CR 1187:2001

SLOVENSKI STANDARD
01-februar-2002
Preskusne metode za strehe, izpostavljene požaru z zunanje strani
Test methods for external fire exposure to roofs
Prüfverfahren zur Beanspruchung von Bedachungen durch Feuer von außen
Méthodes d'essai pour l'exposition des toitures a un feu extérieur
Ta slovenski standard je istoveten z: CR 1187:2001
ICS:
13.220.50 Požarna odpornost Fire-resistance of building
gradbenih materialov in materials and elements
elementov
91.060.20 Strehe Roofs
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

CEN REPORT
CR 1187
RAPPORT CEN
CEN BERICHT
May 2001
ICS 13.220.50
English version
Test methods for external fire exposure to roofs
Méthodes d'essai pour l'exposition des toitures à un feu Prüfverfahren zur Beanspruchung von Bedachungen durch
extérieur Feuer von außen
This CEN Report was approved by CEN on 19 April 2001. It has been drawn up by the Technical Committee CEN/TC 127.
CEN members are the national standards bodies of Austria, Belgium, Czech Republic, Denmark, Finland, France, Germany, Greece,
Iceland, Ireland, Italy, Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland and United Kingdom.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION
EUROPÄISCHES KOMITEE FÜR NORMUNG
Management Centre: rue de Stassart, 36  B-1050 Brussels
© 2001 CEN All rights of exploitation in any form and by any means reserved Ref. No. CR 1187:2001 E
worldwide for CEN national Members.

Page 2
Contents Page
Foreword 3
Introduction 4
1 Scope 5
2 Normative references 5
3 Terms and definitions 5
4 Test 1: Method with burning brands 7
4.1 Test equipment 7
4.2 Calibration 8
4.3 Test conditions (pitch) 8
4.4 Test specimen 8
4.5 Conditioning 10
4.6 Test environment 11
4.7 Test procedure 11
4.8 Observations and measurement 12
4.9 Test report 13
4.10 Direct field of application of test results
5 Test 2 : Method with burning brands and wind 15
5.1 Test equipment 15
5.2 Calibration 16
5.3 Test conditions 18
5.4 Test specimen 18
5.5 Conditioning
5.6 Test environment 19
5.7 Test procedure 19
5.8 Observations and measurements 20
5.9 Test report 20
5.10 Direct field of application of test results 21
6. Test 3 : Method with burning brands, wind and supplementary radiant heat. 21
6.1 Test equipment 21
6.2 Calibration 23
6.3 Test conditions 24
6.4 Test specimen 24
6.5 Conditioning 25
6.6 Test environment 26
6.7 Test procedure 26
6.8 Observations and measurements 27
6.9 Test report 27
6.10 Direct field of application of test results 29

Page 3
Foreword
This CEN Report has been prepared by Technical Committee CEN/TC 127 "Fire safety in buildings",
the secretariat of which is held by BSI.

Page 4
Introduction
The first mandate given to CEN/TC127 on fire resistance testing in support of the Construction
Products Directive required two test methods for external fire exposure to roofs. One test method
was to include the effect of a burning brand, the other was to include the effect of a burning brand
together with wind and supplementary radiant heat. These two draft methods were circulated for the
CEN 6 month enquiry, but many countries did not support the two and several countries requested
that only one method should be prepared by CEN.
The topic was referred to the EC Fire Regulators’ Group. Discussions took place in both the Fire
Regulators’ Group and the Standing Committee on Construction, and in April 1997 CEN/TC 127 was
requested to produce a standard incorporating the two existing draft methods and the Nordtest
method.
The Standing Committee accepted this as a short-term solution and strongly expressed the view that
a truly harmonised test procedure should be developed in the long term i.e. a single test procedure
for these characteristics.
Caution
The attention of all persons concerned with managing and carrying out these tests is drawn to the
fact that fire testing can be hazardous and that there is a possibility that toxic and/or harmful smoke
and gases may be evolved during the test.
An assessment of all potential hazards and risks to health should be made and safety precautions
should be identified and provided. Written safety instructions should be issued. Appropriate training
should be given to all relevant personnel. Laboratory personnel should ensure that they follow
written safety instructions at all times.

Page 5
1 Scope
This CEN Report specifies three methods for determining the performance of roofs to external fire
exposure. The three methods assess the performance of roofs under the following conditions:
test 1 - with burning brands
test 2 - with burning brands and wind
test 3 - with burning brands, wind and supplementary radiant heat.
The tests assess the fire spread across the external surface of the roof, the fire spread within the
roof, the fire penetration (tests 1 and 3) and the production of flaming droplets or debris falling from
the underside of the roof or from the exposed surface (tests 1 and 3).
Tests 2 and 3 are not applicable to geometrically irregular roofs or roof mounted appliances e.g.
ventilators and roof lights.
NOTE  The three tests listed above do not imply any ranking order. Each test stands on its own
without the possibility to substitute or exchange one for another.
2 Normative references
This CEN Report incorporates by dated or undated reference, provisions from other publications.
These normative references are cited at the appropriate places in the text and the publications are
listed below. For dated references, subsequent amendments to or revisions of any of these
publications apply to this CEN Report only when incorporated in it by amendment or revision. For
undated references the latest edition of the publication referred to applies (including amendments).
EN ISO 13943:2000 Fire safety – Vocabulary (ISO 13943:2000).
3 Terms and definitions
For the purposes of this CEN Report, the terms and definitions given in EN ISO 13943:2000,
together with the following apply:
3.1
roof
covering and sealing systems including any insulating layers or vapour barriers normally provided
together with their supporting elements including attachment (glued, mechanically fastened etc.),
and roof lights or other closures for roof apertures that are intended to provide a weatherproof
surface
3.2
material
basic single substance or a uniformly dispersed mixture of substances (e.g. metal, stone, wood,
bitumen, concrete, mineral wool)
3.3
damaged material
material that has been burnt, melted or otherwise visually changed by heat. Discolouration and soot
deposits are not to be regarded as damaged material

Page 6
3.4
burnt material
material that has been destroyed by combustion or pyrolysis
3.5
composite
combination of materials which is generally recognized in building construction as a discrete entity
(e.g. coated or laminated products such as roofing felts)
3.6
assembly
fabrication of materials and / or composites (e.g. sandwich panels)
3.7
product
material, composite or assembled about which information is required
3.8
specimen
representative section of the roof prepared for the purpose of the test
3.9
continuous deck
element with a continuous supporting function in which the nominal gap between adjacent elements
is not greater than 0,5 mm (5,0 mm in the case of wooden planks with plain edges)
3.10
exposed surface
external surface of the specimen which is subject to the heating conditions of the test
3.11
measuring zone
area of the specimen within which measurements are made
3.12
underside
bottom surface of the specimen
3.13
sustained flaming
flames arising from an observed location, which persist for 5 s or longer
3.14
external fire spread
progression and extent of sustained flaming across the exposed surface of the specimen
3.15
fire penetration
appearance on the underside of any sustained flaming or glowing due to combustion including the
occurrence of any flaming droplets falling from the underside. Charring and/or discolouration are not
to be regarded as fire penetration
3.16
opening
appearance during the test of any hole greater than 25 mm
 in area or any crack greater than 2 mm
wide, which penetrates completely through the specimen and which would allow burning materials to
fall through the roof
Page 7
3.17
internal damage
extent of damaged material in each layer inside the specimen, measured from the edges of the
projection of the ignition source in the upwards and downwards directions with respect to the pitch of
the roof
3.18
fire spread within the roof
maximum burnt length measured from the upper edge of the projection of the ignition source in the
upward direction of each layer for sloped roofs, and in any direction for horizontal roofs (see 4.8.1)
3.19
flaming droplets or debris
burning material falling from the specimen that continues to burn on the floor for at least 5 s
3.20
joint
any kind of connection or link within a layer in the roof such as overlap, seams etc.
3.21
roof covering
uppermost layer of a roof. This layer can comprise single layer or multiple layer coverings
3.22
damaged length
the length of damaged material of the roof covering and the substrate respectively, expressed in mm
as measured from the centre of the wood crib position (for test method 2)
3.23
substrate
product which is used immediately beneath the roof covering, about which information is required
(for test method 2)
3.24
pitch
inclination of the roof to the horizontal
4 Test 1: Method with burning brands
4.1 Test equipment
4.1.1 Basket
A basket as shown in figure 1 shall be used. The basket shall be made from 3 mm diameter mild
steel wire forming a mesh of approximately 50 mm x 50 mm. The basket shall be open at the top
and bottom and have four projecting feet 10 mm long, one at each corner. The outer dimensions of
the basket shall be 300 mm x 300 mm x 200 mm deep. The mass of the basket shall be
(650 ± 50) g.
4.1.2 Wood wool
Wood wool shall consist of fibres approximately 2 mm wide x 0,2 mm to 0,3 mm thick and be
manufactured from softwood e.g. spruce, pine or fir.

Page 8
4.1.3 Balance
The balance is used to weigh the wood wool. It shall have a nominal capacity of at least 2 kg and an
accuracy of ± 1 g.
4.1.4 Timing device
A timing device with an accuracy of ± 5 s over 1 h shall be used.
4.2 Calibration
Each new batch of wood wool shall be checked by carrying out a control test as follows. A sample of
calcium board as described in 4.4.2.2 c), measuring 1 m x 1 m shall be supported horizontally at a
height of about 1 m above the floor. A basket filled with conditioned wood wool as described in 4.5.1
and 4.7.2 shall be placed at the centre of the board and ignited as described in 4.7.5. The burning
time shall be measured from the commencement of ignition until the final flame extinguishes.
Three separate tests shall be carried out. The calcium board used shall be cooled down between the
tests. If the range of the burning times recorded is greater than 2 min, a further three tests shall be
conducted. The mean burning time for all performed tests shall be calculated and shall lie within the
range of 4 min to 5 min.
4.3 Test conditions (pitch)
Roofs intended to be installed with pitches up to 20° in practice shall be tested at a pitch of 15°.
Roofs intended to be installed with pitches greater than 20° shall be tested at a pitch of 45°.
In special cases (e.g. where proof for a specific type of roof is to be provided), by way of departure
from the specifications regarding pitches, verification of the resistance of a roof to burning brands
may also be carried out by testing at the actual intended pitch. The test results so obtained shall
then be valid only for the pitch tested.
4.4 Test specimen
4.4.1 General
Four specimens with minimum dimensions 0,8 m in width x 1,8 m in length shall be prepared under
appropriate supervision by the laboratory for each roof pitch to be tested.
Roof lights shall be tested in the original size, with a waterproofing system attached as in practice.
The specimens shall be representative, in all details of practical application (except for standard
supporting deck), with regard to both the support, and the type and number of layers of roofing
materials (including any insulation, vapour barriers, etc), and with regard to the joining of those
layers. An example specimen section is shown in figure 2.
Where appropriate (see table 1), at least one specimen shall incorporate a vertical joint in the top
layer and another specimen shall incorporate a horizontal joint in the top layer. The other two
specimens may be without any joint in the top layer.
In the case of a roof with jointed thermal insulation, a vertical joint in the insulation shall be included
in each of the specimens. This joint shall be located on the longitudinal axis of the test specimen.
Additional guidance on joints is given in 4.4.3.

Page 9
4.4.2 Selection of standard supporting decks
4.4.2.1 General
When the product being evaluated is a roof covering system intended for general application over a
variety of decks (see 4.10), the test deck shall be selected in accordance with either 4.4.2.2 b) or
4.4.2.2 c).
4.4.2.2 Roof coverings intended to be installed over continuous decks
a) In the case of roof coverings intended to be installed over a continuous deck which is not profiled,
a roof deck in accordance with either 4.4.2.2 b) or 4.4.2.2 c) shall be used. If the materials are to be
laid over a profiled steel deck, then a trapezoidal profiled steel deck, in accordance with 4.4.2.2 d)
shall be used.
b) If it is intended to simulate a wooden continuous deck with gaps not exceeding 0,5 mm, the wood
particle board deck shall be constructed from planks 250 mm wide x 16 mm thick running parallel to
the eaves with plain edges and tightly butt jointed so that the gaps between planks do not exceed
0,5 mm.
If it is intended to simulate a deck made of narrow solid wood planks, the gaps between the particle
board panels shall be (5,0
0,5) mm.
The wood particle board shall consist of wood particles bonded with polymer adhesive (e.g.
ureaformaldehyde); it shall have a density of (680 ± 50) kg/m and shall not be treated with fire
retardants.
c) If roofing materials are intended to be laid only on continuous, non-combustible decks with a
minimum thickness of 10 mm, then the deck shall consist of 10 mm thick reinforced calcium silicate
board, dry density (680 ± 50) kg/m.
d) The trapezoidal profiled steel deck shall have the width of the crown approximately equal to 50 %
of the pitch and a trough depth of approximately 100 mm. The corrugations shall run parallel to the
eaves and be open at the ends.
4.4.2.3 Roof coverings intended to be installed without a continuous deck
The spacing of roof supports of any type shall be in accordance with the maximum permissible
spans proposed by the manufacturer for the particular application but not exceeding the minimum
dimensions prescribed for the specimen in 4.4.1.
4.4.3 Positioning of joints
4.4.3.1 Specimen types
The five types are illustrated in figure 3.
Type 1 : Single central joint in the top layer parallel to the roof pitch. No joints are required in any
other layer (including insulation).
Type 2 : Single joint in the top layer at 90° to the roof pitch 100 mm above the lower edge of the
basket. No joints are required in any other weathering layer. A single central joint in the insulation
parallel to the roof pitch shall be provided, where applicable (see table 1). The joint shall extend from
100 mm below the lower edge of the basket to 800 mm above the upper edge of the basket.
Type 3: No joint is required in any of the weathering layers. A single central joint in the insulation
parallel to the roof pitch shall be provided, where applicable. The joint shall extend from 100 mm
below the lower edge of the basket to 800 mm above the upper edge of the basket.

Page 10
Type 4: A single central joint in the weathering layer parallel to the roof pitch next to the insulation.
No joints are required in any other layer (including insulation).
Type 5: A single central joint in the weathering layer parallel to the roof pitch next to the top layer.
No joints are required in any other layer (including insulation).
Where there are overlapping layers, the position of the joints shall be considered to be the edge of
the upper layer.
4.4.3.2 Types to be tested
According to the different compositions of roofs, test specimens described in table 1 shall be tested:
Table 1 - Test specimens
Specimen Type
st nd rd th
Composition of the roof 1 spec 2 spec 3 spec 4 spec
3 or more weathering 124 5
layers
2 weathering layers 1 2 3 4
1 weathering layer 1 2 3 3
Where the dimensions of the elements of any of the layers are such that it requires more than four
pieces to cover the specimen, or if any of the layers are jointless, then the specimens shall be
fabricated in such a way that those layers are representative. In the first instance, at least one joint in
the layer(s) concerned shall be located underneath the basket.
4.4.4 Edge detailing
No special measures shall be taken by the sponsor to protect the edge of the specimen.
4.5 Conditioning
4.5.1 Wood wool
Before being stored, any compressed wood wool shall be loosened by pulling apart, but not to the
extent that the fibres are separated. Large pieces of unshredded wood shall be removed.
The wood wool shall not be selected for tests until the stabilized moisture content of a representative
sample lies within the range of 8 % to 12 % of the dry weight. This shall be determined by drying a
10 g to 20 g sample in an oven at (105° ± 5) °C. Wood wool used for this moisture content
determination shall not be used for the test.
A suitable amount of wood wool selected for the test shall be stored for at least 12 h in a room at a
temperature of (23
2) °C and a relative humidity of (50
5) %.
The wood wool shall not be exposed to a different atmosphere for a period of more than 1 h before
the test is carried out.
4.5.2 Test specimen
At the time of test, the test specimen shall be in equilibrium with the laboratory environment.
If the materials involved are likely to absorb or contain a significant amount of moisture (liable to
influence the test results), then extra care shall be exercised to ensure that the specimen is at this
stable equilibrium. The moisture content of these materials shall be measured and recorded at the
time of test.
Page 11
Special samples may be used for moisture content determination and may be conditioned with the
test specimens. These special samples shall be fabricated so as to represent the loss of water
vapour from the specimen by having similar thicknesses and exposed faces.
4.6 Test environment
Testing shall be carried out in a draught-free area, in an enclosure of not less than 150 m³ in volume.
The lower edge of the top surface of the specimen shall be (0,75 ± 0,25) m above the laboratory
floor. The temperature in the enclosure shall be (20 ± 10) °C prior to the fire test.
NOTE  An exhaust hood may be fitted above the specimen provided it does not create a
draught over the specimen.
4.7 Test procedure
4.7.1 Protection of the edges
It may be necessary for the laboratory to take measures to prevent failure due to flames passing
around the edges of the specimen. These measures shall not affect the fire behaviour within the
measuring zone (see also figure 4) and shall be fully described in the test report. If it is necessary to
protect the lower edge, this shall be done in such a way so as not to interfere with or obstruct
material flowing or falling from the roof surface.
4.7.2 Filling the basket
Wood wool specified in 4.1.2 and conditioned in accordance with 4.5.1 shall be uniformly pressed
into the wire frame basket, in at least six layers, until the mass of wood wool in the basket is
(600 ± 10) g. The packed volume of the wood wool shall correspond to the internal dimensions of the
basket, excluding the feet.
4.7.3 Positioning of the brand
4.7.3.1 The basket filled with wood wool (the brand), shall be suspended with its base (excluding the
feet) parallel to the nominal slope of the specimen 10 mm from the surface of the specimen so that
this position is maintained throughout the test. On a curved surface the centre of the base of the
basket shall be 10 mm above the surface.
In cases where the weight of the basket will not affect the deformation of the roof or roof covering
during the test, the basket may be placed directly on the roof surface. Any device used to retain the
basket shall not interfere with the burning behaviour of the wood wool.
The brand shall be placed in accordance with figure 4.
4.7.3.2 Roof lights shall be tested with the basket placed in turn at:
a) the upper edge of the roof light, where the slope is smallest (see position a, figure 5);
b) the centre of the roof light (see position b, figure 5);
c) the lower edge of the roof light, where the slope is greatest (see position c, figure 5);
d) on the waterproofing attached to the lower side of the roof light and as close as possible to
the upstand of the roof light (see position d, figure 5).
If the roof light consists of several segments joined together, the basket shall be positioned over the
joint closest to the centre in test b.
A new specimen shall be used for each test if interference could exist between successive tests.

Page 12
4.7.4 Start of the test
For each fire test, the brand shall be positioned on the specimen before ignition. The wood wool
shall be ignited along the bottom of all four sides within a period of 10 s, commencing with the upper
side. The timing device shall be started at the commencement of the ignition of the wood wool.
4.7.5 End of the test
The duration of the test is 60 min, measured from the commencement of the ignition of the wood
wool. Surface flames still existing 30 min after the start of the test shall be extinguished e.g. by
applying a fire blanket, gaseous extinguishing agent etc, taking care not to influence any internal
constructions. 60 min after the start of the test, the roof shall be opened and checked for non-flaming
fire propagation (glowing combustion).
A test may be terminated earlier, if:
a) evidently there is no occurrence of fire (flames, glowing combustion, smoke);
b) the flames reached an edge of the sample;
c) penetration occurs; or
d) there is a risk to safety of personnel or impending damage to equipment.
4.8 Observations and measurement
4.8.1 General
During and/or after the test, the following parameters (see 4.8.2 to 4.8.4) shall be observed,
measured and recorded. Dimensions shall be measured in cm.
Progress of sustained flaming shall be estimated with respect to the base of the flame, in contact
with the exposed surface, and not with respect to the flame envelope.
If the roof is a flat roof with a pitch of zero degrees, then the relevant flame spread is that in any
direction. This is determined by reference to concentric circles using the centre of the basket as a
nominal origin. In such cases, if the radius of spread is L cm, the result shall be expressed as
(L - 15) cm, as an approximation to the distance of fire spread.
4.8.2 External fire spread
4.8.2.1 The time when the sustained flaming has progressed upwards 10 cm, 30 cm, 50 cm and
70 cm from the upper edge of the projection of the brand onto the exposed specimen surface and
when reaching the upper edge of the measuring zone, as defined in 3.11 and figure 4.
4.8.2.2 The time when the sustained flaming has progressed downwards 10 cm, 30 cm and 50 cm
from the lower edge of the projection of the brand onto the exposed specimen surface and when
reaching the lower edge of the measuring zone, as defined in 3.11 and figure 4.
4.8.2.3 The time of occurrence and description of any burning material (flaming droplets or debris)
falling from the exposed surface, as defined in 3.19.
4.8.2.4 The extent during the test of the external fire spread upwards and downwards, as defined in
3.14, expressed as the maximum burnt length from the edges of the projection of the brand onto the
exposed surface, measured at the end of the test.

Page 13
4.8.3 Fire penetration and openings
4.8.3.1 The time and nature of fire penetration, if this has occurred, as defined in 3.15.
4.8.3.2 The time of occurrence and description of any burning material (flaming droplets or debris)
falling from the underside of the specimen, as defined in 3.19.
4.8.3.3 The time of occurrence of openings and their dimensions, as defined in 3.16.
4.8.4 Damage
4.8.4.1 The extent and nature of internal damage upwards and downwards, measured after the test
from the edges of the projection of the brand, as defined in 3.17.
4.8.4.2 The maximum length of burnt material upwards and downwards in each layer, measured
after the test from the edges of the projection of the brand, as defined in 3.18.
4.8.4.3 The extent of area of internal damage, as defined in 3.4.
4.9 Test report
The report shall provide all the information relevant to the selected procedures, the products tested
and the results obtained. Photographs are strongly recommended to illustrate the test results.
The test report shall include the following:
a) reference to this CEN Report i.e. CR 1187 test 1;
b) the name of the test laboratory, the address where the test has been performed, and the date of
the test;
c) the names of the sponsor, the product and the manufacturer of the specimen and its component
parts, if known. If unknown, this shall be stated;
d) a statement concerning the degree of supervision exercised by the laboratory during specimen
fabrication. If there was no supervision, this shall be mentioned in the report, together with the
reason why this requirement was not complied with;
e) a full description of the product tested with a description of the test roof deck. This shall include
the method of attaching the roof covering (nails, fixings, spacing, adhesives, etc.), the density or
mass per unit area of materials and, where applicable, moisture content of the materials used.
Building material classes (e.g. with reference to standards), the nature and quantity of adhesives or
their rate of application may also be necessary to define the product uniquely;
f) description of any actions taken to prevent the flames passing around the edges (see 4.7.1). If no
actions were taken, this shall be stated.
g) for each tested specimen, information concerning:
- the pitch
o
- the room temperature at the beginning of the test, in C;
h) for each tested specimen observations recorded on its behaviour during and after the test
including:
(i) external fire spread, expressed by the time of occurrence, in minutes and seconds, and
description where applicable:
Page 14
- for the fire to spread distances of 10 cm, 30 cm, 50 cm and 70 cm, and to the edge of the
measuring zone, in an upwards direction;
- for the fire to spread distances of 10 cm, 30 cm and 50 cm, and to the edge of the
measuring zone, in a downwards direction;
- any falling flaming materials (see 3.19) from the surface of the roof:
-the burnt length upwards and downwards (see 4.2.8.4);
- the damaged area;
(ii) fire penetration, expressed by the time of occurrence in minutes and seconds and
description:
- any falling flaming materials (see 3.19) from the underside;
- openings (see 3.16) formed;
- any fire penetration (see 3.15) if it occurs;
(iii) damage:
- non-flaming propagation (smouldering or glowing combustion);
- the extent of internal damage (see 3.17) upwards and downwards;
- the maximum length of burnt material (see 3.4) upwards and downward in each functional
layer;
- the damaged area;
(iv) time at which the test is terminated and the cause of the termination (see 4.7.5).
NOTE  For classification purposes, not all measurements taken might be needed but they
can be useful for the extension of the field of application, direct or extended.
i) the range of roof pitches for which the product has been evaluated and the types of deck for which
the report is valid (see 4.10).
4.10 Direct field of application of test results
4.10.1 Pitch
o
Test results obtained at 15° apply to roofs with pitches from 0 up to 20°.
Test results obtained at 45° apply to roofs with pitches greater than 20°.
o o
Test results obtained at a single specified pitch other than 15 or 45 apply to roofs for that pitch
only.
Page 15
4.10.2 Nature of the deck
4.10.2.1 Test with standard supporting decks
Test results obtained with standard supporting decks shall apply to all systems with the same
components (including the thicknesses) installed in the same way, but with different decks as
follows.
a) Test results obtained with a wood particle board deck as defined in 4.4.2.2 b) with gaps between
planks not exceeding 0,5 mm shall apply to:
- any wooden continuous deck with a minimum thickness of 16 mm and with gaps not
exceeding 0,5 mm
- any non-combustible continuous deck with a minimum thickness of 10 mm.
b) Test results obtained with a wood particle board deck as defined in 4.4.2.2 b) with gaps of
(5,0 ± 0,5) mm between planks, shall apply to:
- any wooden continuous deck (see 3.7)
- any non-combustible deck with gaps not exceeding 5 mm.
c) Test results obtained with a reinforced calcium silicate board as described in 4.4.2.4 c) shall apply
to:
- any non-combustible continuous deck with a minimum thickness of 10 mm.
d) Test results obtained with a trapezoidal profiled steel deck as defined in 4.4.2.4 d) apply to:
- any profiled and non perforated steel deck
- any non - combustible continuous deck with a minimum thickness of 10 mm.
4.10.2.2 Test with alternative supporting deck
Test results obtained with an alternative supporting deck shall apply only to that roof system (i.e. the
constitution, materials, dimensions of components, thickness etc. are identical).
5 Test 2 : Method with burning brands and wind
5.1 Test equipment
5.1.1 Air channels equipped with fans and dampers
The air channels are made of steel plates. A removable wall channel section made of steel plates
and non-combustible fibre reinforced calcium silicate boards with a thickness of (11 ± 2) mm and a
dry density of (680 ± 50) kg/m is placed between the lower and the upper air channels. A lid which is
hinged to the lower end of the upper air channel covers the top of the removable wall channel
section. This is shown in figure 6.
5.1.2 Crib ignition stand
The crib ignition stand (see figure 7) consists of a support for the wooden crib and a line burner with
five vertically oriented burner tubes. The line burner is mounted centrally on the support with the
axes of the burner tubes and the lowest edge of the crib in the same plane. The distance from the
top of the burner tubes to the lowest edge of the crib is 30 mm. The crib ignition stand shall be
protected from air currents. This can be achieved by the use of a shield in the form of a box 600 mm

Page 16
square and with a height of 700 mm with an open top. The front side consists of a threshold 100 mm
high. The ignition stand shall be placed centrally in the bottom of the box. Alternatively, the crib
ignition stand may be placed in a fume cupboard according to similar principles.
5.1.3 Wood crib
The source of fire is a wood crib made of eight pieces of pine wood containing no knots and having
the dimensions of 10 mm x 10 mm and a length of 100 mm. Six of these pieces of wood are nailed
to the remaining two pieces at a distance of 8 mm between each of them. The nailing shall be done
from the underside of the crib only (see figure 7). Steel nails with a diameter of 1,0 mm and a length
of 15 mm are suitable. The overall dimensions of the crib will then be 100 mm x 100 mm.
5.1.4 Oven
An oven is used to dry the wooden cribs. It shall be ventilated and capable of maintaining a constant
o
temperature of (105 ± 2) C.
5.1.5 Desiccator
The desiccator is used for storing the dried wood cribs.
5.1.6 Balance
The balance is used to weigh the wood cribs. It shall have a nominal capacity of at least 100 g and
an accuracy of ± 0,1 g.
5.1.7 Timing device
A stop watch with an accuracy of ±1 s/h is used.
5.1.8 Air velocity measuring devices
A vane-wheel anemometer and a hot-wire anemometer or a pitot tube are used for measuring air
velocities.
The vane-wheel anemometer shall have an accuracy of ± 0,1 m/s at a velocity of 2 m/s and
± 0,2 m/s at a velocity of 4 m/s. The diameter of the vane-wheel shall be 100 mm.
The hot-wire anemometer and the pitot tube shall have an accuracy of ± 0,2 m/s at a velocity of
6 m/s.
5.1.9 Flowmeter
The propane consumption of the line burner in the crib ignition stand shall be measured with a
flowmeter. This shall have an accuracy of ± 3 %.
5.2 Calibration
5.2.1 Air velocity
5.2.1.1 Basic calibration
The basic calibration shall be carried out following the installation of the apparatus and whenever
changes occur which could affect the performance of the apparatus.
a) With the dummy specimen (see 5.2.1.1 b) properly inserted in the specimen holder (see 5.7.1)
and with the apparatus running and properly adjusted in accordance with 5.2.1.2 the following air
velocities in the direction of the flow shall be obtained:

Page 17
- (2,0 ± 0,1) m/s at point S500 and S850 (see 5.2.1.1b) on the dummy specimen as measured with
the vane-wheel anemometer (see 5.2.1.2a).
- (6,0 ± 0,5) m/s at the central line of the narrow pass (60 ± 2) mm in the upper air channel (see
5.2.1.2a). Furthermore the air velocities shall be verified with the hot-wire anemometer mounted
through the left and right anemometer probe insertion holes, respectively. These are located in the
bottom of the channel.
b) A dummy specimen shall be made of an ordinary particle-board with a thickness of (19 ± 2) mm
and density of (680 ± 50) kg/m at normal conditioning atmosphere (see 5.5.2) and cut to the size of
the specimen (400 mm x 1000 mm). It shall be plane and marked at the following points along the
center line.
S0:  100 mm from the bottom edge
S500: 600 mm from the bottom edge
S850: 950 mm from the bottom edge.
5.2.1.2 Calibration prior to testing
The following calibration procedure shall be done daily or more frequently if necessary. The actual
specimen shall be properly inserted (see 5.7.1) and both fans shall be running.
a) Calibration for testing at an air velocity of 2 m/s
Adjust the air velocity above the specimen surface and in the air channel with the dampers.
Step 1: Measure the air velocity above the specimen in the direction of the flow with the vane-wheel
anemometer (see 5.1.8) at the point where the centre of the wooden crib is to be placed (see 5.7.6).
The measured air velocity shall be (2,0 ± 0,1) m/s.
Step 2: Measure the air velocity in the upper air channel in the direction of the flow with the hot-wire
anemometer or with the pitot tube (see 5.1.8) at the central point of the (60 ± 2) mm narrow pass of
the upper air channel. The probe is inserted through a hole in the bottom of the upper air channel.
The measured air velocity shall be (6,0 ± 0,5) m/s.
Step 3: Adjust the dampers and repeat steps 1 and 2 until further adjustments are no longer
necessary.
b) Calibration for testing at an air velocity of 4 m/s
Step 4: First, ensure that the procedure in a) is followed. Adjust the air velocity above the specimen
and in the air channel with the dampers. Then measure the air velocity above the specimen in the
direction of the flow with the vane-wheel anemometer (see 5.1.8) at the point where the centre of the
wooden crib is to be placed (see 5.7.6). The measured air velocity shall be (4,0 ± 0,1) m/s.
5.2.2 Propane supply
The rate of propane supply to the line burner of the crib ignition stand shall be (145 ± 10) g/h
(1,3 l/min at 101 kPa and 20 °C). Propane having a purity of at least 95 % shall be used. The air
supply through the holes in the burner tubes shall not be restricted. The blue flames with small
yellow tips thus obtained shall be equal and approximately 120 mm high as measured from the top
of the burner tubes.
5.2.3 Mechanical alignment
The following shall be carried out following installation of the apparatus and at all other times as
necessary.
Page 18
a) Lower air channel
The level near the outlet of the channel is verified crosswise and lengthwise, and necessary
adjustments are carried out.
b) Specimen holder
°
The dummy specimen is inserted, and the level and the pitch (30 ) are verified crosswise and
lengthwise, respectively. Necessary adjustments are carried out.
c) Upper air channel
The height of the narrow pass located in the upper air channel approximately 2 000 mm from the
lower edge of the specimen is verified. The height shall be (60 ± 2) mm, and the measurements of
the height are done at the three anemometer probe insertion holes in the bottom of the channel.
5.3 Test conditions
The specimen is mounted at a pitch of 30° to the horizontal plane, and a forced airflow is passing
over the exposed surface. A burning wooden crib is placed on the surface of the specimen.
5.4 Test specimen
5.4.1 Sampling
The sample of the product to be tested shall be sufficiently large to be representative of the bulk
product, particularly in the case of non-homogeneous products.
A product amount for at least ten tests should be available, each test requiring one specimen.
5.4.2 Preparation of test specimens
5.4.2.1 Specimens shall be rectangular with the dimensions of 400 mm x 1 000 mm.
5.4.2.2 The specimens are normally prepared by attaching the product to a standard substrate.
Standard substrates are:
a) Combustible but not treated with fire retardant
- Wood particleboard of density (680  50) kg/m and thickness (19  2) mm or expanded
polystyrene (EPS) of density of (20  5) kg/m and thickness (50  10) mm.
b) Non-combustible
- Fibre reinforced calcium silicate board of dry density (640  70) kg/m and thickness
(10  2) mm or mineral wool of density (150  20) kg/m and thickness (50  10) mm.
In special cases, in accordance with a specific agreement between the manufacturer and the test
laboratory, on a specific substrate for which the product is to be marketed.
5.4.2.3 The specimens shall as far as possible conform to the end use of the product (see 5.4.2.4).
5.4.2.4 The product shall be attached to the substrate (e.g. glueing, nailing) in accordance with the
directions given by the sponsor.

Page 19
If in practice the substrate is made of any material other than wood (e.g. concrete, cellular plastic,
cork) the product shall be attached to this substrate according to the directions given by the sponsor.
For roof coverings which in practice are built up in situ (sprayed or cast systems), the specimens are
made by use of the appropriate substrate and with a dry film thickness according to the directions
given by the sponsor.
5.5 Conditioning
5.5.1 Crib
Before testing, the crib is dried for 24 h in a ventilated oven at (105 ± 2) °C. After drying the crib, it
shall be kept in a desiccator where it is cooled to ambient temperature before use. The mass of the
crib shall then be (40 ± 2) g excluding nails.
5.5.2 Test specimen
Before testing, the specimens shall be conditioned to constant mass at a temperature of (23 ± 2) °C
and a relative humidity of (50 ± 5) %. The specimens are kept in the conditioning atmosphere until
the tests are carried out.
5.6 Test environment
The test shall be carried out indoors in an environment essentially free of air currents. The room
temperature shall be (20 ± 5) °C.
5.7 Test procedure
5.7.1 The specimen is mounted in the specimen holder. Care shall be taken so that the joint
between the lower edge of the specimen and the bottom of the lower air channel is airtight. In
practice, the bottom plate of the lower air channel may overlap the lower end of the specimen by
2 mm to 3 mm. In some cases it may be necessary to pack air gaps with mineral wool. The position
where the burning wood crib is to be placed shall be marked on the specimen.
5.7.2 The two fans are started, and the line burner in the crib ignition stand is ignited. The two fans
yield a constant airflow through the apparatus.
5.7.3 The loose wall channel section is placed on the specimen and the lid is closed.
5.7.4 The air supply to the lower fan is shut off by closing the lid attached to the fan inlet. The fan
shall not be turned off.
5.7.5 A wood crib is removed from the dessicator, weighed and placed on the crib ignition stand.
The crib is ignited and allowed to burn for 30 s.
5.7.6 Immediately thereupon the burning wood crib is placed centrally on the specimen with its
centre 100 mm from the bottom edge and in such a way that the airflow can pass freely under the
upper six wood pieces. Measures shall be taken to ensure that the wood crib remains in position
during the test. Immediately after this the stopwatch is started. After a period of 15 s, the air supply
to the lower fan is opened again. The line burner may now be turned off.

Constant mass is considered to be reached when two successive weighing operations, carried out at an
interval of 24 h, do not differ by more than 0,1% of the mass of the specimen or 0,1 g, whichever is the
greater.
Page 20
5.7.7 During the test the following observations shall be recorded:
- the time at which the specimen ignites;
- the time at which the flames die out; and
- the time at which the glow dies out.
5.7.8 The test is terminated by extinguishing of the fire on the specimen
- 15 min after the start of the test, i.e. the moment when the burning wood crib is placed on the
specimen; or
- when the flame front has reached the upper end of the specimen.
After the termination of the test, the specimen shall be kept in the apparatus for some minutes with
the fans still running. During this period the specimen is allowed to cool and to give off fumes.
Alternatively, the specimen may be removed from the apparatus and placed in a fume cupboard.
5.7.9 The behaviour of the specimen during the test and the condition after the test are recorded.
This includes melting, foaming, charring, expansion, shrinkage, delamination or any other behaviour.
5.7.10 After the test the specimen is examined and the extent of damage done to both the roof
covering and the substrate is measured. Areas which are only discoloured shall be disregarded.
5.7.11 Three tests shall be performed with air velocities along the specimen at each of 2 m/s and
4 m/s, respectively.
5.7.12 The main apparatus shall be allowed to cool off between two successive tests.
5.8 Observations and measurements
5.8.1 All time observations are expressed in minutes and seconds elapsed from the start of the
test. The test results include:
- the length of damaged material of the roof covering and the substrate, respectively, expressed in
mm as measured from the centre of the wood crib position.
- where required for regulatory purposes or for classification, the damaged area of the roof covering
and the substrate, respectively, expressed in square mm.
...