C A S E S T U D I E S O N A P P LY I N G B E S T P R AC T I C E TO I N - S I T U CO N C R E TE F R A M E D B U I L D I N G S

St George Wharf Case Study Early age construction loading

Introduction

The European Concrete Building Project Figure 1: Backpropping installed at St George Wharf
at Cardington was a joint initiative aimed
at improving the performance of the
concrete frame industry. It led to the
preparation of a series of Best Practice Early age construction loading can be the most significant
Guides, giving recommendations for
improving the process of constructing in-
load experienced by multi-storey structures.
situ concrete frame buildings.

Key points
This Case Study looks at the experiences of applying new criteria for
the striking of slabs and the design of backpropping.
• In residential developments the spare load bearing capacity of slabs used
in determining backpropping requirements is very low because of the low
level of imposed load specified. This is important because such spare
As part of a programme to disseminate
and apply what has been learnt from capacity needs to be available to support freshly cast concrete.
Cardington, BRE has subsequently
worked directly with those involved in
• It was found in practice that the levels of preload measured in individual
St George Wharf, a high-profile, 100,000 m2 backprops were high.
mixed-use phased development on the
River Thames. • When the effect of preload was taken into account, the distribution of
BRE worked jointly with the developers, loads for the supporting slabs was found to be close to that predicted
St George (South London), their by conventional approaches assuming an even distribution of load
engineers, White Young Green, between slabs.
and specialist concrete contractors,
Stephenson, to develop and implement • Quality control concerning the type, positioning, sequencing of placement
process improvements tailored to the
St George Wharf site. and removal and tightening of backprops is important and does not
always appear to be exercised.
This work has led to a series of
innovations being trialled, the results of • The temporary works designer and permanent works designer should
which are summarised in this series of
Best Practice Case Studies. work together to assess whether a higher design load should be used to
cater for the construction load conditions.

www.bca.org.uk www.stephenson-ssc.co.uk www.wyg.com www.bre.com www.dti.gov.uk www.construct.org.uk www.stgeorgeplc.com www.concretecentre.com

1
Criteria for early age loading
Work carried out on the in-situ concrete
building at Cardington highlighted the
potential benefits that could be achieved
by adopting revised criteria for striking
and the design of backpropping based on
serviceability. This led to the preparation
of a Best Practice guide, Early striking and
improved backpropping for efficient flat
slab construction (see back cover).
These new criteria can be summarised as:
- (w/wser)(fcu/fc)0.6 ≤ 1.0 Equation 1
- (w/wser) ≤ 1.0 Equation 2
Where
w = construction load
wser = design service load
fc = estimated concrete strength at time
of application of construction load
confirmed by measurement
(e.g. Lok test)
fcu = specified characteristic cube
strength at 28 days
The existing approaches being used by
the contractor were based on BS 5975.
These were compared with the revised
approaches suggested by the work
from Cardington.

Findings in relation to striking

The expectation was that adopting the
new criteria would allow striking at
lower concrete strengths than currently
permitted. However this was found to very
much depend on the assumptions made.
As fairly optimistic assumptions were
used for the existing criteria, it was not
considered prudent to revise the existing
strengths at striking on this project.
Based on a characteristic cube strength
at 28 days of 40 N/mm2 the minimum
strengths required to be achieved for
striking were 22 N/mm2 for slab pours
without balconies and 25 N/mm2 with
balconies.
The minimum age at which striking took
place was three days. The results of air-
cured cubes indicated that these minimum
strengths were exceeded when the slabs
were struck.
The use of Lok tests has also been invest-
igated as an alternative to air-cured cubes
for determining the striking strength.
This is the subject of a companion case
study. The influence of the age of striking
on deflection is covered in a further Figure 2: Backpropping spreadsheet
case study.
2
Findings in relation to Although the contractor was not given difficult to justify increasing
backpropping any instruction to preload the backprops the load bearing capacity of the
and did not present any calculations supporting slab by following the
When designing backpropping the critical
making any assumption about this, recommendations given in the Best
issue is the assumed distribution of load
it was found in practice that the levels Practice Guide, consideration could be
between the levels of supporting slabs.
of preload measured in individual given to applying appropriate levels of
The conventional approach to the design backprops were high [3] and were such pre-load in the backprops. This would
of backpropping is to assume a uniform that, with one level of backpropping, justify the assumption of a more
distribution of load between supporting the uppermost slab was not necessarily even distribution of load between
slabs. The number of supporting slabs predicted to be the most critically loaded. supporting slabs, as in conventional
required is then determined by the spare approaches. However, if this were
In the context of the 15th floor transfer
capacity1 of each of the slabs to support more than a nominal amount it would
slab it would not have been possible to
the additional weight of the next slab to involve specification of a defined level
justify the construction of this slab using
be cast. of preload in individual backprops
the new criteria unless the beneficial
that would be very difficult to control
Work from Cardington suggested that, in effects of preload are taken into account.
in practice.
practice, without the input of significant This again creates a problem since the
levels of preload into the backprops, and temporary works designer would be 2. The Guide to flat slab formwork and
assuming the slabs to remain essentially faced with specifying a level of preload falsework [1] includes a CD with an
elastic, there was very little benefit in in the backprops that the contractor interactive Excel spreadsheet,
having more than one level of back- would not be able to control or verify. illustrated in Figure 2 with sample
propping. Further, the uppermost slab of In practice such problems can be data. This allows the influence of
the supporting slabs carries approximately overcome only by both the designer and cracking of the slabs and the effects
70% of the load during the construction of contractor taking a pragmatic approach [1]. of pre-load to be taken account of in
the slab above. This can be shown Quality control on site concerning the calculations for up to two levels of
theoretically by considering the stiffness type, positioning, sequencing of backpropping. There is evidence to
of the different slabs in relation to the placement and removal and tightening suggest that there may be merit in
props and the arrangements of the false- of backprops is important and did extending the scope of the spread-
work and backprops. It was found to hold not always appear to be exercised at sheet to allow additional levels of
true over a range of different arrangements St George Wharf in accordance with backpropping. However, as stated
of backprops and backprop types (steel or the calculations presented. Because of above, the level of pre-load might
aluminium). the possibility of significant preload prove very difficult to control in
being introduced into the backpropping, practice, especially for multiple
As with many other residential develop-
it is advisable to make allowance for this floors of backpropping.
ments, the spare capacity of the slabs at
St George Wharf (3.1 kN/m2 unfactored), in any assumptions made about the
loads carried by the props themselves. 3. The issue of the design of the back-
is very low because of the low level of
propping will be most acute for
imposed load specified (1.5 kN/m2).
Interpretation of existing situations where low imposed loads
This creates a dilemma since the work
are specified, such as in car parks and
from Cardington would suggest that the Best Practice guidance residential developments, because of
slab immediately beneath that being cast concerning backpropping the limited spare capacity of the slabs.
could theoretically be overloaded unless 1. The assumptions made in Table 1, Exceeding the design service load of
steps are taken to prevent this. which comes from the Best Practice the slabs by a small margin will not be
Assuming an even distribution of load Guide, were shown not to represent a safety issue, but could have some
between the slabs in accordance with typical site practice, in particular impact on serviceability performance.
conventional approaches, the weight of arranging for the backprops to be The permanent works designer should
a freshly cast slab (250 mm thick giving finger-tight and not relying on any therefore be involved in any decisions
6 kN/m2) in addition to allowance for pre-load in them. Where it proves to theoretically overload slabs and
construction loads meant that for most should consider possible implications
slabs two levels of backpropping were Table 1: Load distribution by backpropping for serviceability.
required and this was the backpropping
NO ONE LEVEL OF TWO LEVELS OF
arrangement adopted. This ignored any LOCATION LOAD BACK-
BACKPROPS BACKPROPS
4. If the developer is
PROPS
reduction in the load factor appropriate On slab On slab In props On slab In props
closely involved in
to the loading. New slab being cast total 100% 100% 100% the design and
Falsework/formwork wp 100% 100% 100% construction process,
In the case of the 15th floor transfer slab, On supporting slab(1) 100% wp 70% wp 65% wp as is the case with
which was 600 mm thick with a self- In backprops wb1 30% wp 35% wp
weight of 14.4 kN/m2, three levels of St George, they can
On lower slab (2) 30% wp 23% wp
backpropping were employed. Such a In backprops wb2 12% wp perhaps take a more
transfer slab is often required if the On lower slab (3) 12% wp informed decision as
column grid layout changes above the Notes to the relative merits
1. Assumes lower and supporting floors have been struck, have taken up their
level in question. deflected shape and are carrying their self-weight
of accepting a higher
2. Floor loading from imposed loads and self-weight is not considered design load to cater
3. The strength of particular slabs to carry applied loads will have to be considered
1
Spare capacity is defined as the available separately for the construction
4. All floors are suspended floors load conditions.
service load capacity less the self-weight
3
 C A S E S T U D I E S O N A P P LY I N G B E S T P R AC T I C E TO I N - S I T U CO N C R E TE F R A M E D B U I L D I N G S

Use of the spreadsheet Feff factor. This factor has been introduced 2. Best practice in concrete frame
in recognition that Equation 2 is not construction: practical application at
In the spreadsheet the user may set a St George Wharf, by R. Moss. BRE
relevant if the slab is uncracked. Equation 2
value for the level of preload in individual Report BR462, 2003.
was introduced to limit excessive strains in
backprops. The default value is 6kN per
the reinforcement. This is explained further 3. Backprop forces and deflections in flat
backprop [1], which is believed to be
commonly achieved in practice. in Reference 1. slabs: construction at St George Wharf,
Measurements of the preload in individual by R. Vollum. BRE Report BR463, 2004.
backprops at St George Wharf varied Conclusions
considerably, but averaged 13 kN per prop. 1. The distribution of loads for the Acknowledgements
From the point of view of the uppermost supporting slabs at St George Wharf The support of the DTI for this project
was found to be close to that predicted carried out under its Partners in
supporting slab, it should be recognised
by conventional approaches assuming Innovation scheme is gratefully
that the relieving preload (as measured
an even distribution of load between acknowledged.
in kN/m2) is dependent not only on the
preload in each backprop but also on slabs once the effect of preload was
the number of backprops. taken into account. However preloading
of the props was not achieved in a The Best Practice Guide, Early striking
As an illustration of this, the level of controlled manner and in practice and improved backpropping for efficient
preload chosen in each backprop for would be very difficult to do. This is flat slab construction, summarises work
trial use of the spreadsheet for emphasised by the variations carried out on these topics during the
St George Wharf was 6kN per prop. in prop loads measured. construction of the in-situ concrete
This gave a preload in kN/m2 similar building at Cardington.
to that actually measured. 2. If heavily tightened, loads measured
in individual backprops can be This can be downloaded free from the
The Best Practice Guide recommends the Downloads section of the Concrete Centre’s
significant, although variable, and
installation of backprops at the earliest website at www.concretecentre.com and at
averaged 13 kN for the props
opportunity to assist in the distribution http://projects.bre.co.uk/ConDiv/
instrumented at St George Wharf.
of load between the supporting slabs. concrete%20frame/default.htm.
This is believed to be higher than the It should be read in conjunction with a
In many cases, as with St George Wharf, levels of preload generally measured
flying form systems are used which, companion guide Early age strength
at Cardington. However the overall assessment of concrete on site.
in practice, usually means that the level of preloading achieved at St
uppermost slab carries all of the weight George Wharf (estimated as 1 kN/m2)
of the falsework. The spreadsheet Case Studies in this series of applying
is not believed to be exceptional. best practice:
allows this loading to be specified
(usually 0.5 kN/m2) and automatically 3. Although slabs may be predicted to be • St George Wharf project overview
takes this into account when calculating overloaded, they may very well not be
• Early age concrete strength
the overall distribution of the load so in practice because of the margins
assessment
between the slabs. on the actual construction load
allowed for. • Early age construction loading
An average backprop stiffness of
• Reinforcement rationalisation and
23 kN/mm was used, based on measured 4. Additional margins may be required supply
average values for different types of props. for the design of the backprops
Parameters were chosen to allow the themselves to allow for unintentional • Slab deflections
influence of cracking on the slab to preload induced in them during • Special concretes
be taken into account. These were based installation and as a result of
on past experience. These parameters subsequent temperature changes.
resulted in an equivalent reduced modulus Ref TCC/03/04
of elasticity for a given concrete strength, 5. To achieve a controlled approach to First published 2004
but these calculated values could have early age loading the most reliable Price group A
been overridden if desired. method would be to follow the ISBN 1-904818-04-8
existing guidance given in the Best
The number and location of the falsework Practice Guide, but the penalty of this © The Concrete Centre 2004
supports and backprops was specified on approach is that the slabs will need Published by The Concrete Centre on
the basis of the calculations presented to be designed for a higher loading behalf of the project partners.
for the project. As can be seen from the during construction. www.concretecentre.com.
results for this example, which is for an
edge panel with two levels of backpropping, The work undertaken and the conclusions For further copies of these Best Practice
the distribution of load between the slabs, reached in relation to the innovations Case Studies contact
taking account of the preload in the back- described above should be viewed in the www.concretebookshop.com
props, was predicted to be fairly close context of the particular project on which
All advice or information from The Concrete
to the equal thirds split suggested by the innovations have been trialled.
Centre is intended for those who will
conventional approaches.
This Case Study is underpinned by full evaluate the significance and limitations of
In this example the results indicate that reports [2, 3] giving the background and its contents and take responsibility for its
the slab immediately beneath that being further information on the innovations. use and application. No liability (including
that for negligence) for any loss resulting
cast is subject to a construction load very
from such advice or information is accepted.
marginally in excess of the design References Readers should note that all Centre
service load. publications are subject to revision from time
1. Guide to flat slab formwork and
The spreadsheet allows some interpolation falsework, by Eur Eng P. F. Pallett. to time and should therefore ensure that
they are in possession of the latest version.
between the two criteria set out in Published by The Concrete Society on
Equations 1 and 2 by virtue of an behalf of Construct. Ref. CS 140, 2003.
4