Osu1269631164 Carga Vertical PDF

RESPONSE ESTIMATION OF REINFORCED CONCRETE COLUMNS SUBJECTED
TO LATERAL LOADS
A Thesis
Presented in Partial Fulfillment of the Requirements for
the Degree Master of Science in the
Graduate School of The Ohio State University
By
Muhammad Shadab Lodhi
Graduate Program in Civil Engineering
The Ohio State University
2010
Masters Examination Committee:
Dr. Halil Sezen, Adviser
Dr. Shive K. Chaturvedi
Dr. Ethan Kubatko

Copyright by
Muhammad Shadab Lodhi
2010
ABSTRACT
There is significantly large number of reinforced concrete buildings in the US and many
parts of the world that are not designed according to prevailing seismic code provisions.
These buildings are often characterized by low lateral displacement capacity and rapid
degradation of shear strength and hence are vulnerable to severe damage or even collapse
during strong ground motions. Typically, columns in such buildings have insufficient and
widely spaced transverse reinforcement and lack essential seismic reinforcement details.
These columns exhibit lack of strength and ductility in reverse cyclic loading and are
vulnerable to brittle shear failure and loss of axial load carrying capacity during strong
ground shaking. The need to assess their vulnerability to earthquake damage and hence
suggesting the desired level of retrofit requires evaluation of the expected behavior in
terms of strength and deformation capacity. This can be achieved by estimating the load-
deformation response considering all potential failure mechanisms associated with axial,
flexure and shear behavior. This study evaluates two different macro models to
investigate their capabilities to determine the lateral response. Based on the comparison
of the predicted responses with experimental test data, an analytical procedure is
proposed. In the proposed procedure, flexural and shear deformations are determined
while considering interaction between these mechanisms. Flexural deformations are
calculated by incorporating effect of shear deformation on flexural performance through
ii
compression softening factor. Likewise, shear deformations are calculated by employing
axial strain and shear stresses from axial-flexure model to consider the effect of flexural
deformations on shear behavior. The interaction between axial-flexure and axial-shear
mechanisms allows for accurate response estimation while decoupled flexural analysis
minimizes iterations within the analysis and make the process relatively simple and easy.
In addition, buckling of compression bars is also incorporated by considering separate
stress-strain relationships for reinforcing steel in tension and compression. The
displacement due to reinforcement slip are calculated separately and added to flexural
and shear displacements. Total lateral response is obtained by combining deformation
components due to flexure, shear and reinforcement slip according to a set of rules based
on comparison of the column yield, flexural and shear strength. The comparison of the
predicted responses with experimental test data showed that proposed procedure performs
well in estimating response of reinforced concrete columns to lateral loads.
iii
Dedication
This document is dedicated to my parents, M.N.U.K Lodhi and Hafeez Akhtar
and my wife, Tabassum Shadab
and my children, Muhammad Ibrahim Khan Lodhi and Aliza Lodhi
iv
ACKNOWLEDGMENTS
The completion of this thesis would not have been possible without the help of
many people. First and foremost, I would like to thank my adviser, Dr. Halil Sezen, for
his continuous support and endless patience throughout the years so far. Without his
guidance in our meetings and discussions, I would not have been able to complete my
research. I would like to acknowledge Professors Shive Chaturvedi and Dr. Ethan
Kubatko for giving their time as members of my committee.
I would like to thank my friends, colleagues and family members for their
unconditional support over the years. I also appreciate the support of my friend and
officemate Kyong Yun Yeau who has made the rough times manageable and the good
times memorable.
I am also deeply obliged to Ministry of Defence, Government of Pakistan for
providing me this wonderful opportunity of graduate studies and supporting myself and
my family financially.
v
VITA
June 18, 1971 .. Born Pakistan
1993 B.S. Civil Engineering,

Military College of Engineering Risalpur,
Pakistan
1993 - 2006 .... Commissioned Military Service

Pakistan Army Corps of Engineers
2006 to present. MS/PhD Graduate Student

The Ohio State University, Columbus, Ohio
FIELDS OF STUDY
Major Field: Civil Engineering
Specialization: Structural Engineering
vi
TABLE OF CONTENTS
Page
Abstract .. ii
Dedication... iv
Acknowledgments . v
Vita vi
List of Tables . x
List of Figures xi
Chapters:
1 INTRODUCTION . 1
1.1 Overview 1
1.2 Research Impetus ... 3
1.3 Research Objective 5
1.4 Organization ... 6
2 DISPLACEMENT COMPONENT MODEL...... 11

2.1 Introduction 11
2.2 Deformation Components... 12
2.2.1 Flexural Deformations.. 12
2.2.2 Reinforcement Slip Deformations. 15
2.2.3 Shear Deformations... 18
2.3 Total Response. 20
2.4 Shear Failure Due to High Displacement Ductility.. 24
2.5 Implementation of Displacement Component Model 25
2.5.1 Test Specimens and Material Properties.. 25
2.5.2 Material Constitutive Relationships.. 26
2.5.3 Comparison of Model and Test Results. 28
2.6 Conclusion 29
3 AXIAL SHEAR FLEXURE INTERACTION (ASFI) APPROACH -

THEORY . 39
3.1 Introduction 39
3.2 Background. 41
3.3 Axial-Flexure Model... 42
3.3.1 Effect of Shear Deformations on Flexural Response.. 43
vii
Page
3.3.2 Axial Strain due to Flexure. 43
3.3.3 Flexural Deformations.... 45
3.4 Axial-Shear Model. 46
3.4.1 MCFT Constitutive Laws... 49
3.4.1.1 Concrete in Compression....... 50
3.4.1.2 Concrete in Tension.... 50
3.4.1.3 Reinforcement Stress-Strain Relationship... 51
3.4.2 Consideration of Local Cracks Conditions of Integration.. 53
3.4.3 Material Stiffness Formulations. 56
3.5 Development of Axial Shear Flexure Interaction Approach.. 58
3.5.1 Compatibility Conditions/Relationships. .. 59
3.5.2 Equilibrium Conditions/Relationships 60
3.5.3 Constitutive Laws... 61
3.5.4 Interactions Considered.............. 62
3.5.4.1 Interaction Strain Methodology.. 62
3.5.4.2 Concrete Compression Softening... 64
3.5.5 Stiffness Model in ASFI approach... 64
3.6 Conclusions. 65
4 IMPLEMENTATION OF ASFI APPROACH.. 71

4.1 Introduction.... 71
4.2 Response Estimation of Columns by ASFI approach. 72
4.2.1 Pullout Effect Consideration and Total Deformations... 73
4.2.2 Analytical Procedure for Implementation of ASFI
approach... 75
4.2.3 Solution Technique for Implementing analytical
Procedure 77
4.3 Implementation and Verification of ASFI approach... 80
4.3.1 Details of Test Specimens and Material Properties. 81
4.3.2 Constitutive Laws... 81
4.3.2.1 Constitutive Law for Concrete in Compression... 82
4.3.2.2 Constitutive Law for Concrete in Tension. 83
4.3.2.3 Constitutive Law for Reinforcements. 84
4.3.3 Flexural Deformation Model.. 85
4.3.4 Pullout Deformation Model 85
4.3.5 Buckling of Compression Bars and Post-Peak Analysis 86
4.4 Comparison of the Results...... 87
4.5 Summary and Conclusions. 87
5 COMPARISON OF COMPONENT DISPLACEMENT MODEL AND

ASFI APPROACH..... 99
5.1 Introduction. 99
5.2 Implementation of the Analytical Models.. 100
5.2.1 Test Specimens and Material Properties. 100
viii
Page
5.2.2 Material Constitutive Relationships 101
5.2.2.1 Concrete in Compression 101
5.5.2.2 Concrete in Tension 103
5.2.2.3 Reinforcing Steel ... 103
5.2.3 Flexural Deformation Model...... 104
5.2.4 Pullout Deformation Model.... 104
5.2.5 Buckling of Compression Bars... 105
5.3 Comparison of Predicted and Experimental Response... 105
5.3.1 Flexural Displacements... 106
5.3.2 Reinforcement Slip Displacements 108
5.3.3 Shear Displacements... 108
5.3.4 Total Response 112
5.3 Conclusions. 116
6 DEVELOMPENT AND IMPLEMENTATION OF THE PROPOSED

MODEL.. 130
6.1 Introduction. 130
6.2 Desired Characteristic of Proposed Analytical Procedure.. 131
6.3 Proposed Interaction Methodology. 135
6.3.1 Interaction of Concrete Compression Softening. 135
6.3.2 Interaction of Axial Strains. 139
6.4 Buckling of Compression Bars... 141
6.4.1 Dhakal and Maekawa (2002) Bar Buckling Model... 143
6.4.2 Potger et al. (2001) Bar Model... 144
6.4.3 Implementation of Bar Buckling Models... ... 146
6.4.4 Proposed Model for Buckling of Compression Bars...... 146
6.5 Flexural Analysis 147
6.6 Reinforcement Slip Deformation Model 148
6.7 Shear Deformations.... 148
6.8 Total/Combined Response . 151
6.8.1 Pre-Peak response... ... 152
6.8.2 Post-Peak Response. .. 152
6.9 Analytical Procedure and Solution Technique for Proposed
Model.. 153
6.10 Implementation of the Model and Comparison of Results. 160
6.11 Conclusion.. 161
7 CONCLUSION . 174
7.1 Summary 174
7.1.1 Component Models 175
7.1.2 Combined Model ... 177
7.2 Conclusions 177
7.3 Recommendations for Future Work .. 179
ix
Page
List of References . 181
x
LIST OF TABLES
Table Page
4.1 Geometric and Material Properties of the Test Specimens. 98
xi
LIST OF FIGURES
Figure Page
1.1 Olive View Hospital damaged in the 1971 San Fernando Earthquake
(Steinbrugge, K. V., NISEE).. 8
1.2 Failure of first story columns in the Olive View Hospital (Steinbrugge,
K. V., NISEE). 8
1.3 Imperial County Services Building damaged during the 1979 Imperial
Valley earthquake (Bertero, V. V., NISEE)... 9
1.4 Collapsed reinforced concrete buildings during 2005 Pakistan

earthquake (Nawed, K., eNKay Solutions). 9
1.5 Examples of commonly observed deficiencies and damage in reinforced

concrete construction after 2005 Pakistan earthquake (EEC, Department
of Civil Engineering NWFP UET Peshawar). 10
2.1 Components of lateral deformation in a reinforced concrete column 30
2.2 Plastic Hinge Model for Determining Flexural Deformation 30
2.3 Bar Slip Model ... 31
2.4 Reinforcement slip rotation ... 31
2.5 Shear displacement model . 32
2.6 Spring representation for total response model for fixed ended column 32
2.7 Classification of the column and Lateral load-displacement relationships 33
2.8 Rules governing the combination of deformation components.............. 33
xii
Figure Page
2.9 Constitutive relationships for concrete in compression.. 34
2.10 Constitutive relationship for reinforcing steel 34
2.11 Lateral load flexural deformation relationships. 35
2.12 Lateral load slip deformation relationships.... 36
2.13 Lateral load shear deformation relationships.. 37
2.14 Lateral load total deformation relationships.... 38
3.1 Concrete compression softening... 38
3.2 Axial-Flexural model and determination of axial strain due to flexure 67
3.3 Plastic hinge model for flexural deformations ... 67
3.4 (a) Membrane element subjected to in-plane stresses (b) Average strains
in-plane loading (c) Average strains in cracked concrete (d) Mohrs
circle for average strains. 68
3.5 Average stress-strain relationship for (a) Cracked concrete compression

(b) Cracked concrete in tension (c) Reinforcing Steel..... 68
3.6 (a) Calculated average stresses (b) Local stresses at a crack.. 69
3.7 Parameters influencing crack spacing .... 70
3.8 Coordinate transformation for concrete and steel material stiffness.......... 70
3.9 Spring model for axial-shear-flexural element 70
4.1 Flow chart for implementation of ASFI approach for columns 89
4.2 Constitutive relationships for concrete in compression.. 90
4.3 Constitutive relationship for concrete in tension ... 90
4.4 Constitutive relationship for reinforcing steel ... 91
4.5 Rigid body rotation of column due to reinforcement slip or pullout.... 91
4.6 Pullout deformation model (Okamura and Maekawa, 1991)................. 92
xiii
Figure Page
4.7 Comparison of the results for Specimen-5 (Ousalem et al, 2005).. 93
4.8 Comparison of the results for Specimen-12 (Ousalem et al, 2005)............ 94
4.9 Comparison of the results for Specimen-14 (Ousalem et al, 2005) 95
4.10 Comparison of the results for Specimen-15 (Ousalem et al, 2005). 96
4.11 Comparison of the results for Specimen-16 (Ousalem et al, 2005) 97
5.1 Constitutive relationships concrete in compression 117
5.2 Constitutive relationship concrete in tension.. 117
5.3 Constitutive relationship for reinforcing steel 118
5.4 Plastic Hinge Model for Determining Flexural Deformation. 118
5.5 Lateral load flexural displacement relationship for Specimen-1. 119
5.9 Effect of concrete tensile stress in flexural section analysis... 121
5.10 Effect of concrete compression softening in flexural section analysis... 121
5.11 Effect of compression bars buckling in flexural section analysis... 122
5.12 Lateral load slip displacement relationship for Specimen-1 122
5.16 Lateral load shear displacement relationship for Specimen-1. 124
xiv
Figure Page
5.20 Lateral load displacement relationship for Specimen-1.. 126
5.23 Lateral load displacement relationship for Specimen-4. 128
5.24 Effect of plastic hinge length on total response by ASFI approach

(Specimen-1). 129
6.1 Determination of concrete principal tensile strain. 162
6.2 Dhakal and Maekawa (2002) bar buckling model. 162
6.3 Potger et al. (2001) bar buckling model. 163
6.4 Effect of compression bar buckling Specimen 1. 163
6.7 Proposed bar buckling model. 165
xv
Figure Page
6.23 Lateral load displacement relationship for Specimen-4. 173
6.24 Comparison between compressive concrete constitutive laws... 173
xvi
CHAPTER 1
INTRODUCTION
1.1 Overview
There is a large inventory of reinforced concrete buildings in the US and other
parts of the world that are not designed according to modern seismic design provisions.
In the developed countries, these are the buildings which were constructed between
1930s and 1970s according to codes and standards available at the time. Even today, the
many developing countries that lag behind on seismic code developments continue to
design and build structures which do not have essential details deemed vital to withstand
large lateral loads. These buildings are often characterized by low lateral displacement
capacity and rapid degradation of shear strength and hence are vulnerable to severe
damage or even collapse during strong ground motions.
The reconnaissance of damage observed during the past earthquakes suggests that
poorly designed reinforced concrete columns are the most critical elements to sustain
damage leading to a potential building collapse. Typically, these columns have
insufficient and widely spaced transverse reinforcement with poor details such as 90-
degree end hooks and splicing of the longitudinal bars near the column ends, which are
the regions that typically experience the most inelastic deformations. Due to these
1
deficiencies, the columns may not have sufficient shear strength to develop the plastic
hinges at the ends. In addition, wide spacing of the ties does not provide good
confinement for the core concrete. As a result, the columns exhibit non-ductile behavior
and may fail suddenly.
The examples of the damage caused by earthquakes in non-seismic reinforced
concrete construction are shown in Figures 1.1 through 1.5. Figure 1.1 shows Olive View
Hospital building which was damaged in the 1971 San Fernando Earthquake. The
building sustained significant damage in almost all columns in the first story. Close up
views of the two columns in Figure 1.2 show severe shear damage to the columns. The
concrete is entirely crushed and the columns have virtually lost their ability to carry
lateral and axial loads. The Imperial County Services Building, as shown in Figure 1.3,
also suffered significant damage to its first-story columns during the Imperial Valley
earthquake of 1979. Transverse reinforcement did not provide sufficient confinement to
core concrete, resulting in crushing of the concrete and buckling of the longitudinal
reinforcement.
Notice that both buildings in above examples suffered heavy structural damage
but did not collapse. In developing countries, in addition to the absence of requisite
seismic detailing, the quality of construction is also very poor. Resultantly, catastrophic
failure of the structures is observed after every major earthquake. For example, Pakistan
earthquake of 2005 caused widespread damage to reinforced concrete buildings (Figure
1.4). A reconnaissance team of structural engineers that inspected the damage after the
earthquake found that many of the buildings that collapsed or sustained severe damage
suffered from lack of seismic details, poor quality of construction, or both (Amjad et al.,
2
2005). The non-seismic details in collapsed or damaged buildings included insufficient
lap-lengths, improper location of the lap splicing and insufficient concrete confinement
due to large spacing of ties. The team also observed irregular configurations with
soft/weak story or captive columns, constructions without any engineering design and
poor quality of concrete and workmanship. Figures 1.5 shows several examples of these
deficiencies and damaged reinforced concrete buildings from the Pakistan 2005
earthquake.
Seismic design follows strong column-weak beam philosophy where the
columns have sufficient strength and deformation capacity to push inelastic action and
damage to the beams. Damage of the beams affects mainly the immediate surroundings
whereas the failure of a lower-story column can potentially lead to collapse of the
building. Current design codes require buildings to have details that ensure ductile
behavior to withstand high seismic activity. These details include closely spaced ties in
columns, with 135-degree end hooks. The large amount of transverse steel gives the
columns sufficient shear strength to resist lateral earthquake loads, and the closely spaced
ties keep the core concrete confined at high displacements. Similarly, 135-degree end
hooks perform much better and remain closed during cyclic loading as these are
embedded in the core concrete which remains intact due to good confinement.
1.2 Research Impetus
The large numbers of reinforced concrete buildings that are not designed to
withstand earthquake are vulnerable to damage or collapse in a major seismic event. It is
possible to retrofit an existing building to increase its strength and ductility to enhance its
3
performance during an earthquake. The need to assess their vulnerability to earthquake
damage and hence suggesting the desired level of retrofit requires evaluation of the
expected behavior in terms of strength and deformation capacity. This can be achieved by
estimating load-deformation response of the concrete elements, such as beams and
columns, considering all potential failure mechanisms associated with axial, flexure and
shear behavior.
In addition to the retrofitting requirements, there are many situations where the
structures are required to be analyzed accurately to predict their structural responses.
Even important existing buildings and planned future structures may require performance
evaluation in terms of assessing their maximum load capacity, ultimate deformation
capacity, ductility and failure mechanism.
While much is known about the behavior of reinforced concrete components and
systems, there are still areas that require further understanding. For a structure whose
behavior is dominated by flexural mechanism, issues regarding its performance
evaluations have been studied well and design procedures are relatively well established.
However, for the structures whose behavior is affected by shear related mechanisms,
accurate modeling remains elusive with many available approaches and theories.
Understanding the fact that lateral deformation of a structure is mainly comprised of the
flexure and shear components, it must also be realized that these deformations do not
occur independently and these mechanisms interact with each other. Any analytical
procedure that aims to model overall lateral load-displacement relationship must take this
aspect into account. These are few issues that are the motivation for the research reported
here.
4
1.3 Research Objective
As explained earlier that columns tend to be the most critical structural
component in reinforced concrete structures not designed according to modern seismic
design codes. There are a number of studies investigating structural response of non-
ductile reinforced concrete columns following varying approaches. Some of these
approaches can predict structural behavior with good accuracy but employ very
complicated and computation-intensive analytical procedures which may not be
amenable and are difficult to implement. Contrary, many approaches simplify the
analytical process by making simplifying assumption on the structural response but in
most cases this is done at the cost of accuracy. Therefore, it is aimed in this study to
propose a suitable analytical procedure which can address critical issues in structural
performance. The objective is to predict response of the reinforced concrete column
subjected to lateral loading accurately while keeping overall computational process
simple and easy to implement.
In the research reported here, two of the available models will be examined and
analyzed for their applicability in terms of the accuracy of predicted responses and ease
of applications. Both models will be implemented by writing Matlab (MathWorks, 2009)
computer programs and estimated responses for previously tested reinforced concrete
columns will be compared with test data. Based on the observations, an analytical
procedure will be proposed that can accurately predict monotonic lateral load-
displacement response of the reinforced concrete column subjected to lateral loading. The
predicted responses by the proposed procedure will be compared with experimental test
data to evaluate its performance.
5
One of the models analyzed in this study, designated as displacement component
model, was developed by Setzler and Sezen (2005). The macro model was developed
based on the concept that a typical fixed-ended reinforced concrete column, when
subjected to earthquake loading, undergoes lateral deformation which is comprised of the
three components; flexural deformations, reinforcement slip deformations and shear
deformations as illustrated in figure 2.1. In this model, each of three individual
deformation components is estimated separately and then simply added together to get
total pre-peak response. For post-peak analysis, the understudy column is classified into
one of the five categories based on comparison of its predicted shear and flexural strength
and then individual deformation components are combined together according to a set of
rules specified for each category.
The other model evaluated in this study is Axial-flexure-shear interaction (ASFI)
approach by Mostafaei and Kabeyasawa (2007). This model considers interaction
between axial, flexure and shear mechanisms in terms of concrete compression softening
and axial deformation while satisfying conditions on compatibility of average strains and
equilibrium of average stresses. For shear behavior, this model employs Modified
Compression Field theory (MCFT) which is considered very suitable approach for
determining shear deformations of the reinforced concrete elements subjected to shear
and normal stresses.
1.4 Organization
This thesis is organized into seven chapters. The organization is based on the
investigation process implemented in this study. Chapter 2 presents the summary of the
6
displacement component model. The analytical procedure is implemented and predicted
responses for test specimens used in the original study are compared with test data to
verify correct implementation and working of the computer program. Chapter 3
summarizes the theory of axial-flexure-shear interaction approach. As MCFT employed
in this model is a complex approach, analytical procedure and relevant details are
explained in detail. This chapter can serve as a good summary documentation on theory
and implementation of MCFT. In Chapter 4, the complex analytical procedures of ASFI
approach are implemented and predicted responses for the test specimens used in the
original study are presented and compared with reported response to verify correct
implementation. Both understudy approaches are analyzed by comparing their estimated
response for a common set of test specimens to evaluate their capabilities in terms of
predicting peak strength, post peak response and ultimate deformations in Chapter 5. The
observations made in this chapter form the basis of the proposed analytical procedure. In
Chapter 6, the proposed procedure for response estimation is presented. This chapter
explains the procedure to determine flexural and shear deformations considering
interaction between these mechanisms. Two of the available bar buckling models are
analyzed and recommended procedure is presented to incorporate bar buckling
phenomenon in the proposed procedure. Predicted responses by the proposed procedure
are compared with experimental data to verify the model.
A summary of the research completed is included in Chapter 7. This chapter also
discusses conclusions drawn from this study, and suggests several areas of future
research.
7
Figure 1.1: Olive View Hospital damaged in the 1971 San Fernando Earthquake
(Steinbrugge, K. V., NISEE)
Figure 1.2: Failure of first story columns in the Olive View Hospital
(Steinbrugge, K. V., NISEE)
8
Figure 1.3: Imperial County Services Building damaged during the 1979 Imperial Valley
earthquake (Bertero, V. V., NISEE)
(a) Margallah Towers Islamabad
(b) A low rise commercial building in Muzafarabad
Figure 1.4: Collapsed reinforced concrete buildings during 2005 Pakistan earthquake
(Nawed, K., eNKay Solutions)
9
(a) Insufficient and improper lap splicing (b) Widely spaced transverse reinforcement
(c) Poor quality of the concrete (d) Weak column strong beam construction
(e) Shear failure of the column (f) Plastic hinge formation near column end
Figure 1.5: Examples of commonly observed deficiencies and damage in reinforced

concrete construction after 2005 Pakistan earthquake
(EEC, Department of Civil Engineering NWFP UET Peshawar)
10
CHAPTER 2
DISPLACEMENT COMPONENT MODEL
2.1 Introduction
In order to evaluate structural performance of reinforced concrete columns
subjected to lateral loads, Setzler and Sezen (2008) developed a macro model based on
the concept that a typical fixed-ended reinforced concrete column, when subjected to
earthquake loading, undergoes lateral deformation which is comprised of three
components; flexural deformations, reinforcement slip deformations and shear
deformations as illustrated in figure 2.1. In this model, each of three individual
deformation components are estimated separately and then simply added together to get
total pre-peak response. For post-peak analysis, the understudy column is classified into
one of the five categories based on comparison of its predicted shear and flexural strength
and then individual deformation components are combined together according to a set of
rules specified for each category.
In this chapter, the analytical procedure of the above mentioned displacement
component model is briefly summarized. First, deformation component models for
estimating each component of lateral deformation are presented followed by the details of
total response model that outlines the rules for total response. The approach is
11
implemented in a computer program written in MATLAB (MathWorks, 2009) and the
results for response estimation are presented to verify correct implementation of the
model. The simulated lateral load-deformation responses of the test specimens shall be
used for discussion and comparison in subsequent chapters.
2.2 Deformation Components
As mentioned above, total lateral deformation of a reinforced concrete column
can be determined after estimating lateral deformations resulting from flexural, bond slip
and shear mechanisms. In the procedure discussed in this chapter, these deformations are
calculated separately as per the details given below and then added together depending
upon dominant failure mode determined after comparing flexural and shear strength of
the columns.
2.2.1 Flexural deformations
For reinforced concrete elements subjected to bending moment and axial load,
such as beams or columns, flexural deformations can accurately be determined by
performing section analysis on a fiber model in one-dimensional stress field. This is very
handy and reliable approach if realistic material constitutive relationships and actual
stress distribution across the depth of the cross-section are considered. In this approach,
the reinforced concrete cross-section is discritized into finite number of concrete and steel
fibers. Each of the fibers is idealized as a uniaxial element with its unique stress-strain
relationship. Bernoullis principle that plane section before bending remains plane after
bending is the main hypothesis in the analysis and implies that the longitudinal strain in
12
concrete and steel at any point in the cross section is proportional to its distance from
neutral axis resulting in linear strain distribution. Based upon the resulting strain profile,
stress distribution for concrete and reinforcing steel can be determined in accordance with
their respective stress-strain relationships. By satisfying equilibrium equations at the
cross section, the moment capacity of the section is determined. The process is repeated
number of times by incrementing longitudinal stain until either the concrete or steel fails
as per defined failure criterion. The analysis results in moment-curvature relationship,
which is used to determine flexural deformations.
Typically, fiber section analysis takes into account the enhancement in the
strength and ductility of the concrete due to confinement and ignores concrete behavior in
tension. Also, confined core concrete and unconfined cover concrete are configured
separately with their respective stress-strain relationships. It must also be noted that the
accuracy of the analysis is directly related to the ability of the constitutive material
models to simulate real material behaviors in the cross section and the level of
simplifying assumptions during the analysis.
Knowing the moment-curvature relationship for the reinforced concrete section,
lateral force-flexural deformation relationship can be determined either by force-based or
deformation-based approach. Force-based approach integrates curvature over height of
the column to determine lateral deformation due to flexure as
L
f ( x) xdx (2.1)
0
where x is the curvature distribution at distance x measured along column axis, and
L is the height of the column. This approach results into an accurate response as it
13
captures the non-linearity in actual curvature distribution up to the point corresponding to
the maximum moment capacity. However, after the peak has reached, the deformations
will continue to increase and this approach can no longer predict post-peak deflections, as
no direct relation exists between moment and curvature in this range. In the plastic hinge
method for computing flexural deformations, as shown in figure 2.2, a linear curvature
distribution is assumed in the elastic range and inelastic curvatures are lumped at the
column end over the plastic hinge length L p . Flexural displacements are calculated by
taking the moment of the areas of the curvature diagram about the column end. Unlike
force-based approach, plastic hinge model can predict post-peak flexural deformations
but overestimates the deflections in elastic range as it approximates the linear moment-
curvature relationship in this range.
In view of the discrepancies of both of these methods, a hybrid approach is
adopted for determining flexural deformations employing deflection calculations by
curvature integration up to yield point and plastic hinge method after yielding. The yield
point for the section is defined when the longitudinal steel first yields. Following
equation is used to calculate lateral deformations of the column due to flexure.
Lp
f f , y y L p a for y u (2.2)
2
where f , y is the flexural deflection at yielding calculated using the integration method,
is curvature at column end, y is curvature at yield point, a is shear span equal to L
for a cantilever column and L 2 for fixed ended column. The plastic hinge length L p is
14
taken 0.5h per the recommendations of Moehle (1992), whereas h is depth of cross-
section.
From the moments calculated as per the above-mentioned procedure, lateral load
V corresponding to any moment capacity M can be calculated by dividing moment by
shear span a . The maximum lateral force sustainable by the column is
Mp
Vp (2.3)
a
where, M p is maximum moment capacity from flexural section analysis.
2.2.2 Reinforcement Slip deformations
When a reinforcing bar embedded in the concrete is subjected to a tensile force,
strain accumulates in the embedded length of the bar which causes the bar to extend or
slip relative to the concrete in which it is embedded. The same phenomenon is observed
when a reinforced concrete column is subjected to bending moment. The concrete on
tension face of the cross-section, being weak in tension, cracks at the early stage of
loading and becomes ineffective in anchoring the column to the base. Resultantly, the
reinforcing bars carry tensile loads from column to the anchoring concrete. The bond
stresses at concrete-steel interface in the anchoring concrete cause a tensile stress gradient
in the bars over their embedded length. The steel stresses vary from zero at the dead end
of the embedded bar to a maximum value at the face of beam-column or beam-footing
joint. The length of the bar over which these stresses are distributed, and eventually
transferred to concrete by the bond stresses, is called the development length. The
accumulation of the strain over the development length causes the extension of the bars
15
relative to the anchoring concrete. This extension is commonly known as reinforcement
slip and it leads to rigid-body rotation of the column, as shown in figure 2.1. This results
in lateral displacement that can be as large as 25 to 40 % of the total lateral deformations
(Sezen, 2002).
The flexural deformations as determined through conventional fiber section
analysis (moment-curvature analysis) do not account for lateral deformations caused by
reinforcement slip. Reinforcement slip deformations add another component of lateral
deformation to overall response. Therefore, reinforcement slip deformations must be
calculated separately and added to the other deformation components, such as flexure and
shear to accurately model the total drift.
In this study, lateral displacements due to reinforcement slip are calculated
through a model which was proposed Sezen and Moehle (2003) and further developed by
Sezen and Setzler (2008). The model, as shown in figure 2.3, approximates the bond
stress as bi-uniform function with different values for elastic and inelastic steel behavior,
which allows for the efficient computation of the reinforcement slip and eliminates the
need for the nested iteration loops that are required in some of the existing bond stress-
slip models. The value for the bond stress in the elastic range is taken as ub 1 f c
(MPa) based on a study by Sezen (2002) on 12 test columns. For the inelastic range, the
value for bond stress is adopted from the study by Lehman and Moehle (2000) as
ub 0.5 f c (MPa) where f c is concrete compressive strength. Slip at the loaded end of
the reinforcing bar can be calculated by integrating bi-linear strain distribution over the
development length as follows:
16
ld ld
slip ( x)dx
0
(2.4)
where ld and ld' are development lengths for the elastic and inelastic portion of the bar,
respectively. As bond stresses are uniform in each range, strain distribution is bi-linear as
shown in figure 2.3. Integration of equation 2.3 yields following relations for extension or
slip of the reinforcing bars as
s ld S y
slip
2
for (2.5)
y ld ( s y )l d
slip S y
2 2
where s is the strain at loaded end of the bar, and y is steel yield strain. The
development lengths for elastic and inelastic regions can be calculated as
f s db
ld
4ub
(2.6)
( f s f y )db
ld
4ub
where f s is stress at loaded end of the bar, f y is steel yield stress, and d b is diameter of
the longitudinal bar. The reinforcement slip is assumed to occur in tension bars only and
cause the rotation about the neutral axis. Hence, rotation caused due to reinforcement slip
can be calculated from the following equation figure 2.4.
slip
s (2.7)
d c
where d and c are the distances from the extreme compression fiber to the centroid of
the tension steel and the neutral axis, respectively. The lateral displacement due to slip at
17
the free end of a cantilever column can be calculated as the product of slip rotation s
and length of the column L as
s s L (2.8)
2.2.3 Shear Deformations
Shear deformations in reinforced concrete members are have traditionally been
ignored in design and research due to their lack of complete understanding and being
difficult to measure in an experimental set up or a real structure. For a well-designed
reinforced concrete column, shear deformations are small as compared to the flexural
deformations and are often less than 10 percent of total deformations. Contrary, for a
reinforced concrete column not designed according to stricter seismic design provisions,
shear behavior could be the governing failure criterion. Shear deformations in such shear
critical reinforced concrete column could contribute large percentage towards total
deformations and hence cannot be ignored if an accurate analysis of deformation capacity
is desired.
The basis of the shear model used in this analytical approach is the model
developed by Patwardhan (2005), which uses Modified Compression Field Theory
commonly known as MCFT (Vecchio and Collins, 1986). Patwardhan proposed a
piecewise linear model defining key points in the lateral force-shear deformation
envelope through a parametric study implementing MCFT through a computer program
Response-2000 (Bentz, 2000a). In shear model employed in this approach, as shown in
figure 2.5, pre-peak non-linear shear force-shear deformation response is obtained
18
indirectly from Response-2000 by integrating shear strain distribution over the height of
the column for each load step as following:-
L
v ( x)dx (2.9)
0
where x is the average shear strain over the cross-section at each location x along
the height L of the column, and v is the shear displacement. After the peak strength has
reached, the shear strength is assumed to remain constant at its peak value until the onset
of shear strength degradation. By modifying the equation proposed by Gerin and Adebar
(2004), the shear displacement at the onset of shear degradation v ,u can be calculated as
v
v ,u 4 12 n v ,n (2.10)
f c
where, vn is the shear stress at peak strength V peak bd , f c is the concrete
compressive strength, and v ,n is the shear displacement at maximum strength
determined from Response-2000. The peak strength Vpeak is the minimum of the shear
strength of the column Vn and shear force corresponding to the maximum moment
sustainable by the section Vp . After the shear degradation is initiated, shear strength
decreases linearly with increasing shear deformations to the point of axial load failure,
where lateral strength is assumed zero. The shear displacement at axial load failure v, f
is calculated as
v, f ALF f , f s, f v,u (2.11)
19
where ALF is the total displacement at axial load failure, f , f and s , f are the flexural
and slip displacement at the point of axial load failure, respectively. The total
displacement at axial load failure is determined by the expression proposed by Elwood
and Moehle (2005a), which is based on a shear friction model and an idealized shear
failure plane as:-
ALF 4 1 tan 2
(2.12)
L 100 s
tan P
A f d tan
sv yv c
where is the angle of the shear crack, P is the axial load, Asv is the area of transverse
steel with yield strength f yv at spacing s and d c is the depth of the core concrete,
measured to the centerlines of the transverse reinforcement. In the derivation, is
assumed to be 65 degrees. The values of f , f and s , f in equation 2.11 are determined
according to the expected failure mode and classification of the column into categories as
explained in subsequent subsection.
2.3 Total Response
In order to model the response of a structure subjected to lateral loading, all of the
three deformation components must be combined to predict total lateral deformation with
due regard to their interconnectedness. The procedure for total response models each of
flexure, slip and shear deformation by a spring subjected to the same force and the total
response is the sum of the responses of each spring as shown in figure 2.6. In this model,
each of the deformation components is simply added to obtain the total response up to the
peak strength of the column, defined as lesser of the strength from flexural and shear
20
model. However, for post-peak behavior, the column is classified into one of the five
categories based on a comparison of its shear, yield and flexural strength and rules are
specified for the combination of the deformation components for each category. Yield
strength Vy is defined as the lateral load corresponding to the first yielding of the tension
bars in the column whereas flexural strength Vp is the lateral load corresponding to the
peak moment sustainable by the column during flexural analysis. The shear strength of
the column Vn is calculated by the expression developed by Sezen and Moehle (2004) for
lightly reinforced concrete columns as
6 f
Vn k (Vc Vs ) k 0.80 A Asv f yv d
P
c
1 (2.13)
a 6 f c Ag s
g
where Vc is the concrete contribution to shear strength, Vs is the steel contribution to
shear strength, Ag is gross cross-sectional area and k is a factor related to the
displacement ductility which is the ratio of the maximum displacement to the yield
displacement.
Classification of the column based on comparison of shear, yield and flexural
strength determines expected column behavior. The peak response will be limited by the
lesser of the shear strength from shear model and the flexural strength ( Vp ), however post
peak response is assumed to be governed by the limiting mechanism (i.e., flexure or
shear). The classification system and rules governing the post peak response in each
category are described below and are illustrated in figure 2.7 and 2.8, respectively.
21
Category I. Vn V y
The shear strength is less than the yield strength and column fails in shear
while the flexural behavior remains elastic. The deformation at peak strength (i.e.,
shear strength) is the sum of deformations in each spring at the peak strength.
After the peak strength is reached, the shear behavior dominates the response. As
the shear strength degrades, the flexure and slip springs unload along their initial
responses. The post-peak deformation at any lateral load level is the sum of the
post-peak shear deformation and the pre-peak flexural and slip deformations
corresponding to that load.
Category II. V y Vn 0.95V p
The shear strength is less than flexural strength and column fails in shear,
however inelastic flexural deformation occurring prior to shear failure affects the
post-peak behavior. The deformation at peak strength is the sum of the
deformations in each spring at the peak strength. Shear deformations continue to
increase after the peak shear strength is reached, but the flexure and shear springs
are locked at their peak strength values. The post-peak deformation is the sum of
flexural and shear deformations at peak strength and post-peak shear deformation.
Category III. 0.95V p Vn 1.05V p
The shear and flexural strengths are nearly identical. It is not possible to
predict conclusively which mechanism will govern the peak response. Shear and
flexural failure are assumed to occur simultaneously, and both mechanisms
contribute to the post-peak behavior. The post-peak deformation at any lateral
22
load level is the sum of the post-peak flexure, slip, and shear deformations
corresponding to that load.
Category IV. 1.05V p Vn 1.4V p
The shear strength is greater than the flexural strength and the column may
potentially fail in the flexure, however inelastic shear deformations affect the
post-peak behavior and shear failure may occur as the displacements increase.
The deformation at peak strength is the sum of the deformations in each spring at
the peak strength. After the peak strength is reached, flexural and slip
deformations continue to increase according to their models, but the shear spring
is locked at its value at peak strength. The post-peak deformation at any lateral
load level is the sum of the post peak flexural and slip deformations
corresponding to that load and the shear deformation at peak strength.
Category V. Vn 1.4V p
The shear strength is much greater than the flexural strength and column
fails in flexure while shear behavior remains elastic. The peak strength of the
column is the flexural strength calculated from the flexure model. If the column
strength degrades, flexural and slip deformations continue to increase according
to their models, while the shear spring unloads with an unloading stiffness equal
to its initial stiffness. The post-peak deformation at any lateral load level is the
sum of the post-peak flexural and slip deformations and the pre-peak shear
deformation corresponding to that load.
23
For category-I columns, f , f and s , f values to be used in equation 2.11 are
assumed zero. For the category-II columns, shear strength is lesser than flexural strength
and these values are taken as the flexural and slip deformations at the load equal to the
shear strength of the columns. For categories III, IV, and V specimens, f , f and s , f are
the maximum calculated flexural and slip deformations.
2.4 Shear Failure Due to High Displacement Ductility
Shear strength degrades as the displacement ductility increases (Sezen and
Moehle, 2004), which can cause shear failure in the columns that are initially dominated
by flexure. Such behavior, as encountered in category-IV columns where shear failure
can occur at higher displacements, can not be captured by simply combining three
deformation component models as described above, thus necessitating an alternate
approach to predict delayed shear failure. Elwood (2004) proposed that shear failure,
after occurrence of flexural yielding, will occur if the lateral response intersects the shear
failure surface imposed on the lateral load-total displacement behavior of the column.
The shear failure surface is defined by the empirical drift capacity model proposed by
Elwood and Moehle (2005b) as
SF 3 1 v 1 P 1
4 v (2.14)
L 100 500 f co 40 Ag f co 100
where SF is the drift at shear failure, v is the transverse reinforcement ratio, and v is
the nominal shear stress. f co and v have units of psi. The displacement at shear failure is
calculated according to above equation using the peak strength (Vp) in the model to
24
calculate the shear stress. If the total lateral response envelope exceeds the calculated
drift at shear failure, shear failure is assumed to have occurred. The model is modified to
degrade linearly from the point of shear failure to strength of zero at the displacement at
axial load failure.
2.5 Implementation of the Displacement Component Model
In order to implement above mentioned displacement component model, a
computer program is written in MATLAB and lateral load-deformation responses for
flexural, bar slip, shear and total response are calculated for four test columns tested by
Sezen (2002). These are the same test specimens, which were used by Setzler and Sezen
(2008) to develop their model and verify its response predictions. Details of the test
specimens, material constitutive relationships used for response estimation and response
predictions are presented in following sub-sections.
2.5.1 Test Specimens and Material Properties
Sezen (2002) tested four fixed ended reinforced concrete columns
subjected to unidirectional monotonic and cyclic lateral loading. The test specimens,
designated specimens-1 through -4, were designed with the details common in the
buildings that lack current seismic code requirements and represented lightly reinforced
columns that have shear and flexural strengths very close to each other. These tests
rendered very useful data in terms of experimental force-displacement responses for each
25
of the flexure, slip, and shear components individually, as well as for the overall
response.
These are 18 inch square columns with fixed ends at top and bottom having height
of 116 inches. The longitudinal reinforcement consists of eight No. 9 bars with clear
cover of 2 inches. No. 3 column ties with 90-degree end hooks were spaced at 12 in. over
the height of the column. Specimens-1 and -4 were subjected to a constant 150 kip axial
load, Specimen-2 was tested under a constant 600 kip axial load, and Specimen-3 had an
axial load varying from 600 kip in compression to 60 kip in tension to simulate the range
of forces experienced by an exterior building column in an earthquake. The columns were
tested under unidirectional cyclic lateral loading, except for Specimen-4, which was
tested under monotonic loading. Average concrete strength was 3.077 psi with maximum
aggregate size of 1 inch. Yield strength of longitudinal and lateral steel was 59 and 69
ksi, respectively.
2.5.2 Material Constitutive Relationships
Same material constitutive laws for concrete and reinforcing steel are employed in
the analysis which were used by Setzler and Sezen (2008) for their proposed model. In
flexural section analysis, separate stress-strain relationships are employed for confined
core and unconfined cover concrete. For core concrete, constitutive laws in compression
are defined considering the effects of confinement on core concrete as per the
confinement model by Mander et al. (1988). However, in order to represent the expected
post-peak concrete behavior in shear critical columns, the descending branch is modeled
by the relation developed by Roy and Sozen (1964).

26
The pre-peak confined concrete compressive stress-strain relationship is defined
with the help of following equations.
f cc r c cc
fc
r 1 c cc
r
f cc
cc co 1 5 1 for c cc (2.15)
f co
Ec
r
Ec Esec
where f cc is the peak confined concrete strength calculated according to Mander et al.
(1988), c is the concrete strain, cc is the concrete strain at peak stress for confined
concrete, co is the concrete strain at peak stress in unconfined concrete (taken here as
0.002), f co is the concrete compressive cylinder strength, Ec is the modulus of elasticity
of the concrete and is equal to 57,000 fco (in psi units) for normal weight concrete, and
Esec fcc cc is the secant modulus of the concrete.
The post-peak confined concrete, compressive response is calculated from the
slope of the unloading branch and peak strength point ( cc , f cc ). The slope of the
descending leg is calculated from Roy and Sozen 50% strength point having strain as,
3 0.002 f co
50u (2.16)
f co 1000
For unconfined cover concrete, same set of above mentioned equations is used
except that f cc and cc ,wherever they appear, are replaced with f co and co . Stress-
strain relationships for confined and unconfined concrete are illustrated in figure 2.9.
27
The reinforcing steel behavior is modeled considering a linear elastic behavior, a
yield plateau, and a non-linear strain-hardening region, as per following set of equations.
f s Es s s y
f s f y s sh Es for y s sh (2.17)
p

f s f u ( f u f sh ) s sh sh s u
u sh
where Es is the elastic modulus of steel, s is the steel strain, and the subscripts y, sh, and
u refer to the yield point, the onset of strain hardening, and the ultimate stress,
respectively. p is the order of the curve that defines the strain-hardening region, and is
often taken as 2 for a parabolic curve. is a coefficient that defines the slope of the yield
plateau. For the columns tested by Sezen, these parameters were used to define the
longitudinal steel stress-strain model: f y 59 ksi, sh 0.016 , 0.02 , f u 93.5 ksi,
u 0.23 , p 6 , E s 29,000 ksi. Steel stress-strain relationship used for this model is
illustrated in figure 2.10.
2.5.3 Comparison of Model and Test Results
Component and total responses for specimens-1 through -4, obtained after
implementing Setzler and Sezen (2008) approach, are presented in figure 2.11 to 2.14.
Predicted responses for flexural and slip deformations compare well with the
experimental data. Predicted shear responses follow the initial stiffness of the
experimental responses closely but overestimates post-peak deformations as the shear
strength degrades. For total responses also, generally good correlation with the
experimental data is obtained. It can be concluded that the model overall does a good job
28
in predicting lateral behavior of the columns. These results will be used subsequently for
discussion and comparison of the models applicability and suggesting improvements to
the existing procedure.
2.6 Conclusion
Displacement component model by Setzler and Sezen (2008) has briefly been
summarized in this chapter. This approach considers total deformation of a reinforced
concrete column to lateral loading being composed of flexural, bar slip and shear
deformations which can be combined together depending upon dominant failure
mechanism. Structural response of four test specimen is calculated as per the prescribed
model after implementing analytical routine in MATLAB computer program.
Comparison of predicted response and experimental data shows that the understudy
model has correctly been implemented and model performs well in evaluating lateral
load-lateral deformation responses of reinforced concrete columns.
29
Flexure deformation Slip deformation Shear deformation
Figure 2.1: Components of lateral deformation in a reinforced concrete column
y
x x
Lp L - Lp L
y y
Figure 2.2: Plastic Hinge Model for Determining Flexural Deformation
30
F As f s
Reinforcing bar
Beam-column
interface
b' 0.5 fco fs s ld'
fy y
ld
b 1 fco
b fs s
Figure 2.3: Bar Slip Model
Figure 2.4: Reinforcement slip rotation
31
Maximum
Maximum Beginning
Initiation of of shear
shear
strength point
degradation
( v ,n,V
, V peak )
strength point
( degradation
,V )
v,n peak
v,u peak
v,u peak
, V
load
Lateralload
lateral
Response-2000
Response 2000
Axialload
Axial loadfailure
failure
,0
(
,0) v,f
v, f
shear displacement
Shear displacement
Figure 2.5: Shear displacement model
V
Zero-length slip
rotation spring
Zero-length
shear spring
Flexural spring
L
s
Zero-length slip
rotation spring
V
s f v
Figure 2.6: Spring representation for total response model for fixed ended column
32
Cat. I
V Vy (yield strength)
Vp (flexural strength)
Cat. II
Equation 2.13
Eq. 7 (k=1.0)
Vn
Cat. III Cat. IV
Equation
Eq. 7 2.14
0.7*Vn
Cat. V
0
0 2 6
displacement ductility ()
Figure 2.7: Classification of the column and Lateral load displacement relationship
Category
Category V V Category III, IV, V
Category
Category
Category I,I,II,II,
III III
Category
Category I I III, IV, V
Lateral force
lateral force (V)
lateral force (V)
Category
Category IV IV Category II
Category II
shear displacement ( ) flexural or slip displacement ( f, s )

Shear deformations
v Flexural or slip deformations
Figure 2.8: Rules governing the combination of deformation components
33
3500
Confined concrete
3000 Unconfined concrete
2500
Stress (psi)
2000
1500
1000
500
0
0 1 2 3 4 5 6 7 8
Strain x 10
-3
Figure 2.9: Constitutive relationships for concrete in compression
100
90
80
70
60
Stress (ksi)
50
40
30
20
10
0
0 0.025 0.05 0.075 0.1 0.125 0.15
Strain
Figure 2.10: Constitutive relationship for reinforcing steel

34
80
Specimen - 1
60
Lateral force (kip)

40
20
-20
-40
-60 Test data

Model
-80
-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5
100
Specimen - 2
Lateral force (kip)
50
-50
Test data
Model
-100
-0.9 -0.7 -0.5 -0.3 -0.1 0.1 0.3 0.5 0.7 0.9 1.1
1.12
80
Specimen - 3
60
Lateral force (kip)
40
20
-20
-40 Test data

Model
-60
-3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2
80
Specimen - 4
60
Lateral force (kip)
40
20
-20
-40
-60 Test data

Model
-80
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4
Flexural Displacement (in)
Figure 2.11: Lateral load flexural deformation relationships

35
80
Specimen - 1
60
Lateral force (kip)

40
20
-20
-40
-60 Test data

Model
-80
-1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1 1.25 1.5
100
Specimen - 2
Lateral force (kip)
50
-50
Test data
Model
-100
-0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8
80
Specimen - 3
60
Lateral force (kip)
40
20
-20
-40 Test data

Model
-60
-2.5 -2.25 -2 -1.75 -1.5 -1.25 -1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1
80
Specimen - 4
60
Lateral force (kip)
40
20
-20
-40
-60 Test data

Model
-80
-1 0 1 2 3 4
Slip Displacement (in)
Figure 2.12: Lateral load slip deformation relationships

36
80
Specimen - 1
60
Lateral force (kip)

40
20
-20
-40
-60 Test data

Model
-80
-5 -4 -3 -2 -1 0 1 2 3 4
100
Specimen - 2
Lateral force (kip)
50
-50
Test data
Model
-100
-1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1
80
Specimen - 3
60
Lateral force (kip)
40
20
-20
-40 Test data

Model
-60
-2 -1.5 -1 -0.5 0 0.5 1 1.5
80
Specimen - 4
60
Lateral force (kip)
40
20
-20
-40
-60 Test data

Model
-80
-4 -3 -2 -1 0 1 2 3 4
Shear displacement (in)
Figure 2.13: Lateral load shear deformation relationships

37
80
Specimen - 1
60
Lateral force (kip)

40
20
-20
-40
-60 Test data

Model
-80
-6 -4 -2 0 2 4 6
100
Specimen - 2
Lateral force (kip)
50
-50
Test data
Model
-100
-3 -2 -1 0 1 2 3
80
Specimen - 3
60
Lateral force (kip)
40
20
-20
-40 Test data

Model
-60
-8 -6 -4 -2 0 2 4
80
Specimen - 4
60
Lateral force (kip)
40
20
-20
-40
-60 Test data

Model
-80
-6 -4 -2 0 2 4 6 8
Displacement (in.)
Figure 2.14: Lateral load total deformation relationships

38
CHAPTER 3
AXIAL SHEAR FLEXURAL INTERACTION (ASFI) APPROACH - THEORY
3.1 Introduction
Displacement-based evaluation approach for estimating structural performance of
reinforced concrete elements requires monitoring of their load-deformation responses in
all failure mechanisms associated with axial, flexural and shear loadings. Traditionally,
the analytical process for response estimation is simplified by splitting the total structural
behavior into individual mechanisms of flexure and shear. The analyses for both
mechanisms are carried out individually and dominant failure mechanism is determined
which controls overall response and design considerations.
Flexural deformations are most commonly estimated by performing flexural
section analysis. This is very convenient and accurate analytical tool for flexural
response, but performance of reinforced concrete elements dominated in shear or shear-
flexure cannot be estimated, as shear behavior is not considered in the approach.
Likewise, most of the available models and theories for shear response are good for
evaluating shear behavior only and can not estimate flexural performance directly.
It must be recognized that flexural and shear deformations in a reinforced
concrete element subjected to combined axial, flexure and shear loading, do not occur
39
independently. Any model that aims to capture total response must consider the
interaction between flexure and shear mechanisms. When the analyses for these
mechanisms are carried out independently, this aspect can not be taken care of due to the
inherent inability of the analytical procedure to perform good for the mechanism it has
not been developed for.
Mostafaei and Kabeyasawa (2007) have recently developed a displacement-based
analysis approach for response estimation of reinforced concrete elements by considering
interaction between axial, shear and flexural mechanisms. The approach is designated as
Axial-Shear-Flexural Interaction (ASFI) Approach and consists of two models that
evaluate flexural and shear response simultaneously. In this approach, axial-flexural
behavior is simulated by employing conventional section analysis or fiber model in one-
dimensional stress field whereas axial-shear response is determined by considering one
integration point in in-plane stress conditions employing modified compression field
theory (MCFT). MCFT is one of the many available shear response approaches such as
truss models and empirical models, and is considered most suitable for determining shear
deformations of the reinforced concrete elements subjected to shear and normal stresses.
In ASFI approach, axial-flexural and axial-shear mechanisms are coupled in stress-strain
field considering axial deformation interaction, softening of concrete compression
strength and satisfying compatibility and equilibrium conditions.
In this chapter, theory of ASFI approach is presented by explaining details of the
component models and development of the approach in generalized form for a reinforced
concrete element between two subsequent flexural sections.
40
3.2 Background
As already mentioned above, structural performance of reinforced concrete
column can accurately be estimated if flexural and shear deformations are evaluated
simultaneously with due regards to their interconnectedness. It must also be recognized
that flexural section analysis and MCFT are powerful assessment tools for the mechanism
originally developed for, and deal only with individual component of flexural and shear
deformations respectively. For MCFT, this statement is true in general, but efforts have
been made to determine flexural-shear response by MCFT. To do so, a reinforced
concrete element must be discritized into large number of biaxially loaded elements and
analyzed by conducting a non-linear finite element analysis. The concept of MCFT has
also been extended to sectional analysis approach elements loaded in combined axial,
shear and flexure loads by employing biaxially stressed elements as concrete fibers
instead of uniaxially stressed elements employed in conventional sectional analysis
approach (Vecchio and Collins 1988). By fixing the longitudinal strain in each concrete
fiber by corresponding section strain, concrete elements are analyzed individually for in-
plane stress field based on MCFT. The most recent implementation of this approach can
be seen in the computer program called Response-2000 (Bentz 2000). The application of
MCFT in finite element approach or sectional analysis approach in the way mentioned
above yields very reliable response but at the same times results in fastidious
computations and is not simple enough for practice.
Some studies have also been carried out to combine individual components of
shear and flexure by modeling them as springs in series. Shirai et al (2001) presented a
macro-element approach by decomposing total deformation of reinforced concrete
41
column into flexural and shear components and simulated these component deformations
by layered element model and in-plane shear element respectively. Setzler and Sezens
displacement component model (chapter 2) is also based on the same concept that total
deformation of a reinforced concrete column is composed of three individual components
of flexure, reinforcement slip and shear. Their study models these component
deformations as spring in series subjected to the same force and total response is
determined by combing the component responses according to a set of rules developed
based on the comparison of the columns yield, flexural and shear strength. Although,
these models are simple for practical purposes and estimate response reasonably, they do
not consider interaction of axial deformation and effect of degrading compression
strength of concrete fibers due to increasing shear deformations.
3.3 Axial-Flexure Model.
Axial-flexure model in ASFI approach evaluates flexural response of reinforced
concrete element by employing standard flexural section analysis techniques. The section
analysis is performed in usual way except that constitutive relationships of diagonally
cracked concrete are employed to consider the effect of shear deformations on flexural
performance of the element. This is done by softening the response of concrete in
uniaxial compression by a factor which decreases as shear deformations increase. This
factor, known as compression softening factor, is function of principal tensile strain in
concrete and is obtained from in-plane shear analysis of the flexural element, thus
coupling axial-flexure model with axial-shear mechanism. Due to interaction of
compression softening factor, axial-flexural analysis in ASFI approach can not be
42
performed independently. This aspect is described in detail in subsequent sections. The
other deviation from usual practice of flexural section analysis is consideration of
concrete tensile behavior in the analysis, which is generally ignored in standard
procedure.
Sometimes, in order to simplify the calculations, section analysis with one stress
block is performed for simplified analysis and design. This is done by replacing actual
concrete stress distribution across the section by an equivalent rectangular stress block,
commonly known as whitney stress block. This allows the analysis to be performed with
much ease and simplicity and eliminates need of any iterative calculations required for
more prcised fiber section analysis approach explained in previous chapter.
Section analysis with equivalent stress block or fiber model in one-dimensional
stress field can be employed as axial-flexure model in ASFI approach, provided that
concrete tensile behavior and concrete compression softening is considered in the
analysis. Choice of the particular method depends on the required level of accuracy and
available amount of computational effort. In the analysis presented here, fiber section
analysis approach is adopted as axial-flexural model in ASFI approach. Details of fiber
section analysis have already been presented in chapter 2.
3.3.1 Effect of Shear Deformations on Flexural Response
Conventionally, the flexural section analysis is performed independently
simulating concrete behavior in uniaxial compression. The compressive stress-strain
relationships for the concrete in flexural analysis are usually derived from its response in
standard cylinder test. It must be recognized that the strain conditions for the concrete in
43
the web of a reinforced concrete beam or column subjected to shear differ significantly
from those in a cylinder test. The concrete in standard cylinder test is subjected to only
small tensile strains primarily due to Poissons effect, whereas, the concrete in diagonally
cracked web is subjected to very substantial tensile strains primarily due to shear. As a
result, the concrete in diagonally cracked web is weaker and softer than the concrete in a
cylinder. To account for theses influences, axial-flexural model in ASFI approach
employs constitutive model for cracked concrete in compression. The degradation in
compressive strength and stiffness of cracked concrete, denoted as concrete compression
softening, has been studied by Vecchio and Collins (1986) in their efforts to investigate
stress-strain characteristics of diagonally cracked concrete. Their study on 30 reinforced
concrete panels subjected to in-plane stress conditions indicates that the principal
compressive stress in the concrete is not only a function of principal compressive strain
but also of the co-existing principal tensile strain, such that compressive strength and
stiffness of the concrete decrease as the tensile strains increase. The concrete
compression softening, as shown in Figure 3.1, is one of the interaction terms from axial-
shear mechanism that is incorporated into axial-flexural model coupling both
mechanisms. In ASFI approach, concrete compression softening is employed in flexural
analysis by softening concrete response in uniaxial compression in the manner that,
fcsof fc (3.1)
where, fcsof is concrete compressive response after softening, f c is concrete
compressive stress given by any of the appropriate stress-strain relationships simulating
concrete behavior in uniaxial compression, and is compression-softening factor
defined as
44
1
1.0 (3.2)

0.8 0.34 c1
co
where, c1 is principal tensile strain and co is concrete strain at maximum concrete
cylinder strength.
3.3.2 Axial Strain Due to Flexure
Axial strain due to flexure is another interaction term in ASFI approach that
couples axial-flexure mechanism to axial shear mechanism. Complete details on axial
strain interaction methodology shall be presented subsequently; however, as axial strain
due to flexure can only be determined from flexural section analysis, it is appropriate
here to mention the way it can be calculated. Average axial strain due to flexure xf
between two flexural sections can be determined as illustrated in Figure 3.2 based on
relative centroidal deformation between two sections assuming linear strain distribution.
1
l12
x
xf
l12
0
o2 o1
l12
dx o 2 o1
2
(3.3)
where o1 and o 2 are centroidal strains of two consecutive flexural sections and l12 is
distance between the sections.
3.3.3 Flexural Deformations.
Flexure deformations in axial-flexure model are determined with the help of
plastic hinge model. In this method, elastic and inelastic curvatures are modeled
separately, and added together. A linear curvature distribution is assumed in the elastic
45
range, and the inelastic curvature is lumped at the column end over the plastic hinge
length. The conceptual illustration of the plastic hinge model for determining the flexural
deformation of a cantilever reinforced concrete column is presented in Figure 3.3. The
flexural drift ratio f can hence be calculated as following:
Lin
1
f
Lin x ( x)dx
0
(3.4)
where, Lin is the distance from column end section to inflection point, L p is plastic hinge
length taken equal to the section depth, ( x ) is curvature function at distance x
measured from inflection point and y is yield curvature.
3.4 Axial-Shear Model
Axial-shear model simulates shear mechanism in ASFI approach employing
modified compression theory. MCFT is a suitable displacement-based evaluation
approach for response estimation of reinforced concrete membrane elements subjected to
normal and shear stresses. Vecchio and Collins (1986) developed the theory after
modifying previously proposed Compression Field Theory (Mithell and Collins 1974)
which includes compression-softening effects, contribution of tensile stresses in cracked
concrete and local stress conditions at crack. MCFT is essentially a smeared rotating
crack model in which cracked concrete is treated as a new orthotropic material with its
unique stress-strain characteristics. The theory consists of compatibility, equilibrium and
stress-strain relationships formulated in terms of average stresses and strains. The critical
aspect of MCFT is the consideration of local stress-strain conditions at cracks ensuring
46
that the tension in the concrete can be transmitted across the crack and shear stress on the
surface of the crack does not exceed maximum shear provided by the aggregate interlock.
Thus, load deformation response of the members loaded in shear can be estimated by
considering compatibility of average strains for concrete and reinforcement, equilibrium
relationships involving average stresses in concrete and reinforcement and appropriate
stress-strain relationships for reinforcement and diagonally cracked concrete. Complete
details of MCFT and its formulation can be found in number literature where it is
employed. In this section, only the aspects relevant to ASFI approach shall be presented.
Following assumptions form the basis of the theory and are adopted as such for
ASFI approach as well.
1. Only one stress state exists corresponding to each strain state
2. Stresses and strains can be considered in terms of average values when taken over
areas or distances large enough to include several cracks.
3. The concrete and reinforcing bars are perfectly bonded at the boundaries of the
element, which implies that there is no overall slip.
4. Orientations of principal strains and principal stresses coincide with each other.
5. The longitudinal and transverse reinforcing bars are uniformly distributed over the
element.
6. The cracks are uniformly distributed and rotating.
7. The normal and shear stresses are applied uniformly.
8. The constitutive relationships for concrete and reinforcement are independent.
9. Shear stresses in the reinforcement are negligible and hence are not considered in
equilibrium relationships.
47
Consider an orthogonally reinforced concrete membrane element as shown in
Figure 3.4a. The element consists of smeared reinforcement in longitudinal (x) and
transverse (y) directions, with the corresponding reinforcement ratios x and y . The
yield strengths of longitudinal and transverse reinforcement are f yx and f yy , respectively.
The concrete is characterized by a cylinder compressive strength f c' , a strain at peak co
and a tensile cracking stress f cr . The elements edge planes are subjected to uniform
normal stresses f x , f y and shear stress xy . The deformation of the element is assumed
to occur such that the edges remain straight and parallel.
Under the applied loads, an equilibrium condition is attained resulting in unique
strain condition defined by two normal strains x and y and the shear strain xy as
shown in Figure 3.4b. From the Mohrs circle of the average strains (Figure 3.4c and
3.4d), average concrete principal tensile strain c1 , average concrete principal
compressive strain c 2 and orientation of principal strain filed can be determined as:-
x y 1
y xy
2 2
c1 x (3.5)
2 2
x y 1
y xy
2 2
c2 x (3.6)
2 2
1 xy
p tan 1 (37)

2 x y
48
The inclination of the principal strain p as given by Equation 3.8 can either be
orientation of principal tensile strain t or orientation of principal compressive strain
(crack angle) c depending upon the magnitude of normal strains x and y . Hence,
p t for x y
(3.8)
p c for x y
The principal tensile plan and principal compressive plane are 90 o apart, i.e.,
t c 90o or c t 90o . In addition, positive orientation of principal planes is
considered to make counterclockwise angle with positive x-axis. Hence, following these
rules, the relationship between t and c can be summarized as,
If p t 0 c t 90o
If p t 0 c t 90o
(3.9)
If p c 0 t c 90o
If p c 0 t c 90o
3.4.1 MCFT Constitutive Laws.
In addition to relevant compatibility and equilibrium conditions, constitutive
relationships are required to link average stresses to average strains for concrete and
reinforcement in MCFT. Derivation of these laws, especially for the concrete, is one of
the important aspects in development of the theory and simulates cracked concrete
behavior in compression and tension.
49
3.4.1.1 Concrete in Compression.
The constitutive laws for concrete in compression in MCFT have been derived by
Vecchio and Collins (1986) after investigating experimental stress-strain behavior of
cracked concrete. These laws consider reduction in compressive strength and stiffness of
concrete due to shear after softening the uniaxial response by compression softening
factor. The relationship for compressive concrete behavior in MCFT was suggested as,
and is shown in Figure 3.5a:-
2
f c 2 f co 2 c 2 c 2 (3.10)
co co
where is compression softening factor defined as per Equation 3.2, c 2 is concrete
compressive strain and f c 2 is corresponding concrete stress. The term multiplied with
in Equation 3.10 is the Hognested Parabolic relationship for concrete in uniaxial
compression, often used for normal strength concrete. Although, this equation is
specifically used in MCFT, any uniaxial concrete compressive stress-strain relationship
can theoretically be used as part of axial-shear model in ASFI approach; however, it must
incorporate compression-softening effect as defined by the Equation 3.2.
3.4.1.2 Concrete in Tension.
The constitutive relationship for the concrete in tension was also developed by
Vecchio and Collins (1986) based on their reinforced concrete panel tests. The concrete
tensile stress-strain relationship, as illustrated in Figure 3.5b, relates principal concrete
tensile stress f c1 to principal concrete tensile strain c1 .
50
The relationship suggested prior to cracking i.e., 0 c1 cr is linearly elastic
and is given by the expression:-
f c1 Ec c1 (3.10)
where, Ec is modulus of elasticity of the concrete, cr is cracking strain corresponding to
uniaxial cracking strength of concrete f cr , and f co uniaxial compressive strength of
concrete in MPa. These quantities are given by following expressions
Ec 2 fc' co ; cr fcr Ec ; f cr 0.33 f c' (3.11)
After cracking, the concrete tensile stresses continue to exist due to bond
interaction between concrete and reinforcement and decrease as the principal concrete
tensile strains increase. This phenomenon is known as tension stiffening. The relationship
suggested after cracking, i.e., c1 > cr is
f cr
f c1 (3.12)
1 200 c1
For large reinforced concrete elements, this relationship is modified slightly to the
following expression
f cr
f c1 (3.13)
1 500 c1
3.4.1.3 Reinforcement Stress-Strain Relationship
MCFT adopts bi-linear stress-strain relationship for the longitudinal and
transverse reinforcement in both tension and compression that consists of initial
ascending linear-elastic branch followed by a yield plateau. In the development of the
51
theory, it is assumed that the axial stress in the reinforcement depends only on one strain
parameter, the axial strain in the reinforcement. The shear stress resisted by the
reinforcement on the plane normal to the reinforcement is zero. The relationship for
average axial stress f s and average strain s used in MCFT is shown in Figure 3.5c and
is described by the following equations
f sx Es sx f sxyield
(3.14)
f sy Es sy f syyield
The subscripts x and y represent x -direction (longitudinal) and y -direction
(transverse) reinforcement, respectively. E s is modulus of the elasticity of the steel, and
f sxyield and f syyield are yield stress of the reinforcement in x and y directions,
respectively.
Compatibility of the average strains in MCFT requires that strain in reinforcement
and concrete must match with the in plane strains in x and y -directions i.e., sx x
and sy y . Hence, the Equation 3.14 takes the form as:-
f sx Es x f sxyield
(3.15)
f sy Es y f syyield
Although MCFT uses conventional bi-linear stress-strain relationship for the
reinforcement, other reinforcement relations allowing strain hardening can be used as part
of axial-shear model in ASFI.
52
3.4.2 Considerations of Local Cracks Conditions
Consideration of the local stress conditions at the cracks is another very important
aspect in development of the modified compression field theory. These considerations
ensure that the stresses can be transmitted across the cracks. The formulations
considering average stresses and strains do not capture local variations that may occur at
the cracks. For example, tensile stresses in the reinforcement will be higher than the
average at the cracks and lower than the average midway between the cracks. On the
other hand, the concrete tensile stresses will be zero at the crack and higher than the
average midway between the cracks. These local variations are important as the response
of bi-axially loaded element may be governed by the reinforcements ability to transmit
tension across the cracks or sliding shear failure along the cracks. To address these
possibilities, MCFT limits the local stresses at the cracks and the average concrete tensile
stress between the cracks.
Figure 3.6ashows average stresses at a section between the cracks perpendicular
to the principal tensile stress direction and Figure 3.6b shows local stresses at the free
surface of the crack. At the free surface of the crack, the average concrete tensile stresses
reduce to zero. This causes reinforcement stresses to increase locally at the crack in order
to transmit tensile stresses across the crack. Hence, average concrete tensile stresses must
be limited to avoid failure of the reinforcement at the crack. Static equivalency of the
average and local stresses in the direction normal to the crack surface results in the
condition that limits the concrete average tensile stress to the following upper limit to
insure its transmission across the crack.
f c1 x f scrx f sx cos 2 nx y f scry f sy cos 2 ny (3.16)

53
where, x and y are reinforcement ratio in x and y -directions respectively, f scrx and f scry
are local reinforcement stresses at crack, and nx and ny are the angles between the
normal to the crack and reinforcement in x and y directions, respectively. The values
for nx and ny are defined positive counterclockwise, as shown in Figure 3.6b, and may
be determined as
nx t
(3.17)
ny c
In MCFT, yielding of the reinforcement is the upper limit for local reinforcement
stresses at the crack and hence, the average concrete tensile stresses are limited to-
f c1 x f sxyield f sx cos 2 nx y f syyield f sy cos 2 ny (3.18)
The terms in the parenthesis in above equation represent the reserve capacity of
the reinforcements to transmit concrete tensile stresses before they fail. Hence, Equation
3.18 may be modified by replacing local reinforcement stresses with specified failure
criteria if it is other than yielding of the reinforcement.
The other consideration of local stresses at the crack deals with the shear
stresses, which are present locally at the crack surface due to reinforcement crossing the
cracks at skew angles (Figure 3.6b). This consideration limits the local shear stresses at
the crack surface by the shear resistance provided by the aggregate interlock mechanism.
Static equivalency of the average and local stresses in the direction tangential to the crack
determines local shear stresses as follows
vci x f scrx f sx cos nx sin nx y f scry f sy cos ny sin ny (3.19)
54
This local shear stress value at the crack must not exceed the shear strength
provided by the interface between the cement paste and the aggregate particles due to
aggregate interlock mechanism.
vci 0.18vci max (3.20)
f co
vci max (3.21)
0.31 24w /(a 16)
where w is crack width in millimeters, a is maximum aggregate size in millimeters and
f co is concrete compressive strength in MPA. The crack width w in Equation 3.21
should be the average crack width over the crack surface and can be taken as product of
the principal tensile strain c1 and the average spacing of diagonal cracks in the direction
normal to the crack S cr .
w c1Scr (3.22)
where,
1
Scr (3.23)
sin cos

S mx S my
S mx and S my in the above equation are the crack spacing that indicate crack control
characteristics of x and y reinforcement respectively. These quantities depend on bond
properties and layout of the reinforcement and can be estimated from CEB-FIP Code as
follows,
Sx d
Smx 2(cx ) 0.25k1 bx
10 x
(3.24)
S d
Smy 2(c y ) 0.25k1 by
10 y
55
where cx , cy , and Sx are the parameters determined from the reinforcement layout as
shown in Figure 3.7. k1 is taken as 0.4 for the deformed bars and 0.8 for plain bars. d bx
and dby are longitudinal and transverse the bar diameters, respectively. S is center-to-
center spacing of the transverse reinforcement. The crack control characteristics of the
longitudinal and transverse reinforcement can, alternatively, be also be taken as
Smx 1.5 maximum distance from x-bars and Smy 1.5 maximum distance from y-bars .
The term in Equation 3.24 is inclination of the crack with respect to longitudinal axis
as shown in Figure 3.6a. Its value can be determined as c such that 0 90 o .
If Equation 3.20 is not satisfied i.e., the local shear stress on the crack exceeds the
shear resistance provided by the aggregate interlock mechanism, the average concrete
tensile stress f c1 must be reduced as
f c1 0.18vci max tan c (3.25)
3.4.3 Material Stiffness Formulations.
Vecchio (1989) presented a procedure that showed that nonlinear analysis of
reinforced concrete membrane elements can be performed accurately by simply
modifying existing linear elastic finite element routines based on secant stiffness
formulation provided realistic constitutive relations for concrete and reinforcement are
employed. In constructing an individual element stiffness matrix K , a material stiffness
matrix D is required to relate stresses to the strains. The material stiffness matrix for a
linear elastic isotropic material in state of plane stress can be modified in a form that
56
depends on type of stiffness moduli used to reflect nonlinear behavior of reinforced
concrete according to appropriate set of constitutive laws. Using the same concept in
ASFI approach, the material stiffness matrix D for axial-shear element is determined
by first defining stiffness matrices for concrete and reinforcements with respect to their
respective principal material directions using secant moduli. The total stiffness matrix is
then obtained by combining the component stiffness matrices, after appropriate
transformations to take into account the directional dependence of the materials. Hence,
D T Dc T T Dsi T
T T
(3.26)
i
where, D c is the concrete material stiffness matrix evaluated with respect to principal 1
and 2 axes system corresponding to the direction of the principal tensile strain and
principal compressive strain, respectively.
Ec 2 0 0
D c 0 Ec1 0

(3.27)
0 Gc
0
where, Ec1 and Ec 2 are secant moduli for the concrete and relate to the stress-strain
behavior in principal directions. These are determined as:-
f c1 fc 2 Ec1 Ec 2
Ec1 , Ec 2 , Gc (3.28)
c1 c2 Ec1 Ec 2
For each reinforcement component, a reinforcement material stiffness matrix D si is

evaluated as follows,
i Esi 0 0

D si 0 0 0 i x, y (3.29)
0 0 0

57
where, E si is the secant moduli for the reinforcing steel and relate to the stress-strain
behavior in two orthogonal directions x and y . For a particular stress/strain state,
f sx f sy
Esx , Esy (3.30)
x y
The transformation matrix T Equation 3.26 will differ for each component and is given
by following expression
cos 2 sin 2 cos sin

T sin 2 cos 2 cos sin (3.31)
2 cos sin 2 cos sin cos 2 sin 2

where, c for the concrete component, 0 and 90o for longitudinal and transverse
reinforcement, respectively (Figure 3.8).
Having determined the material stiffness matrix D , it can then be used to relate
stresses f to strains
f D D f
1
or (3.32)
where,
fx x

f f y and y (3.33)

xy xy
3.5 Development of Axial-Shear-Flexure Interaction Approach
The main objective of the ASFI approach is to couple axial-flexural and axial-
shear mechanisms satisfying compatibility and equilibrium conditions while considering
58
their interaction in terms of axial deformations and concrete compression strength
softening, as explained in the following sections.
3.5.1 Compatibility Conditions/relationships
Figure 3.2 illustrates an element between two flexural sections of a reinforced
concrete column subjected to axial load, bending moment and shear force. Total axial
deformation in a reinforced concrete column can be considered as sum of axial
deformations due to three mechanisms of axial, flexural and shear. In an axial-flexural
mechanism, total axial strain is combination of axial strains caused by axial mechanism
xaf and axial strain due to flexural mechanism xf . Likewise, in an axial-shear
mechanism, axial strain is combination of axial strain due to axial mechanism xas and
axial strain caused by shear mechanism xs . Hence, total axial strain x of the column
between the two sections can be obtained by extracting xf from axial-flexural model and
adding to axial deformations of axial-shear model.
x xas xs xf (3.34)
Compatibility of axial deformations is satisfied when axial deformations due to axial
mechanisms in axial-shear and axial-flexural elements are equal to the axial strain due to
only applied axial load xa , that is
xa xaf xas (3.35)
Hence, Equation 3.34 becomes
x xa xs xf (3.36)
59
Furthermore, the total lateral drift of a reinforced concrete column or beam
between two flexural sections is taken as summation of shear strain s and the flexural
drift ratio f .
s f (3.37)
3.5.2 Equilibrium Conditions/relationships
Equilibrium conditions in ASFI are satisfied in average stress-strain field. For
equilibrium, axial stress in axial-flexural element xf should be equal to axial stress in
axial-shear element xs . Also, the equilibrium of shear stress in axial-flexural element f
and shear stress in axial-shear element s is satisfied simultaneously throughout the
analysis.
xf xs o
(3.38)
f s
where o is axial stress due to applied axial load and is resultant shear stress. Shear
stress in axial-flexural and axial-shear mechanisms are calculated as,
1 M1 M 2
f
bd f l12
(3.39)
V
s
bd s
where M 1 is larger moment on one of the flexural section, M 2 is smaller moment on
other flexural section, b is width of the cross-section, d f is the flexural depth of the
section which can be taken as d f h before concrete cracks in flexure and
60
d f d afterwards. h is overall depth of the section, d is effective depth of the section,
V is applied lateral load, d s is shear depth of the section and can be taken as d s h until
concrete tensile crack due to flexure and then d f d .
Normal stress o due to applied axial load P in both axial-flexural and axial-shear
models is calculated as
P P
o (3.40)
Ai bh
where Ai equals the cross-sectional area of fiber i . Normal stress in the direction
perpendicular to the longitudinal axis of the column, or the clamping stresses
y and z are neglected due to inexistence of lateral external forces along theses
directions.
y z 0 (3.41)
3.5.3 Constitutive Laws
In order to satisfy above mentioned compatibility and equilibrium conditions,
same material constitutive relationships must be employed in both axial-flexural and
axial-shear models. In an axial-flexure model, if constitutive laws of confined core
concrete and unconfined cover concrete are defined differently, constitutive law for the
concrete in axial-shear model must also be modified. Equivalent peak concrete
compression stress fc'as and corresponding strain c' as can approximately be calculated as
f c'as
KAcon Auncon f , c' as
KAcon Auncon (3.42)
co co
A A
61
where A is gross cross-sectional area, Acon is area of confined core, Auncon is area of
unconfined cover and K is confinement parameter is taken as ratio of peak confined
concrete strength and concrete cylinder strength. This modification is not required if
confinement effect is applied only on post peak concrete response. In view of the
simplicity and applicability of secant stiffness method in MCFT, it is applied in axial-
shear model of ASFI approach.
3.5.4 Interactions considered
Axial deformation and concrete compression softening are two main interaction
terms used in ASFI approach interconnecting axial-flexural and axial-shear mechanisms.
3.5.4.1 Axial Strain Interaction Methodology
Axial deformations play a very important role in ASFI approach by
interconnecting axial-flexure and shear mechanisms. Axial deformations due to flexure
increases shear crack width and principal tensile strain in the web resulting in lower shear
capacity for the element. As already mentioned, total axial strain in an element between
two sections is sum of axial strain caused by axial, shear and flexural mechanisms
i.e., x xa xs xf . Axial-shear model gives axial strain caused by axial and shear
mechanism xa xs . Therefore, in order to obtain total axial strain x , axial strain due to
flexure xf from axial-flexure model (determined from Equation 3.3) must be added to the
axial strains of axial-shear model. This can be done by simply adding flexibility
component of axial deformation due to flexure to the corresponding flexibility
62
component of axial-shear model. The flexibility component for axial deformation due to
flexure f xf can be determined as,
xf
f xf (3.43)
x
where, x is applied axial stress in longitudinal direction of the column. In case of the
beams or columns where axial load is zero, in order to avoid having an indefinite value in
Equation 3.43, a very small value must be considered for axial stress.
A stress-strain relationship in terms of flexibility matrix for an in plane shear
element (axial-shear model) can be defined as,
f11 f12 f13 x x xas xs

f 21 f 22 f 23 y y (3.44)
f
f33 xy
31 f32
xy
where, fij i, j 1, 2,3 are flexibility components of in plane shear model,
x xas xs is axial strain in axial-shear element and y is strain in transverse
reinforcement. Axial strain due to flexure xf can be taken into account to axial-shear
model by adding flexibility component obtained from Equation 3.43 into Equation 3.44.
Considering x axis as the main longitudinal axis of the column, stress in
x direction is assumed equal to the applied axial stress, i.e. x o . Stresses in
y direction (clamping stresses) are zero due to inexistence of lateral external force
along the column, i.e. y 0 . Hence, flexibility matrix for axial-shear-flexure element
takes the form
63
f11 f xf f12 f13 o x xa xs xf

f 21 f 22 f 23 0 y (3.45)

f31 f32 f33 xy xy

where xa xs is axial strain in axial-shear model considering the compatibility,
xas xa .
3.5.4.2 Concrete Compression Softening
Concrete compression softening is another interaction term in ASFI approach.
However, unlike interaction of axial deformation from axial-flexural model and axial-
shear model, compression softening of axial-shear model is taken into account in axial-
flexural model. Hence, compression softening factor determined from axial-shear
mechanism (Equation 3.2) must be employed in flexural analysis to soften the uniaxial
compressive behavior of the concrete in axial-flexure model.
3.5.5 Stiffness Model in ASFI
In ASFI approach, stiffnesses of axial-shear-flexure element are modeled as
springs in series as illustrated in Figure 3.9. Shear spring of axial-shear model with
stiffness K s is in series with flexural spring of axial-flexural element of stiffness K f .
Three axial springs, i.e., axial spring in axial-flexural model with stiffness k xf , axial
spring in axial-shear model with stiffness k xs , and axial spring of axial mechanism with
stiffness k xa are also in series. Flexural and shear stiffnesses can be determined based on
following expressions:-
64
f s
Kf , Ks (3.46)
f s
Total stiffness K corresponding to total drift ratio is,
1 1 1
(3.47)
K K f K s
Hence, total shear-total deformation relation can be determined as K where
s f .
As axial strain of axial mechanism is contributed in both axial-shear and axial-
flexural models, in order to determine total axial stiffness k x , axial stiffness from axial-
flexural mechanism is considered into axial-shear mechanism. When axial stiffness is
defined for each axial spring individually based on its deformation, total axial stiffness
can be determined as,
1 1 1 1

k x k xa k xs k xf (3.48)
k x x x where x xs xf xa
where, k xa is axial stiffness of axial mechanism, k xs is axial stiffness of shear mechanism,
and k xf is axial stiffness of flexural mechanism
x
k xa , k xs x , k xf x (3.49)
xa xs xf
3.6 Conclusion
In this chapter, theory and conceptual development of ASFI approach is presented
(Mostafaei and Kabeyasawa, 2007). ASFI approach is a macro-model based approach
65
that consists of two models; axial-flexure and axial-shear model evaluating flexural and
shear responses simultaneously by employing standard flexural section analysis
techniques for axial-flexure response and MCFT for axial-shear response. Total response
at any load step is obtained by adding deformations from component models obtained by
considering interaction of axial strains, compression softening and compatibility and
equilibrium conditions.
66
c2
c2
f co c2

f co c2
c1 c2

c2
co c2
Figure 3.1: Concrete compression softening

P
V M4
M3
Flexural Section 2 M2
Flexural Section 1 M1
Inflection Point Mo
-M1
-M2
-M3
-M4
o2
Flexural Section 2
M2 o2
Centroidal
l12
Axis
Flexural Section 1 h o1
M1
P
o1
Figure 3.2: Axial-Flexural model and determination of axial strain due to flexure
67

y
x x
Lin Lp Lin- Lp
y y
Figure 3.3: Plastic hinge model for flexural deformations
y
y
fy
xy y
c1 y
1
.5
x

xy
fx fx
2
c2

.5
xy x xy
x x
fy
(a) (b) (c)
2
y
y
xy / 2

2 1
2 c
x
x
c2
c1
(d)
Figure 3.4: (a) Membrane element subjected to in-plane stresses (b) Average strains
from in-plane loading (c) Average strains in cracked concrete
(d) Mohrs circle for average strains
68
fc2 f c1
2
f c 2 f co 2 c 2 c 2
f co co co f cr
Ec c1
f cr
1 200 c1
co c2 cr c1
(a) (b)
f si
f si Es si f yi
f yi
yi si
(c)
Figure 3.5: Average stress-strain relationship for
(a) Cracked concrete compression (b) Cracked concrete in tension (c) Reinforcing steel
y y
f sy f sy
ny
f c1 ci
fx 1 1
fx
f sx f sx
nx
2 2
xy x xy x
fy fy
(a) (b)
Figure 3.6: (a) Calculated average stresses (b) Local stresses at a crack
69
Cy
dby

Cx
c
Sx d bx
Figure 3.7: Parameters influencing crack spacing
y y
1
90O
0O
x x
C
2
Figure 3.8: Coordinate transformation for concrete and steel material stiffness matrices
P
M2
V Axial-Shear Ks
Model
k xs
Kf
k xa Ks
Kx
Kf k xf
V
Axial-Flexural
M1
Model
P
Figure 3.9: Spring model for axial-shear-flexural element
70
CHAPTER 4
IMPLEMENTATION OF ASFI APPROACH FOR COLUMNS
4.1 Introduction
The theory and conceptual development of ASFI approach were presented in the
previous chapter by explaining the details of its component models (axial-flexural and
axial-shear) and interaction methodology. Although theoretical aspects of the approach
and the details of the analytical models were derived in generalized form for a reinforced
concrete element between two subsequent flexural sections of the member, the approach
can be extended to evaluate structural performance of the reinforced concrete elements
such as beams and columns. In addition to the models considered for axial, flexural and
shear mechanisms in ASFI approach, the deformations caused by reinforcement slip at
the section of maximum moment must also be incorporated into the analysis for
estimating total response.
In this chapter, the analytical procedure for implementation of ASFI approach for
response estimation of reinforced concrete column subjected to lateral load is outlined.
Solution technique for implementing the analytical procedure is based on ASFI approach
by Mostafaei and Kabeyasawa (2007) as described in Chapter 3. The approach is
implemented in a MATLAB (MathWorks, 2009) computer program written in this study
71
and lateral load-lateral drift responses of previously tested columns are presented to
verify correct implementation of the model. The simulated lateral load-deformation
responses of the test specimens shall be used subsequently for discussion and
comparison.
4.2 Response Estimation of Columns by ASFI Approach
In Chapter 3, it was demonstrated that deformation response of an element
between any two subsequent flexural sections of a reinforced concrete element could be
estimated by considering the interaction of axial deformation and satisfying equilibrium
and compatibility conditions in average stress-average strain environments. This is
theoretically admissible approach as equilibrium of internal and applied forces can be
satisfied at any load level. Hence, this approach can principally be applied for response
estimation of reinforced concrete columns subjected to the lateral loads. However,
instead of considering an arbitrary element between two subsequent flexural sections of
the column, an axial-flexure element and an axial-shear element are considered for
modeling the element from its inflection point to one of the end sections. Conventional
flexural section analysis or fiber model approach, as described in Section 3.3, is applied
for modeling flexural behavior of the axial-flexure element and modified compression
theory as explained in Section 3.4 is employed for modeling axial-shear element. Given
the compatibility and equilibrium relationships, axial-flexure and axial-shear elements are
coupled into one-component model considering axial interaction and material constitutive
relationships in accordance with Section 3.5. In addition, a pullout model as explained
72
below is considered to account for deformation caused by reinforcement slip at the
section adjacent to flexural end section.
4.2.1 Pullout Effect Consideration and Total Deformations
Pullout or reinforcement slip deformations are experienced in fixed ended
reinforced concrete column subjected to bending moment due to rigid body rotation of
the column resulting from slip of longitudinal reinforcing bars out of foundation or beam-
column joint. Unlike flexural and shear deformations that occur along the length of the
column, pullout deformations occur at the end section(s). The mechanics of the pullout
deformations has been explained in detail in Section 2.2.2 and illustrated in Figures 2.1
and 4.5. Axial-flexure model (fiber section analysis) and axial-shear model (MCFT) can
not account for slip or pullout deformations, therefore these must be determined
separately and added to flexural and shear deformations for total response.
To account for pullout effects in total performance of the column, total lateral
drift is taken as sum of the flexural drift ratio, shear strain and pullout rotation Rpul . In
addition, centroidal strain due to pullout pul must also be added to the total axial
deformation of the column. End rotation and centroidal strain due to pullout are
illustrated in Figure 4.5. The pullout element is considered as a rotational spring at
column end, which is in series with springs of flexure, and shear mechanisms. Hence,
total stiffness K corresponding to total drift ratio is,
1 1 1 1

K K f K s K pul (4.1)
K
73
where,
f
K pul (4.2)
pul
s f pul (4.3)
For axial deformation,
1 1 1 1 1

k x k xa k xs k xf k xpul (4.4)
k x x x
where,
x
k xpul (4.5)
xpul
xtot xa xs xf xpul (4.6)
In above equations, K f is flexural stiffness, K s is shear stiffness, is resultant
shear stress and f is shear stress in axial-flexure element, as defined in Section 3.5. In
pullout model, K pul is pullout stiffness, pul is pullout drift that equals Rpul and kxpul is
pullout axial stiffness. In Equation 4.4, k xa , k xs , k xf are axial stiffnesses of axial, shear,
flexural mechanisms, respectively and are defined in Section 3.5.5.
Total axial and lateral deformations in ASFI approach are computed based on
summation of corresponding deformation from four mechanisms of axial, axial-flexure,
axial-shear and pullout. Hence, total lateral drift ratio in Equations 4.3 is sum of shear
deformation s from axial-shear model, lateral flexural deformation f obtained from
axial-flexure model and pullout rotation pul from pullout mode. Similarly, total axial
74
strain x tot in Equation 4.6 is sum of axial strain due to axial xa , shear xs , flexure xf
and pullout xpul pul mechanisms, respectively.
It must be noted that any of the available pullout models such as Otani and Sozen
(1972), Hawkins et al. (1982), Morita and Kaku (1984), Alsiwat and Saatcioglu (1992),
Lehman and Moehle (2000) and Setzler and Sezen (2008) etc can be used to determine
rotation and axial deformation due to reinforcement slip and include these effects in
overall response.
4.2.2 Analytical Procedure for Implementation of ASFI Approach
This section summarizes the procedure for estimating the response of laterally
loaded reinforced concrete columns employing ASFI approach.
1. Consider an axial-flexure element of reinforced concrete column from its
inflection point to one of the end section. Perform flexural section analysis on end section
employing compression-softening factor to soften the compressive stress-strain
relationship of the concrete given by any of the appropriate constitutive law for uniaxial
concrete compressive behavior. Importance of considering confinement effects,
contribution of concrete tensile stresses in the flexural analysis, considering confined
concrete model for core and unconfined concrete model for the cover, and choice of
realistic constitutive material models have already been highlighted in Section 3.3
outlining the details of the axial-flexure model in ASFI approach.
2. Perform axial section analysis at the inflection point and determine axial strain
due to applied axial load only xa considering same material constitutive laws employed
75
for flexural section analysis of the end section. Assuming a linear distribution of average
centroidal axial strain between end section and section at inflection point, axial strain due
to flexure xf can be determined as
lin
1 x
xf
lin
0
o xa
lin
dx 0.5 o xa (4.7)
where, lin is the length of the column between end section and inflection point, o is total
centroidal axial strain determined through flexural section analysis of the end section and
xa is the axial strain due to applied axial load only. It may be noted that xf is also the
average value of axial strains due to flexure only between at end section and section at
inflection point and considered constant over the entire length of the axial-flexure
element.
3. Determine flexural drift and pullout deformation employing any of the
appropriate models and calculate respective stiffnesses.
4. Consider an axial-shear element from inflection point to end section and apply
MCFT as explained in Section 3.4 and determine required material stiffness matrices
considering secant stiffness formulation and relevant material axis transformations.
5. Apply axial strain interaction methodology while satisfying compatibility and
equilibrium conditions as per Section 3.5 and convert axial-flexure and axial-shear
elements into one axial-shear-flexure element.
6. Check convergence of the deformations by considering stiffness model in ASFI
approach as outlined in section 3.5.5 and use of Equation 4.8 relating stresses and strains
of ASFI element through material stiffness matrix. Also, include the effects of pullout
deformation in the ASFI stiffness model as explained in Section 4.2.1.

76
f D D f
1
or (4.8)
4.2.3 Solution Technique for Implementing Analytical Procedure
In the previous section, the analytical procedure for implementation of ASFI
approach is presented. The procedure is iterative and outlines only the major steps
required for response estimation. This section summarizes the solution technique for
implementing above-mentioned analytical procedure for analyzing response of a
reinforced concrete column with ASFI approach.
1. Define or input material properties and geometry. Decide on a sign convention for
tensile and compressive stresses and strains.
2. Input axial load P . If there is no applied axial load, then consider a negligible
small value for P .
3. Select a small value of total drift ratio , such as 0.000001 , as a starting value.
4. Consider the variables of iteration as oi ,i , xi , yi , si and assume some very small
values for each of them for the first iteration. These variables are axial centroidal strain at
the end section in the axial-flexure element ( oi ), curvature of the end section in axial-
flexure element ( i ), average normal strain in x -direction for axial-shear element ( xi ),
average normal strain in y -direction for axial-shear element ( yi ) and average shear
strain of axial-shear element ( si ), respectively.
5. Calculate compression-softening factor with the help of Equation 3.2.
77
6. Perform flexural section analysis at the end section of the axial-flexure element
and determine nominal moment capacity M and corresponding axial centroidal
strain oi 1 .
7. Determine flexural-shear stress f by Equation 3.39 modified for axial-flexure
element of the column as,
M
f (4.9)
Bd f lin
8. Perform axial section analysis at the inflection point and determine axial strain,
xa due to applied axial load only. Calculate axial strain due to flexure xf with Equation
4.7 by replacing o with oi 1 determined in step 6.
9. Determine flexural drift ratio f and pullout deformations by any of the
appropriate models. Pullout deformations include pullout rotations pul and axial strain
due to pullout xpul .
10. Determine o , f xf , K f , and K pul from Equations 3.40, 3.43, 3.46, and 4.2
respectively.
11. Apply MCFT, as explained in Section 3.4 and follow following steps
a. Determine average concrete principal tensile and compressive strains
c1 and c 2 by Equations 3.5 and 3.6, respectively. Also, determine
orientation of principal tensile plane t and principal compressive plane
strains c with the help of Equations 3.7 through 3.9.
78
b. Determine average steel stresses f sx and f sy corresponding to the
strains x and y in accordance with employed constitutive laws for
reinforcement in x and y -directions respectively.
c. Determine average principal tensile and compressive stresses in concrete
by considering same material constitutive laws employed in axial-flexural
model (step 6). However, if in axial-flexure model cover and core concrete
are modeled separately then concrete compressive strength and
corresponding strain must be modified according to the Equation 3.42.
d. Check local stress-strain conditions at cracks ensuring that the tension in
the concrete can be transmitted across the crack and shear stress on the
surface of the crack does not exceed maximum shear provided by the
aggregate interlock by Equations 3.18 through 3.24. If Equation 3.20 is
not satisfied and local shear stress at the surface of the crack exceeds shear
strength provided by aggregate interlock mechanism, then average
concrete principal tensile stress must be reduced as per Equation 3.25.
e. Determine secant moduli for concrete and reinforcements and assemble
material stiffness matrices using Equation 3.27 through 3.30. Determine
total material stiffness D matrix by transforming component material
stiffnesses into global x and y -directions using Equation 3.26 and 2.31.
12. Invert total stiffness matrix D to get flexibility matrix f of Equation 3.44 and
add flexibility component of axial deformation due to flexure f xf to f11 component of f
to determine flexibility matrix of coupled axial-shear-flexure element.
79
13. Consider x o as per Equation 3.40, y 0 , and si from initial assumed
value in the iteration, and obtain s and K s using Equations 4.8 and 3.46. Then determine
K and using Equation 4.1.
14. Assume x o as per Equation 3.40, y 0 , and from step 13, and compute

xi 1 , yi 1 , and si 1 for the integration point using Equation 4.8. Determine f and
Kf
calculate i 1 for the flexural section.
15. Assess the convergence of the acquired variables oi 1 , i 1 , xi 1 , yi 1 , and si 1 . If
satisfactory convergence is achieved, then go to the next step otherwise repeat steps 5 to
15 with improved estimate of the iteration variables.
16. Compute lateral load capacity of the column corresponding to the given total drift
by Equation 3.39 replacing s with obtained in step 13.
17. Compute axial deformation by Equation 4.6.
18. Increment total drift ratio and repeat steps 4 to 17 until required response is
evaluated.
These analytical steps are summarized in a flow chart as shown in Figure 4.1.
4.3 Implementation and Verification of ASFI Approach
Mostafaei and Kabeyasawa (2007) developed and verified ASFI approach by
reporting consistent correlation between estimated analytical response and experimental
results for a number of reinforced concrete test specimens. In order to implement the
ASFI approach, a computer program in MATLAB (MathWorks, 2009) is written and
80
lateral load-deformation responses of previously tested reinforced concrete columns are
estimated. The test specimens modeled in this study are the same which were originally
used by Mostafaei (2006) for verification of ASFI method. Calculated responses are
compared with experimental test data to verify correct implementation of the analytical
procedure of ASFI approach. While modeling test specimens, all efforts are made to use
the same data and models used in the original study and appropriate details are assumed
where sufficient information is not available in the literature.
4.3.1 Details of Test specimens and Material Properties
Five reinforced concrete columns tested by Ousalem et al. (2003) are analyzed in
this study. These columns were loaded under constant axial load of 540 kN and
unidirectional cyclic lateral loads. The columns were designed considering 1/3 scale of
actual sized column, representing columns located in the first story of a building with
moderate height. The columns were designed to fail in different failure mechanisms like
shear-tension, shear-compression, shear-flexure and flexural failures. The test columns
were fixed against rotations at top and the bottom. Geometric and material properties of
test specimens are presented in Table 4.1. The complete details of the tests and discussion
of the analytical results can be found in Mostafaeis PhD dissertation (2006).
4.3.2 Constitutive Laws
Same constitutive laws for concrete and reinforcements are used for both of the
axial-flexure and axial-shear models in ASFI approach. The material constitutive laws
81
used to model behavior of the test columns under consideration are presented in
following sections.
4.3.2.1 Constitutive Law for Concrete in Compression
Stress-strain relationship for concrete in compression employed for both axial-
flexure and axial-shear elements is illustrated in Figure 4.2. Pre-peak stress-strain
relationship i.e., c p is given by following expression
c
2

fc f p 2 c (4.10)
p p

Stress-strain relationship for the post-peak response, i.e., c p is defined based
on concrete confinement as follows
f c f p 1 Z m c p (4.11)
where,
f p K f c' and p K co (4.12)
In Equations 4.11 and 4.12, parameters K and Z m are confinement parameters.
These are determined based on modified Kent and Park model for stress-strain
relationship of concrete confined by rectangular steel hoops (Park et al. 1982).
s f yh
K 1 (4.13)
f c'
0.5
Zm (4.14)
3 0.29 f co 3 h"
K co
145 f c' 1000 4
s
sh
82
where, s is ratio of the volume of rectangular steel hoops to volume of concrete core
measured to outside of the peripheral hoop, f yy is yield strength of the steel hoop, h " is
width of the concrete core measured to the outside peripheral hoop, sh is center to center
spacing of hoop set, and co is strain corresponding to concrete compressive cylinder
strength f co . In Equation 4.12, is compression softening factor as defined in Equation
3.2.
4.3.2.2 Constitutive Law for Concrete in Tension
There is not enough clarification in the literature reported by Mostafaei (2006)
about the final choice of the tensile concrete constitutive material model used for
producing load-deformation responses of the test specimens. The reported literature
discusses two models, one developed by Vecchio and Collins (1986) (the model used in
original MCFT, details of which are given in Section 3.4.1.2) and the other based on
model presented by Izumo (1992). Theoretically, any of the appropriate model can be
used to simulate concrete behavior in tension. However, it is assumed here that
constitutive relationship for average concrete tensile stress-strain used in the originally
reported estimated responses is the one based on model by Izumo. The model is
presented below and illustrated in Figure 4.3.
The relationship suggested prior to cracking i.e., 0 c1 cr is linearly
elastic and is given by the expression:-
f c1 Ec c1 (4.15)
83
where, Ec is modulus of elasticity of the concrete, cr is cracking strain corresponding to
uniaxial cracking strength of concrete f cr , and f co uniaxial compressive strength of
concrete in MPa. These quantities are given by following expressions
Ec 2 fc' co ; cr fcr Ec ; f cr 0.33 f c' (4.16)
After cracking, the concrete tensile stresses remain constant up to strain of
2 cr and then decline in post-peak range as given by following expressions.
f c1 f cr for cr c1 2 cr (4.17)
0.4
2
f c1 f cr cr for c1 2 cr (4.18)
c1
4.3.2.3 Constitutive Law for Reinforcements
A tri-linear stress-strain relationship is considered for the longitudinal and
transverse reinforcements as illustrated in Figure 4.4 and expressed by following set of
expressions.
f s Es s for s y
fs f y for y s sh (4.19)
s sh
fs f y +
u sh
f u fy for sh s u
where, E s modulus of elasticity of steel, f y and fu are yield and ultimate strengths of
steel, respectively, y , sh and u are yield strain, strain at onset of strain hardening and
ultimate strain, respectively.
84
4.3.3 Flexural Deformation Model
Plastic hinge model as illustrated in Figure 3.3 is employed to model flexural
deformations by integrating curvatures along the length of the column. A linear curvature
distribution is assumed in the elastic range, and the inelastic curvature is lumped at the
column end over plastic hinge length. Plastic hinge length L p is taken equal to the height
of the column cross-section. Hence, flexural drift ratio can be determined as,
Lin
1
f
Lin x ( x)dx
0
(4.20)
Using Equation 4.20, flexural drift ratio before yielding, i.e., y can be calculated as,
1
f Lin (4.21)
3
and, after yielding i.e., y
1
Lin L p Lin 1 y Lin Lp
Lin
f x ( x)dx x ( x)dx x 2 dx xdx (4.22)
Lin 0

L Lin L L
in p

Lin L p in
0
In above equations, Lin is the distance from column end section to inflection
point, L p is plastic hinge length, ( x ) is curvature function at distance x measured from
inflection point and y is yield curvature.
4.3.4 Pullout Deformations Model
Pullout model by Okamura and Maekawa (1991) is employed in the analytical
process of ASFI approach to model pullout deformations of test specimens. The pullout
85
model and reinforcement slip-strain relations are illustrated in Figure 4.5 and 4.6.
Important relations used in the model are
23
20 MPa
slip SD ' (4.23)
fc
slip
pul Rpul (4.24)
X
pul
eR 0
pul
(4.25)
Lin
where S is normalized slip as defined in Figure 4.6, D is reinforcing bar diameter in
mm, f c' is concrete compressive strength in MPa, pul is pullout drift ratio which equals
pullout rotation R pul , pul is centroidal strain due to pullout, Lin is the distance from the
end joint to the inflection point. Other parameters like slip , X and e are as shown in
Figure 4.5.
4.3.5 Buckling of Compression Bars and Post-peak Analysis
In ASFI approach, buckling or slip of compression bars is considered by lowering
compressive stresses in longitudinal reinforcing bars when unconfined cover concrete
fibers adjacent to the reinforcement reach 30 % of the maximum concrete strength. After
reaching this stage, steel stresses are reduced linearly according to post-peak compression
stiffness of the confined core.
It must also be noted that in post-peak analysis, pullout and shear stiffnesses are
kept constant at their least value during the analysis. In post-peak stage of the analysis,
only flexural stiffness is the main variable. Also, compression softening factor is used
86
with its calculated value until it start to increase. At this stage, it must then be kept
constant equal to its minimum value for rest of the analysis. In ASFI approach, axial
failure or gravity collapse is defined as the stage when equilibrium in vertical direction in
section analysis cannot be satisfied any more under the applied axial load. At this stage,
the column reaches complete loss of lateral load capacity.
4.4 Comparison of the Results
Lateral load-drift responses for specimens No 5, 12, 14, 15 and 16 are presented
in Figure 4.7 through 4.11. Comparison of the predicted response with experimental data
indicates that ASFI approach performs very well in evaluating structural performance of
all of the test columns. Predicted responses follow initial stiffness very well and peak
loads are estimated quite accurately. Estimated post-peak responses also compare well
with experimental post-peak behaviors in general. Good correlation achieved between
calculated and actual behavior indicates that ASFI approach is effective displacement
based evaluation approach for response estimation of reinforced concrete columns. It can
also be concluded that the ASFI approach has correctly been implemented in the analysis
as per analytical procedure described in Section 4.2 above.
4.5 Summary and Conclusion
Analytical procedure and solution technique for implementation of ASFI
approach for response estimation of reinforced concrete columns is presented. The
procedure in incorporated into a programming routine written in MATLAB (MathWorks,
2009) and response of five columns is estimated based on ASFI approach. Comparison of
87
calculated total lateral load-total lateral drift responses of test specimens with their
experimental responses indicates that ASFI approach is an effective tool for response
estimation of columns to lateral loads. The results also show that the analytical procedure
and solution technique of ASFI approach has correctly been implemented in Matlab
computer program.
88
Input material properties and
geometrical dimensions
Input axial load P . If P 0 , then define P as a negligibly small value.

Specify initial/increment total lateral drift value
Consider variables in iteration i as oi , i , xi , yi , si
Apply axial-flexural model and compute

oi 1 , xa , f , f , pul , pul , f xf , K f , K pul
Apply axial-shear model based on MCFT and construct total

stiffness matrix D of axial-shear element.
Compute flexibility matrix f of axial-shear element by inverting D and obtain total

flexibility matrix of axial-shear-flexure element by adding f xf to f11 term of f
Considering x P bh , y 0 , and si obtain s and K s s si . Get K and .
Considering x P bh , y 0 , and , compute xi 1 , yi 1 , and si 1 for the

integration point and then calculate f and i 1 for the flexural section
Kf
Deformation converged No
oi 1 ,i 1 , xi 1 , yi 1 , si 1
Output shear force and axial strain and calculate total load V bd s .
Increment drift ratio and repeat above steps until failure
Figure 4.1: Flow chart for implementation of ASFI approach for columns
89
35
Confined concrete
Stress (Mpa) 25
20
15
10
0
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
Strain
Figure 4.2: Constitutive relationships for concrete in compression
1.8
1.6
1.4
1.2
Stress (Mpa)
0.8
0.6
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1
Strain x 10
-3
Figure 4.3: Constitutive relationship for concrete in tension

90
600
500
400
Stress (Mpa)
300
200
100
0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Strain
Rpul
pul slip
slip
X pul R pul
X
e
pul
e 0
pul
Lin
Figure 4.5: Rigid body rotation of column due to reinforcement slip or pullout
91
S Normalized slip
20MPa
slip S D
23
S
Sy

6 7 f u f y s sh
'
fc 2 100
S y 0.2 fu f y s sh
Sy
S s 6 3500 s
s
y sh
Figure 4.6: Pullout deformation model (Okamura and Maekawa, 1991)
92
350
Reported response
Predicted response
300
250
Lateral force (kN)
200
150
100
50
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Drift ratio
Figure 4.7: Comparison of the results for Specimen-5 (Ousalem et al, 2005)
93
300
Reported response
Predicted response
250
200
Lateral force (kN)
150
100
50
0
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045
Drift ratio
94
350
Reported response
Predicted response
300
250
Lateral force (kN)
200
150
100
50
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
Drift ratio
95
350
Reported response
300 Predicted response
250
Lateral force (kN)
200
150
100
50
0
0 0.05 0.1 0.15 0.2
Drift ratio
96
400
Reported response
350 Predicted response
Lateral force (kN) 300
250
200
150
100
50
0
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045
Drift ratio
97
bxh 2Lin sh g w fyx fyy fco
Specimen
(mm) (mm) (mm) (%) (%) (Mpa) (Mpa) (Mpa)
No. 5 300x300 750 50 2.65 0.43 420 410 28.5
No. 12 300x300 900 150 2.26 0.14 415 410 28
No. 14 300x300 900 50 2.26 0.43 415 410 26
No. 15 300x300 900 50 2.26 0.85 415 410 26
No. 16 300x300 600 50 1.80 0.43 415 410 27
b = Width of the column cross section; h = Height of the column cross section;
Lin = Length of the column from inflection point to end section; sh = Hoop spacing;
g = Longitudinal reinforcement ratio; w =Transverse reinforcement ratio;
fyx = Longitudinal reinforcement yield stress; fyy = Transverse reinforcement yield
stress; fco = Concrete compressive cylinder strength
Table 4.1: Geometric and material properties of the test specimens
98
CHAPTER 5
COMPARISON OF DISPLACEMENT COMPONENT MODEL AND ASFI

APPROACH
5.1 Introduction
In previous chapters, two approaches for response estimation of reinforced
concrete columns were presented. In order to verify correct implementation of each
approach, test specimens were modeled with material constitutive relationships and other
component models as proposed in the original study. Comparison of predicted and
experimental responses indicated that both methods perform equally well despite
following quite different approaches for response estimation of respective set of test
columns. In order to evaluate performance of both understudy approaches, it is necessary
to model the same set of specimens employing same material and deformation
component models within each approach.
In this chapter, a comparative study is conducted to evaluate capabilities and
limitations of displacement component model by Setzler and Sezen (2005) and ASFI
approach by Mostafaei and Kabeyasawa (2007). Both approaches are implemented for
same columns employing same set of models for material stress-strain relationships,
flexural and pullout deformations. This comparison forms the basis of the proposed
model presented in Chapter 6.
99
5.2 Implementation of the Analytical Models
In Chapter 2, displacement component model by Setzler and Sezen (2005) is
outlined and response estimated for four columns (Sezen, 2002) is compared with
experimental responses. In Chapter 4, lateral load-deformation responses of specimens
tested by Ousalem et al. (2003) based on ASFI approach are presented. Within
implementation of each of these approaches, same material models are employed which
were originally used in development and verification of the respective approaches. For
comparing estimated responses by both of understudy approaches, it is imperative that an
appropriate set of test specimens and material models are selected and employed for
response estimation by each approach.
5.2.1 Test Specimens and Material Properties
Reinforced concrete columns tested by Sezen (2002) provide data for flexural,
reinforcement slip and shear deformations in addition to total deformation responses.
This provides a unique opportunity to compare and evaluate capabilities of any of the
analytical model intending to predict overall response. Hence, columns by Sezen (2002)
are employed for estimation and comparison of responses by both approaches. Details of
these test specimens were presented in Section 2.5.1 and key features are repeated here
for easy reference.
These columns, designated as Specimen-1 through -4, are fixed ended columns
with square cross-section of 18 inches and length of 116 inches. The columns have eight
No.9 bars as longitudinal reinforcement and transverse reinforcement of No.3 ties with
90- degree hooks at 12 inches center-to-center spacing. Specimen-1, 2 and 4 were tested
100
with constant axial load of 150, 600 and 150 kip, respectively. Specimen-3 was, however,
tested under varying axial load from 60 kip in tension to 600 kips in compression.
Specimen-1 through 3 were subjected to unidirectional cyclic lateral loading whereas,
Specimen-4 was tested under monotonically increasing load after few initial cycles of
elastic loading.
All of the test specimens are modeled with average concrete compressive strength
of 3077 psi and maximum aggregate size of 1 inch. Yield strength of longitudinal and
transverse reinforcement are taken to be 59 and 69 ksi, respectively.
5.2.2 Material Constitutive Relationships
5.2.2.1 Concrete in Compression
In chapter 2, Setzler and Sezens approach was implemented with segmental
stress-strain relationship for concrete in compression derived from Mander et al. (1988)
confined concrete model till peak, and slope of descending branch of Roy & and Sozen
(1964) model after peak (Figure 2.9). In ASFI approach, concrete compressive behavior
was modeled by modified Kent and Park model, 1982 (Figure 4.2). Theoretically, any of
the available constitutive relationships can be used to model compressive concrete
behavior. In this comparison, Mander et al. (1988) model for confined and unconfined
concrete is employed for both analytical approaches. This model is presented below and
is illustrated in Figure 5.1.
According to Mander et al. (1988) model, compressive stress-strain relationship
for confined concrete is defined as
101
f cc r c cc
fc
r 1 c cc
r
f cc
cc co 1 5 1 (5.1)
f co
Ec f
r ; Ec 57, 000 f co ; Esec cc
Ec Esec cc
where f cc is the peak confined concrete strength, c is the concrete strain, cc is the
concrete strain at peak stress for confined concrete, co is the concrete strain at peak
stress in unconfined concrete (taken here as 0.002), f co is the concrete compressive
cylinder strength (in psi units), Ec is the modulus of elasticity of the concrete for normal
weight concrete, and Esec is the secant modulus of the concrete.
In this study, ultimate compressive strain cu for confined concrete is calculated
from Equation 5.2, which is obtained by slightly modifying the maximum strain formula
for spirally confined concrete in Priestley (1996).
cu 0.004 0.14 x y
f yy
(5.2)
f cc
dt2
x y (for square columns) (5.3)
2 sd c
where f yy is yield strength, d t is diameter and s is center-to center spacing of the
transverse reinforcement, respectively.
For unconfined cover concrete, compressive stress-strain relationship is defined
by following equations
102
f co r1 c co
fc for c 2 co
r1 1 c co 1
r
(5.4)
Ec f
r1 ; Ec 57000 f co ; Esec co
Ec Esec co
After reaching the strain of 2 co , cover concrete is assumed to start spalling and
part of falling branch in the region where c 2 co is assumed to be a straight line which
reaches zero stress at spalling strain sp , taken equal to 0.006 in this study (Figure 5.1).
In ASFI approach, in order to apply compression softening effect, compressive
behavior for confined and unconfined concrete obtained from Equations 5.1 through 5.4
must be multiplied by compression softening factor . The compression softening factor
was defined in Equation 3.2.
5.2.2.2 Concrete in Tension
Concrete behavior in tension is not taken into account in flexural analysis by
displacement component model by Setzler and Sezen (2005). However, in ASFI
approach, concrete tensile stresses are considered in both axial-flexure and axial-shear
models. The model by Vecchio and Collins (1986) defines concrete behavior in tension
for ASFI approach. This model will be used in this study and is presented in Section
3.4.1.2 and is illustrated in Figure 5.2.
5.2.2.3 Reinforcing steel
The reinforcing steel behavior in this study is modeled considering a linear elastic
behavior, a yield plateau, and a non-linear strain-hardening region, for both analytical
103
approaches. This model is defined in Section 2.5.2 and is illustrated in Figure 5.3. In
Setzler and Sezens model, same constitutive laws are used for steel behavior in tension
and compression. In ASFI approach, compressive steel behavior is not the same as tensile
steel behavior, and is modified to consider effect of compression bars buckling. After
reaching buckling strain, compressive steel stresses are reduced linearly according to
slope of post-peak confined concrete behavior.
5.2.3 Flexural Deformation Model
Fiber section analysis is conducted for determining moment-curvature
relationship for both of the analytical approaches. For ASFI approach, however, concrete
tensile stresses and compression softening factor are considered in the section analysis.
Flexural deformations are determined by plastic hinge model as defined in Section 4.3.3
and Figure 5.4. This model assumes a linear curvature distribution in elastic range and
inelastic curvatures are lumped at the column end over plastic hinge length L p . In
displacement component model, plastic hinge length is as one half of the section depth,
whereas ASFI approach considers this length equal to section depth. For comparison,
plastic hinge length is taken as one half of the section depth for calculating flexural
deformations by both approaches.
5.2.4 Pullout Deformation Model
Lateral deformations due to pullout or reinforcement slip are calculated by
employing Sezen and Setzler (2008) model. This model has was explained in detail in
Section 2.2.2 and Figures 2.3 and 2.4.

104
5.2.5 Buckling of Compression Bars
ASFI approach considers buckling of compression bars in flexural
analysis after unconfined cover concrete fiber next to compression steel reaches
compressive strength of about 30% of maximum concrete strength. Stresses in
compression bars are then linearly reduced according to slope of post-peak confined
concrete stiffness. In this study, compression steel stresses were reduced as per the model
given in Figure 6.3. According to this model, compression stresses start to decrease when
unconfined cover concrete start to spall. When this happens, corresponding strain in the
relevant steel layer can be calculated from flexural strain distribution across the cross
section depth. This strain is sp as shown in the figure. This point can fall anywhere on
typical stress-strain relationship for steel depending upon the level of flexural strain. Steel
stresses follow their usual constitutive stress-strain relationship until strain reaches this
limit. Then compression stresses in reinforcement follow new path defined by line
joining peak stress point to residual strength point having slope m, which is the same for
descending branch of concrete compression strength. This model is used for modeling
compression steel behavior for in ASFI approach in this study. Displacement component
model by Setzler and Sezen does not consider the bar buckling effect and employs and
same constitutive laws for steel behavior in tension and compression.
5.3 Comparison of Predicted and Experimental Response
Displacement component model and ASFI approach are implemented following
their respective analytical procedures for same set of test specimens with above-
105
mentioned material and deformation component models. The lateral load-displacement
relationships for component and total responses are presented and discussed as follows.
5.3.1 Flexural Displacements
Lateral load-flexural displacement relationships for Specimen-1 through -4 are
presented in Figures 5.5 through 5.8, respectively. For Specimen-1, both approaches
predict identical pre-peak response, which matches very well with the experimental data.
Peak load and deformation at peak load is also estimated very well by both approaches.
For post peak behavior, however, predicted responses are quite different. Post peak
response by ASFI approach closely follows stiffness of the actual response. The
estimated response by displacement component model initially follows post peak
stiffness and then becomes flat.
Lateral load-flexural displacement relationship for Specimen-2 is presented in
Figure 5.6. Initially, both approaches predict similar responses but after reaching the
strains corresponding to start of compression bars buckling, response by ASFI approach
drops gradually till peak load which is slightly underestimated. Predicted response by
displacement component model follows experimental pre peak response well but slightly
overestimates the peak load in positive direction. Again, post peak responses by both
approaches are diverging shortly after peak is reached. However post-peak stiffness of
ASFI approach somewhat matches experimental slope.
Figure 5.7 shows lateral load-flexural displacement relationships for Specimen-3.
Displacement component model overestimates initial stiffness and peak strength; whereas
ASFI approach does a good job in capturing initial response and peak strength. For
106
Specimen-4, ASFI approach predicts actual response that follows experimental pre and
post peak responses very well, as shown in Figure 5.8. With displacement component
model, response until peak load is estimated almost exactly and then diverges with gentle
slope from experimental behavior and response predicted by ASFI approach.
Comparison of the predicted and experimental responses for flexural
deformations indicates that displacement component model and ASFI approach predict
pre peak response almost identically. However, post peak responses differ significantly. It
should be recalled that lateral load-flexural displacement relationships are calculated by
employing common material constitutive laws, fiber section analysis procedures and
flexural deformation model, except that concrete behavior in tension, compression
softening effect and buckling of compression bars are considered in ASFI approach. It
can be seen from Figure 5.9, that difference in moment-curvature relationship with and
without considering concrete tensile behavior in section analysis is not very significant.
Compression softening factor in ASFI approach for all of the specimens modeled was
calculated to be 1.00 for almost all of the loading steps. It means that increasing shear
deformations did not affect flexural performance and both approaches employed same
uniaxial concrete compressive behavior in flexural analysis. The effect of compression
softening is illustrated in Figure 5.10. It can be seen that, this effect is not very significant
for the test specimens modeled in this chapter. If, however, compression-softening factor
has values less than 1.00, then its effect on moment-curvature relationship and softened
response can be significant.
This implies that the only difference in near-peak and post-peak responses
predicted by displacement component model and ASFI approach is due to considering
107
buckling of compression bars in ASFI approach. The effect of buckling of compression
bars is highlighted in Figure 5.1. It can be seen that the response is softened considerably
at peak and post peak stage after considering compression bars buckling effect.
After comparing responses estimated by both approaches with experimental data,
it can be concluded that ASFI approach does a relatively better job than displacement
component model in predicting pre peak response, peak load and post peak response in
flexure.
5.3.2 Reinforcement Slip Displacements
Lateral load-slip displacement relationships for Specimen-1 through -4 are
presented in Figure 5.12 through 5.15, respectively. After comparison of predicted
responses by displacement component model and ASFI approach and experimental data,
same conclusions are drawn as those of flexural deformations. Both approaches produce
almost identical response up until peak load and then diverge at post peak stage. Again,
this highlights the need for considering buckling of compression bars in the flexural
analysis, or another approach to better represent the post-peak response degradation.
5.3.3 Shear Displacements
Lateral load-shear displacement relationships for Specimen-1 through -4 are
presented in Figures 5.16 through 5.19, respectively. For Specimen-1, both approaches
predict identical initial response that matches initial stiffness very well. Near the peak,
predicted response by displacement component model follows the experimental response
slightly better than response by ASFI approach, which continues with same initial
108
stiffness until peak load. Peak load is captured very well by ASFI approach but the
displacement at the peak is underestimated. With displacement component model, peak
load is slightly overestimated and displacement at peak load is relatively closer to the
experimental value. Shear response by ASFI approach terminates at peak load and post
peak response is not captured, as it employs MCFT for estimating shear response. MCFT
being a force based approach has a limitation of predicting response till peak strength
only. The displacement component model does a fair good job in post peak response and
overestimates the displacements as shear strength degrades.
Figure 5.17 shows lateral load-shear displacement relationship for Specimen-2.
Predicted responses by both approaches are identical until observed peak and follows
experimental data perfectly. The peak load is predicted very well by ASFI approach and
overestimated by displacement component model in positive direction. But in the
negative direction, displacement component model captures shear strength very
accurately and ASFI approach underestimates it. This column experienced a brittle
flexural compression failure and did not show degrading shear behavior. Post peak
response by displacement component model showed the same trend. Both approaches
predict initial response for Specimen-3 reasonably that follows initial experimental
stiffness (Figure 5.18). Peak strength is significantly overestimated by displacement
component model, but post peak response captures slope of degrading branch very well.
Peak strength is accurately predicted by ASFI approach however, displacement at peak
load is underestimated. Similar trends are observed for response estimated for Specimen-
4 as shown in Figure 5.19. Initial strength is estimated fairly well by displacement
component model, but peak strength is slightly overestimated. The degrading portion
109
seems to follow the data well, although the deformations are overestimated. Response
estimated by ASFI approach also matches pre peak stiffness well. The peak strength is
estimated very well but the displacement at peak strength is underestimated.
As already described in Chapter 2, pre-peak shear response in displacement
component model is obtained indirectly from Response-2000. It must be noted that
Response-2000 is basically a section analysis routine that incorporates shear effects in the
analysis by implementing MCFT. In this approach, reinforced concrete cross-section is
considered a stack of biaxially loaded concrete elements and longitudinal steel elements.
Each of the concrete elements is analyzed individually for in-plane stress field based on
MCFT and overall section equilibrium conditions are satisfied. This is very rigorous
procedure and can only be implemented in a sophisticated computer program, such as
Response 2000. Nevertheless, it produces realistic response especially for shear critical
elements. In addition to producing total load-deformation response, Response-2000 also
delivers shear strain distribution along the length of the column for each load step. In
displacement component model, this shear strain distribution is copied manually and then
integrated over the length of the column to obtain shear deformation corresponding to
respective load step. As the analytical procedures adopted within Rsponse-2000 are very
deliberate and involved, a good correlation with experimental data is achieved for pre
peak shear response in displacement component model, as seen in the comparison of the
results.
On the other hand, ASFI approach also employs MCFT to estimate shear
deformations, but in a much simplified way than the analytical procedures adopted within
Response 2000. This is done by considering the column length between inflection point
110
and end section as a single shear element subjected to biaxial state of stresses, which are
assumed constant over the length of the element. The response of this element to average
shear and axial stresses is determined through MCFT satisfying compatibility and
equilibrium conditions described in Section 3.5. While evaluating shear response of the
column, average axial strain due to flexure is added to axial strain of axial-shear element.
This is a relatively simple procedure that can be implemented easily in hand calculations
and reasonably good response can be obtained in few steps. In Figures 5.16 through 5.19,
pre peak shear responses estimated by ASFI approach show satisfactory correlation with
experimental data.
Comparison of responses predicted by both approaches with test data shows that
pre peak response estimated by displacement component model follows experimental
curve slightly better than response estimated by ASFI approach. In view of the simplified
procedure employed within ASFI approach, minor loss of accuracy near the peak strength
is acceptable and does not affect overall response much.
As MCFT is a force based approach, shear response in ASFI approach terminates
at peak strength and does not show shear strength degradation with increasing shear
deformations. After reaching peak load, shear deformations in ASFI approach are
calculated from secant stiffness at peak strength, which is kept constant for post peak
behavior. In displacement component model, after estimating peak strength, shear
strength is assumed to remain constant at its peak value until onset of shear degradation.
From this point, the response degrades linearly to the point of axial load failure, where
strength is zero and drift is calculated from the procedure explained in Section 2.2.3.
111
After comparing experimental data with predicted responses, it can be concluded
that shear response envelope of displacement component model is a better choice for post
peak analysis over ASFI approach, which does not give any information on shear strength
degradation.
5.3.4 Total Response
Figure 5.20 shows the comparison of predicted and experimental lateral load-total
displacement relationships for Specimen-1. Shear strength of this specimen by Sezen-
Moehle equation (Equation 2.13) is calculated to be 69.0 kips and flexural strength from
moment-curvature analysis is 70.0 kips. As per displacement component model, the
column is classified as category-III specimen, for which total displacement at any point in
the response is sum of flexural, slip and shear displacement at that load step (Section 2.3).
The initial response is predicted very well up to the peak strength. Peak strength and
deformation at peak load are captured almost exactly. The post peak response initially
follow the experimental response and then becomes flat and deformations are over-
predicted. Response predicted by ASFI approach follows experimental data and response
predicted by displacement component model closely for most of the pre peak response.
The peak strength is predicted well but deformation at peak load and post peak
deformations are greatly underestimated.
Comparison of predicted and experimental lateral load-total displacement
relationships for Specimen-2 is presented in Figure 5.21. As per displacement component
model, this column has shear and flexural strengths of 92.0 and 72.0 kips, respectively,
thus classifying it into category-IV specimen. For this column, shear deformation is
112
frozen at its value at peak strength (flexural strength, 72.0 kips) and added to flexural and
slip displacements for post-peak response. Predicted response by this approach, slightly
overestimates pre peak stiffness and peak load in positive direction and follows post peak
experimental response fairly in both directions. Predicted response by ASFI approach
also slightly overestimates pre peak response in positive direction and underestimates
peak strength in both directions. The deformations at peak load are, however captured
well. Post peak response in negative side follows observed response better than positive
direction. For specimen-3, lateral load-displacement relationships are shown in Figure
5.22. pre peak response predicted by both approaches is identical and overestimates pre
peak stiffness of the experimental behavior. Peak strength is slightly overestimated by
displacement component model and, is underestimated by ASFI approach. Displacement
at peak load is however, well captured by both methods. Post peak response by
displacement component model matches slope of the observed response, but ASFI
approach significantly underestimates post peak response.
Figure 5.23 shows comparisons of predicted and experimental lateral load-total
displacement relationships for Specimen-4. This column has identical to Specimen-1
except that it was tested under monotonically increasing lateral load after few initial
elastic cycles. Comparison of shear and flexural strength classifies this column into
category-III column as per displacement component model. The predicted response by
displacement component model follow the trend in experimental data but slightly
overestimates initial stiffness, peak strength and post peak response. Predicted response
by ASFI approach also overestimates initial stiffness and captures peak strength well.
113
Deformation at peak strength and post peak deformations are significantly
underestimated.
Comparison of the predicted responses by displacement component model and
ASFI approach with each other and experimental test data shows very similar trends for
all of the modeled test specimens. Both approaches produce identical pre peak responses
that generally follow experimental data well. For post peak response, displacement
component model performs relatively better than ASFI approach and deformations are
slightly overestimated. Whereas, predicted post peak deformations by ASFI approach are
significantly underestimated.
It must be noted that the rules for combining deformation components for
response up to the peak are same in both approaches and flexural, bar slip and shear
deformations are simply added together to obtain total response. That is why predicted
pre peak responses by both of the approaches are identical for all test specimens as shown
above. For post peak analysis, the rules for combining deformation components are
however, quite different in both approaches. In displacement component model, post
peak deformations are obtained by combining flexural, bar slip and shear deformations
depending upon predicted failure mode according to set of rules explained in Section 2.3
and Figures 2.7 and 2.8. In ASFI approach, post peak response is calculated with constant
pullout and shear stiffnesses at the values corresponding to peak load and only axial-
flexural model continues until axial load failure. Reinforcement slip and shear
deformations are calculated from their respective constant secant stiffnesses, and are
added to flexural deformations at any load level in post peak state. Estimation of total
response in this way forces slip and shear deformations to decrease with degrading lateral
114
load capacity of the column. This is the reason that total post peak response by ASFI
approach is underestimated for test specimens modeled in this chapter.
In ASFI approach, plastic hinge length in flexural deformation model is taken
equal to section depth while in displacement component model, it is taken as half of the
section depth. In order to compare response estimations by both approaches, same plastic
hinge length was employed and flexural deformations in this chapter were calculated with
plastic hinge length of one-half of the section depth. Figure 5.24 shows a comparison of
lateral load-displacement relationships by ASFI approach for Specimen-1, calculated
with plastic hinge length taken equal to section depth and half of the section depth. It can
clearly be seen that post peak response improves significantly when total deformations
are calculated with plastic hinge length equal to section depth. On the other hand, if
lateral load-flexural displacement relationships for the same specimen is analyzed (Figure
5.25), it is seen that post peak flexural response calculated with plastic hinge length of
one-half of section depth gives very good match with experimental data. This implies that
response estimation in ASFI approach by taking plastic hinge length equal to section
depth does predict a more accurate response and flexural deformations be calculated with
plastic hinge length equal to half of the section depth, as it conforms to behavior. Also, it
is recommended that an effective way to incorporate for combination of deformation
components in ASFI approach must be considered for total response in post peak state.
115
5.4 Conclusion
In this chapter, responses estimated by displacement component model by Setzler
and Sezen (2005) and ASFI approach by Mostafaei and Kabeyasawa (2007) are presented
by modeling the behavior of same set of test specimens. A detailed discussion of the
results is carried out to get an insight into analytical procedures of both approaches.
Conclusions are drawn after comparison of predicted responses for each of the flexural,
slip, shear and total deformations with respective experimental data. Conclusions drawn
in this chapter will form the basis for a model proposed in the next chapter.
116
3500
Confined concrete
Stress (psi) 2500
2000
1500
1000
500
0
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016
Strain
Figure 5.1: Constitutive relationships concrete in compression
250
200
150
Stress (psi)
100
50
0
0 0.2 0.4 0.6 0.8 1
Strain x 10
-3
Figure 5.2: Constitutive relationship concrete in tension
117
100
90
80
70
60
Stress (ksi)
50
40
30
20
10
0
0 0.025 0.05 0.075 0.1 0.125 0.15
Strain
y
x x
Lp L - Lp L
y y
Figure 5.4: Plastic Hinge Model for Determining Flexural Deformation
118
80
60
40
Lateral load (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
ASFI approach
-80
-3 -2 -1 0 1 2 3
Flexural displacement (in.)
Figure 5.5: Lateral load flexural displacement relationship for Specimen-1
80
60
40
20
Lateral force (kip)
-20
-40
-60
Test data
-80 Setzler and Sezen
ASFI approach
-100
-1.5 -1 -0.5 0 0.5 1 1.5
119
80
60
40
Lateral force (kip)
20
-20
Test data
-40
Setzler and Sezen
ASFI approach
-60
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
80
60
40
Lateral force (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
ASFI approach
-80
-3 -2 -1 0 1 2 3 4
120
4500
4000
3500
3000
2500
M (k-in.)
2000
1500
1000 Specimen-1 (with concrete tension)

Specimen-1 (without concrete tension)
500 Specimen-2 (with concrete tension)
Specimen-2 (without concrete tension)
0
0 0.5 1 1.5 2 2.5 3
Curvature (1/in.) x 10
-3
Figure 5.9: Effect of concrete tensile stress in flexural section analysis
4000
3500
3000
2500
Moment (k-in.)
2000
1500
1000
Specimen-1 (Setzler and Sezen)
Specimen-1 (ASFI approach)
500
Specimen-2 (Setzler and Sezen)
Specimen-2 (ASFI approach)
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
-3
Figure 5.10: Effect of concrete compression softening in flexural section analysis
121
4500
4000
3500
Moment (k-in.) 3000
2500
2000
1500
1000
500 Specimen-1 (with bar buckling)

Specimen-1 (without bar buckling)
0 Specimen-2 (with bar buckling)
Specimen-2 (without bar buckling)
-500
0 0.5 1 1.5 2 2.5 3
-3
Figure 5.11: Effect of compression bars buckling in flexural section analysis
80
60
40
Lateral load (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
ASFI appoach
-80
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
Slip displacement (in.)
Figure 5.12: Lateral load slip displacement relationship for Specimen-1
122
80
60
40
20
Lateral force (kip)
-20
-40
-60
Test data
ASFI approach
-100
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
80
60
40
Lateral force (kip)
20
-20
Test data
-40
Setzler and Sezen
ASFI approach
-60
-1 -0.5 0 0.5 1 1.5 2
123
80
60
40
Lateral force (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
ASFI approach
-80
-2 -1 0 1 2 3 4
80
Test data
60 Setzler and Sezen
ASFI approach
40
Lateral load (kip)
20
-20
-40
-60
-80
-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2
Shear displacement (in.)
Figure 5.16: Lateral load shear displacement relationship for Specimen-1
124
100
Test data
80
Setzler and Sezen
60 ASFI approach
Lateral force (kip) 40
20
-20
-40
-60
-80
-100
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.5
100
80
60
Lateral force (kip)
40
20
-20
Test data
-40
Setzler and Sezen
ASFI approach
-60
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8
125
80
60
40
Lateral force (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
ASFI approach
-80
-1.5 -1 -0.5 0 0.5 1 1.5
80
60
40
Lateral load (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
ASFI approach
-80
-6 -4 -2 0 2 4 6
Displacement (in.)
Figure 5.20: Lateral load displacement relationship for Specimen-1
126
80
60
40
20
Lateral force (kip)
-20
-40
-60
Test data
ASFI approach
-100
-5 -4 -3 -2 -1 0 1 2 3 4 5
Displacement (in.)
80
60
40
Lateral force (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
ASFI approach
-80
-4 -3 -2 -1 0 1 2 3 4 5
Displacement (in.)
127
80
60
40
-20
-40
Test data
-60
Setzler and Sezen
ASFI approach
-80
-6 -4 -2 0 2 4 6 8
Displacement (in.)

80
60
40
Lateral load (kip)
20
-20
-40
Test data
-60
Response with Lp=h/2
Response with Lp=h
-80
-6 -4 -2 0 2 4 6
Displacement (in.)
Figure 5.24: Effect of plastic hinge length on total response by ASFI approach
(Specimen-1)
128
80
60
40
Lateral load (kip)
20
-20
-40
Test data
-60
Response with Lp=h/2
Response with Lp=h
-80
-5 -4 -3 -2 -1 0 1 2 3 4 5
Figure 5.25: Effect of plastic hinge length on flexural response by ASFI approach
(Specimen-1)
129
CHAPTER 6
DEVELOPMENT AND IMPLEMENTATION OF THE PROPOSED MODEL
6.1 Introduction
In Chapter 5, analytical procedures of displacement component model and ASFI
approach were investigated and predicted responses were compared with experimental
test data. Based upon the observations made in the last chapter, a new analytical
procedure is proposed, which is expected to perform better in predicting lateral load-
displacement response of reinforced concrete columns subjected to lateral loads. The
proposed procedure retains features from both of the understudy approaches that help
improve response estimation and make analytical procedure relatively simple and easy to
implement.
In the proposed procedure, flexural and shear deformations are calculated
independently while considering interaction between these mechanisms, and then
combined together depending upon dominant failure mode to obtain total response. The
interaction between axial-flexure and axial-shear mechanisms in terms of compression
softening and axial deformation allows for accurate response estimation while decoupled
flexural analysis minimizes iterations within the analysis and make the process relatively
simple and easy. In addition, buckling of compression bars is also incorporated by
130
considering separate stress-strain relationships for reinforcing steel in tension and
compression.
This chapter presents the development of the proposed analytical procedure by
explaining the details of the component deformation models, interaction methodology
and rules for total response estimation. In order to consider the possibility of compression
bars to buckling under high compressive strains, available steel buckling models are
evaluated and an appropriate model is recommended for use in the proposed model.
6.2 Desired Characteristics of the Proposed Analytical Procedure
It must be acknowledged that in a fixed-ended reinforced concrete column
subjected to lateral loading, flexure and shear mechanisms interact with each other and
corresponding deformations do not occur independently. Previous studies established that
principal compressive stress in the concrete is not only the function of principal
compressive strain but also gets affected by the coexisting principal tensile strain in an
inverse proportion (Vecchio and Collins, 1986). It implies that the concrete subjected to
combined normal compressive and shear stresses is weaker in compression than the
concrete subjected to normal compressive stresses only. In the web of a laterally loaded
reinforced concrete column, the concrete is also subjected to shear stress in addition to
the normal stresses due to axial load and flexure. Due to applied shear stress, the concrete
cracks diagonally and becomes weaker and softer in axial compression. Similarly,
flexural deformations also influence shear behavior of the column. Axial deformation due
to flexure increase width of the shear crack on tension face and hence lower shear
capacity of the column.
131
Any analytical model that aims to capture total response of an element, such as
beam or column, must consider interaction between axial, flexure and shear mechanisms.
In conventional flexure section analysis, the effect of shear stress on degrading
compressive strength of the concrete is ignored and concrete behavior is simulated by its
response in uniaxial compression. The influence of shear mechanism on flexural behavior
can be taken into account by considering compressive stress-strain relationships of
diagonally cracked concrete in flexural section analysis instead of employing constitutive
relationship for concrete in uniaxial compression. Likewise, the effect of flexural
deformations on shear mechanism can also be accounted for by adding axial deformation
component of flexural mechanism to the axial strain of the axial-shear mechanism.
ASFI approach considers interaction between axial, flexure and shear mechanisms
as explained earlier in Chapters 3 and 4. The interaction procedure of the approach is re-
summarized here for easy reference. The ASFI approach employs cracked concrete
behavior in flexural analysis to acknowledge the effect of shear deformations on flexural
performance. This is done by determining compression softening factor from axial-shear
model (through MCFT analysis of shear element) and employing it into axial-flexure
model (through flexural section analysis) to lower uniaxial compressive stress of the
concrete. Similarly, in order to consider the effect of flexural deformation on shear
behavior, ASFI approach incorporates average centroidal axial strain from axial-flexure
model into axial-shear model. While these interactions are taken into consideration,
equilibrium of axial and shear stresses from axial-flexure and axial-shear models are
satisfied for each incremental load step.
132
In previous chapter, the comparison between predicted responses by ASFI
approach and experimental test data was presented. For flexural deformations, predicted
responses (Figures 5.8 through 5.11) show a good correlation with experimental data
indicating that ASFI procedure of considering interaction of compression concrete
softening is effective in capturing effect of shear deformations on flexural performance.
Similarly, predicted shear deformations up to peak load compares well with the observed
shear responses (Figures 5.16 through 5.19), indicating that procedure of axial strain
interaction is also successful in capturing effect of flexural deformations on shear
response. Therefore, it can be concluded that ASFI methodology is effective in capturing
interaction between flexural and shear deformations and must be retained for the
proposed analytical procedure.
It is important to highlight here that identical flexural responses up to peak were
predicted by displacement component model in previous chapter even without
considering interaction between flexural and shear mechanisms. As already explained
that effect of shear deformation on flexural performance is taken into account by
employing compression softening factor from shear model to flexural analysis. For the
columns dominated by flexure or flexure-shear behavior with low shear stresses, it is
most likely that shear strains will not effect flexural deformations. The test specimens
modeled in the Chapter 5 exhibited flexure dominant response up to peak load. The
compression softening factor by ASFI approach was found to be 1.00, eliminating the
need to soften concrete compressive response in flexure. As a result, constitutive law for
concrete in uniaxial compression is employed in ASFI approach, leading to obtain
identical flexural responses by both approaches. It must, however, be noted that if
133
compression softening factor for any column during the analysis by ASFI approach is
found less than 1.00, a significantly different flexural deformations shall be predicted by
ASFI approach and displacement component model.
It must also be noted that the interaction terms in ASFI approach couple axial-
flexure and axial-shear models with each other in such a way that the analysis for any of
the two cannot be performed independently. For example, interaction of shear
deformation with flexural behavior requires determination of compression softening
factor from shear mechanism, coupling axial-shear model with axial-flexure. Similarly,
axial-shear analysis cannot be performed unless axial deformations due to flexure are
determined from axial-flexure mechanism. Thus, the analyses for both mechanisms must
be carried out simultaneously at each loading step. Due to coupling or interlocking of the
mechanisms and requirement on satisfying equilibrium of shear stresses in overall
analysis, the analytical process in ASFI approach is complicated and computationally
intense and solution technique requires a deliberate iteration process.
On the other hand, flexural and shear deformations in displacement component
model are determined independently in a relatively simple analytical procedure. Despite
having limitation on interaction, decoupled analyses methodology of displacement
component model offers the advantage of conducting simple calculations that are easy to
implement. Hence, any analytical process that aims to consider interaction between
flexure and shear can be simplified significantly if analyses for both mechanisms can be
performed independently in decoupled fashion. In view of the need to consider
interaction between flexure and shear deformations in a simplified analytical procedure,
an effort is made in this study to propose a suitable analytical model that performs
134
decoupled flexural and shear analyses offering easy implementation while still
considering interaction between them.
6.3 Proposed Interaction Methodology
As describe above, the proposed procedure considers the interaction between
flexure and shear mechanisms in a manner that flexural section analysis can be performed
independent of shear analysis. Axial strain and shear stress determined from flexure
analysis can then be incorporated to axial-shear model to determine shear deformations.
The proposed interaction methodology is presented as follows.
6.3.1 Interaction of Concrete Compression Softening
One of the objectives of the proposed procedure is to conduct decoupled flexure
and shear analyses while considering interaction between them. If the effect of shear
deformations on flexural performance can be assessed without performing full axial-shear
analysis of the element, it will decouple axial-flexure analysis from axial-shear
mechanism. This will allow flexural deformations to be determined independently
without having the need to perform shear analysis prior to flexure analysis while still
incorporating the effect of shear deformations on flexural performance. Thus, knowing
axial strain due to flexure from flexure analysis, shear analysis can be carried out
considering the interaction of axial deformations. This will make overall analytical
process less complicated and will result in comparatively simple calculations. The effect
of shear deformations on flexural performance is incorporated by lowering concrete
stresses in uniaxial compression by compression softening factor. Hence, compression
135
softening factor must be determined in a simplified way without performing full shear
analysis to represent degradation in concrete strength due to applied shear stresses.
In order to simplify the computation intensive analytical procedure of ASFI
approach, Mostafaei and Vecchio (2008) formulated a new model called Uniaxial Shear
Flexure Model (USFM) by eliminating complex iteration process of the shear model
employed within ASFI approach. In the new approach, single concrete stress block is
employed to represent concrete stress distribution across the cross section depth instead
of fiber section approach. In addition, the axial strain and principal tensile strain of the
element are determined based on average centroidal axial strains and average concrete
strains corresponding to resultant compressive forces of the compression blocks of two
end sections. This simplifies the analytical process significantly and eliminates the
iteration process required for shear analysis in original ASFI approach. The complete
details on formulation, implementation and verification of USFM approach can be found
in Mostafaei and Vecchio (2008).
It must be recalled that the compression softening factor, as defined by Equation
3.2, is a function of concrete principal tensile strain c1 of the element being analyzed. As
USFM approach employs simplified procedure to calculate concrete principal tensile
strain, it is adopted in the proposed procedure with minor modifications. The approach is
based on few fundamental equations of MCFT and two simplifying assumptions on
concrete strains from flexural section analysis and does not require a deliberate shear
analysis (MCFT) of the element. The procedure to determine principal tensile strain and
subsequently compression softening factor is explained below for a fixed ended
reinforced concrete column subjected to in-plane lateral load (Figure 6.1).

136
As already described in Section 3.3.1, compression softening factor , is defined
as
1
1.0 (6.1)
c1
0.8 0.34
co
where c1 is principal tensile strain and co is concrete strain corresponding to the
maximum concrete cylinder strength. For an element considered between inflection point
and one of the end sections of the column, c1 can be determined according to MCFT
equation as follows
c1 x y 2 (6.2)
In above equation, x is average axial strain at the centroid for the element and is
obtained by averaging the values of centroidal axial stain at one of the end section o and
axial strain of the inflection point xa as
x
o xa (6.3)
2
Likewise 2 is concrete principal compressive strain for the element. Its value is
assumed to be the average of the concrete compressive strain corresponding to resultant
compressive force of the stress block at end section c and axial strain at the inflection
point xa . Hence,
2
c xa (6.4)
2
137
The other unknown quantity in Equation 6.2 is strain of the transverse
reinforcement y . Its value is determined from following relationship determined based
on MCFT as
y b2 c b (6.5)
where,
f c1
b 2
2 y Esy 2
c
x 2 fc1 fcx fc1 2
y Esy
f cx f x x f sx
where y is transverse reinforcement ratio, Esy is modulus of elasticity of transverse
reinforcement, f cx is concrete stress in x -direction, f x is applied axial stress, x is
longitudinal reinforcement ratio, f sx is longitudinal steel stress obtained from section
analysis based on average centroidal strain, fc1 0.145 fco is concrete principal tensile
stress, f co is concrete cylinder strength, x is normal strain at the centroid determined
from Equation 6.3 and 2 is concrete principal compressive strains determined from
Equations 6.4.
After calculating concrete principal tensile strain c1 from Equation 6.2,
compression softening factor is determined with the help of Equation 6.1 for a given
curvature. The concrete compressive stresses in flexural section analysis are then lowered
by multiplying concrete stresses given by employed constitutive law with compression
softening factor. The softening of the concrete compressive stress-strain relationship
138
explained in Section 3.3.1 and Figure 3.1. In proposed procedure, compression softening
factor determined with this method is employed till peak flexural load and then a constant
value, corresponding to peak load, is used in post peak analysis.
6.3.2 Interaction of Axial Strains
In order to consider the effect of flexural deformations on shear behavior, the
proposed procedure incorporates axial strain and shear stress due to flexure into in-plane
analysis of the shear element. The proposed procedure is based on the axial strain
interaction methodology of ASFI approach and equilibrium of shear stresses in flexural
and shear mechanisms. In this procedure, interaction of axial strain is taken into account
by adding flexibility component of axial deformation due to flexure to the corresponding
flexibility component of axial-shear model. By employing flexural shear stress to in-
plane stress-strain relationship of the shear element, shear deformations are determined.
For a fixed ended column subjected to lateral load, the procedure for axial deformation
interaction and determination of shear strain is explained as follows.
The length of the column between inflection point and one of the end sections is
considered as a shear element subjected to constant normal stress due to applied axial
load and average shear stresses due to applied lateral load. Performing flexural analysis
on fiber model of the end section and axial model at the inflection point, average axial
strain due to flexure at the centroid of the section for the element can be determined with
following equation. These analyses are carried out incorporating compression softening
factor determined as per the procedure explained above.
139
xf
o xa (6.6)
2
where o and xa are centroidal and axial strain at the column end section and inflection
point, respectively (Figure 6.1). The flexibility component for axial deformation due to
flexure f xf can be determined as
xf
f xf (6.7)
x
where, x is applied axial stress in longitudinal direction of the column and can be
determined by dividing applied axial load P by the area of the cross section
P
x (6.8)
bd
In above equation, b and d are width and depth of the cross section,
respectively. A stress-strain relationship in terms of flexibility matrix for an in plane
shear element (axial-shear model) can be defined as,
f11 f12 f13 x x

f 21 f 22 f 23 y y (6.9)
f
31 f 32 f 33 xy xy

where fij i, j 1, 2,3 are flexibility components of in plane shear model, x is normal
applied stresses in longitudinal direction, y is normal stress in transverse directions, xy
is shear stress, x is normal strain in axial direction, y is normal strain in transverse
direction, and xy is shear strain. In the above equation, longitudinal axis of the shear
element is aligned with horizontal x -axis.
140
Axial strain due to flexure xf can be taken into account in the axial-shear model
by adding flexibility component obtained from Equation 6.7 into Equation 6.9.
f11 f xf f12 f13 x x xf

f 21 f 22 f 23 y y (6.10)

f 31 f 32 f 33 xy xy

In above equation, stresses in transverse direction (clamping stresses) are zero due
to inexistence of lateral external force along the column, i.e., y 0 . In addition, the
applied shear stress xy of the element is taken from flexural section analysis as
M
xy (6.11)
Linbd
where M is the moment obtained from fiber section analysis of the end section and Lin
is length of the element, taken equal to half of the total length for fixed ended column. In
Equation 6.10, knowing the applied stresses, corresponding strains can be calculated. The
flexibility matrix is obtained by inverting material stiffness matrix of the shear element
formulated using secant stiffness methodology as explained in Section 3.4.3.
6.4 Buckling of Compression Bars
The comparison of the responses for flexural deformation in previous chapter
(Figures 5.8 through 5.11) highlighted the need to consider reduction in stresses of the
compression bars due to possibility of their buckling, especially in post peak stage. The
post peak flexural response by ASFI approach compared well with the experimental
behavior for the modeled test specimens, indicating that bar buckling model employed is
effective in capturing the column behavior in flexure. The details of this model were
141
given in Section 5.2.5. In addition to the bar buckling model employed for response
estimation by ASFI approach in previous chapter, two other models are explored to see
their applicability and effectiveness in the analysis. Based upon the analysis results, one
of the models is recommended to be used for response estimation by proposed procedure.
The details on the bar buckling phenomenon, related aspects and proposed bar buckling
model are explained in the following sections.
Reinforced concrete columns in seismic zones are subjected to combined action
mainly due to axial, bending moment and shear forces. Reinforcing bars may experience
inelastic axial compression under severe loading and exhibit lateral deformation known
as buckling. The behavior in the compressive face of a concrete member at overload
depends on a variety of factors such as, size and shape of the cross-section, the amount of
longitudinal compression steel, the amount of transverse reinforcement providing
confinement to the section, thickness of the cover concrete, and stress-strain properties
for the steel and concrete (Potger et al. 2001). The tendency for the compressively loaded
steel bars to buckle and deflect outwards is initially resisted by the lateral restraint
provided by the surrounding cover concrete as well as the transverse steel ties or stirrups.
As the compressive loads increase and approach the section capacity, the concrete
surrounding the compressive bars carries large longitudinal compressive stress, and
eventually becomes prone to longitudinal cracking, and spalling. After the cover concrete
spalls off, stirrups restrain lateral movement and buckling.
The effect of buckling on the load-deformation behavior of reinforcing bars has
been studied under monotonic and cyclic loads (Monti and Nuti (1992), Rodriguez et al.
(1999), Suda et al. (1996), Gomes and Appleton (1997), and Dhakal and Maekawa
142
(2002)). These researchers developed steel constitutive relationships to take into account
the buckling of longitudinal reinforcement, especially focusing on post-buckling load-
deformation behavior.
In this study, two of the available bar buckling models are evaluated to assess
their applicability, accuracy and ease of implementation in the analysis. Flexural
deformations shall be determined by employing these models for the four specimens
tested by Sezen (2002). After comparing predicted responses with test data, appropriate
model shall be employed for analysis.
6.4.1 Dhakal and Maekawa Bar Buckling Model
Dhakal and Maekawa (2002) developed a model to study buckling mechanism of
the reinforcing bars by finite element microanalysis using fiber technique. Their model is
illustrated in Figure 6.2. According to the model, an intermediate point * , * is
established, after which a constant negative stiffness equal to 0.02 Es is assumed until the
average steel stress becomes equal to the residual value of 0.2 f y . The intermediate point
is defined with the help of following set of equations
fy L
* 5.5 2.3 y 7 y
100 db

(6.12)
f L *
* 1.1 0.016 y f s 0.2 f y
100 d
b
where f y is yield stress of longitudinal steel in MPa, L is length of the column which is
taken equal to stirrup spacing, d b is longitudinal bar diameter, y is yield strain of
143
longitudinal bars, f s* is point wise stress on original curve corresponding to * , and is
a factor related to strain hardening region of steel stress-strain relationship. Its value is
1.00 for linear hardening bars and 0.75 for perfectly elastoplastic bars. In modeling test
specimens with non-linearly hardening bars, this factor was assumed 1.00 for
implementing this model in this study. After determining the intermediate point, stress-
strain relationships are then defined as follows
* s y
f sl 1 1 * * f s for y s *

f s y
(6.13)

f sl * 0.02 Es s * 0.2 f y for s *
where f sl is lowered compression steel stress corresponding to steel strain s , f s is point
wise stress corresponding to s , and E s is modulus of elasticity of reinforcing bar.
6.4.2 Potger et al. (2001) Bar Buckling Model
Potger et al. (2001) developed a simplified bar buckling model by using a
predetermined buckling strain, softening slope and stress degradation after modifying
some of the existing models (Kato et al. (1973), Meng et al. (1992), Nakatsuka et al.
(1999), Inoue and Shimizu (1988), Yamada et al. (1993)). The model is illustrated in
Figure 6.3. In this model, buckling and lateral displacement of the bar is assumed to
begin at a critical buckling strain lb indicated by point A in Figure 6.3. After buckling,
the compressive strain in the bar continues to increase and the stress is assumed to fall-off
progressively with a slope of lb Es . Softening continues until a second limiting strain ps
144
is reached at point B. With further increases in compressive strain (past point B), the
stress in the bar is assumed to soften at a less severe rate of 0.005Es .
The critical buckling strain is calculated as follows by the relationships developed
by Nakatsuka et al. (1999). This relationship takes into account the effects on buckling
sh
strain due to lateral reinforcement spacing to confined core diameter ratio , confining
dc
stress, yield strength of lateral reinforcement f yy , shape of reinforcement (circular,
rectangular), and the compressive strength of plain concrete f co .
lb c0 fco 103 f1 f2 f3 f 4 f5 (6.14)
where,
3.6 4.8( sh d c ) 0.1 ( sh d c ) 0.75

f1
0 ( sh d c ) 0.75
f 2 y f yy
2
1.0 for bar in circular column

f3 0.9 for corner bar (6.15)
0.18
for intermediate bar
110
-1 30 MPa f c 0 110 MPa
f4 fc0
0 f c 0 110 MPa

600
f5 0.5 102 f yy 400 MPa
f yy

where co is maximum strain corresponding to peak strength of the plain concrete f co
(MPa), sh is spacing of the transverse reinforcement, d c is smallest side length of
concrete cross section surrounded by lateral reinforcement, y is transverse
reinforcement ratio, and f yy is yield strength of the transverse reinforcement.

145
The post-buckling slope is calculated with following relationship developed by
Inoue and Shimizu (1988). The second post-buckling slope is set at 0.005Es adopted
from the report by Yamada et al. (1993) for the buckling of steel plates.
1
lb 100 yx 1
1 500
2
(6.16)
sh

ir
where yx is yield strain of longitudinal bars, is 1.0 for corner bars and 0.5 for
intermediate bars, and ir is radius of gyration of longitudinal bar.
6.4.3 Implementation of Bar Buckling Models
Both of the above models and bar buckling model proposed in the previous
chapter are incorporated in flexural section analysis and flexural deformations are
calculated for Specimen-1, 2 and 4. The predicted lateral load-flexural displacement
relationships are compared with experimental responses in Figures 6.4 through 6.6.
6.4.4 Proposed Model for Buckling of Compression Bars
Based on the comparison of the predicted and observed responses, the model
explained in Section 5.2.5 is recommended to be used for lowering the compression
stresses of the longitudinal reinforcement to account for bar buckling phenomenon. This
model is also explained Figure 6.7.
It must be noted that diameter of the longitudinal bar and spacing of the transverse
reinforcement are important parameters that affect buckling of the compression bars
146
(Monti and Nuti 1992). Smaller diameter bars contained by widely spaced stirrups are
most likely to undergo lateral deformations and buckling much earlier during loading
history than larger diameter bars confined by closely spaced transverse reinforcement.
Therefore, in the proposed model to account for this fact, for stirrup spacing to bar
diameter ratio sh db of less than 5.00, no buckling is considered and compressive
behavior of the reinforcement is similar to tensile behavior. For sh db ratio above 11.00,
the bars are considered to buckle as soon as reinforcement yields. For sh db ratios
between 5.00 and 11.00, post-buckling softening is considered soon after yielding by
employing the proposed model.
6.5 Flexural Analysis
Conventional fiber section analysis is performed to determine moment-curvature
relationship required to calculate lateral load-flexural displacement response. However,
concrete behavior in tension, compression softening factor determined as per procedure
described in Section 6.3.1 and buckling of compression bars as per procedure explained
in Section 6.4.4 are incorporated in the flexural analysis. Flexural section analysis also
enables to determine axial strain due to flexure xf and shear stress of the axial-flexure
model xy required for shear analysis of the element.
Lateral displacements due to flexure are calculated by plastic hinge model with
the help of following equations. This model is already explained in Section 4.3.3 and
Figure 5.4.
147
1
f a2 for y
3
(6.17)
Lp
f y a 2 y L p a
1
for y u
3 2
where is curvature at column end, y is curvature at yield point, a is shear span equal
to length L of a cantilever column and half the length for a fixed ended column. The
plastic hinge length L p is taken as one-half of the section depth.
The lateral load-flexural displacement relationships for Specimens 1, 2, 3 and 4
are calculated using above mentioned procedure and are presented in Figures 6.8 through
6.11, respectively.
6.6 Reinforcement Slip Deformation Model
In Chapter 5, lateral load-slip displacement relationships for modeled test
specimens were obtained by employing Sezen and Setzler (2008) bar slip model. The
predicted responses showed satisfactory correlation with experimental data (Figure 5.12
through 5.15). Hence, this model is adopted to calculate bar slip deformations in the
proposed procedure. This model is explained in detail in Section 2.2.2 and Figures 2.3,
2.4. The lateral load-slip displacement relationships for Specimens 1, 2, 3 and 4 are
calculated using above mentioned procedure and are presented in Figures 6.12 through
6.15, respectively.
6.7 Shear Deformations
It was shown in the previous chapter that shear response envelope of displacement
component model performs well in predicting shear behavior of the test specimens
148
(Figure 5.16 through 5.19). For pre-peak response, this shear model relies on shear strain
output from Response-2000 computer program. The user is required to input section
configuration, material properties and loading conditions to the program and run full
member analysis. The analysis results in overall lateral load-deformation response up to
the peak load the cross section can sustain. At each load step, the program gives average
shear strain distribution along the length of the column. In order to obtain lateral load-
shear displacement response, the user is required to manually copy this shear strain
distribution at each load step and integrate it over the length of the column to get
corresponding shear displacement. This procedure is repeated for all load steps up to the
peak load, which makes the process very laborious and time consuming. Additionally, as
the load-displacement relationships for shear and flexure/bar slip mechanisms are
calculated separately at different load steps, these responses are required to be
interpolated to a common set of load steps and corresponding displacements to make
their addition possible for total response.
On the other hand, axial-shear model of ASFI approach adopts relatively simple
approach for shear response estimation with reasonably good accuracy. This procedure
enables the interaction of axial strains to capture effect of flexural deformation on shear
behavior. Nevertheless, ASFI approach has a serious limitation of predicting shear
response only up to the peak load and does not predict shear strength degradation with
increasing deformations. This aspect is addressed of by shear response envelope of the
displacement component model that is able to predict post peak behavior up to the point
of axial load failure.
149
The proposed shear displacement model in this study retains the features from
ASFI approach and displacement component model that help estimate response
accurately and make analytical process simple and relatively easy to implement. In the
proposed shear model, shear displacements up to the peak shear load are calculated by in-
plane analysis of the shear element based on direct application of MCFT while adding
axial strain due to flexure to the total axial strain of the shear element. The post-peak
shear behavior is obtained by employing post peak shear response envelope of the
displacement component model.
The procedure for application of MCFT was explained in Section 4.2.3. These
steps along with axial strain interaction methodology explained in Section 6.3.2 are
followed to obtain flexibility matrix in Equation 6.10. This equation is then solved with
applied normal stress due to axial load (Equation 6.8) and shear stress from axial-flexure
model (Equation 6.11) to obtain shear strains up to peak shear load. The shear strength is
then assumed to remain constant at its peak value until the point of shear strength
degradation. From this point, the response degrades linearly to the point of axial load
failure, where strength is zero. The shear failure point and axial load failure point are
calculated according to the procedure explained in Section 2.2.3 and 2.3.
The term peak shear load used above refers to point where pre-peak shear
response by MCFT terminates. This corresponds to the load step at which either shear
failure occurs or peak flexural strength reaches before shear failure. Hence, for the
columns failing in shear prior to experiencing flexural strength, peak shear load will be
the shear strength of the specimen. For the columns having higher shear strength than the
flexural strength, this point will give shear response at peak flexural strength.
150
As the shear analysis in the proposed procedure is performed at the same load
steps employed for flexure and bar slip deformations, lateral load-shear displacement
relationships are not required to be tailored to match the loading intervals of the
flexure/slip load-displacement relationships. This allows for simple addition of all
deformation components for post peak response. The proposed shear model also
eliminates the dependency of displacement component model on Response-2000 and
gives control to the user on choice of material constitutive models and parameters to be
used other than the ones employed in Response-2000.
The lateral load-shear displacement relationships for Specimens 1, 2, 3 and 4 are
calculated using proposed shear procedure and are presented in Figures 6.16 through
6.19, respectively.
6.8 Total/Combined Response
The total lateral response of a reinforced concrete column to the lateral load is
modeled as a set of springs in series (Figure 2.6). The lateral displacement components of
flexure, bar slip and shear are each modeled by a spring. Each spring is subjected to the
same force and total displacement is sum of response of each spring.
The analysis in the proposed procedure is carried out by incrementing load in
steps of appropriate size starting with a small initial value. First, flexural and bar slip
deformations are calculated and then shear response is evaluated at each load step. If at
any load step, shear failure does not occur, the value of shear strain is taken as shear
response at current load level and the shear analysis for the next step is carried out. This
procedure is repeated until either shear failure occurs or peak flexural strength is
151
achieved. The load corresponding to this stage is defined as peak strength. After reaching
the peak, the mechanism limiting the peak strength (flexure or shear) will dominate the
behavior.
6.8.1 Pre-Peak Response
In the proposed procedure, total response up to the peak strength is obtained by
simply adding deformation components due to flexure, bar slip and shear mechanism
determined as per the individual component models explained above.
6.8.2 Post-Peak Response
The rules for combining deformation components for post peak responses in the
proposed procedure are taken from displacement component model. The procedure for
classification of the columns into categories is same as before except that the
determination of shear strength of the specimens is slightly different. It was shown in the
previous chapter that the procedure employed for post peak response in displacement
component model is much realistic and performs better than ASFI approach (Figures 5.20
through 5.23).
The classification of the columns into categories is based on the comparison of
their shear, flexural and yield strength (Setzler, 2005). The yield strength Vy of the
column is defined as the lateral load corresponding to the first yielding of the tension bars
in flexural analysis. The flexural strength V p is the lateral load corresponding to the
peak moment sustainable by the column. The shear strength Vn for the columns failing
152
in shear prior to the reaching flexural strength or failing close to flexural strength is
determined from the proposed shear model in Section 6.7, in which peak shear load is
shear strength of the column. For other columns where peak shear load by proposed shear
model is equal to the flexural strength, Sezen and Moehle (2002) equation is employed to
determine shear strength Vn of the column. The determination of the shear strength
with this procedure minimizes the dependency of the component displacement model on
alternate mean of determining shear strength for classification purpose. The procedure for
combining deformations due to flexure, bar slip and shear are explained in Section 2.3
and are followed as such for post peak behavior.
6.9 Analytical Procedure and Solution Technique for Proposed Procedure
This section summarizes the procedure for estimating the response of laterally
loaded reinforced concrete columns by proposed procedure.
1. Define or input material properties and geometry. Decide on a sign convention for
tensile and compressive stresses and strains.
2. Input axial load P . If there is no applied axial load, then consider a negligible
small value for P .
3. Consider an axial-flexure element of reinforced concrete column from its
inflection point to one of the end section.
4. Select a small value of the curvature , such as 0.00000001 , as a starting
value.
5. Determine an estimate of compression softening factor
153
i. Assume location of the neutral axis c for the first iteration. An initial
h
value of c is good starting point, where h is overall section depth.
2
ii. Apply section analysis procedure at end section and inflection point and
determine centroidal axial stain at end section o , strain corresponding to
resultant compressive force of the stress block at end section c , and axial
strain at the inflection point xa . Refer to Figure 6.1 for better
understanding.
iii. Determine average centroidal strain x and average concrete principal
compression strain 2 for the axial-flexure element using Equations 6.3
and 6.4, respectively.
iv. Determine the average concrete principal tensile strain c1 using
Equations 6.5 and 6.2. The average value of concrete principal tensile
stress fc1 0.145 fco can be used.
v. Calculate compression softening factor using Equation 6.1. Determine
concrete compression stress of the stress block with the help of employed
constitutive law and multiply it with to represent cracked concrete
behavior.
vi. Check for the section equilibrium at the end section. If equilibrium of
axial forces is not achieved, modify the assumed value for depth of the
neutral axis c and repeat steps ii through vi.
154
6. Evaluate axial-flexure mechanism by performing flexural section analysis at the
end section and inflection point.
i. Divide column cross section into appropriate number of concrete fibers.
Apply standard fiber section analysis procedure and obtain strain
distribution across the section depth.
ii. Employ appropriate constitutive laws for compressive behavior of
confined core concrete and unconfined cover concrete, and determine
stress distribution corresponding to the imposed strain profile. Effect of
confinement due to transverse reinforcement on compressive strength and
ductility of the core concrete must be taken into account with the help of
appropriate confinement model.
ii. Multiply compressive stresses of the concrete by compression softening
factor obtained in step 5 above to consider effect of shear deformations
on flexural performance.
iv. Employ appropriate constitutive law to consider concrete behavior in
tension in section analysis.
v. Employ appropriate constitutive law for reinforcing bars and determine
steel stresses corresponding to the section strain at each layer. For
considering effect of compression bars buckling, employ proposed bar
buckling model explained in Section 6.4.4.
vi. Determine forces in each concrete fiber and reinforcing bars by
multiplying stresses by corresponding areas and check for section
155
equilibrium. When section equilibrium is achieved, calculate nominal
moment capacity M and obtain axial centroidal strain o .
vii. Determine flexural-shear stress xy by Equation 6.11.
viii. Perform axial section analysis at the inflection point and determine axial
strain due to applied axial load only xa considering same material
constitutive laws employed for flexural section analysis of the end section.
Calculate axial strain due to flexure xf with Equation 6.6.
ix. Calculate lateral displacements due to flexure f by plastic hinge model
with the help of Equation 6.17. Plastic hinge length is taken equal to one
half of the section depth.
x. Calculate reinforcement slip displacement pull by procedure explained in
Section 6.6. Lateral displacement due to slip is added to flexural and shear
displacements to obtain total response.
xi. Determine x and f xf from Equations 6.8 and 6.7, respectively.
xii. Determine lateral load V corresponding to the nominal moment capacity
M by following equation
M
V (6.18)
Lin
xii. Increment curvature to next load step and repeat all steps until either
concrete crushes or steel ruptures. After reaching peak flexural load Vp ,
defined as load corresponding to the maximum moment capacity of the
156
section, use constant compression softening factor corresponding to its
value at peak load.
7. Evaluate axial-shear mechanism by proposed procedure employing MCFT and
interaction of axial strains for the column element between end section and inflection
point. The procedure for MCFT is explained in Section 3.4.
i. Start with the first load step from axial-flexural analysis explained above.
Consider the variables of iteration as xi , yi , si and assume some very
small values for each of them for the first iteration. These variables are
average normal strain in x -direction ( xi ), average normal strain in y -
direction ( yi ) and average shear strain ( si ) of axial-shear element,
respectively.
ii. Determine average concrete principal tensile and compressive strains
c1 and c 2 by Equations 3.5 and 3.6, respectively. Also, determine
orientation of principal tensile plane t and principal compressive plane
strains c with the help of Equations 3.7 through 3.9.
iii. Calculate compression-softening factor with the help of Equation 6.1
using c1 calculated in above step.
iv. Determine average steel stresses f sx and f sy corresponding to the
strains x and y in accordance with employed constitutive laws for
reinforcement in x and y -directions respectively.
157
v. Determine average principal tensile and compressive stresses in concrete
by considering same material constitutive laws employed in axial-flexural
model (step 6). However, as in axial-flexure model cover and core
concrete are modeled separately, concrete compressive strength and
corresponding strain must be modified according to the Equation 3.42.
vi. Check local stress-strain conditions at cracks ensuring that the tension in
the concrete can be transmitted across the crack and shear stress on the
surface of the crack does not exceed maximum shear provided by the
aggregate interlock by Equations 3.18 through 3.24. If Equation 3.20 is
not satisfied and local shear stress at the surface of the crack exceeds shear
strength provided by aggregate interlock mechanism, then average
concrete principal tensile stress must be reduced as per Equation 3.25.
vii. Determine secant moduli for concrete and reinforcements and assemble
material stiffness matrices using Equation 3.27 through 3.30. Determine
total material stiffness D matrix by transforming component material
stiffnesses into global x and y -directions using Equation 3.26 and 2.31.
viii. Invert total stiffness matrix D to get flexibility matrix f of Equation 6.9
and add flexibility component of axial deformation due to flexure f xf to
f11 component of f to determine flexibility matrix of coupled axial-
shear-flexure element in Equation 6.10.
158
ix. Consider x from Equation 6.8, y 0 , and flexural-shear stress xy from
step 6(vii) and solve Equation 6.10 for xi 1 , yi 1 , and si 1 for the
integration point.
x. Assess the convergence of the acquired variables xi 1 , yi 1 , and si 1 . If
satisfactory convergence is achieved, note the value of si as shear
response at current load level and go to the next load step. Otherwise,
repeat above steps in shear analysis with improved estimate of the
iteration variables. For the next iteration, average of the initial assumed
values and calculated values in last step can be used for fast convergence.
xi. The above mentioned shear analysis steps are repeated for all load steps
until either shear failure is observed or load step corresponding to peak
flexural load Vp is reached. The load corresponding to this stage is defined
as peak shear load.
x. After the peak shear load is reached, shear strength is then assumed to
remain constant at its peak value until the point of shear strength
degradation. From this point, the response degrades linearly to the point of
axial load failure, where strength is zero. The shear failure point and axial
load failure point are calculated according to the procedure explained in
Section 2.2.3 and 2.3.
8. Obtain combined/total response with the help of procedure explained in Section
6.8.
159
6.10 Implementation of the Model and Comparison of the Results
The proposed procedure is implemented to estimate component and total
responses for Specimen-1 through 4. The load-displacement relationships are calculated
by employing same material constitutive laws as explained in the Chapter 5. The pre-
peak flexural and bar slip deformations are generally predicted well (Figures 6.8 through
6.11). The pre-peak shear stiffnesses and peak shear strengths are also captured well.
However, for total response estimation (Figure 6.20 through 6.23), it can be seen that
displacement component model, ASFI approach and proposed procedure, predict almost
identical responses before peak load. After the peak is reached, response predictions by
these models are quite different. ASFI approach generally underestimates post peak
responses for the test specimens. This is mainly due to shear and bar slip displacements
are calculated from constant secant stiffnesses, which follow straight-line path after the
peak and become zero at axial load failure.
The displacement component model generally overestimates the post peak
responses. When the same columns were modeled by this approach in Chapter 2, very
good correlation was obtained between predicted and observed responses. In Chapter 2,
these columns were analyzed by employing constitutive relationship for compressive
concrete that had steep descending slope. This was done to acknowledge the poor
confinement due to widely spaced ties. In this analysis, the concrete compressive
behavior is simulated by Mander et al. (1988) model for both loading and unloading
branches. The constitutive models employed in Chapter 2 and this study are presented in
Figure 6.24.
160
6.11 Conclusion
A procedure is proposed for response estimation of reinforced concrete columns
subjected to lateral loads. The procedure is based on the observations made in the
previous Chapter 5 after evaluating and analyzing displacement component model and
ASFI approach. The proposed procedure determines flexure, bar slip and shear
deformations individually considering interaction between these mechanisms. The axial-
flexure mechanism is decoupled from axial-shear model, allowing relatively simpler
analytical procedure. The flexure model in proposed procedure incorporates concrete
tensile behavior, interaction of compression softening and buckling of compression bars
into the flexural analysis. The shear model evaluates the shear response of the column
while considering effect of flexural deformation on shear behavior. The pre-peak
response is evaluated by employing MCFT, and post peak shear response envelope from
displacement component model is adopted for predicting post peak shear behavior. All
deformation components i.e., flexural, bar slip and shear are added together to get total
response of the column. The total/combined peak response is limited by lesser of the
shear and flexural strength of the column and limiting mechanism governs the post peak
response. The proposed procedure employs relatively simple calculations for overall
response estimation and appears to perform better than displacement component model
and ASFI approach in predicting the component and total response.
161
Concrete stress block
at end section
P NA
End section V o
c
Centroidal
Lin Axis
Inflection Point
xa
x 2
Assumed distribution
h
of x and 2 over
Concrete stress block length of the element
at inflection point
Fixed-ended column
subjected to lateral load
b
h
Column cross section
Figure 6.1: Determination of concrete principal tensile strain
fs
*
s , f s*
fy
,
* *
0.02 Es
0.2 f y
s
y
Figure 6.2: Dhakal and Maekawa (2002) bar buckling model

162
fs
f s ,lb A
f s , ps
s
y
Figure 6.3: Potger et al. (2001) bar buckling model
80
60
40
Lateral force (kip)
20
-20
-40
Test data
Dhakal et al.
-60
Potger et al.
Proposed model
-80
-3 -2 -1 0 1 2 3
Figure 6.4: Effect of compression bar buckling Specimen 1
163
80
60
40
20
Lateral force (kip)
-20
-40
-60 Test data

Dhakal et al.
-80 Potger et al.
Proposed model
-100
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
80
60
40
Lateral force (kip)
20
-20
-40
Test data
Dhakal et al.
-60
Potger et al.
Proposed model
-80
-3 -2 -1 0 1 2 3 4

164
Figure 6.7: Proposed bar buckling model
80
60
40
Lateral load (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
Proposed model
-80
-3 -2 -1 0 1 2 3
165
80
60
40
20
Lateral force (kip)
-20
-40
-60
Test data
Proposed model
-100
-1.5 -1 -0.5 0 0.5 1 1.5
80
60
40
Lateral force (kip)
20
-20
Test data
-40
Setzler and Sezen
Proposed model
-60
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
166
80
60
40
Lateral force (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
Proposed model
-80
-3 -2 -1 0 1 2 3 4
80
60
40
Lateral load (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
Proposed model
-80
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2

167
80
60
40
20
Lateral force (kip)
-20
-40
-60
Test data
Proposed model
-100
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
80
60
40
Lateral force (kip)
20
-20
Test data
-40
Setzler and Sezen
Proposed model
-60
-1 -0.5 0 0.5 1 1.5 2
168
80
60
40
Lateral force (kip)
20
-20
-40
Test data
-60
Setzler and Sezen
Proposed model
-80
-2 -1 0 1 2 3 4
80
Test data
ASFI approach
40 Propose model
Lateral load (kip)
20
-20
-40
-60
-80
-5 -4 -3 -2 -1 0 1 2
169
100
Test data
60 ASFI approach
Proposed model
20
-20
-40
-60
-80
-100
-1 -0.75 -0.5 -0.25 0 0.25 0.5
100
80
60
40
Lateral force (kip)
20
-20
-40 Test data

Setzler and Sezen
-60 ASFI approach
Proposed model
-80
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8
170
80
60
20
-20
-40
Test data
Setzler and Sezen
-60
ASFI approach
Proposed model
-80
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
80
60
40
Lateral load (kip)
20
-20
-40
Test data
Setzler and Sezen
-60
ASFI approach
Proposed model
-80
-6 -4 -2 0 2 4 6
Displacement (in.)

171
80
60
40
20
Lateral force (kip)
-20
-40
-60 Test data

Setzler and Sezen
-80 ASFI approach
Proposed model
-100
-5 -4.15 -3.3 -2.45 -1.6 -0.75 0.1 0.95 1.8 2.65 3.5 4.35 5
Displacement (in.)
80
60
40
Lateral force (kip)
20
-20
-40
Test data
Setzler and Sezen
-60
ASFI approach
Proposed model
-80
-4 -3 -2 -1 0 1 2 3 4 5
Displacement (in.)
172
80
60
40
Lateral force (kip)
20
-20
-40
Test data
Setzler and Sezen
-60
ASFI approach
Proposed model
-80
-6 -4 -2 0 2 4 6 8
Displacement (in.)
4000
Mander confined concrete model
3500 Mander unconfined concrete model
Sezen confined concrete model
3000
Sezen unconfined concrete model
2500
Stress (psi)
2000
1500
1000
500
0
0 0.005 0.01 0.015 0.02
Srain
Figure 6.24: Comparison between compressive concrete constitutive laws
173
CHAPTER 7
CONCLUSION
7.1 Summary
The focus of this research was to suggest a suitable analytical procedure that can
accurately predict the monotonic lateral force-displacement relationship for reinforced
concrete columns by overcoming limitations of available models. The columns found in
many existing buildings not designed according to modern seismic design standards,
often have limited deformation capacity and are vulnerable to brittle shear failure during
earthquakes. Observed damage pattern after past earthquakes has shown that columns,
especially those in the first story, are often the most critical members in non-ductile
concrete buildings, and failure of columns can lead to structure collapse. The model
proposed here is intended for use in the analysis of existing buildings to determine the
required level of retrofit necessary for satisfactory seismic performance. The proposed
procedure can also be used for the properly designed existing and future planned
buildings for their performance evaluation.
Lateral deformation of a column is composed of three deformation components
due to flexure, reinforcement slip and shear mechanisms. Theses deformations do not
occur independently and axial, flexure and shear mechanisms interact with each other.
174
For example, shear deformations tend to lower the compressive strength of the concrete
in the web of the column that affect flexural behavior. Likewise, flexural deformations
increase shear crack width and lower shear capacity of the column. In the proposed
procedure, flexural and shear deformations are determined while considering these
effects. The flexural deformations are calculated separately by employing cracked
concrete behavior to account for shear deformations on flexural performance. After axial
strains and shear stresses are determined from flexural analysis, shear behavior is then
evaluated. The shear response estimation in the proposed procedure employs MCFT and
post-peak shear response envelope from displacement component model. Deformations
due to bar slip are calculated separately and added to flexural and shear displacements to
obtain total response. The rules for combining deformation components for total response
are the same as employed by displacement component model.
7.1.1 Component Models
The flexural behavior in the proposed procedure was evaluated by fiber section
analysis techniques. However, in order to incorporate the effect of shear deformation on
flexural behavior, cracked concrete behavior was considered in the section analysis. This
was done by determining compression-softening factor from axial-shear mechanism and
employing it in flexural section analysis to lower compressive stress-strain relationship
for the concrete in uniaxial compression. The compression-softening factor was
determined as per the suggested procedure that decouples flexural analysis from shear
mechanism, thus allowing flexural deformations to be determined independently and
easily. The concrete behavior in tension was also incorporated in the flexural analysis. In
175
order to consider the phenomenon of compression bar buckling, few of the bar buckling
models were analyzed and a procedure to incorporate this effect was suggested. After
knowing moment-curvature relationship with above considerations, flexural displacement
were calculated with plastic hinge model by taking plastic hinge length equal to one-half
of the section depth. Comparison of the predicted flexural response with experimental
data showed that the proposed procedure performs well in estimating flexural
displacements.
The shear model in the proposed procedure employs MCFT and considers effect
of flexural deformations on shear strength. This was done by extracting axial strain due to
flexure only from flexural section analysis as per the procedures employed in ASFI
approach and incorporating it to the shear analysis. The normal and shear stress obtained
from axial-flexure analysis are employed as applied load for in-plane shear analysis of
the element. The response up to the peak load is determined with this procedure and then
post peak shear response envelope from displacement component model was employed to
determine points of shear and axial load failures. The shear model in the proposed
procedure eliminated the dependency of the displacement component model on
Response-2000 computer program for obtaining shear strain distribution. The comparison
of the predicted responses with experimental data indicated that axial strain interaction
methodology, shear response estimation by MCFT and post peak shear response envelope
were successful in capturing observed shear behavior.
In addition, lateral displacements due to reinforcement slip were calculated
separately by Setzler and Sezen (2008) bar slip model and were added to flexural and
shear displacements at each load step.
176
7.1.2 Combined Model
The overall lateral deformation of a reinforced concrete column was modeled as
three springs in series, one for each of the deformation components. The shear strength of
the column failing at or prior to reaching flexural strength was calculated using the shear
model of the proposed procedure. For the columns having higher shear strengths than the
flexural strength, shear strength was determined using shear strength equation proposed
by Sezen and Moehle (2004), taking the displacement ductility parameter k as 1.0. This
shear strength was compared to the yield and flexural strengths determined from the
flexural analysis to classify columns into one of five categories. The same rules employed
by displacement component model were used for classification of the columns into
categories. Peak load was defined as lesser of the load predicted by shear model and
flexure model. Up to the peak strength, the three deformation components were simply
added together to obtain total response. After the peak strength was reached, post peak
response was governed by dominant failure mechanism and rules for combining post
peak deformations from displacement component model were employed for obtaining
total response. The comparison of the predicted and observed responses indicated that
proposed procedure performs well in predicting overall response of the column.
7.2 Conclusions
The interaction between flexure and shear deformations must be taken into
account to determine total response of the column to lateral load. The recommended
interaction methodology for considering effect of shear deformation on flexural
177
performance and procedure to calculate compression softening factor appeared to work
well in this research. Decoupled flexural analysis allows calculating flexural
deformations accurately and independently while still considering the interaction with
shear mechanism.
Buckling of compression bars affect post peak behavior and hence must be
considered in the analysis. The procedure adopted in this research for incorporating this
phenomenon worked well and is recommended to be used for the analysis.
Moment-curvature analysis employing fiber model approach while considering
concrete behavior in tension, compression-softening factor, and buckling of compression
bars, and the plastic hinge method of calculating flexural deformations worked well in
this research. These tools appear to be appropriate for use in both well reinforced and
lightly reinforced concrete columns.
The reinforcement slip model by Setzler and Sezen (2008) which utilizes a bi-
uniform bond stress over the embedded length of the bar, represented the experimental
slip behavior well and can be used to model the response with least complexity and good
accuracy.
The shear behavior of columns was predicted well by the shear model employed
in the proposed procedure. The Modified Compression Field Theory proved to be an
effective method for predicting shear displacement, up to the maximum strength of a
member. The procedure to analyze shear behavior of the column while considering
interaction of axial strains and applied shear stresses from flexural model resulted in
fairly accurate response estimation and was simple to implement. Only little experimental
shear displacement data was available. More would be required to determine the
178
effectiveness of the model for a wider range of column types. The post-peak shear
response envelope is essentially required to complete the full load-deformation response.
The classification system and rules to combine deformation component to obtain
total response by displacement component model are very effective in capturing observed
behaviors. Theses rules can be used for wide range of the columns failing in various
modes.
The proposed analytical procedure is relatively easy to implement and employs
simple calculations to estimate component and total deformations. The procedure can be
used to accurately estimate monotonic lateral load-displacement relationships for the
columns exhibiting different failure modes such as flexural failure with very limited or no
shear effects, flexure and/or shear failure following the flexural yielding, and brittle shear
failure prior to flexural yielding etc.
7.3 Recommendations for Future Work
The research reported here presents a model for the monotonic lateral deformation
of lightly reinforced concrete columns subjected to lateral loads. However, there remains
a large amount of work to be completed before the behavior of non-ductile concrete
structures in earthquakes can be fully modeled and understood. Several recommendations
for continued research on this topic follow.
The proposed model was used to predict the behavior of few test columns. While
general good agreement was shown between the test data and model, further comparisons
should be made to complete validation of the model over a wide range of column
properties.
179
Several areas warrant more experimental research to improve the knowledge base.
Particularly, more testing of reinforced concrete columns to axial load failure is
necessary. Most test data available stops before loss of axial capacity. More data would
help improve models for shear capacity and axial capacity; both of these are necessary
components of the proposed model. Additionally, more test specimens should be
instrumented to monitor the individual displacement components, particularly shear
displacements. This would aid in improving the component models, which should lead to
improvements in the combined model as well.
The most significant area of research that remains is the modeling of the
hysteretic behavior of columns. Many hysteretic models for flexural deformations exist,
which should be applicable. Several shear models are available as well, however, these
would need to be evaluated in the context of lightly reinforced columns to determine their
applicability. These hysteretic models would need to be combined through a set of rules
to predict the overall cyclic lateral behavior of lightly reinforced concrete columns.
180
LIST OF REFERENCES
Aboutaha, R. S., Engelhardt, M. D., Jirsa, J. O., and Kreger, M. E., 1999. Rehabilitation
of Shear Critical Concrete Columns by Use of Rectangular Steel Jackets, ACI
Structural Journal. V. 96, No. 1, Jan. 1999. pp. 68-78.
ACI Committee 318, 2002. Building Code Requirements for Structural Concrete and
Commentary, ACI 318-02/318R-02. American Concrete Institute, Farmington
Hills, Mich., 2002. 443 pp.
ACI Committee 408, 1979. Suggested Development, Splice, and Standard Hook
Provisions for Deformed Bars in Tension, ACI 408.1R-79. American Concrete
Institute, Farmington Hills, Mich., 1979. 3 pp.
Alsiwat, J. M. and Saatcioglu, M., 1992. Reinforcement Anchorage Slip under

Monotonic Loading, Journal of Structural Engineering. ASCE, V. 118, No. 9,
Sept. 1992. pp. 2421-2438.
Bentz, E., 2000a. Response-2000. http://www.ecf.utoronto.ca/~bentz/home.shtml.
Bentz, E., 2000b. Sectional Analysis of Reinforced Concrete Members, Ph.D.

Dissertation. University of Toronto, 2000. 310 pp.
Berry, M., Parrish, M., and Eberhard, M., 2004. PEER Structural Performance Database
Users Manual (Version 1.0). Pacific Earthquake Engineering Research Center,
University of California, Berkeley, Jan. 2004. 38 pp. http://nisee.berkeley.edu/
spd/performance_database_manual_1-0.pdf.
Chadwell, C. B. and Imbsen & Associates, 2004. XTRACT Cross Section Analysis
Software for Structural and Earthquake Engineering. http://www.imbsen.com/
xtract.html.
Chadwell, C. B. and Imbsen, R. A., 2004. XTRACT: A Tool for Axial Force-Ultimate
Curvature Interactions, Building on the Past, Securing the Future: Proceedings
of the 2004 Structures Congress. ASCE, Nashville, Tenn., May 2004. 9 pp.
181
Cho, J.-Y. and Pincheira, J. A., 2004. Nonlinear Modeling of RC Columns with Short
Lap Splices, Proceedings of the 13th World Conference on Earthquake
Engineering. Vancouver, B.C., Canada, Aug. 2004. Paper No. 1506.
Ciampi, V., Eligehausen, R., Bertero, V. V., and Popov, E. P., 1982. Analytical Model
for Concrete Anchorages of Reinforcing Bars Under Generalized Excitations,
Report No. UCB/EERC 82/23. Earthquake Engineering Research Center,
University of California, Berkeley, Nov. 1982. 111 pp.
Collins, M. P., and Mitchell, D., Shear and Torsion Design of Prestressed and Non-
Prestressed Concrete Beams, Journal of the Prestressed Concrete Institute, V.
25, No. 5, 1980, pp. 32-100.
Dhakal, P.R., and Maekawa K., Modeling for Postyield Buckling of Reinforcement
,Journal of Structural Engineering. ASCE, V. 129, No. 9, Sept. 2002 pp. 1139-
1147.
Eberhard, M., 2003. Structural Performance Database. Pacific Earthquake

Engineering Research Center, University of Washington, Seattle.
http://nisee.berkeley.edu/spd. Accessed 1 Nov. 2005.
Eligehausen, R., Popov, E. P., and Bertero, V. V., 1983. Local Bond Stress-Slip
Relationship of a Deformed Bar Under Generalized Excitations, Report No.
UCB/EERC 83/23. Earthquake Engineering Research Center, University of
California, Berkeley, Oct. 1983. 169 pp.
Elwood, K. J., 2002. Shake Table Tests and Analytical Studies on the Gravity Load
Collapse of Reinforced Concrete Frames, Ph.D. Dissertation. University of
California, Berkeley, 2002. 419 pp.
Elwood, K. J., 2004. Modeling failures in existing reinforced concrete columns,

Canadian Journal of Civil Engineering. NRC Canada, Vol. 31, No. 5, Oct. 2004.
pp. 846-859.
Elwood, K. J. and Moehle, J. P., 2005a. Axial Capacity Model for Shear-Damaged
Columns, ACI Structural Journal. V. 102, No. 4, July 2005. pp. 578-587.
Elwood, K. J. and Moehle, J. P., 2005b. Drift Capacity of Reinforced Concrete

Columns with Light Transverse Reinforcement, Earthquake Spectra. EERI, V.
21, No. 1, Feb. 2005. pp. 71-89.
Filippou, F. C., Popov, E. P., and Bertero, V. V., 1983. Effects of Bond Deterioration
on Hysteretic Behavior of Reinforced Concrete Joints, Report No. UCB/EERC
83/19. Earthquake Engineering Research Center, University of California,
Berkeley, Aug. 1983. 184 pp.
182
Gerin, M. and Adebar, P., 2004. Accounting for Shear in Seismic Analysis of Concrete
Structures, Proceedings of the 13th World Conference on Earthquake
Engineering. Vancouver, B.C., Canada, Aug. 2004. Paper No. 1747.
Gomes, A. and Appleton, J., (1997), Non-linear cyclic stress-strain relationship of

reinforcing bars including buckling, Engineering Structures, Vol. 19(10) 822-826.
Harajli, M. H., 2004. Comparison of Bond Strength of Steel Bars in Normal- and High-
Strength Concrete, Journal of Materials in Civil Engineering. ASCE, V. 16,
No. 4, Aug. 2004. pp. 365-374.
Hawkins, N. M., Lin, I. J., and Jeang, F. L., 1982. Local Bond Strength of Concrete for
Cyclic Reversed Loadings, Bond in Concrete. Applied Science Publishers,
London, 1982. pp. 151-161.
Inel, M. and Aschheim, M. A., 2002. Models for Estimating the Deformation Capacities
of Reinforced Concrete Columns, ECAS2002 International Symposium on
Structural and Earthquake Engineering. Middle East Technical University,
Ankara, Turkey, Oct. 2002.
Kent, D. C. and Park, R., 1971. Flexural Members with Confined Concrete, Journal of
the Structural Division. ASCE, V. 97, No. ST7, July 1971. pp. 1969-1990.
Lee, D. H., Choi, E., and Zi, G., 2005. Evaluation of earthquake deformation and
performance for RC bridge piers, Engineering Structures. Elsevier, V. 27, No.
10, Aug. 2005. pp. 1451-1464.
Lee, D. H. and Elnashai, A. S., 2001. Seismic Analysis of RC Bridge Columns with
Flexure-Shear Interaction, Journal of Structural Engineering. ASCE, V. 127,
No. 5, May 2001. pp. 546-553.
Lehman, D. E. and Moehle, J. P., 2000. Seismic Performance of Well-Confined

Concrete Bridge Columns, Report No. PEER-1998/01. Pacific Earthquake
Engineering Research Center, University of California, Berkeley, Dec. 2000. 316
pp.
Lin, I. J. and Hawkins, N. M., 1982. Anchorage Characteristics for Reinforcing Bars for
Reversed Cyclic Loading, Report No. SM 82-1. Department of Civil
Engineering, University of Washington, Seattle, June 1982. 326 pp.
Lynn, A. C., 2001. Seismic Evaluation of Existing Reinforced Concrete Building

Columns, Ph.D. Dissertation. University of California, Berkeley, 2001. 308 pp.
183
Lynn, A. C., Moehle, J. P., Mahin, S. A., and Holmes, W. T., 1996. Seismic evaluation
of existing reinforced concrete building columns, Earthquake Spectra. EERI, V.
12, No. 4, Nov. 1996. pp. 715-739.
Mander, J. B., Priestley, J. N., and Park, R., 1988. Theoretical Stress-Strain Model for
Confined Concrete, Journal of Structural Engineering. ASCE, V. 114, No. 8,
Aug. 1988. pp. 1804-1825.
Marquez, J. L. G. and Jirsa, J. O., 1975. A Study of Hooked Bar Anchorages in Beam-
Column Joints, ACI Journal. V. 72, No. 5, May 1975. pp. 198-209.
The MathWorks, Inc., 2005. Matlab. http://www.mathworks.com/products/matlab.
McKenna, F., Fenves, G. L., and Scott, M. H., 2004. Open System for Earthquake
Engineering Simulation. Pacific Earthquake Engineering Research Center,
University of California, Berkeley. http://opensees.berkeley.edu.
Melek, M., Wallace, J. W., and Conte, J. P., 2003. Experimental Assessment of
Columns with Short Lap Splices Subjected to Cyclic Loads, Report No. PEER
2003/04. Pacific Earthquake Engineering Research Center, University of
California, Berkeley, Apr. 2003. 176 pp.
Mo, Y. L. and Wang, S. J., 2000. Seismic Behavior of RC Columns with Various Tie
Configurations, Journal of Structural Engineering, ASCE, V. 126, No. 10, Oct.
2000. pp. 1122-1130.
Moehle, J. P., 1992. Displacement-Based Design of RC Structures Subjected to

Earthquakes, Earthquake Spectra. EERI, V. 8, No. 3, Aug. 1992. pp. 403-428.
Moehle, J. P., Elwood, K. J., and Sezen, H., 2002. Gravity Load Collapse of Building
Frames During Earthquakes, S. M. Uzumeri Symposium: Behavior and Design of
Concrete Structures for Seismic Performance, SP-197. American Concrete
Institute, Farmington Hills, Mich., 2002. pp. 215-238.
Monti, G. & Nuti, C., (1992), Non-linear Cyclic Behaviour of Reinforcing Bars Including
Buckling,Journal of Structural Engineering, ASCE, Vol. 118(12), 3269-3284.
Mostafaei, H., and Kabeyasawa, T., A Simple Approach for Displacement-Based

Assessment of RC Columns, The First NEES/Defense Workshop on Collapse
Simulation of Reinforced Concrete Building Structures, University of California
Berkeley, Berkeley, Calif., July 2005, pp. 47-63.
Mostafaei, H., Axial-Shear-Flexure Interaction Approach for Displacement-Based

Evaluation of Reinforced Concrete Elements. PhD dissertation, Faculty of
184
Engineering, Architrave Department, University of Tokyo, Tokyo, Japan, 2006,
255.
Mostafaei, H., and Kabeyasawa,T., Axial-Shear-Flexure Interaction Approach for

Reinforced Concrete Columns. ACI Structural Journal, vol. 104, No.2, Mar-Apr
2007, pp. 218-226.
Mostafaei, H., and Vecchio, F. J. (2008). Uniaxial Shear-Flexure Model for Reinforced
Concrete Elements. ASCE Journal of Structural Engineering, 134(9), 1538-
1547.
Mostafaei, H., Vecchio, F.J., and Kabeyasawa T. (2009). Deformation Capacity of

Reinforced Concrete Columns. ACI Structural Journal, 106(2), 187-195.
Morita, S. and Kaku, T., 1984. Slippage of Reinforcement in Beam-Column Joint of

Reinforced Concrete Frame, Proceedings of the Eighth World Conference on
Earthquake Engineering. San Francisco, July 1984. pp. 477-484.
National Information Service for Earthquake Engineering (NISEE), 2005. The

Earthquake Engineering Online Archive. Earthquake Engineering Research
Center, University of California, Berkeley. http://nisee.berkeley.edu/elibrary.
Accessed 25 Oct. 2005.
Ousalem, H.; Kabeyasawa, T.; Tasai, A.; and Iwamoto, J., Effect of Hysteretic
Reversals on Lateral and Axial Capacities of Reinforced Concrete Columns,
Proceedings of the Japan Concrete Institute, V. 25, No. 2, 2003, pp. 367-372.
Okamura, H., and Maekawa, K., Nonlinear Analysis and Constitutive Models of
Reinforced Concrete, Gihodo Shuppan Co., Ltd., 1991, 182 pp.
Otani, S. and Sozen, M. A., 1972. Behavior of Multistory Reinforced Concrete Frames
during Earthquakes, Structural Research Series No. 392. University of Illinois,
Urbana, 1972. 551 pp.
Ozcebe, G. and Saatcioglu, M., 1989. Hysteretic Shear Model for Reinforced Concrete
Members, Journal of Structural Engineering. ASCE, V. 115, No. 1, Jan. 1989.
pp. 132-148.
Park, R. and Paulay, T., 1975. Reinforced Concrete Structures. John Wiley & Sons,
Inc., New York, 1975.
Park, R.; Priestley, M. J. N.; and Gill, W. D., Ductility of Square Confined Concrete
Columns, Journal of Structural Division, ASCE, V. 108,No. 4, 1982, pp. 929-
950.
185
Patwardhan, C., 2005. Strength and Deformation Modeling of Reinforced Concrete
Columns, M.S. Thesis. The Ohio State University, Columbus, Ohio, 2005. 166
pp.
Pincheira, J. A., Dotiwala, F. S., and DSouza, J. T., 1999. Seismic Analysis of Older
Reinforced Concrete Columns, Earthquake Spectra. EERI, V. 15, No. 2, May
1999. pp. 245-272.
Pochanart, S. and Harmon, T., 1989. Bond-Slip Model for Generalized Excitation
Including Fatigue, ACI Materials Journal. V. 86, No. 5, Sept. 1989. pp. 465-
474.
Rodriguez, M.E., Botero, J.C. & Villa, J. (1999), Cyclic Stress-Strain Behaviour of
Reinforcing Steel Including Effect of Buckling, Journal of Structural
Engineering, ASCE, Vol. 125(6), 605-612.
Potger, G.M., Kawano A., Griffith M.C., and Warner R.F., Dynamic Analysis of RC
Frames Including Buckling of Longitudinal Steel Reinforcement, Proceedings of
the NZSEE 2001 Conference. Paper No. 4.12.01
Roy, H. E. H., and Sozen, M. A., 1964. Ductility of Concrete, International

Symposium on Flexural Mechanics of Reinforced Concrete Proceedings.
Miami, Fla., Nov. 1964. pp. 213-235.
Saatcioglu, M., Alsiwat, J. M., and Ozcebe, G., 1992. Hysteretic Behavior of
Anchorage Slip in R/C Members, Journal of Structural Engineering. ASCE, V.
118, No. 9, Sept. 1992. pp. 2439-2458.
Saatcioglu, M. and Ozcebe, G., 1989. Response of Reinforced Concrete Columns to

Simulated Seismic Loading, ACI Structural Journal. V. 86, No. 1, Jan. 1989.
pp. 3-12.
Sasani, M., 2004. Shear Strength and Deformation Capacity Models for RC Columns,
Proceedings of the 13th World Conference on Earthquake Engineering.
Vancouver, B.C., Canada, Aug. 2004. Paper No. 1838.
Sezen, H., 2002. Seismic Behavior and Modeling of Reinforced Concrete Building
Columns, Ph.D. Dissertation. University of California, Berkeley, 2002. 324 pp.
Sezen, H., Whittaker, A. S., Elwood, K. J., and Mosalam, K. M., 2003. Performance of
Reinforced Concrete Buildings during the August 17, 1999 Kocaeli, Turkey
Earthquake, and Seismic Design and Construction Practise in Turkey,
Engineering Structures. Elsevier, V. 25, No. 1, Jan. 2003. pp. 103-114.
186
Sezen, H. and Moehle, J. P., 2003. Bond-Slip Behavior of Reinforced Concrete
Members, fib-Symposium: Concrete Structures in Seismic Regions. CEB-FIP,
Athens, Greece, May 2003.
Sezen, H. and Moehle, J. P., 2004. Shear Strength Model for Lightly Reinforced
Concrete Columns, Journal of Structural Engineering. ASCE, V. 130, No. 11,
Nov. 2004. pp. 1692-1703.
Sezen, H., Moehle J.P. Seismic Tests of Concrete Columns with Light Transverse
Reinforcement. ACI Structural Journal 2006; 103 (6) 842-849.
Sezen H., and Setzler E.J. Reinforcement Slip in Reinforced Concrete Columns. ACI
Structural Journal. 2008; 105(3) 280-289
Soroushian, P. and Choi, K.-B., 1991. Analytical Evaluation of Straight Bar Anchorage
Design in Exterior Joints, ACI Structural Journal. V. 88, No. 2, Mar. 1991. pp.
161-168.
Soroushian, P., Obaseki, K., Nagi, M., and Rojas, M. C., 1988. Pullout Behavior of
Hooked Bars in Exterior Beam-Column Connections, ACI Structural Journal.
V. 85, No. 3, May 1988. pp. 269-276.
Setzler, E.J. Modeling the Behavior of Lightly Reinforced Concrete Columns Subjected
to Lateral Loads. M.S. Thesis. The Ohio State University, Columbus, Ohio, 2005.
Setzler, E.J., Sezen, H. Model for the Lateral Behavior of Reinforced Concrete Columns
Including Shear Deformations. Earthquake Spectra. 2008; 24(2) 493-511
Suda, K, Murayama, Y. Ichinomiya, T. & Shimbo, H., (1996), Buckling Behaviour of

Longitudinal Reinforcing Bars in Concrete Columns Subjected to Reverse Lateral
Loading, Eleventh World Conference on Earthquake Engineering, 1996.
Ueda, T., Lin, I. J., and Hawkins, N. M., 1986. Beam Bar Anchorage in Exterior
Column-Beam Connections, ACI Journal. V. 83, No. 3, May 1986. pp. 412-
422.
Umehara, H. and Jirsa, J. O., 1984. Short Rectangular RC Columns under Bidirectional
Loadings, Journal of Structural Engineering. ASCE, V. 110, No. 3, Mar. 1984.
pp. 605-618.
Vecchio, F. J. and Collins, M. P., 1986. The Modified Compression-Field Theory for
Reinforced Concrete Elements Subjected to Shear, ACI Journal. V. 83, No. 2,
Mar. 1986. pp. 219-231.
187
Vecchio, F. J. and Collins, M. P., 1988. Predicting the Response of Reinforced
Concrete Beams Subjected to Shear Using Modified Compression Field Theory,
ACI Structural Journal. V. 85, No. 3, May 1988. pp. 258-268.
Vecchio, F. J., Nonlinear Finite Element Analysis of Reinforced Concrete Membranes,

ACI Structural Journal, V. 86, No. 1, Jan.-Feb. 1989, pp. 26-35.
Vecchio, F. J., and Collins, M. P., Predicting the Response of Reinforced Concrete
Beams Subjected to Shear Using Modified Compression Field Theory, ACI
Structural Journal, V. 85, No. 3, May-June 1988, pp. 258-268.
Wehbe, N. I., Saiidi, M. S., and Sanders, D. H., 1999. Seismic Performance of
Rectangular Bridge Columns with Moderate Confinement, ACI Structural
Journal. Vol. 96, No. 2, Mar. 1999. pp. 248-258.
Wight, J. K. and Sozen, M. A., 1975. Strength Decay of RC Columns under Shear
Reversals, Journal of the Structural Division. ASCE, V. 101, No. ST5, May
1975. pp. 1053-1065.
188

Osu1269631164 Carga Vertical PDF

Uploaded by

Document Information

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

Osu1269631164 Carga Vertical PDF

Uploaded by

Copyright:

Available Formats

RESPONSE ESTIMATION OF REINFORCED CONCRETE COLUMNS SUBJECTED

Presented in Partial Fulfillment of the Requirements for

the Degree Master of Science in the

Graduate School of The Ohio State University

Muhammad Shadab Lodhi

Graduate Program in Civil Engineering

The Ohio State University

Masters Examination Committee:

Dr. Halil Sezen, Adviser

Dr. Shive K. Chaturvedi

Dr. Ethan Kubatko

Muhammad Shadab Lodhi

of the predicted responses with experimental test data, an analytical procedure is

while considering interaction between these mechanisms. Flexural deformations are

calculated by incorporating effect of shear deformation on flexural performance through

deformations on shear behavior. The interaction between axial-flexure and axial-shear

In addition, buckling of compression bars is also incorporated by considering separate

stress-strain relationships for reinforcing steel in tension and compression. The

and shear displacements. Total lateral response is obtained by combining deformation

well in estimating response of reinforced concrete columns to lateral loads.

This document is dedicated to my parents, M.N.U.K Lodhi and Hafeez Akhtar

and my wife, Tabassum Shadab

and my children, Muhammad Ibrahim Khan Lodhi and Aliza Lodhi

Kubatko for giving their time as members of my committee.

I am also deeply obliged to Ministry of Defence, Government of Pakistan for

June 18, 1971 .. Born Pakistan

1993 B.S. Civil Engineering,

1993 - 2006 .... Commissioned Military Service

2006 to present. MS/PhD Graduate Student

Major Field: Civil Engineering

Specialization: Structural Engineering

2 DISPLACEMENT COMPONENT MODEL...... 11

3 AXIAL SHEAR FLEXURE INTERACTION (ASFI) APPROACH -

4 IMPLEMENTATION OF ASFI APPROACH.. 71

5 COMPARISON OF COMPONENT DISPLACEMENT MODEL AND

6 DEVELOMPENT AND IMPLEMENTATION OF THE PROPOSED

4.1 Geometric and Material Properties of the Test Specimens. 98

1.4 Collapsed reinforced concrete buildings during 2005 Pakistan

1.5 Examples of commonly observed deficiencies and damage in reinforced

2.1 Components of lateral deformation in a reinforced concrete column 30

2.2 Plastic Hinge Model for Determining Flexural Deformation 30

2.3 Bar Slip Model ... 31

2.4 Reinforcement slip rotation ... 31

2.5 Shear displacement model . 32

2.7 Classification of the column and Lateral load-displacement relationships 33

2.8 Rules governing the combination of deformation components.............. 33

2.10 Constitutive relationship for reinforcing steel 34

2.11 Lateral load flexural deformation relationships. 35

2.12 Lateral load slip deformation relationships.... 36

2.13 Lateral load shear deformation relationships.. 37

2.14 Lateral load total deformation relationships.... 38

3.1 Concrete compression softening... 38

3.2 Axial-Flexural model and determination of axial strain due to flexure 67

3.3 Plastic hinge model for flexural deformations ... 67

3.5 Average stress-strain relationship for (a) Cracked concrete compression

3.6 (a) Calculated average stresses (b) Local stresses at a crack.. 69

3.7 Parameters influencing crack spacing .... 70

3.8 Coordinate transformation for concrete and steel material stiffness.......... 70

3.9 Spring model for axial-shear-flexural element 70

4.1 Flow chart for implementation of ASFI approach for columns 89

4.2 Constitutive relationships for concrete in compression.. 90

4.3 Constitutive relationship for concrete in tension ... 90

4.4 Constitutive relationship for reinforcing steel ... 91