30 / 208
2 8
T H E B U I L D I N G C O N S E R VAT I O N D I R E C T O R Y 2 0 1 5
T W E N T Y S E C O N D E D I T I O N
1
PROFESS IONAL SERV I CES
apart from the physical construction survey,
vibrations in the suspended floor need to
be measured to enable a focussed report to
be communicated to the client. Otherwise
it is possible to spend a lot of money trying
to fix a lively floor with little perception of
improvement.
ASSESSMENT
Every historic timber structure is unique.
Although it should be possible to predict
a beam’s deflection from its dimensions
based on an assumed modulus of elasticity,
in practice such calculations are unreliable.
The size and location of knots, the quality
and strength of the timber, and the presence
of decay, all impact on its performance. The
strength and performance of the structure
as a whole is also affected by the integrity of
connections, whether primary or secondary
(from beam to wall, and from joist to beam,
for example). A repair previously carried out
on one project rarely suits another. Although
the structure may appear to be similar, the
variables are so great that the probability of an
exact fit is low.
Although obvious, it is worth stressing
that surveying the situation is essential. The
uniqueness of the construction needs to be
understood for the repair to be a complete
design that takes account of the performance
required from the structure, the longevity
required, and the ease of maintenance and
accessibility, as the repair might require future
modification or, indeed, reversal. The design
must also be based on an accurate assessment
of how much fabric needs to be disturbed.
A survey starts with the preparation of
accurate drawings, recording the current
structure, noting any obvious defects. An
assessment of existing records can often
shed further light on past alterations. Some
evaluation of the actual deflection is also
required to give a fuller picture of how the
structure is performing.
The standard method of measuring
deflection involves constructing a stable
scaffold to within 50mm of the underside of
the ceiling and then measuring the deflection
under live loads using a dial gauge fixed
between the top of the scaffold and the
underside of the ceiling. If the scaffold is rigid,
the dial gauge is sufficiently accurate to record
deflections of less than 0.01mm. However, to
achieve the stability required it is best if the
scaffold is built off solid ground, which is not
always possible.
Investigation and analysis techniques are
being developed that can model the dynamic
performance of suspended timber floors in
three dimensions. A dynamic floor appraisal
that records accelerations of a structure
against known excitation forces provides
more information than simple static tests. To
explain the difference between
static stiffness
(EI) and
dynamic response
– accelerations
of the structure – imagine standing still in
the middle of a floor while loads are applied;
you experience static stiffness when the floor
deflects but not necessarily the bounce or
response. It is the combination of EI, mass,
boundary conditions and damping which
creates the dynamic response and thus human
perception of it. Therefore the dynamic analysis
is a more accurate method. Using this approach
data can be collected that can identify the
natural frequencies of the floor as a whole that
account for the uniqueness of its construction.
Mann Williams Consulting Engineers
have carried out a number of tests on historic
floors to develop the technology. Data is
provided using an appropriate dynamic
excitation device at intervals across the floor
surface while the vibrations are recorded from
around the room. Deflections, stiffness and
the dynamic responses of the structure can
be modelled from the data gathered, allowing
the true characteristics of the whole floor
assembly to be analysed.
Not only can this provide a clear picture
of the existing structure and its anomalies, but
it can also be used to explore various ‘what ifs’,
such as the effect of better load management
or the introduction of reinforcement, to
give a better understanding of the potential
of the existing structure. Dynamic testing
and modelling recognises and quantifies
the uniqueness of timber structures and
highlights the failures of simple analyses
based on theoretical magnitudes of deflection.
It can see the structure as a whole, identifying
various areas of damping and areas that can
be exploited to reduce vibration. Essentially
it can give a better idea of whether or not the
vibration of the floor can be improved within a
given budget.
INTERVENTION AND MITIGATION
Solutions to excessive deflection under
dynamic load fall under two categories: load
management and structural improvements.
The former includes measures to reduce
the loading or better distribute it across
a wider area, or to bypass defective
elements altogether. The latter includes
repairs and, if necessary, alterations to
Recording vibrations from a dynamic excitation device (in this case a drop-
hammer) to analyse the performance of a timber floor: the data can help assess
deflections, stiffness and dynamic responses
Steel tensioning to improve the structural performance of over-spanned joists
Dynamic analysis of the Octagonal Gallery at
Mount Stewart House: the top three diagrams
illustrate the mode shapes associated with the first
three natural frequencies of the structure, the bottom
image is a finite element model which represents the
predicted deflection of the structure with a uniformly
distributed loading applied to one half of the gallery.