ISO/TR 22099:2025

Technical
Report
ISO/TR 22099
First edition
Application examples for using
2025-05
reaction-to-fire test data for fire
safety engineering
Exemples d'applications de l'utilisation des données des essais de
réaction au feu pour l'ingénierie de la sécurité incendie
Reference number
© ISO 2025
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms, definitions and abbreviated terms . 1
3.1 Abbreviated terms .1
4 Example 1: Using mass loss data for a single burning item as input for computational
fluid dynamics (CFD) calculations . 1
4.1 Introduction .1
4.2 Experimental investigations .1
4.3 Numerical investigations .4
4.4 Discussion of the results .7
4.5 Conclusions .8
5 Example 2: Using cone calorimeter and LIFT apparatus data as input for flame spread
calculations. 9
5.1 Reaction to fire tests for flame spread calculations .9
5.1.1 Overview .9
5.1.2 Cone calorimeter tests .9
5.1.3 Lateral ignition and flame transport (LIFT) test .10
5.2 Methods to derive material properties for flame spread .11
5.2.1 Parameters calculated from cone calorimeter tests .11
5.2.2 Parameters measured by LIFT tests .17
5.3 An example of derivation of properties .19
5.3.1 Specimen .19
[19]
5.3.2 Properties derived from Cone calorimeter tests .19
5.3.3 Properties derived from LIFT tests . 22
5.4 Prediction of flame spread over lining materials . 25
5.4.1 Overview . 25
5.4.2 Schematics of model . 25
5.4.3 Ignition of wall surface . 26
5.4.4 Heat release rate .27
5.4.5 Upward flame spread .27
5.4.6 Lateral and downward flame spread . 28
5.4.7 Smoke layer temperature . 28
5.4.8 Calculation procedure . 28
5.5 Comparison with an experiment . 28
5.5.1 Experimental conditions and procedure . 28
5.5.2 Experimental results . 30
5.5.3 Calculation conditions .32
5.5.4 Calculation results . 33
5.6 Summary of example 2 . 35
6 Example 3: Using mass loss data for a single burning item as input for zone model
calculations including fuel response effects .35
6.1 General . 35
6.2 Experimental investigations . 36
6.3 Numerical investigations .37
6.4 Discussion of the results .37
Bibliography . 41

iii
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.
The procedures used to develop this document and those intended for its further maintenance are described
in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the different types
of ISO documents should be noted. This document was drafted in accordance with the editorial rules of the
ISO/IEC Directives, Part 2 (see www.iso.org/directives).
ISO draws attention to the possibility that the implementation of this document may involve the use of (a)
patent(s). ISO takes no position concerning the evidence, validity or applicability of any claimed patent
rights in respect thereof. As of the date of publication of this document, ISO had not received notice of (a)
patent(s) which may be required to implement this document. However, implementers are cautioned that
this may not represent the latest information, which may be obtained from the patent database available at
www.iso.org/patents. ISO shall not be held responsible for identifying any or all such patent rights.
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and expressions
related to conformity assessment, as well as information about ISO’s adherence to the World Trade
Organization (WTO) principles in the Technical Barriers to Trade (TBT), see www.iso.org/iso/foreword.html.
This document was prepared by Technical Committee ISO/TC 92, Fire safety, Subcommittee SC 1, Fire
initiation and growth.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www.iso.org/members.html.

iv
Introduction
In recent years, fire test data has been increasingly used as input for fire safety engineering (FSE)
calculations. This document provides three different examples of how fire test data can be used for FSE.
Guidance on how data can be derived from reaction-to-fire tests is given in ISO/TR 17252. Background on
reaction-to-fire tests and limitations of data derived from these tests is given in ISO/TS 3814.

v
Technical Report ISO/TR 22099:2025(en)
Application examples for using reaction-to-fire test data for
fire safety engineering
1 Scope
This document provides three examples of the use of reaction-to-fire test data for fire safety engineering (FSE).
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content constitutes
requirements 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
3 Terms, definitions and abbreviated terms
For the purposes of this document, the terms and definitions given in ISO 13943 apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org/
3.1 Abbreviated terms
CFD computational fluid dynamics
FSE fire safety engineering
LIFT lateral ignition and flame transport
MLR mass loss rate
4 Example 1: Using mass loss data for a single burning item as input for
computational fluid dynamics (CFD) calculations
4.1 Introduction
The use of data from open calorimetry or mass loss measurements for a single burning item is used in FSE to
predict, for example, the temperature development, smoke development or the further spread of fire.
This example provides insight in the use of mass loss data for CFD calculations and its limitations.
4.2 Experimental investigations
Three tests with a modern upholstered chair were performed, see Table 1. In all three experiments, the
modern upholstered chair was ignited with a 100-gr paper cushion which was made similar to the standard
train seat ignition source (according to DIN 5510).

Figure 1 shows the three test set-ups and Figure 2 shows the measured mass loss rate (MLR) for all three
experiments. In experiment 1, only the mass loss was measured. In experiments 2 and 3, the mass loss,
temperatures in the room and smoke gas composition were measured.
Table 1 — Overview of experiments with modern upholstered chair
Experiment 1 Experiment 2 Experiment 3
Upholstered chair in big hall Upholstered chair in little room – no Upholstered chair in little room –
front wall with door opening
a)  Experiment 1 b)  Experiment 2 c)  Experiment 3
Figure 1 — Test set-up of the three experiments
a) b)
Key
X time (s)
Y1 mass loss (kg)
Y2 mass loss rate (kg/s)
1 experiment 1
2 experiment 2
3 experiment 3
Figure 2 — Resulting mass losses and mass loss rates
Figure 3 shows the fire development for experiment 2 after 5 minutes and after 9 minutes. Figure 4 shows
the fire development in experiment 3 after 5 minutes and after 8 minutes. After 5 minutes, the fire of

experiment 3 is more developed than in experiment 2. After 9 minutes, in experiment 2, a short period
occurs where the smoke layer burns. In experiment 3, flash-over occurred in the small room after 8 minutes.
This corresponds to the measured temperatures which are shown in Figure 5. Temperatures were measured
under the ceiling at four locations as shown in Figure 6. The average temperatures of the four thermocouples
over time are shown in Figure 5 for experiments 2 and 3. Temperatures under the ceiling exceed in both cases
600 °C. In experiment 3, the maximum temperatures (approximately 900 °C) are significantly higher than in
experiment 2 (approximately 680 °C). The slope of the temperature rise is also steeper for experiment 3 and
the maximum is reached earlier than in experiment 2. Additional temperatures were measured in front of
the chair which are not discussed here.
a)  5 minutes after ignition b)  9 minutes after ignition
Figure 3 — Observations of fire development in experiment 2
a)  5 minutes after ignition b)  8 minutes after ignition
Figure 4 — Observations of fire development in experiment 3

Key
X time (s)
Y temperature (°C)
1 experiment 2
2 experiment 3
Figure 5 — Measured temperatures in experiments 2 and 3
Figure 6 — Locations of thermocouples
4.3 Numerical investigations
The small room configuration was simulated with the CFD software FDS 8,9 (Fire Dynamics Simulator,
version 6.01). The small room and the surroundings were modelled with a cubic grid with 0,05 m cell size
(side length). The small room has the dimensions of 2,5 m × 2,5 m × 2,5 m. The upholstered chair is modelled
[15] 3
under the assumption the material is polyurethane foam with a density of 38 kg/m , a specific heat
of 1, a conductivity of 0,05 and a thickness of the material of 0,2 m. The pyrolysis gas is assumed to be
C6.3H7.1NO2.1, which burns with a soot yield of 0,1. The seat of the upholstered chair is assumed to be the
burning area (0,49 m ). The heat release rate over time curves are calculated from the measured mass loss

[11]
with an assumed heat of combustion of 25 000 MJ/kg for polyurethane foam. The walls of the small room
are assumed to be gypsum plaster with a thickness of 0,1 m. The gypsum plaster is modelled with a density
3 . . [12]
of 1 440 kg/m , a conductivity of 0,48 W/(m K) and a specific heat of 0,84 kJ/(kg K ) .
As can be seen in Figure 7, if the measured mass loss is used to model experiments 2 and 3, it is possible
to predict the fire development in the small rooms satisfactorily. Both fire developments in the room with
door opening and in the room without front wall can be predicted well, as shown in Figure 7, 5 minutes
after ignition. Also, the smoke layer ignition after 8 minutes in the room without front wall and the flash-
over after 9 minutes in the room with door opening are captured well by the calculations. The results of the
calculations are shown without smoke to make the effects better visible.
a)  5 minutes b)  5 minutes
c)  9 minutes d)  8 minutes
Key
> 200 (kW/m )
Figure 7 — Predictions of fire development using mass loss rate from same experiment
In comparison to the calculations above, the small room without front wall was calculated again; the only
difference to the calculations above is the use of the mass loss of the upholstered chair from the experiment
in the hall (experiment 1). Grid size, materials and combustion model are the same as above.
In Figure 8, the predicted fire development for this case between 9 minutes and 10 minutes is shown. The
ignition of the smoke layer does not occur in this model set-up.

a)  9 minutes b)  9 minutes 40 seconds c)  10 minutes
Key
> 200 (kW/m )
Figure 8 — Predictions of fire development using mass loss rate from experiment 1 (open hall)
Predicted temperatures in the small rooms under the ceiling correspond to this result, i.e. no ignition of
the smoke layer. In Figure 9, the averaged temperatures over the four thermocouples under the ceiling are
plotted over time for the different calculated cases: small room with front wall and door and use of the MLR
of corresponding experiment 3 (dotted light grey line black line); small room without front wall and use of
MLR from corresponding experiment 2 (dotted grey line) and small room without front wall and use of MLR
from experiment 1 (dotted black line). Experiment 1 was performed in a big hall with no walls or ceiling
nearby. Predicted temperatures for experiments 2 and 3 using the corresponding measured MLR exceed
600 °C under the ceiling. That corresponds with the smoke layer ignition in experiment 2 and the flash-over
in experiment 3. In the third case where temperatures under the ceiling in the small room without front wall
were predicted using the MLR from experiment 1, which was performed in a big hall, the temperatures do not
exceed 600 °C. This corresponds with the prediction shown in Figure 8 that the smoke layer does not ignite.

Key
X time (s)
Y temperature (°C)
1 MLR, experiment 1
2 MLR, experiment 2
3 MLR, experiment 3
Figure 9 — Predicted temperatures under the ceiling in small room, averaged
4.4 Discussion of the results
Experimental and numerical investigations of the first burning item of the large scale experiments of
the modern living rooms showed that identical chairs burn with significant different mass losses under
different room conditions. The measured mass losses and the derived MLRs (see Figure 2) show significant
differences, although the identical chair and the same ignition method was used. The measured temperatures
which were only recorded for experiments 2 and 3 also show significant differences (see Figure 5) which
corresponds to the observations of the fire development (see Figures 3 and 4). In the small room with front
wall and door, the fire development was quicker and the temperatures higher – only in this case flash-over
in the room occurred. In the small room without front wall, the fire development was slower and only the
ceiling smoke layer ignited for several seconds.
The measured mass loss, more precisely the derived MLR, was used in the numerical models to predict the
fire development in the small rooms and the temperatures in the room under the ceiling. On the one hand,
when the MLR of the identical experiment was used for the numerical predictions, the fire development
could be predicted very accurately. The predicted temperatures under the ceiling were satisfactorily similar
to the measured values in shape and magnitude, although, in the case of the small room with front wall and
door, the predicted temperatures showed a second peak that was not measured in the experiments as can be
seen in Figure 10. Figure 10 shows the predicted and measured averaged temperatures under the ceiling for
experiments 2 and 3 .
Key
X time (in s)
Y temperature (in °C)
1 experiment 2
2 experiment 3
3 MLR, experiment 2
4 MLR, experiment 3
Figure 10 — Predicted and measured temperatures (averaged) under the ceiling in small room
On the other hand, when the MLR of experiment 1 (hall) was used to predict the fire development in the small
room without front wall, the fire development as well as the predicted temperatures are quite different to
the observed fire development and measured values. Figure 6 shows that in this case no smoke layer ignition
is predicted, although it occurs in the experiment and could be predicted when the MLR from the identical
experiment is used. Also, the predicted temperatures under the ceiling are significantly lower than the
measured ones. They stay below 600 °C, which is in accordance to the predictions of the fire development.
4.5 Conclusions
The modern upholstered chair was tested in three different settings: in a hall, a small room without front
wall and the same small room but with front wall and door opening. The measured mass losses of the three
experiments showed significant differences. These mass losses were used as input in the numerical models
of the upholstered chair in the small rooms to see how well the fire development and temperatures could be
predicted. Predictions were satisfying when the mass loss of the corresponding experiment was used. In one
case, the measured mass loss of the first experiment in the hall was used to predict the fire development and
the temperatures in the small room without front wall. In this case, the smoke layer ignition could not be
predicted and the temperature in the room was distinctly under the measured temperatures. As measured
heat release rates – often measured in open calorimetry – are often used as input in numerical models, for
instance for the first burning item, it is important to realize that even with the same chair tested with the
same ignition method the results can differ and influence the model predictions.

5 Example 2: Using cone calorimeter and LIFT apparatus data as input for flame
spread calculations
5.1 Reaction to fire tests for flame spread calculations
5.1.1 Overview
[7]
The necessary material properties for flame spread calculations are ignition temperature, thermal
[2]
inertia, heat release rate, total heat release obtained by cone calorimeter test and flame spread parameter
[3][6]
obtained by lateral ignition and flame transport (LIFT) test .
5.1.2 Cone calorimeter tests
In the ISO 5660-1 cone calorimeter test, cone heater is used to heat up and burn a specimen of 99 mm , as
shown in Figure 11. To simulate the heating and burning of a wall, it is desirable to heat the specimen in
vertical orientation following to ISO 5660-1:2015, Annex E. Ignition time and heat release rate are measured.
Heating intensity is varied in the range of minimum heat flux for ignition and up to flux corresponding with
direct flame impingement. The ignition time and heat release rates are correlated with heating intensity.
Key
1 cone heater
2 sample holder
Figure 11 — View of a cone calorimeter test in vertical orientation
Several measurements are necessary varying heating intensity. The typical time histories of heat release
rate are shown in Figure 12 for various heating intensities. At least one measurement is continued until
burnout in order to measure total heat release.

Key
X time (s)
Y heat release rate (kW/m )
1 intensely heated
2 mildly heated
3 weakly heated
4 total heat release (area under heat release rate curve) (kJ/m )
Figure 12 — Typical time histories of heat release rate under various heating intensities
5.1.3 Lateral ignition and flame transport (LIFT) test
Lateral flame transport is measured by the LIFT apparatus using a rectangular specimen. As shown in
2 2
Figure 13, the specimen is heated by non-uniform intensity in the range of 2 kW/m to 50 kW/m . After the
specimen is ignited at one end, the lateral spread rate is measured by eye. By carrying out one experiment,
arrival time to each point and the maximum distance of spread are measured as shown in Figure 14.
Figure 13 — LIFT test
Key
X horizontal distance from ignited end (mm)
Y arrival time (s)
1 ignition
2 no more spread
3 maximum distance of spread
Figure 14 — Arrival time and maximum distance of spread
5.2 Methods to derive material properties for flame spread
5.2.1 Parameters calculated from cone calorimeter tests
5.2.1.1 Ignition temperature
Repeating the cone calorimeter tests for various heating intensity, correlation between ignition time and
heating intensity is obtained, as shown in Figure 15. The critical heat flux for ignition is determined as the
minimum heating intensity that causes ignition. Surface temperature is calculated by considering the heat
[8]
balance at material surface as shown in Figure 16. Heating intensity is calculated using Formula (1):
qq=−()1 εσ+−()TT +−h ()TT (1)
ee sc∞∞s
where
q is the heating intensity (kW/m );
e
ε is the emissivity of the surface (= 1);
−11 2. 4
σ is the Stefan-Boltzmann constant [= 5,67 × 10 kW/(m K )];
h is the convective heat transfer coefficient between specimen surface and ambient air [= 0,013 kW/
c
2. [9]
(m K)] ;
T is the surface temperature (K);
s
T is the ambient temperature (K).
∞
When the heating intensity is equal to the critical heat flux for ignition, the surface temperature approaches
to ignition temperature at a steady state.
4 4
qq=−()1 εσ+−()TT +−h ()TT (2)
o,ig o,ig ig ∞∞cig
where
q is the critical heat flux for ignition (kW/m );
o,ig
T is the ignition temperature (K).
ig
Approximating that the emissivity of surface ε is unity, the ignition temperature T is determined so that
ig
that the error of Formula (2), R(T ), becomes zero.
ig
4 4
RT()≡−σ()TT +−h ()TT −q (3)
ig ig ∞∞cigo,ig
As the explicit solution of Formula (3) is not given, ignition temperature is calculated numerically by
*
[10]
Newton’s method. Let the assumed value of T be T , the Formula (3), would be
ig ig
* *44 *
RT()=−σ()TT +−h ()TT −q (4)
ig ig ∞∞cig o,ig
*
Linearizing Formula (4) around T , the following formula applies,
ig
dR
* * * *3 *
RT()=+RT() ()TT−=RT()++(4σT h )(TT− ) (5)
ig ig ig ig ig ig ccigig
dT
*
ig
TT=
ig ig
Calculating T to satisfy R(T ) = 0 in Formula (5), the corrected value of ignition temperature is
ig ig
*
RT()
ig
*
TT=− (6)
ig ig
*3
4σT +h
ig c
*
Using T calculated by Formula (6) as a new estimation of T , the whole process is repeated. The converged
ig ig
value is the solution to Formula (3).

Key
X heating intensity, q (kW/m )
e
Y ignition time, t (s)
ig
1 critical heat flux for ignition, q (kW/m )
o,ig
ignited
not ignited
Figure 15 — Ignition time versus heating intensity
Key
1 heated surface
convection, h ()TT− (kW/m )
cs ∞
4 4
re-radiation, εσ()TT− (kW/m )
s ∞
4 external radiation (heating intensity), q (kW/m )
e
5 reflection, (1 − ε)q (kW/m )
e
6 conduction
Figure 16 — Surface heat balance at ignition

5.2.1.2 Thermal inertia
As long as the material is thick enough so that the rear surface temperature is unchanged up to ignition
[11]
time, the temperature profile and conduction heat flux are approximated by those in infinitesimal body.
The ignition temperature and time are correlated with heating intensity, as in Formula (7):
εq
e
= (7)
h()TT−
k k
s ∞
1−exp( tt)erfc()
ig ig
ρρc c
where
.
k is the thermal conductivity (kW/(m K));
2.
h is the overall heat transfer coefficient (kW/(m K));
t is the ignition time (s);
ig
ρ is the density (kg/m );
.
c is the specific heat (kJ/(kg K)).
[12]
Hasemi et al. showed that the above relationship can be approximated by Formula (8) :
1 11, 8ε
*
= ()qq− (8)
eo,ig
tTkcρ ()−T
ig ig ∞
*
In this notation, intercept on q -axis, q , can be the critical heat flux for ignition. However, it differs from
e o,ig
the actual critical heat flux for ignition as the assumption of thermally thick wall does not apply to the
conditions around critical heat flux.
As shown in Figure 17, the inverse of square root of ignition time is plotted against heating intensity. The
slope of the regression line, S, is as in Formula (9):
11, 8ε
S= (9)
kcρ ()TT−
ig ∞
which results in Formula (10) to calculate thermal inertia:
11, 8ε 11,18 1
kcρ = ≈ (10)
()TT− ST()−TS
ig ∞∞ig
Key
X heating intensity, q (kW/m )
e
−1/2 −1/2
Y inverse of square root of ignition time, t (s )
ig
1 critical heat flux for ignition, q (kW/m )
o,ig
*
2 2
apparent critical heat flux for ignition, q (kW/m )
o,ig
ignited
not ignited
Figure 17 — Derivation of thermal inertia from ignition time data under various heating intensities
5.2.1.3 Heat release rate per unit area
The heat release rate after ignition is hardly constant but changes with time. The changes can be
[13] [14]
approximated by a constant value or by exponential decay after a peak. In this example, heat release
rate is approximated by the average over 180 seconds after ignition, as per Formulae (11) to (14) (see also
Figure 18):
00() 
ig

q= qt() 
180 ig b

0 ()tt<

b
t +180
ig
qq= ()tdt (12)
180 m
∫
t
ig
M
t = (13)
b
q
∞
M= qt()dt (14)
m
∫
where
q is the heat release rate per unit area (kW/m );
q is the average heat release rate per unit area after 180 seconds after ignition (kW/m );
q is the measured heat release rate per unit area (kW/m );
m
M is the total heat release per unit area (kJ/m );
t is the burning duration (s).
b
The similar approach was adopted by a research project to predict material flammability used in upholstered
[15]
furniture.
Key
X time (s)
Y heat release rate (kW/m )
1 measured heat release rate (kW/m )
2 total heat release (kW/m )
3 ignition
Figure 18 — Approximation of heat release rate

Considering the net heat flux absorbed by the surface, heating intensity q and average heat release rate q
e 180
[16]
are correlated by Formula (15) :
ΔH
qq=−()σTq+ (15)
180 fl ig e
L
where
ΔH is the heat of combustion (kJ/kg);
L is the heat of gasification (kJ/kg);
q is the flame heat flux to material surface (kW/m );
fl
4 2
is the re-radiation from material surface (kW/m );
σT
ig
q is the heating intensity (kW/m ).
e
Measuring the average heat release rate under various heating intensity and plot them against heating
intensity as shown in Figure 19, the slope of the regression line would be ΔH/L. By extending the regression
line to find the horizontal intercept q (<0), the flame heat flux would be as in Formula (16):
qT=−σ q (16)
fl ig 0
Key
X heating intensity, q (kW/m )
e
Y average heat release rate, q (kW/m )
1 regression (slope = ΔH/L)
measured peak heat release rate (kW/m )
Figure 19 — Correlation of average heat release rate with heating intensity
5.2.2 Parameters measured by LIFT tests
5.2.2.1 Flame spread parameters

[17]
The lateral flame spread rate V (m/s) is calculated by Formulae (17) to (20) :
f
=−Ch[ ()TT −qx()Ft()] (17)
ig ∞ ef
V
f
π 1
C= (18)
2 2
h
δ q
ff
kcρ
q
o,ig
h= (19)
TT−
ig ∞
Tt()−T
h tth
s ∞
Ft()≡ =−1 exp( )erfc() (20)
TT()∞− kcρρkc
s ∞
where
1/2 3/2
C is the flame heat transfer modulus (s m / kW);
F is the correction factor for heating intensity (-)
2.
h is the overall heat transfer coefficient at ignition (kW/m K);
q is the heat flux from flame to preheat zone (kW/m );
f
V is the lateral flame spread rate (m/s);
f
δ is the length of preheat zone (m).
f
Using a result of LIFT test, the inverse of the square root of flame spread rate at each point is correlated
with heating intensity as shown in Figure 20. The slope of the regression line yields the flame heat transfer
modulus −C.
[18]
On the other hand, flame spread rate is correlated with surface temperature instead of heating intensity.
For calculations of flame spread under general conditions, Formula (21) is convenient.
Φ
V = (21)
f
kcρ ()TT−
ig s
where
2 3
Φ is the flame spread parameter (kW /m );
2. 4. 2
kρc is the thermal inertia of material (kW s/m K );
T is the ignition temperature (K);
ig
T is the surface temperature of preheated zone (K).
s
The flame spread parameter Φ and flame heat transfer modulus C can be converted by Formula (22):
kcρ
Φ = (22)
hC
Key
X effective heating intensity, q F (kW/m )
e
− 1/2 1/2 1/2
Y inverse of square root of flame spread rate, V (s /m )
f
1 regression (slope −C)
2 minimum heat flux for flame spread (kW/m )
3 critical heat flux for ignition (kW/m )
measured flame spread rate
Figure 20 — Regression for flame heat transfer modulus
5.2.2.2 Minimum heat flux and temperature for flame spread
The heat flux at the position of maximum flame spread x (m) corresponds with the limit of spread.
max
The heating intensity at x is the minimum heat flux for flame spread q (kW/m ). In Figure 20, the
max s,min
minimum heat flux for flame spread corresponds with the lowest data of measured flame spread rate. The
minimum surface temperature for spread T (K) is calculated by the same procedure used to calculate
s,min
ignition temperature from critical heat flux for ignition. Using q instead of q in Formula (3), T is
s,min o,ig s,min
calculated.
5.3 An example of derivation of properties
5.3.1 Specimen
Plywood for structural use (12 mm thickness) was selected as specimen. The density was 500 kg/m on
average.
[19]
5.3.2 Properties derived from Cone calorimeter tests
Cone calorimeter tests were carried out 7 times, changing heating intensity. The testing conditions and
schematics of test results are shown in Table 2. No ignition took place in two of the tests. One of the tests at
50 kW/m was continued up to burnout in order to measure total heat release.

Table 2 — Specimens for Cone calorimeter tests and schematics of results
No. Mass Thick- Density, Ρ Heating Ignition Average heat release Total heat
ness intensity, q time, t rate during 3 minutes release, M
e ig
after ignition, q
3 2 2 2
(g) (mm) (kg/m ) (kW/m ) (s) (kW/m ) (MJ/m )
1 58,00 11,89 497,7 15,0 No ignition — —
2 59,45 11,81 513,6 16,0 No ignition — —
3 55,53 11,88 476,9 17,0 286 59,7 —
4 59,22 11,88 508,6 25,0 72 66,3 —
5 59,37 11,84 511,6 35,0 39 101,8 —
6 55,68 11,80 481,5 50,0 16 117,5 —
7 58,98 11,82 509,1 50,0 26 115,9 66,4
average — — 500,0 — — — —
NOTE Remark: specimen size 100 × 100 mm.
The critical heat flux for ignition was estimated as 16,5 kW/m . The ignition temperature was calculated by
Formula (3) as:
−11 4 4
RT()=×56,(710 TT−+293 ),0 013()−−293 16,50=
ig ig ig
which results in T = 677,2K (= 404 °C).
ig
The results of measurements of ignition time were plotted against heating intensity in Figure 21. The slope S
of regression line was 0,004 74. Using the value in Formula (10), the thermal inertia was calculated as
11, 81 11, 8 1
kcρ = = =0,649 .
()TT− S (,677 2−293),0 00474
ig ∞
Key
X heating intensity, q (kW/m )
e
−1/2 −1/2
Y inverse of square root of ignition time, t (s )
ig
1 critical heat flux for ignition (kW/m )

2 regression, slope = 0,004 74
ignited
not ignited
Figure 21 — Correlation of ignition time with heating intensity for 12 mm thick plywood
In addition, ignition time at 60 kW/m of heating intensity is required in the calculation of flame spread
over surface. To obtain the value, the regression line was extended to 60 kW/m and the ignition time was
calculated. Using the values shown in Figure 21,
=−0,(00474 q 23,)3
e
t
ig
Thus, the ignition time for 60 kW/m of heating intensity was calculated as
t = =13,4
ig,60
{,0 00474(,60−233)}
The measured results of heat release rates are shown in Figure 22. In all the ignited tests, the peak value
of heat release rate appeared right after ignition. After that, heat release rates were decreased because of
surface charring. As the burning proceeds to the rear surface of material, the second peak appears due to
the collapse of material. In one of the tests at 50 kW/m , measurement was continued up to burnout. The
total heat release was as defined by Formula (14).
∞
Mq==()tdt 66,4MJ/m
m
∫
Key
X time (s)
Y heat release rate (kW/m )
1 heating intensity 15 kW/m
2 heating intensity 16 kW/m
3 heating intensity 17 kW/m
4 heating intensity 25 kW/m
5 heating intensity 35 kW/m
6 heating intensity 50 kW/m
Figure 22 — Measured results of heat release rates by Cone calorimeter
The average heat release rates were calculated by Formula (12) and plotted against heating intensity as shown
in Figure 23. Referring to Formula (15), the fraction of heat of combustion to heat of decomposition was
ΔH
=18, 1 .
L
Applying the value of the horizontal intercept to Formula (16), the flame heat flux was obtained as
4 −11 4
qT=−σ q =×56,(710 ×−677 −=15,)6273, .
fl ig 0
Key
X heating intensity, q (kW/m )
e
Y average heat release rate after 180 seconds since ignition (kW/m )
1 regression, slope = 1,81
2 intercept, q = −15,6 (kW/m )
measured
Figure 23 — Average heat release rate as a function of heating intensity
5.3.3 Properties derived from LIFT tests
[20]
Three LIFT tests were carried out 3 times in the same testing condition. The density of specimens was
500 kg/m . Water content was 6,92 % by weight. The arrival time of the flame spread front is shown in
Figure 24. Dividing the arrival time by the measuring interval, the flame spread rate at each interval was
calculated as shown in Figure 25. By correlating the results with the effective heating intensity shown in
Figure 26, correlation was obtained as shown in Figure 27. By fitting a regression line to major part of the

data, the flame heat transfer modulus resulted in C = 4,04. As measured by cone calorimeter tests, thermal
inertia was kcρ =0,649 . The overall heat transfer coefficient was
q
o,ig 16,5
h= = =0,043 .
TT− 677−293
ig ∞
Using all these values in Formula (22),
kcρ 0,649
Φ == =0,602 .
2 2
hC 0,,043×404
The minimum heat flux for flame spread was extracted as the minimum value of heating intensity where
flame spread rate approached zero as
qq==()xq ()560mm =43, .
s,mine min e
Converting the heat flux to surface temperature,
−11 4 4
RT()=×56,(710 TT−+293 ),0 013()−−293 43, =0
s,mins,min s,min
which results in T = 460 K (= 187 °C).
s,min
Key
X distance from left end (mm)
Y arrival time of flame front (s)
run 1
run 2
run 3
Figure 24 — Arrival time of flame spread front in LIFT tests

Key
X distance from left end (mm)
Y flame spread rate (mm/s)
run 1
run 2
run 3
Figure 25 — Lateral flame spread rate in LIFT tests
Key
X distance from left end (mm)
Y effective heating intensity (kW/m )
1 heating intensity (kW/m )
run 1
run 2
run 3
Figure 26 — Effective heating intensity in LIFT tests

Key
X effective heating intensity (kW/m )
−1/2 1/2 1/2
Y inverse of square root of flame spread rate, V (s /m )
f
1 regression
2 minimum heat flux for flame spread (kW/m )
run 1
run 2
run 3
Figure 27 — Correlation of flame spread rate and effective heating intensity in LIFT tests
5.4 Prediction of flame spread over lining materials
5.4.1 Overview
As to the prediction of surface flame spread over lining materials, basic models have been proposed
...