Paper No.
6501
STEEL FRAMES WITH BRICKWORK AND CONCRETE INFILLING
bY
Malcolm Hohes, Ph.D., BSc., A.M.I.C.E.
Reader in Civil Engineering, Instituteof Technology, Bradford
For written dLrcussion
SYNOPSIS
It has been realizedfor some time past that the stiffness and strength of a steel
frame is greatly increased when it acts compositely with a wall panel. Several
full-size tests1V2 have been carried out to determine the behaviour of infilled
steel frames subjectedto racking or shear loading. In addition the Author has
carried out similar tests using small-scalespecimens.
In this Paper a simple procedure is suggested for calculating the ultimate
(racking)load and sidesway deflexion ofa steel frame with brickwork or concrete
infilling; a comparison is then made with thirteen test results.
NOTATION
THEFOLLOWING notation is used throughout the Paper:-
h, l, ddenote height, width, and diagonal of rectangular steel frame
t ,, thicknesswall
of idling
EZ, EZo ,, flexural rigidities of members of steel frame
H ,, horizontal shear force at failure
fc ,, compressivestrengthofinfillinginthedirectionoftheframediagonal
e', ,, strain in infilling at instant of failure
d
A ,, cross-sectionalarea of equivalent strut = t -3
F ,, compressiveforce in the equivalent strut = A .f,
INTRODUCTION
2. Some of the benefits arising from the plastic design of steel frameworks
may not be fully realized because of simultaneous reduction in elastic stability.
It has been suggested1 that an economical design solution may be obtainedby
including spine or end walls to eliminate side-sway and then to make use of
modern frame design methods which involveno side-sway. Under these circum-
stances great benefit can be derivedfrom a study of the stiffening effects of the
cladding of buildings.
3. The present Paper concernsthe stiffening and strengtheningeffect of wall
panels used compositely with steel frames subjected
to horizontal loading tending
to produce side-sway. The effectiveness ofsuch wall panels is admittedly greatly
1 The references are given on p. 478.
473
474 HOLMES ON STEEL FRAMES WITH BRICKWORK AND CONCRETE INFILLING
dependent on standards of workmanship and materials, but their contribution
to strength may be so great that even a small fraction of this contribution may
be worthy of consideration.
STEEL FRAME WITH CONCRETE INFILLING
4. Fig. 1 shows a steelframe ABCD with a concrete wall infill subjectedto a
horizontal shear force, H, giving rise to complementary vertical shear forcesof
W tan a. The resultants of these shear forces are compressive forces acting
at B and D along the diagonal BD. At failure thewall and frame will only be in
contact inthe vicinity of B and D, as shown in Fig. 1. This suggeststhe equiva-
/
/-
FIG.l.
lent structure shown in Fig. 2(a), where the wall panel has been replaced by an
equivalent strut BD, the cross-sectionalarea (A) of whichis yet to be determined.
The shear forces carried by the steel frame alone are shown in Fig. 2(b). At
failure the deflexion of the composite wall and frame is small in comparison
with the flexibility of the bare steel frame. Hence the change in length of the
diagonal BD of the steel frame, due to the forces shown in Fig. 2(b), may be
calculated elasticallyas:-
S,, = -
( H - FCOS
-
24EI
( t )
a)h3
l+-cota COS^ . . . . (1)
The shortening of the equivalent strut BD at failure is:
S,, = e‘& . . . . . . . . (2)
where e’c= strain in the concreteat failure, i.e.at the instant at which maximum
stress is developed.
HOLMES ON STEEL FRAMES WITH BRICKWORK AND CONCRETE INFILLING 475
5. Values of e‘, have been obtained by Baker3,and vary from 0.002 to 0.005
for strong and weak concrete respectively. Baker suggests a safe limiting value
of 0.002. From equations (1) and (2) the horizontal load to cause failure is
given by:
24EIe’J
H= +A& cosa . . . . (3)
where F has been replaced by the product of the cross-sectional area ( A ) of the
equivalent strut and the crushing strength of the concrete f,. The horizontal
deflexion at failure isgiven by:
S , = SBDCOS a = elcdcosa . . . . . (4)
d
6. A value for the cross-sectional area (A) of the equivalent strut of f3 is
obtained by considering the deflexion of a squarewall panel, assuming that the
reaction between the steel frame and infilling varies linearly from a maximum
at B and D to a zero valueat A and C.
it b /
S
../- (H-Fcora)
(a) EQUIVALENT STRUCTURE (b) SHEARFORCES ON STEEL FRAME ALONE
FIG.2.
STEEL FRAME WITH BRICKWORK INFILLING
7. Equations (3) and (4) may be used for brickwork i d l i n g , provided that
suitable values for e’, and f, are chosen. For brickwork a value of e‘,=0.005
is taken, i.e. a vaIue appropriate to weak concrete. As the compressive stress
is acting at an angle a to the brickwork courses, the “compressive strength” of
the brickwork (fc) is largely a functionof the properties of the mortar joints. A
value for f, of 450 Ib/sq. in. is suggested based on the available test results.
COMPARISON WITH TEST RESULTS
8. Results from thirteen testsare shown in Table 1, where they are compared
with values for ultimate
load and deflexion calculated from equations (3) and (4).
476 HOLMES ON STEEL FRAMES WITHBRICKWORK AND CONCRETEINFILLING
c? m
2 VI
00
m
VI W O
m mm
c-r
B
X
f
X
8
HOLMES ON STEEL FRAMES WITH BRICKWORK AND CONCRETE INFILLING 477
9. For the small-scale specimens 1 and 2, measured values for the concrete
infilling off, =2,400 lb/sq. in.and e', =0.005 were used; for the remaining speci-
menswithbrickworkinfillingvalues of f,= 450 lb/sq.in. and OQO5 were
d
assumed. The area of the equivalent strut (A)was taken as t 3 for all calcula-
tions.
10. The results for specimens 3 to 7 are taken from Wood's Paper', where
the frames consisted of concrete-encased steel joists with cleated connexions.
For calculation purposes the frames were assumed rigidly jointed but the concrete
encasement was neglected.
11. Specimen 8 consisted2 of a steel joist frame with pinned connexions and
no encasement, the frame itself offeringno resistance to horizontal shear loading,
the resistance being provided entirelyby the infilling. Hence the ultimate load
of this specimen was calculated omitting the first term in equation (3), i.e.
H = Af, COS a.
12. Where repeat tests were made on nominally identical specimens-Nos
1,2,3,5--the variations in the experimental failing loads were 16%, 16%, 14%,
and 3%respectively. With the exception of specimen 5c, where hollow clay
blocks were used,the maximum variationbetween experimental and theoretical
failing loadsis 14% with the majority of values well belowthis figure.
13. The theoretical deflexion at ultimate load is generally much lower than
the corresponding experimentaldeflexion, although where values are available
it is seen that the theoretical deflexion at ultimate load corresponds to the
measured deflexion at some point between 9 0%and 100% ultimate load.
CONCLUSIONS
14. Due to the variations in behaviour of nominally identical infilled steel
frames there seems little justification for proposing anything but the simplest
methods of analysis. In fact BenjaminandWilliams4,investigating the be-
haviour of reinforced concrete frames with infilling,state: "Numerous studies
were made involving exact mathematical procedures. The results were in no
way better or more reliable than the most elementary procedures of strength
of materials."
15. The tests recordedin Table 1 were carried out on frames wherethe value
of the angle cz was in the range 45"+ 10". This covers a fairly wide range of
rectangular frames, giving a rangefor ratio of sides h/l=0.7+1-4. For frames
lying far outside this range it is possiblethat some modificationto the procedure
suggested above wouldbe required.
16. Further tests were carried out on specimens similar to Nos 3 and 8, but
with openings provided in the brickwork infilling; these openings resulted in
reductions in ultimate load of 30% and 40% respectively. Had these openings
been suitably reinforcedby framing, however, it is probablethat little reduction
in ultimate load need have occurred.
ACKNOWLEDGEMENT
17. The Author would like to thank Mr M. E. Illingworth and his staff for
assistance in makingand testing the small-scale specimens.
478 HOLMES ON STEEL FRAMES WITH BRICKWORK AND CONCRETEINFILLING
REFERENCES
1. R.H. Wood, “The stability of tall buildings”. Proc. Instn civ. Engrs, vol. 11 (1958).
pp. 69-102.
2. J. R. Benjamin and H. A. Williams, “The behaviour of one-storey brick shear walls”.
Proc. Amer. Soc. civ. Engrs, vol. 84, No. ST4 (1958), Paper 1723 (30 pp.).
3. A. L. L. Baker, “The ultimate-load theory applied to the design of reinforced and
prestressed concrete frames”. 1st edition, London, Concrete Publications Ltd
(1956) (90 PP.).
4. J. R.Benjamin and H. A. Williams, “The behaviour of one-storey reinforced concrete
shear walls”. Proc. Amer. Soc. civ. Engrs, vol. 83, No. ST3 (1957), Paper 1254
(49 PP.).
The Paper, which was received on 26 July, 1960, is accompaniedby one sheet
of diagrams, from whichthe Figures in the text have been prepared.
Written discussion onthis Paper should be forwarded to reach the Institution
before 15 October, 1961, and willbepublished in or after February, 1962.
Contributions should not exceed 1,200 words.-SEC.
