BRITISH STANDARD BS 476-32:

1989

Fire tests on building

materials and
structures —
Part 32: Guide to full scale fire tests
within buildings

UDC 614.841.332:620.1:69.01
BS 476-32:1989

Committees responsible for this

British Standard

The preparation of this British Standard was entrusted by the Fire Standards
Policy Committee (FSM/-) to Technical Committee FSM/1, upon which the
following bodies were represented:

Association of British Roofing Felt Electricity Supply Industry in England and

Manufacturers Wales
Association of Building Component Engineering Equipment and Materials User’s
Manufacturers Ltd. Association
Association of Structural Fire Protection Eurisol (UK, Mineral Wool Association
Contractors and Manufacturers Fibre Building Board Organisation (FIDOR)
British Cement Association Fibre Cement Manufacturers’ Association
British Fire Services’ Association Limited
British Floor Covering Manufacturers’ Fire Protection Association
Association Flat Glass Manufacturers’ Association
British Plastics Federation Flat Roofing Contractors Advisory Board
British Railways Board Gypsum Products Development Association
British Rigid Urethane Foam Manufacturers’ Home Office
Association Institution of Fire Engineers
British Wood Preserving Association Loss Prevention Council
Chemical Industries Association Mastic Asphalt Council and Employers
Chief and Assistant Chief Fire Officers’ Federation
Association National Council of Building Materials
Concrete Society Producers
Department of Education and Science RAPRA Technology Ltd.
Department of the Environment (Building Royal Institute of British Architects
Research Establishment) Steel Construction Institute
Department of the Environment Timber Research and Development
(Construction Industries Directorate) Association
Department of the Environment United Kingdom Atomic Energy Authority
(Property Services Agency) Warrington Fire Research Centre
Department of the Environment for Northern Wood Wool Slab Manufacturers’
Ireland Association
Department of Transport (Marine Directorate) Yarsley Technical Centre Ltd.

The following bodies were also represented in the drafting of the standard,
through subcommittees and panels:
Autoclaved Aerated Concrete Products Queen Mary College Industrial Research
Association Thermal Insulation Manufacturers and
London Scientific Services Suppliers Association (TIMSA)
National GRP Construction Federation University College London
Phenolic Foam Manufacturers’ Association

This British Standard, having

been prepared under the
direction of the Fire
Standards Policy Committee,
was published under the
authority of the Board of BSI
and comes into effect on
30 November 1989
Amendments issued since publication

© BSI 02-1999 Amd. No. Date of issue Comments

The following BSI references

relate to the work on this
standard:
Committee reference FSM/1
Draft for comment 87/41174 DC

ISBN 0 580 17617 7

 BS 476-32:1989

Contents

Page
Committees responsible Inside front cover
Foreword ii
1 Scope 1
2 General 1
3 Definitions 1
4 Design and configuration 1
5 Specimens 3
6 Ignition sources 4
7 Monitoring 5
8 Safety precautions 10
9 Data recording 10
10 Test report 11
Appendix A Fire test facilities 13
Figure 1 — Fire test facilities 13
Publications referred to Inside back cover

© BSI 02-1999 i
BS 476-32:1989

Foreword

This Part of BS 476 has been prepared under the direction of the Fire Standards
Policy Committee.
The guide does not conflict with the concept or recommendations of the
ISO/TC 92, Fire tests on building materials and structures, proposal DP 9705
“Room fire test in full scale for surface products” which specifies one configuration
of test rig and one degree of ventilation. This proposal is being developed to
standardize a test method for the fire performance of surface products under
closely specified conditions.
This guide is designed to enable a quantitative assessment to be made of the fire
behaviour of a given specimen or a total interacting system, based on sound
experimental practice and scientific principles. It will help to eliminate the risk
of providing an incorrect assessment of a product such as may inadvertently occur
if data from tests based on a limited range of experimental variables are
extrapolated to other situations which are not directly comparable.
The complete programme for any test series of full scale enclosed fire experiments
usually involves many different considerations and possible simulations. This
guide reflects the current state of knowledge and suggests a choice of geometry,
ignition sources, ventilation and similar aspects of the experiment, not all of
which will be appropriate to every test situation.
Because of the developments in measuring techniques, this Part of BS 476 has
been couched in terms of advisory rather than mandatory statements in order to
allow the use of up-to-date techniques as they become available.
A British Standard does not purport to include all the necessary provisions of a
contract. Users of British Standards are responsible for their correct application.
Compliance with a British Standard does not of itself confer immunity
from legal obligations.

Summary of pages
This document comprises a front cover, an inside front cover, pages i and ii,
pages 1 to 14, an inside back cover and a back cover.
This standard has been updated (see copyright date) and may have had
amendments incorporated. This will be indicated in the amendment table on
the inside front cover.

ii © BSI 02-1999
 BS 476-32:1989

1 Scope 2.4
test construction
This Part of BS 476 provides comprehensive
recommendations on the conduct of full scale a complete assembly of a specimen and any
experiments simulating fires in buildings. Such associated construction
tests, of varying scale and configuration, can be used 2.5
to evaluate the behaviour of building components, pilot ignition
assemblies or constructions and/or contents under
ignition by a small flame or spark of the flammable
actual fire conditions. The advice on experimental
mixture of air and decomposition volatiles evolved
design and instrumentation and the analysis and
from a heated combustible material that may
reporting of results provides a basis for a
accumulate above its surface
consumer/contractor agreement for the execution of
ad hoc work. 2.6
The guide may be used to establish, as far as is spontaneous ignition
reasonably possible, laboratory conditions that ignition of the flammable mixture of air and
represent a specific set of fire conditions, with an decomposition volatiles evolved from the heated
emphasis on examination of the pre-flashover material that may accumulate above its surface
behaviour and contribution to fire growth of the without provision of additional energy from spark or
product(s) under consideration. flame once the flammable mixture attains its
The choice of ignition source and test construction is spontaneous ignition temperature
based on the objective of the fire experiment which 2.7
may be: critical exposure
a) a comparison of the fire performance of the minimum irradiance at which ignition (pilot or
different materials; spontaneous as specified) can be effected, regardless
b) a comparison of theory and experiment; of duration
c) a simulation experiment;
3 General
d) a measurement of the fire behaviour of
composites, assemblies and finished products; If only one test is envisaged, it is preferable to carry
out tests in the most severe conditions likely to be
e) an investigation of the interaction of
encountered including position, orientation and size
components within the system;
of ignition source. For example, for wall lining
f) an authentication of fire properties measured materials the interrelationship between walls,
in small scale tests. ceilings and ignition source needs careful
consideration. Positioning an ignition source in the
2 Definitions corner of a test arrangement is recommended if it is
For the purposes of this Part of BS 476, the the most severe condition. In such a case, additional
definitions given in BS 4422 apply, together with side wall tests, with the ignition source placed
the following. against a plain wall arrangement would not be
required, but if it is suspected that such an
2.1 arrangement would significantly affect behaviour,
closed test arrangement
then such tests would be necessary.
a form of compartment, capable of imposing a It is thus essential to carry out tests on walls and
restriction on ventilation at some stage during a fire ceilings in the most stringent arrangement
2.2 foreseeable. Because of difficulty which might arise
open test arrangement from variable weather conditions, it is
a form of compartment which enables unrestricted recommended that full scale tests be carried out
ventilation to the fire throughout the duration of the within an adequately dimensioned, roofed
test enclosure.
2.3 4 Design and configuration
associated construction
4.1 General
a form of construction that may be required for the
testing of some elements of construction and to Configuration of specimens and ignition sources is
which the test specimen is connected, e.g. the wall of importance in a large scale fire test; compartment
into which a glazed element would be fitted geometry and ventilation have major effects on test
results.

© BSI 02-1999 1
BS 476-32:1989

In general the test should be carried out full scale. If 4.2 Size and shape of compartment
this is impracticable the scale should be as large as The compartment size, shape and openings should
possible and particular care is needed with be chosen to simulate the nature or type of
interpretation of results. Physical and mechanical compartment in which the subject material,
fixings and jointing elements which are likely to be product, or system is expected to be used in actual
used in practice should be incorporated in the service. If there is a range of sizes, then account
specimen in full scale in order to establish the effect should be taken of the fact that for a given ignition
of such products under fire conditions. exposure, the smaller compartment sizes will
The minimum height of the test rig should be usually provide the most severe fire development
normal storey height (2.3 m) but it may be necessary conditions. Preferably the compartment should be
to exceed this dimension. The ratio between the designed to be symmetrical and as simple as
height and lateral dimensions of the test rig should possible for ease of analysis. The test construction
be realistic. Normally there should be no scaling of could be a wall system or a lining applied to an
thickness although if this is necessary the minimum appropriate realistic support but it has to be noted
thickness of the test rig should be such that there is that the results of the test will only apply to that
no change in the thermal effects on fixings nor particular combination and care would be needed
significant differences in heat loss to the with the application of those results to systems with
surroundings during the early stages of a test. variants of that tested. The floor of the test rig
There should be no scaling of surface irregularities. should have thermal properties approximately
With regard to ignition sources there should be no equal to those which will be encountered in use. The
scaling in terms of impingement area, thermal roof/ceiling of the test rig should be flat, unless a
severity, or application time, although with sloping roof/ceiling is being investigated. If the
experience, it may be possible to introduce time actual roof/ceiling is not being subjected to test, the
limits. roof/ceiling used in the test should have thermal
It is important to avoid limiting ventilation too properties approximately equal to those which will
be encountered in use.
severely. Vents should be either to the full height of
the test rig, or to door height, whichever is In a test for linings with unrestricted ventilation the
preferred. The latter arrangement will cause a horizontal dimensions of the specimen may be
concentration of hot gases in the upper part of the scaled down and are of less importance than the
test room. The size of vent may prevent or delay vertical dimensions, although the horizontal
flashover. In situations where it is necessary to use dimensions should be sufficient to contain the
a closed test arrangement, vent area should be ignition source and obtain information on fire
representative of full scale so that results can be spread, including spread across junctions. A full
indicative of performance in practice. Such tests height rig with a 1 : 1 ratio of height to horizontal
require additional safety measures due to length on both sides of the corner would usually be
explosibility problems, such as careful monitoring of the minimum required.
temperatures and remote venting and exhaust The arbitrary decisions made when designing an
extraction after the fire is considered to be enclosed test will influence the behaviour in the test
extinguished (see 4.4.2). and therefore the ranking order of the material
The choice of configuration of the test facility performance. It is desirable that flexibility of test
depends on the above considerations and on the configuration is maintained but ISO is developing a
required fire scenario. The alternative test with precise requirements for dimensions and if
configurations given in Appendix A provide a such a closed test is required on a standard basis
sufficiently wide range to take account of these then use should be made of the ISO test
points. (ISO/DP 9705).
4.3 Thermal and radiative properties of
associated construction
The thermal conductivity, density and heat capacity
of the wall lining will affect the rate of fire growth
within the test structure. The test structure
therefore should be representative of those
materials with which the structure will be used in
practice and should not influence the fire behaviour
of the material under test in an unrealistic way.

2 © BSI 02-1999
 BS 476-32:1989

During the course of a compartment fire experiment 4.4.3 The location of the ignition source has a
disintegration or cracking of the materials lining significant influence on the air inflow rate, the
the compartment may affect the behaviour of the upper gas temperature, and the oxygen
fire. concentration profiles within the compartment and
In a confined fire test, the vertical pressure gradient should be fixed. If it is varied, the results may be
developed in the presence of the fire will cause influenced and it may be preferable to standardize
smoke and hot gases to leak to the outside, and cool this arbitrarily (see also 6.7).
air to be drawn into the compartment through the 4.5 Instrumentation
cracks in the compartment walls or specimens. The
Detailed information is given in clause 7 on the type
effect of the leakage of hot gases or air will result in of instrumentation to be used in monitoring the
lower gas temperatures and reduced smoke density various parameters of a fire. Its installation within
than would occur without leakage. This effect often
the test enclosure, however, should not affect the
causes pulsations in burning rate and
configuration of the test, the ventilation to the fire,
temperatures.
the air-flow pattern or the burning behaviour of the
4.4 Ventilation material being assessed.
4.4.1 In an open fire test, characterized by an In particular, care should be taken when inserting
uncontrolled and unrestricted degree of ventilation, thermocouples and thermal transducers into
the structure should not compromise the basic aim materials so that fire cannot spread to the back of
of realism and it is important to ensure an the specimen or into any cavities which may exist.
essentially draught-free test environment. In Damage to the material being tested whilst
general, where the air supply is plentiful (high installing instrumentation can adversely affect that
ventilation), the resulting fire will have a rapid material’s fire performance. Conversely, by
growth rate and high maximum temperatures with supporting the materials, the fire performance can
a swift transition from the growth to the decay be apparently improved.
stage. The controlling influence on this type of fire The sensitivity and response time of instruments
will be the amount of fuel (essentially surface area) should also be considered when selecting which type
available. to install. Fine thermocouples (e.g. 1 mm o.d.) will
4.4.2 With a restricted degree of ventilation the be more accurate and respond within a shorter
growth of the fire will be controlled by the available timescale than larger diameter types; this is
air in which the fuel can burn. The difficulties particularly important when monitoring flame
encountered with a ventilation controlled fire will spread and heat output. Instruments should also be
vary with the degree of control and the stage of the calibrated at regular intervals. For radiometers,
fire. this can be achieved by using the procedure in
With all polymeric materials the rate of flame BS 6809.
propagation is influenced by the concentration of
oxygen in the gas phase near the surface of the 5 Specimens
material under consideration. For this reason, in In the range of tests covered by this guide specimens
the early stages of a fire, the rate of growth of the may vary from a room lining material applied to a
fire initiated within a compartment is controlled by standard support structure to a full room/corridor as
the fuel bed characteristics because of the well as complete free standing assemblies.
availability of sufficient air. However, when the fire
The specimen should always be tested in the
grows larger, the rate of air supply to the
orientation in which it will be used. Fixing methods,
compartment may become the limiting factor. If the
joints and surface coatings should reflect typical
air supply becomes the limiting factor, the fire
practice.
growth decreases, until a constant burning rate is
obtained dependent upon the supply of air. The fire Specimens may need to be incorporated into the test
duration will be longer than for an equivalent well facility construction (see 3.3). The choice of the
ventilated fire. associated construction may be significant to the
performance of the test specimen. This relationship,
The degree of ventilation will affect smoke
where the contribution of the wall or ceiling cover
production, heat release, nature and concentrations
may differ, is also influenced by the size of the
of toxic gases etc. and therefore the choice of ignition source and its orientation and position.
ventilation conditions is of vital importance and
should be as close to the end use situation as
possible to enable realistic product assessment
(see also 4.1). The influence of any extraction system
should also be taken into account.

© BSI 02-1999 3
BS 476-32:1989

6 Ignition sources 6.4 Gas burner flames

6.1 Choice Diffusion flames are more representative of actual
fires and premixed flames should only be used for
The choice of an ignition source in a compartment
specific representations.
fire experiment is significant. Either a real ignition
source or a simulation with equivalent 6.5 Liquid fuel pool fires
characteristics could be used. This guide presents a The rate of production of volatiles from liquid pool
list of the important considerations on the choice. It fires is readily determined from their rate of mass
is important to identify what real ignition source is loss or the flow rate necessary to maintain a
being simulated in terms of convection, heat constant depth in the pool and they have an
duration, irradiance, etc. interaction with the fire environment which can be
A typical ignition source could be chosen or a range quantified by their change in heat production rate.
could be selected in order of severity. Occasionally a Their radiation characteristics can be controlled by
worst case example could be employed. the choice of fuel.
The size of the ignition source should not be 6.6 Wood crib fires
excessive in relation to the dimensions, shape and The variability of crib sources can be minimized by
ventilation of the test construction. The ignition controlling the moisture content, species and
source will also have some effect on the ventilation density of the wood. Their use as a simulation of
conditions prevalent in the fire enclosure. burning contents of buildings has proved very
It is important to choose an ignition source which successful. However, it should be noted that the
will not adversely affect measurements (for source does not provide a constant ratio of heat
example, by generating high levels of smoke, or toxic release to mass loss throughout the fire test and the
gases, or reducing the available oxygen for rate may be significantly lower than that for some
combustion). For this reason, it may be necessary to plastics based products.
carry out preliminary tests to estimate the likely 6.7 Electrical sources
effects of the chosen ignition source on such
measurements and to correct these after the test. The main ignition sources created by misuse of
electrical supply and appliances are:
6.2 Characterization
a) overloaded wires and cables, where breakdown
When an attempt is made to simulate a real ignition of the insulation occurs and adjacent combustible
source, it is essential to realise that burning materials are ignited by hot wires;
characteristics may be affected by environmental
conditions and therefore recognise that the design b) mechanical failure of the insulation resulting
parameters chosen may not be correct and may from ageing or physical damage;
require subsequent adjustment. c) heaters, where glowing wires or bars emit high
radiant energy.
Possible ignition sources can be characterized by:
Electrical ignition sources for use in room tests can
a) total fuel content;
be simulated using sources from smaller-scale fire
b) type of fuel content; tests (e.g. glow-wires at temperatures up to 960 °C
c) rate of fuel release; and radiant cones at irradiances up to 70 kW/m2).
d) rate of heat release;
e) height of flame for given location;
f) convective and radiative heat; and
g) time of burning.
6.3 Type and size
The complete character of the ignition source should
be determined, including mass, material
identification, morphology, dimensions, and all
other physical and chemical characteristics that are
necessary to repeat the experiment. Typical ignition
sources may be solid, liquid, or gaseous fuels and
include wood cribs, gas burners and electrical
sources.

4 © BSI 02-1999
 BS 476-32:1989

6.8 Location 7.2 Ignitability and critical exposure

The location of the ignition source is one of the most 7.2.1 General. When assessing the ignitability and
important considerations in conducting critical exposure of products, it is necessary to
compartment experiments. Its position in relation to consider the irradiance at its surface up to the time
the wall can significantly influence the rate of of ignition and whether ignition is spontaneous or
burning. When it is close to the wall there can be pilot. The progress of the flame front and time to
major feedback influences, and the ignition source flashover within the test enclosure are also
will burn more quickly, although this will depend to important.
some extent on the properties of the lining, the 7.2.2 Direct measurement. This is only feasible in
availability of air and the type of ignition source. cases where the measuring device may be inserted
The flame height is affected by the entrainment of into the body of the component so that the sensitive
air into the plume which itself is critically affected head is flush with the exposed surface.
by the position of the ignition source in relation to
Instruments which may be used include the
the wall/corner. For example if the access of air to
following.
the flame is blocked from one side, such as would
occur by placing the ignition source against a wall, a) Heat flux meter (see also 7.4.3). This measures
then an increased flame height for the same rate of combined radiative and convective heat flux
gaseous fuel leaving the source would result. This density on the surface of the specimen. The
analogy can be extended further to an ignition necessity for water cooling often makes the
source in a corner which would give an even higher positioning of the instrument and protection of
flame height. the tubing the critical factors in its selection for a
given purpose. The meter should be mounted so
7 Monitoring that the face of the instrument is flush with the
exposed surface of the product being assessed.
7.1 General
b) Radiometer. Simple robust radiometers that
Monitoring is the process of identifying and may be sacrificial (e.g. copper disc radiometers)
recording the performance of the test specimen may be used in fully developed fires.
during and after the test and involves both visual Alternatively, water-cooled instruments, with a
and instrumental techniques. The factors which given field of view may be positioned flush with
need to be monitored will vary from test to test the face of the product so that a noticeable
depending on the objectives of the test but would response is given to a localized flame on the
normally include ignition, flame spread, heat surface of the product.
output, smoke and toxic gas production, mass loss
c) Thermocouples. A crude method which relies on
and mechanical behaviour.
recording a sudden rise in gas temperature above
Visual observations should be recorded of all the surface of the product and which is probably
important events during the test such as time of the most effective simple means of timing the
ignition, dimensional extent of flame spread, nature onset of ignition. This method however does not
and rate of smoke production, mechanical behaviour provide an indication of the irradiance on the
of the test specimen (such as partial collapse or loss surface of the component at that time, and
of integrity of all or part of the specimen, melting or therefore should be used in conjunction with heat
other unusual behaviour). It is important that a flux meters (see a) above).
visual record is made of the test specimen at the end
d) Video/cinematograph. As with c), no
of the test. A visual record should also be made of
quantification of the irradiance at ignition can be
the test. This should preferably be continuous, such
recorded but an accurate indication of the time to
as film or video (see 9.1).
ignition will be provided. This method relies on
Instrumental measurements are normally used to the ability of the camera to detect the onset of a
monitor the performance of the test specimen for small flame on the surface of the specimen. This
flame spread, temperature, irradiance, mass loss, may not always be achievable and visual
smoke and toxic gas production. A wide range of observation will be necessary.
different techniques are available and discussed
under each heading. Instrumental techniques
should preferably be recording types to give a
permanent record.
In addition, the ambient conditions should be
recorded at the start of the test.

© BSI 02-1999 5
BS 476-32:1989

7.2.3 Indirect measurement. The construction of a 7.3.2.2 Vertical spread: flame spread. Although it is
test specimen may preclude the insertion of a heat difficult to distinguish the position of a vertical
receiver into the body of the product. Provided flame front on a surface every attempt should be
ignition of the component is caused by the primary made to record this phenomenon. The methods
ignition source alone (e.g. crib, burner) then a commonly used are visual observations (which can
calibration experiment using a control material be made by looking along the surface and behind the
carried out earlier can provide information on the flame plume) and recording by thermocouples fixed
exposure conditions during tests. with their hot junctions in contact with the surface
7.3 Flame spread at intervals not exceeding 0.5 m.
7.3.1 General. Three aspects of flame behaviour are 7.3.2.3 Horizontal spread. It may be difficult to
considered: distinguish realistically between the lateral spread
of the flame plume, spread of flame on the surface in
a) the spread of a flame front over a product the early stages of the test and transitory flaming.
caused by progressive ignition of volatiles However, the lateral extent of the flame plume
emitted from the heat surface; including flickering, transitory flaming, should be
b) flames from both the ignition source and the recorded.
burning product combining into a flame plume, The maximum extent of lateral spread on both sides
whose size is influenced by the thermal of vertical reference lines as a function of time and
properties of the product being tested; the height at which this occurs should be recorded.
c) elongated flames travelling horizontally The centre point for the reference of lateral spread
following the deflection by a ceiling or horizontal should be the line projected vertically from the mid
obstruction of a substantial body of flame. point of the side of the ignition source in closest
These aspects of flame behaviour are influenced by proximity to the test products. The distance
the configuration and thermal properties of between reference lines should not be more
materials under test and the test enclosure. than 0.5 m. Records should be made of the
However, the recommendations which follow should maximum extent of spread at given times in both
be used as guidelines and, where appropriate, as directions from the centre reference line to allow a
minimum requirements. meaningful plot of lateral spread against elapsed
7.3.2 Vertical surfaces time to be made.
7.3.2.1 Vertical spread: flame plume development. Trip threads may be used to supplement visual
Characterization of this can be difficult because of observations and video/photographic records.
the speed and often erratic behaviour of the flames Similarly thermocouples may be employed installed
at the tip of the plume. A visual assessment (aided in a lateral array at mid height between floor and
by height reference points at intervals of not more ceiling.
than 0.5 m) should always be made and supported 7.3.3 Horizontal surfaces. The appropriate portions
by photograph or video record. The top of the plume of concentric circles (centred on the vertical
at a given time should be taken as the uppermost reference line) should be marked on the ceiling or
point of a continuous plume averaged subjectively floor surface to facilitate visual assessment and
over 3 s. Any anomalous behaviour of the plume photographic records.
such as detached portions of flame should be Intervals of 1 m on horizontal downward facing
recorded. surfaces and 0.5 m on upward facing surfaces are
At least four more observations of the time/height usually appropriate.
relationship of the flame plume should be recorded The time taken for sustained flaming to reach a
between ignition and the time at which the flame particular distance line at any point should be
plume reaches the ceiling, which should also be recorded together with any major irregularities of
noted. spread. Alternatively, flame spread could be
As additional methods of recording vertical spread, recorded at specific time intervals.
thermocouples may be suspended in an array in the
likely path of the flame plume or cotton trip thread
connected to an event recorder.
Material properties which may influence flame
spread such as blistering, intumescence, cracking
and spalling should be reported.

6 © BSI 02-1999
 BS 476-32:1989

7.3.4 Hot gas layers/flames under ceilings. Due to 7.4.2 Temperature. This basic measurement, made
their buoyancy, flames and/or hot gases tend to in most fire experiments allows relative
accumulate beneath a ceiling or in the apex of a comparisons of the heat output from similar tests to
pitched roof. The depth of this layer as a function of be made, or alternatively, by reference to a
time should be determined either visually, calibration test, can indicate the additional
photographically or instrumentally using vertical contributions made by the presence of the materials,
arrays of thermocouples. Other observations such as components and assemblies under evaluation.
flaming at the base of the hot gas layer should also Sheathed or other appropriately insulated
be recorded. Smoke can impair visual observations thermocouples are commonly used, their signal
and alternative methods may need to be used. leads connected to precalibrated recording media
7.3.5 Other considerations regarding flame spread such as chart recorders or data loggers. The
monitoring. If the test construction contains thermocouples are usually positioned as follows:
concealed cavities in which flame spread may occur, a) within the fire chamber, directly over the
consideration should be given to the detection and centre of the fuel or ignition source and
measurement of such spread by the use of projecting 100 mm from the ceiling;
thermocouples.
b) at the ventilation opening of the fire chamber,
A test construction should contain representative in the centre-line and 25 mm below the top;
joints/ junctions. These can noticeably influence the
c) within a canopy hood and duct system, in the
course of flame spread, particularly when they occur
centre of the duct and positioned where the flow
on vertical surfaces.
can be expected to be uniform;
The occurrence of any secondary ignition remote
d) in the smoke exit stream.
from the ignition source by hot gases, flaming debris
etc. should be noted. Temperatures are usually recorded continuously
throughout the duration of a test with monitoring
A thorough post-fire examination of the test system
commencing 1 min prior to the test.
can be a very valuable aid to interpreting the
observations made during the fire. The extent of 7.4.3 Irradiance. The determination of the
charring, bubbling, melting, spalling or other distribution of energy flux within a fire
physical reaction to the fire should always be compartment is necessary in order to understand
recorded after the fire has been extinguished or has how the fire spread has occurred and also to give a
burnt out. The limits of such damage can usually be good indication of the heat output of the product
related to the passage or proximity of a flame front being tested.
on the basis of preliminary checks on a material’s A wide range of instruments can be used to measure
reaction to exposure to radiation or flaming. irradiance, the two most common being the Gardon
7.4 Heat output and Schmidt Boelter. Radiometers are also
available which measure radiative heat only and
7.4.1 General. A knowledge of the heat output from may have a wide or narrow angle of view.
any fire experiment, whether purely a qualitative
evaluation or a quantitative measurement, is one of Flux meters can either be air or water-cooled, or
the most important properties which should be sacrificial. It is important the instruments are
determined. This fundamental characteristic allows frequently cleaned, blackened and recalibrated
the damage and spread potential of a fire to be since soot deposits will reduce the sensitivity.
assessed and is a necessary requirement in the There are four common positions for siting heat flux
computer modelling of fires. meters. These are:
Except with the more sophisticated methods of a) as close as possible to the specimen initially
measuring heat output, it is normally not possible to ignited;
arrive at a thermodynamic heat balance for a fire b) close to the specimen at a position likely to
experiment, since this involves many complicated become involved by radiative flame spread to
measurements. The total heat output will comprise adjacent materials;
elements of radiation, convection and conduction c) at a position still within the fire chamber but
and depending upon the objectives of the remote from the fire (to determine flashover);
experiment, one or a combination of these, will
assume greatest importance. d) outside the fire chamber door and window
openings (if any) to determine threat to other
There are three main parameters (temperature, buildings.
irradiance and rate of heat release) which can be
used to provide information on heat output.

© BSI 02-1999 7
BS 476-32:1989

7.4.4 Rate of heat release. The measurement of heat In a static situation, where smoke will stratify,
release will give an accurate indication of the heat density should be determined at multiple levels or
output of a fire. Three methods by which heat using a single vertical beam from floor to ceiling.
release can be calculated are by monitoring with 7.5.3 Measurement. Optical density meters as
time, either temperature, mass loss or oxygen described in BS 5446-1 can be used for smoke
depletion. measurement. The most commonly used are quartz
Temperature profiles give only a very crude halogen lamps with tungsten filaments and a
estimate of the rate of heat release and use a series focussing lens assembly in conjunction with an eye
of thermocouples to obtain the raw data necessary to response photodiode detector. The detector should
calculate the heat release rate. be calibrated to various levels of incident radiation
Mass loss can be a useful technique provided the using neutral density filters (range ND 0.1 to 0.4).
items involved in the fire are of a single material A frequent source of error in smoke measurement is
and can be continuously weighed (using either a soot deposition on the windows of the detector
load platform or transducers). However, it has the system. This can be overcome by using a honeycomb
advantage that the method can be used in any test grating or allowing a gentle flow of forced air to flow
facility. over the windows.
Oxygen depletion is probably the most commonly 7.6 Air flow
used method since, although based on calorimetry,
7.6.1 General. Air flow rate is a necessary
it is not limited by the material being burnt. Oxygen
parameter to the calculation of rates of heat release,
depletion techniques are limited to test facilities
smoke and toxic gas generation.
where the fire effluent can be channelled into
instrumented ducts so that volume flow can be 7.6.2 Measurement. The usual methods of
determined and the oxygen concentration can be measuring air flow include the following.
measured by taking a transverse sample of the a) Vane anemometer, which is usually used for
atmosphere across the direction of flow, thus cold, low velocity air inflow determination with
avoiding problems with non-uniform flow patterns. the anemometer head sited in the open doorway
7.5 Smoke production or corridor of a test facility. This device is the
least accurate method of determining air flow.
7.5.1 General. The rate of generation of smoke with
b) Pitot tube, which is a differential pressure gas
time during a fire is important since it allows the
flow measurement device capable only of giving a
average smoke concentration to be calculated for
single point reading.
communicating compartments and corridors thus
giving an indication of the visibility problems which c) McCaffrey probe, which is a bidirectional
may be presented to persons escaping fire. The impact pressure tube capable only of giving a
smoke generation rate can be determined by single point reading.
measuring the change in optical density and the d) Orifice plate, which can only be used in ducted
volume of smoke passing the detection system. systems. The plate has to be precalibrated, for
7.5.2 Positioning of detectors. Changes in the example using a pitot tube and the heat output
intensity of a collimated light beam are used to from a propane burner and taking traverses
measure optical density. The light beams may across the duct.
extend from ceiling to floor, pass horizontally across Pitot tubes, McCaffrey probes and orifice plates can
rooms, doorways, corridors or duct systems or be be used for high velocity, high temperature outflow
limited to a few centimetres in length. recordings and thus can be placed directly in the
In choosing one or more path lengths for effluent gases of the fire either at the top of a
measurement, consideration should be given to the doorway, corridor or in a duct as appropriate.
following factors: Two common problems which need to be considered
a) the normal tendency of smoke to stratify, are the effects of thermal radiation on the
particularly in the early stages of a fire; anemometer and sooting of the pilot tube. The use of
the McCaffrey bidirectional probe or orifice plate,
b) routes of spread as determined by the type of
which are not sensitive to either of these factors, can
test facility used (including leakage, see 4.3).
overcome this problem.
In a dynamic fire test, where the smoke generated NOTE When calculating air flow, a correction has to be made
can be expected to flow from the test facility either for air expansion at the elevated temperatures within the fire
through an open doorway, a duct or a corridor, it is environment, therefore temperature measureements have to be
usually sufficient to determine smoke density using taken in the vicinity of the flow measurement device. Also the
calculation of flow has to take account of the influence of the type
only the one system sited so that the beam traverses of flow measured i.e. turbulent or uniform.
the fire effluents.

8 © BSI 02-1999
 BS 476-32:1989

7.7 Fire gas analysis 7.7.3.2 Oxygen is commonly analysed using

7.7.1 General. Fire gas analysis may be carried out paramagnetic or zirconium oxide instruments.
to quantify burning behaviour or to assess life Ideally response time should be short
hazard from toxic combustion effluents. The (commonly 5 % of final reading in 30 s) and the
chemical species found in fire gases can be range of the analyser at least 0 % to 21 %.
considered as follows: NOTE Zirconium oxide analysers can be subject to interference
from unburnt organic volatiles.
a) oxygen;
7.7.3.3 Carbon dioxide can be continuously analysed
b) carbon monoxide and carbon dioxide; using a fixed wavelength infra-red
c) inorganic species, e.g. hydrogen cyanide, spectrophotometer. The response time should be
hydrogen chloride, oxides of nitrogen, sulphur less than 30 s and the range at least 0 % to 20 %.
dioxide etc.; However depending on size of fire load lower ranges
d) organic volatiles, e.g. aldehydes such as may be more applicable.
acrolein. 7.7.3.4 Carbon monoxide analysers are usually fixed
Carbon monoxide, hydrogen chloride and hydrogen wavelength infra-red spectrophotometers with a
cyanide are the most abundant toxic gases range of 0 % to 5 % for full scale fire testing and a
(depending on fuel type), with carbon monoxide response time of not more than 30 s.
usually being the major toxic agent. 7.7.4 Inorganic species. The concentration of
More specific, extensive advice is given in hydrogen halides, hydrogen cyanide and other
ISO/DTR 9122/2 (in preparation). inorganic species can be estimated using
commercially available colorimetric tubes. The
7.7.2 Sampling. Ideally, all gases should be sampled
gases have to be sampled through heated glasslined
from the same position within a fire chamber. The
stainless steel tubing, the fire effluents being drawn
positioning of the sample lines is critical; the
continuously from the fire chamber. Analysis of
composition of the gases extracted will vary with the
several different species can be carried out by
flow patterns within the experimental rig (i.e. the sampling with various colorimetric tubes from this
position of fire zone and airflow to it). heated line.
Maximum concentrations can be achieved by
Analysis of hydrogen fluoride is more complex due
sampling in the gas plume or layer of decomposition
to its high reactivity but use of PTFE-lined sampling
products either just below the ceiling (100 mm) or in
lines does allow an approximate determination of
an exhaust duct.
concentrations.
The sampling point should not be in the zone of An alternative sampling procedure is the use of a
flaming. probe, carrying the colorimetric tube at the end,
Additional positions for sampling in an enclosed rig inserted into the fire gases. Although there can be
would be at “nose” and “crawl” heights to give an procedural difficulties accuracy can be increased
indication of the concentration of gases likely to be using this technique.
inhaled by persons escaping the fire. NOTE Colorimetric tubes should only be used in situations
For room/corridor rigs, an additional sampling where interference from other chemical species will not affect
results (consult manufacturers’ literature). A knowledge of the
point, 100 mm below the ceiling at the end of the elements present can provide a guide to possible errors, e.g. SO2
corridor, would give information on the hazard will not be present if sulphur is absent. The accuracy of the
presented to persons remote from the fire. method is limited and is affected adversely by hot gas streams
above 30 °C. A more sophisticated technique, ion
The length of the sampling line should be kept as chromatography, is available for extremely accurate analysis.
short as possible and the rate of extraction as rapid
7.7.5 Organic volatiles. Analysis for the presence of
as possible to prevent sample loss, further reaction
these gases, some of which may be toxic in trace
of the chemical species in the fire gases and amounts, often requires specialist techniques.
deposition on the walls of the sampling line. Samples collected in evacuated vessels or on
7.7.3 Permanent gases adsorbent solids can be used for assessment of
7.7.3.1 To prevent deposition of soot and water in organic volatiles by subsequent laboratory analysis
the analysers used to monitor oxygen, carbon using mass spectrometry, gas chromatography or a
dioxide and carbon monoxide the fire gases should combination of these and other techniques.
be cleaned and dried. Glass wool and calcium Total hydrocarbon content can be assessed using
chloride are commonly used for this purpose. equipment sensitive to flammable gases. This
Concentrations of these gases are usually monitored apparatus has been used to monitor continuously
continuously throughout the fire test. concentrations of hydrocarbons throughout a test.

© BSI 02-1999 9
BS 476-32:1989

7.7.6 Special information on toxicity. Chemical c) Precautions should be taken to mitigate the
analysis alone does not give an evaluation of toxicity hazard of explosion (see 4.1).
due to the complicated interaction of many chemical d) Instrument cabins should be provided if
compounds with human physiology. Therefore, the possible. Instrumentation systems can be
use of biological models for evaluating the effects of controlled from instrument cabins fitted with
fire effluents will still continue. Further information positive pressure fresh air feed systems. Such
is available in the various parts of PD 6503 and cabins can also provide a place of safety for
in DD 180. observers. This provision enables operators and
7.8 Mass loss observers to monitor the fire without risk of
7.8.1 General. Mass loss of the product tested can exposure to toxic combustion products and
give a good indication of the rate of burning and without the need for individual respiratory
protection.
provide a crude measure of the rate of heat release
of that product provided that it is a single material e) Adequate extraction and effluent treatment
and not a composite. should be available throughout the test to remove
toxic combustion products. Alternatively, the test
7.8.2 Measurement. In its simplest form, mass loss
assessment can consist of a simple measure of the enclosure should have an adequate volume
mass of a product both before and after the test available to allow dilution of combustion products
to “safe” concentrations.
using a simple balance arrangement.
f) Personnel should be clear of the experimental
This type of assessment can be extended to give a
site before ignition.
continuous record of mass loss, if the balance is
fitted with an analogue output, and if small scale g) Wiring should be checked to ensure electrical
testing only is involved. safety.
The product is placed on the balance which is h) Care should be exercised when using primary
connected to a suitable recording media and mass ignition sources (e.g. flammable liquids and
loss monitored constantly with time. The balance gases).
has to be well protected from the heat generated in i) Personnel should not enter the test rig until the
the fire. fire has extinguished and conditions are declared
If large scale testing is required, the continuous safe.
record of mass loss throughout the fire test can be j) Consideration should be given to the issue of a
achieved using load cells or transducers. Two, three safety memorandum in advance of the test,
or four load cells or transducers, linked together, are detailing, for example, identification of safety
positioned beneath a noncombustible tray or board. officer, evacuation procedures and routes.
The output from these measuring devices can be
read directly from a precalibrated meter or can be 9 Data recording
fed into a logging system or potentiometric recorder.
9.1 General
Conversion factors can be used to calculate the mass
loss from the millivolt signal recorded. The recording, storage and presentation of data
acquired in any fire test are important and
8 Safety precautions necessary parts of any fire test procedure. Visual
records together with any output from the
Attention is drawn to the Health and Safety at Work
instrumentation used in the fire test have to be
etc Act, 1974 and the need to ensure that tests are
obtained in a form which is both durable and usable.
carried out under suitable environmental conditions
to provide adequate protection to personnel against 9.2 Video and photographic recording
the risk of fire and inhalation of smoke and/or toxic Records in the form of still and moving visual
products, especially on the occasions when an recordings are valuable for conveying information to
operator needs to work in the vicinity of the fire. other interested parties and for relating
The following precautions should also be taken into instrumental results to events occuring in the fire,
account. Commonly, two systems are used to record visual
a) Adequate extinguishing equipment should be information.
provided, suitable for the size of the fire.
b) Adequate breathing apparatus and protective
clothing should be issued to all personnel in the
vicinity of the fire [see also clause 8 d)].

10 © BSI 02-1999
 BS 476-32:1989

9.2.1 The photographic system. This commonly 9.4 Data replay and processing
consists of a camera with an automatic The instrumental data stored on the disc or tape can
motor-driven film transport, sited at an observation be replayed and the data fed into the memory of a
port in the wall of the compartment or in front of an microcomputer. A variety of software is available to
open vent but away from the fire. If the camera convert the signal recorded on the disc to
cannot be operated manually an automatic unit temperatures, gas concentrations, smoke density,
which permits photographs to be taken remotely rate of heat release, heat output, irradiance etc.,
can be used in a remote instrument room. using previously determined conversion factors.
9.2.2 The video system. One or more video cameras Presentation of data can be either in graphic or in
can be used to provide a permanent visual record of tabular form depending on requirements and the
the test. Cameras can be positioned to cover all availability of hardware i.e. printers and plotters.
aspects of the fire test and can be coupled to Results can thus be provided in a form suitable for
monitors to enable observers to observe the test inclusion in published reports.
from a safe position. If it is possible to control a
9.5 Special treatment of graphics data
camera during the test, it is often advantageous to
take close-up views of interesting events during the Many of the parameters recorded during real fires
test. display some level of transient variations. Optical
obscuration information is particularly susceptible
The use of a digital time generator to provide a
to rapid fluctuation owing to the turbulent nature of
permanent indication of the elapsed time during the
the smoke plumes measured. Graphic presentation
test on the video recording is recommended. The use
of smoke measurements is often difficult to assess
of video recording enables a detailed examination of
because of this transient noise.
the fire test to be made at a later date including
frame-by-frame measurements related to scale With automatic data handling, however, it is
axes. possible to build curve smoothing characteristics
into the program. These techniques usually use data
Because of the great contrast in light levels which
point averaging.
arises when luminous flames develop, it is usually
preferable to set the exposure control of the camera
before the test is started and to leave it on this
10 Test report
setting even though this will result in overexposure The test report should include the following
of flames. It does, however, facilitate interpretation information.
of the development of the fire. a) Laboratory. The name and address of the
NOTE Photographic or video records of a test should testing laboratory.
supplement the recording of visual observations. They should not
be considered as an alternative. b) Sponsor. The name and address of the sponsor.
9.3 Data acquisition c) Test facility. The design and construction of the
fire test facility should be fully described. The
The output from all the transducers and measuring ventilation available to the fire should be detailed
instruments can be wired to a central display and in terms of size and position of openings.
recording area within an instrumentation
compartment. d) Test specimens. Comprehensive information
should be sought and reported to fully describe
The most common method of data output is a series
the item or items to be tested. Construction
of potentiometric pen recorders which give a
details, materials, masses, dimensions,
simultaneous record during the fire test of each preparation, positioning and fixing within the
parameter measured, thus giving direct information test facility should all be detailed within the
on the progress of the fire.
report.
Data loggers based on microcomputer units are now
e) Instrumentation. A detailed description of all
used either instead of or in addition to chart
apparatus and instrumentation used to monitor
recorders. Data logging instrumentation consists of
the progress of the fire and the performance of the
a scanner which will scan across any number of test specimen during and after the test should be
input signals over a preset time interval. The reported. It is essential that the positioning of the
scanner is connected to an analogue to digital
instruments within the test facility are detailed.
converter which converts each signal to a form in
Calibration procedures and predetermined
which it can be stored on either magnetic disc or
conversion factors which are to be used in the
tape. The speed at which the scanner can operate
interpretation of raw data should be reported.
and the number of input channels used will
NOTE Use of a diagram to provide information as to the
determine the sensitivity of the system to monitor dimensions of the rig and the positioning of instrumentation
fluctuations in fire behaviour. is recommended.

© BSI 02-1999 11
BS 476-32:1989

f) Data recording. The methods by which visual h) Experimental results. A full description of all
observations and raw data from the timed events occurring during the fire should be
instrumentation used are recorded, should be reported together with results from each aspect of
reported. Specialist techniques used to prepare the fire which has been monitored. Results are
and present the final data related to the test commonly presented in a graphical form.
report should also be detailed. Diagrams and/or photographs of the damage
g) Ignition sources. The method of ignition used in should be prepared showing area of scorch/char,
the test should be detailed together with the depth of char, depth, width and length of cracks
positioning of the source. The total fuel used and and holes formed and any specimen distortion. A
its type should be recorded together with an sketch showing severe areas of soot deposits is
approximate timescale of burning. Where the also useful.
ignition source is likely to provide a major
contribution to the parameters measured in the
early stages of the fire, full details of the
proportion of this contribution should be
determined and reported by prior calibration if
necessary.

12 © BSI 02-1999
 BS 476-32:1989

Appendix A Fire test facilities b) The open corner, (walls only or walls plus
ceiling) which allows for open ventilation only
The three most common types of fire facility are as
and is frequently used for testing wall-linings and
follows.
internal insulation on a large scale [Figure 1 (b)].
a) The room/corridor, which usually consists of a
c) The single room, again with a variable vent
room with a variable opening to a long corridor opening and often connected [as in Figure 1 (c)] to
along which air enters and the fire effluents can a canopy and duct system used to give an
flow out [Figure 1 (a)]. Ventilation to the fire can
assessment of the rate of heat release from the
be controlled by altering the width of the vent or
burning contents of the fire chamber. An example
by introducing air via a port directly into the
of this type of facility is the room test currently
room.
being developed within ISO (ISO/DP 9705).

Figure 1 — Fire test facilities

© BSI 02-1999 13
14 blank
 BS 476-32:1989

Publications referred to

BS 4422, Glossary of terms associated with fire.

BS 5446, Specification for components of automatic fire alarm systems for residential premises.
BS 5446-1, Point-type smoke detectors.
BS 6809, Method for calibration of radiometers for use in fire testing.
DD 180, The assessment of toxic hazards in fires.
PD 6503, Toxicity of combustion products.
PD 6503-1, Toxicity testing of fire effluents: The state of the art 19871).
PD 6503-2, Guide to the relevance of small scale tests for the toxicity of combustion products of materials
and composites.

1)
In preparation.

© BSI 02-1999
 |
|
|
|
|
|
|
|
|
BSI Ð British Standards Institution
|
|
|
|
|
|
| BSI is the independent national body responsible for preparing British Standards. It
|
| presents the UK view on standards in Europe and at the international level. It is
| incorporated by Royal Charter.
|
|
| Revisions
|
|
| British Standards are updated by amendment or revision. Users of British Standards
|
| should make sure that they possess the latest amendments or editions.
|
|
| It is the constant aim of BSI to improve the quality of our products and services. We
|
| would be grateful if anyone finding an inaccuracy or ambiguity while using this
| British Standard would inform the Secretary of the technical committee responsible,
|
| the identity of which can be found on the inside front cover. Tel: 020 8996 9000.
|
| Fax: 020 8996 7400.
|
|
| BSI offers members an individual updating service called PLUS which ensures that
|
| subscribers automatically receive the latest editions of standards.
|
|
| Buying standards
|
| Orders for all BSI, international and foreign standards publications should be
|
| addressed to Customer Services. Tel: 020 8996 9001. Fax: 020 8996 7001.
|
|
| In response to orders for international standards, it is BSI policy to supply the BSI
|
| implementation of those that have been published as British Standards, unless
|
| otherwise requested.
|
|
| Information on standards
|
| BSI provides a wide range of information on national, European and international
|
| standards through its Library and its Technical Help to Exporters Service. Various
|
| BSI electronic information services are also available which give details on all its
|
| products and services. Contact the Information Centre. Tel: 020 8996 7111.
|
| Fax: 020 8996 7048.
|
|
| Subscribing members of BSI are kept up to date with standards developments and
| receive substantial discounts on the purchase price of standards. For details of
|
| these and other benefits contact Membership Administration. Tel: 020 8996 7002.
|
| Fax: 020 8996 7001.
|
|
| Copyright
|
|
| Copyright subsists in all BSI publications. BSI also holds the copyright, in the UK, of
|
| the publications of the international standardization bodies. Except as permitted
| under the Copyright, Designs and Patents Act 1988 no extract may be reproduced,
|
| stored in a retrieval system or transmitted in any form or by any means ± electronic,
|
| photocopying, recording or otherwise ± without prior written permission from BSI.
|
|
| This does not preclude the free use, in the course of implementing the standard, of
|
| necessary details such as symbols, and size, type or grade designations. If these
|
| details are to be used for any other purpose than implementation then the prior
| written permission of BSI must be obtained.
|
|
| If permission is granted, the terms may include royalty payments or a licensing
|
| agreement. Details and advice can be obtained from the Copyright Manager.
|
| Tel: 020 8996 7070.
|
|
|
|
|
|
|
|
|
BSI |
|
389 Chiswick High Road |
|
London |
|
W4 4AL |
|
|
|
|
|
|