http://www.paper.edu.

cn

Journal of Constructional Steel Research 59 (2003) 385–403

www.elsevier.com/locate/jcsr

Load carrying capacity of composite slabs with

various end constraints
S. Chen
School of Civil Engineering, Tongji University, Shanghai 200092, People’s Republic of China

Received 4 December 2001; received in revised form 14 April 2002; accepted 16 April 2002

Abstract

In many cases, the load carrying capacity of composite slabs depends on the shear-bond
resistance at the sheet-concrete interface. At the ultimate state, the tension forces in the hogging
region of a continuous composite slab are mainly transferred by the negative bending reinforce-
ment and the shear-bond resistance in the region do not significantly influence the load carrying
capacity of the slab. To identify the shear-bond action in composite slabs, seven simply sup-
ported one-span composite slabs and two continuous composite slabs were tested. Different
end restraints had been used in the simply supported slabs. The slabs with end anchorage of
steel shear connectors were found to bear a higher shear-bond strength than that of slabs
without end anchorage. The shear-bond strength was calibrated based on a linear regression
of the test results of the one-span composite slabs with end anchorage. The prediction of the
shear-bond resistance was also found in close agreement with the vertical shear force at the
onset of the initial shear-bond slip in the two-span continuous composite slabs. It is suggested
that the shear-bond slip model be reasonable to predict the shear-bond resistance of a continu-
ous composite slab. However, the shear span of the continuous composite slabs must be related
to the sagging region, which could be derived on an elastic analysis base, or simply taken as
0.8 L for the side span and 0.6 L for the interior span. At the onset of the initial shear-bond
slip, the mean ratios of the vertical shear force to shear-bond resistance (Ve / Vu) are 1.065 for
the one-span slabs and 1.165 for the two-span continuous composite slabs, which are on the
safe side. Because of the shear-bond failure at the sheet-concrete interface, composite slabs
would not be capable of developing full plastic moments in the spans.
 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Composite slab; Shear-bond; Load carrying capacity

E-mail address: [email protected] (S. Chen).

0143-974X/02/$ - see front matter  2002 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0143-974X(02)00034-2

转载
中国科技论文在线 http://www.paper.edu.cn
386 S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403

Nomenclature
A cross-section area (mm2)
I inertia moment of area (mm4)
L span length (mm)
M bending moment (kN m)
N numbers of the shear connectors
V shear force (kN)
P test load, end anchorage capacity (kN)
b breadth (mm)
d depth of the composite section to the centroid of the deck (mm)
f strength (N/mm2)
q distributed load (kN/m2)
s slip (mm)
t thickness (mm)
x depth of concrete (mm)
w dead load carried by the slab during concrete casting (kN/m)
d mid-span deflection (mm)
j load reduction factor (dimensionless)
g modified value of the shore condition (dimensionless)
l load factor (dimensionless)

subscripts

a related to anchored studs

c in compression, concrete
cr at the shear-bond slip initiation
e vertical shear
h hogging
max maximum
p profiled steel sheeting, plastic
s sagging, shear-bond
t test value
u ultimate, related to shear-bond strength

1. Introduction

The merits of using profiled steel sheeting composite floors have been recognized
for its efficiencies in construction and its higher load carrying capacity over the
traditional steel deck as shuttering. The shear-bond resistance is essential to the inter-
中国科技论文在线 http://www.paper.edu.cn
S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403 387

action between steel sheeting and concrete at the sheet-concrete interface, and gov-
erns the composite slab design.
Most literature on composite slabs concerns full-scale tests of proprietary decking
systems. The tested composite floor slabs are normally one-way, simply supported,
provided with nominal light mesh reinforcement. The test results are believed to
provide the lower-boundary estimation of the static load-carrying capacity, and are
accepted in the design philosophy for most codes and specifications such as British
code for composite construction (1982) [1] and EUROCODE 4 “Design of Com-
posite Steel and Concrete Structures” (1994) [2]. The latest Chinese technical speci-
fication for steel structure of tall buildings [3] documented the similar design clauses
for profiled steel sheeting composite floors.
In a simply supported composite slab, the whole span is in sagging, and the sheet-
concrete interface is in tension. For continuous composite slabs, the span consists
of sagging and hogging regions, where the sheet-concrete interface of the hogging
part is in compression. In the absence of theoretical models and relevant test data,
it is assumed that, only the longitudinal shear in the sagging span governs the load
carrying capacity of a continuous composite slab, because the tension forces in the
hogging region are mainly transferred by the negative bending reinforcement.
An industry funding research program has been carried out in Tongji University
recently to identify the influence of the longitudinal shear-bond to the load carrying
capacity of the continuous composite slabs.
The profiled steel sheeting is trapezoidal shaped which is provided by the domestic
steel industry. A series of one-way full-scale composite slab tests have been carried
out. The slabs were loaded one-span simply supported and continuously two spanned,
with various end constraints. In all the tests, longitudinal shear slip was found prior
to the attained maximum loading. The slabs with end anchored studs had showed a
substantially higher shear-bond resistance than the slabs with no end anchors, whose
shear-bond failure was brittle.
Compared with the simply supported single span composite slabs, continuous com-
posite slabs demonstrated the much higher carrying load, and less deflection. How-
ever, in magnitude, at the onset of the initial shear-bond slip, the vertical shear forces
at the exterior supports of the continuous slabs were close to the vertical shear forces
of a single spanned slab with the same span length.
To supplement the composite floor research, two composite slabs with steel fiber
reinforced concrete were also tested, and were compared with the composite slabs
light-mesh reinforced. The brittle shear-bond failure was found in these slabs with
no end anchorage.

2. Composite slab tests

2.1. Test specimens

Fig. 1 illustrates the geometric shape of the profiled steel sheeting. The properties
of the steel sheeting are given in Table 1. The average width of the slabs was 914
中国科技论文在线 http://www.paper.edu.cn
388 S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403

Fig. 1. Geometric shape of 3W-DECK sheets.

Table 1
Geometry and strength properties of the profiled steel sheeting

Thickness tp Area Ap Weight Width (mm) Ix (×106) Yielding Ultimate

(mm) (mm2) (kg/m) mm4 strength fp strength fu
(N/mm2) (N/mm2)

0.9 1170.8 9.188 914 1.152 275 380

1.2 1561.1 12.251 914 1.537 275 380

Note: Ix is inertia moment of area of the steel sheeting.

mm. An outline of the composite slabs is shown in Fig. 2, and the specimens are
grouped in Table 2.
Specimens A-1, A-2, A-3 and A-4 are simply supported one-span composite slabs
with end anchorage. In construction of the specimens, the end anchorage of the
composite slabs was in the form of steel shear studs, 19 mm in diameter, one or

Fig. 2. 3W-Deck composite slab specimens.

中国科技论文在线 http://www.paper.edu.cn
S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403 389

Table 2
Groups and the details of tested composite slabs

No. Sheeting Slab type Span L Span/depth End anchorage Negative

thickness (m) (per trough) reinforcement
(mm)

A-1 0.9 One-span 2.6 15.76 One 19 Non

dia.stud
A-2 1.2 One-span 2.6 15.76 Two 19 Non
dia.stud
A-3 0.9 One-span 3.2 19.39 One 19 Non
dia.stud
A-4 1.2 One-span 4.2 25.45 One 19 Non
dia.stud
A-5 0.9 One-span 2.6 15.76 Non Non
B-6 1.2 Two-spans 2.6/2.6 15.76/15.76 One 19 ⌽12@100
dia.stud
B-7 1.2 Two-spans 2.6/3.2 15.76/19.39 One 19 ⌽12@100
dia.stud
C-8a 0.9 One-span 2.6 15.76 Non Non
C-9b 0.9 One-span 2.6 15.76 Non Non

a
Concrete reinforced with steel fibre; Dramix RC 65/35BN, fibre content: 20 kg/m3.
b
Concrete reinforced with steel fibre; Dramix RC 65/60BN, fibre content: 20 kg/m3.

two each trough, welded through the steel sheeting to the top flanges of the steel
beams. The specimens A-5, C-8 and C-9 were also simply supported, one-span com-
posite slabs, but with no end anchorage.
The continuous slabs labeled B-6 and B-7, were reinforced with steel bars, 12
mm in diameter and 100 mm in spacing, on the top layer above the interior support.
They were also anchored with the shear studs welded through to the top of the steel
beams, at the ends and at the section above the internal support.
The surface of the deck was greased in all the specimens to eliminate the variable
effect of chemical bond between concrete and steel. Concrete was cast with mid-
span props for specimen A-4. No props were used during the construction for the
other specimens. All specimens except for C-8 and C-9, which were reinforced with
steel fibre, were light steel mesh reinforced.
The specimens were loaded by a hydraulic jack system via the spread beams to
simulate uniformly distributed load. The test rigs are shown in Fig. 3.
Deflections were measured using displacement transducers, two each at the mid-
span, symmetrically centerline of the section, and two each at the 1/4 span. Two
displacement transducers at each end of the specimens were placed to measure the
shear slip between concrete and steel sheeting.
The average concrete cube strength measured is 31 N/mm2, and the cube strength
for the steel fiber reinforced concrete is 33 N/mm2. The steel fibers used are end
hooked, labeled Dramix RC 65/60 BN and RC 65/35 BN, 65 in the nominal aspect
中国科技论文在线 http://www.paper.edu.cn
390 S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403

Fig. 3. Test rigs.

ratio, 60 and 35 mm in length, respectively. The tensile strength of the steel fiber
is 1100 N/mm2.

2.2. Test results

The specimens were loaded with an increment of 3 kN, up to the stage, when the
longitudinal slips were detected, and were then loaded gradually up to the maximum
load. The tests were terminated when either the load dropped 20% from the
maximum values, or the mid-span deflections were close to one fiftieth of the span.
For specimens A-1, A-2, A-3 and A-4, one-span and simply supported, with end
anchors of shear studs, the load-deflection curves behaved almost linear before the
shear-bond slip initiation, which was detected at the slab ends. Increasing the load
further, the shear-bond slippage developed, accompanying de-bonding cracking at
the sheet-concrete interface. The load-deflection curves then became non-linear.
Specimens A-5, C-8 and C-9 were also simply supported one-span slabs, but with-
out end anchorage. The slabs behaved very brittle. After initial slip, the load dropped
sharply accompanying flexural cracks in concrete. The specimens with steel fiber
reinforced concrete (C-8 and C-9) were found equivalent to specimen A-5, which
was reinforced with steel mesh.
In all the tests, shear-bond failure was found, but no concrete split and shear stud
failure occurred at the ends of the slabs. The maximum sagging moments of the slabs
were found smaller than Mp, the plastic moment, which was calculated assuming full
component interaction at the sheet-concrete interface. Table 3 shows the test results
of the one-span composite slabs. In Table 3, P and M are the exerted test load and the
mid-span bending moment, respectively, with subscripts cr and max corresponding to
 中国科技论文在线

Table 3
Test results of the one-span composite slabs

Specimen At longitudinal shear-bond slip When the central deflection d ⫽ L / 250 At the maximum load stage
initiation
Pcr (kN) Mcr (kN m) d(mm) P (kN) Ms (kN m) s (mm) Pmax (kN) Mmax d(mm Mp Mmax/Mp
(kN m) (kN m)

A-1 52.7 20.4 5.68 52.7 20.4 0.60 81.2 31.5 47.88 33.7 0.935
A-2 61.0 23.6 6.32 63.3 24.5 0.56 110.9 42.9 47.27 43.1 0.995
A-3 42.8 20.9 7.83 38.3 18.7 0.67 44.3 21.6 33.72 33.7 0.641
A-4 46.9 29.9 16.68 46.9 29.9 0.07 47.1 30.0 16.64 43.1 0.696
A-5 39.8 15.4 5.16 23.9 9.3 1.37 44.8 17.4 / 33.7 0.516
C-8 32.5 12.6 6.89 18.1 7.0 0.30 32.5 12.6 / 33.2 0.380
C-9 35.8 13.9 3.14 21.8 8.5 1.44 35.8 13.9 / 33.2 0.419

Note: Mp is the positive plastic moment of the cross-section based on the measured dimensions; dis the deflection in the mid-span.
S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403
391
http://www.paper.edu.cn
中国科技论文在线 http://www.paper.edu.cn
392 S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403

the initial shear-bond slip and the maximum load, respectively. S is the slip measured
at the ends of the slabs.
Compared with the simply supported composite slabs, continuous composite slabs
were found be able to sustain higher load but with less deflection. Fig. 4 shows the
load-deflection curves of the tested composite slabs. Similar shear slips were
observed at Pcr, in the continuous composite slabs, however, until the load increased
up to quite high, no flexural cracks were found, and the deflections in the spans
were small. Table 4 shows the test results of the continuous composite slabs. The
similar notations as the single span tests are used.
Two tests had been done on specimen B-7. In test B-7(a), loads were exerted on
the two spans of the specimen, and test was terminated when the longer span failed
at the maximum load. In test B-7(b), load was exerted on the shorter span only, and
the longer span of the slab remained unrestrained vertically. As the longer span of
the specimen was already in a failure mode, a negative moment owing to the weight
of the span was expected at the internal support. In Table 4, load on each span is
P/2 in tests B-6 and B-7(a), and P for one-span load in test B7-(b)
The mid-span span moments (Ms) and the vertical shear forces (Ve) at the exterior
supports of the continuous composite slabs are given in Table 5. When shear-bond
slip initiated, the loads resisted by the two-span composite slabs were found higher
than that resisted by specimens A-1 and A-2, but the vertical shear forces at the
exterior supports and the mid-span moments were lower than those in the one-span
composite slabs. Substantial sagging moments were found having developed in the
continuous spans after shear slip initiation.

Fig. 4. Comparison of load-deflection curves of composite slabs.

 中国科技论文在线

Table 4
Test results of the two-span continuous composite slabs

Specimens When longitudinal shear-bond slip When the central deflection d ⫽ L / 250 At the maximum load stage
initiation at the slab ends

Pcr (kN) d1 (mm) d2 (mm) P (kN) s1 (mm) s2 (mm) Pmax (kN) d1 (mm) d2 (mm)

B-6 149.6 3.28 3.53 214.1 1.18 0.79 293.6 19.62 18.47
B-7(a) 140.8 2.05 5.72 155.5 0 0.884 239.5 2.03 41.00
B-7(b) 61.9 7.35 / 74.6 0.49 / 128.1 44.5

Note: The subscripts 1 and 2 represent span 1 and span 2.

S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403
393
http://www.paper.edu.cn
 394
中国科技论文在线

Table 5
The sagging moments and shear forces in the composite slabs

Specimens P ⫽ Pcr d ⫽ L / 250 P ⫽ Pmax

Ms (kN m) Ve (kN) Ms/Mp Ms (kN m) Ve (kN) Ms/Mp Ms (kN m) Ms/Mp Mp (kN m)

A-1 20.4 26.35 0.605 20.4 26.35 0.605 31.5 0.935 33.7
A-2 23.6 30.50 0.548 24.5 31.65 0.568 42.9 0.995 43.1
B-6(L1) 15.71 25.06 0.371 22.48 35.86 0.531 37.99 0.898 42.3
B-6(L2) 15.71 25.06 0.371 22.48 35.86 0.531 37.99 0.898 42.3
B-7 a (L1) 12.46 21.77 0.295 13.76 24.04 0.325 26.49 0.626 42.3
B-7 a (L2) 22.67 24.29 0.536 25.04 29.25 0.592 43.86 1.036 42.3
B-7 b (L1) 20.35 26.37 0.481 24.53 31.78 0.580 45.99 1.087 42.3

Note: (1) Negative bending resistance at the internal support M⬘p ⫽ 48.8kN / m; (2) specimen B-6 L1 ⫽ L2 ⫽ 2.6m; (3) specimen B-7 L1 ⫽ 2.6m, L2 ⫽
3.2m.
S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403
http://www.paper.edu.cn
中国科技论文在线 http://www.paper.edu.cn
S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403 395

2.3. Discussions

In all the tests, the load carrying capacity of composite slabs was governed by
the longitudinal shear-bond failure. End anchorage was confirmed to have the posi-
tive effect on the shear-bond strength of the slabs. At the onset of the initial shear-
bond slip, the vertical shear forces at the supports increased by 32.4%, from 19.9
kN for specimen A-5 to 26.35 kN for specimen A-1, which were the same in span
length but differed in the end constraints.
The shear-bond failure of composite slabs is governed by the shear-bond slip at
the sheet-concrete interface rather than the strength of the end constraints. Different
in thickness of the sheeting and in the numbers of shear connectors per trough, an
increment by 15.7% in the vertical shear was found from A-1 to A-2 at the onset
of the initial shear-bond slip. However, when the shear-bond slip occurred, the shear
force transmitted to the studs was very small, and two studs per trough in the same
line parallel to the shear-bond force in specimen A-2, were much stiffer than the
sheet-concrete interface of the slabs. Therefore, increase of the numbers of shear
studs per trough would not enhance the shear-bond strength distinctly.
Specimens A-5, C-8 and C-9 were spanned with no end anchorage. At the onset
of the initial shear-bond slip, vertical shear forces at the supports were lower than
that of the specimens with end anchorage. Reinforced with light steel mesh, specimen
A-5 subjected to a slightly higher load than specimens C-8 and C-9, which were
reinforced with steel fibre, however, all these specimens behaved very brittle. The
diverse loads at the slip initiation would be errors caused in the specimen construc-
tion, for instance the final surface greasing. Generally, shear-bond slip developed
much faster than that in the anchored specimens. The specimen reinforced with the
longer fibre likely subjected to a higher load and less deflection than the specimen
with shorter fibre.
Continuous internal supports were provided in specimens B-6 and B-7. Compared
with the simply supported slabs, there are sagging and hogging portions in these
slabs. When the shear-bond slip was detected at the ends of the specimens (C-6 and
C-7), the vertical shear forces at the exterior supports were slightly lower than those
measured in specimens A-1 and A-2, but the load in each span was higher than that
exerted on the simply supported composite slabs.
Three aspects should be considered. First, in continuous composite slabs, existence
of the negative bending at internal supports will reduce the vertical shear force at
the exterior supports. Second, reduced deflections in continuous spans will ease the
shear-bond slippage at the concrete-sheeting interface. Finally, as tension force is
mainly transmitted by the negative bending reinforcement in the hogging region, the
shear-bond, which is in compression in this region, will not influence the resistance
of the cross-section of the slab. Further tests of isolated specimens in hogging may
be necessary to identify this shear-bond action. But in the present work, it is simply
assumed that only the shear-bond in sagging governs the load carrying capacity of
the continuous composite slabs.
中国科技论文在线 http://www.paper.edu.cn
396 S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403

3. Shear-bond strength of the composite slabs with end anchorage

Initiation of shear-bond slip at the sheet-concrete interface reveals a loss in com-
posite action. In simply supported composite slabs, the shear-bond slip develops
with load increment, leading to flexural cracks in concrete, and final failure of the
composite slabs.
In slabs with no end anchors, the shear-bond slip occurred at the approximate
maximum load. The mean value of the vertical shear at the onset of initial shear-
bond slip for the slabs with no end anchors is 18 kN, 8.4 kN lower than that of
specimen A-1, which has the same steel profile but with end anchors of steel studs.
When the shear-bond slips occur, the vertical shear force of specimen A-2 is approxi-
mately 4 kN higher than that of specimen A-1, basically owing to the difference in
the sheeting thickness and in the numbers of studs per trough between the two speci-
mens. The effect of the end anchorage to the shear-bond resistance can be explained
that a restrained tensile strut in the steel sheeting is formed during flexure bending
in the slab, which will influence its shear-bond resistance. As end anchorage is usu-
ally more flexible than the longitudinal shear-bond interface, which is relatively
brittle, it cannot combine directly the effects of the shear-bond and the end anchor-
age.
Shear-bond tests must be carried out to calibrate different types of steel decks.
Normally, the slabs are tested with no shear connectors, so that contribution of the
end anchor is ignored. Jolly and Lawson [4] proposed a method to appraise the
contribution of the end anchor to the shear-bond resistance of the composite slabs.
It is recommended that the shear-bond resistance of a composite slab with end
anchorage of shear studs can be assessed as follows:
Vu ⫽ Vs ⫹ 0.5Va (1)
where Vu is the total vertical shear capacity of the slab, Vs the shear-bond capacity
of the slab, depending on the interaction between the sheeting and concrete only,
and Va is the shear capacity due to the end anchorage provided by the shear connec-
tors, evaluated from the following formula:
Va ⫽ NPa(d⫺xc / 2) / Ls (2)
where N is the number of shear connectors attached to each sheet end, d the effective
depth from the compression fiber to the steel deck centroid, xc the depth of concrete
in compression (or the upper 20 mm of the slab as a conservative assumption), Ls
the shear span, and Pa is the end anchorage capacity.
In Eq. (1), the contribution by the end anchor to the shear-bond strength is 0.5Va.
Table 6 shows the evaluated shear capacity due to the end anchorage in specimens
A-1 and A-2. For a comparison, the measured shear capacity by the end anchorage
derived from the test results of specimens A-1, A-2 and A-5 is also given. Specimens
A-1 and A-5 are identical except for the end constraints. The difference in the meas-
ured vertical shear force at the onset of the initial shear-bond slip, between the two
specimens is 6.45 kN, which is smaller than 11.85 kN (0.5Va), as predicted by Eq.
(2). Probably the reduction factor in Eq. (1) should be reduced from 0.5 to 0.3 or
less to prevent an unsafe prediction.
中国科技论文在线 http://www.paper.edu.cn
S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403 397

Table 6
Measured difference in the shear-bond capacity and 0.5 Va

Specimen Measured d (mm) xc Pa N Ls 0.5 Va Vu⫺Vs kN

Vu (kN) (mm) (kN) (mm) (kN)

A-1 26.35 128.50 20 50 2 500 11.85 6.45

A-2 30.50 128.65 20 50 4 500 23.70 10.60a
A-5 19.90 128.50 20 Non Non 500 Non Non

a
The shear-bond capacity of specimen A-5 is deducted for an approximation.

The shear-bond resistance of a composite slab is related to steel sheeting shape,

deck embossments, span, end-anchorage configuration and etc. The shear-bond
behaviour of composite slabs is normally expressed in curves of shear vs. slip based
on the shear-bond pull-out tests [5]. In design practice, it is more likely to define
the shear-bond resistance in an empirical formula relating Ve, the vertical shear force
of the slabs, to the test-related parameters. Based on the test results of the one-span
composite slabs with end anchorage, a linear regression of the shear-bond resistance
for the composite slabs is expressed in terms of variables given in Eq. (3):

冋冉
Vu ⫽ j bd 84.665
Ap
bLs
⫹ 0.0221冑fc ⫹
2 冊
gw1L
册 (3)

where Vu is longitudinal shear-bond resistance, j the reduction factor (j ⫽ 0.8), b

the slab width, d the effective depth of the slab from the compression fiber to the
steel deck centroid, Ap the cross-section area of the steel deck, Ls the shear span, fc
the design compressive strength of concrete, g the modified value of the shore support
conditions, L the span of the slab, and w1 is the dead load carried by the slab during
concrete casting.
The form of Eq. (3) is as the same as the shear-bond capacity formula for a
standard composite slab with no end anchor. End anchor has positive effect on the
shear-bond capacity of the slabs, and the effect is included in the coefficients in the
formula, valid only when the composite slabs are restrained with end anchorage of
shear studs, 19 mm in diameter. In term of the effective end anchorage, no end
anchorage should fail before the shear-bond failure at the sheet-concrete interface.
Compared with the shear connection failure in the composite beams with profiled
steel sheeting, investigated by Johnson and Yuan [6], the shear-bond failure of the
sheet-concrete interface of a composite slab occurs at a much smaller slip (normally
less than 0.5 mm) than that the end anchors can subject to. Neither concrete split
nor shear stud failure was found in all the tests. It is the shear-bond slip rather than
the strength of the end anchored studs that governs the contribution of the end con-
straints to the shear-bond resistance in the composite slabs. For the typical steel deck,
the favorable position of the studs is at the centre of the trapezoidal trough. Fig. 5
shows the arrangement of shear studs of the tested specimens and the typical slip
curves of composite slabs and of the shear studs. It is identified that increase of the
numbers of the shear studs per steel trough will not reduce the initial slip at the
中国科技论文在线 http://www.paper.edu.cn
398 S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403

Fig. 5. Shear-stud end anchorage and approximate vertical shear-slip relationship for shear-bond and
end anchorage.

shear-bond interface, at which the force transmitted to the studs is low. A better
expression would be derived if the stud parameters were introduced, but more tests
would be needed.
For continuous composite slabs, only the sheet-concrete interface of the sagging
regions is in flexural tension, but over the hogging regions, commonly 20% of the
span, the sheet-concrete interface is in compression. In the sagging region, shear-
bond failure means a loss of composite action between steel sheet and concrete, and
will govern the bending resistance of the slab. In the hogging region, however,
because the tension forces are mainly transferred by the negative bending reinforce-
ment, the shear-bond, that could influence the initial shear-bond slip in the region,
would not affect the ultimate hogging bending resistance of the composite slabs.
Therefore it does not govern the load carrying capacity of the continuous com-
posite slabs.
Two problems may arise when applying the shear-bond test results of the simply
supported slabs to the continuous composite slabs. First, it may be difficult to predict
the position of the contraflexure points, between which the sagging potion is defined.
Second, the bending moment distribution along the spans of continuous composite
slabs may not be capable of developing full plastic moments both in sagging and
hogging regions under the shear-bond slippage.
To appraise the load carrying capacity, Vu, the shear-bond resistance of the sagging
region of the continuous composite slabs (specimens B-6 and B-7), are calculated
based on Eq. (3), and given in Table 7. Ve, the vertical shear forces at the exterior
supports, recorded at the onset of the initial shear-bond slip are also given. For a
comparison, the values of Vu and Ve of the one-span composite slabs are listed in
中国科技论文在线 http://www.paper.edu.cn
S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403 399

Table 7
A comparison of the shear-bond strength between calculation and tests

Specimen Ve (kN) testing value Vu (kN) calculated value Ve/Vu

A-1 26.35 23.55 1.119

A-2 30.50 28.11 1.085
A-3 21.40 20.63 1.037
A-4 23.45 23.05 1.017
B-6 25.06 23.14 1.083
B-7(a) 24.29 19.10 1.272
B-7(b) 26.37 23.14 1.139

the table. To derive the shear-bond resistance, the measured shear span, is identified
based on the sagging region of the specimens both of one-span and two-span.
From Table 7, it is found that the mean values of Ve/Vu are 1.065 for the simply
supported one-span slabs and 1.165 for the two-span continuous composite slabs.
The difference may relate to the hogging moment transferred at the internal support,
which has a restrain effect on the longitudinal slip. However, the value Ve/Vu of the
continuous slabs is on the safe side.

4. Appraisal of load carrying capacity of continuous composite slabs

The internal support can transmit negative moment, as far as the negative steel
reinforcement is placed in. In the tests of the continuous composite slabs, concrete
crack widths at the top of the slabs in hogging bending were less than 0.1 mm. It
is hopefully assumed that the full plastic negative flexural resistance would be
reached at the internal support, but actual situation may not be so. The contribution
of hogging bending transfer at the internal support to the load carrying capacity in
a continuous composite slab is analyzed as the following.
Fig. 6 shows a two-span continuous slab, under a uniformly distributed load. In
static equilibrium, the sagging moment in the mid-span Msi, and the vertical shear
force at the exterior support Vi are derived as:
qL2i Mh
Msi ⫽ ⫺ (4)
8 2
q Mh
Vi ⫽ Li⫺ (5)
2 Li
as far as the shear-bond resistance satisfies:
ViⱕVu (6)
where Mh the hogging moment at the internal support, q uniformly distributed load,
Vu the shear-bond resistance based on the sagging region only, and Li is span of the
中国科技论文在线 http://www.paper.edu.cn
400 S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403

Fig. 6. Analytical model of a two-span composite slab.

slab. As the load carrying capacity is governed by the shear-bond failure, qv, at
which the longitudinal shear slip occurs, can be rewritten as the following:
2Vu 2Mh
q vⱕ ⫹ 2 (7)
Li L1
If the shear-bond slip is restrained, for a two equal span continuous slab, we have
Mh ⫽ qvL2 / 8 at the internal support and lead to: qv ⫽ 8Vu / 3L,Ms ⫽ VuL / 6 and
MhⱖVuL / 3.
The similar expressions can be derived for a one-span slab as: qv ⫽ 2Vu / L, and
Ms ⫽ VuL / 4. The comparison shows that if the slabs possess the same shear-bond
strength, the load carrying capacity of a two-span continuous composite slab is 1.333
times of that for a simply supported composite slab. The mid-span sagging moment
of the two-span slab is reduced by 1/3 of that for the one-span slab. The analyses
agree with the test results of A-1 and B-6 at the onset of the initial shear-bond slip.
Increasing load up to lqv, where l ⬎ 1.0, the shear-bond slip will develop, and
the sagging and the hogging moments will also increase. Let q in Eqs. (4) and (5)
replaced by lqv, the ultimate load of the slab can be derived with the assumption
that the slab is capable of developing full plasitic resistances both in sagging and
in hogging:
8Ms ⫹ 4Mh
lqv ⫽ (8)
L21
or
2Vi Mh
lqv ⫽ ⫹ 2 (9)
Li L1
中国科技论文在线 http://www.paper.edu.cn
S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403 401

The derived load factor and the tested load factor for specimens B-6 and B-7 are
given in Table 8. In Table 8, lu is the theoretical ultimate load factor, assuming the
slabs are capable of plasitic resistance, both in positive and negaitive bending. lt is
the maximum load factor derived from the test results. The theoretical values are
close to the experimental values.
Test results suggested that the shear-bond slip model of simply supported com-
posite slabs be reasonable to predict the shear-bond strength of a continuous com-
posite slab. But the shear span must be related to the sagging region, which can be
derived on an elastic analysis base. After initial shear-bond slip, a substantial sagging
moment can develop further until the maximum loads are attained. Because of the
shear-bond slip along the sheet-concrete interface, the continuous composite slabs
are not capable of developing full plastic moments in the span length.

5. Comments on the continuous composite slab design

In practice, design engineers will be provided with the allowable superimposed

load or the coefficients related to shear-bond behaviour of the composite slabs from
the steel deck manufacturers, who have to carry out performance tests on their pro-
ducts. The loads provided by the manufacturers are basically based on the simply
supported spans, whatever the slabs are one spanned or continuous spanned. Conse-
quently, load carrying capacity of the continuous composite slabs is underestimated.
It is proposed in the Chinese design specification (JGJ 99-98) [3] that the negative
reinforcement is essential if the composite slab is designed continuously over the
supported steel beams. In construction, to enable a composite beam, the transverse
reinforcement should be sufficient to resist the longitudinal shear in the composite
beam, which is formed by the composite slab and the steel beam underneath. This
reinforcement together with steel light mesh in the slab could form a negative bend-
ing contribution over the continuous support.
The shear-bond strength based on the simply supported slabs can be used to predict
the shear-bond resistance of a continuous composite slab. The shear-span is related
to the sagging region of the slab, which can be determined by an elastic analysis.
The existence of the shear-bond in hogging region does not affect the load carrying
capacity of the continuous slabs. For simplicity, under the uniformly distributed load,
the sagging region of a continuous slab can be taken as 0.8 L for the side span and

Table 8
Load factors of the tested two-span composite slabs

Specimen Load type span (m) lu lt lt/lu

B-6 Equivalent UDL 2.6 2.098 1.963 0.936

B-7(a) Equivalent UDL 3.2 1.666 1.701 1.021
B-7(b) Equivalent UDL 2.6 1.929 2.069 1.073

Note: UDL is uniformly distributed Load.

中国科技论文在线 http://www.paper.edu.cn
402 S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403

0.6 L for the interior span, where L is the slab span. However, continuous composite
slabs will be restrained to develop the full plastic resistance both in sagging and
hogging regions because of the shear-bond strength restriction.

6. Conclusions

The shear-bond behaviour of composite slabs with profiled steel sheeting has been
investigated experimentally. Longitudinal shear-bond slip is found in both one-span
and two-span steel profiled sheeting composite slabs. The end anchorage of steel
studs is found to be effective to enhance the shear-bond resistance of the composite
slabs. To enable an effective end anchorage, however, it is the shear-bond slip rather
than the strength of anchored studs that governs the contribution of the end con-
straints to the shear-bond resistance in composite slabs. The shear-bond resistance
of the steel profiled sheeting composite slabs with end anchors of steel shear studs
is derived based on a linear regression of the simply supported composite slab tests.
Prediction of the shear-bond strength is found in close agreement with the test results
of the continuous composite slabs.
In the simply supported composite slabs, the steel fiber reinforced concrete is
found to be equivalent to the steel reinforced concrete. To enable of continuous
composite slabs, negative reinforcement is necessary to transmit the hogging bend-
ing.
Generally, the load-carrying capacity of continuous composite slabs is relevant to
the shear-bond strength in the sagging regions, which can be predicted based on the
simply supported slab tests, and the hogging bending resistance at the internal sup-
port. However, the slabs would not be capable of developing full plastic moment in
the spans.

Acknowledgements

The author acknowledges with thanks financial assistance and the provision of
materials by Huayin Steel Work Company, and the assistance of the technicians at
the Building Structural Laboratory of Tongji University.

References

[1] Structural use of steelwork in building, Part 4:code of Practice for design of floors with profiled steel
sheeting. London: British Standards Institution; 1982.
[2] DD ENV 1994-1-1. Eurocode 4. Design of composite steel and concrete Structures. Part 1—general
rules and rules for buildings. London: British Standards Institution; 1994.
[3] Chinese Standard of Construction. Technical specification for steel structure of tall building, JGJ 99-
98; China Building Industry Press, Beijing, 1999.
[4] Jolly CK, Lawson RM. End anchorage in composite slabs: an increased loadcarrying capacity. Struct
Eng 1992;70(11):202–5.
中国科技论文在线 http://www.paper.edu.cn
S. Chen / Journal of Constructional Steel Research 59 (2003) 385–403 403

[5] Daniels BJ, Crisinel M. Composite slab behavior and strength analysis Part II: comparisons with test
results and parametric analysis. J Struct Eng, ASCE 1993;119(1):36–49.
[6] Johnson RP, Yuan H. Models and design rules for stud shear connectors in troughs of profiled sheet-
ing. Proc Instn Civ Eng Structs Bldgs 1998;128(3):252–63.