ISO/TR 16312-2:2007

TECHNICAL ISO/TR
REPORT 16312-2
First edition
2007-03-01
Guidance for assessing the validity of
physical fire models for obtaining fire
effluent toxicity data for fire hazard and
risk assessment —
Part 2:
Evaluation of individual physical fire
models
Lignes directrices pour évaluer la validité des modèles de feu physiques
pour l'obtention de données sur les effluents du feu en vue de
l'évaluation des risques et dangers —
Partie 2: Évaluation des différents modèles de feu physiques

Reference number
©
ISO 2007
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.

©  ISO 2007
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2007 – All rights reserved

Contents Page
Foreword. iv
Introduction . v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions. 1
4 General principles. 1
4.1 Physical fire model . 1
4.2 Model validity . 2
4.3 Test specimens . 2
4.4 Combustion conditions. 2
4.5 Effluent characterization. 2
5 Significance and use . 2
6 Physical fire models . 3
6.1 Cup-furnace smoke-toxicity test method. 3
6.2 Radiant furnace toxicity test method (United States) . 5
6.3 Closed cabinet toxicity test (international) . 8
6.4 Closed flask test (Israel) . 10
6.5 NES 713 (United Kingdom) . 12
6.6 Japanese toxicity test. 14
6.7 Cone Calorimeter (International). 17
6.8 Flame propagation apparatus (United States) . 19
6.9 University of Pittsburgh tube furnace . 21
6.10 Tube furnace (Germany) . 24
6.11 Tube furnace (France). 27
6.12 Tube furnace (United Kingdom) . 29
Bibliography . 32

Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
In exceptional circumstances, when a technical committee has collected data of a different kind from that
which is normally published as an International Standard (“state of the art”, for example), it may decide by a
simple majority vote of its participating members to publish a Technical Report. A Technical Report is entirely
informative in nature and does not have to be reviewed until the data it provides are considered to be no
longer valid or useful.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO/TR 16312-2 was prepared by Technical Committee ISO/TC 92, Fire safety, Subcommittee SC 3, Fire
threat to people and environment.
ISO 16312 consists of the following parts, under the general title Guidance for assessing the validity of
physical fire models for obtaining fire effluent toxicity data for fire hazard and risk assessment:
⎯ Part 1: Criteria
⎯ Part 2: Evaluation of individual physical fire models [Technical Report]
iv © ISO 2007 – All rights reserved

Introduction
Providing the desired degree of life safety for an occupancy increasingly involves an explicit fire hazard or risk
assessment. This assessment includes such components as information on the room/building properties, the
nature of the occupancy, the nature of the occupants, the types of potential fires, the outcomes to be avoided,
etc.
This type of determination also requires information on the potential for harm to people due to the effluent
produced in the fire. Because of the prohibitive cost of real-scale product testing under the wide range of fire
conditions, most estimates of the potential harm from the fire effluent depend on data generated from a
physical fire model, a reduced-scale test apparatus and procedure for its use.
The role of a physical fire model for generating accurate toxic effluent composition is to simulate the essential
features of the complex thermal and reactive chemical environment in full-scale fires. These environments
vary with the physical characteristics of the fire scenario and with time during the course of the fire, and close
representation of some phenomena occurring in full-scale fires can be difficult or even not possible at the
small scale. The accuracy of the physical fire model, then, depends on two features:
a) degree to which the combustion conditions in the bench-scale apparatus mirror those in the fire stage
being simulated;
b) degree to which the yields of the important combustion products obtained from burning of the commercial
product at full scale are matched by the yields from burning specimens of the product in the small-scale
model. This measure is generally performed for a small set of products, and the derived accuracy is then
presumed to extend to other test subjects. At least one methodology for effecting this comparison has
[1]
been developed.
This part of ISO 16312 provides a set of technical criteria for evaluating physical fire models used to obtain
composition and toxic potency data on the effluent from products and materials under fire conditions relevant
to life safety. This Technical Report comprises the application by experts of these criteria to currently used test
methods that are used for generating data on smoke effluent from burning materials and commercial products.
There are 12 physical fire models discussed in this part of ISO 16312. Additional apparatus can be added as
they are developed or adapted with the intent of generating information regarding the toxic potency of smoke.
For the 12 models in this part of ISO 16312, the first five are closed systems. In these, no external air is
introduced and the combustion (or pyrolysis) products remain within the apparatus except for the fraction
removed for chemical analysis. The second seven are open apparatus, with air continuously flowing past the
combusting sample and exiting the apparatus, along with the combustion products.
To make use of this part of ISO 16312, it is necessary for the user to have present a copy of ISO 16312-1,
which contains much of the context and definitions for the present document. It is also necessary to make
[33] [34] [31] [32]
reference to ISO 19701 , ISO 19702 , ISO 19703, ISO 13344 , and ISO 13571 for discussions of
analytical methods, bioassay procedures, and prediction of the toxic effects of fire effluents.

TECHNICAL REPORT ISO/TR 16312-2:2007(E)

Guidance for assessing the validity of physical fire models for
obtaining fire effluent toxicity data for fire hazard and risk
assessment —
Part 2:
Evaluation of individual physical fire models
1 Scope
This part of ISO 16312 assesses the utility of physical fire models that have been standardized, are commonly
used and/or are cited in national or international standards, for generating fire effluent toxicity data of known
accuracy. It does so using the criteria established in ISO 16312-1 and the guidelines established in ISO 19706.
The aspects of the models that are considered are the intended application of the model, the combustion
principles it manifests, the fire stage(s) that the model attempts to replicate, the types of data generated, the
nature and appropriateness of the combustion conditions to which test specimens are exposed and the
degree of validity established for the model.
2 Normative references
The following referenced documents are indispensable for the application of this document. For dated
references, only the edition cited applies. For undated references, the latest edition of the referenced
document (including any amendments) applies.
ISO 13943, Fire safety — Vocabulary
ISO 16312-1, Guidelines for assessing the validity of physical fire models for obtaining fire effluent toxicity
data for fire hazard and risk assessment — Part 1: Criteria
ISO 19703, Generation and analysis of toxic gases in fire — Calculation of species yields, equivalence ratios
and combustion efficiency in experimental fires
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO 13943 and in ISO 19703 apply.
4 General principles
4.1 Physical fire model
A physical fire model is characterized by the requirements placed on the form of the test specimen, the
operational combustion conditions and the capability of analysing the products of combustion.
4.2 Model validity
For use in providing data for effluent toxicity assessment, the validity of a physical fire model is determined by
the degree of accuracy with which it reproduces the yields of the principal toxic components in real-scale fires.
4.3 Test specimens
Fire safety engineering requires data on commercial products or product components. In a reduced-scale test,
the manner in which a specimen of the product is composed can affect the nature and yields of the
combustion products. This is especially the case for products of non-uniform composition, such as those
consisting of layered materials.
4.4 Combustion conditions
The yields of combustion products depend on such apparatus conditions as the fuel/air equivalence ratio,
whether the decomposition is flaming or non-flaming, the persistence of flaming of the sample, the
temperature of the specimen and the effluent produced, the thermal radiation incident on the specimen, the
stability of the decomposition conditions and the interaction of the apparatus with the decomposition process,
with the effluent and the flames.
4.5 Effluent characterization
4.5.1 For the effluent from most common materials, the major acute toxic effects have been shown to
depend upon a small number of major asphyxiant gases and a somewhat wider range of inorganic and
[32]
organic irritants. In ISO 13571 , a base set of combustion products has been identified for routine analysis.
Novel materials can evolve previously unidentified toxic products. Thus, a more detailed chemical analysis
can be needed in order to provide a full assessment of acute effects and to assess chronic or environmental
toxicants. A bioassay can provide guidance on the importance of toxicants not included in the base set.
[35]
ISO 19706 contains a fuller discussion of the utility of bioassays.
4.5.2 It is essential that the physical fire model enable accurate determinations of chemical effluent
composition.
4.5.3 It is desirable that the physical fire model accommodate a bioassay method.
4.5.4 The use of laboratory animals as test subjects is the only means of insuring inclusion of the impact of
all combustion gases. However, it is recognized that the adoption and use of protocols using laboratory
animals can be prohibited in some jurisdictions. An animal-free protocol captures the effects of known
combustion gases but misses the impact of any uncommon and highly toxic species, those smoke
components that are most in need of identification. Laboratory studies to date have shown that lethality from
smoke inhalation results from the combined effects of a small number of gases and that none of the missing
gases is “supertoxic.” There are also data that indicate incapacitation results from half the lethal exposure for
a wide range of today’s materials, indicating that exotic gases do not affect incapacitation without affecting
lethality as well. The decision to base hazard and risk assessments on analytical or animal-based
measurements resides with the authority having jurisdiction.
5 Significance and use
5.1 Most computational models of fire hazard and risk require information regarding the potential of fire
effluent (gases, heat and smoke) to cause harm to people and to affect their ability to escape or to seek
refuge.
5.2 The quality of the data on fire effluent has a profound effect on the accuracy of the prediction of the
degree of life safety offered by an occupancy design. Uncertainty in such predictions commonly leads to the
use of safety factors that can compromise functionality and increase cost.
2 © ISO 2007 – All rights reserved

5.3 Fire safety engineering requires data on commercial products. Real-scale tests of such products
generally provide accurate fire effluent data. However, due to the large number of available products, the high
cost of performing real-scale tests of products and the small number of large-scale test facilities, information
on effluent toxicity is most often obtained from physical fire models.
5.4 There are numerous physical fire models cited in national regulations. These apparatus vary in design
and operation, as well as in their degree of characterization. The assessments of these models in this part of
ISO 16312 provide product manufacturers, regulators and fire safety professionals with insight into
appropriate and inappropriate sources of fire effluent data for their defined purposes.
5.5 None of the models in this part of ISO 16312 is appropriate for simulation of smouldering combustion.
5.6 The assessments of physical fire models in this part of ISO 16312 do not address means for combining
[31]
the effluent component yields to estimate the effects on laboratory animals (see ISO 13344 ) or for
[32]
extrapolating the test results to people (see ISO 13571 ).
6 Physical fire models
6.1 Cup-furnace smoke-toxicity test method
6.1.1 Application
[2]
This method is designed to generate toxic potency data for materials and, perhaps, end products. It is not a
national or international standard.
6.1.2 Principle
A schematic of the apparatus is shown in Figure 1. The furnace is open to an 0,2 m closed reservoir from
which (air) oxygen is supplied by natural buoyancy. Vitiation in the reservoir is measured. The sample
(approximately 10 g) is cut into pieces and heated conductively, convectively and (at higher temperatures)
radiatively to just below or just above its auto-ignition temperature.
6.1.3 Fire stage(s)
[35]
The fire stage(s) from ISO 19706:2007 , Table 1, are as follows:
⎯ 1.b, oxidative pyrolysis;
⎯ 2, well-ventilated flaming.
6.1.4 Types of data
The standard procedure includes measurement of total mass loss, averaged mass consumed and mass
charged concentrations, gas concentrations and gas yields. The gases to be measured are: CO , CO, O ,
2 2
HCN, HCl and HBr. In addition, the procedure includes measurement of the incapacitation (by hind-leg flexion
or immobilization) and mortality of six rats, the times to these effects and documentation of any physiological
harm, determined post-mortem. Blood samples are taken during and after exposure for subsequent analysis.
6.1.5 Presentation of results
Sufficient tests are performed, at different mass loadings, to determine LC and IC values and their
50 50
confidence limits for within exposure and within-plus-post-exposure periods.

Key
1 furnace 7 thermocouple
2 gas-sampling port 8 1 000 ml quartz beaker
3 pressure-relief panel 9 ceramic
4 animal ports 10 thermocouple well
5 galvanized sheet 11 heating element in bottom
6 insulation
Figure 1 — Schematic of the cup-furnace smoke-toxicity apparatus
6.1.6 Apparatus assessment
6.1.6.1 Advantages
Each test uses a small sample. The apparatus is inexpensive and easy to operate. Data for a wide range of
materials and products have been published. There is a close similarity to the oxidative pyrolysis conditions in
real-scale fires.
6.1.6.2 Disadvantages
The realism of sample exposure is questionable due to the cutting up of the sample, especially for non-
homogeneous products. For well-ventilated combustion, the simulation of real-scale heating, which is primarily
radiative, is poor. Mixing by natural buoyancy makes values of the global equivalence ratio somewhat
uncertain. In common with many physical fire models, no indication is given about the rate of burning;
4 © ISO 2007 – All rights reserved

therefore, additional data input on burning rates at different fire stages are needed for fire safety engineering
calculations.
6.1.6.3 Repeatability and reproducibility
[3]
A successful inter-laboratory evaluation of this method has been performed .
6.1.7 Toxicological results
6.1.7.1 Advantages
The method produces true measures of smoke lethality and incapacitation and identifies instances of extreme
and unusual smoke toxic potency. It also produces data enabling calculation of the yields of measured
toxicants. It can identify cases where unusual toxicity occurs as a result of constituents not identified by the
analytical procedures applied.
6.1.7.2 Disadvantages
The relationship between data for a finished product and data for its component materials has not been
determined. The concentration of combustion products is not truly uniform over the entire animal-exposure
period, introducing some reduction in the precision of the lethality and incapacitation measures.
6.1.8 Miscellaneous
This is primarily an animal-exposure test with chemical instrumentation to quantify the expected major
toxicants. Additional analytical instrumentation can be added with little interference with the standard method.
The apparatus can be used without test animals, but it then loses the ability to identify the principal cases of
real interest.
6.1.9 Validation
[1]
The toxic potency and gas yield data did not replicate real-scale post-flashover test data well . The method
has not been assessed against real-scale test data for oxidative pyrolysis or well-ventilated flaming.
6.1.10 Conclusion
This method is potentially a useful test for screening the toxic potency of materials and homogeneous
products. However, cutting the specimen into pieces makes it unlikely that the test results relate to the real fire
exposure of heterogeneous end products. Thus, with validation, it can produce useful information for hazard
models for oxidative pyrolysis and well-ventilated flaming of homogeneous materials, but not of complex
commercial products.
6.2 Radiant furnace toxicity test method (United States)
6.2.1 Application
[4] [5]
This apparatus, used in NFPA 269 and ASTM E 1678 , was designed to generate toxic potency data for
building and furnishing materials and end products for use in fire and hazard analyses.
6.2.2 Principle
A photograph of the apparatus is shown in Figure 2. A sample, up to 76 mm x 127 mm in area and up to
50 mm in thickness and representative of the end-use configuration of the finished product, is exposed to
thermal radiation. Buoyancy from the burning sample entrains air from a closed reservoir similar to that
described in 6.1.
Key
1 chimney 4 specimen holder
2 combustion cell 5 animal ports
3 radiant heaters 6 expansion bag
Figure 2 — Photograph of the NFPA 269 apparatus
6.2.3 Fire stage(s)
[35]
The fire stage(s) from ISO 19706:2007 , Table 1, are as follows:
⎯ 2, well-ventilated flaming combustion;
⎯ 3, under-ventilated flaming (using a post-flashover correction for the yield of CO);
⎯ 1.b, oxidative pyrolysis, if the sample does not auto-ignite.
6.2.4 Types of data
The standard procedure includes continuous measurement of mass loss and gas concentrations, gas yields,
and atmosphere vitiation. In addition, the procedure includes measurement of the mortality of six rats and
documentation of any physiological harm, determined post-mortem.
6.2.5 Presentation of results
Sufficient tests are performed, at different sample surface areas, to determine LC values and their
confidence limits for within exposure and within-plus-post-exposure periods.
6 © ISO 2007 – All rights reserved

6.2.6 Apparatus assessment
6.2.6.1 Advantages
The representation and exposure of finished products is accurate. The mass burning rate is recorded,
enabling direct use of the yield data in engineering calculations.
6.2.6.2 Disadvantages
Mixing by natural buoyancy makes values of global equivalence ratio somewhat uncertain. Limited sample
intumescence can be tolerated.
6.2.6.3 Repeatability and reproducibility
No inter-laboratory evaluation of this method has been performed.
6.2.7 Toxicological results
6.2.7.1 Advantages
The method produces a true measure of smoke lethality and identifies instances of extreme and unusual
smoke toxic potency. It also produces data enabling calculation of the yields of measured toxicants. The
method can be adapted to measure incapacitation (hind-leg flexion or immobilization).
6.2.7.2 Disadvantages
The use of an empirically derived correction for CO introduces some uncertainty into the LC values.
Furthermore, altered yields of other product gases are not included. However, these factors are included in
the comparison of LC values with room-scale test data in Reference [1]. This correction limits the value of
this method for other sublethal effects in which the uncorrected gas yields play a prominent role.
6.2.8 Miscellaneous
This is primarily an animal-exposure test with limited chemical instrumentation. However, additional analytical
instrumentation can be added with little interference with the standard method. The apparatus can be used
without test animals, but it then loses the ability to identify the principal cases of real interest.
Using a generic, experimentally observed carbon monoxide yield correction, accurate post-flashover LC
[1]
values have been obtained relative to real-scale fire tests of the same combustibles .
6.2.9 Validation
The output of this method for three products has been compared to room-scale data for the same products
[1]
(wall lining configuration) . The post-flashover LC values were well within a factor of 2. No data were taken
for pre-flashover combustion.
6.2.10 Conclusion
This is a useful test for obtaining quantitative toxic potency information for materials and end products for input
to fire hazard models.
6.3 Closed cabinet toxicity test (international)
6.3.1 Application
[6] [7]
This physical fire model is used in ISO 5659-2 and ASTM E 1995 . It was designed to generate smoke
optical density data. The International Maritime Organization (IMO) also requires use of this apparatus for
toxic gas concentration data for qualification of materials.
6.3.2 Principle
A schematic of this closed cabinet test is shown in Figure 3. A horizontally mounted specimen, 75 mm square
and up to 25 mm thick, is exposed to a radiant heater for a minimum of 10 min. A test is conducted at
2 2
25 kW/m with and without pilot flame and at 50 kW/m without pilot flame. The gases are sampled through
probes positioned at the geometrical centre of the smoke box.
6.3.3 Fire stage(s)
[35]
The fire stage(s) from ISO 19706:2007 , Table 1, are as follows:
⎯ 1.b, oxidative pyrolysis;
⎯ 3.a, well-ventilated flaming combustion; see 6.3.6.2.
6.3.4 Types of data
The standard procedure includes measurement of total mass loss, smoke obscuration and specific effluent
gas concentrations (CO , CO, HCN, HCl, HF, HBr, NO , SO ) at the time when the maximum smoke
2 x 2
concentration is reached. In the two aircraft tests, the specific optical density of the smoke and the gas
concentrations are determined at 90 s and 240 s. Fire effluent for cables may be determined after 20 min.
6.3.5 Presentation of results
The specific optical density of the smoke and the combustion fire gas concentrations are compared to
specified values.
6.3.6 Apparatus assessment
6.3.6.1 Advantages
The apparatus is simple to use and widely available. The test specimen can be a reasonable representation of
a finished product.
6.3.6.2 Disadvantages
The combustion conditions are fixed and not well characterized. The test specimen is vertical and melting
materials can flow into the trough below the specimen holder or even onto the floor of the test chamber,
thereby altering the combustion mode or even reducing the amount of specimen destroyed. The gases are
mixed by natural convection and possible stratification can lead to non-representative sampling of the
combustion gases. Vitiation can occur with thick or less thermally stable materials and affects the yields of
combustion products.
6.3.6.3 Repeatability and reproducibility
Inter-laboratory evaluations have been performed for the smoke density test and gave satisfactory results for
a range of materials. No inter-laboratory evaluation of toxic gas production has been reported.
8 © ISO 2007 – All rights reserved

Key
1 photomultiplier-tube housing 4 light source window
2 radiator cone 5 blow-out panel
3 pilot burner
Figure 3 — Schematic of the closed cabinet toxicity test apparatus

6.3.7 Toxicological results
6.3.7.1 Advantages
The initial conditions are few and well prescribed.
6.3.7.2 Disadvantages
Possible vitiation can lead to time-dependent generation of toxicants that are sampled only at a specified time.
Condensation can occur on the wall of the chamber leading to removal of some gases from the sampled
environment. The possibility of condensation on the wall of the chamber can lead to removal of some gases
from the sampled environment. The prescribed set of gases to be measured can be insufficient to estimate
lethal toxic potency.
6.3.8 Miscellaneous
No animals are exposed in the test, nor is the apparatus compatible with such an addition. The use of the
chemical data is typically limited to a comparison with critical concentrations of listed toxic gases.
6.3.9 Validation
There are no reported comparisons of toxic gas generation with data from real-scale fire tests.
6.3.10 Conclusion
While relatively easy to perform, this method is of questionable value for generating smoke toxicity data for
use in fire hazard analysis. It is also necessary to verify its use as a screening tool against real-scale fire test
data. The absence of animal-exposure data means that smoke extreme or unusual toxic potency is not
identified.
6.4 Closed flask test (Israel)
6.4.1 Application
[8] [8]
This apparatus, used in the Israeli standard SI 755 , is for classification of building materials. SI 755:1998 ,
Section 2.6, addresses the total risk of gas toxicity.
6.4.2 Principle
As shown in Figure 4, this apparatus is a closed spherical bulb with a volume of at least 3 l. It has a tubular
extension, about 10 % of the total bulb volume, that is inserted into an oven heated to either 250 °C or 550 °C.
The former temperature is below nearly all auto-ignition temperatures; the latter is above the auto-ignition
−5
temperature for vapours from many materials. The test specimen is about 3 × 10 of the bulb volume. There
is no ignition source. The bulb volume contains excess air. Several gas samples are extracted from the bulb to
find the maximum concentration of each of six gases. The test that gives the highest gas concentration is
used.
6.4.3 Fire stage(s)
[35]
The fire stage(s) from ISO 19706:2007 , Table 1, are as follows:
⎯ 1.b, oxidative pyrolysis;
⎯ 2, well-ventilated flaming;
⎯ 3a, under-ventilated flaming.
6.4.4 Types of data
The reported data are the maximum concentrations of CO, HCN, NO , HCl, SO , and formaldehyde.
x 2
6.4.5 Presentation of results
The output is a simplified toxicity index similar to a fractional effective dose (FED) for the six gases. This is
presumably for incapacitation, since the coefficients for the gas concentrations are based on “critical
concentrations for escape during fires.” The rationale for the equation to calculate the total risk of gas toxicity
in a 300 cm high room is not given.
10 © ISO 2007 – All rights reserved

Key
1 thermocouple 5 device for mixing gases
2 expansion balloon 6 test specimen
3 gas-sampling port 7 oven
4 collection vessel for effluent
Figure 4 — Schematic of the SI-755 smoke toxicity apparatus

6.4.6 Apparatus assessment
6.4.6.1 Advantages
The method uses an inexpensive, easy-to-operate apparatus and a small sample.
6.4.6.2 Disadvantages
The higher temperature should produce the highest effluent concentrations. However, results for different
materials might not be comparable since ignition is not controlled. If ignition occurs in the tubular extension,
the combustion would become vitiated quickly. Manual mixing of air within the bulb is not well defined. Poor
mixing can result in stratification, poor sampling, oxygen depletion and the condensation of certain gases. The
small sample might not be representative of a non-homogeneous material or product. There is a small balloon,
presumably to contain a pressure rise if auto-ignition occurs. It is not clear whether this is sufficient to prevent
leakage. The concentrations of the gases depend on the bulb volume, which is not fully defined. In common
with many physical fire models, no indication is given about the rate of burning, so highly fire-retarded
materials can be forced to burn at the same rate as materials without any fire retardants. Therefore, additional
data input on burning rates at different fire stages is required for fire safety engineering calculations.
6.4.6.3 Repeatability and reproducibility
There are no known published reports of an inter-laboratory evaluation.
6.4.7 Toxicological results
6.4.7.1 Advantages
The calculation of the index is simple and enables estimation of the relative importance of the six gases in
affecting escape.
6.4.7.2 Disadvantages
The realism of sample exposure is questionable due to its small size and the shape of the tubular extension.
The coefficients for the toxicity index calculation are not current. It can be necessary to include additional
gases, as the prescribed set of gases to be measured might be insufficient to estimate lethal toxic potency.
The basis for the risk equation is unclear and does not appear to be related to actual risk.
6.4.8 Miscellaneous
The test does not involve laboratory animals. The gases may be measured by any suitable method.
6.4.9 Validation
No data are provided.
6.4.10 Conclusion
The method has questionable value for screening the toxic potency of materials and products, since some test
specimens flame and others pyrolyse. The pyrolysis or combustion gas concentrations are not likely to be
representative of room-scale effluent.
6.5 NES 713 (United Kingdom)
6.5.1 Application
[9]
This apparatus described in NES 713 is designed to provide values of a toxicity index for use in short-listing
materials and end products for warship marine use.
6.5.2 Principle
A photograph of this closed cabinet test is shown in Figure 5. A specimen of size chosen to provide optimal
analytical precision (typically 1 g to 2 g) is exposed to a premixed Bunsen burner flame. The burner is turned
off after the specimen has burned to completion, and the atmosphere is mixed with a fan before being
sampled for gas measurement.
6.5.3 Fire stage(s)
[35]
The fire stage from ISO 19706:2007 , Table 1, is 2, well-ventilated flaming. However, this might not relate to
a real fire, as the burner is actually a premixed blow-torch-type flame at about 850 °C and not a free-burning
fire.
12 © ISO 2007 – All rights reserved

Key
1 mixing fan
2 gas-sampling tube
3 gas burner
4 specimen support
Figure 5 — Photograph of the NES 713 apparatus

6.5.4 Types of data
The standard procedure includes measurement of CO, CO , formaldehyde, NO , HCN, acrylonitrile, phosgene,
2 x
SO , H S, HCI, NH , HF, HBr and phenol. Corrections are applied for the concentrations of CO, CO and NO
2 2 3 2 x
produced by the gas flame alone burning for the same period as the test specimen.
6.5.5 Presentation of results
The output is an FED-like toxicity index for the 14 gases. The weightings of the gases are the concentrations
considered nominally lethal to a man for a 30 min exposure. The index is calculated for 1 g of sample or for
1 m of wire or cable.
6.5.6 Apparatus assessment
6.5.6.1 Advantages
The apparatus is simple to use. The combustion period is short, so the combustion environment is stable
throughout. The test specimen can be a reasonable representation of a finished product.
6.5.6.2 Disadvantages
Nearly all materials and end products are composed of multiple components. These can gasify at different
times during burning. The test specimen is immersed in a pre-mixed gas flame and is burned to completion
but this might not produce gases representative of the combustion of the sample in real fire conditions. The
test specimen is small and is immersed in the test flame and combusted from all sides and to completion. In
common with many physical fire models, no indication is given about the rate of burning, so highly fire-
retarded materials can be forced to burn at the same rate as materials without any fire retardants. Therefore,
additional data input on burning rates at different fire stages is required for fire safety engineering calculations.
Colorimetric tubes are not a reliable measurement technique for combustion products due to possible
interferences.
6.5.6.3 Repeatability and reproducibility
There are no reported results of an inter-laboratory evaluation. However, repeatability is reported to be
reasonably good, since the specimen is relatively small, is completely immersed in the gas flame and is
burned to completion.
6.5.7 Toxicological results
6.5.7.1 Advantages
The initial conditions are few and well prescribed.
6.5.7.2 Disadvantages
The gases selected are those considered by the originators of the standard to be a hazard in warship fires and
the levels specified are considered to be relevant when the standard was reviewed in 2000. The coefficients
for the toxicity index calculation are not current. The basis for the index equation is unclear.
6.5.8 Miscellaneous
No animals are exposed in the test, nor is the apparatus compatible with such an addition.
6.5.9 Validation
There are no reported comparisons of toxic gas generation with data from real-scale fire tests.
6.5.10 Conclusion
While relatively easy to perform, this method is of questionable value for generating smoke toxicity data for
use in fire hazard analysis because of its unsatisfactory fire model. Its use as a screening tool has not been
verified against real-scale fire test data as it is intended that short-listed materials would be retested with more
relevant tests. The small sample size limits the use for evaluation of finished products. The absence of animal-
exposure data means that smoke extreme or unusual toxic potency will not be identified.
6.6 Japanese toxicity test
6.6.1 Application
This apparatus described in Reference [10] is designed to obtain toxic potency data for slow burning building
and furnishing materials. It is the basis for method 1231 of the Japanese Ministry of Construction.
14 © ISO 2007 – All rights reserved

6.6.2 Principle
This is a two-chamber apparatus (see Figure 6). There is a slow flow of air through the combustion chamber
in order to keep the oxygen in the mouse-exposure chamber above 16 %. The samples, 220 mm square and
not more than 15 mm thick, are exposed in moderately vitiated air to convective and radiative heating. The
exhaust gas is introduced into an animal-exposure chamber in which there are 8 rotary cages, each
containing a mouse. The movement of the mice is monitored and reflected to the evaluation of toxicity.
6.6.3 Fire stage(s)
[35]
The fire stage from ISO 19706:2007 , Table 1, is 3.a, small, under-ventilated fire.
6.6.4 Types of data
The standard procedure is to measure the times to incapacitation of the mice. In addition, gas samples can be
extracted for external analysis. Following the test, blood samples can be extracted for analysis.
6.6.5 Presentation of results
The reported information is the incapacitation times of the eight mice. Usually, 15 min is the maximum.
6.6.6 Apparatus assessment
6.6.6.1 Advantages
The single combustion test condition is well defined. The method could be adapted to produce data enabling
calculation of yields of toxicants.
6.6.6.2 Disadvantages
The test is limited to a single fire stage.
6.6.6.3 Repeatability and reproducibility
[11]
In a 4-laboratory examination of the method for six materials , the inter-laboratory standard deviation of the
times to incapacitation of the mice was under 15 %. The agreement of duplicate tests within each laboratory
was within 5 %.
6.6.7 Toxicological results
6.6.7.1 Advantages
The test provides a direct measure of the incapacitation capability of the smoke. It can identify instances of
extreme and unusual smoke toxic potency.
6.6.7.2 Disadvantages
The test requires specialized equipment for animal exposure.
6.6.8 Miscellaneous
This is primarily an animal-exposure test with limited chemical instrumentation. However, additional analytical
instrumentation can be added with little interference with the standard method.
Key
1 mixing box 8 secondary air
2 stirrer 9 pumps
3 furnace 10 heater and stirrer
4 mixer 11 rotary cages
5 flowmeter(s) 12 exhaust
6 LP gas 13 animal-exposure box
7 primary air
Figure 6 — Schematic of the Japanese smoke-toxicity apparatus

6.6.9 Validation
No comparison against real-scale fire tests has been published.
6.6.10 Conclusion
This method is useful for screening the incapacitation potency of smoke from various products, to the extent
that the mouse response to the effluent is similar to human response. The combustion conditions simulate
only a single fire stage. If the potency varies with the degree of vitiation, it is necessary to perform multiple
tests.
16 © ISO 2007 – All rights reserved

6.7 Cone Calorimeter (International)
6.7.1 Application
[12] [13] [14]
The physical fire model in this apparatus is the one used in ISO 5660-1 , ASTM E 1354 , NFPA 271
[15]
and NFPA 272 . The apparatus is designed to generate measurement of the rate of heat release (RHR) and
smoke from sampl
...