Report of Committee On Fire Tests: ! . E. Miller° Factory Insurance

1376
REPORT OF THE COMMITTEE ON FIRE TESTS Fr-1
Report of Committee on Fire Tests

J a c k A. Bene, C/m/,'mcm,
U n d e r w r i t e r s ' Laboratories, Inc., 806 P f i n g u t e n Road, Northbrook, I L 60062
M. B. A b r a m s , P o r t l a n d Cement Dr. A. F. Robertson" Institute of Applied
Assoeiatlon Teehnolog-y, Nattonal B u r e a u of
I r w i n A. B e n j a m i n . National Bureau of Standards
Standards J o h n ~ Ryan" National Forest Products
J e l m A. Blair, Societ~ of the Plastics Association
Industry Louis 8effal, F i r e Marshals Association of
Rlelmrd W. Biatzaeker, Columbus, O H North America
H e r b e r t B. Carlsen, Gypsum Association Gordon W. Shorter, N a t i o n a l R e e e a l ~
Buell B. Dnttem, Building Officials & Code Council of C a n a d a
A d m i n i s t r a t o r s I n t e r n a t i o n a l , Inc. Louis W. Vaultlmn, C a n a d i a n Sheet Steel
Rielmrd G. Gewaln. A m e r i c a n Iron & Steel Buildin~ Institute
Institute Calvin H. Yuill, Southwest Research
Gerald E. Linffenfciter, A m e r i c a n Institute
I n s u r a n c e Association
Gorald L. M a a t m a n " Chicago, I L Alternat~
W. F. M a r o n | , F a c t o r y Mutual Research David E. Brnckett, Gypsum Assn. (Alter.
Corp n a t e t o H e r b e r t B. Carlsen)
!~. E. Miller° F a c t o r y I n s u r a n c e B. J . Callahan" F a e t e r y Mutual EnWlneer.
Association i n g ( A l t e r n a t e to W, F. Maroni)
l l e a r y Omsen" Mobile Homes S. K. Goodwin, F a c t o r y I n s u r a n c e Assn.
Mann f a e t u r e r s Association ( A l t e r n a t e t o E. E. Miller)
N o r m a n 8. P u r e e , U n d e r w r i t e r s ' G . M . Watson, A m e r i r a n I n s u r a n c e Assn
I~boraturtes of Canada ( A l t e r n a t e t o Gerald F,. Linftenfelter) "
NoavoUnlr
AI Bartesle, Philadelphia, P A
TM8/let r e p r e s e n t s the membereftlp a t tAe time the Committee was balloted oa the text ot
tJti8 edition. Since t h ~ time, cho/nffes iyt the ~nembership m~# h~t~Te~ .
This report has been submitted in two parts to letter ballot of

the Committee on Fire Tests, which consists of 20 voting members.
Part I consists of the standard text (Chapters 1 through 6). Part

II consists of Appendices A, B, C, and D.
On Part 1 18 have voted aO~rmatively, Messrs. Maroni and Ryan

have voted negatively.
On Part 1I, 16 have voted a~rmatively, Messrs. Blair, Maroni,

and Ryan have voted negatively, Mr. Lingen[elter has requested to
be recorded as not voting.
The Committee on Fire Tests recommends the tentative adop-

tion of this Standard Test Method [or Measuring the Smoke
Generated by Solid Materials.
GENERAL
1377
258-5
Tentative
Standard Test Method for Measuring the Smoke
Generated by Solid Materials
NFPA 2 5 8 - T ~ 1974
Chapter I General
1-1 Scope. This method of test covers a procedure for measur-

in~ the smoke generated by solid materials and assemblies in
thickness up to and including 1 inch (25.4mm). Measurement is
made of the attenuation of a light beam by smoke (suspended
solid or liquid particles) accumulating within a closed chamber
due to nonflammg pyrolytic decomposition and flaming combus:
tion. Results are expressed in terms of specific optical density, 1
which is derived from a geometrical factor and the measured Ol)-
tical density (absorbance), the single measurement most character-
istic of the "concentration of smoke." The photometric scale used
to measure smoke by this test method is simihr to the optical
density scale for human vision.
1-2 Significance 2
1-2.1 This method provides a means for comparing the
specific optical density of the smoke generated by materials and
assemblies in the form and thickness tested, and under the speci-
fied exposure conditions.
1-2.2 Values determined by this test are specific to the speci-
men or assembly material in the form and thickness tested and
shall not be considered inherent, fundamental properties of a
given material
1-2.3 The values stated in U. S. customary units shall be re-
garded as the standard. The metric equivalents of U. S. customary
units given in the standard may be approximate.
1A commentary describing the significance of specific optical density and ap-

propriate considerations for application of test results is included as Appendix C.
2Attempts are now underway to relate the results of this test to the measure-
ment of smoke generated under large-scale test conditions. Other test methods
for measuring smoke have been reviewed and summarized in " T h e Control of
Smoke in Building Fires -- A State-of-the-Art Review," Materials Research and
Standards, pp. 16-23, 42, April 1971.
1378
258-6 TEST METHOD FOR SMOKE GENERATED
1-3 Summary of Method
1-3.1 T h i s method for measuring the smoke generated by

materials employs an electrically heated radiant energy source
mounted within an insulated ceramic tube and positioned so as
to produce an irradiance level of 2.2 Btu/sec ft. 2 (2,5 W/ctn 2)
averaged over the central 1.5 in. (38.1 mm) diameter area of a
vertically m o u n t e d specimen facing the radiant heater. T h e nomi-
nal 3 in. by 3 in. (76.2 mm by 76.2 mm) specimen is mounted
within a holder which exposes an area measuring 29/16 in. by
29/16 in. (65.1 mm by 65.1 mm). T h e holder can accommodate
specimens up to 1 in. (25.4 ram) thick. This exposure provides
the nonflaming condition of the test.
1-3.2 For the flaming condition, a six-tube b u r n e r is used to

apply a row of equidistant premixed (air-propane) flamelets
across the lower edge of the exposed specimen area and into the
specimen holder trough. This application of flame, in addition to
the specified irradiance level from the heating element, constitutes
the flaming combustion exposure.
1-3.3 T h e test specimens are exposed to the flaming and non-

flaming conditions within a closed 18 ft. 3 (0.51 m 3) chamber. A
photometric system with a 36 in. (914 ram) vertical light path
measures the continuous decrease in light transmission as smoke
accumulates.
1-3.4 Calibration procedures for the test equipment, as

described in A-2, shall be followed.
1-3.5 T h e light transmittance measurements are used to ex-

press the smoke generated by the test materials in terms of the
specific optical density during the time period to reach the maxi-
mum value)
3Additional parameters, such as the m a x i m u m rate of smoke accumulation,

the time to a fixed optical density level, or a smoke obscuration index may be
more a p p r o p r i a t e in particular situations.
1379
TEST APPARATUS 258-7
Chapter 2 T e s t A p p a r a t u s
2-1 T h e apparatus shall be essentially as shown in Figures 2-1A

and 2-1BA T h e apparatus shall include that given in 2-2
through 2-11.
Figure 2-1A.
4A more detailed description of suggested details is given in A-1.

1380
258--6 T E S T M E T H O D FOR S M O K E G E N E R A T E D
~ , "~'~6e
J
A - - Phototube Enclosure I ~
Autotransfermers Q- Light Source Switch
B --Chamber J ~
Voltmeter (furnace) R - - Line Switch
C - - Blowout Panel K ~
Fuse Holders S - - S u p p o r t Frame
D - - H i n g e d Door with L ~
Furnace Heater T - - I n d i c a t i n g Lamps
Window Switch U - - Photometer Readout
E - - Exhaust Vent Control M ~ Gas & Air Flowmeters V - - Reds
F - - Radiometer Output N ~ Gas & Air Shutoff W - - Glass Window
Jack Valves X - - Exhaust Vent
G u Temperature (Wall) O ~ Light Intensity Y - - Inlet Vent
Indicator Controls Z - - Access Ports
H - - Temperature Indicator P ~ Light Voltage
Switch Measuring Jack
lrzgure 2 - l B . Smoke D e m i V y C h a m b e r Ammbl~,.
1381
TEST ^PPAaATUS 258--9
2-2 Test Chamber

2-2.1 As shown in Figure 2-1B, the test chamber shall be
fabricated from laminated panels 5 t o provide inside dimensions
of 36 in. by 24 in. by 36. in. 4- ~ in. (914 mm by 610 mm by 914
mm --. 3 mm) for width, depth, and height, respectively.
2-2-2 The interior surfaces shall consist of porcelain-enam-
eled metal or equivalent coated metal, resistant to chemical attack
and corrosion, and suitable for periodic cleaning.
2-2.3 Sealed openings shall be provided to accommodate a
vertical photometer, power and signal connectors, air and gas
supply tubes, exhaust blower, inlet and exhaust vents, pressure
and'- gas sampling taps, a pressure relief valve,'a rod for remote
positioning of the specimen holder, an aluminum foil [0.0010 in.
iapprox. 0.025 mm) or less] safety blowout panel at least 125 in. 2
(8{)B cm2) in area, and a hinged front-mounted door with an
observation port or window.
2-2.4 All openings shall be located on the floor of the
chamber.
Exception: The gas sampling taps, the positioning rod, and an inlet vent.
2-2~ When all openings are closed, the chamber shall be
capable, of developing, and maintainin, g positive pressure during
test periods, m accordance wzth 4-10.
2-3 Radiant Heat Furnace
2-3.1 An electric furnace with a 3 in. (76.2 mm) diameter
opening, as shown in Figure 2-3, shall be used to provide a constant
it~radiance on the specimen surface.
2-3.2 The furnace shall be located along the centerline
equidistant between the front and back of the chamber, with the
opening facing toward and about 12 in. (305 mm) from the right
wall.
2-3.3 The centerline of the furnace shall be about 7s~ in.
(195 mm) above the chamber floor.
2-S.4 Furnace Control System
2-3.4.1 The furnace control system shall maintain the re-
quired irradiance level under steady-state conditions with the
chamber door closed to within 4- .04 Btu/sec ft. 2 (4- .05 W / c m 2)
for 20 minutes.
5Commercially available panels of porcelain.enamel steel (interior surface)

permanently laminated to asbestos-cementboard and backed with galvanized
steel (exterior surface),with a total thickness 5/16 in., have been found suitable.
1382
Q l"
®
A m Stainless Steel Tube O ~ Stainless Steel K - - A s b e s t o s Board Rings
B m Asbestos BoQrd Spacing Washers (3) L - - A s b e s t a s Board Cover
C m Ceramic Tube H ~ Stainless Steel M - - Sheet Metal Screws
D -- Heating Element, Reflector W ~ Pyrex Glass Wool
525 W I ~ Stainless Steel
E ~ Stainless Steel Screw Reflector
F - - A s b e s t o s Paper J ~ Asbestos Board
Gasket
Figure 2-3. Furnace Section,
2-3.4.2 The control system shall consist of an autotrans-

former or alternate control device and a voltmeter or other means
for monitoring the electrical output. 6
2.4 Specimen Holder
2-4.1 Specimen holders shall conform in shape and dimen-
sion to that shown in Figure 2-4, and shall be fabricated to expose
a 2~6 in. by 29~0 in. (65.1 mm by 65.1 ram) specimen area. Also
shown in Figure 2-4 are the spring and rods for retaining the
specimen within the holders.
SWhere line voltage fluctuations are present, a constant-voltage transformer

may be required to maintain the prescribed irradiance level.
~o
Alignment of Mocltfl¢*6
Hol~r ' ond B~rner
A --
~ C0/~4
• ' .. - •
o ~s I J n II-M°dif~d Specimen
0 O ~ ° Stc6"~/eis
Steel
~E ruth Spot WL4dl~
Fastenings
~-~'~- ~- ~J RightSide Section
A-A(Not to Scole)
Isometric View
of Specimen Holder
~.~. ~'~
eeto~nlng
4"-'
feto~ng
IT [.,~i rod rod
il I'~,~-
o~
~E
o'oq IsometricView
Modil~ (trough)Specimen HolcJ~r
Specimen Holder
(flaming exposure)
F~,~re 2-¢ Oetails o f S p e d m m H o l d e r s a n d Pilot B u r n e r .
1384
258-12 TEST METHOD FOR S M O K E GENERATED
2-5 Framework for Support of the Furnace and Specimea

Holder. T h e furnace and specimen supporting framework shall
be constructed essentially in accordance with Figure 2-5.
I# . . . . . . . . . . . . . .
• ROOs
STRIPS
FRAME
Figure 2-5. F u r n a c e Support.
2.6 Photometric System

2-6.1 The photometric system shall consist of a light source
and photodetector, oriented vertically to reduce variations in
measurement brought about by stratification of the smoke gen-
erated by materials under test.
2-6.2 The system shall be as shown in Figures 2-6A and
2-6B and include the following:
1385
TEST APPARATUS 258-13
" ~ Photomultipliertube
, _.; ]1 , Filter(removable)
! ~ il "IShutter
Diffuser---~ ~ ~J~- Aperture
-/\-.
IIw,now,
3" dia.
U glass
I I
I ~r- Lens
,' l
I [~.,/~/y//~//~ g/- Topof chamber
~Side o: I' l ~' ~PSideof chamber

chamber I
{1 36"
iI I r~l--3',"0ia,
steelrods n
~''r i "~"~/77'j~ J~l ~'B°tt°m°' chamber

~____~ \4.~ h,ter II
, , 0 v t ~ ~;:v. U ' " II
~ L a m p
( ~1 ~---~< voltage
-~'~----"~ jacks
Opticallamp" I~
110V 4"1 0.3c.v.
trans. r ~J
m
2-6A. Photometer Details.

1386 "I
258-14 TESTMETHOD
FORSMOKEGENERATED
Rear of chamber
T
I 4"
,,
Jo
E¢D
--~) 8"
t-¢J
t4-
O
-_ _ ~ - i~_ I Owe
o~
-34" dia.
metal rods I
®
~----- 4¥' J-- ,,
PARTIAL PLAN VIEW

Figure 2-6B. Photmnetex I m a t i o n .
(a) T h e light source shall be an incandescent lamp operated
at a fixed voltage in a circuit powered by a voltage regulating
transformer. The light source shall be mounted in a sealed and
light-tight box located below the chamber. This box shall contain
the necessary optics to provide a collimated light beam passing
vertically through the chamber.
(b) T h e photodetector shall be a photomultiplier tube, with
an S-4 spectral sensitivity response and a dark current less than
10-gA. A sealed box located directly opposite the light source shall
be provided to house the photodetector and the focusing optics.
A glass window shall be used to isolate the photodetector and its
optics from the interior of the chamber.
2-7 Radiometer
2-7.1 T h e radiometer for standardizing the output of the
radiant heat furnace shall be of the circular foil type, the opera-
tion of which was described by Gardon. 7
7R. Gardott, "An Instrument for the Direct Measm~ment of Intense Thermal
Radiation," Review o[ Scienti/ic Instruments, Vol. 24, pp. ~6-370 (1953).
1387
TEST APPARATUS 258--15
2-7.2 T h e construction of the radiometer shall be as shown

in Figure 2-7.
2-7.3 T h e radiometer shall have a stainless steel reflective
heat shield with a 1 ~ in. (38.1 mm) aperture on the front and a
finned cooler supplied with compressed air mounted on the rear
to maintain a constant body temperature of 200 _+5°F (93 _ 3°C).
~ 8
~,-,, "~ I' I I-,T-.,..,L

•
moN --I~
jw
1388
258-16 T E S T M E T H O D FOR S M O K E G E N E R A T E D
2-8 Thermocouples for Determining Chamber Wail Tern.

perature. A thermocouple shall be provided for determining the
chamber wall temperature prior to testing.
2-9 Portable Recorder or Read-Out Meter

2-9.1 The outputs of the radiometer and the thermocouples
shall be monitored by a suitable recorder or read-out meter.
2-9.2 The photodetector output shall be recorded or moni-
tored with a potentiometer or other suitable instrument capable
of measurement over a range of 5 decades, or more. (see A-I.4.)
2.10 Manometer for Chamber Pressure Measurements

2-10.1 A simple water manometer with a range up to 6 in.
(152 mm) of water shall be provided to monitor chamber pressure
and leakage. (See A-2.3.)
2-10.2 The pressure measurement point shall be through a
gas sampling hole at the top of the chamber.
2-10.3 A simple water column or relief valve shall be pro-
vided to permit control of chamber pressure. (See A-I.8.)
2-11 Multiple Flamelet Burner with Premixed Air-Propane

Fuel
2-11.1 For a flaming exposure test, a six-tube burner, with
construction details as shown in Figure 2-4, shall be used.
2-11.2 T h e burner shall he centered in front and parallel
to the specimen holder.
2.11.3 The tips. of the two horizontal tubes shall be centered
- ~ e in. (6.4 -4- 1.6 ram) above the holder edge and ~ _ ~6
in. (6.4 _ 1.6 mm) away from the specimen surface.
2-11.4 Provision shall be made to rotate or move the burner
out of position during nonflaming exposures.
2-113 A premixed air and propane (95% purity or better)
test gas shall be used.
2-11.6 T h e air and propane shall be metered by calibrated
flowmeters and needle valves at 500 cm3/min, for air and 50 cmS/
rain. for the propane.
TEST S P E C I M E N s
1389
258-17
Chapter 3 Test Specimens
~-1 Specimen Description

3-1.1 Size
3-1.1.1 The test specimens shall be 3 in. by 3 in. __..03 in.
(76.2 mm by 76.2 mm _+ 0.7 mm) by the intended installation
thickness up to and including 1 in. (25.4 mm) thickness.
3-1.1.2 Specimens provided in thicknesses in excess of 1 in.
(25.4 mm), shall be sliced to 1 in. (25.4 mm) thickness and the
original (uncut) surface tested.
3-1.1.3 Multilayer materials greater than 1 in. (25.4 mm)
thick, consisting of a core material with surface facings of different
materials, shall be sliced to 1 in. (25.4 mm) thickness, and each
original (uncut) surface shall be tested separately if required un-
der 3-1.3.
3-1.2 Specimen Orientation
3-1.2.1 If visual inspection of the specimen indicates a
pronounced grain pattern, process-induced surface orientation,
or other nonisotropic property, the specimen shall be tested in two
or more orientations.
3-1.2.2 The highest smoke density value and the test
orientation shall be stated.
3-1.3 Specimen Assembly
3-1.3.1 The specimen shah be representative of the mate-
rial or composite and shall be prepared in accordance with rec-
ommended application procedures,s
Exception: Flat sections o[ the same thickness and composition may
be supplied and tested in place of curved, molded, or specialty parts.
$-1.3.2 Where an adhesive is intended for field applica-
tion of a finish material to a substrate, the prescribed type of ad-
hesive and its spreading rate shall be noted and used for test.
8Substrate or core materials for the test specimens should be the same as
those for the intended application. Where a material or assembly may be
exposed to a potential fire on either side, both sides should be tested.
1390
3-1.3.3 When supplementary tests are necessitated by de.

lamination, cracking, peeling, or other separations affecting smoke
generation, the manner of performing such supplementary tests,
and the test results, shall be included in the report with the con.
ventional test. 9
3-1.3A For comparative tests of finish materials without
a normal substrate or core and for screening purposes only, the
following procedures shall be employed:
. ( a ) Rigid o r semirigid sheet materials shall be tested by
the standard procedure regardless of thickness.
( b ) Liquid films (paints, adhesives, etc.), intended for ap-
plication to combustible base materials, shall be applied to the
smooth face of ~ in. (6.4 ram) thick tempered hardboard, nominal
density 50 lb./ft, s to 60 lb./ft. 3 (0.8 g/cm3 to 0.97 g/cm3), using
recommended (or practical) application techniques and coverage
rates. Tests shall also be conducted on the hardboard substrate
alone, and' these values shall be recorded as supplemental to the
measured values for the composite specimen.
(c) Liquid films (paints, adhesives, etc.), intended for ap-
plication to noncombustible substrate materials, shall be applied
to the smooth face of ~ in. (6.4 ram) thick asbestos-cement board,
nominally 120 lb./ft, s (1.9 g/cmS) in density, using recommended
(or practical) application techniques and coverage rates.
3-2 N u m b e r of Test Specimens. Three tests under flaming
exposure and three tests under nonflaming exposure shall be con-
ducted on each material (total of six specimens) in accordance
with the conditions described herein.
3-3 Specimen Conditioning. Specimens shall be predried for
24 hrs at 140 + 5oF (60 4- 3oC), then conditioned to equilibrium
(constant weight) with an ambient temperature of 73 4- 5°F (23
4- 3°C) and a relative humidity of 50 -4-_5 percent.
3-4 SpecimenMounting
3-4.1 All specimens shall be covered across the back, along
the edges, and over the front surface periphery with a single sheet
of aluminum foil (0.0015 4- 0.0005 in. or approximately 0.04 nun).
9Finish materials, including sheet laminates, tiles, fabrics, and others secured
to a substrate material with adhesive, and compositematerials not attached to
a substrate, may be subject to delamination, cracking, peeling, or other separa-
tions affecting their smokegeneration. To evaluate these effects,supplementary
tests, performed on a scored (slit) exposed surface or on interior layers or
surfaces, may be necessary.
• TEST SPECIMENS
1391
3-4.2 Care shall be taken not to puncture the foil or to in-
troduce unnecessary wrinkles during the wrapping operation.
3.4.$ Foil shall be folded in a way to minimize losses of
melted material at the bottom of the holder.
3.4A Excess foil along the front edges shall be trimmed off
after mounting.
~ 4 ~ All specimens shall be backed with a sheet of asbestos
millboard. (See Section 2-4.)
S-4.6 The specimen and its backing ,s,,hall be secured with
the spring and retaining rod. A modified C shape retainingrod
shall be used with specimens from ~ to 1 in. (1.6 to 2.5 cm) thick.
~4.7 Flexible specimens shall not be compressed below
their normal thickness.
&,4~ It is the intent of this test method to maintain the ]~re-
scribed exposure conditions on the specimen for the test durauon.
If, during a nonflaming exposure test, the specimen tends to melt
or drip and fall away from the specimen holder, it shall be tested
using the modified specimen holder (with trough) designed for
the flaming test. 10
101f under either the flaming or nonflaming exposure, there is an excess of

melted material which overflows the trough, the specimen area should be
reduced, e.g. if the area is reduced to 1~'2 in. wide by 3 in. high ($8.1 mm by
76.2 ram) centrally located, the appropriate area should be used in calculating
Ds.(See S~ction 5-I.) In using tl~e modified holder with the trough, a flap of
f0il shotlh~be cut and bent forward at the spout to permit flow from melting
1392
Chapter 4 T e s t Procedure
4-1 All tests shall be conducted in a room or enclosed space

having an ambient temperature of 73 _ 5°F (23 ± 3°C) and rela.
tive humidity of 50 +_ 20 percent at the time of test.
4-2 E q u i p m e n t Cleaningtt
4.2.1 T h e chamber wall s shall be cleaned whenever periodic

visual inspection indicates the need.t2
4.2.2 T h e exposed surfaces of the glass windows separating
the photodetector and light source housing from m e inter!or of
the chamber shall be cleaned before each test (ethyl alcohol is
generally effective).
4-3 W a r m . u p of FurnacelS
4-B.I During the warm-up period all electric systems (fur.
nace, light source, photometer readout, etc.) shall be on, th~ ex-
haust vent and chamber door closed, and the inlet veht open.
4.3.2 W h e n the temperature on the center surface of the
back wall reaches a steady-state value in the range of 95 +_ 40F
(35 _ 2°C), the chamber is ready for furnace calibration or testing.
4.$.~ T h e furnace o u t p u t irradiance shall be calibrated at
periodic intervals according to test experience (normally twice per
test day).
4.3A A "blank" specimen holder, with the asbestos mill-
board exposed, shall always be directly in front of the furnace.
Exception: When displaced to the side by (1) the specimen holder
during a test or (2) the radiometer during calibration.
llCharred residues on the specimen holder and horizontal rods should

removed to avoid contamination.
12An ammoniated spray detergent and soft scouring pads have been found
effective.
ISTo increase chamber wall surface temperature to the stated level under
adverse conditions, an auxiliary heater may be used. Conversely, to dececececece~.~
this temperature, the exhaust blower may be used to introduce cooler air from
the laboratory.
1393
TEST PROGEDURE 258--21
The specimen holder shall be returned immediately to the

above position when testing or calibration is completed•
4.$.5 During calibration, the radiometer shall be placed on
the horizontal rods of the furnace support framework and accu-
rately positioned in front of the furnace opening by sliding and
. . . the
displacing . blank . specimen
. . holder
. against
. . preposmoned
the
stop. The furnace support framework, stop, and "blank" specimen
holder shall provide for the horizontal and vertical centering
within - 1A6 in. (1.6 ram) with respect to the furnace opening of
the radiometer during calibration and of the loaded specimen
holder during test.
4-S.6 With the chamber door closed and inlet vent opened,
the compressed air supply to the radiometer cooler shall be ad-
justed to maintain its body temperature at 200 ± 5OF (93 ± 3°C).
4-$.7 The autotransformer setting shall be adjusted so as to
obtain the calibrated millivolt output of the radiometer corre-
spo~ndin to a steady-state irradiance of 2 2 ÷ 04 Btu/sec ft 2
g • m • • ~ • •
(~.5 ± .05 W/cm2) averaged over the central 1.5 m. ($8.1 ram)
diameter area.
4-$.8 The recorder or meter described in 2-9 shall be used
to monitor the radiometer output. After the prescribed irradiance
level has reached steady-state, the radiometer shall be removed
from the chamber and replaced with the "blank" specimen holder.
4-$.9 After the system has reached steady-state conditions,
the meter and/or recorder zero shall be adjusted•
4-$.10 The amplifier sensitivity shall be ad'ustedj to obtain
a full-scale reading of the photodetector (100 percent transmit-
tance) on the recorder or readout meter.
4-$.11 The "dark current" (zero percent transmittance) on
the maximum sensitivity range of the read,~ut meter shall b e d e -
termined by blocking the light, and the dark current reading
shall be adjusted to zero.
6.4 Burner Positioning. For nonflaming exposures, the mul-
tlple flamelet burner shall be removed. For flaming exposures, the
burner shall be positioned across the lower edge of the specimen
as described in 2-11. (Check the burner distances relative to the
"blank" specimen before fuel adjustment and ignition.)
4.5 Before positioning the test specimen, the chamber shall
be flushed for about 2 minutes with the door and exhaust and
inlet vents open and the starting temperature of the chamber
shall be verified using the procedure described in 4-$. 1 and 4-~.2.
1394
4-6 Then the exhaust vent and blower shall be closed.
4-7 The loaded specimen holder shall be placed on the bar

support and pushed into position in front of the furnace (with
burner in position for flamm"g exposure) by displacing the "blank"
holder.
4,8 T h e chamber door shall be quickly closed and the timer

and/or recorder chart drive shall be simultaneously started. The
inlet vent shall be closed completely only when the photometer
indicates smoke.
4-9 Light transmittance and the corresponding time shall be

recorded, either as a continuous plot with a multirange recorder
or a t sufficient time intervals with a multirange meter readout.
(Make and note the necessary full-scale range changes in decade
steps.)
4.10 T h e increase in chamber pressure shall be observed with

the manometer described in 2-10. A regulator (see A-I.8) shall be
tised to maintain the pressure in the range of 4 _+ 2 in. (100 _ 50
nun) of water during most of the test. If negative pressure develops
after very intense specimen flaming, the inlet vent shall be open6d
slightly to equalize the pressure. As a result of pressure rise, the
fuel and air valves shall be adjusted during the flaming test to
maintain constant flow rate.
4-11 Any observations pertinent to the burning and smoke

generating properties of the material under test shall be recorded
in accordance with 6-6 and 6-7.
4-12 The test shall continue for 20 minutes or until a mini-

mum light transmittance value is reached, whichever occurs first.
If .the minimum light transmittance does not occur within the 20
mmute exposure period, this shall be noted in reporting the
results.
4.,IS If transmittance falls below 0.01~7o, the chamber window

shall be covered with an opaque screen to avoid possible light
scattering effects from room ligfit. Also, any supplementary optical
filter in the photometer system shall be removed or displaced
in order to extend the measuring range. (If extraneous light can
reflect into the photometer during removal of the filter, turn the
high voltage off or adjust the scale to minimize sensitivity. Replace
the filter before exhausting smoke from the chamber.)
!395
TEST PROCEDURE 258-23
4-14 The burner on flaming exposures shall be extinguished

and exhausting the chamber shall be initiated within one minute
after reaching minimum transmittance.14 The specimen shall be
displaced from the front of the furnace by pushing, the "blank"
specimen holder with the positioning rod. Exhausting shall con-
tinue with the inlet vent open until maximum transmittance is
reached. (Record this transmittance value as the To, "clear beam"
reading, which is to be used to correct for deposits on the photom-
eter windows.)
14In some cases the transmittance m a y increase somewhat and subsequently

decrease to the ultimate m i n i m u m transmittance.
1396
C h a p t e r 5 C a l c u l a t i o n s 15
5-1 C a l c u l a t e specific o p t i c a l d e n s i t y , De, f r o m t h e r e d u c t i o n

in p e r c e n t l i g h t t r a n s m i t t a n c e , T , c a u s e d b y t h e s m o k e g e n e r a t e d
f r o m a n e x p o s e d s p e c i m e n area, A, i n t h e closed c h a m b e r of vol.
u m e , V, a n d o v e r a l i g h t p a t h , L, as follows:
w h e r e 0 r e p r e s e n t s t h e g e o m e t r i c a l f a c t o r associated w i t h the
d i m e n s i o n s of t h e c h a m b e r a n d s p e c i m e n . C o r r e c t i o n s for t h e vol-
u m e of t h e f u r n a c e assembly a n d t h e v o l u m e i n c l u d e d in t h e d o o r
recess a r e g e n e r a l l y less t h a n 1 p e r c e n t a n d may. be n e g l e c t e d .
W h e n i t is necessary to r e m o v e t h e n e u t r a l d e n s i t y filter (see
,t-1.4) to m e a s u r e low levels o[ l i g h t t r a n s m i t t a n c e , t h e specific
15Precision. The extent of agreement between replicate tests by the same

operator and between results from different laboratories is a function of the
properties of the sample, principally its uniformity and its physical reaction to
the test exposure. Experience has shown that for uniform materials which do
not exhibit erratic test behavior, the coefficient of variation is generally less
than 10. percent. For selected (standard reference) materials, the coefficient ot
variation is less than 5 percent. For some materials, induing les~ uniform
materials and those which exhibit erratic test behavior (e.g., cracking, separa-
tion, melting), the single-operator-coefficient of variation may range up to 25
pereent,~ and the multilaboratory coefficient of variation may range up to 80
percent. Causes of excessive variations should be investigated.
The following coefficients of variation were measured in a "round-robin"
testing program on materials exhibiting a variety of test behavior and uni-
formity over the complete range of the measuring apparatus. The round-robin
testing was performed by a total of 22 laboratories on 13 materials in dupli-
cate as follows:
22 laboratories 7 materials nonflaming exposure
22 laboratories 1 material flaming exposure
14 laboratories 5 materials flaming exposure
Exposure Number of Number of Range in Multilaboratory

Laboratories Coefficient of Variation (%) *
Materials in Analysis Low Median High
Flaming 6 12 3.8 I0. 27.
Nontlaming 7 18 2.9 7.2 27,
*Computed as V u ~ + vZa,where
n
~m~is the betweendaboratory coefficient of variation
p, is the within-laboratory coei~cient of variation
n is the number of replicate tests (n ffi=2)
CALC U L A T I O N S
1397
258--25
optical density appropriate for the filter shall be added. The
value t o be added is equal to the known optical density of the
filter (see A-2.1.3) multiplied by G.
5,2 Calculate the maximum specific optical density, Din, using
the formula in 5-1 with a light transmittance corresponding to the
minimum level reached during the test. Correct all maximum
specific optical density values by subtracting the specific optical
density equivalent for soot and other,deposits on the photometer
windows. As described in 4-14, the clear beam transmittance
reading Te is used to calculate a specific optical density equivalent
Dv using the same formula but with different subscript. A cor-
rected maximum specific optical density calculation is expressed
as follows:
D= (corr.) ----D m - De
5-~ For systems without "dark current" cancellation, a correc-
tion shall be made for any percent light transmittance reading T,
approaching the dark current value Ta. The corrected percent
light transmittance T t, shall be obtained from:
T 100[1100100 T] I00[T
and shall be used for the specific optical density calculations de-
scribed in 5-1 and 5-2.
5.4 Determine t.s~m, the time for the smoke to accumulate to
90 percent of the uncorrected maximum specific optical density
value from a plot of specific optical density versus time or from
the tabulated data.
1398
Chapter 6 Report16
6.1 T h e report (see Appendix B) shall include the following:
(a) Complete description of the specimen tested including:

Type, manufacturer, shape, thickness a n d / o r other appropriate
dimensions; weight or density, coloring, etc.
(b) Complete description of the test specimens, including: Sub.
strate or core, special preparation, mounting, etc.
(c) Test specimen conditioning procedure.
(d) N u m b e r of specimens tested.
(e) Test conditions: Type of exposures, type of holder used,
exposure period.
(f) Observations of the burning or smoldering characteristics
of the specimens during test exposure, such as delamination, sag.
ging, shrinkage, melting, collapse, etc.
(g) Observations of the smoke generating properties of the
specimens during exposure, such as, color of the smoke, nature 0[
the settled particulate matter, etc.
(h) A record of the geometrical factor, (3, as calculated from
measured values of chamber volume, V, photometer light path
length, L, and exposed specimen area, A. (See Chapter 5 on Cal.
culations.)
(i) Test results calculated as described in Chapter 5, including
the average and range on each set of specimens for Dm (corr.), t.gDm
and D e.
6.2 If the test is terminated on the basis of a 20-minute ex-

posure limitation, this fact shall be noted when reporting mea-
surements observed at that time.
16Analysis of products of combustion. Although not specifically required

as part of the method, products of combustion may be drawn from the chamber
at various times during the progress of the test for analysis. The physical
properties of the smoke may be investigated by electrostatic or impact collec-
tion and various methods of particle analysis. The presence and concentrations
of various toxic and irritating gaseous products may be determined using
colorimetric gas detector tubes, gas chromatography methods, ion-selective
electrodes, or other techniques.
1399
APPENDIX A 258-27
Appendix A
This Appendix is not part of this NFPA Standard, but is included for in-
formation purposes only.
Apparatus Construction and Cah'bration
A-I Construction Details
A-I.1 Radiant Heat Furnace (see 2-3). The furnace shall

consist of a coiled wire or other suitable electrical heating element
525 W or greater) mounted vertically in a horizontal ceramic tube
in. (76.2 ram) i.d. by 38~ in. (85.7 ram) o.d. by 1 ~ in. (41.3 mm)
long. T h e t u b e is bored out at one end to 3zAa in. (77.0 ram) i.d.
and to a depth of ~ in. (15.9 mm) to accommodate the heating
element. A zAe in. (1.6 ram) asbestos paper gasket, three stainless
steel reflectors are mounted behind the heating element. A s~ in.
(9.5 ram) asbestos millboard disc, provided with ventilation and
lead wire holes, is positioned behind the heating element and
used to center the assembly with respect to front s~ in. (9.5 ram)
asbestos millboard ring by means of a 6-32 stainless steel screw.
The adjustment nuts on the end of the centering screw provide
proper spacing of the furnace components. The cavities adjacent
to the heating element assembly are to be packed with glass wool.
The furnace assembly is housed in a 4 in. (102 ram) o.d. by 0.083
in. (2.1 ram) wall by 4 ~ in. (10.5 cm) long stainless steel tube. Two
additional s~ in. (9.5 ram) asbestos board spacing rings and a rear
cover of s~ in. (9.5 mm) asbestos board complete the furnace. T h e
furnace is to be located centrally along the long axis of the cham-
ber with the opening facing toward and about 12 in. (305 ram)
from the right wall. The centerline of the furnace is about 7sA in.
(195 ram) above the chamber floor.
A-I.2 Specimen Holder (see 2-4). The specimen holder is
to conform in shape and dimension to Figure 2-4 and to be fab-
ricated by bending and brazing (or spot welding) 0.025 in. (0.6
ram) thick stainless steel to provide a 1]~ in. (38.1 ram) depth, and
to expose a 2~e in. by 2~e in. (65.1 mm by 65.1 man) specimen
area. As described in 2-5 the holder is to have top and bottom
guides to permit accurate centering of the exposed specimen area
in relation to the furnace opening. A 3 in. by 3 in. (76.2 mm by
76.2 mm) sheet of ~,~ in. (12.7 ram) asbestos millboard, having a
1400
nominal density of 50 +_ 10 lb/ft. 3 (0.85 _ 0.17 g/cm3), is to be

used to back the specimen. A spring bent from 0.010 in. (approxi.
mately 0.25 mm) thick phosphor bronze sheet is to be used with
a steel retaining rod to securely hold the specimen and millboard
backing in position during testing.
A-1.3 Support of Furnace and Specimen Holder (see 2-5)
The framework as shown in Figure 2-5 has welded to it a 5 in~
(12.7 cm) o.d., ¼ in. (6.4 ram) wall, 2 in. (50.8 ram) long horizontally
oriented steel tube to support the radiant heat furnace described
in 2-3. This support tube is to have provision to accurately align
the furnace opening so that it is: (1) 11~ in. _ ~6 in. (38.1 mm
1.6 ram) away from, (2) parallel to and (3) centered horizontally
and vertically to within -4" 1/16 in. (1.6 ram) with respect to the
exposed specimen area. Three tapped holes with screws equidis-
tantly positioned around the furnace support tube, or one screw
at the top of the support in conjunction with two adjustable (ver-
tically alon.g the support tube) metal guide, strips mounted, hori-
zontally reside to the tube, are to provlde adequate ahgnment.
The framework is to have two ~ in. (9.5 ram) diameter trans-
verse rods of stainless steel to accept the guides of the specimen
holder described in A-1.2. The rods are to support the holder so
that the exposed specimen area is parallel to the furnace opening.
Spacin.g stops are to be mounted at. both. ends of each. rod to ermit
qmck and accurate lateral posiuonmg of the spectmen ho~er.
A.1.4 Photometric System (see 2-6). The photometric sys-
tem is to consist of a tungsten-filament light source, (Type 1630
6.5 volt lamp, maintained at 4 4- 0.2V)and a photodetector with an
S-4 spectral sensitivity response. The photometer is to be oriented
vertically to reduce variations in measurement brought about by
stratification of the smoke generated by the specimens under test.
The system is shown in Figures 2-6.A a n d 2-6.B. The window in
the chamber floor through which the light beam passes is provided
with an electric heater to maintain a temperature of at least 125°F
(52°C) to minimize smoke condensation. The collimated beam
inside the chamber is to have a path length of 36 +_ ~ in. (914 _+
3 ram). T h e approximately circular light "spot" is centered en-
tirely. within the sensing area. of the detector. A typical photomul-
.
upher photometer system wtU require a high-voltage D.C. power

supply and a neutral density filter of sufficient optical density to
produce a convenient signal level for the indicator or recorder.
The photometer system used is to be capable of permitting the
recording of reliable optical densities of at least 6.0, corresponding
to transmittance values of 0.0001 percent of the incident light.
(See A-2.13.)
1401
APPENDIX A 2~8--~Q
The two optical platforms and their housings shall be kept in

alignment with three metal rods, ~ in. (12.7 mm in diameter fas-
tened securely into ~6 in. (7.9 mm) thick extera)ally mounteci top
and bottom plates and symmetrically arranged about the colh-
mated light beam.
A.I.5 Radiometer (see 2-7). The body temperature of the
radiometer shall be monitored with a 100-220°F (38-100°C) ther-
mometer in a ~ in. by ~ in. by 1 ~ in. long (12.7 mm by 12.7 mm
by 38.1 mm) brass well drilled to accept the thermometer with a
dose fit. Silicone grease may be used to provide good thermal
contact.
The circular receiving surface of the radiometer shall be spray-
coated with an infrared-absorbing black paint containing a sili-
cone vehicle. The radiometer is to be calibrated calorimetrically
in accordance with the procedure summarized in A-2.2.
A.I.6 Chamber Wall Thermocouple (see 2-8). A thermo-
couple is mounted with its junction secured to the geometric cen-
ter of the inner rear wall panel of the chamber using a ~ in. (6.4
ram) thick polystyrene foam disk cover and epoxy cement.
A.I.7 Burner (see 2-1.11). The multiple flamelet burner
is a six-tube burner, with construction details as shown in Figure
2-4. The vertical tubes of the six-tube burner are made from ¼
in. (3.2 ram) o.d. by 0.031 in. (0.8 mm) thick-wall stainless steel
tubing. All tubes should be crimped at the tip to reduce the open-
ing diameter to 0.055 in. (1.4 mm). The horizontal manifold sec-
tion of the burner consists of ~ in. (6.4 ram) o.d. by 0.035 in. (0.9
ram) wall stainless steel tubing. The other end is attached to a
fitting in the chamber floor.
A.I.8 Chamber Pressure Regulator. A simple pressure reg-
ulator consists of an open, water-filled bottle and a length of flex-
ible tubing, one end of which is connected to a sampling port on
the top of the chamber. The other end of the tubing is inserted
4 in. (10 ram) below the water surface. The bottle is located at the
same level as the floor of the chamber.
A.2 Calibration of Test Equipment
A.2.1 Photometric System
A-2.1.1 When first assembled and as necessary following
use or when suspicious of a malfunction, calibration of the pho-
tometer should be checked by interrupting the light beam with
calibrated neutral density filters. The filters should cover the full
range of the instrument. Optical density values measured by the
photometer are to be within _ 3 percent of the calibrated values.
1402
A.2.1.2 Shifts in dark current levels between tests, exces-

sive zero shifts during test, or lack of calibration indicate the need
for inspection of the photometer system.
A.2.1.3 The optical density of a supplementary filter used

to extend the measuring range of the photometer is to be known
to an accuracy of _+ 3 percent.
A-2.2 Radiometer. Calibration of the radiometer is ac-

complished by placing it at suitable distances from a radiant en-
ergy source, while maintaining its body temperature at 200 _ 5oF
(93 _+ 3oc) with controlled air flow through the rear-mounted
cooler, and measuring its electrical output as a function of the
irradiance level. The irradiance level is determined calorimetric.
ally by measuring the rate of temperature rise of a blackened thin
coppr disk of known weight, area ( 1 ~ in., 38.1 mm dia.), specific
heat, and absorptivity in place of the radiometer. The measured
millivolt output of the radiometer, at a body temperature of
200°F (93°C) , corresponding to an irradiance level of 2.2 __ .04
Btu/sec. ft. 2 (2.5 _ .05 W/cm2) is used to establish the furnace
control settings discussed in 2-3, 4-3.1, and 4-5.2.
A-2.3 Chamber Pressure Manometer -- Leak Rate Test.

For purposes of standardization, a leakage rate test should be
periodically conducted using the manometer and tubing described
in 2-10. T h e chamber is pressurized to 3 in. (approximately 76
ram) of water by introducing compressed air through a gas sam-
pling hole in the top. The decrease in pressure from 3 in. to 2 in.
(approximately 76 to 50 mm) of water is timed with a stop watch.
This time should not be less than 5.0 minutes.
A.2.4 Standard Smoke Generating Materials. For check-

ing operational and procedural details of the equipment and
method described herein, two standard materials may be used.
Under nonflaming conditions, a single layer of alphacellulose
(cotton linters) paper and under flaming conditions, plastic sheet
should provide repeatable maximum sp&iflc optical d~nsity values
in two portions of the measuring rangeA7 Use of these standard
materials does not obviate the need for following the calibration
and standardization procedure outlined in this standard.
17These reference samples, designated SRM 1006 and SRM 1007, may be
purchased from the Office of Standard Reference Materials, National Bureau
of Standards, Washington, D.C. 20234-
1403
APPENDIX B 258-31
Appendix B
S u g g e s t e d R e p o r t F o r m -- S m o k e D e n s i t y C h a m b e r
Sample Code Test No. Date

Lab Code Operator Time
" " Recorded Data or Curve Operating Conditions

~Time, min. % Trans. D, Radiometer Reading mV;
Irradiance W/cm 2
Furnace Voltage " V
Burner Fuel cc/min air;
.cc/min propane
Thermal Exposure: flaming smoldering
Chamber Pressure inch H20
Chamber Wall Temp. • °C
Chamber Surface Condition
Burner: [] Standard [ ] Special

Sample
Description -
Manufacturer -
Preconditioning: Temp °C;
Duration, jr.
Conditioning: Temp.. °C;
RH %; Duration,
Thickness - ._.__....~n , Density._____-__g/cm3
or lb/ftS
Initial W t . ~ ; Final W t ~ ;
% Lo,~q
Special Conditions -
Results
Min. Trans % at min
Max. Specific Optical Density, D m f f i f . - -
Time to Reach 90% Dm ffi= .min.
Clear Beam Reading ffi= %;
Equiv. D~
D , (corr.) ffi D . - Dc =
Remarks
1404 TEST METHOD FOR SMOKE GENERATED
258-32
Appendix C
Commentary on Measurement of Smoke Production from

Building Furnishing and Construction Materials
C.I General
C.I.1 Foreword
C,I.I.I The many ways in which fire gases influence the
hazard to life have, so far, precluded an exact technical assessment
of them. It can be shown however that there are at least five direct
ways in which a fire fatality can be caused.
(a) Oxygen deficiency
(b) Toxic chemical products
(c) Solid or liquid aerosols
Hot gases
(e) Flames
The smoke hazard considered in this pamphlet is primarily asso-
ciated with solid or liquid aerosols and only b y inference with
chemical products since some of these are readily absorbed and
carried by the dispersed particulate. Is The test.method recom.
mended is directed towards measurement ot the ltgrtt attenuation
characteristics of smoke. Actual vision will never be better than
that indicated by photometric means; usually it will be much
poorer as a result of light scattering and irritating effects on the
eyes caused by decomposition products. While it is evident that
vision impairment is an important contributing [actor it is not
clear that it alone can be considered as a direct cause of fire fatali-
ties. These facts should be recognized in any application of the
recommended procedure.
C.1.1.2 This commentary is intended to assist the user in

wise application of the results of the smoke density chamber Test
Methc~d. The references, including the bibliography in Appendix
D, are provided tO further this goal and, in addition, to provide a
means through which the reader can pursue his studies in the
general field of smoke - its measurement, hazards, and manage-
ment during fires.
C-1.2 Purpose. To provide guidance for those concerned
with use of a smoke test method as ~t basis of reducing the smoke
hazard during fires.
lSCarhart, H., Hazlett, 11. N., Johnson, E., Stone, J. P., "The Transport of
Hydrogen Chloride by Soot from Burning Polyvinylchloride,"Fire Flamma-
bility, 4, 42-51 (Jan 1973).
1405
APPENDIX C 258-33
C-1.$ Scope
C.1.3.1 This commentary provides a brief discussion on
the various components of the fire gas hazard to life. It shows that,
while photometric measurement of light attenuation by smoke
ar~pears to directly measure only a limited fraction of the fire gas
h~zard, it may be a significant indicator of this hazard.
C,1.3.2 A complete description of the smoke density
chamber is included to form the basis for specific reference to a
smoke test method.
C.1.3.3 A brief review was made of various methods used
earlier for smoke measurements and some of their limitations.
Methods are proposed for application of the test method de-
scribed, and emphasis is placed on the importance of recognizing
limitations in their effectiveness in preventing the development
of hazardous smoke conditions.
C-1A Definitions
C,1.4.1 Fire gases. Fire gases include the total gaseous,
particulate or other aerosol effluent from a fire or pyrolysis re-
action.
C-1.4.2 Smoke. Smoke is an aerosol comprising dis-
persed solid and liquid particulate in air or gas. For the purpose
o-fthis paper, gaseous products of combustion are excluded. - -
C-1.4.3 Optical Density. Optical density is the measure
of the degree of opacity; the negative common logarithm of the
transmission.
C-1.4.4 Specific Optical Density. Specific optical density
is a measure of smoke production from a material or product in
terms of optical density normalized with respect to the product of
specimen area and photometric path length divided by the volume
of the smoke collection chamber.
C.1A.5 Attenuation. Attenuation is reduction in light
transmission.
C.1.5 Smoke Hazards
C.1.5.1 There is increasing recognition of the importance
of combustion products in playinga major role in the majority of
fire fatalitiesA9 However, the chain of events during an attempt
19Fristrom, R., Annual Summary Report, July 1972 to 30 June 1973, App.
Phys. Lab. Johns Hopkins Univ. Rept. No. FPP A73 (Aug. 1973.)
1406
to escape a fire which ultimately leads to a fire gas fatality is far

from clear. It probably varies greatly depending on many factors
associated with both the fire and the people involved. Recent Work
at Johns Hopkins University shows that high carboxyhemoglobia
from exposure to CO has been observed in the majority of autop.
sies, which have been performed in their studies of fire fatalities.
In these studies there was little direct evidence of the importance
o£ vision obscuration as a contributing cause of these fatalities.
C,1.5.2 A distinction should be recognized between light
attenuation and vision impairment, especially when the obse~rv.
er's eyes are exposed to the smoke. The former is a physical mea.
surement, while the latter involves physiological reactions as well
as visual acuity. It is a well-known ~act that the eye is most sensi.
tive to any alkaline or acid forming gas. The reflex reactions to
an irritated eye can be so severe that little freedom is provided
for normal vision. Thus, a light smoke with high chemical irritant
content may limit vision much more severely than a more dense
but chemically inert smoke.
C~1.5.$ Nevertheless, there is ample evidence that, even
in the absence of toxic or irritating components, dense smoke can
present real hazards. The ingested solid particulates can cause
coating and blockage of the various respiratory channels and pas-
sageways with resulting severe edema, a serious life hazard. The
limitations on vision, too, can be a major cause for alarm and
panic with resulting disorientation. An important aspect of the
smoke problem is the fact that smoke moves. This is especially
true in modern high-rise buildings. The smoke is not only pumped
by the ventilation system to compartments other than that of its
production, but it is also forced to flow in the vertical direction
between floors by stack action resulting from thermal differences
between the interior and exterior of the building. The sudden
appearance of smoke in an otherwise unaffected room has ap-
parently provided adequate cause for alarm and panic. Further,
the mere evidence of smoke provides circumstantial evidence of
associated toxic fire gases.
Col.5.4 In summary, the hazards associated with smoke
may be enumerated as follows:
(a) Dispersed particulate may pose a life hazard resulting
from massive ingestation in the respiratory tract.
(b) Obscuration of vision can result in disorientation and
inability to escape.
(c) It may cause a psychological threat to those affected
and lead to panic situations.
1407
APPENDIX C 258-35
(d) Smoke usually provides an indication of a fire and may

often b'e accompanied by toxic a n d / o r irritating gases.
C.1.5.5 While the measurement of smoke through light

attenuation m a y not provide a measure for all aspects of the fire
o-as hazard, ~t can provide a useful measure of some components
~fthe hazard associated with fire gases.
C-2 Smoke T e s t Methods
C.2.1 Principles of Measurement. It is beyond the scope

of this commentary to include a comprehensive discussion of
methods used for smoke measurement. However, a brief review
is provided in the background to show the progress which has been
made in its characterization.
C-2.2 Smoke Density C h a m b e r T e s t
C-2.2.1 T h i s test m e t h o d was developed at the National

Bureau of Standards and~was first described in an A S T M research
symposium in 1967. 2° Since that time, there have been numerous
vublications reporting on its application and studies of the corre-
lation of results of interlaboratory tests through its use. 21, 22, 2s, 24
C-2.2.2 T h e method is very similar to. the box. type test.

developed by R o h m and Haas. 25, 26 However, it provides certain
improvements in the capability for quantitative measurement of
the smoke produced. Advantages provided by use of this test
method include:
20Gross, D., Loftus, J. J., and Robertson, A. F., "Method for Measuring
Smokefrom Burning Materials, Fire Test Methods u Restraint and Smoke,"
ASTM STP No. 422 ( 1967).
21Gross, D., Loftus, J. J., Lee, T. G.. Gray, V. E., "Smoke and Gases Pro-
duced by Burning AircraftInterior Mat'erials," NB$ Bldg. $ci. Series, BSS 18,
U.S. Gov't. Printing Office, Washington, D.C. (Feb. 1969).
22Brenden, J. J., "Usefulness of a New Method for Smoke Yield from
Speciesand Panel Products," Forest Prod. ]., 21, 23-8 ( 1971).
23Lee, T. G., "Interlaboratory Evaluation of Smoke Density Chamber,"
bIBS Tech. Note 708 (Dec. 1971).
24Lee, T. G., "The Smoke Density Chamber Method for Evaluating the
Potential Smoke Generation of Materials," NBS Tech. Note 757 (Jan. 1973).
25Bartosic, A. J., Rarig, F. J., "Evaluation of the XP2 Smoke Density
Chamber," (Symposium on Fire Test Methods -- Restraint and Smoke, 1966).
ASTM Spec. Tech. Pub. 422 (1966).
26ASTM Committee D-20, "Standard Method for Measuring the Density
of Smoke from the Burning or Decomposition of Plastics," 1973 Ann. Book o[
ASTM Stds., Designation D2843-70, Vol. 27, 714-722 (1973).
1408
(a) T h e smoke collection box is completely sealed so all

smoke produced during a test is retained.
(b) Only one surface of a test specimen is exposed to fire
or radiant heating, thus providing a measure of the effectiven~
of surface treatment in assisting in control of smoke release.
(c) A vertical photometer is used as a means for avoiding
measurement errors resulting from smoke stratification.
(d) Provision is included for reporting the results of smoke
measurements in terms of specific optical density.
C-2.2.3 T w o burning conditions are simulated b.y the test:

One, radiant heating in the absence of ignition; anct the other,
an open flaming combustion of the specimen in the presence of
supporting radiation. These two conditions were selected as rep.
resentative of two extremes of fire involvement of the product.
Provision was also made to insure that thermoplastics which melt
and fall away from direct exposure to the pilot flame were col.
lected in a trough and forced to burn. This probably represents
the worst of fire exposure conditions likely to be experienced.
C-2.2.4 The concept of specific optical density, while old

in terms of chemical photometric practice, was first introduced for
measuring smoke as part of the smoke density chamber Test
Method. It is based on Bouguer s law and permits reporting smoke
developed in terms which recognize the area of specimen involved,
the volume of the box and the optical path length of the photom-
eter. Specific optical density is without dimensions, but its value
must be recognized as relating to the specimen in the thickness
tested. It has the unique advantage of providing a basis of predict-
ing the smoke density which can be developed by the same product
in other fire-involved areas and in other enclosure volumes, on
the assumption of uniform smoke-air mixing. Figure C,-2.2.4 re-
produces a chart from "Measuring Smoke from Burning Mate-
rials, Fire Test Methods -- Restraint and Smoke, ''27 which may
prove useful in such predictions. The crosshatched area on this
figure represents the approximate limiting range of visibility of a
strong light by eyes protected from contact with the smoke. Details
of the smoke test method are presented in the standard itself.
2?Gross, op. cir.

1409
APPENDIX C 258-37
,oo----T ...... r T. . . . . ,/'-,~...~..~.-~;7, / A ~

/" ..":.".'..\,...,,.",
i o_ 1,,,',.,.:~
,....
,:-.
vo -~~'~-!",",<':-,'. ~
I
I!
I I I I I I
2 5 10 20 50 100 200 500 10oo
Geometricalfactor . v
AL
NOTE: The crosshatched area represents the approximate limit of

visibility of a bright light.
Figure C-22.4. Specificoptical density versus geometrical factor
for selected light Wansmittance values
C-3 Application of Smoke Measurements

C-3.1 T h e smoke problem which develops during unwanted
fires has been recognized for m a n y years. Fire fighters are faced
with it daily in their work. However, three things have tended to
prevent application of standards limiting the acceptability of
materials or products on the basis of smoke production:
(a) Questions on the extent to which the smoke measure-
ment assesses the fire gas hazard. 28
(b) T h e lack of a well defined measurement m e t h o d which
could be shown to provide a technically valid means for smoke
characterization.
(c) Most materials or products, when burning, release large
quantities, of smoke, and there have been only limited ways of
reduction of smoke production.
C-~.2 T h e first problem still exists, although as indicated
in Section C-I, there seems to be a valid basis for consideration
28"The Control of Smoke in Building Fires - - A State-of-the-Art Review,"

Mtls. Res. and Stds. 11, 4, pp. 16-24(1971).
1410
258--38 TEST METHOD FOR SMOKE GENERATED
that smoke measurements do in fact serve to measure an important

component of the fire gas hazard. T h e second problem has been
eliminated to a large degree by development of the smoke density
chamber Test Method. Within the limitations imposed by use of
the method, or in fact any method of test, it appears to be Well
based technically.
The third problem is still an important consideration. Because
of it, as emphasized later, it seems necessary to require large
changes in D, if significant changes in smoke level are to be ex-
pected when fire occurs. The following sections have been devel.
oped to suggest appropriate means and precautions in application
o f smoke measurements.
C-3.3 Nature of the Measurement. It is important to re-
member that for a given thermal exposure condition the smoke
produced when a fire occurs is directly related to the quantity of
material involved." Moreover, the values of D, derived fromthe
test are normalized only with respect to the involved surface area
of the specimen. Thus, they are only applicable to other specimens
of the same thickness. The importance of specimen thickness is
well illustrated in Figure C.:-3.3.~ The deviation from a linear
SO01
oop
300 -
QE
200 -
I00~
ol 1
1/4
__.
1/2
I
3/4
Specimen thickness (in.)
Figure~S~ D m for spruce as a f u n c t i o n of specimen thickness

under nonflamingconditions.
28a"Method for Measuring Smoke f r o m B u r n i n g Materials, Fire Test

Methods - - Restraint a n d Smoke," op. cir.
1411
APPENDIX C 258.-39
relationship of Dm with specimen thickness results from the de-

creasing pyrolysis rate of the specimen as the burning layer pro-
gresses into the specimen and, also, from the increasing rate o f
smoke drop out and condensation as high smoke concentration
develops: Possible smoke development during fire involvement
of a product, of course, is directly related to the surface area af-
fectedby the fire.
C-3.4 Fi~,ure C-3.4~ shows the approximate relationship be-
tween the opucal density per unit path length and the limits of
visual range under two illumination condiuons. The eyes of the
observer were protected from contact with the smoke. This rela-
tionship suggests that the optical density through which lamps or
signs may be observed can vary from about 1.3 to 4, depending on
factors that apparently include but probably are not limited to
the brightness of the light source and whether the object is
viewed by transmitted or reflected light. Other available data s°
suggests a somewhat different range of optical density (0.7 to 3)
and that the high limit seems to be associated with rather short
path lengths while the lower level is with longer light paths but
|ower smoke optical densities per unit path length. T h e hatched
area in Figure 0-2.2.4 represents the approximate region within
which these limits might apply.
0-3.5 It should be recognized that the work on which Figure
0-3.3 was based did not involve exposure of the observer s eyes to
smoke. Recent studies sz suggest that much lower optical density
values would limit visibility under direct smoke exposure condi-
tions. The amount of the reduction would be highly dependent
on the chemical combustion products developed and might vary
from 50 to 5 percent or even less.
C-3.6 AppUcation of the Test -- A Case History
C.3.6.1 A test method, such as the one discussed in this
commentary, has obvious merit as a measurement tool for ass/sting
in research, development, and production quality control of ma-
29Jin, T., "Visibility Through Fire Smoke (Part 2, Visibility of Monoch.ro-

matic Signs through Fire Smoke)," Report of Fire Res. Inst. of ]ap~n, No.
33, 31-48 ( 1 9 7 1 ) . , ,,
S0Rasbash, D. J., 'Smoke and Toxic Products Produced at Fires, Trans.
].Plasticslnst.,55-61 (Jan. 1967). . ,
3tMarcy, John, Private communication resulting from contractual wor~
performed for Federal Aviation Administration.
1412
1.0 I I
L__
\
\
\ \\\\ Beyond
\
\ visual
m \ \ range
"0
\ \ •
0
_J
\ %\
c~
t-
0.1
"0
m
0
E n
t~
0.01 I I I I IIIII I I I [ III

3 5 10 30 50 100
Visual range (ft.)
Figure C-$.4. SmokeDensity and Visual Range.
terials and products. There should be little need to detail its

possible applications in these fields. T h e method has been shown
to possess acceptable precision and accuracy, and it should serve
well for these purposes. Use of the test method for rough analysis
of the smoke production during an actual fire is informative in
showing the magnitude of the smoke problem. T h e example used
will be based on an actual building and the reported behavior of
a fire which occurred in it.
C.3.6.2 Building Features. T h e apartment building in-

volved comprised a 10-story structure with 30 apartments on each
floor. Each apartment had about 500 ft. 2 floor area and ceiling
height of 8 feet, or a gross volume of about 4000 ft. s. T h e air supply
to each floor corresponded to an hourly air ventilation rate for
each apartment of 1~ changes. T h e r e was a corridor of 300 feet
in length on each floor. Since corridors were 6 feet wide and 8
feet high, this corresponds to a floor area of about 1800 ft. 2 and
volume of 14,400 ft. 3. Thus; each floor of the building enclosed a
gross volume of about 134,000 ft. s. Both apartments and corridor
floors were fully carpeted with a polypropylene twist carpet with
1413
APPENDIX C 258-41
jute backing and foam rubber pad. The carpet showed D m values
of 110 and 480 without and with the pad, respectively, under
flaming smoke test conditions.
C.3.6.3 The fire which occurred started in the living room

of one of the apartments. It spread to the corridor and involved
burn-through of carpet and rubber pad on 300 ft. 2 of apartment
floor covering and 720 ft. 2 of corridor floor covering. An addi-
tional 900 ft. 2 of carpet surface was burned in the corridor, the
rubber pad being unaffected. T h e fire burned for about 40 minutes
before fire fighting was successful in its control.
• C.3.6.4 A total of eight people died and autopsies showed

that six o f these had blood carboxyhemoglobin content greater
than 40 percent, a very hazardous level. A total, of.74 people were
rescued by the fire service. Smoke producuon durmg the fire was
very heavy and the smoke traveled through the elevator shafts,
also filling the three floors above that involved.
C-3.6.5 Smoke Production Calculations. T h e data from

smoke tests of the carpeting material provides a basis of estimating
the smoke production from that source alone. The smoke density
D / L is computed from the relationship D / L -- ADm/V, where V
is the volume assumed occupied by the smoke. Several different
assumptions have been made to show the nature of the problem.
The results are presented in Table C-3.6.5.
C-3.6.6 In making these estimates, several assumptions

have been made which are not usually individually valid:
(a) No loss of smoke occurs to spaces other than that as-
sumed to be Occupied by the smoke.
(b) Maximum smoke generation occurs essentially simul-
taneously.
(c) Perfect mixing of smoke and air takes place in the
volume occupied by the smoke.
(d) All the smoke produced resulted from involvement of
the floor covering.
Assumption (a) was probably valid during early stages of. the fire
because of the low building ventilation rate. At later periods after
the apartment window opened this assumption would obviously
be in error. This assumption results in very significant errors in
smoke density estimates on the high side.
Table C-S.6~
Results of C a l c u l a t i o n o f S m o k e D e n s i t y R e s u l t i n g f r o m Fire
I n v o l v e m e n t of C a r p e t i n g of V a r y i n g A r e a a n d S m o k e D i l u t i o n in
V a r y i n g V o l u m e s of a n A p a r t m e n t B u i l d i n g
Area of Carpet Space Assumed Remdflng Smoke

Involved Occupied by Smoke Density DIL Remarks
A 10 f t . 2 Ap.artment of fire 1.20D/ft. Smoke of this density would be intolerable, ,~
in apartment ongm severely limiting vision which if possible i[~
at all would be limited to less than 1.5 ft.
B " 10 f t . 2 Apartment of fire .26 OD/ft. Identification of bright objects possible
in apartment origin and all at distances of about 4-10 ft. m the absence
adjacent connecting of eye irritants.
corridors ]!
300 f t . 2 Apartment of fire 7.80D/ft. Smoke levels such as fixls would be oppre~
all apartment origin and all sive and prevent identification of bright
floor covering adjacent connecting objects at distances of more than a few [~
corridors inches. Probably a life hazard.
D 300 ft.2 Volume of halls and 1.07OD/ft. Bright objects might be identified at a
all apartment all apartments on range of two ft. People would have ditii-
floor covering the floor of origin culty in self-rescue.
E 300 ft.2 Volume of halls and .27 OD/ft. Identification of bright objects at distances
all apartment apartments on four of about 4-10 ft. possible in absence of eye
floor covering, floors irritants.
300 ft.2 Volume of all halls 1.10D/ft. An intolerable density limiting identifica-
in apartment and apartments on ~ tion of bright objects to ranges of less than
and 750 and four floors, about 1.5 ft. "
900 ft.2 in
corridor the
latter not
including pads.
APPENDIX C
1415
~-,~8---~3
C-$.6.7 The assumption that smoke generation is in-

stantaneous and maximum values of D8 are applicable is an error.
Not only ~s me release of smoke a time dependent function of
exposure ourauon, but the development and spread of fire in the
building was time dependent. Thus, in the absence of loss of
smoke: the smoke densities would build up gradually, during the
_l~erioaot t~e nre to the estimates reported. At early times during
the fire and to me degree that make agglomerates and condenses,
this assumption will tend to yield high values.
C-$.6.8 The assumption that perfect mixing of the smoke
and air occurs is obviously incorrect in this instance, since air in
the building was relatively cjuiescent. T h e atmosphere in a burn-
ing building is usually stratified with buoyant forces carrying the
sm~oke and fire gases to the upper portion of the space occupied.
Horizontal mixing of smoke and air in the corridors would be
accelerated by the ventilation system of the building, by winds,
and by the thermal convection of the fire gases themselves, but
will usually be far from thorough. This assumption will tend to
result in low estimates of maximum smoke density.
C-3.6.9 The assumption that the floor covering was the
only source of smoke is highly unlikely, The carpet involved
would not have burned as it did in the absence of an exposure
fu'e. While cellulosics,, probably, the p.rimary material burned,
often burn under acuve flaming condiuons with less smoke pro-
duction than many polymerics, it is to be expected that large
volumes of fire gases including smoke would result. Thus, this
assumption will tend to result in low estimates of maximum smoke
density.
C-3.6.10 It is believed that these four assumptions are
roughly self-concealing and the estimates of smoke density pro-
vided above, while not effective for the whole of the volumes
assumed occupied, could be expected to be effective over a very
significant portion of the space involved, especially in the vicinity
of the fire.
C.3.6.11 The implications of these estimates emphasize
the magnitude of the smoke problem. The particular floor cover°
ing material involved was not an unusually prolific smoke pro-
ducer for this class product under the flaming conditions assumed.
In spite of this the estimates of smoke density which have been
made show the very serious vision-obscuring situation resulting
from burning even a small amount of material in a very large
volume.
1416
C-~.7 Conclusions
C-3.7,1 Atpresent, we are unable to predict and limit the
size of accidental fires that may occur. Because of this, it seetns
clear that token measures of limiting smoke production, such as
the establishment of a level (e.g., Dmm 400) of extensively used
building components, will provide little if any protection against
development of high smoke levels during building fires even in
the unlikely event that the occupancy fire load is not a potential
smoke producer itself. It will be necessary to consider imposition
of major reductions of either D m or the exposed area of the smoke
producing product involved below that of the worst performers
before it is likely that a significant improvement in smoke produc-
tion can be expected. Of course, the preferable solution, though
not always possible, is to limit the fire growth and spread through
other than strict smoke considerations.
C-3.7.2 The smoke density chamber provides a means for
characterizing smoke production with an accuracy far in excess
of any application requirements which should be recommended.
It also provides a means for reporting rate of smoke production
and time at which specific smoke levels are reached under the test
conditions applied. The original paper describing thig test meth.
ods2 suggestedone way these properties could be combined with
Dm to yield a smoke obscuration index (SOI). Since then, others
have suggested refinement of the classification method, s3 It seems
that refinements such as these are only likely to be of significant
value under the most carefully controlled fire safety conditions.
They are not recommended for use in general application of the
test method.
C-3.7.3 The applications discussed so far are primarily
directed towards safety of building occupants when fire occurs.
The fmdings suggest that small fires may yield large volumes of
smoke even with cellulosic materials. However, the fire services
have found that such fires, when well ventilated, can burn with
greatly reduced smoke production. This is not generally true for
plastic products. 34 These materials often yield increased smoke
production under active flaming conditions. If it is considered
desirable to limit smoke production under these fire-fighting con-
ditions, the obvious solution would be to limit the Dm value for
32Gross, op. cir.

S3Gaskill and Veith, "Smoke Opacity from Certain Woods and Plastics,"
Fire Technology, 4, 3, pp. 185-195 (1968).
34Robertson, A. F., "Tests Indicate Venting Increases Smoke Production from
Some Polymerics," Fire Engineering, 126, pp. 97-8 (Sept. 1973).
1417
APPENDIX C 258-45
all building and furnishing items likely to be extensively fire

involved to a value roughly equal to that of similar items of
cellulosic materials, m troth instances Dm would be observed under
actual flaming conditions. A D~ value of about 100 to 150 would
seem roughl.y appropriate. However, it should be remembered
that precauuons of this type, while likely to make the problem
of fire. fighting easier, cannot be expected to. si.gnificantly-modify
the hfe safety hazard posed by smoke to bmldmg occupants.
C.3.7.4 Application considerations of the smoke density
chamber Test Method for selection of building construction and
occupancy products suggest the following:
(a) Any requirement limiting the maximum Ds value of
a building product, to be valid, must be based on the assumed
quantity of the product likely to be fire exposed.
(b) In any application of the method it should be recog-
nized that the D m values of the various materials tested are not
material properties and representative of' the product in the thick-
ness and assembly nature under the test conditions imposed.
(c) Relatively small quantities of materials, when involved
in fire, may produce very large quantities of smoke. Furthermore,
the normal combustible fire load of the occupancy is likely to
provide high smoke concentrations. Therefore, significant reduc-
tion in smoke development validly can be expected during fires
only if stringent limitations are placed on acceptable levels of ADm
for the product of the area of the building element likely to be
fire exposed and the specific optical density of this element.
(d) Most benefit is likely to be achieved through applica-
tion of controls based on this test method when careful considera-
tion is maintained over all aspects of fire prevention and safety
associated with building construction occupancy and use.
over(e)
all In the absence of careful and continuing consideration
of fire prevention and protection procedures in building
construction and occupancy, the application of controls on smoke
production of a few building or occupancy items or both, is likely
to mislead the public with regard to the degree of control of the
smoke hazard when fires occur.
(D The application of smoke limitations on building occu-
pancy items such as wastebaskets, curtains, chairs, and other igni-
tion-likely, products would a pp ear to serve a useful
. purpose . These
1terns are most likely subject to involvement m early fire develop-
ment and for that reason the smoke they produce occurs at a ume
during which escape is most likely to be practical.
(g) If prime concern is to be accorded the fire fighter in his
need to vent and clear the smoke from buildings, a requirement
1418
258--46 TEST METHOD FOR SMOKE GENERATED
limiting the Dm value of material and occupancy products, likely

to be extensively fire involved, to levels of the order of 100 to 150
under flaming condition would seem appropriate. However, these
limitations will not necessarily reduce the hazard posed by smoke
to the building occupants. This recommendation assumes that
wood, paper, and other cellulosic construction and occupancy
items will be involved in a significant way in any fire which does
Occur.
(h) It seems clear that, u n d e r present accidental fire con-
ditions, large quantities of smoke are to be expected. It is recom-
mended that more attention be given to plans for smoke manage-
ment and disposal when a fire occurs. Such planning is likely to
yield greater benefits with regard to life safety of building occu-
pants than can be achieved by any b u t the most rigid material or
product fire development and smoke control limitations.
CA Review of Smoke Test MethodS5
C-4.1 One of the early concerns for smoke measurement
involved the problem of discharge of furnace gases to the atmos-
phere through stacks. T h e Ringlemann charts which essentially
provide cards of different darkness varying in uniform steps of
reflectance differing from each other by 20 percent. T h e y are
designated as n u m b e r 0, I, 2, 3, 4 and 5. A subjective matching of
the apparent darkness of a smoke plume with the Ringlemann
cards provides a basis for characterizing the effluent from a stack.
A shortcoming of the early use of this classification m e t h o d was
the fact that neither the area of the stack flue nor the discharge
velocity were considered. T h e m e t h o d also suffers from the use
of visual darkness rather than optical density as a measure of
smoke production. Thus, there was little if any basis of reporting
the quantity of smoke production.
C-4.2 T h e measurement m e t h o d was improved later by
various groups through use of an electric p h o t o m e t e r and a light
source. H e r e again in most applications, although some subjective
problems have been reduced, the measurement is still largely
relative since volumetric flow rate and photometer p a t h length
are seldom recognized in reporting the data.
SSRasbash, D. J., "Smoke and Toxic Products Produced at Fires," Trans. ].

Plastics Inst., 55-61 (Jan. 1967), provides a lucid analysis and review of the
smoke and toxic hazard presented by fire gases. It will prove well worth study
by those concerned with the general problem. "The Control of Smoke in
Building Fires - - A State-of-the-Art Review" provides a review of the smoke
problem in building fires with special emphasis on smoke test methods. For those
who do not want to locate this reference, a brief historical review of the develop-
ment of smoke test methods is provided herein.
1419
APPENDIX C 258-47
C-4.$ T h e Steiner tunnel Method for Measuring Surface

Burning Characteristics of Materials (NFPA 255 - 1972) makes
use of a nPnh°lt.°smnetercross axial to the flue at the discharge end
of the tu ,, = " e a 24-foot length of specimen is involved and
only a smart traction ot this length is forceably exposed to the test
burner flame, the smoke measurements resulting from its use will
reflect in some manner the importance of flammability in modify-
ing the observed smoke production. This may be a distinct ad-
vantage of the test method. T h e tunnel procedure, however,
suffers from a technically deficient method of reporting smoke
rOduction. T h e ratio of areas under an attenuation-time curve
r the material as compared with that of red oak is reported.
The difficulty here is that attenuation is not linearly related to
the quantity of smoke produced so the resulting ratm has little
physical significance. T h e scale of smoke production is not a linear
one.
C-4.4 T h e Rohm and Haas Company sponsored develop-

ment of a smoke chamber method of measurement. This is cur-
rendy the basis of ASTM Procedure D2843. This involves burn-
ing a specimen in a partially enclosed chamber. T h e smoke is
allowed to accumulate and photometric measurements are made
of light attenuation over a one-foot path length. T h e procedure
has the advantage of a more quantative measurement of total
smoke produced since it is allowed to accumulate within an en-
closure. T h e method of reporting the data involves reference to
the m a x i m u m attenuation observed while another reporting
method involves calculation of the fractional percentage repre-
sented by the area of the attenuation versus time plot for a four-
minute test period. T h e first procedure yields a measurement
based on the maximum quantity of smoke accumulated. T h e col-
lection process provides for integration of the production rate
involved. T h e significance of the average attenuation over a four-
minute [~eriod is not clear. It probablyprovides a mix of measure-
ments ot the quantity accumulated, the rate of initial production
and the degradation characteristics of the smoke produced. T h e
smoke density chamber Test Method described in this commen-
tary was developed after rather extensive study and modification
of the R o h m and Haas smoke test method. Although it introduced
numerous improvements from the point of view of material char-
acterization, the greatest single innovation was the introduction
of the concept of specific optical density.
1420
Appendix D
D-1 Rderenced Publications. The [allowing is a list O[ publica.
tions ~e[erenced in the standard, but is not a part o[ this standard.
(a) Bartosic, A. J., Rarig, F. J., "Evaluation of the XP2 Smoke
Density Chamber," (Symposium on Fire Test Methods-Restraint
and Smoke, 1966), A S T M Spec. Tech. Publ. 422 (1966).
(b) Brenden, J. J., "Usefulness of ,a, New Method for Smoke
Yield from Species and Panel Products, Forest Prod. J., 21, pp. 23-8
(1971).
(c) Carhart, H., Hazlett, R. N., Johnson, E., Stone, J. P., "The
Transport of Hydrogen Chloride by Soot from Burning Polyvinyl~
chloride," Fire Flammability, 4, pp. 42-51 (Jan. 1973).
(d) "The Control of Smoke in Building Fires -- A State-of-the.
Art Review," Mtls. Res. and Stds., 11, 4, pp. 16-24 (April 1971).
(e) Fristrom, R., Annual Summary Report (July 1972 to 30 June
1973), App. Phys. Lab., Johns Hopkins Univ., Rept. No. FPP A75
(Aug. 1973).
(t) Gardon, R., "An Instrument for the Direct Measurement of
Intense Thermal Radiation," Review o[ Scientific Instruments,
Vol. 24, pp. 366-370 (1953).
(g) Gaskill and Veith, "Smoke Opacity from Certain Woods and
Plastics,' Fire Technology, 4, 3, pp. 185-195 (1968).
(h) Gray, V. E., Gross, D., Lee, T. G., Loftus, J. J., "Smoke and
Gases Produced by Burning Aircraft Interior Materials," NBS
Bldg. Sci. Series, BSS 18, U.S. Gov't. Printing Office, Washington,.
D.C. (Feb. 1969).
(i) Gross, D., Loftus, J. J., Robertson, A. F., "Method for
Measuring Smoke from Burning Materials, Fire Test Methods-
Restraint and Smoke," A S T M S T P No. 422 (1967).
(j) Jin, T., "Visibility Through Fire Smoke (Part 2, Visibility of
Monochromatic Signs through Fire Smoke)," Report o/Fire Res.
Inst. o[ Japan, No. 33, pp. 31-48 (1971).
(k) Lee;, T. G., "Interlaboratory Evaluation of Smoke Density
Chamber, NBS Tech. Note 708 (Dec. 1971).
(1) Lee, T. G., "The Smoke Density Chamber Method for Evalu-
ating the Potential Smoke Generation of Materials," NBS Tech.
Note 757 (Jan. 1973).
(m) Marcy, John, Private communication resulting from con-
tractual work performed for Federal Aviation Administration.
(n) Method o[ Test o[ Surface Burning Characteristics o/Building
Materials, NFPA 255, 1972.
1421
A P P E N D I X D 258-49
(o) ,Rasbash, D. J., "Smoke and Toxic Products Produced at

Fires, Trans. 1. Plastics Inst., pp. 55-61 (Jan. 1967).
(p) Robertson, A. F., "Tests Indicate Venting Increases Smoke
production from Some Polymerics," Fire Engineering, 126, 9, pp.
• . . °
9f~(St'~SPetta~9!nr3!ngethodofOpoMtasUro?gpltht~csD,e/~3y of Smoke
Ann. Book
d A S T M Stds., Designation D2843-70, Vol. 27, pp. 714-722 (1975).
1)-2 Selected Papers for Further Study
]9-2.1 General. T h e papers enumerated below have been
selected as an introduction to a much broader list of documents
relating to smoke problems during building fires. In selection of
these references emphasis has been placed on USA and Canadian
work. There is a considerable body of foreign language publica-
tions of great merit, but with one or two exceptions these have not
been included.
I)-2.2 Review Papers
(a) Little, Arthur D., "Fire Gas Research Report," NFPA
(2uarterly45, 3, pp. 280-306 (Jan. 1952).
(b) Auticu, J., Toxicology Aspects of Flammability and
Combustion of Polymeric Materials," Fire and Flam., pp. 259-268
(July 1970).
(c) Galbreath, M., "Fire in High Buildings," National Re-
search Council of Canada, Fire Study, 21, NRC No. 10081 (Ottawa,
April 1968).
(d) Wood, P. G., "The Behavior of People in Fire," Joint
Fire Res. Ors. Fire Research Note No. 953, p. 113 (Nov. 1972).
D-2.3 Fire Gas and Smoke Hazard
(a) Bono, J. A., Breed, B. K., "Smoke Ratings in Relation to
Visual Observations, Fire Technology, 2, 2, pp. 146-158 (May
1966~)
x Fiorca, V., Higging, E. A., Thomas, A. A., and Davis,
H. V., "Acute Toxicity of Brief Exposures to HF, HC1, NO 2 and
HCN with and without CO," Fire Technology, 8, 2, pp. 120-129
(May 1972).
(c) Shern, J. H., "Smoke Contribution Factor in Fire Hazard
Classification of Building Materials in Fire Test Methods-Re-
straint and Smoke," A S T M STP 422 (1966).
(d) Smith, E. E., "Evaluation of the Fire Hazard of Duct
Materials, Fire Technology, 9, 3, pp. 157-170 (Aug. 1973).
(e) Tsuchiya, Y., and Sumi, K., "Combined Lethal Effect of
Temperature CO, CO 2 and O 2 of Simulated Fire Gases," Fire
andFlam., 4, 132-140 (April 1973).
1422
(f) Tsuchiya, Y., and Sumi, K., "Evaluation of the Toxicity

of Combustion Produc,ts, Fire and Flare., 3, pp. 46-50 (Jan. !772)
(g) Yuill, C. H., The Life Hazard of Bedding Fires, Fire
Technology, 1, pp. 312-325 (Oct. 1970).
1)-2.4 Smoke Test Methods
(a) Birky, M., Comeford, J. J., "A Method for the Measure.
ment of Smoke and HCI Evaluation from Poly(vinyl-chloride),,,
Fire Technology, 8, 2, pp. 85-90 (May 1972).
(b) Christian, W. J., and Waterman, T. E., "Ability of Small.
Scale Tests to Predict Full-Scale Smoke Production," Fire Tech.
nology, 7, 4, pp. 332-344 (Nov. 1972).
(c) "Method of Test for Surface Flammability of Material~
Using a Radiant Heat Energy Source," E-162, Annual Book of
A S T M Standards, ASTM, Philadelphia, PA (1973).
(d) Silversides, R. G., 'Measurement and Control. of Smoke
in Building Fires," (Symposium on Fire Test Methods-Restraint
and Smoke, 1966) A S T M STP 422 (1966).
(e) Smith, E. E., "Measuring Rate of Heat, Smoke, and Toxic
Gas Release," Fire Technology, 3, 3, pp. 257-245 (Aug. 1972).
D-2:5 Smoke Behavior During Fires
(a) Bell, D., Lieberman, P., "Smoke and Fire Propagation in
Compartment Space," Fire Technology, 9, 2, pp. 91-100 (May 1973).
(b) Fackler, J. B., "Fire and Smoke Invasion of Apartments
in Fire Test Methods-Restraint and Smoke," A S T M STP 422,
pp. 205-219 (1966). ,,
(c) Heselden, A., Hinkley, P., Smoke Travel in Shopping
Malls -- Experiments in Cooperation with Glasgow Fire Brigade,'
Part I, Joint Fire Res. Org., Fire Research Note 832 (July 1970),
see Fire Res. Abs. and Rev,,12, 3, p. 234 (1970).
(d) Hinkley, P. L., The Flow of Hot Gases Along an En-
closed Shopping Mall - A Tentative Theory,' Joint Fire Res.
Org., Fire Research Note 807 (March 1970), see also Fire Research
Abstracts, and Reviews 12, 3, p. 235 (1970).
(e) Movement of Smoke on Escape Routes in Buildings,"
(Proceedings of the Symposium held at Watford College of Tech-
nology), Joint Fire Res. Org., Her Majesty's Stationery Office,
London (1971 ). , ,,
(f) Shorter, G. W., 'The St. Lawrence Burns, NFPA Quar-
terly, 53,4, pp. 300-16 (April 1958).
D-2.6 Smoke Management During Fires
(a) "Explanatory Paper on Control of Smoke Movement in
High Buildings," National Building Code of Canada, Nat. Res.
Council, Canada, NRC, No. 11413.
1423
APPENDIX V 258-51
(b) Haggerty, J. R., "A New Look at Smoke Ston Partitions"

Fire 7"echnology, 6, 2, pp. 102-110 (May 1970). r ,
(c) McGuire, Tamura, and Wilson, "Factors in Controlling
smol~e'in High Buildings," ASHRAE Symposium, Fire Hazards in
Buildings, PP. 8-13 (Jan. 1970).
(d) Shaw, C., Tamura, G., "Basis for the Design o£ Smoke
Shafts," Fire Technology, 9, 3, pp. 209-222 (Aug. 1973).
(e) shern, J. H., Smokeproof Towers in Center Core Struc-
tures," Fire Technology, 2, 4, pp. 303-307 (Nov. 1966).
(f) Shorter, G. W., Wilson, A. G., "Fire and High Buildings."
Fire Technology, 6, 4, pp. 292-304 (Nov. 1970).
D-2.7 Chemical Analysis of Fire Gases
(a) DiPietro, J., and Stepniczka, H., "A Study o[ Smoke Den-
sity and Oxygen Index of Polystyrene ABS and Polyester Systems,"
Fire and Flare., 2, pp. 36-53 (Jan. 1971).
otb) Gaskill, "Smoke Development in Polymers During
PYr ysis of Combustion," Fire and Flammability, 1, pp. 183-216
(1970). ,,
(c) Hilado, C., The Effect of Chemical and Physical Factors
on Smokc E',vaba uation from Polymers," Fire and Flare., I, pp. 217-
238 (July 1970).
(d) Hilado, C. J., "Smoke from Cellular Polymers," Fire
Technology, 5, 2, pp. 130~139 (May 1969).
(c) Wagner, J. P., Su~cy of Toxic Species in the Pyrolysis
of Combustion of Polymers, Fire Research Abstracts and ReViews
14, 1, pp. 1-23 (1972).
(t) Wodley, Ann I., Woolley, W. D., The Thermal Decomposi-
tion Products of Phenol-Formaldehyde Laminates.

Report of Committee On Fire Tests: ! . E. Miller° Factory Insurance

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Report of Committee On Fire Tests: ! . E. Miller° Factory Insurance

Uploaded by

Copyright:

Available Formats

1376

REPORT OF THE COMMITTEE ON FIRE TESTS Fr-1

Report of Committee on Fire Tests

This report has been submitted in two parts to letter ballot of

Part I consists of the standard text (Chapters 1 through 6). Part

On Part 1 18 have voted aO~rmatively, Messrs. Maroni and Ryan

On Part 1I, 16 have voted a~rmatively, Messrs. Blair, Maroni,

The Committee on Fire Tests recommends the tentative adop-

1-1 Scope. This method of test covers a procedure for measur-

1A commentary describing the significance of specific optical density and ap-

1-3 Summary of Method

1-3.1 T h i s method for measuring the smoke generated by

1-3.2 For the flaming condition, a six-tube b u r n e r is used to

1-3.3 T h e test specimens are exposed to the flaming and non-

1-3.4 Calibration procedures for the test equipment, as

1-3.5 T h e light transmittance measurements are used to ex-

3Additional parameters, such as the m a x i m u m rate of smoke accumulation,

2-1 T h e apparatus shall be essentially as shown in Figures 2-1A

4A more detailed description of suggested details is given in A-1.

2-2 Test Chamber

5Commercially available panels of porcelain.enamel steel (interior surface)

2-3.4.2 The control system shall consist of an autotrans-

SWhere line voltage fluctuations are present, a constant-voltage transformer

2-5 Framework for Support of the Furnace and Specimea

Figure 2-5. F u r n a c e Support.

2.6 Photometric System

~Side o: I' l ~' ~PSideof chamber

~''r i "~"~/77'j~ J~l ~'B°tt°m°' chamber

2-6A. Photometer Details.

PARTIAL PLAN VIEW

2-7.2 T h e construction of the radiometer shall be as shown

~,-,, "~ I' I I-,T-.,..,L

2-8 Thermocouples for Determining Chamber Wail Tern.

2-9 Portable Recorder or Read-Out Meter

2.10 Manometer for Chamber Pressure Measurements

2-11 Multiple Flamelet Burner with Premixed Air-Propane

Chapter 3 Test Specimens

~-1 Specimen Description

3-1.3.3 When supplementary tests are necessitated by de.

101f under either the flaming or nonflaming exposure, there is an excess of

4-1 All tests shall be conducted in a room or enclosed space

4.2.1 T h e chamber wall s shall be cleaned whenever periodic

llCharred residues on the specimen holder and horizontal rods should

The specimen holder shall be returned immediately to the

4-6 Then the exhaust vent and blower shall be closed.

4-7 The loaded specimen holder shall be placed on the bar

4,8 T h e chamber door shall be quickly closed and the timer

4-9 Light transmittance and the corresponding time shall be

4.10 T h e increase in chamber pressure shall be observed with

4-11 Any observations pertinent to the burning and smoke

4-12 The test shall continue for 20 minutes or until a mini-

4.,IS If transmittance falls below 0.01~7o, the chamber window

4-14 The burner on flaming exposures shall be extinguished

14In some cases the transmittance m a y increase somewhat and subsequently

5-1 C a l c u l a t e specific o p t i c a l d e n s i t y , De, f r o m t h e r e d u c t i o n

15Precision. The extent of agreement between replicate tests by the same

Exposure Number of Number of Range in Multilaboratory

6.1 T h e report (see Appendix B) shall include the following:

(a) Complete description of the specimen tested including:

6.2 If the test is terminated on the basis of a 20-minute ex-

16Analysis of products of combustion. Although not specifically required

Apparatus Construction and Cah'bration

A-I Construction Details