Professional Documents
Culture Documents
TO LATERAL LOADS
A Thesis
By
2010
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
proposed. In the proposed procedure, flexural and shear deformations are determined
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
mechanisms allows for accurate response estimation while decoupled flexural analysis
minimizes iterations within the analysis and make the process relatively simple and easy.
displacement due to reinforcement slip are calculated separately and added to flexural
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
iii
Dedication
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
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.
providing me this wonderful opportunity of graduate studies and supporting myself and
my family financially.
v
VITA
FIELDS OF STUDY
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
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
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
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
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
2.6 Spring representation for total response model for fixed ended column 32
xii
Figure Page
2.9 Constitutive relationships for concrete in compression.. 34
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
xiii
Figure Page
4.7 Comparison of the results for Specimen-5 (Ousalem et al, 2005).. 93
xiv
Figure Page
5.18 Lateral load shear displacement relationship for Specimen-3. 125
xv
Figure Page
6.15 Lateral load slip displacement relationship for Specimen-4 169
xvi
CHAPTER 1
INTRODUCTION
1.1 Overview
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
The reconnaissance of damage observed during the past earthquakes suggests that
poorly designed reinforced concrete columns are the most critical elements to sustain
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
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
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
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.
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.
The large numbers of reinforced concrete buildings that are not designed to
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
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
Even important existing buildings and planned future structures may require performance
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
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
design codes. There are a number of studies investigating structural response of non-
approaches can predict structural behavior with good accuracy but employ very
amenable and are difficult to implement. Contrary, many approaches simplify the
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
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
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
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
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
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
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
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
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
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
interaction between these mechanisms. Two of the available bar buckling models are
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)
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
2.1 Introduction
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
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
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
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.
For reinforced concrete elements subjected to bending moment and axial load,
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
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
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
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 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
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,
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
Mp
Vp (2.3)
a
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
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
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
(Sezen, 2002).
calculated separately and added to the other deformation components, such as flexure and
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
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
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
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
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
s s L (2.8)
ignored in design and research due to their lack of complete understanding and being
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
is desired.
The basis of the shear model used in this analytical approach is the model
piecewise linear model defining key points in the lateral force-shear deformation
18
indirectly from Response-2000 by integrating shear strain distribution over the height of
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
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
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
and Moehle (2005a), which is based on a shear friction model and an idealized shear
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,
according to the expected failure mode and classification of the column into categories as
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
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
displacement ductility which is the ratio of the maximum displacement to the yield
displacement.
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
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
The shear strength is less than flexural strength and column fails in shear,
however inelastic flexural deformation occurring prior to shear failure affects the
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.
The shear and flexural strengths are nearly identical. It is not possible to
predict conclusively which mechanism will govern the peak response. Shear and
22
load level is the sum of the post-peak flexure, slip, and shear deformations
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
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
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
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
Moehle, 2004), which can cause shear failure in the columns that are initially dominated
can occur at higher displacements, can not be captured by simply combining three
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
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
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
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
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.
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
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
of the concrete and is equal to 57,000 fco (in psi units) for normal weight concrete, and
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
u 0.23 , p 6 , E s 29,000 ksi. Steel stress-strain relationship used for this model is
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
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
2.6 Conclusion
Displacement component model by Setzler and Sezen (2008) has briefly been
concrete column to lateral loading being composed of flexural, bar slip and shear
mechanism. Structural response of four test specimen is calculated as per the prescribed
Comparison of predicted response and experimental data shows that the understudy
model has correctly been implemented and model performs well in evaluating lateral
29
Flexure deformation Slip deformation Shear deformation
y
x x
Lp L - Lp L
y y
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
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
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)
Category
Category IV IV Category II
Category II
33
3500
Confined concrete
2500
Stress (psi)
2000
1500
1000
500
0
0 1 2 3 4 5 6 7 8
Strain x 10
-3
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
20
-20
-40
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
80
Specimen - 4
60
Lateral force (kip)
40
20
-20
-40
20
-20
-40
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
80
Specimen - 4
60
Lateral force (kip)
40
20
-20
-40
20
-20
-40
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
80
Specimen - 4
60
Lateral force (kip)
40
20
-20
-40
20
-20
-40
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
80
Specimen - 4
60
Lateral force (kip)
40
20
-20
-40
3.1 Introduction
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
section analysis. This is very convenient and accurate analytical tool for flexural
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.
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
interaction between axial, shear and flexural mechanisms. The approach is designated as
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.
component models and development of the approach in generalized form for a reinforced
40
3.2 Background
column can accurately be estimated if flexural and shear deformations are evaluated
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
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
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
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
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
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
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
concrete element by employing standard flexural section analysis techniques. The section
cracked concrete are employed to consider the effect of shear deformations on flexural
concrete and is obtained from in-plane shear analysis of the flexural element, thus
42
performed independently. This aspect is described in detail in subsequent sections. The
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
stress field can be employed as axial-flexure model in ASFI approach, provided that
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
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
softening, has been studied by Vecchio and Collins (1986) in their efforts to investigate
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-
fcsof fc (3.1)
defined as
44
1
1.0 (3.2)
0.8 0.34 c1
co
cylinder strength.
Axial strain due to flexure is another interaction term in ASFI approach that
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
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
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
normal and shear stresses. Vecchio and Collins (1986) developed the theory after
modifying previously proposed Compression Field Theory (Mithell and Collins 1974)
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
stress-strain relationships formulated in terms of average stresses and strains. The critical
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
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
2. Stresses and strains can be considered in terms of average values when taken over
3. The concrete and reinforcing bars are perfectly bonded at the boundaries of the
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.
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
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
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
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
(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.,
considered to make counterclockwise angle with positive x-axis. Hence, following these
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
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
49
3.4.1.1 Concrete in Compression.
The constitutive laws for concrete in compression in MCFT have been derived by
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,
2
f c 2 f co 2 c 2 c 2 (3.10)
co co
compressive strain and f c 2 is corresponding concrete stress. The term multiplied with
compression, often used for normal strength concrete. Although, this equation is
can theoretically be used as part of axial-shear model in ASFI approach; however, it must
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
50
The relationship suggested prior to cracking i.e., 0 c1 cr is linearly elastic
f c1 Ec c1 (3.10)
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
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
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
f sx Es sx f sxyield
(3.14)
f sy Es sy f syyield
f sxyield and f syyield are yield stress of the reinforcement in x and y directions,
respectively.
and concrete must match with the in plane strains in x and y -directions i.e., sx x
f sx Es x f sxyield
(3.15)
f sy Es y f syyield
reinforcement, other reinforcement relations allowing strain hardening can be used as part
52
3.4.2 Considerations of Local Cracks Conditions
Consideration of the local stress conditions at the cracks is another very important
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
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
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
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-
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
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
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
f co
vci max (3.21)
0.31 24w /(a 16)
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
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
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
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
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
modifying existing linear elastic finite element routines based on secant stiffness
formulation provided realistic constitutive relations for concrete and reinforcement are
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
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
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
f c1 fc 2 Ec1 Ec 2
Ec1 , Ec 2 , Gc (3.28)
c1 c2 Ec1 Ec 2
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
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
where, c for the concrete component, 0 and 90o for longitudinal and transverse
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
The main objective of the ASFI approach is to couple axial-flexural and axial-
58
their interaction in terms of axial deformations and concrete compression strength
concrete column subjected to axial load, bending moment and shear force. Total axial
mechanism, total axial strain is combination of axial strains caused by axial mechanism
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
x xas xs xf (3.34)
mechanisms in axial-shear and axial-flexural elements are equal to the axial strain due to
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)
analysis.
xf xs o
(3.38)
f s
where o is axial stress due to applied axial load and is resultant shear stress. Shear
1 M1 M 2
f
bd f l12
(3.39)
V
s
bd s
other flexural section, b is width of the cross-section, d f is the flexural depth of the
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
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
y and z are neglected due to inexistence of lateral external forces along theses
directions.
y z 0 (3.41)
concrete and unconfined cover concrete are defined differently, constitutive law for the
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
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
Axial deformation and concrete compression softening are two main interaction
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
62
component of axial-shear model. The flexibility component for axial deformation due to
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.
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.
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
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
xas xa .
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-
mechanism (Equation 3.2) must be employed in flexural analysis to soften the uniaxial
springs in series as illustrated in Figure 3.9. Shear spring of axial-shear model with
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
1 1 1
(3.47)
K K f K s
s f .
flexural models, in order to determine total axial stiffness k x , axial stiffness from axial-
defined for each axial spring individually based on its deformation, total axial stiffness
1 1 1 1
k x k xa k xs k xf (3.48)
k x x x where x xs xf xa
x
k xa , k xs x , k xf x (3.49)
xa xs xf
3.6 Conclusion
65
that consists of two models; axial-flexure and axial-shear model evaluating flexural and
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
equilibrium conditions.
66
c2
c2
f co c2
f co c2
c1 c2
c2
co c2
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
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
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
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
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
such as beams and columns. In addition to the models considered for axial, flexural and
the section of maximum moment must also be incorporated into the analysis for
In this chapter, the analytical procedure for implementation of ASFI approach for
Solution technique for implementing the analytical procedure is based on ASFI approach
71
and lateral load-lateral drift responses of previously tested columns are presented to
responses of the test specimens shall be used subsequently for discussion and
comparison.
between any two subsequent flexural sections of a reinforced concrete element could be
satisfied at any load level. Hence, this approach can principally be applied for response
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
72
below is considered to account for deformation caused by reinforcement slip at the
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
column end, which is in series with springs of flexure, and shear mechanisms. Hence,
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)
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
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,
Total axial and lateral deformations in ASFI approach are computed based on
axial-shear and pullout. Hence, total lateral drift ratio in Equations 4.3 is sum of shear
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
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.
This section summarizes the procedure for estimating the response of laterally
inflection point to one of the end section. Perform flexural section analysis on end section
relationship of the concrete given by any of the appropriate constitutive law for uniaxial
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
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
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.
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
equilibrium conditions as per Section 3.5 and convert axial-flexure and axial-shear
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
approach is presented. The procedure is iterative and outlines only the major steps
required for response estimation. This section summarizes the solution technique for
1. Define or input material properties and geometry. Decide on a sign convention for
2. Input axial load P . If there is no applied axial load, then consider a negligible
3. Select a small value of total drift ratio , such as 0.000001 , as a starting value.
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-
average normal strain in y -direction for axial-shear element ( yi ) and average shear
77
6. Perform flexural section analysis at the end section of the axial-flexure element
strain oi 1 .
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
appropriate models. Pullout deformations include pullout rotations pul and axial strain
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
78
b. Determine average steel stresses f sx and f sy corresponding to the
model (step 6). However, if in axial-flexure model cover and core concrete
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
not satisfied and local shear stress at the surface of the crack exceeds shear
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
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
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
satisfactory convergence is achieved, then go to the next step otherwise repeat steps 5 to
16. Compute lateral load capacity of the column corresponding to the given total drift
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.
results for a number of reinforced concrete test specimens. In order to implement the
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
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
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
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.
c
2
fc f p 2 c (4.10)
p p
f c f p 1 Z m c p (4.11)
where,
These are determined based on modified Kent and Park model for stress-strain
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
3.2.
about the final choice of the tensile concrete constitutive material model used for
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
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
f c1 Ec c1 (4.15)
83
where, Ec is modulus of elasticity of the concrete, cr is cracking strain corresponding to
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
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
84
4.3.3 Flexural Deformation Model
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
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
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.
23
20 MPa
slip SD ' (4.23)
fc
slip
pul Rpul (4.24)
X
pul
eR 0
pul
(4.25)
Lin
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.
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
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,
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
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
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
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
30 Unconfined concrete
Stress (Mpa) 25
20
15
10
0
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
Strain
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
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
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
Figure 4.8: Comparison of the results for Specimen-12 (Ousalem et al, 2005)
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
Figure 4.9: Comparison of the results for Specimen-14 (Ousalem et al, 2005)
95
350
Reported response
250
Lateral force (kN)
200
150
100
50
0
0 0.05 0.1 0.15 0.2
Drift ratio
Figure 4.10: Comparison of the results for Specimen-15 (Ousalem et al, 2005)
96
400
Reported response
350 Predicted response
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
Figure 4.11: Comparison of the results for Specimen-16 (Ousalem et al, 2005)
97
bxh 2Lin sh g w fyx fyy fco
Specimen
(mm) (mm) (mm) (%) (%) (Mpa) (Mpa) (Mpa)
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
98
CHAPTER 5
5.1 Introduction
approach, test specimens were modeled with material constitutive relationships and other
experimental responses indicated that both methods perform equally well despite
following quite different approaches for response estimation of respective set of test
to model the same set of specimens employing same material and deformation
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
99
5.2 Implementation of the Analytical Models
outlined and response estimated for four columns (Sezen, 2002) is compared with
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
appropriate set of test specimens and material models are selected and employed for
Reinforced concrete columns tested by Sezen (2002) provide data for flexural,
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
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-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
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
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
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
cylinder strength (in psi units), Ec is the modulus of elasticity of the concrete for normal
from Equation 5.2, which is obtained by slightly modifying the maximum strain formula
cu 0.004 0.14 x y
f yy
(5.2)
f cc
dt2
x y (for square columns) (5.3)
2 sd c
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).
behavior for confined and unconfined concrete obtained from Equations 5.1 through 5.4
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
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
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
employing Sezen and Setzler (2008) model. This model has was explained in detail in
analysis after unconfined cover concrete fiber next to compression steel reaches
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
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.
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
Figure 5.6. Initially, both approaches predict similar responses but after reaching the
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
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
deformations indicates that displacement component model and ASFI approach predict
pre peak response almost identically. However, post peak responses differ significantly. It
employing common material constitutive laws, fiber section analysis procedures and
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
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
This implies that the only difference in near-peak and post-peak responses
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.
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.
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
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,
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
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
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
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
component model, but post peak response captures slope of degrading branch very well.
load is underestimated. Similar trends are observed for response estimated for Specimen-
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
Response-2000 is basically a section analysis routine that incorporates shear effects in the
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
Response 2000. Nevertheless, it produces realistic response especially for shear critical
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
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
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
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.
Figure 5.20 shows the comparison of predicted and experimental lateral load-total
Moehle equation (Equation 2.13) is calculated to be 69.0 kips and flexural strength from
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
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
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
5.22. pre peak response predicted by both approaches is identical and overestimates pre
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
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
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.
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
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
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
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
components in ASFI approach must be considered for total response in post peak state.
115
5.4 Conclusion
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
2000
1500
1000
500
0
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016
Strain
250
200
150
Stress (psi)
100
50
0
0 0.2 0.4 0.6 0.8 1
Strain x 10
-3
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
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.)
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
Flexural displacement (in.)
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
Flexural displacement (in.)
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
Flexural displacement (in.)
120
4500
4000
3500
3000
2500
M (k-in.)
2000
1500
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
Curvature (1/in.) x 10
-3
121
4500
4000
3500
2500
2000
1500
1000
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.)
122
80
60
40
20
Lateral force (kip)
-20
-40
-60
Test data
-80 Setzler and Sezen
ASFI approach
-100
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
Slip displacement (in.)
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
Slip displacement (in.)
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
Slip displacement (in.)
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.)
124
100
Test data
80
Setzler and Sezen
60 ASFI approach
20
-20
-40
-60
-80
-100
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.5
Shear displacement (in.)
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
Shear displacement (in.)
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
Shear displacement (in.)
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.)
126
80
60
40
20
Lateral force (kip)
-20
-40
-60
Test data
-80 Setzler and Sezen
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.)
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
Flexural displacement (in.)
Figure 5.25: Effect of plastic hinge length on flexural response by ASFI approach
(Specimen-1)
129
CHAPTER 6
6.1 Introduction
approach were investigated and predicted responses were compared with experimental
test data. Based upon the observations made in the last chapter, a new analytical
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.
combined together depending upon dominant failure mode to obtain total response. The
softening and axial deformation allows for accurate response estimation while decoupled
flexural analysis minimizes iterations within the analysis and make the process relatively
130
considering separate stress-strain relationships for reinforcing steel in tension and
compression.
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.
subjected to lateral loading, flexure and shear mechanisms interact with each other and
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
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.
compressive strength of the concrete is ignored and concrete behavior is simulated by its
deformations on shear mechanism can also be accounted for by adding axial deformation
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
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
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
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
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 between flexural and shear deformations and must be retained for the
employing compression softening factor from shear model to flexural analysis. For the
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
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
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
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
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
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
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
and shear analyses while considering interaction between them. If the effect of shear
without having the need to perform shear analysis prior to flexure analysis while still
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
135
softening factor must be determined in a simplified way without performing full shear
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
end sections. This simplifies the analytical process significantly and eliminates the
iteration process required for shear analysis in original ASFI approach. The complete
3.2, is a function of concrete principal tensile strain c1 of the element being analyzed. As
strain, it is adopted in the proposed procedure with minor modifications. The approach is
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
as
1
1.0 (6.1)
c1
0.8 0.34
co
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
x
o xa (6.3)
2
Likewise 2 is concrete principal compressive strain for the element. Its value is
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
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
analysis based on average centroidal strain, fc1 0.145 fco is concrete principal tensile
from Equation 6.3 and 2 is concrete principal compressive strains determined from
Equations 6.4.
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
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
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
and shear mechanisms. In this procedure, interaction of axial strain is taken into account
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
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
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
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,
where fij i, j 1, 2,3 are flexibility components of in plane shear model, x is normal
direction, and xy is shear strain. In the above equation, longitudinal axis of the shear
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.
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
(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
The details on the bar buckling phenomenon, related aspects and proposed bar buckling
mainly due to axial, bending moment and shear forces. Reinforcing bars may experience
inelastic axial compression under severe loading and exhibit lateral deformation known
depends on a variety of factors such as, size and shape of the cross-section, the amount of
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
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
deformation behavior.
In this study, two of the available bar buckling models are evaluated to assess
deformations shall be determined by employing these models for the four specimens
tested by Sezen (2002). After comparing predicted responses with test data, appropriate
the reinforcing bars by finite element microanalysis using fiber technique. Their model 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
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
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-
* 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 *
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
144
is reached at point B. With further increases in compressive strain (past point B), the
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
where,
f 2 y f yy
2
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
Both of the above models and bar buckling model proposed in the previous
chapter are incorporated in flexural section analysis and flexural deformations are
relationships are compared with experimental responses in Figures 6.4 through 6.6.
Based on the comparison of the predicted and observed responses, the model
stresses of the longitudinal reinforcement to account for bar buckling phenomenon. This
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
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
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
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
are calculated using above mentioned procedure and are presented in Figures 6.8 through
6.11, respectively.
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.
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
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
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
response only up to the peak load and does not predict shear strength degradation with
displacement component model that is able to predict post peak behavior up to the point
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
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
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
deformation components for post peak response. The proposed shear model also
gives control to the user on choice of material constitutive models and parameters to be
calculated using proposed shear procedure and are presented in Figures 6.16 through
6.19, respectively.
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
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.
simply adding deformation components due to flexure, bar slip and shear mechanism
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).
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
This section summarizes the procedure for estimating the response of laterally
1. Define or input material properties and geometry. Decide on a sign convention for
2. Input axial load P . If there is no applied axial load, then consider a negligible
value.
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
resultant compressive force of the stress block at end section c , and axial
understanding.
Equations 6.5 and 6.2. The average value of concrete principal tensile
concrete compression stress of the stress block with the help of employed
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
154
6. Evaluate axial-flexure mechanism by performing flexural section analysis at the
ductility of the core concrete must be taken into account with the help of
on flexural performance.
155
equilibrium. When section equilibrium is achieved, calculate nominal
viii. Perform axial section analysis at the inflection point and determine axial
constitutive laws employed for flexural section analysis of the end section.
with the help of Equation 6.17. Plastic hinge length is taken equal to one
Section 6.6. Lateral displacement due to slip is added to flexural and shear
M by following equation
M
V (6.18)
Lin
xii. Increment curvature to next load step and repeat all steps until either
156
section, use constant compression softening factor corresponding to its
interaction of axial strains for the column element between end section and inflection
i. Start with the first load step from axial-flexural analysis explained above.
small values for each of them for the first iteration. These variables are
respectively.
157
v. Determine average principal tensile and compressive stresses in concrete
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
not satisfied and local shear stress at the surface of the crack exceeds shear
vii. Determine secant moduli for concrete and reinforcements and assemble
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
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.
response at current load level and go to the next load step. Otherwise,
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
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
6.8.
159
6.10 Implementation of the Model and Comparison of the Results
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
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,
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
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
into the flexural analysis. The shear model evaluates the shear response of the column
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
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
fs
*
s , f s*
fy
,
* *
0.02 Es
0.2 f y
s
y
f s ,lb A
f s , ps
s
y
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
Flexural displacement (in.)
163
80
60
40
20
Lateral force (kip)
-20
-40
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
Flexural displacement (in.)
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
Flexural displacement (in.)
165
80
60
40
20
Lateral force (kip)
-20
-40
-60
Test data
-80 Setzler and Sezen
Proposed model
-100
-1.5 -1 -0.5 0 0.5 1 1.5
Flexural displacement (in.)
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
Flexural displacement (in.)
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
Flexural displacement (in.)
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
Slip displacement (in.)
60
40
20
Lateral force (kip)
-20
-40
-60
Test data
-80 Setzler and Sezen
Proposed model
-100
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
Slip displacement (in.)
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
Slip displacement (in.)
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
Slip displacement (in.)
80
Test data
60 Setzler and Sezen
ASFI approach
40 Propose model
Lateral load (kip)
20
-20
-40
-60
-80
-5 -4 -3 -2 -1 0 1 2
Shear displacement (in.)
169
100
Test data
80 Setzler and Sezen
60 ASFI approach
Proposed model
Lateral force (kip) 40
20
-20
-40
-60
-80
-100
-1 -0.75 -0.5 -0.25 0 0.25 0.5
Shear displacement (in.)
100
80
60
40
Lateral force (kip)
20
-20
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
Shear displacement (in.)
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.)
60
40
20
Lateral force (kip)
-20
-40
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
173
CHAPTER 7
CONCLUSION
7.1 Summary
The focus of this research was to suggest a suitable analytical procedure that can
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
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
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
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
The flexural behavior in the proposed procedure was evaluated by fiber section
flexural behavior, cracked concrete behavior was considered in the section analysis. This
determined as per the suggested procedure that decouples flexural analysis from shear
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
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
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
separately by Setzler and Sezen (2008) bar slip model and were added to flexural and
176
7.1.2 Combined Model
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
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
177
performance and procedure to calculate compression softening factor appeared to work
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
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
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
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
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.
simple calculations to estimate component and total deformations. The procedure can be
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
The research reported here presents a model for the monotonic lateral deformation
of lightly reinforced concrete columns subjected to lateral loads. However, there remains
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.
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
displacements. This would aid in improving the component models, which should lead to
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.
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.
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. and Moehle, J. P., 2005a. Axial Capacity Model for Shear-Damaged
Columns, ACI Structural Journal. V. 102, No. 4, July 2005. pp. 578-587.
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.
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.
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.
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.
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., 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.
184
Engineering, Architrave Department, University of Tokyo, Tokyo, Japan, 2006,
255.
Mostafaei, H., and Vecchio, F. J. (2008). Uniaxial Shear-Flexure Model for Reinforced
Concrete Elements. ASCE Journal of Structural Engineering, 134(9), 1538-
1547.
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
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.
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
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., 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