r r-. '""1 · ... v r-',.---~--.

-~ r~'~~

L• ~,'
-- ..~.'" .... b ...... ~~ ~ .. ''~J101_ ..... ~"III~
I

Fire Research Note

No.155

AN INVESTlGATlON INTO THE FIRE RESISTANCE

OF TIMBER DOORS
"

by

w. A.• MORRIS

January 1971

FIRE
RESEARCH
STATION

© BRE Trust (UK) Permission is granted for personal noncommercial research use. Citation of the work is allowed and encouraged.
 F.R. Note No. 855.
Jsnuary. 1971.

AN INVESTIGATION INTO THE FIRE RESISTANCE OF TIMBER DOO15

by

"w. At. Morris

SUMMRY

An investigation has been carried out into the performance of eighteen"

timber door sets under the conditions of the British standard! fire test.
1Iin attempt has been made to establish some of the parameters which are
significant in determining the" fire resistance of timber doors. Some of "the
important factors which have been examined are: the effect of the depth" of
rebate, the'fit' of the door, and the effect of using an intumescent strip to
seal the edges of the door; the effect of door thickness. and the effect of
ineorporating glazed vision panels in the" doors.

The tests have shown that the 'fit' of a" door can be more important than
the frame dimensions as a factor in deciding the effectiveness of a" door as a
fire barrier. With normal clearances a fire resistance of t hour can be "
achieved by providing a seal at the edges of a door using intumescent
materials. In the case of til sw,ing door having no rebated. frame a seal is
esaential if the door is to achieve recognised fire protection standards" "
Panels of wired glass up to 0.9 m2 in size may be incorporated in tiinber
fire doors. These may be retWinedby untreated timber beading if thr fire
check standards are required. but for a full t hour fire resistance it is
necesaary to protect the beading. Glazing in 1 hour fire doors must be
located in a suitably designed non-combustible frame.""

KEY VORIS: Doors. fire resistance tests. glazing. wood.

Crown copyright

This report has not been published and

should be considered as confidential advance
information. No reference should be made
to it in any publication without the written
consent of the Director of Fire Research"

DEPARTMENT OF THE ENVIRONMENT AND FIRE OFFICES'COMMITTEE

JOINT FIRE RESEARCH ORGANIZATION

 F.R. Note No. 855.
January, 1971.

AN INVESTIGATION INTO THE FIRE RFoSISTANCE OF TIMBER DOORS

by

Ii. AI. Morris

1. Introduotion

Doors form a vital link in the ohain of fire proteotion measures for
buildings. They provideolosure for the openings in walls and it is important
that their presenoe should not substantiallly lower the level of" prot eot ion for
whioh the wall has been designed. The nature of doors with the.faoilit"ies for
eceess. is. such that some inherent weak features are Lnt roduced whioh require
attention to ensure that a satisfaotory barrier to the passage of fire will be
provided, the most important being the olearanoe between the' door and the
frame. without,whioh a door cannot functd on , For this reason n is ·neoessa.:i:y:-
to ,examine the performanoe of a door/frame assembly for its fire proteotion
oharaoteristios rather than the door by itself. The space rLn.Tr ont of, doors'
is kept olear henoe there is some justifioation for relaxing the.rigid
re~uirements whioh are appropriate for walls.

The' majority of 'doors used in residential and of:ti'ice type bUildings 81Xe
made of timber. Previous experienoe has shown that these oan be designed'to'
1,
provide proteotion for a maximum period of 1 hour. A British Standard
prepared after .the last war when there was soaroity of timber,- speoifies the
design features for +and 1 hour 'fire-oheok' doors of hollowoore oonstruction.
Sinoe its publioation many manufacturers have produoed successful designs for'.
doors made entirely· of wood. There are many problems requi:!:'ing resolution in
judging ,the performanoe of doors par-tLcu.Laz-Ly i f they incorporate glazing.
This note desoribes a oo-operative programme of testing with· the 'Doors Committee'
I of the British Woodwork Mamufaoturers Assooiation (BWMA) to resolve some of the
t problems.

2. The test programme

2.1. Scope of the investigation

Eighteen docirs were included in the programme. These were

manufactured mnd supplied complete with frames by members of the British
W,oodwork Manufacturers Asaooililltion. The following factors have been
studied in the programme.
 1• Effect of depth of rebate
2. Effect of fit between do0X and frame
3. Ef:i'act of using an intumescent strip at the edges of the door
4. Effect of thickness of the door
5. Effect of positive pressure
6. Effect of glazing the door.

Sixteen of the doors were designed to give protection against fire for
a period' of up to t hour and' the other -mo for one hour.

2.2. Door constructions

The doors tested were·nominally 1980 mm x 760 mm (6 ft 6 in x 2 ft· 6 in)

and of the type normally found in residential and commercial buildings. The'
essenti!liJ.,features .of each door are summarised in Table 1. The construction
specified in British Standard 459 : part 3, illJ:ustrated in Fig. 1,. 'was used
for. the IIlIl.jority' of the specimens. Two types of door constructed from wood..
particle board, were included .for comparison purposes. In alll, ten different·
types of door were made and these were referred to as types 'A' to"J'.
Types 'A~' to 'H' wer€» nominally t
hour doors and doors type 'I' and IJI, one
,',
hour doors. Types 'A' to 'H' wereo supplied in pairs, giving a total of
eighteen specimens including the one hour doors. Drawings of the doors
'.
tested ~re provided in figures 2 to 19 and a brief description of each door
is given in the Alppendix.

Each door WS!S supplied. complete'with frame; the frames were of

softwoo~of the size specified in B.S. 459, the depth of rebate being
achieved by means of a planted stop. The swing doors type 'D' had no stops •.
The doors were hung on 1t pairs of 102 mm (4 in) butt hinges with the
exception of doors 'E' which were hung on a single pair of hinges. The
.swing door 'D' pivo1ied on pins and was provaded with a spring closure
mechanism located beneath a floor plate cast into the threshold. Vhere I

appropriate the doors were provided with a mortice latch and handles, locks t
were not fitted to any of the doors tested.

- 2 -
 3. The test method and performancre criteria

The doors were tested in pairs, built into a 230 rom (9 in) fletto;n brick
WiaJ.l and were provided with a concrete, threshold. After the frames had been
fi tted, any gaps at the edges of the frame were point ed with vermiculite
plaster., The' tests were caz-r-Led out in the sequence given :iin Table 2.

Table 2 - Seguence of door tests

TesJG Doors '~ ' and 'B '

1
Test 2 Doors 'c 1 ' and 'D '
1
Test 3 Doors 'G ' smd' 'H '
1 1
Test, 4 Doors 'E 1 ' and 'F 1 '

Test 5 Doors 'B ' andt 'D '

2 2
Test 6 Doors '~' and 'G '
2
T'est 7 Do<I>:!'s 'E ' and 'H
,
2 2
Test 8 Doors 'I' andl 'J'
Test 9 Doors 'C ' and 'F '
2 2

The doors were tested in accordance with British Standard 476 : Part 1 .
1953~, one face of the door being subjected to the heating conditions specified
,
in the Standard.

T~ performance of the doors was judged by the criteria of stability and

integrity. Stability is the resistance of the door against collapse and
integrity failure may be defined as the ap psazance of cracks, fissures, or
orifices throUgh which the flames and hot gases may pass. The criterion of
limi ting the temperature rise on the unexposed face is usually waived in tests
on doors on the assumption that combustible materials would not be stored,in
contact with the face of the door.

• The' d~'teTInination of integrity failure can be aomewhat subjective as

B.S. 476 does not specify a precise procedure for this purpose. In the
present investigation integrity failure has been construed as:

The appearance of flaming on the unexpos edt face of the door

persisting for 15 seconds or more, or, the development of a gap
estimatelli to be greater than 6 rom. (t in) wide, accrompanied by
clleirring and glowing in the vicinity of the gap.

- 3 -
 IDle flue arrangements for the furnace are such the t there is' a sli'ght
negative pressure on the fire side of the specimen and the typical pressure
2
tiifferentials across a door of 2 m height are -5 N/m (-:-0.5 mm va) at the top'
2
decreasing to -15 N/m (-1.5 mmwa) at the sill. To determine the effect
of having a positive pressure on t;l1e furnace side test No.6 was conducted
wi th a damper in the fluee outlet of the furnace. The pressure was controlled
to give a value of 10 N/~2 (1.0 mmwa)
at the top of the door and the pressure
2
at the sill level was found to be -5 N/m (-0.5 mnn we) with the neutral axis
oCCiW.rr.i.ng at a height of approldmately 800 mm above the sill.

4. Test results

The records of each test are given in the appendix and are summarised in
Tables 4, 5 arrd 6.

Door tyPe'!, ,

Integrity failure occunred at 20 minutes due to sustained ignition on the

unexpos ed face. This occurred at the head of the door, at ' the gap betweeen
the cicor and the frame. The glazing bead ignited due to heat transfer through
the glass at 2'H- minutes. The door retained its stability for the t hour
period at which'time the test was terminated.

Door tYpe 'B I

1

. Integrity failure occurred at 22 minutes due to the formation of openings

aX, the edge of the dO<ll>r. Ignition of the glazing bead. due to heat transfer
through the glass occUrred at 22t minutes. The door retained its stability
for the t hour period at which time the test was terminated.

Door tYpe '° 1 '

Integri ty failure occur-red at 12 minutes due to the formation of openings
at the edges. Sustained ignition of the unexposed face occurred at 15
minutes (top edge). The test was terminate<1 at 20 minutes with the door
still in position. Ignition of the glazing beam had; not occurred at the
end of the test.

Door tYpe I D1 '

Integrity failure OCCllITed at 11t minutes due to the development of

openings, at the edges of the door. Sustained ignition ocmurred on the
unexposed face of the door on the hinge edge towards the top of the door,
at 1# minutes. The test was terminated at 20 minutes with the door still
in position. Ignition of the glazing beam he<1 not occurred. at the end of
teat.
Door tYpe tEl'
. '- ......

This door (solid particle board core) deformed more than the other doors
am.d integrity failure occurred at 30 minutes due to gap formation; flame
penetration dieL not occur until 35 minutes, at a point near the lower edge
where the bowing was greatest, about 12 mIll (t in). Flames also penetrated
close to the lateh at 38 minutes.

Door tYpe IF 1'

No significa.nt gaps developed between the door and its frame during the
first 30 mrinutes of the test. Fire penetrated through the face of the door
however at 26 minutes and ignition of the facing occurred at 38 minutes. The
test was terminated at 39 minutes.

The door retained its integrity for a ,20 minute, period. Ignition of
the glam.ng bead due to heat transfer through the glasa occurred at 21t minutes.
The test was terminated at 35 minutes with the door retaining its stability.

Door tYpe' 'H l '

Ignition of the glazing bead. occurred at 1st minutes, the'glEJ.S~.startedii

to fallout at 3.3 minute~i14~ At 26 minutes, large gaps had develope(t"iit the.
perimeter,of the door. The test was stopped at 35 minutes with the door
retaining its stability. J, '.'

Integrity failure coeur-red at 12t minutes due to the flame penetrating

the: top edge of the door, (positive preasure}, the flames spread to the
glamng bead at ~ 7 minutes. The door retained its' stability for the t hour
test period.

Glowing of the frame near the latch occurred at 23 minutes and a gap
12 mm in width had. developed by 26 minutes. Ignition of the top glazing
bead occurred. a.t 3<>!- minutes and 15 seconds later the t rea t ed beading around
the centre pane ignited (flames appeared... to have been transmitted from the
upper bead.). The test was terminated at 31 minutes with the doer retad.ning
its stability.

- 5 -
Door tyPe' 02'

Integrity fad.lure occurred. at 28 minutes when gaps developed between the

gl&zi.ng bead! and the door, sustained flaming had occurred by 29 .mtnut ea, The;
door retained its stability for the t hour test pezi.od.,

Door type 'D Z'

The intumescent strip had expanded by 13 minutes, this expansion did not
appear to start until 9 minutes. The upper glazing bead, ignited. at 25 minutes,
and the beading at the side ignited after 29 minutes. The test was stopped
at, 31 minutes, the door retaining its stability.

Door tYpe fE2'

Severe bowing oceurned, with this door (solid particle board) Ell. significant
gap developing by 21 minutes. The top hinge had broken away from the frame
at 24 mi.nut es , This was followed by ignition of the face of the door-, The
door collapsed completely at 30 minutes •

. D6~r·>yPe 'F2'
~

A~though some distortion of this door occurred it retained its stability

amd integrity for a period of 30 minutes.

:Door tyPe 'G2 '

Integrity fad.lure of the door occurred at 17 minutes when the unexposed

facr.e of the door adjacent to the aluminium glazing bead ignited. A.t 24
minutes the glazed vision panel collapsed into the furnace. Significant
gaps had formed at the top edge of the door by 25 minutes.

Ga.ps occur-red, at the edges of the door at 18 minutes. At 23t minutes

ignition of the unexposed face of the door occurred adjacent to the NO
glaz.i.ng bead, flaming also penetrated around the door at the top edge at
about this time. The dOor retained its stability for the 30 minute test
period.

Door tyPe 'I'

Integrity failure occurred at 34t minutes when distortion of the door

enabled flames to penetrate at the top edge. Severe distortion of the
welded steel glazing surround was evident and ignition occurred beneath the
,.
beading at 55t minutes. The door retained its stability for the 11 hour
test period.
- 6 -
Door tYpe 'J'
Ignition of the unexposed face of the door oceuzred at 40 minutes when the
r
flames penetrated behind the steel gla~ng bead. The door retained its
stability for the full test period.
5. Discussd.on of results

5.1. General
Timber fire doors are used in buildings· where. the fire resistance
requirements are for t hour or 1 hour. Of these, doors for the t hour
requirement represent the major usage. Some typical examples of their
use are the front doors of flats, doors in corridors and to staircases,
and doors to service ducts.

The performance of a timber door is measured on the basis of the

criteria given in B.S~ 476 for stability, integrity, and insulation.
The last named require\Ilent is usually waived in the case of doors on the
grounds that no combustible contents will be stored against the door
and hence there is no hazard of their ignition by conducted heat. The
duration for which a door complies with the stability and integrity
requirements enables it to be classified either as a 'fire check' or
as a ·'fire resisting' construction. The foIl-owing specification is
used for this purpose.

Table 3 - Classification of doors

Type of door Minimum duration for compliance
(minutes)
Stability Integrity
t hour fire cheak 30 20
t hour fire resisting 30 30
1 hour fire check 60 45
1 hour fire resisting 60 60

:'·,131;> It is apparent that a fire check door poaeasses lower integrity

than a c:orJresponding fire resisting door. In the majority of cases the
integrity failure occurs by the formation of orifices or openings along
the edges of the doCll<r, (Plate 1) an inherent weakness; due to the nature
of the door construction. The present tests have highlighted the
importance of detailing at the edges of doors and have shown some ways
of improving the resistance· to fire penetration.

- 7 -
5.2. The effect of the depth of rebate

+hour and 1 hour fire

British Standard 459 specifies that for both
check doors the depth of the rebate should be 1 in. The +hour door can
have: plan-tted stops whereas for the one hour door ,they mti.st be cut 'from'
solid material. AdditionalJLy, the frame for the one hour door is 'required.
to be impregnated with flame retardant chemicals.

In the tests on +hour doors, rebate depths of 12.5, 19, end 25 mm

(t, t and 1 in) were used. However as the specimens with the 19 mm (t in)
rebates were p;ovided with additional means for sealing the gaps, the test
data cannot be used for comparative purposes.

The door specimen ~ with 25 mm (1 in) rebates and specimens B1

Bald 01 with 12.5 mm (t in) rebates were tested with their stop sides
exposed to the furnace. The times at which the three doors suffered
loss of integrity by the formation of orifices at the 'edge were 20, 22 and
12 minutes respectively. Measurements on the doors (see appendix) showed
that 'the initial gap sizes we~e 2.8-3.5 mm for ~(at the top of the door
where fail,:re first occurred), 1.5 mm for B1, and' 2.~3.0 mm for 1 • °
The close fit in the case of door B1 enabled it to give a much better
°
performance than, door 1 , slightly better' even than door ~ • A di'rect
comparison between doors A1 and 01 is possible and it is apparent that
with a normal fit, i.e. gaps up to 3 mm between the door and the frame,
it is not possible to satisfy the integrity requirements for 20 minutes
(~ appropriate for t hour 'fire check' doors) if the frame rebate is
12.5 rom (t in). Even with 25 mm (1 in) rebates and a: good. fit, flillilure'
willi. probably occur at the ed.ges before 30 minutes.

The sUing door specimen D1 had, no stops but it ~as a very tight
fit in its frame (average clearance less than 0.5 mm (j64'!:l·n~). The
penetration of the fire through this door occurred at 11t minutes and the
failure would have occurred earlier had the clearances been larger.

5.3. The effect of the fit of the door

Specimens B and 01 allow a direct comparison to'be made on the

1
effect of door fit, i.e. the size of the gaps between the door' and the
frame. Both d.oors had stops prOViding 12.5 mm (t in) rebates with
averSlge gaps of 1.5 I!lIIIl (1/16 in) and 3.0 mm (t in) respectively. The
,
integrity failures were recorded at 22 and 12 minutes respectively.

-8,"
 .: ...
I t is appar-ent the penetration of a' door. by fire .ilS , influenced
~hat

not onlJ; by the depth of rebate but also by the gap sizes,. the latta:
probably being the more import~n~, faCtor. When co~paring.specimens .~
and B it, can ~e seen tha~ with a sma.Her stop, but much tighter fit,:.i;
1
is possible to achieve a comparable performance •

. ' If doors.are to perforIri'their fire .proteation purposes reliably it is

essential that not only should the.frame members be of the right dimensions
but the fit of the door should be tip .bo a specified standard. This could'
be achieved in practice if such doors were hung in their' frames ,at the
.factory and delivered to site as a completeassembly~' This would eliminate
.the normal procedure ·of trimming the door edge to 'fit the 'frame, 'a '·process,
difficult to control. _, .

It is also apparent from the test on 'door'D that a swing d6o;' having
1
no .et ops and no special sealing arrangements cannot ·reach the',standard
required of.a t hour fire .check 'door even if·extremelycarefull.y·made.
If the gaps between the door edge and the' frame, or between meeting edges
'.:" had been 3 mm or. so, penetration of fire. might have occurred in·less··.than
10 minutes • .- , ~ "

. : .
. 5.4•. The effect of sealing the 'door edge
• I .~.

The first penetration ,of fi,e.throug~.a timber. door. g~nerally oc.curs

a.t the edge of the dcea-, usuallly in the upper part. Failure is caused
by the fire exploiting the gaps between the'drior-edge'and the' frame.
Wider gaps. quicken the ,fire penetration as shown in 5.3.' A. door IIlSJy

have a very"good fit and give· & good. performance in a.standard test, but
unless control 'is exercised' .to ensure that"a similar fit is IIlBlintained in'
practice·the behaviour of the door in an actual fire'may not be equally
reliable., One. Cif the ways in which the influence of the gaps' can be
minimised and a good. performance assured'is by the use of'a special
• material along the ,edges of' the door .which have the property of
intumescing at high temperatures. One type of intumescing material is
available in the shape of a' 3 mm x .12.!IlJlll· (t x t'in) strip and this was
employed in eight of. the specimens tested" two of which were the one
hour doee types '.I', and 'J'. There was a ·significant improvement in
the performance of allL the doors where the strip was used (Table 4) with
the exception of· door E ··Thisdoor had a wood pa·rticle board: core and
2•
was only fitted with a single pair of hinges. The door was tested with
the hinge side exposed to the furnace and bowing of the door caused the
top hinge to pull away from the frame. (Plate 2).

- 9 -
 The most significant improvements were in the cases of loosely fitted
doors and the swing doors having no rebated stop. Door C was ~imilar to,
1
Iillld was expos edt in the same manner as, . door C The fire peneh'ation time
2,
for the latter door, which incorporated an intumescent strip, more than
doubled. (Plate 3). The swing door Diwas a very good fit in the frame
(average gap lesa than 0.5 mm (4tt'64 in)); fire penetration occurred at
11t minutes without the strip, but when the strip was fitted no "penetration
hallll occuzred with a similar door lllfter 11. test duration of 31 minutes.
The strips used in the 55 mm door • J', which had 25 mm (1 in) rebated
stops, enabled the door to retain its integrity at the perimeter for 11

65 minute test period, even though the frame had not been impregnated.
The other 55 mm door, doer 'I', failed when the fire penetrated the edge
of the door at 35 minutes. Failure in this case however was due to
excessive warping of the door.

The intumescent strips used in these tests did not·swe11 sufficiently

to seal.the gap~ until the door has been subjected to direct fire attack
for a period of 10 to 15 minutes. In many .situations where a door is
required to' prevent the passage of smoke it is located in & position where'
it may not be exposed to the full severity of fire' and the temperatures
may not be sufficient to cause intumescence of the sealing strip. Hence
the use of these stripe cannot be taken as a means of reducing the smoke
penetration during the early stages of a fire.

5.5. The effect of door thickness

Sd.x of the doors tested were of 40 mmJ thickness as opposed to the

45 mm required by B.S. 459. Four of these doors were otherwise of
B.S. 459 construction and the other two doors had a solid wood particle
board core. The latter t~o doors suffered badly from distortion during
the test and this waa aggravated! by the fact th&t only two hinges had been
fitted to these doors. Door 'Ein faet collapsed into the furnace
2'
30 minutes after the start of, the test." Of the six doors tested, five
were prOVided with intumescent strips at the edge. Door 'G 1 ' with a
25 mm stop and no seal just attainedta t hour fire check performance.
The 40 mm doors appeared adequate for fire check reqUirements and in the
case of five of the doors the fire had not penetrated through the
thickness, of the door panel at 30 minutes. However, in the caee of
door 'Fl' .&40 mm door of B.S. 459 construction, charring and

- 10 -
penetration over a small area occurred on the face of the door at 26 minutes,
(Plate 4) this a'ppearedto, be due to III localised fault in the plasterboard
core. Even on 'the other doors of similar construction there we;~ ~har '
patterns forming at about 30 minutes and therefore there is no safety
. , I . , :
margin in hand for a door of 40 mm thickness and of hollow construction.
It is suggested that for 30 minute fire check purposes a 40 mm door of
either type ia adequate, but for a fully fire resisting door for tne same
perioQ a thicknesa of 45 mm (1 in) may be necessary if this type of
construction is used.

Two doors of 55 mnn thickness, were tested one similar to a B.S. 459
door in construction and the other having a solid chipboard core. ,Door 'I'
the ~.S. 459 opened towards the furnace and suffered integrity,
doo~,

failUre, at 35 minutes, because of excesaive distortion. Both of these:

doors could provide one hour fire resistance if adequate att.ention is ,paid
to sealing the edges of the door. To be certain of a full hour;'s fire, ,
resistance with the door of B.S. 459 construction it may be necessary to,
design a latch me~nism which reatrains the free edge of the door ,more
~ffectively. The distortion may have been exaggerated by the presence of
the offset glazing panel. When eXFosed from the opposite direction the
distortion would be, reduced and an improved performance could have been
expect.ed ,

5.6. Door furniture

An investigation into the performance of door furniture was not

directly included in the programme. It is, however, worth stressing
that the performance of a door is very dependent on the nature of the
fittings used. A lock was not included in any of the doors tested and
previous experience, has shown that unless & lock is fitted satisfactorily
LoceL<Lcea. of integrity can occur. Two of the doors supplied for test
were fitted! with plastic, latches; this was an error which ws;s rectifiedl
before test. It does, however, draw attention to the fact that
materials of low melting point ahou.Ld 'not be used in the furniture of
fire doors where a failure of the'component could lead to a failure of
the door. In addition to plastics, these remarks also apply to
components made from low melting point alloys such as brass and aluminium•

.,. 11 -
5.7. Glazing

If a door incorporates glazed panels, integrity failure may occur

by the ignition of the beading or the frame on the unexposed face due to
heat, being transmitted! through the glass. (Plete 5). This is usually
due to the combined effects of conducti~radiation, and convective
heating. In the tests, ignition of untreated timber beading occurred
in every case except for doors 'e 1 ' and ID1 ' where the tests were
terminated at 20 minutes. The ignition time for the beading varied
between 17 and 31 minutes, and in only one case, i.e. door IB ' was the
2
ignition time just in excess, of 30 minutes. It can be concluded that
a, door having glazing retained by untreated timber beads is unlikely to
attain a t hour fire resistance. For the majority of the doors ignition
occurred between 20 and 30 minutes. One glazed section in door IB2 ' had
the bead on both sides of 'the door painted with an intumescent 'paint.
The failure time of 30t minutes showed that this type of paint can be
t
used successfully to protect the beading of & hour fire resisting door.
(PlaUes 6 and 7). Such a treatment however'would often be outside the
direct control' of the manufacturer. As yet no information is available
on-the cont{nued effectiveness of this method 0t,protection if overpainted
with normal decorative trestmente.·~~a later' date. A similar
performance could be achieve& by providing a decorative metal capping. over
the bead. It therefore seems that to achieve the t hour fire check
standard in a glazed wooden door the timber beading and the frame m~mbers

do not require any flame retardant treatment, but to attain IiIL t hour
fire resisting standard some preventative treatment is necess&ry. To
eli~nate doubts about the durability of flame retardant treatments the

heading should be either non-combustible or covered with a non-comb~stible

material.

No direct relationship was noted between the ignition times and the
size of the glass. There was perhaps a tendency for the smaller sizes
of glazing to result in earlier ignition probably due to a higher level
of radiation falling on the g1al2t:ing bead. This increase in radiation
was attribut&ble to the increased flaming on the exposed face of the door
due to the combustion of the unglazed area.
Glazing beads of 13 mm section were sufficient to retain the glass
in the door even when the panel size was 0.93 :l!l~':. If multiple glazed
panels are used it is important that the intermediate glazing bars are
of sufficient thickness to prevent the fall of glass. Test observations
indicate that the minimum depth of glazing bar should be 60 mm and its
thickness should be at least 45 mm, the rebates for the glass being
worked from the solid material.
- 12 -
 The timber glazing bead. of door C WIllS treated with an in-tumescent
2
paint'but premature failure occurred when the fire 'penetrated between
the planted. beads and the door frame. (Plate 8). It is preferable to
locate the glass in a rebated frame rather than by planted beading on
both-'sides of the glass.

In tw.o tests (doors IG2' and 'H 2') the glazing was retained by
a Lunnzri.um and PVC beading. Door 'G with the aluminium beading failed
2'
) at 17 minutes the glass collapsing before th~ end of the test. (Plate 9).

'Dg2~ IH with PVC beading failed at 27t minutes. The aluminium

2'
beadi~ on the exposed side of the door deformed and came away and the

PVC beading was destroyed early in the test. Neither of these methods
of ret~ining,the glass are considered suitable for fire doors.

'In the case of the two 1 hour doors ('I' and IJI)'~ne glazing was
retained in position by steel beads with inserts of asbestos tape.
. .: )
Ignition in the vicinity of the beading occurred at 5~and 40 minutes
respectively. The buckling of the metal beading was, primarily,
respon~ible for flame penetration (Plate 10) and allowance for expansion
would have minimised this effect. With this modificart;ion this method
of fixing would appear to be adequate for 1 hour fire check doors, but to
prevent ignition for 60 minutes it will be ,necessary to improve :Curther the
method of fixing. The use of stee2 beads with better insulating and
masking arrangements to minimise heat transmission and to prevent the
formation of gaps, or the employment of a separate framing of a sbe's t oa
insulation board or similar ,material seem possible solutions.

The tests have shown that i t is not necessary to limit the size
of gla'zed panels, to less, than 0.9 m2 (10 ft 2) prOVided that adequate
precautions are taken to prevent the ignition of beading and the

r framework. When no precautions are taken even with panels as small as

2,
l 0.2 m ignition of the beading can occur in about 20 minutes. The
factor likely to limit the size of the glass in a glazed door is the
amount of radiant heat being transmitted through the glass. It has
been stated elsewhere 3 that -if the ignition of combustible materials is
the criterion the maximum ac~ptable radiation level is 3.3 kW/m2•
Since the expected maximum radiation level from a compartment fire
e~valen~ to a t hour fire resistance test is 1.~ kW/m2 it would seem
unnecaesary to limit the size of glass in t hour fire doors.
,-'
The stailiility of large glazed panels in one hour fire tests has
not been a subject of this investigation.

- 13 -
 5.8. Direction of exposure

Six types of door were tested both opening towards and away from the
furnace. In only three _cases (types B, E and H) could a direct comparison
be made of the effect of ,exposing one or other of the faces to fire
conditions, the interaction of variables influencing the result in the
other instances. Docrs type 'B' performed similarly in both directions,
doors type 'H' however showed a difference, door 'H 1 ,- openang awaJy from
the furnace failing at 26 minutes and door' H opening t owards the furnace
2'
'failing at 18 minutes. Doors' 'E1 ' and lE ' 'also showed a difference in
2
performance, 'E 1' failing at 30 minutes and IE failing at 21 minutes.
2'
It appeared that if a door had a tendency to deform when SUbjected to
heating, failure occurred at an earlier time when tested opening towards
the fire. The direct fire atrt ack-son the hinges and the latch plate
assists this tendency, This finding is contrary to the cOlIllDionly held
view that exposure of the stop side represents an equivalent 4 or more
ssvere test condition.

5.9. The effect of positivs pressure

Doors '~' and IG were tested under positive pressure of
2'
10 N/m2 (1.0 mm wg) applied to the fire side of the door. The penetration
time of the 'A' type door was reduced' from 20 minutes to 12 minutes.
Door 'G achieved fire' check standard but this result was influenced by
2'
the PVC strip used to seal the edge of the door.
6. Conclusions
An examination has been made of the performance of timber doors subjected'
to the fire resistance test of B.S. 476 : Part 1 : 1953 a limited number of
tests being carried out to establish the effect of certain design variables
and to ascertain the influence of changing the test conditions. 'A total of
eighteen doors .has been examined. Fourtesn of these doors incorporated
1
vision panels of 6 mm (t in) Georgian wired glass, sixteen were designed to
give protection against fire for a t hour periqd and two for a period of J
hour.
This investigation has not been concerned with the use of 'doors and no
attempt has been made to establish the performance requirements for a door
that is to serve a particular purpose. There is a definite need for a
rationalisation of the requirements concerning the fitting of fire doors in
various types of occupancies and in particular it is necessary to draw a
cLea'rer- distinction between those doors which are required as barriers to
contain a spreading fire, and those doors whose main function will be to
restrict the flow of smoke and hot gases.
- 14
 v t
" ' ". I

From this investigation"the following conclusions are, drawn:

1. A door with 12.5 rom rebates is not adequate for 'fire check' purposes
unless tolerances on fit are controlled to better than 1.5 rom.

2. A door having 25 rom rebates will achi,eve the t hour fire check standard
Wi thgaps 'of up to 3 rom but will not generally provt.de a full t hour
fire resistance unless some additional precautions are taken.

3. Doors having no rebated frame, i.e. swing doors, would haws a low fire
resistance ewen if very close tolerances are specified.

4. The fit of a door is relatively more important than the dimensions of

the rebated, frame in determining the fire performance. A door having
clearances in excess of 3 rom is likely to fail before 20 minutes'e-WeD:
,if the rebates are 25 rom in d e p t h . , , '

5. The use of an intumescent 'strip to seal the edges of a' door under fire
conditions greatly enhances the performance of timber doors, including
swing doors without rebated frames.

6. Intumescent seals provide a possible method of achieving a 1 hour fire

resistance without the need for impregnation of the door frame.

7. Use of intumescent strips should not, however, be regarded as a'

substitute for a poor fit as the door would not then be an efficient
smoke bar,ier until the seal had been activated.

8. A 40 rom thick door of B.S. 459 construction can meet the requirements
for a t hour 'fire check' door.

9. If the thickness is increased to 45 rom the door will retain its

integrity for the 30 minute period providing attention is paid to .the
rebates and to sealing the gaps at the edge.

10. Doors 55 rom in thickness can provide a 1 hour fire resistance if care
is taken in sealing and restraining the edges of the door.

1,1. The door frame sizes examined in the investigation were adequate 'for the
periods of fire resistance for which they were designed.

12. Materials of low melting point should not be used for doors furniture
as,their collapse may result in a premature failure of the door.

13. Glazing retained in a door by untreated timber beads would not generally
impair the ability of the door to attain t hour 'fire check' standard.

- 15 -
1.4. II) order to achieve a t ·hour fire resistance standard it is necessary to
treat the timber glazing· beads. Intumescent-paints proved satisfactory
for this purpose but metal ·trim or non-combustible capping would provide
a more durable and reliable protection. Aluminium and PVC are not
satisfactory materials to use for beading.
2
15. Beading 13 mm in depth is adequate to retain glass up to 0.9 m in area
in t hour doCllTs.

16. If the glass is retained by planted beads on both sides flames, can
penetrate beneath the beads. It is therefore recommended that the'
gla~g should always be located in a rebated frame.

17. .Glazed panels up to 0.9 m2 size are considered satisfactory for t hour
doors.

18. Glazed panels can be used in 1 hour doors providing the glass is 'retained
in a suitably designed non-combustible frame.

19. Intermediate bars when using multiple glazed panels can be a source of
weakneaa, These should be at least 60 mm in depth and 45 mm thick for
t hour doors.

20. Doors subject to marked distortion will fail at an earlier time when'
tested opening towards the heat source.

Acknowledgements

The author would like to express his thanks to Mr. H. L. Malhotra for
his guidance throughout the test programme and Messrs. J. S. Hopkinson and
G. D. Simpson for undertaking much of the.experimental work. Thanks are also
due to the British Woodwork Manufacturer's Association who provided the
door-sets for test.

References

1• British Standard 459 Part 3 1951. Fire Check .Flush Doors and Fraines.-
•
2. British Standard 476 Part 1 1953. Fire Tests on Building Materials
and Structures.

3. Safe distances from Wlired glass screening a fire. l'r. Law Institute of
Fire Engineers Quarterly No. 73. March 1969.

4. National Building Studies Technical Paper No.6. Investigations on

Building Fires. Part IV.

- 16 -
 EXPERIMENTAL OBSERVATIONS

Door Type A - 1 Construction - B.S. 459

Thickiiess: - 45 mm
Rebate - 25 mm
2
Glaz:!,ng - One panel 0.37 m , 510 mm x 740 mm
Beading: chamfered hardwood
Exposure - Opening away from furnace

Time to - ~O:OO integrity failure

failure

Test Results
Time from start
of test Observations
min seo
05.00 Flaming of exposed beading

07.00 Exposed face delaminating

08.00 Gonsiderable amount of smoke from top of door'

12.00 Outer skin, exposed faces fallen away completely"

20.00 Occasional flames from top "left gap

22.00 .Top rail alight

27.15 Ignition of window frame, bottom and L.R.S.

29.00 Vertical edges of door burning well

3q.OO Test stopped

~I

- 17 -
Door Type, A 2 Construction - B.S. 459
Thickness 45 mm
Rebate 25 mm

Glazing One panel m2 , 510 mm x 740 mm

Beading: chamfered hardwood treated with fire retardant. paint
Exposure - Opening towards the furnace
A positive pressure was maintained throughout the test
Time to - 12.15 integrity failure
failure

Test Results

Time from start

of test Observations

min sec
06.15 5moke from top and sides to glazing level

08.15 Beading fallen into furnace from top of window

10~45 All panelling and beading alight - exposed face

10.50 Smoke extended to bottom half of door

12.15 Flames from top of door - unexposed face

17.00 Flames spread down to glazing

27.30 Plasterboard core fallen into furnace

30.00 Test concluded, glass still in place. Flaming limited

to glazing area on unexposed face

I~

- 18 -
 Door Type B-1 Construction - glazed , "
.. .. i" .;. f ..'.

Thickness .- In'~45 nun . ..; • • oil ...

Rebate - 12.5 nun

2
Glazing - Three panels of 0.31 m , 530 IIl1Ilt x 585 mm
2
Total &rea 0.93 m
Beading: hardwood. 19 nun x 19 nun on upper panels
19 mm x 13 nun on lower panel

E:x;posure Opening away from the furnace

Time to, - 22.00 integrity failure

fad. lure

Test Results

Time from start Observations

of test I .......

min see
·il....-" r-
05.00 Flaming of exposed face beading

11.00 Slight smoke from top of door

.. .. . ,
,~
..
15.00 Stop breaking away. Lower pane beading falling away...
P • "J\ ", ~. '~I ••

20.00 Furnace visible down closing edge, crack r: 3 mm wide

22.00 Furnace visible along edges below handle level ~~J .. ;;;.
22.30 Ignition of wind~ be~ding on all panes, "
. ~, .

25.00 Ignition of closing edge below handle

27.00 20 mm gap along hf.nge edge, top rail well alight

I 29.00 Vertical edges well alight
30.00 Test stopped. Cross rails were nearly burnt through
and the glass of the top panel was free at the upper
edge, being close to falling out

- 19 -
Door Type B-2 Construction - glazed

Thickness - 45, mm
I '. ~: :. j
Rebate - 12.5 mm

G1a:zing - Three panels of 0~31, m2 , 530 mm x 585 mm

i "

Total area 0.93 m2

Beading: Upper panel, untreated hardwood

Clentre panel, hardwood with intumescent paint

Lower panel, asbestos 19 mm x 19 mm

Exposure - Opening towards the furnace

Time of - 26.00 integrity failure

failure

Test Results

Time from start

Observations
of test

min sec
06.00 'Upper wind.ow beading ignited on exposed face

10.00 Centre beading ignited on exposed face

I:: "
23.00 Glowing of frame near latch

25.00 Gap by latch 70 mm x 5 IDID approximately

"
26.00 G&p increased. to 12 IIUm

30.30 Top 'panel bead.ing, ignited

30.45 Centre panel beading ignited

31.00 Test concluded

- 20 -
Door Type C - 1 Construction - B.S. 459

Thiclmess 45 IIIIlIl

RebateC: - 12.5 mm
2
Glazing One panel 0.37 m , 300 mm x. 1200 mm

Beading: chamfered hardwood

Exposure - Opening away from the furnace

Time to - 12.00 integrity failure

failure

Test Results

Time from start

Observations
of test
., ".
; min sec;
05.00 Warping top right hand corner r
, . ~.
I.', "'..
06.00 Smoke from top of door

08.40 Intennittentflaming from top left hand side of deal:

12.00 Furnac.e visible all round door, gaps exceed 6 mm

15.00 Sustadned ignition on unexposed face

16.00 Top of door ~ell alight

. ..~.

17.00 Top and sides of 'doo:; well alight'

;.'j,
1,8.00 Gap along closing ed€e about 25 mm

20.00 Test' stopped.

- 21 -
Door Type C 2 Construction - B.S. 459

Thickness 45 mm

Rebate - 12.5 mm vith intumescent strip fitted to perimeter of door

2
Glazing - One peneL 0.37 m , 300 mm x 1200 mm

Beading: chamfered hardwood treated with fire retardant paint

on fire side

Exposure - Opening awa.y from the furnace

Time to - 28.00 integrity failure

failure

Test Results

Time from start

of test Observations

min sec
06.00 Smoke from top of door

12·09 T9P.left strip began to operate

12·45 ',. Strip starting to seal at top right

,::.' .;:

13.45 Strip sealed top of door.

16.00 Beacling falling on exposed face

16.15 Sealing strip appears effective down to handle

22.00 Smoke from beacling

26.45 Intermi titent flaming of top left beading

28.00 Failure between frame and beading bottom left

28.55 Ignition of top left beading

32.40 L.R.S. nndow beading ignited

33.00 Extensive flaming on both vertical beads

33.15 Test concluded

- 22 -
Door Type D - 1 Const~~t,ion - glazed.

Thickness - 45 mm, ,single ,leaf of double swing door ,

I • • " . ~ • ,.
~:.~.~L.., _ .J. :

2
Glazing - One panel 0.93
. '
m , 550 Ill.IIJi. x 1675.mm
.

Beadi:ng: ~er~d hp-rdwood, 19 mm x .l2.? mm

Time to - 11.30 integrity failure
fad. lure
r sr ,
Test Results

Time from start

of test .Observations
.:
min sec .. j.:-

08.00 Intermittent flaming from top L.H.S.

11.00 Smoke from. top

. of door,
11.30 Gap a~. closing edge about 6 mIn . , .
14.00 Gaps of 10-15 mm

14.30 Flaming on hinge edge at. top •• I

17.30 Gap~:. ~~ lower half about .25 ..1D.lJI t, I .... - ./1

20.00 Test stopped

- 23 -
Door TyPe D - 2 .Construction - glazed
.... ....
I

Thicknesa - 45 ~ single leaf of double swing' door

.. 2 , "
Glazing - One panel 0~93 m , 550 mm x 1675 mm
".f J .', .. : ":~. -r r t

Beading: ch8.mfered hardwood, 19 ~. x ;12.5 ncii.

Intumescent strip fitted. to perimeter of door :.,.

Time to - 25.00 integri ty failur~ ."

failure

Test Results , .
Time. from start
of test Observations

min sec
09.00 Strip at top of door intumescing
13.00 Moat of closing edge sealed by strip
25.00 Ignition of top bead, - uliexposed face
29.00 Ignition of side bead
30.00 10-1 5 rom inward. warp at closing edge ..
31.00 Test stopped

~ 24 -
 Door Type E - 1

Thickness "'" 40 mm, solid wood particle board core ,'

Rebate - 12.5 !DIn

Intumescent strip fitted to frame

Exposure - Opening awa;y from the furnace '.

Tiime to - 30 minutes
fad.lure

'Test Results

Time from start

of start Observations

min sec
07.00 Furnace visible along hinge. edge and top left of:.,closing
. '.
edge
,. , " ,.
11 .00 L6llI'ge gap top left vertical ediga

12.00 Smoke from top left

.' .
1.4.00 Strip sealing at top
: ".1 .: ., I,

20.00 . 4 !DIn gap lower L.R.S.

.)' :. :. ~ .
30.00 . Bottom edge deformed outwards beyond .!ltrip

35.00 Bottom corner burnt, t.hrm:gh ." ,.

38.00 Flaming around latch arEla

39.00 'Pe'st stopped : ;: I

I
l
f

- 25 -
Door Type E. - 2
Thiclmess
l'
Rebate 12.5 mm
.- " :.
Intumescent strip fitted to frame
Exposure Opening towards the furnace
Time to - 21 .00 integrity failure
failure

Test R.e~ults
Time from start
of test . Observations

min sec
06~30 . ',SmokS from top of door

09.00 Flaming visible (not penetrating) at bottom R.H.S.

13.00 Glowing visible centre L.H.S o
16.00 Smoke from centre LoR.S.
21.00 Bowing at top right corner, with consequent gap formation
, "

23.00 Bowing at top about 10-15 mm with flaming along

. .. . ~ ~

, \
top and sides
24.00 Top hinge broken away
30.00 ~oor fallen into furnace
30.30 Test concluded

. ...
~

- 26 -
Door Type F - 1 Const~ction -:~ • .f? 459." ~ •. ' II~

Thickness - 40 mm

Rebate - 19 mm..
Intumescent strip fitted to frame " I • U

Exposure - Opening away from the furnace • ":. r' 'i.

Time to 26.00 integrity failure

failure :+'

Test Result s . -'f·

Time from start

Observations
of test

min sec '. ,

10.00 Plywood falling from exposed face ........

Plasterboard·: cracked on exposed face "

22.00 So~e plasterboar4 fal1ing~.into furnaCEF, i. .! ~ ~ .' •

25.00 Char:t?-ng of. unexposed race, -·lower .half. centrally.:.

26.00 Small pe~tration on face ... »», .~

"'~'. ~ ...
j' ::"'

30.00 Bottom edge deformed beyond. strip leaVing a gapi]<.. ~

34.00 Strong char. pattern. Strip not fOI'ming smoke seal:: .

38.20 Ignition:of.unexposed faqe--

39.00, . ~est conc luded .. .....

- 27 -
Door- Type F 2 construction...; B.S.' 459'

Thickness; 40 mm

Rebate 19 mm ,J • )',

Intumescent strip fitted to frame , , . " :!"~ . ; '.f: • ~:

, ,
Exposure - Opening towards ':b~~fu,rnace

Time to - No failure , ..' • ~,.. It. '_" <:.

,
', ...
'

failure .... , , .

Test Results ., ,_ ~

Time from'~start .... '. I

of test 'Observations

min sec
CY7 .40 Fierce flames on -exposed ',face + }.

15.00 Gap at '!top R.R.S.' about 5 rum ,;

17.00 Plasterboard panels visible on eip'osed face .......

21.05:,·!. s ",!; :-'R.R.S. t-op 'corner distorted' about' 5-mID

21.40 Smoke from top of ,door momentartly .......

23.00· BoWihg increased,to 10 nnn

", Smoke from aU\}., along top 'edge

32.00 Unexposed panel's beginning to darken""

33.15 Test concluded - door intact except for scorching'of

panels and at top R.R.S. corner

- 28 '-
 Door Type G - 1 Construction -·B.S. 459

Thickness - 40 mm " .....

'-

Rebate - 25 DIJm

2;
Glazing - One panel 0.23 m 480 mm x 480 mm

Beading: chamfered hardwood 19 mm·x 13 IIlJIL .

Exposure - Opening away from the furnace .

Time to - 21.30 integrity failure

fad.lure

Tltst Results

Time from start

Observations
of test

min sec····
11.00 Inner plywood skin fallen into furnace.
r ......'; .
19.30 Gap along hinge edge about 3 mm •• I~,

21.30 Ignition of beading on all but R.R.8., unezpcsed face .

23.00 Ignition of R.R.S. bead, gap at edge 6 IIlJIL

33.00 Glasa falling out

35.00 Test stopped

· .." C·

t" r

. ,
.
~
,l
I

.- 29 -
 ~
Door Type G 2 Construction"- B.S. 459 I, • ":.. • " ..

Thickness 40 mm .....·:·".·:t·.

Rebate 25 mm with 5 mm dia. tubular P.V.C. sealing-strip.

Glazing - One panel 0.23 ni~,480 mm" x 480 mm .: • J

Beadillek -A·1Uminium' 'angle, 1} m x'13 mni..\..x··3.2- mm thick on

both sides
. '....' ........~.
....~
J'

""'
.' ". /lo ,;r I _I.' .••. ,
I;
- . :- t "
Exposure Opening towards the furnace with positive pressure maintained
throughout the test -.,.~ _.: .... _':

Time to 17.00 integrity fad.lure

.. >. ~. .
failure

Test Results
...
: t. " .•. ,
.. '. .~ ~ ..
Time from start
Observations
of test

min
.;.' sec...•..
,.
10.00 Beading buckled exposed face

10.55 Smoke all along R.R.S.

I,

11 .15 Beading bowing at top, unexposed face

.~ .. 2 .

16.30 Beading falling away into furnace from top of panel

17.00 Flames from behind beading, top of panel, unexposed face

22.15 All exposed face beading fallen off

24.00 Glasa fallen into furnace

25.00 Large gaps at top edge of door

30.00 Test concluded - flames not extended beyond glazing

area. Charring just visible R.H.S. unexposed panel

- 30 -
 Door Type H - 1 Co~truction
.. ~.
:":" B.~.•
- ...459..
-.. .... .

Thickness - 45 mID

Rebate - 25 mID ..',,: ".....

',--
2,
Glazing - One panel 0.23 '.m ' .759
.
l\I;D1 x :300 mm
, '. '. ..'
~ .J" .

.: Bea~ng: chamfered hardwood, 19 ~ x . ,12.5.mm

.. ~,

Exposure - Opening away from the furnace

Time to - 18.45 integrity failure .:

failure
_ 'I'
. l_.".·.. , " . 'j

Test Results

Time from start . .-

Observations
of test -._1., -'.. I ~ ~'.
"
'" ."
min sec :: • -~ '.~ .\0.

11.00 Slight charring of window bead

Hole developed ..behind·.'bead .. · . " . :" - '.. '

Igniti.on· of. beading. _L '"

.......
26.00'.; !-,; "'-Frame glowing at -top" v· ..

.. ,'. '- ..
. .~ :, ~ '·l,..l To \ .' •

35.00 Test concluded - .':

• \'. II •• , '• • ~ • ... I 11.:. • r" • ..

r;
" ~.. ...
.".:.
~

.. .
, . • ~ .~. ,". -.".: I' ':' '...J. :: .

t, .:' . . . . .

• ~ -. t I..\. . : c~ :.. .'. .

~... .' ~"."' -.

- 3~
Door TyPe H - 2

Thickness 45 mm
." • • J
Rebate 25 m.m

Glazing One panel 0.23 'm 2, 7'50 mm x ' 300 min'

, :.~

Beadin8:"' 'P;'V .c. q,liadrarit, 9.5:' mm racliu;' on: both faces of

glazing. secured with panel pf.ne. , , ..... .:.. '.

Exposure Opening towards the furnace'

Time to - 18.00 integrity feilure

failure . , ,.r·'

Test Results
p' .1
Time from start
Observations
of test
'0. . ",
min sec
10.15 P.V.C. beading peeling away from', frame:.,

12.00 Beading melting at top' of glazing, unexposed fac,e,:,·

18.00 Bowing towards furnace at :top R.H •. corner, with>~.,~,·

conse~uent gap.~o,rmatton•. B~a~ng burning &nID ,

.• . • • t ~. . ., . . I ' . . " . ",. . I ... ' .

fallling into furnace. Beadd.ng smoking on unexposed

~.

face

20.00 Flames visible top R.H. corner

21.00 Glowing under beading, bottom left of panel

22.00 Beading fallen R.H.S.

23.30 Ignition top L.H.S. by glazing

Test concluded

..

- 32 -
 Door Type I Construction - B.S. 459

Thickness - 55 mm

Rebate - 25 mm.' . Frame ...,. one hour type but no impregnation -:-:'
. ~- ,.'
2
Glaz.:i.ng - One panel 0.36 m , 1200 mm x 300 mID

' .. ". Beading: stainless steel, secured 'wi th' 'screWs, embedded on
woven asbestos • : t._

Intumescent st~ps ~ere"I_ocate4 in; the door perimet~;r and the frame

Exposure - Opening towards the furnace

Time to - 34~45 integrity failure • i.

failure
' .....
Test Results
.. '"t
J { •

Time from start

Observations
of test

min sec
07.30 Exposed face steel beading distorted L.H.S. and .; c,

bottom corners

., '.08' -30 :" ...

•r, • ~ t •• ~ •
- .
Spalling of asbestos facing on exposed face
~!
~ .;.\

Small q.u.~nti.ty-.of smoke.rt op R.H.&•. " .r ...

.'f

15.45 •. i,. ~ .V,erti~l ste.elbeadiIlg buckLed by 20-30 mm on exposed face

~O.~Oo.i' p~' Exposed,-.:t:ace asbeat oe board,.p,eeling· away

32.30 Unexposed R.~.S. bead bu.c.k1iD8

33.20 Char visible t.hrough gap at «top. R.H.S.

34.10 OCCSiSi0IlB;l f1aI!les. frol11 top L .li.S.. .

34.45 Ignition unexposed face t,Op .ccrnera , DOQlr bowed away

from stop. Serious distort~on of beading
~
40.00 Gap to hinge from top

41.30 Top edge flaming

46.10 Window beading buckled away from wood on vertical sides

53.00 Flames around top corners of window beading

55.00 TOp gap 70-80 mm, Extends to 150 mm above handle

55.45 Beading area ignition at centre of R..H.S.

60.00 Flames around beading down to handle

65.00 Test concluded. Exposed face plasterboard sti11 in positi~n

- 33 -
 :4 s !. ' •.
- .(~
... "\.-

• ... r .,.-
,- ''': _~ __ I .~ ... l •

Door Type, J Construction - soli~ ~rticle, boa~ ~ . ~-', .. '. J'

.' '
Thickness - 55 mm
Rebate
I .~ ;: •
- 25 mm, Frame. - one hour construction but ~~o impregnation
2,
Glazing - One, panel 0.5 m 1 OOO"mm :i '500 mm

Beading:" stainless: steel; secured 'by-screws'; 'embedded; on':

woven asbee'tos

Intumescent strips were located in the, door perime~er and: the fram~-, "
." .. .' .o• .

Exposure - Opening away from the furnace

Time to 40.00 integrity failure

failure '-,
. .:.'
"

" .
Test Results

Time from':,;start
,Observations .,
of teat

min sec
, '
,. \, • ..1
14.00 Slight smoke from top L.R.S.

16.15 Smoke emitted R.R.S. above handle

17.20 Smoke from window 'frame' both" sides ,',

..' 4~. -~ r

, ,

.~; '18.1.5.... . Smoke increasing on Ii.H.S. and az-ound- handle \ I • ,- I

29.00 'Steel,beading not buckled; , Smoke from lower·L'.E;~S.

40.00 Ignition of Unexposed face above top beading

Jillames confined to beading at top

48.00 steel buckling only at top edge" ; }

I
.\ • J '',
':':,:~ '52.00 ' ~ ,Bowing outwards at' 'handle 4
I

'1
53.00 Smoke from top L.H. corner
..
65.00 Test concluded

.. , .

34
 Table 1. SummarY of Door Constructions
i

I ,
"
, ,, ,
"
,
! , , , GLAZING BEAD
Depth'
I 'FIT', Type Approximate Area
Door Constrtictign Thickness " of (Approx) of ' of "
Pane
Size
,
Section
Rebated Type of
Ref. (mm) Rebate ( ;mm ;) Seal 2) (mm) (nmi) .. one treatment
(mm): Glas!l. (m side
, I !
..
, ,
A1 459* 45 ,25 >'3. 0 , None 0.37 ' 510' x 740 1,9 Yes None
459 45 25 1.5 None 0.37 510 x 740 1·9 : Yes' None
~
B Fully Glazed 45 12.5 . 1.5 None 3 x 0.31 = 0.93 (530 x 585)x 3 19 Yes None
1 ,
1• None
B Fully Glazed 45 12.5 1,5 None 3 x 0.31 = 0.93 (530 x' 585)x 3 19 Yes 2. Int. Paint+
2
.. 3• Non Combustible
C1 459 45 12.5 3. 0, None 0.37 300 x 1200 1'3 None
C2 459 45 12.5 3.0 IS'" 0.37 300 x 1200 13 Int. Paint+
D1 Fully Glazed 45 Swing Lo1.0 None 0.93 1675 x 550 13 None
D Fully Glazed 45 Swing ~1.0 IS 0.93 1675 x 550 13 None I
2 u-,
E1 Chipboard 40 J2.5 1.5 IS Nil - - - - ""I
E
2
Chipboard 40 12.5 1,5 IS Nil - - - -
F1 459 4D 19 1.5 IS Nil .- - - -
F
2 459 40 19 1,5 IS Nil - - - -
G1 459 40 25 1,5 None 0.23 480 x 480 13 None
l
G
2 459 40 25 1.5 p.v.c. 0.23 480 x 480 Aluminium
H
1 459 45 25 1.5 None 0.25 '750 x 300 13 None
H2 459 45 .25 1.5 None 0.25 750'x 300 P.v.C.
I 459 55 25 1.5 2 x IS ' 0.36 1200 x.300 Steel
J Chipboard' 55 25 1.5 .2 x IS 0.50 1200 x 300 Steel

• 459 - Denotes door to B.S. 459 Construction where applicable

I IS - Denotes Intumescent Strip Seal
+ Int. Paint - Denotes Fire Retardant Intumescent Paint
 (

. ': ~;.,. . ~ .... : to:"

~ ~'i

..... _'. 4'" , _.,

Table 4. sUmmary of· results showing performance of door/frame assemb.IY

.~: ~

- ,.
, "

i
s "
c_
r Time of ignition
Time of gap: Test'
J)oor Direction Pressure of unexposed
i . : formation --face at edge -, .opE3r,ati4n
". ,I
(min) (min) ,(min)
• r:
, , "

A1 ~
,

20 - 22 30
" ! .1

AZ R,; Positive' Not observed ""

121- "'
30' ~'; ... , '.:
, ,
; "
30'('
,B1 , 22- t '
r 25
.,
1. •••

. .. , , - . .. "
.. .. . ' . . "
i,

B'"2 "R 26
"

None 31 ,
:
"
I \ .,
: .. .:
,

. ., . , ,
I
r',
-,
",

15, .
,~

"
P1
"

12' , "
,~20, ! :
';

, ,. .. , ':.. o.
.. :-:':'"
C '
2 , . '.
" Not observed ," Not observed .. ,.'
"

.
"33'" ,
,t~

H~
~
,,
: ,
".,
11t
'0
'" "

~1 ,
i ,)
14f .
'20:\'
.,
., ,
"

:
D
2
;
;
Not observed Not observed 31
, ;
E1 j -e-, 30 ,35 -39,,'
, ' , .-
\

;,
, '
., -
, ,R ;
21
~2 Not observed- '! ' "l ..'-
',. "
'; I',
'30 '. 2
: .;
F1 - 30 26 (on face) 39 I "

"

... -.. R .. ., .... .. ".

,
~2
. ' _ ... ~n

Not observed Not observed 33

"
;
,
: ' ;

G1 ,. " ~'23 Not observ~d - ;35 ~:. \

G2 .. "R Positive Not obse rved 25 ."

30 t
.. ,
Not observed "
\
H1 26 35 ; .,
"

.... -
H R 18 20 30
2 ,
,
I ...~
R 55 34-t 65
..
J Not observed Not observed 66

R denotes door opening. ~owaI'~s furnace (stops ,.expos.ed to hea~i,.ng)

po. 36 - . ~ ... '~

'. .
"
 Table 5. Summary of test results showing performance of glazing

Radiation ..
Time of Test falling -
Door ignition duration on Comments
of beading . -
b(:1~ (min) (It 202f
kW m

A1 27t 30. 2.7 Radiati9I\ failure

A2 17 30 . Positive pressure, fl_ame transfer from

top of door
B1 22t 30 4.6 Radiation failure

B2 3<!t 31 Radiation faiiure'(treated bead flashed

from untreated 'bead above) .
C1 Not observed 20 9.6 Test stopped 20 min
- ...
C2 28 33 Flames under bead
.. . . .. ."- .. , -.
G1 21t 35 Radiation failure, glass fell out at
end of test -
G2 17 30 Positive· pressure, . glass ··fell at 25 min

H1 1st 35 Flames under bead .. .. ..- . , . ..

H2 23t 30t Flames' under bead

.'
D,, Not observed 20 .. ..
5.4 Test stopp~d 20 min
D2 25 31 . ,

..
I 55f 65 1 lii' door
J 40 65 '1 hr door ' . ;

- 37 -
 Table 6. Ranking of doors in order of failure times

. .. - .-. - ~.-

Overall failure Edge failure Glazing failure

Door Time (min) Door Time (min)' Door Time (min),

1 D1 11+ D1 11+ A2 17

2 C1 12 C1 12 G2 17

3 A2 12t A2 - 12-t H1 18t

,":4 G2 17 H2 18 G1 21+

.. 5, H2 . ' 18 ,A1 ' -, 20 'B1 22.1-2

__,6 , H1 1st E2* 21 H2 23t
, ,
" 7 A1 20 B1 22 D2 25
-, ' .
8 G1 21+ G1 23 A1 27t
.. . ~ . . _.
9 E2 21 G2 25 C2 _ 28
--
10 B1 . 22 B . 26 B2 30t
2
·11-- D
2-, 25 . - H1 . 26· C1 ' 20+

12 -B 2 --26 F 1* 30 D1 20+
. 13 F1 26 E1* 30 I 5st
_,' .14 C2 28 F2* 33+ , J 40

.. 15 E1 ,30. ,_ C2* 33+

16 F2 33+ , D2* 31+
17 _ _I 34t 1* -- 34-t. -- -
18 J 40 J* 65+

* Denotes use of intumescent strip •

+ Denotes test terminated at this time

- 38 -
 INDEX

1• Introduction .. 1
2. The test programme
2.1 Scope of investigation .. .. 1
2.2 Door constructions .. 2
3. The test method and performance criteria 3
4. Test results 4
5. Discussion of results
5.1 General .. 7
5.2 The effect of the depth of rebate 8
5.3 The effect of the fit of the door 8
5.4 The effect of sealing the door edge 9
5.5 The effect of door thickness .... .. 10
5.6 Door furniture .. •• .. .. .. 11
5.7 Glazing 12
5.8 Direction of exposure 14
5.9 The effect of positive pressure .. .. 14
6. Conclusions .. .. 14
"Experimental observations - Appendix 17-34
Table 1 Summary of Door Constructions 35
Table 4 Summary of results - performance of door/frame assembly .. 36
Table 5 Summary of results - performance of glazing .. 37
Table 6 Ranking of doors in order of failure times .. 38
 914 max

3'2 minimum
Plywood Top rail
faci ng and s tlles
38x95'3

44·5-_~..
Interrned iot tl
rail

o
co
9·5 Plcster
....
(7)
board
protect tvcz
infilling

Plaster board
.~l-------H--4-protcc t iv C2
infill ing

'-+-~lr--T'-~~--++-~Plywoodfac i n9

Bottom
raB
38x95'3

All dimensions in rnltttmet res - 25·4mm=1 in

FIG. 1. HALF -HOUR TYPE FIRE-CHECK FLUSH DOOR

TO B.S. 459 : PART 3: 1951
 .........•• Tight "
All dimensions in millimcztrczs- 25-4mm=1in
762 ~_20min-gap with occasional

r
florne czmczrging
508

.-
3·8 I· 28
/
2·0
22min- rail alight

005
~52 2·0
Fainy hecvy
smoke from

Z •
· 8min

743
0·6
=l 686
02 butts ;
·
6·4 Gaorqian
w ire d glOss

19·1 FIRE
12'
SIDE

9 BO
.€)-- +- o o
1118
r

i
991
1 686
WINDOW BEAD
DETAIL
Fumoce vis; b
from 22min

:J I·e

152
2·0 ~O Lo- --..I 2·0
~8 ~8 EXPOSED FACE SHOWING UNEX POSED FACE SHOWIN<:
UNEXPOSED FACE SHOWING
GAP A GAP B POINTS OF FAILURE
FIG.2 DOOR TYPE A1
 ~ ~- ~--------------------------------
~GaPA +FIRE
II" ".-------, __

102 butts
-. SIDE

44·45
All dlrnenslons in milli rnetr-es-ze- 4mm =1in .

762
8 'i~~~_ _l ~ Flomas from 12min15s
--.......... Smoka from
2·7
6min 15s
,0:6 0-97 2-7
152
J

r-
686 .
2·5

Flcma sprecd
17min from ~~-#-_..J
198 top

2·0

2-6 0 o
I
1·6

1118 WINDOW BEAD smoke --L-.;:-T"

6 86 DETAIL 10min 50s

L
~1 )
2·5
Intumescent poirn
on fira side of
beading
I
1
1
12 oe
<4'1 4-l
3.5
'----------' ,-6
401 4-1
UNEXPOSED FACE SHOWING EX POS E 0 FACE SHOWING UNEXPOSED FACE SHOWING_
GAP 8 GAPA POIN TS OF FAILURE
FI G.3 DOOR TYPE A2
 •• '~ 'j -... -~ .. ~-. --.,-~-----------------------------
.......

FIRE SIDE
Gap 8

............ Tig ht
All dtrnanslons in mnurnetr-es - 25·4mm =1in
~ 762
('I") 584 Smoke at top of
Q) door from 11 min

1 ',5 {·5 ........

• '·0

102 butts 0'8

6·4 G<Zorgian
wired glass 22min 305
533 19·1 2'8 Ignition of
b<Zading
!
730
2'0

F
o 533 -l S
t>
25 min
Ignition""

~ -Ll==::t=.~~~~~~
1'7

-1 1·5
Furnnce

'I
visible from
20 .
10
610 cra~~'2
i wide
991 i,
I 1 I ·0

I 1
5461
1·5
~
1·5 WINDOW BEAD
DETAIL
27min
gap12·7 to
25·4widcz
...-I=:::::f==:-t!:======d
1~4
~216
.
•
~
1
* 1·0
2'5 2·9 2·5
EXPOSED FACE SHOWING GAP B UNEXPOSED FACE SHOWING
UNEXPOSED FACE SHOWI N G POINTS OF FAILURE
GAP A FIG. 4. DOOR TYPE 81
 . -.... . -----.. -

31·8 12·7

--_.. l
t1-- 7
762
5 8 4 .. _-..--.
--_._
0'7 J
..... Gap A
FIRE SIDE
0·8 0·7
44·5
_1
0-8
...........Tight
All dimensions in mi!limtltres-25'4mm:1in

JAMB DETAIL Jo4r-------~l·e

1 !
[
.:
:
08
1102
102 butts
6·4 GeQr~ian

1
1
.---wIred as
i /'

I
533
I

1 73 o FIRE
Ignition
30min
305
- V'

SIDE
I ..-...-
1-[-
J Ignition
-
J
o I Beads 30min
co -..... painted with 455
0> 533 Intumescent
..... I
Of.'
0 paint "6 lAO
I
-.
.'

~l
1°'7 Glowing
.~

23min
6·4 ga P
-,-j- ~-
at 25m In
i i
610
~ I =I9·1
508:
11
991
, I ,-.-J_
~l I""'" 546--·.. WINDOW BEAD
1 I~ DETAIL
I
J ,.I 3·9 3'0
216
lo-eJ.
3·0
1·0
3·0 3'0 3'9
2·8

UNEXPOSED FACE SHOWING GAP A EXPOSED FACE SHOWING GAP B UNEXPOSED FACE SHOWING
POI NTS OF FAILURE
FIG. 5. DOOR TYPE B2
 , -.,. • »

All dimansions in mlnirnetr es- 25'4mm=lin

+FIRE SIDE
Gop B 6 min puffs of

\U
smoke
102 b tts
762
.
305 229 15min25s
2 29 15min 25s
smoke ----.... smoke
3·6 3·7 ).9 ),7
DETA1~
,
L
JAMB

~
2.Q
152' 3·5 Z·8
2k9 ioturnescent
strip
26min45s
spcsrnodk
GapA fl amas;----+-JL-.)r:=:::±:==i1

r=
6 58
Z·3
6'4 G<Zorgian
wir-ed glass
3·0
17rnin26s
2.7 flaming

28min55s

L 3·7 ] --& !2·5-

1219

I
1 '2.8
ignition
33mine--of"'""""I
flaming
o

r
6 58
1980

WINDOW BEAD
DETAIL 3·1 2·5

L
2·7
1066 B0Jds 30min

1 533
pointed with floma
rcztordont paint
panal---L-I
dorkaning --
•
152 2·7
-.J.
2·0 2·0 2·7
V
2·0 2·7 2·0 z·o 2·7 z-c
UNEXPOSED FACE SHOWING EXPOSED FACE SHOWING UNEXPOSED FACE SHOWING
GAP A GAPS POINTS OF FAILUR E

FIG. 7 DOOR TYPE C2

All dimensions in rniuirnetres -Y .-' r: ~ - •
Radius to suit floor spring
~_~.~~.~.~_._~_ i:~ht 6 /~ l~~~'.-m,-
_~J ~"'>""'>"~~)I/

-,
C\J • 8 min occasional flame from gap
755 ~ "102 -L
I I· Singla Jaaf
Simulated second Smoka at top of door

&4, I
552
6-4 I ... '
I
6.4 -L.5 L '\. ~
/from 11 min

0·5 ! I

.........16min30s
------"' flames

6,4 Georg ian

wired glass
\
\1 4min30s
flaming

«)
CO

-
<D 0·25

Gap 6·4 at
11 min 30s -,
12·7 at 14 min -, .
25-4 at 17min "-
30s

0·25 WINDOW BEAD

DETAIL
- ---
('I')
0
(\f 0.127 I I I
-
9·5 9'5 ~. 5 9'5 -
UNEXPOSED FACE EXPOSED FACE UNEXPOSED FACE SHOWING
SHOWING GAPS SHOWING GAPS POINTS OF FAI LURE .

FIG.S. DOOR TYPE 01 (SINGLE LEAF OF A SWING DOOR)

 FIRE SIDE Simulatad second laof
JAMB DETAIL

_ Tight
102-
C\J -
- 755 l -
fo-102
All dimansions in millimatras -25'4mm =1in
Radius to suit
--
0
- 552
- floor spring Ignition 25 min
/

~
0-3 0'3 0·3 0·3 /
1'4
I /
I
I
1

I! j I

I
I , I

. I
1
I
I

o I

CO 1686
6·4mm
Q')
~ ,. 5 Gaorgian wlred
g l ass 1-5
Ii It
j

I !I

J
9·5.a--'~~~....L1,2·7
i ,
i
1

~.- WINDOW BEAD

('I"l DETAIL
0
C\J . I o· 5 2·0~
I
-.&._ _
I
8'1 8·1 5·5 5·5 8·1 8·1
UNEXPOSED FACE EXPOSED FACE UNEXPOSED FACE SHOWING
POINTS OF FAI LURE
FIG. 9. DOOR TYPE D2
 I __ .,_ ,- ...

All dimensions in millimatrczs- 25-4mm =1 in

762

HI
,,6 J 2·0
'229
•
7·6
2oO
0·8
ETAIL
solid

~L butts
I
Int u rnescent strip

1980

38min fJami ng~

1219
\·4 rv
""
- 1'4 2·8 o 0:' "- 0

991
35min flamin
9\

2-7
]., 3'6 3-6
2·5
I
I
229
30min bowin
to produce 9 ~;\
,"
UNEXPOSED FACE SHOWING EXPOSED FACE SHOWING UNEXPOSED FACE SHOWING
GAP A GAP B POINTS OF FAILURE
FIG.10 DOOR TYPE El
()
 40 "'------
Intu rnescent
Chipboard - - - / strip
solid

All di mansions in mill tmetres- 25-4mm =1in

JAMB
762 6min 30s 23m;n Gap
r
Smok~ 12·7 inwards
5·' t'5 3-0 300
t· o • ..
\

4. ~9 23min __ 23min} Top

L.I.
:J'"
I
76 butts Flom"s\ 457 warped
. ~

2 .,

2·1
1-0
13min
0 Flames -_~r
visiblcz o
(J)
())
..-
··--·------0 t· o o '·0
Smoke
21-0
!I
!
I
!
991 1
i 0·6
1-8 1-0

I
9min Flumes
-visibla but not
~~9 ,~
3-0 2,' 0'4
pene trot in 9
2·1 '-5 ,,0 "0 2·0
UNEXPOSED FACE SHOWING GAP B EXPOSED FACE SHOWI NG GAP A UNEXPOSED FACE SHOWI NG
POINTS OF FAILURE
FIG.11. DOOR TYPE E2
 ............. . ... -~-~~.-------------

Intu rnescent
strip All dlrnensions in millimatras - 25mm =1in
.......... ··Tight

762

,,4
t 1·8 ·152
I
I [
L 102 butts
1'6

1980

1·6
.e- r
J .

··
·.·
··•· 0
0'4

0 [
· Pancztrotion
zernm, <,
Sign',ficanl

~
658 30min ~
991

J
Flo('O(ld
38min

r ·
···•
[
f,52
2·,) 1·4 l ·
5'6 2·0 4,4 4-4 2·0 4·4
UNEXPOSED fACE SHOWING EXPOSED fACE SHOWING UNEXPOSED fACE SHOWING
GAP A GAP B POINTS Of FAILURE
FIG.12 DOOR TYPE Fl
All cnrnensions In mnumetres > ,-"

25'4mm = 1 in
--------r. Gap A 21min40s smoke momantarily

r
I ntumescent stri P . f
102 butts 762 I ~ 7"-...dII'" ~-+-..:lo.-~~
40mm 459 construction
l smoke
23 min 20s
Gap 6·4 at 15 min
Gap 4·8 at 21 min 5s

• 10·7 2·8 '3-7

3·7
1 .', , ,--..J..,r----......
1· 4
---..,NS
1'6
GoP 9'6 at 23mifj

~1.8J'9.1 L 33 min 155

JAMB DETAIL scorchi ng at
(2nd of tczst

3-0 0-7
2·0

32 min
charring on
3·1 O· 2-3 o unexposed face

an
Q)
CD 1·1
2·5
~

0>
(J) 3'7 3·5 ..
L.
,7_
I
I
~.
~

3·1 3·5

UNEXPOSED FACE
3'9
3'0
- 2·7
4·0 3·6

EXPOSED FACE
3'1
1·0

UNEXPOSED FACE SHOWING

SHOWING GAP B SHOWING GAP A POINTS OF FAILURE
FIG. 13. DOOR TYPE F2
 T FIRE SIDE
35
Gap 8
82-6

All dimensions in mttltmetr-es- 25·4 = 1in

- 762 - 19 min 30s

3'2 gap
I· ,
483
'·0
- ,.... DETAIL 1-8
\

'-0 I"BT
~2
26min
19 gap - \
T--- I
If
102 butts

----.
r
483
II
J
II
33min
glass
falling
-
~- ..
!
1'51
-, " • >...... -
685 6·4 Georgian

J
~~ wired glass 21min 3 05- ~--
Ignition 01
0 19·1 beading
Q) I·
(J)
0r-

~~ '·0 O I o 1·0 0 [
:-aT
I

1345 I
685
~40~
991 J WINDOW BEAD
i
I DETAIL

I I~ [
,. 1,,06 3·0 4-7
'152
o-s, I-g
.. ·7 3·0
I-a

UNEXPOSED FACE SHOWI NG GAP A EXPOSED FACE SHCM'I NG GAP B UNEXPOSED FACE SHOWING
POINTS OF FAILURE
FIG.14. DOOR TYPE "G1
 -FIRE SIDE •
--------. -- --I -.,..
Gap A

........-.. Tight
All dimensions in millimetres - 25·4mm=1in

~ 762 ..... From 11min

spasmodic smoke
102 butts
-,
~ 483
~·o
.. JAMB \
().9 3·'" 4·6 f ·1
• 0·7 '·4
~-~ Flames
\
T---
from
17 min

J 3 , ·6 25min

J
glass
558 fallan int0
0'3 tur-nnce
.~ -_...,4
1·8
o --' ._ ..._....... 2-0
a)
L-- .. ... ..
- - ........ _-{) 4·6
FIRE
SIDE
0 0

r
Aluminium
1345 alloy angle
13x13x3'2
5· ..
0'7 2-0
991
"8
658

L
-f·~-
-
._~
II
6·4 WINDOW BEAD
DETAIL t'1
1'6

I' 152• ,1,8

3·4 3,9 3'5
18'4~ O·7.....
3'5
---_.a:
3'9 3·6
UNEXPOSED FACE SHOWING GAP B EXPOSED FACE SHOWING GAP A UNEXPOSED FACE SHOWING
POINTS OF FAILURE
FIG.15. DOOR TYPE G2
 All dimenstons in millimatras - 25·4mm =, in
•••...•..•. Tight
762 butts
762 26 min
. f 01'I ure
~P"dga
[ .. 305 -I
J'8
-\ 45
\-8 JJAMB "-
I
, <,
0'S
2 03

! ~

r 6-4 GfZorgian
18min
--- _.----
ignitio n of
be odin 9
wirad gloss
6 60 761

L 0'6
~ -
'-7
006
"

, 198o
a ] 0

I
6 60
FIRE
SIDE
... 30minI~
12,7 go P
~

L
1
)
991
WINDOW BEAD
DETAIL
]
2 29
j J'O I'" 3-0
5'1 5'1 ~5'-1-------=5~'1
UNEXPOSED FACE SHOWING EXPOSED FACE SHOWING UNEXPOSED FACE SHOWING
GAP A GAP 6 POINTS OF F'AILURE
FI G. 16 DOOR TYPE Hl
 ::-- ---.,; --- ..... -, . _ - - - - - - - - - - - -
Gap A
fiRE SIDE ~
,- AU dimansions in milli metres - 25·4 =1in

45 20min nome
25 min flo ming
76 butt j visibla

~I~
~
762
18 min bowing

J
203
2'0
)
I
32'4
0'5
- J "0
G
JA
2·5
23min~Os
.inwards

j
r igni tion,r1t--rr==:::;="

r:
660
10()

19 eo
Quadrant PVC
b<zading

6-4 Gaorgian
wire d gla ss 1-1

L 1 0
1·4 -9-5
l
ITfiRE o
21 min
glowing
visible
o
r
660
2·0

WINDOW BEAD
SIDE
.......-

L 1
1'4
991

I
DETAIL

1
L 4'8
UNEXPOSED FACE SHOWING
, 4-8
JJ
1·4
4'8
EX POSED FACE UNEXPOSED FACE SHOWING
GAP B GAPA POINTS OF fAILURE
FIG.17 DOOR TYPE H2
 . -- Gap A SIDE
'--~-~-.--------------

-,.. ... -
55 (1 hour) ..
All dtrnonstons in millimetres-25·4mm=1in
13 min Smoke
33 rnln 20s char
~~~~~~-- exposed through gap

L...o 762 -. 18mln 30s 38min 40s

102 b\tts 330 114,. 34min 1Ossmokin~ fl~mczs
',0 2tf 3" 0_5 '-4 '·6'·4 flames
~ 2£>3\6 3-9 " sposrno die
34min4 5s'--
~ ~-_.-----.
Ignition - V
• 55min
76 gap
/'
0'7 IlL---lr-
2.0 53min_·
6'~Georqian tlcrnes
1·8 3'9 WIred glass spraad
635
55 min 45~s~I--_ _--f"-- ___
0
CO
Q)
......
L ,.$l ... .
~
. _ ..._---- -0
121 9

3·5
1 ·1

o 2·0
Ignition

60min 0
II r- flo mas ---I-1I1----~H---___iBt

T
l I
spraad

! 635 • .
2,0
55min 45~
~1
\ 1·8

J:9~'7
,
I Stolnlass steal flpmClS ---v
, framcz Visible
;
! WINDOW BEAD
I 63 5 DETAIL
!

I
i
:

--.- I-
I
294 t·A
3,5 5-2 3·5
~ 1~
3·5. S'2
t~
3·5
UNEXPOSED FACE SHOWI NG GAP B EXPOSED FACE SHOWING GAP A UNEXPOSED FACE SHOWING
POIN TS OF FAILURE
FIG. 18. DOOR TYPE I
 ..... ... ~ - ~ -- -..-..--,.--------------

I FIRE
t SIDE
Gap. 1I:8~~~~~
Chipboard cora -r--"'t:: 87
(1 hour) 55 -..c.....£..

L ....-~""'_"_..£......tC._.L......C__~.~~--L.- All dimensions in rnllltrnetr-es -25·4mm =1in

....
"'"
762 - ·'''-·Intumescent 14min
102\utts 127. _
"
~
520
0·6
-... ()~
JAMB
..:...·...
8
strips
---:.1·...
8
Slight smoke
0·7 2'0r- 1'4
2t>3'\0

-
l
40m iil
\ .- f---, .....
Igniti on- -_. ..
18min 15s

T
0·6
"4 Smok a
17min 205
635 97 7 6-4 GaorQ i an 0·7 Smok <l ___
0'5 l'S wi red g I(iss
1.'8
---..
16min 155
, SliQht
0 srrfoke
CO 1·0 I ...... 1·6 0'5
(J) _J ...."
',6 0 29min 0
~

T
~ Ir
-- -=4 ~ Smok a

635 \ 52mi n--

Bowin
outwa ~d5
----------
991
I 07
88
\
Stainless
steel frame
...JL.-
~4~ WINDOW BEAD
DETAIL

n
,;-
2~3
35 3-5 \,07
. ' ..
. 1·4
3·5 3·5
0'5
UNEXPOSED FACE SHOWING GAP A EXPOSED FACE SHOWING GAP B UNEXPOSED FACE SHOWING
. POINTS OF FAI LURE
FIG. 19. DOOR TYPE J
 _._-~--~---------------

FLAMES PENETRATING AROUND THE DOOR EDGES 17 MINUTES

AFTER THE START OF A TEST
PLATE 1
•

•
 f '-- .... _~

,
.<

DOORS HAVING THE EDGES SEALED WITH AN INTUMESCENT

STRIP AFTER 30 MINUTES OF TEST. DOOR 'C' IS A POOR FIT
PLATE 3

".......
 ~
8
e
e
0
~

«
~

p::
[g. 0
0
8
~
0 ~
~
~
o E-'"
«
~ j
~ ~

~
:I:
o
.. W
t- ~
0
p::
~
z
-
0
E-'"
~
) E-'"
~
~ Z
I ~
P-l
....
~
\
"
 FLAMING OF GLAZING BARS 23 MINUTES AFTER
START OF TEST
PLATE 6

.,

FLAMING OF TIlE UNPROTECTED BEADING AT 30! MINUTES

AFTER THE START OF TEST - PROTECTED BEADS STILL NOT IGNITED
PLATE 7
t
I
'tll
L,.
~
.........---

FLAME PENETRATION BENEA TH THE BEAD WHEN THE

GLAZING IS NOT LOCATED IN A REBATE
PLATE 8
'IIIIIIIIIIIP" ~~,...........,

, ---... ~-~. ------------------

COLLAPSE OF THE GLAZING RETAINED BY ALUMINIUM BEADS (DOOR G

2)
PLATE 9
 Ii ......

EXCESSIVE DISTORTION OF THE GLAZING SUBFRAMES

BECAUSE NO ALLOWANCE HAS BEEN MADE FOR EXPANSION

PLATE 10