Ermiyas Ketema PDF

ADDIS ABABA UNIVERSITY
SCHOOL OF GRADUATE STUDIES

FACULTY OF TECHNOLOGY
DEPARTMENT OF CIVIL ENGINEERING
DESIGN AID FOR COMPOSITE COLUMNS
A thesis submitted to the school of Graduate Studies in Partial fulfillment of the

Requirements for the Degree of Master of Science in Structural Engineering
By
Ermiyas Ketema
Advisor: Dr. Shifferaw Taye
July 2005
ADDIS ABABA UNIVERSITY
SCHOOL OF GRADUATE STUDIES
FACULTY OF TECHNOLOGY
DEPARTMENT OF CIVIL ENGINEERING
DESIGN AID FOR COMPOSITE COLUMNS

(CONCRETE FILLED TUBES)
By
Ermiyas Ketema
July 2005
Approved by Board of Examiners
______________________ _____________________ _________________

Advisor Signature Date
______________________ _____________________ __________________

External Examiner Signature Date
______________________ _____________________ _________________

Internal Examiner Signature Date
______________________ _____________________ __________________

Chairman Signature Date
AKNOWLEDGMENT
I would like to thank my advisor, Dr. Shifferaw Taye, for his valuable advice and guidance
for my work as well as providing the necessary materials for the research work.
My thanks also go to Jimma University, for being the sponsor for my M.Sc. study.
I would also like to use this opportunity to convey my gratitude to academic staff of the
Department of Civil Engineering, Faculty of Technology, Addis Ababa University and to my
friends. Without their support and encouragement I couldn’t have this opportunity to
complete my study.
i
TABLE OF CONTENT
AKNOWLEDGMENT ...........................................................................................................................................i
TABLE OF FIGURES......................................................................................................................................... iii
NOTATIONS ........................................................................................................................................................v
ABSTRACT ........................................................................................................................................................vii
INTRODUCTION..................................................................................................................1
CONCRETE-FILLED COMPOSITE COLUMNS .................................................................3
TYPES OF COMPOSITE COLUMNS .........................................................................................................................3
MERITS AND DEMERITS OF CFT COMPOSITE COLUMNS ......................................................................................3
APPLICATIONS .....................................................................................................................................................5
STRUCTURAL BEHAVIOUR ...................................................................................................................................7
Effect of confinement ......................................................................................................................................7
Bond between concrete and steel ...................................................................................................................8
Section failure behaviour ...............................................................................................................................9
Earthquake resistance ..................................................................................................................................10
COLUMN LOAD CAPACITY ............................................................................................12
AXIAL COMPRESSION ........................................................................................................................................13
Resistance of cross-section...........................................................................................................................13
Relative slenderness .....................................................................................................................................14
Buckling resistance ......................................................................................................................................16
RESISTANCE TO COMPRESSION AND BENDING ...................................................................................................16
Cross-section resistance under uniaxial moment and axial compression ....................................................16
The influence of shear force .........................................................................................................................20
Slender member resistance under axial compression and uniaxial bending................................................20
Member resistance under axial compression and biaxial bending ..............................................................22
SHORTCOMINGS OF EBCS 4-1995 DESIGN PROCEDURE ....................................................................................23
CHART DEVELOPMENT ..................................................................................................24
CALCULATION METHOD AND SCOPE .................................................................................................................24
SELECTED REPRESENTATIVE SECTIONS & MATERIALS .....................................................................................25
INTERACTION CHART FOR AXIAL COMPRESSION AND UNIAXIAL BENDING .................................................26
Fundamental equations ................................................................................................................................26
Rectangular (Square) section .......................................................................................................................27
Circular section............................................................................................................................................32
Hexagonal section ........................................................................................................................................36
Octagonal section.........................................................................................................................................47
BIAXIAL CAPACITY FOR RECTANGULAR SECTION .............................................................................................54
Fundamental equations ................................................................................................................................54
Moment and axial load capacity for different neutral axis positions ...........................................................56
Biaxial chart.................................................................................................................................................59
EXAMPLES ........................................................................................................................61
CONCLUSIONS..................................................................................................................70
APPENDIX: Uniaxial and Biaxial Charts ..........................................................................................................72
REFERENCES ....................................................................................................................................................81
ii
TABLE OF FIGURES
Fig. 2.1 Typical cross-sections of composite columns............................................................. 4
Fig. 2.2 Photos of the Millennium Tower, 1st Street Plaza and Sannomia Grand buildings in
respective order from left to right................................................................................ 6
Fig. 2.3 Tie-bar stiffening scheme .......................................................................................... 7
Fig. 2.4 Cyclic load-deflection behavior of a CFT beam–column.......................................... 11
Fig. 3.1 M-N interaction curve for uniaxial bending.............................................................. 17
Fig. 3.2 Development of stress blocks at different points on the interaction curve................. 19
Fig. 3.3 Resistance to axial compression and uniaxial bending ............................................. 21
Fig. 3.4 Member resistance under compression and biaxial bending ..................................... 23
Fig. 4.1 Composite cross-section regions used for computing section capacity ..................... 26
Fig. 4.2 Rectangular cross-section ....................................................................................... 27
h h
Fig. 4.3 Neutral axis position − t ≤ hi ≤ ........................................................................ 29
2 2
h
Fig. 4.4 Neutral axis position 0 ≤ hi ≤ − t .......................................................................... 30
2
h
Fig. 4.5 Neutral axis position 0 ≤ hi ≤ − t .......................................................................... 31
2
Fig. 4.6 Circular cross-section ............................................................................................. 32
d
Fig. 4.7 Neutral axis position for 0 ≤ hi ≤ − t ................................................................... 33
2
Fig. 4. 8 Segment of a Circle ................................................................................................ 34
d
Fig. 4.9 Neutral axis position with 0 ≤ hi ≤ − t ................................................................. 35
2
Fig. 4.10 Hexagonal CFT cross-section ................................................................................ 36
Fig. 4.11 Hexagon of side length s........................................................................................ 37
3s 3s
Fig. 4.12 Neutral axis position with − t ≤ hi ≤ ....................................................... 38
2 2
3s
Fig. 4.13 Neutral axis position with 0 ≤ hi ≤ − t ............................................................ 39
2
3s
Fig. 4.14 Neutral axis position with 0 ≤ hi ≤ − 1 ........................................................... 41
2
iii
Fig. 4.15 Hexagon of side length s........................................................................................ 42
s
Fig. 4.16 Neutral axis position with ≤ hi ≤ s − 1.155t ........................................................ 43
2
s'
Fig. 4.17 Neutral axis position with 0 ≤ hi ≤ .................................................................... 44
2
s'
Fig. 4.18 Neutral axis position with 0 ≤ hi ≤ .................................................................... 45
2
s
Fig. 4.19 Neutral axis position with ≤ hi ≤ s' ..................................................................... 47
2
Fig. 4.20 Octagonal CFT cross-section ................................................................................. 47
Fig. 4.21 Octagonal section .................................................................................................. 47
s
Fig. 4.22 Neutral axis position with ≤ hi ≤ 1.207 s − t ....................................................... 49
2
Fig. 4.23 Neutral axis position with 0 ≤ hi ≤ 0.5s − 0.414t .................................................... 51
Fig. 4.24 Neutral axis position with 0 ≤ hi ≤ 0.5s − 0.414t .................................................... 52
s
Fig. 4.25 Neutral axis position with ≤ hi ≤ 1.207 s − t .................................................... 53
2
Fig. 4.26 Rectangular section and axes used for capacity computation.................................. 54
Fig. 4.27 Neutral axis position for t < s < b , and t ≤ r ≤ h ................................................... 56
Fig. 4.28 Neutral axis position for t < s < b and h ≤ r ....................................................... 57
Fig. 4.29 Neutral axis position for t < r < h , and b ≤ s ....................................................... 58
b' h'
Fig. 4.30 Neutral axis position for h ≤ r , b ≤ s and + ≥ 1 ............................................ 59
s' r '
Fig. 5.1 Stress block for zero axial compressive force........................................................... 63
Fig. 5.2 Stress block for maximum moment.......................................................................... 64
Fig. 5.3 Normalized interaction curve................................................................................... 65
iv
NOTATIONS
Ac Cross-sectional area of concrete

Acc Cross-sectional area of concrete under compression
As Cross-sectional area of steel profile
Asc Cross-sectional area of steel profile under compression
Ast Cross-sectional area of steel profile under tension
Av Sheared area of structural section
e Eccentricity of axial compressive force
Es Elastic modulus of steel
Ecm Secant modulus of the concrete
fck Characteristic concrete strength
fy Yield strength of structural steel
fcd, fyd Design strength for concrete and steel
hi Neutral axis distance from centroidal axis
Ic Second moment of area of uncracked concrete section
Is Second moment of area of steel section
l Buckling length of a column
Mmax,Rd Maximum moment resistance
Mu Bending resistance of the cross-section
Mxu Bending resistance of the cross-section about X-axis
Myu Bending resistance of the cross-section about Y-axis
Ncr Elastic critical load
NSd Axial design loading
NG.Sd Permanent part of NSd.
Npl.Rd Axial plastic resistance of a cross-section
Npl.Rk Npl.Rd where partial safety factors are taken as 1
Npm,rd Compressive resistance force for the whole area of concrete
Nu Axial compressive resistance
Qcx First moment of concrete area under compression about X-axis
Qcy First moment of concrete area under compression about Y-axis
v
Qsxc First moment of steel area under compression about X-axis
Qsxt First moment of steel area under tension about X-axis
Qsyc First moment of steel area under compression about Y-axis
Qsxt First moment of steel area under tension about Y-axis
v Normalized Axial compressive resistance
Vs,Sd Design shear force
Vpl,s,Rd Plastic shear resistance
w Ratio of contribution to over all axial plastic resistance of steel section to
concrete section.
Wpc,Wpa Plastic section moduli for the total concrete and structural steel sections
Wpcn, Wpan Plastic section moduli for parts of concrete and structural steel sections
x Ratio of side length to thickness of steel section
y Height-to-width ratio for rectangular cross-section
δ Relative contribution of the steel section to the overall axial plastic resistance
γ M2 , γc Partial safety factors for steel and concrete, respectively
λ Relative slenderness ratio
µ Normalized bending resistance of the cross-section
ρw Reduction factor on cross-sectional area due to influence of shear
ηs, ηc Coefficients for considering effect of confinement for circular cross-section
χ Factor taking account for the influence of imperfection and slenderness
vi
ABSTRACT
Composite columns are in an increased usage for construction of high-

rise and medium-rise buildings, bridges and other structures. However,
their designs involves tiresome calculations and drawing of interaction
charts. In addition, the results are approximate. This thesis presents
design aids for concrete-filled steel tubes to simplify the design and at
the same time increase the accuracy of results. It also summarizes
important behaviors of concrete-filled tube columns as compared to
other column types.
General approaches have been presented on how to prepare design

charts and such charts have been drawn for selected cross-sectional
shapes. Uniaxial charts have been drawn for square, rectangular,
hexagonal and octagonal shapes whereas biaxial chart is drawn for
square section. This thesis also provides guidelines to easily develop
other charts for other shapes and material types that were not
considered.
Numerical examples have been presented to illustrate the application

of the charts for the design of composite cross-sections. Results have
been compared with those obtained using the procedure outlined in
EBCS 4-1995 and have shown close similarity. The accuracy of
vii
biaxial charts as compared with the approximate method in EBCS 4-
1995 has also been shown with the help of a numeric example.
Keywords: Concrete-Filled Tubes, Uniaxial Bending, Biaxial Bending,

Plastic Capacity, Interaction Chart
viii
INTRODUCTION
With the increasing use of composite construction world wide, there is a growing interest in utilizing
Concrete - Filled Tubes (CFTs) as a primary column member. The interest develops from the fact that
properties of steel and concrete in the CFTs are fully utilized, so that the strength, stiffness and
ductility of the structures constructed from CFTs can be enhanced simultaneously. Since the function
of longitudinal reinforcement and transverse confinement can be acquired due to presence of the steel
tubes, the traditional longitudinal and transverse reinforcement may be eliminated. This type of
column also maintains sufficient ductility when high strength concrete is used.
CFT columns can replace conventional structural columns like reinforced concrete, structural steel
with reinforced concrete and structural steel alone with enhanced performance and at the same time
reducing costs to a minimum. It is especially useful in high-rise buildings where high strength is
required and flexibility of open space is desired for a maximum range of applications.
The purpose of this research is to develop design charts for easy determination of the necessary
section dimension and strength requirement for a given load or to determine the capacity of a given
cross-section. The charts may be used both for short and long columns. To make the chart usable for
any cross section dimension of a given shape, capacity equations used to draw the chart are
normalized.
This thesis consists of six chapters of which chapters two and three deal with literature review of the
subject matter. The second chapter briefly summarizes composite column types, relative advantages,
practical application made so far and their structural behavior. The third chapter discusses about
computation of section capacity for short and slender columns subjected to axial compression and
uniaxial or biaxial bending. Problems in the design process which are the reasons for this thesis work
are also presented.
The forth chapter contains the core of research work. It includes development of strength equations
for both the uniaxial and biaxial section capacities. Equations are presented for different cross-
1
sectional shapes considered in this thesis. The equations and procedure for drawing the chart are also
shown.
Chapter Five, presents illustrative examples to show the application of the charts. The results from the
charts developed have been compared to give similar results with the procedures given in EBCS 4-
1995. It also includes examples for cross-checking the similarity of results from uniaxial and biaxial
chart. This is done by checking the results from the biaxial chart for zero moment about one of the
axes to have similar value with the corresponding result from uniaxial chart. Further more, a
numerical example is presented to show the advantage of using biaxial chart over the approximate
method.
The last chapter consists of conclusion, summary of contributions and future research areas related to
this work.
All charts produced are given in the appendix.
2
CONCRETE-FILLED COMPOSITE COLUMNS
Types of Composite Columns
Different types of composite columns are principally in use which can be categorized as concrete-
encased steel and concrete-filled steel tube columns.
Concrete–encased composite columns have structural steel component that could be either one or
more rolled steel sections. In addition to supporting a proportion of the load acting on the column, the
concrete encasement enhances the behavior of the structural core by stiffening it, and so making it
more effective against both local and overall buckling. The concrete encasement can be either full
encasement (Fig.2.1a) or partial encasement (Fig. 2.1b).
Fig. 2.1c and 2.1d show cross-section of concrete-filled steel tube. These types contain steel hollow
tubes of different shape filled with either plain concrete or concrete with longitudinal and transverse
reinforcement. In this paper concrete-filled steel tubes without reinforcement bars are covered.
Merits and Demerits of CFT Composite Columns
There are many relative advantages of using CFTs rather than using the other column
types. These include:
• Concrete and steel are completely compatible and complementary to each other as they have
almost same thermal expansion and they have an ideal combination of strengths with the
concrete efficient in compression and the steel in tension.
• Composite construction, particularly that uses CFTs, allows rapid construction as there is no
need for form work construction. In addition, waiting for curing time of concrete is not
necessary as construction of upper storey can proceed before curing of concrete in the lower
storey.
3
a) Fully encased composite b) Partially encased
column composite columns
c) Concrete-filled column with reinforcement d) Concrete-filled tube without

reinforcement
e) Concrete-filled tube with structural steel sections
Fig. 2.1 Typical cross-sections of composite columns
• Steel is positioned at the furthest location of the cross-section to produce the maximum
flexural moment due to the maximum lever arm. Whereas, in reinforced concrete the possible
position of reinforcement is within the cross-section allowing for cover requirement.
• The steel section can replace the function of longitudinal reinforcement and transverse
confinement by carrying longitudinal and transverse load and at the same time providing
confinement (even in a better way). As a result, provision of longitudinal and transverse
reinforcement may be eliminated.
4
• Continuous confinement provided by the steel tubes prevents excessive spalling of concrete
and on the other hand, the concrete filled inside the steel tubes prohibits local inward buckling
of the steel tube wall, consequently providing efficient system.
• Suitable for high rise structures or structures that need wider free space because of their high
strength and flexibility.
• The ductility requirement for earthquake resistant design is easily achieved by using CFT than
the common reinforced concrete column or encased composite column.
• Vibrations caused by earthquakes and winds can be reduced due to its higher rigidity than that
of steel columns.
• Column sections in the composite structural system can be reduced because of its high
strength. It is also possible to keep the same column dimensions over several storey of a
building, which provides both functional and architectural advantages.
The excellent performance of structures with CFTs with respect to flexibility, suitability for high-rise
and ease of construction relative to other column types have been shown in Table 2.1 as presented by
Tarenaka Corporation [Tarenaka, 2001]. In the table comparison is made between CFT, RC
(Reinforced Concrete), EC (Encased Composite), and S (Steel) columns.
The main disadvantage of composite construction is the need to provide corrosion and fire resistant
coatings. Another minor drawback is that it is somewhat more complicated than other methods to
design and construct. Theses drawbacks need some consideration; however, are far outweighed by
the significant advantages that can be gained.
Applications
The beginning of frequent use of composite columns in tall building dates to about 1980. The
columns are designed to resist both gravity and lateral loads either alone or in combination with other
columns and shear walls [Ivan,1997].
5
RC EC S CFT
Flexibility
Rigidity, habitability
Fire resistance
Suitability for high-rise
structures
Workability
Excellent
Good
Fair
Table 2.1 Properties of structures with different columns
A number of buildings have been constructed with CFT columns worldwide. To mention a few are
the 1st First Street Plaza in San Francisco California [Roeder, et al], Sannomiya Grand Building in
Japan [Tarenaka, 2001], Millennium Tower in Austria Vienna [Gerald], and The Two Union Tower
in Seattle, USA [Ivan,1997]. (See Fig. 2.2)
CFTs are also used as bridge piers in different parts of the world [Hajjar,2000]. In Ethiopia there is a
limited use till today. One case worth mentioning is the Addis Ababa airport terminal, which makes
use of circular CFT columns.
Fig. 2.2 Photos of the Millennium Tower, 1st Street Plaza and Sannomia Grand buildings in
respective order from left to right
6
Structural Behaviour
Effect of confinement
The initial Poisson’s ratio of concrete (approximately 0.15 to 0.25) is below that of steel
(approximately 0.3). Thus, as the strain in the materials during the early loading is different there is
often little initial confinement of the concrete in a CFT. However, as the concrete begins to crack, it
expands faster than the steel tube and becomes well confined at higher load value [Ivan, 1997].
This confinement results in a higher load caring capacity. EBCS 4-1995 allows an increase of
concrete compressive capacity factor from 0.85fck to fck for all CFTs due to effect of confinement.
Confinement has a significant effect of increasing load capacity for circular sections. However, the
effect reduces with increase in slenderness of the column. The increase in concrete strength due to
confinement of the concrete core by the steel tube is valid only for columns with a slenderness ratio
below a certain limiting value as given in section 3.1.1.
Rectangular or square cross-sections, however, don’t have same capacity increase due to confinement
as circular ones. This is because, hoop stress is not significant as the plane faces deform when the
concrete expands.
Recently, a scheme has been proposed [Lliu, et al] as an alternative for improving the performance of square
CFTs. This scheme, called the “tie-bar stiffening scheme”, is carried out by welding sets of stiffeners , each of
which consists of four tie bars, with equal spacing along the tube axis as shown in Fig.2.3.
Fig. 2.3 Tie-bar stiffening scheme
This stiffening method results in increased confinement resulting in higher axial load capacity, enhanced
ductility and possibility of using thinner steel sections than allowed in codes.
7
Although the effect of confinement to increase load carrying capacity is limited for many cases, confinement
has a significant effect on the ductility of columns. After the maximum concrete compressive strength has
been reached, the steel tube prevents the concrete from spalling and the concrete core continues to
carry high stresses with increased strains, thereby increasing the ductility of the CFT column [Lahlou,
1999].
Bond between concrete and steel
Bond develops either from adhesion between concrete and steel or from friction due to normal stress.
Since adhesion is active mainly at the early stage of loading, it is assumed that bond between the
concrete and the steel is mainly contributed by friction [Ivan, 1997]. Friction develops between the
concrete core and the steel tube due to the coefficient of friction and normal contact pressure, which is
caused by lateral expansion of the concrete core when subjected to compressive loading. The
magnitude of the friction force developed in CFT columns depends on the rigidity of the tube walls
against pressure perpendicular to their plane.
The significance of bond on strength of the composite section depends of the condition of load
applications. When the load is applied at the column end only to the steel section or only to the
concrete section, the load must be transferred over the contact surface from the concrete core to the
steel tube or vise versa. For this transfer sufficient bond is necessary. Otherwise, the load will not be
shared between the concrete and steel in proportion to their strength.
When the load is applied to the entire section, the contributions by the concrete core and steel tube to
the total load carrying capacity are in proportion to strength and will be constant along the height of
the column, and thus, bond will not affect the strength significantly.
Experiments [Johansson, 2001] show that when the load is applied to the steel section, the steel tube
carried almost the whole load through the column and a higher coefficient of friction gave just a small
increase in the load resistance. That is, transfer of load from steel to concrete is hardly achieved by the
8
natural bond between concrete and steel. A better load transfer is observed when the load is applied to
the concrete, although not sufficient to assume full composite strength.
The transmission length of the shear force should not be assumed to exceed twice the transverse
dimension. If the bond is not sufficient to transfer the load within this length, it is necessary to provide
the top region of the steel tube with mechanical shear connectors at the inside to ensure full composite
action. Thus, effect of bond has to be considered for the design of connections. For the purpose of
calculation, the design shear strength due to bond for concrete-filled tube shall not be taken more than
0.4 Mpa. To take larger value, it should be verified by tests [EBCS 4-1995].
Experiments also indicate that bond strength is larger in circular versus rectangular CFTs, and that it
decreases with an increase in the depth to steel thickness ratio [Hajjar, 2000].
Section failure behaviour
The compressive axial strength of CFTs is governed by a combination of yielding of the steel and
crushing of the concrete. CFTs with small length to depth ratios typically fail near their cross-section
strength. Intermediate length or long (slender) CFT columns are governed by flexural instability,
usually involving at least some crushing of the concrete and yielding of the steel prior to buckling.
Failure of CFTs having relatively low width-to-thickness ratios and low-to-moderate strength concrete
typically occurs through a combination of yielding of the steel, local buckling of the steel, crushing of
the concrete, and flexural buckling of the member as a whole if it has sufficient length. Ductile
behavior generally results regardless of whether it is the steel or concrete which initiates inelasticity.
Failure of thin-walled steel tubes like CFTs with tubes having a width to thickness ratio greater than
approximately 60 or CFTs having high-strength concrete tends more towards local buckling of the
steel tube combined with a shear failure of the concrete. While the steel tube helps to delay shear
failure of the concrete, this is still a more brittle mode of failure. Consequently, most CFT
specifications throughout the world, currently specify a steel yield strength limit of the order of 380
MPa, a concrete strength limit of the order of 55 MPa, and limits on the width to thickness ratios of
9
the steel tube so as to help to insure that some ductile yielding of the steel generally occurs prior to its
local buckling or to crushing of the concrete [Hajjar, 2000].
Earthquake resistance
CFTs are suitable for structures which are constructed in seismic areas. Majority of the structures
constructed around the globe using CFTs are also in seismic prone areas of the world.
Researches conducted [Hajjar, 2000] to study the cyclic behavior of axially loaded CFT specimens
showed that the addition of concrete delayed local buckling and at the same time increased the
number of cycles to failure and the amount of energy dissipated. Concrete-filled steel tube beam–
columns subjected to combined axial force plus uniaxial or biaxial flexure typically show very full
hysteresis loops indicating large energy dissipation.
Fig. 2.4 illustrates typical behavior of a square CFT beam–column subjected to cyclic loading,
showing the load-deformation response [Hajjar, 2000]. A schematic illustration of the test setup is
shown in the inset. The figure shows that CFTs exhibit stable hysteretic behavior, although with some
evidence of degradation. Failure occurs through a combination of concrete crushing, steel yielding
and buckling, and in some cases eventual fracture in the steel tube in the regions of plastic hinges
10
Fig. 2.4 Cyclic load-deflection behavior of a CFT beam–column
Earthquake resistant design of reinforced concrete columns needs closely spaced stirrups to get the
required ductility. When high strength concrete is used the spacing needs to be further reduced
resulting in congestion of stirrups. However, in CFTs this problem is eliminated as there is no need
for stirrups.
CFTs with high strength concrete have excellent performance under cyclic loading. Experimental
investigation has been made [Lahlou, 1999] to measure the amount of stiffness degradation due to
dynamic loading and static loading. The stress-strain curves from dynamic test, which has a
characteristic of an elasto-plastic curve with strain hardening behavior, shows no loss of stiffness on
the ascending linear branch when compared with the test results from static loading. The residual
strength of the column that was severely tested under dynamic loading and then reloaded with a
monotonic static loading was observed to be around 80% of the strength of similar columns subjected
to static loading conditions only. This result indicates to what extent concrete columns confined in
steel tubes show enhanced structural behavior.
Cyclic loading observation on spirally confined reinforced concrete showed larger degree of stiffness
degradation caused by micro cracking leading to progressive damage. Whereas, CFTs during cyclic
11
loading, the concrete core consolidates in the tube instead of softening progressively through damage.
The concrete in the tube becomes stiffer through consolidation. The column continues to show a
linear behavior during unloading and reloading cycles. The cyclic behavior observed in CFTs is
similar to that of an elasto-plastic material such as steel rather than a brittle material such as
unreinforced concrete.
CFTs also show a great capacity for absorbing and dissipating energy input from dynamic load
excitations.
COLUMN LOAD CAPACITY
12
Axial Compression
Resistance of cross-section
Resistance of a composite cross-section subject to different loading is discussed in this chapter based
on EBCS 4, 1995. The cross-sectional resistance of a composite column to axial compression is the
aggregate of the plastic compression resistances of each of its constituent elements as follows:
N pl . Rd = As f yd + Ac f cd (3.1)
fy
where: f yd =
γM2
f ck
f cd =
γc
As and Ac are respectively the cross-sectional areas of the steel profile and the
concrete;
fy and fck are characteristic steel and concrete strength in accordance with EBCS 3-1995
and EBCS 2-1995, respectively;
γ M 2 and γ c are partial safety factors for steel and concrete respectively;
γ M 2 = 1.10
γ c= 1.50
The increase of concrete resistance from 0.85fck to fck for concrete-filled hollow sections as compared
with reinforced concrete and concrete encased columns is due to the effect of confinement (section
2.4).
For a concrete-filled circular hollow section, a further increase in concrete compressive resistance is
caused by hoop stress in the steel section. This happens only if the hollow steel profile is sufficiently
rigid to prevent most of the lateral expansion of the concrete under axial compression. This enhanced
concrete strength may be used in design when the relative slenderness λ (section 3.1.2) of the
composite column does not exceed 0.5 and the greatest bending moment Mmax.Sd (calculated using
first-order theory) does not exceed 0.1NSdd, where d is the external diameter of the column and NSd is
the applied design compressive force. The plastic compression resistance of a concrete-filled circular
section can then be calculated as:
13
 t 
N pl . Rd = Asη s f yd + Ac f cd 1 + η c f yd  (3.2)
 d 
in which t represents the wall thickness of the steel tube. The coefficients ηs and ηc are defined as
follows for 0 < e < d/10, where e=Mmax.Sd /NSd is the effective eccentricity of the axial compressive
force:
 e
η s = η so + (1 − η s 0 )10  (3.3)
 d
e
η c = η c 0 (1 − 10 ) (3.4)
d
When e > d/10 it is necessary to use ηs = 1.0 and ηc = 0. In equations (3.3) and (3.4) above the
terms ηso and ηc0 are the values of ηs and ηc for zero eccentricity e. They are expressed as functions
of the relative slenderness λ as follows:
η s 0 = 0.25(3 + 2λ ) <1 (3.5)
2
η c 0 = 4.9 − 18.5λ + 17λ >0 (3.6)
The presence of a bending moment MSd has the effect of reducing the average compressive stress in
the column at failure thus, reducing the favorable effect of hoop compression on its resistance. The
limits imposed on the values of ηs and ηc, and on ηs0 and ηc0, represent the effects of eccentricity and
slenderness respectively on the load-carrying capacity.
Relative slenderness
The elastic critical load Ncr of a composite column is calculated using the usual Euler buckling
equation
π 2 ( EI ) e
N cr = (3.7)
l2
in which (EI)e is the bending stiffness of the composite section about the buckling axis considered,
and l is the buckling length of the column. If the column forms part of a rigid frame this buckling
length can conservatively be taken equal to the system length.
14
For short-term loading the effective elastic bending stiffness (El)e of the composite section is given
by:
( EI ) e = E s I s + 0.8 E cd I c (3.8)
in which :
Is and Ic are the respective second moments of areas, for the bending plane considered, of the steel
section and the uncracked concrete section;
Es is the elastic modulus of steel;
Ecd = Ecm / γ c ;
Ecm is the secant modulus of concrete;
γc =1.35 is the safety factor for stiffness;
For long-term loading the bending stiffness of the concrete is determined by replacing the elastic
modulus Ecd with a lower value Ec which allows for the effect of creep and is calculated as follows:
N G , sd
E c = E cd [1 − 0.5 ] (3.9)
N sd
where NG.Sd is the permanent part of the axial design loading NSd.
Long term loading modification of the concrete modulus is only necessary if:
the relative slenderness λ , for the plane of bending considered, is greater than 0.8/(1-δ),
As f yd
where δ = is the relative contribution of the steel section to the overall axial plastic
N pl . Rd
resistance. It should be noted that the calculation of λ requires knowledge of an initial value
of the elastic modulus Ec of concrete. For checks against the limits given above it is
permissible to calculate λ without considering the influence of long-term loads.
the relative eccentricity e/d (d being the depth of the section in the plane of bending
considered) is less than 2.
15
This limiting value applies in the case of braced non-sway frames. It is replaced by 0.5/(1-δ) in the
case of sway frames or unbraced frames. The relative slenderness λ of a composite column in the
plane of bending considered is given by:
N pl . Rk
λ= (3.10)
N cr
in which Npl.Rk is the value of the plastic resistance Npl.Rd calculated using material partial safety
factors γMa and γc set equal to 1.0 (or, using the characteristic material strengths).
Buckling resistance
A composite column has sufficient resistance to buckling if, for each of the planes of buckling, the
design axial loading NSd satisfies the inequality:
N Sd ≤ χN pl . Rd (3.11)
in which the value of χ, the strength reduction factor in the plane of buckling considered, is a function
of the relative slenderness λ and the appropriate buckling curve. The buckling curves which apply to
composite Concrete filled columns is “curve a” which is given in [EBCS 3-1995].
It is also possible to calculate the value of the strength reduction factor χ using:
1
χ= 2
≤1 (3.12) in
φ + [φ 2 − λ ]1 / 2
which
2
φ = 0.5[1 + α (λ − 0.2) + λ ] (3.13)
where α=0.21 is a generalized imperfection parameter which allows for the unfavorable effects of
initial out-of-straightness and residual stresses.
Resistance to Compression and Bending
Cross-section resistance under uniaxial moment and axial compression
16
It is necessary to satisfy the resistance requirements in each of the principal planes, taking account of
the slenderness, the bending moment diagram and the bending resistance in the plane under
consideration. The cross-sectional resistance of a composite column under axial compression and
uniaxial bending is given by an M-N (moment – axial force) interaction curve as shown in Figure 3.1.
N
A
Npl,Rd
E
Npm,Rd C
0,5 Npm,Rd D
B
M
0 M pl,Rd M max,Rd
Fig. 3.1 M-N interaction curve for uniaxial bending
The above interaction curve can be determined point by point, by considering different plastic neutral
axis positions in the principal plane under consideration. The concurrent values of moment and axial
resistance are then found from the stress blocks. Fig. 3.2 illustrates this process for four particular
positions of the plastic neutral axis corresponding respectively to the points A, B, C, D marked on
Fig. 3.1.
• Point A : Axial compression resistance alone:
N A = N pl .Rd MA = 0
• Point B : Uniaxial bending resistance alone:
NB = 0 M B = M pl .Rd
• Point C : Uniaxial bending resistance identical to that at point B, but with non-zero resultant
axial compressive force:
N C = N pm. Rd
where: Npm,Rd = Compressive resistance of the concrete section

=Ac fcd
17
M C = M pl .Rd
t fy
Note: fcd may be factored by [ 1 + η c ] for a circular concrete-filled hollow section.
d f ck
• Point D : Maximum moment resistance
1 1
N = N pm. Rd = Ac f cd
D
2 2
1
M D = W pa f yd + W pc f cd
2
in which Wps and Wpc are the plastic moduli respectively of the steel section and the concrete.
Point D corresponds to the maximum moment resistance Mmax,Rd that can be achieved by the
section. This is greater than Mpl.Rd because the compressive axial force inhibits tensile cracking of
the concrete, thus enhancing its flexural resistance.
• Point E : Situated midway between A and C.
The enhancement of the resistance at point E is little more than that given by direct linear
interpolation between A and C, and determination of this point can therefore be omitted.
It is usual to substitute the linearised version AECDB (or the simpler ACDB) shown in Figure 3.1 for
the more exact interaction curve, after doing the calculation to determine these points.
18
N pl,Rd
M pl,Rd
N pm,Rd
M pl,Rd
N pm,Rd/2
M max,Rd
Fig. 3.2 Development of stress blocks at different points on the interaction curve
19
The influence of shear force
It is permissible to assume for simplicity that the design transverse shear force VSd is completely
resisted by the steel section. Alternatively it is possible to distribute it between the steel section and
the concrete; in this case the shear force carried by the concrete is determined by the method given in
EBCS 2-1995.
The interaction between the bending moment and shear force in the steel section can be taken into
account by reducing the limiting bending stresses in the zones which are affected by significant shear
force. This reduction of yield strength in the sheared zones can be represented, for ease of
calculation, by a reduction in the thickness of the element(s) of the steel section which carries the
shear force. This influence need only be taken into consideration if the shear force carried by the
steel section, Vs.Sd , is greater than 50% of its plastic shear resistance:
V pl , s , Rd = Av f yd / 3 (3.15)
where Av is the sheared area of the steel section as given in [EBCS 3-1995]. The reduction factor
which may need to be applied to this area is:
  2V  
2
ρ w = 1 −  s . Sd
− 1 
 
(3.16)
  V pl , s. Rd  

Therefore, the effective area of part of the section assumed resisting the shear is given by: ρ w Av .
When the reduced area ρ w Av is used, the method described in Section 3.2.1 for determination of the
resistance interaction curve for the cross-section can be applied freely.
Slender member resistance under axial compression and uniaxial bending
The principle of the EBCS 4-1995 calculation method for member resistance under axial load and
uniaxial moment is demonstrated schematically in Fig. 3.3, which is a normalized version of the
interaction diagram of cross-sectional resistance in Fig. 3.1. For a design axial compression NSd
the plastic section resistance MRd, which is a proportion µ d of the fully plastic resistance Mpl.Rd, is
indicated by the interaction curve.
20
N/Npl.Rd
Resistance locus of the cross-section
1.0
χ d= NSd/Npl.Rd
Limiting value
MSd/Mpl.Rd < 0.9 µd
µd= MRd/Mpl,Rd
0 1.0 M / Mpl.Rd
Fig. 3.3 Resistance to axial compression and uniaxial bending
The design moment MSd is the maximum moment occurring within the length of the column,
including any enhancement caused by the column imperfections and amplification of the total first-
order moments due to the second-order “P-D” effect.
Under the design axial force NSd, a composite column has sufficient resistance if
M Sd ≤ 0.9 µ d M pl , Rd (3.17)
The 10% reduction in resistance indicated by the introduction of the factor 0.9 compensates for the
underlying simplifications in the calculation method. For example, the interaction curve has been
established without considering any limits on the deformations of concrete. Consequently, the
bending moments, including the second-order effects, are calculated using the effective bending
stiffness (EI)e determined on the basis of the complete concrete cross-sectional area.
It is evident from Figure 3.3 that values of µd taken from the interaction diagram may be in excess of
1.0 in the region around point D, where a certain level of axial compression increases the moment
capacity of the section. In practice, values of µd above 1.0 should not be used unless the moment MSd
is directly caused by the axial force NSd, acting at a fixed eccentricity on a statically determinate
column. This is because of the fact that the moment may be coupled with an axial force that is lower
than the assumed one.
21
Member resistance under axial compression and biaxial bending
When a composite column is subjected to axial compression together with biaxial bending, it is first
necessary to check its resistance under compression and uniaxial bending individually in each of the
planes of bending. This is not however sufficient, and it is necessary also to check its biaxial bending
behavior. In doing so it is only necessary to take account of imperfections in the plane in which
failure is likely to take place. For the other plane of bending the effect of imperfections is neglected.
This can be represented by the two simultaneous conditions:

M y .Sd ≤ 0.9 µ y M pl . y .Rd (3.18)
M z .Sd ≤ 0.9 µ z M pl . z .Rd (3.19)
If there is any doubt about the plane of failure the designer is recommended to consider the effect of
imperfections in both planes.
To take account of the peak stresses caused by moments between the limits given by the inequalities
(3.18) and (3.19), acting about two orthogonal axes, a linear interaction formula must also be satisfied
between the two design moments. The design moments are calculated including both imperfections
and the amplification due to second-order “P-D” effects.
M y , Sd M z , Sd
+ ≤ 1.0 (3.20)
µ y M pl , y , Rd µ z M pl , z , Rd
These three conditions of equations 3.18, 3.19 & 3.20 together define the ultimate strength locus in
terms of the orthogonal design moments at the design axial compression value NSd as shown in
Fig. 3.4(c).
22
NRd/Npl.Rd NRd/Npl.Rd
1.0 (a) 1.0 (b)
NSd/Npl.Rd NSd/Npl.Rd
0.9µ y 0.9µ z
My,Rd /Mpl.y.Rd Mz,Rd/Mpl.z.Rd

0 µ d 1.0 0 µ d 1.0
0 0.9µ y
My.Rd/Mpl.y.Rd
µy (a) Section resistance interaction
diagram - (y-y)
(b) Section resistance interaction
diagram – (z-z)
(c) Biaxial bending resistance
(c)
locus of the column section
0.9µ z
under axial compression NSd.
µz
Mz.Rd/Mpl.z.Rd
Fig. 3.4 Member resistance under compression and biaxial bending
Shortcomings of EBCS 4-1995 Design Procedure
One of the reasons for the limited applicability of CFTs is attributed to the tiresome calculations in
design.
Designing CFT cross-section under axial compression and uniaxial moment using the Code [EBCS 4-
1995] has the following problems.
- Drawing charts is necessary which is time consuming and difficult
- Design procedure is trial and error by which one need to draw an interaction curve for each
trial section.
- The result is approximate since the curves are constructed from only four or five points.
- It is not easy to calculate the plastic moment capacity as it needs computation of
corresponding neutral axis position.
- The computation of section capacity for section like hexagon and octagon is more difficult.
- The equation for circular section given in the code is approximate.
23
For the design of biaxially loaded CFT the method used in the Code is approximate. This inaccuracy
will make the design more conservative and uneconomical. The problem will be even more amplified
as CTF columns are more expensive in comparison to reinforced concrete sections.
Solutions to the above problems are provided in the next chapter.
CHART DEVELOPMENT
Calculation Method and Scope
24
The cross sections considered in this thesis are those that fulfill the criteria for simplified method
of analysis given in the Code [EBCS 4-1995]. The criteria are:
• The column cross-section must be prismatic and symmetric about both axes over its whole height,
with its ratio of overall orthogonal cross-sectional dimensions in the range between 0.2 and 5.
Particularly, the sections considered are square, rectangular, circular, hexagonal and octagonal.
• The relative contribution of the steel section to the design resistance of the composite section,
given by δ = As f yd / N pl . Rd , must be between 0.2 and 0.9 or w given by w = As f yd / Ac f cd must be
between 0.25 and 9.

• The relative slenderness λ of the composite column must be less than 2.0;
In addition, the slenderness of the elements of the steel section must satisfy the following
conditions.
• for circular hollow sections of diameter d and wall thickness t, d / t ≤ 90 ε 2 ;
• for rectangular hollow sections of wall depth h and thickness t, h / t ≤ 52 ε ;
ε = 235 / f y , where fy is the yield strength of the steel section in Mpa.
Selected Representative Sections & Materials
From the allowable steel ratios w those below 4 have been selected for drawing chart as this range
utilize a smaller area of steel which the writer assumes will be more economical sections than those
sections with higher steel ratio. The rectangular sections considered are those with height-to-width
ratio of 0.5 and 2.
Circular cross-sections considered are those with slenderness ratio λ greater than 0.5 or those for
which increase in strength due to confinement is not considered.
Biaxial design chart is prepared for columns with square cross-section. The axial load ratios selected
are 0 and 1 where the axial load ratio is given by:
25
N sd
ν= (4.1)
Ac f cd
However, the method outlined and equations written may be extended to the other cross-
sections and material types.
The structural steel grades that can be used are recommended by the Code [EBCS 4-1995]. From the
steel grades, Fe360 with thickness of the steel section less than or equal to 40mm has been selected.
The Code allows use of concrete grade up to C-60. From these, concrete grade of C-30 has been used
in this research.
Interaction Chart for Axial Compression and Uniaxial Bending

Fundamental equations
Interaction charts are drawn using the stress blocks that show the plastic section capacity of composite
cross-sections. The fundamental equations used are given below with respect to typical composite
cross-section shown in Fig. 4.1.
Region
Fig. 4.1 Composite cross-section regions used for computing section capacity
As f yd
Steel Ratio; w = (4.1)
Ac f cd
f cd
Moment Capacity; M u = (W pa − W pan ) f yd + (W pc − W pcn ) (4.2)
2
Axial Capacity; N u = Acc f cd + Asnet f yd (4.3)
26
Where: As and Ac are area of the total steel and concrete sections, respectively;
Wpa and Wpc are plastic section moduli of the total steel and concrete sections, respectively;
Wpan and Wpcn are plastic section moduli for the steel and concrete sections parts with in
region 2, respectively;
Asnet = Asc- Ast
Asc and Acc are area of part of the steel and concrete sections in compression, respectively
and,
Ast is area of part of the steel section in tension.
Rectangular (Square) section
Fig. 4.2 Rectangular cross-section
Determining value of x=b/t for a particular steel ratio w

As f yd
w=
Ac f cd
=> As f yd = wAc f cd
but, Ac = b' h' where b' = b − 2t and h' = h − 2t
As = bh − A c
Let h/b = y
Substituting the above in the equation for w and dividing both equations by t2
[ x 2 y − ( x − 2)( xy − 2)] f yd = w( x − 2)( xy − 2) f cd
27
Simplifying the above equation and solving for x
( f yd + wf cd )(1 + y ) + {( f yd + wf cd )(1 + y )}2 − 4wf cd ( wf cd y + f yd )

x=
wf cd y
For square section where y = 1
2( f yd + wf cd ) + {2( f yd + wf cd )}2 − 4wf cd ( wf cd + f yd )

x=
wf cd
Moment and axial load capacity for different neutral axis positions
Moment and axial load capacity can be computed from assumed neutral axis positions. Each position
of the neutral axis represents one point in the interaction curve for section capacity. Sufficient points
are developed to get a smooth curve that represents the capacity of a given cross-section. Four
different cases of neutral axis position are selected.
Case i
The whole cross section under compression.
a) Moment capacity
Since the whole part is in compression the moment capacity is zero.
b) Axial load capacity
N u = N pl ,rd
N pl , rd = A c f cd + A s f yd
Ac = b'h' & As = bh − Ac
Ac f cd + As f yd
=> ν =
Ac f cd
Dividing equation by t2 and simplifying
( x − 2)( xy − 2) f cd + {x 2 y − ( x − 2)( xy − 2)} f yd
v=
( x − 2)( xy − 2) f cd
Case ii
Some part of the flange of the steel section on tension while other parts of the cross-section in
compression.
28
h h
Fig. 4.3 Neutral axis position − t ≤ hi ≤
2 2
a) Moment Capacity
f cd
M u = (W pa − W pan ) f yd + (W pc − W pcn )
2
but in this case only the steel contributes for the moment
M u = (W pa − W pan ) f yd
bh 2
W pa =
4
b(h − 2u ) 2
W pan =
4
h
where u = − hi
2
Mu 0.25{(bh 2 − b(h − 2u ) 2 } f yd
=> µ = =
Ac f cd h' b' h' f cdh'
Diving both the numerator and denominator by t3
0.25{x 3 y 2 − x( xy − 2u / t ) 2 } f yd
µ=
( x − 2)( xy − 2) 2 fcd

N u = N pl ,rd − 2buf yd
29
Ac = b' h' & As = bh − b' h'
Ac f cd + As f yd 2buf yd
=> ν = −
Ac f cd Ac f cd
Dividing equation by t2
( x − 2)( xy − 2) f cd + {x 2 y − ( x − 2)( xy − 2)} f yd 2 x(u / t ) f yd
v= −
( x − 2)( xy − 2) f cd ( x − 2)( xy − 2) f cd
c) Values of hi/t used
Values of hi/t used are 0.5xy-1, 0.5xy-2/3 and 0.5xy-1/3
Case iii
More than half the area under compression
h
Fig. 4.4 Neutral axis position 0 ≤ hi ≤ −t
2
a) Moment capacity
f cd
2
bh 2 b ' h ' 2
W pa = −
4 4
2 b' h' 2 2
W pan = (b − b' )hi , W pc = , W pcn = b' hi
4
bh 2 b' h' 2 2 2
{( − ) − (b − b' )hi } f yd + (0.25b' h' 2 −b' hi ) f cd / 2
Mu 4 4
=> µ = =
Ac f cd h' b' h' 2 f cd
30
0.25{x 3 y 2 − ( x − 2)( xy − 2) 2 − 2(hi / t ) 2 } f yd + {0.25( x − 2)( xy − 2) 2 − ( x − 2)(hi / t ) 2 } f cd / 2
µ= b)
( x − 2)( xy − 2) 2 f cd
Axial load capacity
N u = b' (h' / 2 + hi ) f cd + 4hi tf yd
Nu b' (h' / 2 + hi ) f cd + 4hi tf yd
=> v = =
Ac f cd b' h' f cd
( x − 2)( xy / 2 − 2 + hi / t ) f cd + 4(hi / t ) f yd
v=
( x − 2)( xy − 2) f cd

Values of hi/t used are 0, 0.1xy, 0.2xy, 0.3xy and 0.4xy
Case iv
Less than half the area under compression
h
Fig. 4.5 Neutral axis position 0 ≤ hi ≤ −t
2
a) Moment capacity
Moment capacity for this case is equal to that given in case iii

N u = b' (h' / 2 − hi ) f cd − 4hi tf yd
Nu b' (h' / 2 − hi ) f cd − 4hi tf yd
=> v = =
Ac f cd b' h' f cd
31
( x − 2)( xy / 2 − 1 − hi / t ) f cd − 4(hi / t ) f yd
Diving both the numerator and denominator by t2 v = ( x − 2)( xy − 2) f cd

Values of hi/t used are 0.1xy, 0.2xy, 0.3xy and 0.4xy
Circular section
Fig. 4.6 Circular cross-section
Determining value of x=d/t for a particular steel ratio

As f yd
w=
Ac f cd
π d '2
but, Ac = where d ' = d − 2t
4
π d2
As = − Ac
4
π
Substituting the above in the equation for w and dividing both equations by t2
4
[ x 2 − ( x − 2) 2 ] f yd = w( x − 2) 2 f cd
2( f yd + wf cd ) + {2( f yd + wf cd )}2 − 4 wf cd ( wf cd + f yd )
x=
wf cd
This is same as the corresponding equation for square section
32
Case i
The whole cross section under compression.
a) Moment capacity
N u = N pl ,rd
π d '2 π d2
Ac = & As = − Ac
4 4
Ac f cd + As f yd
=> ν =
Ac f cd
Dividing equation by t2 and simplifying
( x − 2) 2 f cd + {x 2 − ( x − 2) 2 } f yd
v=
( x − 2) 2 f cd
Case ii
c'
c
d
Fig. 4.7 Neutral axis position for 0 ≤ hi ≤ −t
2
Area and Section Modulus for a Segment
33
Segment
Fig. 4. 8 Segment of a Circle
From Fig. 4. 8
2hi
c = cos −1 ( )
d
Area of segment: Aseg = c (d / 2) 2 − hi2 tan c
2 d3
Section modulus of the segment: Wseg = ( sin c − hi3 tan c)
3 8
Thus, applying these equations for concrete filled tubes (Fig. 4.7)
2 d '3
Section modulus of concrete part of the segment: Wsegc = ( sin c'− hi3 tan c' )
3 8
2hi
where c' = cos −1 ( )
d'
2 d3
And, section modulus of steel part of the segment is given by Wsega = ( sin c − hi3 tan c) − Wsegc
3 8
a) Moment capacity
M u = 2 × Wsega f yd + Wsegc f cd
1 2 3 2 d '3
2 × { ( d 3 sin c − d ' 3 sin c' ) − (tan c − tan c' )hi } f yd + ( sin c'− hi3 tan c' ) f cd
Mu 12 3 3 8
=> µ = =
Ac f cd
d' π d '3
f cd
2 8
34
1 2 2 ( x − 2) 3
2 × { ( x 3 sin c − ( x − 2) 3 sin c' ) − (tan c − tan c')(hi / t ) 3 } f yd + ( sin c' − (hi / t ) 3 tan c') f cd
=> µ = 12 3 3 8 where
π ( x − 2) 3
f cd
8
2(hi / t )
c = cos −1{ }
x
2(hi / t )
c' = cos −1{ }
( x − 2)
N u = Acc f cd + Asnet f yd
Acc = π (d ' / 2) 2 − {c' (d ' / 2) 2 − hi2 tan c'}
Asnet = 0.25π (d 2 − d ' 2 ) − 2[{c (d / 2) 2 − hi2 tan c} − {c' (d ' / 2) 2 − hi2 tan c'}]
Nu
=> v =
Ac f cd
Substituting the above equations into the equation for v and dividing both the numerator and
denominator by t2
π (( x − 2) / 2) 2 − {c' (( x − 2) / 2) 2 − (hi / t ) 2 tan c'} f cd

=> v = +
π (( x − 2) / 2) 2 f cd
.25π ( x 2 − ( x − 2) 2 ) − 2[{c ( x / 2) 2 − (hi / t ) 2 tan c} − {c'(( x − 2) / 2) 2 − (hi / 2) tan c'}]
π (( x − 2) / 2) 2 f cd
Values of hi/t used are 0, 0.1x, 0.2x, 0.3x and 0.4x
Case iii
d
Fig. 4.9 Neutral axis position with 0 ≤ hi ≤ −t
2
35
a) Moment capacity
The moment capacity is the same as that given in case ii
Acc = {c' (d ' / 2) 2 − hi2 tan c'}
Asnet = [0.25π (d 2 − d ' 2 ) − 2[{c (d / 2) 2 − hi2 tan c} − {c' (d ' / 2) 2 − hi2 tan c'}]
Nu − {c' (( x − 2) / 2) 2 − (hi / t ) 2 tan c'} f cd
=> v = = +
Ac f cd π (( x − 2) / 2) 2 f cd
0.25π ( x 2 − ( x − 2) 2 ) − 2[{c ( x / 2) 2 − ( hi / t ) 2 tan c} − {c'(( x − 2) / 2) 2 − (hi / 2) tan c'}]
π (( x − 2) / 2) 2 f cd
Values of hi/t used are 0, 0.1x, 0.2x, 0.3x and 0.4x
Hexagonal section
Fig. 4.10 Hexagonal CFT cross-section
Determining value of x=s/t for a particular steel ratio w
As f yd
w=
Ac f cd
but, Ac = 2.598s ' 2 where s ' = s − 1.155t
36
As = 2.598s 2 − A c
Substituting the above equations in to the equation for w and dividing the numerator and denominator
by 2.598t 2
[ x 2 − ( x − 1.155) 2 ] f yd = w( x − 1.155) 2 f cd
2.31( f yd + wf cd ) + {2.31( f yd + wf cd )}2 − 5.333wf cd ( wf cd + f yd )

x=
2 wf cd
1. Bending about Y-axis
Area and Section Modulus
For hexagonal section shown in Fig.4. 11:
- area of the whole section is A = 2.598s 2 ,
- the area of the polygon ABCDEF is A = 3.464 s 2 − (1.732 s − hi )(2 s − 1.155hi )
A F
B E
D
C
Fig. 4.11 Hexagon of side length s
- section modulus of the whole section is W = s 3 , and

- section modulus of the polygon ABCDEF is W = 2 s 3 − (1.732 s − hi )(2 s − 1.155hi )(0.577 s + .667 hi )
Case i
The whole cross-section is under compression.
a) Moment capacity
37
a) Axial load capacity

N u = N pl ,rd
Ac = 2.598s ' 2 & As = 2.598s 2 − Ac
Ac f cd + As f yd
=> ν =
Ac f cd
( x − 1.155) 2 f cd + {x 2 − ( x − 1.155) 2 } f yd
=> v =
( x − 1.155) 2 f cd
Case ii
Some part of the flange of the steel section on tension while other parts of the cross section in
compression.
3s 3s
Fig. 4.12 Neutral axis position with − t ≤ hi ≤
2 2
a) Moment capacity
f cd
2
but, in this case only the steel contributes for the moment
M u = (W pa − W pan ) f yd
38
W pa = s 3
W pan = 2 s 3 − (1 . 732 s − h i )( 2 s − 1 . 155 h i )( 0 . 577 s + . 667 h i
Mu [ s 3 − {2s 3 − (1.732s − hi )(2 s − 1.155hi )(0.577 s + .667 hi )}] f yd

=> µ = =
Ac f cd s ' 2.598s ' 3 f cd
[ x 3 − {2 x 3 − (1.732 x − (hi / t ))(2 x − 1.155 (hi / t ))(0.577 x + .667 ( hi / t ))}] f yd
µ=
2.598 ( x − 1.155) 3 f cd

N u = 2.598s' 2 f cd + [{3.464 s 2 − (1.732 s − hi )(2s − 1.155hi )} − 2.598s ' 2 ] f yd
2.598s ' 2 f cd + [{3.464s 2 − (1.732s − hi )(2s − 1.155hi )} − 2.598s ' 2 ] f yd

=> ν =
Ac f cd
2.598( x − 1.155) 2 f cd + [{3.464 x 2 − (1.732 x − (hi / t ))(2 x − 1.155(hi / t ))} − 2.598( x − 1.155) 2 ] f yd
v= c)
2.598( x − 1.155) 2 f cd
Values of hi/t used

Values of hi/t used are 1.732x-1, 1.732x-2/3 and 1.732x-1/3
Case iii
3s
Fig. 4.13 Neutral axis position with 0 ≤ hi ≤ −t
2
a) Moment capacity
f cd
2
W pa = s 3 − ( s − 1 . 155 ) 3
39
2
W pan = 2.31hi t
W pc = s '3
W pcn = 2 s ' 3 −{(1.732 s '− hi )(2 s '−1.155s ' )(0.557 s '+0.667 hi )}
Mu
=> µ =
Ac f cd s '
2
{s 3 − s' 3 −2.31hi t} f yd + {− s' 3 +{1.732 s'− hi )(2 s'−1.155hi )(.577 s '+.667 hi )} f cd / 2
=
2.598s ' 3 f cd
hi 2
{x 3 − ( x − 1.155) 3 − 2.31( ) } f yd
µ= t +
2.598( x − 1.155) 3 f cd
hi h h
{−( x − 1.155) 3 + {1.732( x − 1.155) − ( ))( 2( x − 1.155) − 1.155( i ))(0.577( x − 1.155) + 0.667( i )} f cd / 2
t t t
2.598( x − 1.155) 3 f cd

N u = [2.598s' 2 −0.5{(2s '−1.155hi ) + ( s − 1.155t )}{.866 s − t − hi }] f cd + 4.62hi tf yd
[2.598s '2 −0.5 * {( 2 s '−1.155hi ) + s '}{0.866 s − t − hi }] f cd + 4.62hitf yd

=> ν =
Ac f cd
hi h h
[2.598( x − 1.155) 2 − 0.5{( 2( x − 1.155) − 1.155 ) + ( x − 1.155)}{0.866 x − 1 − i }] f cd + 4.62 i f yd
v= t t t
2.598( x − 1.155) 2 f cd

Values of hi/t used are 0, 0.2x, 0.4x, 0.6x, 0.8x, x, 1.3x and 1.5x
Case iv
40
3s
Fig. 4.14 Neutral axis position with 0 ≤ hi ≤ −1
2
a) Moment capacity
The moment capacity is the same as that given in case iii
N u = .5{(2 s '−1.155hi ) + ( s − 1.155t )}{.866 s − t − hi } f cd − 4.62hi tf yd
0.5{(2 s '−1.155hi ) + s ' }{.866 s − t − hi } f cd − 4.62hi tf yd

=> ν =
Ac f cd
hi h h
0.5{(2( x − 1.155) − 1.155 ) + ( x − 1.155)}{.866 x − 1 − i } f cd − 4.62 i f yd
v= t t t
2
2.598( x − 1.155) f cd
Values of hi/t used are 0, 0.2x, 0.4x, 0.6x, 0.8x, x, 1.3x and 1.5x
2. Bending about X-axis

For hexagonal section shown in Fig.4.15,
- area of the total section is A = 2.598s 2 ,
- area of the polygon ABC is Atri = 1.732( s − h' i ) 2 ,
- the area of the polygon DEFG is given by A = 3.464 shi ,
41
Fig. 4.15 Hexagon of side length s
- section modulus of the total section is W = 1.01s 3 ,

- section modulus of the polygon ABC is Wtri = 0.577( s − h' i ) 2 ( s − 2h'i ) ,
2
- section modulus of the polygon DEFG is W = 1.732shi .
Case i
a) Moment capacity

N u = N pl ,rd
Ac = 2.598s ' 2 & As = 2.598s 2 − Ac
Ac f cd + As f yd
=> ν =
Ac f cd
( x − 1.155) 2 f cd + {x 2 − ( x − 1.155) 2 } f yd
=> v =
( x − 1.155) 2 f cd
Case ii
More than half the area under compression (Fig. 4.16)
42
s
Fig. 4.16 Neutral axis position with ≤ hi ≤ s − 1.155t
2
a) Moment capacity
f cd
2
W pa − W pan = 2 × Wtri = 1.155( s − hi ) 2 ( s − 2hi ) − 1.155( s'−hi ) 2 ( s'−2hi )
W pc − W pcn = 2 × Wtri = 1.155( s'−hi ) 2 ( s'−2hi )
Mu 1.155( s − hi ) 2 ( s − 2hi ) − 1.155( s '− hi ) 2 ( s '−2hi ) f yd + {1.155( s '−hi ) 2 ( s'−2hi )} f cd / 2

=> µ = = Diving
Ac f cd s ' 2.598s ' 3 f cd
both the numerator and denominator by t3

hi 2 h h h
1.155( x − ) ( x − 2 i ) − 1.155( x − 1.155 − i ) 2 ( x − 1.155 − 2 i ) f yd
µ= t t t t +
3
2.598( x − 1.155) f cd
hi 2 h
{1.155( x − 1.155 − ) ( x − 1.155 − 2 i )} f cd / 2
t t
3
2.598( x − 1.155) f cd
Acc = Ac − Atri = 2.598s ' 2 −1.732( s '− hi ) 2
Asnet = Asc − Ast = As − 2 Atri
Asnet = 2.598( s 2 − s ' 2 ) − 3.464{( s − hi ) 2 − ( s '− hi ) 2 }
Nu {2.598s '−1.732( s '− hi ) 2 } f cd + [2.598( s 2 − s ' 2 ) − 3.464{( s − hi ) 2 − ( s '− hi ) 2 }] f yd
v= =
Ac f cd 2.598s ' f cd
43
hi 2
{2.598( x − 1.155) 2 − 1.732( x − 1.155 − ) } f cd
v= t
2.598( x − 1.155) 2 f cd
hi 2 h
[2.598{x 2 − ( x − 1.155) 2 } − 3.464{( x − ) − ( x − 1.155 − i ) 2 }] f yd
+ t t
2
2.598( x − 1.155) f cd

Values of hi/t used are 0.5x, 0.6x, 0.8x and x-1.155
Case iii
s'
Fig. 4.17 Neutral axis position with 0 ≤ hi ≤
2
a) Moment capacity
f cd
2
W pa = 1.01( s 3 − s ' 3 )
W pan = 1.732( s − s ' )hi2 = 2thi2
W pc = 1.01s ' 3
W pcn = 1.732 s ' hi2
Mu {1.01( s 3 − s '3 ) − 2thi2 } f yd + {1.01s' 3 −1.732 s ' hi2 } f cd / 2

=> µ = =
Ac f cd s ' 2.598s ' 3 f cd
44
hi 2 h
{1.01( x 3 − ( x − 1.155) 3 ) − 2( ) } f yd + {1.01( x − 1.155) 3 − 1.732( x − 1.155)( i ) 2 } f cd / 2
µ= t t b) Axial
3
2.598( x − 1.155) f cd
load capacity
Acc = 0.5 × 2.598s ' 2 +1.732s ' hi = 1.299 s ' 2 +1.732 s ' hi
Asnet = 4hi t
Nu {1.299 s ' 2 +1.732 s ' hi } f cd + 4hi tf yd
v= =
Ac f cd 2.598s ' 2 f cd
hi h
{1.299( x − 1.155) 2 + 1.732( x − 1.155) } f cd + 4 i f yd
v= t t
2.598( x − 1.155) 2 f cd

Values of hi/t used are 0x, 0.1x, 0.3x and (x-1.155)/2
Case iv
Less than half the area under compression (Fig. 4.18)
a) Moment capacity
The moment capacity is the same as that given in case iii
s'
Fig. 4.18 Neutral axis position with 0 ≤ hi ≤
2
45
Acc = 0.5 × 2.598s ' 2 −1.732s ' hi = 1.299 s ' 2 −1.732 s ' hi
Asnet = −4hi t
Nu {1.299 s ' 2 −1.732 s ' hi } f cd − 4hi tf yd
v= =
Ac f cd 2.598s ' 2 f cd
hi h
{1.299( x − 1.155) 2 − 1.732( x − 1.155) } f cd − 4 i f yd
v= t t
2.598( x − 1.155) 2 f cd

Values of hi/t used are 0, 0.1x, 0.3x and (x-1.155)/2
Case v)
a) Moment capacity
The moment capacity is the same as that given in case ii
Acc = Atri = 1.732( s '−hi ) 2
Asnet = 2.598( s 2 − s ' 2 ) − 3.464{( s − hi ) 2 − ( s '− hi ) 2 }
Nu 1.732( s '− hi ) 2 f cd + [2.598( s 2 − s ' 2 ) − 3.464{( s − hi ) 2 − ( s '− hi ) 2 }] f yd
v= =
Ac f cd 2.598s ' 2 f cd
hi 2 h h
1.732( x − 1.155 − ) f cd + [2.598( x 2 − ( x − 1.155) 2 ) − 3.464{( x − i ) 2 − ( x − 1.155 − i ) 2 }] f yd
v= t t t
2.598( x − 1.155) 2 f cd

Values of hi/t used are 0.5, 0.6x, 0.8x and x-1.155
46
s
Fig. 4.19 Neutral axis position with ≤ hi ≤ s '
2
Octagonal section
Fig. 4.20 Octagonal CFT cross-section
tra
Fig. 4.21 Octagonal section
47
For Octagonal section shown in the above figure,
- area of the whole section is A = 4.828s 2
- the area of the polygon is ABCD A = 5.328s 2 − 0.5(3.414 s − 2hi ) 2
- the area of polygon ABEF (Atra) is A = 0.5(4.414 s − 2hi )(1.207 s − hi )
- section modulus of the whole section is W = 2.545s 3

- and section modulus of polygon ABCD is
W = 2{1.616 s 3 − .25(3.414 s − 2hi ) 2 (0.569 s + 0.667 hi )}
Determining value of x=s/t for a particular steel ratio
As f yd
w=
Ac f cd
but, Ac = 4.828s ' 2 where s ' = s − 0.828t
As = 4.828s 2 − A c
Substituting the above equation in equation of w and dividing both equations by 4.828t 2
[ x 2 − ( x − 1.155) 2 ] f yd = w( x − 1.155) 2 f cd
1.656( f yd + wf cd ) + {1.656( f yd + wf cd )}2 − 2.744 wf cd ( wf cd + f yd )

x=
2 wf cd
Case i
a) Moment capacity
As the whole part is in compression the moment capacity is zero.
48
N u = N pl ,rd N pl , rd = A c f cd + A s f yd
Ac = 4.828s ' 2 and As = 4.828s 2 − Ac
Ac f cd + As f yd
=>ν =
Ac f cd
Substituting equations and dividing equation by t2 and simplifying
( x − 0.828) 2 f cd + {x 2 − ( x − 0.828) 2 } f yd
v=
( x − 0.828) 2 f cd
Case ii
s
Fig. 4.22 Neutral axis position with ≤ hi ≤ 1.207 s − t
2
a) Moment capacity
f cd
2
W pa = 2.545( s 3 − s '3 )
W pan = 2{1.616 s 3 − .25{3.414 s − 2hi ) 2 (0.569 s + 0.667 hi )} − W pcn
W pc = 2.545s '3
W pcn = 2{1.616 s'3 −.25{3.414 s '−2hi ) 2 (0.569 s '+0.667 hi )}
49
Mu
=> µ =
Ac f cd s '
2.545( s 3 − s' 3 ) − {2{1.616( s 3 − s' 3 ) − .25{3.414 s − 2hi ) 2 (0.569 s + 0.667 hi )} f yd
=
4.828s ' 3 f cd
2
0.5{3.414 s '−2hi ) (0.569 s '+0.667 hi )} f yd
+
4.828s ' 3 f cd
{2.545s' −2{1.616 s ' −.25{3.414 s '−2hi ) 2 (0.569 s '+0.667 hi )} f cd / 2

3 3
+
4.828s ' 3 f cd
hi 2 h
2.545( x 3 − ( x − .828) 3 ) − {2{1.616( x 3 − ( x − 0.828) 3 ) − .25{3.414 x − 2 ) (0.569 x + 0.667 i )} f yd
µ= t t
4.828( x − 0.828) 3 f cd
hi 2 h
0.5 × {3.414( x − 0.828) − 2 ) (0.569( x − 0.828) + 0.667 i )} f yd
+ t t
4.828( x − .828) 3 f cd
hi 2 h f
{2.545( x − 0.828) 3 − 2{1.616( x − 0.828) 3 − 0.25{3.414( x − .828) − 2 ) (0.569( x − 0.828) + 0.667 i )} cd
+ t t 2
4.828( x − 0.828) 3 f cd

Acc = Ac − Atra = 4.828s ' 2 −0.5(4.414 s − 2hi − 3.656t )(1.207 s − t − hi )
Asnet = Asc − Ast
Asnet = 5.328( s 2 − s ' 2 ) − 0.5{(3.414 s − 2hi ) 2 − (3.414 s '−2hi ) 2 }
Nu {4.828s ' 2 −0.5(4.414 s − 2hi − 3.656t )(1.207 s − t − hi )} f cd

v= =
Ac f cd 4.828s ' 2 f cd
[5.328( s 2 − s ' 2 ) − 0.5{(3.414 s − 2hi ) 2 − (3.414 s '−2hi ) 2 }] f yd
+
4.828s ' 2 f cd
hi h
{4.828( x − .828) 2 − 0.5(4.414 x − 2 − 3.656)(1.207 x − 1 − i )} f cd
Nu t t
v= = 2
Ac f cd 4.828( x − .828) f cd
hi 2
[5.328( x 2 − ( x − .828) 2 ) − 0.5{(3.414 x − 2 ) − (3.414( x − .828) − 2hi ) 2 }] f yd
+ t
4.828( x − .828) 2 f cd
50
Values of hi/t used are 0.5x, 0.6x, 0.8x, x and 1.207x-1
Case iii
Fig. 4.23 Neutral axis position with 0 ≤ hi ≤ 0.5s − 0.414t
a) Moment capacity
f cd
2
W pa = 2.545( s 3 − s ' 3 )
W pan = 2hi2 t
W pc = 2.545s ' 3
W pcn = (2.414 s − 2t )hi2
Mu
=> µ =
Ac f cd s '
{2.545( s 3 − s ' 3 ) − 2hi2 t} f yd + {2.545s '3 −(2.414 s − 2t )hi2 } f cd / 2
=
4.828s ' 3 f cd

hi 2 h
{2.545( x 3 − ( x − .828) 3 ) − 2( ) } f yd + {2.545( x − .828) 3 − (2.414 x − 2)( i ) 2 } f cd / 2
µ= t t
3
4.828( x − .828) f cd
51
Acc = 0.5 * 4.828s ' 2 +2hi (1.207 s − t )
Asnet = 4hi t
Nu {0.5 × 4.828s ' 2 +2hi (1.207 s − t )} f cd + 4hi tf yd
v= =
Ac f cd 4.828s ' 2 f cd
hi
{0.5 × 4.828( x − .828) 2 + 2hi (1.207 x − 1)} f cd + 4 f yd
Nu t
v= =
Ac f cd 4.828( x − .828) 2 f cd

Values of hi/t used are 0, 0.2x, 0.4x and 0.5x-0.414
Case iv
a) Moment capacity
The moment capacity is similar with that of case iii
Fig. 4.24 Neutral axis position with 0 ≤ hi ≤ 0.5s − 0.414t

Acc = 0.5 × 4.828s '2 −2hi (1.207 s − t )
Asnet = −4hi t
Nu {0.5 × 4.828s ' 2 −2hi (1.207 s − t )} f cd − 4hi tf yd

v= =
Ac f cd 4.828s ' 2 f cd
52
hi
{0.5 × 4.828( x − 0.828) 2 − 2hi (1.207 x − 1)} f cd − 4 f yd
Nu t
v= =
Ac f cd 4.828( x − 0.828) 2 f cd

Values of hi/t used are 0, 0.2x, 0.4x and 0.5x-0.414
Case v)
a) Moment capacity
Moment capacity is same as case ii
s
Fig. 4.25 Neutral axis position with ≤ hi ≤ 1.207 s − t
2
53
Acc = Atra = 0.5(4.414 s − 2hi − 3.656t )(1.207 s − t − hi )
Asnet = Asc − Ast
Asnet = −[5.328( s 2 − s ' 2 ) − 0.5{(3.414 s − 2hi ) 2 − (3.414 s '−2hi ) 2 }]
Nu {0.5(4.414 s − 2hi − 3.656t )(1.207 s − t − hi )} f cd
v= =
Ac f cd 4.828s ' 2 f cd
− [5.328( s 2 − s ' 2 ) − 0.5{(3.414 s − 2hi ) 2 − (3.414 s '−2hi ) 2 }] f yd
+
4.828s ' 2 f cd
Dividing the equation by t2
hi h
{0.5(4.414 x − 2 − 3.656)(1.207 x − 1 − i )} f cd
Nu t t
v= = 2
Ac f cd 4.828( x − .828) f cd
hi 2
− [5.328( x 2 − ( x − .828) 2 ) − 0.5{(3.414 x − 2 ) − (3.414( x − .828) − 2hi ) 2 }] f yd
+ t
4.828( x − .828) 2 f cd

Values of hi/t used are 0.5x, 0.6x, 0.8x, x and 1.207x-1
Biaxial Capacity for Rectangular Section

Fundamental equations
Y
Fig. 4.26 Rectangular section and axes used for capacity computation
54
For the rectangular section under axial compression and biaxial bending the section capacity for a
given neutral axis position can be determined in a similar fashion as that for sections under uniaxial
bending.
The axial load and moment capacities are given by:

N u = ( Ast − Asc ) f yd − Acc f cd (4.4)
Nuh (4.5)
M xu + = (Q sxt − Q sxc ) f yd − Q cxc f cd
2
N b (4.6)
M yu + u = ( Q syt − Q syc ) f yd − Q cyc f cd
2
where:
N u = Axial compressive capacity
Ast = Area of steel under tension
Asc = Area of steel under compression
Acc = Area of concrete under compression
M xu = Moment capacity about x-axis
h = Height of cross section

Qsxt = First moment of steel area under tension about X-axis
Qsxc = First moment of steel area under compression about X-axis
Q cxc = First moment of concrete area under compression about X-axis
M yu = Moment capacity about y-axis
b = Width of cross section
Qsyt = First moment of steel area under tension about Y-axis
Qsyc = First moment of steel area under compression about Y-axis
Qcyc = First moment of concrete area under compression about Y-axis
55
Other terms used to compute capacity are:
Q cx = First moment of concrete area about X-axis
= b ' h' 2 + b ' h' t

4
Qcy = First moment of concrete area about Y-axis
= b' 2 h' + b ' h' t

4
Q sx = First moment of steel area about X-axis
2
= bh + Qcx
4
Qcy = First moment of steel area about Y-axis
2
= b h + Qcy
4
The biaxial moment and compressive capacities of a given section for a particular neutral axis
position can be computed using the above basic equations. To get a closed form solution for the axial
capacity and the corresponding moment capacities different cases of neutral axis position are
considered and terms corresponding to the equations are computed for each case.
Case i
t < s < b , and t ≤ r ≤ h
N.A
Fig. 4.27 Neutral axis position for t < s < b , and t ≤ r ≤ h
56
s' r ' sr
Acc = Asc = − Acc Qsxt = Qsx − Qsxc
2 2
s' r ' 2 Ast = As − Asc Qsyt = Qsy − Qsyc
Qcxc = + Acc t
6 sr 2
Qsxc = − Qcx
s' 2 r ' 6
Qcyc = + Acc t
6 s2r
Qsyc = − Qcy
6
Case ii
t < s < b and h ≤ r
N .A
Fig. 4.28 Neutral axis position for t < s < b and h ≤ r
s' r ' 2
Acc = {1 − k 2' }
2
s' r '2 2 3
Qcxc = {1 − 3k 2' + 2k 2' } + Acc t
6
s'2 r ' 2 3
Qcyc = {1 − 3k1' − k 2' } + Acc t
6
sr
Asc = (1 − k 22 ) − Acc , Ast = As − Asc
2
sr 2 2 3 s2r 2 3
Qsxc = {1 − 3k 2 + 2k 2 } − Qcxc , Qsyc = {1 − 3k 2 − k 2 } − Qcyc
6 6
Qsxt = Qsx − Qsxc
Qsyt = Qsy − Qsyc
r '− h' r−h

Where k 2' = and k 2 =
r' r
57
Case iii
t < r < h , and b ≤ s
N.A
Fig. 4.29 Neutral axis position for t < r < h , and b ≤ s
The values of the terms for this case can be obtained by interchanging vales of r and s, h and b, and
corresponding x and y subscripts of the symbols of case ii.
Case iv
b ' h'
h ≤ r , b ≤ s and + ≥1
s' r '
s' r ' 2 2 sr
Acc = (1 − k1' − k 2' ) Asc = (1 − k12 − k 22 ) − Acc
2 2
s' r ' 2 3 2 3 s' 2 r ' 3 2 3
Qcxc = (1 − k1' − 3k 2' + 2k 2' ) + Acc t , Qcyc = (1 + 2k1' − 3k 2' − k 2' ) + Acc t
6 6
Ast = As − Asc
sr 2 2 3
Qsxc = (1 − k13 − 3k 2 + 2k 2 ) − Qcxc
6
s2r 2 3
Qsyc = (1 + 2k13 − 3k 2 − k 2 ) − Qcyc
6
Qsxt = Qsx − Qsxc and Qsyt = Qsy − Qsyc
58
s−b r − h ' s '−b' ' r '− h'
where k1 = , k2 = , k1 = , k2 =
s r s' r'
y
N.A
Y
b ' h'
Fig. 4.30 Neutral axis position for h ≤ r , b ≤ s and + ≥1
s' r '
Case v
b h
+ <1
s r
In this case the whole section is under compression.
Thus, the moment in both directions is zero and the axial compressive capacity is computed using the
basic equation with
Ac = Acc = b' h'
As = Asc = bh − b' h'
while the other terms are zero.
Biaxial chart
Biaxial chart shows the strength envelope for a given axial force N u by varying the neutral axis
position to get different corresponding values of moments M xu and M yu .
To draw the curve the normalized axial capacity is represented by:
59
Nu
v=
Ac f cd
and, the normalized moment capacities can be computed as
M ux
µx =
Ac f cd h'
M uy
µy =
Ac f cd b'
To get normalized biaxial chart all terms in the basic equations are made non-dimensional by dividing
the terms by t as was shown for uniaxial chart.
To accurately compute the moment capacities for a given axial force ratio sufficient pairs of moment
capacities are computed for different neutral axis positions. First the neutral axis position
corresponding to the given axial force ratio is computed from the basic equation for axial load
capacity (equation 4.4). This is done assuming r/t or s/t value and substituting into the equation and
solve for the unknown s/t or r/t value, respectively. The computation is done by using solver tool from
Microsoft Excel Program. After determining the neutral axis position, the moment capacities can be
computed from the basic equations (4.5 and 4.6).
These steps give one point on the interaction diagram. The steps are repeated for different neutral axis
positions and curves are drawn for different steel ratios.
60
EXAMPLES
Example 1
Design a square column subjected to Uniaxial Bending.
Given:
Action Effect: Factored loads allowing for initial eccentricity and slenderness effect.
Nsd = 435.20kN
Msdx = 174.08kNm
Materials Data: Concrete Grade; C30
Steel Grade; Fe360
Required:
Dimension of cross-section and thickness of steel
Solution:
Assume column size, column concrete size, b’ × h’ = 200mm × 200mm
f ck
f cd = = 16
γc
Acfcd = 200 × 200 × 16 = 640,000N=640kN
Nsd 435.20
v= = = 0.68
Ac f cd 640
Msd 174.08
µ= = =1.36
Ac f cdb' 640 ×0.20
Using chart for square section
w = 3.15
As f yd wAc f cd 3.15 × 640,000
w= => As = = = 9,436.6mm 2
Ac f cd f yd 213.64
As + Ac = 200 2 + 9,437 = 49,437

b = h = 49,437 = 222.34mm
t = 0.5 × (222.34 − 200) = 11.17 mm
Thus, use t=12mm and square column of overall dimension 224mm × 224mm.
61
Example 2
Verify the answer of example 1 using the procedure in EBCS 4-1995.
Given
Overall depth (d) = 224mm
Thickness of steel section (t) = 12mm
Action Effect: Factored loads allowing for initial eccentricity and slenderness effect
Nsd = 435.2kN
Msdx = 174.08kNm
Materials Data: Concrete Grade C30
Steel Grade Fe360
Required:
Check adequacy of the section for the given loading using the procedure in EBCS 4-1995.
Solution:
To draw the interaction curve the following section capacities are determined
- Plastic resistance to compression:
Npl,rd = Acfcd + Asfyd
=200 × 200 × 16 + (224 × 224-200 × 200) × 213.64
=2814kN
- Compressive resistance of the concrete part
Npm,rd = Acfcd
= 200 × 200 × 16 × 10-3
=640kN
- Neutral axis position for zero axial compression (hn).
From Fig.5.1 for zero axial force:
Fcc = Fst (5.1)
Fcc = (100-hn) × 200 × 16
62
ys2
ys1
yc ys1
ys2 M pl,Rd
Fig. 5.1 Stress block for zero axial compressive force
Fst = 2 × 12 × 2hn × 213.64

Subsisting the above equations in (5.1) and solving for hn
hn = 23.78mm
- Plastic moment capacity for zero axial force, Mpl,rd
Using the value of the neutral axis depth computed above:
Mpl,rd = Fcc × yc+2Fsc1 × ys1+2Fsc2 × ys2 (5.2)
Fcc = (100-hn) × 200 × 16 × 10-3
= 243kN
yc = 61.89mm
Fsc1 = 2 × 12 × (100-23.78) × 213.64 =390.8kN
ys1 = 61.89mm
Fsc2 = 224 × 12 × 213.64 = 574.26 kN
ys2 = 112mm
substituting the above values into equation (5.2)
Mpl.rd= 192.05 kNm
- Maximum moment capacity, Mmax,rd
Maximum moment capacity occurs when the neutral axis passes trough the centroid.
Mmax,rd =Fcc × yc+2Fsc1 × ys1+2Fsc2 × ys2 (5.3)
63
ys2
ys1
N pm,Rd/2
ys1
yc M max,Rd
ys2
Fig. 5.2 Stress block for maximum moment
Fcc= 100 × 200 × 16 =320 kN

yc=50 mm
Fsc1 = 2 × 12 × 100 × 213.64 = 512.7 kN
ys1 = 50 mm
Fsc2 = 224 × 12 × 213.64 = 574.26 kN
ys2 = 112 mm
Substituting the above values into equation (5.3)
Mmax,rd =195.9 kNm
The four points (as discussed in section 3.2) for drawing the interaction curve can be computed as
follows:
Point A: µ=0; v=1
Point B: µ=1; v=0
N pm ,rd 640
Point C: µ=1; v= = = 0.23
N pl , rd 2814
M max, rd 195.9 0.5 × N pm ,rd

Point D: µ = = = 1.02 ; v= = 0.115
M pl ,rd 192.05 N pl ,rd
Using this points interaction curve is drawn as shown in fig.5.3.

Point corresponding to the applied load:
M sd
µ sd = = 0.91;
M pl ,rd
N sd
v sd = = 0.15
N pl ,rd
This point lies within the interaction curve as shown in Fig. 5.3. Thus, the section is sufficient.
64
Interaction Chart
1.2
1.0
0.8
0.6
Nu
N pl , Rd
0.4
0.2
*
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
M u
M pl , Rd
Fig. 5.3 Normalized interaction curve
Example 3
Check for the adequacy of load capacity of a circular cross section against given loading.
Given:
Overall depth = 250mm
Thickness of steel section = 10mm
Action Effect: Factored loads allowing for initial eccentricity and slenderness effect.
Nsd = 500kN
Msd = 125kNm
65
Materials Data: Concrete Grade is C30
Steel Grade is Fe360
Required:
Check whether the section is sufficient or not.
Solution:
f ck
f cd = = 16 Mpa
γc
f yk 235
f yd = = = 213.636 Mpa
γs 1.10
π × 230 2
Ac f cd = × 16 × 10 −3 = 664.424 kN
4
Nsd 500
v= = = 0.754
Ac f cd 664 .424
M 125
µ = sd = =1.636
Ac f cd r' 664 .424 ×.115
From chart for circular column the required steel ratio is
w = 2. 2
While the available steel ratio is
As f yd π / 4 × (250 2 − 230 2 ) × 213.636
w= = = 2.423
Ac f cd 664.424 × 1000
Thus, the section is sufficient.
Example 4
Compare the similarity of values from uniaxial and biaxial curves at conditions common for both.
Given:
w = 4 , v = 1.0 and µ b = 0
Required: µ h
Solution:
66
a) Using uniaxial chart for square section
µ h = 1 .7
b) Using biaxial chart
µ h = 1 .7
Thus, both charts give similar values at common conditions.
Example 5
Design a column subjected to Biaxial Bending.
Given:
From analysis, including load factors and slenderness effect,
N sd = 1,440kN
M sd ,h = 400kNm
M sd ,b = 300kNm
Materials: Concrete C30

Steel Fe360
Required: Thickness of steel
Solution: Assume column concrete size, b’ × h’ = 300mm × 300mm

N sd 1440000
v sd = = = 1 .0
f cd A c 16 × 300 × 300
M sd ,h 400 × 10 6
µ sd ,h = = = 0.926
f cd Ac h 16 × 300 × 300 × 300
M sd ,b 300 × 10 6
µ sd ,b = = = 0.694
f cd Ac b 16 × 300 × 300 × 300
From the biaxial chart
w = 1 .9
67
As f yd
w=
Ac f cd
wAc f cd 1.9 × 1,440,000
=> As = = = 12807mm 2
f yd 213.64
As + Ac = 300 2 + 12,807 = 102,807

b = h = 102,807 = 320.6
t = .5 × (320.6 − 300) = 10.3mm
Use t = 11mm
Example 6
Design the above example using uniaxial charts by the approximate method given in EBCS 4-1995.
Given:
Same as Example 5
Required:
Same as Example 5
Solution:
M b, Sd M h, Sd
+ ≤ 1 .0 (5.4)
µ b M pl ,b, Rd µ h M pl ,h ,d
Trial 1: Let t = 11mm
=> w = 1.9
For an axial load ratio v = 1 using uniaxial chart
µ b = 0.82 = µ h
M pl ,b , Rd = M pl ,h, Rd = 0.82 × Ac f cd h' = 0.82 × 300 × 300 × 16 × 300 = 354.24 kNm
Substituting in equation (5.4)

300 400
+ > 1 .0
0.82 × 354.24 0.82 × 354.24
=> 2.41 > 1.0 Not Ok!
Trial 2: Let, t=17
68
As f yd (334 × 334 − 300 × 300) × 213.64
=> w = = = 3 .2
Ac f cd 300 × 300 × 16
For an axial load ratio v = 1 using uniaxial chart
µ b = 1.32 = µ h
M pl ,b , Rd = M pl ,h, Rd = µ pl , Rd × Ac f cd h = 1.32 × 300 × 300 × 16 × 300 = 570.24 kNm
Substituting in equation (5.4)
300 400
+ = 0.93 < 1.0
1.32 × 570 .24 1.32 × 570 .24
Thus, use t=17mm
Note: Comparing the results from the above two examples, one can see there is a significant saving
can be achieved in the amount of steel provided by using the biaxial chart rather than applying the
approximate method.
69
CONCLUSIONS
Concrete-filled steel tubes used as structural columns have significant economic, structural and
functional advantages. However, their design has been difficult and time consuming as it needs
drawing interaction curves for each trial cross-section. That is, the design procedure that was given in
EBCS 4-1995 follows a trial and error procedure to determine the necessary cross-section for a given
load. To alleviate this problem, normalized charts have been produced that simplify the design
calculation. The charts can be used to directly compute the amount of steel required for a given cross-
section without any trial and error. The charts produced also give more accurate values than using the
method in the Code.
According to EBCS 4-1995, design of cross-sections subjected to axial compression and biaxial
bending is done using an approximate and conservative way by using results from uniaxial chart to
estimate biaxial capacity. In this thesis, biaxial charts were developed from stress resultants of cross-
sections to accurately execute design of cross-sections subjected to biaxial bending. The charts also
simplify the calculation.
In design practice, there may be architectural and functional requirement to use columns of different
shape. The design using shapes other than the common ones is usually handled in an approximate way
as the required computation efforts are more demanding. However, for economical and efficient
design, provision of design chart is necessary. Thus, charts for octagonal-shape and hexagonal-shape
cross-sections have been developed as a starting work from many other possible cross-sectional
shapes.
Summary of contribution:
In this thesis charts have been developed for computing the axial compression and uniaxial moment
capacity of square, rectangular, circular, hexagonal and octagonal sections. The charts were drawn for
normalized axial compression and bending moment capacities. The values in the chart were
normalized by dividing with terms that involve only concrete cross-section properties are shown in
section 4.3 rather than using Mpl,rd and Npl,rd as given in the Code [EBCS 4-1995]. This makes the
design more simplified as one can directly compute the area of steel required by assuming a certain
70
concrete cross-section. However, in the code one needs to take a trial section for both the steel and the
concrete and check for capacity and then repeat for another trial section if the section is not adequate.
For rectangular sections, width-to-depth ratios of 0.5 and 2 were considered. The other polygons
considered are equal-sided. The steel ratio (w) used varies between 1.5 and 4. The materials used are
steel Fe360 with plate thickness less than 40mm and concrete C30.
Columns subjected to axial compression and biaxial bending are also treated to get a more simplified
and more accurate method for design or checking capacity. Procedures necessary to develop charts for
any rectangular section with normalized values similar to those for uniaxial chart showing the section
capacity were developed. These charts substitute the approximate methods of design of biaxial
capacity given in the code. A chart is produced for axial compression ratio (v) equal to 0 and 1 for a
square section. But, the equations developed can be used for any rectangular section. The steel ratio
(w) varies between 1.5 and 4. The materials used are also similar with that for uniaxial chart.
This thesis also gives summery of important equations and procedures for determining CFT column
capacity. Other behaviors related to section capacity, confinement, bond and seismic resistance are
also covered briefly.
Future Research
Other uniaxial and biaxial charts can be developed based of the procedures given in this thesis to
cover the remaining cross-sectional shapes and material types.
Concrete-filled steel tubes with reinforcing bars and concrete encased composite columns can also be
studied in a similar fashion as the principles for computing section capacity are similar.
Another area for further research can be verification of the results of this thesis work by experimental
tests as this work is based on theoretical guidelines.
Structural efficiency and behavior of composite columns within frames could also be studied. This
may include ductility and strength properties within frames under static or dynamic loading. The
71
connections of CFTs with beams or other structural components and their property can also be
studied.
APPENDIX: Uniaxial and Biaxial Charts
72
Square Section
6.0
5.5
5.0
4.5
N u 4.0
v=
Ac f cd
3.5
3.0 w=4
w = 3.5
2.5
w=3
2.0 w = 2.5
1.5 w=2
w = 1.5
1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0
M u
µ =
A c f cd h '
Chart No. 1 Uniaxial Chart for Square Section
73
Rectangular Section
h/b=0.5
6.0
5.5
5.0
4.5
4.0
Nu
v=
Ac f cd 3.5
3.0 w=4
2.5 w = 3.5
w=3
2.0 w = 2.5
1.5 w=2
w = 1 .5
1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0
M u
µ =
A c f cd h '
Chart No. 2 Uniaxial Chart for Rectangular Section with Height-to-Width (h/b)
Ratio of 0.5
74
Rectangular Section
h/b=2
6.0
5.5
5.0
4.5
4.0
Nu
v=
Ac f cd 3.5
3.0 w=4
w = 3. 5
2.5
w=3
2.0 w = 2 .5
w=2
1.5
w = 1.5
1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0
M u
µ =
A c f cd h '
Chart No. 3 Uniaxial Chart for Rectangular Section with Height-to-Width (h/b) Ratio of 2
75
Circular Section
6.0
5.5
5.0
4.5
4.0
Nu
v= 3.5
Ac f cd
3.0
w=4
2.5
w = 3 .5
2.0 w=3
w = 2. 5
1.5
w=2
1.0
w = 1.5
0.5 w =1
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Mu
µ =
Ac f cd d ' / 2
Chart No. 4 Uniaxial Chart for Circular Section
76
Hexagonal Section
Bending about Y-axis
6.0
5.5
5.0
4.5
Nu
v= 4.0
Ac f cd w=4
3.5 w = 3.5
3.0 w=3
2.5 w = 2.5
2.0 w=2
1.5 w = 1.5
1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Mu
µ =
A c f cd s '
Chart No. 5 Uniaxial Chart for Hexagonal Section with Bending about Y-axis
77
Hexagonal Section
Bending about X-axis
6.0
5.5
5.0
4.5
N u 4.0
v=
Ac f cd 3.5
w=4
3.0 w = 3 .5
2.5 w=3
2.0 w = 2.5
1.5 w=2
1.0 w = 1.5
0.5
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
M u
µ =
A c f cd s '
Chart No. 6 Uniaxial Chart for Hexagonal Section with Bending about X-axis
78
Octagonal Section
6.0
5.5
5.0
4.5
Nu 4.0
v=
Ac f cd 3.5 w=4
w = 3.5
3.0
w=3
2.5
w = 2.5
2.0 w=2
1.5 w = 1.5
w =1
1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Mu
µ =
A c f cd s '
Chart No. 7 Uniaxial Chart for Octagonal Section
79
Biaxial Chart
for
Square Section
v = 0 or v = 1
2.0
1.5
w=4
Mb
µb = 1.0 w = 3. 5
Ac f cd b ' w=3
w = 2 .5
0.5 w=2
w = 1.5
0.0
0.0 0.5 1.0 1.5 2.0
Mh
µh =
Ac f cd h '
Chart No. 8 Biaxial Chart for Square Section
80
REFERENCES
1. Boa-Jun Sun and Zhi-Tachu, 1992’, "Design Aids for Reinforced Concrete Columns", Journal of
structural engineering, ASCE Vol.118, pp 2986-2995
2. Charles W. Roeder, et al, “Seismic Behavior of Steel Braced Frame Connections to Composite
Columns”, University of Washington, www.aisc.org
3. EBCS 2-1995, “Structural Use of Concrete”, Ministry of Works and Urban Development, Addis
Ababa, pp 43-46
4. EBCS 3-1995, “Design of Steel Structures”, Ministry of Works and Urban Development, Addis
Ababa, pp57,59-91
5. EBCS 4-1995,Composite Column in, “Design of Composite Steel and Concrete Structures”,
Ministry of Works and Urban Development, Addis Ababa pp53-79
6. Gee-Yu Lliu, Yeoug-Kae Yeh and Keh-Chyuan Tsai, “The Axial and Flexural Load Behavior of
Concrete-Filled Steel Thin-Walled Tubes with Stiffened Square Sections”, National Center for
Research on Earthquake Engineering, Taipei, Taiwan, www.ncree.org.tw/iwscc
7. Gerald Huber, “Semi-Continuous Beam-to-Column Joints at the Millennium Tower in Vienna,
Austria, www.aisc.org.
8. Girma Zereyohanes,1998, "Biaxial Design Chart for Reinforced Concrete in Accordance with The
Ethiopian Building Code Standard EBCS 2 1995", Journal of EAEA, Vol.15 pp14-22
9. Ivan M.Viest et al, 1997,Composite columns in: "Composite Construction Design for Buildings",
ASCE, Mc Graw -Hill, New York pp 4.1-4.51
10. JF Hajjar, 2000, "Concrete-Filled Steel Tube Columns under Earthquake Loads", Prog. Structural
Engineering Material, www.hosted.ce.umn.edu
11. K. Lahlou, M. Lachemi, and P.C. Aitcin, 1999, "Confined High Strength Concrete Under
Dynamic Compressive Loading", Journal of structural engineering, ASCE, Vol.125, pp 1100-1108
12. Mathias Johansson and Kent Gylltoft,2001, "Structural Behavior of Slender Circular Steel-
Concrete Composite Columns Under Various Means of Load Application, Steel and Composite
Structures", Vol. 1, No. 4, Chalmers University of Technology, Sweden, 2001,
www.ste.chalmers.se
13. Tarenaka Corporation, 2001,”Features of CFT structure system”, Concrete-Filled steel tube
structural system CFT, www.Tarenaka e/quake e/cft/cft-htm
81
82

Ermiyas Ketema PDF

Uploaded by

Copyright:

Available Formats

Ermiyas Ketema PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ermiyas Ketema PDF

Uploaded by

Copyright:

Available Formats

ADDIS ABABA UNIVERSITY

SCHOOL OF GRADUATE STUDIES

DESIGN AID FOR COMPOSITE COLUMNS

A thesis submitted to the school of Graduate Studies in Partial fulfillment of the

Advisor: Dr. Shifferaw Taye

DESIGN AID FOR COMPOSITE COLUMNS

Approved by Board of Examiners

______________________ _____________________ _________________

______________________ _____________________ __________________

______________________ _____________________ _________________

______________________ _____________________ __________________

Fig. 4.24 Neutral axis position with 0 ≤ hi ≤ 0.5s − 0.414t .................................................... 52

Ac Cross-sectional area of concrete

Composite columns are in an increased usage for construction of high-

General approaches have been presented on how to prepare design

Numerical examples have been presented to illustrate the application

Keywords: Concrete-Filled Tubes, Uniaxial Bending, Biaxial Bending,

All charts produced are given in the appendix.

Merits and Demerits of CFT Composite Columns

c) Concrete-filled column with reinforcement d) Concrete-filled tube without

e) Concrete-filled tube with structural steel sections

Fig. 2.1 Typical cross-sections of composite columns

Fig. 2.3 Tie-bar stiffening scheme

Bond between concrete and steel

Section failure behaviour

COLUMN LOAD CAPACITY

Resistance to Compression and Bending

Cross-section resistance under uniaxial moment and axial compression

Fig. 3.1 M-N interaction curve for uniaxial bending

• Point A : Axial compression resistance alone:

• Point B : Uniaxial bending resistance alone:

where: Npm,Rd = Compressive resistance of the concrete section

• Point D : Maximum moment resistance

Slender member resistance under axial compression and uniaxial bending

Fig. 3.3 Resistance to axial compression and uniaxial bending

This can be represented by the two simultaneous conditions:

M z .Sd ≤ 0.9 µ z M pl . z .Rd (3.19)

1.0 (a) 1.0 (b)

My,Rd /Mpl.y.Rd Mz,Rd/Mpl.z.Rd

Fig. 3.4 Member resistance under compression and biaxial bending

Shortcomings of EBCS 4-1995 Design Procedure

Solutions to the above problems are provided in the next chapter.

Calculation Method and Scope

between 0.25 and 9.

• for circular hollow sections of diameter d and wall thickness t, d / t ≤ 90 ε 2 ;

• for rectangular hollow sections of wall depth h and thickness t, h / t ≤ 52 ε ;

ε = 235 / f y , where fy is the yield strength of the steel section in Mpa.

Selected Representative Sections & Materials

Interaction Chart for Axial Compression and Uniaxial Bending

Rectangular (Square) section

Fig. 4.2 Rectangular cross-section

Determining value of x=b/t for a particular steel ratio w

but, Ac = b' h' where b' = b − 2t and h' = h − 2t

( f yd + wf cd )(1 + y ) + {( f yd + wf cd )(1 + y )}2 − 4wf cd ( wf cd y + f yd )

2( f yd + wf cd ) + {2( f yd + wf cd )}2 − 4wf cd ( wf cd + f yd )

b) Axial load capacity

c) Values of hi/t used

b) Axial load capacity

c) Values of hi/t used

____________ _ _______

____________ _ ________

____________ _ _______

____________ _ ________