NBS

Publi -

Reference cations
A lllDti MD7bbb

IMBSIR 82-2597

Heat Release Rate Properties of

Wood-Based Materials

July 1983

•QC
100 U S. DEPARTMENT OF COMMERCE

.156 NATIONAL BUREAU OF STANDARDS

02-2597
1983
 MATIOWJU. BUREAU
OF STAKDA-AQ6
UiiRART

AUG 8 1983

o-c^voO

iQO-i
NBSIR 82-2597
t • 0

HEAT RELEASE RATE PROPERTIES OF

WOOD-BASED MATERIALS

David L. Chamberlain, Research Associate

National Forest Products Association

U S. DEPARTMENT OF COMMERCE
National Bureau of Standards
National Engineering Laboratory
Center for Fire Research
Washington, DC 20234

July 1983

U.S. DEPARTMENT OF COMMERCE, Malcolm Baldrige, Secretary

NATIONAL BUREAU OF STANDARDS. Ernest Ambler. Director
 -
’

•
..
•
-
e

l ‘3 >J* /SOITAK
 TABLE OF CONTENTS

Page

List of Figures V

List of Tables vii

Abstract 1

1. Introduction 1

2. Objectives 3

3. Historical Review of Heat Release Rate Calorimetry 3

4. Operating Characteristics of Existing Heat Release Rate Calorimeters . 4

4.1 The Factory Mutual Construction Materials Calorimeter 4
4.2 The Parker/Long NBS-I Heat Release Rate Calorimeter 5
4.3 The NBS-I Rebuilt Calorimeter 5
4.4 The Ohio State-Smith Heat Release Rate Calorimeter 6
4.5 The SRI International Heat Release Rate Calorimeter 6
4.6 The Forest Products Laboratory (FPL) Heat Release Rate
Calorimeter 7

4.7 Recent Developments 7

5. Materials and Methods 8

5.1 Wood-Based Materials 8
5.2 Conditioning 9
5.3 Calorimeter Operation 9
5.4 Heat Release Rate Computation 11

6. Results and Discussion 12

6.1 Characteristic Heat Release Rate Curves 12
6.2 Heat Release Rates of Lumber and Other Wood Products 13
6.2.1 Lumber 13
6.2.2 Plywood 14
7.1 6.2.3 Particle Board 14
6.2.4 Acoustical Tile 15
6.2.5 Hardboard 15
6.2.6 General Effect of Density 15
6.2.7 Effects of Other Physical Properties 17
6.2.8 Effect of Fire-Retardant Treatment 18
6.2.9 Effect of Moisture Content 19
6.2.10 Effect of Imposed Radiant Heat Flux 21
6.2.11 Comparison of First 1-Minute and First 5-Minute Heat
Release Rates with the Corresponding Maximum Rates. . . 21

7. Comparison of Heat Release Properties Measured by Different

NBS Calorimeters 22
Correlation of NBS-I, NBS-I Rebuilt, and NBS-II Calorimeter
Values at 60 kW/m 2 Exposure Flux 22
7.1.1 NBS-I Rebuilt Calorimeter 23
7.1.2 NBS-II Calorimeter 23
7.1.3 Calorimeter Correlations 24
7.1.4 Conversion Equations 25

iii
 Page

7.2 Effect of Density as Measured in the NBS-I Rebuilt

and NBS-II Calorimeters 25
7.2.1 NBS-I Rebuilt Calorimeter 26
7.2.2 NBS-II Calorimeter 27
7.3 Effect of Imposed Radiant Flux as Measured in the
NBS-I Rebuilt and NBS-II Calorimeters 27
7.3.1 NBS-I Rebuilt Calorimeter 27
7.3.2 NBS-II Calorimeter 28

8. Conclusions 28

9. Acknowledgments 30

10. References 30

iv
 LIST OF FIGURES

Page

Figure 1. Diagram of FM Construction Materials Calorimeter 33

Figure 2. Diagram of NBS-I Calorimeter 34

Figure 3. Artist's View of NBS-I Calorimeter 35

Figure 4. Diagram of OSU Calorimeter 36

Figure 5. Drawing of SRI Calorimeter 37

Figure 6. Diagram of FPL Calorimeter 38

Figure 7. Photograph of FPL Calorimeter 39

Figure 8. Artist's View of NBS-II Calorimeter 40

Figure 9. Diagram of NBS Oxygen Consumption Calorimeter 41

Figure 10. Typical HRR Curves: Douglas Fir 42

Figure 11. Typical HRR Curves: Southern Pine 43

Figure 12. Typical HRR Curves: Medium-Density, Paper-Faced Hardboard. . .

44

Figure 13. Typical HRR Curves: Medium- Density Hardboard 45

Figure 14. Typical HRR Curves: High-Density Hardboard 46

Figure 15a-d. Heat Release Rate Data Comparison of Materials Conditioned

at 70°F and 50% RH, Exposed at 60 kW/m 2 47

Figure 16a and 16b. Effect of Density on HRR Properties:

NBS-I Calorimeter 51

Figure 17. Effect of Conditioning Room RH on HRR Properties 53

Figure 18. Effect of Radiant Flux on Heat Release Properties 54

Figure 19a. Comparison of First 1-Minute HRR with Maximum 1-Minute HRR . .
53

Figure 19b. Comparison of First 5-Minute HRR with Maximum 5-Minute HRR . .
56

Figure 20a-i. Heat Release Rate Properties of Wood Products

as Measured by Different NBS Calorimeters 57

Figure 21. Correlations Among NBS HRR Calorimeters 66

Figure 22. Correlations Among NBS HRR Calorimeters (Cont.) 67

68
Figure 23. Effect of Density on HRR Properties: NBS-I R Calorimeter ...

v
 LIST OF FIGURES (continued)

Page

Figure 24. Effect of Density on HRR Properties: NBS-II Calorimeter. . • •

69

Figure 25. Effect of Radiant Heat Flux on HRR Properties:

NBS-I Calorimeter 70

vi
 LIST OF TABLES

Page

Table 1. Heat Release Rate Calorimeters 71

Table 2. Physical Properties of Wood Products 72

Table 3. Format for Recording HRR Data 73

Table 4. Heat Release Properties of Wood-Based Materials 74

Table 5. Comparison of Heat Release Properties of Fire Retardant

Treated Douglas Fir Plywood and Lumber, Untreated
Controls, and Untreated Southern Pine Lumber 75

Table 6. Effect of Conditioning Environment on Heat Release Properties. .

76

Table 7. Effect of Radiant Flux Exposure on Heat Release Properties ... 77

Table 8 . Comparison of First 1-Minute and First 5-Minute HRR

with the Corresponding Maximum Rates 78

Table 9. Comparison of Mean Heat Release Characteristics of Different

NBS Calorimeters 79

Table 10. Relationships Among Heat Release Values Determined by

Different NBS Calorimeters 81

Table 11. Effect of Radiant Flux on Heat Release Properties Measured

in NBS-I Rebuilt and NBS-II Calorimeters 82

Table 12. Summary of Heat Release Rate Data 83

vii
 HEAT RELEASE RATE PROPERTIES OF WOOD-BASED MATERIALS

D. L. Chamberlain*

Center for Fire Research

National Bureau of Standards

ABSTRACT

A background to the present heat release rate

calorimetry is presented. Heat release rates and cumu-
lative heat release were measured for 16 different lumber
and wood products, using three different heat release rate
instruments. The effects of moisture content, exposure
heat flux, density of product, and fire retardant on rate
of heat release were measured. The three small-scale heat
release rate calorimeters were compared, and equations
relating the data from each were developed.

Key Words: Acoustical tile; Douglas fir; fire retardants;

hardboard; heat release rate; heat release rate calorimeters;
irradiance; particle board; plywood; redwood; southern pine

1 . INTRODUCTION

Few problems have proved as intractable as that of unwanted fire. Fire

is a threat that has had society on the defensive since even before man learned
how to produce it.

Bihr [1]"*" cited early Roman laws that regulated building materials
on the basis of their combustibility. As long as materials of construction
were limited to wood and stone or brick, the qualitative definitions of
combustible and noncombustible were adequate. As synthetic materials
(including composite materials such as wood fiber-reinforced mineral
board, gypsum board, and glass fiber-reinforced plastics) found useful
places among construction materials, the need for quantitative definitions
of the terms combustible and noncombustible became, and still is,
urgent. In 1965, ASTM E 136-65, Standard Method of Test for Noncombustibility
of Elementary Materials, was adopted for the evaluation of materials for
their performance in fire, thereby permitting the use of those desirable
materials that failed to meet the simple criterion of noncombustibility.
This test defined noncombustibility in terms of (a) the temperature rise
within a standard sample subjected to a specified heating cycle, (b) the
absence of flaming after the first 30 seconds of exposure, and (c)
specific weight-loss criteria.

*Research Associate, National Forest Products Association, 1973-1982.

“''Numbers in brackets refer to the literature references listed at the
end of this paper.

1
 The meaning of "noncombustible" within the fire community was
further broadened in 1973 by the Standard Building Code [2], which
defined the terms in two parts: (a) in terms of the ASTM E 136 test and
(b) in terms of the thickness of flammable material permitted on the
surface of a noncombustible structural base. Thus, while it is probably
clear to most individuals, on a qualitative basis, what is meant by the
term "combustibility," the difficulties encountered in attempting to
develop a precise and quantitative description of the word as it applies
to the behavior of solid materials in real fires have not been overcome.

In recent years there has been a growing appreciation that an

important criterion by which to evalute the fire behavior of materials
is the rate at which heat is released during burning. Three stages
occur in the course of total involvement of a solid material in fire:
ignition, flamespread, and sustained burning. Ignition requires that
the material receive sufficient heat from an outside source to raise the
temperature of the material to its decomposition or pyrolysis temperature.
Volatile pyrolysis products must diffuse into the surrounding air suffici-
ently to provide a mixture that lies within the flammability limits of
the gaseous fuel. This mixture must then be heated to the temperature
where "ignition" (the appearance of flame) occurs. Finally, heat must
be supplied at a sufficient rate to maintain the pyrolysis process at a
minimum rate in order that burning continues. If heat is supplied in
excess of this minimum, additional material will become involved through
the process of flamespread. Continued flame spread to total involvement
of the sample, and subsequent destruction will continue only if sufficient
heat is fed back from the flame or other sources of radiation to maintain
the pyrolysis process in the solid phase of the material.

The development of a fire in the flamespread and sustained burning

stages is dependent upon rate of heat release and is not inherently
related to the total heat that would be released if the specimen burned
to completion. An important aspect of the flamespread and sustained
burning stages is the production of smoke and toxic gases. These
stages, culminating in flashover when the fire is in an enclosure, are
dominant as far as life safety is concerned. The important aspect of
the third stage is fire endurance, in which efforts to contain the fire
and save the structure depend in part upon the rate of combustion or
rate of heat release of building contents and materials of construction.

Test methods to evaluate the performance of building materials with respect

to flamespread, smoke production, and fire endurance have assumed major impor-
tance in efforts to control fires. Such methods have been developed into
ASTM, NFPA, and other national standards. Rate of heat release is a relatively
new criterion for evaluating fire behavior and is expected to assume similar
importance in research, material assessment, and building regulations.

2
 2 . OBJECTIVES

The objectives of this research were:

o to review the development of heat release rate calorimetry;

o to establish by appropriate calorimetric methods the heat

release properties of a broad range of commercially important
wood-based construction materials;

o to determine the effects of external heat flux and moisture

content upon heat release rate of these materials; and,

o to compare heat release rate values measured by different

National Bureau of Standards' calorimeters.

A review of the history of heat release rate calorimetry and of the operating
features of available heat release rate calorimeters provided a very helpful
background for this work.

3. HISTORICAL REVIEW OF HEAT RELEASE RATE CALORIMETRY

Development of any test method may be characterized by three stages the :

development of a concept, the development of an instrument to measure a quantity

based on the concept, and the application of the measurements to a problem.
The concept of rate of heat release has been described by several workers over
the past 20 years. Instruments have been developed in five different U. S.
laboratories to measure the rate at which heat is released from a burning
material specimen. At some of these laboratories, efforts have been made to
apply the measurements to the problems of controlling fires in structures and
their contents.

The concept of heat release rate calorimetry is simply the measurement of

the rate at which a known weight (volume or area) of a carefully prepared
specimen releases heat when exposed to a prescribed and controlled heating en-
vironment. It is difficult to determine the original published suggestion that
the rate of release of heat from a burning material should provide a good measure
of burning behavior. In 1943, Steiner [3] described the measurement of "fuel
contributed" during the flamespread measurement in his newly developed tunnel
test. "Fuel contributed" was obtained by integration under a recorder plot of
time versus flue-gas temperature at the end of the tunnel. While Steiner did
not describe the plot as one of rate of heat release resulting from flamespread
over the surface and subsequent burning of the sample, a "heat contributed
value" was calculated from the temperature rise in the gases entering the stack.

In a similar manner, Robertson, Gross, and Loftus [4] in 1956 described a

flamespread index in what is now ASTM E 162, Standard Test for Surface Flammability.
The index was a combined value of the rate of progress of the flame front together
with a factor involving the heat generation by the material under test. Neither
the Steiner nor the Robertson/Gross/Lof tus test was designed to provide infor-
mation on rate of heat release.

3
 , .

Thompson and Cousins [5] in 1959, described the FM Construction

,

Materials Calorimeter developed at Factory Mutual Laboratories, apparently

the first instrument intentionally designed for determining rate of heat
release from burning materials. The importance of heat release rate
began to be more widely recognized in subsequent years. British Standard
476: Part 6: 1968 [6] accounted for "the amount and rate of heat evolved
by the specimen while it is being subjected to heat in an enclosed space
under prescribed conditions." In practice, a performance index was
calculated from data obtained from plots of stack-gas temperature versus
time. ASTM E 286-65T, Surface Flammability of Building Materials Using
an 8-foot Tunnel Furnace [7], also develops a time- temperature curve
from which the total heat release is obtained. This same curve can be
used to determine rate of heat release, if desired.

The rate of weight loss of a burning material has also been used as
a measure of the rate of energy release during burning [8,9,10]. Weight-
loss rate is directly related to rate of heat release only in those
cases where fuel composition and combustion efficiency remain constant
as burning progresses.

Two fresh approaches to the measurement of heat release rates by

burning materials were presented simultaneously in 1972 by Parker and
Long [11] at the National Bureau of Standards (NBS) and by Smith [12] at
Ohio State University (OSU) Subsequently, Martin and coworkers [13] at
.

Stanford Research Institute (now SRI International) scaled up and

modified the NBS calorimeter design. In 1973, Brenden [14,15] reported
on heat release rate equipment developed at the Forest Products Laboratory
(FPL). More recently, a second NBS calorimeter, similar in principle of
operation to the Parker/Long instrument, was designed and built by
Tordella [16]. Finally, a heat release rate instrument was developed by
Sensenig and Parker [17] at NBS, based upon the principle of oxygen
consumption during the combustion process [18]

4. OPERATING CHARACTERISTICS OF EXISTING

HEAT RELEASE RATE CALORIMETERS

A number of different methods have been developed for measuring

heat release rates of materials under conditions that may be expected in
actual fires. The size of samples used and operating features of these
different methods vary widely. Those measurement methods designed for
possible use as standard test methods are described below.

4 . 1 The Factory Mutual Construction Materials Calorimeter

The Factory Mutual Construction Materials Calorimeter, referred to

as the FM calorimeter [5] Figure 1, has been in use for a number of
years for evaluating interior finish materials and floor, roof, and wall
assemblies. The exposure imposed on the relatively large sample (1220 x
1220 mm) is obtained from a burning heptane-air mixture and corresponds,
roughly, to the standard ASTM E 119 time-temperature curve compressed
into a much shorter time. Maximum heat flux is about 150 kW/m 2 .

4
 .

Interior finish materials are usually exposed in the instrument for

10 minutes, while construction assemblies are exposed for 30 minutes.
The flue gas temperature is recorded throughout the test run. In a
subsequent evaluation run, an inert (or reference) sample is substituted
for the test sample and given the identical exposure. The flue gas,
time-temperature curve is reproduced by burning propane to make up the
difference between the test and reference samples. The rate of heat
release of the test sample is thus obtained from the rate of consumption
of propane during the substitution run. This instrument is described in
detail in FM Bulletin 1929-1 [19].

4 2
. The Parker/Long NBS-I Heat Release Rate Calorimeter

The Parker/Long calorimeter, referred to as the NBS-I calorimeter

[11], Figures 2 and 3, employs gas-fired radiant panels as the exposure
source. It has the advantages of small sample size (114 x 152 mm),
uniformity of exposure, and wide range of exposure levels (up to 100
kW/m z ). Thus, materials may be evaluated for their response to a wide
range of fire conditions. A pilot flame is required for ignition at
exposures below about 40 kW/m z. Above this exposure, ignition of most
materials (plastics and cellulosics) occurs in 5 to 60 seconds. The
back side of the sample holder is water cooled to remove heat from the
back face of the sample. This instrument was designed to burn pyrolysis
products completely by means of a propane diffusion flame in the secon-
dary burner. This instrument is not provided with smoke-measurement
equipment. During operation, a pre-set stack temperature is maintained
through the use of a secondary burner supplied with propane gas that is
metered through a control valve. The controller is activated by a bank
of four parallel-connected thermocouples that sense small changes in
stack temperature. The change in flow through the control valve is
proportional to the change in stack temperature, which is, in turn,
proportional to the rate of heat release.

The NBS-I calorimeter, as described above, no longer exists. It

was modified into the NBS-I Rebuilt Calorimeter as described in the next
paragraph.

4.3 The NBS-I Rebuilt Calorimeter

The NBS-I Rebuilt Calorimeter, referred to as the NBS-I R, was a

1978 modification of the original Parker/Long instrument. The major and
only significant change was the removal of the baffled/vent stack above
the auxiliary burner and the replacement of the four sensing thermocouples
with an array of 48. The purpose of the baffle was to accomplish mixing
of the exhaust gases so that their temperature gradient was negligible.
The electrical equivalent of mixing is supplied in the rebuilt instrument
by the grid of 48 thermocouples uniformly spaced across the exit plane
of the exposure chamber. These thermocouples are arranged in parallel
so that the resultant emf is the average of the emf for each of the 48
couples

5
 The major effects of the NBS-I modification were to reduce the
response time of the instrument and to increase the magnitude of the
initial response of the instrument. This last effect probably resulted
from removal of the metal baffle which constituted a significant heat
sink. The small sample size for both NBS-I calorimeters provides the
convenience of economy and ease of operation. However, the HRR results
for small samples of materials cannot be reliably extrapolated to much
larger samples. Results from small samples of a single material of
regular geometry cannot be used to predict the heat release rate of an
assembly of materials in other geometries.

4.4 The Ohio State-Smith Heat Release Rate Calorimeter

The Ohio State-Smith Heat Release Rate Calorimeter, referred to as

the OSU [12], Figure 4, is unique in that it was designed to permit
simultaneous measurement of flame spread rate, heat release rate, smoke
release rate, and toxic gas production rate as a sample is exposed to a
known heat flux obtained from electric radiant panels. The apparatus
was constructed to have a low heat capacity and low heat losses in order
that the rate of heat release from a sample could be calculated from the
rate of temperature rise in the well-mixed products of combustion. The
instrument accepts a 152 x 152 mm sample in the vertical position and a
sample of approximately 102 x 152 mm in the horizontal position. A
pilot ignition is generally used.

Uses of the OSU instrument to characterize materials and to predict

material behavior in certain fire situations have been variously reported
[20-24]. The instrument’s main disadvantage is that rate of heat
release due to flame spreading over the surface cannot be dissociated
from inherent rate of heat release, up until the time when the flame
spreading process is completed. For charring materials, initial heat
release rate characteristics may be completely obliterated. Furthermore,
continuous application of a pilot flame at the bottom edge of the sample
may influence heat release rates. The instrument has been modified [25]
with silicon carbide resistance heaters that permit operation at exposures
up to 100 kW/m 2.

4.5 The SRI International Heat Release Rate Calorimeter

The SRI International Heat Release Rate Calorimeter, referred to as

the SRI calorimeter [13], Figure 5, is very similar to the NBS-I calorimeter
in basic design but is scaled to accept a sample 16 times larger. Thus,
the sample size is 610 mm wide x 460 mm high. Both the radiant heating
panels and the secondary burner are fired with a premixed natural gas
flame. This instrument has the unique features of a wide range of
available radiant fluxes and a capacity to accept large size samples,
including model wall and other construction assemblies. By installing a
smoke meter in the stack and using a bypass around the secondary burner,
smoke measurements may be made.

6
4.6 The Forest Products Laboratory (FPL) Heat Release Rate Calorimeter

The Forest Products Laboratory Heat Release Rate Calorimeter,

referred to as the FPL Calorimeter [14,15], Figures 6 and 7, has a
water-cooled sample holder which, like the SRI calorimeter, can accept
large size (460 x 460 mm) samples, including model construction assemblies
The exposure imposed on the sample is obtained from burning a premixed
natural gas-air fuel mixture. The instrument operates on the substitution
principle, similar to the FM calorimeter. Its disadvantages are that it
requires two runs to evaluate each sample and is limited to heat flux
2
exposures below 40 kW/m .

4 7
. Recent Developments

Two new instruments recently have been developed at NBS. The first
is a redesigned and scaled-up version of the NBS-I calorimeter. This
instrument, the NBS-II calorimeter. Figure 8, was designed and built by
Tordella [16] . It operates on the same basic principle as the NBS-I
model but differs from the original version in (a) use of a water-cooled
metal sample holder, (b) use of a larger sample size (300 x 300 mm), (c)
provision for exposure of samples in the horizontal position, (d) use of
an air lock and mechanized sample injection device that greatly reduces
heat losses and baseline disturbances that occur when a sample is
introduced in the NBS-I calorimeter, (e) use of a positive, non-extinguish
able spark-ignition device to insure that combustible pyrolysis products
could never form an explosive mixture in the calorimeter, and (f) use of
a load cell to measure rate of weight loss of the sample. This instrument
will be described in a forthcoming NBS/CFR report [16].

The second new instrument at NBS is a prototype design for obtaining

heat release rates by measurement of the oxygen consumed in the burning
of a sample, Figure 9. The instrument, known as the Oxygen-Consumption
Calorimeter, was designed and built by Sensenig [17] and was modified by
Babrauskas, as shown in the figure. This technique [18] assumes that
the quantity of heat released by a burning material is proportional to
the amount of oxygen consumed. Rate of heat release is calculated from
measurement of the flow of air and products of combustion and of the
oxygen concentration in the exhaust stack. Both the NBS-II and the
oxygen-consumption instruments will be fully described in forthcoming
reports and will not be further described here.

Important operating characteristics of the seven heat release rate

calorimeters described in this section are compared in Table 1.

Development of fire test methods appears to have proceeded along

two parallel lines: (a) ignition, flamespread, and smoke tests involving
relatively low heat fluxes applied to interior finish materials; and,
(b) fire endurance tests involving high heat fluxes or temperatures
applied to structural materials in wall, roof, and floor assemblies.
The research reported herein was concerned with the performance of wood-
based construction materials across the full range of conditions obtained

7
in actual fires. Because the design of the NBS-I calorimeter permits
study of material behavior under controlled conditions in both phases of
the fire problem (initiation and growth at low heat fluxes and fire
endurance at higher incident heat fluxes) and because this instrument
has been successfully used in earlier research, the NBS-I calorimeter
was used in this research to establish benchmark heat release rate data
for wood-based materials.

The project was initiated on the original NBS-I calorimeter and

most of the heat release rate data developed were obtained on this
instrument. Before the project was completed, construction of a new
calorimeter (NBS-II) that would accept much larger samples and function
more reliably than the NBS-I was initiated at NBS. Measurement of heat
release rates of wood-based materials, therefore, was interrupted for a
20-month period during construction of the new calorimeter. When the
new instrument was operational, the original calorimeter was restored to
operating condition but with several permanent design changes. As
previously described, these changes included elimination of the baffled
vent stack and addition of a grid of 48 sensing thermocouples.

Heat release rates for a number of wood materials and exposure

conditions were repeated on the NBS-I R and the new NBS-II calorimeters
to establish the relationship between data obtained from these instruments
and the original NBS-I design.

5. MATERIALS AND METHODS

5.1 Wood-Based Materials

The heat release rate properties of lumber (three species), soft

plywood, particleboard, hardboard, and acoustical ceiling tile were
determined in this investigation. The effect of fire retardant treatment
on the heat release responses of lumber and plywood also was evaluated.

The Douglas fir, redwood, and southern pine lumber samples were cut
from two or three 3 to 4.3 m (10 to 14 ft) long pieces of each species
that had been specially selected by the supplier to minimize variation
in natural grade characteristics (knots, pitch pockets, etc.) and radial
variation in age that might confound test results. Douglas fir 2x6
lumber samples, both treated and untreated, were cut from pieces taken
from regular production.

The material used to provide test specimens for the panel products
was obtained from regular production except for the 19/32 in treated and
untreated Douglas fir plywood. This plywood, consisting of two 1.2 x
2.4 m (4 x 8 ft) panels, was made by the American Plywood Association
using specially selected clear veneer. Cross-bands in each half panel
were end-matched to those in the other half. Each panel was cut in half
after lay-up, and one half was fire-retardant treated.

8
 . . C

All fire-retardant material was pressure-treated and dried at

treating-company facilities. Physical properties of these lumber and
panel products are given in Table 2. Moisture content values are based
on sample weight at test conditions and oven-dry weight. Density values
are based on volume at test conditions and oven-dry weight.

Sample sizes tested in the NBS-I and NBS-II calorimeters were

nominally 114 x 152 mm (4.5 x 6 in) and 305 x 305 mm (12 x 12 in),
respectively. All panel test samples and all lumber 114 x 152 mm samples
were single pieces. The 305 x 305 mm lumber samples were constructed by
edge-gluing three pieces with a resorcinol adhesive.

5 . 2 Conditioning

The basic conditioning environment used in this work was 21°

(70°F) and 50 percent relative humidity (RH) [26]. These conditions
were obtained in a controlled -environment conditioning room and, as
shown in Table 2, produced equilibrium moisture contents of from 6 to 10
percent

To evaluate the effect of moisture content on heat release rate,

some materials were conditioned at 100°C (212°F) and 0 percent RH and
21°C (70°F) and 80 percent RH. The 100°C (212°F)/9 percent RH environment
was accomplished with an air-circulating oven. The 21°C (70°F)/80
percent RH environment was obtained in a cabinet with air circulating
over a large surface of a saturated solution of ammonium sulfate [27].
The vapor pressure of this solution at 20°C (68°F) is such that the RH
of air in equilibrium with the solution is 81 percent. Further, the RH
is essentially constant with small changes in temperature. The 0 and 80
percent nominal conditions produced average equilibrium moisture contents
of less than 1 and 15 percent, respectively.

Samples were conditioned until the weight did not change more than
0.1 percent in a three-day period and were then transported to the
calorimeter in sealed plastic bags.

5 . 3 Calorimeter Operation *

The main heating panels in the calorimeter that provide radiant heat
flux to the samples are operated with a natural gas-air mixture. The
mixture ratio is regulated by an "atmospheric regulator" that supplies
air in a constant ratio to the fuel supply. The fuel supply is adjusted
to obtain the desired irradiance to the sample.

The radiant heat flux to the sample is measured by the temperature

rise in a copper-slug calorimeter that is mounted in the center of a
refractory board of the same dimentions as the sample. Heat-flux measure-
ments are then taken with the copper slug in the sample position.

*The procedure described applies, generally, to both the NBS-I and NBS-II
calorimeters
9
 The secondary burner, used to burn combustion products completely
and to maintain a constant temperature in the exhaust stack, is supplied
with commercial grade propane. The control temperature is set to a
value appropriate to the radiant flux in the sample exposure chamber.
The control temperature/irradiance relationship is readily determined
experimentally by plotting fuel consumption of the secondary burner
against control temperature. The control temperature is chosen from the
linear portion of such a plot, a separate plot being made for each
desired value of radiant exposure.

The calorimeter is calibrated at a particular control temperature/radiant

flux value by adding heat at the sample position in the form of a 99+
percent purity methane diffusion flame. The methane is supplied through
a suitable meter at a constant rate. A calibration constant for the
instrument is developed from a known total volume of gas supplied at a
constant rate, for a measured period of time, and from a known net heat
of combustion of the calibrating fuel. With this constant, heat release
rates are calculated from the strip-chart record of secondary fuel
consumption during a heat release rate run.

In practice, the calorimeter is operated at the desired exposure

flux with an unreactive blank sample in the sample holder until a steady
baseline (constant fuel consumption in the secondary burner) is obtained.
The door is opened, the blank is replaced by the sample, and the door is
quickly closed.

The NBS-I/NBS-I R calorimeter is equipped with a sample holder

having a water-cooled brass backing-plate. The flow of water through
this plate may be adjusted to a desired rate, and the temperature of the
water leaving the plate may be measured and recorded. This arrangement
serves to permit the estimation of heat released from the unexposed face
of the sample and to prevent the rise in temperature of this unexposed
face, thereby simulating the conditions of a semi-infinite slab. Two
—
operational problems the build-up of a tar-like deposit on the cooled
—
plate and the need to insert a backing plate behind thin samples led to
a modified operating procedure. The water flow was maintained at a
constant rate, but the water temperature was not measured for each
sample. Also, for samples less than 38 mm (1.5 in) thick, a backing
plate of mineral board (of thickness equal to, or greater than, sample
thickness) was placed behind the sample. Thus, the back-face environment
for each sample was not constant. For samples greater than 10 to 12 mm
and exposure time 10 minutes or more, this condition should be of no
consequence. For the thinner materials (two medium-density hardboards
of 10.5 and 9.0 mm and oil-tempered hardboard of 6.4 mm thickness) the
usable exposures were 5 minutes with few exceptions. Thus, while the
true effect upon heat release rate due to the use of a backing plate is
not known, it is assumed to be negligible.

10
 . ,

Loss of heat through the open door causes the control-burner fuel-
rate to increase rapidly. This rate decreases just as rapidly when the
door is closed, but the sudden increase in heat release rate results in
an over-compensating decrease in fuel to the secondary burner. As a
consequence, a "door-opening transient" is recorded. This transient
sometimes can be confused with the heat released from the sample at the
time of ignition. This point will be discussed later.

The fuel consumption of the secondary burner changes linearly and

inversely with the heat released from the sample after ignition occurs.
The strip chart record of changes in fuel consumption is also the record
of heat release rate. Two other values may be obtained from this
record, with variable accuracy: time to ignition (t. ), if that time is
not too short, and time-to-peak heat release rate, ftlually, the t.
can also be measured visually with a stop watch. Although more accuracy
is obtained when higher chart speeds are used, this practice produces
much longer chart records. Chart speeds of 2 min/ in or 5 min/in were
customarily used in the work.

An exposed sample in the calorimeter may be allowed to burn to

completion, although it was usually removed after 20 minutes or when its
physical integrity was lost, depending upon the radiant heat flux and
the material involved. Heat release data are no longer valid after the
sample begins to crumble, warp, or burn on the edges and back, because
heat release rates are quoted on a unit-area basis and calculated from
the initial area of the exposed surface of the specimen.

5 . 4 Heat Release Rate Computation

Heat release rates and total heat released for a given time period
are obtained from the strip chart record of changes in the fuel consumption
required to maintain constant temperature in the exhaust gases from the
particular instrument. Typical heat release rate traces are shown in
Figures 10 to 14. These traces record the decrease in fuel flow that is
equivalent to the fuel contribution of the sample. An instantaneous
heat release rate, expressed in watts/cm 2 area of sample, is obtained by
multiplying the net scale deflection at a given time by the calibration
constant

The total heat release for a particular time interval was derived
by graphic integration of the area under the curve between the desired
time limits. The convention was established that zero time was the
moment the sample "saw" the radiant flux in the exposure chamber (i.e.,
time was measured from the moment the door was closed).

The average heat release rate for a given time period was obtained
by determining the area under the curve (i.e., total heat released) and
dividing by the time period of interest. In practice, this averaging
was done for each successive 1-minute interval under the curve. Average
rates were calculated and recorded for the first 1-minute, first 5-
minute, first 10-minute, first 15-minute, and first 20-minute time
periods. Time to ignition, time-to-peak (maximum) heat release rate,
and the values of total or cumulative heat released for the 5-, 10-
15- and 20-minute time periods also were determined.
,

11
 The format for recording basic heat release rate (HRR) is shown in
Table 3.

The time to ignition value is that for spontaneous ignition in

those cases where spontaneous ignition occurs. For the NBS-I and the
NBS-I R calorimeters, ignitions at incident heat fluxes of less than 35
kW/m^ were not spontaneous, and a pilot ignition source was required.
Where such an ignition pilot was used, three options were available: to
place the pilot at a point above the sample so that it would ignite the
gases produced by pyrolysis; to place the pilot so that it contacted the
sample at the center of the top edge; or, to place the pilot so that it
contacted the sample at the center of the bottom edge. All data obtained
by the use of a piloted ignition source are identified. Piloted ignition
exposure of some materials was made with bottom as well as top pilot
flames for the NBS-I and the NBS-I R calorimeters. Top ignition was
accomplished by the second option, with the pilot flame in contact with
the sample.

All data for the NBS-II calorimeter, except where noted, were
obtained with a spark-ignition pilot placed above the sample in the
boundary layer of the pyrolysis products.

6. RESULTS AND DISCUSSION

6.1 Characteristic Heat Release Rate Curves

HRR curves for a given material obtained from a particular instrument

are quite reproducible in a qualitative sense. In Figures 10 and 11,
HRR curves are shown for Douglas fir and southern pine lumber, respectively,
as measured on the NBS-I calorimeter. Time is measured from the moment
that the specimen becomes exposed, i.e., from the moment of door closing.
Time to ignition was measured with a stop watch or was taken from the
strip-chart record. It is the time from door closing to the appearance
of flames as viewed through the window in the door or it is the time
from door closing to the time of heat release as indicated on the strip-
chart record. The peak heat release rate for Douglas fir occurs immedi-
ately upon ignition, while the peak for southern pine occurs later
following a period where the initial rapid rate subsides. This behavior
of southern pine may be due to the distillation of terpenes and resins
as the bulk of the wood is heated following ignition.

Figure 12 shows the HRR curve for medium density hardboard surfaced
with a thin layer of paper (to facilitate painting, etc.). Rapid
ignition and burning of the paper results in a high initial HRR that
subsides slowly to a minimum, or a plateau, at about 5 minutes. (The
rate at this minimum is still relatively high for wood products.) After
this time, the rate increases due to disintegration of the sample and
the consequent increase in the actual surface of burning. Measurements
beyond this point have no numerical meaning because the area of burning
surface has increased by some large and unknown factor.

12
 Figure 13 is the HRR curve for medium density hardboard having an
embossed surface to represent weathered flat-grain lumber. The HRR
after ignition is almost constant for about 5 minutes, at which time the
sample fails catastrophically.

Lastly, another typical type of behavior is illustrated in Figure

14. This figure shows the HRR for high density, oil-tempered hardboard.
For this material, ignition was delayed for about 30 seconds. Initial
HRR was low but increased rapidly as the bulk of the material became
heated and the board disintegrated and fell from the sample holder.
Factors which contributed to the response of this material are the high
density, which extends time to ignition; the low thickness, 5.6 mm (0.22
in) which allows the bulk temperature to rise rapidly; resin binder
,

(about 5 percent), which supplies volatile fuel; and thermal relief (or
annealing) of fiber stresses, produced in the manufacture of the board,
which precedes the loss of physical integrity.

6.2 Heat Release Rates of Lumber and Other Wood Products

Average heat release properties of wood-based materials, conditioned

at 21°C (70°F) and 50 percent RH and exposed at 60 kW/m 2 incident radiant
heat flux in the NBS-I calorimeter, are shown in Table 4. Properties
tabulated are time to ignition, peak HRR, time to peak HRR, first 1-
minute average HRR, first 5-minute average HRR, and cumulative 10-minute
total heat release. To facilitate comparison of the responses of
different materials, mean values are also expressed in Table 4 as a
percentage of corresponding values for southern pine lumber. Mean HRR
properties and the range in individual property values, based on three
tests for each material, are compared in Figures 15a to 15d.

6.2.1 Lumber

Rate of heat release from a material, once it is burning, is a

function of a complex interaction of chemical and physical properties
including density, thermal conductivity, heat capacity (or specific
heat), the chemical structures responsible for the effective heat of
vaporization (the heat required to convert unit mass of solid to volatile
fuel), and the effective heat of combustion per unit mass, thickness,
and mechanical stability. Since cellulose and lignin constitute the
bulk of all woods, it would be expected that heat of vaporization and
heat of combustion values would not vary greatly among wood species.

Mean heat release property values for untreated southern pine,

Douglas fir, and redwood lumber appear in the following ranges: t. 11
to 17 seconds; peak HRR, 98 to 134 kW/m 2 first 1-minute HRR, 59 to®96
;

kW/m 2 first 5-minute HRR, 69 to 109 kW/m 2 and 10-minute total HR,
; ;

38080 to 57710 kJ/m z (3352 to 5080 Btu/ft z ). Mean time to peak HRR
ranged from 20 to 60 seconds.

13
 In all cases, heat release property values for southern pine were
higher than those for the other two species. This is attributed to the
relatively higher density and high resin and volatile hydrocarbon
(terpene) content of the species.

Redwood generally exhibited the lowest HRR values of the three

species. Time to ignition, peak HRR, 1-minute HRR, 5-minute HRR, and
10-minute total heat release were 65, 88, 99, 65, and 66 percent,
respectively, of corresponding southern pine values. The relatively
high (99 percent) ratio of redwood to southern pine first 1-minute
average HRR is attributed to the fact that the peak HRR for southern
pine was not reached until 60 seconds into the test, while that for
redwood was reached in 20 seconds. In addition to its low density,
redwood contains relatively high percentages of tannins and tannic acid.
These constituents are responsible in part for the resistance of redwood
to fungus and insect attack. These tannins also are believed to function
as natural fire-retardant agents, produced in-situ, and to account for
the resistance of the species to forest fires.

All heat release values for the 2x8 Douglas fir sample lie
between those of southern pine and redwood. However, the 2x6 sample
of this species exhibited lower values of peak, 1-minute average HRR,
and 5-minute average HRR than redwood. The density of both samples 5f
Douglas fir was intermediate to that of the other two species.

6.2.2 Plywood

Heat release values for untreated 15.1 mm (19/32 in) and 19.1 mm
(3/4 in) Douglas fir plywood samples were similar to those for Douglas
fir lumber. Because the five-ply test panels were made of natural wood
veneer, the similarity in response is not unexpected. The density of
the plywood, however, was slightly greater than that of the Douglas fir
lumber. This is attributed, in part, to the plywood manufacturing
process which involves bonding the component veneers with phenolic resin
adhesives under elevated temperature and pressure.

6.2.3 Particle Board

In line with its significantly higher density, the 15.9 mm (5/8 in)
particle board had a significantly higher time to ignition and 10-minute
—
total heat release than plywood 26 seconds versus 8 to 14 seconds and
58840 kj/m 2 (5180 Btu/ft z ) versus 38850 to 43400 kj/m z (3420 to 3820
Btu/ft x ), respectively. However, peak and 5-minute HRR values were only
slightly higher than those for plywood, and 1-minute HRR values were
essentially the same. Time to peak HRR was a long 104 seconds for
particle board, compared to 17 to 19 seconds for plywood. Compared to
southern pine lumber, t. for particle board was longer (26 versus 17
seconds) , and time to pelk heat release rate was significantly longer
(104 seconds versus 60 seconds). However, particle board heat release
rate properties were very similar to those of southern pine lumber,
corresponding values ranging between 99 and 102 percent.

14
 . .

6.2.4
Acoustical Tile

The two acoustical tile materials had the lowest density and
exhibited the lowest heat release rate properties of any of the untreated
wood products. Time to ignition, peak HRR, 1-minute average HRR, and 5-
minute average HRR were 55 to 57 percent, 72 to 75 percent, 71 to 86
percent, and 56 to 60 percent of the corresponding values for southern
pine lumber. Times to peak HRR were relatively high, however, being 42
6.2.562 seconds compared to 60 seconds for southern pine and only 17 to
and
19 seconds for plywood.

Hardboard

The hardboard products had, as a class, the highest heat release

properties and also the highest densities of all the materials assessed
in the study. Times to ignition of the medium-density embossed, medium-
density paper-faced, and high-density oil-tempered hardboard samples
were 124, 147, and 194 percent of the southern pine lumber ignition
time

Peak and first 1-minute average HRR values for these three materials
were 118, 265, and 284 percent and 98, 156, and 117 percent, respectively,
6.2.6
of corresponding southern pine lumber values. The paper-faced product
reached a peak HRR in only 29 seconds, while the embossed and high-
density oil-tempered products reached peak HRR in 197 and 153 seconds,
respectively.

General Effect of Density

The general correlations of peak, first 1-minute average, first 5-

minute average, and first 10-minute average HRR with density are shown
in Figures 16a and 16b. The data illustrated in these figures are given
in Tables 2 and 4.

It can be seen from Figure 16a that the paper-faced and high-
density hardboards exhibited a much higher peak HRR for their densities
than the other products. In the case of the paper-faced hardboard, the
"flashing" of the paper face was undoubtedly responsible for the high
value. In the case of the high-density product, the high rate may be a
result of the oil-tempering process used in manufacturing and also the
thickness of this material (5.6 mm, 7/32 in). The low thickness caused
this material to disintegrate before a 5-minute mean rate could be
measured

The relationship between first 1-minute average HRR and density

shown in Figure 16b is distorted by the fact that the peak HRR of five
of the 12 products was not reached until 60 or more seconds into the
test. In these cases, first 1-minute average HRR values tend to be
lower for a particular density than products which had much shorter time
to peak HRR. This distortion does not appear to be significant in the

15
5-minute and 10-minute average HRR versus density relationships, shown
in Figure 16b.

Linear regressions were fitted to the HRR-density relationships by

least square analysis. Regression constants and coefficients of deter-
mination (r 2 ) are shown below.

Regression Constants
Number of Intercept, Slope, r2 ,

HRR products kW/m 2 kW*m/g %

Peak 12 -24.2 0.327 50

b
io£ 72.3 0.092 78

a
1-minute average 12 52.6 0.073 39
7 7.8 0.176 66
d
c ,d
6 49.2 0.077 20

5-minute average iij 29.8 0.109 82

10 '

32.3 0.102 80
a
10-minute average 7 29.6 0.091 80

a/ All products for which rates are available

b/Excluding paper-faced and high-density hardboards
c/Excluding products with time to peak HRR of 60 seconds or greater
d/Excluding paper-faced hardboard

It can be seen that when the paper-faced and high-density oil-

tempered hardboards are excluded from the analysis, the slopes of the
regression equations for density versus the peak, the 5-minute average,
and the 10-minute average HRR are approximately the same. In these
cases, density accounts for between 75 and 80 percent of the variation
in HRR between products. The same regression slope also adequately
describes the relationship between 1-minute average HRR and density when
those materials exhibiting time to peak HRR of 60 seconds or longer and
the paper-faced hardboard are excluded.

In general, these results indicate a reasonably constant effect of

density on the HRR properties of untreated non-overlaid wood products
greater than 6.4 mm (0.25 in) thick. This relationship applies to any
HRR having a time period base which includes the peak rate. At 60 kW/m 2
radiant flux in the NBS-I calorimeter, an increase or decrease in density
of 600 kg/m 3 causes an increase or decrease in peak, first 5-minute
average, and first 10-minute average HRR of approximately 10 kW/m 2 .

16
 .

The HRR-density regression for the peak rate had a significantly

higher intercept than the intercepts for the 5-minute and 10-minute
average regressions when the two atypical hardboard products were
excluded. This simply indicates that, after adjusting for density
differences, heat release rates for untreated, non-overlaid wood
products greater than 6.4 mm thick generally are lower after the burning
process is well established. Such behavior is considered reasonable as
the peak HRR includes the process of ignition and flamespread, as well
as sustained burning. The former process is different from that of
sustained burning which includes, among other characteristics, the
presence of a protective char layer.

6.2.7 Effects of Other Physical Properties

Grain angle, number of growth rings per inch (radially), and

percent of summerwood are considered likely to have some influence on
the heat release response of lumber. These properties are tabulated in
Table 2 for the three species evaluated. The effect of these variables,
however, cannot be determined from the available data because any one of
the properties cannot be isolated from the others in a series of controlled
experiments

Although the effects of growth rate, grain angle, and percent of

summerwood on heat release may, in fact, be small, relative to the
variability in test values, it is reasonable to assume these properties
affect or relate to both charring and heat release rates. As a wood
surface receives heat, a thin layer of char is rapidly formed. The
insulative and reradiative effects of intact char layers are well known
[28,29]. The cells of the wood shrink as char forms causing stresses to
be set up in the char layer. The properties of the char and of the
underlying wood determine the time at which these stresses in the char
layer are relieved by cracking and spalling, with subsequent reduction
in the protective effects of the char layer. The finer and more uniform
the texture of the lumber (i.e., slower growth and least pronounced
difference in summerwood and springwood) , the longer the char layer is
expected to remain intact. This was the case for the three species
tested in this study. (Redwood was the finest textured of the three
lumber species tested and had the most stress-resistant char layer.)

Schaffer [28] showed that charring rates of wood decrease as the

density increases. McLean [30] attributed this effect to the variation
of thermal diffusivity with density. Thermal diffusivity, a measure of
how quickly a material absorbs heat from its surroundings, is defined as
the ratio of thermal conductivity to the product of density and specific
heat: a = (k/pC). A limited number of charring-rate measurements were
made on the three lumber species in the course of this study. Observed
char rates for samples conditioned at 21°C and 50 percent RH and exposed
at 60 kW/m 2 radiant flux were as follows:

17
 Char Rate
Material imn/min in/min

Southern pine 0.65 0.025

0.79 0.031
Douglas fir 0.69 0.027
0.71 0.028
Redwood 0.74 0.029
0.76 0.030
0.73 0.031

The general trend shown by these data of increased charring rate

with decreased density of the species involved is in agreement with the
previously cited literature. However, in view of the positive correlation
found in this study between heat release rates and density, it would
appear that other factors, in addition to charring rate, contribute to
the relative heat release rates of different wood species and wood
products.

6.2.8 Effect of Fire-Retardant Treatment

The effect of the pressure-process fire retardant treatment on

Douglas fir 2x6lumber (treatment A), 15.1 mm (19/32-in) Douglas fir
plywood (treatment B), and 12.7 mm (1/2-in) Douglas fir plywood (treat-
ment B) was evaluated. The 15.1 mm treated plywood was taken from the
same panel as the untreated plywood of this thickness. Therefore, a
comparison of the heat release properties of these matching materials
provides a very sensitive evaluation of the treatment effect. The 12.7
mm treated plywood was provided by a treating company. Because matching
untreated plywood was not tested at the same 60 kW/m 2 exposure in the
original NBS-I calorimeter, control data are not available for making an
optimum assessment of the effect of treatment on this particular plywood
sample.

The treated (incised) and untreated 2x6

Douglas fir lumber also
was supplied by a treating company. It is assumed that the treated and
untreated samples were originally obtained from the same inventory lot
and, therefore, a comparison of the data for the two samples does
provide a reasonably good measure of the treatment effect. The large
difference observed in the densities of the treated and untreated lumber
(540 versus 450 kg/m 3 .Table 2) is attributed to the high loading of dry
salt in the former. The fire retardant treatment used with the 15.1 mm
and the 12.8 mm plywoods did not cause such an increase in density.

Heat release data for the treated materials, untreated controls,

and untreated southern pine lumber are compared in Table 5. It can be
seen from this table that the fire-retardant treatment of the 15.1 mm
Douglas fir plywood caused a reduction in peak, first 1-minute average,
and first 5-minute average HRR values by 30, 71, and 63 percent,
respectively. Ten-minute total heat release was reduced 59 percent

18
relative to the matching untreated material. Although no untreated
control data are available for the 12.7 mm plywood, it can be seen that
the treated plywood of this thickness exhibited heat release rates
similar to those of the 15.1 mm treated material. The same commercial
fire retardant treatment was used for both materials.

The treated Douglas fir lumber material exhibited even greater

reduction in HRR than the treated plywood. Peak, 1-minute average, and
5-minute average rates were reduced to 33, 12, and 17 percent, respec-
tively, of the untreated material. Ten-minute total heat release was
reduced to 24 percent.

The use of fire-retardant treatment for wood products, thus, is

shown to significantly reduce the rate at which heat is contributed
under a moderately severe exposure. The char from treated materials
appeared to be more dense, less friable, and in greater yield, compared
to untreated wood materials.

6.2.9 Effect of Moisture Content

Wood and other cellulosic materials are hygroscopic. Water occurs

in wood as free water, absorbed in the larger voids and capillaries, and
as adsorbed or bound water held within the submicroscopic structure of
the wood. Adsorbed water is of two types, surface-adsorbed and capillary-
condensed. The amount of such water (moisture) in a particular wood
product increases or decreases directly with change in relative humidity
and inversely with change in temperature.

Absorbed water may be reversibly removed without affecting the

physical or chemical properties of wood. Adsorbed water may be rever-
sibly removed, but its removal causes shrinkage and changes in mechanical
properties of wood. The last few percent of adsorbed water may be held
so firmly that its removal (by heating) causes degradation of the
structural components of wood. One process in the thermal degradation
of cellulose is the dehydration of hydroxyl groups, resulting in the
non-reversible loss of water of constitution.

The effect of three different relative humidity and temperature

conditioning environments on the heat release properties of southern
pine lumber, redwood lumber, Douglas fir plywood, acoustical tile, and
hardboard is shown in Table 6. It can be seen from this table that time
to ignition and time to peak HRR generally increased and peak HRR, 1-
minute average HRR, 5-minute average HRR, and 10-minute total heat
release generally decreased with increase in relative humidity. This
behavior is consistent with that expected. The higher the moisture
content, the greater the amount of energy required to remove the bound
water and the longer the exposure time required before pyrolysis is
initiated.

19
 Regression equations for the relationship between heat release
response (expressed as a percentage of the response at 50 percent RH)
and relative humidity were determined for peak and 5-minute average HRR
and 10-minute total heat release of each product. These least square
regressions are compared with experimental values in Figure 17. It can
be seen that, with the exception of peak HRR for paper-faced hardboard,
all products exhibited a consistent trend of decreasing heat release
rate with increasing relative humidity between 0 and 80 percent.

The slopes of the regressions illustrated in Figure 17 are tabulated

below.

Regression Slope, Heat Release Response (%

of 50% RH Value) Against % Relative Humidity
First 5-min. 10-min. total
Material Peak HRR Average HRR heat release

Hardboard, paper-faced +0.320 -0.2800 —

Hardboard, embossed -0.999 — —
Southern pine lumber -0.353 -0.440 -0.450

Redwood lumber -0.199 -0.174 -0.174

Douglas fir plywood -0.448 -0.485 -0.510

Acoustical tile A -0.120 -0.307 —

Acoustical tile B -0.667 -0.667 —
Average (excluding
paper-faced hardboard) -0.4644 -0.414 -0.378

It can be seen from Figure 17 that the average regression slope

(excluding paper-faced hardboard) provides a reasonable description of
the effect of relative humidity on each of the remaining products.
Further, relative to the variation between products, the average slopes
for the peak HRR, 5-minute average HRR, and 10-minute total heat release
appear to be reasonably similar.

In general, a 10 percent increase in relative humidity during the

conditioning period causes an approximate 4 percent decrease in the peak
HRR, the 5-minute average HRR, and the 10-minute total heat release for
untreated, non-overlaid wood products of thickness greater than 6.4 mm
(1/4 in) exposed at 60 kW/m 2 in the NBS-I calorimeter.

20
 These results are in general agreement with those obtained by
Parker and coworkers [31].

6.2.10 Effect of Imposed Radiant Heat Flux

Heat release properties for woo‘d-based materials exposed at 25, 40,

2
60, and 80 kW/m in the NBS-I calorimeter are given in Table 7. A pilot
ignition source, located either at the top or bottom edge of the test
sample, was used at the lowest flux level exposure.

Heat release rates were expected to show a positive correlation

with radiant flux exposure because the rate of pyrolysis is a function
of temperature. As can be seen from Table 7 and Figure 18, this
relationship was generally observed between the 25 and 60 kW/m 2 exposure.
However, consistently lower HRR values were observed at 80 kW/m 2 than
at 60 kW/m 2.

Mean values of peak HRR, first 5-minute average HRR, and 10-minute
total heat release at 25, 40, and 60 kW/m 2 radiant flux were found to be
reasonably well described by a line connecting the 40 and 60 kW/m 2 data
points (Figure 18) . For all test materials combined, average response
at 40 kW/m 2 flux expressed as a percent of that at 60 kW/m 2 was 83
percent (7 materials), 67 percent (6 materials), and 79 percent (5
materials) for peak HRR, 5-minute average HRR, and 10-minute total heat
release, respectively. Longer times to ignition and to peak HRR account
for the 5-minute average and 10-minute total percentage values being
lower than the percentage for peak HRR.

As expected, time to ignition for top ignition at 25 kW/m 2 flux was

consistently longer than that for bottom ignition at this same flux
level. The long times to ignition for the former case and for some
samples at the non-piloted 40 kW/m 2 exposure account for the low or zero
first-minute average HRR given in Table 7.

No reasonable explanation has been found as to why the 80 kW/m 2

exposure values were consistently lower than those at 60 kW/m 2. On the
basis of data obtained from the NBS-I calorimeter, it is assumed that
the results at 80 kW/m 2 exposure are low because of some unidentified
measurement malfunction of the original calorimeter at this high radiant
flux.

6.2.11 Comparison of First 1-Minute and First 5-Minute Heat Release

Rates With the Corresponding Maximum Rates

While most materials exhibit the highest HRR in the early period of
burning, some materials exhibit the highest HRR later in the period.
The tables of HRR data show time-averaged heat release rates calculated
from time zero. That is, the first 1-minute average heat release was
calculated from the heat release rate curve starting at time zero, the
time when the sample first "saw" the radiant heat source. Heat release
rates derived in this manner will be affected by the time to ignition
(t .). Thus, a long t. may have a significant effect upon the first 1-
min§te and the first 5-liinute average heat release rates. Beyond five
minutes, t. would be expected to have insignificant effect.

21
 .

This section presents a comparison of 1-minute and 5-minute average

heat release rates calculated as the first and maximum average rates.

Table 8 summarizes the 1-minute average and 5-minute average heat

release rates for ten wood products. These products were conditioned at
21°C (70°F) and 50 percent RH and were exposed in the NBS-I calorimeter.
Heat release rates were calculated for the first 1-minute and maximum 1-
minute periods and for the corresponding 5-minute periods. The ratio of
maximum to first HRR is also shown for each material.

The data in Table 8 are also plotted in Figure 19a for the 1-minute
values and in Figure 19b for the 5-minute values. The straight lines
represent the linear regression plots of the data. The departure from
the dashed line illustrates the differences between these two methods of
calculation

The 1-minute data show that the values for (8) (9) and (10) (two
,

medium-density hard boards and the particle board) are much out of line,
with a correlation coefficient of 0.77. When the data for these three
materials are deleted, the least squares plot has a correlation coefficient
of 0.95. The differences between the first and maximum values for the
5-minute average HRR are much less, with a slope of 1.11 and a correlation
coefficient of 0.998 for all samples. The slope and correlation coefficient
are 1.07 and 0.999, respectively, when samples 8, 9 and 10 are deleted.

The densities of the two hardboards and the particle board, as

shown in Table 2, are substantially higher than for the remaining seven
materials. Therefore, one may conclude that the much higher ratio of
maximum to the first 1-minute average HRR is due to their longer times
to ignition caused by their greater heat capacities.

The above discussion illustrates the importance of choosing a

method of data reduction that is appropriate to the intended use of HRR
data.

7. COMPARISON OF HEAT RELEASE PROPERTIES

MEASURED BY DIFFERENT NBS CALORIMETERS

7 . 1 Correlation of NBS-I, NBS-I Rebuilt, and NBS-II Calorimeter Values at

60 kW/m z Exposure Flux

Mean time to ignition, time to peak HRR, peak HRR, first 1-minute
average HRR, first 5-minute average HRR, and 10-minute total heat release
values for thirteen materials as obtained in the NBS-I, NBS-I R, and
NBS-II calorimeters are shown in Table 9. Mean values and ranges in
individual values are compared in Figures 20a to 20i.

22
7.1.1 NBS-I Rebuilt Calorimeter

Both the NBS-I calorimeter (which is no longer available) and the

NBS-I R calorimeter accept 115 x 152 mm (4.5 x 6.0 in) samples. The
differences between these two instruments are that the rebuilt instrument
does not have a baffle in the stack, which reduces by 30 percent the
distance between sample and sensing thermocouples and that the rebuilt
instrument has 48 sensing thermocouples instead of four. The effects of
these changes should be the reduction of heat storage during periods of
very high heat release rate or flame impingement on the baffles and the
electrical equivalent of more thorough mixing (temperature uniformity)
of the exhaust gases as they pass the temperature sensor.

Times to ignition, from Table 9, for the NBS-I R calorimeter range

from 29 to 100 percent of those for the NBS-I calorimeter, with an
average of 52 percent for the 13 materials listed.

The NBS-I R calorimeter gave consistently higher heat release

values than the original NBS-I calorimeter, for a given material. The
values in Table 9 include a tabulation of results for the rebuilt
calorimeter as the percent of the corresponding results for the original
NBS-I calorimeter.

For untreated wood, these values are 106 to 221 (average 176)
percent for peak HRR, 105 to 262 (average 145) percent for the first 1-
minute average HRR, 112 to 167 (average 132) percent for the first 5-
minute average HRR, and 114 to 154 (average 131) percent for the 10-
minute total heat release.

For treated wood, the average values are 239 percent for peak HRR,
192 percent for first 1-minute average HRR, 163 percent for the 5-minute
average HRR, and 141 percent for the 10-minute total heat release.

7.1.2 NBS-II Calorimeter

The NBS-II calorimeter differs from the NBS-I calorimeter in these

major features: sample size, 305 x 305 mm versus 114 x 152 mm; sample
holder, hollow water-cooled metal frame versus solid mineral-board
frame; ignition, spark ignition for all exposures versus pilot-flame
ignition for exposures below 40 kW/m 2 and sensors, 48 thermocouples
;

(in parallel) versus four thermocouples (in parallel). The NBS-II

instrument is also equipped with a load cell for measuring rate of
weight loss, concurrently with rate of heat release. Another difference
was the use of a mineral-board backing plate in the NBS-I and NBS-I R
calorimeters, for samples of 19 mm (0.75 in) or less. It was
anticipated that these differences among the instruments would affect
heat release values, but the nature and direction of the effects could
not be predicted. The effect of a mineral-board backing on the heat
release rate of a material should be apparent only on thin samples, or
when the combustion zone approaches the back face of thicker samples.

23
 Heat release data for nine materials as measured in the NBS-II
calorimeter are compared with corresponding values as measured in the
NBS-I calorimeter in Table 9. Times to ignition for eight of the nine
materials were greater in the NBS-II instrument than those for the NBS-
I. Individual times ranged from 80 to 471 percent and averaged 227
percent of NBS-I calorimeter values. Time to peak HRR from the NBS-II
calorimeter ranged from 58 to 432 percent of those from the NBS-I
calorimeter, with an average of 184 percent.

Six of nine materials exhibited higher peak heat release rates in

the NBS-II than in the NBS-I. Values ranged from 69 to 194 percent
(average 135) of those from the latter instrument. First 1-minute
average HRR values from the NBS-II calorimeter ranged from 0 to 136
percent (average 72) of values obtained from the NBS-I. Five-minute HRR
and 10-minute total heat release values showed smaller differences,
ranging from 92 to 148 percent and from 63 to 121 percent, respectively.
The average response for each of these two properties was 106 percent
and 97 percent. Whereas, HRR values for fire-retardant treated wood
products were consistently higher in the NBS-I R calorimeter than in the
original NBS-I, these values were consistently lower in the NBS-II.

7.1.3 Calorimeter Correlations

As previously indicated, differences in absolute response values

obtained in different calorimeters are not unreasonable. Robertson [32]
classifies test methods into two general categories, property tests and
system tests.

A property test would involve determination of one specific

physical, chemical, or behavioral characteristic of a material,
product, or system. A system test would characterize the overall
behavioral reaction of a material, product, or system with the
environmental as well as internal variables which influence its
performance. When classifying test methods on this basis, it is
important to remember that a system test involves interactions be-
tween the material, product, or system with its surroundings. Heat
release rate test methods are among those classified as system
tests.

While individual calorimeters may differ in absolute values of

response, it is reasonable to expect that values from two or more
different instruments should be correlated, assuming that each apparatus
is designed to account for the known and important variables.

Regression analyses were made of the relationship among heat

release property values obtained from the three NBS calorimeters.
Regression of mean NBS-I heat release values on mean NBS-I R and on
mean NBS-II values are shown in Figure 21. It can be seen from this
figure that good correlations were found between NBS-I and NBS-I R
calorimeters for all four properties analyzed: peak HRR, first 1-minute
average HRR, first 5-minute average HRR, and 10-minute total heat release.
The coefficients of determination, r 2 for these relationships were 74,
,

71, 92, and 94 percent, respectively. These high values of r 2 indicate

that the response of one calorimeter to a material is generally matched
by a parallel response of the other calorimeter to that material.

24
 Five-minute average HRR and 10-minute total heat release values
from the NBS-I and NBS-II calorimeters also were highly correlated. The
coefficients of determination, r 2 for these two cases were 88 percent
,

and 98 percent, respectively. As shown in Figure 21, the relationship

between NBS-I and NBS-II peak HRR values, although significant, was
reduced because of the value for paper-faced hardboard. Excluding this
datum point from the regression analysis increased the r z from 48 percent
to 85 percent.

Virtually no correlation was found between NBS-I and NBS-II first

1-minute HRR values (r 2 = 0.3 percent). This reflects the effects of
major differences in time to ignition and time to peak HRR between the
two calorimeters and the fact that the time to ignition enters into the
calculation of the first 1-minute average heat release rate.

As shown in Figure 22, the degree of correlation between NBS-II and

NBS-I R heat release values was similar to that found between NBS-I and
NBS-II values. Correlations of NBS-II and NBS-I R peak HRR, 1-minute
average HRR, 5-minute average HRR, and 10-minute total heat release
values, as measured in terms of r 2 were 65, 3, 91, and 96 percent,
,

respectively.

7.1.4 Conversion Equations

To facilitate use of existing and future new information on the

heat release properties of construction materials, equations for estimating
HRR values related to a particular NBS calorimeter from values experimentally
determined on a different NBS calorimeter are desirable. Regression
equations and related statistics based on least-squares analysis of the
experimental data obtained in this study are presented in Table 10 for
this purpose. The equations are applicable to materials conditioned at
21°C and 50 percent RH and exposed at 60 kW/m 2 radiant flux.

7 . 2 Effect of Density as Measured in the NBS-I Rebuilt and NBS-II Calorimeters

Relationships between heat release rates and density for untreated

materials, conditioned at 21°C and 50 percent RH and exposed at 60 kW/m 2
radiant flux in the NBS-I R and NBS-II calorimeters, are shown in Figures
23 and 24. It can be seen from these figures that the same positive
high correlation between density and peak, 5-minute average, 10-minute
average HRR observed in the NBS-I calorimeter also was observed in the
NBS-I R and NBS-II calorimeters. Due to the confounding effect caused
by significantly different times to ignition, the 1-minute HRR values
for these instruments showed no correlation with density. The same
effect was observed in the corresponding NBS-I calorimeter data.

Linear regression data for the HRR-density relationships shown in

Figures 23 and 24 are tabulated below.

25
 .

Number Regression Constants

of Intercept Slope, r2 ,

HRR Products kW/m 2 kW *m/ g %

NBS-I R

Peak 10 +113.6 +0.220 48

b
9 +138.0 +0.153 60

1-minute average
K9
b c
+101.1
+106.4
+0.047
+0.032
19
19
+ 59.9 40.146 51

5-minute average 10 + 30.0 40.161 90

b
9 + 33.0 40.153 90

10-minute average + 18.0+0.160 89

K 8
b
+ 19.7+0.156 86
*********************************************************************************
NBS-II

Peak it + 69.7 +0.275 92

b
6 + 81.1 +0.243 92

1-minute average 7* +123.0 -0.104 25

b
6 + 94.7 -0.025 3
4 + 81.0 +0.032 89

5-minute average 7 + 17.7 +0.164 81

;
b
6 + 14.4 +0.173 78

a
10-minute average 4 - 11.7 +0.181 88

a/All untreated products for which rates are available.

hj Excluding paper-faced hardboard.
c/Excluding products with time to peak HRR of 57 seconds or greater.

7.2.1 NBS-I Rebuilt Calorimeter

It can be seen from the regression data for the NBS-I R calorimeter
that when paper-faced hardboard is excluded from the peak HRR data set,
the slope of the HRR-density relationship for peak, 5-minute average,
and 10-minute average HRR is essentially the same, with values ranging
between 0.15 and 0.16 kW*m/kg. When the paper-faced hardboard and those
materials exhibiting time to peak HRR of about a minute or more are
excluded, the results are very similar to those obtained in the NBS-I
calorimeter

26
 In general, these data indicate that at 60 kW/m 2 radiant flux in
the NBS-I R calorimeter an increase or decrease in density of 100 kg/m 3
causes an increase or decrease in peak, 5-minute average, and 10-minute
average HRR of approximately 15 kW/m 2 The corresponding density effect
.

in the NBS-I calorimeter was 10 kW/m 2 In view of the greater sensitivity

.

of the rebuilt instrument, this difference appears reasonable.

7.2.2 NBS-II Calorimeter

The NBS-II HRR values do not show the close correlation with
density that is found with the other two calorimeters. When paper-faced
hardboard is excluded, the peak, 5-minute average, and 10-minute average
HRR versus density regression curves had slopes of 0.24, 0.16, and 0.18
kW*m/kg, respectively. The high coefficients of determination for these
relationships (92, 81, and 88 percent, respectively) suggest that the
differences among the peak, 5-minute, and 10-minute regression values
may be due not only to the small number of samples but to other factors
as well.

Thus, in general, the effect of density on the heat release rate of

wood-based materials exposed at 60 kW/m 2 flux in the NBS-II calorimeter
is approximately a change of 25 kW/m 2 in peak HRR and 17.5 kW/m 2 in
both the 5-minute and 10-minute average HRR for each 100 kg/m 3 change in
density.

7 . 3 Effect of Imposed Radiant Flux as Measured in the NBS-I Rebuilt and

NBS-II Calorimeters

7.3.1 NBS-I Rebuilt Calorimeter

Heat release values for nine wood-based materials exposed at 40,

60, and 80 kW/m 2 in the NBS-I R calorimeter are given in Table 11 and
illustrated in Figure 25. It can be seen that heat release rates increased
with an increase in imposed flux, as expected, with most materials
exhibiting a greater incremental increase from 60 to 80 kW/m 2 than from
40 to 60 kW/m 2 The fact that the 80 kW/m 2 exposure values were higher
.

in all cases than those at 60 kW/m 2 suggests that the reverse trend
observed in the NBS-I calorimeter data was the result of some malfunction
at the high radiant flux exposure.

For the combined data for eight materials, the average ratio of
heat release property at 40 kW/m 2 to that at 60 kW/m 2 was 74, 79, and 81
percent for peak HRR, first 5-minute average HRR, and 10-minute total
heat release, respectively. A comparison of average ratios of 40 kW/m 2
to 60 kW/m 2 values from the NBS-I R and NBS-I calorimeters, for the same
test materials, shows reasonable agreement between the two data sets.
Average peak HRR, 5-minute average HRR, and 10-minute total heat release
for four products tested in both the NBS-I R and NBS-I calorimeters were
73, 68, and 74 percent (rebuilt) and 84, 64, and 78 percent (original),
respectively. These differences between matching property values are
considered within experimental error. (See correlation equations, Table 10.)

27
 .

The average ratio of heat release at 80 kW/m 2 to that at 60 kW/m 2

for six materials tested at the higher flux in the NBS-I R calorimeter
was 148, 169, and 180 percent for peak HRR, first 5-minute average HRR,
and 10-minute total heat release, respectively. These ratios are 1.85,
2.65, and 4.21 times larger than those obtained by simple linear extrapolation
of the 40 to 60 kW/m 2 exposure values. However, there is no justification
for expecting simple linear relationships among these variables.

As expected, time to ignition and time to peak HRR generally decreased

as imposed radiant flux increased. Only in two instances, 3/4-in plywood
and 5/8-in particle board, was this behavior reversed. The reversal in
these instances is attributed to experimental error.

7.3.2 NBS-II Calorimeter

Heat release properties for one material, acoustical tile B, were

determined at 80 kW/m 2 in the NBS-II calorimeter in order to compare the
response of this instrument and that of the NBS-I R at this radiant flux
level. Results of these tests and matching tests at 60 kW/m 2 are given
at the bottom of Table 11. It can be seen that the ratios of heat
release at 80 kW/m 2 to those at 60 kW/m z from the NBS-II calorimeter
were lower than those from the NBS-I R but still above 100 percent.
Ratios for peak HRR and first 5-minute average HRR were 128 and 115
percent for the NBS-II calorimeter, as compared to 150 and 136 percent
for the NBS-I R. These differences between calorimeter responses are
considered to be relatively small and within the standard deviation of
the relationship between the two calorimeters. For example, estimated
NBS-II calorimeter peak and 5-minute average HRR at 80 kW/m 2 based on
NBS-I R calorimeter values and the correlation equations in Table 10 are
256 ± 49 and 104 ± 15 kW/m 2 respectively, compared to NBS-II calorimeter
,

observed values of 217 and 85 kW/m 2 .

Ratios of time to ignition and time to peak HRR at 80 kW/m 2 to

those at 60 kW/m z from the NBS-II calorimeter also follow reasonably
well those from the NBS-I R (64 versus 67 percent and 81 versus 58
percent)

These limited data indicate that the effect of imposed radiant flux
measured in the NBS-II calorimeter is similar to that measured in the
NBS-I R.

8. CONCLUSIONS

(1) There is a general high level of correlation between heat release

rate properties of wood-based products measured in the NBS-I, NBS-I R,
and NBS-II calorimeters.

Equations relating peak heat release rate (HRR), first 1-minute

average HRR, first 5-minute average HRR, and 10-minute total heat
release mean values for a range of wood materials measured at 60 kW/m 2

28
radiant flux in the NBS-I calorimeter and similar values measured in the
NBS-I R calorimeter have coefficients of determination (r 2 ) of 74, 71,
92, and 94 percent, respectively. Equations relating these mean property
values measured at 60 kW/m z in the NBS-II calorimeter and similar values
measured in the NBS-I R calorimeter have r 2 values of 65, 3, 91, and 96
percent, respectively. The low correlation existing between the NBS-II
and NBS-I R 1-minute average HRR values is attributed to differences in
the construction and operation of the two calorimeters. These differences
significantly affect time to ignition and time to peak HRR of certain
materials. The time to ignition, in turn, is a determining factor in
the first 1-minute average heat release rate. This is demonstrated by
the high ratio of maximum 1-minute HRR to the first 1-minute HRR.

(2) Density accounts for a significant percentage of the difference in

peak HRR, first 5-minute average HRR, and first 10-minute average HRR of
untreated lumber and other wood-based materials. The effect of density
is approximately constant for the peak, the 5-minute and 10-minute
average HRR, and for the 10-minute total heat release.

At 60 kW/m 2 radiant flux, an increase or decrease in density of

untreated wood-based products of 100 kg/m 3 is associated with an average
increase or decrease in peak, 5-minute average, and 10-minute average
HRR of approximately 15 kW/m 2 when measured in the NBS-1 R calorimeter
and of approximately 10 kW/m 2 when measured in the NBS-I calorimeter.
At the same radiant flux, a 100 kg/m 3 change in density of untreated
wood materials exposed in the NBS-II calorimeter is associated with a
change of 25 kW/m z in peak HRR and 175 kW/m 2 in 5-minute and 10-minute
average HRR.

(3) Fire-retardant treatment causes a significant reduction in the rate

at which heat is released from wood products. Based on NBS-I and NBS-I
R calorimeter measurements on treated lumber, on plywood, and on matching
controls, fire-retardant treatment caused a 30 percent reduction in peak
HRR values and about 60 percent reduction in the values for first 1-
minute average HRR, first 5-minute average HRR, and the 10-minute total
heat release.

(4) Heat release rate values of wood-based materials decrease with

increase in moisture content. Based on NBS-I calorimeter measurements
at 60 kW/m 2
,
a 10 percent increase in relative humidity during conditioning
causes a decrease of approximately 4 percent in the peak HRR, in the
first 5-minute average HRR, and in the 10-minute total heat release of
untreated, non-overlaid wood products of thickness greater than 1/4
inch.

(5) Heat release rates increase non-linearly with an increase in

imposed heat flux. Based on NBS-I R calorimeter data, peak HRR, first
5-minute average HRR, and 10-minute total heat release values decrease
by an average of 26, 21, and 19 percent, respectively, when the imposed

29
flux is reduced from 60 to 40 kW/m z The values also increase an
.

average of 48, 69, and 80 percent, respectively, when the imposed flux
(6) increased from 60 to 80 kW/m 2
is .

Basic heat release properties for wood-based materials, conditioned

at 70°F and 50 percent relative humidity and exposed at 60 kW/m 2 in the
NBS-I R calorimeter, are summarized in Table 12.

9 . ACKNOWLEDGMENTS

Mr. William J. Parker and Mr. Daniel Gross provided consultation and
many helpful discussions throughout the course of this work. Dr. John
Tordella, NBS Research Associate from E. I. duPont de Nemours and
Company during 1977 and 1978 was very helpful with instrumental trouble-
shooting. Special thanks are due to Mr. Charles Veirtz and Mr. William
Twilley for help with sample preparation and operation and maintenance
of the heat release rate calorimeters. Special credit is due Dr. Edward
King, National Forest Products Association, for the detailed statistical
correlations among the various sets of data.

10. REFERENCES

[1] Bihr, J. E., Ignition, Heat Release, and Noncombustibility of

Building Materials, ASTM STP 502, p. 310, American Society for
Testing and Materials, Philadelphia, PA, (1972)

[2] Standard Building Code, pp.2-6, Southern Building Code Congress

International, Inc., Birmingham, AL (1973)

[3] Steiner, A. J., Method of Fire Hazard Classification of Building Materials,

ASTM Bulletin No. 121, pp. 19-21, American Society for Testing and
Materials, Philadelphia, PA (1943)

[4] Robertson, A. F. D. Gross, and J. Loftus, A Method for Measuring Surface

,

Flammability of Materials Using a Radiant Energy Source, Proc. Am. Soc Testing
.

and Materials, 56^, pp. 1437-1453 (1956)

[5] Thompson, Norman J., and E. W. Cousins, The FM Construction Materials Calorimeter,
NFPA Quarterly, 52_, No. 3, pp. 186-196 (Jan. 1959)

[6] British Standard 476: Fire Tests on Building Materials and Structures; Part 6.
Fire propagation tests for materials (1968)

[7] Bruce, H. D.,

and V. P. Miniutti, Small Tunnel-Furnace Test for Measuring
Surface Flammability, Report No. 2097, Forest Products Laboratory, U. S.
Department of Agriculture, Madison, WI (1957)

[8] Blackshear, P. L., Jr., B. D. Wood, N. J. Alvares, J. W. Matthews, and N. J.

Barsic, Combustion Laboratory Technical Report No. 7, Final Report OCD
Contract No. N0022869C1172 University of Minnesota, pp. 68-93 (Nov. 1969)
,

30
 ,

[9] George, Charles W. and Aylmer D. Blakely, Energy Release Rates in Fire
,

Retardant Evaluation, Fire Technology, j>. No. 3, 203-210 (Aug. 1970)

[ 10 ] Fransden, William H. , and Richard C. Rothermel, Measuring the Energy-Release

Rate of a Spreading Fire, Combustion and Flame, 1£, pp. 17-24 (1972)

[ 11 ] Parker, W. J. and M. E. Long, Development of a Heat Release Rate Calorimeter

,

at NBS. Ignition, Heat Release, and Noncombustibility of Materials, ASTM

STP 502, pp. 135-151, American Society for Testing and Materials,
Philadelphia, PA (1972)

[ 12 ] Smith, E. E. ,
Heat Release Rate of Building Materials, ibid, pp. 119-134

[13] Amaro, A. J. A. M. Kanury, A. E. Lipska, and S. B. Martin, Thermal Indices

,

From Heat Release Rate Calorimetry, Paper No. 37, Presented at the Western
States Section of the Combustion Institute Fall Meeting, California
State University at Northridge, CA (1974)

[14] Brenden, John J. An Apparatus Developed to Measure Rate of Heat Release

,

from Building Materials, FPL Research Paper No. 217, USDA Forest Service,
Forest Products Laboratory, Madison, Wisconsin (1973)

[15] Brenden, John J. Rate of Heat Release from Wood-Based Building Materials
,

Exposed to Fire, FPL Research Paper No. 230, USDA Forest Service, Forest
Products Laboratory, Madison, Wisconsin (1974)

[16] Tordella, John, and William H. Twilley, Development of a Calorimeter

for Simultaneously Measuring Heat and Mass Loss Rate, Nat. Bur. Stand.
(U.S.) NBSIR 83-2708

[17] Sensenig, D. L. , An Oxygen Consumption Technique for Determining the

Contribution of Interior Wall Finishes to Room Fires, Nat. Bur.
Stand. (U.S.), NBS Technical Note 1128 (July 1980)

[18] Huggett, Clayton, Estimation of Rate of Heat Release by Means of

Oxygen Consumption Measurements, Fire and Materials, 4_, pp. 61-65
(1980)

[19] Anon., Fire Hazard Test Procedures, Class I Building Materials, Bulletin No.
1929-1, Factory Mutual Research Corporation, Norwood, Massachusetts

[ 20 ] Smith, E. E. , An Experimental Determination of Combustibility, Fire Technology,

]_, No. 2, pp. 109-119 (May 1971)

[ 21 ] Smith, E. E. An Experimental Method for Evaluating Fire Hazard, Proc. 4th

,

Int. Fire Protection Seminar, II , Zurich, pp. 133-146 (1973)

[ 22 ] Smith, E. E., Evaluation of the Fire Hazard of Duct Materials, Fire Technology,
9_, No. 3, pp. 157-170 (1973)

31
[23] Smith, E. E., Model for Evaluating Fire Hazard, J. Fire and Flammability, 5,
pp. 1979-1989 (July 1974)

[24] Smith, E. E., Application of Release Rate Data to Hazard Load Calculations,
Fire Technology, 10^, No. 3, pp. 181-186 (1974)

[25] Smith, E. E., private communication

[26] ANSI/ASTM E-171-63, Standard specification for Standard Atmospheres

for Conditioning and Testing Materials, Annual Book of ASTM Standards,
Part 41, American Society for Testing and Materials, Philadelphia, PA

[27] Perry, J. H., (ed.). Chemical Engineers’ Handbook, Third Edition, McGraw-Hill,
New York, p.797 (1950)

[28] —
Schaffer, E. L., Charring Rate of Selected Woods Transverse to Grain,
FPL Research Paper No. 69, pp. 14-15, U. S. Forest Service, Madison, WI
(April 1967)

[29] Schaffer, E. L. Review of Information Related to the Charring Rate of

,

Wood, U. S. Forest Service Research Note, FPL 0145, U. S. Forest

Service, Madison, WI (Nov. 1966)

[30] McLean, J. D., Rate of Temperature Change in Laminated Timbers

Heated in Air Under Controlled Relative Humidity Conditions,
Mimeograph No. R1434, Forest Products Laboratory, Madison, WI (1943)

[31] Parker, W. J. D. C. Brackett, R. E. Willard, and R. H. Zile, A Preliminary

,

Investigation of the Effect of Humidity on the Ignition, Heat Release and

Smoke Density Tests for Typical Room Finishing Materials, NBSIR 73-139,
Tables V-IX (March 1973)

[32] Robertson, A. F. Test Method Categorization, ASTM Standardization

,

News, _3, No. 11 (Nov. 1975)

32
 K-20

W
&
B
4"

M
a)
u
Q)
2
Ch
O
•
iH
»6 .Z cd
u
(0
rH
co
•H
C '1
U
<y

s
Castable

O)
c c
'£ o
•H
® 4J
a O
Narcocrete
O / 3
Refractory

_® M
a >1m
CO
•u
CO
C
E I
-*
CO
ra 00 o
00 £ 4- u
J /
CO O

0-1
O
e
M
oo
CO
iH
^ O

ai
M
a
00
•H
Cm

Evaluating

Burners

33
 SPECIMEN

COMBUSTION

FOR

AIR

34
Figure 3. Artist's View of NBS-I Calorimeter

35
 Smoke Detector

Figure 4. Diagram of OSU Calorimeter

36
CrA S S AM
A mo Thermoprs.
Smoke Dewsitt

H IX1N&
Chamber

AUXILIARY
B U Rwt«

Figure 5, Drawing of SRI Calorimeter

37
 t

<u
4-1
<u
B
•H

O
t—
CO
U
h4
IX
is Cn

e
CO
M
&0
(0
*H
n

vO
0)

3
t>0
•H
IX

38
Figure 7. Photograph of FPL Calorimeter

39
Figure 8. Artist's View of NBS-II Calorimeter

40
 line]

center

stack

on

Calorimeter

(located
PROBE

PRESSURE

//THERMOCOUPLE

Consumption

STATIC

Oxygen

EXHAUST.

NBS

of

Diagram

9.

Figure

41
 Fir

Douglas

Curves:

HRR

LU
Typical

IN.)
2
(1.5

104kW/m

38.1mm 10.

60kW/m2 111kW/m2

HRR:
Figure

FIR.

HRR:
AVG.

DOUGLAS EXPOSURE:

MIN.
PEAK

I J_
CD CD
CD LO
CO CNI

(z^/m) uuh
42
 CD
CNI

; i

Pine

Southern

Curves:

HRR

IN.) LU
(1.5
Typical

2 2 2
104kW/m 11.
PINE.-38.1mm 122kW/m
60kW/m

Figure

HRR:
HRR:

SOUTHERN
EXPOSURE:

MIN.
PEAK
1

I L
CD CD
O
CO
LO
CM

(
z
w/m) yuH
43
400 PAPER-FACED, MEDIUM DENSITY HAROBOARD-II.Imm (7/16 IN.)

EXPOSURE: 60kW/m2
PEAK HRR: 274kW/m2
350 1 MIN. AVG. HRR: 164kW/m 2

300

250

200

150

100

50

0 1 1
5 10

TIME, MINUTES

• Typical HRR Curves: Medium-Density, Paper-Faced Hardboard

44
350
MEDIUM DENSITY HARDB0ARD-9.5mm (3/8 IN.)

EXPOSURE: 60kW/m 2
PEAK HRR: 200kW/m 2
300 1 MIN. AVG. HRR: 110kW/m 2

250

200

150
L

100

50

0
10

TIME, MINUTES
igure 13. Typical HRR Curves: Medium-Density Hardboard

45
 350
HIGH-DENSITY HARDB0ARD-6.4mm (1/4

NBS STD FLAME SPREAD REFERENCE

(ASTM E-162 FLAME SPREAD INDEX 190)
300
EXPOSURE: 60kW/m2
PEAK HRR: 360kW/m2
1 MIN. AVG. HRR: 102kW/m2
250 -

CNI

200 -

150

100 -

50 -

0
0

TIME, MINUTES

Figure 14. Typical HRR Curves: High-Density Hardboard

46
 I i

Conditioned

I O «

m3—
1

—
i

2
hO
Materials

kW/m

CO CO of 60
LTJ CM

at

£_0t x asea|ay leay |eioi ainui|/\|-iiai

Comparison

Exposed

Data RH,

50%

Rate

and

70°F
Release

at

Heat

15a.

Legend:

Figure

w/M>|-yHH
'
aiey asea|ay leay
z
 spuo3as (6n) uoi;m6|-oi-auj!j_
Conditioned

Materials 2
kW/m

of 60

CO LT5 CO CVJ •- O at

£_0l x asea|ay leay |eioi ainmifl-uai

Comparison

Exposed

Data RH,

50%

Rate

and

70°F
Release

at

Heat

15b.

Figure
Legend:

48
 spuo33S (6u) uo|i|uB|-ox-a«JU|X
Conditioned

Materials

kW/m

of 60

co ir> *3- co csi •— o at

g.Ql x ase9|ay \ea\\ |eioi aj.nu!|/\|-u3j_ Comparison

Exposed

Data RH,

50%

Rate

and

70°F
Release

at

Heat

15c.

Figure
Legend:

49
 < — l c H

CD CD CD CD
o o CD
CO CM CO
CO CO CM
OO
X0)
C
o

•H
X
c
o
u

MCM
<u B
4-1
->
J9
——2
m vO
o
O
c 4-1

o CO
GO

x as68|au lean |Bjoj_ 3inu!i/\|-u0i

’H X
^_0l u <u
CO w
a o
B c.
O X
o w
to
CC 4J 33
CC
-Q 03 os
nz
CNI o Q
03 6s?

KH >
CD
a)
4J
O
m
ca
03 hOl E
OS X
_Q c
E K> LTD a)
GO
CO

03
CO PL,
03 DC cuo
o 03 i-h o
CO CD
X 00
00 CO cu
OS
CM CC
CO « CO CC 4-J

CD GO I
4-1 CD
- CD
c CO
a)
Q_ 33
c k_ u-

03
r—
-O
03
C3
CD
O
-O
“O
CD
O
_o
*o
CD
O
-O
“O
—CH
•

XI
O *c k. u. k— Lf~)

o CD CD CD CD rH
CO Q_ n: IE 3 :
CU

Legend:
o n < o O lfr=
K*
9C t-
3
50
xeoi ueaiu
o
= UILU
DC
•H

lq =
m
xeiu
— g

ueaiu
ge=
i
h-

HD—
DC
fii

-OH

CD CO CD CD CD CD CD CD CD
CD CD *3- CD CD
CD CM CD CO
CNI CM CM CM

,ui//\A>|-aaH
'
a l B H aseaiay leay

50
 PEAK
400
1. SOUTHERN PINE. 2><8 12

350 2. DOUGLAS FIR. 2*8

3. DOUGLAS FIR, 2*6
4. REDWOOD. 2*8
2
300 5. PLYWOOD. 19/32 in
kW/m 6. PLYWOOD. 3/4 in

7. PARTICLE BOARD HRR = 0.328 P -24.3

8. HAROBOARD. PAPER-FACED (CORR. = 0.71)
RATE, 250
9. HARDBOARD. EMBOSSED-FACE
10. ACOUSTICAL TILE A
RELEASE
200 11. ACOUSTICAL TILE B
12. HARDBOARD, HIGH DENSITY

HEAT
150

AVERAGE
100 HRR = 0.092 P + 722
(CORR. = 0.88)

50

0
400 500 600

DENSITY, kg/m 3

FIRST ONE-MINUTE AVERAGE

400

1.
350
2.

2 3.

4.
kW/m 300
5.

6.
RATE,
7.
250
8.

9.
RELEASE

10.
200
11.

12.
HEAT
8
150
= 0.57)
p + 602 (CORR
HRR = 0060
AVERAGE

100

50

I
-L _L

100 200 300 400 500 600 700 800 900 1000

DENSITY, kg/m 3

Figure 16a. Effect of Density on HRR Properties: NBS-I Calorimeter

51
 400

I. SOUTHERN PINE. 2*8

350 - 2. DOUGLAS FIR. 2*8 FIRST FIVE MINUTE AVERAGE
3. DOUGLAS FIR 2*6
4. REDWOOD, 2*8
300 PLYWOOD, 19/32 in.
5.
kW/m2

6. PLYWOOD. 3/4 in.

250 _ 7. PARTICLE BOARD
RATE.
8. HARDBOARD. PAPER-FACED
9. HARDBOARD. EMBOSSED-FACE
200 - 10. ACOUSTICAL TILE A
RELEASE

II. ACOUSTICAL TILE B

__ 12. HARDBOARD. HIGH DENSITY (disintegrated < 5 min.)
150
HEAT

AVG.
100

50

I I I I I I I I I L
0 100 200 300 400 500 600 700 800 900 1000

DENSITY. kg/m3

2
kW/m

RATE.

RELEASE

HEAT

AVG.

Figure 16b. Effect of Density on HRR Properties; NBS-I Calorimeter

52
 peak FIRST FIVE-MINUTE AVG. TEN-MINUTE TOTAL
HEAT RELEASE
200

RH

50%

at

HRR

of

PERCENT

0 50 80 0 50 80 0 50 80

RELATIVE HUMIDITY. %

Figure 17. Effect of Conditioning Room RH on HRR Properties

53
 PEAK

Southern Pine Lumber FIRST FIVE MINUTE AVG TEN-MINUTE TOTAL

HEAT RELEASE

Redwood Lumber

60kW/m2

at

HRR

of

PERCENT

Acoustical Tile "A"

T= Top ignition

B= Bottom ignition

Acoustical Tile "B"

100 B
„^
60
1 1 L 1 I I I

20 40 60 80 20 40 60 80 20 40 60 80

RADIANT HEAT FLUX, kW/m2

Figure 18. Effect of Radiant Flux on Heat Release Properties

54
250

200

150

100

50

0
50 100 150

FIRST ONE-MINUTE HRR, kW/m 2

'igure 19a. Comparison of First 1-Minute HRR with Maximum 1-Minute HRR

55
250 T T

1. REDWOOD LUMBER. 2x8

2. DOUGLAS FIR LUMBER. 2x8
3. SOUTHERN PINE LUMBER. 2x8
4. PLYWOOD. MARINE 3/4 in
-

200 5. PLYWOOD. SPECIAL 19/32 in

-

6. ACOUSTICAL TILE "A”

7. ACOUSTICAL TILE "B"

150

100

50

0
50 100 150
FIRST FIVE-MINUTE HRR, kW/m 2

Figure 19b. Comparison of First 5-Minute HRR with Maximum 5-Minute HRR

56
 • i

Heat Release Rate ,

HRR-kW/m 2
—- P'O hO ro ro
ro
o O <D
CD
co
o o
o
co
CD
cd
O ro
CD o
-P* CD
CD
CD

T
I
—o—
Peak
(140%)*- O 1

(163%)HX
HRR

> o Legend:

I z z z
I
O — 00 03 03
C/5CO CO
— —
1-min (127%)i
o O- CO
1

30 CD
CD
avg.
(108%) l 0 1
Qj
o
O
3
HRR 3 o CD
CD CD
*4 CD
CD
O
I “
3'
CD
© CD

5-min
Z
CD
<
cp
(109%)i —>h CD
cp
— O
< Z
IC CD
00
zo C 00
30 CD

<CD
c
CD

3
Ten-Minute Total Heat Release x 10

cp
3 (115%) i-Q- H

o (115% ) l- ->1230
CD

Time-To-Ignition (tig)-Seconds

Figure 20a. Heat Release Rate Properties of Wood Products as

Measured by Different NBS Calorimeters

57
 O —

Heat Release Rate .

HRR-kW/m^
NO IO io ro
ro CD OO
O CD
O NJ
O ^
O CD
O OO
O cd
o ro
o
-o CD
o
CD O O CD

Peak

i— h
I
o 1

(221)1- —
- ISJ
05

HRR

I 0 0 o Legend:

z z z
00 ro ro
C/5 C/5 C/5
1-min
(258)1 O o ro O
Q) CD Q)
avg.
-H>-i O O' O
HRR
- 0 “ 3
C-5 2.
5L »
O
I 3
CD

5-min (149)1
avg. +—>*
HRR
o
3
CT
CD 3
cr
CD
Ten-Minute Total Heat Release x 10 3

cd — * ro co cn od

Figure 20b. Heat Release Rate Properties of Wood Products as

Measured by Different NBS Calorimeters

58
 Heat Release Rate .
HRR-kW/m 2
NJ ro ro
05
O
oo o
f'O

o KJ
O o 05
o
o cz>

—
I
o-
rsj
05
(189) -D-
Peak co
(144)
HRR

> O Legend:

I
00 00 00
1—0 1
CO CO CO

1-min
(124) l 1
o O
O)
a>
avg.
( 86 ) «>|
~ 3
HRR co
— £?
os
S
? O
I I'
CO

5-min

avg.
(94) H>

HRR

3
Ten-Minute Total Heat Release x 10

o —* ro oj cn 05

CO CO
00
o o
o 05
o ro
o oo

-o
m
-rt>

Figure 20c. Heat Release Rate Properties of Wood Products as

Measured by Different NBS Calorimeters

59
 D •

1
Heat Release Rate . HRR-kW/m^
nj ro
OO
O CD
O NJ
O £»
0 05
O OO
O O
O
r-o
1^0
O 0 CD
O
CD O o
-e*
O
CJ5

r
o
1 1 1 1 1 1

Peak
(218)1 Oo
(193), 1> '

HRR

I
> o Legend:

o
1-min
(157)1- a— 00
OO
DO
OO
OO
OO
avg.

CO S’ s?
CD
HRR
=
~ 3
f'

3
I CD co 2.
EL »
o
O 3
5-min (167)1 — CD

avg.
(117)>

HRR

3
Ten-Minute Total Heat Release x 10

co — rsj to -e*. cn os

OOOOOOOOO
00-C-O03N000
03IV0_ CO OO O0

Figure 20d. Heat Release Rate Properties of Wood Products as

Measured by Different NBS Calorimeters

60
 i I

Heat Release Rate .

HRR-kW/m 2
ISO ro ro
00 O ro A 05
O 00 o
o
fsO
ro 05
O CD O O o CD CD CD CD

i r
I—OH
(228)i—Dh
Peak

(69)>
HRR

> O Legend:

I
00 DO 00
00 00 00
1-min
(188)1 — — O
avg. & *
s? i-r-
05

O C ^ '

1' 3
HRR =
CO o ^ ro
CO
CD ZL
I o
3
o re

5-min

05
- (131) —
—
i —
< (96)>
CO

JO
JO

3
Ten-Minute Total Heat Release x 10

o — ^ oj £» cn <x>

Figure 20e. Heat Release Rate Properties of Wood Products as

Measured by Different NBS Calorimeters

61
 i

Heat Release Rate ,

HRR-kW/m 2
rsj ro ro ro
05 CO O -p*
o 05
o OO
o o
o ro 05
o
O
-P*
o CD o CD CD

—CH
I

(189)1-
Peak

(194)
HRR

> o
I
z z
CD CD 00
CO CO CO
1-min
(133)h -D-
— — —
avg.
(93)
o 33 r>
05 ro 05

o o- O
c 3'
HRR 3 CD
CD
CD
o
05
CD

I O
3
CD

5-min (140)»— —
avg. (142)>

HRR

3
Ten-Minute Total Heat Release x 10

CD ISJ OO 05 05

CDOOCSOOOOO
- o
05 ro
_
oo
rsj oj
CD
oj
05 ro CD

-O-I
>

Figure 20f. Heat Release Rate Properties of Wood Products as

Measured by Different NBS Calorimeters

62
 »

Heat Release Rate ,

HRR-kW/m'
hO ro hO N9

o-t- OO
o O
o NO
o o
-t*
o
2?
00
o CD
O ro
o o
-S*
O
CT>

"i i i i r
o 1

(148)1 -I

Peak

(139) l
—1>—
HRR

> o Legend:

I
00 03 00
C/5 C/5 C/5

1-min

C5
3D
CD
O
Q>
avg. 03
(125)K>*
© E. 2.
HRR 3 O ^ CD
CD
CD CD
CD
S’
I
3
CD

5-min

avg.
(126)0

HRR

3
Ten-Minute Total Heat Release x 10'

O _ NO 00 CJ1 CT5

Time-To-Ignition (tig)-Seconds

Figure 20g. Heat Release Rate Properties of Wood Products as

Measured by Different NBS Calorimeters

63
 o

Heat Release Rate ,

HRR-kW/m^
no NO NO NO
OO CD
CD
-P*
CD
05
CD
OO
O o
CD
NO
CD
-P*
CD
05
CD
CD CD

I
—
(208)1- -D-
Peak

(170)>
HRR

> O Legend:

I
t-CH ro oo on
C/5 CO C/3

1-min
(139)i— co
cj ro
avg. (127) h>H CD CO
=> o'
C -I
3'
3
HRR
2. o 2.
» 5L 2
I
3
CD

5-min (134)1

avg. (114)H>t

HRR

3
Ten-Minute Total Heat Release x 10'

cp
3 I— -Q-
=j’

—I
o
ST

Time-To-Ignition (tig)-Seconds

Products as
Fieure 20h. Heat Release Rate Properties of Wood
Measured by Different NBS Calorimeters

64
 Heat Release Rate, HRR-kW/m :

—* — ro ro ro co
ro
£ 05 ro oj oo
*
o ro <x>
°
on
on

ro co cn 05

T"
05 ~0
CO r— OO
4*r1—
i

r—I-J
i

o
i

O
10-min

—4
CO
CO
Total

Time-To-Ignition (tig)-seconds

—t to CO CO
05 to 00
o 4*
o O
o 05 ro £
n 1 1

- Cl

-a
OQ &

Figure 20i. Heat Release Rate Properties of Wood Products as

Measured by Different NBS Calorimeters

65
 PEAK HRR

2
kW/m

NBS-I.

Excluding paper-faced
hardboard

kW/m 2
NBS-II

2 FIRST ONE-MIN. AVG HRR

kW/m

NBS-I.

2 FIRST FIVE MIN. AVG. HRR

kW/m

NBS-I.

MJ/m2

NBS-I,

Figure 21. Correlations Among NBS HRR Calorimeters

66
 Peak HRR, kW/m 2

NBS-I Rebuilt

NBS-I Rebuilt

Figure 22. Correlations Among NBS HRR Calorimeters (Cont.)

67
 FIRST ONE-MIN. AVG.

RATE.kW/m

RELEASE

HEAT

AVERAGE

DENSITY 2
(p), kg/m

Figure 23. Effect of Density on HRR Properties: NBS-I R Calorimeter

68
 3

FIRST ONE-MIN. AVG

HRR=0.032 p+ 81.0

(Excl. paper-faced hard board, and products

with an avg time-to-peak > 58 sec. 1 ^)

(Excl. paper-faced hard board 1

)

kW/m2

1000
HRR.

AVERAGE

HRR=0.164 p + 17.7 or. HRR=0.173 P+ 14.4

(Excluding paper-faced hard board

DENSITY, kg/m

Figure 24. Effect of Density on HRR Properties: NBS-II Calorimeter

69
 kW/m2

60

AT

VALUE

THE

OF

PERCENT

THE

AS

PROPERTY,

RELEASE

HEAT

EXPOSURE. kW/m 2 RADIANT FLUX

Figure 25. Effect of Radiant Heat Flux on HRR Properties: NBS-I Calorimeter

70
 ,

Table 1. Heat Release Rate Calorimeters

Exposure

Sample Size
Location mm Type Source Intensity
(in) (cm) (kW/m 2 )

FM 1219 x 1219 Substitution, Hot combustion t/T curve,

(48 x 48) equiv. gas flow gases 0-120

NBS 114 x 152 Direct, equiv. Gas-fired 15-80

(4.5 x 6.0) gas flow radiant panel

OSU 152 x 152 Direct Electric Up to

(6 x 6) temperature rise radiant panel 35

SRI 457 x 610 Direct, equiv. Gas-fired Up to

(18 x 24) gas flow radiant panel 80

FPL 457 x 457 Substitution, Gas-fired Up to

(18 x 18) equiv. gas flow panel 40

NBS-II 305 x 305 Direct, equiv. Gas-fired Up to

(12 x 12) gas flow radiant panel 80

Oxygen 102 x 102 Change in oxygen Electrically Up to

Consumption (4 x 4) consumed during heated radiant 100
combustion panel

71
 1 i 1 1 1 i —

TJ
o
4-1o
O
l l 1

s S CM <r
0) u m 1
l -a-
-a-
i

i
CT\
co < 1

1
co
u cu
M E 00
co
l 1

1 ml i

i
1

ON ON
i 1

1 m 1

a) E 1 CM 1 rH CM 1 CM
Ph 3 1 l 1

c/3 l l 1
i 1

1 1 1

a) 1 1 1

rH 1 1 1

oo co
e CU m 1

1 O 1

1 o 00
1

1 o
RH)
< a)
Vi
CO 1 00 1 00 so 1
00

c
•H
00
CU
o 1 1

1
m1

co
1

i
1
00 CO
sO
1 1

1
o
so
1

50%
cd X) 1 i 1
JH 1 i 1

o 1 l 1

1 i 1

and

70°F

CO
00 iH
c
ON
1 m
rH
i r-~ 00
CM rH
1

00
1 1 1

at c i 1 i l i 1 1 l

•H Vi i-~ 1 sO 1 00 ON 1 CN
Pi ai 1 1 rH 1 CO
a. 1 1 1

(Conditioned

X Jn<o
4-> B O 1 O 1 o O 1 O 1 o <r
O 1 o O o O
m O O O O o
O
1 1 1 1 1 1 1

•H <r ON 00 <3* rH sO rH O' CN >d*

CO
c X
00 m i

l
-a-
1

i
'Om i

1
CO
1

1
m in
i

l
mm 1

l
m I

i
n-
i

i
CM
1

1
co
1

1
|

i
00
l

1
ON
Cl) 1 i 1 1 l l i 1 1

Q
i i 1

1 1 1 1 1 1 i i 1 1 i 1
1 1 1 1 i l i i 1 1 i 1

1 1 1 1 1 l i i 1 1 i 1
Products

1 i t 1 1 1 i i 1 1 i 1 indicated

i l 1 1 1 1 i i 1 i 1

S'? l i 1 1 l 1 i i 1 1 i 1

0) « 1 i 1 1 1 l i i 1 1 i 1

S-4C0
Wood
3 4-1 <r
i

i SO
l

1 CO ON
1

1 >a-
1
i i— O 1

1 CM
l

1 rH
i

i
i

i m 1
i I-"
1

1 CO
i
| O 1

1 CN
ih e •
C0
•H U
0) o
rH
i

l SO 1 ON 1 00 1 ON 1 ON On 1 00 1 i 1 1 SO 1 sO 1 sO
of 1 1 1 1 1 1 i i i 1 1 1 control

o c 1 1 1 1 1 1 i i i 1 1 1
S O 1 1 1 1 1 1 i i 1 1 1
o 1 1 1 1 1 1 i i i 1 1 1
1 1 1 1 1 1 i i i 1 1 1

1 1 1 1 I 1 i i i 1 1 1

1 | | j | i i i 1 | |

Properties
1 1 1 1 1 1 i i i 1 1 1

1 1 1 1 i 1 i i i 1 1 1

/~v \ rs /—\ /-N /—s rs ^s /r, /—N /"V /—\ moisture

CO
CO /-V
ON
1

i
O
m
1

i
T—
m m
CO
1

i
00
MT
1

1
o
sO
CM
sO
i

1
O
m
ON
<3-
1

1
>a-
r-
1

1
CO
SO
1

i
rH
m
1

I
co
m
1

i
rH
-a-
|

1
m
co
I

i
CM
CM
a) 3 •
3
Xo
*H
s_x
rH
V'-'/
1

1
rH 1

1
rH rH
v-/ ^ 1

1
rH 1

1
O O
v-x 'w'
1

|
o V—
O '
1

1
o 1

1
O~
V '
1

1
O
V— '
1

1
O
N— '
1

1
O
S
-"'
1

1
O
'w'
1

|
O
Vw'
at
m 00 m m m O
1 1 1 1 1 1 1 1 1 l 1 1
•iH ^ ON 1 CN I co ON 1 i r-f 1 i 00 1 n- i ON 1 i i 1 sO
x B • 1

m m m o
Physical

H co
| 00
CO
l 00 00
CO CO
1

co
l

rH
l CM CM
rH rH
1 00 1 i CM 1 co i

rH
l ON 1 in
I | 1 i rH 1 1 rH 1 rH 1 rH 1 rH 1 1 1
volume

2.
and

X 1 00 1 sO 1 1 CO
I
1
1 CM <r 1 4-1
1 i

X00
1 1 1 1 l
—
Table CN 1 X 1 X 1 1 ON 1 1 1 1 i 1

1 1 1 1 rH 1 rH 1 cn 1 1 CM CM i 1

CM CM 00 v— * * iH weight
#s
1 1 1 1
A
1
A
1
1
| I

— u i

A
I
o
Vi 1 1 1 1 1 1
•s
1 1 rH 1 1 i CU
CU 1
#s
1
#N
1 X 1 'cd 1 Td -cd z Td • 'O &
X
1 1 l i I

o o o 00 A A • ad
rH B
3
1

I
V4
a)
X
1
1

1 X
V4
cu
1

1
CM
1

&
o
1

1
o
5
1

1
1
o
&
1

i
\
m
1

i
< PQ B
i

B
• 1 X Jo
V
a
1
r i 1
~
•H rH 1 B 1 B 1 1 >N 1 >N 1 So i i CU a> i
—
1 'cd
U 1 3 1 3 Td 1 Vi 1
rH Td rH Td rH #» 1 rH —
i SO i CM -cd 1

a au
1 1 1

<u (U rH rH a) cu 1 cu (U 1 a TJCU "O •H •H rH 00 co cu 3

1 1

XE 1
1 1 | i 1

4—1
c 4-1 TO 4-1 'O Hd V 4-1 4-1 — ad - V <U
oven-dry

cd •H
1

1 Vi
1

1 iH cd a)
1

1
1

1 Vi cd <u
1

1 Vi cd CU
1

1 Vi cd
1

i cd
1

i r-'- CU
i

|
co
j

1 m cu >
S O. 1

1
iH
14H
1

1
iH
4H
cu
Vi
4-1
cd
1

1 1—
0 1

1
tH
14H
a)
V
4-1
cd
1
1
iH
14H
(U
V
4-1
cd
1

1
•rl
14H
V4
00
1

1
O
X
1

1
i —
cd
rH
cd Td
U
cd
1

i Td
Td
cu
1

i 'cd
a.
B
O
c 1 1 4-1 cu 1 1
4-1 0) 1 4-1 a> 1 1 CU 1 o a V 4h | V CO l v cu 3 on
Vi 1 CO 1 CO a V 1 -o 1 CO 3 Vi 1 CO c Vi 1 CD cu 1 rH | •H iH cd V | cd CD i cd 4J O
X
<U 1

1
cd
rH
1

1 I
cd
—
33 H 1

1
o
o
1

1
cd
rH
33 Eh 1

1
cd
rH
33 eh 1

1
cd
rH •H
e 1

1 iH
O 1

I
4-1
CO
4-1

CO X a
o <U 1

i
o o
X X 1

1
o
X rH Td
4-1 1 00 1 00 1 s 1 oo 1 00 1 00 Vi 1 4-1 | 3 3 Td cd i Td B i T3 •H a)
3 1 3 1 3 1 ad 1 3 1 3 1 3 cd 1 V 1 O o V a- i V CD 1 V o CO Based

O 1 O 1 o 1 cu 1 O 1 O 1 o B i cd i o o cd i cd i cd cd
CO 1 Q 1 a 1 Pi 1 Q 1 a 1 Q Pk 1 c < Pd l PC i Pd PQ
X
1

I
cd

72
Table 3. Format for Recording HRR Data

73
 , 1 i 1 1i i *

4-1 o 1

1 vO
1

1 Oh
i

i i o
i 1

m 1

CM
1 i 1 1 1

o o 1
i 1 1 1 i i 1 1 1 1 1 1

o Cl, 1 1 i vO i iO 1 1 1 1 i i 1 1 1 1 1 1

C/3 rH 1 1 i i 1 1 rH 1 i 1 1 1

rH QJ 1 1 i i 1 1 1 i 1 1 1

03 1 1 i i 1 1 1 i 1 1 1

4-> CO 1 1 i i 1 1 1 i 1 1 1
o 0) 1 1 i i 1 1 1 i 1 1 1

t-l rH /'“N 1 1
/^\ i i /—N 1 1 /-s 1 i 1 1 1
0) Oh cn rH <r oo CN
o
1 1 i i 1 1 1 i 1 1 1

• «! i£> 1 00 1 CO i i oo 1 cn 1 00 1 i 1 1 1

e CsJ • • • • • •
1

m
1 i 1 1
i 1 i 1 1 1
•H 4-J 4J CM cn 00 oo cn oo
e cd
0J
4-4 e m
'w'
1

1 sr
'w'
1

1 "d’
V-/
i

i cn
—
'
i

V-/
cn
1
1 'd'
1

1 m 1

1 1 t
i

i
1

1 1
1

1 1
1

1 1

o 1 i
i 1 i 1 1 1 1 1 1 1 1 | 1 | |

3 *n
rH 4-J

eq
X
w O
oo
1

1 O
vO
1

1 o
Oh
i

i CN
m
i

i o
CN
1

1 o
CN
1

1 o
00
1

1
i

i
1

1
1

1
1

o 1 i 1 i
1 i 1 1 1 1 1

oo Oh cn <r oo rH
m 1

1 cn
1

1 CO
i

1 cn 1
i

cn
1

1 cn
1

1 m 1

1 i
i 1

1
1

1 1
1

1 1 i i 1 1 1 i 1 1 1

Calorimeter 00 1 1 i 1 1 1 1 i 1 | |

>
< 4-1 o
o
1

CM
1

CM
i

m 1

^d-
1
m 1

o 1
vO
i

o 1

00
1

o 1

•
o PL|
CO rH
1

1
1

1
i

i
vO
1

1
vO
1

1
1

1
o
rH
1

1 m i

i P 1

1 o
rH
1

1 rH 1
1 1

1 i 1 1 1
c 1 1 i l 1 1 1 i 1 1 1

•H 1 1 i l 1 1 1 1 1 | |

s 1 1 i 1 1 1 1 i 1 1 1

m CM 1
1 i
O m
1 1 1

o
1 1 1 1 1
NBS-I

s Oh Oh 00 rH CN Oh rH 00
4-J " — o 1

1 N 1

1
i

i
l

i
1

1 00
1

1 o 1

1 vO
i

P 1

1 rH
1

CN
1 1

s
i 1 1

CO rH 1 1 i l 1 1 rH 1 i 1 rH 1 1
Mi -X 1 1 i i 1 1 1 i
in •H
1 1 1

1 1 i i 1 1 1 i 1 I 1

Ua 1 1 i i 1 1 1 i 1 1 1
1 1 i l 1 1 1 i 1 1 1
1
1 1 1 1 1 1
|
flux 1 1 1

1 1 1 1 1 1 l 1 1 l 1

Materials 1 | I | | 1 | 1 | |

00 1 1 i 1 1 1 1 i 1 I |

pc >
Ph < 4-4 o 1

cn
1

—
i
Oh
l 1

CN
o
1
Oh
1
rH
i
vO
1
NO
1

00
1

o 1

m
i
1 i 1 1 1 1 i 1 1 1

Pd o PH 1 Oh 1 hO i Oh l O', 1 1 Oh 1 i oo 1 1 Oh 1 rH
radiant
• CO rH 1 1 i i 1 rH 1 1 i 1 rH 1 1

c c 1 1 i l 1 1 1 i 1 1 1
Cd •H 1 1 i 1 1 1 1 i 1 1 1
qj s 1 1 i l 1 1 1 i 1 1 1

2 X i

rH CM
1 1 i 1 1 1 1 i 1 1 1

Wood-Based

4-J \6 VO
Oh
1
1

Oh
00
1

1
Oh
m
i

i
m
Oh
1

l
o
Oh
1

1 oo
Oh
1
1

m
Oh
1

1 oo
P
i

i cn
CO
1

1 o
m
1

1
ON
1

1 CN
rH
1 1 1 1 i 1 1 1
kW/m rH
CO 12 1 1 i i 1 1 1 i 1 1 1

M ,X 1 1 i i 1 1 1 i 1 1 1
•H 1 1 i i 1 1 1 i 1 1 1
60 1 1 i 1 1 1 1 i » 1 1

1 1 i 1 1 i 1 1
of i 1
1
1

1 j 1 j i j j

at 4-1 o
o
1

1 cn 1
1

cn
i

i 00
1

1 i D 1
1

ON
1

1 Oh
1
1 CN i
i

m i
i m 1
1

00 1
1

<r
o
CO
P 1—
1 Oh 1 i 00 l 00 1 00 1 Oh 1 i i iO 1 rH 1 oo
1 1 i 1 1 1 1 i i CN 1 rH 1 CN
1 1 i l 1 1 1 i i 1 1

1 1 i i 1 1 1 i i 1 1
Properties

Exposed
03 1 1 i 1 1 1 1 i i 1 1

<u 1 1 i 1 1 1 1 i i 1 1

PU CM
s
1 1

oo
i

00
1

m 1

Oh
1

CN
1

P i
o i
m 1

oo
1

o
CO
rH
1

1
CM
rH
1
1

Oh
i

i rH
rH
l

l —
i

rH
1

1 rH
rH
1

1 cn
rH
1

1 ON i
i

o
rH
i

i m
cn
1

1 m
rH
1

1 oo
cn
1 1 1 i i
3 1 1 1 1
i

and 1 1 i 1 1 1 1 i i 1 1

Release
4-1
o PH
o
o 1

VO
1
>d*
i

i
cn
cn
1 oo
CN
1 CN
cn
1 cn
r>.
1 cn
o i oo
hO
i oo 1 oo
CN
1 m
m
1 1 i 1 1 1 i i 1 1
RH
o CO rH 1 1 i i 1 1 rH 1 rH i i 1 cn 1 CM
4-J s*? 1 1 i i 1 1 1 i i 1 1
pc 1 1 i 1 1 1 1 i i 1 1
50%
CU 1 1 i i 1 1 1 i i 1 1

e rX
•H 03 • o 1

O 1

00
i

o 1 1

Oh
1

>d*
1

es|
i
rH
i
Oh
1
r>-
1

cn
Heat

and
H a)
PH
o
aj
vO
1

1
1

1 CM
i

i CN
l

i rH
1

1 rH
1

1 o
rH
1

1 NO
i

i ^d-
i

i CN
1

1 Oh
rH
1

1 m
rH
1 1 i 1 1 1 1 i i 1 1

CO 1 1 1 1 1 1 1 1 1 1 1

o m
1 | | | j 1 | | j | |

4. 70°F 4-4 K Oh 00 rH cn ON cn >d*

o C o PH o 1

1
1
1 CO
i

i NO
1

i
1
1 ^d-
1

1 m 1

1 m i

i m i

i
1
1 CN
1

1
-d-
Oh
4-J o CO rH 1 1 i l 1 1 rH 1 i i rH 1 rH 1 «H
•H S'S 1 1 i l 1 1 1 i i 1 1

a) •U 1 1 i l 1 i 1 i i 1 1

Table
at B •rl 1 1 i l 1 1 1 i i 1 1
•H Pi 1 1 i i 1 1 1 i i 1 1

H 00 •

M o
01 rH
1

1
>d*
rH
1

1
m
rH
i

i
rH
—
i
1

1
CN
rH
1

1
00
1

1
hO
CN
1

1
o
rH
i

i
Oh
i

i
m
CN
1

1
rH
CN
1

1
cn
cn
CO 1 1 i 1 1 1 1 i i 1 1

4-4 CO 1 1 1 1 1 1 1 1 1 1 1

o 4-J 1
m 1 i 1 1

m i

m i 1 1
Conditioned i 1

CO 1 cn 1 i ON i CN 1 cn 1 cn 1 i i 1 cn 1 cn
• 0) 1 1 i 1 i 1

H
i 1 1 i 1

o 1 1 i l 1 1 1 i i 1 1

iz 1 1 i 1 1 1 1 i i 1 1

’
mM 1

1
1
|

1
1
|

i
|

i r—
#

>N
|

1
1
•
1

1
|

1
|

i
i
|

i
|

1
|

1
1

rO 1
• 1
• i l PH 1 1 1 i 1 1
•

Samples rH 1 M M i 1 rH 1 T3 • T3
P
1 i 1 1 i 1 1

00
1 -O 1 i l z 1 PH 1 z 1 < i PQ 1 1
rH rH l CN oo 6 • JC
1 1

— 1 i
1 i 1 l 1 1

X • cn z '

B pine

—
1 1
CM
1

1
00
X
1

1
lO
X
i

i JO
M
l

l Oh
1

1
1

1
m 1

1
i

a)
i

i
1

a)
l

1
z
vO
1

1 z
1

1
z
CN
* cn rH
•H
03

M
a)
c:
1

1
CN
•»
1

1
CN
**
i

i
rH

00
l

l
rH 1

1
*
1

1
-a
M
1

1
•H
4-J
i

i
rH
•H
4-J
1

1
rH T3
\ QJ
O
1

1
\
oo
cn
1

1
cn
in
OJ •H 1 M 1 M i X l M 1 M 1 (0 1
— i i 3 1 "O 1

4-J PH •H •H CN •H •H O i rH 4-1 #N QJ 1
*N
1

P 1
1 1 i i 1 1 i 1 1

o3 1
4-4 1
4-4 i l 4-4 1 4H 1 1 cd i (0 1 XI i 1 TJ 03 1 "d
Southern

s P3 1 1 i l 1 1 <u 1 a i o 1 M M 1 M 03 1 H
M CO 1 CO T3 1 CO 1 C/3 rH 1 •H i •H 3 3 cd o cd
p
1 i 1 1 1 1

<U 1 cd 1 CO i o l cd 1 cd o 4-J i 4-J O PH o 1 o

P
1 1 1 1

-C 1 rH 1 rH i o 1 rH 1 —
i 1 •H 1 C/3 i CO 1 cd 1 x> B 1 JP
=
4-J 1 00 1 00 i s 1 00 1 00 1 4-J 1 3 i 3 1 X) PH 1 X) QJ 1 'd
3 1 3 1 3 i T3 1 3 1 3 1 M 1 O i O 1 M 1 H 1 H
O 1 O 1 o i aj l O 1 o 1 cd 1 o i o (0 1 cd 1 cd
p p
1

CO 1 P 1 i pi i P 1 1 PH 1 c i < 1 PS 1 PC 1 PC
*SP

1 1 i i 1 1 1 i 1 1 1

74
 3 1 / s
' 1 I I

1 1

1 1

rH 0) CM /-N 1
/—s /—v 1

CO 03 0 03 rH r- <t iH CO
VJ CO vO
1

1
CO vO
/^
1

1 00
C3
O m
O a) •“3 • /-N /—S • • /—s
H tH x r--
m
1

1
LD 0> o <1-ON
1

1 00 I'- VO rH 00
— <C CM
1

| 03

C
X
03

<N
w 1

1
<f r-
V-/ V-/
rH
S-M v-/
CM rH
w 1

1
CO vO
—/
' v-r
i

' ' v-r

I

|
i— I co

•H 4-1 4-1 O 1

1 o
On
o 1

1 o
CM
o o
6 CO 44 00 >H
03 —-- o
m
1

o> a\
1

<r MJ- r*.

o
l 1 1

rH
X 3
4-1
1

1
CO 1

1
CO iH

PQ 1 1

Plywood 1 1

1 1

1 1
1 1

1 1

1 1

c 1 1

Fir •H 1 1

0 03 1 1

6C — s
|

o
1 1 j

LO CO 03 00 CM co cm <r vO n- <f 1 r— 00I

o
(Conditioned 1 1

Vi 1 iH rH rH 1 lo CM CO CM |
CO CM
Calorimeter) -U 0) rH 1
s-/ v-' 1
3_/ |

Douglas W > 1 1

u < 1 1

N' •rH 1 1

6 pH 1 1

1 1

&
-X
1

1
1

1 1
1 1

Treated
Lumber
C 1 1
NBS-I
g 1
1
1
1

X g 0)
00
1

/*“N S
1

C3 ^3
in 3
i

rH cO
u
VO
03
1

1 CTN
m
rH
vD
r» cm r^
rH ^
1

1 O
03
O’
03
vO 03 C"
CM CM CM
j

1 O' i—
co sr
Pine
cO
3 4-1 03
1

1
s-' v—
1

1
v_ v — ' s-r
1

X 03 > 1 1

Flux
Vl
•H
< 1

1
1

Pm 1 1

1 1
Fire-Retardant

1 1
Southern
“N / V /“N V~\
X CO
1

1 00 co
on
CM co <r
co CO CM
1

1 m
H
vo
00
OO O
00
I

1 i—( 00
Radiant CO
0) iH
1

1
1

1 rH — vo
— ' ' '
'
i

j
O' vO
Pm 1 i,. 1

1 1

1 1

of
2 1 1
1

1 1

Untreated o X 1 1

kW/m 4-1 X 1 rs 1

a)
X CJ O 1

oo r- m m
0>
oo
rs 1

r- oo
/ '

03 CM CM
/—\ 1

rH in
00 o
1 1

60 e
•H
X
cO
0)
CO
vO 1

1
CM <f
S—r m v-/
CM On
1

1
rH CM
v-<
rH rH CO
tH 3_r
1

|
CM CO
Properties

H 0) 1 1
MX j

and
at Pm 1 1

1 1

1 1

1 1

1 1

O c 1 1

Exposed 4-1o /-\

•H
1
r~i O vO
1

m
1

o o
1
1
Controls,
3 o
4-1 in oo CO vO <* 00 CM oo
Release

0 •H 0 rH
1

1 iH 00 CO CM rH
1

1 r-i
V iH 00 O ^
I

1 rH 00
•H 3 CO
H W00
1

1
A v-'
A A
w 1

1
tH
V—y
1

and 1 1

1 1

1 1
Heat
1 1

RH 1 1

Untreated 44 CO 1 1

O 4-3
of 50%
CO
01
m 1

1
<r '3-
1

1
CM CM CO
O H 1 1

53 1 1

and 1 1

| |

1 1
Comparison

70°F 1 /-N t /*N 1

Lumber,
00 1 03 <1) 1
— 3 3 1

1 3 3 1 CM 3 3 1 pine
X CO

CM
1

1
vO

X
•H
P- /—s
T3
•H
Cl
1

1
\
03
•H
a /—N
X
•H
3-
|

1
at
and
*
1 3 3 3 1 iH 3 3 3 1

5. 1 CM Vi 4-3Vl 1 Vi 4-1 Vl 1

Vl 03 3 3 •» 3 3 3
X 3 X
1 1

3 X
1

03 1
* 1 O
r
-3 T3
U
1

1 Vi 4-1 Vl 4-3 1 o Vl 4-3 1 O Southern

tH 03 3 4-1 3 o 3 Vl 3
X
1 1 |

Table
co 3 1 O < 3 O
3 C/3
1 £ o PQ 3 O
3 c/3
1 PQ
•H rH 1 0 C/3 1 >3 C/3 1

IM 3 T3 | 1 rH 33 1

a
1 I

03 03 tH 03 44 44 44 CU 3 44 44 44 | 4-1
U O
1

4-3 3 1 o o 1 4-3 O -a O O 1 T3 O
cO •H 1 Vi 3 03 1 V 3 3 Vi 3
X CL 1

1
•H
4M
0)
V
S'? 4J
3
S^ 6s?
V-M \ •
1

1
•H
44
3 S'?
V
VI
3
S'? S'?
V-r '
|

|
•H
<4-1
4J S'?
CO ^
3 1
4-1 3 1 4-3 3 01
H 1 CO ts V in 3 Vi a Vi
01 1 CO X E-m
i
i 3 33 H to H
X
4-1
1
1
i—
60
i

i
rH
60
r—
oc
3 1 3 i 3 3
O 1 O i O o
C/3 1 a i Q Q
1 i

75
4
 ii 1

T
l
T
1
1

1
1
1
1

1
T
i

o
m 1
I
l

1
1

1
1
1
1
1
l

O o 1

O
i 1 1

o. *3 00 on rH l cn r- i

o
1

O
i

o
1

o
1

03 </ C ON CTN I rH l 1 1 1 i

U o rH rH l rH rH rH 1 rH 1 1 1 l

O
H —
a u* o I l 1 1 1 «

o i I 1 1 1 i

a i l 1 1 1 i

X i

o l l 1 1 1

C NT rH '3- 00 rH 00 00 cn
om
I l 1 1 1 i

•H 4- CN vd 00 vd i vd 1 cn cn 1 1 1 i

e n 4-3 CN 1 i

m m on
1 1

a *44 B 00 cn 00 cn r— cn
m
1 1 1 1 i

o
i

X m i

i cn 'T cn l NT cn 1 1 1 i

3 l 1 1 1 1 i

« z
4J
vdO o
i

o O o
CN o
l

o oO 1 1 1

1
i

o CN
i

m
i 1 1

r«* CN vd NT
Om 00 m cn
i I 1 1 1 i

r". on 00 I
m vd
1

I
cn cn cn
1
cn CN
>T
1

1 1
|
1
1
1

1
|
1
1

1 T
1

T
1

T
l

o
am
.
1

1
1

1
1 1

1
I

1
l

> \ l i

1
i 1 1 i

om
o
l

o o 1

o CN m
1

O o
1

o
i

< 00 CN rH
o 00 >T
•— T)
o o
i

ON C CN 00
l

1 cn l o «H r".
1

1 o O 1

1 o 00 1

1
i

o l 1 rH
c *44 u l i
1

1
1

1
1

1
i

*H o i l 1 1 1 i

E l 1 1 1 I i

fr-e
ml i

I
1

1
1

1
i

1
i

1
i

N 1 1 1

m CN
i
1 i

u E cn NT vd rH r* CN CN 00 rH CN vd 00
o
l

m m
l 1 l 1 i

w on cn l r- r«- vd 1 00 cn 1 vd vd NT 1 vd l cn i

n rH rH i 1 1 1 1 rH i

*H -X i I 1 1 1 i
l 1 1 1 1 i
1 1 1 1 1
j

o
m
i

i
1

1
1

1
1

1 1
i

ot i 1 1 1 1 i

>
< i— 03 O
OOm
NT m
i

o
o
oo oo
1

o cn cn
1

o
o
vd cn
1

O m
i

o cn
i

o m cn
om o
i 1 1 i

o o cn
1 i
CN e i 00 vd 1 00 1/3 1 'O' l l r*' i vd
o rH rH i rH 1 rH 1 rH rH i

c *4-1 u i 1 1
1

1
I

i i

•H o i 1 1 1 i i

c B i 1 1 1 i i

Properties
03 1 5^ i 1 1 1 i i

03 rH i 1 1 1 i i

e CN
id o on
i

m m 1

00 rH 00
1 l

cn vd
i

o O i

NT cn m
4-1 B 00 CN cn
c/3 o> o
rH
m
i

i O'* 00 vd
1

I cn 00 m 1

1 vd mcn
1

i 00 -O’
i

i m i

i cn 00 vd
i 1 1 1 i rH rH i

•H -X i 1 1 i i i

i 1 1 1 l i

i 1 1 1 i i

Calorimeter)
i 1
Release

O
m i 1
1

1
l

1
i

i
i

i 1 1 i i i
i i 1 l l i

rH
ON 03 o
o
ON rH
i

i o
o
cn NT
I

o CN
1

O
o
CN CO
I

o
o 00 i

o
o 'd-
i

i o CN
o o
l 1 1 i

Heat
UN
C
o rH
rH CTi
rH
i

rH
cn l

l
CN cn
rH rH
1

rH
00 vd l

rH
l

rH
00 i o
rH rH rH
o
l i

o
i 1 l

03 o i 1 1 1 l i

NBS-I 03 i l 1 I i i

P-> i 1 1 1 i i
on
CN

m o
B NT o
CN
i

oo o
i"-
1

cn rH
1

vd <n m 1

o o i

m 00
i

00o rH
i

m
1
in
on Id CN
rH rH rH
i rH rH 00
rH rH
1

l rH rH
rH rH rH
1 cn r>. vd
1

1 o 00 l

l
cn
cn
CN
i

i m 00
CN
vd
i l 1 1 1 i
-X i 1 1 1 i i

i 1 1 1 i i
Flux
Environment 1 1
1 1 i
1 1 1 1 1 1

i 1 1 l i

O
m i
1

l 1 1 l i

i 1 1 l i i

O o m o i
l

O o 1
o
oO o o
i 1 i

rH 03 on 00 CN 00 r>. CN 00 rH CN
m 00 m
Radiant

o o o o
i i

O m o
1 1 l i

o ON C i l cn 1 'O’ 1 CN i cn i <n rH
u o rH rH rH i rH rH 1 rH rH rH 1 rH rH 1 l rH rH i rH V rH
*4-4 cj i 1 1 1 i i V
03 o 1 1 l i
2 E -X
i

i 1 1 1
i

i i

•H 6N! i 1 1 1 l i

Conditioning

kW/m
H 3
a,
i l 1 l i i

o o ON
i

o O 1

cn mo 1

1 CN 00 00
1

rH cn
i

cn o i
i oO
O i 1 1 i

03 vO CT> i CN rH cn 1 rH CN cn 1 vd >3- 00 1
'0- -O’ l CN i cn 00 CN
60 C/3 rH i 1 1 l i i rH rH CN
i 1 1 1 l i A
1

1
1
1

1
1

1
I

1
i

1
i
of at 1

i l 1 1 i i 4-3

O
m i

i
1

1
1

1
1

1
i i d
0)
i i

03 i 1 1 1 i i u
iH u
O m O m cn O o 1

o O O
1

o l

O NT i

O m
i

o
i

Effect
(Exposed
O c CN 03 in 00 vd 00 03
u o
•H *44
d
o
O
rH
id on
rH
i

i O
rH
NT o
rH
1

I o o
rH rH
cn
CN
1

1 o CN 00
rH rH rH
1

1 o m i

l o
rH
i

i O
rH
cn cn
V
a
1
o
i 1 l i

o
l

03 4J o i 1 1 1 i i
•H i 1 1 1 i i in
6. •S c i 1 1 1 l i

H M o( i 1 1 1 i i no

a rH on
i

i rH m CN
1

1 00 00 cn
1
1 o CN 00
1

1 cn
i

i m o i
i rH 00 O §
03 rH rH CN i rH rH 1 rH 1 1 rH i CN rH i CN V CN
Table (/) i 1 1 1 l i CJ
i 1 O
i 1
1
1 I
i
rH
1 t 1 1 1 1 CN
<D 4J i 1 1 i *44
H c i l 1 1 i i O
3 0)
u 4-3 fi'S o o m i

i 00 o CN
1

1 CO o 1

1 00 o 1

1 00 1
i

l Vd o i

i Vd o l
4-»

cn c rH rH i rH 1 1 1 1 i i l d
*H o i 1 1 1 i i 03
O CJ i l 1 1 l i S
E i

i
I

l
1

1
1

1
i

i
i

i
d
o
i 1 1 1 l i u
i 1 1 1 i i >
Cm cn i 1 1 1 i i d
O 4-1 03
m m m
i 1 1 1 l i

co on on i ON on cn l cn CN cn 1 CN cn 1 cn l cn i cn CN cn
03 i 1 1 1 i i B
o H i 1 1 1 i i o
z i

i
1

l
1

1
1

1
i

l
i

i
o
u
1 1 1
1 | i

I 1 1 1 1 i OO

CO 4-1
i

i
1

1
1

1
1

1
l

l
i

i 5
d d d
o o 00 o o o
m o 00
o o
m o 00
o o O o
m o o
m o 00
o
1 i

•H 03
m O i

m o 00
1

1 1
1

m CO
i

i i o
C F i l 1 1 l i *H
o C i l 1 1 l i
4-1
•H o i i 1 1 l i

4-1 a *0
•H •H F CJ o
o
rH
i

i o
o
I

1 o
o
rH
1

1 rH O
O
rH
1

1 rH rH
l

i rH o
o
i

i rH o
o
rH c
o
T3> <13 0 CN CN i ON CN l CN CN 1 CN CN 1 CN CN i CN i CN CN
C d H rH i rH 1 rH 1 rH 1 i rH i rH o
o w i 1 1 1 l i

cj i 1 1 1 i i o
i 1 1 1 i i 4-3
i 1 1 1 l i

i l 1 1 i i U
i 1 1 1 i no i 03
i 1 1 1 rH l 03 i
*4-4

rH i I 1 CO 1 CO l *T3 O i T3 03

H <
03 d u i 1 1 u 1 U CQ i d CO i u U
•H u rO 'O cn >v *H CO *4-4 CO 03
1

o
i 1 1 i i

M 0) rH o U CO rH 4J 1 4-1 O O C/3

03 SL
H
00
i

i o 03 00
l

1 rH a
1

1 W
1

1 03
1 i

i X3 u
i

i -O C/3 m
4-1 03 i 2 JO 1 01 1 3 03 1 3 03 i T3 03 i T3 O
03 3 d X i T3 S X 1 3 NT 1 O rH 1 O rH i ^4 a i U 23 rH
E O •H
a cn
i

X
03 3 1 O
Q
•H 1 O
<
*H 1 u •H i CO <0
a
i CO
X
E CN
C/3 i

i
rH CN 1

1
*4-4 cn 1
1
4J 1

1
< 4J l
i
33 i

i
03 CO

76
 . i i 1 i

T~ -r
1 i

1 1

1 i
CM
0 i

o uo so rs CM Om rH CM rs O rH 00 os O
O
•4- rH 1 1 1 1 l i 1 l 1 1

O
1 1

s
I

I o <r m
is Os
1

i o
rH
co so is 00 1

rH
sO 00
rH
1

a i— -H 1 rH 1 1

o
1

cd a
4-> it sO
o <U
H rH
_____ |
__ ^o | ^
rs
N
cr*
^^ -S
m CO
is
1 /-S ^-s /-s
<r uo rs rH
/•“N N /-s /-S
rH CM
1

1
i

•H 4-» CM /-S vO CO M1
*0 *0 l co rs is co CM
1

1 O CM H CM as
i

l 00 oo MTo O Os
i

f
cd
O i oo co co rs CM i oo co -*4 4 sO rH CM 1 1

1 <r H S/
CM ro *4
s-'
1 CO rH CM CM CO
S-r s-' WW |
s-/ s_^
— rH rH
I

* * * * 1
rH 1 1

S
w omooO CM O O O o oo
OoOo O o Mf
O
4-»
i

rs <r o o
1

m
1
PQ i 1

rH
vO CO CO co <r CM
1

m
I

1 rO -H O 'J as 1
1

O CO
*4 rH <r rH
m cm co co uo 1
CO rH CM CM co 1
CO rH CM CM CM rH rH rH 1

1
\
i

1 i

1 1 i

1 1

1 i

CM 1 i

0 O
o <r 4
1 i

o o* co co rs 1 o o\ CO o as so O
o
CO rs rH MT '4 CM O CM *4
O
1 i

6 2 o *4 m so as 1
i

O co vO 00 i o
rH
mm 00
I

l o
rH
rs rs 00
—
in rs 1 O
rH
U0 rs 00 00 i o
rH
rs
1

1 rs
.x 1 rH 1 1
1
1 i

3 S'S 1 i

•H o
sO
1

1
i

f
m 1

1
i

4J CM
CO a OS rH OO as rH O 00 CM OS O UO
is m
CO as SO uo O 1

rH rH uo rH O i

uo rs 00
m 1 1 i 1

o o 1 1

u Mf so is i is co co sO l rs co *4 uo i uo in CM — CM * 1 sO CO Mf ui UO i sO *4 1 *4
•H
U*
3
rX
i— i i— 1

1 i
i

1 i

1 i

1 i

1 i

1 i

ci CM 1 i

0
H OO rH O rH 0070 <r oom cm CO ooo IS O CO rH OO 00 CO CM O as CM o
O o O O O
i

o
l 1

O
1

O 12 00 sO CM rH sO sO (N CO 1 CO <r sO CO sr sO
3 i 1 1 i

1 rH 1 rH rH 1 rH i rH
Properties

3 1 i

•H O 1 i

3 s sO 1 i

Calorimeter) cd 1 i

0) 1 i

s 4J CM
1 i

CO 0
s^
sO O rH O CO 1 O* O CM O rs
m 1 UO O -4 rH o 1 o ooo
sO
sD sO rH
1

1 00 o sO as CM
i

i CO CM uo O
H cr* rs 1 00 <f 1 Os CM rH sO 1 OS CM CM 1 sO CM CM sf i 00 CO CO uO
Release
•rH s
rX
1

1
i

1 i

1 i
NBS-I

1 i

1 i
Heat
CM 1 i
in
0
14-1 s* O O SO CO rH O co 00 Mo o oo *4 sO m O CM UO OS O Os U0 O rs rs rs rs O 00 rs Mf
on
o
rX
O
rH
is sO 00 ON
1

1 O
rH
so rs oo rs
i

1 O
rH
rs uo co is
1

1 O
rH
00 rH rs
rH
o
rH
1

1 O
rH
sO rs rs sO i
i

O
rH
00 00 IS
1 1 1 1 i

1 1
Tested

it
O
sO
1

1
i

<u 1 i

Exposure
cn 1 i

1 i

and CM
0 *cf
CO
M
OS
CO CM
00 rH CM
H i

I
sr co rs <r rs
cm rs as 00 o
1

I
O
rH
vO os <t *4 J
oo m
as 00 1
uo 4
CM
rH O* CO as
o
00
Os so
CO CO
1

1
1 SO MJ- <f <r <r
as sO rs rs so
i

i
O
o 00 IS >4
00 00 is
RH &
rX
rH rH rH 1 rH rH 1 rH 1 rH rH 1

1
1

1
rH

1 1
Flux
1 1
50%

1 1

! 1 1
CM
0
Radiant
and
*4-4

o
\ O
O O
O
CM CM rs
O'* «-H
l o om
O rs
r>-
m rs
O rs.
i o
O m ooO
rs ro o 1 O CO
O uo
rH as
-4 CM
o
o
CO <r
in 00
1 o
o O
rH
cr> <r rs
so as 00
1 O
O
00 so CO
00 -4 Os
J2 O'* 1
rH o
1 1 1 1

70°F 0 rH CO rH rH rH 1 rH CN rH iH 1 rH sO CO CO rH 1 <r rH rH rH 1 rH CM 1 rH rH
4-1 S>S rH 1
O
1 1

1 1

<D sO 1 1
of 1

at
H 3 1

o o omm o
*D CO H H is
1 O 00 O O oo O uo id 00 O 1 rs o
uo CM Os o U0 CM Oco 00 •sf rH SO O
00
Effect
<D
rH rH rH
1 <T O
rH
is |s CM
I

1 CM co lO *D CM
CO
1 rH CO IS CM
rH
CO 1 sO CO
rH
uo uo
1

1 <r co sO CO
CO 1 1
1
1 1

1 1

1 1

(Conditioned

7. 1 1

1 1
CM 1 1

0
4H -s o
O m rs oo m O
O O
O
VO *4- sO 1 O
O
vj CO CM uo o
O
rs uo O o i co Oo Ooo o IS CO >4
O o 00 o
1 1 1

Table

C 3
° ^ H sO
IS
Mf 00 CO
C4 vT
1

i *H fs>
00 sO CO
—
i
1

l h
UO rH 00 *4
<r cm <r
1

1 rH
sO CM U0
rH UO
l 1

1
co CM 00 sO
CM co
1

1
SO CO *4
CO
4- C CM 1 1

o
1

•r 1 1

CD 4J sO 1 1

•H 1 1

3
H H 1
1

0J 1 1

U
0)
rsO
H CO
cn co *o
sT co
i

1
4 «
rH OS
vo
CM sO
m m 1

1
rH O
*4 co UO
rH rs cm uo
1

i hO
CM CO SO
vo
o sO
1

1 o CO CM 00 sO
00 CM co
1

1 as UO O
CO
1 1
CO rH 1 CM 1 rH 1 1
sample
1 1

1 1

sample

of

m «h iH co co co cm rH co co 0> rH rH ro CO m co cm co co UO H CM CM
of
bottom

top

at at
<D
M cd CM
3 X 0
w 3
O rH
—
2 flame flame

a m
wx
ignition ignition

3 m pilot pilot

M ,0 w * 03 * CO T3
0) rH cd M O M cd = a)
,3 h aJ oo o <u oo CM
I i->

5 *
60 ,0
3 S X o5 rO
0 X
00 CO
3 -s
oo co
3 's <U
cd II II

O 3
Q rH N <U
C*S
3
«H CM Q
O OS
rH
O ON M
Q rH 4J
H
cd
CO

77
 Table 8. Comparison of First 1-Minute and First 5-Minute HRR
with the Corresponding Maximum Rates

Mean Mean
No. Material One-minute HRR Five- minute HRR
(Conditioned at 70°F/50% RH, kW/m 2 k.W/m 2
and Exposed at 60 kW/m 2 in
w Max. Max.
NBS-I Rebuilt Calorimeter) First Max. First Max.
First First

1 Redwood lumber, 2x8 102 124 1.22 81.0 84.3 1.04

2 Douglas fir lumber, 2x8 126 136 1.08 112 119 1.06
3 Southern pine lumber, 2x8 122 154 1.26 121 126 1.04
4 Plywood, Marine, 3/4 in 132 158 1.20 88.0 92.0 1.05
5 Plywood, Special, 5/8 in 141 168 1.19 117 122 1.04
6 Acoustical Tile "A" 84. 3 101 1.20 71.3 73.3 1.03
7 Acoustical Tile "B" 115 131 1.14 86.7 89.7 1.03
8 Hardboard, med. density. 119 222 1.87 161 175 1.09
embossed face
9 Hardboard, med. density, 157 223 1.42 158 163 1.03
paper face
10 Particle board, 5/8 in 146 243 1.66 163 178 1.09

Slope 1.72 1 . 11
Intercept -48.0 -5. 88
Corr. Coefficient 0.77 0 9976
.

y = 1.72 x -48.0 y = 1.11 x -5.9

1 Omitting samples 8,9,&10 that

are pressed to higher than
normal densities.

Slope 1.14 1 07
.

Intercept 5.0 -2. 63

Corr. Coefficient 0.95 0 9991
.

y = 1.14 x +5.0 y = 1.07 x -2.63

78
 1 1 S

1 I 1 1

1 1

1
O co
1

O CM o st <t O NO rH om co o r*»

O rH 00
I

o st
1

o oo
i

I O m rH
1

1 O CM 00
rl H
1 i i

t— i rH iH j 1 rH rH 1 rH rH rH 1 rH rH rH rH rH rH
rH <u iN? 1 1

(TJ w
44 cd

£
<u
^^^ ^^ /-S /-s /— /-N I /S/-S/-S ^>/-n/s /-S /-N
N
1 |
I I

rH Sf N 1 <J NO O' 1 rH rH NO
O
m
1

M nO O CO I o oo
1

1 CO 00 j
NO Sf j
00 on CM j
CM ON

$ 00 Sf CO m <3- o on 00 O' >3- nO cm O ON CM CM co m

cd in VO MD
1

1 CO >3 CO
i

1
>3- NO
i

1 H rl i

i co m <3
i

1 *H CM rl rH CM rH
? <u 3 •->

O 32 44
CQ s-/
X ooo OO OO o oo 1 O O O o oo oo
CM 00 O
1 1 1
1
rH 00 -3-
O' N
1 1

ONN St CM ON rH 00 00
co m co
1

CO m
1 1

mmm S%S
1

| 1 rH 1 1 pH rH rl rH rH rH
!

1 1

1 1

1 1

1 1

oo 1 1

O CM ON O Sf CM O O' O in o r* n. O rH CM o cm m om
4-1
o O rH o
1

O r-4 O'
!

O >3-
1

O rH
I

O NO rH
1

1 O co ON O 3 NO O rH
M 1

1 iH »H
1

i rH rH
1

i rH CM
|

j H rl rl 1 rH rH rH rH rH rH
1 1

5 1 1 1

1 1

m?
1 1

C\J

a
O
HHm
N
1

00 NO
1

1 CO 00 1 O
N n*
H
CM 1 NO Sf *J rH Sf O CM *3-

HN
I 1

44 H
rl rl
1 00 NO 1 r-H
rH
| rH CM 1

rH
CO 1 CM CO CM CO -3- CM 00 ON
co 1 1 1

M -X 1 1

*H 1 1

1 1
Calorimeters

1 1

Flux)
i i

i i

i i

oo i i

-§ 4-1 ON CO O ON CM o cm ion* O rH O 00 CM O rH ON O NO
NBS
o
1

O rH 00
l

i o so i o m 1

1 O m <r
1

1 O 00 rH O CO N O co
Radiant

w 1

1 rH rH 1 rH CM 1 rH CM i rH rH 1 rH rH rH rH rH rH
1 1

§ 5 1 1

0) a 1 1

2 X i

CM
1 1

Different

B no cm <r m co oo
1

oo cm
1

co O rH !". nO ON CO ON rH rH
m 00 CO
N mm 1 1

O
i 1 1

kW/m 4-1 "s. On CM 1 OiH I l rH 1 ON >3 CO 1 CM St co CO ON CO

CO rH i— ! rH 1 rH 1 1 r-H rH
H 1 1

60 1 1

pH i 1
of

at
1 I

44 O O CO O
o r- CO
1

1 O rH
1

O CM O
O
00 co 1 O
O
00 ON O
O 00 CO O
O
o st
1 1

o no
1

oo sr 1 O CM 1 O r-. 1 rH ON I CM nO rH ON r-
Tested

6M2
M 1

1 rH CM 1 rH CO 1 rl CM rl 1 rH CM rH rH rH rH
1 1

Characteristics 4* 1 1

cd

a. CM
and
a Pn CO
CO 00 H
1

HH
00
CM
ON
nO
1

1 CO
ON rH
N 1

1 CM ON
CO rH
1 m
rH
rH CM
m
CM
i o cm
co oo
m
m h
on
rs
o
m
00
ON rH
rH rH
1 1 1 i l

1 CM 1 rH 1 rH CM CM 1 rH rH rH CM
RH .X 1 1

1 1

1 1
50%
1 j

Release 1 1

1 1

and
44 O NO
O CO o m m 1

O CO
1

o o O ON rH O rH CM O ON sf O CM
o O N i

l o n- m
1

i o st
1

i o on
1

1 o m St
1

1 O
HH
rH CO O CM rH O CO
70°F o H4 rH 1 rH rH 1 rH 1 rH 1 rl rl CO 1 «t rH rH rH rH CM
Heat
44
a
0) 1

$ *
1

Mean
at H s o
<D
O
NO
NO
CO >3-
1 O
CM
in
»H CO
H I oo cm
CM H i m oo
00 CM
1 N N CO
rH cm m
1 ON rH CM
rH CM 00
H
CM
|N «t
CM CM
ON
rH
>3-
-3-
CO
1 1

I
1

i mm I 1

i i

i i
of

(Conditioned
i
i

i i

44 om o'
co m
om mo i

O ON
i
o o O O NO O CO rH O CO O oo
o cd o
i

i o
rH
o
CM
1

i o
rH
cm
1

i o
rH
st
1

1 o m
rH
no
CO
1

1 O St r-'
St
o m oo o o
Comparison

C c H4 r— 1 i—l 1 1 | 1 1 rH rH rH rH
4- o 1 1

1 1

a 4-1 1 1

B •H 1

9.
H M0C u
<D
MON
rH CM
1

1
rH NO CM
rH CM
1 sf *3- 1 O CM 1 CM NO St 1

1
-S3
rH
-
NO NO
no
m oo cm 00 00
CO 1 1

1 1

1 1

Table

1 f

1 1

1 1

1 1
of ts

•
0)
0)
m <r co 1 O' N CO
1

i
sr st
1

i
sr co 1 CM CO CM 1 CM CM CM CO CO rH CO CO

z HO i i

i i

i 1

i i

i i

i i

i i

M i i

1 D i i

CO O
H *
44 O'
CQ rH 0) M M 1
M M" H f H H M fH t—1 <fw M
z Cd E H-
|
1
1

j
H
|
1 1
!

\
H
1

j
W
!

hH !

;
mH M - I *
<f
W
|4 1

1 1

1 1

1 1

1 i

1 1

1 1

1 1

U < u M
HH U
| NO I 14 NO 1 • 1 • CQ CQ
-H -H *H Ps r •H
cc
1

f X
|

f X !
1

1 44 rH 1 44 rH 1 44 • | 44 •
T- C M 1 1 1 CL i o. >> >N
u >4 43 1 TO * 1 CO CM 1 CO CM TD i co 1 CO TO CO rH TD W rH
a 0) rH 1 o u i

Hcd i cd
H60 0) i cd : I cd r a) cd a 4) Cd Cu
4- X oo i o a) oo 1 » 1 * 44 1 rH CM I rH CM *4 rH 44

S
cc 44
3 C X
0) i 5
-a
^S x
1 00
3 M
• 1

3 M
• cd 1

3
00 CO
^ i oo co
3^4) cd oo:
3 cm
cd oo:
O *H
co a. cm
i

i
0)
m hn 3
1

I
O 43
Q rH
1

1
O
Q
43
rH
0)
u
44
1

1
0
Q
ON
rH
1

1
1 O
Q
On
rH
M
44 Q
O\ rH 44
<U
M o
a
\
co
1 1

1 1

1 .1 -L- J, - _L
79
(Cont.

Table

80
 N ' '

i
i

i i

i
i

i i

CM 0.3 0.3

U rH sr
i

sr rH 92 94 00 00 00 m 00 00 NO
65 CO 91 ON
i

i cn —
i

o\ 96
S'? i
48 85 65
1

r- 92 On i O' 00 ON si- 00 00 ON i

i
i

i i

i
i

i i

i i

i i

3 3 i i

TJ o
u •H •H r-' CO o i o «H o o o on sr O' o m 00 00 CM o ON sr m o i

1 ON om o
cd -U a CO
o
cd o CO on 04 rH ON i 04 rH 00
rH
NO <N <f H ON NO CO rH rH '3- *<r rH VO i si- rH ON
c •H 3 04 O' r-' rH
CO > 4-t
0)
3 n
i

i <r CO CM m i

i sr
Calorimeters

3 0) OC i i

CO « 3
Pi
i

i
i

i
i

i i

Flux)
i i

i i

NBS i i

Pi
/'“S
Pi Pi pi
\ s rs /—N /^\ /*“N Pi
/-N
Pi Pi
/•""S

Pi
i

i
/^N /-N /x, /-\
Radiant

S /— /^N /*“N
i

/-S /*-N M X X H H i

X X X X
X XV X X V-/
i

M X H M H H H H v-/
X X X
1 1 l i

H w w M w X v—
M 'w'
i 1 i

X X v-/
V w
H V-/ '
i
V-/ V-/ s-/ v-x V 'wr' V— V-r/
i

i
'w' 'w''
i

o CM o 00 o
Different 1

2 NO NO 'O- CO NO 04 O' O' sr NO ON ON r-'NO rH NO

m mo O o
1

O NO CM rH
.
1

o
00
CO LO CO
ON CNI 04
i

i
04 04 CO NO
ON O' O'
CO
NO
O' NO 04 00
04 co co O O' CM
sr r'
NO r'
ON 00 in
00 rH ON o i

i 00 rH ON ON
kW/m

3 o o rH rH i O o oo o CO O rH rH o oo o o o oo rH i ooO o
O 1

by 60 •H + + b + 1 + + + + + + + + + + + -b + + + + + + 1 + + + +
4-1 i i

CO CO
at
CO
3 ON ON nO rH
•
i

i <r o O' rH in 04 NO ON 00 r-' ON o f' ON m tH p' o i

i o m rH
cr
w ON cO CO On rH iH o rH —
CO CO OJ rH m 00 r-' ON m 00 rH rH o X ON CM
o co
i

i nO ON
1

on
«H
1

rH rH m NO I-' rH CO rH rH
i

i ON rH rH
Determined

rH i i

1
Exposed
+ + + + 1 1 i + i + i + i i + + + + + + + i i i + + + +
ii II II ii i II ii 11 ii II n ii n n ii it II u II ll ii ii ii 1 ii II ii ii

pi Pi Pi Pi M X w M x X M w w M w M M IH H H M X pi pi pi pi
i

H M W M X H X H X i
i

X X X X
l 1 i i

M M X H
1 1 1 i

and i i
Values
i i

i i

RH i
i

i i

CO CM CM CM CM i CM CM CM CM CM C4 C4 CM C4 CM CM CM CM CM CM CM CM CM i CM CM CM CM
50%
3 a a a a i a a a a a a a a a a a a a a a a a a i a a a a
•H **''*.

a a a X a a a X
Release i i

3 a a a X i E2 IS x a a a ax a a a ax
xa
i

and
3> a
as AS as i

i
AS AS AS a AS S
A* AS Ai AS a
AS AS AS AS AS i

i
a
AS AS AS

i i

i i

21°C
Heat
i i

i i

i i
NBS-II
i
• • i

at
i xl
-Q
X
A3
i

X3
i

X3
i
Among
i i

<u CU 3 3 3 3 3 3 3 3 3 3 =
00 00 rH
i

3
PC 00 X3 00 3 00 00
3
X oc 3 00 00 rH 00 00 X i

i 00 00 X
fo cO cO 3 3 3 A3 3 3 3 AS 3 3 3 3 3 3 3 3 3
4-1 3 1-4 4-1
i

i 3 J-t 3 •H 3 3 4-t X 3 3 4-t 3 3 3

i

3 3 3 II
X3 X X3
i

M CU CU o i 3 3 O XI 3 3 O 3 3 o 3 3 O i 3 O 3
(Coordinated CU > > 4-) i > > 4-J 3 > > 4-t 3 > > 4-1 > > 3 i > 3 >
o. cO cO i 3 3 O rH 3 3 U
rH 3 3 3 3 i 3 3
o CU i 3 o 3 3 o 3 3 3 i 3
3 CU (U 4J i 3 3 4-t X 4-t 3 3 4-1 X 4-4 3 3 4-t 3 3 3 i 3 3 3
Relationships

Ph 4-1 3 4J i 4J 4-1 3 3 1 3 4J 3 3 1 4-t 4-t3 3 4-1 3 i 3 3 3 Rebuilt,

3 3 3 3 3 3 3 3 3 3 V-/ 3 3 3 3 3 3 3 3 3 3
3 3 X
i i

3 3 •H 3 3 •H 3 3 3 -3 3 3 3 •H 3 3 •H
AS X H AS X X a AS X •H a
i i

as •H •H a i a AS AS (X *H •h a AS AS a •H •H a i

cO a a 3 a a 3 3 3 a a 3 3 3 a a 3 a a 3 a a
0) o i i

3 o i

3 3 (X o i

3 3 a o 3 o i i

3 o i

Ph rH
i i

rH
i

i
P4 rH
i

m rH i

P-l P4 rH
1

m rH 1

Ph P4 rH
i

m i

rH Ph rH
i

m i

rH
i

i Ph rH
i

m i

rH
i NBS-I
i i
10.
i i

i
=
i

i i

i i

Table cd i i I-R
3 Pi Pi
3 o
i

X M M X i

3 X M 1 i

i X M
i

i X
1
,
3 i i

xa i
i
i

i
NBS-I
3 B i i

o o Pi Pi
W X
i

*H 3 i

X 1 X i

X
CO Pn i X M 1 i

i X
CJ> i i

i i

i i

81
 < 1 O

CN
s
— — —— 1

mO o
3 OO m O O' O' SO O O rH O CO
m
r-.om O
00 o O o 8 O o o O o 1 1

.X 00 O eg Os e*. m o co O o 1 1

01
w.
No rH eg rH rH rH »H rH CN rH rH rH rH rH »H rH rH rH rH

u sO
O 3
H rH
0>

e SO >3 O 00 sO >3 vO vO eg •H 00 O 00 sO o 3 co 3 CN sO o CO
•H r*» O sO 00 00 r». os Os rH O' CM sO 3 3 3 CO m s£> CO OS r*.r- O'
E
o M E m m3m 3
mm
O' •H O
rH 00 eg rs*
m m
m rH O »H r-
3
m eg CO
3 3 sO
O
1

X m sO >3
rH
CO 3 00 -3 os rH CN -3 CN 00 00 CO 1

t— 3 1

? Om o Ooo O oo O 00
oo o O o o o o o o o OO
CQ
rH O sO m m
Os m
sO 00 00 m
O e>* 3 00 sO r^> 00 o sO CN co sO CO
o O co sO
3 mo
as 00 •-H Os CN CO r- so r*. O' 00 •H rH r>- CN
co CO CO CO m co rH rH CO rH m3 e*. eg 3 3 sn
rH

00 CN
> B
< CM
mO m m o
O 3 Os o eg
m 3 o oo m o eg O eg O O m 00 O sO o m
.X o
00 o o
r*» 00 sO O -3 O SO eg o O O os O O sO O' O co o
c •h eg rH rH rH iH rH rH rH rH rH rH rH H rH rH rH rH

Calorimeters
0 sO
ir>

4-* eg
to 0 -3- eg o rH rH co co r- 00 m -3 r- <H 3 sO 3 rH CO rH m 00 3 m
n O eg m sO 00 m r>» rH r- rH co m •H 3 Os Os 3 m f-i 00 00
•H rH rH CM rH rH rH rH H rH
X 2
NBS-II

00 eg
and > B
< CM
<j- o co O
O
-3 CN O in OO CO eg O 00 00 O sO O sO o o CO 00 O 00 O m
X a
-X
rH
co
eg
sO o
rH
r"-
rH
O
sO
rH
so
rH
rH O
rH
r-
rH
O
rH
00 O
rH
o o
rH
00 OS O 00
rH
O co
rH rH rH rH rH
Rebuilt c 0 vO
0)
£ 4-1 eg
CO 0 Os CM sO Oa 00 m rH O SO Os e* 3 rH 3 CO sO sO 3 3 co m
sO m
OO CM 00 r-. Os 00 *3 CO *3 00 in rH CO O' eg 00 m rH o 3
*H rH eg rH rH rH CM rH rH rH rH rH rH CM rH rH
NBS-I
3
RH)

In
50Z

eg
B
o
0 3
oO O O O m 8 m o O O o
oO
o
00 CO r- sO CO rH
o 00
and rH O' CO 00
Measured

M o as O in 00 O
rH
CM
rH
CM o m Os o 00 o O3 00 m o eg
*"* *H *“*
O ’ 1 1 1

70°F -X
03 sO
0)
a.

eg
at
B P*^ rH eg rH 00 so •H oo
m o
O' CN SO O'
m o «H CO o
m eg CN eg 00 o r-.
Properties
00 00 f".
rH rH co
sO eg >3
rH eg CO
rH
CM eg co
CO 00 os
rH rH
O
rH
o
eg CN
O
eg eg
3
rH
rH
eg
o
rH CN co rH CN
3
-

Conditioned

7s
Release

o
o w oO
00
O
Os
sO
CO
m O o
O so
sO
eg
O
co
O
co 3 O
00
O
CO
CN sO
o
o o O
O
co
eg
O
O Oo
O
o sO
O
O 00
m o
o 00
u
o
rH rH rH rH rH rH CN o
rH
rH
<U vO
r* '

(Samples a
Heat
H 3
X m m
u Os co r, Os *3 r- os o rH 00 e* rH 3 Om r*» rH •H so 33 r>* o
on
o
00
co CO CM eg rH CO CN *H CN *3 00 eg
CM
CO 3 CN eg rH eg co CO

Flux

eg
^B

O C
o V
-X
co
00
o
o
co
co
O
O o
o
CO
CO
O
O oo <n
co
1

1
O
o
CO
co
00
CO
O
0
o O
oO
co
00
o
o o
o
00
00 oO
o O sO
O
O3sO
CN CN CO *""*

o 1
Radiant 1 1 1

•H
<D 4-1 SO
•H
a c
of
H M04 .

<u
a
o
rH so eg CN SO eg CN so CN c sO CN rH 00 00 00 m 00 00 Os co CN 3 O'
rH rH rH 0 rH rH rH
oo e
Effect
-
L . _ —
CM 10
o 4J
11.
CO co •3 CO CO f''* CO CO CO CO CO eg co CN co eg CO rH co CN co CO co co CO CO
a)
o H
z
- -
J - - 1

Table

o
3 X 6 '

<0 3 ooo o SO o
o CO o o o oOO Oo Oo o sO
o o 00
o OOO OO
o <— 3
Cl cm .x
3 sO 00 •3 so 00 *3 sO oo 3 S 3 sO 3 v£> 3 sO 00 sD 00

UJ

• =
U • U • CQ CQ 00
rH •H •H >s *H *H rH
03 SM rH SM •H 1 SM 1 SM • 1 m CO CO CO
•H c M CL a JO >s Q) U < U CQ O CQ
u X T3 <0 C/3 TO CO rH TO CO rH rH - -H *H *H
<D a> •H O u <0 z <0 r <o CO a <D CO a U a 4-1 1 4-1
W
1 4-1 1

Ml JC 00 O <D 00 rH CN >H eg 4-> rH 4-1 rH •H co CO

3 X 00 co 00 CO oc oc r 3 3 3
&
4-1

3 C X
O •H
0)
T3 B X
<13 3
3
o O'
3
o
^
O'
CO
0)
u
3
O
eg
03
a)
u
3
0
3
4-1

U 0
cO
cO

X O
O
0)
rH
“H
0
O
rH
•H
0J
O
U
0)
rH
*H
CO a eg OS rH CM Q rH Q rH 4J O rH •D Q co X < 4J <! 4J c 4J

[-SRN II--S9N

82
 N
l / — /
1 s \

o
NO rH rH O 1 ON rg LT) NO rH
i

|
er
—
i

1 00 H i

eg U rH rH pH 1 pH cn gf i ( 1 1
cn CO 1

e
-g.
1 1 1 1

4J 1 j
1
1
1
1

3 X /— 1

rs
1

^N
1 1

3 S en
/—>.
g g rs
rs pP~N /^s p-s /^S
O
/~N

o o oo 00
3 PQ NO
1

1 NO
ON cn
o CN 00 rH
I

1 cn
1

cn
|

•H • 1 •S’ eg i NO 1 i CN
• • • • •
0 rH eg • r-~ • *
3 4-4 3 no m g gf ON m eg O O rH CM CO
O1

4-4 X 3 NO NO gf '
—
1

1 mm rH eg
v-/ N—
eg
N^
1

i CO
—
1

| CO
1

1 00
pH O "g 3 > ' > N_^ j
v / |
'
j
v-/ v-/ |

H 3 a o o o o O O 1

oO 1

o
m oO 1 1

4J eg
PQ pH meg rH
1

00 rg rH 00 00
1

I O
X
1

1 vO vO
1

1 CO
oo o r- ON
i

eg 00 (ON r"- CO
m in m cn m gf
1 1 1 l

i
rH pH rH 1 rg | CM co |

o
X
HH
CO cr> eg eg fg r-g cn no NO NO cn rg
0 0)
00
U rH rl eg rH cn eg eg

m1

co
i-i

3 rO
> eg in no H n- <r co gf gf cn
m
rH IT 00 rH
m
X<
ph eg
rl rl rl
ox
oo rH ON cg gf en n- oo no

Ph
ON

<v

o
i 1 1 i i X
3 pQ
X 3 ONO CO i"~
i

Ov en
1

o CN eg
1

X l

o m 1

NT e^- 3
0 3 U X eg CN X i

1 eg
i

i eg CO en
1

1 en
1

1 eg eg
l

1 cn EH
00
X i 1 1 l 1

3
X3
i 1 1 l 1

Ph i 1 1 1 1

4-1 3 3 X>
3 > 3 eg on CN cn
eg — LT) X
i

i rH en
1

i 00
X
— ON
i
1

1 NO
1

1 gf m
X
1

m ON
X 3
Ph <3 3 en LO gf eg 00 O
X X X X X
i 1

X rH X
i 1 1

X
1

•H a rH 1 rH 1 1 l 1

Ph l l 1 1 3
l 1 1 1 l 3
l l 1 1 1 3
1 1 1 l Ph
O 00
3 O O Ov eg
X
i

i 00 NO
1

1 gf
X
NO
1

1 00
X
1

1 NO 00
1

1 O NO
rH X
3
u eg eg rH 1 rH eg l rH 1 1 eg 1 Ph
-X i 1 1 1 1

3 l 1 1 1 1 X
3 i 1 1 1 1 |

Oh 3 -a CO
3 n- on X i

rH
m X 1

On eg
1

o
l

X
eg
1
«n ON PQ

Xo
1

X 1

Xo 1 1

Xo
1

3 00 rH eg 00 in gf NO pa
a eg CN eg
|

l CN eg
1

l X rH X 1

1 eg
1

1 eg
1

1 co eg
3
3
CO
pH
3
> estimate

PC

o no
n eg m fg gf 00 rH
Cr- rg. X gf o NO
m
I

3 rl rl eg gf eg eg eg eg eg cn rg of
-i<5 cn

eg
co
0)
in
I
m eg

Ph
x
G percent

3
3
3
3 as
Jn rH
4 J cn 3
X 0 o o OO
gf ON m 00
1

o O o Oo o •

o O o o >
gr X
1

m
1 1
ON O
1 1

CO gf VO gf eg rg 4-4
3 60 m gf en 1

i mm
1

i mmm 1

1 fg
l

l eg en
1

1 fg 00 3 Ph
3 4c! 1 i 1 l 1 3 3
n 1 i /r 1 l
— /g
p 1 /T-S a 4->
regression

1 i 3 3 3 3 3 Ph 3
X X
1 i

X X
1

1 i •H 3
I
1 i 1

X X3
1 l eg eg eg eg NO PH
1
—-
i 1

1 l en 1 1
h*. 1 Ph
rH rH 00 O
G rH
NO cn O 3
iH O from
4-4
3
00 Ph rH •
• m •rl
Ph
I
co
•rl rH ON 3 PQ
X oo no 4-1 oo <J PQ
X3 X > S
eg X X CO m 1

X X 3 3 1 3 X e deviation

cO CJ 3 o o
X X
1 1

X3 3O
d) eg eg oo rH 33 1 1 Ph
3 00 Ph 1 +J
4-1 1 4-1
•rl ph Ph X 3 3 -o 3
3< ’rl
x x
-rl O O
1

1 XX 1 1

Ph 0 3 X3
eg Q pQ
3
1 3 3
O a
1

1 3 3 X
o-
Xo
4-4
e CO CO 33 33 x) Xa 1

X X 1

3 •»
3
X
1 1

)H 0) cfl (0 O O o 4-1 4-4 ns

px, 0
3 4= rH — i I O O eg
CO
4-1
3
X
U
1

1 3 3
1

1 pO X Standard

4-1 00 00 |3 1 3 3 1 3} 35 3
3 3 3 31 3 Ph O O' -O Ph O 3
1
J
o O O
q a
III ON |H
H 3
1

1 O o
1

1 35 3 O H
CO Pi PH PH 1
1
< < 1
1
EC EC 3

S3
NBS-1 14A (REV. 2-80
U.s. DEPT. OF COMM. 1. PUBLICATION OR 2. Performing Organ. Report No 3. Publication Oate
4. REPORT NO.
BIBLIOGRAPHIC DATA
NBSIR 82-2597 July 1983
SHEET (See instructions)
TITLE AND SUBTITLE
5.

Heat Release Rate Properties of Wood-Based Materials

6.

AUTHOR(S)
9. D. L. Chamberlain
PERFORMING ORGANIZATION (If joint or other than NBS, see in struction s) 7. Contract/Grant No.

national bureau of standards

DEPARTMENT OF COMMERCE 8. Type of Report & Period Covered
WASHINGTON, D.C. 20234
10.

SPONSORING ORGANIZATION NAME AND COMPLETE ADDRESS (Street, City. State. ZIP)
National Forest Products Association
1619 Massachusetts Avenue, NW
11.
Washington, D. C. 20036

SUPPLEMENTARY NOTES

|
Document describes a computer program; SF-185, FIPS Software Summary, is attached.
ABSTRACT (A 200-word or less factual summary of most significant information. If document includes a significant
bi bliography or literature survey, mention it here)

A background to the present heat release rate calorimetry is presented. Heat

release rates and cumulative heat release were measured for 16 different lumber
and wood products, using three different heat release rate instruments. The
effects of moisture content, exposure heat flux, density of product, and fire
retardant on rate of heat release were measured. The three small-scale heat
12.
release rate calorimeters were compared, and equations relating the data from
each were developed.

KEY WORDS (Six to twelve entries; alphabetical order; capitalize only proper names; and separate key words by semicolons)
Acoustical tile; Douglas fir; fire retardants; hardboard; heat release rate; heat
release rate calorimeters; irradiance; particle board; plywood; redwood; southern
pine
13. AVAILABILITY 14. NO.OF
PRINTED PAGES
(XX| Unlimited
| |
For Official Distribution. Do Not Release to NTIS 90
Order From Superintendent of Documents, U.S. Government Printing Office, Washington, D.C.
20402. 15. Price

Order From National Technical Information Service (NTIS), Springfield, VA. 22161
$11.50
USCOMM-DC 6043-P80