ASIAN JOURNAL OF CIVIL ENGINEERING (BUILDING AND HOUSING) VOL. 6, NOS.

1-2 (2005)
PAGES 21-33

DESIGN OF A DOUBLE CORBEL USING THE STRUT-AND-TIE

METHOD
Bhupinder Singh1, Yaghoub Mohammadi2 and S.K. Kaushik3
Department of Civil Engineering, National Institute of Technology,
Jalandhar-144 011, India,
2
Department of Civil Engineering, University of Mohaghegh Ardebili, Ardebil, Iran,
3
Department of Civil Engineering, IIT Roorkee, Roorkee-247 667, India

ABSTRACT

ch

ive

of

SI

The strut-and-tie method can be used for the design of regions of structures where the basic
assumptions of flexure theory, namely plane sections remaining plane before and after
bending, are not applicable. Such regions occur near force discontinuities arising from
concentrated forces or reactions and near geometric discontinuities such as abrupt changes
in cross section etc. The strut-and-tie method of design is based on the assumption that
appropriate regions in concrete structures can be analysed and designed using hypothetical
pin-jointed trusses consisting of struts and ties connected at nodes. Although IS 456:2000
recommends the strut-and-tie method for design of corbels, no guidelines are given for
determination of concrete strut and node dimensions and for the allowable stresses for
design. The ACI Code 318-02 on the other hand, gives explicit and practical
recommendations related to the design of structural members using the strut-and-tie method.
To amplify the application of these recommendations to the design of structural concrete a
complete example on the analysis and design of a double corbel using the strut-and-tie
method has been presented.

Ar

Keywords: corbel, strut, tie, node, truss, detailing

LIST OF NOTATIONS

Pv = Design Vertical Load on Corbel

Ph = Design Horizontal Load on Corbel
fck = Characteristic Cube Compressive Strength of Concrete
fc = Specified Cylinder Compressive Strength of Concrete
fy = Specified Yield Strength of Reinforcement
fcu = Effective Compressive Strength of Concrete in a Strut or a Nodal Zone
n = Factor whose values are specified in Clauses A.5.2.1 through A.5.2.3 of ACI 318-02

Email-address of the corresponding author: [email protected]

www.SID.ir

22

Bhupinder Singh, Yaghoub Mohammadi and S.K. Kaushik

SI

= Strength Reduction Factor

a = Shear Span, Distance between Concentrated Load and Face of Support
d = Distance from Extreme Compression Fibre to Centroid of Longitudinal Tension Reinforcement
D = Overall Depth of Corbel at Column face
h = Overall Thickness of Member
Ws = Width of Strut
b = Out-of-plane Dimension of Corbel
Ast = Area of Tension Reinforcement
Astmin. = Required Minimum Area of Tension Reinforcement
s = Factor whose values are specified in Clauses A.3.2.1 through A.3.2.4 of ACI 318-02
AH = Area of Shear Reinforcement Parallel to Flexural Tension Reinforcement
An = Area of Reinforcement in Bracket or Corbel Resisting Tensile Force Nuc
Nuc= Tension Force Applied at Top of Bracket or Corbel
Asi = Area of Surface Reinforcement in the ith layer crossing a strut
Si = Spacing of Reinforcement in the ith layer Adjacent to the Surface of the Member
i = Angle Between the Axis of a Strut and the Bars in the ith layer of Reinforcement
Crossing the Stru

INTRODUCTION

ch

ive

of

The corbel geometry and the loads for which the corbel has to be designed are shown in Figure
1. The principal dimensions of the corbel may not be known a priori and the fixing of such
dimensions is explained later on. The structure under consideration is a region of discontinuity
since it encompasses abrupt changes in geometry and is in the vicinity of concentrated forces.
The conventional flexure theory is not applicable to regions of discontinuities and hence the
application of the strut-and-tie method for the design of the corbel is justified. The structure
will be designed as per recommendations of Appendix-A of ACI 318-02, Ref. [1] (here-inafter referred to as the Code) and other applicable clauses of the same Code.

LOADS AND MATERIAL PROPERTIES

Ar

Design vertical load on each corbel= Pv = 325 kN (Figure 1)

Design horizontal load on each corbel = Ph = 75 kN (Figure 1)
Design axial load on column = 950 kN (Figure 1)
Characteristic cube compressive strength of concrete (assumed) = fck = 25 MPa
Take cylinder compressive strength of concrete = fc = 0.80 fck = 20 MPa
Yield strength (0.2% proof stress) of reinforcement bars = fy = 415 Mpa

DESIGN OF BEARING PLATES

The sizes of the bearing plates located at the loading points have to be determined. The

www.SID.ir

DESIGN OF A DOUBLE CORBEL USING THE STRUT-AND-TIE METHOD...

23

of

SI

bearing plates at the load locations, as will be seen later on, are resting on the underlying
CCT (Compression-Compression-Tension) nodes of the strut-and-tie model. The applied
load on the bearing plates exerts a bearing stress on a face of the hypothetical nodal zone,
which is assumed to enclose the CCT node. This bearing stress should be within Code
specified limits for CCT nodes.

Figure 1. Loads on the double corbel

ive

Assume the size of the bearing plates as 300 x 200 mm, the longer dimension being along
the plane of the corbel.
Hence, the bearing stress in concrete at loading points =

325 103
300 200

= 5.42 MPa

Ar

ch

As per Clause A.5.2 eq. A-8 [1], the effective compressive stress at the face of a
node=fcu=0.85n fc .
The values of the parameter n is specified in Clauses A.5.2.1, A.5.2.2 and A.5.2.3 [1].
For the CCT node which anchors one tie (the T in CCT), Clause A.5.2.2 [1] is applicable
wherein n = 0.80.
Therefore, fcu= 0.85 x 0.80 x 20 = 13.60 MPa
The allowable bearing stress is equal to fcu where is the strength reduction factor [1].
As per Clause 9.3.2.6 [1], for strut-and-tie models, = 0.75
Hence, fcu = 0.75 x 13.60 =10.20 > 5.42 MPa, ok.
Hence, the assumed size of the bearing plates is adequate.

SELECTION OF CORBEL DIMENSIONS

As per Clause 11.9.1 of Ref. [1], for application of the recommendations related to the strutand-tie method given in Appendix-A of Ref. [1] for the design of corbels, the span-to-depth

www.SID.ir

Bhupinder Singh, Yaghoub Mohammadi and S.K. Kaushik

24

(Figure 2), should be less than 2. Further, Clause 11.9.2 of Ref. [1]
d
requires the corbel depth beyond the bearing area to be at least 0.5 d.

SI

ratio of the corbel,

Figure 2. Structural action of a corbel

Ar

ch

ive

of

Select overall depth of corbel at column face, D, equal to 500 mm. Select a depth of 250
mm at the free end of the corbel.
Allow a clearance of 100 mm from the outer edge of the bearing plate to the free edge of
the corbel. The selected dimensions of the double corbel are presented in Figure 3.

Figure 3. Selected dimensions of the double corbel

www.SID.ir

DESIGN OF A DOUBLE CORBEL USING THE STRUT-AND-TIE METHOD...

25

SELECTION OF STRUT-AND-TIE MODEL

Ar

ch

ive

of

SI

Multiple strut-and-tie models can be developed for a single load case. Some options for the
double corbel being designed are shown in Figure 4. The moot question at this stage is how
to select the optimal model. Usually, that model is the best in which the loads follow the
path with the least force and the least deformation [2]. At the same time, since ties are more
deformable than concrete struts, the model with the least number and the shortest ties is
likely the best. This requirement can be quantified as FIlimi = minimum, where Fi is the
force in the strut or tie, li is the length of the member i and mi is the mean strain in member i
Ref. [2]. In addition to the above requirements, the selected strut-and-tie model should be
such that the angle between the axes of the struts and ties acting at a node should be as large
as possible so as to mitigate possible cracking and to avoid incompatibilities due to
shortening of the struts and lengthening of the ties occurring in almost the same directions.
The ACI Code of Ref. [1], recommends that the angle between the axes of any strut and tie
entering a single node shall not be less than 25.
In light of the above recommendations, the strut-and-tie model of Figure 4 (a) is selected
for modeling the double corbel.

Figure 4. Strut and tie models for a double corbel

www.SID.ir

26

Bhupinder Singh, Yaghoub Mohammadi and S.K. Kaushik

DETERMINATION OF TRUSS FORCES

ive

of

SI

The forces in the members of the strut-and-tie model are determined from conditions of
static equilibrium. The location and orientation of the struts and ties is defined by the
position of the nodes. The selected strut-and-tie model for the double corbel is shown in
Figure 5. The center of the top horizontal tie BC (Figure 5), is assumed to be at a distance of
75 mm from the top of the corbel. Since the tie capacity is assumed to be furnished
exclusively by the reinforcement bars, the assumed cover takes care of requirements for
clear cover to the tie reinforcement bars and for the provision, if required, for two layers of
reinforcement bars for the tie BC. Like wise, all ties are assumed to have an effective cover
of 75 mm.

ch

Figure 5. Selected strut and tie model for the double corbel

Ar

Hence, d = 500 75 = 425 mm.

a 200
= 0.47 < 2, ok..
Therefore, =
d 425
The strut AD is assumed to lie on an imaginary horizontal line running through the point
of intersection of the sloping face of the corbel and the vertical face of the column.
As shown in Figure 5, the column axial load of 950 kN is resolved into two equal loads
of 475 kN each acting in line with strut AG and DH respectively.
The position of the centerline of the strut AE can be found by calculating the width w s
of strut AE (Figure 5).
Let FAE be the required compressive force in strut AE. As per Clause A.3.1 of Ref. [1],
the nominal compressive strength of a strut = Fns = fcuAc , where AC is the cross-sectional

www.SID.ir

DESIGN OF A DOUBLE CORBEL USING THE STRUT-AND-TIE METHOD...

27

area at one end of the strut and fcu is the effective compressive strength of concrete in the
strut.
As per Clause A.3.2 eq. A-3 [1], fcu = 0.85s fc .
Assuming strut AE to have a uniform cross-sectional area (prismatic strut), the value of
s as per Clause A.3.2.1 of Ref. [1], is equal to 1.0.
AC = wsb, where b is the out-of-plane dimension of the corbel and is equal to 350 mm in
this case.
Hence, FAE = 0.75 x 0.85 x 1.0 x 20 x 350 x w s
--------------(i)
By inspection of the conditions of loading symmetry in Figure 5, it can be easily
determined from conditions of equilibrium that the force in the strut
950
+ 325 =800 kN
AE=FAE=
2
Substituting for FAE in (i) and solving, gives w s = 179.27, say 180 mm.
Hence, the centerline of the strut AE and the center of node A are located at

180
2

= 90

Ar

ch

ive

of

SI

mm from the vertical face of the column.

Therefore, the geometry of the strut-and-tie model is fixed.
The forces in all the members of the strut-and-tie model are determined from conditions of
static equilibrium and are summarized in Table 1 and also illustrated in Figure 6.

Figure 6. Computed forces in the members of the strut-and-tie model

www.SID.ir

Bhupinder Singh, Yaghoub Mohammadi and S.K. Kaushik

28

Table 1. Computed truss forces

Member

AB = CD

BC

AD

AE = DF

AG = DH

Force (kN)

+386.77

- 284.68

+221.84

+800

+475

+Compression, -Tension

DESIGN OF TIES

SI

The capacity for tie BC is furnished by steel reinforcement and concrete is not assumed to
carry any tensile loads.
F
The area of reinforcement required for tie BC is equal to Ast = BC , where FBC is the
Y
tensile force in tie BC and y is the permissible tensile stress in the steel reinforcement and
is equal to fy f y where is the strength reduction factor for the reinforcement yield
stress fy f y and is taken as 0.75 as per recommendations of Clause 9.3.2.6 of Ref. [1].

284.68 10 3
= 914.63 mm2
0.75 415
per Clause 11.9.5 [1], the required minimum area of tensile reinforcement =
'
fc
bd = 0.04 20 350 425 = 286.74 mm2 < 914.63 mm2, ok.
f y
415

of

Therefore, Ast =

ch

ive

As

0.04

Clause A.4.2 of Ref. [1] requires tie reinforcement to be distributed approximately

uniformly over the width of the tie. This may entail furnishing the tie reinforcement in
several layers rather than concentrating the rebars in a section normal to the plane of the
member being designed, as is usually done in the case of beams.
Provide 6 bars of 16 mm diameter in two layers of 3 bars each as tie reinforcement.
Area of steel provided = 6 x 201 = 1206 > 914.63 mm2, ok.

Ar

CHECK ON NODAL ZONES AND ANCHORAGES

The nodal zones of interest are at nodes A, B, C and D. The dimensions of the nodal zones
have to be such that the stresses acting on the faces of the nodal zones are within permissible
limits.
The width w s of strut AE was so determined that the stresses in the strut are within
permissible limits. Strut AE bears on one face of nodal zone A and Appendix-1 of Ref. [1]
suggests that the faces of nodal zones loaded in compression shall have the same widths as
that of the ends of the struts bearing on them. In consequence, the stresses in nodal zone A
(and D) are deemed to be within permissible limits.

www.SID.ir

DESIGN OF A DOUBLE CORBEL USING THE STRUT-AND-TIE METHOD...

29

of

SI

Therefore, only nodal zones B (and C) need to be checked. Nodal zone B encompasses a
CCT node. If any of the forces acting on a node is tensile, the required minimum width of a
side of the nodal zone containing such a node is calculated from the width of a hypothetical
bearing plate anchoring one end of the tie which is assumed to exert a uniform bearing
pressure on the back side of the nodal zone, Figure 7. The width of the hypothetical bearing
plate in turn is equal to the width of the tie anchored in the node.

ive

Figure 7. Tie width in a CCT node

The permissible bearing pressure for a CCT node has already been computed as = f cu =
10.20 MPa

ch

Therefore, the required width of the tie to be anchored in the CCT node =

FBC
f cu b

Ar

284.68 103
= 79.74 mm
10.20 350
Available tie width = 75 x 2 = 150 > 79.74 mm, ok.
To provide positive anchorage to tie BC, weld the six 16 mm bars to a steel angle ISA
150 x 150 x12 located at the free end of the corbel, as shown in Figure 8.
CHECK ON STRUTS
The check on struts involves determination of strut widths required to shoulder the
computed strut forces and to determine whether the required strut widths will fit in the
geometry of the structure.

www.SID.ir

Bhupinder Singh, Yaghoub Mohammadi and S.K. Kaushik

SI

30

Figure 8. Strut and tie dimensions

ive

of

Consider strut AD
The effective compressive strength of strut AD = fcu, where fcu = 0.85sfc.
Strut AD is expected to have a uniform cross-sectional area throughout its length
(prismatic strut). For such struts, as per Clause A.3.2.1 [1], the parameter s =1.0.
F
221.84 103
= 49.71 mm.
Therefore, required width of strut AD = AD =
f cu b 0.75 0.85 1 20 350

ch

Select a width of 100 mm for strut AD. It can be verified that this width can be
accommodated in the geometry of the structure, Figure 8.

Ar

Consider struts AE and DF

The required widths of struts AE and DF have already been calculated as 180 mm each and
it can be verified that this width can be accommodated in the geometry of the structure,
Figure 8.
Consider struts AG and DH
The effective compressive strength of strut AG = fcu, where fcu = 0.85sfc .
Strut AG is expected to be a bottle-shaped strut as the width of the compressed concrete
in the strut at its mid-length can spread laterally. If no transverse crack control reinforcement
to resist the bursting forces is provided in this strut then as per recommendations of Clause
A.3.2.2 [1], s = 0.60.
F
475 103
Therefore, required width of struts AG (and DH)= AG =
f cu b 0.75 0.85 0.60 20 350

www.SID.ir

DESIGN OF A DOUBLE CORBEL USING THE STRUT-AND-TIE METHOD...

31

= 177.40 mm.
Select a width of 180 mm for each of the struts AG and DH such that their widths are
also equal to the widths of the underlying struts AE and DF respectively. It can be verified
that these widths can be accommodated in the geometry of the structure, Figure 8.
Consider struts AB and CD
The effective compressive strength of struts AB and CD = fcu, where fcu = 0.85sfc .
Struts AB and CD are non-prismatic and assuming that the required transverse crack
control reinforcement is provided in them, the parameter s as per recommendations of
Clause A.3.2.2 [1] can be taken equal to 0.75.
F
386.77 103
Therefore, required width of strut AB (and CD) = AB =
f cu b 0.75 0.85 0.75 20 350

of

SI

= 115.56 mm.
The above required strut width is compared with the available strut width at the smaller
end of the strut AB. A perusal of the geometry of strut AB in Figure 8 reveals the width of
the smaller end of the strut to be 147 mm> 115.56 mm. Hence, the available strut width is
adequate.
All the strut widths fit within the geometry of the corbel and thus the proposed strut-andtie model is acceptable.

CRACK CONTROL REINFORCEMENT

Ar

ch

ive

For adequate crack control in the vicinity of a tie, Clause 11.9.4 of Ref. [1] requires the
provision of reinforcement in the form of horizontal stirrups, the plane of these stirrups
being parallel to the reinforcement in the tie. The Clause further specifies this crack control
2
2
reinforcement to be distributed uniformly within of the effective depth ( 425 = 283.3 ,
3
3
say 280 mm) of the corbel adjacent to the tie BC.
As per Clause 11.9.4 of Ref. [1], the area of these horizontal stirrups must not be less
than Ah = 0.5 (Ast An), where An is the reinforcement area required to resist the horizontal
load at the bearings and Ast is the reinforcement provided in the tie BC.

75 10 3
= 496.69 mm2
Therefore, Ah = 0.5 1206
0
.
85

415

Provide 4 numbers of 10 mm diameter two legged horizontal stirrups within 280 mm

280
from the top edge of the corbel at an average spacing of
= 93.33 mm, say 90 mm c/c.
3
Area of stirrups provided = 628.31 > 496.69 mm2, ok.
Since s has been assumed as 0.75 while computing the strength of struts AB and CD,
confining reinforcement is required for these struts to take care of the transverse tensile
stresses. According to Clause A.3.3.1 of Ref. [1], for concrete strengths less than 42 MPa,
the requirement of confining reinforcement shall be deemed to be satisfied if the axis of the

www.SID.ir

32

Bhupinder Singh, Yaghoub Mohammadi and S.K. Kaushik

strut under consideration is crossed by reinforcement that satisfies the following

relationship:

Asi

bs

sin i 0.0030

where Asi is the total area of reinforcement at a spacing s i making an angle of i to the

axis of the strut, i not being less than 40.

For this double corbel, the crack control reinforcement provided above crosses the axis of
the struts AB and CD at an angle of 55 (Figure 8).
A
628.31
sin 55 = 0.016 > 0.003, ok.
Hence, si sin i =
350 90
bs i

Ar

ch

ive

of

SI

Therefore, the provided reinforcement serves the purpose of crack control as well as
confinement.
Also provide three numbers of 12 mm diameter framing bars in the double corbel.
The detailing of reinforcement in the double corbel is presented in Figure 9.

Figure 9. Detailing of reinforcement in the double corbel

Acknowledgement: Generous assistance has been derived from ACI SP-208 of Ref. [3]
while preparing this manuscript which is gratefully acknowledged

www.SID.ir

DESIGN OF A DOUBLE CORBEL USING THE STRUT-AND-TIE METHOD...

33

REFERENCES

D
SI
of
ive

3.

ch

2.

ACI 318-02, Building Code Requirements for Structural Concrete (ACI 318-02) and
Commentary (ACI 318R-02) , American Concrete Institute, Michigan, USA, 2002,
443 pp.
Schlaich, J., Schafer, K. and Jennewin, M., Toward a Consistent Design of Structural
Concrete, PCI Journal, May-June 1987, pp. 75-146.
SP-208, Examples for the Design of Structural Concrete with Strut-and-Tie Models,
American Concrete Institute, Michigan, USA, 2003, 242 pp.

Ar

1.

www.SID.ir