figure (4.42). The load –deflection and load – angle of twisting curves at
Figure (4.41) Mode of Failure and Cracks Pattern for SCB.Icfrp

Curved Beam
b)
a)
Figure (4.42) Mode of Failure and Cracks Pattern for SCB.Icfrp Curved Beam
a) 1st Opening Region
b) 2nd Opening Region
111
Figure (4.43) Load-Midspan Deflection Curve SCBIcfrp Curved Beam
Figure (4.44) Load-Midspan Twisting Angle Curve SCB.Icfrp Curved Beam
112
4.2.4.4 Summary of Test Result for SCB.I Beams

Table 4.3 shows the cracking load, ultimate load, rotation and
deflection at midspan. Also, Figure (4.45) and (4.46) show a comparison of
load-midspan deflection and load- angle of twisting curves for curved beams
SCB.P, SCB.Io, SCB.Isr and SCB.Icfrp.
It can be concluded that, the presence of the openings near interior
support reduces the ultimate load capacity (compared with control beam
SCB.P) to the half, also a significant reduction in twisting angle and
deflection were noticed. On the other hand, strengthening of the opening by
internal reinforcement or external CFRP laminates increased ultimate load
capacity (compared with SCB.Io) by about 31% and 32% respectively, also
angle of twist was increased with an interested ratio because of the
confinement of the beam at opening region which postpone failure at
opening for both types of strengthening. This is because of the increasing of
post cracking stiffness of the beam at the opening which could be seen
clearly in Figures (4.45) and (4.46).
Table 4.3 Summery of Tested SCB.I Beams

*
Ultimate Max. θ ×10-3 Max. Δ at
Cracking Load , kN Ultimate Failure
Specimen Load (rad) at (mm)
Corner Flex. Tor. Load, kN Diff. % midspan midspan
Mode
SCB. P -- 62.3 62.3 152.3 100 44.6 20.4 Torsional

SCB. Io 34.6 48.5 48.5 76.2 -- 21.65 13.35 shear
Torsional
SCB. Isr 41.5 55.9 76.7 99.5 31 30.02 12.1
and shear
SCB. Icfrp 31 51.9 76.1 100.9 32 28.01 17.1 shear
Difference = (Pu(Specimen) − Pu(SCB.Io) )/Pu(SCB.Io)
113
Figure (4.45) Comparison of Load-Midspan Deflection Curves for SCB.P,


SCB.P, SCB.Io, SCB.Isr and SCB.Icfrp Curved Beams
114
4.2.5 Summary of Tested Semicircular Curved Beams

As a summary of test results for full circular beam with strengthened
and unstrengthened opening, the following notes were obtained:
1. For the specimens SCB.Eo SCB.Mo and SCB.Io, which were
unstrengthened, the decrease in ultimate load capacity was about
18%,83% and 100%, respectively when compared with the corresponding
control specimen SCB.P.
2. For internally strengthened specimens SCB.Esr SCB.Msr and SCB.Isr, the
ultimate load capacities were enhanced by about 3%,21% and 31%, when
compared with unstrengthened specimens FCB.Eo, FCB.Mo and FCB.Io
respectively.
3. The use of CFRP laminates as external confinements in spacemens
SCB.Ecfrp SCB.Mcfrp and SCB.Icfrp, increased ultimate load capacities
by about 11%,39% and 32% when compared with unstrengthened
specimens FCB.Eo, FCB.Mo and FCB.Io respectively.
4. A reliable enhancement appears in post-cracking behavior of specimens
SCB.Mcfrp, SCB.Msr and SCB.Icfrp, SCB.Isr when compared with
unstrengthened specimens SCB.Mo and SCB.Io, respectively, and this is
more than that observed from SCB.Ecfrp, SCB.Esr if compared with
SCB.Eo.
5. The use of internal confinement change the failure mode from beam type
failure to frame type failure, while the use of CFRP laminates retains the
failure mode to beam type failure.
4.3 Full Circular (Ring) Concrete Beams (Group II)

This section deals with the overall behavior of four horizontally
full circular curved (ring) concrete beams with and without openings, and
strengthened with internal reinforcements or with external CFRP laminates.
115
4.3.1 Control Specimen (FCB.P)

A specimen FCB.P is a circular full curved beam and without opening
as shown in Figure (4.47). The beam was loaded gradually until the first
crack was observed. The first crack (flexural and torsional) appeared in the
position of maximum positive moment (under point load) and at the section
of maximum torsional moment respectively, at load of approximately 103.8
kN. As the load was increased further, several flexural and torsional cracks
initiated in the positions of midspan loads, spatially at tension zones of the
beams. Torsional cracks spread and began to propagate more than flexural
cracks with load increments. As the load approaches to its ultimate value a
negative flexural crack above the support appears, while torsional cracks
directed to propagate at the compression zone above the support and under
point loads. The value of the ultimate load was 380.8 kN. Figure (4.48)
show failure mode and cracks pattern, also Figures (4.49) and (4.50) show
load –deflection and load – angle of twisting curves at midspan, respectively.
Figure (4.47) Full Circular Beam FCB.P
116
a) b)
Figure (4.48) Mode of Failure and Cracks Pattern for Circular Beam FCB.P
a) Overall Crack Pattern
b) Point Load Region
Figure (4.49) Load-Midspan Deflection Curve for FCB.P Circular Beam
117
Figure (4.50) Load-Midspan Twisting Angle Curve for FCB.P Circular Beam
4.3.2 Full Circular Beam with Opening (FCB.Mo)

A specimen FCB.Mo is a full circular curved beam with opening
spaced at distance d/2=106.5 mm from the applied load as shown in Figure
(4.51). The beam was loaded gradually until the first crack was observed.
The first crack appears at corners of the opening due to a stress concentration
of torsional moment, at load of approximately 58.8 kN. As the load was
increased further, and at a load about 95 kN torsional cracks initiated near
the corners of opening closed to applied load and support. Also flexural
cracks start to appear at a load about 110.7 kN. Torsional cracks spread and
began to propagate while no considerable propagation of flexural cracks
occurs up to last load increments. As the load approaches failure a negative
flexural cracks appears at the top face of maximum negative moment above
the support, while torsional cracks directed to propagate at the compression
118
zone around support and under point loads. Abeam type failure occurs at a
load of 190 kN. Figure (4.52), show mode of failure and cracks pattern,
while Figures (4.53) and (4.54) show load –midspan deflection and load–
midspan angle of twisting curves, respectively.
Figure (4.51) Full Circular Beam FCB.Mo
Figure (4.52) Mode of Failure and Cracks Pattern for FCB.Mo Circular Beam
at Different Positions
119
Figure (4.53) Load-Midspan Deflection Curve FCB.Mo Circular Beam
Figure (4.54) Load-Midspan Twisting Angle Curve FCB.Mo Circular Beam

120
4.3.3 Internally Strengthened Full Circular Beam (FCB.Msr)

The curved beam FCB.Msr includes opening spaced at distance
d/2=106.5 mm from applied load as shown in Figure (4.55), and
strengthened by six pairs of closed stirrups Ø6 mm for each top and bottom
chords of opening and two diagonal bars Ø6 mm for each corner of opening,
also two Ø6 mm stirrups at each side of the opening. In this beam cracks
were first observed at the corner of the opening at the cord of compression
stresses at load of 55.4 kN, while crack in the skew position of the opening
at cord of tension stresses appeared at load of 69.2 kN. As the load increased
a flexural crack appear at tension zone of maximum negative and torsional
moments at a load of approximately 79.8 kN. As the load increased to 89.9
kN, torsional cracks began to appear and propagate and widened more than
flexural and corner cracks. In spite of forming diagonal corner and flexural
cracks at earlier stages of loading for FCB.Msr, the beam failed due to
torsional effect (torsional cracks )at load of 305 kN in a frame type failure
mechanism, as shown in Figure (4.56). The extension of cracks at the
opening corners had a small propagation because of the internal
reinforcement which prevents their extension or widening at opening. The
load –deflection and load –angle of twisting curves at midspan are shown in
Figures (4.57) and (4.58), respectively
Figure (4.55) Full Circular Beam FCB.Msr
121
a) b)
Figure (4.56) Mode of Failure and Cracks Pattern for FCB.Msr Circular Beam
a) Load and Support Region
b) Exterior Face of Failed Opening
Figure (4.57) Load-Midspan Deflection Curve FCB.Msr Circular Beam
122
Figure (4.58) Load-Midspan Twisting Angle Curve FCB.Msr Circular Beam
4.3.4 Externally strengthened Full Circular Beam (FCB.Mcfrp)

As mentioned previously, the full curved beam FCB.Mcfrp includes
opening spaced at distance d/2=106.5 mm from the applied load as shown in
Figure (4.59), and strengthened by full wrap of CFRP laminates of 0.131mm
thickness (pair of 40 mm width on the each side of the opening and three
pairs of 25mm width for both top and bottom chords of opening). The beam
was loaded gradually until the first crack was observed. The first crack
appears at a load of 83 kN at the corner of the opening. After more loading
increments (at load about 100.4 kN) a flexural crack appears at position of
maximum positive moment, while torsional cracks at an angle of 45° began
to appear and propagate at a load of 128 kN. As load was increased further,
corner cracks increased and widened till CFRP laminates, also torsional
cracks propagated and spreads in zones that have no CFRP up to failure load.
123
A beam type failure mode occurred through torsional forces at load of 333.5
kN, as shown in Figure (4.60). The load –deflection and load – angle of twist
curves at midspan are shown in Figures (4.61) and (4.62), respectively.
Figure (4.59) Full Circular Beam FCB.Mcfrp
Figure (4.60) Mode of Failure and Cracks Pattern for FCB.Mcfrp

Circular Beam at Different Region
124
Figure (4.61) Load-Midspan Deflection Curve FCB.Mcfrp Circular Beam
Figure (4.62) Load-Midspan Twisting Angle Curve FCB.Mcfrp Circular Beam
125
4.3.5 Summary of Test Results for FCB.M Beams

Table 4.4 shows the cracking load, ultimate load, Percentage of
Ultimate Load with Respect to FCB.P. Figure (4.63) and (4.64) show a
comparison of load-midspan deflection and angle of twist curves for circular
curved beams FCB.P, FCB.Mo, FCB.Msr and FCB.Mcfrp. It can be
concluded that, the presence of the openings near the applied load reduce the
ultimate load capacity (compared with control full curved beam) to the half,
also a significant reduction in twisting angle was noticed because of the
ultimate load reduction. On the other hand, strengthening of the opening by
internal reinforcement or external CFRP laminates increased ultimate load
capacity (compared with FCB.Mo) by about 60% and 75% respectively, also
angle of twist and deflection were increased with an interested ratio because
of the confinement of the beam at opening region which postpone failure at
opening for both types of strengthening. This is because of increasing of post
cracking stiffness of the beam at opening which could be seen clearly in
Figures (4.63) and (4.64).
Table 4.4 Summery of Tested FCB Beams

Cracking Load, kN *
Ultimate Max. θ ×10-3 Max. Δ at
Ultimate
Specimen Load (rad) at (mm) Failure
Corner Flex. Tor. Load, kN
Diff. % midspan midspan mode
FCB. P -- 103.8 103.8 380.8 100 74.8 14.34 shear
Torsional
FCB. Mo 58.8 110.7 95 190 -- 31.7 16.71
and shear
Torsional
FCB. Msr 55.4 79.8 89.9 305 60 46.4 26.04
and shear
Torsional
FCB. Mcfrp 83 100.4 128 333.5 75 27.4 23.9
and shear
Difference = (Pu(Specimen) − Pu(FCB.Mo) )/Pu(FCB.Mo)
126

FCB.P , FCB.Mo, FCB.Msr and FCB.Mcfrp Curved Beams

FCB.P, FCB.Mo, FCB.Msr and FCB.Mcfrp Curved Beams
127
4.3.6 Summary of The Tested Curved Beams

As a summary of test results for full circular beam with strengthed and
unstrengthed opening, the following notes were obtained:
1For the specimen FCB.Mo, which was unstrengthened, the decrease in ultimate load
capacity was about 100% when compared with the corresponding control specimen
FCB.P.
1. For internally strengthened specimen FCB.Msr, the ultimate load
capacities were enhanced by about 75%, when compared with
unstrengthened specimens FCB.Mo.
2. The use of CFRP laminates as external confinements in spacemen
FCB.Mcfrp, increases ultimate load capacity by about 60% when
compared with unstrengthened specimens FCB.Mo, and 15% less than
FCB.Mcfrp.
3. A reliable enhancement appears in post-cracking behavior of specimens
FCB.Msr and FCB.Mcfrp when compared with unstrengthened
specimens FCB.Mo.
4. The use of internal confinement changes the failure mode from beam type
failure to frame type failure, while the use of CFRP laminates retain the
failure mode to beam type failure.
5. The use of CFRP laminates as external confinement makes the beam
more brittle than using internal confinement.
128
Chapter Five Finite Elements Modeling and Nonlinear Solution Techniques
CAPTER FIVE
FINITE ELEMENTS MODELING AND
FORMULATION FOR NONLIEAR ANALYSIS OF
TESTED CURVED BEAMS
5.1 Introduction
It is well known that laboratory tests require a great amount of effort
and time, in many cases these tests are very expensive and even impractical.
On the other hand, the finite element method has become, in recent years, a
powerful and useful tool for the analysis of a wide range of engineering
problems. A comprehensive finite element model permits a considerable
reduction in the number of experiments. Nevertheless, in a complete
investigation of any structural system, the experimental phase is essential .
Taking into account that numerical models should be based on reliable test
results, experimental and numerical/theoretical analyses complement each
other in the investigation of a particular structural phenomenon..
The finite element method is a numerical procedure that can be
applied to obtain solutions to a variety of problems in engineering. Steady,
transient, linear, or nonlinear problems in stress analysis, heat transfer, fluid
flow, and electromagnetism problems may be analyzed by finite element
methods (Witte and Kikstra,2002).
The Present study applies a nonlinear finite element analysis on
continuous horizontally curved concrete beams with web openings and
unstrengthen and strengthened with internal reinforcement or CFRP
laminates and subjected to two points loads at midspan by using
ANSYS(version 12.1) computer program. The objective is to explore the
129
overall response, such as: load-deformation, ultimate load and cracking load
, ……..etc.
5.2 Material Modeling
5.2.1 Concrete Modeling

Reinforced concrete is a composite material consisting of steel
reinforcement and concrete, these two materials having vastly different
properties. The required mechanical properties of reinforcing steel are
generally known. However, those for concrete are more difficult to define
depending upon the particular condition of mixing, placing, curing, nature,
rate of loading and environmental influences.
Concrete contains a large number of micro-cracks, especially at
interfaces between coarse aggregate and mortar, even before any load has
been applied. This property is decisive for the mechanical behavior of
concrete. The propagation of these micro-cracks during loading contributes
to the nonlinear behavior of concrete at low stress level and cause volume
expansion at failure.
Many of these cracks are caused by segregation, shrinkage, or thermal
expansion in the mortar. Some micro-cracks may be developed during
loading because of difference in stiffness between aggregate and mortar. The
differences can result in a strain at interface zone several times larger than
the average strain. Since the aggregate-mortar interface has a significantly
lower tensile strength than the mortar, it constitutes the weakest link in the
composite system. This is the primary reason for low tension strength of the
concrete material. From the preceding discussion one can expect that the size
and texture of aggregate have a significant effect on the mechanical behavior
of concrete under various types of loading.
130
Many experimental studies of the behavior of concrete under uniaxial

and multiaxial loading conditions have been performed. The aims of such
investigations have made to understand the complex response of concrete for
various imposed stress conditions and to provide the necessary data required
to develop accurate numerical models to be used in nonlinear finite element
analysis of concrete structures.
5.2.1.1 Uniaxial Compression Behavior for Concrete

A typical stress-strain behavior for concrete under uniaxial
compression is shown in Figure 5.1. It is nearly linear up to about (0.3-0.5)
( )
times the ultimate strength of concrete f c' .
The stress-strain curve shows a gradual increase in curvature that
( ) ( )
occurs up to a stress level of 0.75 f c' to 0.9 f c' , after which the stress-strain
( )
curve bends sharply and approaches the peak point at f c' (Chen,1982, Chen
and Saleeb ,1981). Then, the stress-strain curve descends until failure occurs
due to the crushing of concrete at the ultimate strain (ε u ) .
The shape of the stress-strain curve is similar for concrete of low,
normal, and high strengths.
High strength concrete behaves in a linear fashion to a relatively
higher stress level than the low strength concrete. The strain at the maximum
stress is approximately (0.002) (although high strength concretes have
somewhat a little higher strain at peak stresses). On the descending portion
of the stress-strain curve, higher strength concretes tend to behave in a more
brittle manner, with the stress dropping off more sharply than it does for
concrete with lower strength (Wischers,1978).
131
Figure 5.1: Uniaxial Compressive Strain curve for

concrete with different strength (Wischers,1978).
The modulus of elasticity of concrete (Ec ) is generally taken to be a
function of the compressive strength  f c'  .

 
For normal weight concrete based on a dry unit weight (2200-2500

kg/m3), (Ec ) can be permitted to be taken as follows (ACI Committee
318M,2011).
Ec = 4700 f c' in (MPa)
Where, ( Ec ) is the modulus of elasticity of concrete in (MPa), ( f c' ) is

the ultimate strength of concrete in (MPa).
The Poison's ratio (ν ) for concrete under uniaxial compression ranges
between (0.15 to 0.22). It remains constant up to about  0.8 f c'  (Chen,1982,

 
Mackava and Okamura,1983).
132
Beyond this level it begins to increase rapidly and values in excess of

(1.0) have been measured (Mackava and Okamura,1983) due to the rapidly
increasing transverse strain increases rapidly owing to internal cracking
parallel to the direction of loading then the longitudinal strain. In this study a
value of (0.2) is adopted for analyzing reinforced concrete arches.
The multi-linear isotropic material uses Von Mises failure criterion
along with (Willam and Warnke, 1974) model to define failure of concrete.
Compressive stress-strain relationship for the concrete model was obtained
by using the following equations to compute the multi-linear isotropic stress-
strain curve for the concrete,( MacGregor, 1992).
Ec ε
f = (5.1)
2
 ε 
1 +  
 εo 
2 f 'c
εo = (5.2)
Ec
f
E= (5.3)
ε
where:
f: Stress at any strain, ε
ε: Strain of concrete
ε 0 : Strain at ultimate compressive strength f' c
E: Tangent modulus of elasticity
Figure 5.2 shows the stress-strain relationship used in this study. Point
1, defined as 0.30 f c' , is calculated in the linear range Eq.(5.3), Points 2, 3 and
4 are calculated from Eq.(5.1). With ε 0 obtained from Eq.(5.2), strains were
selected and stresses were calculated for each strain value. Point 5 is defined
at f c' , and point 6 is the termination with ε cu = 0.003 mm/mm indicating
traditional crushing strain for unconfined concrete.
133
Figure 5.2: Stress-Strain Curve for Concrete in Compression ( MacGregor, 1992)
5.2.1.2 Tensile Behavior of Concrete

The general mechanical behavior of concrete under uniaxial tensile
loading shows many similarities to the behavior observed in uniaxial
compression. Typical stress – strain curves for concrete in uniaxial tension
are shown in Figure 5.3. In general, at a stress less than (60 %) of the tensile
strength, the creation of new micro-cracks is negligible. So, this stress level
will correspond to a limit of elasticity. Above this level of stress, the bond
micro-cracks start to grow (Hughes and Chapman,1966).
The direction of crack propagation for uniaxial tension is transverse
(normal) to the stress direction. The growth of every new crack will reduce
the available load - stress capacity and this reduction causes an increase in
the stresses at critical crack tips. The failure in tension is caused by a few
connected cracks rather than by numerous cracks, as it is for compressive
states of stress.
134
The ratio between uniaxial tensile strength ( ft ) and compressive
( )
strength f c' may vary considerably but usually ranges from (0.05) to (0.1).
The modulus of elasticity under uniaxial tension is somewhat higher and
Poisson's ratio somewhat lower than in uniaxial compression.
The direct tensile strength of concrete is difficult to measure and is
  
normally taken as (0.3) to  o.4 f c'  . Many times, either the modulus of
  
rupture ( f r ) or the split cylinder strength ( f t ) is used to approximate the

tensile strength of concrete. The value of the modulus of rupture of concrete
 
varies quite widely but is normally taken as  0.62 f c'  (ACI Committee
 
318M,2011). The split cylinder tensile strength is usually somewhat lower,

  
at approximately (0.45) to  o.55 f c'  in (MPa) (Chen,1982).
  
Figure 5.3:Typical Tensile Stress-Strain Curve for Concrete

(Hughes, B. P. and Chapman,1966)
135
5.2.1.3 Biaxial Stress Behavior Concrete

Kupfer and Grestle (1973) had performed extensive experimental tests
on concrete plates made with three different concrete strengths (19, 31 and
58 MPa) and loaded in orthogonal directions. The ultimate strength data
were reported in terms of a biaxial stress envelope as shown in Figure 5.4.
Under biaxial compression, the compressive strength increases
approximately (25 %) over that of the uniaxial strength at the stress ratio of
(σ 2 = 0.5 σ 1 ) and this is reduced to about (16 %) at an equal biaxial
compression state (σ 2 = σ 1 ). Under biaxial tension - compression, the
compressive strength is decreased almost linearly as the applied tensile stress
is increased. Under biaxial tension, the concrete strength is almost the same
as that of the uniaxial tensile strength.
Figure 5.4: Biaxial State of Loading( Kupfer and Grestle ,1973)
136
5.2.1.4 Triaxial Stress Behavior of Concrete

Under triaxial loading, experimental studies indicated that concrete
has a consistent failure surface which is a function of the three principal
stresses (Chen and Saleeb, 1981). Because of concrete isotropy, the elastic
limit and failure limit can be represented as surfaces in three dimensional
principal stress space as shown in Figure 5.5.
For increasing hydrostatic compressions along the (σ 1 = σ 2 = σ 3 ) axis, the
deviatoric sections (planes perpendicular to the axis (σ 1 = σ 2 = σ 3 ) of the
failure surface are more or less circular, which indicates that the failure in
this region is independent of the third stress invariant. For smaller
hydrostatic pressure, these deviatoric cross sections are convex and
noncircular. The failure surface can be represented by three stress invariants.
Figure 5.5:Failure Surface of Concrete in 3-D Stress Space ( Chen and Saleeb,1981).
5.2.1.5 Post - Cracking Model (Tension Stiffening Model)

Upon cracking, the stresses normal to the cracked plane are released as
the cracks propagate. To simulate this behavior in connection with the finite
element modeling of reinforced concrete members, the tension stiffening
concept is usually used (Banzant, 1983). This concept is based on the fact
that some of the tensile stresses can be carried by the concrete between the
137
cracks due to the bond action between the steel bars and the surrounding
concrete. This ability is progressively weakened because of the formation of
new cracks.
Furthermore, in order to improve the numerical stability of the
solution, the tension stiffening effect was introduced in several models.
In the present work, the tension stiffening of reinforced concrete after
cracking was represented by providing a linearly descending branch as
shown in Figure 5.6.
Figure 5.6: Pre and Post-Cracking Behavior of Normal Strength Concrete (Banzant,1983).
This model is given by:
(a) For ε cr ≤ ε n ≤ α 1 ε cr
 α f cr 
fn =  2 α − ε  (5.4)
 α 1 − 1  1 ε cr 
 
(b) For ε n  α 1 ε cr
fn = 0 (5.5)
where, ( f n , ε n ) is the stress and strain normal to the crack plane, ( f cr , ε cr )

is the cracking stress and strain, (α1 ) is the rate of stress released as the crack
widens and (α 2 ) is the sudden loss of stress at the instant of cracking.
138
5.2.1.6 Crushing Modeling

If the material at an integration point fails in uniaxial, biaxial, or
triaxial compression, the material is assumed to crush at that point. Under
this condition, material strength is assumed to have degraded to an extent
such that the contribution to the stiffness of an element at the integration
point under consideration can be ignored (ANSYS Manual,2009).
5.2.1.7 Shear Transfer Model

The concrete is assumed to behave linearly in tension up to the onset
of cracking. When concrete cracks, its shear stiffness is reduced. However,
cracked concrete can partially transmit shear across the crack due to
aggregate interlock and dowel action of the reinforcement. The shear transfer
mechanism depends on the reinforcement ratio, bar size, bar arrangement,
the amount of concrete cover, the type of concrete and aggregate size. To
estimate such an effect, a shear transfer coefficient (β ) is introduced which
represents a shear strength reduction factor for concrete across the crack
face. When the crack is formed, only a constant value of a shear transfer
( )
coefficient β 0 for the opened crack is introduced, and if the crack is closed,
( )
the shear transfer coefficient (β c ) is used. The values of β 0 and (β c ) are
( )
always in the range 1  β c  β 0  0 (ANSYS Manual,2007).These values
( )
depend on the texture of the cracked surface. In this study, β 0 is assumed
to be (0.18) and (β c ) is assumed to be (0.72).
5.2.1.8 Cracking Modeling

The main feature of plain concrete is its low tensile strength compared
with the failure stress in compression. In the finite element methods, two
139
main approaches have been used for cracked section representation, the
discrete cracked model, and the smeared crack model.
a) Discrete Crack Model

This approach was first used by Ngo and Scordelis (1967) to analyze a
simply supported reinforced concrete beam. In this approach, the cracks are
restricted to occur at the boundary of the elements by separation of nodal
points initially occupying the same position in space, see Figure (5.20). This
means that when any crack occurs the topology of the mesh varies. This
makes the analysis expensive. These difficulties have resulted in a very
limited acceptance of this model in the general structural applications.
( a)
(b)
Figure (5.7) Cracking representation in discrete cracking modeling approach

(a) one directional cracking (b) two directional cracking
b) Smeared Crack Model
In this approach, the cracked concrete is assumed to remain a continuum;
the cracks are smeared out in a continuous fashion. It is assumed that the
concrete becomes orthotropic after first cracking has occurred. In the
smeared-cracking model, a crack is not discrete but implies an infinite
number of parallel fissures across that part of the finite element, see Figure
(5.8). After cracking has occurred, the cracked concrete becomes an
orthotropic material and a new relationship must be derived. This is
accomplished by modifying the stiffness of the element; the modulus of
140
elasticity is reduced to zero in the direction normal to the crack. Further, a

reduced shear modulus is assumed on the cracked plane to account for the
aggregate interlocking.
In the smeared crack simulation, two different models are used for defining
the crack direction. The first is the fixed crack model. In this model, the
crack is fixed. A perpendicular crack plane is allowed if the tensile stress in
this direction exceeds the tensile strength. The direction of crack is held and
fixed at all subsequent time steps. The second approach is the rotating crack
model, in this approach the cracks are permitted initially to be perpendicular
to the principal tensile stress direction when the stress reaches the specified
limiting value. With further increment of loading the principal stress
changes, the crack is assumed to rotate and orthotropic material axes are set
in a new crack direction.
Figure 5.8: Smeared Crack Modeling (Chen and Saleeb,1981).

For ANSYS computer program, crack modeling of concrete, depends
on smeared crack. This model is described in terms of shear transfer model
and closing and reopening of cracks.
5.2.1.9 Failure Criteria for Concrete

The actual behavior and strength of concrete materials are very
complex because they depend on many factors such as the physical and
141
mechanical properties of the aggregate, cement paste and the nature of

loading. No single mathematical model can describe the strength of real
concrete materials completely under all conditions, so, simple models or
criteria are used to represent the properties that are essential to the problem
being considered (Chen,1982).
Willam and Warnke (1974) developed a mathematical model capable
of predicting failure for concrete materials under multiaxial stress state. Both
cracking and crushing failure modes are accounted for. This model is
represented by the following equations:
F
−S ≥0 (5.6)
f c'
where, (F ) is the function of principal state (σ xp , σ yp , σ zp ) , (S ) is the failure

surface expressed in terms of principal stresses and five input parameters
(f c
'
) ( )
, f t , f cb , f1 and f 2 , f c' is the ultimate uniaxial compressive strength, ( f t )
is the ultimate uniaxial tensile strength, ( f cb ) is the ultimate biaxial

compressive strength, ( f1 ) is the ultimate compressive strength for a state of
biaxial compression superimposed on hydrostatic stress state (σ ha ) , ( f 2 ) is the
ultimate compressive strength for a state of uniaxial compression
superimposed on hydrostatic stress state (σ ha ) , and (σ )
a
h is the ambient
hydrostatic stress state.
The failure surface is separated into hydrostatic (change in volume)
and deviatoric (change in shape) sections as shown in Figure 5.9. The
hydrostatic section forms a meridianal plane which contains the equisectrix
(σ 1 = σ 2 = σ 3 ) as an axis of revolution.
The deviatoric section lies in a plane normal to the equisectrix (dashed
line). The deviatoric trace is described by polar coordinate (r , θ ) , where (r )
is the position vector locating the failure surface with angle (θ ) . The failure
surface is defined as:
142
1 σm 1 τm
f (σ m , τ m , θ ) = + −1= 0 (5.7)
2 f c' r (θ ) f c'
(a) Hydrostatic Section (θ = 0) (b) Deviatoric Section

Figure 5.9:Failure Surface (Chen,1982).
where, (σ m and τ m ) is the average stress components defined as:

1
σm = (σ 1 + σ 2 + σ 3 ) = 1 I1 (5.8)
3 3
2
τm
2
= J2 (5.9)
5
Where, (I1 ) is the first stress invariant, ( J 2 ) is the second deviatoric stress
invariant and (ρ ) is the apex of the surface.
The free parameters of failure surface (ρ ) and (r ) are identified from
the uniaxial compressive strength ( f c' ) , biaxial compressive strength ( f cb )
and uniaxial tensile strength ( f t ) . If Eq. (5.6) is not satisfied, there is no
attendant cracking or crushing.
Otherwise, the material will crack if any principal stress is tensile,
while crushing will occur if all principal stresses are compressive.
143
Willam and Warnke (1974) succeeded in finding an expression for the

failure cross section, since it can meet not only the conditions of symmetry,
smoothness and convexity, but also it degenerates to a circle if (r1 = r2 ) . This
means that the cylindrical Von Mises model and the conical Drucker-Prager
model are all special cases of Willam and Warnke failure formulation
(ANSYS Manual,2009).
5.2.1.9.1 Determination of the Model Parameters

The failure surface can be specified with a total of five strength
( )
parameters f c' , f t , f cb , f1 and f 2 in addition to an ambient hydrostatic
( ) ( )
stress state σ ha , as shown in Figure 5.9. f c' and ( f t ) can be specified from
two simple tests, and the other constants can be determined from Willam and
Warnke (1974):
f cb = 1.2 f c' (5.10)
f1 = 1.45 f c' (5.11)
f 2 = 1.725 f c' (5.12)

However, these values are valid only for stress states where the
condition stated below is satisfied:
σ h ≤ 3 f c' (5.13)
where
(σ h ) is the hydrostatic stress state = 1 (σ xp + σ yp + σ zp ) (5.14)

3
The condition of Eq. (5.7) applies to stress situations with a low hydrostatic
stress component.
144
Figure 5.10, the lower curve represents all stress states in which
(θ = 0 ), while the upper curve represents stress states for (θ = 60 ). The axis
(ξ ) represents hydrostatic length.
The failure of concrete is categorized into four domains:
1st domain: (0 ≥ σ 1 ≥ σ 2 ≥ σ 3 ) (compression – compression – compression).
2nd domain: (σ 1 ≥ 0 ≥ σ 2 ≥ σ 3 ) (tension – compression – compression).
3rd domain: (σ 1 ≥ σ 2 ≥ 0 ≥ σ 3 ) (tension – tension – compression).
4th domain: (σ 1 ≥ σ 2 ≥ σ 3 ≥ 0 ) (tension – tension – tension).
The concrete will crack if any principal stress is a tensile stress, while
crushing will occur if all principal stresses are compressive.
Figure 5.10 Profile of the Failure Surface as Function of Five

Parameters (ANSYS Manual,2007)
5.2.2 Reinforcement Modeling

Since the reinforcing bars are normally long and relatively slender,
they can generally be assumed to be capable of transmitting axial forces
only. For the finite element models, the uniaxial stress - strain relation for
steel was idealized as a bilinear curve, representing elastic-plastic behavior
with strain hardening. This relation is assumed to be identical in tension and
in compression as shown in Figure 5.11.
145
Figure 5.11 Modeling of Reinforcing Bars.
In the present work, the strain hardening modulus (ET ) is assumed to
be (0.01Es).
( )
The value of steel yield strength f y and ultimate tensile strength
( f u ) , corresponding to steel bar, are carried out from tensile test. Table 3.7,
gives details about these values.
In developing a finite element model, three alternative representations
of reinforcement can usually be used (Wolanski,2001, Yousifani,2004),
these are:
1. Discrete representation
Discrete representation has been widely used. The reinforcement in
the discrete model uses one dimensional bar or beam elements that are
connected to concrete mesh nodes as shown in Figure 5.12a. Therefore,
the concrete and the reinforcement mesh share the same nodes and the
same occupied regions. Full displacement compatibility between
reinforcement and concrete is a significant advantage of the discrete
representation. Their disadvantages are the restriction of the mesh and the
increase in the total number of elements.
146
2. Embedded representation
The embedded representation is often used with high order
isoperimetric elements. The bar element is built in a way that keeps
reinforcing steel displacements compatible with the surrounding concrete
elements as shown in Figure 5.12b. When reinforcement is complex, this
model is very advantageous. The stiffness matrix of the reinforcement steel
is evaluated separately and then added to that of the concrete to obtain the
global stiffness matrix.
3. Smeared (Distributed) representation
The smeared model assumes that reinforcement is uniformly spread in a
layer throughout the concrete element in a defined region of the finite
element mesh as shown in Figure 5.12c. This approach is used for large scale
models where the reinforcement does not significantly contribute to the
overall response of the structure.
In the present study, the steel reinforcements were represented by using
2-node discrete representation (LINK8 in ANSYS) and included within the
properties of 8-node brick elements. The reinforcement is assumed to be
capable of transmitting axial forces only, and perfect bond is assumed to
exist between the concrete and the reinforcing bars. To provide the perfect
bond, the link element for the steel reinforcing bar was connected between
nodes of each adjacent concrete solid element, so the two materials share the
same nodes.
-a- -b- -c-

Figure (5.12), Models for reinforcement in R/C (Tavarez 2001).
(a) discrete; (b) embedded; and (c) smeared.
147
5.2.3 CFRP Composite Modeling

CFRP composites are materials that consist of two constituents. First
constituent is the reinforcement, which is embedded in the second
constituent, a continuous polymer called the matrix (Kaw,1997). The
reinforcing material is in the form of fiber such as carbon or glass, which are
typically stiffer and stronger than the matrix. The CFRP composites are
modeled as orthotropic elastic materials; that is, their properties are not the
same in both directions. Figure 5.13, shows a schematic of CFRP composites
(Gibson,1994).
Figure 5.13:Schematic Properties of CFRP Composites (Gibson,1994).
5.3 Material Idealization
5.3.1 Element Types

The elements types shown in Table 5.1 are used to model the tested beams.
Table 5.1: Element Types Used in Modeling of R.C. Beams
Element Element Representation
No. Type
1 SOLID65 Concrete
2 LINK8 Longitudinal (circumference) steel reinforcement
(φ 12 mm ) and Radial reinforcement (stirrups) (φ 6 mm )
3 SOLID45 Steel plate
4 SHELL41 CFRP
148
5.3.1.1 Element SOLID65

SOLID65 is used for the 3-D modeling of solids with or without
reinforcing bars (rebars). SOLID65 is capable of cracking in tension and
crushing in compression. In concrete applications, for example, the solid
capability of the element may be used to model the concrete while the rebar
capability is available for modeling reinforcement behavior. The element
shown in Figure 5.14 is defined by eight nodes having three degrees of
freedom at each node: translations in the nodal x, y, and z directions. Up to
three different rebar specifications may be defined.
The most important aspect of this element is treatment of nonlinear
material properties. The concrete is capable of cracking (in three orthogonal
directions), crushing, plastic deformation, and creep. The rebars are capable
of tension and compression, but not shear.
Figure 5.14: Geometry of Element SOLID65 (ANSYS help, 2007)
5.3.1.2 Element LINK8

LINK8 shown in Figure 5.15 is a spar (or truss) element which may be
used in a variety of engineering applications. This element can be used to
149
model trusses, sagging cables, links, springs, etc. The 3-D spar element is a
uniaxial tension - compression element with three degrees of freedom at
each node: translations in the nodal x, y, and z directions. As in a pin-jointed
structure, no bending of the element is considered. Plasticity, creep,
swelling, stress stiffening, and large deflection capabilities are included.
Figure 5.15: Geometry of Element LINK8 (ANSYS help, 2007)
5.3.1.3 Element SOLID45

SOLID45 is used for the 3-D modeling of solid structures, the element
shown in Figure 5.16 is defined by eight nodes having three degrees of
freedom at each node; translations in the nodal x, y, and z directions.
The element has plasticity, creep, swelling, stress stiffening, large
deflection, and large strain capabilities.
Figure 5.16: Geometry of Element SOLID45. (ANSYS help, 2007)
150
5.3.1.4 Element SHELL41

SHELL41 shown in Figure 5.17 is a 3-D element having membrane
(in-plane) stiffness but no bending (out of plane) stiffness. It is intended for
shell structures where bending of the elements is of secondary importance.
The element has three degrees of freedom at each node: a translation in the
nodal x, y, and z directions. The element has variable thickness, stress
stiffening, and large deflection.
Figure 5.17: Geometry of Element SHELL41.(ANSYS help, 2007).
5.4 Nonlinear Finite Element Analysis

In any three-dimensional finite element analysis, the performance of
any structural member under load depends on the behavior of the type of
material used to construct the member. In concrete members which are made
of different materials, concrete and reinforcing bars are brought together to
behave as a composite system. The steel can be considered as a
homogeneous material that exhibits a similar stress-strain relationship in
tension and compression. While, the behavior of concrete is monitored to
have grossly heterogeneous internal structure because it is dependenting on
151
the properties of each of its components; namely, cement mortar, aggregates

and air voids.
One of the main objectives of the finite element analysis of structures
is to determine the response of the structure under loading. A typical load-
deformation response for a monotonically loaded member is essentially
linear up to a certain limit of load. Beyond this limit a nonlinear load-
deformation response occurs. Such response is due to a combination of
nonlinear material behavior (material nonlinearities), large deformation in
the structure (geometric nonlinearities), and interface nonlinearity for
composite members (changing status) (Dawlat, 2007).
In the analysis of reinforced concrete structures, at nonlinear stage of
behavior, it is not possible to solve the governing equilibrium equations
directly; therefore, resort has to be made to more sophisticated solution
strategies. In the present work, ANSYS computer program (ANalysis
SYStem) is used to create the finite element model.
5.4.1 Numerical Integration

In most finite element analyses, the element stiffness matrix [Ke]
cannot be obtained analytically. Thus, to perform the integration required to
evaluate the element stiffness matrix, a suitable scheme of numerical
integration has to be used.
The application of the three-dimensional finite element analysis in
connection with the nonlinear behavior of structures needs a large amount of
computation time due to frequent evaluation of the stiffness matrix.
Therefore, it is necessary to choose a suitable integration rule that
minimizes the computation time but with sufficient accuracy.
In the current study, the “Gauss-Legendre quadrature scheme” is
used and it has been found to be accurate and efficient (Cervera et al. 1987).
152
Hence, the equations of the stiffness matrix element in two dimensions can
be expressed as:
(5.15)
When extended to three-dimensional cases, the equations of the stiffness

matrix element are expressed as (ANSYS Theory,2007)
(5.16)
Where: n, m, and l are the number of Gaussian points in the ξ, η and ζ-

directions respectively.
In general, "n, m, and l" are taken to be equal in three directions. F (
ξ,η,ζ) is a function which represents the matrix multiplication “
[B]T [D] [B] J ”.
Wi , Wj and Wk are weighting factors, see Table (5.1)

In a similar manner the reinforcing bar element stiffness matrix can
be written as:
(5.17)
The integration rule, which has been used in present work, is 2×2=4
points for the shell element, and 2×2×2 =8 point Gauss quadrature for the
brick element and 1x2=2 for reinforced bar. Figure (5.18) shows the
distribution of the sampling points over the volume of the 8 node and their
weighting factors are given in Table (5.2). The integration points are also the
sampling points for stresses and material state determination
(Mottram,1996).
153
Figure (5.18) Distribution of integration points (ANSYS Theory,2009)
Table 5.2 Sampling Points Locations and Weighting Factors for (2x2x2),
(2x2) and (2x1) Gauss Quadrature (ANSYS Theory,2007)
5.4.2 Procedure for Solving Nonlinear Finite Element Equations

The finite element discrimination process yields a set of simultaneous
equations:
[K ] {U } = {F a } (5.18)
where, [K ] is the stiffness matrix, {U } is the vector of nodal displacements
{ }
and F a is the vector of applied loads.
154
For linear elastic problems Eq. (5.13) is used to find the solution of the
unknown displacement {U }. In the case of nonlinear system, the stiffness
matrix [K ] is a function of the unknown displacement (or their derivatives).
Then the Eq. (5.13) cannot be exactly computed before determination of the
unknown displacement {U }.
There are several techniques for solving the nonlinear Eq. (5.18); the
basic technique can be classified into:
(1) Incremental or stepwise procedure, Figure (5.19a).
(2) The iterative or Newton-Raphson procedure, Figure (5.19b).
(3) Incremental- Iterative procedure, Figure (5.19c).
(
(a) (b)
Figure 5.19:Basic Technique for Solving the Nonlinear Equation

(a) Incremental (b) Iterative (c) Incremental–Iterative (Maurizio et al,2002)
In the incremental procedure, the load is applied in several small

increments, and the structure is assumed to respond linearly within each
155
increment with its stiffness recomputed based on the structural geometry and
member end actions at the end of the previous load increment. This is a
simple procedure, which requires no iterations, but errors are likely to
accumulate after several increments unless very fine increments are used,
Figure 5.19a.
In the iterative procedure, the total load is applied in one increment at
the first iteration, the out of balance forces are then computed and used in the
next iteration until the final converged solution would be in equilibrium,
such that the internal load vector would equal the applied load vector or
within some tolerance. This process can be written as:
[KiT ] {ΔU i }= {F a }− {F nr } (5.19)
{U i+1}= {U i }+ {∆U i } (5.20)
[ ] is the stiffness matrix, (i) is the subscript representing the

where, K iT
current equilibrium iteration and {F nr } is the internal load vector.

This procedure fails to produce information about the intermediate
stage of loading. For structural analysis including path-dependent
nonlinearities increments are in equilibrium in order to correctly follow the
load path. This can be achieved by using the combined incremental-iterative
method.
In the combined-iterative procedure, a combination of the incremental
and iterative scheme is used. The load is applied incrementally, and
iterations are performed in order to obtain converged solution corresponding
to the stage of loading under consideration, as shown in Figure 5.19c.
The incremental-iterative solution procedures have been used in this
study. Full Newton-Raphson procedure is applied. The stiffness matrix is
156
formed at every iteration. The advantage of this procedure may give more
accurate result. The disadvantage of this procedure is that a large amount of
computational effort may be required to form and decompose the stiffness
matrix, as shown in Figure 5.20.
Figure 5.20: Incremental-Iterative Procedures Full

Newton-Raphson procedure (Maurizio et al,2002)
5.4.3 Convergence Criteria

For every incremental load the iteration continues until convergence is
achieved. The convergence criterion for the nonlinear analysis of structural
problems can be classified as:
(1) Force criterion.
(2) Displacement criterion.
(3) Stress criterion.
The displacement criterion has been used in this study. In
displacement criterion, the incremental displacements at iteration (i) and the
total displacements are determined. The solution is considered to be
converged when the norm of the incremental displacements is within a given
157
tolerance of the norm of the total displacements; infinite norm is used and
takes the form:
{∆U i } = (max ∆U i ) ≤ Tn (max ∆U i ) (5.21)
where, U may equal u, v, w or θ z .

For force criteria the norm of the residual forces at end of each
iteration are checked against the norm of the current applied forces as:
0 .5 0.5
{R} =  ∑ Ri2  ≤ Tn  ∑ F 2  (5.22)
   i 
where, {R} is the residual vector:
{R} = {F a }− {F nr } (5.23)
In this study, the tolerance (Tn ) is taken equal to (1 %) near the ultimate load
for loads control.
5.4.4 Analysis Termination Criteria

In the physical test under load control, collapse of a structure takes
place when no further loading can be sustained. This is usually indicated in
the numerical tests by successively increasing iterative displacements and a
continuous growth in the dissipated energy. Hence, the convergence of the
iterative process cannot be achieved. A maximum number of iterations for
each increment are specified to stop the nonlinear solution if the
convergence limit has not been achieved for this study. It has been observed
that a minimum number about (20) of iterations is generally sufficient to
predict the solution divergence or failure. This maximum number of
iterations depends on the type of the problem, extent of nonlinearities, and
on the specified tolerance. In the present study a maximum number of
iterations equal (100) is adopted for load control problems.
158
Chapter Six Finite Element Analysis
CHAPTER SIX
FINITE ELEMENT ANALYSIS
6.1 Introduction
The aim of the present chapter is to make a comparison between the
F.E. analysis results and the experimental results that explore the adequacy
of elements type, material modeling, real constants and convergence criteria
to model the response of the horizontally curved reinforced concrete beams
with and without openings unstrengthened and strengthened by CFRP
laminates or internal reinforcement.
This chapter includes the numerical analysis of the curved beams
tested in chapter four and parametric study of some important variables by
using a powerful nonlinear finite element method package (ANSYS 12.1
software).
6-2 Description of The Tested Specimens in Finite Element

The actual dimensions of the tested curved beams are shown in Figure
(3-1) for semicircular curved beam and Figure (3-2) for full circular beams.
By taking advantage of the symmetry for both beams in geometry and
loadings, a half of the semicircular curved beam, and a quarter of the full
circular beam were used for modeling finite element analysis as shown in
Figure (6-1). This approach reduces computational time and the computer
disk space requirements significantly.
159
Solid45
Brick element of mesh
Solid65
Solid 45
θ=90
Solid 45
a)
Brick element of mesh

Solid65
b)
Figure (6-1) Adopted Descriptions of Curved Beams

a) Half of Semicircular b) Quarter ofFull Circular
6.3 Mesh Refinement
An important step in finite element modeling is the selection of the

mesh density. A convergence of results is obtained when an adequate
number of elements are used in a model. This is practically achieved when a
160
mesh density increases for a reasonable amount to match the results;

therefore, a convergence study was carried out for a half part of SCB.P (solid
semicircular curved beam) by making use of symmetry. Four types of mesh
are used to find the best mesh size for SCB.P as following:
Mesh 1: total number of elements =1200
Mesh 2: total number of elements= 1482
Figure (6.2) shows detail of mesh size for four cases. From the test results, it
was found that models with number of elements equal to 4017 element gives
the best results compared with experimental Analysis as shown in Figures
(6.3) and (6.4).
J
Y X Y X
Y X
Mesh (b)1482 Element

Mesh(a)1200Element Mesh (c)2338 Element
Y X
Mesh (d)4017 Element Mesh (d)7600 Element
Figure (6.2) Mesh density (Cross Section and Top View).

161
Figure (6.3) Effect of Number of Elements on Load-Midspan deflection
Figure (6.4) Effect of Number of Elements on Load- Midspan Twisting Angle
162
6.4 Reinforcing Steel Modeling

Discrete representation was used to model all types of reinforcement
(longitudinal and web reinforcement). Link 8 element was employed to
represent the steel reinforcement for group (I) and (II), as shown in Figure
(6-5). In this study, a perfect bond between concrete and steel reinforcement
is assumed.
b) a)
Figure (6.5) Reinforcing Steel Bars Modeling

a) Half of Semicircular Beam b) Quarter of Full Curve Beam
6.5 Modeling of CFRP Laminates

No mesh is needed for the CFRP laminates. These elements (shell41)
were added by using the existed nodes of concrete. Figure (6.6) shows the
CFRP laminates for both semicircular curved and full curved beams.
CFRP Laminates
(Shell 41)
CFRP Laminates
(Shell 41)
Figure (6.6) CFRP Laminates Arrangement of Tested Concrete Curved Beams.
163
6.6 Loads and Boundary Conditions Representations

Displacement of boundary condition used to constrain the all reinforced
concrete beam models for the purpose of getting a unique solution. These models
constrained in the z-direction, and y-direction (Uz=Uy= 0 ) at the hinge support,
while constrained in the y-direction and x-direction (Uy= Ux=0) at roller support
as illustrated in Figure (6.7-a).
In the experiment, a steel plate of (125x50x8 mm) is used at the support and
loading locations to avoid the situation of stress concentration. The external
applied load is represented by the equivalent nodal forces on the top elements face
at midspan points of beam, as shown in Figure (6.7-b).
The application of the loads up to failure was done incrementally as
required by Newton-Raphson procedure as described previously in details in
chapter five of this study. Failure for each of the models is defined when the
solution for a minimum load increment still does not converge (convergence fails).
The program then gives a message specifying that the models have a significantly
load displacement (rigid body motion).
Figure (6.7) Boundary Conditions and Applied Loads Arrangements
164
6.7 Results of Finite Element Analysis

All tested curved beams have been analyzed by using ANSYS computer
program to determine the validity of this numerical method for the analysis of
horizontally reinforced concrete curved beams with web opening strengthened
externally with CFRP laminates or internally with steel reinforcement. The overall
behavior and specifications illustrated before in chapter three for these
strengthened materials have been taken in the consideration during the built up and
input data of ANSYS computer program.
The load-midspan deflection curves, load-midspan twisting angle curves,
cracking and ultimate loads for all analyzed curved beams have been illustrated
through the results below .Also, crack pattern and stress distribution through the
concrete and CFRP laminates have been listed in appendix C.
6.7.1 Load – Deformations Curves

Figures 6.8 to 6.35 include a comparison between the load-midspan
deflection and the load-midspan twisting angle curves of the experimental and the
numerical results. The variation of midspan deflection and midspan twisting
angle with the step-by-step loads applied for all curved beams are all recorded
through these curves. The finite element load-deflection curves for most beams
showing a stiffer response rather than the experimental results. Microcracks
produced by drying shrinkage and handling are presenting in the concrete; these
would reduce the stiffness of the actual beam, while the F.E. does not include the
effect of microcracks. The F.E. analysis assumes that concrete is a homogenous
material but, the true it is a heterogeneous material. Also, a perfect bond between
the concrete and steel or CFRP Laminates is assumed in the F.E. analysis.
The comparison shows the validity of the FEM results of the program used
in application (ANSYS) by showing a good agreement with the experimental
results discussed in chapter four previously.
165
Figure (6.8) Load-Midspan Deflection Curves for Control Beam
Figure (6.9) Load-Midspan Twisting Angle Curves for Control Beam
166
Figure (6.10) Load-Midspan Deflection Curves for SCB.Eo Beam
Figure (6.11) Load-Midspan Twisting Angle Curves for SCB.Eo Beam
167
Figure (6.12) Load-Midspan Deflection Curves for SCB.Esr Beam
Figure (6.13) Load-Midspan Twisting Angle Curves for SCB.Esr Beam
168
Figure (6.14) Load-Midspan Deflection Curves for SCB.Ecfrp Beam
Figure (6.15) Load-Midspan Twisting Angle Curves for SCB.Ecfrp Beam
169
Figure (6.16) Load-Midspan Deflection Curves for SCB.Mo Beam
Figure (6.17) Load-Midspan Twisting Angle Curves for SCB.Mo Beam
170
Figure (6.18) Load-Midspan Deflection Curves for SCB.Msr Beam
Figure (6.19) Load-Midspan Twisting Angle Curves for SCB.Msr Beam

171
Figure (6.20) Load-Midspan Deflection Curves for SCB.Mcfrp Beam
Figure (6.21) Load-Midspan Twisting Angle Curves for SCB.Mcfrp Beam
172
Figure (6.22) Load-Midspan Deflection Curves for SCB.Io Beam
Figure (6.23) Load-Midspan Twisting Angle Curves for SCB.Io Beam
173
Figure (6.24) Load-Midspan Deflection Curves for SCB.Isr Beam
Figure (6.25) Load-Midspan Twisting Angle Curves for SCB.Isr Beam
174
Figure (6.26) Load-Midspan Deflection Curves for SCB.Icfrp Beam
Figure (6.27) Load-Midspan Twisting Angle Curves for SCB.Icfrp Beam
175
Figure (6.28) Load-Midspan Deflection Curves for FCB.P Beam
Figure (6.29) Load-Midspan Twisting Angle Curves for FCB.P Beam
176
Figure (6.30) Load-Midspan Deflection Curves for FCB.Mo Beam
Figure (6.31) Load-Midspan Twisting Angle Curves for FCB.Mo Beam
177
Figure (6.32) Load-Midspan Deflection Curves for FCB.Msr Beam
Figure (6.33) Load-Midspan Twisting Angle Curves for FCB.Msr Beam
178
Figure (6.34) Load-Midspan Deflection Curves for FCB.Mcfrp Beam
Figure (6.35) Load-Midspan Twisting Angle Curves for FCB.Mcfrp Beam

179
6.7.2 Cracking and Ultimate Loads

The comparison between the theoretical (FEM by ANSYS) and
experimental results of cracking [P cr)theo , P cr)exp ] ,ultimate [P u)theo , P u)exp ]
loads is shown in Table 6.1. The table indicates a good agreement between
the theoretical and experimental results. The average difference in ultimate
loads is about 4.14%.
6.7.3 Deflection of Service Loads

Also, Table 6.1 includes the comparison of the theoretical (ANSYS)
midspan deflection δ theo and experimental midspan deflection δ exp at service
load, the average difference between the theoretical to experimental service
load deflections was 16%.
6.7.4 Crushing and Stresses in Curved Beams.

The theoretical F.E. analysis by (ANSYS) showed that the overall
compressive stress in concrete along the curved beam did not exceed the
compressive strength of concrete until failure occurred, except the points of
compression zone in touch with internal support and the corners of the
openings, which reached the crashing stress.
The effect of shear stresses were also found to be significant under the
applied load and internal support region.
Also, it was found that the maximum stress in CFRP laminates in
curved beams with opening near midspan still less than the allowable tensile
stress of CFRP laminates.
Appendix C shows the local stress distribution (σ x and τ xz ) along the
curved beam for concrete and σ x for CFRP laminates obtained from
theoretical analysis by Ansys for all tested specimens
180
Table 6.1: Theoretical and Experimental Cracking and Ultimate Loads

Cracking Load Ultimate Load Midspan Defl. at
Beam Pu(𝑎𝑛𝑠𝑦𝑠) δtheo
(kN) (kN) Service Load+,(kN)
Symbol Pu(exp) δexp
P cr)exp P cr)Ansys P u)exp P u)theo δexp δtheo
SCB.P 62.3 43.5 147.2 152.3 1.02 6.01 5.42 0.89
SCB.Eo 55.4 35 128.1 125 0.94 6.40 5.73 0.89
SCB.Esr 58.8 38.5 132.0 140 0.99 6.71 3.96 0.60
SCB.Ecfrp 72.7 45 141.9 144 0.99 6.6 4.19 0.63
SCB.Mo 34 21 83 84 0.89 7.15 8.89 1.24
SCB.Msr 41.5 30 100.4 112.5 0.96 4.05 3.10 0.76
SCB.Mcfrp 32 21 116 121 0.96 4.75 6.03 1.26
SCB.Io 34.6 27 76.2 79 0.92 4.21 5.31 1.26
SCB.Isr 41.5 33 99.5 108 0.92 5.44 3.95 0.72
SCB.Icfrp 31 25 100.9 110 0.93 4.91 5.02 1.02
FCB.P 103.8 73 380.8 408 0.99 4.31 3.78 0.87
FCB.Mo 58.8 32 190 192 0.94 6.15 7.4 1.20
FCB.Msr 55.4 36 305 325 0.95 8.19 9.63 1.17
FCB.Mcfrp 83 45 333.5 350 1.02 7.49 6.23 0.83
+Service load =0.70 * Pu ) exp
6.8 Parametric Study

A parametric study is performed to assess the influence of several
important parameters on the behavior of ultimate load investigation of
horizontally curved concrete beams with opening under the effect of
concentrated loading.
The selected parametric study to be discussed in this chapter

can be summarized as follows:
• Effect of beam curvature.

• Effect of type and size of opening
181
• Effect of opening length.

• Effect of opening Hight.
• Types of wrapping schemes of CFRP (U- wrap and full wrap)
6.8.1 Effect of Curvature

To study the effect of curvature on ultimate load capacity of
horizontally curved concrete beam with openings, four types of curved R/C
beams with different radius were analyzed in the present study (1m, 2m, 3m
and ∞). The study carried out on semicircular curved beam with opening
near interior support as shown in Figure (6.36). The load- midspan deflection
curves of all types are shown in Figure (6.37). A significant difference in
general response and ultimate load were noticed and recorded. An increase
in ultimate load of (103%, 130%, and 172%) in curved beams of radius (2, 3
and ∞) m respectively, compared with curved beams of radius 1 m. Table
6.2 shows that the ultimate load for these specimens.
Figure (6.36) SCB.Io Beams With Variable Radiuses
182
Figure (6.37) Load-Midspan Deflection Curves for SCB.Io Beams

With Variable Radiuses
Table 6.2 Ultimate Load for Different Types of SCB.Io with Variable Radiuses.
Increase In
Beam radius Ultimate Load(kN)
Ultimate Load %
Exp. R=1 m 76.16 --
F.E. R=1 m 79 --
F.E. R=2 m 160.5 103
F.E. R=3 m 182.1 130
F.E. R=∞ m 215.4 172
6.8.2 Effect of Type and Size of Opening

The effect of height/Length ratio of the opening on the load-deflection
curve, load-angle of twist curve and ultimate load capacity of a semicircular
curved beam with interior opening (SCB.Io) was also studied here in. In this
183
study three aspect ratios of 1:3(80*250),1:2(100*200) and 1:5(66.67*300)

were taken with constant area for all openings. A semicircular curved beam
with circular opening of radius 79.8mm at interior support which gives the
same area of rectangular section was also analyzed.
Figures (6.38) and (6.39) show the numerical results of the F.E.
analysis with experimental results of load-deflection and load-twisting angle
curves for SCB.Io curved beam with different opening type and dimensions.
It could be conclude that, as the height/length ratio of the opening decreases,
the load carrying capacity increase, furthermore a considerable increasing in
load carrying capacity in beam with circular opening was found. A summary
of the values of collapse loads obtained from F.E. analysis and experimental
test are listed in table (6.3).
Figure (6.38) Load-Midspan Deflection Curves for SCB.Io Beams With Variable
Opening Dimensions
184
Figure (6.39) Load-Midspan Twisting Angle Curves for SCB.Io Beams with
Variable Opening Dimensions
Table 6.3 Ultimate Load Capacity for Different Opening Dimensions of

SCB.Io Curved Beam
Beam Opening Opening Ultimate 𝑷𝒖

𝑷𝒖𝑺𝑪𝑩.𝑰𝒐
(SCB.Io) Length (mm) Height (mm) Load(kN)
(Exp.)200 100 76.16 --
200 100 79 1
Rectangular
250 80 85 1.07
300 66.67 85 1.07
Circular 79.8 79.8 111.8 1.41
185
6.8.3 Effect of Opening Length

The variation in load-deflection curve, load-angle of twist curve and
ultimate load capacity of a semicircular curved beam with interior opening
(SCB.Io) due to the variation of opening length were numerically carried out
herein. The height of the opening was kept constant (100 mm ) while its
length was increased from 100 mm to 300 mm by increment of 100mm.
Figures (6.40) and (6.41) show the results of the F.E. analysis to
gather with experimental results of load-deflection and load-twisting angle
curves for SCB.Io curved beam. It could be noticed that decreasing in
opening length leads to an increasing in ultimate load capacity of the beam.
The collapse loads obtained from this study are listed in table (6.4).
Figure (6.40) Load-Midspan Deflection Curves for SCB.Io Beams With

Variable Opening Length
186
Figure (6.41) Load-Midspan Twisting Angle Curves for SCB.Io Beams With
Variable Opening Length
Table 6.4 Ultimate Load Capacity for Different Opening Length of

SCB.Io Curved Beam
Opening Opening Ultimate 𝑷𝒖

Length (mm) Height (mm) Load(kN)
(Exp.)200 100 76.16 --

Beam
(SCB.Io) 100 100 90 1.14
200 100 79 1
300 100 70 0.88
187
6.8.4 Effect of Opening Height

The effect of opening height on the response of load-deflection curve,
load-angle of twist curve and ultimate load capacity of a semicircular curved
beam with interior opening (SCB.Io) was numerically carried out by fixing
the length of opening to 200mm and decreasing the depth from 50mm to
100mm by increments of 25mm.
Figures (6.42) and (6.42) show the results of the F.E. analysis with
experimental results of load-deflection and load-twisting angle curves for
SCB.Io curved beam with different opening height. The results show that
when opening height decreased, the ultimate load capacity of the beam
increased. The predicted values of collapse loads obtained from F.E. analysis
and experimental test are listed in table (6.5).
Figure (6.42) Load-Midspan Deflection Curves for SCB.Io Beams With

Variable Opening Height
188
Figure (6.43) Load-Midspan Twisting Angle Curves for SCB.Io Beams

with Variable Opening Height
Table 6.5 Ultimate Load Capacity for Different Height of SCB.Io Curved
Beam
Opening Opening Ultimate 𝑷𝒖

Length (mm) Height (mm) Load(kN)
Beam
(Exp.)200 100 76.16 --
(SCB.Io) 90
200 50 1.14
200 75 85 1.07
200 100 79 1.0
189
6.8.5 Wrapping Schemes of CFRP (U and Full Wrapping)

The effect of wrapping schemes on the behavior of semicircular
horizontally curved reinforced concrete beams with opening at midspan or interior
support had been studied. A full and U wrapping of CFRP laminates of thickness
0.131 mm were used for specimens SCB.Io and SCB.Mo. Figures (6.44), (6.45),
(6.46) and (6.47) show the numerical results of the finite element analyses of load-
deflection and load-twisting angle curves for both SCB.Io and SCB.Mo curved
beams. These figures reveal that the full wrapping CFRP laminates gives higher
ultimate load capacity than U wrapping CFRP laminates. Also, post-cracking
stiffness of curved beam with opening at midspan (SCB.Mo) was increased more
than post-cracking stiffness of curved beam with opening at interior support
(SCB.Io). This is because torsional forces at midspan region are greater than those
at internal support, also full wrapping CFRP laminates can resist higher torsional
stresses than U wrapping CFRP laminates. The numerical values of these
specimens are listed in table (6.6)
Figure (6.44) Load-Midspan Deflection Curves for SCB.Io Beams with

Different Wrapping Schemes
190
Figure (6.45) Load- Midspan Twisting Angle Curves for SCB.Io Beams
with Different Wrapping Schemes
Figure (6.46) Load-Midspan Deflection Curves for SCB.Mo Beams with

Different Wrapping Schemes
191
Figure (6.47) Load- Midspan Twisting Angle Curves for SCB.Mo Beams
with Different Wrapping Schemes
Table 6.6 Ultimate Load Capacity of Wrapping Schemes for SCB.Io and
SCB.Mo Curved Beams
Spacemen Full Wrapping U - Wrapping
Name Ultimate Load(kN) Ultimate Load(kN)
SCB.Io (Exp.) 100.93 --
SCB.Io 110 96
SCB.Mo(Exp.) 116 --
SCB.Mo 121 110
192
Chapter Seven Conclusions and Recommendations
CHAPTER SEVEN
CONCLUSIONS AND RECOMMENDATIONS FOR
FURTHER WORKS
7.1 General
The main objective of this research is to study the behavior of horizontally

reinforced concrete curved beams with and without openings, internally
strengthened by stirrups or externally strengthened by CFRP laminates. This
research included an experimental program as well as a nonlinear finite element
analysis by (ANSYS version 12.1 ) in order to predict the ultimate strength and
overall behavior of these tested specimen.
In this chapter, conclusions that are obtained from experimental and
analytical evidences as well as some recommendations for future extension
of the work will be presented.
7.2 Conclusions
The main conclusions observed from each phase of investigation

(experimental program and finite element analysis) for horizontally reinforced
concrete curved beam with and without openings, internally strengthened by
steel reinforcement or externally strengthened by CFRP laminates will be
presented in this section which represents the summary and benefit of this
research:
1- The presence of opening near midspan or interior support reduced the
ultimate load capacity about 35% for semi-circular beams and about
50% for full circular (ring) beam, if compared with control beam without
opening.
193
2- The presence of opening near exterior support of curved beam had lesser
effect on ultimate load capacity, about 13% , if compared with control

beam without opening.
3- The internal strengthening of the opening region by steel reinforcement
(stirrups), increased the ultimate load capacity about (3% to 30%) for
semi-circular curved beams, and about 60% for full circular (ring) beam,
if compared with unstrengthed beams.
4- The external strengthening (confinement) of the opening region by
CFRP laminates enhanced the ultimate load carrying capacity about

(11% to 40%) for semi-circular curved beams, and about 75% for full
circular (ring) beam, if compared with unstrengthed beams.
5- The simple method of design suggested by Mansure (1998) for internally
strengthening of opening region by steel reinforcement (stirrups) gave

generally a good response of strengthed curved beams in terms of
cracking patterns, deflection of service load as well as the mode of
failure changed from mode of opening failure to beam type failure
6- The simple method of design proposed here for externally strengthening
with CFRP laminates of opening region gave good result. The mode of
failure changed from opening mode failure to a beam type failure.
7- The proposed simple method of design for external strengthening with
CFRP laminates of the opening region, enhanced the general behavior of

strengthened curved beams in terms of cracking patterns, deflection of
service load as well as the mode of failure changed from mode of
opening failure to beam type failure.
8- Both internal strengthened and external confinement of curved beams
with opening near exterior support had a small effect on the maximum
deflection and rotation (angle of twist ) of midspan section at ultimate
load value. While for beams with opening at midspan a decrease of (33%
and 66%) occurred for deflection and rotation of midspan section.
194
Furthermore, the deflection and rotation for beams with opening near
interior support increased about 28% and 38% respectively.
9- The general response of externally strengthened specimens by CFRP
laminates was approximately in agreement with specimens of internally

strengthened by steel reinforcement (stirrups) in terms of deflection and
ultimate loads.
10- The numerical F.E. analysis by ANSYS package is valid for the
analysis of horizontally curved concrete beams with openings,
unstrengthened or strengthened by internal reinforcement or CFRP
laminates. The general response of load deformation curves (deflection
and rotation) of F.E. analysis gave a good approchment with those of
experimental curves. The comparison with experimental results
confirmed the validity of the analysis with maximum deviation by about
11% in ultimate load capacity.
11- The average difference of deflections at service loads and ultimate load
capacity of all specimens between numarical and experimental results is

about 4.8% and 4.1%, respectively, which insure the validity of the F.E.
solution.
12- The ultimate load capacity of the curved beam with opening increased
with decreasing the curvature (1/R) for the same length by about
(103%,130% and 172%) for curvature ( 0.5,0.33 and 0.0).
13- Circular shape of opening gave an increase in ultimate load capacity
about 41% more than rectangular openings.

14- Opening with length more than height for the same area gave a best
ultimate load capacity than those of length less or equal height by about
7%.
15- Using of U-scheme for CFRP laminates gave a decreasing in load
capacity of the curved beam reached to 12% if compared with full

scheme of CFRP laminates.
195
7.3 Recommendations for Further Works
1. Investigate the behavior of horizontally curved reinforced concrete

beams with and without opening subjected to dynamic and impact
loading.
beams with different height/radius (h/R) ratios and/or different shapes
(like parabolic or elliptical).
3. Investigate experimentally effect of type of FRP on behavior of
horizontally curved reinforced concrete beams without openings.
4. Include effect of bond slip between concrete and reinforcing steel bars.
5. Investigate the behavior of tapered horizontally curved reinforced
concrete beams with and without openings.
beams with opening under repeated loads.
7. Investigate experimentally effect of other types of supports of semi-
circular curved beam.
196
References
References
(1) ACI Committee 318, (2011) "Building Code Requirements for Structural
Concrete (ACI318M.11) and Commentary", American Concrete
Institute, Farmington Hills, Michigan, USA, 473 pp.
(2) ACI Committee 440, (2002) "Guide for the design and Construction of
externally Bonded FRP Systems for Strengthening of Concrete
Structures", American Concrete institute, Michigan, USA.
(3) ACI 211.1-97, (Re-approved 1997), "Standard practice for selecting
proportions for normal, heavy weight and mass concrete", ACI Manual
of Concrete Practice, American Concrete Institute, Detroit, 1997.
(4) Ali A.Y., (2010), " Three-Dimensional Nonlinear Finite Element
Analysis of Reinforced Concrete Horizontally Curved Deep Beams",
Journal of Babylon University, Vol.18, No.1.
(5) Ali A.Y. et el , (2011), " Analysis of Reinforced Concrete Beams with
Openings and Strengthened by (CFRP)", Journal of Babylon University,
Vol.19, No.3.
(6) Al-Mutairee H.M , (2013), " Effect of Non-Uniform Distribution of
Longitudinal Reinforcement on the Behavior of Reinforced Concrete
Horizontally Curved Beams with Fixed-Ends", Journal of Babylon
University, Vol.21, No.3.
(7) Abdulla, H. A., Torkey A. M., Haggag H. A. and Abu-Amira A. F. ,
(2003) ”Design Against Crack at The Opening in The Reinforced
Concrete Beam Strengthened With Composite Sheet”, Composite
Structure Journal, No.60, pp.197-204.
(8) Al-Temeemi M.A,(2002) "Nonlinear Analysis of Reinforced Concrete
Horizontally Curved Beams on Elastic Foundation", M.Sc. Thesis,
University of Babylon.
(9) "ANSYS Manual", Version (12.1), USA, 2009.
197
References
(10) American Society of Testing and Material (ASTM), "Standard

Specification for Concrete Aggregates"(2002), ASTM C33-02, West
Conshohocken, PA.
(11) ASTM Designation C370-05a "Standard specification for testing
method and definitions for mechanical testing of steel products", (2005)
Annual Book of ASTM Standards, American Society for Testing and
Material, Philadelphia, Pennsylvania, Section 1, Vol. (1.01), pp 248-287.
(12) ASTM Designation C39-01 "Standard specification for testing method
for compressive strength of cylindrical concrete specimens", (2001)
Annual Book of ASTM Standards, American Society for Testing and
Material, Philadelphia, Pennsylvania, Section 4, Vol. (4.02), pp 20-24.
(13) ASTM Designation C496-96 "Standard specification for splitting
tensile strength of cylindrical concrete specimens", (1996) Annual Book
of ASTM Standards, American Society for Testing and Material,
Philadelphia, Pennsylvania, Section 4, Vol. (4.02), pp 1-4.
(14) Al-Kaisy, A.I.D,(2005),''Nonlinear Finite Element Analysis of
Reinforced Concrete Beams with Large Opening under Flexure'' M.Sc.
Thesis,University of Technology, Iraq.
(15) Al-Dolaimy, A.T., (2011), " Structural Behavior of Continuous
Reinforced Concrete Beams with Openings and Strengthened by CFRP
Laminates'' M.Sc. Thesis, university of Babylon.
(16) Al-Tai, S., L., (2010) "Nonlinear Behavior Of Reinforced Concrete T-
Beams Strengthened With CFRP Subjected To Shear". M. Sc. Thesis,
College of Engineering ,University of Babylon, IRAQ.
(17) Badawy, H. E. I, Jordaan, I. J., and McMullen, A. E., "Effect of Shear
on Collapse of Curved Beams", Journal of the Structural Division,
ASCE, Vol. 103, No. ST9, Proc. Paper 13185, September 1977, pp.
1849-1866.
198
References
(18) Bandyapadhya J.N et el ,(2011), " Strengthening and rehabilitation of

reinforced concrete beams with opening", International Journal of Civil
and Structural Engineering" , Vol 2, No 1.
(19) Banzant, Z. P., (1983)"Comment on Orthotropic Models for Concrete
and Geomaterials", Journal of Engineering Mechanics Division, ASCE,
Vol. (106), No. 3, pp. 849-865.
(20) Burton, K.T,(1965),'' Influence of Embedded Service Ducts on The
Strength of Continuous Reinforced Concrete T-Beams''ACI Journal,
Vol. 62, October , pp.1327-1344. Cited by (Al-Thabhawee,20012)
(21) Bisby, L.A. and Williams, B.K. (2004) ,"An Introduction to FRP
Strengthening of Concrete Structures” Department of Civil
Engineering, Queen’s University, Canada, pp.4.
(22) Cervera, M. Hinton., E. and Hassan, O., (1987),''Nonlinear Analysis of
Reinforced Concrete Plate and Shell Structures Using 20 – Noded
Isoperimetric Brick Elements'', Computers and Structures, Vol.25, no.6,
pp.845 – 869
(23) Chen, W. F., (1982) "Plasticity in Reinforced Concrete", McGraw-Hill,

pp. 471.
(24) Chen, W. F. and Saleeb, A. F., (1981)"Constitutive Equations for
Engineering Materials", West Lafayette, Indiana, pp. 580.
(25) Chu, K., and Thelen, A., "Plastic Analysis of Circular Balcony Girders",
Journal of Structural Division, ASCE, Vol. 89, No. ST6, Proc. Paper
3726, December 1963, pp. 159-185.
(26) Carolin, A., (2003)'' Carbon Fiber Reinforced Polymers for
Strengthening of Structural Elements '' Ph.D. Thesis, University of
Lulea, Sweden. Cited by(Al-Tai,2010).
199
References
(27) Dalal, M.S. , (1969),'' Behavior and Design of Opening Reinforced

Concrete T-Beams'' Ph.D. Thesis, University of Taxas Austin, U.S.A.
Cited by (Al-Thabhawee,2012).
(28) Dawlat, D. A., (2007), "Experimental and Theoretical Investigation of
the Behavior of Reinforced Concrete Beams Strengthened by Fiber
Reinforced Polymer", Ph. D. Thesis, University of Baghdad/College of
Engineering, January
(29) Federico M. Mazzolani (2001) "The Use of FRP Materials for The
Seismic Upgrading of Existing RC Structures” University degli, Facoltà
di Ingegneria
(30) Gibson, R. F., (1994) "Principles of Composite Material Mechanics",

McGraw-Hill, Inc., New York.
(31) Hamzah B.A,(2012) " Behavior of RC Curved Beams with Openings
and Strengthened by CFRP Laminates ", P.hd. Thesis, University of
Basrah.
(32) Hanson, J.M., (1969),''Square openings in webs of continuous joists''.
PCA Research and Development Bulletin. RD100 .01 D, Portland
Cement Association, Stokie, IL, pp. l-14. (Cited by Mansur, M.A.
1998).
(33) Hii, A. K. Y., and Al-Mahaidi, R., (2006) "An Experimental and
Numerical Investigation on Torsional Strengthening of Solid and Box-
Section RC Beams Using CFRP Laminates", Department of Civil
Engineering, Monash University, Melbourne, Vic. 3800, Australia.
(34) Hsu, T. T. C., Inan, M., and Fonticiella, L., "Behavior of Reinforced
Concrete Horizontally Curved Beams", ACI Journal, Vol. 75, No. 4,
April 1978, pp. 112-123.
200
References
(35) Hughes, B. P. and Chapman, G. P., (1966) "The Complete Stress-Strain

Curve for Concrete in Direct Tension", Bulletin RILEM, No. 30, pp.
95-97.
(36) Iraqi Specification No.5, "Portland Cement",Baghdad,(1984).
(37) Iraqi Specification No.45, "Natural Sources for Gravel that is used in
concrete and construction", Baghdad,(1984).
(38) Jindal, R.L.,(1976),''behavior of reinforced concrete rectangular beam
having large openings''zanco ;sul university vol.2, pp.11-25.Cited by
(Al-Thabhawee, 2012).
(39) Jordaan, I. J., Khalifa, M. M. A., and McMullen, A. E., "Collapse of
Curved Reinforced Concrete Beams", Proceedings, ASCE, Vol. 100,
ST11, November 1974, pp. 2255-2269.
(40) Kaw, A. K., (1997) "Mechanics of Composite Materials", CRC Press
LLC, Boca Raton, Florida, USA.
(41) Khalifa, M. M. A., "Collapse of Reinforced Concrete Beams Curved in
Plan", thesis presented to the University of Calgary, at Calgary, Alberta,
Canada, in 1972, in partial fulfillment of requirements for the degree of
Master of Science.
(42) Kupfer, H., and Grestle, K., (1973) "Behavior of Concrete Under
Biaxial Stress", ASCE-Journal of Engineering Mechanics, Vol. (104),
No. (EM4).
(43) Mackava, K., and Okamura, H., (1983) "The Deformation Behavior
and Constitutive Equation of Concrete Using the Elasto-Plastic and
Fracture Model", Journal of Faculty of Engineering, University of
Tokyo (b), Vol. (xxxvii), No. 2.
(44) MacGregor J. G., (1992) "Reinforced Concrete Mechanics and Design",
Prentice-Hall, Inc., Englewood Cliffs, NJ.
201
References
(45) Maurizio, G., Kypros, P., and Peter, W., (2002) "Shear Performance of
FRP Reinforced Concrete Beam", Journal of Reinforced Plastics and
Composites, No. 22, pp. 1389.
(46) Mansur, M.A., and Hasnat A. (1979) “Concrete Beams with Small
Openings under Torsion” Journal of the Structural Division, ASCE,
Vol. 105, No. ST11, November, pp. 2433-2447.
(47) Mansur, M.A., and Alwis W.A.M., (1984), ''Reinforced Fiber Concrete
Deep Beams with Web Opening'' The International Journal of Cement
Composites and Lightweight Concrete, Vol. 6, No. 4, November, pp.
263-271.
(48) Mansur, M. A., and Rangan, B. V., "Study of Design Methods for
Reinforced Concrete Curved Beams", ACI Journal, Vol. 78. No.3,
May-July 1981 pp. 226-231.
(49) Mansur, M.A., Tan K.H., and Lee, S.L., (1984) “Collapse Load of
Reinforced Concrete Beams with Large Opening” Journal of the
Structural Division, ASCE, Vol. 110, No. ST11, November, pp. 2602-
2618.
(50) Mansur, M.A., and Paramasivam P., (1984) “Reinforced Concrete
Beams with Small Opening in Bending and Torsion” ACI Structural
Journal, Vol. 81, No. 2, March-April, pp.180-185.
(51) Mansur, M.A., Tan K.H., and Lee S.L., (1985) “Design Method for
Reinforced Concrete Beams with Large Openings” ACI Structural
Journal, Vol. 82, No. 4, July-August, pp. 517-524.
(52) Mansur, M.A., Tan K.H., and Lee, S.L., (1991) “Test on Reinforced
Concrete Continuous Beams with Opening” Journal of the Structural
Structural, ASCE, Vol. 117, No. ST6, June, pp. 1593-1606.
(53) Mansur M.A., Hung L.M., Tan K.H., and Lee S.L., (1992) “Deflections
of Reinforced Concrete Beams with Web Openings” ACI Structural
Journal, Vol. 89, No. 4, July-August, pp. 391-397.
202
References
(54) Mansur, M.A., (1998) “Effect of Openings on the Behavior and

Strength of R/C Beams in Shear” Cement and Concrete Composites,
20, pp. 477-486.
(55) Mansur M.A. and Tan K.H., (1998),"Concrete Beam with Opening
Analysis and Design'' CRC press LLC, Florida, U.S.A, 33441.
(56) Mansur, M.A., (2006) '' Design of Reinforced Concrete Beams with
Web Openings '' Proceeding of the 6th Asia-Pacific Structural
Engoneering and Construction Conference (APSEC 2006), 5-6
September, Kuala Lumber, Malaysia.
(57) Mohammed, T.J., (2004),'' Analysis of Reinforced Concrete Beams
with Large Opening Under Torsion ''M. Sc. Thesis, University of
Technology , Iraq.
(58) Mottram, T., and Shaw, C., (1996), ''Using Finite Element in
Mechanical Design'', McGraw-Hill Book Company, London
(59) Nasser, K.W., Acavalos, A. and Daniel, H.R., (1967), ''Behavior and
Design of Large Opening in Reinforced Concrete Beams'' ACI Journal,
Vol. 64, No.1, January, pp.25-33. Cited by (Mansur et al,1985)
(60) Najim, W.N., (2009), ''Experimental and Theoretical Investigation of
the Behavior of R.C. Deep Beams with Openings Strengthened by
CFRP Laminates''M.Sc. Thesis, university of Babylon.
(61) Neff, C., Schnellenbach-Held, M., Daus, S., (2002) “RC Beams with
one Large Rectangular Opening-Experimental Investigations” DaCon
Vol.17, Darmstadt.
(62) Ngo, D. and Scordelis, A.C. (1967). "Finite Element Analysis of
Reinforced Concrete Beams," Journal of ACI,Vol. 64, No. 3, pp. 152-
163.
(63) Patel and Saksena (2013),:" Experimental of Reinforced Concrete
Beams With Web Opening", International Journal of Advanced
Engineering Research and Studies", Vol. II/Issue III/pp 66-68.
203
References
(64) Salam, S.A., (1977), ''Beams with openings under different stress
conditions'', Proceedings of 3rd Conference on Our World in Concrete
and Structures, CI-Premier, Singapore, 25-26 Aug., pp. 259-267. (Cited
by Mansur, M.A. 1998)
(65) Shah, S. P., Swartz, S. E., and Ouyang, C., (1995) " Fracture Mechanics
of Concrete ", Jhon Wiley & Sons, Inc., New Yourk.
(66) Somes, N.F. and Corley, W.G., (1974), ''Circular openings in webs of
continuous beams'', SP-42, American Concrete Institute, Detroit, MI,
pp. 359- 398. (Cited by Mansur, M.A. 1998).
(67) Sika CarboDur FRP Composites Constrection for Repair &
Strengthening of Structures .. MARCH 2003
(68) Suzuki, A., Masashi, A., Kengo, H. and Shoichi, O., (2007), ''High
Strength Deformed Bar-in-Coil “HDC800” for Shear Reinforcement of
RC Beam with Web-opening'' Nippon Steel Technical Report, No. 96
July, pp.67-74.
(69) Tan K.H., Mansur M.A., and Huang L.M., (1996) “Reinforced
Concrete T-Beams with Large Web Opening in Positive and Negative
Moment Regions” ACI Structural Journal, Vol. 93, No. 3, May-June,
pp. 277-289.
(70) Tan K.H., and Mansur M.A., (1996) “Design Procedure for Reinforced
Concrete Beams with Large Web Openings” ACI Structural Journal,
Vol. 93, No. 3, July-August.
(71) Tavarez, F.A., (2001), ''Simulation of Behavior of Composite Grid
Reinforced Concrete Beams Using Explicit Finite Element Methods'',
M.Sc. Thesis, University of Wisconsin-Madison, Madison, Wisconsin
(72) Weng, W., Tan, K.H., and Mansur, M.A., (1999), '' Effects of Creating
an Opening in Existing Beams''ACI Structural Journal, V. 96, No. 6,
November-December, pp.899-906.
204
References
(73) Witte, F. C. D. and Kikstra, W. P., (2002) "DIANA Finite Element Analysis -
User's Manual, Release 9.0", TNO Building and Construction research,
Netherlands.
(74) Wischers, G., (1978) "Application of Effects of Compressive Loads on
Concrete", Betontechnische Berichte, No. 2 and 3, Duesseldorf,
Germany.
(75) Willam, K., and Warnke, E., (1974) "Constitutive Model for the
Triaxial Behavior of Concrete", Proceedings, International Association
for Bridge and Structural Engineering, Vol. (19), ISMES, pp. 174,
Bergamo, Italy.
(76) Wolanski, I. J., (2001) "Flexural Behavior of Reinforced and
Prestressed Concrete Beams Using Finite Element Analysis " , M.Sc.
Thesis, Marquette University, U.S.A.
(77) Yousifani, A. H., (2004) "Investigation of the Behavior of Reinforced
Concrete Beams with Construction Joints Using Nonlinear Three –
Dimensional Finite Element", M. Sc. Thesis, University of Technology,
pp. 123.
205
Appendix – A -
Appendix – A -
Analytical Solution and Design of Semicircular Curved Beam (Control Beam)
P
y'
c P
y'
Radius of Semicircular Curved Beam = r

Applied Load = p …… for each midspan
Reaction at A = ra = rb =r1
Reaction at C = r2
Taking moment of forces about y'-y' axis (through support C) :
2×r1×r= 2× p (r - r× sin (45))
r1= r2 = 0.2929 p
ƩFy = 0
r2 = 1.4142 p
By taking a section before and after the applied load we can find the shear, bending
moment and torque equations.
• Shear, Moment and Torsion calculation

1. For θ < 45
V = 0.2929p
Mθ = 0.2929p×r×sin (θ)
Tθ = 0.2929p×r(1-cos(θ))
2. For 45 < θ < 90

V= 0.2929p – p = -0.707p
Mθ=0.2929p×r×sin (θ) - p×(r×sin(θ-45))
Tθ = 0.2929p×r(1-cos(θ)) - p×r×(1- cos(θ-45))
A1
Appendix – A -
• Checking flexural capacity of the section
Mu =ρ.b.d2.fy(1-0.59ρ.fy/fc') Ø6 @ 4.5°
ρ= As/bd = 113×2/(125×213) = 0.008488263

2Ø12
fc'= 31 Mpa
fy = 550 Mpa 250 mm
Mu= 24.123 kN.m
2Ø12
for θ = 45 the value of the load p could be found from:
Mθ = 0.2929p×r×sin (45)
p= 119.3 kN
125 mm
also, for θ =90 the value of the load p could be found from:
Mθ = 0.2929p×r×sin (90) - p×(r×sin(90-45))
p= 59.64 kN
• Checking shear capacity of the section
Vc= 0.17× (fc')0.5×b×d

=0.17× (31)0.5×125×213/1000 = 25.2 kN
Vs= Av.fy.d/s
=2×25.5×520×213/90/1000 = 62.7 kN
Vu = Vc +Vs = 87.9 kN
Maximum shear occurs at interior support = 0,707p
0.707p = 87.9
P = 124.4 kN
Therefore , the control ultimate load is the least of the upper three values of ' P '
which equal to 59.64 kN
The shear force, bending moment and torque diagrams could be drawn for this load as
shown below:
A2
Appendix – A -
Shear Force Diagram For Half of

Semicircular kN
30
20
10
0
-10 0 10 20 30 40 50 60 70 80 90 Shear
-20
-30
-40
-50
Moment Diagram For Half of

Semicircular kN.m
15.00
10.00
5.00
0.00
-5.00 0 10 20 30 40 50 60 70 80 90
Moment
-10.00
-15.00
-20.00
-25.00
-30.00
Torqu Diagram For Half of

Semicircular kN.m
8.00
7.00
6.00
5.00
4.00
Torqu
3.00
2.00
1.00
0.00
0 10 20 30 40 50 60 70 80 90
A3
Appendix – A -
• Checking torsional reinforcement of the section
The value of maximum torque could be found from the toque diagram which equal to
6.5 kN.m
The section properties of the section as required from ACI code -318 are :
Acp = 31250 mm2

Pcp = 750 mm2
Xo = 75 mm
Yo = 200 mm
Aoh = 15000 mm2
Ao = 0.85 Aoh = 12750 mm2
Ph= 550 mm
At/s = Tu/(2.Ao.fy)
= 6.5×106/(2*12750*550) = 0.423 mm2/mm
Vc = 0.17× (31)0.5×125×213/1000 = 25.2 kN
Shear force v = 42.17 kN
Vs = V-Vc = 42.17 - 25.2 = 16.9 kN
Av/s = Vs/fy.d = 16.9×1000/( 520×213) = 0.15 mm2/mm
A (v+t) /s = 0.15+0.423 = 0.573 mm2/mm
Assume Φ6 mm stirrups Av= 55 mm2
S = 55 / 0.573 = 95 mm
So, use Φ6 mm stirrups @ 90 mm
Al = At/s × Ph = 0.423× 550 = 232 mm2
So, the use of 4 Φ 12 mm longitudinal reinforcement will satisfy the longitudinal

torsion and bending moment requirements which provide area of steel equal to
(4*113.0 mm2 = 452 mm2).
• For ring beam, a Fortran code program was written to get the forces along the
beam and then the same procedure used to for the analysis and design
A4
Appendix – B -
Appendix – B -
1- Design of Steel Reinforcement for strengthening of

Internal Opening of Semicircular Curved Beams
R = 1.0625 m
fc'= 30.3 Mpa
fy= 586 Mpa
fyv = 550 Mpa 4Ø12
h = 250 mm 250 mm
b = 125 mm
co = 25 mm
d = 250 – 25 – 5.7 – 12/2 = 213 mm
dv = 250 – (2*25+2*5.7+10/2+10/2) = 178.6 mm
μ = 0.0375 125 mm
2Ø12 mm , As provided =226 mm2 along the curved beam for each top and
bottom.
Dimension of opening = (100 × 200) mm

Vu (at center of opening) = 40.7 kN
Mu (at center of opening) = 17.53 kN.m
Tu (at center of opening) = 3.76 kN.m
• Beam Type Failure

• Longitudinal Reinforcement
The required longitudinal reinforcement can be found from :
M o(1) = (3.76 +40.7 *0.0375)*�1 + 2/2 + 17.53 = 25.00 Kn.m
→ As = 249.0 mm2
M o3 = (3.76+ 40.7*0.0375)�1 + 2/2 – 17.53 = -10.05 kN.m
A5
Appendix – B -
Checking Mode 2 failure from the following equation:
Meq(2) = (3.76+ 40.7*0.075/2)�1 + 2/2 = 7.5 kN.m

As = 69 mm2
Required area of steel reinforcement for whole section

As total =69+0+249
= 318 mm2
Available area of steel reinforcement for whole section

As available = 4*113
= 452 mm2
then no additional reinforcement required.
• Design of Stirrups
a) For torsion
Xo = 125-2×25 = 75 mm
Yo = 250 - 2×25 = 200 mm
M eq(1) = 25.00 – 17.53 = 7.5 Kn.m
𝐴𝑤 𝑀𝑒𝑞(1) 7.5 × 106

= = = 0.226 𝑚𝑚2 /𝑚𝑚
𝑠 4𝑋𝑜. 𝑌𝑜. 𝐹𝑦𝑣 4 × 75 × 200 × 550
b) For Shear
Vc = 1/6 �𝑓𝑐′ bw.d
Vc = 1/6 √30 ×125× 213 × 10−3 = 17.5 𝑘𝑁
Vu(sc)= Vu+1.2*Tu/b
Vu(sc)=40.7+1.2*3.67/0.125=75.9 kN
Vu (eq) = Vu(sc)-Vc = 75.9 -17.5 = 58.4 kN
𝐴𝑤 58.4 × 103
= = 0.498
𝑠 213 × 550
𝐴𝑤 2∗28.3
𝑠= = = 112.33 𝑚𝑚
0.498 0.498
A6
Appendix – B -
then use two bar Ø6mm at each side of the opening spaced at a distance
of 40 mm which will satisfy the shear requirement at that section.
• Frame Type Failure

a) Bottom Chord
Vertical Shear
V ub = Vu / 2 = 40.7 /2 = 20.35 kN
𝐴𝑣 𝑉𝑢𝑏 20.3 × 103
= = = 0.66 𝑚𝑚2/ 𝑚𝑚
𝑠 2𝑑. 𝐹𝑦𝑣 38 × 550
2 ∗ 28.3
𝑆 = = 85.5 𝑚𝑚
0.66
Lateral Shear
Vuz= Tu/(θt+θb) = 3.67/(0.175) = 20.97 kN

𝐴𝑣 𝑉𝑢𝑧 20.97 × 103
= = = 0.42 𝑚𝑚2/ 𝑚𝑚
𝑠 2𝑑𝑤. 𝐹𝑦𝑣 91 × 550
𝐴𝑤 2∗28.3
𝑠= = = 52.4 𝑚𝑚
0.72+0.66 0.42+0.66
Use Ø6mm @ 40 mm bars for bottom chord, which will satisfy the shear
requirement.
a) Member above the opening ( compression chord)

Since the dimensions of the section of this chord match’s the bottom
chord dimensions and forces, but with a compression axial force, then the
use of the same spacing of the stirrups will be considered as shown in
Figure (B-2).
• Design for Crack Control
𝑉𝑢 40.7 × 103
𝐴𝑑 = = = 104.65 𝑚𝑚2
𝐹𝑦. 𝑠𝑖𝑛𝛼 550 × sin(45)
Use 4Ø6mm diagonal bars in each direction for each corner of the
opening
A7
Appendix – B -
2Ø6mm 2Ø6mm
6Ø6mm
2Ø6mm
4Ø6mm
6Ø6mm 4Ø6mm
4Ø6mm Diagonal bar
75 2Ø12mm
6Ø6mm
Closed
100 Opening Stirrups
2Ø6mm
75 2Ø12mm
125
A8
Appendix – B -
2- Design of CFRP for External Strengthening of Internal

Opening of Semicircular Curved Beams
R = 1.0625 m
fc'= 30.3 Mpa
fy= 586 Mpa
fyv = 550 Mpa 4Ø12
h = 250 mm 250 mm
b = 125 mm
co = 25 mm
d = 250 – 25 – 5.7 – 12/2 = 213 mm
dv = 250 – (2*25+2*5.7+10/2+10/2) = 178.6 mm
μ = 0.0375 125 mm
2Ø12 mm , As provided =226 mm2 along the curved beam for each top and
bottom.
Dimension of opening = (100 × 200) mm

Vu (at center of opening) = 40.7 kN
Mu (at center of opening) = 17.53 kN.m
Tu (at center of opening) = 3.76 kN.m
• Beam Type Failure
• Longitudinal Reinforcement
The required longitudinal reinforcement can be found from :
M o(1) = (3.76 +40.7 *0.0375)*�1 + 2/2 + 17.53 = 25.00 Kn.m
→ As = 249.0 mm2
M o3 = (3.76+ 40.7*0.0375)�1 + 2/2 – 17.53 = -10.05 kN.m
Checking Mode 2 failure from the following equation:
A9
Appendix – B -
Meq(2) = (3.76+ 40.7*0.075/2)�1 + 2/2 = 7.5 kN.m

As = 69 mm2
Required area of steel reinforcement for whole section

As total =69+0+249
= 318 mm2
Available area of steel reinforcement for whole section

As available = 4*113
= 452 mm2
then no additional CFRP Laminates required.
• Design of Stirrups
a) For torsion
Xo = 125-2×25 = 75 mm
Yo = 250 - 2×25 = 200 mm
M eq(1) = 25.00 – 17.53 = 7.5 Kn.m
Aw/s= 7.5/(4*0.075*0.2*4300)=0.029 mm2/mm

s =50*0.131/0.029 = 252.8 mm and this is too much spacing, check shear
requirement
b) For Shear
Vc = 1/6 �𝑓𝑐′ bw.(d-do)
Vc = 1/6 √30 ×125× (213 − 100) × 10−3 = 17.5 𝑘𝑁
Vf = Vu – Vc = 40.7 - 17.5 = 23.2 kN
23.2 × 103
𝐴𝑓 = = 5.39 𝑚𝑚2
4300
𝐴𝑓 6
𝑏𝑓 = = = 41.8 𝑚𝑚
0.131 0.131
use bf = 25 mm full wrapping on each side of the opening, which will

satisfies torsional requirement too.
use bf = 25 mm full wrapping on each side of the opening
A10
Appendix – B -
• Frame Type Failure

b) Bottom Chord
Vertical Shear
Vc = 0 because of tension state of bottom chord
Vub = Vu / 2 = 40.7 /2 = 20.35 kN
Vf = Vu = 20.35 kN
Avf = Vf / Ffu
Avf = 20.35× 103 / 4300 = 4.7 mm2
bf = (Avf / 2) / tf = (4.7 / 2) / 0.131 = 18.06 mm
Lateral Shear
Vuz= Tu/(θt+θb) = 3.67/(0.175) = 21 kN
Vf = Vuz = 21 kN
Avf = Vf / Ffu
Avf = 21× 103 / 4300 = 4.877 mm2
bf = (Avf / 2) / tf = (4.877 / 2) / 0.131 = 18.6 mm
use three 20 mm CFRP laminates full wrapping for bottom chord
b) Member above the opening ( compression chord)

Since the dimensions of the section of this chord match’s the bottom
chord dimensions and forces, but with a compression axial force, then the
use of the same spacing of the stirrups will be considered as shown in
Figure (B-3).
25cm 25 cm
2 cm
A11
Appendix – C -
Cracks Pattern And Stress In Concrete And CFRP Laminates
Figure (C-1) Axial Stress in x-direction of specimen SCB.Eo on outer face
Figure (C-2) Shear Stress τ xz of SCB.Eo Specimen on outer face
A12
Appendix – C -
Figure (C-3) Cracks Pattern at Internal Support of SCB.Eo Specimen

at Load 125 kN
Crashing
Figure (C-4) Axial Stress in x-direction of specimen SCB.Esr on outer face
A13
Appendix – C -
Figure (C-5) Shear Stress τ xz of SCB.Esr Specimen on outer face
Figure (C-6) Cracks Pattern of SCB.Esr Specimen
A14
Appendix – C -
Figure (C-7) Axial Stress in x-direction of specimen SCB.Ecfrp on outer face
Crashing
Figure (C-8) Shear Stress τ xz of SCB.Ecfrp Specimen on outer face
A15
Appendix – C -
Figure (C-9) Stresses in Y-direction of CFRP for Specimen SCB.Ecfrp
Figure (C-10) Cracks Pattern of SCB.Ecfrp Specimen
A16
Appendix – C -
Figure (C-11) Axial Stress in x-direction of specimen SCB.Mo on outer face
Figure (C-12) Shear Stress τ xz of SCB.Mo Specimen on outer face
A17
Appendix – C -
Crashing
Figure (C-13) Axial Stress in x-direction of specimen SCB.Msr on outer face
Crashing
Figure (C-14) Shear Stress τ xz of SCB.Msr Specimen on outer face
A18
Appendix – C -
Figure (C-15) Cracks Pattern of SCB.Ecfrp Specimen
Figure (C-16) Axial Stress in x-direction of specimen SCB.Mcfrp on outer face
A19
Appendix – C -
Crashing
Figure (C-17) Shear Stress τ xz of SCB.Mcfrp Specimen on outer face
Figure (C-18) Stresses in Y-direction of CFRP for Specimen SCB.Mcfrp
A20
Appendix – C -
Figure (C-19) Cracks Pattern of SCB.Mcfrp Specimen
Figure (C-20) Axial Stress in x-direction of specimen SCB.Io on outer face
A21
Appendix – C -
Crashing
Figure (C-21) Shear Stress τ xz of SCB.Io Specimen on outer face
Figure (C-22) Cracks Pattern of SCB.Io Specimen
A22
Appendix – C -
Figure (C-23) Axial Stress in x-direction of specimen SCB.Isr on outer face
Figure (C-24) Shear Stress τ xz of SCB.Isr Specimen on outer face
A23
Appendix – C -
Figure (C-25) Cracks Pattern of SCB.Isr Specimen
Figure (C-26) Axial Stress in x-direction of specimen SCB.Icfrp on outer face
A24
Appendix – C -
Figure (C-27) Shear Stress τ xz of SCB.Icfrp Specimen on outer face
Figure (C-28) Cracks Pattern of SCB.Icfrp Specimen
A25
Appendix – C -
Figure (C-29) Stresses in Y-direction of CFRP for Specimen SCB.Icfrp
Figure (C-30) Axial Stress in x-direction of specimen FCB.Mo on outer face
A26
Appendix – C -
Figure (C-31) Shear Stress τ xz of SCB.Mo Specimen on outer face
Figure (C-32) Cracks Pattern of SCB.Mo Specimen
A27
Appendix – C -
Figure (C-33) Axial Stress in x-direction of specimen FCB.Msr on outer face
Figure (C-34) Shear Stress τ xz of SCB.Msr Specimen on outer face
A28
Appendix – C -
Figure (C-35) Cracks Pattern of SCB.Msr Specimen
Figure (C-36) Axial Stress in x-direction of specimen FCB.Mcfrp on outer face
A29
Appendix – C -
Figure (C-37) Shear Stress τ xz of SCB.Mcfrp Specimen on outer face
Figure (C-38) Stresses in Y-direction of CFRP for Specimen FCB.Mcfrp
A30
Appendix – C -
Figure (C-39) Cracks Pattern of SCB.Mcfrp Specimen
A31
Ξ΋ΎΗϧϟ΍ϭ ΓΩΩΣϣϟ΍ έλΎϧόϟ΍ ΔϘϳέρ ϥϣ ΔϠλΣΗϣϟ΍ Ξ΋ΎΗϧϟ΍ ϥϳΑ ϕϓ΍ϭΗ ϰϠϋ ϝϭλΣϟ΍ ϡΗ ϡΎϋ ϝϛηΑϭ
ϭ %4,1ϕέϓϝΩόϣΑϭ ϯϭλϘϟ΍ϝΎϣΣϻ΍ϭϕϘηΗϟ΍ϝΎϣΣ΃ϭϝηϔϟ΍ρϣϧϭϝϭρϬϠϟ ΓέϓϭΗϣϟ΍ΔϳέΑΗΧϣϟ΍
.ϲϟ΍ϭΗϟ΍ϰϠϋϝϭρϬϟ΍ϭϯϭλϘϟ΍ϝΎϣΣϼϟ %16
ΓΩΎϳίΑ %12ΔΑγϧΑ ϝϘϳΎϬϠϳϠΣΗϡΗϲΗϟ΍ΕΎΑΗόϠϟϰλϗϻ΍ϝϣΣϟ΍ϥ΃ΓΩΩΣϣϟ΍έλΎϧόϟ΍ΔϘϳέρΕΩϛ΃

ϥΎϓϙϟΫϛϭ .%50ΔΑγϧΑΔΣΗϔϟ΍ϝϭρϝϳϠϘΗΩϧϋ14ΔΑγϧΑΩ΍ΩίΗϭ50έ΍ΩϘϣΑΔΣΗϔϟ΍ϝϭρ
ΓΩΎϳί ϥΎϓ ϙϟΫ ϰϟ΍ ΔϓΎο΍ .%50 ΔΑγϧΑ ωΎϔΗέϻ΍ ΓΩΎϳί Ωϧϋ 14 ΔΑγϧΑ Ω΍Ωίϳ ϰλϗϻ΍ ϝϣΣϟ΍
ΔΣΎγϣϟ΍αϔϧΑϭΔόΑέϣϟ΍ϝΩΑΔϳέ΋΍ΩΔΣΗϓϡ΍ΩΧΗγ΍ΩϧϋΕΩΟϭϰλϗϻ΍ϝϣΣϟ΍ϥϣ41έ΍ΩϘϣΑ
ϥΎϓ ϭ 200,%100 Ώ (1/R) αϭϘΗϟ΍ ΔΑγϧ ΓΩΎϳί Ωϧϋϭ ΏΗόϟ΍ ϝϭρ αϔϧϟϭ ϪϧΎϓ Ύυϳ΍ϭ
.%172ϭ130,%103ΕϧΎϛϰλϗϻ΍ϝϣΣϟ΍ϲϓΓΩΎϳίϟ΍
‫اﻟﺧﻼﺻﺔ‬
‫ث ﺗﻣت دراﺳﺔ ﺗﺻ�رف وأداء اﻻﻋﺗ�ﺎب اﻟﺧرﺳ�ﺎﻧﯾﺔ اﻟﻣﺳ�ﻠﺣﺔ اﻟﻣﻘوﺳ�ﺔ اﻓﻘﯾ�ﺎ وﺑﻧوﻋﯾﮭ�ﺎ‬
‫ﻓﻲ ھذا اﻟﺑﺣ ِ‬
‫اﻟﻐﯾر ﺣﺎوﯾﮫ ﻋﻠﻰ ﻓﺗﺣﺎت واﻟﺣﺎوﯾﺔ ﻋﻠﻰ ﻓﺗﺣﺎت ﻏﯾر ﻣﻘواة او ﻣﻘواة و ﺑﺎﺳﺗﻌﻣﺎل )اﻟﻠداﺋن اﻟﻛرﺑوﻧﯾ�ﺔ‬
‫اﻟﻣﺳﻠﺣﺔ او ﻗﺿﺑﺎن ﺣدﯾد اﻟﺗﺳﻠﯾﺢ ( وذﻟك ﻣن ﺧﻼل ﺗﻘدﯾم دراﺳﺔ ﻣﺧﺗﺑرﯾﮫ وﺗﺣﻠﯾﻠﯾﺔ‪.‬‬
‫ﯾﺗﺿﻣن اﻟﻌﻣل اﻟﻣﺧﺗﺑري اﻋداد وﻓﺣص ارﺑﻌﺔ ﻋﺷر ﻋﺗﺑﺎ" ﺧرﺳﺎﻧﯾﺎ" ﻣﻘوﺳ�ﺎ" ﺑﺎﻻﺗﺟ�ﺎه اﻻﻓﻘ�ﻲ ﻣﻘﺳ�ﻣﺔ‬
‫اﻟﻰ ﻣﺟﻣوﻋﺗﯾن‪ .‬اﻟﻣﺟﻣوﻋﺔ اﻻوﻟﻰ ﺗﺗﺿﻣن ﻋﺷرة ﻧﻣﺎذج ﻋﻠﻰ ﺷﻛل ﻧﺻف داﺋرة )ﻧﻣوذج واﺣد ﺑ�دون‬
‫ﻓﺗﺣﺎت واﻟﺗﺳﻌﺔ اﻻﺧرى ﺗﺣﺗوي ﻋﻠﻰ ﻓﺗﺣﺎت(‪ ,‬ﺑﯾﻧﻣﺎ اﻟﻣﺟﻣوﻋﺔ اﻟﺛﺎﻧﯾﺔ ﺗﺗﻛون ﻣ�ن ارﺑﻌ�ﺔ ﻧﻣ�ﺎذج ﻋﻠ�ﻰ‬
‫ﺷ��ﻛل داﺋ��رة ﻛﺎﻣﻠ��ﺔ )ﻧﻣ��وذج واﺣ��د ﺑ��دون ﻓﺗﺣ��ﺎت واﻟﺛﻼﺛ��ﺔ ﺗﺣﺗ��وي ﻋﻠ��ﻰ ﻓﺗﺣ��ﺎت( ‪ .‬اﻟﻣﺗﻐﯾ��رات اﻟﺗ��ﻲ‬
‫ﯾﺗﺿﻣﻧﮭﺎ اﻟﺑرﻧﺎﻣﺞ اﻟﻌﻣﻠﻲ ھﻲ ﺗﺄﺛﯾر ﻧ�وع اﻟﻌﺗ�ب وﻛ�ذﻟك ﻣوﻗ�ﻊ اﻟﻔﺗﺣ�ﺔ ﺧ�ﻼل اﻟﻣﻘط�ﻊ اﻟط�وﻟﻲ ﻟﻠﻌﺗ�ب‬
‫اﻟﻣﻘوس وﺗﻘوﯾﺔ اﻻﻋﺗﺎب اﻟﻣﻘوﺳﺔ اﻟﺣﺎوﯾﺔ ﻋﻠ�ﻰ ﻓﺗﺣ�ﺎت ﺑﺎﺳ�ﺗﻌﻣﺎل )اﻟﻠ�داﺋن اﻟﻛرﺑوﻧﯾ�ﺔ اﻟﻣﺳ�ﻠﺣﺔ او‬
‫ﻗﺿﺑﺎن ﺣدﯾد اﻟﺗﺳﻠﯾﺢ (‪ .‬ﺗم ﻓﺣص اﻻﻋﺗ�ﺎب ﺗﺣ�ت ﺗ�ﺄﺛﯾر ﺣﻣ�ل ﻣرﻛ�ز )‪ (points load‬ﻣﺳ�ﻠط ﻋﻠ�ﻰ‬
‫اﻟطﺑﻘﺔ اﻟﻌﻠوﯾﺔ ﻟﻣﻧﺗﺻف اﻟﻔﺿﺎء وﺑﺈﺳﻧﺎد ﻣﻔﺻﻠﻲ وﻣﺗدﺣرج‪.‬‬
‫اظﮭرت اﻟﻧﺗﺎﺋﺞ اﻟﻣﺧﺗﺑرﯾﺔ ﻧﻘﺻﺎن ﻣﻠﺣوظ ﻓﻲ ﻗﺎﺑﻠﯾﺔ اﻟﺗﺣﻣل اﻟﻘﺻوى ﻟﻼﻋﺗﺎب ﺑوﺟود اﻟﻔﺗﺣﺔ ﻏﯾر‬
‫اﻟﻣﻘواة ﺑﺣدﯾد ﺗﺳﻠﯾﺢ او اﻟﯾﺎف ﻛرﺑوﻧﯾﺔ ﺑﺣواﻟﻲ ‪ %35‬ﻟﻠﻌﺗب اﻟﻣﻘوس اﻟﻧﺻف اﻟداﺋري و‪ %50‬ﻟﻠﻌﺗب‬
‫اﻟداﺋري اذا ﻣﺎ ﻗورﻧت ﻣﻊ اﻟﻧﻣﺎذج اﻟﻣرﺟﻌﯾﺔ اﻟﻐﯾر ﺣﺎوﯾﺔ ﻋﻠﻰ ﻓﺗﺣﺎت اﺿﺎﻓﺔ اﻟﻰ ﺗﻐﯾﯾر ﻓﻲ ﻧﻣط ﻓﺷل‬
‫ﺑﻘوى اﻟﻘص )‪ .( shear failure‬ان ﻗﺿﺑﺎن ﺣدﯾد اﻟﺗﺳﻠﯾﺢ او اﺳﺗﺧدام اﻟﻠداﺋن اﻟﻛرﺑوﻧﯾﺔ‬
‫اﻟﻣﺳﻠﺣﺔ ﯾﺣﺳن ﻣن ﻣﻘﺎوﻣﺔ اﻟﺗﺣﻣل اﻻﻗﺻﻰ ﻟﻠﻌﺗب ﺑﻧﺳﺑﺔ )‪ %3‬اﻟﻰ ‪ (%30‬و )‪%11‬‬
‫اﻟﻰ ‪ (%40‬ﻋﻠﻰ اﻟﺗواﻟﻲ ﻟﻸﻋﺗﺎب اﻟﻣﻘوﺳﺔ اﻟﻧﺻف داﺋرﯾﺔ ‪ ,‬ﺑﯾﻧﻣﺎ ﻟﻼﻋﺗﺎب اﻟداﺋرﯾﺔ‬
‫اﻟﻣﻘوﺳﺔ ﻛﻠﯾﺎ ﻓﺎن ﻧﺳﺑﺔ اﻟزﯾﺎدة ﺗﺻل اﻟﻰ ‪ %60‬و‪ %75‬ﻟﻠﺗﻘوﯾﺔ اﻟداﺧﻠﯾﺔ واﻟﺧﺎرﺟﯾﺔ‬
‫ﻋﻠﻰ اﻟﺗواﻟﻲ‪ .‬ﻛذﻟك ﻓﺎن اﺳﺗﺧدام ﻛﻼ اﻟﻧوﻋﯾن )ﺑﺎﺳﺗﺧدام ﺣدﯾد ﺗﺳﻠﯾﺢ او اﻻﻟﯾﺎف اﻟﻛرﺑوﻧﯾﺔ(‬
‫ﯾﻘﻠل ﻣن اﻟﺗﺷﻘﻘﺎت وﯾﺣﺳن ﻣن ﺻﻼدة اﻻﻧﺣﻧﺎء واﻻﻟﺗواء ﻟﻠﻌﺗب ﺑﻌد اﻟﺗﺷﻘق‪.‬‬
‫ﺑﺎﻟﻧﺳﺑﺔ ﻟﻠﺗﺣﻠﯾل اﻟﻧظري ﻓﺎﻧﮫ ﻗدم ﻧﻣوذﺟﺎ ﻻﺧطﯾﺎ ﺛﻼﺛﻲ اﻷﺑﻌﺎد ﻟﻠﻌﻧﺎﺻر اﻟﻣﺣددة ﻣﻼﺋﻣﺎ ً ﻟﺗﺣﻠﯾل‬
‫اﻻﻋﺗﺎب اﻟﺧرﺳﺎﻧﯾﺔ اﻟﻣﺳﻠﺣﺔ اﻟﻣﻘوﺳﺔ اﻓﻘﯾﺎ اﻟﺣﺎوﯾﺔ ﻋﻠﻰ ﻓﺗﺣﺎت وﻏﯾر اﻟﺣﺎوﯾﮫ ﻋﻠﻰ ﻓﺗﺣﺎت واﻟﻣﻘواة‬
‫واﻟﻐﯾر ﻣﻘواة ﺑﺎﺳﺗﻌﻣﺎل )اﻟﻠداﺋن اﻟﻛرﺑوﻧﯾﺔ اﻟﻣﺳﻠﺣﺔ او ﺣدﯾد اﻟﺗﺳﻠﯾﺢ ( ﺗﺣت ﺗﺄﺛﯾر أﺣﻣﺎل ﺗزاﯾدﯾﮫ‬
‫ﺑﺎﺳﺗﺧدام ﺑراﻣﺞ اﻟﺣﺎﺳوب )‪ (ANSYS 12.1‬وﻛذﻟك دراﺳﺔ ﺑﻌض اﻟﻣﺗﻐﯾرات اﻻﺿﺎﻓﯾﺔ اﻟﻣﮭﻣﺔ‪.‬‬
‫ﺠﻤﻬورﯿﺔ اﻟﻌراق‬
‫وزارة اﻟﺘﻌﻠﯿم اﻟﻌﺎﻟﻲ و اﻟﺒﺤث اﻟﻌﻠﻤﻲ‬
‫ﺠﺎﻤﻌﺔ ﺒﺎﺒل‬
‫ﻛﻠﯿﺔ اﻟﻬﻨدﺴﺔ‬
‫ﻗﺴم اﻟﻬﻨدﺴﺔ اﻟﻤدﻨﯿﺔ‬
‫اﻟﺘﺤﻠﯿﻞ اﻟﻼﺧﻄﻲ ﻟﺴﻠﻮك اﻷﻋﺘﺎب اﻟﺨﺮﺳﺎﻧﯿﺔ‬

‫اﻟﻤﺴﻠﺤﺔ اﻟﻤﻘﻮﺳﺔ اﻓﻘﯿﺎ واﻟﺤﺎوﯾﺔ ﻋﻠﻰ ﻓﺘﺤﺎت‬
‫ﻣﻘﻮاة ﺑﺄﻟﯿﺎف اﻟﻜﺎرﺑﻮن اﻟﺒﻮﻟﻤﺮﯾﺔ‬
‫ِ‬
‫درﺠﺔ اﻟدﻛﺘوراﻩ‬ ‫ِ‬
‫اﻟﺤﺼول ﻋﻠﻰ‬ ‫ﺠﺎﻤﻌﺔ ﺒﺎﺒل ﻛﺠزء ِﻤ ْن ﻤﺘطﻠﺒﺎت‬
‫ِ‬ ‫رﺴﺎﻟﺔ ﻤﻘدﻤﺔ ﻟﻛﻠﯿ ِ‬
‫ﱠﺔ اﻟﻬﻨدﺴﺔ ‪-‬‬
‫ِ‬
‫اﻟﻤدﻨﯿﺔ )اﻹﻨﺸﺎءات(‬ ‫ﻓﻲ اﻟﻔﻠﺴﻔﺔ ﻓﻲ اﻟﻬﻨدﺴﺔ‬
‫ﻣﻦ ﻗﺒﻞ‬
‫ﺳﺠﺎد ﻋﺎﻣﺮ ﺣﻤﺰة‬

‫ﺑﻜﺎﻟﻮرﯾﻮس ﻋﻠﻮم ﻓﻲ اﻟﮭﻨﺪﺳﺔ اﻟﻤﺪﻧﯿﺔ )‪ (2000‬م‬
‫ﻣﺎﺟﺴﺘﯿﺮ ھﻨﺪﺳﺔ ﻣﺪﻧﯿﺔ ‪ /‬اﻧﺸﺎءات )‪ (2003‬م‬
‫ﺃﺷﺮﺍﻑ‬
‫اﻷﺳﺘﺎذ اﻟﺪﻛﺘﻮر ‪:‬ﻋﻤﺎر ﯾﺎﺳﺮ ﻋﻠﻲ‬
‫‪2014‬‬

Behaviour of R.C. Horizontally Curved Beams With Openings Strengthened by CFRP Laminates-2014

Uploaded by

Document Information

Original Description:

Original Title

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

Behaviour of R.C. Horizontally Curved Beams With Openings Strengthened by CFRP Laminates-2014

Uploaded by

Copyright:

Available Formats

Republic of Iraq

Ministry of Higher Education and Scientific Research

Behavior of R.C. Horizontally Curved Beams

Sadjad Amir Hemzah

Prof. Dr. Ammar Yasir Ali

‫ﻭ‪‬ﻓَﻮﻕَ ﻛُـﻞﱢ ﺫ‪‬ﻱ ﻋ‪‬ﻠﻢٍ ﻋ‪‬ﻠ‪‬ﻴﻢ‪‬‬

‫ﺻدق ﷲ اﻟﻌﻠﻲ اﻟﻌظﯾم‬

‫ﺳﻮﺭﺓ ﻳﻮﺳﻒ ‪ /‬آﻳﺔ ‪76‬‬

I certify that the preparation of this dissertation titled

We certify that we, the examining committee, have read the

Approved by the Head of the Civil Engineering Department

Sadjad Amir Hemzah

1.2 General Guidelines to Selection of the Size and Location 5

Analysis for Ultimate Strength Under Combined Torsion 7

Failure Modes Under Combined Torsion, Bending, and 10

1.6 Design for Torsion 17

1.6.1 Design for Beam type Failure 18

1.6.2 Design for Frame Type Failure 21

1.7 Fiber Reinforced Polymer FRP 23

1.7.1 Fiber Materials 23

1.7.3 Properties of FRP 25

1.8 Aims of the Study. 27

1.9 Layout of Thesis 28

Experimental and Analytical Studies on Reinforced 29

Experiential and Analytical Works on Reinforced 39

2.4 Concluding Remarks 54

CHAPTER THREE:EXPERIMENTAL WORK

3.2 Description of Specimens 57

3.2.1 Semicircular Curved Beams (Group I) 60

3.2.1.1 Semicircular Curved Beams without Opening 60

3.2.1.2 Semicircular Curved Beams with Opening 60

3.2.2 Ring Curved Beams (Group I) 60

3.2.2.1 Ring Curved Beams without Opening 60

3.2.2.2 Ring Curved Beams with Opening 60

3.3 Strengthening System 62

3.3.1 Internal Strengthening by Internal Reinforcement 62

3.3.2 External Strengthening by CFRP Straps 62

3.4 Material Properties of Tested Specimens 65

3.4.1.2 Coarse Aggregate (Gravel) 66

3.4.1.3 Fine Aggregate (Sand) 66

3.4.1.4 Mixing Water 68

3.4.1.6 Mixing Procedure 69

3.4.2 Reinforcing Steel 69

3.4.3 Carbon Fiber (CFRP) Strengthening System 71

3.5 Mechanical Properties of Hardened Concrete 72

3.5.1 Compressive Strength (fc') 72

3.5.2 Splitting Tensile Strength (ft) 73

3.6 CFRP Installation 74

3.7 Instrument and Test Procedure 75

3.8 Loading and Support Condition 78

3.9 Steel and wood mold 79

CHAPTER FOUR:EXPERIMENTAL RESULTS AND

4.2 Semicircular Curved Concrete Beams (Group I) 82

4.2.1 Pilot and Control Semicircular Curved Beam (SCB.P) 82

Semicircular Curved Beam with Opening near External 85

4.2.2.1 Unstrengthened Semicircular Curved Beam (SCB.Eo) 85

Internally Strengthened Semicircular Curved Beam 88

Externally Strengthened Semicircular Curved Beam 90

4.2.2.4 Summary of tested SCB.E Beams 93

Semicircular Curved Beam with Opening near Midspan 95

Internally Strengthened Semicircular Curved Beam 98

Externally Strengthened Semicircular Curved Beam 101