Alexandria Engineering Journal (2014) xxx, xxx–xxx

Alexandria University

Alexandria Engineering Journal

www.elsevier.com/locate/aej
www.sciencedirect.com

ORIGINAL ARTICLE

Prediction of the behavior of reinforced concrete deep

beams with web openings using the finite element method
Ashraf Ragab Mohamed *, Mohie S. Shoukry, Janet M. Saeed

Structural Engineering Department, Faculty of Engineering, Alexandria University, Alexandria 21544, Egypt

Received 8 December 2013; revised 11 February 2014; accepted 2 March 2014

KEYWORDS Abstract The exact analysis of reinforced concrete deep beams is a complex problem and the pres-
Deep beam; ence of web openings aggravates the situation. However, no code provision exists for the analysis of
R.C.; deep beams with web opening. The code implemented strut and tie models are debatable and no
Opening; unique solution using these models is available. In this study, the finite element method is utilized
Finite element; to study the behavior of reinforced concrete deep beams with and without web openings. Further-
Reinforcement more, the effect of the reinforcement distribution on the beam overall capacity has been studied and
compared to the Egyptian code guidelines. The damaged plasticity model has been used for the
analysis. Models of simply supported deep beams under 3 and 4-point bending and continuous deep
beams with and without web openings have been analyzed. Model verification has shown good
agreement to literature experimental work. Results of the parametric analysis have shown that
web openings crossing the expected compression struts should be avoided, and the depth of the
opening should not exceed 20% of the beam overall depth. The reinforcement distribution should
be in the range of 0.1–0.2 beam depth for simply supported deep beams.
ª 2014 Production and hosting by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria
University.

1. Introduction gion is the D- or Disturbed regions, where the strain distribution

Structural bending members can be broadly divided into two re-
gions. The first region is the Bernoulli regions (B-Region), where
the strain distribution across the section is linear. The second re-
* Corresponding author. Tel.: +20 1005424456; fax: +20 34240456.
E-mail address: ashrafrm04@yahoo.com (A.R. Mohamed).
Peer review under responsibility of Faculty of Engineering, Alexandria
University.
Production and hosting by Elsevier
is nonlinear as the case of deep beams. Reinforced concrete deep
beams have many useful applications in building structures such
as transfer girders, wall footings, foundation pile caps, floor
diaphragms, and shear walls. The use of deep beams at the lower
levels in tall buildings for both residential and commercial pur-
poses has increased rapidly because of their convenience and
economical efficiency. It is recognized that the distribution of
the strain across the section of deep beams is nonlinear and
hence, these structural elements belong to the D-Regions,
Nagarajan and Madhavan [1]. Traditionally, the D-Regions
have been designed using empirical formulae or past experience.
Recently, the Strut-and-Tie Model (STM) has been recognized
as an effective tool for the design of both B- and D-Regions
and it has found place in many design codes.

1110-0168 ª 2014 Production and hosting by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria University.
http://dx.doi.org/10.1016/j.aej.2014.03.001

Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
2 A.R. Mohamed et al.
The strut and tie model (STM) provides design engineers
with a more flexible and intuitive option for designing struc-
tural elements. The complex stress flows in a cracked concrete
structure are approximated with simple truss elements that can
be analyzed and designed using basic structural mechanics.
Though the STM is effective for the design of D-Regions,
the method has not yet been widely implemented due to many
reasons such as: (1) the difficulty in fixing an optimum truss
configuration for a given structural member with given load-
ing, (2) the complexity and approximation of the solution
and the inability of the STM to predict the failure modes of
deep beams, Tan et al. [2] and Yang et al. [3].
It has been recognized that the finite element method can
provide realistic and satisfactory solutions for nonlinear
behavior of reinforced concrete structures. Therefore, the finite
element software, ABAQUS [4], has been used to study the
behavior of reinforced concrete deep beams with and without
web opening under monotonic loading actions. First, the mod-
eling technique has been verified by comparing the model pre-
diction to experimental work in the literature. A parametric
study has been conducted to predict the behavior of simply
supported and continuous reinforced concrete deep beams un-
der 3-points and 4-points bending configurations. Also, it
examines the effect of the location of web openings in both
simple and continuous deep beams. Finally, the effect of the
reinforcement distribution on the overall capacity of the beam
has been conducted. The results of this study have been com-
pared with the ACI 318-08-Appendix A [5], and the Egyptian
Code (EC 203-2006) [6] recommendations.
2. Background
Deep beams are defined as members loaded on one face and
supported on the opposite face so that compression struts
can develop between the loads and the supports. Their clear
spans are either equal to or less than four times the overall
member depth; or regions with concentrated loads within twice
the member depth from the face of the support, ACI 318-08
[5]. The EC 203-2006 [6] adopts the same definition as ACI
318-08, whereas the Euro Code [7] defines a deep beam as a
member whose span is less to or equal to 3 times the overall
section depth. These structural elements belong to D (Dis-
turbed) regions, which have traditionally been designed using
empirical formulae or using past experience.
STM is a recent development in the analysis and design of
reinforced concrete structural elements. In STM, the rein-
forced concrete member is replaced by an equivalent truss,
where the compression and tension zones are converted into
equivalent struts and ties connected at the nodes to form a
truss resisting the applied loads.
Design codes provide an extensive explanation and illustra-
tion of the struts, ties and nodes’ shapes, classification and
detailing. In addition to the permissible stresses in struts and
nodes and the corresponding cross sectional areas of struts
and nodes [5–7]. Fig. 1 illustrates a schematic representation
of the STM developed for deep beams under 3- and 4-points
bending configurations respectively.
The STM has been subjected to ongoing debates due to the
difficulty in constructing the optimum truss configuration for a
given loading. Traditionally, STM has been developed using
load path method or with the aid of stress trajectories.
Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
However, this STM is not unique and varies with the de-
signer’s intuition and past experience. In order to overcome
the limitations associated with the development of the STM,
the Finite Element Method (FEM), is applied in the present
study to predict the behavior of reinforced concrete deep
beams. Results are compared to the corresponding code provi-
sions for the design of deep beams using the STM.
FEM has proven to be a versatile tool for studying the non-
linear behavior of reinforced concrete structures. Current ad-
vances in computational capabilities have motivated the
development of large number of commercial finite element
codes. These codes have shown the adequate reliability and
accuracy to study the behavior of reinforced concrete struc-
tures. In the present study, the damaged plasticity model, as
implemented in the general purpose finite element software
ABAQUS [4], is used to study the behavior of reinforced con-
crete deep beams. This constitutive modeling has proved to be
the most stable regime for modeling concrete nonlinear behav-
ior. It shows the ability to capture the whole concrete behavior
up to failure with reliable accuracy when compared to the
experimental results, Saeed [8].
The concrete damaged plasticity model in ABAQUS [4] is
based on the models proposed by Lubliner et al. [9] and Lee
and Fenves [10]. The model uses the concepts of isotropic dam-
aged elasticity in combination with isotropic tensile and com-
pressive plasticity to represent the inelastic behavior of
concrete. The model consists of the combination of non-asso-
ciated multi-hardening plasticity and scalar (isotropic) dam-
aged elasticity to describe the irreversible damage that occurs
during the fracturing process. The elastic behavior of the mate-
rial is isotropic and linear. The model is a continuum, plastic-
ity-based, damage model for concrete. It assumes that the
main two failure mechanisms are tensile cracking and com-
pressive crushing of the concrete material. The evolution of
the yield (or failure) surface is controlled by two hardening
variables linked to failure mechanisms under tension and com-
pression loading, respectively. Fig. 2 shows the uniaxial tensile
and compressive behavior of concrete, respectively, used in the
concrete damaged plasticity model. As depicted from the fig-
ure, if the concrete is unloaded at any point on the softening
branch, the elastic stiffness is reduced. The effect of the dam-
age is different in tension and compression, and the degraded
response of concrete is taken into account by introducing two
independent scalar damage variables for tension and compres-
sion respectively.
3. Research program
The research program includes two parts; the first part is the
validation of the proposed model using experimental data
from literature. The second part is concerned with the para-
metric study.
3.1. Model validation
In order to validate the ability of the selected concrete model
to study the tensile and compressive behavior of reinforced
concrete deep beams, a benchmark test has been carried out
using one of the deep beams (Beam SS-1), studied by Hong
et al. [11] for the evaluation of shear strength of deep beams.
This test serves as a source for comparison with the existing
Reinforced concrete deep beams with web openings 3

Figure 1 Schematic representation of STM.

Figure 2 Response of concrete due to (a) uniaxial tension, (b) uniaxial compression [4].

experimental results. In the study conducted by Hong et al.
[11], simply supported beams were instrumented to measure
the mid span deflections and loads. Fig. 3 illustrates the cross
section and loading configuration of the tested beam. An 8-
node solid element with one point integration was utilized to
create the concrete beam mesh. An embedded truss reinforce-
ment a 2-node linear 3D truss element was used to model steel
rebars. The mesh used in this validation is shown in Fig. 4.
Fig. 5 illustrates the load–deflection response of the studied
beam in comparison with the experimental results obtained by
Hong et al. [11]. The modeled response verifies the ability of
the selected model to capture the whole beam’s behavior up
to failure and shows a good agreement to the experimental re-
sults. The results of the model can be used in validating and
guiding experimental work, in addition to exploring concrete
response under complicated loading conditions such as the
behavior of reinforced concrete deep beams with and without
web opening introduced in the current study.
3.2. Parametric study
The research program for the parametric study conducted in
this paper consists of (9) deep beams. All the studied beams
had a clear span (lc) of 6000 mm, a depth (d) of 2000 mm deep
and a width (b) of 500 mm. Fig. 6 shows the studied beams’
geometry and dimensions, while Fig. 7 shows typical example
for the meshing of deep beam DS3-1W.
The studied beams were categorized in three main groups;
the first group is for simply supported deep beams with and
without web openings under 3-points bending (Beams DS3-0,

Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
4 A.R. Mohamed et al.

P P DS3-1W, DS3-2W). The second group includes simply sup-

ported deep beams with and without web openings under 4-
Stirrups points bending (Beams DS4-0, DS4-1W. and, DS4-2W). On
2 D= 10 mm 10mm Diam.
the other hand, the third group contains continuous deep
h = 600 mm
beams with and without web openings (Beams DC-0, DC-
Bottom Rebars 1W and DC-2W). The mechanical properties of the concrete
(4 D=19mm) and reinforcing bars for all studied beams are summarized in
Table 1. The material properties of concrete are taken as pro-
150mm
posed by Hong et al. [11].
1.40m
The beams were designed in accordance with the recommen-
Figure 3 Cross section and loading configuration of beam SS-1 dation of the EC 203-2006 [6]. Table 2 summarizes the rein-
[11]. forcement detailing for the beams. The deep beams have been
modeled using the damaged plasticity model in ABAQUS/Stan-
dard. Concrete tensile behavior has been modeled as a quasi-
brittle material and the beam reinforcement has been modeled
as embedded truss elements. The considered variables in this
parametric study are the loading scheme, the location of web
openings and the reinforcement distribution. The obtained re-
sults are compared to both the ACI 318-08 [5] and the EC
203-2006 [6] provisions, in terms of the strut and tie widths
and the effective reinforcement distribution depth.

4. Results and discussions

4.1. Load–deflection response

Figs. 8–10 show the load deflection response of the simply sup-
ported deep beams under 3-points and 4-points bending and
continuous deep beam respectively. It can be depicted from
these figures that the presence of web openings crossing the
compression struts developed between the load and the sup-
ports (case of beams DS3-2W, DS4-2W, DC-1W) resulted in
a substantial reduction in the failure load and the beam’s over-
all capacity. This reduction ranges from 33% in case of the
simply supported deep beam under 3-points bending to 36%
in case of the simply supported deep beam under 4-points
Figure 4 The applied mesh for beam SS-1.
bending. However, when the web openings were located out-
side the expected line of action of compression struts, the
reduction in the failure load ranged from 3% in case of beam
DS4-1W to 15% in case of beam DS3-1W
7 10 5
In order to study the effect of the size of the web opening on
the beam’s overall capacity, four different web openings were
6 10 5
considered in Beam DS3-1W with length ranging from 0.1 lc
to 0.2 lc and a total depth from 0.075d to 0.3d. Fig. 11 illustrates
5 10 5 the load–deflection response of Beam DS3-1W for different web
opening sizes and the corresponding failure load. As depicted
Load (KN)

4 10 5 from this figure, for deep beams having the same web opening
depth, the reduction in the failure load due to the increase in
3 10 5 the web opening width from B1 = 600 mm to B2 = 2B1 =
1200 mm is negligible. On the other hand, for deep beams hav-
2 10 5
ing a constant web opening width and a variable depth from H1
to H2 = 2H1, the reduction in the failure load and the beam
Model
Experimental capacity was about 7%. Therefore, the most important observa-
1 10 5
tion that can be deduced from Fig. 11 is that, as long as the
opening does not interfere with the load path or stress trajecto-
0
0 1 2 3 4 5 6 7 ries, i.e. compression struts, the depth is the most important
Mid-Span Deflection (mm) parameter influencing the beam’s overall capacity. Hence, for
an overall reduction in the beam’s capacity not exceeding
Figure 5 Comparison of load–deflection response for the applied 10%, the depth of the central opening should not exceed 20%
material model vs. experimental results. of the beam overall depth (0.2d).

Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
Reinforced concrete deep beams with web openings 5

(a) Beam DS3-0 (b) Beam DS3-1W (c) Beam DS3-2W

(d) Beam DS4-0 (e) Beam DS4-1W (f) Beam DS4-2W

(g) Beam DC-0 (h) Beam DC-1W i) Beam DC-2W

Figure 6 Beams geometry and dimensions.

greater than zero, and the maximum principal plastic strain

is positive. Based on these criteria, the direction of the vector
normal to the crack plane is assumed to be parallel to the
direction of the maximum principal plastic strain. This direc-
tion can be viewed in the visualization module of the post-pro-
cessor ABAQUS/CAE.

4.3. The developed struts’ width

Compression struts have developed between the load and sup-

Figure 7 The applied mesh for beam DS3-1W.
4.2. Crack pattern
The general failure modes of reinforced concrete members are
represented by the occurrence and development of cracks and
crushing of concrete. Fig. 12 illustrates the crack pattern at
failure for the studied beams. It should be noted that the con-
crete damaged plasticity model does not have the notion of
cracks developing at the material integration point. However,
it is possible to introduce the concept of an effective crack
direction with the purpose of obtaining a graphical visualiza-
tion of the cracking patterns in the concrete structure. The cri-
teria adopted in this model are the assumption that cracking
initiates at points where the tensile equivalent plastic strain is
ports. These struts’ widths obtained from the model are com-
pared with the corresponding values evaluated using the ACI
318-08 [5] and the EC (203-2006) [6] equations. Tables 3–5
summarize the comparison between the calculated and the ob-
tained strut’s widths from the model, for the studied deep
beams without web openings.
These results show that the code provisions for the design
of reinforced concrete deep beams substantially underestimate
the strength of the reinforced concrete members. Therefore the
actual capacity of the structure is greater than that of the ide-
alized truss, and, hence, designs based on STM are always on
the safer side. Moreover, as the current study has shown that
the evaluated strut width is wider than the expected strut
width, as obtained by the model, this will allow more openings’
width to be considered in deep beams.
It is noteworthy that no code provisions have been pro-
vided in the ACI 318-08, EC 203-2006 or BS EN 1992-1-1:
2004. Hence, for the design of deep beams with web openings,
no comparison could be made for the stress distribution or the
developed struts widths for the studied reinforced concrete
deep beams with web opening. Fig. 13 illustrates the developed
compression struts between the load and supports for all the
studied deep beams.

Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
6 A.R. Mohamed et al.

Table 1 Material properties for the studied deep beams [11].

Concrete properties:
1. Elastic properties
Young’s modulus 20 GPa
Poisson’s ratio 0.2
2. Uniaxial compression values
Characteristic compressive strength ( f 0c ) 23.5 MPa
3. Uniaxial tension values
Cracking failure stress 2 MPa
a
4. Details of softening behavior Tensile stress (rt) Displacement across crack (uo)
2.00 0
1.5753 5.3633E005
1.0559 9.9559E005
0.70781 0.00014054
0.47446 0.00017843
0.31804 0.00021451
0.21319 0.00024966
0.1429 0.00028447
0.095791 0.00031939
0.064211 0.00035472
0.043042 0.00039072
5. Details of tension damage (dt)a Damage parameter (dt) Displacement across crack (uo)
0 0
0.381217 5.3633E005
0.617107 9.9559E005
0.763072 0.00014054
0.853393 0.00017843
0.909282 0.00021451
0.943865 0.00024966
0.965265 0.00028447
0.978506 0.00031939
0.9867 0.00035472
0.99177 0.00039072
Steel (rebar) properties:
Young’s modulus 230 GPa
Poisson’s ratio 0.3
Yield stress 392 MPa
a
Not listed by the reference.

4.4. The stress and strain distributions
The stress and strain distributions for all the studied deep
beams at failure, have been extracted and evaluated. Typical
example of these distributions is shown in Fig. 14. As expected,
the strain distribution is nonlinear and the stress distribution in
tension is limited to the assumed tensile strength of concrete,
while the compressive stress simulates that assumed by the
code.
4.5. Effect of reinforcement distribution
Since 2003, the EC 203-2006 Committee has issued a hand-
book for the detailing of structural members. This handbook
introduces some guidelines for reinforcement detailing of dif-
ferent structural members. For simply supported deep beams
under uniformly distributed loads, it is recommended that
the main tension reinforcement should be distributed along a
depth ranging from 0.15 to 0.3 of the overall beam depth. This
range varies according to the type of the applied loads and the
boundary conditions. In the present study, this guideline was
reviewed for simply supported deep beams by distributing
the main tension reinforcement over varying depths ranging
from 0.05 to 0.4 of the total beam depth (H). The effect of rein-
forcement distribution on the beam capacity has been studied
and the stress distribution in reinforcing steel has been moni-
tored. Fig. 15 illustrates the different reinforcement distribu-
tion applied in the present study.
The simply supported reinforced concrete deep beam under
3-points bending, (Beam DS3-0) illustrated in Fig. 6a, was
studied for the different reinforcement distribution. Fig. 16
illustrates the load–deflection response of the beam and the ob-
served reduction in the beam capacity with increasing rein-
forcement depth.
From the figure, it can be depicted that no significant reduc-
tion in the beam’s capacity was observed when the reinforce-
ment depth increased from 0.05H to 0.1H. However, a
reduction of 6.0% in the beam capacity was detected when
the upper limit specified in the handbook for reinforcement

Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
Reinforced concrete deep beams with web openings 7

Table 2 Reinforcement of tested deep beams.

Beam ref. Tension longitudinal Compression longitudinal Horizontal Vertical
reinforcement reinforcement reinforcement reinforcement
DS3-0 12 /16 6 /16 /16@275 mm /12@275 mm
DS3-1W 12 /16 6 /16 /16@275 mm /16@275 mm
DS3-2W 12 /16 6 /16 /16@275 mm /16@275 mm
DS4-0 12 /16 6 /16 /16@275 mm /12@275 mm
DS4-1W 12 /16 6 /16 /16@275 mm /16@275 mm
DS4-2W 12 /16 6 /16 /16@275 mm /16@275 mm
DC-0 12 /16 12 /16 /16@275 mm /12@275 mm
DC-1W 12 /16 12 /16 /16@275 mm /16@275 mm
DC-2W 12 /16 12 /16 /16@275 mm /16@275 mm

5000 12000

Beam DC-0
10000
4000
Beam DC-1W

8000
Total Load (KN)

Total Load (KN)

3000 Beam DS3-0 Beam DC-2W

Beam DS3-1W 6000

2000
Beam DS3-2W
4000

1000 Beam DC-0

Beam DS3-0
Beam DS3-1W 2000 Beam DC-1W
Beam DS3-2W
Beam DC-2W
0
0
0 2 4 6 8 10 0 2 4 6 8 10
Mid-Span Deflection (mm) Mid-Span Deflection (mm)

Figure 8 Load–deflection response of simply supported deep Figure 10 Load–deflection response of continuous deep beams.
beams under 3-points bending configuration.

3000

5000 2500

Beam DS4-0

4000 2000
Total Load (KN)

Beam DS4-1W
Total Load (KN)

3000 1500
Beam DS4-2W

2000 1000

Beam DS4-0 600x150 Web Opening

Beam DS4-1W 500 600x300 Web opening
1000 Beam DS4-2W 1200x300 Web opening
1200x600 Web Opening

0 0
0 2 4 6 8 10 0 1 2 3 4 5 6 7 8
Mid-Span Deflection (mm) Mid-Span Deflection (mm)

Figure 9 Load–deflection response of simply supported deep Figure 11 Load–deflection response of beam DS3-1W for
beams under 4-points bending configuration. different web openings.

Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
8 A.R. Mohamed et al.

(a) Beam DS3-0 (b) Beam DS3-1W (c) Beam DS3-2W

(d) Beam DS4-0 (e) Beam DS4-1W (f) Beam DS4-2W

(g) Beam DC-0 (h) Beam DC-1W (i) Beam DC-2W

Figure 12 Visualization of crack pattern.

(a) Beam DS3-0 (b) Beam DS3-1W (c) Beam DS3-2W

(d) Beam DS4-0 (e) Beam DS4-1W (f) Beam DS4-2W

(g) Beam DC-0 (h) Beam DC-1W (i) Beam DC-2W

Figure 13 The developed compression struts.

0 0

500 500
Beam Depth (mm)

Beam Depth (mm)

1000 1000

1500 1500

2000 2000
4 2 0 -2 -4 -6 -8 0.003 0.0025 0.002 0.0015 0.001 0.0005 0 -0.0005
Stress (MPa) Strain

(a) Stress Distribution of Beam (b) Strain Distribution of Beam

DS3-0 at failure DS3-0 at Failure
Figure 14 Stress and strain distribution of simply supported deep beam without web opening under 3-points bending (Beam DS3-0) at
beam mid span at failure.

Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
Reinforced concrete deep beams with web openings 9

H = 2000 mm

0.4 H

0.3 H
0.2 H

0.1 H
0.05 H

B = 50mm

(a) (b) (c) (d) (e)

Figure 15 Reinforcement distribution.

distribution; 0.3H, was applied. Table 6 summarizes the ap- 2 10 6

plied reinforcement distribution depths and the corresponding
reduction in the beam’s overall capacity. From this compari-
son, it is suggested that the reinforcement distribution should
be in the range of 0.1–0.2H for simply supported deep beams. Central Load (N) 1.5 10 6

Table 3 The strut and tie width for beam DS3-0. 1 10 6

Current study ACI 318-08 EC (203-2006)

Strut width (Ws) mm 471.73 670.3 761.8 0.05H
Tie width (WT) mm 278.1 444.6 677 5 10 5
0.1H
0.2H
0.3H
0.4H

0
Table 4 The strut and tie width for beam DS4-0. 0 0.5 1 1.5 2 2.5 3 3.5 4
Current study ACI 318-08 EC (203-2006) Central Deflection (mm)

Strut width (Ws) mm 518.3 659.76 726.3

Figure 16 Load–deflection response of simply supported deep
Tie width (WT) mm 381.05 320.8 488.6
beam with various tension reinforcement distributions.

400

Table 5 The strut and tie width for beam DC-0.

350
Current ACI EC
study 318-08 (203-2006) 300
Strut width (Ws1) mm 549.35 316.76 482.2
Strut width (Ws2) mm 607.4 950 1446.7 250
Stress (MPa)

Tie width (WT1) –top mm 357.5 412.5 628.25

Tie width (WT2) – bottom mm 356.8 275 419 200

150
R1 @ 75mm
R2 @ 100mm
100
Table 6 Reduction in beam capacity due to reinforcement
distribution. 50

Tension reinforcement depth % Reduction in beam capacity

0
0 1000 2000 3000 4000 5000 6000 7000
0.1H (200 mm) 0.34
0.2H (400 mm) 2.1 Length (mm)
0.3H (600 mm) 5.87
0.4H (800 mm) 12.16 Figure 17 Stress distribution in rebars (distribution
depth = 0.05H).

Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
10 A.R. Mohamed et al.

400 400

R1@75mm
350 350 R2@256mm
R3@437mm
R4@619mm
300 300
R5@800mm

250 250

Stress (MPa)
Stress (MPa)

200 200

150 150

R1@75mm
100 R2@137mm 100
R3@200mm

50 50

0 0
0 1000 2000 3000 4000 5000 6000 7000 0 1000 2000 3000 4000 5000 6000 7000

Length (mm) Length (mm)

Figure 18 Stress distribution in rebars (distribution Figure 21 Stress distribution in rebars (distribution
depth = 0.1H). depth = 0.4H).

400 Furthermore, the stress distribution in the tension rein-

forcement was monitored and recorded for different reinforce-
350 ment distribution schemes. Figs. 17–21 illustrate the stress
distribution in each rebar layer along the rebar length.
300 From this set of figures, it can be observed that, for smaller
distribution depth (0.05H), the stresses in the rebar layers are
250 almost identical. A reduction in the reinforcement stresses is
Stress (MPa)

observed when the rebar depth increases and this reduction

200
is proportional to the rebar position.
150
5. Summary and conclusions
R1 @ 70mm
100 R2 @ 157.5mm
R3 @ 240mm
R4 @ 322.5mm Based on this study, it is recognized that the exact analysis of
50 R5 @ 400mm reinforced concrete deep beams is a complex problem and the
presence of web openings aggravates the situation. The appli-
0
0 1000 2000 3000 4000 5000 6000 7000 cation of the damaged plasticity model implemented in ABA-
Length (mm) QUS/Standard for the analysis of simply supported deep
beams under 3-points and 4-points bending and continuous
Figure 19 Stress distribution in rebars (distribution deep beams with and without web openings provided useful
depth = 0.2H). information about the responses of reinforced concrete deep
beams under monotonic loadings. The most important conclu-
400 sions of the conducted parametric study can be summarized as
follows:
350

1. The model validation with experimental work from litera-

300
ture has shown that the model is capable of capturing the
entire response reasonably.
250
Stress (MPa)

2. The web openings crossing the expected compression struts

200
developed between the load and the supports cause approx-
imately about 35% reduction in the beam’s capacity in all
150 the studied cases and hence it should be avoided.
3. When the introduced web opening does not interfere with
100 R1@75mm the load path or stress trajectories, i.e. compression struts,
R2@206mm the observed reduction the beam’s capacity ranged from
R3@337m
50 R4@469mm 6% to 8% depending on the opening dimensions.
R5@600mm 4. The depth of the opening is the most important parameter
0 influencing the beam’s overall capacity. Therefore for an
0 1000 2000 3000 4000 5000 6000 7000
Length (mm) overall reduction in the beam’s capacity not exceeding
10% of the capacity the beam without web opening, the
Figure 20 Stress distribution in rebars (distribution depth of the opening should not exceed 20% of the beam
depth = 0.3H). overall depth (0.2d).

Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
Reinforced concrete deep beams with web openings
5. The effect of reinforcement distribution on the beam capac-
ity was studied and the stress distribution in reinforcing
steel has been monitored. A reduction, ranging from
0.3% to 12% in the beam capacity, was observed, when
the tension reinforcement distribution depth was increased
from 0.1H to 0.4H. Thus, it is suggested that the reinforce-
ment distribution should be in the range of 0.1–0.2H for
simply supported deep beams.
References
[1] P. Nagarajan, P.T.M. Madhavan, Development of strut and tie
models for simply supported deep beams using topology
optimization, Songklanakarin J. Sci. Technol. 30 (5) (2008)
641–647.
[2] K.H. Tan, K. Tong, C.Y. Tang, Consistent strut-and-tie
modeling of deep beams with web openings, Mag. Concr. Res.
55 (1) (2003) 572–582.
[3] K. H Yang, H.C. Eun, H.S. Chung, The influence of web
openings on the structural behavior of reinforced high-strength
concrete deep beams, Eng. Struct. 28 (13) (2006) 1825–1834.
Please cite this article in press as: A.R. Mohamed et al., Prediction of the behavior of reinforced concrete deep beams with web openings using the
finite element method, Alexandria Eng. J. (2014), http://dx.doi.org/10.1016/j.aej.2014.03.001
11
[4] ABAQUS User Manual, Version 6.7, ABAQUS Inc.,
Pawtucket, Rhode Island, 2007.
[5] ACI 318-08, Building Code Requirements for Structural
Concrete and Commentary.
[6] EC, Egyptian Code for the Design and Construction of
Concrete Structures (203-2006).
[7] BS EN1992-1-1: 2004, Eurocode 2: Design of concrete structures
– Part 1-1: General rules and rules for buildings.
[8] J.M. Saeed, Modeling of the Mechanical Behavior of Concrete,
M.Sc. Thesis, Structural Eng. Dept., Faculty of Engineering,
Alexandria University, 2012, p. 163.
[9] J. Lubliner, J. Oliver, S. Oller, E. Onate, A plastic-damage
model for concrete, Int. J. Solids Struct. 25 (3) (1989) 299–326.
[10] J. Lee, G.L. Fenves, A plastic-damage model for cyclic loading
of concrete structures, J. Eng. Mech., ASCE 124 (8) (1998) 892–
900.
[11] S. Hong, D. Kim, S. Kim, N. Hong, Shear strength of reinforced
concrete deep beams with end anchorage failure, ACI Struct. J.
99 (1) (2002) 12–22.