Corner Reinforcement in Reinforced Concrete Slabs: Mechanics, Importance, Estimation and Use
Corner Reinforcement in Reinforced Concrete Slabs: Mechanics, Importance, Estimation and Use
net/publication/324976359
CITATIONS READS
0 3,816
1 author:
Hisham Qasrawi
Hashemite University
39 PUBLICATIONS 1,115 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Hisham Qasrawi on 06 May 2018.
Hisham Y. Qasrawi1
ABSTRACT
The paper summarizes the mechanics, importance, estimation and use of corner
reinforcement in reinforced concrete slabs on linear supports. The paper discusses
both elastic and inelastic approaches for the estimation of the bending moments that
are exerted in the vicinity of the corners of slabs. Special simple plots that can directly
be used by professional engineers, for the estimation of corner moments are provided.
Recommendations regarding corner reinforcement given by some international codes
are presented, discussed and compared with the recommendations of some Arab
codes. Numerical examples are presented whenever necessary.
The results of the analysis of a survey showing the opinion of some professional
structural engineers, regarding the use of corner reinforcement in practical conditions
is presented. Useful recommendations are given at the end.
KEY WORDS
INTRODUCTION
The following paragraphs attempt to provide the professional engineer with the most
used methods for the evaluation of the corner levers and provide him with simple
charts for the evaluation of the corner effects using the least possible effort.
Discussion of the requirements of various codes and the opinion of some of the
1
Senior Lecturer, Civil Engineering Department, Faculty of Engineering, The Hashemite University,
Zarqa 13115,JORDAN.
experts in this field follows. Examples are also provided to show the possible effects
and the consequences of corner levers. A questionnaire was distributed among
structural professional engineers in order to arrive at the best possible conclusions.
Results of the analysis of the questionnaire are provided.
ELASTIC ANALYSIS
Elastic analysis of plates with small deformations and laterally loaded has been
extensively discussed by many authors. Timoshenko and Woinowsky-Kreiger (1)
provided full analysis and solution of the problem. In the solution of the problem, it
was concluded that when a rectangular plate is subjected to a transverse gravity load,
the corners in general have the tendency to rise and this is prevented by concentrated
reactions at the corners. These reactions are basically due to the twisting moments that
form on the edges. It has been shown that the reactions at each corner can be obtained
by solving the equation:
w
R 2 (M xy ) 2 D( 1-p)( ) (1)
xy
where:
Eh3
D
12(1 p2 )
R = Reaction at each corner
Mxy = twisting moment at the edges; per unit length
p = Poisson’s ratio
w = deflection of the plate
E = modulus of elasticity of the material of the plate
h = overall depth of plate
Simply supported rectangular slabs carrying uniformly distributed load has
been solved by M. Levy (1), and the reaction R can be calculated using the equation:
4(1 p)
R 2( M xy ) qa 2
1
[(1 m tanh m ) sinh m m cosh m ] (2)
3
m 1.3,5 ,... m cosh
3
m
where:
q = uniformly distributed load
a = shortest dimension of the slab
b = longest dimension of the slab
The previous equation can be written in the form R = n q a2 . The results for p =1/6 is
shown in Figure
Gould (2) presented an approximate analysis for uniformly distributed loading. In the
analysis, a single harmonic approximation of a uniformly distributed loading was
used. A value of qz = (16 / 2) q ,was used in the solution. The value of R can be
found according to the following equation:
16 a3b3
R 2(M xy ) 2[ 3 q 2 (1 p)] (3)
(a b 2 )
According to Gould (2), the force R in each corner in a square plate will be about 10%
of the total load. This finding is close to those results shown in Figure 1, which is
based on the equation developed by Levy. Once R is determined, the twisting moment
at corners can be obtained using the relationship Mxy = ½R.
From previous analysis it can be concluded that corner reinforcement is necessary in
order to resist the torsional moment and also and to prevent the corners of the slab
from lifting up.
Furthermore, if corners of a rectangular plate are not properly secured against lifting,
the clamping becomes ineffective and the bending moments in the center portion of
slabs increase accordingly. Maximum moments in slab must be, therefore, multiplied
by the approximate coefficient k shown in Figure 1, which is based on the equation
described by Marcus (4).
Tralhair et al (3) calculated an increase of 35% for a square plate, which is compatible
with the previous results.
However, in the case of a polygonal plate with simply supported edges, no single
reactive forces arise at a corner point provided that the angle between both adjacent
sides of the plate is other than a right angle (5).
The presence of the reactions at corners has been established in various types of
simply supported slabs irrespective of the type of the applied load (1). For example,
the R value is 0.122P for a square slab supporting a batch load in the middle; hence
the moment is 0.061P; where P is the value of the concentrated load.
On the other hand, no corner reactions are obtained if the transverse shear
deformation is taken into consideration (1). In view of the strongly reactive forces this
shear deformation obviously is no longer negligible, and the customary thin-plate
theory disregarding it completely must be replaced by a more exact theory. Such
analysis would lead to distribution of the reactive forces concentrated at the corners of
the plate. Instead, the forces are distributed over a small boundary adjacent to the
corner, yielding at the corner itself a finite pressure acting downward. Results of
Kromm, shown in reference 1, showed results for a simply supported plate and
compared results with those obtained by customary theory.
INELASTIC ANALYSIS
The yield line theory is usually used for inelastic analysis and design of reinforced
concrete slabs. Extensive analysis can be found in reference (6). Yield line analysis is
usually performed by assuming a certain failure pattern and then analyzing the slab
according to that particular assumed pattern. The usual assumed pattern is shown in
Figure 2a. Yield line analysis showed that a simply supported slab at a corner may
not follow the usual pattern shown in Figure 2a. Instead, if the corners are not
fastened down, they will rise off the supports as the slab is loaded and the diagonal
yield line will split into two branches to form a Y shape as shown in Figure 2b (7).
This forms an additional corner segment with yield line moments on only two faces
and with support only on the two points where the yield lines cross the boundary. This
condition is called a corner pivot, since the segment pivots about these two points.
If the corners are held down but are not specially reinforced, similar yield lines form,
but in addition a corner crack opens along the pivot line as shown in Figure 2c. If the
corner is specially reinforced for negative moment, this adds a yield line moment
(m’), acting simultaneously with the positive moment (m), across the boundary of the
triangular segment which increases its capacity and moves the junction of the Y
farther from the corner. With a large amount of reinforcement (large enough (m’), the
triangular segment fails to form and the simple diagonal yield line into the corner,
(Figure 2a) is correct without modification (7).
In continuous slabs, the corners have top steel which provides an (m’) for the corner
segment. Whether the Y (Figure 2b and 2c) pattern or the straight-line diagonal
(Figure 2a) yield line occurs depends upon the amount of this reinforcement.
Assuming symmetry, analysis of the corner segment gives:
xhqh
M ab m' m 0.5( 2 )0
3
wh 2
m m' ;
6
6(m m' )
hence h (5)
q
It is clear that the length (h) is independent on (x). Hence, if in a given slab, the
moment-load ratio establishes a distance (h) much less than the length of the diagonal
it lies, this corner pattern will control; that is it does not reduce the slab capacity. If
the calculated (h) is large enough to push Y intersection beyond the limits of its
particular diagonal, it means that the corner segment does not form. Intermediate
values of (h) indicate that the corner element controls but have smaller effect on the
ultimate load or moment strength of the slab (7).
For a simply supported square slab, the ultimate moment (m) would be qa2/ 24 if no
corner segment forms. This the true condition with the corner held down and the
moment (m = m’). With no moment (m’) or with corner free, the ultimate moment
would be qa2 / 22. Thus, in square panel, the maximum effect of the corner segments
is slightly less than 9%, increasing the required resisting moment or decreasing the
allowable load.
Moreover, the corner triangles might be replaced by a small fan (Figure 2d), which
might control slab strength. For example, analysis of a square slab fixed on four edges
and which resists the same moment as that of the positive moment, the unit moment
M = qa2/ 48. A lower-bound solution as proposed by Mansfield’s failure pattern (fan
failure pattern), gives M = qa2/ 42.88 (8,9). Hence, for a uniformly loaded square slab
with load intensity q and degree of fixity (i) on all sides, an upper bound solution can
be obtained in the form of: q a2 = M {24 (1+i)}. A better result is qa2 = M {21.75
(1+i)} (10) when using fan yield pattern.
Wang and Salmon (11), pointed that it might be proper to discount the results of a
regular yield analysis (straight lines to corners) for most rectangular slabs by 8% to
10% for reason of corner effects. Vazirani and Ratwani (12), suggested that instead of
performing the tedious calculations of corner levers, a design engineer can increase
mid-span moment by a percent of (10 + ½A), where (A) is the difference between 90
and the corner angle of the slab..
MacGreger (13), stated that the corner segments or corner levers which reduce the
uniform load capacity of a simply supported slab by up to 9% compared to analysis
that ignores them. In actual slabs, this reduction is offset by increase in capacity due
to strain hardening of the reinforcement and membrane action of slabs. Because of the
previous effect, Kong and Evans (14) stated that the “so-called” upper bound collapse
load obtained by yield line analysis tends in practice to be much lower than the actual
value. Thus the search for the worst yield-line pattern need not be carried out
exhaustively; and that for design purposes, trying a few simple and obvious patterns is
usually sufficient.
According to Reynolds and Steedman (15), tests and elastic analysis of slabs show
that the negative moments along the edges reduce to zero near the corners and
increase rapidly away from these points. In slabs that are fixed or continuous at their
edges, negative moments tend to form across the corners, in conjunction with pairs of
positive yield lines, result in the formation of additional triangle slab elements known
as corner levers, which reduces the overall strength of the slab. Reynolds and
Steedman (15) summarized the factors that affect the decrease of the overall strength
as shown Table (1).
Table (1): Factors Affecting Corner Levers and Overall Strength
Also, Reynolds and Steedman (15) pointed out that for a corner having an angle of not
less than 90 degrees, the reduction in strength is unlikely to exceed 8% to 10%. Such
cases can therefore be treated by neglecting corner-lever action, increasing the amount
of main reinforcement slightly, and providing top steel at the corners to resist
cracking. For acute-angled corners, the decrease in strength is more serious. For a
triangular slab ABC where no corner is less than 30 degrees, Johansen (6) suggested
dividing the calculated strength without corner lever action by approximation:
K = 0.25 (7.4 – sin A – sin B – sin C) (6)
Thus, for an equilateral triangle, K = 1.2.
a1 k1 M / q and a2 k2 M / q (7)
Finite element analysis may be used to analyze the slabs and obtain corner lever
effects. Although the method is quite suitable for analysis of complex systems, it is
still an elastic method of analysis and such methods have distinct disadvantages is that
they do not account for the elastic-plastic behavior of members at the ultimate limit
state (18); there is usually a great deal of ductility and hence a considerable capacity
to resist load after initial yielding.
REQUIREMENTS OF VARIOUS CODES:
JBC 5/93, Jordanian Code for Plain and Reinforced concrete, gives special
recommendations regarding corner reinforcement (21). The recommendations are
given in article 5/4/3 and they are identical to those of BS 8110.
The Swedish Code of Practice recommends that top steel should be provided at the
corners of simply supported slabs. The steel must resist a moment per unit width
equals qab / 40. The bars should be parallel to the diagonal of the slab (oriented at an
angle of 45 degrees in a square slab). The bars used must extend to a distance of two-
fifths the longer span in the diagonal direction and be uniformly spaced about the
diagonals in the vicinity of the corners to a distance of one-eighth the longer span
from each side (15).
The Egyptian (22) Code of practice describes the design of two-way slabs in article
6-2-2. The method assumes that the slab is monolithically cast with the supporting
beams, and that certain amount of the load is resisted by the beams as a twisting
moment. Despite this assumption, the code describes no corner levers and provides no
special reinforcement for the corners. However, in the discussion of the ECP, Hilal
(23) proposed using corner reinforcement typical to that of the ACI 318.
Unfortunately, since then, nothing changed in this respect.
The UBC (24) gives recommendations for corner reinforcement in slabs in article
1913.4.6. The recommendations are identical to those of the ACI 318.
The Indian Code for (ISI) gives special recommendations regarding corner
reinforcement. The recommendations are identical to those of BS 8110 (12,25).
The previous Soviet’s code (26) requires that additional reinforcement should be
provided at simple supports where corners tend to rise. The reinforcement consists of
a mat of steel in both directions. The amount of steel should be 0.6 of that used for
maximum moment at mid-span and should extend to a distance of one-sixth the longer
span.
N.B.: The percentage was calculated as the number of engineers who gave a specific
answer to a certain question to the total number of engineers who answered
that particular question.
In the view of the previous study, the following conclusions and recommendations
can be drawn out:
1. The use of the plots provided in this paper is quite useful for professional design
engineers. It eliminates the troublesome trial and error calculations needed for
estimation of corner levers.
2. Compared with elastic analysis, inelastic analysis is more suitable for evaluating
corner levers because corner levers would probably form at collapse stage.
However, elastic analysis can still be used as a guide for engineers who should be
aware that elastic methods may over- or under-estimate effects.
3. Corner reinforcement is needed at discontinuous corners only because usual top
steel available in slabs deals with corner levers at continuous edges.
4. When designing slabs, engineers can decide among three alternatives:
a. Estimate moment at corners and provide the corresponding reinforcement
using the plots provided in this paper.
b. Use directly the recommendations of the codes.
c. Increase mid-span positive moment and the corresponding reinforcement by a
certain ratio depending on long to short span ratio of the slab.
However, Alternative a would be more economical because only short bars in a
limited area are used. Alternative b would be susceptible in the view of comment
number 6. Alternative c may not be economical because it requires long
additional bars to be embedded in the middle of the slab for the whole span; also it
may result in an increase in the depth of the slab required for flexure. Moreover,
the use of alternative c may result in considerable cracking at corners unless
minimum reinforcement necessary to control such cracking is provided.
5. Corner reinforcement moment is higher when long to short span ratio of the slab is
lower, being maximum for square slabs.
6. Large variations exist in codes recommendations regarding corner reinforcement.
A simple comparison between ACI 318 and BS 8110, the well-known and most
used codes, shows this. The difference between the two codes increases by
increase of long to short span ratio.
7. Unless slabs are heavily loaded, moments at corners are small and can be
neglected. Thus slabs carrying small loads can be designed as usual (without
calculating corner moments) and then only minimum amount of steel necessary to
control cracking in the serviceability limit state is to be provided at corners.
8. Slabs having corner angles greater than 90 degrees need not be specially
reinforced.
9. Special attention should be paid for slabs having acute-angled corners. The easiest
method to analyze such slabs is the method suggested by Vazirani and Ratwani
(Increase mid-pan moment by [{10 + ½ A}]%). However, in addition, minimum
steel should also be used at the corners of slabs in order to control cracking.
10. Corner effects and the corresponding corner reinforcement must be included in
the coarse of reinforced concrete structures taught for undergraduates. It is clear
from the questionnaire that the majority of recently graduated engineers knew
nothing about it. Even among engineers with more than seven years experience,
about 10% still knew nothing about corner reinforcement.
References:
APPENDEX 1 : QUESTIONNAIRE
Occupation:
1. Professional (leading position in a firm) 2. Design Engineer
3. Academic with practical experience 4. Other (Mention………….)
d. Do you think that the code you used is sufficient for corner reinforcement?
1. Yes 2. No
b. What precautions do you take in order to prevent corner lifting and strength
reduction of simply supported slabs?
1. I add additional steel reinforcing bars in the middle of the slab.
2. I add special torsional provisions in the surrounding beams and at the
edges of the slab.
3. I do nothing because it is unimportant.
4. I only add special minimum reinforcement to control cracking in vicinity
of corners.
(4) Have you ever encountered any problem in slabs because of lack of corner
reinforcement?
1. Yes 2. No
APPENDIX 2: EQUATIONS
Basic Equations Presented By Jones and Wood
1 2
k1 sec [ 6(1 i ){ 1]} k1 ( sec )[ C 2 2 (1 i ) ]
2 3 2
C1 sin2
2
where : where :
C1 [4 3 cot2 ] 1 C 2 4 i 3 cot
2 2
Also :
k1
k2 k2 is as for fixed edges.
[cos cot sin ]
2 2
In both cases : cot (C1 1) tan
2
1.50 0.120
1.40 0.110
1.35 0.105
Magnifying Coefficient ( k )
Corner Reaction ( R )
1.30 0.100
1.25 0.095
1.20 0.090
1.15 0.085
1.10 0.080
1.05 0.075
1.00 0.070
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
Long to Short Span Ratio ( b/a)
Is Not Provided
Fig. (2a): Straight-Line-Diagonal Fig. (2b): Y-Branch
q
h
x
m
k1
1.40 k2 6.00
1.30 5.50
1.20 5.00
k1 1.10 4.50 k2
1.00 a1 4.00
a2
a1
-i
M
0.90 3.50
0.80 3.00
Corner Lever
0.70 2.50
0.80 3.60
a1
0.70 3.50
Corner Lever
0.60 3.40
0.50 3.30
k1 k2
0.40 3.20
0.30 3.10
0.20 3.00
k1
0.10 k2 2.90
0.00 2.80