Latest Trends on Engineering Mechanics, Structures, Engineering Geology

Finite Element Analysis of Reinforced Concrete Corbels

MEHDI REZAEI, S.A.OSMAN AND N.E. SHANMUGAM
Department of Civil & Structural Engineering,Faculty of Engineering & Built Environment
Universiti Kebangsaan Malaysia (UKM)
43600 UKM Bangi, Selangor, Darul Ehsan
Malaysia
Mehdi_mcr@yahoo.com

Abstract:- In view of the increasing need of housing these days, the authorities in Malaysia have become
interested to use efficient, faster and safer methods of building. One of such methods is by adopting
Industrialized Building System (IBS) started in 2002. It was found from the last earthquake in Turkey that most
collapse of precast buildings was caused by failure of connections. Therefore, an extensive research should be
carried out to improve precast concrete connections. A common simple precast concrete connection is corbel.
Reinforced concrete corbels are structural elements widely used in practice. The complex response of these
elements is described in design codes in a simplified manner. These formulations are not sufficient to show the
real behavior. Finite element models were carried out on 254 x 405 x 254-mm reinforced concrete corbels. Ratio
of primary and secondary reinforcement was varied. The results indicate that corbels with neither primary
reinforcement nor secondary reinforcement fail suddenly and catastrophically. It seems that the mode of all
failures was by diagonal splitting. Increase in ratio of primary and secondary reinforcement generally resulted in
enhancement both strength and ductility of corbels. This increase also enhances the ultimate shear load until the
ratio of primary and secondary reinforcement reach 0.35% and 0.30% respectively.

Key-Words: Precast Construction, Connection, Corbel.

1 Introduction transfer device. Secondary reinforcements are

In reinforced concrete constructions, lateral loads, normally used to improve their shear capacities and
such as wind and earthquake loads are mainly reduce the likelihood of sudden failure. However,
resisted by shear walls and connections. Failure of contribution of stirrups has been shown to be
precast constructions is mostly caused by variable when corbels are subjected to combined
connections, in which a corbel could be used. vertical and horizontal loads. Furthermore, most
Corbels are widely used in precast concrete corbels containing stirrups as a secondary
structures in view of the advantages such as; reinforcement fail in shear that displays no ductility.
improved production speed and lower construction Also, distress of corbels in the field has been
costs. During the past century, theories (Holnicki- attributed to poor detailing of reinforcement. Such
Szulc & Gierlinski 1997, Ali & White 2001) have detailing may generally include bending and
been proposed to describe corbel behavior, and anchorage of reinforcement and cover to
several experimental studies have been carried out to reinforcement. These difficulties may be
investigate the behavior of corbels from the practical accentuated by the use of stirrups (especially in
point of view. This paper aims to examine the load small-size corbels), as a more complex detailing
carrying capacity of reinforced concrete (RC) procedure is normally required. Additional
corbels with different ratios of primary and secondary reinforcements would improve the
secondary reinforcement. cracking resistance of concrete, and modify or
Yong and Blaguru (1981) conducted experimental contain the explosive nature of high-strength
studies using normal-strength concrete and concrete at failure.
concluded that the shear strength is a function
of: (1) shear span- to-depth ratio (2)
reinforcement ratio (3) concrete strength and (4) 2 Finite Element Modeling
the ratio of the horizontal to vertical components With the development of high-powered computers,
of the applied loads. It is widely assumed that together with state-of-the-art finite element (FE)
reinforced concrete corbels are principally shear software and user-friendly graphical interfaces,

ISSN: 1792-4294 448 ISBN: 978-960-474-203-5

 Latest Trends on Engineering Mechanics, Structures, Engineering Geology

three-dimensional (3-D) FE analysis has become a

popular choice to predict the behavior of structural
elements. Finite element software LUSAS version
14.1 has been used in this study. The mesh size of
30X30 mm was chosen based on convergence
studies carried out to determine the optimal mesh
that gives a relatively accurate solution and one that
takes low computational time. It has been found
that this mesh is capable of producing results close
to the actual behavior of corbel connection. In this
study steel was assumed to behave as an elastic-
perfectly plastic material in both tension and
compression. The idealized stress-strain curve used
in the numerical analysis is shown in Fig. 1. The
material properties of steel were specified using the
elastic and the metal plasticity with plastic options.
LUSAS requires input of the Young’s modulus, E,
Poisson’s ratio, υ, and yield stress of steel, σy. Fig 2 Stress-Strain Curve of Concrete from Damia

Seven specimens tested by Stephan (1996) were

considered in this study to verify the modeling by
σ
LUSAS. All specimens were modeled and analyzed
Tension
with LUSAS and the results presented in Figure 3.
σy The results for ultimate load obtained from LUSAS
are compared with the corresponding test results by
Stephan (1996). It is found from the figure that the
difference in ultimate load is less than 15 %. The
-εy difference could be attributed to the assumptions
ε that were used in modeling the specimens. Since
ε
some of the properties are not given in the paper,
reasonable assumptions were made to achieve close
result between the simulation and the test results.
The 45˚ line indicates the accuracy of the results
σy
Compression thus results from LUSAS and those from the
experiment would lie on the line if they match
exactly. Since it has been found that the LUSAS
model could predict the experimental results to an
acceptable accuracy it has been decided to use
Fig 1 Stress-Strain Curve of Steel LUSAS for further analyses.
1200
Solid elements are capable of predicting the
1000
Ultimate load (test)

nonlinear behavior concrete. The element

characteristic is able to describe elastic, isotropic, 800
plastic and multi-crack concrete behavior. The 600
Multi-crack concrete model is a plastic-damage-
400
contact model in which damage planes form
according to a principal stress criterion and then 200
develop as embedded rough contact planes. 0
Concrete is a quasi-brittle material and has different 0 200 400 600 800 1000 1200
behaviors in compression and tension. The tensile
strength of concrete is typically 8-15% of the Ultimate load (LUSAS)
compressive strength as shown below. Stress-strain
relationship for concrete is shown in Figure 2. Fig 3 Comparison between finite element results
and test results.

ISSN: 1792-4294 449 ISBN: 978-960-474-203-5

 Latest Trends on Engineering Mechanics, Structures, Engineering Geology

3 Details of Corbels Geometry and Table 1 Details of Corbels

Reinforcement
As it is shown in Figure 4, the column supporting Steel Area (mm2)
the corbel cantilevering on either side was 254 by
405 mm in cross section and 254 mm long. Column Designation
Secondary Primary
was reinforced with four 16-mm-diameter Reinforcement Reinforcement
longitudinal bars and 9-mm- diameter stirrups
spaced at 216 mm center to center as shown in the P-0 127.2 0.0
Figure 4. The load was gradually applied on either P-0.15 127.2 154.3
side of the corbels. The reinforcement details for the P-0.20 127.2 205.7
corbels are presented in Table 1. In Table 1, P P-0.30 127.2 308.6
represents the corbels that are associated with P-0.35 127.2 360.0
different primary reinforcement, and S represents P-0.40 127.2 411.5
the corbels associated with secondary P-0.45 127.2 462.9
reinforcement. For instance, P-0.15 represents a P-0.50 127.2 514.4
corbel in series P with 0.15 as the percentage of P-0.55 127.2 565.8
primary reinforcement (steel area of primary S-0 0.0 265.2
reinforcement over cross section).Two corbels
S-0.15 77.2 265.2
without either primary reinforcement or secondary
S-0.20 102.9 265.2
reinforcement were analyzed to study the behaviors.
In Series P, the area of primary reinforcement was S-0.30 154.3 265.2
varied from 0 mm2 to 565.8 mm2, and the area of S-0.40 205.7 265.2
secondary reinforcement was kept constant. On the S-0.45 231.5 265.2
contrary, in series S, the steel area of primary
reinforcement was kept constant, and the steel area
of secondary reinforcement was varied from 0 mm2
to 257.2 mm2. 4 Finite Element Analyse of the
Corbels
The finite element program LUSAS version 14.1
was employed to simulate the behavior of the
corbels. In all these models the boundary conditions
at the top and bottom of the column were assumed
fixed and the load was applied incrementally on a
bearing pad made of steel as shown in the Figure 4.
Bearing pad was used to prevent local crushing of
concrete. Except the ratio of primary and secondary
reinforcements, details of all the models were kept
the same. For each specimen, only a quarter of the
specimen was modeled in view of the symmetry, in
respect of geometry, loading and support conditions.
Typical finite element model is shown in the
Figure5. Analyses were carried out on each of the
models and the results presented in the form of
load-displacement plots as shown in Figures 6 and
7. Displacements plotted on the horizontal axis
correspond to those measured under the load.

Fig 4 Reinforcement Detailing

ISSN: 1792-4294 450 ISBN: 978-960-474-203-5

 Latest Trends on Engineering Mechanics, Structures, Engineering Geology

Table 2 Corbel Series P Test Results

Percentage Free End

Ultimate
Increase of Deflection
Designation Shear
Ultimate at Ultimate
Load(kN)
Load Load (mm)

P-0 272 0.0% 0.42

P-0.15 557 104.9% 0.45

P-0.20 667 145.2% 0.89

P-0.30 736 170.5% 0.94

P-0.35 829 204.7% 1.01

Fig 5 A Quarter of Meshed Corbel Model, using P-0.40 930 241.9% 0.91
LUSAS.
P-0.45 989 263.5% 1.09
5.1 Series P
Fig 6 shows load-deflection curves for Series P. P-0.50 970 256.5% 1.28
Addition of primary reinforcements to corbels
enhance the ductility and toughness. However, the P-0.55 1029 278.2% 1.66
degree of enhancement was more evident in corbels
reinforced with lower ratios of main bars. It can be
seen that the improvement of load carrying capacity 5.2 Series S
of corbel is not significant for reinforcement ratio The effect of addition of secondary reinforcements
0.3 and after. As predicted, corbel P-0 fails on the behavior of corbels, are summarized in Table
catastrophically in brittle manner. The results of 3. Load-deflection curves for corbel series S are
series P is tabulated in Table 2. The results show shown in the Figure 7 in all corbels, it seems that
that by increasing the percentage of primary the first cracks to appear were flexural cracks
reinforcement steel, the ductility increases, but are starting at or near the junction of the tension face of
mostly attributed to up to 0.45%. As expected, the corbel and face of the column. The result in
ultimate load carrying capacity of the corbels is Table 2 shows that the presence of additional
improved by increase in percentage of primary secondary reinforcement resulted in an increase in
reinforcement steel, even though it is mostly load-carrying capacity and ductility of corbel.
attributed for lower ratios of main reinforcement. Corbels S-0 with no secondary reinforcement, failed
in an explosive manner. The increase in load-
carrying capacity of corbel is significant until the
percentage of secondary reinforcement reaches to
0.3%. As expected, ultimate load-carrying capacity
of corbel is improved by increase in percentage of
secondary reinforcement steel, even if it is mostly
pronounced for lower ratios of reinforcement.
Increase in percentage of secondary reinforcement
would influence in the resistance of corbel to lateral
load.

Figure 6: Load Deflection Curve Series P

ISSN: 1792-4294 451 ISBN: 978-960-474-203-5

 Latest Trends on Engineering Mechanics, Structures, Engineering Geology

1000 improved by increase in percentage of primary

900 reinforcement steel, although it is mostly
pronounced for lower ratios of main reinforcement,
800
nevertheless, causes an increase of ductility.
700

600 3- The load-carrying capacities of corbels are

Load (kN)

considerably enhanced by the addition of secondary

500
S-0
reinforcements. The enhancement is noticeable until
400 the percentage of secondary reinforcement reached
S-0.15

300
to 0.3%.
S-0.20

200 S-0.30 4- As high strength concrete fails in brittle manner,

100 S-0.40 and when it comes to corbel, secondary
S-0.45 reinforcement would enhance the ductility of corbel.
0
0 0,5 1 1,5 2 2,5

Displacement (mm) References

[1] Stephen j. Foster, Rex E. Powell, and Hani S.
Fig 7 Load Deflection Curve Series S Selim. Performance of High-Strength Concrete
Corbels, ACI Structure Journal , Vol. 93, No.3,
1996, pp. 93-S52.
[2] Yook-Kong Yong, and P. Blaguru, Behavior of
Table 3.Corbel Series P Test Results Reinforced High-Strength- Concrete Corbels,
American Society of Civil Engineers (ASCE),
Vol.120, No.4, 1982, pp. 1182-1201.
Percentage [3] Alfred Strauss, Andrea Mordini, and Konrad
Free End Bergmeister. Nonlinear Finite Element Analysis of
Ultimate Increase
Deflection at Reinforced Concrete Corbels at Both Deterministic
Designation Shear of
Ultimate Load and Probabilistic Levels, Computers and Concrete,
Load(kN) Ultimate
(mm) Vol.3, No. 2/3, 2006, pp,123-144.
Load
[4] Nijad I. Fattuhi. Reinforced Corbel Made With
S-0 530 0.0% 0.58 High-Strength Concrete and Various Secondary
Reinforcement, ACI Structural Journal, Vol.91,
S-0.15 677 27.7% 0.83 No.3, 1994, pp.345-368.
[5] Mauricio Posada and Sharon L. Wood, Seismic
S-0.20 724 36.4% 0.85 Performance of Precast Industrial Buildings in
Turkey, 7th U.S. National Conference on
S-0.30 770 45.0% 0.95 Earthquake Engineering, Vol.2, No.57, 2002, pp.
42–59.
S-0.40 876 65.1% 0.99 [6] S.J. Foster, R.E. Powell, and H.S. Selim,
Performance of High Strength Corbels, ACI
S-0.45 886 67.0% 0.93 Structural Journal 93, 1996, pp. 555–563.
[7] Kriz, L. B. and Paths, C. H., Connections in
Precast Concrete Structures: Structures-strength of
Corbels, PCI Journal 10 ,1965, 16-60.
6 Conclusions [8] Mattock, A. H., Chen, K. C., and Soonswang,
The following conclusions can be drawn based on K. (1976). The Behavior of Reinforced Concrete
the results: Corbels. Prestressed Concr Journal. Inst., 21(3),
18-42.
1-The failure mode of corbels with neither
secondary reinforcement nor primary reinforcement
was brittle and explosive.

2-Ultimate load-carrying capacity of corbel is

ISSN: 1792-4294 452 ISBN: 978-960-474-203-5