›

Part 2
www.thestructuralengineer.org

Technical TheStructuralEngineer 25
Composite/Steel compendium February 2014

Composite and Steel

Construction compendium
Part 2: Shear connection in composite beams
This article is part of a series
that will gradually build to form a
S Figure 1
Shear stud in solid and composite slabs

Composite (steel-concrete) and

Steel Construction compendium.
Written by leading experts
from the SCI, the articles aim
to provide a real insight into
composite and steel construction,
to help practicing engineers get
the most from these materials.

a. Shear stud in solid slab

a. Shear stud in solid slab
Introduction
The first article in this series set the scene by talking about
developments of products and design codes. It highlighted how
these developments are not as coordinated as one might expect.
In this second article we look specifically at the issue of shear
connection. More general information on composite construction
can be found at: www.steelconstruction.info.

Shear connection in steel-concrete composite beams has been a

subject of some interest over recent years, with several international
research groups reconsidering both the resistance and the ductility

SCI
of shear studs. b. Shear stud fixed through profiled sheeting

Testing and analysis carried out in the UK has shown that the
resistance values given in Eurocode 4 are reasonable, although the In BS 5950-3.1, which also appeared in the mid-1980s, values
assumptions made about ductility in that code may be conservative for shear connection resistance are defined in a simple table as
in some cases. The work also resulted in some reduced shear stud a function of stud (shank) diameter and concrete strength. In
resistances being included in an amended version of BS 5950-3.11, Eurocode 4 the connection resistance in a solid concrete slab is
and highlighted some areas where previous UK practice needs defined by two equations; one considering the strength of the steel
to be revised. stud itself and the other considering the surrounding concrete as
governing failure. The lower of the two values is adopted. When,
It is important for designers to clearly understand three things about the as is normally the case for buildings, profiled decking is present,
shear connection in composite beams. Firstly, what is an appropriate a reduction factor is applied to the solid slab value. The reduction
shear stud resistance to use in a given situation? Secondly, how factor is a function of the stud and decking geometry.
should the shear connection, comprising stud plus surrounding
concrete and reinforcement, be detailed? Thirdly, what should you Figure 1 shows that as well as the stud resistance potentially
do if the total shear connection that a beam can accommodate is reducing, the physical behaviour also changes when decking is used.
less than the minimum degree of shear connection required? The volume of concrete most in compression moves from the base
of the stud to nearer its head. The idea of reducing a solid slab value,
The purpose of this article is to help designers gain that clear rather than giving values directly for composite slabs, betrays the
understanding, based on the most up-to-date technical background. background to Eurocode 4. The shear connection resistance values
are defined by empirical equations, based on many hundreds of tests
Shear connection resistance from the late 1970s to the mid-1990s. Many of these tests were on
Codified guidance solid slabs, which was the typical form of construction, not just for
Eurocode 4 underwent a lengthy development period, finally bridges but also for buildings, during the early part of that period.
appearing as a full standard, EN 19942, in 2004. Useable drafts The relatively small number of tests on composite slabs, inevitably
had been available, as ENV documents, as early as the mid-1980s. used the decking that was available at the time.

TSE26_25-27 CSCC-2 v3.indd 25 23/01/2014 08:16

 ›
Part 2

26 TheStructuralEngineer Technical
February 2014 Composite/Steel compendium

W Figure 2
Concrete failure surface
in push test with two studs and
Tests have shown that there is no need to place the reinforcement
so low, and SCI recommends that the Eurocode rule does not
transverse trapezoidal decking
apply to typical UK composite slabs with only one layer of fabric.
For slabs that have only a single layer, complementary guidance
can be used.

SCI
Placing reinforcement below the head of the studs so that it
crosses the potential concrete failure surface (Figure 2) will
improve performance, although tests have shown that even when
the fabric is above the studs, the connection resistance and
ductility may still be adequate. This is reassuring, as common site
N Figure 3
Geometry of typical 80mm
trapezoidal decking
practice is to place the fabric at nominal cover from the upper
surface of the slab (for reasons of crack control, and for it to act
more efficiently as fire reinforcement). The 2010 amendment to BS
5950-3.1 conservatively assumes this scenario.
Latest findings
Over the past decade or so, the decking profiles used in the UK Number of studs in a transverse trough
have changed. Whereas the early re-entrant shapes formed slabs Code requirements
that were very similar to a solid slab, modern trapezoidal profiles BS 5950-3.1 as amended in 2010, only gives reduction factors for
have been developed to reduce the volume of concrete and use with either one or two studs per trough, and it is therefore
therefore enable greater spanning ability, particularly during the implicit that more than two should not be considered. The wording
construction stage when the decking acts as permanent formwork. of Eurocode 4 is less clear. The reduction factor is defined as
Testing of full scale beams and local ‘push out test’ specimens, a function of the number of studs per trough, but this number
adopting modern decking profiles in a transverse orientation, has (according to EN 1994-1-1 6.6.4.2(1)) is ‘not to exceed two in
revealed two headline results: calculation of the reduction factor kt’. Numerous designers have
interpreted this as meaning if they have more than two studs, then
• Resistances are lower than previously recommended in BS 5950- the resistance of each is the same as if only two were present
3.1, hence the amendment issued in early 2010. The resistances (because a higher value than two should not be used ‘in calculation’).
for single studs are close to those presented in EN 1994. The Consideration of Fig. 2 shows why such an interpretation is
resistances for double studs are lower than those given in EN 1994, erroneous – were three (similarly spaced) studs to have been
but that is because of less onerous detailing requirements (fabric present in that test, the concrete failure surface would not have
does not have to be below the heads of the studs to validate the been 50% larger.
BS 5950-3.1 resistances). The results did not suggest the need for
any immediate changes to EN 1994 UK practice
• The ductility, defined in terms of slip capacity, is considerably It has not been uncommon to find three or more studs used
greater than the 6mm considered when developing the rules for per trough, normally in an endeavour to satisfy the requirements
minimum degree of shear connection in both codes for a minimum degree of shear connection, rather than to
increase the moment resistance of the composite beam.
Positioning of fabric reinforcement Given the considerations noted, it is clear that not much will have
Code requirements been achieved by using so many studs, even though the designer
BS 5950-3.1 requires the studs to project at least 35mm above may have thought differently.
the decking. It explicitly defines the height of decking as the
dimension to the shoulder, therefore not including the c.10mm What heights to consider?
deep stiffener present on the top of the rib for many decks. EN A final point on detailing that requires some clarification refers to
1994-1-1 6.6.5.8(1) requires the studs to project at least 2φ (where the dimensions that should be used for:
φ is the diameter of the shank of the stud) above the top of
decking. It is not clear from the code which height of stud to use, • The height of the decking
but SCI recommends the nominal (pre-welded) height for clarity • The height of the shear stud
and simplicity (to be compared with the height to the shoulder
of the decking). Eurocode 4 also requires that the ‘bottom layer’ Most modern trapezoidal decking has two heights, one to the
of fabric (or mesh as it is called in the Eurocode) should be at shoulder, and one to the top of a stiffener in the upper flange
least 30mm below the head of the stud (EN 1994-1-1 6.6.5.1(1)). (Figure 3). When Eurocode 4 was written there was only one
However, this requirement should be read in the context of a height, so there is only one definition and one variable. It is
document that has many diagrams showing solid slabs with top recognised that a designer should use the height to the
and bottom fabric. shoulder when checking requirements for shear stud
detailing and resistance (Fig. 1). When working out the depth
UK practice of solid concrete, the total depth of decking should be
A typical UK combination of a nominal 60mm trapezoidal deck, taken into account.
which has a total depth in excess of 60mm, with a 95mm as-
welded shear stud, satisfies the requirement from BS 5950-3.1 Defining the height of a shear stud is more complex. Stud height
noted previously, leaving a distance of approximately 15mm is defined in different ways by different stud manufacturers, and is
between the head of the stud and the very top of the decking. With not clearly defined in all codes. Moreover, when a stud is welded
such products, the apparent Eurocode requirement to place the directly to a beam flange, the loss of height (length) due to welding
fabric 30mm below the head cannot be satisfied. is about 5mm less than if it were welded through profiled decking

TSE26_25-27 CSCC-2 v3.indd 26 23/01/2014 08:16

 www.thestructuralengineer.org

27

W Figure 4
Non-
composite and
fully composite
beams (with
infinitely
stiff shear
connection)

(as is common practice). So a given stud has more options than A question often posed, is whether one can specify an effective
simply its height before and after welding. concrete flange width which is less than that specified in the code.
By doing so, the number of studs needed to achieve full shear
BS 5950-3.1 as amended in 2010 is clear, stating that the nominal connection would reduce (there is less concrete to compress), so
‘before welding’ stud height should be used when determining the a given number of studs would appear to provide a greater degree
reduction, to apply to the resistance of a stud in a decking trough. of shear connection. This could be attractive when unable to
Whilst Eurocode 4 is not clear, the SCI design guide to complement meet the code requirement for minimum degree. The answer is an
it (P359)3 goes further by stating that the as-welded height should emphatic ‘no’. Using less studs would result in more slip, perhaps
be used. Although apparently conflicting, it should be remembered more than the 6mm (or 10mm) available.
that the codes apply this reduction factor to a solid slab value
(which is not the same in the two codes), and that the factors are
a result of back analysis of tests (by different organisations). The
correct definition of stud height is therefore part of the context in References and further reading
which it is being used. If anything, the Eurocode definition (as given
in P359) is conservative. Further work is underway to simplify stud 1) British Standards Institution (1990) BS 5950-3.1:1990+A1:2010
resistances and detailing rules in Eurocode 4. Structural use of steelwork in building. Design in composite
construction. Code of practice for design of simple and continuous
In addition to using stud height in the definition of a resistance composite beams London: BSI
reduction factor, the codes use it to define certain geometrical
criteria that should be met (for example EN 1994-1-1 6.6.5.1(1)). SCI 2) British Standards Institution (2005) Eurocode 4. Design of
recommends that nominal (before welding) dimensions are used to composite steel and composite structures London: BSI
avoid a slightly under height stud failing a pass/fail detailing check,
when the real impact of the lesser height on resistance will be 3) Steel Construction Institute (2011) Composite design of steel
negligible. framed buildings (P359), Ascot Berkshire: SCI

Minimum degree of shear connection Further reading

One of the most misunderstood aspects of composite construction Rackham J. W., Couchman G. H. and Hicks S. J. (2009) Composite
is the need for a minimum degree of shear connection. Slabs and Beams using Steel Decking: Best Practice for Design and
Construction [Online] Available at: www.steelconstruction.info/
Composite beams ‘work’ by tying together the steel beam and a (Accessed: January 2014)
concrete flange. Without that connection the flange would simply
slip over the steel as it deflected, and the beam resistance (and Johnson R. P. (2011) Designers’ Guide to Eurocode 4: Design of
stiffness) would be the sum of those of the two parts. To keep this Composite Steel and Concrete Structures (2nd ed.) London: ICE
slip to a manageable level (one below which the connection will Publishing
not be broken, in an ‘unzipping’) the shear connection must have
a certain stiffness (Figure 4). This manageable level of slip has NCCI Document PN001a-GB Resistance of headed stud shear
traditionally been taken as 6mm, although recent work suggests connectors in transverse sheeting [Online] Available at: www.steel-
that for some configurations, 10mm is achievable. ncci.co.uk

In simple terms then, there must be enough studs on a beam to NCCI Document PN002a-GB Modified limitation on partial shear
ensure the total connection has sufficient stiffness (the sum of the connection in beams for buildings [Online] Available at: www.steel-
stiffnesses of all the studs) to limit slip. However, the stiffness of a ncci.co.uk
stud is not quantified in Eurocode 4 or BS 5950-3.1, so an explicit
check is not possible. Instead, the codes indirectly achieve this
check by looking at resistance rather than stiffness. The minimum
degree of shear connection rules ensure there will be enough studs Web resources
to satisfy a certain resistance requirement, in order to indirectly
ensure there are enough to satisfy the stiffness requirement. The Steel Construction Institute: www.steelconstruction.info

TSE26_25-27 CSCC-2 v3.indd 27 23/01/2014 08:17