UDC 624.012.45: 624.046.

The provision of tension and compression

reinforcement toresist in-plane forces
L. A. Clark, BEng, PhD,MICE,MIStructE

C E M E N TA N DC O N C R E T EA S S O C I A T I O N : R E S E A R C HA N DD E V E L O P M E N T DIVISION

SYNOPSIS Introduction
This paper presentsequations f o r proportioning skew In the recent past, considerable attention has been
or ortllogonal tension andlor compression reinforcement given to theapplication of plasticity theory to the
to resista triadof in-plane forces.Theproblem of problems of proportioning reinforcement to resist
obtaining a ductile failure is discussed, and limits to the bending and twisting moments and to resist in-plane
applicability of mild and higlwtrength steel as com- direct forces and shears.
pression reinforcementare suggested. It isalso indicated With regard to theproblem of in-plane forces,
how the design equations might be used to proportion Nielsen"tZ)has presented the yield criterion for a
reinforcement to resist combinations of moments and section having known orthogonal isotropic or ortho-
in-plane forces. tropic reinforcement which can carry either tension or
compression and is symmetrically placed with respect
tothe middlesurface of thesection. Nielsen also
derived the equations for the determination of ortho-
reinforcement areas per unit length gonal tension reinforcement to resist a particular in-
'yield' stress of concrete plane force triad. He assumed that, if both principal
tensionandcompression steel yield in-plane forces were compressive, concrete of sufficient
stresses strength was provided to preclude the use of com-
reinforcement stresses pression reinforcement. NielsenI2) has alsoconsidered
section thickness the case of skew tension reinforcement, when the
in-plane direct and shearforces per unit applied forces are referred to skew co-ordinate axes.
length (in-plane force triad) M ~ r l e y ' has
~ ) treated the general case of combina-
rectangular co-ordinate system tions of moments and in-plane forces by considering
orientation of skew reinforcement to x strain-rate fields which give equal positive strain rates
axis in each reinforcement position. As a particularization
dimensionless term defined in text of this problem, M ~ r l e y ' ~ derived
) equations for the
concrete crushing strain determination of skew tension reinforcement to resist
concrete ultimate strain a particular in-plane force triad.For the case of ortho-
reinforcement strains gonal
reinforcement, these equations
reduce to
compression steel yield strain Nielsen's'l] equations.
principal concrete strains If, on substitution into Morley's four general equa-
principal strain rates tions (i.e. two equations for bottom and two for top
orientation of major principal concrete reinforcement), negative values for the area of rein-
stress to x axis forcement required in one or more positions are ob-
reinforcement ratios tained,this implies that either no reinforcement or
direct and shear stresses compression reinforcement only is required in those
positions. In either case, a strain-rate field which does
not give the same strain rate in each reinforcement
principal concrete stresses position is required.Morleydid not considersuch

3
Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.
 Magazine of Concrete Research : Vol. 28, No. 94 :March 1976

strain-rate fields in his general case because they imply

thatthe direction of principalstrain rate varies
through the depthof the section and explicit equations
for reinforcement areas cannot be derived. However,
in the absence of bending moments, this problem does
not arise and explicit equationscanbe derived.
Nielsen") has derived the equations for the casewhen
no reinforcement is provided in one direction, and the
equations for thecase in which compression reinforce-
ment is provided in one or both directions are pre- Figure I : Sign convention for in-plane direct ana' shear forces
sented in this paper. per unit length.
Theprobablereasonforthe tendency to ignore
compression reinforcement in the past is that it has
not been considered to be of practical interest. How-
ever, the recent publicationby Subedi(4)of a graphical A.. f,

approach to thedesign of compression reinforcement,

which was originally presented by R o ~ e n b l u e t h ( ~ ) ,
would suggest that it is of practical interest.

Assumptions and definitions

(1) It is emphasized that all stresses, strains and strain
Y
rates are taken to be tension positive.
(2) The sign convention for in-plane direct and shear
Fgure 2 : Directions of reinforcement and principal concrete
forcesperunitlength is shown in Figure 1. Shear stresses.
forces perpendicular to the plane of the section are
ignored.
(3) The reinforcement is assumed to be positioned
symmetrically with respect to themiddle surface of the
section and to be in two non-orthogonal directions.
The x axis of the rectangular x-y co-ordinate system is
made to coincide with one reinforcement direction and
theother reinforcementdirection is atan angle a
(0 < a < T)to the x axis (measured clockwise positive
as in Figure 2).
(4) The reinforcement is conservatively assumed to
carry only uniaxial stress in the original bar direction.
Thus kinking of the bars across cracks and any shear
carried by the bars are ignored. There is both experi-
mental
and theoretical to suggest that
kinking of the bars across cracks can be ignored, al- Figure 3: Yield criteria for concrete in plane stress.
though contradictoryevidence has been p u b l i ~ h e d ( ~ J ~ ) .
The small influence of shear carried by the reinforce- stress (as shown by the solid lines of Figure 3) and to
ment on the yield criterion for slabs has been taken be perfectly plastic. These assumptions obviously do
into accountby various r e s e a r c h e r ~ ( ~but
~ 9is~ ignored
~) not reflect thetruebehaviour of concrete andthe
in this paper. The reinforcing bars are also assumed to implications of this are discussed later.
be perfectly plasticand to yield at stresses o f f , in (8) The principal concrete stresses are taken to be o1
tension andf,' in compression. and G* with the majorprincipal stress at 0 to the x axis
(5) The barspacing is assumed to be small in compari- as shown in Figure 2. is always algebraically greater
son with the over-all structure dimensions so that the than G*.
reinforcement can be considered in terms of area per (9) Failure occurs by unrestricted pIastic flow and not
unit length rather than as individual bars. by buckling of the section.
(6) All stresses, strains and strain rates are taken as
their mean continuous values. Thus the discontinuous
cracked reinforced concrete slab is considered as an Theory
homogeneous continuous plate. Let the areasof reinforcement per unit length in the
(7) The concrete is assumed to have zero tensile x and a directions be A , and A , respectively and their
strength, to exhibit a .square yield criterion in plane associated stresses f , and f a . Then, by considering

4
Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.
 Reinforcement to resist in-plane forces

Figures 1 and 2, the three following equilibrium equa- possible combinations to be considered. Table 1 sum-
tions may be written: marizes which of the variables in equations 4 are pre-
determinedforeach of these nine cases and also
+ + o,h cos20

I
n, = Axfx A,f, cos2 a indicatesthemethod of solving forthe remaining
+ o,h sin 0 variables. It can be seen that a direct solution can be
ny = A a f a sin2 a + o,h sin2 0 + 02hcos2 0 . . . .(1) obtained except for cases 1 and 4 where there are four
unknowns to be obtained from three equations. There
nxy = - A , f , sin a cos a - olh sin 8 cos 0 is an infinite number of solutions to these cases, but
+ o,h sin 0 cos 0 the solution which gives the minimum total amount of
On dividing through by theslabthickness ( h ) and reinforcementcan be obtained by considering the
defining the direct and shear stresses as additional equation:

ox = nx/h X P X + Pa) -
- 0.. . . . .. .. . . . . . .(5)
b(tan 0)
oy = ny/h . . . . . . . . . . . (2)
*
In Table 1, o1 is given as zero when tension rein-
Txy = nxylh forcement is provided becauseconcrete
the
must be
and reinforcement
the ratios as cracked, and o2 = f, when compression reinforcement
is provided so as to make optimumuse of the concrete.
Qx = Ax/h Table 2 summarizes
the expressions forareas
of
the
...........
Pa = &/h '(3) reinforcement,principalconcrete stresses and 8 for
each case. The following symbols are introduced in
we obtain: this Table :
0
, = p,fx + p a f a COS' a + 01 COS* 8
= ox - f,
+ o, sin2 0 Oxf

= - f,
oy = p a f a sin2 a + a1 sin20 + c2cos2 0 . . . . .(4)
Gyf oy

4
T~~ = - pafa sin a cos a B = 1 - 7( T + cy cot a ) ( T , ~ + oyfcot a )
~ ~
+ (02 - 0,)sin 0 cos 0
fc

Now the reinforcement in each direction can be ten- A S S O C I A T E DS T R A I N - R A T EF I E L D S

sion reinforcement, or compression reinforcement or The above are lower-boundsolutions(the design
therecan be no reinforcement;therearethus nine equations giving upper bounds to the reinforcement

T A B L E 1 : Summary of various possible combinations of reinforcement.

Case I Reinforcement
description 1 Known values I Method of solution

1 Both tension
'
f x = f a = fs, c
rl = 0 Minimization of (px + pa)
No x
2 f a = h, px = 0,01 = 0 Direct solution
a tension

No a
Direct solution
x tension

Both compression Minimization of (px + pa)

No x
Direct solution
a compression

No a
6 Direct solution
x compression

x tension
7 f x = h , f = h', 0 1 = 0,0 2 = f c Direct solution
a compression

x compression
fx = f S ' , f = h, 5 1 = 0,0 2 = f c Direct s.olution
GI tension

No reinforcement = pa = 0 Direct solution

5
Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.
 TA B L E 2 : Design equations for skew reinforcement.

FX tan 0
~~~~

1(ox + 2Txy Cot a + 5y cot’ a

fs
1
+ cot a
0 +
- 2 ( ~ ~oy cot
~ a) (cot a cosec a) - (cot a & cosec a)

l I)
‘xy 5y

+ sin u

2 0 0

3 f,1 (5. - 5) 0 0 oy +- T2xy

5Y
-
‘xy

4
I
( oYfcosecz a - I T~~ + oyfcot a fc - (- cot a i cosec a)-1
sin a

oxfayf - T x y 2
5 0
-L x f + 2Txy cot a + 5yfCOt*x fc

6 fs.
1 (DXf - 5) 0 fc

7 0 fc

8 1[ m y
fs
- Txy cot a - - (1 - p)
2 1 0 fc

ay cot* a + TX> cot a + fc-2 (1 - p)

9 0 0
2

NOW: Case 1 : alternative sign is the same as that of ( T ~ oY

~ cot a). +
Case 4 : alternativesign is thesameasthat of + ~ ~ f ca).
ot

Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.
 Reinforcement to resist in-plane forces

areas) but, if admissible strain-rate fields correspond- Morley(15)when considering moments, and this ap-
ing to the stress fields can be found, they are unique proach has been applied in the present context in the
solutions(14). form of a graph with axes ox. 1 I
T~,, and D/, T~,, . I I
If the normality rule forplastic strains(14)is applied Typical graphs are shown in Figure 4 for a = 60" and
to the square' yield criterion of Figure 3, it can be I
f,/ T~,,I = -4. Unlessthesymmetrywhich is ap-
shown thatthe followingstrain-rate fieldsgive the parent in Figures 4a and 4b is taken advantage of by
required strain rates in the reinforcement directions. using sign changes similar tothoseproposed by
Morley(ls), itis necessary to construct separate graphs
Cases I , 2 and 3. Uniaxial extension in ol direction.
for positive and negative T~,,. The boundary curves are
Cases 4, 5 and 6. Uniaxial contraction in o2 direction. labelled 1 to 14 and their general equations are given
in Table 3. In general, it is required that:
Case 7. ,Field having a positive principal strain rate
(c1)in the o1 direction and a negative principal strain f, 6 -2 I -rXy I cosec a. . . . . . . . . . '. .(8)
rate ( c 2 ) in the o2 direction where
The rathercomplicated graphs shownin Figure 4 need
tan2 (0 - a) > -
4 tan2 8 . . . . . . . . .(6) not be constructed accurately since, as Morley(15) has
E2
pointed out, the equations for steel areas are contin-
uous across the case boundaries so that onlysmall
and 0 can be obtained from Table 2 . errors result from allotting a borderlinestress triad to
Case 8. As case 7 but where thewrong case. Theboundary lines paralleltothe
I 1
ox/ -rxY axis extend to & m.

tan2 (8 - a) < -
4 tan2 8. . . . . . . . .(7)
E2
Orthogonal reinforcement
CASE BOUNDARY CURVES The complex expressions in Tables 2 and 3 simplify
Having established the equation relevant to each of considerably with orthogonal reinforcement (a = n/2,
the nine cases, it is now necessary to establish a means subscript a subscript y ) to those given in Tables4
of determining which set of equations should be used and 5. In addition, the case boundary graphs illustrated
for a particular stress triad. This can be achieved by in Figure 4 reduce to one graph (Figure5 ) for any T~,,
deriving the surfaces in stress space which form the for a particular value off,, providedf, 6 - 2 T ~ , , . I I
boundaries to the regions pertinent to each case. A In Figure 5, each of the vertical and horizontal bound-
pictorially simpler approach has been usedby ary lines extends to m. *

Or

t-
I TV I
inadmissible

inadmissible
I

inadmissible I
(a) TX,, positive ( S ) T ~ negative
Y
Figure 4 : Case boundarygraphs for a = 60", f c = -4 I 1 Plainnumeralsrepresentboundarycurve Nos. Ringednumerals
represent case Nos.

7
Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.
 Magazine of Concrete Research : Vol. 28, No. 94 :March 1976

T A B L E 3 : Boundary curves for skew reinforcement.

Curve Equation

3 ~-
5y - - (cosec x cota)-’
I Txy I

5y
xf - l
9
I ‘xy I I Txy I

10

11

12

13

14

L
NOTE:. Alternative sign is the same as that of T ~ ~ .

Implications of assumptions However, the practical implication is that the expres-

sionsforcompression reinforcement given in this
Y I E L DC R I T E R I O N OF CONCRETE paper are conservative. The current British Standard
In theprevious derivations, it has been assumed that for the structural use of concrete(”) implies that the
the concrete has a square yield criterion in principal ‘yield’ stress of concrete is 60% of its characteristic
stress space as depicted by the solid lines in Figure 3. cube strength (h,).A partial safety factor of 1.5 has
The adoptionof a more realistic criterion, suchas that to be applied to this, givingf, = -0.4f,, as the value
obtainedexperimentally by Kupfer,Hilsdorf and to be used in practice with the square yield criterion.
Riisch(16) and shown as a brokenlineinFigure 3,
would imply that less compressionreinforcement CONCRETESTRESS-STRAINCURVE
would be required and also that the strain-rate fields It has been assumed up to now that the concrete is
mentioned previously wouldbe slightly different. perfectly plastic, whereas it exhibits strain-softening as

Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.
 TA B L E 4 : Design equations for orthogonal reinforcement.

Case Px PY

1 1
1 - (.x
f,
+ I Txy I ) - (ay + I Txy I 1 0
f,

2 0 0

3 0 0 cry + 2
‘SY 5Y
~~ ~~~~ . ~~~ ~~

1 1 2
4 - ( 5 x f - l ‘xy 1 ) - (ayf - 1 Txy 1 )
fs‘ h’ I ‘xy I

5 0 5x +-
Txy‘

5xf
.fc

6 0 ay +-
Txy2
fc
5Yf
~___ ~

7 fc
fs‘

8 fc
.fc(l - P)

9 0 0
ox + fly - - ay)* + 43; ax - ay - 2/(5x - fJY)’ + 4Txy2 S
Y
2 27, E

Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.
 Magazine of Concrete Research : Vol. 28, No. 94 :March 1976

TABLE 5 : Boundarycurves for orthogonal I 8

0,
I l
reinforcement.

Curve Equation

l
Figure 5 : Case boundary graph for orthogonal reinforcement,
fc = - 4 1 T~~I . Plain numerals represent boundary curve Nos.
Ringed numerals represent case Nos.
8

5xf 5yf - l
9
I t x y I l ‘xy l
STRAIN
10

l1

12 Inapplicable

13 Inapplicable

14

Figure 6: Stress-strain curve f o r concrete in compression.

shown in Figure 5 andalsohasamaximumstrain being used. In a similar situation, Wood(18) has sug-
capacity (E”). The applicability of plasticity theory to gested applying a ‘material stress factor penalty’ when
suchamaterial is obviously dubious;but, in most applying limit analysis to non-ductile material..
previous applications with tension reinforcement, the If it is assumed that, atincipient failure, the concrete
ductility of the latter ensures that the slab element principal total strains are and c2 (cl being algebraic-
itself has sufficient ductility and the strain-softening ally greateF than c2) and act in the directions of the
properties of the concrete are masked.If compression plastic principal strain rates, the steel strains (c, and
reinforcement is to beprovided, sufficient ductility E,) can be expressed as:
will not necessarily be obtained; however,provided
+
that it is ensured that the compression reinforcement
yields before the concrete beginsto crush, an adequate
E,

E,
=

=
cos20 E~

cos2 (0 - a )
sinz 0
+ E~ sin2(e - a )
) ...... .(9)
design should result. In this context, it is worth con-
sidering applying an additional partial safety factor Now, in order to obtain a ductile failure, itis desirable
which reflects the inapplicability of the design method that either or E, should reach the compression steel

IO
Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.
 Reinforcement to resist in-plane forces

yield strain ( E ~ ’ ) as or before E~ reaches the strain (E,) possible to obtain a strain-rate field having positive
at which crushing begins, i.e. and negative principal values which will give a positive
strain ratein the directionof the tension reinforcement
2 E~ cos2 0 + E, sin2 0
. . . . . .(lo) and a negative strain rate in the directionof the com-
E ~ ’ 2 E~ cos2 (0 - a ) + E, sin2 (0 - a) pression reinforcement.
In reality the concrete will crack and thus the situa-
Since E~ is negative in the absence of tension reinforce-
tion of a reinforcing bar in compression crossing a
ment, it is conservative to take E~ = 0 in order to ob-
crackcan arise. However, asthe design equations
tain a bound on E ~ ’ ,i.e.
given in this paper are based upon equilibrium con-
> E,
i
E ~ ’ sin2 0 siderations, they give a safe solution and this anomaly
..................... .(l 1) can be overlooked. In this respect, it is worth quoting
2 E, sin2 (0 - a )
Nielsen(2)on the subject of strain fields:
Codes of p r a c t i ~ e ( ~ ’ *imply
’ ~ ) that concrete begins to “When we are considering the strainfield at all, it is
crush under short-term direct compression, at a strain only to secure that thecalculations are in accordance
of - 0.002, and then inequalities 1 l imply that a duc- with the mathematical model used, viz. the theory of
tile failure cannot be obtained if the magnitude off,’ perfectly plasticmaterials.Takingthestrain field
exceeds 400 N/mm2 and thatmild steel can only give a into consideration we secure that no distribution of
ductile failure if the greatest orientation of o2 to the stresses giving higher carrying capacity canbe found,
compressionreinforcementdirections is less than which is the most essential thing.”
about 38”. Thus,fortheoptimumsituation of o2
bisecting the reinforcementdirections,the angle bi- BUCKLING
sected should be less than about 76”. This means that
Theequations derived in this paperassume that
mild steel should not be used for orthogonalreinforce-
failure does not occur either by buckling of the com-
ment. However, since:
pression reinforcement or by buckling of the section
(a)has been ignored ;
as a whole. Detailed discussion of these points is out-
(b) the value of -0.002 for the crushing strain is in-
side the scope of this paper. Guidance on restraining
tended for columns and is not necessarily applic-
compression reinforcement and on section slenderness
able to plate-type elements on account of the
ratios is usually given in codes of practice. This guid-
lateral
restraint
provided by thesurrounding
ance is intended for columns and is not necessarily
concrete; and
applicable to plates subjected to in-plane forces. Ex-
(c) any enhancement of the concrete crushing strain
perimental data are thus required for such situations.
due to the biaxial stress state of the concrete is
ignored ;
it seems reasonable to suggest that: Application to combined bending and
the adoption of mild-steel compression reinforce- in-plane forces
ment will, in general, lead to ductile failures in the The provision of reinforcement to resist combined
cases of orthogonalreinforcement and of skew bending and in-plane forces is extremely complex. A
reinforcement,provided that o2 lies within the design solution is usually obtained by adopting a sand-
acute angle; wich approach in which the six stress resultants are
a design, using skew compression reinforcement, resolved into two sets of in-plane stress resultants
in which o2 lies within the obtuse angle should not acting in the two outer shells of the sandwich. Such
be adopted and,indeed, this indicates a bad initial approaches have been suggested by M ~ r l e y ‘ ~ and
)
choice of the reinforcement directions; Brm~dum-Nielsen(~~); Morley has indicated in a pri-
high-strength steel (400 to 450 N/mm2) should be vate communication that he is currently developing
used with caution as compression reinforcement this approach. The equations derived in the present
unless the principal appliedstresses are in the rein- paper can be used to obtain the required amounts of
forcement directions. reinforcement to resist the above two sets of in-plane
stress resultants.Regardingthe ductility of such a
section, which is subjected to a strain gradient rather
D I S C O N T I N U I T I E SI NT H ES T R A I N - R A T E than a uniform strain distribution through the depth
FIELD of the section, the extreme-fibre compressive strain in
When plasticity theory is applied to reinforced con- the concrete will be the ultimate strain E, rather than
cretestructures, it is generally assumed that all E,. Hence the concrete strain at the level of any com-

stresses, strains and strain rates can be taken as their pression reinforcement can vary, with a limiting value
mean continuous values and thus the discontinuous, tending to E,. If the strain at the level of the compres-
cracked nature of the concrete is ignored. This has an sion reinforcement is known, suggestions astothe
anomalousimplication when both compression and applicability of mild and high-strength steel can be
tensionreinforcement are provided, since it is then made in a manner similar to thatpresented earlier for

11
Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.
 Magazine of Concrete Research : Vol. 28, No. 94 :March 1976

in-plane forces. Thispoint is not discussed further, 8. THOMAS, P. La deviation angulaire des armatures dans la

other than tosay that the value of E, of -0.0035 given zone de rupture des dalles en btton arm& (Angular devia-
tion of the reinforcement in the fracture zoneof reinforced
in codes of p r a ~ t i c e ( l ~ is
* ’int’ended
~) for beams and concrete slabs.) Annales de l’lnstitut Technique du Britiment
that there is unpublished experimental evidence at the et des Travaux Publics. No. 318. June 1974. pp. 93-106.
Cement and Concrete Association to suggest that this 9. KWIECINSKI, M . Some tests on the yield criterion for a rein-
strain is greater in a slab. In addition, thebiaxial stress forced concrete slab. Magazine of Concrete Research. Vol.
state in a slab compared with the uniaxial state in a 17, No. 52. September 1965. pp. 135-138.
IO. BAUS, R . and TOLACCIA, c. Calcul a la rupture des dalles en
beam influences the ultimate crushing strain. beton arme, et etude experimentale du critkre de rupture en
flexion pure. (Calculation of the failure load of reinforced
Conclusions concrete slabs and experimental study of the yield criteria
(1) Equations have been presented for the proportion- in pure bending.) Annales de l’lnstitut Technique du Briti-
ment etdesTravauxPublics. No. 189. September 1963.
ing of skew or orthogonal tension and/orcompression pp. 871-894. TranslationfromtheFrench by G . S. T.
reinforcement to resist an in-plane force triad. ARMER. Garston, BuildingResearch Station,May 1964.
( 2 ) In order to obtain adequate ductility of the section, pp. 30. Library Communication No. 1210.
it is suggested that the following restrictions be placed 11. PRINCE, M . R . and KEMP, K. 0. A new approach to the yield

upon the use of compression reinforcement: criterion

for isotropically
reinforced concrete slabs.
Magazine of Concrete Research. Vol. 20, No. 62. March
(a) a design, using skew compression reinforcement, 1968. pp. 13-20.
in which theminorprincipalconcretestress at 12. HOLMES, M. and ARNAOUTI, c. Theoretical moments at yield
failure lies within the obtuse angle should not be in a reinforced concrete slab. Building Science. Vol. 8, No.
adopted; 4. December 1973. pp. 363-371.
13. MILLS, G . M . A partial kinking yield criterion for reinforced
(b) in other cases, only mild-steel compression rein-
concrete slabs. MagazineofConcreteResearch. Vol. 27,
forcementshouldbe used unless theprincipal No. 90. March 1975. pp. 13-22.
applied stresses are in the reinforcement directions, 14. PRAGER, W. A n introduction to plasticity. Reading (Mass.),
in which case high-strength (400 to 450 N/mm2) Addison-Wesley Publishing Co., 1959. pp. 148.
steel may be used. 15. MORLEY, c. T. Skew reinforcement of concrete slabs against
bending and torsional moments. Proceedings of the Insti-
tution of Civil Engineers. Vol. 42. January 1969. pp. 57-74.
REFERENCES 16. KUPFER, H . , HILSDORF, H. K. and R ~ S C H ,H. Behavior of
1. NIELSEN, M. P. Yield conditions forreinforced concrete shells concreteunder biaxialstress. Journal oftheAmerican
in the membrane state.Non-classical Shell Problems: IASS Concrete Institute. ProceedingsVol. 66, NO. 8. August 1969.
Symposium, Warsaw, 1963. Editors: W. OLSZAK and pp. 656-667.
A . SAWCZUK. Amsterdam, North Holland Publishing Co., 17. BRITISH STANDARDS INSTITUTION. C P 110: Part 1 : 1972.
1964. pp. 1030-1038. The structural use of concrete. PartI : Design, materials and
2. NIELSEN, M . P. On the strength of reinforced concrete discs. workmanship. London. pp. 154.
Aalborg, Danmarks Ingenimakademi Bygningsafdelingen, 18. WOOD, R. H. Non-linearity and limit-state designof complete
1971. Bulletin No. B2. Reprinted from: Acta Polytechnica composite structures. Paper presented at a Symposium on
Scandinavica, Civil Engineering and Building Construction Non-linear Techniques and Behaviour in Structural Anal-
Series No. 70, 1969. pp. 254. ysis. Crowthorne, Transport and Road Research Labora-
3. MORLEY, c. T. Optimumreinforcement of concreteslab tory, December 1974.
elements against combinations of moments and membrane 19. COMITE EUROPEEN DU BETON AND FEDERATION INTER-
forces. Magazine of Concrete Research. Vol. 22, No. 72. NATIONALE DE LA PRECONTRAINTE. Internationalrecom-
September 1970. pp. 155-162. mendations for the design andconstructionofconcrete
4. SUBEDI, N . K . Design of reinforced concrete sections sub- structures. English edition. London, Cement and Concrete
jected to membrane forces. The Structural Engineer. Vol. Association, 1970. Principles and recommendations:
53, NO. 7. July 1975. pp. 289-292. pp. 80. Appendixes: pp. 47.
5. ROSENBLUETH, E. Shell reinforcement not parallel to prin- 20. BRBNDUM-NIELSEN, T. Optimum design of reinforced concrete
cipal stresses. Journal of the Anlerican Concrete Institute. shells and slabs. Technical University of Denmark, Struc-
Proceedings Vol. 52, No. 1 . September 1955. pp. 61-71. turalResearchLaboratory, 1974. pp. 190-200. Report
6. MORLEY, c. T . Experiments on the distortion of steel bars No. R44.
acrosscracksin reinforced concreteslabs. Magazine of
Concrete Research. Vol. 17, No. 54. March 1966. pp. 25-34.
7. LENSCHOW, R . J. and SOZEN, M . A. A yield criterion for rein-
forced concrete under biaxial moments and forces. Urbana,
University of Illinois, July 1966. pp. 527. Civil Engineering Contributions discussing the above paper should be in the hands
Studies, Structural Research Series No. 31 1 . of the Editor not later than 30 June 1976.

12
Downloaded by [ York University] on [22/09/16]. Copyright © ICE Publishing, all rights reserved.