Designation: D 2863 – 00

Standard Test Method for

Measuring the Minimum Oxygen Concentration to Support
Candle-Like Combustion of Plastics (Oxygen Index)1
This standard is issued under the fixed designation D 2863; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (e) indicates an editorial change since the last revision or reapproval.

1. Scope * when using the Type A gas measurement and control device accuracy as
described in 6.4.
1.1 This test method covers a fire-test-response procedure.
This test method describes a procedure for measuring the 2. Referenced Documents
minimum concentration of oxygen that will just support 2.1 ASTM Standards:
flaming combustion in a flowing mixture of oxygen and D 618 Practice for Conditioning, Plastics and Electrical
nitrogen. Insulating Materials for Testing2
1.2 Methods are provided for testing materials that are D 1071 Test Methods for Volumetric Measurement of Gas-
structurally self-supporting in the form of vertical bars or sheet eous Fuel Samples3
up to 10.5 mm thick. These methods are suitable for solid, D 1622 Test Method for Apparent Density of Rigid Cellular
laminated or cellular materials characterized by an apparent Plastics2
density greater than 15 kg/m3. The methods may also be D 2444 Test Method for Impact Resistance of Thermoplas-
applicable to some cellular materials having an apparent tic Pipe and Fittings by Means of a Tup (Falling Weight)4
density of less than 15 kg/m3. A method is provided for testing D 4802 Specification for Poly(Methyl Methacrylate)
flexible sheet or film materials while supported vertically. Acrylic Plastic Sheet5
NOTE 1—Although this test method has been found applicable for E 176 Terminology of Fire Standards6
testing other materials, the precision of the test method has not been E 691 Practice for Conducting an Interlaboratory Study to
determined for these materials, or for specimen geometry’s and test Determine the Precision of a Test Method7
conditions outside those recommended herein.
2.2 ISO Standards:
1.3 This test method may be used to measure and describe 4589-2 Plastics—Determination of Flammability by Oxy-
the properties of materials, products, or assemblies in response gen Index—Part 2, Ambient Temperatures8
to heat and flame under controlled laboratory conditions and 7823-1 Poly(Methylmethacrylate) Sheets—Types, Dimen-
shall not be used to describe or appraise the fire hazard or fire sions and Characteristics—Part 1—Cast Sheets8
risk of materials, products, or assemblies under actual fire
conditions. However, results of this test may be used as 3. Terminology
elements of a fire risk assessment which takes into account all 3.1 Definitions:
of the factors which are pertinent to an assessment of the fire 3.1.1 ignition—for the purpose of this standard shall imply
hazard of a particular end use. the initiation of flaming combustion.
1.4 This standard does not purport to address all of the 3.1.2 oxygen index (OI)—the minimum concentration of
safety concerns, if any, associated with its use. It is the oxygen determined by the method in 9.1, expressed as volume
responsibility of the user of this standard to establish appro- percent, in a mixture of oxygen and nitrogen that will just
priate safety and health practices and determine the applica- support flaming combustion of a material initially at 23 6 2°C
bility of regulatory limitations prior to use. Specific hazards under the conditions of this test method.
statement are given in Note 2. 3.1.3 Definitions of terms relating to fire are in accordance
NOTE 2—Warning: During the course of combustion, gases or vapors,
with Terminology E 176.
or both, are evolved which may be hazardous to personnel. 3.2 Symbols Specific To This Test Method:
NOTE 3—This test method and ISO 4589-2 are technically equivalent
2
Annual Book of ASTM Standards, Vol 08.01.
3
Annual Book of ASTM Standards, Vol 05.05.
4
Annual Book of ASTM Standards, Vol 08.04.
1 5
This test method is under the jurisdiction of ASTM Committee D20 on Plastics Annual Book of ASTM Standards, Vol 08.03.
6
and is the direct responsibility of Subcommittee D20.30 on Thermal Properties Annual Book of ASTM Standards, Vol 04.07.
7
(Section D20.30). Annual Book of ASTM Standards, Vol 14.02.
8
Current edition approved July 10, 2000. Published October 2000. Originally Available from American National Standards Institute, 11 W. 42nd St., 13th
published as D 2863 – 70. Last previous edition D 2863 – 97. Floor, New York, NY 10036.

*A Summary of Changes section appears at the end of this standard.

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.

1
 D 2863
3.2.1 Co —oxygen concentration in percent volume. Therefore, the results are valid only for the fire-test-exposure
3.2.2 CF —final value of oxygen concentration in percent conditions described in this procedure.
volume. 6. Apparatus
3.2.3 Ci —each of the oxygen concentration percentages
used during measurement of the last six responses in the NT 6.1 Test Chimney, consisting of a heat-resistant glass tube of
series. 75 to 100 mm inside diameter and 450 to 500 mm height. The
3.2.4 O—neither the period or extent of burning exceeds the bottom of the chimney or the base to which the tube is attached
relevant limit specified in Table 1. shall contain noncombustible material to mix and distribute
3.2.5 X—the period or extent of burning exceeds the rel- evenly the gas mixture entering at this base. Glass beads 3 to
evant limit specified in Table 1. 5 mm in diameter in a bed 80 to 100 mm deep have been found
3.2.6 NL —series of “X” or “O” results. suitable. The chimney shall be mounted securely on the base to
3.2.7 NT —series of “X” or “O” results plus five (NT = N L prevent air leaks. See Fig. 1.
+ 5). NOTE 4—For tubes of 75 to 100 mm diameter, a cap converging to an
3.2.8 s*—standard deviation of the oxygen concentration. outlet of 40 mm diameter at a level at least 10 mm above the top of the
3.2.9 d—internal between oxygen concentration levels in cylindrical chimney has been found satisfactory for restricting the column
percent volume. opening.
NOTE 5—It is helpful to place a wire screen above the noncombustible
3.2.10 k—a factor to be determined from Table 2. material to catch falling fragments and aid in keeping the base of the
3.2.11 n—number of measurements of oxygen concentra- column clean.
tion.
6.2 Specimen Holder—Any small holding device that will
4. Summary of Test Method support the specimen at its base and hold it vertically in the
center of the chimney is acceptable. For physically self-
4.1 A small test specimen is supported vertically in a
supporting specimens, a typical arrangement (see Fig. 1)
mixture of oxygen and nitrogen flowing upwards through a
consists of a laboratory thermometer clamp inserted into the
transparent chimney. The upper end of the specimen is ignited
end of a glass tube held in place by glass beads or otherwise
and the subsequent burning behavior of the specimen is
firmly supported. For supported film or sheet test specimens,
observed to compare the period for which burning continues,
the specimen shall be supported by both vertical edges in a
or the length of specimen burnt, with specified limits for each
frame equivalent to that illustrated by Fig. 2, with reference
burning. By testing a series of specimens in different oxygen
marks at 20 and 100 mm below the top of the frame. The
concentrations, the minimum oxygen concentration is deter-
profile of the holder and its support shall be smooth to
mined.
minimize induction of turbulence in the rising flow of gas.
5. Significance and Use 6.3 Gas Supplies, comprising pressurized sources of oxygen
or nitrogen, or both, not less than 98 % pure or clean dry air, or
5.1 This test method provides for the measuring of the
both, (containing 20.9 % oxygen), as appropriate.
minimum concentration of oxygen in a flowing mixture of
6.3.1 The gas mixture entering the chimney shall have a
oxygen and nitrogen that will just support flaming combustion
moisture content of < 0.1 %, unless the results have been
of plastics. Correlation with burning characteristics under
shown to be insensitive to higher moisture levels in the gas
actual use conditions is not implied.
mixture. The gas supply system shall incorporate a drying
5.2 In this procedure, the specimens are subjected to one or
device, or provision for monitoring or sampling the gas supply
more specific sets of laboratory test conditions. If different test
for moisture content, unless the moisture content of the gas
conditions are substituted or the end-use conditions are
supplies is known to be acceptable.
changed, it may not be possible by or from this test to predict
changes in the fire-test-response characteristics measured. NOTE 6—It should not be assumed that bottled oxygen or nitrogen will
always contain < 0.1 % of water; moisture contents of 0.003 to 0.01 % are
typical for commercial supplies as filled bottles > 98 % pure, but as such
TABLE 1 Criteria for Oxygen Index MeasurementsA
bottled gases are depressured to below about 1 MPa, the moisture content
Alternative Criteria of the gas drawn off may rise above 0.1 %.
Period of
Test Specimen 6.4 Gas Measurement and Control Devices, suitable for
Ignition Procedure Burning
Type (See Table 3) Extent of BurningB
After establishing the following accuracies when the gas velocity
Ignition(s)
I, II, III, IV and VI A 180 50 mm below the top
through the chimney is 40 6 2 mm/s at 23 6 2°C;
(top surface ignition) of the specimen 6.4.1 Type A—Maintaining the volumetric concentration of
B 180 50 mm below the oxygen in the gas mixture entering the chimney with an
(propagating ignition) upper reference
mark accuracy of 60.5 % of the mixture and for adjusting the
V propagating ignition 180 80 mm below the concentration with a precision of 6 0.1 % of the mixture;
upper reference 6.4.2 Type B—Maintaining the volumetric concentration of
mark (on the frame)
A
oxygen in the gas mixture entering the chimney with an
These criteria do not necessarily produce equivalent oxygen index results for
specimens of differing shape or tested using different ignition conditions or
accuracy of 61.0 % of the mixture and for adjusting the
procedures. concentration with a precision of 60.5 % of the mixture.
B
The extent of burning is exceeded when any part of the visibly burning portion
of a specimen, including burning drips descending the vertical faces, passes the NOTE 7—Systems of measurement and control that have proved satis-
level indicated in the column. factory include the following:

2
 D 2863
TABLE 2 Determination of k
1 2 3 4 5 6

Responses for the Last Five Values of k for which the first NL determinations are: Responses for the Last Five
Measurements (a) O OO OOO OOOO MeasurementsA

XOOOO −0.55 −0.55 −0.55 −0.55 OXXXX

XOOOX −1.25 −1.25 −1.25 −1.25 OXXXO
XOOXO 0.37 0.38 0.38 0.38 OXXOX
XOOXX −0.17 −0.14 −0.14 −0.14 OXXOO
XOXOO 0.02 0.04 0.04 0.04 OXOXX
XOXOX −0.50 −0.46 −0.45 −0.45 OXOXO
XOXXO 1.17 1.24 1.25 1.25 OXOOX
XOXXX 0.61 0.73 0.76 0.76 OXOOO
XXOOO −0.30 −0.27 −0.26 −0.26 OOXXX
XXOOX −0.83 −0.76 −0.75 −0.75 OOXXO
XXOXO 0.83 0.94 0.95 0.95 OOXOX
XXOXX 0.30 0.46 0.50 0.50 OOXOO
XXXOO 0.50 0.65 0.68 0.68 OOOXX
XXXOX −0.04 0.19 0.24 0.25 OOOXO
XXXXO 1.60 1.92 2.00 2.01 OOOOX
XXXXX 0.89 1.33 1.47 1.50 OOOOO
A
Values of k for which the first NL determinations are (b) X, XX, XXX, and XXXX are as given in Table 2 opposite he appropriate response in Column 6, but with the
sign of k reversed, that is: OI = CF − kd (see 9.1).

(a) For Type A—Needle valves on individual and mixed gas sup- NOTE 9—It is likely that, for materials where the oxygen index is
ply lines, a paramagnetic oxygen analyzer that continuously known to within 62 % by volume 15 test specimens will be sufficient.
samples the mixed gas, and a flowmeter to indicate when the gas However, for materials of unknown oxygen index, or which exhibit erratic
flow through the chimney is within the required limits; burning characteristics, between 15 and 30 test specimens are likely to be
(b) For Type A or B—Calibrated orifices, gas pressure regulators
required.
and pressure gages on the individual gas supply lines; or
(c) For Type B—Needle valves and calibrated flowmeters on the NOTE 10—If non-standard size specimens are used, a difference in
individual gas supply lines. oxygen index may result.
6.4.3 These systems require calibration after assembly to 7.1.1 Ensure that the surfaces of the specimens are clean and
ensure that the compounded errors of the component parts do free from flaws that could affect burning behavior, for example,
not exceed the requirements of 6.4. peripheral molding flash or burrs from machining.
6.4.3.1 Means shall be provided for checking or ensuring 7.1.2 The edges of the specimens shall be smooth and free
that the temperature of the gas mixture entering the chimney is from fuzz or burrs of material left from machining or molding.
23 6 2°C. If this involves an internal probe, its position and 7.1.3 Record position and orientation of test specimens with
profile shall be designed to minimize induction of turbulence respect to any asymmetry in the sample material (see Note 12).
within the chimney. NOTE 11—Oxygen index results are likely to be significantly affected
6.5 Flame Igniter, comprising a tube, with an inside diam- by differences in burning behavior, due to material inhomogeneity (for
eter of 2 6 1 mm, that can be inserted into the chimney to example, different levels of shrinkage when heated for specimens cut in
apply the test flame. different directions from asymmetrically-oriented thermoplastics film).
6.5.1 The flame fuel shall be methane or natural gas of at 7.2 For preparation of Type VI specimens, use the rolling
least 97 % purity, without premixed air. The fuel supply shall tool described in 6.8.
be adjusted so that the flame projects 16 6 4 mm vertically 7.2.1 The rolled film is obtained by first inserting one corner
downwards from the outlet when the tube is vertical within the of the film into the slit of the thin film rolling tool (see 6.8) and
chimney and the flame is burning within the chimney atmo- then winding the film around the wire in a spiral of 45°, as
sphere. shown in Fig. 3. Ensure that the 45° angle is maintained during
6.6 Timing Device, capable of measuring periods up to 5 the winding process so that the film reaches exactly to the end
min with an accuracy of 60.5 s. of the tool, to produce a test piece of the correct length. After
6.7 Fume Extraction System, having sufficient ventilation or the winding is finished, tape the last end of the roll while the
exhaust to remove fumes or soot expelled from the chimney material is still on the stainless steel wire to prevent loosening.
without disrupting the gas-flow rate or temperatures in the Then pull the wire out of the rolled film.
chimney. 7.2.2 Cut off the rolled film at a distance of 20 mm from the
NOTE 8—If soot-generating materials are being tested, the glass chim- top end. See Fig. 4.
ney may require cleaning to maintain good visibility, and the gas inlets, or 7.3 For monitoring the distance over which burning occurs,
inlet screen may also require cleaning to function properly. mark the specimen with transverse lines at one or more levels
6.8 Thin Film Rolling Tool—A 2 6 0.1 mm stainless steel which are dependent upon the specimen form and the ignition
wire with a 0.3 6 0.05 mm slit at one end, equivalent to that procedure to be used. Structurally self-supporting specimens
illustrated in Fig. 3. are preferably marked on at least two adjacent faces. If wet
inks are used, the marks shall be dry before the specimen is
7. Test Specimens ignited.
7.1 Cut or mold at least 15 specimens. Use Table 3 to 7.3.1 Test specimens of Type I, II, III, IV or VI are to be
determine specimen dimensions. tested in accordance with Procedure A (see 8.8), and shall be

3
 D 2863

1. Burning Specimen 6. Glass Beads in a Bed 11. Pressure Gage

2. Clamp with Rod Support 7. Brass Base 12. Precision Pressure Regulator
3. Igniter 8. Tee 13. Filter
4. Wire Screen 9. Cut-Off Valve 14. Needle Valve
5. Ring Stand 10. Orifice in Holder 15. Rotameter
FIG. 1 Typical Equipment Layout

marked 50 mm from the end to be ignited. humidity (RH) immediately prior to use.
7.3.2 The reference marks for testing specimens of Type V 7.5 For cellular materials, the density shall be determined in
are carried by the supporting frame (see Fig. 2), but it is accordance with Test Method D 1622.
acceptable to mark thermally stable materials at 20 mm and at NOTE 12—It is possible that the oxygen index samples of cellular
100 mm from the end to be ignited, for convenience. materials that contain volatile flammable blowing agents that diffuse from
7.3.3 If specimens of Type I, II, III, IV and VI are to be the sample will change with time.
tested in accordance with Procedure B (see 8.8 and 8.10), they
shall be marked at 10 mm and at 60 mm from the end to be 8. Procedure A
ignited. 8.1 Calibrate the flow-measuring system using a water-
7.4 Unless otherwise specified, each test specimen shall be sealed rotating drum meter (wet test meter) in accordance with
conditioned for at least 88 h at 23 6 2°C and 50 6 5 % relative Test Method D 1071 or by equivalent calibration devices. The

4
 D 2863

FIG. 2 Frame Design

TABLE 3 Test Specimen Dimensions

Test Dimensions
Specimen Length, Width, Thickness, Material Form
TypeA mm mm mm
I 80 to 150 10 6 0.5 4 6 0.25 for molding materials
II 80 to 150 10 6 0.5 10 6 0.5 for cellular materials
IIIB 80 to 150 10 6 0.5 # 10.5 for sheet materials
IVC 70 to 150 6.5 6 0.5 3 6 0.25 alternative size for
self-supporting molding
or sheet materials
VB 140 6 5 52 6 0.5 #10.5 for flexible film or sheet
VIBD 140 to 200 20 0.02 to for thin film; limited to film
0.10 that can be rolled by the
wire specified in 6.8
A
Test specimens of Types I, II, III, and IV are suitable for materials that are
self-supporting at these dimensions. Test specimens of Form V and VI are suitable
for materials that require support during testing.
B
Results obtained using Type III, V, and VI test specimens may only be
comparable for specimens of the same form and thickness. It is assumed that the
amount of variation in thickness for such materials will be controlled by other
standards.
C
The Type IV (ASTM) specimen will eventually be discontinued in favor of the
Type 1 (ISO) specimen.
D
The test specimen of Type VI is suitable for thin film that is self-supporting
when it is rolled. Dimensions in the table are of the specimen size from which the
rolled form is made. If the film is very thin, it is possible that proper results will only
FIG. 3 Wire With a Slit be obtained if two or more layers are combined in the preparation of the roll to
obtain proper results.

maximum interval between recalibration shall be six months. A

cast PMMA specimen shall be used as a verification material at 8.2 Maintain the ambient temperature for the test apparatus
least once a month. See Annex A1 for calibration method. at 23 6 2°C. If necessary, keep the test specimens in an

5
 D 2863
NOTE 14—Some materials exhibit a non-flaming type of combustion
(for example, glowing combustion) instead of, or at a lower oxygen
concentration than that required for, flaming combustion. When testing
such materials, it is necessary to identify the type of combustion for which
the oxygen index is required or measured.
8.9 Test Method A—Top Surface Ignition:
8.9.1 For top surface ignition, the igniter is used to initiate
burning only on the top surface of the upper end of the
specimen.
8.9.2 Apply the lowest visible part of the flame to the top of
the specimen using a sweeping motion, if necessary, to cover
the whole surface, but taking care not to maintain the flame
against the vertical faces or edges of the specimen. Apply the
flame for up to 30 s, removing it every 5 s, just briefly, to
observe whether or not the entire top surface of the specimen
FIG. 4 Rolled Film
is burning.
enclosure at 23 6 2°C and 50 6 5 % RH and take the test 8.9.3 Consider the specimen to be ignited, and commence
specimens out of the enclosure just before testing. measurement of the period and distance of burning, as soon as
8.3 Recalibrate equipment components, more frequently if removal of the igniter, after a contact period increment of 5 s,
necessary (see 8.1 and Annex A1). reveals, burning supported by the whole of the top end of the
8.4 Select an initial concentration of oxygen to be used. specimen.
Whenever possible, base the initial concentration on experi- 8.10 Test Method B—Propagating Ignition:
ence of results for similar materials. Alternatively, try to ignite 8.10.1 For propagating ignition, the igniter is used to
a test specimen in air, and note the burning behavior. If the produce burning across the top and partially down the vertical
specimen burns rapidly, select an initial concentration of about faces of the specimen.
18 % oxygen; if the test specimen burns gently or unsteadily, 8.10.2 Lower and move the igniter sufficiently to apply the
select an initial oxygen concentration of about 21 %; if the visible flame to the end face of the specimen and also, to a
specimen does not continue to burn in air, select an initial depth of approximately 6 mm, to its vertical faces. Continue to
concentration of at least 25 %, depending upon the difficulty of apply the igniter for up to 30 s, with interruptions for
ignition or the period of burning before extinguishing in air. inspection of the specimen every 5 s, until its vertical faces are
8.5 Ensure that the test chimney is vertical (see Fig. 1). burning steadily or until the visibly burning portion first
Mount a specimen vertically in the center of the chimney so reaches the level of the upper reference mark on the support
that the top of the specimen is at least 100 mm below the open frame or, if used for specimens of Type I, II, III, IV or VI on
top of the chimney and the lowest exposed part of the specimen the specimen.
is at least 100 mm above the top of the gas distribution device 8.10.3 Consider the specimen to be ignited, for the purpose
at the base of the chimney (see Fig. 1 or Fig. 2 as appropriate). of measuring the period and extent of burning, as soon as any
8.6 Set the gas mixing and flow controls so that an oxygen/ part of the visibly burning portion reaches the level of the
nitrogen mixture at 23 6 2°C, containing the desired concen- upper reference mark.
tration of oxygen, is flowing through the chimney at a rate 40 NOTE 15—The burning portion includes any burning drips that run
6 2 mm/s. Let the gas flow purge the chimney for at least 30 down the surface of the specimen.
s prior to ignition of each specimen, and maintain the flow
without change during ignition and combustion of each speci- 8.11 Assessing the Burning Behavior of Individual Test
men. Specimens:
8.7 Verify the temperature at the lower end of the chimney 8.11.1 Commence measurement of the period of burning as
to be 23 6 2°C and record the oxygen concentration used as soon as the specimen has been ignited in accordance with 8.9
the volume percent calculated according to the equations given or 8.10 as applicable, and observe its burning behavior. If
in Annex A2. burning ceases but spontaneous re-ignition occurs within 1 s,
8.8 Select one of two alternative ignition procedures which continue the observation and measurements.
are dependent upon the specimen form as follows: 8.11.2 If neither the period or extent of burning exceeds the
8.8.1 For specimen Types I, II, III, IV and VI use Method A, relevant limit specified in Table 1 for the applicable specimen,
top surface ignition, as described in 8.9. note the duration and extent of burning. This is recorded as an
8.8.2 For specimen Type V, use Method B, propagating “O” response. (See Appendix X1.)
ignition, as described in 8.10. 8.11.3 Alternatively, if either the period or extent of burning
exceeds the relevant limit specified in Table 1 note the burning
NOTE 13—For tests on materials that exhibit steady burning and spread behavior accordingly, and extinguish the flame. This is re-
of combustion in oxygen concentrations at, or close to, their oxygen index
value, or for structurally self-supporting specimens of #3 mm thickness,
corded as “X” response. (See Appendix X1.)
Procedure B (with specimens marked in accordance with 7.3.2) may be 8.11.4 Note also the burning characteristics of the material,
found to give more consistent results than Test Method A. Test Method B for example, dripping, charring, erratic burning, glowing
may then be used for specimens of Type I, II, III, IV or VI. combustion or after-glow.

6
 D 2863
8.11.5 Remove the specimen and clean, as necessary, any Appendix X1, Part 2).
surfaces within the chimney or on the igniter that have become NOTE 20—Where experience has shown that the requirements of 8.14.6
contaminated with soot, etc. Allow the chimney to regain a are usually satisfied by a value of d other than 0.2 %, that value may be
temperature of 23 6 2°C, or replace it with another so selected as the initial value of d.
conditioned. Install the next specimen. 8.14.4 Test four more specimens, in accordance with 8.4-
NOTE 16—For screening purposes, it is acceptable to invert sufficiently 8.12 inclusive, maintaining d = 0.5 %, and note the Co used
long specimens or trim them to remove the burnt end, and re-use them. for, and response of, each specimen. Designate the oxygen
Results from such specimens can save material when establishing an concentration used for the last specimen as CF .
approximate value for the minimum oxygen concentration required for
8.14.5 These four results together with the last result from
combustion, but cannot be included among those used for estimation of
the oxygen index, unless the specimen is reconditioned at the temperature 8.14.2 (that is, that which differed in response from that of
and humidity appropriate for the material involved. 8.14.1) constitute the remainder of the NT series, so that:
8.12 Selecting Successive Oxygen Concentrations: NT 5 N L 1 5
8.12.1 The procedure described in 8.13 and 8.14 is based See example in Appendix X1, Part 2.
upon the “up-and-down method for small samples,” using the 8.14.6 Calculate the estimated standard deviation, s*, of the
specific case where NT − N L = 5 (see 8.14.2 and 8.14.4), with oxygen concentration measurements from the last six re-
an arbitrary step size for certain changes to be made in the sponses in the NT series (including CF), in accordance with 9.3.
oxygen concentration used. If the condition is satisfied, calculate the oxygen index in
8.12.2 During the testing, select the oxygen concentration to accordance with 9.1, otherwise:
be used for testing the next test specimen as follows: 2s*
8.12.2.1 Decrease the oxygen concentration if the burning 3 , d, 1.5s*
behavior of the preceding specimen gave an “X” response,
8.14.6.1 If d < 2s*/3, repeat 8.14.2-8.14.6 using increased
otherwise
values for d, until the condition is satisfied, or
8.12.2.2 Increase the oxygen concentration if the preceding
8.14.6.2 If d > 1.5s*, repeat 8.14.2-8.14.6 using decreased
specimen gave an “O” response.
values for d, until the condition is satisfied, except that d shall
8.12.3 Choose the size of the change in oxygen concentra-
not be reduced below 0.2 % unless so required by the relevant
tion in accordance with 8.13 and 8.14, as appropriate.
material specification.
8.13 Determining the Preliminary Oxygen Concentration:
8.13.1 Repeat the procedures specified in 8.5 to 8.6 inclu- 9. Calculation
sive, using oxygen concentration changes of any convenient 9.1 Oxygen Index:
step size, until the oxygen concentrations, in percent volume, 9.1.1 Calculate the oxygen index ( OI), expressed as a
have been found that differ by #1.0 % and of which one gave percentage by volume, from the following relationship:
an“ O” response and the other an “X” response. From this pair OI 5 CF 1 kd
of oxygen concentrations, note that which gave the “O”
response as the preliminary oxygen concentration level and where:
then proceed in accordance with 8.14. CF = the final value of oxygen concentration, in percent
volume to one decimal place, used in the series of N
NOTE 17—The two results, at oxygen concentrations #1.0 % apart,
T measurements performed in accordance with 8.14
which give opposite responses, do not have to be from successive
specimens. and noted in accordance with 8.14.4,
NOTE 18—At times, the concentration that gave the “O” response will d = the interval, in percent volume to at least one
not be lower than that which gave the “X” response. Such apparent decimal place, between oxygen concentration levels
inconsistencies, that are likely to be caused by the variability of the test, used and controlled in accordance with 8.14, and
the equipment, or the material, are not uncommon. k = a factor to be obtained from Table 2 as described in
NOTE 19—A format convenient for recording the information required 9.2.
by this and subsequent clauses is illustrated in Appendix X1. 9.1.2 For the purpose of calculation of s*, as required by
8.14 Oxygen Concentration Changes: 8.14.6 and 9.3, the OI shall be calculated to two decimal
8.14.1 Using, again, the preliminary oxygen concentration places.
8.13, test one specimen by repeating 8.5 to 8.11 inclusive. 9.1.3 For the purpose of reporting OI results, express OI
Record both the oxygen concentration (Co) used and the values to the nearest 60.1 %, with exact intermediate results
response, “X” or “O”, as the first of the NL and of the NT series being rounded downwards.
of results. 9.2 Determination of k:
8.14.2 Change the oxygen concentration, in accordance 9.2.1 The value and sign of k are dependent upon the pattern
with 8.12 using concentration changes (d) of 0.2 % (see Note of the responses of specimens tested in accordance with 8.14.
21) of the total gas mixture to test further specimens in Determine them from Table 2 as follows:
accordance with 8.4-8.12 inclusive, noting the values of Co and 9.2.1.1 If the response of the specimen tested according to
corresponding responses, until a different response to that 8.14.1 was “O”, so that the first contrary response (see 8.14.2)
obtained in 8.14.1 is recorded. was an “X”, refer to Column 1 of Table 2 to select the row for
8.14.3 The result from 8.14.1 plus those of like response which the last four response symbols correspond to those found
from 8.14.2 constitute the NL series of results. (See example in when testing in accordance with 8.14.4. The value and sign of

7
 D 2863
k will be that shown in Column 2, 3, 4 or 5 for which the 11.1.1 A reference to this test method;
number of “O”s shown in row (a) of the table corresponds to 11.1.2 Date of testing;
the number of “O” responses found for the NL series, in 11.1.3 A statement that test results relate only to the
accordance with 8.14.1 and 8.14.2, or behavior of the test specimens under the conditions of this test
9.2.1.2 If the responses of the specimen tested according to method and that these results must not be used to infer the fire
8.14.1 was “X”, so that the first contrary response was an “O”, hazards of the material in other forms or under other fire
refer to the sixth column of Table 2 to select the row for which conditions;
the last four response symbols correspond to those found when 11.1.4 Identification of the material tested, including, where
testing in accordance with 8.14.4. The value of k will be that relevant, the type of material, density, previous history, the
shown in Column 2, 3, 4, or 5 for which the number of “X”s specimen orientation with respect to any anisotropy in the
shown in row (b) of the table corresponds to the number of “X” material or sample, and the date of manufacture with lot
responses found for the NL series, in accordance with 8.14.1 number;
and 8.14.2 but the sign of k must be reversed, so that negative 11.1.5 The oxygen index (OI) as determined in 9.1;
values shown in Table 2 for k become positive, and vice versa. 11.1.6 The test specimen type or dimensions;
11.1.7 The gas measurement and control device accuracy
NOTE 21—An example of the determination of the calculation of an OI
is given in Annex A2. (Type A or B);
11.1.8 The ignition procedure used (Test Method A or B);
9.3 Standard Deviation of Oxygen Concentration Measure- 11.1.9 When Procedure B is used, indicate the relevant
ments: specified minimum oxygen index (OI) of the material and
9.3.1 For the purposes of 8.14.6 calculate the estimated report if the material tested had a lower or higher oxygen index
standard deviation, s*, of oxygen concentration measurements (OI);
from the relationship: 11.1.10 If applicable, the estimated standard deviation and
s* 5 F (~Ci 2 OI! 2
n21 G |n$ the oxygen concentration increment used, if other than 0.5 %;
11.1.11 A description of any relevant characteristics or
behavior, such as charring, dripping, severe shrinkage, erratic
where: burning, after-glow; and
Ci = in turn, each of the percent oxygen concentrations
11.1.12 Any variations from the requirements of this test
used during measurement of the last six responses in
method.
the NT series of measurements;
OI = the oxygen index value, calculated in accordance 12. Precision and Bias
with 9.1; and 12.1 Table 4 is based on a round robin9 conducted in 1999
n = the number of measurements of oxygen concentra- in accordance with Practice E 691, involving 8 materials tested
tion contributing to ((Ci − OI) 2. by 12 laboratories. For each material, the samples were
NOTE 22—For this test method, n = 6, in accordance with 8.14.6. For prepared by the supplier of the material and conditioned at the
n < 6, the test method loses precision. For n > 6, alternative statistical laboratories that tested them. Each laboratory obtained 2 test
criteria would apply. results for each material. All laboratories utilized gas measure-
Table 2 values of k for calculating oxygen index concentration from ment and control devices in accordance with 6.4.1 (Type A) for
determinations made by Dixon’s “Up-and-Down” method.

10. Procedure B—Comparison With a Specified 9

Supporting data are available from ASTM Headquarters. Request RR: D20-
Minimum Value for Oxygen Index (Short Procedure) 1031.

10.1 If the actual oxygen index of a material is needed or in

TABLE 4 Oxygen Index (OI), %
case of a dispute, Procedure A shall be used.
10.2 Set up the apparatus and test specimen in accordance Specimen
Material Procedure Average SrA SRB rC RD
Type
with 8.1-8.7 except that the specified minimum concentration
PMMA-1 III A 17.7 0.10 0.23 0.28 0.65
of oxygen shall be selected for the purposes of 8.4. PMMA-2 III A 17.8 0.13 0.25 0.37 0.70
10.3 Ignite the test piece in accordance with 8.8. PVC, plasticized I A 38.4 0.60 2.03 1.67 5.68
10.4 Using up to three specimens, assess the burning ABS, FR I A 26.8 0.58 1.09 1.61 3.07
PF, thermoset I A 49.7 0.36 1.74 1.01 4.87
behavior of each specimen in accordance with 8.11. PS, foam II A 20.9 0.44 0.97 1.22 2.71
10.4.1 If for at least two out of the three specimens thus PC, sheet V B 26.1 0.31 1.42 0.88 3.97
tested the flame is extinguished before the relevant criteria PET, film VI A 21.9 0.64 1.48 1.79 4.15
from Table 1 are exceeded, that is, an “O” response is recorded, A
Sr is the within-laboratory standard deviation for the indicated material. It is
obtained by pooling the within-laboratory standard deviations of the test results
then record that the oxygen index of the material is not less from all of the participating laboratories:
than the specified value. Otherwise, record that the oxygen Sr 5 @@~S1!2 — 1 ~Sn!2#/n#1/2
index of the material is less than the specified value or B
SR is the between-laboratories reproducibility, expressed as standard devia-
determine the oxygen index in accordance with Section 8, as tion:
Sr 5 @S12 1 SL2#1/2
appropriate. where SL is the standard deviation of laboratory means.
C
11. Report r is the within-laboratory critical interval between two test results = 2.83 Sr.
D
R is the between-laboratories critical interval between two test results = 2.8 3
11.1 Report the following information: SR.

8
 D 2863
accuracy and precision. been calculated from a large enough body of data, and for test
12.2 Table 4 does not include three laboratories that partici- results that were averages from testing 2 specimens for each
pated in the round robin and utilized measurement and control test result, then:
devices in accordance with 6.4.2 (Type B) for accuracy and 12.3.1 Repeatability—Two test results obtained within one
precision. The results from these laboratories could not be laboratory shall be judged not equivalent if they differ by more
incorporated into this precision statement, due to the limited than the “r” value for that material. The concept of “r” is the
number of participants to comply with Practice E 691 guide- interval representing the critical difference between two test
lines. Therefore, the resulting precision precision is provided in results for the same material, obtained by the same operator
Annex A3. using the same equipment in the same laboratory.
NOTE 23—Two statistically designed interlaboratory round robins for 12.3.2 Reproducibility—Two test results obtained by differ-
precision evaluation were conducted earlier, one with 18 laboratories and ent laboratories shall be judged not equivalent if they differ by
5 materials (supporting data are available from ASTM Headquarters,
request RR: D20-0102) and one with 29 laboratories and 12 materials.9
more than the “R” value for that material. The concept of “R”
The first study indicated a higher standard deviation for specimens with is the interval representing the critical difference between two
higher oxygen index, while the second study indicated a dependence of test results for the same material, obtained by different opera-
precision with the type of specimen used. tors using different equipment in different laboratories.
NOTE 24—Caution: The explanation of “r” and “R” (12.3-12.3.3) are 12.3.3 Any judgement in accordance with 12.3.1 or 12.3.2
only intended to present a meaningful way of considering the approximate
precision of this test method. The test results and precision in Table 4 would have an approximate 95 % (0.95) probability of being
should not be applied to acceptance or rejection of materials, as these data correct.
apply only to the materials tested in the round robin and are unlikely to be 12.4 Bias—There are no recognized standards on which to
rigorously representative of other lots, formulations, conditions, materials, base an estimate of bias for this test method.
or laboratories. Users of this test method should apply the principles
outlined in Practice E 691 to generate data specific to their materials and
laboratory (or between specific laboratories). The principles of 12.3-12.3.3 13. Keywords
would then be valid for such data. 13.1 candle-like combustion; minimum oxygen concentra-
12.3 Concept of “r” and“ R” in Table 4—If Sr and SR have tion; oxygen; oxygen concentration; oxygen index

ANNEXES

(Mandatory Information)

A1. CALIBRATION OF EQUIPMENT

A1.1 Leak Tests—Leak tests shall be carried out thoroughly 0.1 % of mixture, either by sampling the chimney atmosphere
on all joints where leaks could change the oxygen concentra- for analysis or by using an independently calibrated analyzer in
tion levels in the chimney from the concentration levels set or situ. If an oxygen analyzer is incorporated in the equipment,
indicated. this shall be calibrated using the following gases, each of
which shall conform with 6.3 for purity and moisture content:
A1.2 Gas-Flow Rates:
A1.3.1.1 Any two gases selected from the following: nitro-
A1.2.1 The system for indicating the gas-flow rate through
gen; oxygen; or clean air; and
the chimney, to satisfy 6.4 and 8.6 shall be checked using a
calibrated flow meter, or an equivalent device, with an accu- A1.3.1.2 A mixture of any two of the preceding gases
racy equivalent to 60.2 mm/s flow rate through the chimney. having an oxygen concentration within the range of concen-
A1.2.2 The flow rate shall be estimated by dividing the total trations to be used for most test specimens.
gas-flow rate through the chimney by the cross sectional area
of the bore of the chimney, for example, by using the following A1.4 Verification of Complete Equipment—For monthly
equation: verification, in accordance with 8.1, use Type III specimens of
qv the PMMA. The PMMA shall be a non-modified transparent
F 5 1.27 3 10 6 cast sheet based on a homopolymer of methylmethacrylate in
D2
accordance with Specifications D 4802, Category A-1
where: (ISO7823-1 for Cast Sheets). The PMMA shall have an oxygen
F = flow rate through the chimney, mm/s, index (OI) between 17.0 and 18.5.
qv = total gas-flow at 23 6 2°C through the chimney, L/s,
D = diameter of the bore of the chimney, mm. NOTE A1.1—Other PMMA sheets such as cast sheets based on copoly-
mer of methylmethacrylate, extruded or melt calendered PMMA sheets
A1.3 Oxygen Concentration: may give a different burning behavior depending on the comonomer used,
A1.3.1 The concentration of oxygen in the mixture of gases its contents or molecular weight which effects melt behavior when being
flowing into the chimney shall be checked to an accuracy of burned.

9
 D 2863

A2. CALCULATION OF OXYGEN CONCENTRATION

A2.1 Oxygen concentrations required for the purposes of 98.5 % purity or with nitrogen containing 0.5 % of oxygen, the oxygen
Section 8 shall be calculated according to the equation: concentration, in percent by volume, should be calculated using the
relationship:
100 Vo
Co 5 V 1 V (A2.1) 98.5 V8o 1 20.9 V8A 1 0.5 V8N
o N
Co 5 V8o 1 V8A 1 V8N (A2.2)
where:
Co = oxygen concentration, in percent by volume, where:
Vo = volume of oxygen per volume of mixture, at 23°C, V8o = volume of oxygen stream used, per volume of
and mixture,
VN = volume of nitrogen per volume of mixture, at 23°C. V8A = volume of air stream used, per volume of mixture,
and
V8N = volume of nitrogen stream used, per volume of
NOTE A2.1—If an oxygen analyzer is used, the oxygen concentration mixture; assuming that the streams are at the same
should be determined using the readout from the particular instrument pressure at 23°C.
used.
NOTE A2.2—If the result is calculated from flow or pressure data for For mixtures based on two gas streams, V8o, V8A , or V8N
individual gas streams contributing to the mixture, it is necessary to allow becomes zero, as appropriate.
for the proportion of oxygen present in streams other than a pure oxygen
supply. For example, for mixtures made using air mixed with oxygen of

10
 D 2863

A3. PRECISION AND BIAS USING TYPE B CONTROL DEVICE ACCURACY AND PRECISION

A3.1 Table A3.1 is based on a limited round robin9 A3.3.3) are only intended to present a meaningful way of
conducted in 1999, along with the round robin in Section 14. considering the approximate precision of this test method. The
This limited round robin does not comply with Practice E 691 test results and precision in Table A3.1 should not be applied to
because only 3 laboratories participated on a limited number of acceptance or rejection of materials, as these data apply only to
materials using the Type B control devices for accuracy and the materials tested in the round robin and are unlikely to be
precision. For each material, the samples were prepared by the rigorously representative of other lots, formulations, condi-
supplier of the material and conditioned at the laboratories that tions, materials, or laboratories. Users of this test method
tested them. Each laboratory obtained either 1 or 2 test results should apply the principles in Practice E 691 to generate data
for each material. All laboratories utilized gas measurement specific to their materials and laboratory (or between specific
and control devices in accordance with 6.4.2 (Type B) for laboratories). The principles of A3.3-A3.3.3 would then be
accuracy and precision. valid for such data.
A3.2 Caution—The explanations of “r” and “R” (A3.3- A3.3 Concept of “r” and “R” in Table 4—If Sr and SR have
been calculated from a large enough body of data, and for test
TABLE A3.1 Oxygen Index (OI), %A results that were averages from testing 2 specimens for each
Specimen test result, then:
Material Procedure Average S rB SRC rD RE
Type
A3.3.1 Repeatability—Two test results obtained within one
PMMA-1 III A 17.1 0.20 0.43 0.57 1.21 laboratory shall be judged not equivalent if they differ by more
PMMA-2 III A 17.4 0.00 0.61 0.00 1.72
PVC, plasticized I A 48.0 0.27 7.84 0.77 22.0 than the “r” value for that material. The concept of “r” is the
ABS, FR I A 26.5 0.27 11.1 0.95 31.1 interval representing the critical difference between two test
PF, thermoset I A 52.2 0.15 0.82 0.42 2.30 results for the same material, obtained by the same operator
PS, foam II A 23.3 0.00 3.04 0.00 8.50
A
using the same equipment in the same laboratory.
Based on data from only two laboratories.
B
Sr is the within-laboratory standard deviation for the indicated material. It is
A3.3.2 Reproducibility—Two test results obtained by dif-
obtained by pooling the within-laboratory standard deviations of the test results ferent laboratories shall be judged not equivalent if they differ
from all of the participating laboratories: by more than the “R” value for that material. The concept of
Sr 5 @@~S1!2 — 1 ~Sn!2#/n#1/2
C
SR is the between-laboratories reproducibility, expressed as standard devia-
“R” is the interval representing the critical difference between
tion: two test results for the same material, obtained by different
Sr 5 @S12 1 SL2#1/2 operators using different equipment in different laboratories.
where SL is the standard deviation of laboratory means. A3.3.3 Any judgement in accordance with A3.3.1 or A3.3.2
D
r is the within-laboratory critical interval between two test results = 2.8 3 Sr.
E
R is the between-laboratories critical interval between two test results = 2.8 3 would have an approximate 95 % (0.95) probability of being
SR. correct.

APPENDIX

(Nonmandatory Information)

X1. TYPICAL TEST RESULTS SHEET

11
D 2863

12
 D 2863

SUMMARY OF CHANGES

This section identifies the location of selected changes to this test method. For the convenience of the user,
Committee D20 has highlighted those changes that may impact the use of this test method. This section may also
include descriptions of the changes or reasons for the changes, or both.

D 2863 – 00: (3) Added requirements for the PMMA verification material in
(1) A round robin was conducted in 1999 and Section 12 A1.4.
(Precision and Bias) was replaced with the precision results for (4) Deleted Appendix X2, which cited a previous round robin,
the Type A measurement and control device. and added Note 23 to indicate that this old round robin data is
(2) The 1999 round robin also included limited data using the available from ASTM Headquarters.
precision results for the Type B measurement and control
device, which was added as Annex A3.

ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@astm.org (e-mail); or through the ASTM website
(www.astm.org).

13