Design of Composite Haunch Beams and Connections For Long Span Applications PDF

DESIGN OF COMPOSITE
HAUNCH BEAMS AND CONNECTIONS

FOR LONG SPAN APPLICATIONS
BY
NG YIAW HEONG (BEng.(Hons.), MEng)
DEPARTMENT OF CIVIL ENGINEERING
A THESIS SUBMITTED
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
NATIONAL UNIVERSITY of SINGAPORE
2004
ACKNOWLEDGEMENTS
The author would like to take this opportunity to acknowledge various
individuals for their guidance and encouragement in the course of this research.
Firstly, the author would like to express his appreciation for the constant guidance and
encouragement provided by his research supervisors, Professor N.E. Shanmugam and
Associate Professor J.Y. Richard Liew. This research work would not have been
completed without their continuous support.
Secondly, the author is fortunate to have received moral support and
understanding from Sze Ching; his wife and his parents. He would like to express
gratitude to them. For the authors 3 years-old and 1-year-old sons, Yan Zhang and
Ding Jie; the author could only apologize for not being able to keep them company
most of the time especially during the final stage of the study.
Last but not least, the assistance given by the lab officers during the
experimental testing in the Concrete and Structural Laboratory, National University of
Singapore is gratefully appreciated.
This research project was funded by the National University of Singapore
under a research grant (RP 930648). The support from Yongnam Engineering &
Construction Pte Ltd, Singapore who supplied the test specimens is gratefully
acknowledged.
ii
TABLE OF CONTENTS
TITLE PAGE i
ACKNOWLEDGEMENTS ii
TABLE OF CONTENTS iii
SUMMARY vi
LIST OF TABLES vii
LIST OF FIGURES xiii
LIST OF SYMBOLS xiv
CHAPTER 1 INTRODUCTION 1
1.1 BACKGROUND 1
1.2 RESEARCH OBJECTIVES 4
1.3 SCOPE OF WORK 5
1.4 STRUCTURE OF THE THESIS 6
CHAPTER 2 LITERATURE REVIEW 9
2.1 BACKGROUND 9
2.2 INTERNAL FORCES AND MOMENTS IN CONTINUOUS COMPOSITE HAUNCH BEAM 11
2.3 GLOBAL ELASTIC ANALYSIS OF NON-SWAY FRAME 12
2.4 PLASTIC HINGE ANALYSIS OF NON-SWAY FRAME 14
2.4.1 Rigid-Plastic Analysis 15
2.4.2 Elastic-Plastic Analysis 16
2.5 ANALYSIS OF HAUNCH SECTION 16
CHAPTER 3 EXPERIMENTAL INVESTIGATION

- HAUNCH CONNECTION 21
3.1 GENERAL 21
3.2 MATERIAL PROPERTIES 22
3.2.1 Beam and column sections 22
3.2.2 Reinforcement bar 23
3.2.3 Concrete 23
3.3 FABRICATION OF TEST SPECIMENS 24
3.4 TEST SET-UP 25
iii
3.5 INSTRUMENTATION 26
3.6 TESTING PROCEDURE 26
3.7 DATA ASSESSMENT 28
3.7.1 Beam rotation, b 28
3.7.2 Column rotation, c 28
3.7.3 Connection rotation, 28
3.7.4 Inelastic Rotation, ie 29
3.8 JOINT STIFFNESS, RKI 29
3.9 JOINT ULTIMATE MOMENT, MU 29
3.10 JOINT ROTATIONAL CAPACITY, CD 30
3.11 HAUNCH CONNECTION CAPACITY 30
3.12 HAUNCH TOE MOMENT CAPACITY 31
3.13 JOINT TEST RESULTS AND DISCUSSION 31
3.13.1 Comparison of test results 32
3.13.2 Connections H1 and H2 32
3.14 EFFECT OF SLAB REINFORCEMENT RATIO 40
3.15 EFFECTS OF HAUNCH LENGTH 40
3.16 CONCLUSIONS 41
CHAPTER 4 EXPERIMENTAL INVESTIGATION

- HAUNCH BEAM 59
4.1 INTRODUCTION 59
4.2 MATERIAL PROPERTIES 60
4.2.1 Beam and column sections 60
4.2.2 Reinforcement bar 61
4.2.3 Concrete 61
4.3 FABRICATION OF TEST SPECIMENS 61
4.4 TEST SET-UP 62
4.5 INSTRUMENTATION 65
4.6 TESTING PROCEDURE 65
4.7 BEAM TEST RESULTS AND DISCUSSION 66
4.7.1 Beam Specimen B1 66
4.8 CONCLUDING REMARKS 73
CHAPTER 5 ANALYTICAL MODEL 105
5.1 GENERAL 105

5.2 COMPARISON OF PLASTIC HINGE ANALYSIS AND TEST RESULTS 105
5.3 ROTATION CAPACITY 108
5.3.1 General 108
5.3.2 Calculation of available rotation capacity of composite section 109
5.4 BEAM ANALYSIS 111
5.4.1 Composite Haunch Beam Properties 111
5.4.2 Composite Haunch Beam analysis 115
5.5 FINITE ELEMENT MODELLING 116
5.5.1 Nonlinear Analysis Software: - USFOS 116
iv
5.5.2 Modeling of Composite Haunch Beam 116
5.5.3 Results 117
5.6 LATERAL TORSIONAL INSTABILITY 117
5.6.1 General 117
5.6.2 Lateral Distorsional Buckling Design Method 118
CHAPTER 6 DESIGN RECOMMENDATIONS

AND DESIGN EXAMPLE 131
6.1 INTRODUCTION 131

6.2 DESIGN RECOMMENDATIONS 131
6.3 ELASTIC GLOBAL ANALYSIS 133
6.4 PLASTIC HINGE ANALYSIS 135
6.5 DESIGN PROCEDURE FLOW CHART 138
6.6 DESIGN EXAMPLE 140
CHAPTER 7 CONCLUSIONS AND

PROPOSALS FOR FUTURE WORK 141
7.1 GENERAL 141

7.2 BEHAVIOUR OF THE COMPOSITE HAUNCH CONNECTION 141
7.3 BEHAVIOUR OF THE COMPOSITE HAUNCH BEAM 142
7.4 SECTION PROPERTIES AND FRAME ANALYSIS 142
7.5 FUTURE WORK 144
APPENDIX A DESIGN EXAMPLE 146
REFERENCES 154
LIST OF PUBLICATIONS 160
v
SUMMARY
This thesis is concerned with the behaviour of steel-concrete composite
haunch connections and beams. Experiments were carried out to investigate the
moment rotation characteristics and ultimate capacity of these connections and
beams. Details of the experiments giving information on test specimens,
instrumentation, test set-up and test procedures are described. There are a total
of 10 haunch connections and 3 continuous composite haunch beam specimens
tested to failure. Results obtained for connection moment capacity, rotation
capacity and failure modes are presented. It is found that through proper design
and detailing, these connections display the characteristics of a rigid connection.
Optimum design of composite haunch beam can be achieved when plastic hinge
occurred at haunch toes followed by at the mid-span to form a plastic collapse
mechanism. Haunch toe could be designed as the weakest section to form a
plastic hinge with suitable amount of reinforcement in the slab and range of
haunch length. Experimental results show that composite haunch connection
exhibits a ductile moment-rotation behaviour and is able to redistribute moment
to the mid-span by loss of stiffness due to cracking of concrete slab and yielding
of either steel reinforcement or cross section. Study also has been carried out to
investigate the parameters that influence the stiffness, strength and rotation
capacity of composite haunch connections. Design guidelines for composite
haunch joints and beams are provided.
vi
LIST OF TABLES
Chapter 2
Table 2.1 Limits to redistribution of hogging moment to reduce (EC4)
Table 2.2 Maximum redistribution of negative moment in composite haunch
beam at ultimate limit state [Lawson 1989]
Chapter 3
Table 3.1 Details of joint test specimens
Table 3.2 Summary of universal section properties and tensile test results
Table 3.3 Summary of reinforcement bar properties and tensile test results
Table 3.4 Summary of concrete cube test results
Table 3.5 Summary of test results
Chapter 4
Table 4.1 Summary of concrete cube test results for beam specimen
Table 4.2 Details of beam test specimens
Chapter 5
Table 5.1 Comparison of test results with Plastic Hinge Theory
Table 5.2 Comparison of rotational capacity at Haunch Toe
vii
LIST OF FIGURES
Chapter 1
Figure 1.1 Haunch as Taper Section in Universal Beam
Figure 1.2 Cutting of Taper Section in Universal Beam
Chapter 2
Figure 2.1 Cross Section of Continuous Composite Beam
Figure 2.2 Sub-frame of Beam and Column in Non-sway Frames Analysis
Figure 2.3 Relation between Haunch Beam Elastic Resistance and Parent Beam
Plastic Resistance
Figure 2.4 Distorsional Buckling of Composite Beam
Chapter 3
Figure 3.1 Building Plan Layout for Joint Specimens'Design
Figure 3.2 Cruciform Joint Specimen
Figure 3.3 Test Specimen Ready for Concrete Casting
Figure 3.4 Typical Joint Test Specimen
Figure 3.5 Joint Test Specimen Ready for Testing
Figure 3.6 Instrumentation of Test Specimen
Figure 3.7 Definition of Rotation in a Joint
Figure 3.8 Moment-rotation Curve of Connection
Figure 3.9 Stress-strain Block of Haunch Connection
Figure 3.10(a) View after Failure of Specimen H1 and H2
Figure 3.10(b) View after Failure of Tension Bolt of Specimen H1 and H2
viii
Figure 3.11 Moment-Rotation Curve of H1 and H2
Figure 3.12 View after Failure of H3 and H4
Figure 3.13 Moment-Rotation Curve of H3
Figure 3.15 View after Failure of H5
Figure 3.25 Premature Failure of H10
Figure 3.26 Comparison of Load-Displacement Curve of H2,H4 and H6
Figure 3.27 Comparison of Load-Displacement Curve of H3, H4, H7 and H8
Chapter 4
Figure 4.1 Haunch Beam Test Specimen
Figure 4.2 Wooden Formwork of Beam Test Specimen
Figure 4.3 Beam Specimen Ready for Concrete Casting
Figure 4.4 Beam Specimen Ready for Testing
Figure 4.5 Isometric View of Haunch Beam Test Specimen
ix
Figure 4.6 Schematic Loading of Haunch Beam Test Specimen
Figure 4.7 Loading Frame in Haunch Beam Test Specimen
Figure 4.8 Loading Frame Connected to Hydraulic Actuator
Figure 4.9 Larger Haunch Connection at Cantilever Beam Side
Figure 4.10 Instrumentation of Beam Specimen
Figure 4.11 View after Failure of Specimen B1
Figure 4.12 Load-Displacement Curve of Specimen B1
Figure 4.13 Inelastic Buckling in Compression Flange of the Beam B1
Figure 4.14 Crushing of Concrete Slab at Loading Point in Beam B1
Figure 4.15 Concrete Slab Cracking Pattern of the Beam B1 at Haunch Toe
Figure 4.16 Strain Reading of Cross Section for Beam B1 at Left Haunch Toe at
Different Load Stage
Figure 4.17 Strain Reading of Cross Section for Beam B1 at Right Haunch Toe at
Figure 4.18 Strain Reading of Cross Section for Beam B1 at Left Loading Point at
Figure 4.19 Strain Reading of Cross Section for Beam B1 at Right Loading Point at
Figure 4.20 Strain Reading of Tension Reinforcement for Beam B1 at Left

Haunch Toe at Different Load Stage
Figure 4.21 Strain Reading of Tension Reinforcement for Beam B1 at Right

Figure 4.22 Strain Reading of Concrete Slab for Beam B1 at Left Loading Point at
Figure 4.23 Strain Reading of Concrete Slab for Beam B1 at Right Loading Point
at Different Load Stage
Figure 4.25 Inelastic Buckling of Compression Flange of the Beam B2
x


Figure 4.36 View after Failure of Specimen B3
Figure 4.37 Inelastic Buckling of Left Haunch Flange of the Beam B3
Figure 4.38 Inelastic Buckling of Right Haunch Flange of the Beam B3
Figure 4.42 Strain Reading of Cross Section for Beam B3 at Right Haunch Heel at
Figure 4.43 Strain Reading of Cross Section for Beam B3 at Left Haunch Heel at
xi


Figure 4.50 Comparison of Load-Displacement Curve for Specimens B1, B2 and

B3
Chapter 5
Figure 5.1(a) Haunch Beam with 2 Point Loads
Figure 5.1(b) Collapse Mechanism in Haunch Beam
Figure 5.1(c) Moment Diagram of 2 Point Loads
Figure 5.1(d) Moment Diagram of Haunch Toe Loads
Figure 5.1(e) Moment Diagram of 2 Point Loads and Haunch Toe Loads
Figure 5.2 Moment-Rotation Curve according to Kemp, (1991)
Figure 5.3 Plastic Region near the Internal Support of Continuous Beam
Figure 5.4 Bi-Linear Moment-Rotation Curve for Connection H3
Figure 5.10 Cross Section of Haunch Beam with PNA at Beam Flange
xii
Figure 5.11 Cross Section of Haunch Beam with PNA at Beam Web
Figure 5.12 Modeling of Haunch Beam in Finite Element Software USFOS
Figure 5.13 Comparison of USFOS and Experimental Load-Displacement Curves

for Beam B1

for Beam B1

for Beam B1
xiii
LIST OF SYMBOLS
A = Section area
Af = Flange area
Aw = Web area = A - 2Af
B = Beam Width
Be = Concrete effective width
D = Diameter
D = Beam Depth
Dh = Depth of Haunch
E1 = Strain energy absorbed in lateral bending of the bottom flange
E2 = Torsional energy absorbed in twisting the bottom flange
E3 = Bending energy absorbed in displacing the web
E4 = Torsional energy absorbed in twisting the web
E5 = Work done by the compressive force in the bottom flange
E6 = Work done by the forces in the web
= Connection rotation
a = Inelastic available rotation
C = Parameter IcL/(Ibch)
Cd = Joint rotational capacity
fcu = Concrete Compressive Strength
e = Required Elastic rotation
FEM = Fixed-ended moment of the beam under the same loading condition
r = Required Inelastic rotation
fu = Steel Ultimate strength
u = Rotational capacity
fuf = Ultimate strength of flange
xiv
fur = Rebar Ultimate strength
fuw = Ultimate strength of web
fy = Steel Yield strength
fyf = Yield strength of flange
fyr = Rebar Yield strength
fyw = Yield strength of web
mr = Partial material factor
h = Length of column from floor to floor
LT = Perry Coefficient
Ibc = Second moment of area of the uncracked composite beam
Ic = Second moment of area of the column
L = Length of the beam (including the haunch)
= Slenderness of the beam length between restraints
L = Length between hinges at both ends
Le = Span of the beam between the end of the haunches
Li = Length between maximum moment and adjacent point of inflection
LT = Equivalent Slenderness
Lp = Plastic region of the flange
Mhe = Elastic Moment Resistance of Haunch Section
Mhu = Moment capacity of composite haunch connection
Mn = Negative moment
Mnc = Negative moment resistance of the composite beam
Mp = Design moment resistance
Mpc = Positive moment resistance of the composite beam
Mph = Plastic moment capacity of hogging region
Mps = Plastic moment capacity of sagging region
Ms = Elastic Moment Resistance of Steel Beam Section
xv
Mu = Moment resistance
n = Slenderness correction factor (for shape of bending moment diagram)
P = Point Load
py = Design strength for steel
b = Beam rotation
c = Column rotation
ie = Inelastic rotation
ra = available rotational capacity a / e (Non-dimension)
Rb = Bolt in tension
Rhf = Haunch flange capacity
Rhw = Haunch web capacity
Rki = Joint stiffness
Rki = Rotational stiffness
Rr = Reinforcement in tension
rr = Required rotation capacity r / e (Non-dimension)
Thf = Thickness of haunch flange
thw = Thickness of haunch web
u = Buckling parameter (0.9 for universal sections)
vt = Slenderness factor (including torsional stiffness and other effects).
wu = Factored design load on the beam
yc = Distance from the top of haunch flange
xvi
CHAPTER 1
INTRODUCTION
1.1 Background
Composite structures comprising of steel frames and concrete floors were
introduced in the last century and are now commonly used in modern buildings. These
kinds of construction method are widely used now because of structural economy
with fast speed of construction. Interaction between steel and concrete is achieved by
connecting them by means of shear connectors. Metal decking may be used as
permanent formwork to support any load during construction and later to act
compositely with the hardened concrete slab to form a composite slab.
There are a number of advantages in using composite beams. Firstly, the
total steel weight reduces significantly by 30 to 50 % compared with non-composite

beams (Narayanan, 1991, Lawson, 1995, Uy & Liew, 2003). Composite beams also
provide larger stiffness that will reduce the depth of the beam for the same span. This
results in lower storey heights and savings in cladding costs or, alternatively,
permitting more headroom for services. Another benefit of composite construction is
the metal decking which supports construction loads and acts as a working platform.
The decking also acts as transverse reinforcement to the composite beams and
distributes shrinkage strains and prevents serious cracking of concrete.
Besides the advantages mentioned above, a strong demand for large column-
free space in buildings in recent times has necessitated further research into the
behaviour of haunch beams since they are considered to be an efficient and
economical form for long span construction. This system is able to offer more variety
to the designer in planning the usage of the column-free space. There are several
types of structural options for achieving long span and incorporating of services
within normal floor zones. These include (Lawson and Rackham, 1989, Owen, 2000):
i) Beams with web openings
ii) Castellated beams
1
iii) Fabricated beams with tapered web
iv) Trusses
v) Stub girders
vi) Parallel beam grillage systems
vii) Haunch beams
Haunch beams in this thesis are defined as beams being stiffened at two ends
with a tapered triangular T-Section as shown in Fig.1.1. The tapered section is usually
cut from a similar section. Fig.1.2 shows two typical tapered sections being cut from a
universal beam. These tapered sections will then be welded to the beam ends and
usually there are end plates at both ends of the beam as shown in the diagram.
Haunch beams are designed by assuming a rigid moment connection between
the beams and columns. Depth and length of a haunch are chosen so that they result in
an economical method of transferring moment into the column and in a reduction of
beam depth to a practical minimum. Haunch composite beams in which steel beams
are designed to act in conjunction with a concrete slab of definite width could result in
shallow beams, provide sufficient rotation capacity of the connection that will permit
a redistribution of the moment and thus mobilise a full sagging capacity of the beam
resulting in an economical design. Furthermore, haunch beam systems could also
provide a long unobstructed space for services and increase speed of construction.
One of the common scenarios in steel construction is opening for services. Usually
web of the steel beam need various sizes of penetration for mechanical and electrical
services. Those penetrations normally are required to be strengthened by extra
stiffeners which directly increase the fabrication cost. Therefore, it is not cost
effective to create openings unless really there are no other choices. However, with
the haunch beam system, the space at the haunch region could offer more freedom for
2
Mechanical & Electrical services and less co-ordination between the steel contractor
and the M&E engineers during the construction stage. This will definitely increase the
production of steelwork and indirectly increase the construction speed.
Haunch composite beams may offer continuity at the beam-column support
and hence increase the structural performance of the system as a continuous beam. A
continuous beam could offer about 33% of strength compared with a simply
supported beam system. The continuity in composite beams provides benefits at both
the ultimate and serviceability limit states for long span structures. For instance, the
deflection of the beam could be easily 50% less for a continuous beam compared with
the simply supported beam system. However, one of the shortcomings in continuous
composite beams is that the composite sagging section capacity is always larger than
the hogging moment. For a continuous composite beam such as a parallel beam
grillage system, the negative bending at internal supports is generally significantly
less than the resistance in positive bending in the midspan region. Therefore, the
introduction of a haunch may be an option to overcome the shortcoming because it
will increase the hogging section capacity. And if necessary, tension reinforcement
could be added thus increasing the hogging capacity. Test results show that the
hogging capacity is as high as the sagging section capacity when sufficient tension
reinforcement is placed at the concrete slab at the hogging region. The ultimate
strength of composite beams under sagging moment has been well established and
Eurocode 4 has offered detailed design guideline. However, under hogging moment,
many tests have been conducted (Hamada 1976) and the results have shown that the
majority of beams failed as a result of local buckling. Tests have shown that the
width-thickness ratio for the flange of the steel section and the amount of longitudinal
slab reinforcement are significant factors affect local flange buckling. Therefore, it is
3
important to find out the factors that affect the design of haunched composite beams
so that the structural system will be utilised more efficiently.
Eurocode 4 defines composite connection as the one, which the reinforcement
in the joint is intended to provide resistance in tension. The tensile action of the slab
reinforcement increases both the resistance and stiffness compared with the structural
steel connection. However, the connections will usually be partial-strength
connections relative to the composite section next to the connection. Therefore,
haunches could be introduced to provide full continuity, which strengthens the
connection between the steel sections. For economy in composite beam design, both
the hogging end resistance and the mid-span sagging resistance should be well utilised
as it will be shown in the proposed experimental program that the hogging and
sagging resistance of the composite haunch beam can be proportioned to achieve an
optimum design. It is also noticed that by introducing the haunch in the steel
connection, the rotational capacity at the joint is almost not required because both the
hogging and sagging section capacity are reached at the same time. The philosophy of
this design concept is that the ductility (i.e. rotation capacity) is no longer important if
the hogging and sagging section capacity is achieved simultaneously. This is unlike
the composite joint without haunch which requires that they have both sufficient
strength and ductility. In addition, the connection moment capacity should be greater
than the applied moment, and the connection capacity should be larger than that
required to develop the moments in the beam at the ultimate limit state.
1.2 Research Objectives
The main objective of this research is to study experimentally the behaviour of
composite haunch connections and composite haunch beams with tension
4
reinforcement subjected to negative moment condition in order to simulate the joint in
non-sway composite frames. Parameters such as reinforcement ratio and haunch
length are varied in the experimental program. The effects of these parameters with
respect to moment capacity, rotational stiffness, rotation capacity are studied. The
results will be used to develop analytical and design guidelines for composite haunch
beams. The key joint properties, i.e. moment resistance Mu, rotational capacity u,
rotational stiffness Ki are evaluated for global frame analysis.
1.3 Scope of works
In this thesis, literature related to composite haunch connections and
composite haunch beams are reviewed. The scope of literature study is not only
limited to haunch connections. Non-haunch connections were also studied and
comparisons made between haunch and non-haunch composite connections.
A series of composite haunch connections and composite continuous haunch
beams were tested to failure in the laboratory. All the experimental results are
reported in detail. The experimental study also includes the behaviour of the
composite haunch connection and haunch beam illustrated by their moment-rotation
curves. The effects of parameters such as reinforcement ratio, haunch length and the
moment-rotation curve are investigated.
Analytical models for the prediction of moment resistance Mu, rotational
capacity u and rotational stiffness Ki of the composite haunch beam are established.
Results obtained from experiments are compared against the analytical model. And
finally, design guidelines for composite haunch beam are provided.
5
1.4 Structure of the Thesis
The thesis contains seven chapters. Chapter 1 gives the general description of
the advantages of composite construction and in particular composite haunch beam
construction. The need for further research on composite haunch beam construction in
long span application is presented and the objectives and scope of the research are
highlighted in the chapter.
Chapter 2 reviews the literature on haunch beam construction; both
experimental and analytical studies for braced and sway frame since 1972 are
covered. Various types of constructions other than haunch beam constructions are also
studied here and the pros and cons of these construction methods are presented. This
chapter also describes different types of analyses for composite haunch beams.
Considering the studies carried out by the previous researchers, the direction for the
present study is illustrated.
Chapter 3 describes the experimental program for haunch connections in non-
sway composite frames. Details of the test set-up and parameters varied in the
investigation are given. Materials for the test specimens with their mechanical test
results is presented here. This chapter also explains the loading procedure for the
testing. Test results obtained from the experiments is also presented which includes
the beam behaviour from the initial stage to the ultimate stage. The actual behaviour
of composite haunch connections is discussed systematically by comparing among the
test specimens. Failure modes of those specimens are identified and the effects of the
parameters illustrated.
Chapter 4 describes the experimental program for haunch beam construction.
Three composite haunch beam specimens of 8m span were tested to failure. Details of
the test set-up and the parameters varied in the experimental program are given.
6
Besides, this chapter explains the loading procedure for the beam testing. Test results
obtained from the experiments are presented covering the response behaviour from
the initial load stage to the ultimate load stage. The actual behaviour of composite
haunch connections is discussed by comparing among the test specimens. Failure
modes of those specimens are identified and the effects of the key design parameters
are illustrated.
Chapter 5 presents analytical models to predict the moment capacities,
rotational capacity and initial stiffness of composite haunch connections. The results
obtained in the experimental program are compared with those obtained using the
analytical models proposed, thus verifying the models. In addition, non-linear finite
element analysis is used to confirm further the experimental results and analytical
model.
Chapter 6 presents design recommendations and an example for the composite
haunch construction. Conclusions and recommendations for future research are given
in Chapter 7.
7
Universal Column
Universal beam
Tapered Section Tapered Section
Fig. 1.1 Haunch as Tapered Section in Universal Beam
Universal Beam
Tapered Section Cutting Line
Fig.1.2 Cutting of Tapered Section in Universal Beam
8
CHAPTER 2
LITERATURE REVIEW
2.1 Background
Researchers (Aribert and Raoul, 1992, Cosenza et al., 1995a, Climenhaga and
Johnson, 1972a, Couchman, 1996, Dekker et al., 1995, Fabbrocino et al., 2001,
Hamada and Longworth, 1976, Hamada and Longworth, 1974, Hope-Gill and
Johnson, 1976, Johnson and Chen, 1991, Kemp and Dekker, 1991, Leon, 1990, Liew
et al., 2001, Lukey and Adams, 1969, Price and Anderson, 1992, Tehami, 1997) have
proposed design methods for simple or continuous composite beams, the cross section
of which is as shown in Fig. 2.1. Required and available rotation capacities for the
section have been considered and the accuracy of the prediction method has been
assessed by comparing the theoretical and experimental results. Research works
referred above comment on the composite beam behaviour but seldom consider the
sub-assemblies of composite frame. Early work by Kitipornchai and Trahair, 1972 has
shown that uniform beams are not always the most efficient choice and often great
material economy can be achieved by using non-uniform beams such as haunch beam.
The research work reported herein is to incorporate composite haunch connection as
the joint in a sub-assembly. Unlike the Reduced Beam Section (often referred to as
dogbone (Plumier, 1997)) which is accomplished through an engineered gradual
transition of the beam flanges to the intended reduced section at a given location, the
haunch connection strengthens the connection and allows the formation of plastic
hinges at a designated location (Iwankiw, 1997). Haunch composite beams are
designed in a similar manner to continuous beams of uniform section (Lawson and
Rackham, 1989). The critical section for design is at the haunch toe, and the depth of
9
the haunch is prefixed to develop the required moment in the beam to column
connection.
Intensive research is in progress on composite frame design and methods have
been proposed to achieve an optimum and economical design. One of the most
popular design directions is to incorporate the concepts of semi-rigid and partial
strength connections and semi-continuous framing of EC4 (Nethercot, 1995). In
semi-continuous construction, the support moments are limited to the capacities of the
beam to column connections and the plastic rotations are required to develop the
beams sagging moment capacity to achieve the design moment (Nethercot et al.,
1995). A comprehensive guideline (Li et al., 1995) has been proposed for the design
of semi-continuous composite beams in braced frames where special attention is given
to the effects of joint rotational stiffness.
Tests were carried out (Aribert and Raoul, 1992, Hope and Johnson, 1976) in
order to calibrate analytical models (Tehami, 1997) and to investigate rotation and
moment capacities in composite beams. Local Buckling and moment redistribution in
composite beams have also been studied (Climenhaga and Johnson, 1972b, Johnson
and Chen, 1991); it has been concluded that the redistribution of elastic bending
moments allowed by Eurocode 4 is safe, economical and reflects the real behaviour of
two span composite beams. For beams continuous over more than two spans the
method is believed to be slightly conservative.
Another alternative of composite frame design is to provide composite
connections up to the full hogging resistance of the beam. This is accompanied by
sufficient rotation capacity at the connection to permit the moment redistribution in
continuous composite construction to mobilize the full sagging capacity of the beam
at mid-span to reach an optimum design. A series of tests on composite beam-to-
10
column connections have been carried out by Anderson, (1994) and the results show
that increasing the amount of reinforcement will not only increase the moment
resistance but also increase the rotation capacity of the composite section. Besides,
investigations (Nethercot, 1995, Nethercot and Li, 1995) into the behaviour of
composite connections and continuous composite frames have shown that properly
designed and detailed composite connections are capable of providing moment
capacity up to the full hogging resistance of the beam. It is also concluded that elastic
analysis assuming full continuity is not acceptable for composite frames because it
fails to meet the moment capacity requirement at the support section and it is over-
conservative for sections within the span.
Despite the detailed studies on composite beams, information available on
composite haunch beams is limited. Works by Rackham, (1992) and Boswell, (1992)
have shown that haunches are sufficiently stiff as full strength rigid connections and
the toe is restrained from distorsional buckling when full depth stiffeners are provided
on both sides of the web. Failure modes of haunch toes often involves local buckling
of the compression flange. Investigation of this local buckling has been carried out by
many researchers (Climenhaga and Johnson, 1972a & b, Kitipornchai and Trahair
1975a & b, Lay, 1965, Lay and Galambos, 1965, Nethercot, 1975 & 1983, Nethercot
and Trahair, 1976, Trahair and Kitipornchai, 1972, Trahair, 1983). In order to study
further applications of haunch connections in long-span composite construction a
study has been undertaken by the author on the behaviour of haunch connections
2.2 Internal forces and moments in continuous composite haunch beam
The internal forces and moments in a continuous haunch composite beam may
generally be determined using either:
a) Global elastic analysis or
11
b) Plastic hinge analysis
2.3 Global elastic analysis of Non-Sway Frame
Elastic analysis may be used for determining the forces and moments in
continuous beams. The assumption used in global elastic analysis is that the stress-
strain relationship for the material is linear elastic but the tensile strength of concrete
is neglected. This assumption is valid for first-order and second-order elastic
analyses, even though section capacity is evaluated based on plastic resistance.
Referring to Fig. 2.2, a sub-frame can be assumed in an analysis of the beam
members of non-sway frames under vertical loads; the column bases are assumed to
be fixed or pinned at foundations. The sub-frame is then analysed elastically under
various load combinations.
The magnitude of the negative moment largely depends on the relative
stiffness of the adjacent column and beam. If the beam stiffness is underestimated, the
negative beam moments and the column moments are over estimated. The stiffness of
the haunch largely compensates for any loss of stiffness of the beam due to concrete
cracking. Ignoring both effects is generally conservative for braced frames as it is
usually the consideration of the negative moment region that determines the sizing of
the steel beam.
Taking the simple case of a single-bay haunch beam with column above and
below the beam being analysed, the negative moment at the beam end is given by:
4 c
Mn = FEM
4 c + 1
(Eq 2.1)
Mn = Negative moment
FEM = the fixed-ended moment of the beam under the same loading condition
12
c = the parameter IcL/(Ibch)
Ic = the second moment of the area of the column
h = the length of column from floor to floor
Ibc = the second moment of area of the composite beam (assumed
uncracked)
L = the length of the beam (including the haunch)
In global elastic analysis, certain percentages of moment redistribution from
the hogging (negative) to sagging (positive) moment regions of the beam is allowed.
The redistribution of moment arises from cracking and loss of stiffness of the
composite section and local yielding of the steel beam. The degree of the local
yielding is influenced by classification of the composite section.
Parametric studies show that the length of the haunch does not significantly
affect the bending moment distribution in the beam from elastic global analysis
(Lawson, 1989). Therefore, the haunch length can be varied so that the moment
resistance of the beam is compatible with the designed moment. The haunch toe is the
potential zone subjected to loss of stiffness due to steel yielding and concrete
cracking, the maximum moment redistribution therefore applies at this region.
Equilibrium is maintained by increasing the positive moment by the magnitude of
redistributed moment from the haunch toe.
The elastic bending moment for a continuous composite beam of uniform
depth within each span may be modified by reducing maximum hogging moments by
amounts not exceeding the percentages given in Eurocode 4 as shown in Table 1. In
addition, Lawson, (1989) proposed a different redistribution of moments for
continuous composite haunch beams as shown in Table 2.2.
13
Table 2.1 Limits to redistribution of hogging moment to reduce. (Eurocode 4)
Class of cross section in hogging moment Class 1 Class 2 Class 3 Class 4
region
For "Uncracked" elastic analysis (Haunch 40 % 30 % 20 % 10 %

ignored)
For "cracked" elastic analysis (Haunch 25 % 15 % 10 % 0%

ignored)
Table 2.2 Maximum redistribution of negative moment in composite haunch beam at ultimate
limit state. (Lawson, 1989)
Class of cross section in hogging moment Class 1 Class 2 Class 3 Class 4
region
For "cracked" elastic analysis (Haunch 30 % 20 % 10 % 0%

included)
By comparing the percentages of moment redistribution at the haunch joint
proposed in Eurocode 4, Lawson (1989) suggested additional 5% moment
redistribution for class 1 and class 2 sections.
2.4 Plastic hinge analysis of Non-Sway Frame
Plastic hinge analysis may be carried out using either
- Rigid-Plastic Methods or
- Elastic-Plastic Methods
When using the Plastic global analysis, it is essential to make sure that
restraint be provided within a distance along the member from the theoretical plastic
hinge location not exceeding half the depth of the member (Eurocode 3, 1992).
Experimental results (Rackham, 1992) show that the haunch toe position is restrained
when a full depth stiffener is provided both sides of the web at haunch toe, and when
minimum shear connection is maintained over the hogging region (for haunch length
less than twice the depth of beam)(Boswell, 1992).
14
2.4.1 Rigid-Plastic Analysis
It is assumed in the rigid-plastic analysis approach that elastic deformations of
the member are neglected and plastic deformations are assumed to be concentrated at
plastic hinge locations. Rigid-Plastic Analysis can only be used where the section is
'
Plastic'or '
Class 1'
. This is one of the requirements in Eurocode 4 and it is assumed
that a '
Plastic'section has sufficient rotation capacity to enable the required hinge to
develop. However, the code also recognises some loss of rotation capacity due to local
buckling will be offset by the beneficial effect such as strain hardening and the finite
length of plastic regions. Due to this effect, the cross sections away from the
theoretical location are also in Class 1, or at least Class 2. Class 2 cross sections are
defined as sections that can develop the plastic moment capacity although local
buckling limits the rotation capacity and prevents full redistribution of moment at
such sections, (Price, 1992).
The test results show that the composite haunch connection is very rigid and the
connection rotation is negligible. However, failure does not occur at the composite
haunch connection because the weaker component of the composite haunch joint is at
the haunch toe. Thus, instead of the haunch connection, the haunch toe is tested to
failure. There is sufficient rotation capacity at the haunch toe for a plastic mechanism
to form in a beam even though '

Compact'or '
Class 2'section classified by Eurocode 4
is used.
The collapse load of a uniformly loaded beam is defined by the plastic failure
mechanism of the beam between the tips of the haunches, such that:
L2e
M pc + M nc wu
8
(Eq. 2.2)
15
Mpc = the positive moment resistance of the composite beam (or Mc taking
into account partial shear connection)
Mnc = the negative moment resistance of the composite beam at the tip of the
haunch
wu = the factored design load on the beam
Le = the span of the beam between the end of the haunches
The plastic failure load of other load arrangements of a beam may be
determined from first principles.
2.4.2 Elastic-Plastic Analysis
Elastic-Plastic Analysis consists of two different methods. The first method is
'
Elastic Perfectly-Plastic'which assumes that the cross-section remains fully elastic
until the plastic resistance moment is reached and then becomes fully plastic. The
second method is '

Elasto-Plastic'which shall take account of the load/slip behaviour
of the shear connection. So far, there are no application rules given for these methods
in the Eurocode 4.
2.5 Analysis of Haunch Section
In the continuous beam design, most of the approaches are based on either
elastic or plastic design. In the elastic analysis, a structure is analyzed based on elastic
global analysis and a moment envelope is obtained to design the structure. The design
has to satisfy both the ultimate and serviceability limit states. Moment redistribution is
allowed for the structure and the percentage of moment redistribution depends on
section classification. The second approach, plastic analysis is valid when critical
cross-sections are capable of developing and sustaining their plastic resistance until
the sections have fully yielded for a mechanism of plastic hinges to be present. This
16
analysis requires sufficient rotation capacity to develop a plastic hinge. Thus, class
1 or class 2 sections have to be used although local buckling limits the rotation
capacity and prevents full moment redistribution in class 2 sections, (Price, 1992).
Composite haunch beam design is based on the concept of rigid connection
thus avoid failure at beam-column connections. By strengthening the connection with
haunch, the failure mode of the joint will not occur at the connection. Instead, it shifts
the failure to the haunch toe. As long as the haunch toe is sufficient to redistribute the
moment to the sagging mid-span causing the formation of plastic hinge at mid-span,
an optimum design is achieved. In the proposed method, the beam-column connection
is the haunch connection.
Additional reinforcement in the concrete slab provides more tension resistance
at the haunch toe section. According to the classification system in Eurocode 4, large
amounts of reinforcement result in shifting of the plastic neutral axis. The steel beam
is subjected to more compression and the section may become a non-compact or
slender section, and the available rotation capacity is reduced. The percentage of
moment redistribution is, therefore, reduced further. Thus, there is always an optimum
amount of reinforcement to be used in a composite section. The increase in
reinforcement will result in an increase in moment capacity and drop in the available
rotation capacity. A balance must, therefore, be achieved between the available
rotational capacity and moment redistribution. An increase of reinforcement in a
section also increases the second moment of inertia. It carries larger moment when
moment redistribution occurs in a section with large reinforcement, the percentage of
moment redistribution may be less, but the moment that is transmitted to the mid-span
becomes more and hence the load carrying capacity is enhanced.
17
In practice, the bending resistance of the haunch section is evaluated
elastically to ensure the formation of a plastic hinge at the haunch toe with sufficient
rotational capacity. The problem of instability can be treated by conventional theory.
An approximate relationship between the elastic resistance of a haunch beam and the
plastic resistance of the parent beam is shown in Fig. 2.3. (Lawson, 1989)
In the composite condition the upper flanges of the steel beams are assumed to
be laterally and torsionally restrained by the concrete or composite slab to which they
are attached. In continuous beams, the lower compression flange is unrestrained
except the distorsional stiffness of the cross section. This is illustrated in Fig. 2.4. The
effective slenderness of the beam in lateral torsional buckling is designed as per BS
5950:Part 1 (2000) as:
LT = n u vt
(Eq 2.3)
= slenderness of the beam length between restraints

n = slenderness correction factor (for shape of bending moment diagram)
u = buckling parameter (0.9 for universal sections)
vt = slenderness factor (including torsional stiffness and other effects).
18
Detail 1 Secondary
Beam Concrete slab
Primary
Beam
Slab
Reinforcement Shear
Stud
concrete slab
Universal Beam
Detail 1
Fig. 2.1 Cross Section of Continuous Composite Beam
Column
Span Considered Considered
a) Sub-Frame Used for b) Sub-Frame Used for

Analysis of Beam Analysis of Column
Fig. 2.2 Sub-frame of Beam and Column in Non-sway Frames Analysis
19
4.0
M he D
3.5 Dh
Ms
3.0
2.5
2
Dh Dh
2.0 0.17 + 0.7
D D
1.5 M h e = elastic mom ent resista nce

of haunch se ction
1.0
M s = e lastic m ome nt re sistance
0.5 of stee l beam sec tion
0.0
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
Dep th of haunc h D h
D epth of se ction D
Fig. 2.3 Relation between Haunch Beam Elastic Resistance and Parent Beam Plastic
Resistance (Lawson and Rackham, 1989)
=0
Fig. 2.4 Distorsional Buckling of Composite Beam
20
CHAPTER 3
EXPERIMENTAL INVESTIGATION
- HAUNCH CONNECTION
3.1 General
The primary aim of carrying out full-scale joint tests is to study the behaviour
of composite haunch connections. The behaviour of haunch connections and its
ultimate capacity predominantly depend upon haunch length, haunch depth, amount
of reinforcement in the slab and number of shear studs. Test samples were, therefore,
chosen to reflect the variation in these parameters. Connection specimens were
designed with reference to a building plan layout shown in Fig. 3.1. Based on global
elastic analysis for typical design load of an office block (Refer to Beam 3/A-E), the
point of contraflexure was found to be at about 2 m from the column centreline
(Column C3). Joint specimens of cruciform section were used to simulate the internal
joint. 120 mm thick floor slab was made from normal weight concrete designed to 30
N/mm2. The cross-sectional area of slab reinforcement was determined based on the
span length, 8 m of the beams tested in the study.
The slab reinforcement was chosen as 1.34 and 2.62% relative to the effective
concrete area, which depends upon the effective slab width determined as per
Eurocode 4. Five test specimens of cruciform section were fabricated with each
specimen consisting of two different connections having different haunch length. The
depth of the haunch for all specimens was chosen equal to the depth of the universal
beam. The length was, however, varied from 250 to 968 mm in order to obtain
haunch lengths equivalent to 3.12, 5.41, 8.84 and 12.10 of the 8m beam span,
respectively. One specimen consisting of two connections was tested as a plain steel
specimen whilst the remaining four specimens were tested as composite connections.
21
Shear connection was provided by 19 diameter and 100 long studs, placed at 150
centres. One or two shear studs per group were adopted depending upon the
percentage of reinforcement viz. 1.34 or 2.62, respectively. It is expected that the
variation of slab reinforcement and haunch length selected will provide sufficient
information regarding the effects of these parameters on the behaviour, ultimate
capacity and failure mode of the joints. The ten connections are identified in the text
as H1 to H10 and the details are summarised in Table 3.1.
Table 3.1 Details of test specimens
Specimen 1 Specimen 2 Specimen 3 Specimen 4 Specimen 5
Connection H1 H2 H3 H4 H5 H6 H7 H8 H9 H10
Reinforcement, % None None 1.34 1.34 2.62 2.62 1.34 1.34 2.62 2.62
Haunch Depth 250 250 250 250 250 250 250 250 250 250
Haunch Length 250 433 250 433 250 433 707 968 707 968
Shear stud D19x95 None None 1 1 2 2 1 1 2 2
mm at 150mm c/c per
group (Total Studs) (13) (13) (26) (26) (13) (13) (26) (26)
3.2 Material Properties
3.2.1 Beam and column sections
The steel members in the test specimens were all BS Grade 43 steel. Only one
size of universal beam and column was used in this project. They are summarised as
follow:
Universal beam : 254 x 146 x UB 37

Universal column : 203 x 203 x UC 60
To obtain the yield strength of the steel members, coupons were cut from the
flanges and webs of each beam and column. They were tested in accordance with the
ASTM specification (1979). The tensile test results of the specimens are listed in
22
Table 3. 2. The yield and ultimate strength of an I or H section was calculated by
using expressions as follow:
Aw 2 Af
fy = f yw + f yf
A A
(Eq. 3.1)
Aw 2 Af
fu = f uw + f uf
A A
(Eq. 3.2)
Where,
fy Yield strength fu Ultimate strength
fyw Yield strength of web fuw Ultimate strength of web
fyf Yield strength of flange fuf Ultimate strength of flange
Aw Web area = A - 2Af Af Flange area
A Section area
Table 3.2 Summary of universal section properties and tensile test results
Section Attribute Reference Depth Width Yield Ultimate
Strength Strength
D(mm) B(mm) (N/mm2) (N/mm2)
254 x 146 UB 37 B2 256 146 309 414
203 x 203 UC 60 C2 210 205 328 498
3.2.2 Reinforcement bar
The entire reinforcement bar in the test specimens was high strength deformed
bar. Table 3.3 shows the properties of the reinforcement bars:
Table 3.3 Summary of reinforcement bar properties and tensile test results
Ref. Dia- Area Yield strength Average Ultimate strength Average
2
Meter fyr(N/mm ) fyr fur fyr
D (N/mm2) (N/mm2) (N/mm2) (N/mm2)
(mm)
1 2 3 1 2 3
T20 20 314 563 569 564 565 697 689 694 693
T16 16 201 475 498 480 484 585 588 580 584
T10 10 79 496 487 485 489 591 579 573 581
23
3.2.3 Concrete
Concrete in all the specimens was normal weight concrete with fcu designed to
be 45 N/mm2 at 28 days. The slump of the ready mixed concrete was designed to be
125 mm. Table 3.4 shows a summary of the cube test results of all the specimens on
the day of testing.
Table 3.4 Summary of concrete cube test results

Specimens Date of Date of Day of Concrete strength
Concrete Testing Testing on the day of testing
casting N/mm2
Cube 1 Cube 2 Cube 3 Average
H1 & H2 NA 03/01/97 NA Not Applicable NA
H3 & H4 10/01/97 20/01/97 10 days 44 42 44 43

H5 & H6 03/03/97 13/03/97 10 days 42 43 42 42
H7 & H8 14/04/97 28/04/97 14 days 44 42 45 43
H9 & H10 09/05/97 19/05/97 10 days 43 40 38 40
3.3 Fabrication of test specimens
Details of a typical test specimen are shown in Fig. 3.2. A universal beam
section 254 x 146 x UB37 and column section 203 x 203 x UC60 were used to
fabricate all test specimens. The column of 3480 mm long was first fixed to the top
and bottom girders of the testing rig.
A 20mm thick endplate was welded by means of 10mm fillet weld to the beam
end that is to be connected to the column. Beams 2020 mm long were then connected
on either side of the column through endplates, selected haunch section and high
strength bolts of BS 4390 Grade 8.8, 20 mm diameter. The bolts were tightened with
a torque wrench to 200 Nm. Care was taken to ensure that the column and the beam
sections lie in the same vertical plane. For composite specimens viz H3 to H10, shear
studs were welded to the top flange of the beam sections before being connected to
24
the column section. Once the fabrication of plain steel connections was complete,
formwork of required size i.e. slab depth and width, were built to the beam section;
reinforcement bars to achieve the selected proportion were laid and preparation for
casting concrete slabs were made. The depth and width of the concrete slab were
kept, respectively, as 120 mm and 1050 mm for all composite beams. Grade 30
concrete, made from locally available materials, was poured into the form work in
stages ensuring adequate compaction by means of vibrator. For concrete, 28-day
strength was achieved in 7 days by adding an admixture, trade named Rapidart.
Concrete cubes of sufficient number were cast along with the test specimens and they
were tested on the same day as that of the specimens. A typical test specimen, ready
for concreting, is shown in Fig. 3.3.
3.4 Test set-up
All specimens were tested to failure in a rig, 6 m long and of 1,000 kN capacity.
Two 50 tonne stroke-controlled hydraulic actuators (250mm stroke), attached to an
overhead reaction beam were used to apply the load at the free end of the specimen.
Each of the two actuators was positioned at a distance of 1.8 m from the face of the
column thus giving a moment arm of 1.8 m for the cantilever beam. These actuators
were operated by computer controlled pumps in order to ensure that the load
application was gradual and the increment properly controlled. The two ends of the
column in the sub-assembly being tested were connected to the longitudinal beams at
the top and bottom of the testing rig by means of pins. The testing frame is illustrated
in Fig. 3.4 and a typical assembly in which a specimen is mounted is shown in Fig.
3.5.
25
3.5 Instrumentation
The main objective of the joint tests was to obtain the full response of the joints in
terms of moment-rotation relationship. Moment was calculated using the equilibrium
of force and rotation and measured by using inclinometers and counter checked by the
results obtained from displacement transducers. Five inclinometers were placed along
the centreline of the beam section. One inclinometer, to measure the column rotation,
was located at the intersection of column and beam centrelines and in addition, two
inclinometers were placed one on each side of the column at a distance of 100 mm
from the haunch toe. Three 50mm displacement strain gauge type transducers
(SGTD) were used to measure the displacement of the joint so that the joint rotation,
j could be calculated. They were attached to a rod parallel to the haunch at distances
of 100 mm, 200 mm and 300 mm measured from the column flange along the rod.
Details of the instrumentation are shown in Fig. 3.6. A 200 mm displacement
transducer was used to measure the vertical displacement at the loading point.
Electrical resistance strain gauges were used to measure the strain in the steel so as
to monitor yielding and to determine the failure modes. They were placed at points of
high stress intensity, at the top and bottom of the beam flanges near the column
flange, at the haunch toe and reinforcement bars as shown in Fig. 3.6. Besides, strain
gauges were also placed on some bolts connecting the beams to the column flange.
This was intended to enable the tensile forces in the bolts at each of the load steps to
be determined.
3.6 Testing procedure
After the specimen was positioned in place, loading and instrumentation
devices were connected to the data processing unit. All the readings, inclinometers,
26
transducers and strain gauges were initialized. Ten percent of the estimated failure
load was first applied to the specimen and all readings were checked for continuity
and proper recording by the data processing units. The load was then released and
reapplied in order to remove any slack that may exist at the support before the actual
testing of the joint. This process of repeated loading is expected to ensure proper
functioning of the load application and other measurement devices. The entire load
application was performed in three stages. In the first stage, load was applied until the
first crack was observed in the concrete and, in the second and third stages load was
increased up to 60% and 90% of the estimated load, respectively. In each stage, the
load was released after achieving the intended load and then reapplied. This process
of loading helps to obtain the rotational stiffness of the connection and to compare the
unloading stiffness at different loading stages. In the final stage, loading was
continued until the failure of the specimen.
As mentioned earlier each of the specimens consisted of two connections,
one with a shorter haunch length and the other with a longer haunch length.
Therefore, the load application and other measurements were monitored separately.
Load was applied in equal increments to each of the connections at the initial stages of
loading. Once the weaker connection attained the load close to the failure load, care
was taken to balance the load on both connections. The weaker connection was
allowed to fail at its maximum capacity and the load on that connection was
maintained at that level whilst the load on the other connection was continued until it
reached its failure. The ultimate load and the failure mode for each of the connections
in a particular specimen were thus noted at the end of the test on that particular
specimen. The same procedure was repeated for all the five specimens.
27
3.7 Data assessment
Proper representation of moment-rotation curves obtained from joint tests is
essential for the analysis and design of continuous composite beams. It is, therefore,
essential to study in detail and select carefully to represent joint behaviour adequately.
Readings for rotation can be categorized into four types:
a. Beam rotation b,
b. Column rotation c,
c. Connection rotation ,
d. Inelastic rotation ie
3.7.1 Beam rotation, b
The rotation along the beam varies from the face of the column flange to the
end of the beam. It is assumed that connection rotation measured near the column
flange does not include the rotation due to beam flexure. This is because the beam
flexure is assumed to be very small and can be neglected at the section near the
column flange. Furthermore, with a haunch at the connection, the sections become
more rigid and thus, the beam flexural rotation is negligible.
3.7.2 Column rotation, c
Column rotation, c was measured by means of an inclinometer placed at the
intersection of column and beam centrelines. As a result of unbalanced moment or
unsymmetrical geometry about the centreline of the joint, the column will rotate and
the measured rotation is with reference to the vertical axis.
3.7.3 Connection rotation,
Connection rotation is defined as the relative rotation of the column and beam
at the joint as shown in Fig. 3.7. The connection rotations in these joint tests were
28
calculated based on the inclined transducers mounted on the column flanges. This
arrangement enabled the measurement of relative rotation between the columns and
the beams by subtracting the total joint rotation tot from c.
3.7.4 Inelastic Rotation, ie
Inelastic rotation, ie is defined as the rotation at a particular section after any
of the extreme fibres had yielded. This is especially applied at the haunch toe section.
During the elastic stage, no rotation occurred at the haunch toe because the tangent
line at this point will remain almost horizontal. Soon after the first yield, this point
will start to rotate. Although the entire section has not achieved the full plastic stage,
inelastic rotation starts to take place at this level of load. Inelastic rotation at the
haunch toe, ie can be seen in Fig. 3.7.
3.8 Joint stiffness, Rki
The joint stiffness of the joint can be divided as uncracked and cracked
stiffness. The uncracked stiffness is the gradient of the moment rotation curve before
cracking. The stiffness corresponding to cracking shall be obtained from the
unloading path as shown in Fig. 3.8. The stiffness of haunch joint can be divided into
two parts, first is the stiffness at connection and second is the stiffness at haunch toe
section.
3.9 Joint ultimate moment, Mu
The ultimate moment resistance of the joint Mu is equal to the peak value of
the moment-rotation characteristic as shown in Fig 3.8. For a full strength composite
haunch joint, the moment capacity of the haunch connection has to be greater or equal
29
to the hogging moment capacity of the composite section at the haunch toe. However,
certain criteria has to be followed to make sure failure occurred at the haunch toe.
3.10 Joint rotational capacity, Cd
The joint rotation capacity of a beam-to-column connection is taken as the
rotation achieved at the ultimate moment resistance of the joint (see Fig. 3.8).
3.11 Haunch Connection Capacity
The capacity of the haunch connection is calculated based on stress-strain theory:
i) Haunch connection without slab reinforcement

The steel haunch connection tested in the project is shown in Fig 3.9. The
capacity of this joint can be determined as follows:
The bolt will fail in this haunch connection is accordance with the guideline by
the SCI/BCSA Connection Group (BCSA, 1995).
Only one row of bolt is used, thus the full tension capacity of the bolts is:
T = Rb < Rhf
Thus, taking moment about haunch flange
M hu = Rb x ( D Db + Dh T / 2)
(Eq. 3.3)
ii) Haunch connection with slab reinforcement

Referring to Fig. 3.9, it is found that for specimens with 1.34 and 2.62% slab
reinforcement, the PNA lies in the haunch web when,
Rr + Rb < Rhf + Rhw

Thus, moment can be determined as follow :
Rr + Rb Rhf
yc =
1.2 p y thw
Rhw = 1.2 yc p y t hw
M hu = Rr ( Dr + D + Dh Thf / 2) + Rb ( D Db + Dh Thf / 2) Rhw ( yc + Thf / 2)

(Eq. 3.4)
30
Mhu = Moment capacity of composite haunch connection
py = Design strength for steel
Rb = Bolt in tension
Rhf = Haunch flange capacity
Rhw = Haunch web capacity
Rr = Reinforcement in tension
thw = Thickness of haunch web
Thf = Thickness of haunch flange
yc = Distance from the top of haunch flange
Note : The factor 1.2 is used because strain hardening contributes 20% of dispersion
into the web and the root contributes 20% of bearing strength. If the contribution of
compression comes only from the haunch flange, a factor 1.4 can be used.
(SCI/BCSA 1995)
Although for end-plate composite connections, the first bolt row seldom
achieves its full tension capacity, for haunch composite connection the first bolt row
is always able to achieve it tensile yield capacity. This is because the PNA hardly over
the lever of first bolt row. And this is proved in one of the test in specimen H8.
3.12 Haunch toe moment capacity
Haunch toe moment capacity can be obtained based on Eurocode 4, clause 4.4.
The resistance of cross-sections of beams may be determined by plastic theory only if
the section is in Class 1 or Class2. And the following assumptions shall be made:
- the tensile strength of the concrete is neglected
- plane cross-section of the structural steel and reinforced concrete parts of a
composite member each remain plane.
31
3.13 Joint Test Results and Discussion
3.13.1 Comparison of test results
Ultimate moment obtained from the experiments along with those predicted by
the method given in Section 3.11 for all the test specimens are summarised in Table
3.5 (Shanmugam et. al., 2002). Also, the experimental values are compared with the
predicted results. Extensive measurements were made for strain and displacement at
various locations in the test specimens. However, only typical results at selected
locations are presented for dicussion. Detailed discussion is presented in the following
sections for each of the connections tested.
Table 3.5 Summary of Test Results
Specimen 1 2 3 4 5
Connection H1 H2 H3 H4 H5 H6 H7 H8 H9 H10
(1) Ultimate load (kN) 138 117 162 181 222 241 258 312 306 -
(2) Haunch Toe Section - - 251 248 344 330 282 - 334 -
Capacity, kNm
Test
(3) Haunch Heel 247 211 - - - - - 562 - -

Connection Capacity,
kNm
(4) Ultimate load (kN) 130 130 165 187 206 233 233 296 292 -
(5) Haunch Toe Section - - 255 255 319 319 255 - 319 319
Prediction
Capacity, kNm
(6) Haunch Heel 234 234 - - - - - 533 - -
Connection Capacity,
kNm
Failure Mode con Con toe toe toe toe toe con toe Na
Ratio (1)/(4) 1.06 0.90 0.98 0.97 1.08 1.03 1.10 1.05 1.05 Na
*Failure of Joint H8 occurred at haunch heel. The value shows in the column are haunch heel failure moment. The
prediction is based on the plastic stress block theory as shown in Figure 3.9.
3.13.2 Connections H1 and H2
Views after failure of connections H1 and H2 (specimen 1) are shown in Fig.
3.10(a) and the corresponding moment rotation curves in Fig. 3.11. This specimen
was a plain steel haunch joint with haunch depth 250 mm and lengths 250 mm and
433 mm, respectively, for connections H1 and H2.
32
The moment at failure for the connection H1 was 247 kNm at the connection and
the corresponding value at the toe was 214 kNm. The connection lost the capacity to
carry further load due to the failure of bolts in tension (Refer 3.10(b)). The recorded
strain in the bolt was 2800 at failure. Yielding was also observed at several
locations in the steel beam prior to bolt failure. The first yielding was detected at the
compression flange near the haunch toe followed by tensile yielding of the web at the
haunch heel. High stress concentration was found to occur in the beam flange at the
intersection of the haunch toe with the beam. The steel beam section at the haunch
toe reached the calculated plastic capacity (150 kNm) prior to the bolt failure. All
strain gauges in the beam at the haunch toe section yielded at an applied moment
equal to 180 kNm. The connection continued to resist additional load until the failure
occurred due to tensile fracture of the bolts. The increase in moment capacity beyond
the beam plastic moment capacity may be attributed to strain hardening. Negligible
rotation was measured at the beam-to-column connection. The calculation of rotation
based on transducer readings showed that the rotation of the haunch connection was
less than 2 milliradians. Based on experimental values of strength and rotation, it is
concluded that this steel haunch joint is a rigid full-strength connection.
The maximum moment at failure in the case of connection H2 was 211 kNm at the
connection. Failure was triggered by tensile fracture at the bolt thread as in the case
of the connection H1. The recorded strain at the time of failure was 2900. The first
yield occurred in tension region at the haunch heel and was followed by compressive
yielding at the intersection of haunch toe with the beam flange. The distribution of
stress at the intersection of haunch toe with the beam flange was different from that of
H1. With longer haunch length, it was observed that web stiffener and inner beam
flange were subjected to large stress concentration compared to H1. Test results show
33
that when yielding occurred at compression beam flange, the stiffener and inner beam
flange at haunch toe intersection also yielded extensively. Corresponding applied
moment on the beam section at the haunch toe before the bolt failure was found as
160 kNm. This is in excess of plastic capacity (150 kNm) of the beam and it may be
due to strain hardening. Before the haunch toe section reached its ultimate capacity,
the haunch connection failed. As in the case of connection H1, there was negligible
rotation measured for this connection. The calculated connection rotation for this
steel haunch connection was less than 2 milliradians. This joint is also a rigid full-
strength connection as per Eurocode 3 connection classification.
Moment rotation curves in Fig. 3.11 show that H2 with longer haunch length is
less stiff compared to H1 with shorter haunch length. This may be due to the fact that
initial slackness on H2 side compared to H1 resulted in larger deflection on H2 side
and thus showed larger rotation. Also, no meaningful results could be obtained for
ultimate loads since failure occurred on both connections due to excessive tension in
the bolts. The ultimate moment for H1 and H2, therefore, do not compare favourably
with the corresponding predicted values.
Views after failure for connections H3 and H4 are shown in Fig. 3.12 and the
corresponding moment rotation relationships presented in Figs. 3.13 and 3.14,
respectively. Specimen 2 is a steel-concrete composite haunch joint. The concrete
slab was 120 mm thick, 1050 mm wide and, 1.34 % slab reinforcement consisting of 8
numbers of T16 deformed bars were used in the slab. The haunch depth and length are
250 mm.
The ultimate moment capacity for connection H3 was found as 251 kNm. Failure
occurred at the haunch toe where a plastic hinge was found to have formed. It can be
34
seen from Fig. 3.12 that the compression beam flange at the point of intersection with
the haunch toe has buckled. All the strain readings at the section, both in steel beam
and reinforcement bars, showed extensive yielding thus confirming the formation of
plastic hinge. Despite the formation of this hinge, the stress at haunch sections away
from the toe was generally found to be less than yield. No rotation was measured in
the haunch connection and it is, therefore, concluded that this steel-concrete haunch
connection is a rigid full-strength connection. Test results showed that the ultimate
moment capacity of the composite section (251 kNm) at haunch toe section was close
to the plastic capacity (255 kNm) determined as per Eurocode 4 thus establishing the
good correlation between the experimental and codal predictions. However, the code
does not seem to predict the stiffness of the composite section. One should appreciate
the fact that stiffness of the section is not a requirement in analysis since Plastic
Analysis requires only the ultimate moment and rotational capacities. As long as the
section is able to resist the limit load and provide sufficient rotation which allow
moment redistribution, connection stiffness is not a requirement in a rigid frame
analysis. It can be seen from Fig. 3.13 that the moment-rotation curve is steep in the
elastic range with rotation practically zero. The inelastic rotation at ultimate moment
is 27 milliradians.
The haunch depth and length for the connection H4 are 250 mm and 433 mm
respectively. The failure moment in this case was 248 kNm which is close to the
predicted capacity as per Eurocode 4. First yield was detected in the compression
region at the haunch toe near the beam flange and was followed by yielding in one of
the tension reinforcement bars. The compression beam flange at the junction with the
haunch toe was found to have buckled inelastically (Fig. 3.12). At ultimate stage,
yielding was detected only at the compression beam flange and tension reinforcement.
35
The haunch length in H4 was larger compared to H3. No yielding was noticed in
haunch flange and the compressive force could have been distributed to the stiffener
and inner beam flange. No noticeable rotation was measured in the haunch
connection. Fig. 3.14 shows the moment-rotation curve for a section at the haunch
toe. The inelastic rotation at ultimate moment was 55 milliradians, more than the
rotation for connection H3.
Views after failure of the connection H5 and the corresponding moment-
rotation curve are shown in Figs. 3.15 and 3.16, respectively. Similarly for the
connection H6, view of the tested specimen and the moment-rotation curves are
shown in Figs. 3.17 and 3.18. Specimen 3 consisting of connections H5 and H6 was
the same in all respects as the specimen2 except that the concrete slab was reinforced
with 2.62 %, 10 numbers of T20 deformed bars in this case.
The connection H5 failed at an ultimate moment of 344 kNm and the failure
occurred at the haunch toe where plastic hinge was formed. It can be seen from
Fig.3.16 that there is inelastic buckling in beam flange near the haunch toe. The
tension reinforcements were found to have yielded as shown by strain gauge readings.
Test results also indicated an extensive yielding in the compression beam flange and
the haunch near the toe was relatively unaffected until haunch toe moment reached a
value of 304 kNm. The rotation measured by the inclinometer C in the column was
very small and hence can be neglected. The ultimate moment capacity of composite
section (344 kNm) at haunch toe section is found to be close to the calculated plastic
capacity (319 kNm) in accordance with Eurocode 4. Fig. 3.16 shows the moment-
rotation curve at the haunch toe section from which it is found that the inelastic
rotation corresponding to the ultimate moment is 52 milliradians.
36
The haunch length in connection H6 was longer than that in H5. Fig. 3.18
shows the inelastic buckling in the beam flange and extensive yielding was observed
form the strain gauge readings located at the intersection of haunch toe with the beam
flange. The failure moment for this connection was 330 kNm, close to the moment
capacity predicted by using Eurocode 4. As in the other cases, the first yield was
detected at compression region at the haunch toe near the beam flange followed by
yielding in the beam web. The strain gauges in the tensile reinforcement also showed
yielding at a load corresponding to the ultimate condition. At the intersection of
haunch toe, yielding was observed in the compression beam flange and in the
stiffeners; the inner beam flange towards the haunch heel, however, did not show any
sign of yielding. Moment-rotation curve for a section at haunch toe shown in Fig.
3.18 shows an inelastic rotation of 45 milliradians corresponding to ultimate moment.
Figs.3.19 and 3.21 show the views after failure of connections H7 and H8.
Moment-rotation relationships for the connections are presented, respectively, in
Figs.3.20 and 3.22. These two connections have been tested as parts of the specimen
4 in which the haunch lengths were chosen approximately equal to three times as in
H7 and four times the depth as in H8. The reinforcement in the slab was kept the
same as in specimen 2.
The haunch length, 707 mm in connection H7 is equivalent to 8.84 % of the
design span. The failure moment for this connection was 282 kNm. The failure
occurred at the haunch toe (as shown in Fig. 3.19) where inelastic buckling was
observed. First yield was detected at the compression beam flange followed by
yielding in the beam web. Reinforcement in the slab were also found to have yielded
almost at the same time when yielding occurred in the compression zone of the beam
37
web. Thus the whole section at the haunch toe yielded leading to formation of plastic
hinge. Following the yielding at the compression beam flange at the toe, the beam
flange close to the haunch heel also yielded. Yielding was also noticed at the haunch
flange and web. The ultimate moment capacity for this connection was 282 kNm. It
is found that it is possible to control the failure mode by varying the haunch length
and that longer haunch length shifts the failure from haunch toe to haunch heel. Test
results revealed that the ultimate capacity of composite section (282 kNm) at haunch
toe section was close to the plastic capacity (255 kNm) calculated by Eurocode 4.
Fig. 3.20 shows the moment-rotation curve for a section at the haunch toe and that the
inelastic rotation at ultimate moment (282 kNm) is 21 milliradian.
Connection H8 is the same as H4 except that the haunch length in this case
was 968 mm or 12.10 % of an 8 m span beam. The view after failure of the specimen
is shown in Fig 3.21 in which it can be seen that failure occurred at the haunch heel.
The failure moment for this connection was found as 562 kNm close to the predicted
capacity of 533 kNm. The first yield was detected at the compression beam flange
near the haunch toe and it was followed by yielding at the beam web. Compression
yielding of beam web continued towards the neutral axis. No further yielding was
detected at the haunch toe section. Yielding was found to occur next at the haunch
flange near the end-plate. Reinforcement bars in the slab were also found to have
yielded. Fig. 3.22 shows the moment-rotation curve at the haunch heel.
Fig.3.23 and Fig. 3.24 show the view after failure and the corresponding
moment-rotation curves of the connections H9. The specimen 5 was the same as
Specimen 4 except that the reinforcement in the earlier was 2.62 % consisting of T20
38
deformed bars. The difference between the connections H9 and H10 lies in the
haunch length which was 707 mm in H9 and 968 mm in H10.
Failure of the connection H9 occurred at the haunch toe and the maximum
moment at collapse was 334 kNm. Fig. 3.23 shows the inelastic buckling at the beam
flange near the haunch toe. The first yield was detected at the compression beam
flange followed by yielding in the beam web. Yielding of the reinforcements was also
noticed from strain gauge readings corresponding to ultimate load. The whole section
at the haunch toe was thus found to have fully yielded and plastic hinge formed.
Yielding was also observed at the inner beam flange at the haunch toe intersection.
The experimental value (334 kNm) of the ultimate capacity for the composite section
at the haunch toe is close to the calculated plastic capacity of 319 kNm as per
Eurocode 4. Fig. 3.24 shows the moment-rotation curve for the section at haunch toe.
The inelastic rotation corresponding to ultimate moment (334 kNm) is 43 milliradian.
toe. The inelastic rotation corresponding to ultimate moment (334 kNm) is 43
milliradian.
The connection H10 was the same as H9 except that the haunch was longer
i.e. 968 mm in H10. First yield was detected in the compression flange at the haunch
toe followed by yielding in the beam web. Strain gauges placed on the slab
reinforcements showed no yielding. The concrete cover for tensile reinforcement on
the H9 side started to give way due to excessive load, Fig. 3.25, the resistance to load
dropped rapidly and the excess load shed on to the H10 side. The progressive failure
on H10 side soon after the collapse of H9 was so rapid that no meaningful readings
could be made.
39
3.14 Effect of slab reinforcement ratio
Results corresponding to the connection H2, H4 and H6 are compared in
Fig.3.26 in order to investigate the effect of reinforcement ratio on moment rotation
characteristics. These connections have same haunch depth (D) and haunch length
(2D), but of different slab reinforcement percentage viz. 0, 1.34 and 2.62%,
respectively. Higher percentage of reinforcement in the slab shifts the failure from the
steel connection to the haunch toe of the composite section. Failure of H2 connection
was triggered by tensile fracture at the bolt thread. Failure of H4 occurred at the
haunch toe in which the composite beam section has almost fully yielded, and the
compression beam flange at the point of intersection with the haunch toe buckled
inelastically. H6 was the same in all respects as the specimens 2 and 4 except that the
concrete slab was reinforced with 10 numbers of T20 deformed bars. Failure in this
case occurred at the haunch toe, as shown in Fig 3.17 where the composite beam
section in negative bending has almost fully yielded in compression. Further increase
in reinforcement will not result in any significant improvement of load carrying
capacity since the limit of resistance for the steel section in compression has been
reached with plastic neutral axis shifted to the concrete slab.
3.15 Effects of haunch length
The haunch length was varied from a value equal to the depth as in H3,
approximately equal to two times the depth as in H4, time times the depth as in H7
and four times the depth as in H8. The reinforcement in the slab was kept the same as
in Specimen 2. Fig.3.27 shows the comparison between the load-displacement curves
obtained for these specimens. Failure occurred at the haunch toe in the case of H4 and
H7. For H7, the ultimate capacity of composite section (282kNm) at haunch toe
section was close to the calculated plastic capacity (255kNm) by Eurocode 4.
40
Connection H8 is the same as H4 except that the haunch length in this case was 968
mm or 12.10% of an 8-m span beam. Failure occurred at the haunch heel near to the
connection as shown in Fig.3.21. Failure moment for this connection was found to be
562kNm, which is close to the predicted value of 533kNm. The test results show that
it is possible to control the failure mode by varying the haunch length and that longer
haunch length shifts the failure from haunch toe to haunch heel. Further increase in
haunch length will not result in an enhancement of load carrying capacity since the
failure is controlled by the haunch connections.
3.16 Conclusions
Experiments on composite haunch connections are described and results
corresponding to ultimate moment capacity, moment-rotational characteristic and
rotation capacity are presented. These connections are classified as a full strength
rigid connection in accordance with Eurocode 4. It is confirmed by the test results
which show that the measured moment capacity for all connections is larger than the
plastic capacity of the beams and beam-column or connection rotation in all tests was
very small less than 2 milliradians. The prediction method is found to estimate the
ultimate capacity of composite beams fall within 10% of the predicted value. Haunch
toe can be strengthened effectively by means of web stiffener to the full depth of the
beam. No lateral distorsional buckling was observed in all the rest specimens.
However, the length of the haunch is limited to 12.10% of the beam span.
Experimental observations show that the failure is localized at the haunch toe section.
Haunch length has no significant effect on rotation capacity and it is found that
rotation at the ultimate capacity always falls within 30 to 45 milliradians. Increase in
reinforcement from 1.34 to 2.62% does not reduce rotation capacity significantly but
it increases the ultimate moment capacity of the composite section. Longer haunch
41
length tends to shift the failure to the haunch heel of the connection. Hence rotation
capacity of the connection becomes critical if plastic analysis is required.
42
Loading Data:
a) Concrete slab = 2.88 kN/m2
b) Construction Load = 0.50 kN/m2
c) Building services load = 0.70 kN/m2
d) Imposed Load = 5.00 kN/m2
Fig. 3.1 Building Plan Layout for Joint Specimens'Design
43
Fig. 3.2 Cruciform Joint Specimen
44
Fig. 3.3 Test Specimen Ready for Concrete Casting
Fig. 3.4 Typical Joint Test Specimen
45
Fig. 3.5 Joint Test Specimen Ready for Testing
Concrete slab
Shear stud Reinforcement bars
1/2/3/4/5 6/7/8/9/10 11/12/13/14/15 16/17/18/19/20
59
21 26 40 49
62/63
22 27 64/65 41 50
A 23 B 28 C 42 D 51 E
24 29 43 54 52
35 60
25 36 30 44 55 53
37 31 45 56
38 57
32 46
33 47
34 48
61
39 58
Note:
a)
A B ...... Inclinometer for rotation measurement
b)
1/2/....20 Strain gauges on slab reinforcement bars
c)
21,22,23......61 Strain gauges on structural steel
d)
62/63/64/65 Strain gauges embedded in tension bolts
e)
Displacement transducer
Fig. 3.6 Instrumentation of Test Specimen
46
c P kN
Cracking of concrete slab
Shear stud Reinforcement bars
ie
Universal beam
Haunch
Universal column
Fig. 3.7 Definition of Rotation in a Joint
kNm
Mu
M2
M 2 M1
R ki =
R ki 2 1
M1
1 2 Cd radian
Fig. 3.8 Moment-rotation Curve of Connection
47
Fig. 3.9 Stress-strain Block of Haunch Connection
48
Fig. 3.10 (a) View after Failure of Specimen H1 and H2
Fig. 3.10(b) View after Failure of Tension Bolt of Specimen H1 and H2
49
300
250 247kNm
Moment, kNm
211kNm
200
150
100
H1
50
H2
0
0 5 10 15 20 25 30 35 40 45 50 55 60
Rotation, miliradians
Fig.3.11 Moment-Rotation Curve of H1 and H2
50
Fig. 3.12 View after Failure of H3 and H4
500 First Yield at Compression Flange

450 at 138kNm
400 ie
Moment, kNm
350
300 First Crack at 89kNm 250 kNm
250
200 Rki
150 =30
100 Rki= 19 Haunch Toe
50 Rki=26
Haunch Heel
0
0 5 10 15 20 25 30 35 40 45 50 55 60
Fig. 3.13 Moment-Rotation Curve of H3
51
500
450
400
First Yield at Compression Flange
at 117kNm
Moment, kNm 350
ie
300
First Crack at 69kNm 248 kNm
250
200
Rki
150 =26
Rki= 17
100 Haunch Heel
50 Rki=21 Haunch Toe
0
0 5 10 15 20 25 30 35 40 45 50 55 60
Fig. 3.15 View after Failure of H5
52
500 at 128kNm
450 ie
400 First Crack at 98kNm 344 kNm
Moment, kNm 350

300
250
200
150 Rki Rki=27
100 =43 Haunch Toe
50 Haunch Heel
0
0 5 10 15 20 25 30 35 40 45 50 55 60
53
First Yield at Compression
500 Flange at 130kNm
450 ie
400
Moment, kNm 350
300
250
200
150
100 Rki Rki=20 Haunch Toe
=24 Haunch Heel
50
0
0 5 10 15 20 25 30 35 40 45 50 55 60
54
500 Flange at 164kNm
450 i
400 e
Moment, kNm 350 First Crack at 79kNm 289 kNm
300
250
200
Rki
150 =29 Rki Rki= 20
100 Haunch Toe
=22
50 Haunch Heel
0
0 5 10 15 20 25 30 35 40 45 50 55 60
55
Haunch Flange at 435kNm
700
562 kNm
600
500
Moment, kNm
Flange at 201kNm
400
300
200
Haunch Heel
100
Haunch Toe
0
0 5 10 15 20 25 30 35 40 45 50 55 60
56
500 at 159kNm
ie
450
400
Moment, kNm 350
300
250
Rki
200 =25
150
100 Rki= 16 Haunch Toe
Rki=20
50 Haunch Heel
0
0 5 10 15 20 25 30 35 40 45 50 55 60
Fig. 3.25 Premature Failure of H10
57
300
250
200
Load, kN
150
100 H6- 1D2D 2.62%

H4- 1D2D 1.34%
50
H2- 1D2D 0%
0
0 20 40 60 80 100 120 140
Displacement, mm
Fig. 3.26 Comparison of Load-Displacement Curve of H2,H4 and H6
350
300
250
Load, kN
200
150
H3- 1D1D 1.34%
100 H4- 1D2D 1.34%
H7- 1D3D 1.34%
50
H8- 1D4D 1.34%
0
0 20 40 60 80 100 120 140
Displacement, mm
Fig. 3.27 Comparison of Load-Displacement Curve of H3, H4, H7 and H8
58
CHAPTER 4
EXPERIMENTAL INVESTIGATION
- HAUNCH BEAM
4.1 Introduction
An experimental study on three continuous haunch beams is reported in this
chapter. The continuous composite haunch beams were designed based on a series of
haunch connection tests described in chapter 3. Haunch connection can be designed as
rigid full-strength connection with proper detailing. The haunch toe is planned as the
weakest section to enable the formation of plastic hinge. When the first failure
happened at haunch toe, the moment will be re-distributed to mid-span in order to
obtain a failure mechanism. The degree of moment redistribution depends on the
ductility of the section during failure. An optimum design of composite beam is
achieved if both the hogging and sagging capacity of the composite beam is fully
utilized.
In a composite continuous beam, plastic hinge failure mechanism will only
occur if the rotation capacity at the connection is adequate. Insufficient rotational
capacity in a connection will cause non-ductile failure. Experimental results show that
haunch connection is able to redistribute moment to the mid-span to form plastic
hinge failure mechanism. The results show that plastic hinge failure mechanism can
be achieved in designing continuous composite haunch beam. However, experimental
results also show that plastic hinge failure mechanism is achievable without looking
into the availability of rotational capacity of a beam section. This situation takes place
when all the required plastic hinges form at the same time under the same loading.
The objects of these beam tests is to study the behaviour of composite haunch
beam by modelling as a non-sway continuous steel-concrete composite beam. The
59
behaviour of haunch connection and its ultimate capacity predominantly depend upon
haunch length, haunch depth and the amount of reinforcement in the slab and number
of shear studs. Test specimens were, therefore, chosen to reflect the variation in these
parameters. Beam specimens as shown in Fig. 4.1 were used to simulate the
continuous beam. The slab reinforcement was chosen as 1.34 and 2.62% relative to
the effective concrete area, which depends upon the effective slab width determined
as per Eurocode 4. Three specimens were fabricated with each specimen consisting of
connections having different haunch length and reinforcement combination. Depth of
the haunch for all specimens was chosen equal to the depth of the universal beam.
The length was, however, chosen to be varied from 433 mm and 968 mm in order to
obtain haunch lengths equivalent to 5.41 and 12.10% of 8 m beam span, respectively.
These beam tests could provide a clear picture of the behaviour of continuous
beam. Moment redistribution within the span and the details of formation of plastic
hinges is also studied under ultimate loading. The results could then be compared with
plastic theory and finite element results to establish the design method. The specimens
are identified in the text as B1, B2 and B3 and the details are summarized in Table
4.2.
4.2 Material Properties
4.2.1 Beam and column sections
Steel members chosen for the beam test specimens are same as those used in the
joint tests. One size of universal beam and column was used, i.e. universal beams
254x146xUB37kg/m and universal columns 203x203xUC60 kg/m. Same batch of
steel with the relevant material heat number were used as per the joint test. Therefore,
60
the yield strength and other mechanical properties were the same as those reported in
Table 3.2 in Chapter 3.
4.2.2 Reinforcement bar
The reinforcement bar in beam test specimens was high strength deform bar as
per joint test and, the tensile test results are shown in Table 3.3. Same batch of
reinforcement used in the joint test was chosen for the beam test. Material properties
of the rods and the configuration of the reinforcement were be of the same as those in
joint tests. Therefore, a direct comparison of the joint test results can also be made to
the connection in the beam test results.
4.2.3 Concrete
Concrete in all the specimens was normal weight concrete with fcu designed as
45 N/mm2 at 28 days. The slump of the ready mixed concrete was designed to 125
mm. Table 4.1 shows summary of the cube test results for all the three specimens on
the day of testing.
Table 4.1 Summary of concrete cube test results for beam specimen
Beam Date Date of Day of Concrete strength N/mm2
Specimens of Concrete Testing Testing
on the day of testing
casting
Cube 1 Cube 2 Cube 3 Average
B1 03/12/97 11/12/97 8 days 32 32 33 33

B2 13/02/98 20/02/98 7 days 31 33 31 32
B3 11/05/98 22/05/98 11 days 39 40 39 39
4.3 Fabrication of test specimens
Fabrication of beam specimens complied with BS 5950: Part 2. Details of the
beams are summarised in Table 4.2. Similar to the joint test in Chapter 3, universal
61
beam section 254 x 146 x UB37 and column section 203 x 203 x UC60 were used to
fabricate all beam test specimens.
The steel material is the same as that used in joint tests. Specimens were not
painted or coated with any chemical.
Table 4.2 Details of test specimens
Beam Specimens Specimen B1 Specimen B2 Specimen B3

a. Reinforcement, % 1.34 (8T16) 2.62 (10T20) 1.34 (8T16)
b. Haunch Depth, mm 250 250 250
c. Haunch Length, mm 433 433 968
d. Slab width, mm 1400 1400 1400
e. Slab thickness, mm 120 120 120
f. Shear studs (per trough) 1 2 1
After the erection of universal column and beam section, wooden formwork
was built to the beam section; reinforcement bars to achieve selected proportion were
laid and preparation for casting concrete slabs were made as shown in Fig.4.2. Fig.4.3
shows a typical beam test specimen ready with reinforcement bar ready for
concreting. Grade 30 concrete was poured into the formwork in stages ensuring
adequate compaction by means of vibrator. For concrete, 28 day strength was
achieved in 7 days by adding an admixture trade named Rapidart. Concrete cubes of
sufficient number were cast along with the test specimens and they were tested on the
same day as that of the specimens. A typical beam test specimen, ready for testing, is
shown in Fig.4.4.
4.4 Test Set-up
A typical test specimen consisting of a beam marked B2 spanning 8 meter
between two columns marked C1 and two cantilever beams of 2 metre marked B1
connected to the columns C1 are as shown in Fig. 4.5. This test set-up is to model a
62
continuous beam in a non-sway frame. Design of beam specimens was referred to a
building plan layout shown in Fig. 3.1. The primary and the secondary beams span 8
m and 12 m, respectively. Based on elastic global analysis and typical design load
(Refer to Beam 3/A-E), the point of contraflexture is 1.8 m or about 22.5 % of span
length away from the column C-3. The loading points at the cantilever beam were
assumed to be the points of contraflexure.
The column length is not modeled in this experimental set-up because it is
assumed that the moment at both sides of column could be balanced by the 50 ton
capacity counter reaction actuators at the tip of the cantilever beams as shown in Fig.
4.1. Columns were prevented from out-of-plane sway by using a diagonal bracing
marked S1 in Fig.4.5. Columns marked C1 and diagonal bracing marked S1 were
bolted to base plate marked P1, fixed to the laboratory strong floor rails. It is assumed
that the strong floor was able to provide a rigid support to those members.
The schematic arrangement of the loading systems is shown in Fig.4.6 and
four independent displacement controlled actuators were employed, two for the main
span and two for the cantilever beams. These actuators were operated by computer
controlled pumps in order to ensure that the load application is gradual and the
increment properly controlled. The loads were applied to the specimen through a
loading frame as shown in Fig.4.7. Actuators that fixed to the laboratory strong floor
were attached to the loading frame and the load was applied as a pulling force from
the bottom of the beam as shown in Fig.4.8.
Design calculations were carried out in accordance with the BS5950 Part 1 and
Part 3.1. A plastic collapse mechanism was expected in the main beam, while the
columns were designed to remain elastic. This experimental set up is to test the inner
span to its ultimate capacity. Thus, care had been taken to prevent failure at the beam-
63
to-column connection at the cantilever beam by providing a stronger connection.
Fig.4.9 shows a stronger and larger haunch connection for the reason as mentioned.
The beam was designed to act compositely in the hogging and sagging region.
Additional reinforcement was provided over the haunches to react compositely with
universal beam at hogging region. Concrete slab and universal beam were also
assumed to act compositely in the sagging region. The span to overall depth ratio was
22, which will usually satisfy the strength and serviceability design limits. (Lawson
1989).
Same concrete grade as per the joint specimen was chosen to construct the
120mm thick concrete slab. The floor width of 1.4 m was chosen as the effective
width, Be, for the mid-span in accordance to BS5950: Part 3.1: 1990:
Clause 4.6:
(a) Internal Span, sagging moment region
Be = Lz/4 Where Lz = 0.7Lspan = 0.7 x 8 = 5.6 m
= 1.4 m
(b) Internal Span, hogging moment region
Be = Lz/4 Where Lz = 0.25(Lspan 1 + Lspan 2) = 0.25(8+8)= 4 m
= 1.0 m
Although the effective width of the hogging moment region is 1.0 m, the
concrete slab in the specimen is provided with 1.4 m throughout. Shear connectors
used in these beam specimens are the same as those referred in the joint test in
Chapter 3, ie, 19 mm diameter and 100 mm nominal height, 95 mm after weld height.
Full shear connectors design approaches as per BS5950 Part.3.1 were provided in all
beam test specimens.
64
4.5 Instrumentation
Displacement strain gauge type transducers (SGTD) were used to measure the
displacement of the beam specimen. Two numbers of long travel displacement
transducers (Refer Fig.4.10, T2 & T4) were used to measure displacement at the
loading point and another transducer (T3) was placed at the middle of the beam to
capture the maximum deflection at the beam. Two more displacement transducers (T1
& T5) were positioned at the column to monitor the movement of the columns.
Ten strain gauges (PL-60-11) with gauge resistance at 120 0.3 were
positioned on the concrete slab at the loading point where the formation of plastic
hinge is expected. (Refer Fig.4.10, strain gauge no.: 1,2,3.to 10).
Seventy electrical resistance strain gauges (YFLA-5) from the same
manufacturer were positioned around the specimen to measure strain in steel (strain
gauge no.: 11,12,13to 80). The strain gauges have a gauge resistance at 120 0.3.
Strain gauges were used to monitor yielding and to determine the failure modes. They
were placed at high stress points such as the top and bottom beam flanges near the
column, haunch toe and reinforcement bar as shown in Fig.4.10.
4.6 Testing procedure
After the specimen was positioned in place, loading and instrumentation
devices were connected to the data processing unit. All the readings, transducers and
strain gauges were initialised. All readings were checked for continuity and proper
recording by the data processing units by applying ten percent of the estimated failure
load to the specimen. The load was then released and reapplied in order to remove any
slack that may exist at the support before the actual testing. This process of repeated
loading is expected to ensure proper functioning of the load application and other
measurement devices. The entire load application was performed in three stages. In
65
the first stage, load was applied until the first crack was observed in concrete and, in
the second and third stages load was increased up to 60% and 90% of the estimated
load, respectively. In each stage, the load was released after achieving the intended
load and then reapplied. This process of loading helps to obtain the stiffness of
composite beam and to compare the unloading stiffness at different loading stages. In
the final stage, loading was continued until the failure of the beam specimen.
Load was applied in equal increment to each of the beams until failure. The
ultimate load and the failure mode for each of the beam specimens were thus noted at
the end of the test on that particular specimen. The same procedure was repeated for
all the three beam specimens.
4.7 Beam Test Results and Discussion
4.7.1 Beam Specimen B1
This test was carried out 8 days after casting the floor slab and the concrete
strength on the day of testing was 33 N/mm2. The Specimen B1 was made with
haunch connection H4 referred in Chapter 3 (Table 3.1). The concrete slab was
reinforced with 1.34 %, 8 numbers of T16 deformed bars. The haunch length was
433mm and the depth was 250mm.
During the test, the columns monitored with displacement transducers T1 &
T5, (Fig.4.10) were adjusted back the to the original position by pulling the loading
frames at the cantilever beams. This process is to ensure the verticality of the column
and to minimize the moment transfer from the beam to the column. This operation is
necessary to prevent the failure at the column panel zone due to the unbalance
moment.
66
Views after failure and the corresponding Load-deflection curves for the beam
are show in Fig.4.11 and Fig.4.12 respectively. The registered maximum deflection of
the displacement transducer at the mid-span was 273 mm and the registered maximum
load was 540 kN (270 kN x 2). The deflection shown by displacement transducer T1
and T5 were negligible which means the verticality of the column was always
maintained to prevent unequal moment in the column panel zone.
The failure was observed at the haunch toes and at the loading points and
plastic hinges were formed in these areas. Fig.4.13 shows that there is inelastic
buckling in beam flange near the haunch. The tension reinforcements were found to
have yielded as shown by the strain gauge readings. Meanwhile, the strain gauges on
composite beam sections at the loading points also showed yield values. This is
further confirmed by the crushing of concrete slab as shown in Fig.4.14.
The position of contraflexure point can be measured from the cracking pattern
as shown in Fig.4.15 because concrete will crack under hogging moment, which will
cause tension in the concrete slab. The measured value is about 1.8 m from the
column, which is in good agreement with the value proposed earlier. The figure shows
that no cracking occurred between the contact point of column flange face and
concrete (connection area). Cracking of concrete occurred only at the haunch toe area.
Figs.4.16, 4.17, 4.18 and 4.19 show strain values at various cross sections at
different load stages. The 0 datum at the Depth of Section is referring to the
bottom of the concrete slab. Figs. 4.16 and 4.17 represent the beam cross section at
the left and right haunch toe respectively. As shown in the figure, the neutral axis of
the section is about 70 mm below the concrete slab. Neutral axis will shift up further
if more tension reinforcement is employed. This phenomenon is demonstrated in the
next beam specimen B2 where 2.62% of reinforcement was used in the specimen. The
67
neutral axis shifted up toward tension reinforcement compared to specimen B1 with
1.31% of reinforcement.
During 75% of the ultimate load, almost all strain gauge readings (including
tension reinforcement) show yield value except strain gauge no. 41 & 42 in the left
haunch toe and 62 & 63 in the right haunch toe. But when the load reached 100% of
the ultimate load, strain gauges 42 and 63 showed yield values. Strain gauges 41 and
62 did not reach yielding stage because both strain gauges were positioned near the
neutral axis and a full plastic stress-strain has not fully developed.
On the other hand, beam cross sections near the loading points were subjected
to sagging moment, the universal beam experiencing tension force and the concrete
slab resisting the compression force. Fig.4.18 and 4.19 show the registered strain in
the section at various depths of cross section. When reaching ultimate load, strain
gauge readings showed that the concrete component reached it compression yield
capacity. All other strain gauge readings at this cross section showed yield value as
shown in the figure. Once again, strain gauges near the neutral axis such as strain
gauges 50, 56 & 57 at the loading point did not show any yield.
Studies have also been carried out in this test to determined the effective width
of composite section in the hogging and sagging zone. At hogging haunch toe section,
effective width was checked by placing strain gauges 21-24 & 26-29 on the tension
reinforcement bar, which arranged on both sides of column and parallel to the main
beam. Experimental results suggested that the tension reinforcement bars on both
sides of the column are effective. Figs. 4.20 & 4.21 show the strain distribution of the
tension reinforcement bar at various load stages. At the sagging zone, effective width
of the concrete slab is investigated by strain gauges no.1-10, positioned across the
concrete slab. Test results show that at 75 % of the ultimate load, the whole section is
68
effective in resisting compression force. But during the ultimate loading condition,
shear lag seems to have occurred as shown in Figs. 4.22 & 4.23. It can be concluded
that the effective width of a composite beam at the mid-span will become smaller
under higher moment forces due to the shear lag effect.
This test was carried out 7 days after casting the floor slab and the concrete
strength on the day of testing was 31.7 N/mm2. The Beam Specimen B2 was designed
with connection H4, as shown in Table 3.1. The concrete slab was reinforced with
2.62 % of reinforcement bar, equivalent to 10 numbers of T20 deformed bars. The
haunch length was 433 mm or 5 % of the span and the haunch depth was 250mm.
The maximum deflection registered at the mid-span was 186 mm and the
maximum load was 604 kN (302 kN x 2) for both actuators at the center span. The
Load-deflection curve is shown in Fig.4.24. The deflection shown by the
displacement transducers T1 and T5 were negligible and the verticality of the column
was thus confirmed. In comparison to Specimen B1, Specimen B2 had a larger load
carrying capacity, but the maximum deflection was lower than B1. The ratio of the
load capacity of B2 to B1 is 604/540 = 1.12 but the deflection ratio is only 186/273 =
0.68. This is due to the fact that specimen B2 is stronger at the haunch toe with
additional tension reinforcement bar compared to Specimen B1. The beam stiffness
increased due to additional reinforcement bar, therefore the deflection was lower
compared to Specimen B1. Nevertheless, plastic failure mechanism still can be
observed and an optimum design still achievable with higher percentage of
reinforcement bar. The plastic moment resistance of this composite beam is 319 kNm
and 310 kNm for hogging and sagging section, respectively. While on the contrary,
specimen B1 has only 255 kNm and 310 kNm of plastic moment resistance for
69
hogging and sagging section, respectively. The difference in the hogging capacity at
the haunch toe is because there is additional tension reinforcement bar at the haunch
toe in specimen B2.
Failure of specimen B2 also occurred at the haunch toe and at the one-third
span loading point in which plastic hinge was formed. Fig.4.25 shows that there is
inelastic buckling in beam flange near the haunch toe. The tension reinforcements
were found to have yielded as shown by strain gauge readings and the concrete slab at
the loading points was also crushed as shown in Fig.4.26.
Figs.4.27, 4.28, 4.29 and 4.30 show strain gauge readings in a beam cross
section at different load stages. Figs. 4.27 and 4.28 show the strain values at the left
and right haunch toe sections, respectively. The neutral axis of the section was found
lie right below the concrete slab as shown in the figures. As discussed earlier, neutral
axis will be shifted up if more tension reinforcement was in used. The reinforcement
in B2 was 2.62%, thus, the neutral axis moved up, more towards the tension
reinforcement compared to B1.
At 75% of the ultimate load, almost all strain gauges shows yield value except
strain gauges 16-20, 41 & 42 in left haunch toe and 21-25 & 62 at the right haunch
toe. But when the load reached the ultimate load, all strain gauges showed yield value
except gauges 41 and 62, located near the neutral axis.
Figs.4.29 and 4.30 show the registered strain beam cross section near the
actuator loading points. At ultimate load, strain gauge values at concrete flooring near
the actuator load points already exceeded the concrete yield value. Plastic hinge was
believed to have formed at those sections because all other strain gauges at the same
sections already showed yield value. The strain gauges near the neutral axis showed
no yielding at the section.
70
The effective width in the hogging zone was evaluated by placing strain
gauges 21-25 & 26-30 on the tension reinforcement at the haunch toe area. Test
results showed that all the tension reinforcement bars on both side of the column are
effective. Figs. 4.31 & 4.32 showed strain values at the tension reinforcement bar at
various stages of loading. At the sagging zone, effective width of the concrete slab
was investigated by strain gauges 1-10, positioned across the concrete slab. Shear lag
occurred in Specimen B1 was also observed in B2 and the resulting stress distribution
is shown in Figs. 4.33 & 4.34. Thus, it can be concluded that the effective width of a
composite beam at the sagging section becomes less effective due to shear lag at
higher moment force.
This test was carried out 11 days after casting the floor slab. The concrete
strength on the day of testing was 38.7 N/mm2. The specimen was made with
connection H8 as shown in Table 3.1. The concrete slab was reinforced with 1.34 %,
8 numbers of T16 deformed bars. The haunch length was 968 mm and the depth
250mm.
Views after failure and the corresponding Load-deflection curves for the beam
are shown in Fig.4.35 and Fig.4.36 respectively. The registered maximum deflection
of the displacement transducer at the mid-span was 197 mm and the registered
maximum load at the center span actuators was 674 kN (337 kN x 2).
The failure could be observed at the haunch heel and also at the loading points.
Figs.4.37 & 4.38 show that there is inelastic buckling in the beam flange near the
haunch heel. The tension reinforcements were also found to have yielded as shown by
strain gauge readings. At the loading points at one-third span, concrete slab was also
crushed as shown in Fig.4.39 and the strain gauges at these sections also showed a
71
yield values. The failure mode of this specimen was different from B1 & B2, in which
plastic hinges occurred at haunch toe (hogging zone) prior to the failure at one-third
span loading point (sagging zone). Beam Specimen B3 failed at the sagging zone
before the failure at hogging zone. In the earlier case, moment was redistributed from
hogging zone to sagging zone. However, in the latter case, redistribution of moment
occurred from sagging to hogging region. The disadvantage of moment redistribution
from sagging to hogging moment is that the sagging region where concrete is in
compression could not distribute much moment. This is because the ductility of
concrete is weaker compared to reinforcement bar. Besides, plastic mechanism failure
could not occur in specimen B3 because plastic hinge could not develop at the haunch
connection.
In specimen B3, although failure occurred at the haunch heel and at the
loading points, yielding was also detected at the haunch toe. This can be seen from
Figs.4.40 & 4.41 where the strain gauge readings are plotted against the cross section
at the haunch toe. Nevertheless, no further failure occurred because at that stage of
loading condition, inelastic buckling took place at the beam flange at the haunch
heels.
Figs. 4.42 and 4.43 show strain values at the left and right haunch heel cross
sections for different load stages. It can be seen from the curves that all the strain
gauges had shown yield values. The neutral axis of the section is about 200 mm below
the concrete slab as shown in the figures.
Figs.4.44 and 4.45 show the registered strain in various depths of section near
the center actuator loading points. At ultimate load, strain gauges in concrete slab
showed a yield value. Formation of plastic hinge occurred.
72
. Figs. 4.46 & 4.47 show the strain value at the tension reinforcement bar for
various load stages. Similar to specimen B1 and B2, test results show that all the
tension reinforcement bar on both sides of the column are effective. Investigation
carried out at the sagging zone also showed that effective width of the concrete slab
for a composite section at the sagging section may become less effective when the
applied moment become larger. This could be observed in Figs. 4.48 & 4.49.
Fig. 4.50 shows a comparison of load-deflection curve of all the three
specimens B1, B2 & B3. The rigidity of the three beams is almost the same. Specimen
B3 has a higher load carrying capacity compared to the other beams. However, B1 is
the most ductile compared to B2 & B3. The maximum deflection recorded is 273mm,
compared to 186mm for B2 and 196 mm for B3.
4.8 Concluding remarks
In composite haunch beams, the haunch toe could be planned as the weakest
section to allow a plastic hinge to form. The plastic failure mechanism occurs when
plastic hinges form at the haunch toe and at the mid-span. An optimum design of
composite beam can be achieved when failure happens at the haunch toes and moment
will be redistributed to mid-span.
Experimental results for B1 and B2 show that the haunch connection is able to
redistribute moments to the mid-span to form a plastic hinge failure mechanism. The
results show that plastic hinge failure mechanism can be achieved in designing
continuous composite haunch beam.
Studies on the effective width in continuous composite beams show that the
effective width recommended by BS 5950:Part 3: Section 3.1 and Eurocode 4 is
satisfactory at the sagging region. However, care should be taken because the shear
lag effect will come in during the ultimate load as shown in the test results. On the
73
other hand, the effective width for the concrete floor recommended by the code in the
hogging region is not studied because the concrete strength in tension is neglected.
Nevertheless, test results show that the tension reinforcement bars in the hogging
region, which distribute evenly in 1.4-metre width are all effective. This opens up the
possibility of increasing the effective width beyond the codes recommended value;
however more experimental work is required to come up with a design
recommendation.
74
Fig.4.1 Haunch Beam Test Specimen
75
Fig.4.2 Wooden Formwork of Beam Test Specimen
Fig.4.3 Beam Specimen Ready for Concrete Casting
76
Fig.4.4 Beam Specimen Ready for Testing
77
Fig.4.5 Isometric View of Haunch Beam Test Specimen
78
Fig.4.6 Schematic Loading of Haunch Beam Test Specimen
79
Fig.4.7 Loading Frame in Haunch Beam Test Specimen
80
Fig.4.8 Loading Frame Connected to Hydraulic Actuator
Fig.4.9 Larger Haunch Connection at Cantilever Beam Side
81
Concrete slab Shear stud Reinforcement bars
11/12/13/14/15 16/17/18/19/20 1/2/3/4/5 6/7/8/9/10 21/22/23/24/25 26/27/28/29/30
31 41 50 56 62 71
32 42 51 57 63 72
33 43 52 58 64 73
34 46 44 53 59 65 67 74
35 47 45 54 60 66 68 75
36 48 55 61 69 76
37 49 70 77
38 78
39 T2 T3 T4 79
40 80
T1 T5
Fig.4.10 Instrumentation of Beam Specimen
82
Fig.4.11 View after Failure of Specimen B1
Load vs Deflection (B1)

800
700
600 T5
Load (P2 + P3), kN
T1
500
400 T4
300 T2
T3
200
100
0
0 50 100 150 200 250 300
Deflection, mm
Fig.4.12 Load-Displacement Curve of Specimen B1
83
Fig.4.13 Inelastic Buckling in Compression Flange of the Beam B1
Fig.4.14 Crushing of Concrete Slab at Loading Point in Beam B1
84
Fig.4.15 Concrete Slab Cracking Pattern of the Beam B1 at Haunch Toe
85
150
100
Rebar yielding
50
Section depth, (Top of flange = 0 mm)
Zone(Tension)
0
-15000 -10000 -5000 0 5000
-50
UB yielding
UB yielding
Zone(Tension)
Zone(Compression) -100
-150 0% UL
25% UL
-200 50% UL
-250 75% UL
100% UL
-300
Strain,
Fig. 4.16 Strain Reading of Cross Section for Beam B1 at

Left Haunch Toe at Different Load Stage
150
100
Rebar yielding
50 Zone(Tension)
0
-15000 -10000 -5000 0 5000
-50
UB yielding UB yielding
Zone(Compression) -100 Zone(Tension)
0% UL
-150
25% UL
-200 50% UL
75% UL
-250
100% UL
-300
Strain,
Right Haunch Toe at Different Load Stage
86
150
0% UL
100 25% UL
50% UL

50
Concrete yielding 75% UL
Zone(Compression) 100% UL
0
-5000 0 5000 10000 15000
-50
UB yielding
-100
Zone(Tension)
-150
-200
-250
-300
Strain,

Left Loading Point at Different Load Stage
150
100
Section Depth (Top Flange = 0 mm)
50
Concrete yielding
Zone(Compression)
0
-5000 0 5000 10000 15000 20000
-50
0% UL
-100 UB yielding 25% UL
Zone(Tension) 50% UL
-150
75% UL
-200 100% UL
-250
-300
Strain,

Right Loading Point at Different Load Stage
87
Fig. 4.20
800
700
Rebar yielding
Distance from beam center, mm

600 Zone(Tension)
500
400
0% UL
300
25% UL
200 50% UL
75% UL
100
100% UL
0
0 1000 2000 3000 4000 5000 6000 7000 8000
Strain,
Fig. 4.20 Strain Reading of Tension Reinforcement for Beam B1 at Left

800
700
Rebar yielding
Zone(Tension)
600
500
400
300 0% UL
25% UL
200
50% UL
75% UL
100
100% UL
0
0 1000 2000 3000 4000 5000 6000 7000 8000
Strain,
Fig. 4.21 Strain Reading of Tension Reinforcement for Beam B1 at Right

88
Concrete yielding
800
Zone(Compression)
600

400
200
0
-2500 -2000 -1500 -1000 -500 0 500
0% ULL
-200
25% UL
-400
50% UL
75% UL -600
100% UL
-800
Strain,
Fig. 4.22 Strain Reading of Concrete Slab for Beam B1 at Left

Loading Point at Different Load Stage
800
Concrete yielding
Zone(Compression)
600
400
200
0
-2500 -2000 -1500 -1000 -500 0 500
0% ULL -200
25% UL
50% UL -400
75% UL
-600
100% UL
-800
Strain,
Fig. 4.23 Strain Reading of Concrete Slab for Beam B1 at Right

89
700
T1
600
T5
500
Load(P2+P3), kN
T4
400
T2
300
T3
200
100
0
0 20 40 60 80 100 120 140 160 180 200
Deflection, mm
Fig.4.24 Load-Displacement Curve of Specimen B2
Fig.4.25 Inelastic Buckling in Compression Flange of the Beam B2
90
91
150
100
Rebar yielding
Section depth, (Top of flange=0 mm) Fig Zone(Tension)

.
0
-15000 -10000 -5000 0 5000
-50
UB yielding
UB yielding
Zone(Tension)
-150 0% UL
25% UL
-200
50% UL
75% UL
-250
100% UL
-300
Strain,
Fig. 4.27 Strain Reading of Cross Section for Beam B2 at Left

150
100
Rebar yielding
Section depth, (Top of flange=0 mm)
50
Zone(Tension)
0
-15000 -10000 5000 0 5000
-50
UB yielding
UB yielding
-100 Zone(Tension)
Zone(Compression)
0% UL
-150 25% UL
50% UL
-200
75% UL
-250 100% UL
-300
Strain,
Fig. 4.28 Strain Reading of Cross Section for Beam B2 at Right

92
150
Left Loading Point (Specimen B2)
100
150
mm)
10050
Concrete yielding
of flange=0
mm)))))
Zone(Compression)
mm)))
Concrete yielding 50 0
-4000Zone(Compression)
-2000 0 2000 4000 6000 8000 10000
flange=0
ofofflange=0
0
-4000 -2000
-50
0 2000 4000 6000 8000 10000
(Top(Top
-50
-100 UB yielding
0% UL
(Top
depth,
-100 UB Zone(Tension)
yielding 0% UL 25% UL
depth,
-150 Zone(Tension)
depth,
25% UL50% UL
-150
Section
50% UL75% UL
Section
-200
Section
-200 75% UL
100% UL
-250 100% UL
-250
-300
-300
Strain,
Strain
X 10E-6

150
100
Section depth, (Top of flange=0 mm)
Concrete yielding 50
Zone(Compression)
0
-4000 -2000 0 2000 4000 6000 8000 10000
-50
0% UL
-100 UB yielding
25% UL
Zone(Tension)
50% UL
-150
75% UL
-200 100% UL
-250
-300
Strain,
93
800
700
Rebar yielding
Zone(Tension)
600
500
400
300 0% UL
25% UL
200
50% UL
100 75% UL
100% UL
0
0 500 1000 1500 2000 2500 3000 3500 4000
Strain,

800
700
600 Rebar yielding

Zone(Tension)
500
400
300 0% UL
25% UL
200
50% UL
100 75% UL
100% UL
0
0 500 1000 1500 2000 2500 3000 3500 4000
Strain,

94
800
Concrete yielding
0% ULL Zone(Compression)
600
25% UL
Distance from beam center , mm 50% UL 400

75% UL
100% UL 200
0
-5000 -4000 -3000 -2000 -1000 0 1000
-200
-400
-600
-800
Strain,
Fig. 4.34
800
Concrete yielding
0% ULL
Zone(Compression)
25% UL
600
50% UL
400
75% UL
100% UL 200
0
-5000 -4000 -3000 -2000 -1000 0 1000
-200
-400
-600
-800
Strain,

95
Fig.4.35
800
T1 T5 Fig.4.36
700
600
Load (P2+P3), kN
500
T4
400
300 T3
200 T2
100
0
0 50 100 150 200 250
Deflection, mm
Fig. 4.35 Load-Displacement Curve of Specimen B3
Fig.4.36 View after Failure of Specimen B3
96
Fig.4.37 Inelastic Buckling of Left Haunch Flange of the Beam B3
Fig.4.38 Inelastic Buckling of Right Haunch Flange of the Beam B3
97
98
150
100
Rebar yielding
Section depth, (Top of flange = 0 mm) 50 Zone(Tension)
0
-15000 -10000 -5000 0 5000
-50
UB yielding
-100 Zone(Tension)
Zone(Compression)
-150 0% UL
25% UL
-200 50% UL
75% UL
-250
100% UL
-300
Strain,

150
100
Rebar yielding
50 Zone(Tension)
0
-15000 -10000 -5000 0 5000
-50
UB yielding
UB yielding
Zone(Tension)
0% UL
-150
25% UL
-200 50% UL
75% UL
-250
100% UL
-300
Strain,
99
150
Rebar yielding

50 Zone(Tension)
-30000 -20000 -10000 -50 0 10000 20000
-150 UB yielding
UB yielding
Zone(Tension)
Zone(Compression)
-250
0% UL
-350 25% UL
50% UL
-450 75% UL
100% UL
-550
Strain,

Haunch Heel at Different Load Stage
150
Rebar yielding
50
Zone(Tension)
-30000 -20000 -10000 -50 0 10000 20000
UB yielding
UB yielding -150
Zone(Tension)
Zone(Compression)
-250
0% UL
25% UL
-350
50% UL
75% UL
-450
100% UL
-550
Strain,

Haunch Heel at Different Load Stage
100
150
100
Concrete yielding
50
Section depth (top Flange = 0 mm)
Zone(Compression)
0
-5000 0 5000 10000 15000
-50
0% UL
-100 UB yielding 25% UL
Zone(Tension) 50% UL
-150
75% UL
-200 100% UL
-250
-300
Strain,

150
100
Concrete yielding
50
Zone(Compression)
0
-5000 0 5000 10000 15000
-50
UB yielding 0% UL
-100 25% UL
Zone(Tension)
50% UL
-150
75% UL
-200 100% UL
-250
-300
Strain,

101
800
700
600
Rebar yielding
500 Zone(Tension)
400
0% UL
300
25% UL
50% UL
200
75% UL
100 100% UL
0
0 500 1000 1500 2000 2500 3000
Strain,

800
700
600
Rebar yielding
500
Zone(Tension)
400
0% UL
300
25% UL
200 50% UL
75% UL
100
100% UL
0
0 500 1000 1500 2000 2500 3000
Strain,

102
800
Concrete yielding
Zone(Compression) 600
400
200
0
-3500 -3000 -2500 -2000 -1500 -1000 -500 0 500
-200
0% ULL
25% UL
-400
50% UL
75% UL -600
100% UL
-800
Strain,

800
Fig.
Concrete yielding 4.49
Zone(Compression)
600
Fig. 4.49 400

200
0
-4000 -3000 -2000 -1000 0 1000
0% ULL
-200
25% UL
-400
50% UL
75% UL
-600
100% UL
-800
Strain,
103
800
Specimen B3
700
Specimen B2
600 Specimen B1
Load (P2+P3), kN
500
400
300
200
100
0
0 50 100 150 200 250
Deflection, mm
Fig. 4.50 Comparison of Load-Displacement Curve for Specimens B1,B2 and B3
104
CHAPTER 5
ANALYTICAL MODEL
5.1 General
All the three beam specimens demonstrated ductile behaviour when they were
subjected to concentrated loads, and plastic failure mechanisms have been formed.
The occurrence of plastic failure mechanism is dependent on the formation of plastic
hinges at connection and at load points. The plastic analysis of composite haunch
beam will be illustrated in this chapter.
Besides, derivation of the section properties of composite haunch beam is
presented in this chapter so that further analysis can be carried out based on the
section properties. In addition, lateral distorsional buckling and the available
rotational capacity of reinforced composite beam will be investigated. This will bring
further understanding of the behaviour of the reinforced composite haunch beam.
The test results will be compared with those obtained by the finite element
software USFOS in order to establish the accuracy of the numerical model.
Subsequently, the numerical model would be used to carry out parametric studies.
5.2 Comparison of Plastic Hinge Analysis and Test Result
Fig.5.1(a) represents a haunch beam which is statically indeterminate. When
plastic hinges developed at the loading points, the beam becomes statically unstable
and a collapse mechanism will be developed in the beam. When plastic hinges are
formed in the haunch toe, the beam becomes statically determinate. The beam will
then become a simply supported beam with a constant moment Mph at haunch toe to
carry the load P as shown in Fig.5.1(b). This will be followed by the formation of
plastic hinge at haunch toe. Figs.5.1 (c) & (d) show the corresponding moment
105
diagram due to P and Mph, respectively. By applying the principle of superposition,
the final moment diagram is as shown in Fig.5.1(e).
Thus,
Mps = PL/3 - Mph

(Eq. 5.1)
P = 3 (Mps + Mph) / L
(Eq. 5.2)
Where,
P = Point Load
Mps = Plastic moment capacity of sagging region
Mph = Plastic moment capacity of hogging region
L = Length between hinges at both ends
Table 5.1 Comparison of Test Results with Plastic Hinge Theory
Beam Mps Mph L PE, Experimental PT, Theory PE/PT

Specimen (kNm) (kNm) (m) Failure Load Failure Load Ratio
(kN) (kN)
B1 340 256 7.174 270 249 1.08
B2 339 322 7.174 302 276 1.09
B3 485 343 8.000 337 310 1.09
Table 5.1 shows the comparison of test results with plastic hinge theory.
Plastic hinge theory is able to predict the ultimate load within 10% of the test results.
The first plastic hinge in Specimen B1 & B2 occurred at the haunch toe (hogging)
followed by hinge at the loading point (sagging). Thus, moment is redistributed from
haunch toe to mid-span. Failure of B3 occurred at the connection and at the mid-span
almost at the same time. Referring to the strain gauge readings, four plastic hinges
formed together and no moment redistribution is required in beam test B3. This
phenomena shows that optimum design is already achieved. Hence, it could be
concluded that the failure mode of composite beam might be controlled either at
106
haunch toe or haunch heel. The design procedure for the failure at haunch toe as
shown in B1 and B2 is more straightforward compared to the failure at haunch heel as
shown in B3. However, beam specimen B3 shows that optimum design is possible by
adjusting the haunch length and the amount of reinforcement.
For beam specimen B1 and B2, the designer has to make sure the available
rotation capacity at the haunch toe is sufficient to enable moment redistribution in the
plastic hinge analysis. The failure mode will be controlled at haunch toe as long as the
connection (haunch heel) capacity is larger than the haunch toe capacity under the
moment gradient. This is one of the alternatives in designing composite haunch beam
structure. However the disadvantage of this failure mode is that the connection
(haunch heel) needs to be strengthened and most of the time the strengthening will be
required at the column, which might involve extra welding of stiffener plates. Thus,
the fabrication cost of the steelwork based on this failure mode may be more costly.
Failure at the connection (haunch heel) in Beam B3 demonstrates an ideal
situation where no moment redistribution is required if all the plastic hinges form at
the same time. The design of this failure mode required only the section capacity
under hogging and sagging moment where plastic hinges formed. Rotational capacity
is not required because no moment redistribution occurs. However, if the beam cross
section capacity at hogging and sagging region are very close to the required strength,
this plastic failure mechanism will still occur and it requires only a small amount of
rotation capacity to allow for moment redistribution. If the initial failure is detected at
the connection (haunch heel), the available rotation capacity of the connection need to
be studied carefully to ensure that moment could be redistributed from connection to
mid-span area. The available rotation capacity in composite haunch beam is
107
dependent on the component on the connection such as amount of reinforcement bar,
column and beam properties.
5.3 Rotation Capacity
5.3.1 General
According to Kemp, (1991), the following limit states condition need to be
satisfied in order to provide adequate ductility and sufficient moment redistribution to
a structure.
(a / mr) > r
(Eq. 5.3)
Referring to Fig. 5.2 (Kemp, 1991), a is the inelastic available rotation prior to
the moment below the design moment resistance Mp. It may be provided either by the
end connection or by the member over the length Li between the section of maximum
moment and adjacent point of inflection.
r is the inelastic rotation required to achieve an identified level of redistribution
of moments in a structures.
mr is a partial material factor to allow for many uncertainties.
It is common practice to express the equation to non-dimensional form by
dividing the rotation by a hypothetical elastic rotation r determined for the same
moment resistance Mp over the same length of member Li. Thus,
ra / mr > rr
(Eq. 5.4)
where:
ra = available rotational capacity = a / e

rr = required rotation capacity = r / e
and for a linear moment gradient used in most test as in Fig.1:
e = 0.5 Mp Li / EI
(Eq. 5.5)
108
in which EI is flexural rigidity of moment/unit curvature.
A large partial material factor mr should be introduced because it is very
difficult to predict inelastic rotations from a moment-rotation curve which is nearly
horizontal. Kemp, (1991) proposes a value of 2 for relatively ductile modes of failure
with local and lateral buckling and a value of 3 for sudden fractures.
5.3.2 Calculation of available rotation capacity of composite section
An empirical formula proposed by Kemp, (1991) is used to identify the
available inelastic rotation capacity. The theoretical model accounts for local and
lateral buckling of the steel section.
Fig. 5.3 represents the region near to an internal support of a continuous beam
(Kemp and Dekker, 1991). The local buckling is assumed to develop when the length
of plastic region of the flange (Lp in Fig.5.3) extends sufficiently far to accommodate
the full wavelength of the buckle. Following are the empirical formula that Kemp
proposed to estimate the plastic length and available rotational capacity for steel
beam.
Lp = 0.067Li(60/e)1.2
(Eq. 5.6)
ra = 3(60/e)1.5/2
(Eq. 5.7)
e = Kf/Kw(Li/izc)
Kf = (b/tf)/20 for class 1 & 2 flanges
Kw2= dw/33tw for class 2 web, 33 < (dw/ tw) 40
Kw1= [460-(Li/izc)]Kw2/400 for class 1 web, (dw/ tw) < 33
where:
Li = Length of load point to haunch toe as shown in Fig.5.3.
e = Effective slenderness ratio = Kf/Kw(Li/izc)
Kf & Kw = Empirical factors to allow for the actual flange and web slenderness
= Proportion of the depth of section in compression between the
centers of the two flanges.
izc = Radius of gyration of the flange and portion of web in compression
109
To apply the above empirical equation to a composite section, Kemp and
Dekker, 1991, has pointed out 5 factors which affect the available rotational capacity
namely:-
a) Ratio of moment resistance at support to midspan region
b) Elastic properties used in calculation of rotation capacity
c) Axial force balancing reinforcement force
d) Cracking of concrete adjacent to supports
e) Restraint to lateral buckling by slab
The following are the comparison of result between the joint tests and the model
proposed by Kemp and Dekker, (1991), after taking into account of the 5 factors
mentioned above: -
Table 5.2 Comparison of rotational capacity at Haunch Toe
Specimen 1 Specimen 2 Specimen 3 Specimen 4 Specimen 5
(a) Connection H1 H2 H3 H4 H5 H6 H7 H8 H9 H10
(b) Li 1550 1367 1550 1367 1550 1367 1093 832 1093 832
(c) e NA NA 11.7 10.3 11.7 10.3 8.2 6.3 8.2 6.3
(d) ra(exp) (Test Results) NA NA 9.7 16.7 6.7 14.3 17.0 NA 13.3 NA
(e) ra(kemp) (Kemp Model) NA NA 9.1 12.0 9.3 12.0 14.0 NA 12.0 NA
Ratio of (d)/(e) NA NA 1.07 1.39 0.72 1.19 1.21 NA 1.11 NA
The test results presented in Table 5.2 are based on the combination of Kemps
proposal and a Bi-Linear Curve as shown in Figs. 5.4 to 5.9. The available rotational
capacity of the composite section at haunch toe is obtained by dividing the Platic
Rotation, p with the Elastic Rotation, e. Table 5.2 shows that the ratio of test
results and Kemps Model, ranging from 0.72 to 1.39.
110
Bi-Linear Curve is introduced because it is not possible to obtain the ideal curve
as per Kemps model where the Mp will be achieved after the linear behaviour. The
actual behaviour of such composite section is that after the concrete crack or first
yield, the section begins to lose its stiffness before the Mp. Loss of stiffness may
increase the rotational ability of the section. The experimental Moment-Rotation
curves show that there is an inelastic region where the rotation could not be defined.
When first yield occur at the extreme fiber, the section under consideration only
begins to behave inelastically but a full plastic section has not been fully developed.
Therefore, it shall not be considered as part of the plastic rotational capacity.
However, the introduction of the bi-linear curve is to simplify and standardize the
definition of Rotational Capacity.
5.4 Beam Analysis
5.4.1 Composite Haunch Beam Properties
Fig. 5.10 shows a typical cross section of a composite haunch beam. The
tapered section is formed by taking the required haunch length of universal beam
section, cut it diagonally and welded to a universal beam. The following sections
present the derivation of section properties of composite haunch beam.
5.4.1.1. Section Properties
5.4.1.1.1. Elastic Section Properties
Universal Section
A UB = 2 BT + ( D 2T )t
2 BT 3 D T ( D 2T ) 3 t
I UB
x = + 2 BT ( ) 2 +
12 2 2 12
3 3
2TB ( D 2T )t
I UB
y = +
12 12
2 1
J UB = BT 3 + ( D 2T )t 3
3 3
111
Where:
A T = BT + ( D t T ) t
BT 2
(Dt T )
+ ( D t T )t T +
2 2
c Tx =
AT
2 2
BT 3 T (Dt T )3 t Dt T
I xT = + BT c Tx + + ( D t T )t T + c Tx
12 2 12 2
3
TB (D t T ) t 3
I Ty = +
12 12
1 1
J T = BT 3 + (D t T ) t 3
3 3
AUB = Area of Universal Beam
IxUB = Second Moment of Universal Beam at x-x axis
IyUB = Second Moment of Universal Beam at y-y axis
JUB = Torsional Constant of Universal Beam
Tee Section Haunch

Where:
AT = Area of Tee Section Haunch
cxT = Distance from Bottom of flange to PNA of Tee Section Haunch
IxT = Second Moment of Tee Section Haunch at x-x axis
IyT = Second Moment of Tee Section Haunch at y-y axis
JT = Torsional Constant of Tee Section Haunch
Reinforcement Bar
2
AR = N
4
4
I R
x =N
64
where,
AR = Area of Reinforcement
IxR = Second Moment of Reinforcement at x-x axis
112
Total Composite Haunch Section
ACH = AUB + AT + A R
D
AUB Dt + + AT cTx + A R (Dt + D + Dr )
2
c CH =
ACH
2
4
I CH
x =I UB
x +A
UB D
Dt + c CH (
+ I xT + AT c CH cTx )
2
+N + A R (Dt + D + Dr )
2
2 64
CH
I
Z xCH top flange = x
(D + Dt + Dr c CH )
CH
I
Z xCH bottom flange = x
CH
c
I yCH = I UB
y + Iy + Iy
T R
I yCH
ry =
ACH
J = J UB + J T + J R
where:
ACH = Area of Composite Haunch Section

cCH = Distance from Bottom of flange to PNA of Composite Haunch Section
IxCH = Second Moment of composite Haunch Section at x-x axis
IyCH = Second Moment of Composite Haunch Section at y-y axis
J = Torsional Constant of Composite Haunch Section
ZxCH = Elastic Modulus of Top/Bottom Flange at x-x axis
ry = radius of gyration at y-y axis
5.4.1.1.2. Plastic Section Properties
Unhaunch Universal Section
A UB = 2 BT + ( D 2T ) t
D T ( D 2T ) 2 t
S UB
x = 2 BT ( ) +
2 2 4
where: SxUB = Plastic Modulus of Universal Beam
113
Composite Haunch Universal Section, Neutral Axis in Middle Flange
2
A CH = 2 BT + ( D 2T ) t + BT + ( D t T ) t + N
4
A CH
Half Total Area =
2
( D 2T ) t BT ( Dt T )t 2
= BT + + + +N
2 2 2 8
3 BT t 2
= + (D 2T + D t T ) + N
2 2 8
Half Total Area = BT + ( Dt T )t + (Cx Dt ) B
3BT t 2
+ (D 2T + Dt T ) + N = BT + ( Dt T )t + (Cx Dt ) B
2 2 8
BT t 2
+ (D + Dt 3T 2 Dt + 2T ) + N = (Cx Dt ) B
2 2 8
t 2
T 2
( D Dt T ) + N
+ 8 = (C D )
x t
2 B
t 2
T
( D Dt T ) + N
Cx = Dt + + 2 8
2 B
Which is only valid if:

Dt Cx Dt + T
For this case, the plastic section modulus of the composite haunch section is given by:
T ( Dt T ) B
Sx
CH
= BT ( C x ) + ( D t T )t C x D t + + (C x D t )2
2 2 2
B D 2T T
+ ( D t + T C x ) 2 + ( D 2T ) t D t + T C x + + BT D + Dt C x
2 2 2
2
+ N ( D + Dt C x + Dr )
4
(Eq. 5.8)
Haunch Universal Section, Neutral Axis in Web of Universal Section
Refer to Figure 5.11, half total area in terms of Cx is:
ACH
Half Total Area = = 2 BT + ( D t T )t + (C x D t T )t
2
Solution for C x :
( ACH 4 BT )
Cx = + 2T
2t
which is only valid if :
D t + T C x ( Dt + D T )
114
For this case, the plastic section modulus is given by : -
T D T T
S CH
x = BT C x + (D t T )t C x T t + BT C x D t
2 2 2
t 2 t
+ (C x D t T ) + (D + D t C x T )
2
2 2
T N 2
+ BT D + D t C x + (D r + D + D t C x )
2 4
(Eq. 5.9)
5.4.2 Composite Haunch Beam analysis
There are a few alternatives to model composite haunch beam, which include
(Hogan & Syam, 1997):
a) Modeling the haunch as a single beam element type over the full length of
the haunch with the elements section properties based on that of the
average depth of the haunch.
b) Dividing the haunch length into a number of segments with each segment
having uniform section properties which are representative of that
segment, either maximum or average or minimum.
c) Using a single beam element of equivalent stiffness, calculated using
(PCA, 1958)
d) Obtaining a stiffness matrix and fixed end reaction formulation for a
tapered member and use a computer program which allows a stiffness
matrix for a member to be input.
Based on the alternative (b), a study has been conducted (Hogan & Syam,
1997) and it is found that there is no benefit in using more than two segments in
modeling of haunch section. SCI, (1995) also provides some reasonable guidance and
suggests that haunches may be satisfactorily modeled by using two haunch elements.
115
The two haunch elements are modeled with average section properties for lengths
corresponding to 1/3 and 2/3 of the haunch.
Based on the present information on haunch section modeling, alternative (b)
is selected for the analysis by using computer software USFOS.
5.5 Finite Element Modelling
5.5.1 Nonlinear Analysis Software: - USFOS
USFOS is a nonlinear analysis software in which load can be applied in steps.
After each load step, the new position of nodal coordinates due to displacement will
be updated and the structure stiffness is assembled at each load step. The element
stiffnesses are then calculated from the updated geometry. At every load step each
element is checked to see whether the forces exceed the plastic capacity of the cross
section. If such an event occurs, the load step is scaled to make the forces comply
exactly with the yield condition. A plastic hinge is inserted when the element forces
have reached the yield surface.
5.5.2 Modeling of Composite Haunch Beam
As described in Section 4.5.2 earlier, there is no advantage in using more than
two segments in modeling of haunch section in the analysis. Fig.5.12 shows a
structure model of the test specimen in USFOS. Beam is modeled as a non-prismatic
member with the section properties as shown in the figure. The haunch sections are
modelled using 2 beam elements with length Lh. Determination of contra-flexure
point is based on 22% of span length which is about 1.8m from the column.
116
5.5.3 Results
Fig.5.13 shows the test results for the beam B1. The predicted ultimate load is
568kN (284kN x 2) for 2 load points, which is about 4.2 % more compared to the test
value. The stiffness of Beam B1 predicted by the finite element method is higher than
the test value. Fig.5.14 shows that the predicted ultimate load of the Beam B2 is
600kN (300 kN x 2), which is about 4.5% more compared to the test value.
Comparison of stiffness reflected that the stiffness of Beam B2 is higher than the test
value. Finite element result in Fig.5.15 for Beam B3 shows that the predicted ultimate
load is 674kN (337kN x 2), about 10 % difference compared to the test value.
Comparison of stiffness shows that the stiffness of Beam B3 is also higher than the
test value. It is believed that the higher stiffness prediction by USFOS may be because
of the fact that model did not account for the concrete cracking thus resulting in a
higher stiffness. However, based on the comparison between finite element and
experimental results, the finite element model is sufficiently accurate.
5.6 Lateral Torsional Instability
5.6.1 General
It is essential to check the buckling capacity when designing a structural steel
member. Sometimes buckling failure in a structural member is allowed to happen but
generally it is not economical for this type of failure mode. For steel member such as
I-Beam, two types of buckling mode usually occurred. Firstly, local buckling due to
compression force (the criteria for the local buckling is already well established by the
recent codes, which utilized the "section classification" to limit the local buckling).
Secondly, lateral torsional buckling which involves the cross section rotation and
117
displacement. This type of buckling usually can be avoided by the provision of lateral
and torsional restraints.
However, in steel-concrete composite beam design, the shear connection
between the steel beam and the concrete slab provides the lateral and torsional
restraint to the top steel flange. Therefore, the top flange is prevented from buckling
and only the web and the bottom flange are subjected to torsional buckling. This kind
of buckling requires more energy to induce than the lateral torsional buckling in steel
beam because it involves distorsion of the cross section. It is usually referred as
"Lateral Distorsional Buckling".
Lawson & Rackham, (1989) has proposed a method that is readily usable with
the British Code BS5950 for composite beam. However, their method did not
consider the contribution of top reinforcement. Therefore, the author has proposed a
calculation method for lateral distorsional buckling in composite haunch connection,
accounting for the effects of top reinforcement.
5.6.2 Lateral Distorsional Buckling Design Method
Rackham (1992) concluded in his work that the destabilising effect due to the
addition of reinforcement is likely to be minimal and that the buckling strength will,
in fact, be enhanced by the increased bending strength that the reinforcement
provides. He has shown in his work the equation to predict lateral distorsional
buckling for steel haunch connection. By modifying the energy method used by
Rackham (1992), derivation to obtain the capacity of composite haunch connection
accounting for lateral distorsional buckling is given below:
118
E1 + E 2 + E 3 + E 4 = E 5 + E 6
(Eq. 5.10)
E1= Strain energy absorbed in lateral bending of the bottom flange
E2= Torsional energy absorbed in twisting the bottom flange
E3= Bending energy absorbed in displacing the web
E4= Torsional energy absorbed in twisting the web
E5= Work done by the compressive force in the bottom flange
E6= Work done by the forces in the web
Therefore
E1 + E 2 + E 3 + E 4 = E 5 + E 6
EI y 4V 2
1 9 GJ f V
2 2
1 3 EI wV 2 L 1 3GJ w 2V 2
+ + +
4 L3 2 8 h s2 L 2 2 h s3 2 5 h s2 L
1 M cr BT f 2V 2
1 107 . 5 2V 2 M cr t w h s
= +
2 2Z eL 2 1120 LZ er
Multiplying throughout by 4L/2V2 and collecting terms,
the elastic buckling equation is obtained thus:
2 2
BT f 43 t w h s 2G 9J f 3 3 EI w L
M cr + = EI y + 2 + Jw +
Ze 224 Z er L hs 8 5 h s3
(Eq. 5.11)
Based on the British Code 5950,
M p 2 E
M cr =
2LT p y
where
LT = Equivalent Slenderness
p y = Yield Stress
M p = Plastic Moment
(Eq. 5.12)
119
And
( M cr M b )( M p M b ) = LT M cr M b
where
LT = Perry Coefficient
M b = Buckling Re sis tan ce Moment
M p = Sx py
(Eq. 5.13)
M p 2 E
LT =
M cr p y
S x 2 E
=
M cr
and
2 2
2G 9 J f 3J w 3EI L
EI y + + + 3w
L 2
hs 8 5 hs
M cr =
BT f 43 t w hs
+
Ze 224 Z er
Therefore
BT f 43 t w hs
S x 2 E +
Ze 224 Z er
LT = 2 2
2G 9 J f 3 J w 3EI L
EI y + + + 3w
L 2
hs 8 5 hs
(Eq. 5.14)
The above derivation can directly be used in the code BS 5950:2000 and hence
the Mb or the lateral torsional buckling moment could be obtained.
120
Fig. 5.1(a) Haunch Beam with 2 Point Loads
Fig. 5.1(b) Collapse Mechanism in Haunch Beam
Fig. 5.1(c) Moment Diagram of 2 Point Loads
Fig. 5.1(d) Moment Diagram of Haunch Toe Loads
Fig. 5.1(e) Moment Diagram of 2 Point Loads and Haunch Toe Loads
121
a
e
m
1.4
M
1.2
Mp
Design Moment resistance = Mp
1.0
0.8 Rotation Capacity = ra = a/e
Li
0.6
Slope = Elastic stiffness C
0.4
Rotation
0.2
Max Moment, M
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
End Rotation Ratio = /e
Fig. 5.2 Moment-Rotation Curve according to Kemp, (1991)
122
Lateral Restraint
Lb Lb
Li Li
Lp Lp
Mp
Mp (At Lateral Restraint)
Mm = mMp
= MpLp/(LI-Lp)
Li Li
Lp Lp
ye Lp/(LI-Lp)
Curvature = Strain/'
df s=sy
Lp Le
Lip=CpLi Lie=CeLi
Actual lateral deflection
Lateral defl.of equiv.strut
Fig. 5.3 Plastic Region near the Internal Support of Continuous Beam
123
Rebar Yielded at 238kNm
300 4.5 13. 15
5
250 kNm
250
224 kNm
200 Section Yielded at 242kNm
150
Moment, kNm
First Yield detected at Compression

Flange at 138kNm
100
First Crack observed at 89kNm
Haunch Toe Moment rotation Curve
50
Bi-Linear Moment Rotation Curve
0
0 5 10 15 20 25 30 35 40 45 50
Rotation, mrad
Fig. 5.4 Bi-Linear Moment-Rotation Curve for Connection H3
300
3.5 18. 20.
250 247 kNm
224
200
Moment, kNm
150
100 Flange at 117kNm Haunch Toe Moment-Rotation
Curve
50 Bi-linear Moment Rotation
First Crack observed at 69kNm Curve
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65
Rotation, mrad
124
400
Rebar Yielded at 302kNm 344 kNm
350
8.5 6.5 25.0
300 282 kNm
250 Section Yielded at

294kNm
200
Moment, kNm
150
First Yield detected at
Compression Flange at 128kNm Haunch Toe Moment Rotation
100
Curve
First Crack observed at Bi-Linear Moment Rotation
50
98kNm Curve
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
Plastic Rotation, mrad

400
350 330 kNm
300 282
kNm
250
Moment, kNm
Section Yielded at 298kNm

200
150
100 Compression Flange at 130kNm Haunch Toe Moment Rotation Curve
50 First Crack observed at 72kNm Bi-Linear Moment Rotation Curve
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
Rotation, mrad
125
350
289 kNm
300
250 224 kNm

Moment, kNm
200 Section Yielded at 279kNm
150
100
Compression Flange at
164kNm Haunch Toe Moment Rotation Curve
50 First Crack observed at 79kNm
0 Bi-Linear Moment Rotation Curve

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
Rotation, mrad
400 Rebar Yielded at

341 kNm
350
300 282 kNm
250
Moment, kNm
Section Yielded at 311kNm

200
150
Flange at 159kNm
100
Haunch Toe Moment Rotation Curve
50 First Crack observed at 87kNm Bi-Linear Moment Rotation Curve
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
Rotation, mrad
126
Dr
t D
C t Dt
Fig. 5.10 Cross Section of Haunch Beam with PNA at Beam Flange
Dr
t D
t Dt
Figure 5.11 Cross Section of Haunch Beam with PNA at Beam Web
127
Node18 Node 20
H H
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
A B C D E F G F E D C B A
Lh Lh
H Lh = 216 mm for B1 H
19 21
Lh = 216 mm for B2
Lh = 354 mm for B3
Fig. 5.12 Modeling of Haunch Beam using non-linear Finite Element Analysis
4 3
Section I, mm S, cm Remarks
B1 B2 B3 B1 B2 B3
A 8.56E+07 8.56E+07 8.56E+07 1219 1581 1219 Composite Beam (Hog)
B 4.56E+08 4.56E+08 4.56E+08 1261 1261 1261 Composite Haunch (Hog)
C 9.56E+08 9.56E+08 9.56E+08 1308 1308 1308 Composite Haunch (Hog)
D 6.80E+08 6.80E+08 6.80E+08 1526 1526 1526 Composite Haunch (Hog)
E 3.92E+08 3.92E+08 3.92E+08 1178 1178 1178 Composite Haunch (Hog)
F 8.56E+07 8.56E+07 8.56E+07 1219 1581 1219 Composite Beam (Hog)
G 1.93E+09 1.93E+09 1.93E+09 1619 1614 1619 Composite Beam (Sag)
H 6.13E+07 6.13E+07 6.13E+07 656 656 656 Composite Beam (Sag)
Fig. 5.12 Modeling of Haunch Beam using Non-Linear Finite Element Analysis
128
USFOS Vs Experiment BEAM B1
700
600 568 kN 544 kN

500
Load, kN
400
300
200
USFOS
100 Experiment
0
0 50 100 150 200 250
Displacement, mm
Fig. 5.13 Comparison of USFOS and Experimental Load-Displacement

Curve for Beam B1
129
700
600 kN
600
500 573 kN
Load, kN
400
300
200
USFOS
100
Experiment
0
0 50 100 150 200 250
Displacement, mm

Curve for Beam B2
800
700 674 kN
600
628 kN
500
Load, kN
400
300
200 USFOS
100 Experiment
0
0 50 100 150 200 250
Displacement, mm

Curve for Beam B3
130
CHAPTER 6
DESIGN RECOMMENDATIONS
AND DESIGN EXAMPLE
6.1 Introduction
Based on the information available in the literature and experimental and
analytical studies presented in the previous chapters, design recommendation for
composite haunch beams will be made herein. It is important to provide structural
designers with a design concept with some specific guidelines so that the designers
could take precaution or necessary steps to minimize design errors. However,
engineers shall always make necessary judgements while encountering the situations
not covered in these design recommendations.
6.2 Design recommendations
1) The ratio of length of the beam to depth should be in the range of 25 to 30 for
most efficient design. The experimental program presented earlier shows that
with a 7.8m clear span and 250mm beam depth, L/D ratio equal to about 30,
an optimum design is achieved with appropriate amount of reinforcement.
However, with this large L/D ratio, serviceability in deflection may be a
control factor. Introduction of haunch may help to control the deflection
because it reduces the "clear span" by 5% to 10% when haunch length fixed
between 1.5 to 3 times the depth of the steel beam. Beams of these proportions
when designed on strength would usually satisfy serviceability. It is
recommended that the depth of the haunch is taken as the depth of the steel
section, which allow the haunch be cut from the parent beam. Larger haunch
depth require additional type of beam size and increases the fabrication cost.
131
2) It is recommended that for efficient design of continuous haunch beams, the
length of the end span should be approximately equal to the length of the
adjacent span. This is to ensure the moment diagram for the internal span be
similar to the current study and allow for similar moment redistribution
pattern. Care shall be taken at the end-span when anchorage of reinforcement
at external column is not sufficient to act compositely with the haunch
connection.
3) Since long span beam design is likely to be controlled by serviceability, grade
S275 steel may be more economical than grade S355 for a span to depth ratio
of 25 to 30. The second moment of inertia that controls the deflection will be
the same regardless of the steel grade used.
4) The bending resistance of the composite bolted connection should be designed
to be more than the composite haunch toe section. Therefore, failure can be
assumed to occur at the haunch toe.
5) End-plates are to be welded to the ends of the beam. These plates are
approximately 20% to 30% thicker than the beam flange. This will usually
avoid the failure at the end plate and provide a rigid connection. The bolt
diameter is approximately equal to the end plate thickness. (M24 or M30 bolts
are expected to be the preferred size) (Lawson and Rackham, 1989).
6) The moment resistance of the composite connection should be determined by
the method described in Chapter 3.
132
7) Determine the number of bolts and the amount of slab reinforcement as
follows :
- Determine the number of bolts needed to resist the factored shear
force. Assume the bottom bolt groups (between bottom haunch flange
to bottom beam flange) to resist shear force and upper row bolts to take
tension forces arising from the applied moment.
- Choose the number and diameter of reinforcing bars within the
effective width at the supports (beam span/8). The reinforcement could
be from 1% to 2.5%. Larger percentage of reinforcement will actually
bring up the plastic neutral axis, which is not effective and not
economical.
8) Determine the column size in accordance with the following guideline to avoid
use of column web stiffeners which may be costly:
- Column flanges should be Class 1 or 2 (plastic or compact).
- There is a maximum percentage of reinforcement (limit to 2.5% of
effective width at hogging area) that may be used before requiring
column stiffeners. Consider use of a heavier column section if more
tension reinforcement is used, or alternatively, provide column
stiffeners.
6.3 Elastic global analysis
(1) Determine the moment resistance of the composite section in accordance with
BS 5950: Part 3 or Eurocode 4, for the assumed beam size selected in Section
6.1.
133
(2) Check the construction condition for the design of steel beams. This is
normally not critical but attention to be given to where lateral torsional
buckling of primary beams may occur during construction.
(3) Determine the moment resistance of the composite haunch connection and
composite section (at haunch toe).
(4) Calculate the second moment of area of the composite section, using
uncracked section properties.
(5) Carry out an elastic global analysis, using the appropriate composite beam
section stiffness (at sagging and hogging moment regions) and column section
stiffness. Determine the connection moments for the pattern loads used in the
analysis.
(6) Redistribution of negative (hogging) moments at the haunch toe section (not at
the connections) is normally not required or very minimum as per Table 1 in
Chapter 2. Increase the positive (sagging) moment in the spans to maintain
equilibrium. Check the capacity of the composite section (sagging) as in Step
(1), and the composite section (hogging) at haunch toe as in Step (3). Proceed
further if the applied moments are less than the moment resistance. Modify the
size of the steel section, or the amount of reinforcement, as necessary if the
applied moment is larger than the resistance moment.
134
(7) Check the design of the columns subject to axial force in combination with the
moments and shear forces transferred at the face of the column. The moment
resisted by the columns above and below the connection is divided in
proportion to their stiffnesses (or lengths). Take the moment variation factor
in BS 5950: Part 1 as 0.5 and treat the effective length as equal to the column
length for a non-sway column.
(8) Determine the number of shear connectors necessary to achieve the required
force transfer between the beam and the slab for hogging and sagging moment
region. Check the minimum spacing of the shear connectors. If inadequate,
increase the size of the steel beam or reduce the percentage of reinforcement.
(9) Calculate the deflection of the composite beam using the second moment of
area, as in Step (4). Check the imposed load deflections against serviceability
limits.
(10) Calculate the deflection of the steel beam after construction. For long span
beams, it may be necessary to precamber or prop the beams to reduce the total
deflection. If the serviceability performance is not adequate, increase the size
or depth of the steel beam.
6.4 Plastic hinge analysis
This procedure is only appropriate for Class I beams and Class I or 2 column
sections.
135
(1) Repeat Steps 1, 2 and 3 of Section 6.3.
(2) Combine the moment resistance of the haunch toe section and the composite
beam section in positive(sagging) bending, in an equivalent plastic hinge
analysis. For an internal span subject to equally spaced loads, add the two
moment resistance directly to determine the failure load, For an external span,
ignore the moment resistance of the composite connection, but include the
moment resistance of the steel connection, if appropriate. This is because at the
end span, the anchorage that required for reinforcement is not sufficient to
contribute tension resistance in the composite beam section.
(3) Compare the free moment at the ultimate limit state to the combined moment
determined as in Step 2. If the combined moment resistance exceeds the applied
free moment, proceed. If not, increase the size of the steel beam.
(4) Check the shear resistance of the bolt group for the applied shear, taking
account of the final bending moment variation, and the reduced effectiveness of
the bolts in tension.
(5) Repeat Step (7) of Section 6.3 for the design of the columns. For internal
columns, ignore pattern loading and consider an applied moment equal to half
that of an equivalent column loaded from one side only by the same connection.
Divide this moment between the columns above and below. Include the axial
load due to full loading on all spans.
136
(6) Repeat Step (8) of Section 6.3 for the numbers of shear connectors.
(7) Repeat Steps (9) and (10) of Section 6.3 for serviceability performance.
137
6.5 Design Procedure Flow Chart
Design of Non-Sway
Composite Haunch
Structures
Elastic Global Plastic Hinge

Analysis Analysis
Identify Design
Load
Try Beam &

Column as per
Section 6.2
Analyse Frame with Analyse Frame with

homogeneous Plastic Hinge
section Analysis
Allow negative
moment redistribution
as per Table 2.2
Determine Sagging (Mid-
Span) and Hogging (Haunch
Toe) Moment Resistance of
Determine Sagging (Mid- Composite Section as per BS
Span) and Hogging (Haunch 5950 Part3 Section 3.1.
Toe) Moment Resistance of Total the sagging and hogging
Composite Section as per BS moment to compare with
5950 Part3 Section 3.1 and design moment that required
compared with design moment for plastic failure mechanism
Continue Next Page Continue Next Page
138
Continue From Previous Page Continue From Previous Page
No Mcapacity > Mapplied No

Vcapacity > Vapplied
Yes
No Lateral Distorsional No
Buckling Check
(Eq 5.11)
Yes
Haunch Connection
Design as per Eq. 3.3
or Eq. 3.4
Serviceability Check
End
139
6.6 Design Example
A design example is presented in Appendix A to illustrate the design method
developed herein. The design example is based on a commercial office building,
which is braced against lateral sway. The floor grid consists of a 8 m main beam span
with 12 m span secondary beams at 2.27m centre to centre.
The main beam test B2 was specifically designed to model this frame
arrangement and the section sizes used in the experimental set-up is exactly same as
in the design example. Therefore, a direct comparison could be made between the
results of this calculation and the test results. The maximum failure load in the test
was 604 kN (2 x 302 kN), compared to 500 kN (2 x 250 kN, refer Design Loading in
Appendix A) factored design load. The test frame sustained 20% more load than the
design calculations proposed, which represents a generous safety margin.
140
CHAPTER 7
CONCLUSIONS AND
PROPOSALS FOR FUTURE WORK
7.1 General
Behaviour of composite haunch beams has been investigated through
experimental program and analytical study. The research work carried out has
provided a perception into the behaviour of composite haunch beam (with tension
reinforcement), which has not been explored sufficiently in the past research.
Concluding remarks with regard to experimental and analytical study will be
presented in the following section.
7.2 Behaviour of the Composite Haunch Connection
It is concluded from the tests that composite haunch connections could be
designed as full strength connections. The rotations of the composite haunch
connections are very small and can be neglected. A conventional stress block in
accordance with Eurocode 4 is reasonable to predict accurately the capacity of the
connections. The additional reinforcement added in the slab contributes significantly
to the connection without compromising the ductility. This is evident from the
moment-rotation curves obtained from the connection tests. However, only sections
with "Plastic" classification are recommended for this behaviour.
Besides, the moment-rotation curves of the composite connections also show
that the stiffness of the connections drops after the cracking of the concrete. The
stiffness of the initial '

uncracked'hogging region concrete section is higher than the
values of the '

cracked'hogging region.
141
Experimental observations show that failure at the haunch toe is governed by
compression failure of the flange and web plates or tensile failure of reinforcement
bars depending on the neutral axis, which is affected by the amount of slab
reinforcement. The higher the slab reinforcement ratio, the compression flange and
web plates will fail before the yielding of the tension reinforcement.
7.3 Behaviour of the Composite Haunch Beam
For composite haunch beam, plastic hinge could be designed at haunch toe
which is the weakest section when subjected to hogging moment. Optimum design of
composite beam can be achieved when plastic hinge occurred at haunch toes followed
by a plastic mechanism at the mid-span. Experimental results show that haunch
connection is able to redistribute moment to the mid-span by losing stiffeness due to
cracking of concrete slab and yielding of either steel reinforcement or cross section.
Studies on the effective width in continuous composite beam show that the
effective width 0.25L recommended by BS 5950:Part 3: Section 3.1 and Eurocode 4 is
satisfactory at the sagging region. On the other hand, since concrete strength in
tension is negligible, its contribution is ignored and only the reinforcement within the
effective width is considered. Studies have been carried out for the effectiveness of
the tension reinforcements at the hogging region and test results show that the tension
reinforcements recommended by BS 5950:Part 3: Section 3.1 and Eurocode 4 are
conservative. It is proposed an additional factor of 1.4 to be used while calculating the
effective width using the codes.
7.4 Section Properties and Frame Analysis
The section properties of composite haunch section have been derived for the
frame analysis. It is not necessary to model more than 2 average sections for a haunch
142
connection in order to obtain reasonable results. The finite element software USFOS
produces reasonably accurate prediction for the composite haunch beam capacity. It is
recommended that this software be used to carry out parametric study.
The lateral distorsional buckling design method has been developed and the
method could be used in conjunction with the code. Design for the treatment of lateral
distorsional buckling has been shown by the tests to be conservative when applied to
haunch composite beams. From the analysis of the results of the Sub-Assembly Test
in Chapter 4, it is suggested that, for haunch composite beams, the value of the critical
buckling length Lcr, should be multiplied by 0.6. It is also suggested that the value of
the slenderness parameter, vt, calculated from equations 4 and 5, should be multiplied
by 0.75 to produce realistic design values. However, on the basis of the tests, it is
clear that when a full depth stiffener is provided both sides of the web at the haunch
toe, and when the minimum shear connection is maintained over the whole hogging
region, the haunch is sufficiently stiff to assume that the haunch toe position is
restrained. The possibility of lateral distorsional buckling need only then be checked
beyond the haunch toe, towards the span.
The moment-rotation characteristics of the haunch region were specifically
examined in the tests. The degree of rotation obtained from the Sub-Assembly Tests
was compared with theoretical requirements specified by Kemp, (1991). It was found
that only those beams with full depth stiffeners fitted both sides of the web at the
haunch toes passed the criterion. These specimens continued to sustain their design
plastic moments of resistance up to a value of 45 rads.
It is interesting to note that the maximum load was achieved in the main test
when the haunch toe rotations were of the order of 35 milli radians. It is therefore
143
concluded that it is possible to design and detail the composite haunch beam so that
sufficient rotational capacity will be achieved.
7.5 Future Work
The behaviour of composite haunch beams and connections has been explained
and demonstrated with experimental testing and an analytical model. Certainly,
composite construction with haunch beam structures could provide designers an
option to maximise the usage of steel and concrete so that optimum and economical
design could be achieved. Nevertheless, the design of composite structures involves
many considerations and processes that could still be further developed and verified.
Therefore, besides the work reported in this thesis, the following areas for future study
are recommended.
1. To study the behaviour of haunch composite connections with tension
reinforcement at the end span which is always subjected to unbalanced
moment. In a composite haunch connection where the tension
reinforcement is ignored, and when the column stiffness is a lot larger than
the steel beam, additional reinforcement may contribute extra tension
resistance to achieve a higher capacity at the connection as well as the
haunch toe area. However, at the end span, the anchorage of the
reinforcement could be an issue to be explored. A detailed study shall be
conducted in this area.
2. To study the behaviour of haunch composite connection in sway frame
where the connection is not only subjected to hogging but also sagging
moment. The bolts designed at connection to resist shear force is now
exposed to tension force when the connection is subjected to reverse
144
moment. It is necessary to understand the haunch connection behaviour and
its application in sway frame.
The finite element USFOS shows reasonable predictions of the behaviour of
composite haunch beam as a structural frame. It could be further developed to zoom
into the component design which is able predict the failure mode of the haunch beam
construction system. Further development will be worthwhile because when
reasonable accurate computer simulation is possible, full scale testing will be the least
option for researchers to study the behaviour of any structural system.
145
APPENDIX A
DESIGN EXAMPLE
A) Design Data
i) Structure Data
A 4 storey building size 24m x 72m as shown in Figure 3.1 braced against side-sway
is considered in this design example.
Main Beam Span : 8 m (Composite Haunch Beam Rigid Connection)

Secondary Beam Span : 12 m (Composite Beam Simply Supported)
Storey Height : 4.2 m
Column Base : Fixed
Slab Thickness : 120 mm
ii) Material Data:
Shear Stud : Dia 19mm x 95mm as welded

Concrete : fcu = 30N/mm2 , 2400 kN/m3
Slab thickness : 120mm
Steel Grade : S275
iii) Design Loading
Concrete Slab : 2.88 kN/m2

Construction Load : 0.50 kN/m2
Building Services : 0.70 kN/m2
Imposed Load : 5.00 kN/m2
Design Loading : 1.4 (2.88+0.7) + 1.6(5) = 13.01 kN/m2
Secondary beam self weight, say 0.50 kN/m. Design Load = 1.4 x 0.50 = 0.70 kN/m
146
Therefore, loading from secondary beam transferred to main beam, P
= ((2.67m x 13.01kN/m2) + 0.70 kN/m) x 12m /2
= (41.74 kNm) x 12m/2
= 250 kN
B) Plastic Design of Main Beam
To form a plastic mechanism at the haunch toes and loading points:
Lh A h
L/3-L L/3 L/3-Lh Lh
B C D E
A F
L/3-Lh
Mph
B C
Mps
Where,
Mph = Plastic Moment Capacity at Haunch Toe (Hogging)
Mps = Plastic Moment Capacity at Load point (Sagging)
Moment about haunch toe,
Mph + Mps = P (L/3-Lh)
Therefore, to design the haunch beam,
Mph + Mps > 1
P (L/3-Lh)
147
Beam size - Span to Depth ratio 25 to 30
Using 30, D = 8000/30
= 266 mm
Try UB 254 x 146 x 37 kg/m, plastic section for both sagging and hogging condition
and hence is suitable for plastic design.
A Data:
1 Slab Thickness , Ds = 120 mm
2 Concrete strength, fcu = 30 N/mm2
3 Concrete Cover, Dc = 20 mm
4 Effective width,sagging,Bes = 1400 mm
5 Effective width,hogging,Beh 1000 mm
6 Number of rebars = 10 nos
7 Diameter of rebar = 20 mm
8 Area of rebar, Abar = 3142 mm2
9 Yielding strength of rebar = 460 N/mm2
10 Lever arm of rebar to beam flange, Dr = 90 mm
12 Universal beam dimension
Depth, D = 255.9 mm
Width, B = 146.4 mm
Flange thickness, T = 10.9 mm
Web thickness, t = 6.4 mm
Web depth, d = 218.9 mm
Area = 4740 mm2
Yielding strength, fy = 275 mm2
13 Constant, (Pfix/Py)^(1/2) = 1
B Calculation
1 Resistance of concrete flange, Rc = 5040 kN
2 Resistance of steel flange, Rf = 439 kN
3 Resistance of steel beam, Rs = 1304 kN
4 Resistance of clear web depth, Rv = 385 kN
5 Resistance of overall web depth,Rw = 426 kN
6 Resistance of slender web, Ro = 428 kN
148
7 Resistance of reinforcement, Rr = 1445 kN
8 Resistance of slender steel beam,Rn = 1346 kN
9 d/t = 34.2
10 76*e/(1+Rr/Rv) = 16.0
11 76*e*/(1+Rc/Rv) = 5.4
Case 2 : Plastic neutral axis in concrete slab (Sagging)

Rc > Rw and Rs < Rc
Mc = Rs*(D/2+Ds-Rs/Rc*Ds/2) = 303 kNm BS 5950: Part 3
Case 6 : Plastic neutral axis in concrete slab (Hogging)

Rr > Rw and Rr > Rs
Mc = Rs*(D/2 +Dr) = 284 kNm BS 5950: Part 3
Try Haunch Length 5% of span, hence haunch length Lh = 0.05 x 8000 = 400 mm
Check ultimate condition,
Mph + Mps > 1
P (L/3-Lh)
303 + 284 > 1

250(8/3-0.4)
587 = 1.04 > 1

567
:- Ultimate condition satisfied
Checking of lateral distorsional buckling of beam at haunch toe

Refer to equation (Eq 5.14) in the thesis LT = 17.7 and refer to Table 16 of
BS5950:2000,
pb = 275 N/mm2
Therefore, no reduction to Rs to obtain the Mb = Mc = 284kNm Okay !
149
C) Main Beam Elastic composite Properties
I) Calculation of Moment Inertia, Ig (Sagging) Uncracked
UB254 x 146 x 37kg/m

Universal Beam Properties
D = 255.9 mm
B = 146.4 mm
t = 6.4 mm
T = 10.9 mm
A = 4740.0 mm2
d = 218.9 mm
Sxx = 484363.0 mm4
Ms = 0.0 kNm
1 Steel Beam Inertia Moment Ixx = 55600000.0 mm4
2 Effective Width of Concrete (Sagging) Be = 1400.0 mm
3 Overall slab depth Ds = 120.0 mm
4 Depth of the deck profile Dp = 0.0 mm
5 Modular Ratio e e = 13.2
6 Cross section area of steel beam A = 4750.0 mm2
7 Depth of steel beam D = 256.0 mm
Ig = 1.93E+08 mm4
D) Main Beam Serviceability Deflection Check
Design Load from Secondary Beam (Imposed Live Load Only)

= 5.0 kN/m2 x 8/3 x 12m/2
= 80kN
150
80kN 80kN
Lh L/3-Lh L/3 L/3-Lh Lh
Max Deflection = 0.00772 PL3/EI

= 0.00772 *80,000*80003/(205,000*1.93E08)
= 8.0mm
Hence, Span/Def = 8000/8

= 1000 > 360, Live Load Serviceability Deflection Okay!
E) Haunch Connection Design
Haunch Connection is designed to remain elastic. Therefore, the connection is to

resist 1.1 time the haunch toe moment Mph.
V 250 kN 250 kN V
Lh L/3-Lh L/3 L/3-Lh Lh
322kNm 322kNm
V V
250 kN
1.1 Mps = 1.1 x 322kNm = 354 kNm
Le = 400 mm
Connection Moment = 250 kN x 0.4 + 354 kNm

= 454 kNm
Try M20 Grade 8.8 Bolts with 10T20 Tension Reinforcement
151
Rr = 0.87 x 10 x 3.14 x 202/4 x 460
= 1256 kN
Rb = 2 x 91.9 kN
= 184 kN
Rhf = 146 x 10.9 x 1.2 x 275

= 525 kN
Column Web Bearing (Try column UC 203 x 203 x 60 kg/m)

Pc = (b1 + n2) x tc x py
b1 = Tbeam + 2Tplate
= 10.9 + 2x20
= 50.9 mm
n2 = 2(Tcolumn + rcolumn) x 2.5

= 2(14.2 + 10.2) x 2.5
= 122 mm
Therefore, Pc = (50.9 + 122) x 9.1 x 275

= 436 kN
Column Web Buckling (Try column UC 203 x 203 x 60 kg/m)

Pc = (b1 + n1) x tc x py
n1 = 2 x 209.5/2
= 209
= 2.5 d/tcolumn
= 2.5 x 189.1/9.1
= 51.7
152
from Table 27(c) BS 5950: Part 1
pc = 217 N/mm2
Therefore, Pc = (50.9+209.5) x 9.1 x 217

= 518 kN
Comparing Rhf and Pc (Bearing or Buckling), Column Bearing Controlled

So, Rhf is limited to 436 kN
Rhw = 234 x 6.4 x 1.2 x 275

= 494 kN
Rf = 146 x 10.9 x 1.2 x 275

= 525 kN (Limit to 436 kN Column Bearing Controlled)
Rw = 234 x 6.4 x 1.2 x 275

= 494 kN
Rhf + Rhw + Rf + Rw < Rr + Rb , PNA at beam web
((yc - Dh) x 1.2 x 275)+ Rhf + Rhw + Rf = Rr + Rb
((yc - 256) x 1.2 x 275) + 436 + 494 + 436 = 1256 + 184
yc = 256 mm
Take moment about reinforcement
Mu = 525 (90+256+256-10.9/2) + 494 (90+256+234/2) + 525(90+256-10.9/2)
= 313163 + 228722 + 178762
= 720647 kNmm
= 721 kNm > 454 kNm
Haunch Connection Moment Satisfied !!
Checking of shear bolt, design shear force = 250 kN
Try 4 M20 Bolt Grade 8.8, shear capacity = 4 * 91.9 kN

= 368 kN > 250 kN Okay !!
153
REFERENCES
1. Anderson, D. and Najafi, A.A. (1994) "Performance of Composite Connections:

Major Axis End Plate", Journal of Constructional Steel Research, Vol. 31, 31-57.
2. Anderson, D., Aribert, J.M., Bode, H. and Kronenburger (2000) "Design Rotation
Capacity of Composite Joints", The Structural Engineer, Vol. 78, No.6 March, 25-
29.
3. Aribert, J.M. and Raoul, J. (1992) "Two full-scale tests of class 3 composite
beams", Proceedings of the Engineering Foundation Conference on Composite
Construction ASCE, Potosi, MI, 14-19.
4. ASTM E8-1979 (1979) "Standard Methods of Tension Testing of Material",

American Association State Highway and Transportation Officials Standard,
AASHTO No.:T68.
5. Boswell, L.F. (1992) "The Structural Behaviour of Haunch Composite Beams in

Long Span Building Applications", Final report to SCI in respect of British Steel
Market Development Grant No. MDF P9/90, City University, London.
6. Climenhaga, J.J. and Johnson, R.P. (1972a) "Local Buckling in Continuous

Composite Beams", The Structural Engineer, 50(9), 367-374.
7. Climenhaga, J.J. and Johnson, R.P. (1972b) "Moment-Rotation Curves for Locally
Buckling Beams", Journal of the Structural Division, Proceedings of the ASCE,
Vol. 98, No. ST6, June, 1239-1254.
8. Cosenza, E, Mazzolani, S. and Pecce, M. (1995a) "Non Linear Analysis of

Continuous Composite Beam", 287-296.
9. Cosenza E, Mazzolani, S and Pecce, M. (1995b) "The Influence of the rebar

ductility on the rotational capacity of composite beams", 879-886.
154
10. Couchman, G. (1996) "A New Design Method for Continuous Composite
Beams", Structural Engineering International, 2/96, 96-101.
11. Dekker, N., Kemp, A.R. and Trinchero, P. (1995) Factors Influencing the
Strength of Continuous Composite Beams in Negative Bending Journal of
Constructional Steel Research, Vol. 34, 161-185.
12. Eurocode 3 ENV 199-1-1:1992 (1992) "Design of Steel Structures, Part 1.1
General Rules and Rules for Building", Comite European de Normalisation
(CEN), Brussels.
13. Eurocode 4 DD ENV 1994-1-1:1992 (1994) "Design of Composite Steel and

Concrete Structures, Part 1.1 General Rules and Rules for Building", British
Standard Institution, London.
14. Fabbrocino, G., Manfredi, G. and Cosenza, E. (2001) "Ductility of Composite

Beams under Negative Bending: An Equivalent Index for Reinforcing Steel
Classification", Journal of Constructional Steel Research, Vol. 57, 185-202.
15. Hamada, S. and Longworth, J. (1976) "Ultimate Strength of Continuous

Composite Beams", Journal of The Structural Division, July, 1463-1479.
16. Hamada, S. and Longworth, J. (1974) "Buckling of Composite Beams in Negative

Bending", Journal of The Structural Division, Nov, 2205-2222.
17. Hogan, T.J. and Syam A.A. (1997) "Design of Tapered Haunch Universal Section
Members in Portal Frame Rafters", Steel Construction, Vol. 31, No 3, Sept, 2-27.
18. Hope-Gill, M.S. and Johnson, R.P. (1976) "Tests on Three Three-span continuous
composite beams", Proc. Instn. Civ. Engrgs, Part 2, 61, June, 367-381.
155
19. Iwankiw, N. (1997) "Ultimate Strength Considerations for Seismic Design of the
Reduced Beam Section (Internal Plastic Hinge)", Engineering Journal, First
Quarter, 3-16.
20. Johnson, R.P. and Chen, S. (1991) "Local Buckling and Moment Redistribution in
Class 2 Composite Beams", Structural Engineering International, April, 27-34.
21. Kemp, A.R. (1995) "Interaction of Plastic Local and Lateral Buckling", Journal of
Structural Engineering, Vol. 111, No.10, Oct, 2181-2195.
22. Kemp, A.R. and Dekker, N. (1991) "Available Rotation Capacity in Steel and
Composite Beams", The Structural Engineer, Vol. 69, 88-97.
23. Kemp, A.R. and Nethercot, D.A. (2001) "Required and Available Rotations in
Continuous Composite Beams with Semi-rigid Connections", Journal of
24. Kitipornchai, S. and Trahair, N.S. (1975a) "Buckling of Inelastic I-Beams under
Moment Gradient", Journal of the Structural Division, Proceedings of the ASCE,
Vol. 101, No. ST5, May, 991-1004.
25. Kitipornchai, S. and Trahair, N.S. (1975b) "Inelastic Buckling of Simply

Supported Steel I-Beams", Journal of the Structural Division, Proceedings of the
ASCE, Vol. 101, No. ST7, Jul, 1333-1678.
26. Kitipornchai, S. and Trahair, N.S. (1972) "Elastic Stability of Tapered I-Beams,
Journal of the Structural Division, Proceedings of the ASCE, Vol. 98, No. ST3,
Mar, 713-728.
27. Lawson, R.M. and Rackham, J.W. (1989) "Design of Haunch Composite Beam in
Buildings", Steel Construction Institute, United Kingdom.
28. Lawson, R.M. (1995) "Developments in steel framed commercial building in th

UK, The Structural Engineer, Vol. 73, No. 11, 186-192.
156
29. Lay, M.G. (1965) "Flange Local Buckling in Wide-Flange Shapes", Journal of the
Structural Division, Proceedings of the ASCE, Vol. 91, No. ST6, Dec, 95-116.
30. Lay M.G. and Galambos T.V. (1965) "Inelastic Steel Beams Under Uniform
Moment", Journal of the Structural Division, Proceedings of the ASCE, Vol. 91,
No. ST6, Dec, 67-93.
31. Leon R.T. (1990) "Semi-rigid Composite Construction", Journal of Constructional

Steel Research, Vol. 32, 99-119.
32. Li T.Q, Choo B.S. and Nethercot D.A. (1995) "Determination of Rotation
Capacity Requirements for Steel and Composite Beams", Journal of
33. Li T.Q, Nethercot D.A. and Lawson, R.M. (2000) "Required Rotation of
Composite Connections", Journal of Constructional Steel Research, Vol. 56, 151-
173.
34. Liew J.Y.R., Looi K.L. and Uy B. (2001) "Practical Design Guideline for Semi
Continuous Composite Braced Frames", International Journal of Steel and
Composite Structures, 1(2), 213-230.
35. Liew J.Y.R. (2001) "A Resource Book for Structural Steel Design and
Construction", Singapore: SSSS/BCA Joint Publication.
36. Liew, J.Y.R., Teo, T.H. and Shanmugam, N.E. (2004) "Composite Joints Subject
to Reversal of Loading- Part 1: Experimental Study", Journal of Constructional
37. Liew, J.Y.R., Teo, T.H. and Shanmugam, N.E. (2004) "Composite Joints Subject
to Reversal of Loading- Part 2:Analytical Assessments", Journal of Constructional
157
38. Lukey, A.F and Adams, P.F. (1969) "Rotation Capacity of Beams under Moment
Gradient", Journal of the Structural Division, Proceedings of the ASCE, Vol. 95,
No. ST6, Jun, 1173-1188.
39. Narayanan, R. (1991) "Design Development in Long Span Flooring System in the
UK", Proceedings, International Conference on Steel and Aluminium Structures,
ICSAA91, Singapore, 22-24 May, ed. by Lee, S.L. and Shanmugam, N.E.,
Elsevier Applied Science.
40. Nethercot, D.A. and Trahair, N.S. (1976) "Inelastic Lateral Buckling of
Determinate Beams", Journal of the Structural Division, Proceedings of the
ASCE, Vol. 102, No. ST4, Apr, 701-717.
41. Nethercot, D.A. (1995) "Design of Composite Connections", The Structural

Engineer, vol. 73, no. 13/4 July, 218-219.
42. Nethercot, D.A. and Li, T.Q. (1995) "Design of Semi-Continuous Composite
Frames", Structural Stability and Design, ed. by Kitipornchai, Hancock &
Bradford, 217-282.
43. Nethercot, D.A. (1975) "Inelastic Buckling of Steel Beams under Non Uniform
Moment", The Structural Engineer Vol. 53, No.2, Feb, 73-78.
44. Nethercot, D.A. (1983) "Elastic Lateral Buckling of Beams, Beams and Beam
Columns-Stability and Strength", ed. by Narayanan, R., Applied Science
Publishers, London and New York, 1-33.
45. Owens, G. (2000) "State of the art report: Basic Problems, Design Concepts and
Codification of Steel and Composite Structures", Journal of Constructional Steel
Research, Vol. 55, 7-27.
46. Plumier, Andre. (1997) The Dogbone: Back to the Future, Engineering Journal,
Vol. 34, No. 2, 61-67.
158
47. Price, A.M. and Anderson, D. (1992) "Composite Beams", Constructional Steel
Design: An International Guide, ed. by Dowling, Harding, J.E. and Bjorhord, R.,
Elsevier Applied Science, London, 501-522.
48. Rackham, J.W. (1992) "The Design of Haunch Composite Beam Frames for
Buildings", Thesis, Dept. of Civil Engineering, City University, London.
49. Shanmugam, N.E., Ng, Y.H. and Liew, J.Y.R. (2002) "Behaviour of Composite
Haunch Beam Connection", Engineering Structures, Vol. 24, 1451-1463
50. SCI/BSCA Connection Group (1995) "Joints in Steel Construction Moment

Connections", Steel Construction Institute, United Kingdom.
51. Tehami, M. (1997) "Local Buckling in Class 2 Continuous Composite Beams",

Journal of Construction in Steel Research, Vol. 43, Nos.1-3, 141-159.
52. Trahair, N.S. and Kitipornchai, S. (1972) "Buckling of Inelastic I-Beams under
Moment Gradient", Journal of the Structural Division, Proceedings of the ASCE,
Vol. 99, No. ST11, Nov, 991-1004.
53. Trahair, N.S. (1983) "Inelastic Lateral Buckling of Beams, Beams and Beam
Columns-Stability and Strength", ed. by Narayanan, R., Applied Science
Publishers, London and New York 1983, 35-69.
54. Uy, B. and Liew, J.Y.R., (2003) Composite Steel-Concrete Structures, The Civil
Engineering Handbook, ed. by Chen, W.F. and Liew, J.Y.R., CRC Press, 51-1/62.
159
LIST OF PUBLICATIONS
1. Shanmugam, N.E., Ng, Y.H, Richard Liew, J.Y., Behaviour of Composite

haunch beam connection, Engineering Structures 24(2002), pp 1451-1463.
2. Ng, Y.H., N.E. Shanmugam, Richard Liew, J.Y. and Yu, C.H., Haunch
Connections in Composite Construction, Proceeding of the Fifth Pacific
Structural Steel Conference, 13-16 October 1998, Seoul, South Korea, pp.
717-722.
3. Richard Liew, J.Y., Ng, Y.H. and N.E. Shanmugam, Design of Haunch
Composite Connections for Long-Span Beam Construction, Connections in
Steel Structures IV: Behaviour Strength and Design, edited by Roberto Leon
and W.S. Eastering, Chicago: AISC 2002, pp. 424-433.
4. Ng, Y.H., N.E. Shanmugam and Richard Liew, J.Y., Tests To Failure of
Continuous Composite Haunch Beams, Proceeding of the International
Conference on Structural and Foundation Failures August 2-4, 2004,
Singapore.
5. Ng, Y.H., N.E. Shanmugam and Richard Liew, J.Y., Behaviour of Composite
Haunch Beam, Journal of Constructional Steel Research (Accepted for
Publication)
160

Design of Composite Haunch Beams and Connections For Long Span Applications PDF

Uploaded by

Copyright:

Available Formats

Design of Composite Haunch Beams and Connections For Long Span Applications 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

Design of Composite Haunch Beams and Connections For Long Span Applications PDF

Uploaded by

Copyright:

Available Formats

What is the thesis about?

What is the thesis about?

What are the objectives of the research?

What are the objectives of the research?

DESIGN OF COMPOSITE

HAUNCH BEAMS AND CONNECTIONS

NG YIAW HEONG (BEng.(Hons.), MEng)

DEPARTMENT OF CIVIL ENGINEERING

The author would like to take this opportunity to acknowledge various

encouragement provided by his research supervisors, Professor N.E. Shanmugam and

completed without their continuous support.

Secondly, the author is fortunate to have received moral support and

experimental testing in the Concrete and Structural Laboratory, National University of

Singapore is gratefully appreciated.

This research project was funded by the National University of Singapore

TABLE OF CONTENTS iii

LIST OF TABLES vii

LIST OF FIGURES xiii

LIST OF SYMBOLS xiv

CHAPTER 2 LITERATURE REVIEW 9

CHAPTER 3 EXPERIMENTAL INVESTIGATION

CHAPTER 4 EXPERIMENTAL INVESTIGATION

CHAPTER 5 ANALYTICAL MODEL 105

5.1 GENERAL 105

CHAPTER 6 DESIGN RECOMMENDATIONS

6.1 INTRODUCTION 131

CHAPTER 7 CONCLUSIONS AND

7.1 GENERAL 141

APPENDIX A DESIGN EXAMPLE 146

LIST OF PUBLICATIONS 160

moment rotation characteristics and ultimate capacity of these connections and

beams. Details of the experiments giving information on test specimens,

of 10 haunch connections and 3 continuous composite haunch beam specimens

tested to failure. Results obtained for connection moment capacity, rotation

and detailing, these connections display the characteristics of a rigid connection.

occurred at haunch toes followed by at the mid-span to form a plastic collapse

mechanism. Haunch toe could be designed as the weakest section to form a

haunch length. Experimental results show that composite haunch connection

exhibits a ductile moment-rotation behaviour and is able to redistribute moment

capacity of composite haunch connections. Design guidelines for composite

haunch joints and beams are provided.

Table 2.1 Limits to redistribution of hogging moment to reduce (EC4)

Table 2.2 Maximum redistribution of negative moment in composite haunch

beam at ultimate limit state [Lawson 1989]

Table 3.1 Details of joint test specimens

Table 3.4 Summary of concrete cube test results

Table 3.5 Summary of test results

Table 4.2 Details of beam test specimens

Table 5.1 Comparison of test results with Plastic Hinge Theory

Table 5.2 Comparison of rotational capacity at Haunch Toe

Figure 1.1 Haunch as Taper Section in Universal Beam

Figure 1.2 Cutting of Taper Section in Universal Beam

Figure 2.1 Cross Section of Continuous Composite Beam

Figure 2.2 Sub-frame of Beam and Column in Non-sway Frames Analysis

Figure 2.4 Distorsional Buckling of Composite Beam

Figure 3.1 Building Plan Layout for Joint Specimens'Design

Figure 3.2 Cruciform Joint Specimen

Figure 3.3 Test Specimen Ready for Concrete Casting

Figure 3.4 Typical Joint Test Specimen