Behavior of Beams With Dowel Action

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

Engineering Structures 29 (2007) 899903

www.elsevier.com/locate/engstruct
Behavior of beams with dowel action
Bilal El-Ariss

United Arab Emirates University, Civil and Environmental Engineering Department, P.O. Box 17555, Al Ain, United Arab Emirates
Received 17 February 2004; received in revised form 25 June 2006; accepted 7 July 2006
Available online 6 September 2006
Abstract
The phenomenon of dowel action as a shear transfer mechanism across cracks has long been recognized as an important component of
the overall shear resistance capacity of reinforced concrete beams. In this paper, a simple analytical model for the dowel action of reinforcing
bars crossing cracks is developed for analysis of reinforced concrete beams. This model is incorporated into a computer program that uses
the displacement method and the initial stiffness procedure. The nonlinear behaviors of several reinforced concrete beams tested by others are
analyzed. The beams are analyzed rst with the dowel action neglected and then with the dowel action considered. It is found that in certain cases,
the dowel action can have signicant effects on the shear strength of reinforced concrete beams and that the theoretical results of the proposed
model generally agree better with the experimental values when the dowel action is accounted for.
c 2007 Published by Elsevier Ltd
Keywords: Reinforced concrete beams; Dowel force; Elastic foundation
1. Introduction
Cracking in concrete beams may result in a signicant
reduction in their stiffness and strength. Although the exural
behavior of cracked reinforced concrete beams can generally
be well predicted, accurate prediction of the shear behavior of
reinforced concrete beams remains a formidable task due to the
complexity of the shear transfer mechanism in the reinforced
concrete. Shear resistance in reinforced concrete beams is
provided by the shear transfer in the compression zone,
aggregate interlock across the crack face, stirrups crossing the
shear crack, and dowel action of longitudinal reinforcing bars
crossing the crack in the concrete. The contributions of the
compression zone, the stirrups, and the aggregate interlock are
fairly well modeled in the literature, but so far the dowel action
of the reinforcing bars has not been explicitly represented,
despite the implicit belief in much of the current design
thinking that dowel action is an important component of the
shear resistance. The dowel action of reinforcing bars can play
an important role if other contributions to shear transfer are
relatively small as in the case of a beam with a small amount
of web reinforcement or the case of a post-peak stage of the

Tel.: +971 50 663 4601, +971 3 762 4601; fax: +971 3 762 3154.
E-mail address: [email protected].
loading process. It may contribute signicantly to the post-
peak resistance and hence contribute to the shear ductility of
concrete members.
From a literature survey on nite element analysis of
reinforced concrete structures covering papers published from
1985 to 1991 [1], it is noted that modeling of the dowel action
has not been mentioned in any of the papers surveyed. This
reects to some extent the difculties involved in modeling
the behavior of the dowel action. There are some major
difculties in modeling the dowel action of reinforcement
bars for nite element analysis. In experimental tests, the
shear force transferred by the dowel action is quite difcult
to measure because it is embedded with other shear transfer
components. In fact, since the dowel action involves interaction
between the reinforcement bars and the concrete near the
cracks and the interaction stresses are extremely difcult to
measure, many details of the dowel action have never been
investigated. Consequently, experimental results on the dowel
action have been rather limited. Even in nite element analysis,
the mechanism of the dowel action is too complicated to
describe. To analyze the details of the dowel action, the steel
bars need to be individually modeled by nite elements and a
very ne mesh has to be used for the concrete. As a result, the
number of elements required would be very large. Furthermore,
such individual modeling of the steel bars and concrete is not
0141-0296/$ - see front matter c 2007 Published by Elsevier Ltd
doi:10.1016/j.engstruct.2006.07.008
900 B. El-Ariss / Engineering Structures 29 (2007) 899903
compatible with the common practice of modeling the concrete
and the steel together in the analysis of reinforced concrete
structures. To consider the dowel action in the global analysis of
reinforced concrete structures, a simplied model of the dowel
action that is compatible with the crack and reinforcement
models is required. Thirdly, since the dowel action is usually
more signicant near peak load and at the post-peak stage,
experimental testing or theoretical analysis extending into the
post-peak range are needed to investigate the full effects of the
dowel action, but such testing and analysis are generally quite
difcult.
To incorporate the effects of the dowel action in the
nonlinear analysis of reinforced concrete beams, a simplied
analytical model for the dowel action is described in this
paper. The model is incorporated into a computer program
that employs the displacement method and the initial stiffness
procedure.
2. Analytical model
The analytical model used to predict the behavior of a dowel
bar embedded in concrete is based upon the work presented by
Timoshenko and Lessels [2] for the analysis of beams on an
elastic foundation. A beam on an elastic foundation is made
of discrete springs that connect a beam to a rigid base. The
dowel action behavior of the reinforcement bars crossing cracks
in the concrete is analyzed by treating each reinforcement bar
as a beam and the surrounding concrete as a bed of springs so
that the reaction force of the foundation at any point may be
assumed to be proportional to the deection of the beam at that
point. According to Timoshenko and Lessels, the differential
equation for the deection of a beam on elastic foundation is
written as follows:
EI
d
4
y
dx
4
= ky (1)
where:
k = stiffness of the elastic foundation (the concrete represents
the exible foundation);
y = deection.
The solution to this differential equation is given by:
y = e
x
(A cos x + B sin x)
+e
x
(C cos x + D sin x) (2)
=
4
_
k
4E
s
I
s
(3)
where:
E
s
= modulus of elasticity of the steel bars;
I
s
= moment of inertia of the bar (equal to
d
4
b
64
in which d
b
is
the diameter of the bar);
A, B, C, and D = constants determined from the boundary
conditions for a particular problem.
Cutting the reinforcing bar at the face of the crack, the bar
may be treated as a semi-innite beam resting on an elastic
foundation and subjected to concentrated dowel force V
d
and
Fig. 1. Semi-innite beam on an elastic foundation.
Fig. 2. Forces acting on the dowel bar.
moment M
o
applied at its end, as shown in Fig. 1. For a semi-
innite beam on an elastic foundation, the constants A and B
are equal to zero and Eq. (1) becomes:
y =
e
x
2
3
E
s
I
s
[V
d
cos x M
o
(cos x sin x)] . (4)
Differentiating Eq. (4) with respect to x gives the slope,
dy/dx:
dy
dx
=
e
x
2
2
E
s
I
s
[(2M
o
V
d
) cos x V
d
sin x] . (5)
Applying the solution for a semi-innite beam on an
elastic foundation to dowel bars crossing a joint in a concrete
pavement, Fribergs [3] developed equations for the slope and
deection of a dowel at the face of the joint are used to
determine the slope and deection of a dowel at the face of
a crack in concrete. In Fribergs analysis, the stiffness of the
elastic foundation, k, was replaced with the expression K
o
b.
The modulus of the dowel support, K
o
, denotes the reaction
when the deection is equal to unity and b represents the
dowel bar diameter. Fribergs developed equations were derived
assuming a dowel bar of semi-innite length. However, many
engineers view Fribergs work as the authoritative analysis
on the behavior of dowel bars to date. Therefore, Fribergs
equations were used in accomplishing the theoretical work
associated with this research.
Assuming that an inection point exists in the dowel at the
center of the crack, the forces acting on the portion of the dowel
within the crack width, z, are shown in Fig. 2. Substituting
(
V
d
z
2
) for M
o
and setting x equal to zero, Eqs. (4) and (5)
become Eqs. (6) and (7) for the slope and deection of the
dowel at the face of a crack in concrete, as shown in Fig. 3:
dy
o
dx
=
V
d
2
2
E
s
I
s
(1 +z) (6)
y
o
=
V
d
4
3
E
s
I
s
(2 +z) . (7)
The stiffness of the elastic foundation (concrete surrounding
the dowel bars) is an important parameter in the equations
B. El-Ariss / Engineering Structures 29 (2007) 899903 901
Fig. 3. Slope and deection of dowel at the face of the crack.
presented in this paper. Before these equations can be used,
a value for the elastic foundation stiffness is needed. For the
elastic foundation stiffness of the surrounding concrete, k,
the following data-tting expression proposed by Soroushian
et al. [4] is used:
k =
127c
1
_
f

c
d
2/3
b
(8)
where:
f

c
= compressive strength of the concrete in N/mm
2
;
d
b
= diameter of the bar in mm;
c
1
= coefcient ranging from 0.6 for a clear bar spacing of
25 mm to 1.0 for larger bar spacing.
The loaddeection response for dowel bars embedded in
concrete proposed by Millard and Johnson [5] is adopted in this
research. Although it has been suggested that dowel strength
across a shear plane is owing to a combination of direct shear,
kinking and exure of the reinforcing bars, Millard and Johnson
have illustrated that exure of the bars predominates, since
there is a signicant amount of deformation in the underlying
concrete cover. They proposed the following loaddeection
response for dowel bars embedded in concrete:
V
d
= V
u
_
1 exp
_
k
V
u
__
(9)
where
V
d
= dowel force at a shear displacement at a crack;
V
u
= the ultimate dowel force.
When the dowel deformation is not too large and none of the
materials have yielded, the dowel forcedisplacement relation
is linearly elastic. However, when the elastic limit is exceeded,
the dowel action becomes plastic. At the ultimate limit state,
local crushing of the surrounding concrete and/or yielding of
the dowel bar occurs. Based on experimental results, Dulacska
[6] has given the following equation for estimating the dowel
force at ultimate limit state V
u
:
V
u
= 1.27d
2
b
_
_
f

c
_ _
f
y
_
(10)
where:
f
y
= yielding strength of the dowel bar.
The dowel displacement used in Eq. (9) can be assumed
equal to the distance from the inection point in the dowel at
the center of the crack to the level of the bar in the concrete;
therefore:
= 2
_
y
o
+
_
dy
o
dx
_
_
z
2
_
_
=
V
d
2
2
E
s
I
s
_
(2 +z)

+(1 +z) (z)


_
. (11)
Substituting Eqs. (10) and (11) in Eq. (9) will yield the dowel
force of the reinforcing bar:
V
d
= 1.27d
2
b
_
_
f

c
_ _
f
y
_

_
1 exp
_
_
_
kV
d
_
(2+z)

+(1 +z) (z)


_
2.54
2
E
s
I
s
d
2
b
_
_
f

c
_ _
f
y
_
u
_

_
_

_
. (12)
The forces acting on the portion of the dowel within the
crack width, z, are as shown in Fig. 2 where M
o
is equal to
(
V
d
z
2
). The moment produced by the dowel force, V
d
, will
tend to reduce the moment applied at the section where the
crack intersects the reinforcing bar. The resultant moment is
used in the standard section analysis to compute the strain
and curvature. In the standard section analysis, the resultant
moment is applied at its corresponding beam cross section.
Due to the application of the moment a change in the strains
and stresses will occur at the section. Two parameters, strain
and stress, are used to dene the strain and stress distributions.
These two parameters are then obtained from the equilibrium
requirements. The analysis is repeated for a number of sections.
An arbitrary number of sections along the beam is chosen and
incorporated in the computer program to perform the analysis.
When the dowel action of the reinforcing bars is not considered,
the standard section analysis is performed using the moment
applied at the section and not the resultant moment above.
It is worth noting that the section analysis employed in the
computer program along with the displacement method and the
initial stiffness procedure has an advantage over the standard
nite element method. The essential feature of the analysis is
that the actual deected shape is obtained by integrating the
actual strains and curvatures. In the nite element method, the
deected shape of a member is usually assumed as a function
of the displacements at the nodes and equilibrium between the
external and internal forces is satised only at the nodes.
3. Verication of the analytical model
To verify the reliability of the proposed analytical model, a
comparison with experimental and analytical work conducted
by other researchers is carried out. The deep beams tested by
Ashour [7] are analyzed and the analytical results are compared
to the experimental results. The beams are analyzed twice, rst
with the dowel action neglected and then again with the dowel
action incorporated, in order to study the signicance of the
dowel action of the main reinforcement bars contained in these
beams. Beams, CDB1, CDB2 and CDB3, are selected for the
analysis. The details of the beams are shown in Fig. 4. The
top and bottom longitudinal reinforcement bars have yielding
902 B. El-Ariss / Engineering Structures 29 (2007) 899903
(a) Beam CDB1.
(b) Beam CDB2.
(c) Beam CDB3.
Fig. 4. Details of the beams analyzed (dimensions in mm).
strengths of 500 MPa and 400 MPa, respectively. The web
reinforcement was 8 mm diameter steel bars with yielding
strength of 370 MPa. The compressive strengths of the concrete
for the beams CDB1, CDB2 and CDB3 were 30.0 MPa,
33.1 MPa and 22.0 MPa, respectively.
The loaddeection curves are plotted in Fig. 5. It is
seen that when the dowel action is not taken into account,
the predicted strengths are lower than the corresponding
experimental values. However, when the dowel action is taken
into account, the loaddeection responses of the beams are
in better agreement with the test results. This reveals that the
contribution of the dowel action has a signicant effect on
the behavior of the beams. The effect of the dowel action
becomes evident when the applied load approaches the peak.
Beyond the peak, the effect of the dowel action is even more
signicant especially when the aggregate interlock action along
the cracks drops due to gradual increase of crack widths. The
importance of the dowel action increases as the amount of
web reinforcement decreases. In beam CDB1, almost all of
the vertical shears are resisted by the web reinforcement and
thus the contribution of the dowel action is relatively small. In
beam CDB3, there is no web reinforcement and consequently
the dowel action plays a more important role in resisting the
applied shear force, which agrees with He and Kwan [8].
They modeled the dowel action of reinforcement bars for nite
element analysis of concrete structures and showed that the
dowel action can have signicant effects on the shear strength
and ductility of reinforced concrete beams.
Fig. 5. Loaddeection curves (DA is dowel action).
4. Conclusions
A simple analytical model for the dowel action of
reinforcing bars crossing cracks in concrete is developed and
incorporated into a computer programfor the nonlinear analysis
of reinforced concrete beams. The behavior of the dowel bar
is derived based on the beam on an elastic foundation theory.
Application of the dowel action model to the analysis of
deep reinforced concrete beams tested by others veried that
the proposed dowel action model can be used to predict the
behavior of shear critical reinforced concrete members. The
analytical results also showed that the dowel action could
have signicant effects on the behavior and ductility of the
reinforced concrete beams especially when the amount of web
reinforcement in the beam is small. Therefore, in the nonlinear
analysis of shear critical reinforced concrete members, the
dowel action should be taken into account.
It is recommended to carry out a parametric study to
investigate the inuence of some parameters such as bar
diameters and amount of reinforcement on the dowel action
as a shear transfer mechanism across cracks. Comparison with
other experimental results would provide more complete and
satisfactory results.
B. El-Ariss / Engineering Structures 29 (2007) 899903 903
Further research on the bearing capacity of dowel action
in concrete structures reinforced with ber reinforced plastic
(FRP) bars would provide some insight on the importance of
the dowel action.
References
[1] Darwin D, editor. Finite element analysis of reinforced concrete structures
II. New York: ASCE; 1993. p. 20332.
[2] Timoshenko S, Lessels JM. Applied elasticity. Pennsylvania: Westinghouse
Technical Night School Press; 1925.
[3] Friberg BF. Design of dowels in transverse joints of concrete pavements.
Transactions, American Society of Civil Engineers 1940;105(2081).
[4] Soroushian P, Obaseki K, Rajos MC. Bearing strength and stiffness of
concrete under reinforcing bars. ACI Materials Journal 1987;84(3):17984.
[5] Millard SG, Johnson RP. Shear transfer across cracks in reinforced
concrete due to aggregate interlock and dowel action. Magazine of
Concrete Research 1984;36(126):921.
[6] Dulascka H. Dowel action of reinforcement crossing cracks in concrete.
ACI Structural Journal 1972;69(12):7547.
[7] Ashour AF. Tests of reinforced concrete continuous deep beams. ACI
Structural Journal 1997;94(1):312.
[8] He XG, Kwan AKJ. Modeling dowel action of reinforcement bars for nite
element analysis of concrete structures. Computer & Structures 2001;79:
595604.

You might also like