U D C 620.176: 624.072/073.012.

45

Reducing the shear reinforcement

in reinforced concrete beams and slabs
by Professor Dr lng. F. Leondardt

TECHNICAL UNIVERSITY OF STUTTGART

SUMMAR Y C internal compressive force in compression chord

The shear tests reported in the present paper were of " truss" due to bending
carried out in Stuttgart from 1961 to 1963. The mechan- Cs compressive force in the inclined struts of the
ism of failure at ultimate shear load is explained in '.' truss"
terms of the truss analogy, and the ratio between the Tx tensile force in the longitudinal reinforcement
width of the compression flange and the thickness of the (tension chord of" truss ") due to bending
web is shown to have a considerable influence on the Ty tensile force in stirrups
inclination of the top chord and of the diagonal struts of P load on test beam
the truss and, consequently, on the tensile forces in the Pp prestressing force
web. From the results, a simple method for reducing V total shear force
the shear reinforcement is derived. For short beams with Vj shear force carried by the compression members
a slenderness ratio ljd < 8 or for concentrated loads of the truss
close to the supports, a further reduction in shear cover- V' shear force carried by the web reinforcement
age is possible. The determination of the points for cut- v = Vjb'jd = shear stress
off and bent-up bars must be adapted to the variations Vj = V;fb)d

in the compression members of the truss. The tests Vp shear stress resulting from inclined prestressed
showed clearly that stirrups are preferable to bent-up tendons
bars for shear reinforcement, particularly if high-tensile Iv tensile stress in web reinforcement
steel is used. fe' compressive cylinder strength
.hc compressive stress in prestressed concrete due to
prestressing force after all losses at centroid of
cross-section
Notation h yield strength of reinforcement
t total depth of section flo j~ principal stresses according to bending theory of
d distance from extreme compression fibre to cen- beams
troid of tensile reinforcement Av area of web reinforcement within the length s of
j ratio of distance between centroid of compression the beam
and centroid of tension to depth d Ac gross area of section
b' width of web 0; angle between web reinforcing bars and longi-
b width of compression flange tudinal axis of beam
s spacing of stirrups or bent-up bars in a direction (3 angle between compression strut and longitudinal
parallel to the longitudinal reinforcement axis of beam in truss analogy
h amount of horizontal movement of the Mjjd line r Avjb's
to get the T line "y] factor of shear coverage, related to shear rein-
span length forcement according to Morsch's analysis
a distance of single loads from support x. second reduction factor for shear coverage for
M bending moment short beams or loads close to support

187
Magazine of Concrete Research: Vol. 17, No. 53 : December 1965

WITH VEP-TlCAl STIRRUPS, n~fold truss, n= if

stirrup forces T1 = V 1 related to the beam

diagonal compression forces C1 = Vy2 j length ill = jd
LOAD P-tons'
o 5 10 15
4.-----,,----~-.--~--.-~~--~~---,
20 25
WITH INCLINED STIRRUPS

'
C, . -C1 c II T
). st-----~------_+------_+_------+__----___1

- j C=T=~ --..:. - - 20
" T,/.

~~o:......""':"-""~~=---lI4~s·~~......l'J.~=.:Li:l""""-_T
~
diagonal forces Ts = Cs = ~2 related to the beam -lengthj,1 = jd x
2'5

Figure 1: Assumptions of classical truss analogy according to

..
E
~
.><
I
I
15 o.!
Morsch. o.! 2
on
0..
:J B
'"'" 1'5 9
i=
v> I
I 10 v>
~ I
v>
w
v>
v>
w '"
l-
v>

'"
l-
v>

1 T A
/::1
T
V
0·5

(b)

-0·5 L-____ ~ _______ L_ _ _ _ _ _J -______L __ _ _ _ ~O

0 6 12 18 H )0
(e)
LOAD P-kg x 103 (tonnes)

Figure 2: Tied-arch action in beams with thick webs. Figure 3: Average stresses in stirrups, measured in beams with
(a) Tied-arch action (b) Total shear force in C (c) Decrease in the one variable bIb' = width o/compression zone to width of
tie force due to bending resistance of concrete" teeth". web, with equal and constant shear reinforcement (1) = 0'38)
compared to stresses calculaied according to Morsch.

Introduction Variability of the behaviour in the cracked

The Stuttgart shear tests, which have been published state
in seven reports O -7), have brought considerable new The classical shear analysis of Ritter and Morsch
knowledge because each of the many parameters in- explains the shear strength in the cracked state by a
fluencing the ultimate shear load have been studied truss analogy, using a truss with parallel chords, com-
separately. It has been found, for example, that the pression struts at an angle of 45 0 and vertical, or with
most unfavourable shear condition is given by one or tension bars inclined at -45 (Figure 1). It is assumed
0

two concentrated loads per span with a moment/shear that the shear force V will be fully carried by the web
ratio, MjVd, between 2·4 and 3·5. All other load pat- members; this leads to the following percentage of
terns can be considerably more favourable. The pre- shear reinforcement in the web for the length s
sent paper deals only with this unfavourable loading;
v V .h Av
short beams or loads near the support with M/Vd < 2·4 r = fv = b'jd sin yfv Wit r = b's
are t~eated separately. High-tensile deformed steel bars
with a yield point, fy, of 60,000 Ib/in 2 were used in all This amount of shear reinforcement may be called
the Stuttgart tests. full shear coverage; a reduced amount will be defined
188
 Reducing the shear reinforcement in reinforced concrete beams and slabs

with rectangular cross-section they remain very small

or are even negative (compression) up to 0·7Pu (ulti-
mate load).

EFFECT OF REDUCING THE WEB

6-05 In.
THICKNESS
Figure 4: The resultant force C in the compression chord ofT If, in T beams, the compression and tension chords
beams can be inclined from J/12 to 1/20 in spite of shear cracks are kept equal and only the thickness of the web is
coverinf{ the whole depth of the web.
reduced, then the tensile stresses of the equal stirrups
increase almost in proportion to the ratio b/b' (Figure
3). The measured tensile forces in stirrups, however,
by the reduction factor 'YJ < 1, referred to this full remain far below those calculated with Morsch's truss
coverage. analogy even for webs as thin as 2 in. with b' /b = 1/6.
With this classical truss analogy, the equilibrium This means that, even in beams with very thin webs, a
conditions for the inner forces are fulfilled, but the part of the shear force must be carried by the chords
compatibility conditions are adversely affected because and not by the web members. In fact, we have mea-
the compression members of concrete are 3 to 20 times sured tensile stresses on top of the compression flange
stiffer than the tension members of thin steel bars. near the supports. Therefore, even in T beams the
Nature is not aware of this truss analogy and acts resultant compressive force C, which can be con-
according to the law of the minimum of deformation sidered as the top chord of an imaginary truss, must
work; the inner forces in reinforced concrete beams be inclined. Figure 4 shows that a slope of 1/12 to 1/20
therefo're change considerably as soon as cracks appear between the load and the upper end of the last shear
and largely depend upon the stiffness ratio between crack is easily possible and has been observed. This
the concrete and the steel members. A beam with a slope means that 25% to 15% of V is carried by the
thick web will show internal forces very different from compression chord in a beam with a shear span of 3.
those of a beam with equal chord members but with In the bottom chord (longitudinal reinforcement), a
a thinner web. Even the crack pattern changes with small part of V might be carried by the dowel effect of
this stiffness ratio and so influences the behaviour. We the reinforcing bars; however, this effect is surely
shall try to explain these various behaviours by our over-estimated in reference 9, as soon as stirrups are
test results, using again trusses as an analogy. used.
Further, it has been observed that the inclination of
the shear cracks is not simply 45° but varies consider-
BEHAVIOUR OF SLABS AND BEAMS ably with the stiffness ratio expressed by b' /b and can
WITH THICK WEBS (FIGURE 2) be as low as 30° (beam ETI of reference 2)-see Figure
Beams or slabs with rectangular cross-section be- 5. This means that the diagonal struts of our imagin-
have as a tied arch and the shear forces are carried ary truss are inclined less and, therefore, a smaller
totally by the inclined upper chord (arch or strut). The quantity of vertical tension bars, the stirrups, will be
tie force decreases towards the support because of the sufficient for equilibrium. Simultaneously, the tensile
unreliable bending resistance of the concrete between force in the bottom chord near the support will be
the cracks, as shown by Kani(S): unreliable, because larger than in a truss with 45° struts. Figure 6 explains
this resistance decreases almost to zero when the ratio by simple truss analysis how the vertical tensile forces
between crack spacing and crack depth becomes small Ty and the chord tension Tx are influenced, when the
in beams with t > 15 in. At ultimate load, this decrease inclination of the top chord varies from 113 to zero
in the tie force can become small and therefore the tie or the angle f3 of the diagonal struts varies from 45 °
should continue unreduced to the supports and be well to 30°. Ty can vary between zero and V, according to
anchored. the beam length 8/ = jd.
As soon as cut-off bars are used in rectangular cross- The inclination of the shear cracks can also be in-
sections with their length determined by the moment fluenced to a smaller extent by the percentage of shear
diagram, then the shear strength will be reduced con- coverage, expressed by 'YJ. We have found that when
siderably. (6) b'lb = 1/6, the angle f3 of shear cracks can vary be-
Shear reinforcement in such cross-sections does not tween 3r for 'YJ = 0·25 and 44° for 'YJ = 1'0; the
alter this behaviour under working load; stirrups re- average variation, however, is small. On the other
main without tension up to very high loads and begin hand, 'YJ has no influence on the cracking load or on
to carry load only after shear cracks enter the inclined the distance between lines of calculated and actual
compression chord C. The test results shown in steel stresses in stirrups as shown in Figure 7 for T
Figure 3 prove that the stresses in stirrups are strongly beams with constant b'lb and only 'YJ variable.
influenced by the thickness of the web and in beams Even in large beams with extremely thin webs and
189
Magazine of Concrete Research: Vol. 17, No. 53 : December 1965

Figure 5 (lefl and above): Crack pal/ern of rectangular beam

ET 1 with bIb' = I compared wilh pal tern of T beam £T 4 with
blb'= 3. f2I Both beams have equal longitudinal and shear
reinforcement bul the average angles ~ of shear cracks differ
considerably.

Tensile forces
No. Truss system In th;, web In th~ chord
T,
related to Id near the support

L. x I1I ~=V
\ :
'}
/
/ ....
jd
I ,/ ~/ /.i
1 0 3·0V
8 = V '-T,

T, tllllllllllllllllllll1 3V

.~
f-, p_ V

with b'lb = IllS, as we have tested in our beams Tl 2

O. 5jd r ~. 40· 45 1/1 0 ·6V 1·5V
and T2 o ,5), where the angle of the cracks was about 8 = V t-jd-!-.jd-l--jd-

45 °, the forces in the stirrups do not reach the values T, tl 1111. "111 1 hiil, Ilrl I
of Morsch's truss analogy (Figure 8). This means that
even for the highest shear stresses, and in I beams,
part of the shear force is carried by a slight inclination
~-'T
~5 jd
3 1·0V 1·0V
of the resultant compressive force in the top chord and 8= V jd-+- jd --l--jd-J ...1
by" frame action" between the web members and the
chords.
T, tIIliIffill&liI 3V

Figure 9 hows how the portion of the shear force ~P - tV

carried by stirrups varies with increasing load for
~
~)tP -tJ T~5·
h f 2V 0 · 5V
different values of b' /b and two classes of shear cover- 8 = V -jd- jd-->-jd~
= o·nv
age. These lines show that the load intensity has con- T, 'I ~U5V' 1IIIIqililMt
siderable influence on the behaviour; the web members P= V
carry more shear with increasing load and only ulti-
mate load conditions dictate the safety and give the
5
1~ /1
8 = V '-1-5jd~1.5jd- 0·67V 1· 5V
design criteria.
In Figure 10, the influence of the ratio between web T, tlllll1;~r:ll lllllilll l
stiffness and chord stiffness expressed by bIb' on the
P= V
behaviour is shown in a different way; the lines give
l'
the distribution of the shear force on the web mem-
bers, the compression chord and on the "frame 6
I /I~
8= V't-1 .73jd--l-1 .73jd_
jd
...i 0·57V 1·73V

action" due to the fixed connexion between struts and

chords. In normal T beams where bIb' is between 3
T, fllllil WrlllllllWHI
and 6, the web members carry less than half of the
total shear force even under ultimate load conditions. Figure 6: The influence of an inclillation of the top chord or
of the angle ~ ~ 45 ° of the compression struts in simple trusses
The result is that the behaviour changes with decreas- on the vertical tensile force Ty and on the tie force Tx near
ing web thickness and increasing load intensity from the support.

190
 Reducing the shear reinforcement in reinforced concrete beams and slabs

SHEAR STRESS-Ib/in' P P

0
I
100 200
I
300
I
400
i
500
i
600
i
700
i
5
0 10

a-!tf-a,
20
LOAD P-ton.
30 40

I
50 il% i* iii II
T2 Tl

12" ! III.f
30
I 4-5
4-5

25
~
x 3-5 c

1 c
0 3-5
~ 'j'
I ..!
..! 20
'"
Q. ~
'"
Q. ::> x
c
::> '"'"
'"'"
~
2-5
15
~
~
l I
~
g
..! 2-5
~ 15
..!
'"
'" '"
'"
::]
W Q.
'"
10 '"
!;;
::> Q.
::>
'"
!;; 1-5 '"'" '"'"
!;; !;;
~ ~
10
5 '"
'"
w 1- 5 '"
::]
'"
!;; '"
!;;

LOAD P-kg x 10l (tMne.)

10 20 30 40 50 0- S f---;L--M-!I-----i
SHEAR STRESS-kg/em'

Figure 7: Average stresses in stirrups, measured in beams with 40 80 120 160 200
bIb' = 6 and A. constant, the only variable being the
percentage of shear coverage 1), compared with stresses LOAD 2P-kgxl0 (tonne.)
'
calculated according to Morsch.
Figure 8: Measured and calculated stresses in stirrups of beams
with extremely thin webs (bIb' = 15) and shear stresses v
reaching 2,500 Iblin 2 • Beam T 1 had vertical stirrups, T 2
inclined stirrups.

the tied arch to a truss with a curved or inclined top

chord and with diagonal struts inclined between 30°
1
and 45°. This means that the trusses which must be 1-75
considered in order to understand the shear problem
must have variables as shown in Figure 11.
0-8
BEHA VIOUR OF CONTINUOUS BEAMS
The system of internal forces in continuous beams
0-6
is generally assumed to have a simple beam between
V'
the points of inflexion (M = 0) suspended from a V
cantilever beam (Figure 12). The Stuttgart tests and 0-4
others(lO,ll) have shown that, for rectangular cross-
sections or T beams with thick webs, the inclined com-
pression chord extends directly to the support so that 0-2

the bottom tie reinforcement must also here be ex-

tended partially towards this support. Correspond-
ingly, there are very flat shear cracks near the inner
support with angles as low as 30° (Figure 13). As a o PIP, (P, = ultimate load in bending)

result, stirrups are more necessary near the support

than in the region of the small moments, where there Figure 9: Portion of shear force v' carried by web
were no cracks up to high loads, so that stirrups in reinforcement for different widths of webs expressed by b'lb
and for different shear coverages or, plotted against load
this region showed only small stresses. intensity PIPu.

191
Magazine of Concrete Research: Vol. 17, No. 53 December 1965

FOR WORKING LOAD p P

~
'"f2
80
~A~------'----~-------1~[ ~l
'"~
I
V>
~

o
60

~o
T
~~-300-38"
b'
b large

w
\:J
-<
f-
Z
w
U 20 b'
small
...'"
w
T ~-38"_450
b

1 2 3 bib' 15
I' Figure 11: The actual mechanism of internal forces can be
range of common use compared with stich trusses with inclined compression chords
and diagonal web struts with angles ~ .,; 45 °, the angles of
BELOW ULTIMATE LOAD
chord and struts mainly depending on bIb' (extended truss
analogy).

~ 80
'"f2
'"~ 60
I
V> wrong mechanism

o
w 40
\:J
-<
f-
Z
w
U 20

...'"
w
true mechanism

2 3 bib' 15
Figure 12: Popular idea of the mechanism of continl/ous beams
I' range of common use . compared with the findings by tests for rectangular
cross-sections.
Figure 10: Distribution of the shear force V carried by
1. web members = stirrups and struts (V')
2. compression chord (Vi)
3. frame action due to stiffness of junction between web
and chord members
as a function of bIb' for working load and just below ultimate
load.

Figure 13: Small inclination of shear cracks of continuous beams near inner support, when bIb' .,; 3.

192
 ReduCing the shear reinforcement in reinforced concrete beams and slabs

! CUBE STRENGTH. (,ube'-Ib/ln' x 10'

"" U*
critical
;&

/
I
/
according to Morsch
(,= vir
35
0

zone

~
30
/ 400 c
'o! / observed I :a-
/ V-VI .'
2S 1
'"0..::J fV=-r-
~
.'
/ 300
at:
"" / .- 20
20 em
~
;-
E; / '" 15
/ '"
w 200 '"
~ ""~ '"
w
at:
'" / 10 ~
'"
w
/ at:
...'" /
-<
w
::x:
100 at:
::i::x:
'" '" o 'I; for beams with web reinforcement
/ o 'IIJ for beams without web reinforcement '"
0
/ 0 200 300 400 500
'------J<--- depends on b' and d CUBE STRENGTH. (,ube'-kg/em'

Figure 15: Shear stress Vi due to portion of shear force

carried by compression members of single-span beams with
Pcrack
shear reinforcement plotted against the cube strength of
LOAD P concrete, compared with the ultimate shear stress Vu of beams
without shear reinforcement for different depths t.
Figure 14: Characteristic line of stresses in web reinforcement, [2'5 < (M/Vd) < 3'2]
compared with classical analysis of Morsch.

Also in continuous beams, the behaviour changes value however, is influenced by the portion of
Vi,
with the stiffness ratio b'lb in a similar way as shown Vi = which represents the portion of V carried
Pcrack
for single-span beams. In beams with thin webs, there by the compression members of the truss and, there-
is truss action under high loads with a smaller influence fore, shows a straight-line function with the compres-
of the inclination of the chords but a larger influence sive strengthfc' (Figure 15). We have found this value
of the angle f3 of the struts near inner supports. The to be different for single and continuous beams:
ratio V'I V is larger there than near outer supports. for single-span beams, Vi = is fe'}
.......... (1)*
for continuous beams, Vi = ~ fe'
Calculation of the reduction in shear
Pcrack increases with increasing b' and jd, because the
reinforcement
corresponding shear force is Vi = Vib'jb. The thicker
SLENDER BEAMS AND SLABS the web, the larger the distance between the actual
The types of behaviour described make it clear that stress line of the stirrups and the one calculated
full shear coverage according to Morsch's theory is not according to Morsch. This is caused by the inclination
necessary, because the tensile forces in the web are of the compression chord and the angles f3 < 45° for
considerably smaller and their magnitude depends the diagonal struts. From the characteristic line of
primarily upon the stiffness ratio b'lb. The sound Figure 14, equation 2 for the stresses of the stirrups
design principle for reinforced concrete-to carryall can be derived.
tensile forces by steel--can, therefore, be fulfilled by a V - Vi
reduced shear coverage YJ < I corresponding to the fv = - - . . . . . . . . . . . . . . . . (2)
r
requirements of the A.C.1. Code 318-63. Consequently, the factor for the necessary percentage
In all the tests, the measured stresses in stirrups of shear coverage can be written
showed a characteristic pattern (Figure 14). However,
YVD+L - Vi
the line of the stresses in the stirrups as a function of 'lJ = . . . . . . . . . . . . . . (3)
yv D+L
the load runs almost parallel to the one calculated by
Morsch's truss analogy, in a horizontal distance equal with D+L = working load, Y = factor of safety. This
to the load Pcrack. This load Pcrack is the one under equation corresponds in principle to the A.c.I. Code
which a shear crack has reached the stirrup. We have 318-63, and the Stuttgart tests confirm fully the sound-
defined Pcrack as the load which is found by continuing ness of this new code as regards shear.
the measured line towards the abscissa. This load has If we plot vulfc' for all our test results for the men-
a corresponding shear stress Vi, which we found to be tioned range of aid and the results of reference II
different from Vc for beams without shear reinforce-
ment. The American results give Vc = f( vic'), in- *A common and safe value of Vi = is fe' can be recommended
fluenced by the tensile strength of the concrete. The for practical use.

193
Magazine of Concrete Research: Vol. 17, No. 53 : December 1965

MOMENTISHEAR RATIO. MIVd = aid

- - A.C.1. 318-{,3 with" = 1·1v'f,'for f<' = 3.600 Ib/ln'
o single.span 'beans, Stuttgart T 0 T, at o
120
o
11
I
continuous beams. Stuttgart
I
two-span beams. Bryant et al.
'

t
~=O·6
(cube'
l 1
I
we web crushing failure I 100
1
~ ~
WC .'

~
o I ~:::Id
0.3r----------r--------~r-----~O~~--+-+---~
:;;w'" 80
I
~
b
....
\ .\:
concentrated
o o o WC ~

0 loads I ~

'" 60
o
~ '"
w

\\
a:
o o :;;
a:
1~'1'51 ~~d
«w 40

o 0
I

w
~~ / uniform load

.... ~
« 10 ~

L """-<
;::
--'
:J

o 1 12 16 20 24 28
0·1 ~--------_+-"-----__:7/F_--__..~'----_+=.::.::support

6 SLENDERNESS RATIO lid

Figure 18: The increase in shear-carrying capacity without

shear reinforcement with decrease of the slenderness ratio lid
or of the shear span ratio aid for single loads.
for v, = nfc'
0·25 0·5 0·75

FACTOR OF SHEAR COVERAGE. "Yj

Figure 16: The ultimate shear stresses of test beams with shear
failures plotted against percentage of shear coverage compared to?7A!AtXA
~I-------I
with curves 1) = tvu - vi)lvu.

X 2·5
a:
....o
u
~
0·1 .--------r-------,--------~-----,_------_,
z o· 5 1-----+----~'97''---'---+_----+----~----_____1
maximum v = 0-19 fe '
o
for inclined stirrups
....
U
:J
o
w
a: 6 10 12
lid

maximum v = 0-15 f('

0·15 1-------+-------+ for vertical stirrups ---"'-...I-f--~ Figure 19: Reduction factor x for the portion of the shear
and bent-up bars force to be carried by web reinforcement or for the shear
single spans coverage depending on lid for uniform load or on aid for
concentrated loads.

continuous beams

Y'D+l-"
~ 0·1 r -_____1)...'-= Y vO+L +--------+--+---ft---------j
+ against the percentage of shear coverage "I), we get
~
Figure 16. The lines for Vi = 0/16) fe' and (I /22) Ie' are
below all corresponding points with two exceptions:
three values for continuous beams, which did riot fail
by shear in the web but by bond failure of the top
o·05 r-------~------..,..."-------7"9f--------_t_------__1
chord bars (which have only about half the bond
strength of bottom bars owing to segregation of the
concrete after compaction) and one value for a simple-
span beam in which the web was reinforced with
bent-up bars only, which is not allowed. The graph
0·1 0·4 0·6 0·8 proves that the percentage of shear coverage "I) calcu-
FACTOR OF SHEAR COVERAGE. 1)
lated by equation 3 using Vi from equation I gives
sufficient safety; VD+L refers to the maximum value of
Figure 17: The required shear coverage for working load design V for the beam and "I) should be constant for the shear
and upper limits of allowable shear stresses. zone belonging to this maximum.

194
 Reducing the shear reinforcement in reinforced concrete beams and slabs

An upper limit of v must be chosen in order to avoid or with the percentage of shear reinforcement:
failure by diagonal compression in the web. In our Av v
tests we found that the diagonal compression is not r = X1) b's sin ex = X1).r.
only influenced by the direction of the web reinforce-
ment (as can be derived from the truss analogy) but
is further influenced by the angle of the shear cracks.
Structural conditions for reduced shear
We found that, for vertical stirrups, the principal reinforcement
diagonal compressive stress for high shear intensity PREFERENCE FOR STIRRUPS
can reach The reduced shear coverage should preferably be
fz = 2·4v (for low shear stresses it can befz - 3·5v). made with closely spaced stirrups, the spacing de-
creasing with increasing values of the shear stress from
For closely spaced inclined stirrups (ex = 45°) it
s < tl2 to t16. Bent-up bars of high-strength steel are
reaches
less suitable and lead to wider shear cracks. Inclined
fz = I·7v
stirrups are the most efficient (Figure 20). Bent-up bars
Therefore, by using a factor of safety of 2·1 and a are acceptable in continuous beams in the region of
reduction for sustained load, the upper limit of the small moments for changing some bars from bottom
shear stresses due to the working load can be chosen to top, so avoiding the large anchorage length espec-
for vertical stirrups and ially in the top region. Stirrups must be well anchored,
stirrups combined with preferably with hooks (Figure 21); those made of a
bent-up bars max VD+L = 0·I5fe' welded-wire mesh with a wire spacing of 2 to 6 in. and
anchored at the top by welded longitudinal wires, are
for inclined stirrups
especially suited for ribs, joists and small beams in
ex = 45° to 55° max VD+L = 0·19f/
buildings.
(since compression is critical and not tension, we refer
v to fe' and not to Vfc'). THE DETERMINATION OF THE POINTS
The maximum shear stresses for working load give
FOR CUT-OFF AND BENT-UP BARS
the necessary percentage of shear coverage 1) for single-
It is known that the tie force T is not proportional
span and continuous beams according to Figure 17.
to the bending moment as soon as the shear stress is
The upper limits are governed by web compression;
low. For higher shear stresses, the distribution of the
the lower limits show the low values which can be
tie force depends also upon the direction of the shear
allowed without web reinforcement and with constant
reinforcement. For low values of v and rectangular
ties. For T beams with b'lb < 1/3, the minimum
cross-sections or slabs, the tensile force in the longi-
requirements for shear reinforcement must be ob-
tudinal reinforcement can be as high as T = Mfjd +
served.
1·5 V near the support. For vertical stirrups, Morsch's
truss analogy gives a tie force T = (Mfjd) + (VI2). This
FUR THER REDUCTION OF THE SHEAR
value increases according to the types of behaviour
COVERAGE FOR SHORT BEAMS, SHORT
described.
SLABS OR BRACKETS
The length of longitudinal bars cannot therefore be
The Stuttgart tests have confirmed the results of found with the M fjd line, but with a T line, which is
other investigators(13-1S), concerning the influence of found by moving the M fjd line horizontally an amount
Mlvd or of aId for single loads and of the slenderness h (Figure 22). The value of h must be dependent upon
ratio lId for distributed loads as shown in Figure 18. v and, since the shear stress v is connected with the
According to these test results, the shear strength of factor of shear coverage 1) according to Figure 17, it
beams without shear reinforcement increases rapidly will be sufficient to determine h from 1) and the direc-
for concentrated loads as soon as aId < 3 or for dis- tion of the web reinforcement. The two following
tributed loads as soon as lId < 12. For deep beams values cover practical needs (Figure 23):
with lId = 1, such high shear stresses had been ob- h = 0·2 - 0·9 1))d ~ 0·5d
tained that shear reinforcement would make no sense, for vertical stirrups or vertical stirrups combined with
as several tests have actually proved (7,16). bent-up bars;
This fact can be fully explained by the arching action h = 0·2 - 1·2 1))d ~ 0·2d
of the concrete; it allows the shear coverage to be for inclined stirrups with ex = 45° - 55°. Top bars for
further reduced by the factor x, which can be taken negative moments over intermediate supports of con-
from the straight line given in Figure 19. The necessary tinuous beams must have a longer anchorage than
shear reinforcement (Iv = permissible stress in web bottom bars if the beams are concreted in normal posi-
reinforcement) will finally be calculated by tion, because the bond strength of top bars is low
X1) VS sin IX owing to segregation of the compacted concrete. The
Av = jdfv bond stress of top bars should always be calculated,
195
Magazine of Concrete Research Vo!' 17, No. 53 December 1965

LOAD P-tons
20 ~O 60
0·8
I
I
0·3
11 11
E
0·6
'I
/ E
I I
I
I:SSS
E
I
I
f0-
e
I
I l 0·2
I
I
f0-
e
7741 I
~
'"-<
u
I
I / /
~
'"
u
I I
I

I;11111111
o·~

'"
U
I
I /
V
Jill
-<
'"
u
III
1111111'11
:/ /
:E :E
:::J :::J
:E I
0·1 ~
X
-< 0·2 V x
-<
I

I~'»~
Figure 20: Maximum crack
Mdil
:E

l'h ~V
:E
/ IV
IV width of shear cracks for
average types of shear
~ reinforcement. Comparison
~ for equal cross-sections
o Ac and As.
20 ~O 60
LOAD P-kg x 10'

u0 INCORRECT
111 INSUFFICIENT
1·2

~
,,~ v}= 1·2-0·91)for vertlc.1
stirrups with bent·up bars

0·8 i'...

/~ ,~
welded·wlre mesh

u u DlT CORRECT FORMS OF STIRRUPS

hooks for deformed as well as smooth bars

h
d 0·6
;= 1·2-1·21)for
stirrups Inclined at
45-600 and almost
in direction of principal
tensile stresses
,
" '" "
~,

Figure 21: Different forms of stirrups .. the short upper anchorage

" ,,
is very important. 0·2

0·2 O·~ 0·6 0·8

FACTOR OF SHEAR COVERAGE, 1)

large anchorage length for top bars

Figure 23: Values of h for T line, depending on the factor 0/
shear coverage 1) and the type of web reinforcement .

..I
I
I UNDESIRABLE CONCENTRATION RECOMMENDED DISTRIBUTION
OF BARS OF BARS
I
I
I

Figure 22: The determination of the length of cut-off bars

must be made for a tie force diagram with the T line found by Figure 24: Top bars for negative moments of continuous beams
moving the MUd line horizontally a distance h, given by should be distributed sidewards into the slab, using medium
Figure 23. diameter bars.

196
 Reducin[( the shear reinforcement in reinforced concrete heams and slahs

0·22

0-0·1

0·2
0·15
unsafe. owing to
diagonal compression
in web above
horizontal line 0·2
0·18

0·25
0·16

'"6
+ -. 0·3
~ ~
+~
0·14
.f....
0·35
0
..(
0 -I 0·12
-'
f-
w
f- '-'
..( Z
w
1: <£
>= f-
~
-' 0-1
:J w
<£
'"
:J
~ U
~
w
~ f-
w w
<£ <£
f- u 0·08
~
Z
<£ 0
..( u
w
I
~

0·06

0·04 1-----::;;;;o"""'F'-------'
Figure 25: The necessary shear
coverage for prestressed concrete
girders, considering the degree
0·02 I------+-----+-----f---------+----- of prestressing by the average
compression stress due to Pp ,
depending on the maximum
shear stress due to loads and Pp.
Limits to prevent web crushing
failures.
0-2 0-6 0·8
FACTOR OF SHEAR COVERAGE, 1)

and it will often be found that small diameters must Torsion and prestressing
be used, well distributed horizontally in the slab Two tests on large prestressed hollow box girders,
(Figure 24). reported by Leonhardt and Walther(17), have shown
that there is no inclination of the compression chord
ANCHORAGE for shear forces due to torsion. It is therefore not
In beams and slabs without shear reinforcement, possible to reduce reinforcement dimensioned accord-
especially in beams with a slenderness of lId < 8 or ing to the classical truss analogy. Furthermore it was
with concentrated loads close to the supports with found that the principal compression stress due to
aId < 2, longitudinal bars must be safely anchored. torsion can become as high as 3·5 times the value cal-
In many cases, the bond length of straight deformed culated by usual analysis and can become critical
bars is inadequate and hooks may be necessary, pre- earlier than is generally suspected. One must therefore
ferably hooks placed horizontally, or loops or special be careful about the diagonal compression in the webs
anchor pieces. of hollow box girders. On the other hand, the principal
197
Magazine of Concrete Research: Vol. 17, No. 53 : December 1965

tension does not have to be limited in design for 8. KANI, G. N. J. The riddle of shear failure and its solution.
torsion, because it is a simple matter to cope with the Journal of the American Concrete Institute. Proceedings
tensile forces by reinforcement or by prestressing in Vol. 61. April 1964. pp. 441-467.
9. KREFELD, W. J. and THURSTON, C. w. Studies of the shear
the webs and slabs. strength of simply supported reinforced concrete beams.
For shear forces due to bending and prestressing, New York, Columbia University, June 1962. Report of the
similar relations for reduced web reinforcement as Department of Civil Engineering and Engineering Mech-
before can be established. Figure 25 shows a first anics.
tentative proposal for the determination of the neces- 10. BRYANT, R. H., BIANCHINI, A. C., RODRIGUEZ, J. J. and
KESLER, C. E. Shear strength of two-span continuous rein-
sary shear coverage in prestressed concrete beams, forced concrete beams with multiple point loading. Journal
depending upon the shear stress v due to loads and of the American Concrete Institute. Proceedings Vol. 59.
shear force Vp due to prestressing force and upon the September 1962. pp.
average normal stress due to the final prestressing 11. RODRIGUEZ, J. J., BIANCHINI, A. C., VIEST, I. M. and KESLER,
C. E. Shear strength of two-span continuous reinforced
force he = Pp/Ae. Further confirmation by tests is
concrete beams. Journal of the American Concrete Institute.
needed along these lines.
Proceedings Vol. 55. April 1959. pp. 1089-1130.
12. AMERICAN CONCRETE INSTITUTE. Building Code requirements
REFERENCES for reinforced concrete. Detroit, June 1963. pp. 144. ACI
1. LEONHARDT, F. and WALTHER, R. Versuche an Plattenbalken 318-63.
mit hoher Schubbeanspruchung. Berlin, Deutscher Ausschuss 13. LAUPA, A., SIESS, C. P. and NEWMARK, N. M. Strength in shear
flir Stahlbeton, 1962. No. 152. of reinforced concrete beams. University of Illinois, March
2. LEONHARDT, F. and WALTHER, R. Schubversuche an einfel- 1955. Experiment Station Bulletin No. 428.
drigen Stahlbetonbalken mit und ohne Schubbewehrung. 14. GURALNICK, s. A. Shear strength of reinforced concrete
Berlin, Deutscher Ausschuss fur Stahlbeton, 1962. No. 151. beams. Proceedings of the American Society of Civil
3. LEONHARDT, F. and WALTHER, R. Schubversuche an Platten- Engineers. Vol. 85, No. ST 1. 1959. pp. 1-42.
balken mit unterschiedlicher Schubbewehrung. Berlin, 15. FRANZ, G. and NIEDENHOFF, H. Die Bewehrung von Kon-
Deutscher Ausschuss fur Stahlbeton, 1963. No. 156. pp. solen und gedrungenen Balken. Beton- und Stahlbetonbau.
4. LEONHARDT, F. and WALTHER, R. Schubversuche an Durch- Vol. 58, No.5. May 1963. pp. 112-120 and 148. The
lauftriigern. Berlin, Deutscher Ausschuss fur Stahlbeton, reinforcement of brackets and short deep beams. London,
1963. No. 163. Cement and Concrete Association, December 1964. pp. 17.
5. LEONHARDT, F. and WALTHER, R. Beitrage zur Behandlung Library Translation Cj. 114.
der Schubprobleme im Stahlbetonbau. Beton- und Stahl- 16. CHOW, L., CONWAY, H. D. and WINTER, G. Stresses in deep
betonbau. Vol. 56, No. 12. 1961. Vol. 57, Nos. 2, 3, 6, 7 and beams. Proceedings of the American Society of Civil
8. 1963. Vol. 58, Nos. 8 and 9. 1963. Engineers. Vol. 118. May 1952.
6. LEONHARDT, F. and WALTHER, R. Schubversuche an Platten 17. LEONHARDT, F. and WALTHER, R. Torsions- und Schubver-
mit geschweiten Bewehrungsmatten. Beton- und Stahl- suche an vorgespannten Hohlkastentriigern. Dusseldorf,
betonbau. Vol. 59, Nos. 4 and 5. 1964. Beton- und Monierbau A.G., October 1964. pp. 35-86.
7. LEONHARDT, F. and WALTHER, R. Untersuchungen an wand-
artigen Triigern mit unterschiedlicher Bewehrung und
Belastung. Berlin, Deutscher Ausschuss fur Stahlbeton, Contributions discussing the above paper should be in the hands
1965. No. 178. pp. of the Editor not later than 30 April 1966.

198