See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/282574773

Structural Behavior of Concrete Encased Steel Beams

Article  in  Ain Shams Engineering Journal · January 2005

CITATIONS READS
0 2,518

1 author:

O. El-Mahdy
Benha University
37 PUBLICATIONS   5 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Performance of FRP Strengthened RC elements in Fire View project

Progressive Collapse of Reinforced Concrete Structures View project

All content following this page was uploaded by O. El-Mahdy on 08 November 2017.

The user has requested enhancement of the downloaded file.

 SCIENTIFIC BULLETIN
Received on : 20/12/2005
AIN SHAMS UNIVERSITY Accepted on : 3/1/2006
Vol. 40, No. 4, December 31, 2005
FACULTY OF ENGINEERING pp. : 87-99

STRUCTURAL BEHAVIOR OF CONCRETE ENCASED

STEEL BEAMS
Osama O. EL-MAHDY1

ABSTRACT

Concrete encased steel beams have been used extensively in bridges and
buildings construction for many years. Investigations showed that such
composite beams lead to a reduction of both structure weight and the resulted
deformations. In the present paper, an accurate modeling for concrete encased
steel beams is achieved by using the nonlinear three-dimensional finite element
method through the general purpose computer program DIANA. The analysis
takes into consideration the interaction between the steel and concrete to
simulate the behavior of the encased beams well. The analysis has been
performed for both linear and nonlinear stages. The results obtained from the
present work are compared with the corresponding ones of previous available
experimental works. The comparison showed that the present model is capable
of introducing a good representation for the concrete encased steel beams
behavior and strength.
KEYWORDS: Steel beams; Concrete; Composite beams; Interface bond;
Nonlinear modeling; Finite element method.

! " # $ !% &'
()* +$ , - .
12 3 45 6 7# !" # $8 2, .$ ! 9 , :;% / 0
"
# $ ! 9 , :;? @A B 0/ C # :; . "D "' < $5 =>
D# ! ?H' "' ! @A , I < ? " & G , E'F % / 0 "D
? H OP Q G , % / M'J
"R 0&?!2 2 "JKLM .N "' < ?H / 0 "
$<!2 ?!S" - 1 $ & TQ C * 7 U1 7&/ 0 " "' V WI
7E Y:; " " X H * I * $ 7 U1,/E Y:; 7E /"DZ PE "E /0 ""
' .I< @A , & Q !' " 0 ! Y:;?H'B E @A , J@ , /
\$ ! Y:;[
INTRODUCTION
The combination of steel and reinforced concrete, thereby utilizing the unique
characteristics of the two materials, generally results in structures of greater

1
Associate Professor, Faculty of Engineering at Shoubra, The University of Banha, EGYPT.

87
economy and safety than either material alone could achieve. Because of this,
engineers have been continually interested in finding practical and effective
ways of joining the materials, in developing new design concepts, and in
establishing requirements for satisfactory performance. In the recent years, very
significant advances have been made in all these areas, thus leading to a
widespread use of combined steel and concrete elements in construction of
buildings, bridges, nuclear power plants, and other types of engineering
structures.

Basically, there are two ways that steel and concrete or elements made of them
can be effectively combined. The first is to combine steel shapes with concrete
to form individual members which can then be joined to make up a complete
structure. The individual members or structures thus produced are called
composite structures; typically examples are composite beams and floor slabs,
steel reinforced concrete (SRC) beams or columns, concrete filled tubular
columns, and steel section with partial concrete encasement. The second way is
to develop efficient structural systems by combining steel frames with
reinforced concrete elements, such as shear walls, and interior or exterior cores.
These systems are called mixed or hybrid structural systems. Extensive research
has been carried in many countries to study the behavior of composite and
mixed structures and to develop criteria for the design (Metwally et al. 2000)
and (Noguchi and Uchida 2004).

38 mm
50 mm

Mesh or
other
reinforcing

50 mm

50 mm 50 mm

Fig.1. Encased Beam Minimum Cover Requirements

In this study, concrete encased steel beam shown in Fig. 1 is considered. These
types of beams are widely used in buildings where the steel beams can be either
totally or partial encased in concrete for fire protection. The fire resistance is due

88
to the fact that the concrete part prevents the inner steel parts - structural steel as
well as reinforcing bars- from heating up too fast. In addition to the encased fire
resistance, crippling and local buckling of the steel web is prevented and the
resistance of the steel beam against lateral-torsional buckling is significantly
increased. These beams also have greater stiffness under bending and vertical
shear which results in a reduction of final deflection. Under some circumstances
these encased beams can be designed using two alternatives as per ECP’01: a)
The composite section properties shall be used in calculating bending stresses,
neglecting concrete in tension and b) The steel beam alone is proportioned to
resist all loads, live and dead, neglecting the composite action. No shear
connectors, in the usual sense, are considered in this type of composite beam.
The horizontal shear is transmitted from steel beam to concrete by friction and
bond. In order to qualify as a composite beam, the concrete encasement must
have mesh reinforcement throughout the whole depth and across the soffit of the
steel beam, to prevent spalling of the concrete. The concrete encasement must
also meet the minimum cover requirements according to LRFD Specifications’
1999 (refer to Fig. 1). The slab and the encasement must be cast integrally.

In actual practice, zero composite or non-integral action is impossible because

there is always some degree of natural bond caused by chemical adhesion and
mechanical friction between the concrete and steel beam. Similarly, 100 % or
fully-integral composite action is impossible because there is always some small
degree of slip. There are advantages to the encased beam which are usually
overlooked. Only the compression area of the concrete is considered in the
design. However, all of the concrete contributes a valuable reserve of shear
strength to the beam. Also, the shrewd designer will utilize the additional
stiffness of the encased beam when analyzing the building for combinations of
vertical and lateral loads.

The main purpose of the present study is to summarize the results of the analysis
that have been conducted to investigate the structural behavior of concrete
encased steel beams using DIANA (Version 7) a general purpose finite element
program. The nonlinear behavior of concrete in compression and tension,
progressive cracking, and the bond-slip between steel and concrete are
incorporated in the model. Predefined subroutines for material properties are
coded. The validity of the model is studied by comparing analytical and
experimental results which show an excellent agreement.

89
 Flexural Span
158 I section 200x100x10x7

29
10
180 258
10
29
100
1500
29 29

1700

All Dimensions are in mm

Fig. 2. Steel-Concrete Beam Reference Specimen

REFERENCE SPECIMEN AND FINITE ELEMENT MODELING

A three-dimensional finite element model of concrete encased steel beam shown

in Fig. 2 is presented. This beam is selected for the analysis. It was
experimentally tested by Miura et al. (1984). The symmetrical arrangement of
the specimen made it possible to select only one-fourth of the concrete encased
steel beam to be considered in the analysis. Appropriate boundary conditions are
imposed on the symmetrical planes of the model. Figures 3 and 4 illustrate the
finite element meshing that used in the analysis. Eight-node brick elements
having three degrees of freedom at each node (HX24L in DIANA) are used to
model both the concrete and steel parts. The concrete-steel behavior at the
interfaces is idealized by using nodal and surface contact elements (N6IF and
Q24IF in DIANA). The nodes of the finite element mesh are doubled at the
steel-concrete interface for taking into consideration the bond. Frictional forces
at the interface are neglected. The loads are applied to the specimen in the
analysis. The model has 612 elements and 882 nodal points.

NONLINEAR MATERIAL MODELS

Modeling of Concrete
In tension, the concrete material is idealized as a linearly brittle material,
assuming that cracks occur when the principal tensile stress exceeds the tensile
strength. The concrete model is a smeared crack model in the sense that it does
not track individual macro cracks. Bearing in mind that cracks may be closed
and reopened again, a cracked element is checked in every iteration step for
closure and reopening; a crack is closed or opened up if the stress normal to the
crack surface is compressive or tensile, respectively.

90
 Boundary Conditions
Constraints Nodes
v=0 99, 108, 121 & 137
u=0 On face CDHG
w=0 On face ADHE

Fig. 3. Finite Element Meshing for One-Fourth of the Concrete Encased Steel Beam

91
 /

Fig. 4. Cross-Section Showing Interface Elements and the Double Noding

92
Before cracking, the material is assumed to be isotropic. After cracking, due to
the presence of the crack surface, the material becomes anisotropic. For a
cracked element, The Young’s modulus normal to the crack surface and the
shear modulus parallel to the crack surface are taken as zero. When a crack is
closed, the Young’s modulus normal to the crack surface is restored to the
uncracked value and shear stress of the crack surface is assumed to be taken up
by friction in a way similar to the interface elements. In addition, Poisson’s
effect can be considered negligible after cracking (Vecchio and Selby 1991).

In compression, the concrete material exhibits extensive nonlinearity in the

stress-strain relation. As the solid elements are characterized by three-
dimensional stress situation, the concrete material is under triaxial stresses.
However, the triaxial stress state is approximately uniaxial (i.e. two of the
principal stresses is much smaller than the third one). Therefore, the analysis can
be simplified with the following assumptions: 1) The Young’s modulus varies as
a function of the larger compressive principal strain, as in the uniaxial case; 2)
The Poisson’s ratio remains constant at any stress level; and 3) The concrete is
crushed when the principal compressive strain exceeds the ultimate limit and
once the concrete fails in compression, it can never regain any strength.

Stress

fcu

0.3 fcu

0.002 0.0035 Plastic Strain

Strain Strain

Fig.5. Stress-Plastic Strain Curve for Concrete under Uniaxial Compression

The concrete constitutive model in DIANA requires input of several parameters.

These include points from the stress versus plastic strain curve in uniaxial
compression, concrete uniaxial tensile strength, concrete biaxial compressive
strength, principal component of plastic strain at ultimate state in biaxial
compression and principal stress at cracking in plane stress when the other
nonzero principal stress component is at its ultimate compressive stress value.
The stress-plastic strain relationship for concrete under uniaxial compression is

93
shown in Fig. 5. DIANA default values are used for those parameter requiring
biaxial test data. The concrete failure criteria follow Drucker-Prager in the space
for principal stresses flow σ1, σ2, and σ3, as shown in Fig. 6. The yield point has
to be on the cone surface or inside it.

Modeling of Steel

The steel material is idealized as an isotropic elasto-plastic model based on the

von Mises criterion. σ2
Hydrostatic Axis

σ1=σ2=σ3

C cot φ

σ1

σ3

Fig. 6. Drucker-Prager Yield Criterion

Modeling of Bond

Plane linear quadrilateral and three dimensional nodal interface elements are
utilized in the present analysis (refer to Fig. 3). These elements can be arranged
along the entire steel-concrete interface and allow more rational use of load-slip
relationships derived from previous experimental works (ASCE, 1982 and Plauk
and Hees, 1981). The element constitutive relationship is formulated in terms of
relative displacements of its top and bottom surfaces according to

Fs = K s ( ∆ s − ∆ r ) (1)
Fn = K n ∆ n (2)

where Fs, Fn = Shear and normal forces at interface.

∆s , ∆n = Relative slip and relative normal displacement of interface.
∆r , ∆r' = Previous and new residual slips at interface.
Ks , Kn = Shear and normal stiffness of interface.

94
 Shear force (Fs)
Slip (∆s) Concrete
Interface

Steel

Normal force (Fn)

Separation (∆n)
Fig. 7 Representation of Interface in Finite Element Analysis

Based on the experimental results for pull-out tests on flat plates with clear
covers of 29 mm to the surface of the plate, a bilinear bond stress-slip
relationship shown in Fig. 8 is used in the present analytical model. If the
calculated friction is greater than the limiting friction, the friction developed is
equated to the limiting friction and the residual slip ∆r is reduced to

Fs
∆ r′ = ∆ s − (
) (3)
Ks
As cracks develop and propagate, the interface normal and shear stiffness are
changed. Furthermore, the interface elements are removed when the interface
had separated.

Fs
µFn
ks

ks

∆r ∆r' ∆s

−µFn

Fig. 8 Idealized Bond-Slip Relationship at the Interface

It may be assumed that complete interaction between steel and concrete is

maintained under serviceability limit state. The complete interaction between
steel and concrete shall be ensured in such a way that bond stresses between

95
steel and concrete shall not exceed 20 percent of bond strength under
serviceability conditions (JSCE, 1986).

LOADING AND SOLUTION STRATEGY

Following the application the applied concentrated live load. A sufficient

number of iterations are allowed during the analysis until a converged solution is
reached. Convergence in DIANA is attained when the maximum residual nodal
forces are less than a user-specified tolerance, which is defined as a small
fraction of the applied nodal forces. The applied load is increased from zero to
10 KN in four steps. The load step size is then decreased to 1.50 KN up to a total
load of 16 KN. Beyond this load and up to a load of 32 KN, a load step size of
1.0 KN is used. This is further reduced to 0.50 KN and the analysis is carried out
up to a load of 52 KN. However, a singularity problem is encountered when the
load is increased from 52 KN to 52.5 KN. The DIANA program indicated that
the "plasticity algorithm did not converge at some node”. The load step size is
then decreased to 0.30 KN and the analysis is successfully performed to 56 KN.
Once again, no converged solution is obtained between 56 KN and 56.30 KN.
The solution is restarted from 56 KN using a load increment of 0.15 KN.
However, the numerical problems are encountered as the load is increased to 64
KN. The load step size is decreased to 0.10 KN and the analysis is continued
from a load of 64 KN. Converged solutions are reached as the load is increased
beyond this pressure. Several unsuccessful attempts are made to continue the
analysis beyond this load, but convergence could not be achieved.

ANALYTICAL RESULTS

Three analytical cases are investigated. In case 1, perfect bonding is assumed

between the structural steel and the concrete beam. In case 2, zero bond is
assumed between steel and concrete. In case 3, bond-slip model shown in Fig. 8
is utilized. The normal and shear stiffness of the plane and node interface
elements are assumed to have either zero or infinite values in order to simulate
the cases of no or perfect bond, respectively. Table (1) illustrates the obtained
analytical results for the steel tensile stress and vertical displacement at the
nodal point numbers 874 and 872, respectively (see Fig. 3).

Table 1. Tensile Stresses and Vertical Displacements Results

Case Perfect Bond Zero Bond Bond-Slip
2
Steel Tensile Stress (N/mm ) 340.48 425.60 383.04
Vertical Displacement (mm) 13.683 24.033 16.823

96
Ultimate Strength Results

The ultimate moment obtained using the present model is shown in Table 2.
Furthermore, the comparison of the calculated value with those measured
experimentally by Miura et al. 1984 is presented in this table. From this
comparison, it can be concluded that the measured values are approximately
greater than the calculated value only by about 7%.

Table 2. Results of the Obtained and Measured Ultimate Strength

Specimen Measured Value Mum Calculated Value Muc

Mum/Muc
No. (KNm) (KNm)
1 79.990 73.26 1.09
2 78.284 73.26 1.07
3 76.224 73.26 1.04

Load-Deflection Relationship

Figure 9 shows the load-deflection curve at the nodal point of the mid-span
obtained from the present analysis using bond-slip model. The degradation of
the slope of this curve represents the effect of the reduced stiffness and the
interface bond-slip. The analytical ultimate concentrated load is 256 KN.

80
70
60
Load (KN)

50
40
30
20
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Deflection (mm)

Fig. 9. Analytical Load-Deflection Curve

97
CONCLUSIONS

Although there are various types of composite structure, the common design
method has not yet been established. Concrete encased steel beams are widely
used in buildings. However, there have been few experimental or analytical
studies on this area. The reluctance of practicing engineers to take the more
accurate behavior of these beams has been, firstly, due to lack of knowledge
concerning the composite behavior and secondly, due to lack of practical
methods for predicting the stiffness and strength. This paper concerns the
development of a three-dimensional nonlinear finite element model for studying
the behavior of concrete encased steel beams beam specimen that was
previously tested by Miura et al. with regards to both serviceability and ultimate
limit states. The nonlinear behavior of component materials (concrete cracking,
nonlinear behavior of concrete, yield of steel beam, and bond-slip relationship at
the interface), is adequately considered in the model. The results obtained from
this model are compared with the measured experimental ones. The comparison
demonstrates that the proposed finite element model is accurate enough to
model this type of composite beams. This model may therefore, serve as a
simulation tool or as a means to improve design codes as well.

REFERENCES

1. Noguchi, H. and Uchida, K., (2004), “Finite Element Method Analysis of

Hybrid Structural Frames with Reinforced Concrete Columns and Steel
Beams”, Proceedings of Journal of Structural Engineering, ASCE, Vol.
130, No. 2, pp. 328-335.
2. Metwally, M.E.A., Allam, H.M., and Atia, A.H., (2000), “Shear Behavior
of Concrete Beams with Embedded Steel Sections”, Proceedings of The
Second International Conference on Civil Engineering, Faculty of
Engineering, Helwan University, pp. 175-183.
3. ECP’01, (2001), “Egyptian Code of Practice for Steel Construction and
Bridges”, Ministerial Decree No. 279-2001.
4. LRFD Specification for Structural Steel Buildings, (1999), American
Institute for Steel Construction, INC., Chicago, Illinois, 60601-2001.
5. DIANA 7 “Finite Element Analysis”, TNO Buildings and Construction
Research, Netherlands.
6. Vecchio, F.J. and Selby, R.G., (1991), “Toward Compression-Field
Analysis of Reinforced Concrete Solids”, Proceedings of Journal of
Structural Engineering, ASCE, Vol. 117, No. 6, pp. 1740-1758.
7. ASCE Task Committee on Finite Element Analysis of Reinforced Concrete
Structures, (1982), “State-of-the-Art Report on Finite Element Analysis of
Reinforced Concrete”, American Society of Civil Engineers, New York.

98
 8. Plauk, G., and Hees, G., (1981), "Finite Element Analysis of Reinforced
Concrete Beams with Special Regard to Bond Behavior", Proceedings of
IABSE Colloquium on "Advanced Mechanics of Reinforced Concrete",
Vol. 34, pp. 655-670.
9. Standard Specifications for Design and Construction of Concrete Structures
(1986) Part 1 (Design), Japan Society of Civil Engineers (JSCE), Tokyo,
Japan.
10. Miura, T., Sano, M. and Sato, M., (1984), “Properties of Concrete Beams
Reinforced with Deformed Steel Bars, Plates or Shapes”, Transactions of
The Japan Concrete Institute, Vol. 6, pp. 449-456.

99

View publication stats