Designation: C 1363 – 05

Standard Test Method for

Thermal Performance of Building Materials and Envelope
Assemblies by Means of a Hot Box Apparatus1
This standard is issued under the fixed designation C 1363; 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 box apparatus has been designed, constructed and operated in

1.1 This test method establishes the principles for the design the ideal manner, no further calibration or adjustment would be
of a hot box apparatus and the minimum requirements for the necessary. As such, the hot box is considered a primary method
determination of the steady state thermal performance of and the uncertainty of the result is analyzed by direct evalua-
building assemblies when exposed to controlled laboratory tion of the component measurement uncertainties of the
conditions. This method is also used to measure the thermal instrumentation used in making the measurements.
performance of a building material at standardized test condi- 1.3.1 In an ideal hotbox test of a homogenous material there
tions such as those required in material Specifications C 739, is no temperature difference on either the warm or cold
C 764, C 1224 and Practice C 1373. specimen faces to drive a flanking heat flow. In addition, there
1.2 This test method is used for large homogeneous or would be no temperature differences that would drive heat
non-homogeneous specimens. This test method applies to across the boundary of the metering chamber walls. However,
building structures or composite assemblies of building mate- experience has demonstrated that maintaining a perfect guard/
rials for which it is possible to build a representative specimen metering chamber balance is not possible and small corrections
that fits the test apparatus. The dimensions of specimen are needed to accurately characterize all the heat flow paths
projections or recesses are controlled by the design of the hot from the metering chamber. To gain this final confidence in the
box apparatus. Some hot boxes are limited to planar or nearly test result, it is necessary to benchmark the overall result of the
planar specimens. However, larger hot boxes have been used to hot box apparatus by performing measurements on specimens
characterize projecting skylights and attic sections. See 3.2 for having known heat transfer values and comparing those results
a definition of the test specimen and other terms specific to this to the expected values.
method. 1.3.2 The benchmarking specimens are homogeneous pan-
els whose thermal properties are uniform and predictable.
NOTE 1—This test method replaces Test Methods C 236, the Guarded These panels, or representative sections of the panels, have had
Hot Box, and C 976, the Calibrated Hot Box which have been withdrawn.
their thermal performance measured on other devices that are
Test apparatus designed and operated previously under Test Methods
C 236 and C 976 will require slight modifications to the calibration and directly traceable or have been favorably compared to a
operational procedures to meet the requirements of Test Method C 1363.2 national standards laboratory. For example, a Test Method
C 177 Hot Plate, a Test Method C 518 Heat Meter or another
1.3 A properly designed and operated hot box apparatus is
Test Method C 1363 Hot Box will provide adequate specimens.
directly analogous to the Test Method C 177 guarded hot plate
Note that the use of Test Method C 518 or similar apparatus
for testing large specimens exposed to air induced temperature
creates additional uncertainty since those devices are calibrated
differences. The operation of a hot box apparatus requires a
using transfer standards or standard reference materials. By
significant number of fundamental measurements of tempera-
performing this benchmarking process, the hot box operator is
tures, areas and power. The equipment performing these
able to develop the additional equations that predict the
measurements requires calibration to ensure that the data are
magnitude of the corrections to the net heat flow through the
accurate. During initial setup and periodic verification testing,
specimen that account for any hot box wall loss and flanking
each measurement system and sensor is calibrated against a
loss. This benchmarking provides substantial confidence that
standard traceable to a national standards laboratory. If the hot
any extraneous heat flows can be eliminated or quantified with
sufficient accuracy to be a minor factor of the overall uncer-
1
This test method is under the jurisdiction of ASTM Committee C16 on Thermal tainty.
Insulation and is the direct responsibility of Subcommittee C16.30 on Thermal 1.4 In order to ensure an acceptable level of result uncer-
Measurement. tainty, persons applying this test method must possess a
Current edition approved May 1, 2005. Published June 2005. Originally
approved in 1997. Last previous edition approved in 1997 as C 1363 – 97. knowledge of the requirements of thermal measurements and
2
Footnotes in the text are supplied to clarify the discussion only, and as such, are testing practice and of the practical application of heat transfer
not mandatory.

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

1
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
theory relating to thermal insulation materials and systems. 1.12 This test method does not specify all details necessary
Detailed operating procedures, including design schematics for the operation of the apparatus. Decisions on material
and electrical drawings, shall be available for each apparatus to sampling, specimen selection, preconditioning, specimen
ensure that tests are in accordance with this test method. mounting and positioning, the choice of test conditions, and the
1.5 This test method is intended for use at conditions typical evaluation of test data shall follow applicable ASTM test
of normal building applications. The naturally occurring out- methods, guides, practices or product specifications or govern-
side conditions in temperate zones range from approximately mental regulations. If no applicable standard exists, sound
−48 to 85°C and the normal inside residential temperatures is engineering judgment that reflects accepted heat transfer prin-
approximately 21°C. Building materials used to construct the ciples must be used and documented.
test specimens shall be pre-conditioned, if necessary, based 1.13 This test method applies to steady-state testing and
upon the material’s properties and their potential variability. does not establish procedures or criteria for conducting dy-
The preconditioning parameters shall be chosen to accurately namic tests or for analysis of dynamic test data. However,
reflect the test samples intended use and shall be documented several hot box apparatuses have been operated under dynamic
in the report. Practice C 870 may be used as a guide for test (non-steady-state) conditions after additional characterization
specimen conditioning. The general principles of the hot box (1). Additional characterization is required to insure that all
method can be used to construct an apparatus to measure the aspects of the heat flow and storage are accounted for during
heat flow through industrial systems at elevated temperatures. the test. Dynamic control strategies have included both peri-
Detailed design of that type of apparatus is beyond the scope of odic or non-periodic temperature cycles, for example, to follow
this method. a diurnal cycle.
1.6 This test method permits operation under natural or 1.14 This test method does not permit intentional mass
forced convective conditions at the specimen surfaces. The transfer of air or moisture through the specimen during
direction of airflow motion under forced convective conditions measurements. Air infiltration or moisture migration can alter
shall be either perpendicular or parallel to the surface. the net heat transfer. Complicated interactions and dependence
1.7 The hot box apparatus also is used for measurements of upon many variables, coupled with only a limited experience in
individual building assemblies that are smaller than the meter- testing under such conditions, have made it inadvisable to
ing area. Special characterization procedures are required for include this type testing in this standard. Further considerations
these tests. The general testing procedures for these cases are for such testing are given in Appendix X1.
described in Annex A11. 1.15 This standard does not purport to address all of the
1.8 Specific procedures for the thermal testing of fenestra- safety concerns, if any, associated with its use. It is the
tion systems (windows, doors, skylights, curtain walls, etc.) are responsibility of the user of this standard to establish appro-
described in Test Method C 1199 and Practice E 1423. priate safety and health practices and determine the applica-
1.9 The hot box has been used to investigate the thermal bility of regulatory limitations prior to use.
behavior of non-homogeneous building assemblies such as
structural members, piping, electrical outlets, or construction 2. Referenced Documents
defects such as insulation voids. 2.1 ASTM Standards: 4
1.10 This test method sets forth the general design require- C 168 Terminology Relating to Thermal Insulating Materi-
ments necessary to construct and operate a satisfactory hot box als
apparatus, and covers a wide variety of apparatus construc- C 177 Test Method for Steady-State Heat Flux Measure-
tions, test conditions, and operating conditions. Detailed de- ments and Thermal Transmission Properties by Means of
signs conforming to this standard are not given but must be the Guarded-Hot-Plate Apparatus
developed within the constraints of the general requirements. C 236 Test Method for Steady-State Thermal Performance
Examples of analysis tools, concepts and procedures used in of Building Assemblies by Means of a Guarded Hot Box
the design, construction, characterization, and operation of a C 518 Test Method for Steady-State Heat Flux Measure-
hot box apparatus is given in Refs (1-34).3 ments and Thermal Transmission Properties by Means of
1.11 The hot box apparatus, when constructed to measure the Heat Flow Meter Apparatus
heat transfer in the horizontal direction, is used for testing C 739 Specification for Cellulosic Fiber (Wood Base)
walls and other vertical structures. When constructed to mea- Loose-fill Thermal Insulation
sure heat transfer in the vertical direction, the hot box is used C 764 Specification for Mineral Fiber Loose-fill Thermal
for testing roof, ceiling, floor, and other horizontal structures. Insulation
Other orientations are also permitted. The same apparatus may C 870 Practice for Conditioning of Thermal Insulating Ma-
be used in several orientations but may require special design terials
capability to permit repositioning to each orientation. What- C 976 Test Method for Steady-State Thermal Performance
ever the test orientation, the apparatus performance shall first of Building Assemblies by Means of a Calibrated Hot Box
be verified at that orientation with a specimen of known
thermal resistance in place.
4
For referenced ASTM standards, visit the ASTM website, [Link], or
contact ASTM Customer Service at service@[Link]. For Annual Book of ASTM
3
The boldface numbers in parentheses refer to the list of references at the end of Standards volume information, refer to the standard’s Document Summary page on
this standard. the ASTM website.

2
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
C 1045 Practice for Calculating Thermal Transmission Section 7. For purposes of this method, a single material whose
Properties from Steady-State Heat Flux Measurements properties are being evaluated is also defined as a building
C 1058 Practice for Selecting Temperatures for Reporting element.
and Evaluating Thermal Properties of Thermal Insulations 3.2.2 metered specimen—the element that fills the boundary
C 1114 Test Method for Steady-State Thermal Transmission of the metering chamber opening. The metered specimen can
Properties by Means of the Thin-Heater Apparatus be: (1) the entire building element when it is the same size as
C 1130 Practice for Calibrating Thin Heat Flux Transducers the metering chamber opening dimensions; (2) the building
C 1132 Practice for Calibration of the Heat Flow Meter element and the surround panel in the case when the building
Apparatus element is smaller than the opening; (3) a portion of the
C 1199 Test Method for Measuring the Steady State Ther- building element when the building element is larger than the
mal Transmittance of Fenestration Systems Using Hot Box opening.
Methods
3.2.3 test specimen—that portion of the metered specimen
C 1224 Specification for Reflective Insulation for Building
for which the thermal properties are to be determined. The test
Applications
specimen can be: (1) the entire building element when it is the
C 1373 Practice for Determination of Thermal Resistance
same size as the metering chamber dimensions; (2) the building
of Attic Insulation Systems Under Simulated Winter Con-
element only in the case when the building element is smaller
ditions
than the opening; (3) that portion of the building element that
C 1558 Guide for Development of Standard Data Records
is within the metered area when the building element is larger
for Computerization of Thermal Transmission Test Data
than the opening.
for Thermal Insulation
E 230 Standard Temperature-Electromotive Force (EMF) 3.2.4 surround panel—the surround panel, often called the
Tables for Thermocouples mask, is a uniform structure having stable thermal properties
E 283 Test Method for Rate of Air Leakage Through that supports the building element within the metering area.
Exterior Windows, Curtain Walls and Doors The material shall be homogeneous and low thermal conduc-
E 1423 Practice for Determining the Steady State Thermal tivity that both supports the test specimen and provides a
Transmittance of Fenestration Systems uniform, reproducible heat flow pattern at the edges of the
E 1424 Test Method for Determining the Rate of Air Leak- metering chamber perimeter.
age Through Exterior Windows, Curtain Walls, and Doors 3.2.5 self-masking—a hot box configuration which occurs
Under Specified Pressure and Temperature Differences when the metering chamber opening is less than the building
Across the Specimen element dimensions. This configuration may be used when the
2.2 Other Documents: thermal behavior of the building element is such that it is
ASHRAE Handbook of Fundamentals, Latest Edition, “self-masking.” This means that the lateral heat flow at the
American Society of Heating, Refrigerating and Air Con- edges of the metering chamber can be minimized. With proper
ditioning Engineers, Inc.5 design and control of the metering chamber, this condition is
ISO Standard 8990 Thermal Insulation Determination of easily obtained for test specimens that are homogeneous, or
Steady State Thermal Properties—Calibrated and Guarded while not homogeneous, do not contain highly conductive
Hot Box, ISO 8990-1994(E)6 elements that extend beyond the boundary of the metering
ISO Standard 12567 Thermal Performance of Windows and chamber. This configuration was previously known as a
Doors—Determination of Thermal Transmittance by Hot “guarded hot box.”
Box Method, ISO 12567-20006 3.2.6 masked—a hot box configuration which occurs when
the metering chamber opening is the same or greater than the
3. Terminology test specimen dimensions. This configuration must be used
3.1 Definitions—The definitions of terms relating to insu- when the test specimen cannot be “self-masking.” Here, the
lating materials and testing are governed by Terminology perimeter of the test specimen requires a separate mask, called
C 168, unless defined below. All terms discussed in this test a surround panel, constructed to eliminate lateral heat flow.
method are those associated with thermal properties of the Note that the hot box wall acts as a mask when the test
tested specimen, unless otherwise noted. specimen and the metering chamber dimensions are the same.
3.2 Definitions of Terms Specific to This Standard: The case where the hot box walls act as the mask was
3.2.1 building element—a portion of a building assembly, previously known as a “calibrated hot box.”
selected for test, in the expectation that it will exhibit the same 3.2.7 heat transfer—the energy transfer that takes place
thermal behavior as the larger building assembly that it between material bodies as a result of a temperature difference.
represents. Guidance for the selection process is given in 3.2.8 metering box wall loss, Qmw—the time rate of heat
exchange through the walls of the metering box.
[Link] Discussion—The metering box wall loss must be
5
Available from American Society of Heating, Refrigerating, and Air- subtracted from, or added to, the heat input to the metering
Conditioning Engineers, Inc. (ASHRAE), 1791 Tullie Circle, NE, Atlanta, GA
chamber as part of the determination of the net heat flow
30329.
6
Available from American National Standards Institute (ANSI), 25 W. 43rd St., through the metered specimen. A more complete discussion of
4th Floor, New York, NY 10036. the Metering Box Wall Loss is provided in Annex A3.

3
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
3.2.9 flanking loss, Qfl—the time rate of heat exchange from Qmw = time rate of heat flow from the metering chamber
the metering chamber to the climatic chamber and or guard to the guard chamber through the metering box
chamber that is due to the two-dimensional heat transfer at the walls, W
interface of the test specimen and the surround panel or Qrad = time rate of heat flow to a surface by radiation, W
metering box wall. Qs = time rate of heat flow through the metered
[Link] Discussion—The flanking loss must also be sub- specimen, W
tracted from, or added to, the heat input to the metering Qsp = time rate of heat flow through the surround
chamber as part of the determination of the net heat flow panel, W
through the metered specimen. A more complete discussion of R = surface to surface thermal resistance, m2·K/W
Rc,env = surface to environment thermal resistance, cold
the Flanking Loss is provided in Annex A4.
side, (m2·K)/W
3.3 Symbols—The following are symbols, terms, and units Rh,env = surface to environment thermal resistance, hot
used in this test method. side, (m2·K)/W
3.3.1 Some of these symbols can be modified for a particu- Rs = surface to surface thermal resistance, (m2·K)/W
lar application by the subscript attached. Ru = overall thermal resistance, m2·K/W
S = heat flux transducer calibration factor (a function
of temperature), W/(m2·V)
A = metering box opening area, m2 ta = volume averaged temperature of ambient air, K
Aeff = effective area of the metering box wall, m2 or °C
Ain = inside surface area of the metering chamber, m2 tb = area weighted average temperature of the baffle
As = effective area of the test specimen, m2 surface, K or °C
C = surface to surface thermal conductance, tc = volume averaged air temperature 75 mm or more
W/(m2·K) from the cold side surface, K or °C
E = voltage output of heat flux transducer or thermo- tenv = the effective environmental temperature includ-
couple, V ing radiation, conduction, and convection effects,
hc,env = surface to environment heat transfer coefficient, K or °C (see Annex A9)
cold side, W/(m2·K) th = space averaged air temperature 75 mm or more
hconv = convective surface heat transfer coefficient, from the hot side surface, K or °C
W/(m2·K) tm = average specimen temperature, average of two
hh,env = surface to environment heat transfer coefficient, opposite surface temperatures, K or °C
hot side, W/(m2·K) t1 = area weighted average temperature of specimen
hrad = radiative surface heat transfer coefficient, hot surface, K or °C
W/(m2·K) t2 = area weighted average temperature of the speci-
HC = equivalent heat capacity of an object, (W·h)/ men cold surface, K or °C
(kg·K) th = panel thickness at the location of the flanking
L = length of the heat flow path (usually, the thick- loss path, m
ness of the test panel), m Dt = temperature difference between two planes of
m = the slope of the metering box thermopile equa- interest, K or °C
tion, W/V Dta-a = temperature difference—air to air, K or °C
M = mass of an object, kg Dts-env = temperature difference—surface to the environ-
q = time rate of heat flow through a unit area, W/m2 ment, K or °C
Q = time rate of net heat flow through the metering Dts-s = temperature difference—surface to surface, K or
box opening, W °C
Qcp = time rate of heat flow through a known calibra- U = thermal transmittance, W/(m2·K)
tion panel, W l = apparent thermal conductivity, W/(m·K)
Qconv = time rate of heat flow to a surface by convection, e = total hemispherical surface emittance, (dimen-
W sionless)
Qcool = time rate of heat input to the metering chamber s = Stefan-Boltzmann Constant for Thermal Radia-
by the cooling coils, W tion, 5.673 3 10-8 W/( m2·K4)
Qf = time rate of heat input to the metering chamber teff = effective thermal time constant of the combined
by the fans, W apparatus and specimen, s
Qfl = time rate of heat flow from the metering chamber Sei = total edge length on the inside walls of the
to the climatic chamber, other than that through metering chamber, m
the metering box walls or metered specimen, W 3.3.2 Subject Modifiers:
Qh = time rate of heat input to the metering chamber
by the heaters, W
Qin = the net time rate of heat flow into the metering 1 = hot side surface
chamber, equals the algebraic sum of the heat 2 = cold side surface
from the fans, heaters and cooling coils, W a = ambient condition

4
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

a-a = air to air difference A · ~t2 2 tenv,c!

Rc,env 5 Q (7)
ap = apparatus
b = baffle 3.4.6 Surface Heat Transfer Coeffıcient, hi,env—Often called
c = cold surface conductance or film coefficient. The subscripts h and c
conv = convection are used to differentiate between hot side and cold side surface
cool = cooling energy heat transfer coefficients respectively. The coefficients are
eff = effective or equivalent property calculated as follows:
env = environment
fl = flanking path Q
hh,env 5 (8)
h = hot A · ~tenv,h 2 t1!
i = index Q
in = inside hc,env 5 (9)
A · ~t2 2 tenv,c!
m = mean or average value
NOTE 4—The surface heat transfer coefficient, hi,env, and the corre-
mw = metering box wall
sponding surface resistance, Ri,env, (see 3.4.5) are reciprocals, that is, their
o = null or zero condition product is unity.
out = outside
rad = radiation 3.4.7 Surface Coeffıcient Determination—An expanded dis-
s = surface cussion of the interactions between the radiation and convec-
sp = surround panel tive heat transfer at the surfaces of the test specimen is included
s-a = surface to air difference in Annex A9. The material presented in Annex A9 must be used
s-env = surface to the environment difference to determine the magnitude of the environmental temperatures.
s-s = surface to surface difference These temperatures are required to correct for the radiation
t = test heat flow from the air curtain baffle.
u = overall 3.4.8 Whenever the heat transfer is greatly different from
3.4 Equations—The following equations are listed here to one area to another or the surface area of one surface of the test
simplify their use in the Calculations section of this test specimen is significantly larger than the projected area, or the
method. detailed temperatures profiles are unknown, only the net heat
3.4.1 Overall Thermal Resistance, Ru—The overall thermal transfer through the specimen is meaningful. In these cases,
resistance is equal to the sum of the resistances of the specimen only the calculation of the overall resistance, Ru, and transmis-
and the two surface resistances. It is calculated as follows: sion coefficient, U, are permitted.
A · ~tenv,h 2 tenv,c! 3.4.9 Apparent Thermal Conductivity of a Homogeneous
Ru 5 Q 5 Rc 1 R 1 Rh (1) Specimen, l:
3.4.2 Thermal Transmittance, U—(sometimes called overall Q·L
l5 (10)
coefficient of heat transfer). It is calculated as follows: A · ~ t1 2 t 2 !
Q NOTE 5—Materials are considered homogeneous when the value of the
U5 (2) thermal conductivity is not significantly affected by variations in the
A · ~tenv,h 2 tenv,c!
thickness or area of the specimen within the range of those variables
I/U 5 ~1/hh! 1 ~1/C! 1 ~1/hc! (3) normally used.
NOTE 2—Thermal transmittance, U, and the corresponding overall
thermal resistance, Ru, are reciprocals, that is, their product is unity.
4. Summary of Test Method
4.1 This test method establishes the principles for the design
3.4.3 Thermal Resistance, R:
of a hot box apparatus and the minimum requirements for the
A · ~ t1 2 t 2 ! determination of the steady state thermal performance of
R5 Q (4)
building assemblies when exposed to controlled laboratory
3.4.4 Thermal Conductance, C: conditions. At the minimum, the hot box apparatus shall be
Q able to measure the rate of heat flow through a building
C5
A · ~ t1 2 t 2 !
(5) element of known area for known test conditions while limiting
extraneous heat flows. The apparatus is required to establish
NOTE 3—Thermal resistance, R, and the corresponding thermal con- and maintain a desired steady temperature difference across the
ductance, C, are reciprocals; that is, their product is unity. These terms
test specimen for the period of time. The elapsed time required
apply to specific bodies or constructions as used, either homogeneous or
heterogeneous, between two specified isothermal surfaces. is that necessary to ensure constant heat flow and steady
temperatures, and, for an additional period adequate to measure
3.4.5 Surface Resistance, Ri,env—The surface resistance is these quantities to the desired accuracy.
the resistance, at the surface, to heat flow to the environment 4.2 To determine the conductance, C, the transmittance, U,
caused by the combined effects of conduction, convection and or the resistance, R, of any specimen, it is necessary to know
radiation. The subscripts h and c are used to differentiate the area, A, the net heat flow,Q and the temperature differences,
between hot side and cold side surface resistances respectively. Dt, all of which shall be determined under such conditions that
Surface resistances are calculated as follows: the flow of heat is steady.
A · ~tenv,h 2 t1! 4.3 The area and temperatures are measured directly. The
Rh,env 5 (6)
Q ent heat flow Q, however, cannot be directly measured. To

5
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
determine the net heat flow through the metered specimen, a 5.3 This test method does not establish test conditions,
five-sided metering box is placed with its open side against one specimen configuration, or data acquisition details but leaves
face of the metered specimen. these choices to be made in a manner consistent with the
4.4 If there were no net heat exchange across the walls that specific application being considered. Data obtained by the use
of the metering box and the flanking loss around the metered of this test method is representative of the specimen perfor-
specimen is negligible, then the heat input from the fan and mance only for the conditions of the test. It is unlikely that the
heaters minus any cooling coil heat extraction from the test conditions will exactly duplicate in-use conditions and the
metering box is a measure of the net heat flow through the user of the test results must be cautioned of possible significant
metered specimen. differences. For example, in some specimens, especially those
4.5 Since it is difficult to achieve the condition described in containing empty cavities or cavities open to one surface, the
4.4, the hot box apparatus must be designed to obtain an overall resistance or transmittance will depend upon the
accurate measure of the net metered specimen heat flow. The temperature difference across the test specimen due to internal
net heat transfer through the metered specimen is determined convection.
from the net measured heat input to the metering chamber, 5.4 Detailed heat flow analysis shall precede the use of the
corrected for the heat flow through the metering chamber walls hot box apparatus for large, complex structures. A structure that
and flanking loss for the specimen at the perimeter of the contains cavity spaces between adjacent surfaces, for example,
metering area. Where the metering chamber opening contains an attic section including a ceiling with sloping roof, may be
a building element smaller than the opening masked by a difficult to test properly. Consideration must be given to the
surround panel, the net heat transfer through the surround panel effects of specimen size, natural air movement, ventilation
is subtracted from the metered specimen heat flow in order to effects, radiative effects, and baffles at the guard/meter inter-
determine the net heat flow through the building element. face when designing the test specimen.
4.6 The heat flow rate through the metering chamber walls 5.5 For vertical specimens with air spaces that significantly
is limited by the use of highly insulated walls, by control of the affect thermal performance, the metering chamber dimension
surrounding ambient temperature, or by use of a temperature- shall match the effective construction height. If this is not
controlled guard chamber. possible, horizontal convection barriers shall be installed inside
4.7 The portion of the building element or specimen frame the specimen air cavities at the metering chamber boundaries to
outside the boundary of the metering area, exposed to the prevent air exchange between the metering and guarding areas.
guarding space temperature, constitutes a passive guard to The operator shall note in the report any use of convection
minimize flanking heat flow in the building element near the barriers. The report shall contain a warning stating that the use
perimeter of the metering area (see Annex A2). of the barriers might modify the heat transfer through the
4.8 Both the metering chamber wall flow and the flanking system causing significant errors. For ceiling tests with low
loss corrections are based upon a series of characterization density insulations, the minimum lateral dimension of the
tests, using specimens of known thermal properties. These tests specimen shall be at least several times the dimension of the
cover the range of anticipated performance levels and test expected convection cells.
conditions. While it is possible to estimate the magnitude of 5.6 Since this test method is used to determine the total heat
these corrections using numerical techniques and material flow through the test area demarcated by the metering box, it is
properties of the components, the accuracy of those corrections possible to determine the heat flow through a building element
must be verified by characterization measurements. (See An- smaller than the test area, such as a window or representative
nex A2 through Annex A11 for details.) area of a panel unit, if the parallel heat flow through the
remaining surrounding area is independently determined. See
5. Significance and Use Annex A8 for the general method.
5.7 Discussion of all special conditions used during the test
5.1 A need exists for accurate data on heat transfer through shall be included in the test report (see Section 12).
insulated structures at representative test conditions. The data
are needed to judge compliance with specifications and regu- 6. Apparatus
lations, for design guidance, for research evaluations of the 6.1 Introduction—The design of a successful hot box appa-
effect of changes in materials or constructions, and for verifi- ratus is influenced by many factors. Before beginning the
cation of, or use in, simulation models. Other ASTM standards design of an apparatus meeting this standard, the designer shall
such as Test Methods C 177 and C 518 provide data on review the discussion on the limitations and accuracy, Section
homogeneous specimens bounded by temperature controlled 13, discussions of the energy flows in a hot box, Annex A2, the
flat impervious plates. The hot box test method is more suitable metering box wall loss flow, Annex A3, and flanking loss,
for providing such data for large building elements, usually of Annex A4. This, hopefully, will provide the designer with an
a built-up or composite nature, which are exposed to appreciation of the required technical design considerations.
temperature-controlled air on both sides. 6.2 Definition of Location and Areas—The major compo-
5.2 For the results to be representative of a building con- nents of a hot box apparatus are (1) the metering chamber on
struction, only representative sections shall be tested. The test one side of the specimen; (2) the climatic chamber on the other;
specimen shall duplicate the framing geometry, material com- (3) the specimen frame providing specimen support and
position and installation practice, and orientation of construc- perimeter insulation; and (4) the surrounding ambient space.
tion (see Section 7). These elements shall be designed as a system to provide the

6
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
desired air temperature, air velocity, and radiation conditions resistivity, low heat capacity and high air flow resistance.
for the test and to accurately measure the resulting net heat Polystyrene or other closed cell foam materials have been used
transfer. A diagram of the relative arrangement of those spaces since they combine both high thermal resistivity, good me-
is shown in Fig. 1. chanical properties, and ease of fabrication. One potential
6.2.1 The basic hot box apparatus has been assembled in a problem with some foam is that they exhibit time dependent
wide variation of sizes, orientations and designs. Two configu- thermal properties that would adversely affect the thermal
rations have been historically used for a majority of the stability of the apparatus. Problems associated with the use of
designs. The first is the self-masking hot box which has a these materials are avoided by using materials that are initially
controlled “guard” chamber surrounding the metering box. An aged prior to assembly, or by periodic chamber verification, or
example of this configuration is presented in Fig. 2. by using impermeable faced foam materials with sealed edges
6.2.2 The second configuration is the masked hot box. This to greatly minimize the aging effects.
configuration can also be considered as a special case of the 6.5 Metering Chamber:
guarded hot box in which the surrounding ambient is used as 6.5.1 The minimum size of the metering box is governed by
the guard chamber. An additional design consideration for the the metering area required to obtain a representative test area
masked hot box design is that the metering chamber walls shall for the specimen (see 7.2) and for maintenance of reasonable
have sufficient thermal resistance to reduce the metering box test accuracy. For example, for specimens incorporating air
wall loss to an acceptable level. The masked design is generally spaces or stud spaces, the metering area shall span an integral
used for testing of large specimens. Fig. 3 shows an example of number of spaces (see 5.5). The depth of the metering box shall
a masked apparatus for horizontal heat transfer. be no greater than that required to accommodate the air curtain,
NOTE 6—The two opposing chambers or boxes are identified as the radiation baffle and the equipment required to condition and
metering chamber and the climatic chamber. In the usual arrangement, the circulate the air. Measurement errors in testing with a hot box
temperature of the metering chamber is greater than that of the climatic apparatus are, in part, proportional to the length of the
chamber and the common designations of “hot side” and “cold side” perimeter of the metering area and inverse to metering area.
apply. In some apparatus, either direction of heat flow may apply. The relative influence of the perimeter length diminishes as
6.3 Apparatus Size—The overall apparatus shall be sized to metering area is increased. Experience on testing homogeneous
match the type of specimens anticipated for testing (see 7.2). materials, has demonstrated that for the “guarded,” self-
For building assemblies, it shall accommodate representative masking hot box configuration, the minimum size of the
sections. Generally, the maximum accuracy is obtained when metering area is 3 times the square of the metered specimen
the specimen size matches that of the metering chamber while thickness or 1 m2, whichever is larger (18). From the same
the climatic chamber also matches or is larger. experience base, for the “calibrated,” masked box configura-
tion, a minimum metering area size is 1.5 m2. For non-
NOTE 7—A large apparatus is desirable in order to minimize perimeter
effects in relation to the metered area, but a large apparatus may also homogeneous specimens, the size requirements are more
exhibit longer equilibrium times, thus, a practical compromise must be significant.
reached. Typical heights for wall hot boxes are 2.5 to 3 m with widths 6.5.2 The purpose of the metering chamber is to provide for
equal to or exceeding the height. Floor/ceiling hot boxes up to 4 by 6 m the control and measurement of air temperatures and surface
have been built. coefficients at the face of the specimen under prescribed
6.4 Construction Materials—Materials used in the con- conditions and for the measurement of the net heat transfer
struction of the hot box apparatus shall have a high thermal through specimen. The usual arrangement is a five-sided

FIG. 1 Typical Hot Box Apparatus Schematic—Definition of Locations and Areas

7
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

FIG. 2 Typical Guarded Hot Box Schematic

FIG. 3 Typical Calibrated Hot Box Apparatus

chamber containing airflow baffles, electrical heaters, cooling electrical power to the heaters and blowers minus the cooling
coils (if desired), and an air circulation system. At steady state energy extraction, corrected for the heat passing through the
conditions, the heat transfer through the specimen equals the chamber walls and flanking the specimen. Both the metering

8
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
box wall loss and flanking loss are determined from character- contact surface of the metering chamber with the specimen
ization measurements (see Section 8 and Annex A2-Annex shall be narrowed to the minimum width necessary to hold the
A9). seal. A maximum width of 13 mm, measured parallel to the
6.5.3 To minimize measurement errors, several require- specimen surface plane, shall be used as a guide for design.
ments are placed upon the metering chamber walls and the Periodic inspection of the sealing system is recommended in
adjoining ambient space: order to confirm its ability to provide a tight seal under test
[Link] The metering chamber heat flow corrections, which conditions.
are estimated for design purpose using the equations of Annex 6.5.4 Since one basic principle of the test method is to
A2-Annex A4, must be kept small, by making the metering box measure the heat flow through the metering box walls, ad-
wall area small, keeping its thermal resistance high or by equate controls and temperature-monitoring capabilities are
minimizing the temperature difference across the wall (see essential. Small temperature gradients through the walls occur
Note 8). due to the limitations of controllers. Since the total wall area of
[Link] With proper design, the metering box wall loss are the metering box is often more than twice the metering area of
controlled to be as low as 1 or 2 % of the heat transfer through the specimen, these small temperature gradients through the
the specimen. The metering box wall loss shall never be greater walls cause substantial heat flows totaling a significant fraction
than 10 % of the specimen heat transfer. In any case, the of the heat input to the metering box. For this reason, the
minimum thermal resistance of the metering chamber walls metering box walls shall be instrumented to serve as a heat
shall be greater than 0.83 m2K/W. flow transducer so that heat flow through them can be
NOTE 8—The 10 % limit is based upon design analysis of existing hot minimized and measured. A correction for metering chamber
boxes. The choice of construction of the metering chamber can only be wall loss shall be applied in calculating test results. The use of
made after review of the expected test conditions in which metering box one of the following methods is required for monitoring
wall loss and associated uncertainties are considered in relation to the metering box wall loss.
anticipated energy transfer through the metered specimen and its desired
maximum uncertainty. The influence of the guarding temperature upon the NOTE 10—The choice of transducer types and mounting methods used
ability to maintain steady temperatures within the metering chamber must to measure the heat flow through the metering chamber walls is guided by
also be considered in choosing between highly insulated walls and a the hot box design. However, they must provide adequate coverage and
tightly controlled guard space conditioning. output signal to quantify the metering box wall loss during testing (see
[Link]).
[Link] However large the metering box wall loss is, the
uncertainty of the resulting metering box wall loss correction to [Link] The walls may be used as heat flow transducers by
the net heat flow shall not exceed 0.5 % of the net heat flow application of a large number of differential thermocouples
through the specimen. In some designs, it has been necessary connected between the inside and outside surfaces of the
to use a partial guard to reduce the metering chamber box wall metering chamber walls. Care must be taken when determining
loss. locations of the differential thermocouples, as temperature
[Link] For best results, the heat transfer through the meter- gradients on the inside and outside of the metering box walls
ing chamber walls shall be uniform so that a limited number of are likely to exist and have been found to be a function of
heat flux transducers or differential thermocouples can be used metering and climatic chamber air velocities and temperatures.
to characterize the heat flow from each representative area. Care must also be taken when determining the number of
This goal is best approximated by the use of a monolithic, differential thermocouples. Based upon a survey of hot box
uniform insulation uninterrupted by highly conducting struc- operators (18), a minimum of five differential thermocouple
tural members, and by eliminating any localized hot or cold pairs per m2 of metering box wall area shall be used. The
sources from the adjoining space. No highly conductive thermocouple junctions shall be located directly opposite each
structural members shall be within the insulation. Thermal other and, preferably, located at the centers of approximately
bridges, structural cracks, insulation voids, air leaks and equal areas. Small pieces of foil, having surface emittance
localized hot or cold spots from the conditioning equipment matching the remainder of the box walls, may be attached to
inside the metering chamber walls shall be avoided. the thermocouples to facilitate the thermal contact with the
wall surface. The junctions and the attached thermocouple
NOTE 9—One method of constructing satisfactory chamber walls is by
gluing together large blocks of an aged, uniform low thermal conductivity
wires shall be flush with, and in thermal contact with, the
cellular plastic insulation such as extruded polystyrene foam. A thin surface of the wall for at least a 100 mm distance from the
covering of reinforced plastic or coated plywood is recommended to junctions. The thermocouple pairs are connected in series to
provide durability, moisture and air infiltration control. In addition to form a thermopile in which the individual voltages are summed
using a high thermal resistance, the designer must also recognize that wall to give a single output or read out individually in cases where
heat storage capacity is also a governing factor in hot box wall design. significant differences may occur or be expected in the local
[Link] To ensure uniform radiant heat transfer exposure of heat flow levels.
the specimen, all surfaces which exchange radiation with the [Link] As an alternative, separate heat flux transducers are
specimen shall have a total hemispherical emittance greater placed on the metering chamber walls. Care must be taken in
than 0.8. choosing and installing the transducers to ensure that the
[Link] In applications where the metering chamber contacts thermal resistance of the wall and its surface emittance remain
the specimen, an airtight seal between the specimen and essentially unchanged. The transducers shall be initially cali-
metering wall shall be provided. The cross section of the brated separately to ensure that the relative sensitivities are

9
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
approximately the same. Since the transducer sensitivity is also used in the flanking paths. The thickness and conductance of
temperature sensitive, temperature sensors shall be installed at skins shall be limited to minimize the flanking loss potential.
the same or adjacent location. The outputs from these trans- 6.8 Air Circulation:
ducers are measured separately or as a group. If measured 6.8.1 The measured overall resistance, Ru, and, when appli-
separately, the transducers shall be detachable from the surface cable, the surface resistances, Rh or Rc, depend in part upon the
so their calibrations, at energy flux levels typical of use, may be velocity, temperature uniformity, and distribution patterns of
checked periodically (see Practice C 1130). If the measurement the air circulated past the specimen surfaces.
procedure is to calibrate the chamber with the heat flux 6.8.2 Air temperature differences of several degrees exist
transducers in place, the transducer outputs shall be connected from air curtain entrance to exit due to heating or cooling of the
in series to provide a single reading. The designer must air curtain as it passes over the specimen surface. The
recognize that the calibration factors for the heat flux trans- magnitude of this difference is a function of the heat flow
ducer will be different due to shunting effects when calibrated through the specimen and the velocity and volume of the air
in-situ versus calibrated alone. flow. When natural convection is desired, the temperature
[Link] Regardless of the method of hot box metering wall differences will be larger. A forced air flow reduces the
instrumentation used, the metering box wall heat flow shall be magnitude of this difference. Specific airflow conditions are
correlated with the signal outputs during the characterization established by the specification requirements for the material
process. See Section 8 and Annex A5 and Annex A6 for this being tested. The paragraphs below describe some specific
process. details required for maintenance of an acceptable air circula-
6.6 Climatic Chamber: tion within the hot box.
6.8.3 Test specifications sometimes require that near natural
6.6.1 The purpose of the climatic chamber is to provide
convection conditions be used in a wall test apparatus or in a
controlled conditions on the side of the specimen opposite the
floor/ceiling test apparatus. When required, these tests shall be
metering chamber. The test conditions specified are generally
run using forced convection at near natural convection condi-
those associated with standardized or normal outdoor condi-
tions. However, the air velocity shall be below 0.5 m/s if
tions. The instrumentation shall be capable of the control and natural convective air conditions are to be approximated with
measurement of the air temperature and velocity and surround- some forced airflow to maintain temperature control.
ing surface temperatures in order to maintain the desired
6.8.4 The design of the air circulation system will have an
surface heat transfer coefficient. In the usual arrangement, it
impact on the entrance to exit air temperature difference.
consists of a five-sided insulated chamber with internal dimen-
Tradeoffs during design must be made between the desired
sions matching or greater than the metering chamber opening uniformity of the air curtain temperatures and the operational
and with sufficient depth to contain the required cooling, mode of convective flow. A velocity of approximately 0.3 m/s
heating and air circulation equipment. An acceptable alternate has proven satisfactory for a wall test apparatus of 3 m height
is to utilize a large environmental chamber with an opening when testing wall systems.
matching the metering chamber opening size. This arrange-
6.8.5 When more uniform air temperatures are desired, it is
ment is especially suited for a floor/ceiling test apparatus in necessary to provide curtains of forced air moving past the
which large roof/attic structures are to be tested. specimen surfaces. For test purposes, the curtain air velocities
6.6.2 The walls of the climatic chamber shall be well shall be measured 75 mm away from the surface at the center
insulated to reduce the refrigeration capacity required and to of the specimen in the direction of airflow as specified in
prevent the formation of condensation on the outside of the [Link].
chamber walls. 6.8.6 For uniform test results, the maximum point to point
6.6.3 Heaters, fans and cooling coils shall be shielded or air temperature variation across the test panel, perpendicular to
placed behind an air baffle to maintain the uniformity of the the air flow direction at the center of the test panels, shall be
surface temperatures radiating to the surface of the specimen. less than 2 % of the overall air to air temperature difference, or
The internal surfaces of the climatic chamber shall also meet 2 K, whichever is greater.
the criteria of [Link] for surface emittance. 6.8.7 The direction of airflow in a hot box apparatus is
6.7 Specimen Frame: determined by the test design and may be parallel, that is, up,
6.7.1 A specimen frame shall be provided to support and down, or horizontal, or perpendicular to surface. However, less
position the specimen and to provide the needed perimeter fan power is required to maintain air movement in the direction
insulation. The frame opening shall have dimensions at least of of natural convection (down on the hot side, up on the cold)
those of the metering chamber opening. In the direction of heat and that direction is recommended. In some situations the test
flow, the frame shall be at least as thick as the thickest specification requires a specific direction to evaluate the system
specimen to be tested. In the outward direction perpendicular performance.
to the normal energy flow direction, the wall thickness of the 6.8.8 Air velocities greater than 1m/s are permissible when
specimen frame shall be at least equal to that of the metering their effect upon heat transfer is to be determined. Velocities
chamber walls or 100 mm, whichever is greater. commonly used to simulate parallel or perpendicular wind
conditions on the exterior side are 2.75 m/s for summer
6.7.2 Care must be taken in the design and construction of
conditions and 5.5 m/s for winter conditions.
specimen frames so that flanking losses are minimized. Con-
ductive plates, fasteners or structural members shall not be NOTE 11—Distinction is made between the effects and requirements of

10
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
air velocity parallel to the specimen surface and those for velocity and an outlet slot across the opposite. The inlet header shall
perpendicular to it. Parallel velocities simulate the effect of the cross incorporate adjustable slots or louvers to aid in obtaining
winds, and may be achieved by moving a small amount of air confined in uniform distribution.
a narrow baffle space and therefore require relatively little blower power.
Perpendicular velocities, simulating direct wind impingement, require [Link] After construction of an air circulation system, the
moving larger amounts of air with corresponding larger power require- air velocity profile shall be measured across the area perpen-
ments. The baffles in the second case must be placed further from the dicular to the direction of airflow in the proximity of the
specimen surface and should have a porous section (a set of screens or a specimen. The test shall be conducted with a flat, homogeneous
honeycomb air straightener) that directs the air stream to the specimen panel in place so that the surface of the test panel has minimum
surface. Fig. 4 shows an example of climatic chamber arrangement for
effect on the velocity profile. The air velocity profile shall be
perpendicular flow.
defined as uniform if all measurements from the profile scan
6.8.9 Air Baffles—For parallel flow, a baffle, parallel to the are within 10 % of the mean of all measurements. For parallel
specimen surface, shall be used to confine the air to a uniform air curtains, the air flow measurements shall be made at 0.3 m
channel, thus aiding in maintaining an air curtain with uniform intervals across the specimen face, perpendicular to the air flow
velocities. direction, at the centerline of the metering chamber. For air
[Link] The baffle thermal resistance shall be adequate to flow perpendicular to the specimen face, the air flow measure-
shield the specimen surface from radiative heat exchange with ments shall be made in the radial direction at a density of one
any energy sources located behind it. A baffle thermal resis- per every 30 degrees around the outlet of the diffuser at a
tance of 1 (m2 K /W) is recommended for this purpose. Other distance from the center of the metering area equal to the outlet
baffle designs that maintain temperature uniformity of the diameter of the air supply diffuser. If the profile is not uniform,
baffle surface seen by the test specimen are acceptable. additional adjustments shall be made to the inlet header slot or
[Link] An adjustable baffle-to-specimen spacing is one louvers or in the placement of fans or blowers to achieve an air
means of adjusting the airflow velocity. For purpose of curtain with uniform velocity across the face of the specimen.
maintaining a well-mixed and characterized air curtain, a The velocity profiles shall be verified, whenever modification
spacing of 140 to 200 mm is recommended. or repairs of the distribution system are made that might cause
[Link] A baffle also serves as a radiation exchange surface a change in flow patterns. Also, the profiles shall be verified
with a uniform temperature only slightly different than that of during characterization checks.
the air curtain. The baffle surface facing the specimen shall
have an emittance greater than 0.8. NOTE 12—Linear air diffusers designed for ceiling air distribution
systems have been found satisfactory to use as distribution headers. For
6.8.10 Air Velocity Uniformity—Uniform air flow profile large floor/ceiling testers it may be necessary to use more than one set of
across the specimen width, perpendicular to the air flow fans or inlet and outlet headers creating opposing zones to obtain the
direction, is achieved by use of multiple fans or blowers or by required temperature uniformity. Tangential fans have also been found to
use of an inlet distribution header across one edge of the baffle provide uniform temperatures.

FIG. 4 Hot Box Arrangement for Perpendicular Air Flow

11
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
6.8.11 Air Velocity Measurement—The apparatus design circulated by a separate blower to the specimen air curtain and returned.
shall provide a means for determining mean air velocity past An air reheat and control system provides fine control of air temperature
both the hot and cold faces of the specimen during each test. at the distribution header inlet. Other proven configurations utilize only a
single air circuit containing both cooling and reheat elements. Under
Acceptable methods are as follows: certain conditions, a desiccant may be needed to remove moisture from
[Link] One method is to measure the volumetric airflow in the air stream.
the duct to the inlet distribution header by using a calibrated
orifice or other flow-measuring device. The average baffle 6.9.4 Metering chamber blowers shall be small and efficient
space velocity is then calculated from the volume flow and the since, without cooling, they determine the least possible net
size of the space between the specimen and the parallel baffle. energy input to the metering chamber. If large fans or blowers
The baffle must be well sealed for this technique to work. are necessary, then compensatory cooling with inherent loss in
[Link] Another method is to calculate the velocity from an accuracy shall be used. Some heat is removed by locating the
energy balance. The rate of loss, or gain, of heat by the air as blower motor outside of the metering chamber and accurately
it moves through the baffle space, as indicated by its tempera- measuring the heat equivalent of the shaft power. Precautions
ture change, will match the rate of heat transfer through the shall be taken to prevent air leakage around the shaft.
metering chamber opening, average values of which can be 6.9.5 When cooling of the metering chamber is required, it
determined from the test data. must be done in a manner in which the amount of heat
[Link] The best method is to locate velocity sensors extracted can be measured accurately. One method is to
directly in the air curtain. For test purpose, wind velocity shall circulate a chilled liquid through a heat exchanger located in
be measured at a fixed location that represents the average free the metering chamber air circuit. The rate of heat extraction is
stream condition. For both perpendicular and parallel flow controlled by the inlet to chamber air temperature difference,
patterns, this location shall be a distance out in the air stream the airflow rate, the liquid properties, and the heat exchanger
such that the wind speed sensor is not in the test specimen efficiency. The amount of cooling used shall be limited to that
surface boundary layers or wakes. A distance of 75 to 150 mm necessary to overcome any excess blower or other heating
out from the test specimen surface at the center point is loads since test accuracy will be lost if excessive heating must
typically used. On the room side, where low circulation be used to compensate for large cooling. For example, assume
velocities are generally used, a properly located sensor is also that the heater input was 400 Btu/h out of an overall heater
required. The operator’s experience and knowledge of the air capacity of 2000 Btu/h and is known to within 1 % of capacity
distribution system obtained in the profiles from 6.8.10 shall be or 6 20 Btu/h. Also assume a concurrent cooling load of 320
used to determine the optimum sensor location. Btu/h out of an overall cooling capacity of 1600 Btu/h which
6.9 Air Temperature Control: is known to within 1 % of capacity or 6 16 Btu/h. Since these
6.9.1 The temperature of the air entering the air curtains loads oppose each other, the net load is 80 Btu/h but the
shall be within 6 1 K of the setpoint temperature across its uncertainty of the net could be as large as 6 36 Btu/h or 45 %
width and, for steady-state tests, shall not change during the of the net load. For this reason, care must be observed in
measurement period. obtaining the correct test setup.
6.9.2 One method of providing controlled, heated air is to 6.9.6 Special Considerations, Humidity Control—Moisture
install open wire, low thermal mass electrical heaters in an migration, condensation, and freezing within the specimen can
insulated, low emittance section of the blower duct or other also cause variations in heat flow. To avoid this, the warm side
part of the air circulation system and to control these heaters relative humidity shall be kept below 15 %.
using a sensor located at the inlet to the air curtain. 6.10 Temperature Measurement:
NOTE 13—Another method of heater control is to use several individual 6.10.1 When surface temperatures are required, specimen
heaters that are switched on to provide fixed levels of energy. Fine-tuning surface temperature sensors shall typically be located opposite
is provided by an additional heater modulated by a controller. Another each other on the two faces of the specimen. However, when
satisfactory method is to use a controller that varies the power to all the
heaters. placement opposite each other is not possible, the sensors shall
be placed to represent the correct area weighting for each
6.9.3 Methods for cooling the climatic chamber include the surface. These sensors shall be chosen and applied to the
installation of a refrigeration system evaporator inside the surface in a manner such that the indicated temperature is
chamber, ducting in chilled air from an external source or within 6 0.2 K of the temperature that would exist if the sensor
injecting liquid nitrogen. Usually the evaporator or external had not been applied. This requirement is met by thermo-
chilled air is controlled at a constant temperature a few degrees couples if: (1) the wire is no larger in diameter than 0.25 mm
(typically < 5°C) below the desired setpoint. Then, a reheat and (No. 30 AWG.); (2) the wire meets, or is calibrated to, the
control system, similar to that for obtaining heated air (see special limits of error as specified in the Tables E 230; (3) the
6.9.2) is used to achieve fine control of the temperature at the junctions, not larger than two times the wire diameter, are
inlet to the specimen air curtain. When liquid nitrogen is used twisted and welded or soldered; (4) 100 mm of adjoining wire
a valve regulating its flow is pulsed or modulated to obtain fine are taped, cemented or otherwise held in thermal contact with
temperature control. the surface using materials of emittance close (6 0.05) to that
NOTE 14—One proven configuration for a climatic chamber utilizes of the surface; and (5) they are electrically insulated, or
two air circuits created by suitable baffles. The evaporator fan creates one otherwise protected, so that the electrical junction is at the
circulation path that includes a mixing chamber from which air is location of the thermocouple bead. Application of alternate

12
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
temperature sensor systems may be used if comparative surface temperature sensors in 6.10.2. These sensors must be
measurements or calculations show that the basic requirements radiation shielded or otherwise protected to provide an accurate
are met. indication of the temperature of the air curtain. Sensors shall be
NOTE 15—Metal foil tape, which has been painted to make the
small to ensure fast response to changing temperatures. Resis-
emittance greater than to 0.80, is an effective means to attach thermo- tance wires, if used, shall be distributed uniformly in the air
couple sensors to most high emittance test specimens. curtain.
6.10.2 If the specimen construction, and therefore its ther- NOTE 17—One suitable radiation shield is made by using 12 mm
mal resistance, is uniform over its entire area, then a minimum diameter, 75 mm long pieces of thin walled plastic tubing covered on the
number of sensors, spaced uniformly and symmetrically over outside with aluminum foil tape. The air thermocouple is placed at the
the surface, are sufficient. The required minimum number of center of the tube to measure the air stream temperature and yet be
shielded from radiation sources.
sensors per side shall be at least two per square meter of
metering area but not less than nine (24). [Link] The best location for temperature sensors depends
[Link] If each element of the specimen construction is upon the type of air curtain convection (natural or forced). In
relatively uniform in thermal resistance and is repeated several natural convection situations, it is usually possible to identify
times over the entire surface, the number of sensors specified in the temperature of still air outside the boundary layer. Conse-
6.10.2 may still be sufficient. In this case, the sensors shall be quently, when natural convection is established, air tempera-
located to obtain the average surface temperature over each ture sensors shall be located in a plane parallel to the specimen
type of construction element and, for each type of element, surface and spaced far enough away from it that they are
shall be distributed approximately uniformly and symmetri- unaffected by temperature gradients of the boundary layer. For
cally over the specimen area. The average surface temperature minimum velocities required to attain temperature uniformities
of the specimen shall be calculated by area weighting of the (see 6.8 and Note 12), the minimum spacing from the specimen
averages for the different types of construction elements. surface is 75 mm. At velocities greater than 1 m/s, the required
[Link] If the surface temperatures are expected to be, or minimum spacing is greater. The boundary layer thickness
found to be, greatly non-uniform, additional sensors shall be increases sharply at the transition from laminar to turbulent
required. Often a great number, such as three or more times the flow. With fully developed turbulent flow, the boundary layer
normal amount as determined by trial and error, is required to occupies the full space between the specimen and the baffle.
adequately sample the different temperature areas so that a When forced convection is established and the flow is fully
reliable area weighted mean surface temperature may be developed, the sensors shall be located at a distance from the
obtained. Some research has been published on the subject of specimen surface corresponding to 2⁄3 up to 3⁄4 of the specimen-
testing highly conductive member that might be used as to-baffle distance. This is to detect a temperature approaching
guidance for this determination. For example, see the work on the airflow bulk temperature.
steel framed buildings (29). [Link] Thermocouple sensors used for measurement of air
[Link] If an accurate determination of the average surface temperatures shall meet the requirements of Items (1), (2), (3),
temperatures cannot be obtained, the hot box apparatus can and (5) in 6.10.1. Other sensors are acceptable if they have
accurately measure only the thermal transmittance, U, or the similar time response and are calibrated so that the measure-
overall thermal resistance, Ru. The average panel resistance, R, ments are accurate within 6 0.5 K.
of the specimen can be estimated by subtracting off the
6.10.4 The surface temperature of the baffles in the metering
previously determined surface film thermal resistances estab-
and climatic chambers, where required, shall be measured by
lished using a transfer standard of equal thermal resistance,
placing sensors on all surfaces seen by the specimen. A
size, surface configuration and roughness. Note that the geom-
minimum area density of three sensors per square meter of
etry, average temperatures, and energy exchange conditions
baffle area, but not less than one sensor per baffle surface, is
must be similar for the calibration transfer standard (CTS) and
required. These data (1) can be used to determine any differ-
test panel for this technique to have reasonable accuracy. (See
ence between the baffle surface and air curtain temperatures;
Test Method C 1199 for discussion on CTS design.)
(2) permits corrections to be made to the radiation component
NOTE 16—Tests on specimens containing thermal bridges require of the surface film conductance due to differences in these
special care because of the possible great differences in thermal resistance temperatures; and (3) is a necessary component of the data
and temperatures between the thermal bridge areas and those of surround- analysis for specimens such as windows which have a high
ing insulated structures. Added complications arise when tests are run at
higher air velocities since temperatures and energy transfer can depend
thermal conductance. (See the discussion on the environmental
significantly upon bridge geometry relative to the overall sample as well temperature determination in Annex A9.)
as the velocity and direction of air movement. If test results are to be 6.11 Specimen Pressure Difference:
comparable for competing systems, they must be run under similar 6.11.1 For some tests, it is necessary to establish and
conditions. This method does not attempt to standardize such conditions.
measure the air pressure differential between the faces of the
6.10.3 The temperature of the air on each side of the test specimen. This is especially important for window and
specimen shall be measured by thermocouples, temperature other samples where the airflow resistance between the speci-
sensitive resistance wires, or similar temperature sensors. men surfaces is low. The specimen pressure difference is
[Link] The minimum number and locations of sensors defined as the difference in the local static pressure, on either
used to measure air temperatures shall be that specified for side of the specimen, measured at a location at the geographic

13
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
center of the metered area, at a distance 75 mm from the 7.2.1 The building element shall be sized for the apparatus.
surfaces of the sample. Normally the outside dimensions of the building element shall
6.12 Instruments: match the dimensions of the metering chamber opening.
6.12.1 All signal conditioning and data logging instruments 7.2.2 Wherever possible, the percent framing and insulated
shall be located outside of the apparatus. All instruments shall cavity space dimensions of the building element shall be the
be calibrated to the specified accuracy, traceable to a national same as the building assembly it represents.
standards laboratory, and shall meet the following additional 7.2.3 For elements such as an opaque envelope section, the
requirements: building element is defined by an integral number of structural
[Link] All instrumentation shall have adequate sensor sections. For example, a residential wall section constructed of
response so that the scanning speed does not adversely effect 0.41 m on center framing by 2.44 m wall height, would have a
the measurement results. specimen size of at least 1.22 m wide by 2.44 m high. Metal
building sections shall have a specimen width equal to the
[Link] Temperatures shall be readable to 6 0.05 K and be
framing dimension, often 1.52 m.
accurate within 6 0.5 K.
7.2.4 If smaller elements must be tested, a surround panel
[Link] Heat flux transducer outputs shall be measured to shall be used to fill out the required size. The surround panel
the precision required to limit the error in estimation of the aperture for test purpose shall be sufficiently small relative to
metering box wall loss to less than 6 0.5 % of the specimen the metering area such that the minimum distance between the
energy transfer. This requires a heat flux transducer calibration metering area boundary and the aperture boundary is greater or
accuracy of 5 percent or better. equal to 100 mm (see Annex A5).
[Link] Many methods of air velocity measurement are 7.2.5 For building elements having limited dimensions such
possible depending on the specific box design and test condi- as windows, doors, etc., the test specimen shall be the complete
tions. However, an accuracy of 6 5 % of the reading is component plus the necessary surround panel.
required. A sensor whose signal can be processed by automatic 7.2.6 For a building element having thermal behavior that is
data acquisition equipment is recommended. mostly independent of its horizontal and vertical dimensions,
[Link] Pressure difference measurements shall be accurate the test specimen size is at least that necessary to obtain an
to within 6 5 % of reading or 6 1 Pa, whichever is greater. average performance for the material system. For example,
[Link] Total average power (or integrated energy over a insulated systems such as foam sandwich structure panels, are
specified time period) to the metering box shall be accurate to relatively uniform in the cross directions but may be non-
within 6 0.5 % of reading under conditions of use. Power uniform through the thickness. The test specimen for this type
measuring instruments shall be compatible with the power of material shall be large enough to obtain an average value
supplied whether ac, dc, on off, proportioning, etc. Voltage which accounts not only for manufacturing variability but also
stabilized power supplies are strongly recommended. Metered includes the effect of joint details between adjacent panels
cooling instruments shall be calibrated together as a system to when tested as a system.
similar accuracy. 7.2.7 For the characterization of homogeneous or nearly
[Link] Temperature controllers for steady-state tests shall homogeneous materials that are self-supporting, the test speci-
be capable of controlling temperatures constant to within 6 men shall consist of a single layer of material. However,
0.25 K (see 6.9). specimen assembly precautions such as sealing the surfaces
shall be observed.
7. Sampling and Test Specimens 7.2.8 Three-dimensional structures may be tested if the
apparatus size permits.
7.1 Building elements shall be representative of typical field
assemblies. As such, the metered specimen is usually a portion 7.3 Sensors—The temperature sensors for the measurement
of a building assembly that has been selected for test due to the of surface temperatures shall be installed as directed in 6.10.
expectation that it will exhibit the same thermal behavior as the Additional sensors may be installed throughout the interior of
larger building element that it represents. Tests on apparatus the specimen for special investigations of local temperature
requiring smaller than representative specimens shall be variations.
avoided. The construction details of the building elements to be 7.4 Mounting—The building element shall be located in the
investigated may be modified but only if necessary for test same position in test frames as the specimen was during
purposes. It must be recognized that modifications to the characterization tests so that flanking geometry is duplicated.
construction result in conditions that do not represent true field 7.5 Sealing—The building element shall be gasketed,
conditions. Conduction and convection paths that have consid- caulked, taped, or otherwise sealed in place to prevent air
erable effect on the performance of the building elements must movement around its perimeter. The procedures and material
be left intact. During specimen design the following shall be for sealing shall be chosen to minimize flanking loss. If the
considered. building element is suspected of being porous so that a
significant energy transfer results from air infiltration through
NOTE 18—Reduced scale elements shall not be tested with the intent of the building element, then tests shall be run before and after
extrapolating results to larger elements unless detailed modeling analysis sealing both faces. If the overall resistance changes signifi-
clearly shows the validity of the extrapolations.
cantly, then the building element does not possess unique
7.2 Building Element Sizing: properties independent of the imposed conditions. Results from

14
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
all tests must be reported. Thin, air impervious sheets of paper fasteners, or other less obvious heat flow paths. Thus, the net
or plastic, may be glued on to seal surfaces without signifi- metered specimen heat transfer shall be determined from the
cantly affecting thermal conduction. Some building elements measured heat input by applying a correction for these flows.
are sealed with suitable paint. In all cases, the surface emit- This correction, which is determined by characterization pro-
tance of the sealed building element shall be within 6 0.1 of cedures, is different for each set of operating conditions and for
the emittance of the original unsealed building element. metered specimens of different thickness or thermal resistance.
7.6 Perimeter Insulation—Insulation shall be used at the The accuracy of the test results depends upon the accuracy of
building element perimeter. This insulation normally is incor- this correction. In a properly designed apparatus, however, the
porated into the re-usable specimen frame but may be newly flows are a relatively small fraction of the metered specimen
installed for each building element. If newly installed, it shall heat transfer under steady-state conditions and any error in the
be fully characterized in order to account for the surround correction is reduced by a similar fraction in its effect upon the
panel flanking loss. final result.
7.7 Internal Air Barriers—Testing of a building element,
with uninterrupted internal air cavities that extend beyond the NOTE 19—A discussion of the characterization for the metering cham-
ber walls is presented in Annex A3. A discussion of Flanking Loss
boundaries of the metering section is not permitted. To
characterization for one apparatus is given by Lavine et al. (12) and in
characterize building elements having uninterrupted air cavi- Annex A7. The overall test matrix for the characterization is discussed in
ties that are larger than the metering chamber, it is necessary to Annex A6. Examples of typical characterization matrices are presented in
alter the element by placing an internal convection barrier in that section.
each cavity where it crosses the boundary of the metering
chamber. These barriers are required to prevent undesired air 8.3 In principle, if all details of the hot box construction and
exchange between the metering and guard areas of the speci- all material thermal properties are known, it is possible to
men. For example, such barriers are required for vertical wall calculate all extraneous flows for a particular set of test
cavities extending above or below the metered area when the conditions and then apply this calculated correction to mea-
cavity is insulated with reflective insulation having no internal sured data for unknown test specimens. However, because of
air barriers. Any modifications to the building element shall be the uncertainties involved, a wholly calculational characteriza-
reported. tion procedure, without experimental verification at the test
7.8 High Lateral Conductance Building Elements—For all conditions, shall not be used for this method. In general, such
building elements, it is necessary to maintain a near zero lateral calculations are practical only with monolithic walls made of
energy flow between any guard and the metering areas of the homogeneous material. If calculated corrections, after initial
specimen. This can be achieved by maintaining a near zero experimental verification, are used, then the chamber wall heat
temperature difference on the building element surface be- flow meter or thermopile outputs are used as a check to indicate
tween the metering and guard areas. However, in building any future changes in wall material properties. Calculations are
elements incorporating an element of high lateral conductance, useful in estimating the magnitude of the major heat flows so
such as a metal sheet, it is necessary to separate the highly that characterization procedures may be better directed. Indeed,
conductive element with a thermal break. One form of thermal the most practical characterization technique uses corrections
break is a narrow gap caused by a saw cut at the metering determined experimentally for a limited set of conditions, but
chamber boundary. modified on the basis of calculated estimates for use under
7.9 When testing high thermal resistance specimens that are somewhat different conditions of test. In general, the charac-
smaller than the metering area using a surround panel, the heat terization procedure of 8.5, using a correction developed
loss through the surround panel may approach or exceed the statistically from tests on standard reference materials shall be
heat loss through the specimen. In this case, the operator shall used. The choice of the characterization procedure details shall
determine the uncertainty of the test result and include that be made only after a review of the expected accuracy judged
uncertainty value in the report. against the accuracy needed and against the practicability of
the various procedures available.
8. Apparatus Characterization 8.4 Characterization Specimens—The accuracy of the char-
8.1 All fundamental measurement devices used in the hot acterization specimen measurements will depend upon the
box control and data acquisition systems shall be individually variability of the material, the means of sampling and the
maintained and calibrated to meet their design accuracy accuracy of the apparatus used to measure it. The accuracy
specifications. In general, this requires that each device be required will depend upon the contemplated use. For highest
traceable to standards obtained from a national standards accuracy, a characterization specimen having a known thermal
laboratory. Records of this calibration and periodic calibration resistance over the range of test mean temperatures is required.
verification checks shall be maintained in the laboratory files. Such specimens shall be impervious to air and thermal radia-
Frequency of validation checks will be dependent on the tion transfer, shall be free of internal air spaces that would
purpose, style and stability of the equipment used. affect the thermal resistance or allow internal convection, and
8.2 Hot box apparatus characterization is necessary since shall be stable over the time period of use. Additionally, such
the measured net heat input to the metering chamber includes specimens shall possess a thermal resistance that is essentially
not only the heat transfer through the specimen, but also constant over all areas of the specimen so that properties
metering box wall loss, flanking loss, and other such heat flows determined on smaller areas will be representative of those of
as through gaskets, penetrations for wires or pipes, mechanical the whole area. Any joints necessary in large specimens shall

15
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
be designed to minimize deviations in thermal resistance (as Rucker and Mumaw (9), by Lavine, et al (12) and in Annex A3 through
verified by small scale tests of specimens with and without Annex A7.
joints). Characterization specimens shall be self-supporting 8.6 In addition to the initial characterization sequence, it is
and capable of being transported, repeatedly mounted and necessary to repeat selected measurements at times dictated by
tested, and stored for future use without change in thermal either the known aging characteristics of the materials used in
resistance. These properties are also required for specimens the metering chamber wall construction or, more often, as
used in inter-laboratory comparison tests (round robins). The required by contractual or certification regulations. A single
thermal resistance of characterization specimens shall be test may often be sufficient to verify that properties have not
determined by measurements in proven apparatus conforming changed. The maximum time between verification of charac-
to Test Methods C 177 or C 518 or another hot box that has terization shall be one year.
been verified or calibrated by specimens traceable to a national 8.7 It is recommended that the performance of an apparatus
standards laboratory. Generally, the hot box characterization be periodically confirmed by successful measurements on
specimen will be larger than the apparatus used in these specimens traceable to a national standards laboratory, previ-
measurements; thus, it will be necessary to measure smaller ously measured building specimens of known thermal perfor-
representative pieces. Such pieces are cut from the character- mance or as part of a laboratory accreditation program.
ization specimen if they can be replaced without change in the Participation in inter-laboratory round robin programs and
average thermal properties, or they are selected from compan- comparisons with another proven hot box apparatus are other
ion pieces of the same lot of material used to fabricate the methods to demonstrate continued satisfactory operation.
specimen.
NOTE 20—Suitable characterization specimens have been constructed 9. Conditioning
from molded glass fiberboard of approximately 100 to 125 kg/m3 density
or aged cellular polystyrene (XPS) board. During the tests both surfaces of
9.1 Conditioning requirements specified by code or con-
the characterization panel shall be faced with an air impervious skins struction specifications shall govern for the test, where avail-
having an emittance greater than 0.8. able. Normally pre-test conditioning shall be in ambient air, for
a period long enough to come to practical equilibrium. If the
8.5 Metering Box Wall Characterization—An Overview— requester does not provide specific conditioning instructions,
Since significant heat flows may exist which are not directly use Practice C 870 as a guide for conditioning. The recom-
related to heat flow through the chamber walls and therefore mended condition is in air at 24 6 2°C with 50 6 5 % relative
not related to the voltage output of the wall energy flow meters humidity. To avoid abnormally long conditioning periods,
or thermopiles, a full experimental characterization is neces- building materials may be preconditioned at laboratory condi-
sary. This characterization involves running a series of tests tions prior to specimen assembly. Other conditioning may be
over the expected operating temperature range using a charac- used as, for example, long term exposure to cold dry (outside
terization specimen of known thermal resistance (see 8.3). For winter) air on one side and warm, moderately humid (inside)
each test, determination is made of the difference between the air on the other to investigate the effects of moisture or ice
measured heat input to the metering chamber and the heat build up.
transfer through the characterization specimen, calculated from
the measured temperature drop across it and its known resis-
10. Test Procedure
tance. It is impractical to run a sufficient number of tests to
cover all possible sets of operating conditions. Since some of 10.1 Detailed written operating procedures for each test
the extraneous heat flows included in the measurement are not apparatus shall be developed and shall be available to ensure
metered separately (and indeed may be unknown), it is that the tests are conducted in accordance with the require-
necessary to utilize statistical techniques to develop a usable ments of this test method.
correlation between the corrections and the test conditions. A 10.2 Test Conditions:
useful procedure is to relate the correction to the test variables 10.2.1 Whenever available, product or system specifications
using a multiple linear regression. The significant test vari- or applicable code requirements for all test conditions shall be
ables, or combinations of test variables, can often be deter- used.
mined from physical models. Those variables may include the 10.2.2 Specimen orientation and direction of heat transfer,
mean temperature of the specimen and of the metering cham- hot-side and cold-side air temperature and velocities and
ber walls, the temperature difference across the specimen, and differential pressure, when not specified, shall be chosen to
across the metering chamber walls (related to the output of the meet requirements of the building element investigation, usu-
chamber heat meters or thermopiles) and the temperature ally to match in-use conditions.
difference across any partial guards used. The regression 10.2.3 When not otherwise directed, the air velocities shall
correlation coefficients can be used to judge the statistical be the minimum required to achieve the desired temperature
quality of the regression relation and the choice of variables. uniformity under the requirements of 6.8.4, be in the direction
For greatest accuracy, it is necessary to run characterization of natural convection, and the metered specimen pressure
specimens covering the expected range of specimen thickness differential shall be essentially zero.
and thermal resistance and to include these variables in the
10.2.4 Whenever the temperature conditions are not other-
regression analysis.
wise specified, Practice C 1058 shall be used as a guide for
NOTE 21—Examples of characterization procedures are given by selecting the appropriate test temperature conditions.

16
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
10.2.5 When testing fenestration products, Test Method 37 % of the final temperatures. The thermal time constant of the constant
C 1199 and Practice E 1423 shall be used as a guide for temperature mode is the time required to come to within 37 % of the final
selecting the appropriate test environmental conditions. power level. The thermal time constant of a simple system can be
estimated from knowledge of the thermal diffusivities of the components
10.3 Construct the building element in the specimen frame of the system, but it is more readily determined experimentally for
opening as specified in Section 7 including installation of all complex systems.
required sensors.
10.3.1 Some metered specimens require adequate time to 10.10.3 Annex A10 contains a suggested procedure for
come to thermal and moisture equilibrium after assembly. estimating the thermal time constant of a test system.
These shall be conditioned at laboratory conditions as long as 10.10.4 Normally, the thermal capacity of either the appa-
necessary to establish equilibrium, that is, constant weight. ratus or metered specimen will be the controlling factor.
One example would be concrete walls or wet applied insula- However, since this test method is applicable to low conduc-
tions in a frame wall. tance specimens, the time to reach steady state is on the order
10.4 Place the test frame, with the metered specimen of hours. Even with this information, it may be difficult to
installed, in the opening between the climatic and metering judge whether stability has been reached and the operator shall
chambers. rely upon previous test experience and observations or upon
computer assisted statistical prediction of trends. The follow-
10.5 Make all necessary electrical connections and check
ing guidelines are recommended but shall not be regarded as
out the data acquisition system for measurement continuity.
sufficient criteria in all cases.
10.6 Complete sealing of the hot box system in preparation
10.11 Test Data Acquisition and Completion:
for the test. Check the installed metered specimen for air
10.11.1 Data Acquisition—After the final test temperature
leakage, if possible (see 7.5 and Note 24).
conditions are reached, five successive repeated data acquisi-
10.7 Start conditioning systems and set temperature controls
tion sets shall be obtained. These sets shall be obtained at a data
to the appropriate temperature set points to yield the desired
set time interval equal to the approximate time constant, teff, of
temperature conditions.
the measured system but not less than 30 minutes. In some
10.8 Begin data acquisition scanning of the test apparatus laboratories, an individual data set is developed from the
and continue the operation until the steady conditions de- average value for each variable obtained from multiple, evenly
scribed in 10.10 are obtained. spaced, data scans during the permitted time interval.
10.9 As specified in 6.9.6, avoid test conditions that cause 10.11.2 Test Completion Criteria—This combination of five
condensation on the metered specimen surfaces during the test. data acquisition runs shall constitute a valid test if the datum
This requirement becomes more difficult to satisfy when obtained for each measured variable differs from its mean by
testing building elements with highly conductive components no more than the uncertainty of that variable. If the data
such as steel stud walls or single glazed windows. For example, obtained during this period is changing monotonically with
the National Fenestration Rating Council (NFRC) window test time, the test shall also be considered incomplete and further
operators have observed that the relative humidity in the repeated runs shall be conducted until the steady drift is no
metering chamber must be below 15 % to prevent the forma- longer observed. Such a drift, even at low levels, indicates that
tion of frost on highly conductive window frames tested at the specimen characteristics are changing or that the system is
−18°C cold side and 21°C hot side air temperatures. not at steady state within its test capabilities. In either event,
10.10 Stabilization and Test Times: serious errors may result. (See Note 23 for an example of the
10.10.1 Thermal Steady-State—For purpose of this test criteria for stabilization.)
procedure the definition of thermal steady-state is identical to 10.11.3 Continued Testing—For the purpose of determining
that described as steady-state (thermal) in Terminology C 168. test completion, it is necessary to repeat the testing in five time
10.10.2 The required time to reach stability for a steady- constant blocks (5 · teff) until all the required criteria have been
state test depends upon the properties of both the metered satisfied. For test analysis, a sliding 5 · teff time range shall be
specimen and of the apparatus as well as upon the initial and used. Upon acquisition of each additional data set, an analysis
final conditions of the test. Since these factors can vary over of the last five sets shall be performed to see if the criteria of
wide ranges, a single specification of required stabilization 10.11.2 are met. As soon as these criteria are met, the test is
time and the test period for data acquisition couldn’t be judged complete and the reported result is determined from the
provided. A combined apparatus and metered specimen time averages of the last five readings.
constant, teff, calculated from dimensions and estimated physi-
NOTE 23—Operator experience on different types of wall sections has
cal properties, can be helpful in estimating stabilization times.
shown that the time to stabilized conditions can range from several hours
NOTE 22—The thermal time constant, teff, of the system is the time for lightweight building components to several days for thick massive
required to come to within 1/e (37 %) of the final value of the thermal constructions. Specific test practices have been written and used that
resistance after a step temperature disturbance of the system. This time is reference the hot box test procedure. In these cases, alternate procedures
strongly dependent on the mode of operation. Two modes of operation have been written that specify specific requirements for steady state
have been used for a hot box operation. They are (1) constant power to the determination and frequency of data collection intended to meet the intent
metering chamber and (2) constant temperature control of the metering of these sections. An example of a modified stabilization procedure
chamber. The constant temperature operation mode is usually used since developed for fenestration testing is presented below:
it has a considerably shorter time constant because it is not significantly “After essentially steady state temperature and heat flow conditions
dependent on the thermal mass of the metering chamber. For the constant have been reached, a measurement period of five continuous time
power mode, the thermal time constant is the time required to come within constants shall produce five successive measured data sets in which the

17
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
following conditions exist: 11.2.1 When operated under steady-state conditions with
(1) The average room side and exterior test specimen individual surface temperatures held constant during a test, the results shall be
temperatures (if measured) do not change by more than 6 0.25°C over the expressed as thermal resistance R, thermal conductance C,
entire test period.
(2) The average metering box ambient air temperatures do not vary by
overall thermal resistance Ru, or thermal transmittance U. This
more than 6 0.25°C over the test period. method permits use of either of two procedures for determining
(3) The average metering box wall heat flow does not vary more than the average surface temperatures used in the calculations. The
6 1 % of the specimen heat flow and does not change monotonically over choice between the two procedures depends, to some extent,
the entire test period. upon the uniformity of the specimen and thus upon whether
(4) The net energy input to the metering box shall be recorded by sufficiently uniform surface temperatures exist that can be
automated data acquisition equipment at five minute intervals or less and
measured by temperature sensors and a representative average
shall not deviate more than 6 1 % from the average net energy readings
at any time during the entire test period. The net energy input to the obtained. For some specimens, the choice shall be arbitrary and
metering box also shall not change monotonically during the test period. must be made by the user of the method, by the sponsor of the
(5) The thermal transmittance, as calculated from the data, for the test, or it may be specified in applicable regulations or
sample shall not vary more than 6 1 % when comparing any time period specifications. In all cases, the procedure used shall be fully
with any other period within the five data sets. The data sets shall not reported. The two procedures are:
overlap. [Link] For uniform and nearly uniform metered specimens
(6) The final, calculated test result shall be the average result calculated
the average surface temperatures shall be determined from
for the last five time constant periods of the stabilized test period.”
area-weighted measurements from the temperature sensors
10.12 Recorded Test Data: installed as directed in 6.10. The thermal resistance, R, is then
10.12.1 The data acquired during the testing period shall calculated using the measured energy transfer and the differ-
include, but not be restricted to, the following items. ence in the average temperatures of the two surfaces.
[Link] The total net energy or average power transferred [Link] For very non-uniform metered specimens (see
through the metered specimen, during a measurement interval. [Link]), meaningful average surface temperatures will not
This includes all metering box heating and cooling, power to exist. In this case the thermal resistance, R, is estimated by
fans or blowers, any significant power to transducers, correc- subtracting the surface resistances for the two surfaces from the
tions for metering chamber wall energy transfer and flanking measured overall thermal resistance, Ru. These surface resis-
loss, any other extraneous flow, and corrections for the en- tances shall be determined from tests conducted under similar
thalpy of infiltration air entering the metering chamber. conditions (see Note 25), but using a uniform metered speci-
[Link] All air and surface temperatures specified in 6.10. men of approximately the same overall thermal resistance.
[Link] The average air velocity on each side of the
metered specimen (see 6.8.11). NOTE 25—Surface resistances have been found to depend significantly
on the magnitude of the energy flux as well as the ambient conditions
[Link] The pressure differential across the metered speci- affecting the surface. When using the procedure of [Link], it is important
men, if different from zero, (see 6.11), and the infiltration flow that the energy flux for the uniform metered specimen be similar to that
rate required to maintain it. through the non-uniform metered specimen and that air temperature, air
velocity, and the temperature of surfaces that exchange radiation with the
NOTE 24—For either parallel or perpendicular forced air velocity
specimens also be similar.
conditions, care shall be taken to quantify the amount of air leakage
between the climatic and metering chambers. This may be done by several 11.3 Calculation of Thermal Properties:
techniques, including: (1) tracer gas methods or (2) calibration of the 11.3.1 Generally, the overall thermal resistance, Ru, or the
airflow rate as a function of the pressure difference using Test Method thermal transmittance, U, shall be determined for the specimen
E 1424. under the conditions of interest.
[Link] The effective test specimen dimensions and me- 11.3.2 For very non-uniform specimens where the energy
tered area (the projected area perpendicular to the direction of transfer is greatly different from one area to another, for
energy flow). It may also be helpful to determine and report the example, metal frame building section or windows, and if
hot and cold side surface areas if they are different from the detailed temperatures profiles are not known, only the net
projected areas. For example, detailed windows have surface energy transfer through the specimen (see 10.11) shall be
areas as much as 50 % greater than the projected areas. meaningful. In these cases, only the overall resistance, Ru, and
[Link] The metering area of the hot box. transmission coefficient, U, are permitted.
[Link] Any other conditions specific to this test such as 11.3.3 For a relatively uniform specimen having only minor
modifications to the normal specimen design required to thermal bridging such as wood framed walls, floors, ceilings,
assemble the test specimen for test purpose. etc., the thermal properties that shall be calculated are: the
resistance R, conductance C, overall resistance Ru, transmit-
11. Calculations tance U, surface resistances Rc,env and Rh,env, and surface
11.1 For steady-state tests, the average thermal transmission conductances hc,enc and hh,env.
properties appropriate for metered specimen are calculated by 11.3.4 For a homogeneous specimen of insulation material,
using one of the equations given in 3.4.2, using the average the apparent thermal conductivity, l, shall also be calculated if
data obtained in 10.10 and 10.11. Practice C 1045 should be the specimen meets the uniformity requirements of Terminol-
used to resolve the test results for variable temperature ogy C 168. Available test data shall demonstrate that the
difference testing. thermal resistance of the material under test is linearly propor-
11.2 Average Temperature Determination: tional to thickness within the range of temperatures and

18
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
thicknesses under consideration. An expected error of these NOTE 27—The primary units used in this method are SI, but either SI
assumptions shall be assigned to the thermal conductivity or inch-pound units may be used in the report, unless otherwise specified.
result as part of the report. Table 1 provides conversion factors between inch-pound and SI units.
[Link] Where there is a question as to the uniformity of the 12.1.1 Identification of the test laboratory with address and
tested material, multiple tests at the same temperature condi- telephone number, the responsible person in charge, the test
tions but at different thicknesses shall be made. If the material operator (optional), the date and duration of test, and the test
has been tested at thicknesses greater than the representative sponsor, if appropriate.
thickness, the calculation of the apparent thermal conductivity 12.1.2 Name, and any other identification or description of
for those tests will yield the same result within the uncertainty the test construction, including if necessary, a drawing showing
of the measurements. If the result is not the same within the important details, dimensions, and all modifications made to
uncertainty for these tests, then the test results are applicable the construction, if any, and specimen orientation. Photographs
only at the thickness of the test at that thickness. This fact shall and drawings are helpful as are statements explaining how the
be included as part of the report. specimen represents or differs from typical constructions. It is
11.3.5 For a specimen smaller than the metering chamber also desirable to include in the description of the test construc-
opening, the properties that apply to that specimen, as per the tion a complete and detailed description of all materials. This
distinctions of 11.3.1 through 11.3.4, shall be calculated if includes the generic names of all construction materials and
surround panel calibration tests have been run that permit the their densities. For hygroscopic materials, such as concrete and
specimen energy transfer to be determined. Annex A8 presents wood, the moisture content shall also be given. If the thermal
considerations for these calculations. conductivity of these materials, at the test conditions, has been
11.3.6 When directed by applicable agreements or regula- measured, these values shall also be included.
tions, the overall resistance at standardize conditions, Ru, shall
be determined from the estimate of the thermal resistance, R, NOTE 28—A detailed description of the test materials in addition to the
obtained as directed in [Link] by adding standardized surface brand name shall be reported, where possible. For example: preformed,
cellular polystyrene, Type II with a density of 22 kg/m3; spruce-pine-fir
resistances. One source of standardized resistances is the with moisture content of 12 % and a dry density of 486 kg/m3.
ASHRAE Handbook of Fundamentals.
12.1.3 Any pertinent information regarding the specimen
NOTE 26—Overall resistances, Ru, obtained from measured resistances,
R, by adding standardized surface resistances typical of different condi-
preconditioning.
tions, may not agree with overall resistances that would be measured 12.1.4 Information shall be provided that describes the test
directly under those conditions. Discrepancies are especially likely for apparatus, apparatus number, test configuration, mode of op-
non-uniform specimens with high conductance surface elements con- eration, etc. including the dimensions of the metered area and
nected to thermal bridges when measured resistances, R, are obtained its relationship to the overall test specimen dimensions and to
under nearly still air conditions on one or both sides, and the standardized
principle elements of the specimen.
outside surface resistances are typical of high wind velocities. The user is
cautioned to be aware of such possible discrepancies. 12.1.5 Test specimen orientation and the direction of energy
transfer during the test.
12. Report 12.1.6 Average air velocity and direction on both sides of
12.1 The report shall include information on the following: the test specimen and air velocity distribution, if non-uniform.

TABLE 1 Thermal Properties Conversion Factors (International Table)

NOTE 1—Conversion factors for thermal resistivity and thermal conductance or transmittance can be found by using these tables in reverse direction.
NOTE 2—Units are given in terms of (1) the absolute joule per second or watt; (2) the calorie (International Table) = 4.1868 J; or the British thermal
unit (International Table) = 1055.06 J.
NOTE 3—Example of table use: To convert from thermal conductivity of 0.05 W/mK (SI units) to thermal conductivity (IP units), multiply by 6.9330
to yield 0.35 (Btu in./h ft2 F).
Thermal Conductivity
W/m K W/cm K cal/s cm K kcal / h m K Btu / h ft F Btu in/hr ft2 F
W/m K 1.0000 0.0010 2.388E-3 0.8598 0.5778 6.9330
W/cm K 100.0000 1.0000 0.2388 85.9800 57.7800 693.3000
W/cm K 418.7000 4.1870 1.0000 360.0000 241.9000 2,903.0000
cal/s cm K 1.1630 1.163E-2 2.7788E-3 1.0000 0.6720 8.0640
Btu/h ft F 1.7310 1.731E-2 4.134E-3 1.4880 1.0000 12.0000
Btu in/h ft2 F 0.1442 1.442E-3 3.445E-4 0.1240 8.333E-2 1.0000

Thermal Resistance
K m2/ W K cm2/W K cm2 s/cal K m2 h/kcal F ft2 h/Btu
2
Km /W 1.0000 1.0000E4 4.187E4 1.1630 5.6780
K cm2/W 1.000E-4 1.0000 4.1870 1.163E-4 5.678E-4
K cm2 s/cal 2.388E-5 0.2388 1.0000 2.778E-5 1.356E-4
K m2 h/cal 0.8598 8.598E3 3.600E4 1.0000 4.8820
F ft2 h /Btu 0.1761 1.761E3 7.272E3 0.2048 1.0000

19
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
12.1.7 Latest apparatus characterization verification date loss, in relation to metering box heat input, is achieved by
and procedure used. References for the characterization re- using large boxes where the ratio of perimeter to area is less,
port(s) shall also be included. and by minimizing of any highly conductive layers or skins
12.1.8 Average pressure differential across the test specimen flanking the specimen at its perimeter. Also in general, for a
and the average airflow volume rate, if applicable. particular apparatus, the uncertainty will decrease as the heat
12.1.9 Report temperatures, both air and surface, on each transfer through the specimen increases.
side of the test specimen as follows: NOTE 29—As an example, an outline of the procedure for an uncer-
[Link] For uniform test specimens, report the average tainty analysis for thermal resistance, R, is as follows:
temperatures over the test specimen area. From 3.4.3, R = (t1 – t2)A / Q where, Q, is the power through the
[Link] For non-uniform test specimens, including test specimen. The net energy input to the metering box is, Qin = Qh + Qf +
elements, separate measured temperature averages for each Qcool; the energy into the metering box through its walls is, Qmw; and the
different area or element must be given. Areas for each element flanking loss power is, Qfl; such that Q = Qin + Qmw + Qfl (other terms such
shall also be reported. as blower input or cooling may be added as needed). Combining these
equations, the relation for resistance is R = (t1 – t2)A / (Qh + Qf + Qcool +
12.1.10 The net heat transfer through the test specimen,
Qmw + Qfl). The individual uncertainty for each item in this equation must
steady-state average rate. Include values for metering box loss, be estimated. Such estimates shall be made from knowledge of individual
flanking loss, surround panel heat flow, surround panel opening instrument and transducer uncertainty or from the results of characteriza-
flanking loss, and other energy flows included in the net energy tion experiments designed to investigate such uncertainties. Then, follow-
calculation. ing the propagation of errors theory that assumes the errors to be
12.1.11 Any thermal transmission properties calculated in independent and not systematic, the uncertainties are combined by adding
11.3, and their estimated uncertainty (see 13.1 and Note 29). in quadrature (square root of the sum of the squares) the absolute
uncertainties for sums and the relative uncertainties (fractional or percent-
12.1.12 A full description of test procedures and data age of the variable) for the products or quotients.
analysis techniques used. NOTE 30—Uncertainty estimates for existing apparatus range from 1 to
12.1.13 The test-start date and time, the time required to 10 %. Published estimates include 0.75 to 1.0 % according to Mumaw (2)
establish steady temperature conditions, the time to reach and to Miller et al (4) and from 1.5 to 3 % according to Rucker and
steady state, the data acquisition time period and frequency, Mumaw (9). A 5 % agreement with guarded hot boxes was also reported
and the test-end date and time. by Miller et al (4). Unpublished estimates range from less than 2 % for a
12.1.14 Include a statement of laboratory accreditation of large box operated with a temperature difference of 56°C to 10 % when
the same box is operated with a temperature difference of 14°C for a high
the test facility used, if applicable. resistance (5.3 K m2/W) specimen. Also see Refs (29) and (32) for recent
12.2 Precaution: Where this test method might be specifi- discussions of the uncertainty determination.
cally referenced in published test reports and published data
claims, and where deviations from the specifics of the test 13.2 Interlaboratory Comparison Results:
method existed in the tests used to obtain said data, the 13.2.1 Background—During the 1980’s, a round robin for
following statement shall accompany such published informa- guarded and calibrated hot boxes was conducted with 21
tion: “This test did not fully comply with following provisions laboratories participating, 15 boxes were guarded while 6 were
of Test Method C 1363.” (followed by a listing of specific calibrated hot boxes. The design of the round robin is described
deviations from this test method and any special test conditions by Powell and Bales (14). Data were reported for 100 mm thick
that were applied) homogenous specimens of expanded polystyrene board. Each
laboratory received material from a special lot whose produc-
13. Precision and Bias tion was specially controlled to ensure a uniform product
13.1 Uncertainty Estimation—The precision and bias of this density. At a mean temperature of 24°C, the average R-value
method depends upon the test equipment and operating proce- was determined to be 2.81 K m2/W. The regression equation
dures, and upon the test conditions and specimen properties for each data set was:
(24, 34). For this reason, no simple quantitative statement can Rguarded 5 3.146 2 0.016 · Tmean (11)
be made that will apply to all tests; however, in order to comply Rcalibrated 5 3.265 2 0.016 · Tmean (12)
with the requirements of 12.1.11, it is necessary to estimate the
uncertainty of results for each test to be reported. Such over a mean temperature range of 4 to 43°C. The mean
estimates of uncertainty can be based upon an analysis using specimen density ranged from 20.2 to 23.9 kg/m3. The report
the propagation of errors theory discussed in textbooks on of this round robin was prepared by Bales (19).
engineering experimentation and statistical analysis; see for NOTE 31—These results are for hot boxes built to C 236 and C 976
example Schenck (13) or ISO Standard 8990. These estimates specifications. These two standards were combined in the development of
can be augmented by the results of interlaboratory test com- this test method. Additional refinements were added here in hopes of
improving testing performance.
parisons (round robins), and by the results of experiments
designed to determine repeatability of the effect of deviations 13.2.2 Precision—At a specimen thermal resistance of R =
from design test conditions and by measurements of transfer 2.81 K m2/W and on the basis of test error alone, the difference
standards from appropriate standards laboratories. In general, in absolute value of the test results obtained from two
the best overall accuracy will be obtained in apparatus with low laboratories on this same specimen material lot will be ex-
metering box wall loss and with low flanking loss. Low pected to exceed the reproducibility interval only 5 % of the
metering box wall loss is achieved by using highly insulated time. The reproducibility intervals based upon this round robin
walls subjected to low temperature differences. Low flanking are presented in Table 2. For example, measurements from two

20
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
TABLE 2 Reproducibility Test Results—Homogeneous building panel, which contains large thermal bridges. The
Specimens—ASTM Hot Box Round Robin (19) accuracy of the results of the overall thermal transmission, U,
Mean Reproducibility Difference at the test conditions are expected to be equal to that of other
Temperature Interval (%) in Resistance
(°C) Calibrated Guarded (m2K/W)
specimens. The problem is with the determination of the
4 13.6 14.6 6 0.22 surface-to-surface thermal resistance, R, which is expected to
24 14.4 15.6 6 0.22 have greater uncertainty due to problems with defining the true
43 15.4 17.2 6 0.22
surface temperatures (30).
13.2.5 The precision and bias of the hot box apparatus used
different laboratories using a calibrated hot box on this same for testing windows has been evaluated and the results pub-
specimen lot would be expected to differ less than 14.4 % at a lished by the National Fenestration Research Council (NFRC).
mean temperature of 24°C, 95 % of the time. These results are from their annual round robin testing of the
13.2.3 Bias—Based on guarded hot plate data, (see Test NFRC accredited laboratories which perform testing using the
Method C 177), from the National Institute of Standards and C 1363 and C 1199 test procedures. The results of the most
Technology and supported by measurements from other labo- recent published survey show an uncertainty (two standard
ratories, the average value for the round robin specimen is a deviations units) of 6 0.23 W/m2K for a non-thermally broken,
thermal resistance of 2.81 K m2/W at an average density of aluminum framed, horizontal slider window having an average
20.8 kg/m3 (19). The mean value as measured by the composite thermal transmission of 3.2 W/m2K. The 2001 results on a
of the calibrated hot boxes was 2.88 (K m2/W) or 2.7 % greater thermally improved, aluminum fixed window with high per-
than expected from the hot plate tests. The mean value as formance glazing (Ust = 2.3 W/m2K, showed a reproducibility
measured by the composite of the guarded hot boxes was 2.78 limit of 13.8 % at 95 % confidence level and a coefficient of
(K m2/W) or 1.1 % less than the expected value. All measure- variance of 4.92 %. Eight testing laboratories participated in
ments were made at a mean temperature of 24°C. these round robins. No bias was calculated since the “true”
NOTE 32—Both round robins used quasi-homogeneous specimens as- value was not known (28, 32).
sembled from multiple pieces of the polystyrene board stock. While this 13.3 No interlaboratory comparison exists for this latest
specimen approximates an ideal wall section, it cannot be represented by version of the hot box method. An interlaboratory comparison
the homogeneous board stock material due to the presence of joints and of this test method is planned as soon as it is available and the
surface treatment. The precision and bias statement above gives an laboratories have had time to modify their apparatus to meet
indication of those values expected for this specimen lot only and may not
the requirements of this test method, if necessary.
represent the values expected for either a non-homogeneous wall section
(that is, real walls) or for a specimen that is truly uniform in density and
material properties. 14. Keywords
13.2.4 The precision and bias of the hot box apparatus has 14.1 building assemblies; building materials; hot box; test
not been confirmed for building sections, such as a metal method; thermal properties; thermal resistance

21
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

ANNEXES

(Mandatory Information)

A1. INTRODUCTION TO THE ANNEXES

A1.1 This introduction provides a brief description of each A1.1.7 Annex A7: An Example of a Hot Box Characteriza-
of the Annexes provided. tion Test Program—Provides an example of characterization
A1.1.1 Annex A1: Introduction to the Annexes—Provides a test results used to determine the metering box wall transducer
brief summary of all the Annexes. output and flanking loss coefficients.
A1.1.2 Annex A2: Heat Balance in a Hot Box Apparatus— A1.1.8 Annex A8: Using the Hot Box To Determine the Heat
Provides a general overview of the heat balance within a hot Transfer Through Specimens Smaller Than the Metering
box apparatus. Area—Explains how to use a surround panel to measure the
A1.1.3 Annex A3: Estimating the Metering Box Wall Loss— thermal resistance of specimens smaller than the metering area
Describes the physics of the metering box wall loss. Also of the hot box.
describes the characterization tests required to determine the A1.1.9 Annex A9: Determination of the Environmental Tem-
heat flow through the metering box walls in relation to the perature in the Hot Box Environment—Describes how to
metering box wall transducer output. calculate the environmental temperature in both chambers of
A1.1.4 Annex A4: Estimating the Flanking Loss—Defines the hot box. These values are used to determine the thermal
the concept of the flanking loss. Also describes methods for resistance of all specimens.
modeling and model verification of the flanking loss in a hot A1.1.10 Annex A10: Recommended Practice for Estimation
box apparatus. of the Testing System Time Constant—Provides a methodology
A1.1.5 Annex A5: Preliminary Hot Box Characterization— in which the time constant of the thermal chamber and
Outlines the initial testing required for the initial setup of the specimen can be estimated and measured.
metering box wall transducers. A1.1.11 Annex A11: Design and Construction of the Hot
A1.1.6 Annex A6: Experimental Determination of the Box Characterization and Surround Panels—Specifies how to
Flanking Loss and Metering Box Wall Loss Model assemble and instrument characterization and surround panels.
Coeffıcients—Describes the development of the testing matrix These panels are used in the characterization tests specified in
for establishing the relationships between the hot box heat Annex A5 through Annex A7 and in testing specimens smaller
flows and the instrumentation output signals. than the metering area as described in Annex A8.

A2. HEAT BALANCE IN A HOT BOX APPARATUS

A2.1 Hot boxes are designed to measure the heat transfer scribes the development of a testing matrix for characterization
through a specimen when the environmental conditions on both and then Annex A7 provides an example of the use of a text
sides of the specimen are held constant. Tests are typically matrix to obtain the characterization coefficients. Annex A8
performed with a significant temperature difference across the describes the additional steps required to measure specimens
specimen, and with the air temperatures on both sides being smaller than the metering area. Annex A11 describes the design
held at fixed values. In addition, the air velocities on both sides and construction of the characterization and surround panels
of the specimen are measured and held constant during the test. required for testing in this apparatus.
Once the environmental conditions are stable, as defined by the
steady state criteria, the net heat flow into the metering box is A2.3 The tests described in Annex A5 through Annex A8
carefully measured. depend upon establishing a heat balance between the metering
chamber and the surrounding environment. These tests are
A2.2 The measured value of heat flow is then adjusted performed using homogeneous characterization panels, which
based on the results from characterization tests described in are instrumented on both sides to determine the surface
these Annexes. Annex A2-Annex A4 describe the basics temperature difference across them. A schematic of the heat
relating to the metering box wall heat flow and the flanking flows in a hot box and their sources is shown in Fig. A2.1.
loss, respectfully. Annex A5 outlines the tests necessary to A2.3.1 The equation that describes the total heat balance of
initialize the metering box wall transducers. Annex A6 de- the metering box is:

22
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

FIG. A2.1 Hot Box Heat Flow Diagram

Qaux 1 Qmw 1 Qfl 5 Q 5 A · Dt / R (A2.1) metering box wall thermal resistance. One of these tests shall
be performed with the guarding temperature above the meter-
where: ing box air temperature. The second test has the guarding
Qaux = net heat flow due to the fan, heater, and cooling
temperature approximately equal to the metering box air
coil, W,
temperature. And finally, one test shall be performed with the
= Qc + Qh + Qf, W,
Qc = net heat removed by the cooling coil, W, guarding temperature below the metering box air temperature.
Qh = net heat added by the heaters, W, All control parameters shall be held constant during each test.
Qf = net heat added by the fans, W, A2.4.2 Once the value for Qmw is determined for each test
Qmw = metering box wall loss, W, using Eq A2.1; the results are plotted versus the transducer
Qfl = flanking loss, W, output E. The slope of the line, m, and the y-intercept, Eo, as
= Qfl, m-g + Qfl, m-c, W, described by Eq A2.2 are determined from the plot or by fitting
Qfl, m-g = flanking loss from the metering chamber to the the data to Eq A2.2.
guard, W, A2.4.3 Ideally, a set of tests shall be performed where the
Qfl, m-c = flanking loss from the metering chamber to the heat flow through the specimen was negligible (Dt across the
climate chamber, W, specimen is zero), and any heat flow measured in the metering
Q = heat flow through the specimen, W, box is attributed to the metering box wall heat flow. Unfortu-
R = thermal resistance of the specimen, m2·K/W, nately, not all hot boxes operate at environmental conditions
A = metered area of heat flow, m2, and where the temperature differences across the specimen are
Dt = surface temperature difference across the speci- close to zero. Therefore, this test method specifies a character-
men, K. ization methodology in Annex A6 where the coefficients
A2.3.2 Recall that Qmw is a function of the transducer representing the metering box wall loss and flanking loss are
output, E, described by Eq A2.2. From an operational stand- combined, and may not be individually identified.
point, the objective of proper metering box operation is to A2.4.4 For those hot boxes that have metering boxes with
make Qmw equal to or nearly zero. active metered refrigeration, the thermopile zero offset is
Qmw 5 ƒ ~E! 5 m · E 1 Eo (A2.2) determined separately from the thermal chamber flanking loss
coefficient. Hot boxes with active metered refrigeration can
where: determine the thermopile zero offset by setting the temperature
E = thermopile voltage, V,
difference across the specimen surface equal to zero (Q = 0).
m = slope of the metering box loss versus thermopile
Substituting Eq A2.2 into Eq A2.1 and setting Q = 0 reduces Eq
output relationship, and
Eo = zero offset for the thermopile output, V. A2.1 to:
Qaux 5 2~m · E 1 Eo! (A2.3)
A2.4 Metering Box Wall Loss Determination: A2.4.4.1 Notice that setting the temperature difference
A2.4.1 To quantify m in Eq A2.2, three, or more, test runs across the test specimen to zero also forces the flanking loss
shall be performed with differing levels of E. In practice, the Qfl, m-c to be equal to zero.
adjustment of the value of E is accomplished by adjusting the A2.4.5 Those thermal chambers that cannot perform the
guard temperature while holding the other temperatures con- metering box wall heat flow test with zero temperature
stant. The level of change required for analysis of the relation- difference across the specimen, shall use the procedure out-
ship will depend upon the transducer sensitivity and the lined in Annex A6 to determine a combined metering box wall

23
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
heat flow and flanking loss coefficient. The coefficient is valid A2.5.3 As previously mentioned in A2.4.2, many hot boxes
only at the environmental conditions at which the tests were are not configured to measure the flanking loss separately from
performed. Therefore, separate tests shall be performed at the the thermopile zero offset. The process used to establish the
environmental conditions expected during actual testing. combined flanking loss and metering box wall thermopile
coefficients is similar to the procedure used to establish the
A2.5 Flanking Loss Determination: metering box wall transducer coefficients when the temperature
A2.5.1 Typically, flanking losses can occur at two locations difference across the specimen surface is equal to zero. When
in a hot box. One source of flanking loss is the heat transfer the temperature difference across the specimen is not zero, the
between the metering box and the guard or climate chambers flanking loss coefficient and the heat flow through the specimen
around the contact point where the metering box wall touches shall be included in Eq A2.3 to produce a new heat balance
the specimen. Examples of this flanking loss are shown in Fig. equation shown in Eq A2.4. By solving Eq A2.4 at the range of
A2.2, for thermal chambers with a guard chamber, and in Fig. testing conditions using a variety of panels, temperatures, and
A2.3, for hot boxes where the metering box has the same sized wind speeds, the hot box flanking loss can be fully character-
aperture as the climate chamber. This extraneous heat flow is ized. As a minimum, solving Eq A2.4 requires a minimum of
considered to be the flanking loss associated with that specific three tests for each characterization panel. At each environ-
metering chamber at that environmental condition, and deter- mental condition, the guarding temperature is set to different
mining its value is the primary focus of Annex A4. The second levels, but all other temperatures remain constant.
area where flanking loss occurs is between the metering box ~A · Dt / R! 2 Qaux 5 m · E 1 @Eo 1 Qfl# (A2.4)
and the climate chamber through the edge of the aperture of the
surround panel holding a specimen smaller than the metering A2.5.4 The results from performing numerous tests at the
area. This flanking loss only occurs when a test specimen range of heat flows, temperatures and wind speeds experienced
smaller than the metering aperture is mounted in a surround during testing will generate different values of the combined
panel. Although the specimen flanking loss can be estimated by coefficient, [Eo + Qfl] in most hot boxes. The slope of the linear
computer simulation or testing, typically the heat flow associ- equation, m, remains relatively constant throughout these tests,
ated with test specimen flanking loss is included with the test but the flanking loss may vary with changes in the environ-
specimen heat flow, which slightly decreases the measured mental conditions or heat flow. The values of m, and [Eo + Qfl]
thermal resistance of the test specimen (33). See A8.4 for a shall be analyzed in relation to the changes in environmental
more detailed discussion of test specimen flanking loss. conditions and heat flow to establish a unified and consistent
A2.5.2 The metering box wall flanking loss is a source of methodology to adjust the measured heat flow of actual
error when measuring the heat flow through a metered speci- specimens based on these tests. Since the actual specimen will
men. This method requires that the flanking loss be determined have a significantly different construction than the relatively
using the tests described in Annex A6. Unfortunately, the heat homogeneous characterization panels, the test operator shall
flow due to flanking loss cannot be measured directly. In have to make a judgment as to which values of m, and [Eo +
addition, the flanking loss has been shown to vary with the test Qfl] should be used for a particular test. In some cases, it shall
conditions. In some of these hot boxes, the flanking loss varies be necessary to interpolate between different values of m, and
more with the change in air velocities than with the change in [Eo + Qfl] based on the assembly of the specimen or test
air temperatures. In these circumstances, a series of tests shall conditions.
be performed at all the environmental conditions experienced A2.5.5 One of the consequences of using the results from
during testing to properly quantify the heat flow due to flanking the characterization tests to adjust the results from testing
loss. actual specimens is that any systematic errors present in the hot

FIG. A2.2 Guard Chamber/Metering Box Interface

24
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

FIG. A2.3 Hot Box Wall or Frame Interface with Specimen

box and instrumentation shall be included in the combined the cause of the flanking loss and thermopile zero offset within
flanking loss and thermopile zero offset coefficient, [Eo + Qfl]. their specific thermal chamber. To help understand the reasons
For this reason, it is critical to accurately measure the heat for flanking loss, compare the heat flow that is calculated using
input into the metering chamber by the heaters, fans or the procedures outlined in A7.4 to the results generated by the
instrumentation, Qaux. Not only is it important to have a test analysis as shown in Fig. A7.3. The reason for any
combined coefficient that is relatively small compared to the differences shall be identified.
net heat flow, but also the test operator should try to understand

A3. ESTIMATING THE METERING BOX WALL LOSS

A3.1 The heat flow through the metering box walls is where, the effective area normal to heat flow, m2, is given by:
estimated by various means, which differ in accuracy and the Aeff 5 Ain 1 0.54 · L · S ei 1 0.60 · L2 (A3.2)
level of effort required. The heat flow of the metering box shall
be estimated during the design of the hot box to refine the final where:
construction. In addition, the predicted heat flow shall then be Ain = metering chamber inside surface area, m2,
compared to the actual values measured in Annex A6 as a L = metering chamber wall thickness, m,
gauge of meeting the design goals. The procedures described leff = metering chamber effective wall thermal conductiv-
below assume that the hot box apparatus is designed to have ity, W/mK,
generally uniform airflow and temperatures at each surface of tin = metering chamber inside wall surface temperature,
the metering chamber walls. K,
tout = metering chamber outside wall surface temperature,
A3.2 Model Prediction: K, and
S ei = sum of all (total of 8) metering chamber interior
A3.2.1 The following equations represent one method of edge lengths formed where two walls meet, m.
estimating the heat flow through the walls of a five-sided A3.2.2 There are numerous two-dimensional computer
rectangular metering box made of homogeneous material. analysis tools that can be used to estimate the heat flow through
Langmuir (15) estimates the metering chamber wall heat flow the metering box walls. Typically, these computer programs
to be equivalent to one-half that of a closed six sided box require detailed cross sections of the metering box wall at all
formed by placing two of the open sided boxes together. The locations that are representative of the metering box wall
heat flow per unit time for the five-sided box is given by: construction. The thermal conductance and emittance of all the
leff · Aeff · ~tin 2 tout! building components in those cross sections are input into the
q5 (A3.1)
L computer models to determine the heat flow through those

25
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
sections. The total heat flow through the metering chamber grams allow detailed analysis of the convection and radiation
walls is determined by area weighting the computed heat flow environments encountered in the hot box. Typically, these
through the various cross sections. computer programs require detailed three-dimensional com-
A3.2.3 The most accurate method of calculating the heat puter drawings of the metering box wall construction, as well
flows in and out of the metering box is by using three- as the thermal conductance and surface heat transfer coeffi-
dimensional computer analysis tools. These computer pro- cients of all the building components.

A4. ESTIMATING THE FLANKING LOSS

A4.1 The flanking loss is defined to be the quantity of heat, A4.4.1 By modeling the flanking loss, a better understand-
which flows between the metering and climatic chambers ing of the mechanisms of extraneous heat flows for a particular
through the surround panel or test frame, which holds the hot box is achieved. The objective is to characterize the
specimen. The flanking loss from the metering chamber to the flanking loss using a simplified representation of the extrane-
guard chamber that passes through the specimen, Figs. A2.2 ous heat flow based on actual measurements. Typically, the
and A2.3, have been discussed in A2.5. This analysis is flanking loss shall be represented using regression equations
applicable to the flanking loss through the surround panel at its based on known thermodynamic properties.
interface with the specimen, Fig. A4.1. This loss also occurs at A4.4.2 For ease of calculation of the flanking loss correc-
the opening when the surround panel thickness is different tion, the heat flow along two paths is lumped together and
from the specimen thickness. The flanking loss is expected to described as a single path with an effective thermal conductiv-
be a function of the construction through which the flanking ity, length, and area. The exact form of this equation will be
loss passes, the temperatures, the air velocities in both cham- guided by the modeling results. In thermal chambers where the
bers, and the thickness and construction of the specimen. air velocities on both sides are always constant, the flanking
loss has been successfully predicted using the following
A4.2 Before the flanking loss tests are performed, a equation form:
preliminary analysis must be made to predict the magnitude of Qfl 5 leff · ~A/L!eff · Dta2a (A4.1)
the flanking loss as a function of the appropriate variables. For
example, the air-to-air temperature difference between the where:
chambers; associated mean temperature; surface heat transfer Qfl = flanking loss, W,
coefficients; and the specimen construction and thickness all leff = effective thermal conductivity of base insulation
may be significant. and the skin material, W/(m · K),
(A/L)eff = effective area/path length of entire frame around
A4.3 Fig. A4.1 shows a cross-section of the joint between its perimeter, m, and
the surrounding panel opening and the specimen. The primary Dta-a = air-to-air temperature difference, K.
direction of the flanking heat flow is parallel to the surround A4.4.3 Strictly speaking, the effective thermal conductivity
panel opening surface skin. Since the skin has a fairly high is a function of temperature, since the thermal conductivity of
thermal conductivity compared to the internal insulation, it the base insulation and skin vary with temperature. The
cannot be ignored as a heat flow path. The flanking loss occurs effective path length and area will clearly be a function of
through both the skin and the insulation beneath. For this specimen thickness, since varying the specimen thickness will
analysis, the use of a two-dimensional or three-dimensional change the geometry of the problem. As the specimen thick-
finite element or finite difference model is recommended. ness is increased, the path length for flanking loss will increase.
So, the function (A/L)eff will decrease with increasing specimen
A4.4 Modeling the Flanking Loss: thickness.

FIG. A4.1 Specimen—Surround Panel Opening Interface

26
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
A4.4.4 The model above presupposes one-dimensional heat factors. An example of this analysis is presented in Annex A7.
flow through the specimen. In actuality, the heat flow will be Each variable shall be tested at its range of expected values.
two-dimensional or three-dimensional near the frame. For this This would include, as a minimum, tests at several thicknesses,
reason, a two-dimensional or three-dimensional model is mean temperatures, temperature differences and air speeds.
preferred. There are numerous two-dimensional and three-
dimensional computer analysis tools that can be used to model A4.6 Limitations—Consideration shall be given to various
and estimate the flanking loss. These models typically require possible sources of errors in the flanking loss calibration
that a representation of the metering box wall construction be procedure (33). The three listed below are highlighted for
input into the computer including the thermal conductivity, consideration.
surface emittance, air temperatures and surface heat transfer
A4.6.1 The flanking loss equation developed using one
coefficients of all the appropriate components. Much of the
particular frame may differ slightly for other frames of the
difficulty in modeling the flanking loss is assigning the proper
same general construction.
air temperatures and surface heat transfer coefficients to use in
the analysis. A4.6.2 The data analysis assumed that the specimen heat
A4.4.5 Once the computer models are operational, a sensi- flow can be calculated as Q = C · A · Dts-s. This presupposes
tivity analysis shall be performed to determine the effects of one-dimensional heat flow through the specimen. In reality, the
variation in the specimen construction, air temperatures, and heat flow is two-dimensional near the interface.
surface heat transfer coefficients on the flanking loss heat flow. A4.6.3 Finally, the testing and analysis are generally per-
formed on homogeneous specimens. It is not known whether
A4.5 Experimental Model Verification: flanking loss would be greatly different for a non-homogeneous
A4.5.1 Once the relationships between the various factors specimen. It is conceivable that a multi-layer wall, in which the
controlling the magnitude of flanking loss is determined, it is layers vary significantly in thermal conductivity, would behave
necessary to conduct a series of tests on known specimens in differently. The model used in this calibration can be used to
order to determine the equation coefficients for the various investigate these concerns for the particular box construction.

A5. PRELIMINARY HOT BOX CHARACTERIZATION

A5.1 The procedure given in this section outlines the steps zero when there is no heat flow through the metering chamber
required to verify the proper output of the metering box wall walls. This method helps determine if the thermopile used to
thermopile, and to obtain the initial relationship between measure the temperature difference (and heat flow) across the
metering chamber wall heat flow, metering box loss, and its metering box wall is wired properly. The construction of the
transducer output. The latter series of tests addresses the metering box wall thermopile is described in 6.5.4, and the
technique that will yield the heat flow relationship as a function proper operation of the thermopile shall be verified before
of the transducer output including a zero offset, if present. In additional calibration tests are performed.
addition to the verification tests described in this Annex, the
A5.4.2 Install a characterization panel, as specified in An-
flanking loss characterization tests described in Annex A6 shall
nex A11, in the thermal chamber. Do not start any fans, heaters,
be performed before testing actual specimens.
or instrumentation, which generates heat (that is, hot wire
NOTE A5.1—Alternate procedures to evaluate the slope and offset of anemometers). Record the ambient laboratory air temperature,
the metering box heat flow and flanking loss are acceptable, if documented
the temperatures of the air, baffle surfaces, and surround panel
and verified experimentally.
in the hot box, and the output from the metering box wall
A5.2 To perform the required tests, a characterization transducer (thermopile) for at least 24 h after the hot box has
panel, as described in Annex A11, shall be instrumented and reached steady state conditions with the surrounding laboratory
installed in the hot box. This panel shall fill the available environment.
dimensions of the test frame. The metering wall characteriza- A5.4.3 Once the hot box has reached steady state conditions
tion cannot be performed using a test specimen smaller than the with the ambient environment (this may take days to achieve);
metering chamber opening. the surface and air temperatures in the climate chamber,
A5.3 It is essential that the air velocities, power inputs and metering chamber, and guard chamber (if present) should be
temperatures for the metering, guard and climatic chambers be close to each other. Therefore, the output from the metering
held constant throughout each test. By holding all the control box wall transducer should be close to zero. There may be a
parameters constant, the operator decreases the variability of small cyclic output from the metering box wall transducer
the surface heat transfer coefficients on the panel during the based on the diurnal fluctuation of temperature in the surround-
test. ing laboratory, but the average output over 24 h shall be nearly
zero. If the average output from the metering box wall
A5.4 Verifying Metering Box Wall Transducer Null Offset: transducer is not close to zero, then the wiring of the metering
A5.4.1 This procedure outlines a verification test required to box wall thermopile shall be checked and repaired, if neces-
confirm that the metering chamber wall transducer output is sary, before additional tests are performed.

27
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
A5.5 Preliminary Characterization of Metering Wall Trans- such that the fans operate at minimum levels and the heaters
ducer: are barely activated. Use the minimum heat in the metering box
A5.5.1 This describes the process to determine the relation- to maintain temperature control. Adjust the climatic chamber
ship between the output from the metering box wall transducer temperature to match the metering chamber temperature. In
and the heat flow through the metering box walls when the this configuration, no heat, Q, is flowing through the specimen,
temperature difference across the characterization panel is and thereby, all the net heat into the metering box is lost (or
close to zero. The environmental conditions generated during gained) through the metering box walls.
this test are significantly different from actual test conditions, A5.5.2.2 As described in A2.4 and A2.5, separate tests shall
and therefore the results from these tests are only used to to be performed with the guarding temperature set at different
establish the initial value of the coefficient that is multiplied by values, but with the metering room air temperature held
the output from the metering box wall transducer to determine constant. The fans generating the airflow on both sides of the
the metering box wall loss. Any offset due to flanking loss or thermal chamber shall also be set at constant speeds. It is
other anomalies is determined by the tests are described in recommended that, as a minimum, one test be performed with
Annex A6. the guarding temperature above the metering chamber air
A5.5.2 Hot boxes that do not have the capacity to cool the temperature, one test with the guarding temperature equal to
metering chamber with a metered, active refrigeration system the metering room air temperature, and one test be performed
will have difficulty performing the test described in this with the guarding temperature below the metering box air
section. For this reason, it is not mandatory to perform this test, temperature.
but it is recommended. If active metering chamber cooling is
A5.5.2.3 For the condition where the surface temperature
not available, then it is possible to reach steady state conditions
by installing a characterization panel with a low thermal difference across the known panel is close to zero, the flanking
resistance. This panel shall be installed and instrumented as loss is also zero, and the heat balance can be determined by Eq
specified in Annex A11. An alternate procedure to determine A2.3. By plotting the heat flow versus the output from the
the relationship between the output from the metering box wall metering box wall thermocouple, the slope and the zero offset
transducer and the heat flow through the metering box walls is as described by A2.2 can be determined.
described in Annex A7. A5.5.3 The measured metering box wall heat flow shall also
A5.5.2.1 Perform a minimum of three tests with the meter- be compared to the theoretical value calculated in Annex A3. If
ing box air temperature equal to the climate chamber air there is a significant discrepancy between the measured and
temperature, but with the guarding temperature set to different calculated heat flow, conduct an investigation to identify the
values. If so equipped, the metering box air temperature shall reason for this discrepancy. If all systems are operating
be controlled with the assistance of a metered refrigeration satisfactorily, use the measured coefficients when performing
system. Otherwise the metering box controls shall be adjusted the calibration tests specified in Annex A6.

A6. EXPERIMENTAL DETERMINATION OF THE FLANKING LOSS AND METERING BOX WALL LOSS
MODEL COEFFICIENTS

A6.1 Characterization of the hot box apparatus is required conditions anticipated. The example in Table A6.1 is for a
before testing of products can begin. The complete character- window test apparatus having a window smaller than the
ization of the apparatus serves to verify the assumptions made metering chamber opening. The apparatus is to be used only at
in the design and to quantify the extraneous heat transfer paths one set of temperature conditions and only one mode of
seen during operation of the apparatus. The objective of this operation but for three different surround panel thicknesses.
annex is to provide examples for the test condition combina- For this matrix the set of calibration tests is small. For the
tions of the metering chamber, surround chamber and climatic example presented in Table A6.1 as few as 6 tests may be
chamber temperatures that are required to fully characterize the adequate. If only one surround panel thickness is used, fewer
apparatus. The choice of the test matrix is based upon the style tests are possible. On the other hand, in the example of Table
of apparatus construction, the mode of operation and the test A6.2, for an apparatus used for walls and windows over a wide

TABLE A6.1 Test Matrix for Characterization of an ASTM C 1363 Hot Box Hot Box—For Fenestration Testing per ASTM C 1199A
Metering Chamber Guard ChamberB Climate Chamber Metering Wall Surround
Test No. Air Temperature, Air Temperature, Air Temperature, Thermopile Output, Panel Thickness,
°C °C °C Volts mm
Nul Ambient Ambient Ambient 0 114
1 21.1 21.1 −17.8 0 114
2 21.1 18.3 −17.8 + 114
3 21.1 23.9 −17.8 − 114
4 21.1 21.1 −17.8 0 152
5 21.1 21.1 −17.8 0 203
A
This matrix does not include the tests required to characterize the surface coefficients needed for window testing as specified in Test Method C 1199.
B
Guard chamber or surrounding laboratory environment.

28
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
TABLE A6.2 Test Matrix for Calibration of an ASTM C 1363 Hot Box
Hot box used for general testing at different environmental conditions including multiple air velocities.
Metering Chamber Guard ChamberA Climate Chamber Metering Wall Calibration Metering Chamber Climate Chamber
Test No. Air Temperature, Air Temperature, Air Temperature, Thermopile Output, Panel Thickness,B Air Velocity, Air Velocity,
°C °C °C Volts mm m/s m/s
1 21.1 21.1 −17.8 0 50 0.2 5.4
2 21.1 18.3 −17.8 + 50 0.2 5.4
3 21.1 23.9 −17.8 − 50 0.2 5.4
4 21.1 21.1 −17.8 0 50 0.4 5.4
5 21.1 18.3 −17.8 + 50 0.4 5.4
6 21.1 23.9 −17.8 − 50 0.4 5.4
7 21.1 21.1 −17.8 0 50 0.2 1.3
8 21.1 18.3 −17.8 + 50 0.2 1.3
9 21.1 21.1 −17.8 − 50 0.2 1.3
10 21.1 21.1 −17.8 0 50 0.4 1.3
11 21.1 18.3 −17.8 + 50 0.4 1.3
12 21.1 23.9 −17.8 − 50 0.4 1.3
13 21.1 21.1 −17.8 0 114 0.2 5.4
14 21.1 18.3 −17.8 + 114 0.2 5.4
15 21.1 23.9 −17.8 − 114 0.2 5.4
16 21.1 21.1 −17.8 0 152 0.2 5.4
17 21.1 18.3 −17.8 + 152 0.2 5.4
18 21.1 23.9 −17.8 − 152 0.2 5.4
Hot box used for wall and fenestration testing at different environmental conditions.
Metering Chamber Guard ChamberA Climate Chamber Metering Wall Calibration Metering Chamber Climate Chamber
Test No. Air Temperature, Air Temperature, Air Temperature, Thermopile Output, Panel Thickness,B Air Velocity, Air Velocity,
°F °F °F Volts in. mph mph
19 21.1 21.1 −17.8 0 203 0.2 5.4
20 21.1 18.3 −17.8 + 203 0.2 5.4
21 21.1 23.9 −17.8 − 203 0.2 5.4
22 37.8 37.8 10.0 0 50 0.2 1.3
23 37.8 35.0 10.0 + 50 0.2 1.3
24 37.8 40.6 10.0 − 50 0.2 1.3
25 37.8 37.8 10.0 0 50 0.4 1.3
26 37.8 35.0 10.0 + 50 0.4 1.3
27 37.8 40.6 10.0 − 50 0.4 1.3
28 37.8 37.8 10.0 0 114 0.2 1.3
29 37.8 35.0 10.0 + 114 0.2 1.3
30 37.8 40.6 10.0 − 114 0.2 1.3
31 37.8 37.8 10.0 0 152 0.2 1.3
32 37.8 35.0 10.0 + 152 0.2 1.3
33 37.8 40.6 10.0 − 152 0.2 1.3
34 37.8 37.8 10.0 0 203 0.2 1.3
35 37.8 35.0 10.0 + 203 0.2 1.3
36 37.8 40.6 10.0 − 203 0.2 1.3
A
Guard chamber or surrounding laboratory environment.
B
Or continuous surround panel.

range of temperatures, air velocities, and specimen and speci- with different panels and at different environmental conditions
men thicknesses, many more tests are required for full charac- to fully characterize the apparatus at the conditions at which
terization. The examples given below are intended to provide testing takes place.
an outline for the concept only. Each hot box shall have its own
characterization matrix that will depend upon its design and A6.3 It is essential that the air velocities, power inputs and
planned operation. temperatures for the metering, guard and climatic chambers be
held constant throughout each test. By holding all the control
A6.2 Of interest here is how to measure the combined parameters constant, the operator reduces the variability of the
flanking loss and metering box thermopile coefficient. As surface heat transfer coefficients on the specimen during the
discussed in Annex A2, the heat flow due to flanking loss is test.
difficult to measure and quantify. By solving Eq A2.1 at the
range of testing conditions using a variety of characterization A6.4 The first step in characterizing a hot box is to develop
panels, temperatures, and wind speeds, the hot box flanking a matrix identifying all the test conditions and specimens
loss shall be fully characterized. The results from performing a anticipated for testing in the apparatus. If testing includes
minimum of three tests at different guarding temperatures can fenestration products or other specimens that are smaller than
be used to solve for the coefficients in Eq A2.2 for a simple hot the metering aperture, all the surround panels constructions and
box. This matrix shall be repeated when operating the hot box thicknesses to be used shall be identified. Next, arrange the

29
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
matrix so that the environmental conditions and surround need a testing matrix between these two examples. A careful
panels are ordered from the lowest to highest or least conduc- examination of the testing conditions anticipated might limit
tive to most conductive. This matrix shall be used to identify the number of tests required. For a research apparatus, full
the range of environmental conditions and characterization characterization of the apparatus is required since the exact
panels that will be used to perform the tests. conditions of the test cannot always be anticipated.
A6.5 For the example illustrated in Table A6.1, the appa- A6.8 Perform a test at each of the environmental conditions
ratus is assumed to operate only at a 21.1°C metering chamber in the characterization matrix. As a minimum, one test, in the
air temperature and a −17.8°C climatic chamber air tempera- matrix, is performed with the guarding temperature above the
ture. The air velocities are held constant on both sides as metering chamber air temperature, the second test has the
specified in Test Method C 1199. The fenestration specimens guarding temperature equal to the metering room air tempera-
are always mounted in one of three surround panels, but of ture, and the third test is performed with the guarding tempera-
different thickness. The thinnest and thickest surround panel ture below the metering box air temperature. All other tem-
shall be tested first, and if the differences between the metering peratures and air velocities shall be held constant during a test.
box wall transducer and flanking loss coefficients are negli- Each test must meet steady state conditions as specified by
gible, then the middle thickness of surround panel may not Section 10.
have to be tested. Note that for Test Method C 1199 the
flanking loss at the interface between the window or door and A6.9 Solve Eq A2.4 for the metering box wall transducer
the surround panel is assigned to the U-factor of the fenestra- and flanking loss coefficients at each environmental condition.
tion unit and therefore does not need to be evaluated separately. By plotting the heat flow versus the output from the metering
box wall thermocouple, determine the slope, m, and the
A6.6 For the test matrix illustrated in Table A6.2, the y-intercept, [Eo + Qfl].
apparatus is operated over a wide range of temperatures, air
velocities and specimens. This would be the case for a hot box A6.10 Test the parameter values at the extremes of the test
used for testing walls at one set of temperatures and air matrix first. The test laboratory operator can then identify those
velocities, and then testing windows installed in multiple environmental conditions that do not have significant influence
thicknesses surround panels at different environmental condi- on the metering box wall transducer and flanking loss coeffi-
tions. For this configuration, the metering chamber heat flow, cients. Any parameters, which are shown to not significantly
and thermal chamber flanking loss shall be evaluated over a change the metering box wall transducer and flanking loss
wide range of test conditions. This is a most complex system coefficients can then be removed from the calibration testing
and shall require a matrix of tests up to or exceeding the 36 matrix.
listed in Table A6.2. Completion of all the tests in the matrix is
A6.11 The measured metering box wall heat flows and
not necessary if it can be shown that there is no significant
flanking loss shall also be compared to the theoretical values
variation in the metering box wall transducer and flanking loss
calculated in Annex A3 and Annex A4. If there is a significant
coefficients seen upon testing at the extremes values of a
discrepancy between the measured and calculated heat flow,
particular environmental condition.
conduct an investigation to identify the reason for this
A6.7 Hot boxes operating in many commercial laboratories discrepancy.

A7. AN EXAMPLE OF A HOT BOX CHARACTERIZATION TESTING PROGRAM

A7.1 The following example of the application of Annex the specimen for some typical conditions. Consider a 110 mm
A4 through Annex A6 to an actual hot box is presented here for thick wall with an overall thermal resistance of 2.5 m2K/W,
illustration purposes. More detailed information is available in tested at a 10°C mean temperature. Under these conditions, for
the referenced materials. This is only an example of the process the example hot box, the flanking loss (Qfl) is estimated to be
required and not a specific guideline for its application. 6 % of the specimen heat flow (Q). This is a small percentage,
but is not negligible. If Qfl could be calculated to within 10 %
A7.2 Test Apparatus—This example is based upon the
error, then the resultant error in Q would be 0.6 %. The
flanking loss discussion by Lavine et al (12) that was used for
magnitudes of Qfl and Q are strongly related, since both are
the calibrated hot box described by Mumaw (2). That hot box
proportional to the Dt across the specimen. For this example,
is a vertical wall tester with a metering area of 2.7 by 4.3 m.
the value of 6 % is typical for Qfl relative to Q. This magnitude
The chambers and specimen frame are constructed of urethane
could be significantly different for another test construction or
foam (0.5 m thick) with glass fiber reinforced polyester (GRP)
a different specimen area. In contrast to this example, if a
skins (1 to 3 mm thick). The example is specific to that facility,
plywood skin were used as the skin for the frame, it will
however the development procedure and calibration results are
provide a low thermal resistance flanking path for the flanking
useful as a guide for other hot box users.
loss. For a 13 mm thick, continuous plywood skin, the flanking
A7.3 Perspective—It is informative to note the approximate loss would exceed 10 % of the specimen heat flow under many
magnitude of the flanking loss relative to the heat flow through test conditions.

30
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
A7.4 Effect of Specimen Thickness—For the example, the from 28 to 58 K, and mean temperatures varied from −13 to
thickness dependence of the flanking loss was investigated 49°C.
theoretically using HEATING 5, a finite difference heat con-
duction program (17). A cross-section at the joint between the A7.6 Variation with Effective Thermal Conductivity—From
frame and the specimen was modeled. The metering chamber, the tests, a strong linear trend, Fig. A7.2, can be observed for
climatic chamber, and room air temperatures were taken to be both of the specimens. Since the flanking loss had been
24, −4, and 24°C, respectively. Since the metering chamber predicted to be proportional to the independent variable,
and room air temperatures were chosen to be equal, there was straight lines were fit through the data, constrained to go
no metering box wall loss, and all heat leaving the metering through the origin. This was done separately for the single and
chamber ended up in the climatic chamber. Thus, the flanking triple thickness specimens. A statistical analysis indicated
loss was simply the quantity of heat leaving the metering acceptable agreement between the data and the regression
chamber through the frame, integrated over the perimeter of lines. Thus, the predicted temperature dependence of the
the frame. flanking loss had been validated. In this series, however, the
slopes of the two regression lines indicated two values of
A7.4.1 Modeling runs were made on the example facility to (A/L)eff, one for each specimen thickness. This demonstrates
determine the effect of thickness. The thickness of the speci- the predicted thickness dependence of the flanking loss.
men ranged from 19 to 300 mm, and the specimen thermal
conductivity was held constant. (A few runs were made which A7.7 Thickness Variation—Notice that the regression of Qfl
determined that varying the specimen thermal conductivity did versus leff · Dta-a may also provide an experimental estimate of
not strongly affect the flanking loss.) Fig. A7.1 illustrates the the function (A/L)eff. In Fig. A7.3, the experimental flanking
shape of flanking loss per unit temperature difference as a loss and the theoretically predicted flanking loss are now
function of specimen thickness, as predicted by the model. plotted versus specimen thickness. If the general shape of the
Since Dta-a and leff were constant for these runs, this plot can experimental and theoretical results is in agreement, then the
be used to define the thickness dependence of the flanking loss, appropriate coefficients can be determined by regression. In
(A/L)eff. Once the functions leff and (A/L)eff had been defined, this example, the theoretical model results and the two experi-
the predicted flanking loss equation was complete. It could mental estimates of (A/L)eff do not fall on the theoretical curve,
then be compared to experimental results to determine the but that the general shape of the curve appears to be correct.
exact coefficients for the equation. Observe that the theoretical curve predicts flanking loss to be
A7.4.2 Using the model, the temperature dependence of the inversely proportional to thickness for large thicknesses (150 to
materials was estimated to have less than a 10 % effect on the 300 mm). For smaller thicknesses, the flanking loss curve
flanking loss. Since the flanking loss for the example hot box becomes very steep. The difference between the tested results
was on the order of 6 % of the specimen heat flow, temperature and the model results was attributed to differences between the
dependence of the effective frame thermal conductivity has assumed and actual physical properties and dimensions.
only a minor influence on the specimen heat flow. It was, A7.7.1 From the modeling results, it is probable that the
however, included in the final characterization equations. flanking loss dependence on thickness has the general equation
form of Eq A7.1:
A7.5 Characterization Tests—For the example character-
a
ization procedure, a series of hot box tests was run on ~A/L!eff, th 5 (A7.1)
~b 1 th!
homogeneous specimens with known thermal characteristics.
Single thickness (35 mm) and triple thickness (105 mm) where:
specimens were constructed for flanking loss as a function of a and b = model constants, and
specimen thickness. To investigate the temperature dependence th = the specimen thickness.
of flanking loss, a series of tests was performed on each A7.7.2 The two constants were solved for using the two
specimen. Temperature differences across the specimen ranged experimental estimates of (A/L)eff. The regression curve, also

FIG. A7.1

31
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

FIG. A7.2

FIG. A7.3

plotted in Fig. A7.3, gives a reasonable representation of where:

flanking loss as a function of thickness. leff = a function of mean temperature.

A7.8 Final Results—Combining the results of the regres- A7.9 Summary—The results of the analysis for the example
sions on the individual effects from our experiments yields the hot box are plotted versus mean specimen temperature for the
final equation for correction of the flanking loss as a function two test specimens, CS1 And CS2, in Fig. A7.4. The known
of the experimental variables. Thus, for the example hot box, curve of conductance versus temperature is also shown. The
the flanking loss can be described by an equation of the form: root mean square of the percentage error between the test and
Qfl 5 leff · ~a / ~b 1 th!! · Dta2a (A7.2) known values was only 0.8 %.

32
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

FIG. A7.4

A8. USING THE HOT BOX TO DETERMINE THE HEAT TRANSFER THROUGH SPECIMENS SMALLER THAN THE
METERING AREA

A8.1 General Considerations: surround panel material. For both of these reasons, specimen
A8.1.1 Hot boxes are also used to measure the thermal will have a flanking loss associated with its installation in that
resistance of specimens that are smaller than the metering area. particular surround panel. Flanking loss is shown in Fig. A8.1.
For this type of testing, the specimen consists of the specimen Although the heat flow associated with the test specimen
installed and sealed in a surround panel built in accordance flanking loss is typically assigned to the heat flow through the
with Annex A11. In this use, the specimen of area As is located test specimen, there are means of estimating its magnitude, and
centrally in the metering area, A, and is surrounded by a adjusting the final measured results. If the heat flow associated
homogeneous surround panel of area Asp = A – As. The total with test specimen flanking loss is subtracted from the speci-
heat flow rate, Q, is determined by the hot box measurement. men heat flow to calculate the reported thermal resistance of
Assuming no interaction between the two heat flow rates in the specimen, that test specimen flanking loss shall be clearly
parallel, the relationship between the individual heat flows is identified in the test report.
described by Eq A8.1:
A8.2 Surround Panel Construction—The construction de-
Q 5 Qs 1 Qsp (A8.1)
tails for the surround panels are presented in Annex A11.
where:
Qs = the total heat flow through the specimen area As, and A8.3 Characterization of the Surround Panel as a Heat
Qsp = the heat flow through the surround panel area Asp. Flow Transducer—The need to determine the surround panel
A8.1.1.1 To determine Qs, measurement is made of Q, using heat flow, Qsp, accurately requires that the surround panel be
the regular hot box procedure. The surround panel heat flow, designed to act as a heat flux transducer with a temperature
Qsp, is calculated from measurements of the temperature difference, Dt, proportional to the total heat flow through it.
difference between the surround panel surfaces and multiplying Before surround panels are used for testing actual specimens,
that value by the ratio of the area and thermal resistance of the the surround panel wall heat flow transducer and flanking loss
surround panel. The surround panel thermal resistance is coefficients shall be determined for that surround panel using
determined by means of hot box tests of the same surround the characterization tests described in Annex A4 through
panel either before the aperture for the specimen is cut out or Annex A6. These tests require that the surround panel first be
with a blank of identical thermal conductance and thickness as instrumented and calibrated with the specimen aperture filled
the surround panel installed in place of the specimen. The with material of the same thickness, conductance and assembly
characterization tests shall be performed on the surround panel as the surround panel as described in Annex A11. After the tests
at similar environmental conditions that the specimen is tested. specified in Annex A6 are performed, then the surround panel
The method of performing characterization tests on surround opening flanking loss tests described in A8.4 can be performed.
panels is described in Annex A5.
A8.1.2 As specified in Annex A11, the surround panel is the A8.4 Estimating the Surround Panel Opening Flanking
same thickness or somewhat thicker than the specimen (see Loss:
A11.3.4.3). In addition, the materials around the perimeter of A8.4.1 As described in Annex A3 and Annex A4, there are
specimen may have a greater thermal conductance than the numerous two-dimensional and three-dimensional computer

33
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

FIG. A8.1 Test Specimen/Surround Panel Interface

analysis tools that can be used to model and estimate the from the first result. An example of a characterization matrix
surround panel opening flanking loss. These models typically for a single thickness surround panel is given in Table A8.1.
require that a representation of the surround panel and the
specimen be input into the computer including the thermal NOTE A8.1—Additional uncertainty may arise due to the possible
influences of the specimen in causing two or three dimensional heat flow
conductivity, emittance, air temperatures and surface heat
at its boundary with the surround panel. The surround panel heat flow,
transfer coefficients of all the appropriate components. Much determined under a given set of conditions with a transfer standard in
of the difficulty in modeling the flanking loss is assigning the place, may change when the actual specimen is installed, even though the
proper air temperatures and surface heat transfer coefficients to test conditions remain unchanged. The user of this procedure shall attempt
use in the analysis. to evaluate the impact of this uncertainty on the desired accuracy of the
A8.4.2 The surround panel opening flanking loss shall be test.
estimated by performing a test on a transfer standard of known
thermal properties, which is the same thickness as the speci- A8.5 Uncertainty Estimation of Measuring Specimens
men to be tested. An example of such a transfer standard is Smaller Than the Metering Area:
described in detail in Annex A1 of Test Method C 1199. The A8.5.1 From Eq A8.1, the uncertainty in Qs is equal to the
transfer standard of known thermal properties is instrumented,
difference of the uncertainty in Q and Qsp. The fractional
installed and sealed into the hole in the surround panel, and a
uncertainty is given by:
characterization test is performed at the same environmental
conditions, as the test specimen will be tested. It is recom- DQs / Qs5~DQ 2 DQsp! / ~Q 2 Qsp!
mended that the transfer standard be positioned in the same 5 @~DQ / Q! 2 ~DQsp / Q!# / ~1 2 Qsp / Q! (A8.2)
position as the test specimen at the juncture with the surround where:
panel aperture. An estimate of the surround panel opening DQs = the uncertainty in Qs, etc.
flanking loss is calculated by first subtracting the expected heat
flow through the transfer standard, as determined by multiply- A8.5.1.1 An estimate of the fractional uncertainty, DQsp/Q,
ing the temperature difference across the panel by its area and is dependent upon the method used to calibrate the surround
thermal conductance, from the measured heat flow through the panel. If the characterization is made before the aperture for the
metering chamber opening. The final result is then determined specimen is cut out then:
by subtracting the heat transfer through the surround panel DQsp / Q 5~DQt / Q! 3 ~Asp / A! (A8.3)

TABLE A8.1 Test Matrix for Characterization of an ASTM C 1363 Hot Box for a Single Thickness Surround Panel
(Includes Flanking Loss)
Metering Chamber Guard ChamberA Climate Chamber Metering Wall Surround Transfer
Test No. Air Temperature, Air Temperature, Air Temperature, Thermopile Output, Panel Thickness, Panel Thickness,B
°C °C °C Volts mm mm
Nul Ambient Ambient Ambient 0 152 No Opening
1 21.1 21.1 -17.8 0 152 No Opening
2 21.1 18.3 −17.8 + 152 No Opening
3 21.1 23.9 −17.8 − 152 No Opening
4 21.1 21.1 −17.8 0 152 25
5 21.1 21.1 −17.8 0 152 76
6 21.1 21.1 −17.8 0 152 127
A
Guard chamber or surrounding laboratory environment.
B
Installed in the surround panel opening.

34
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

where: box apparatus and upon the accuracy of determination of heat

DQsp = the uncertainty in heat flow measured during the flow through the blank, but it is evident that the systematic
characterization test. portion of the uncertainty DQs/Qs is reduced as DQsp/Qt is
A8.5.1.2 If a blank of known thermal conductance is used to made small. Also, as Qcp is made small, the term DQcp/Qt is
calibrate the surround panel then: presumably also made less significant. Thus, the fractional
DQsp / Q 5~DQt 2 DQcp! / Q (A8.4) systematic uncertainty possible in the determination of Qs is
reduced by increasing either the area of the specimen (if
where: feasible), or the total thermal resistance of the surround panel.
DQcp = the uncertainty in determination of heat flow
through the characterization panel.
A8.5.1.3 Little can be said in general about the magnitudes
of the fractional uncertainties DQt/Q and DQsp/Q since these
depend on the quality and management of the particular hot

A9. DETERMINATION OF THE ENVIRONMENTAL TEMPERATURE IN A HOT BOX ENVIRONMENT

A9.1 General Considerations: of specimen and the surfaces in view of the specimen must be determined
by a more detailed analysis. See ISO 12567 for a more detailed analysis
A9.1.1 Background—The heat transfer environment seen of how to determine the radiation exchange between the specimen and the
by the specimen surfaces within a hot box apparatus are surround panel edge.
generally controlled by two types of heat transfer, convection
and radiation. The air conduction heat transfer is small and can A9.2 Heat Flow Components:
be neglected when compared to radiation and convective heat
transfer at the surface boundary. For purposes of this method, A9.2.1 Convective Heat Transfer—The convective heat
it is lumped with the convective component. The measured transfer is an exchange of heat from the surface to the
surface and air temperatures control the convective heat surrounding air by convective means. This heat flow is a
transfer. The radiation heat transfer is a function of the function of the system geometry, air flow properties, and air
measured surface temperatures of the surrounding enclosure, velocity, and is generally expressed by Eq A9.1:
including the baffle. Although it is desirable to have the Qconv 5hconv · As · Dts2a (A9.1)
surrounding surface temperatures as close to the air tempera-
ture as possible, that condition does not always exist, especially where:
if the specimen contains highly conductive components (that Qconv = heat flow by convection from the specimen sur-
is, steel studs, single glazed window, etc.). Therefore, it is more face, W,
accurate to describe the heat flow (or thermal transmittance) hconv = convective heat flow coefficient, W/m2 K,
through a specimen in terms of the environmental temperature As = specimen projected surface area., m2, and
Dts-a = the temperature difference between the specimen
difference as opposed to the air temperature difference alone.
area weighted average surface temperature (ts),
A9.1.2 Need—Calculation of the environmental tempera- and the surrounding average air temperature (ta);
ture for a hot box test is important where the average surface where, for the metering side: Dts-a = (th − t1), and
temperature is not easily defined. Generally, this is due to the for the climatic side: Dts-a = (t2 − tc).
presence of thermal bridging within the specimen. The defini-
tion of environmental temperature permits the surface coeffi- A9.2.2 Radiation Heat Transfer—The radiation heat trans-
cient to be defined as a function of one temperature variable, fer is an exchange of heat between the specimen surface and
Tenv, which replaces both the air and equivalent radiative the surrounding enclosure by radiation. This heat flow, is also
surface temperatures. The determination of environmental a function of the system geometry, and the surrounding surface
temperature is required to enable the use of this Test Method’s temperatures, and is generally expressed by Eq A9.2:
(C 1363) test results in calculating the heat transfer parameters Qrad 5 hrad · As · Dts2b (A9.2)
required by the equivalent ISO 8990 Hot Box Procedure.
A9.1.3 Introduction—The following equations are ex- where:
pressed in general terms. These equations are used for both the Qrad = heat flow by radiation from the specimen surface to
climatic side and the metering side of the specimen surfaces in that of the surrounding enclosure, W,
Dts-b = the temperature difference between the average test
the hot box by inserting the appropriate surface and environ-
specimen surface (ts) and the surrounding enclo-
mental parameters.
sure surfaces area weighted average temperature
NOTE A9.1—Eq A9.2 assumes that the view factor between the baffle (tb), K; where, for the metering side: Dts-b = (tb1 –
surfaces and the specimen is unity, and therefore the specimen surfaces are t1), and for the climatic side: Dts-b = (tb2 – t2), and
assumed to only “view” the baffle and other surfaces in the chamber on hrad = radiation heat transfer coefficient for the surface as
which surface temperatures are measured. If the specimen views a defined in Eq A9.3 if temperatures are in °C, W/m2
relatively large areas of the surround panel or itself (that is, the metering
side of a curb mounded skylight), the view factor, the radiation exchange
K.

35
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
hrad 5 eeff · s · @~273.15 1 ts!2 1 ~273.15 1 tb!2# · @~273.15 1 ts! Qtotal 5 Qconv 1 Qrad (A9.6)
1 ~273.15 1 tb!# (A9.3)
or, in Eq A9.4, if temperatures are in absolute K, A9.4 Environmental Temperature:

hrad 5 eeff · s · @ts2 1 tb2# · @ts 1 tb# (A9.4) A9.4.1 Calculation of Effective Environmental
Temperature—Eq A9.7 defines the effective environmental
and: temperature as that temperature that yields the same net heat
s = Stefan-Boltzmann constant = 5.673 3 10-08 W/m2 exchange in the simple convective mode as the combination of
K4, convective and radiation exchange seen in the test situation.
eeff = effective emittance of the specimen surface and
surrounding enclosure surface as defined in Eq A9.5, Qtotal 5 ~hrad 1 hconv! · As · Dts2env (A9.7)
1 where:
eeff 5 (A9.5)
~1/es 1 1/eb 2 1! Dts-env = temperature difference between the average test
eb = area weighted emittance of the surrounding enclo- specimen surface (ts) and the effective environ-
sure and baffle surfaces as seen by the specimen mental temperature (tenv); where, for the meter-
surface. For the metering side, eb = eb1, and for the ing side, Dts-env = tenv1 − t1, and for the climatic
climatic side, eb = eb2, and side, Dts-env = t2 − tenv2.
es = area weighted emittance of the specimen surface, or
the metering side, es = e1, and for the climatic side, A9.4.1.1 By substituting Eq A9.1, Eq A9.2, and Eq A9.6
es = e2. into Eq A9.7:
Dts2env 5 @hconv · Dts2a 1 hrad · Dts2b# / ~hrad 1 hconv! (A9.8)
A9.3 Total Heat Flow:
A9.3.1 Total Heat Flow—The total heat exchange from the NOTE A9.2—Additional discussion of the environmental temperature is
specimen surface is then the sum of the two modes of heat flow found in ISO Standard 8990.
from the surfaces defined in Eq A9.6.

A10. RECOMMENDED PRACTICE FOR ESTIMATION OF THE TESTING SYSTEM TIME CONSTANT

A10.1 General Considerations: A10.1.2.1 Also, any transient effects such as residual mois-
A10.1.1 The time required to conduct a hot box test is ture change, latent heat effects, or the onset of convection
determined, in part, by the speed of response of the testing within specimen will increase the time for stabilization for a
apparatus and the specimen’s response to changes in its test.
environment. One measure of this response to change is the
time constant, t, of the system. As defined in Note 22, the time A10.2 Testing System Time Constant Evaluation—The hot
constant of the system is the time required for the system to box apparatus time response is controlled by either the appa-
respond to within 37 % (1/e) of its final value of response, ratus design or the assembled properties of the specimen. For
usually heat flow, after a step change in forcing condition, test purposes, if the apparatus time constant, tap, is greater than
usually temperature difference. As specified in 10.11, a mini- specimen time constant, ts, the test will be controlled by the
mum of five time constants of consecutive, uniform data shall value of tap. If however, tap < ts, then the specimen response
be collected to determine if steady state conditions exist. will be the controlling factor in determining whether the test is
Therefore, it is necessary that an accurate measure of the complete. The apparatus time constant, tap, is determined by
effective time constant, teff, of the operating hot box system be experimental measurement as described in A10.3, and the
determined. specimen time constant, ts, is calculated as specified in A10.4.
A10.1.2 The operation of the hot box apparatus is an heat Note, however, that the two time constants may not be
transfer problem. Therefore, it appears logical that the time completely distinct and independent.
controlling factors for the hot box test would include:
(1) The heating and cooling capacity for the apparatus; A10.3 Response of the Apparatus:
(2) The air circulation patterns and velocity; A10.3.1 The design of the apparatus shall include consid-
(3) The internal heat storage capacity of the test chamber eration of the speed of response of the test chambers to
equipment; changing test conditions and the thermal lag caused by the heat
(4) The thermal diffusivity and resistance of the materials capacity of the internal equipment. The speed of response of
used to construct the chambers; the apparatus, or time constant, tap, is fixed by the design and,
(5) The specimen geometry; for a properly designed system will be less than the specimen
(6) The specimen thermal diffusivity and resistance; and time constant. Since the test apparatus is generally complex
(7) The specimen heat storage capacity. compared to the specimen, and since it does not change with

36
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
the specimen, the apparatus time constant, tap, can be deter- A10.3.3.6 Continue monitoring the test data until steady
mined by experimental means. state is reached. For this determination use five consecutive 1-h
A10.3.2 Experimental Determination of the Apparatus Time time averages to establish steady state.
Constant: A10.3.3.7 Plot the time versus temperature and net sample
A10.3.2.1 The time constant of the apparatus, tap, can be heat flow rate (for the usual case of constant temperature
empirically determined by measuring the speed of response of control) for the period from shortly before the temperature
the hot box with a specimen installed. As discussed in A10.1.2, change to the second time the hot box reaches steady state.
for any experimental setup, the measured system time response (See the example, Fig. A10.1.)
is the sum of the time responses of the individual parts. Any A10.3.3.8 Determine the elapsed time from the temperature
attempt to measure, experimentally, the effective time constant, change, in which the 5-min averages of temperatures and heat
teff, will, in fact, be determining the combined response of the flow was 63.2 % of the final value.
apparatus constant, tap, and the specimen time constant, ts. A10.3.3.9 Determine the elapsed time from the temperature
Therefore, if the time constant of the specimen can be forced to change, in which the 5-min averages of temperatures and heat
be significantly less than the time constant of the apparatus, flow was 85.6 % of the final value.
then the apparatus time constant, tap, can be approximated A10.3.3.10 The maximum difference in times for A10.3.3.8
using the simple experiment outlined in A10.3.3. and A10.3.3.9 is equal to the time constant for the test system,
teff.
A10.3.2.2 Although it is impossible to create a specimen
that has zero specific heat capacity, a specimen can be NOTE A10.1—For most circumstances, the time constant is independent
developed that has a low thermal resistance and low heat of the magnitude of the temperature shift or the heat flow of the system.
capacity. By examination of Eq A10.1, the specimen sample The controlling factor for the time constant will be the heat capacity of the
air handling systems and thermal resistance of the thermal chamber walls
will have a lower time constant if the specific heat capacity and specimen. In thermal chambers that only have one mode of tempera-
(Ms · Cs) is kept low, since As and h are fixed by the apparatus ture control (that is, a metering chamber with electrical heaters, but no
design. Therefore, to establish a good estimate of the minimum active mechanism of cooling), the rate of temperature increase may occur
time constant for the apparatus, one shall use a homogeneous, faster than the rate of temperature decrease. In this circumstance, the rate
lightweight, low thermal resistance specimen. This specimen of heat input by the heaters is greater than the rate of heat flow that is lost
design shall produce the shortest test time constant for the through the metering chamber walls and specimen. When the air tempera-
testing system. ture in the metering chamber is increased, the metering chamber is
considered to be in active mode in that the temperature controllers are
A10.3.2.3 Therefore, the recommended practice is to mea- adding heat to metering chamber by activating the heaters. On the other
sure the apparatus response to a step change in temperature hand, when the air temperature in the metering chamber is decreased, the
using a low mass specimen, and then use those results to metering chamber is considered to be in passive mode in that the
determine the shortest time constant of the system. The time temperature controllers do not activate the heaters, and the metering
constant of the system would then have to be increased if the chamber loses heat through the metering chamber walls and specimen.
The measured time constant of such a hot box is different depending on
time constant of the specimen is determined to be greater than whether the temperature in the metering chamber is increased or decreased
the time constant of the apparatus. during the time constant test. Since the chiller and heaters are typically
A10.3.3 Procedure for Experimental Time Constant activated during a steady state test, the apparatus time constant shall be
Determination—The following experimental procedure is rec- determined while both the climate and metering chambers have their
ommended for determining the time constant for a hot box temperature control in active mode, where the heaters or the chiller system
are actively used to change and control the air temperatures. For this
apparatus.
reason, it is best to perform time constant tests where the metering
A10.3.3.1 Construct a specimen having the lowest R-value chamber air temperature is suddenly increased, not decreased.
and the lightest weight that can be tested within the practical
limits of the test apparatus. A10.3.4 An Example for a Typical Hot Box Apparatus—An
example of an actual time constant test is provided in Fig.
A10.3.3.2 Install and seal the specimen in the hot box, close
A10.1 and Table A10.1. The determination of time constant of
the system, initiate test conditioning. For the initial test
the climate side baffle temperature is graphically shown. Table
conditions, set the air temperatures in the climate and metering
A10.1 presents the results of analysis for all the critical
chambers 5 to 10°C below the typical set point (see Note
parameters. In this example, the climate side air temperature
A10.1).
was suddenly increased 22.2°C (from −12.2 to 10°C), and the
A10.3.3.3 Set up the data acquisition system to record all metering side air temperature was simultaneously increased
test parameters at a minimum of 5 min intervals and begin 11.1°C (from 26.6 to 37.7°C). The specimen used for this
recording data. experiment was a surround panel constructed of 127 mm thick
A10.3.3.4 Continue monitoring the test data until steady polystyrene foam faced on both sides with 3 mm high-density
state is reached. For this determination use five consecutive 1-h polystyrene sheet. The time constant for the chamber with this
time averages to establish steady state (refer to 10.11.2). specimen was finally considered to be 1 hour. However, the test
A10.3.3.5 Once steady state conditions have been achieved, operators have chosen to use 6 time constants of steady state
quickly change the test conditions in both the climate and data since the time constant for power was 1 h and 10 min (5t
metering chamber so that the air temperatures increase and = 5 h and 50 min ~ 6 h).
stabilize at higher values. Record the time at which this change
occurs, and continue to monitor test data. A10.4 Calculation of Specimen Time Constants:

37
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

FIG. A10.1 Example Data Time Constant Determination

TABLE A10.1 Time Constant Example—133 mm Surround Panel

Description Symbol 63.2 % Time 85.6 % Time Difference
Cold Air Temperature Tc [Link] [Link] [Link]
Cold Baffle Temperature Tb2 [Link] [Link] [Link]
Cold Surround Panel Temperature Tsp2 [Link] [Link] [Link]
Cold Surround Panel Guard Temperature [Link] [Link] [Link]
Warm Air Temperature Th [Link] [Link] [Link]
Warm Baffle Temperature Tb1 [Link] [Link] [Link]
Warm Surround Panel Temperature Tsp1 [Link] [Link] [Link]
Guard Surround Panel Temperature [Link] [Link] [Link]
Guard Air Temperature [Link] [Link] [Link]
Total Heat Flow into Metering Chamber Qs [Link] [Link] [Link]

A10.4.1 Since the value of the overall time constant, teff, HCs = equivalent composite specific heat, W h/kg K; equal
determined in the previous section is for the low thermal to the sum, for the test specimen, of the product of
resistance, low heat capacity specimen, it is necessary to the individual component’s heat capacity and
evaluate the magnitude of the time constants for other speci- weight, divided by the total weight of the specimen,
men constructions. Of course, one could repeat the experimen- As = heat transfer area, m2,
tal procedure of A10.3.3 for every specimen. This approach, h8 = the composite surface coefficient which includes an
however, is expensive. estimate of the internal heat flow resistance, W/m2
A10.4.2 One alternative is to calculate the time constant of K,
the specimen based upon the simple formula shown in Eq and:
A10.1. Fortunately, the time constant of a homogeneous
1/h8 5 ~1/hs! 1 ~R! (A10.2)
system, such as a simple specimen, can be approximated by a
first order equation, Eq A10.1: where:
Ms · HCs hs = the surface coefficient, W/m2 K, and
ts 5 h8 · A
s
(A10.1) R = the estimated specimen resistance, m2 K/W.
A10.4.3 This procedure still may be too complex for a
where: typical building construction that has many structural members
ts = specimen effective time constant, h, with significantly different heat flow rates. A further simplifi-
Ms = mass of the composite specimen, kg,
cation for our purpose is to estimate the time constant for each

38
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
of the simple heat flow paths and then combine them into an is 12.5 min, use 10 min. Remember this estimate is a guide for
“averaged” time constant for the complex structure. Review of testing and an exact determination is not required.
the ASHRAE Fundamentals volume and other resource books
A10.6 Alternative Methods—Often a laboratory tests only
on transient heat transfer, shows that the common method for
one type of specimen. In these cases, a simplified method of
combining the heat transfer parameters for a complex structure
determining the system time constant can be utilized. The
is to add the system path effects together using a parallel path
following paragraphs list two possible alternate methods.
technique. Applying this principle to the calculation of the time
constant yields the following: A10.6.1 One alternate approach utilizes a high thermal
resistance, high heat capacity system to determine the system
As / ts 5 A1 / ts1 1 A2 / ts2 1 ... 1 Ai / tsi (A10.3)
time constant. By a line of analysis similar to that illustrated
where: above, a well insulated concrete wall, for example, would yield
As = overall specimen area, m2, a very long specimen time constant. This time constant would
Ai = component heat path area, m2, significantly exceed the time constant of the apparatus. There-
ts = specimen composite time constant, h, and fore, this alternate method is to measure the time constant of
tsi = specimen path component time constant, h. the apparatus with the highest-mass specimen installed, and
use that time constant for all specimens that are less massive.
A10.5 Overall Test Time Constant: While this would eliminate the need to calculate the time
A10.5.1 The effective overall time constant is used to fix the constant of massive systems, it also would increase the time of
time periods required for data acquisition and determination of testing required for less massive specimens.
final system stability. Above, we have established estimates for A10.6.2 A second alternative approach has been used for
the apparatus time constant, tap, and the composite specimen fenestration testing. Fenestration test specimens are typically
time constant, ts, for our test setup. As outlined in A10.2, the mounted in homogeneous surround panels, which have an
remaining step is to choose the effective overall time constant aperture cut in them for installation of window products. Since
that controls our process. This choice is made as follows: the calculation of the time constant of most fenestration
products would be time consuming, if not impractical, the
A10.5.1.1 If ts >> tap, then use teff = ts, or
results from measuring the time constant of the thickest
A10.5.1.2 If tap >> ts, then use teff = tap, or continuous surround panel is often used as the time constant of
A10.5.1.3 If tap >> ts, then use the larger of tap or ts. most fenestration specimens. Using this approach, the time
A10.5.2 To simplify the calculations and data logging, constant of the fenestration specimen only needs to be calcu-
round the scan period time down to the nearest simple fraction lated if the thermal resistance is higher than the equivalent area
of 1 h for the test. For example, if the time constant is of surround panel (that was used to measure the time constant),
determined to be 33.5 min, use 30 min; or, if the time constant or if the fenestration specimen is excessively massive.

A11. DESIGN AND CONSTRUCTION OF THE HOT BOX CHARACTERIZATION AND SURROUND PANELS

A11.1 The procedures outlined in Annex A4-Annex A6 specific recommendations, which follow.
specify the steps required to quantify the relationships for
metering box wall loss and flanking loss. For the experimental A11.3 Construction and Instrumentation of the Panels:
analysis of these parameters, a characterization panel that fits A11.3.1 The panels shall be constructed from a uniform
the metering box opening is required. The surround panels thickness of a homogeneous and stable material of low thermal
required for measurement of specimens smaller than the conductivity. Suitable materials are high-density glass fiber or
opening of the metering chamber are identical in construction polystyrene boards laminated together as necessary. The as-
to the characterization panels. The exception is that the sembled panel shall be non-hygroscopic to minimize changes
characterization panel is continuous and the surround panel has in its thermal resistance with ambient humidity conditions.
a hole, at its center, large enough to hold the specimen. Since A11.3.2 Surround panels have also been fabricated by
the construction, but not necessarily the thickness, is identical sandwiching layers of homogeneous insulation between layers
for both panels, this section presents instructions on the of rigid materials such as plywood or plastic. Such surround
fabrication and instrumentation of both characterization and panels, though non-homogeneous, are uniform in the direction
surround panels. For purposes of this discussion, the word perpendicular to the direction of heat flow and are character-
“panel” shall apply to both types. ized in the same manner as homogeneous panels. Surround
panels shall have adequate strength to support the weight of the
A11.2 The need to determine the panel heat flow, Q, specimens to be tested.
accurately requires that the panel be designed to act as a heat A11.3.3 If the panel is assembled from multiple pieces of
flux transducer with an transducer output proportional to the identical material, thickness and thermal conductivity, then the
temperature difference, Dt, which is in turn proportional to the joints between the pieces shall be sealed with tape or caulk that
total heat flow through it. This consideration is the basis for the is at the same emittance (6 0.1) as the panel surface to which

39
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
it is attached. Tape shall not be placed more than 50 mm (2.0 cases, to incorporate framing in the surround panel to support
in.) from the edge of the joint. If rigid insulation is used as the heavy specimens such as heavy-duty metal frame windows or
core material, there is an opportunity to use a “tongue and masonry sections. Framing members shall be kept away from
grove” or a lapped joints to help minimize the air infiltration the specimen aperture and away from the point of contact of
through the joint. the metering walls so as not to contribute excessively to lateral
heat transfer at these locations. Such non-uniform surround
NOTE A11.1—A recommended surround panel core material is ex-
panded polystyrene (bead board) having a density in excess of 20 kg/m3, panels shall be characterized after the hole is cut using
which has been aged unfaced in the laboratory for a minimum of 90 days. calibration blanks of the same thickness and thermal conduc-
Polyisocyanurate or other fluorocarbon-expanded cellular foam insula- tance as the insulated part of the surround panel. In those
tions are not recommended as their thermal conductivity has been shown specific situations where the surround panel is not homoge-
to significantly change over time. Suitable facing materials are approxi- neous, detailed drawings and description of the surround panel
mately 3 mm thick heat-resistant rigid ABS thermoplastic sheets with construction, along with the measured results shall be included
smooth or matte finish faces or similar thickness high-impact polystyrene
plastic sheets. The surround panel needs to have some horizontal and with the test report.
vertical saw cuts made in the cold side facing material to minimize the
effects of differential thermal expansion between the cold and hot side A11.4 Instrumentation of Characterization and Surround
faces. The thin cuts should be covered with similar emittance tape strips Panels:
to provide a smooth surface to the weather and room side air streams.
A11.4.1 The surface temperature sensors used to measure
A11.3.4 Surround Panels: the temperature difference across the panel shall be perma-
A11.3.4.1 Surround panels are required for testing speci- nently installed uniformly flush with or just under its surfaces.
mens smaller than the metering area. When thermocouples are used, they shall be connected; (1) as
A11.3.4.2 The surround panel aperture, in which the speci- a differential thermopile for determination of the surround
men is installed, shall fit the specimen snugly. Cracks, greater panel temperature difference, or, (2) as individual thermo-
than 3.2 mm width, shall be filled with insulation and caulked couples for exploring temperature distributions on the faces of
or taped at the surround panel surfaces to prevent air leakage. the panel. At a minimum density, there shall be five tempera-
It is desirable that the insulation used to fill cracks has the same ture sensors per square meter installed on each panel surface.
thermal conductivity and thickness as the surround panel The temperature sensors shall be placed in the center of equal
assembly. The edge of the opening in the surround panel shall sized areas, or their output shall be area weighted to determine
be covered with non-metallic tape to minimize surface damage the average temperature of the surround panel surface. As a
of the exposed core insulation. Surround panels used for minimum, there shall be eight temperature sensors on each face
characterization testing shall have the specimen aperture filled of the surround panel. Four located at positions bisecting the
with the same material, thickness, thermal conductivity and four lines from the corners of the specimen aperture to the
assembly as the adjacent surround panel during the character- corresponding corners of the metering area and an additional
ization tests. The joint between the perimeter surround panel four at positions bisecting the sides of the rectangle having the
and the panel filling the aperture shall be flush and sealed with first four thermocouples at its corners. A suitable temperature
tape or caulk as described above. sensor arrangement shall be chosen for non-uniform surround
A11.3.4.3 The thickness of the surround panel shall be at panels that provide representative average surface tempera-
least the maximum thickness of the specimen, and shall be in tures. This is particularly important when natural convection is
no circumstances less than 100 mm. Also, the maximum used and air temperatures and film coefficients vary over the
thickness of the surround panel shall be no more than 25 mm metering surface. If framing members are used, an area-
greater than the maximum thickness of the test specimen. That weighted average of temperatures measured over the members
is, for test specimen maximum thickness less than or equal to and away from them is necessary. The panel, which acts as a
100 mm, the surround panel thickness shall be 100 mm. For heat flow meter, shall be calibrated so that the heat flow is
test specimen maximum thickness greater than 100 mm, the known as a function of the average temperature difference (or
surround panel thickness should be equal to the specimen thermopile output voltage) across it or as indicated by the
thickness rounded to the next higher 25 mm. permanently installed thermocouples.
A11.3.4.4 The restriction of surround panel thickness is to A11.4.2 Surround panels being used as characterization
limit the flanking loss through the surround panel at the panels (that is, the specimen aperture is filled with a known
uncovered areas of its aperture. Other special instances, for specimen) shall have a uniform layout of temperature sensors
example, a building element designed to be set a few centime- across the surround panel surfaces and the surfaces of the
ters outward from the plane of the inner surface of a wall, material filling the specimen aperture. It is sometimes more
requires special characterization of the surround panel. In this difficult to uniformly instrument the surround panel when the
case, a panel of known thermal conductance shall be in the specimen aperture is filled with an actual specimen, which
same position at the juncture with the surround panel aperture often has its own instrumentation scheme (that is, as specified
as the window. in Practice E 1423). As a general practice, the hot side surface
A11.3.4.5 Unless specifically required for test specimen temperature sensors are place directly opposite the cold side
mounting purposes (very high mass test specimens), no ther- sensors. The array of surface temperature sensors are arranged
mal anomalies (that is, thermal bridges like wood or metal) to produce the area weighted average surface temperature for
shall exist in the surround panel. It may be necessary, in some each surface of interest. See 6.10 for details.

40
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
A11.4.3 To protect the panel and the permanently installed A11.5 Characterization of the Panel as a Heat flow Trans-
thermocouples, the surfaces must be impervious to air. A ducer:
permanent coating or thin facing on each face of the panel is A11.5.1 Characterization of any panel material, whether
desirable. However, the coating or facing shall be of low lateral used for characterization, surround panel, or as a transfer
conductance so that it does not contribute excessively to lateral standard for windows testing (see Test Method C 1199) shall
heat transfer at the juncture with the specimen or at the be made by means of thermal tests on a representative sample
boundary of the metering area. The emittance of the panel of the assembled panel, their individual components, or tests on
surfaces shall be uniform and unchanged after testing. In all the entire panel. For this reason, it is required that the thermal
cases, the emittance of the panel surfaces shall be high (e > resistance of a sample assembly of the characterization or
0.8). The adhesive, caulk or tape used to mount the temperature surround panel be measured in an apparatus conforming to Test
sensor instrumentation shall have the same emittance as the Methods C 177 or C 518 at a minimum of three temperatures
surrounding surface (e 6 0.1). over the range of conditions at which the panel will be used. An
A11.4.4 It is probable that many specimens to be tested are alternative is to measure the thermal resistance of a larger panel
inhomogeneous or non-uniform in construction for structural in a hot box apparatus and then subsequently reducing the
reasons, and in consequence that the local thermal conductance panel to the size required to fit the surround panel aperture.
differs considerably at different frontal areas of the element.
A11.5.2 The characterization tests should cover the range of
The variations are inherent, and the result of the test is an
mean temperatures at which the panel will be operated during
average conductance or transmittance value for the total
the testing. At any one surround panel mean temperature, there
construction, provided that the conductance variations at edges
should be little variation of Qsp/Dt with Dt, but Qsp/Dt may
do not seriously impair the validity of using the surround panel
vary slightly with mean temperature due to the change of
as an adequate heat flow meter. This matter varies with each
thermal conductivity to the surround panel material.
case and therefore must rest on the judgment and technical
experience of those conducting the test measurement. A useful NOTE A11.2—Additional uncertainty may arise due to the possible
guiding principle is that nothing shall be incorporated in, or influences of the specimen in causing two or three-dimensional heat flow
omitted from, a specimen being tested that would make it not at its boundary with the surround panel and thus affecting the surround
representative of the assembly that would be found in actual panel heat flow in regions adjacent to the element. Surround panel heat
flow, determined under a given set of conditions with a transfer standard
installation in service. For example, if a metal window ordi- in place, may change when the specimen is installed, even though the test
narily is installed with inset wood framing, the test specimen conditions remain unchanged. If the specimen is expected to have this
shall include just so much of the wood framing as is properly influence, an attempt shall be made to evaluate its impact on the desired
chargeable to it. accuracy of the test.

41
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05

APPENDIX

(Nonmandatory Information)

X1. AIR AND MOISTURE MASS TRANSFER

X1.1 General the metering chamber to the climatic chamber, the heat carried
X1.1.1 Heat transfer through an insulation or insulated with the air entering the metering chamber will directly add to
structure is significantly increased by air infiltration or mois- (or subtract from) the metered heat and a correction must be
ture migration into or through the specimen. Since such made which equals the product of the air mass flow rate, its
phenomena can occur in field applications, it is desirable to specific heat, and the temperature difference between the
duplicate the conditions in the laboratory hot box and to test for incoming air and that in the metering chamber. If the direction
heat transfer due to air and moisture transfer combined with is from the climatic chamber to the metering chamber, no
that due to the imposed temperature difference. In principle, correction is necessary since the heat balance for the climatic
such testing is possible and indeed some hot boxes have been chamber is not determined. In either case, the air shall be so
designed for these tests. Such tests are not included in the introduced that it is thoroughly mixed to achieve the chamber
scope of this method because of the limited experience with air temperature before impinging upon the specimen.
them and because of the uncertainties of relating the results to X1.2.4 Measurements of heat flow made while a pressure
the performance that occurs in field applications. While this differential is imposed can, in some respects, simulate the
method does not recommend such tests, the following guidance effect on thermal performance due to air infiltration caused by
is given for those researchers who might attempt such tests. wind impingement. It is difficult, however, to relate such data
to field conditions of actual wind impingement upon buildings
X1.2 Air Infiltration or specimens because of the variable effects due to size, shape,
X1.2.1 Provisions have been made in some hot box appa- and orientation and the interaction with surrounding surfaces.
ratus for the measurement of both heat transfer and air flow It must also be recognized that a wind will not necessarily
under simultaneous temperature and air pressure differentials impose a pressure differential across a wall equal to its velocity
imposed across the specimen. In such cases, the apparatus was pressure. Thus, it is only possible to conduct tests under
constructed to meet all requirements of Test Method E 1424 specified air pressure differentials and to report the results
with recommended capabilities, in either direction, of flow without direct relation to wind velocities. Surface thermal
rates up to 0.005 m3/s for each square meter of specimen area resistance, Rs, as a function of wind velocity may be found in
and pressure differentials to 125 Pa. Pressure taps were the literature (see, for example, (16). Such values, when used
installed at mid height of the metering chamber and at the same for the added outside surface resistance as directed in 11.3
height in the climatic chamber. along with the thermal resistance measured under the pressure
X1.2.1.1 Caution: Pressure differentials across the speci- differential and an appropriate inside surface resistance, can
men and across box walls shall be limited to values which will give an estimate of the overall thermal resistance, Ru, and
not cause physical damage to the apparatus. Adequate precau- transmittance, U, under wind impingement.
tions shall be taken to prevent excessive pressures and to
protect personnel against possible injury in case of accidental X1.3 Moisture Migration
failure. X1.3.1 Modifications to the hot box apparatus have been
X1.2.1.2 The air supply equipment shall maintain the dew attempted for the measurement of heat transfer due to the
point of air entering the hot side below that of the cold side combined effects of moisture migration and to the imposed
temperature in order to prevent condensation within or on temperature differential (and to an imposed pressure differen-
specimen. Air entering the cold chamber shall be dried suffi- tial, if desired). Moisture driven behavior is complicated to
ciently to prevent undue frosting of evaporator coils. measure. It seems reasonable to expect that strict steady-state
X1.2.2 The apparatus and specimen perimeter shall be thermal conditions will be established only if the specimen and
gasketed or otherwise sealed to limit leakage both to the the air on the hot side are completely dry or if a constant rate
environment and around the specimen. Checks using an of moisture is introduced on the hot side under conditions that
impervious specimen shall show negligible leakage for the it flows through the specimen at that same rate without change
metering chamber. A small leakage for the climatic chamber is in state.
allowable but shall be calibrated and corrections made if the X1.3.2 Non-steady state phenomena may also be of interest.
flow to or from the climatic chamber is being metered. If moisture is introduced on the hot side at an excessive rate
X1.2.3 Corrections to the test heat balance for the enthalpy and if flow to the cold side is prevented or restricted by vapor
of the infiltration air are necessary. The magnitude of the barriers or other impervious or semi-permeable layers, an
correction will depend upon the temperature of the incoming accumulation of moisture will occur, either by condensation or
air and the direction of its movement. If the direction is from by freezing, depending upon conditions. These effects are of

42
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
interest and have been studied in the calibrated hot box. Other the melting of ice in a specimen. In all these cases, changes
moisture effects are also of interest such as heat transfer during occur slowly enough that quasi-equilibrium is established for a
the drying of a moist specimen under the influence of a period sufficiently long enough to obtain the required thermal
temperature gradient or during the evaporation of moisture or test information.

REFERENCES

(1) Burch, D. M., Licitra, B. A., and Zarr, R. R., “A Dynamic Test Method Furnace Walls,” Transactions American Electromagnetic Society, Vol
for Determining Transfer Function Coefficients for a Wall Specimen 24, 1913, pp. 53-84.
Using a Calibrated Hot Box,” Insulation Materials, Testing and (16) Heilman, R. H., “Surface Heat Transmission,” Transactions—Society
Applications, ASTM STP 1030, 1990, pp. 345–361. of Mechanical Engineers, Vol 1, Part 1, FSP - 51 - 91, 1929, pp.
(2) Mumaw, J. R., “Calibrated Hot Box: An Effective Means for Measur- 289-301.
ing Thermal Conductance in Large Wall Sections,” Heat Transmission (17) Turner, W. D., Elrod, D. C., and Siman-Tov, I. I., “HEATING5—An
Measurements in Thermal Insulation, ASTM STP 544, 1974, pp. IBM 360 Heat Conduction Program,” Oak Ridge National Labora-
193-211. tory, Computer Sciences Division, Documents ORNL/CSD/TM-15,
(3) Sabine, H. J., Lacher, M. B., Flynn, D. R., and Quindry, T. L., ORNL, Oak Ridge, TN., 1977.
Acoustical and Thermal Performance of Exterior Residential Walls, (18) Miller, R. G., “Hot Box Operating Techinques and Procedures: A
Doors, and Windows, NBS Building Science Series 77, U.S. Depart- Survey,” Journal of Testing and Evaluation, Amercian Society of
ment of Commerce, National Bureau of Standards, Washington DC, Testing and Materials, May 1987, pp. 153-166.
November 1975. (19) Bales, E., “ASTM/DOE Hot Box Round Robin,” ORNL/Sub/84-
(4) Miller, R. G., Perrine, E. L., and Linehan, P. W., “A Calibrated/ 97333/2, Oak Ridge National Laboratory, Oak Ridge, TN, November
Guarded Hot Box Test Facility,” Thermal Transmission Measurements 1988.
of Insulation, ASTM STP 660, 1978, pp. 329–341. (20) VanGeem, M. G., “Workshop for Hot Box Operators—Overview,”
(5) Mumaw, J. R., “Thermal Research Facility—A Large Calibrated Hot Journal of Testing and Evaluation, American Society of Testing and
Box for Horizontal Building Elements,” Thermal Insulation Perfor- Materials, May 1987, pp. 136-144.
mance, ASTM STP 718, 1980, pp. 195-207. (21) Broderick, T. B., “Design and Calibration of a Guard Added to an
(6) Achenbach, P. R., “Design of a Calibrated Hot Box for Measuring the Existing Hot Box,” Journal of Testing and Evaluation, American
Heat, Air, and Moisture Transfer of Composite Building Walls,” Society of Testing and Materials, May 1987, pp. 145-152.
Thermal Performance of the Exterior Envelopes of Buildings, (22) Zarr, R. R., Burch, D. M., Faison, T. K., Arnold, C. E., and
ASHRAE SP28, American Society of Heating, Refrigerating, and Air- O’Connell, M. E., “Calibration of the NBS Calibrated Hot Box,”
Conditioning Engineers, Inc., New York, NY, 1981, pp. 308–319. Journal of Testing and Evaluation, American Society of Testing and
Materials, May 1987, pp. 167-177.
(7) Fiorato, A. E., “Laboratory Tests of Thermal Performance of Exterior
Walls,” Thermal Performance of the Exterior Envelopes of Buildings, (23) VanGeem, M. G., “Measuring Thermal Performance of Wall Assem-
ASHRAE SP28, American Society of Heating, Refrigerating, and Air- blies Under Dynamic Temperature Conditions,” Journal of Testing
Conditioning Engineers, Inc., New York, NY, 1981, pp. 221-236. and Evaluation, American Society of Testing and Materials, May
1987, pp. 178-184.
(8) Perrine, E. L., Lineham, P. W., Howanski, J. W., and Shu, L. S., “The
(24) Goss, W. P., and Miller, R. G., “Hot Box Instrumentation, Calibration
Design and Construction of a Calibrated/Guarded Hot Box Facility,”
and Error Estimation—A Survey,” Insulation Materials: Testing and
Thermal Performance of the Exterior Envelopes of Buildings,
Applications, 2rd Volume, ASTM STP 1116, 1991, pp 326-354.
ASHRAE SP28, American Society of Heating, Refrigerating, and
Air-Conditioning Engineers, Inc., New York, NY, 1981, pp. 299-307. (25) Gerace, R. R., Dererian, G. D., Cirignano, P. C., Orlandi, R. D., and
Shu, L. S., “An Operational Procedure for Guarded Hot Box
(9) Rucker, J. L., and Mumaw, J. R., “Calibration Procedures and Results
Testing,” Journal of Testing and Evaluations, Vol 15, No. 3, May
for a Large Calibrated Hot Box,” Thermal Performance of the Exterior
1987, pp. 138-144.
Envelopes of Buildings, ASHRAE SP28, American Society of Heating,
(26) Orlandi, R. D., Howanski, J. W., Derderian, G. D., and Shu, L. S.,
Refrigerating, and Air-Conditioning Engineers, Inc., New York, NY,
“Development of a Testing Procedure for a Guarded Hot Box
1981, pp. 237-249.
Facility,” Symposium on Thermal Insulation Materials and Systems
(10) Palfey, A. S., “Thermal Performance of Low Emittance Building for Energy Conservation in the 80’s, ASTM STP 789, 1982, pp.
Sheathing,” Journal of Thermal Insulation, Vol 3, January 1980, pp. 205-214.
129-141.
(27) Rohsenow, W. M., and Choi, H. I., Heat, Mass and Momentum
(11) Goss, W. P., and Olpak, A., “Design and Calibration of a Rotatable Transfers, Prentice-Hall, Inc., Englewood Cliffs, NJ, 1961, p. 112.
Thermal Test Facility,” Thermal Insulations, Materials, and Systems (28) Wise, D. J., Sixth Annual Inter-laboratory Round Robin Evaluation of
for Energy Conservation in the 80’s, ASTM STP 789, 1983, pp. NFRC-Accredited Testing Laboratories, 2000 NFRC Test Laboratory
215-233. Round Robin, National Fenestration Rating Council, Silver Spring,
(12) Lavine, A. G., Rucker, J. L., and Wilkes, K. E., “Flanking Loss MD, 2000.
Calibration for a Calibrated Hot Box,” Thermal Insulations, Materi- (29) Kosny, J., and Childs P., “Accuracy of Hot Box Testing of Steel Stud
als, and Systems for Energy Consevation in the 80’s, ASTM STP 789, Walls,” Insulation Materials: Testing and Applications: 4th Volume,
1983, pp. 234-247. ASTM STP 1426, 2002, pp. 147-158.
(13) Schenck, H., Theory of Engineering Experimentation, McGraw Hill, (30) Petrie, T. W., Kosny, J., Atchley, J. A., and Desjarlias, A. O., “Effect
New York, NY, Third Edition, 1979, p. 53. of Steel Faming in Attic/Ceiling Assemblies on Overall Thermal
(14) Powell, F. J., and Bales, E. L., “Design of Round-Robin Tests Using Resistance,” Insulation Materials: Testing and Applications: 4th
Guarded/Calibrated Hot Boxes—Guarded Hot Plates/Heat Flow Volume, ASTM STP 1426, 2002, pp. 159-175.
Meters,” Thermal Insulations, Materials, and Systems for Energy (31) Yuan, S., Russell, G. A., and Goss, W. P., “Uncertainty Analysis of a
Conservation in the 80’s, ASTM STP 789, 1983, pp. 248-264. Calibrated Hot Box,” Insulation Materials: Testing and Applications:
(15) Langmuir, I., Adams, E. Q., and Meikle, F. S., “Flow of Heat Through 4th Volume, ASTM STP 1426, 2002, pp. 189-202.

43
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.
 C 1363 – 05
(32) Wise, D. J., and Shah, B. V., “An Assessment of Interlaboratory cations: 4th Volume, ASTM STP 1426, 2002, pp. 221-238.
Repeatability in Fenestration Energy Rating: 2001 NFRC Inter- (34) Gatland II, S. P., Goss, W. P., Baumgardner, R. L., Williams, R. G.,
laboratory Test Round Robin,” Insulation Materials: Testing and and Miller, R. G., “A Wall and Edge Guarded Hot Box for Thermal
Applications: 4th Volume, ASTM STP 1426, 2002, pp. 203-220. Transmittance Measurements,” Insulation Materials: Testing and
(33) Yuan, S., Gatland II, S. D., and Goss, W. P., “Calibration Procedure Applications, 3rd Volume, ASTM STP 1320, 1997, pp 46-60.
of a Calibrated Hot Box,” Insulation Materials: Testing and Appli-

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@[Link] (e-mail); or through the ASTM website
([Link]).

44
Copyright by ASTM Int'l (all rights reserved); Thu Apr 16 [Link] EDT 2009
Downloaded/printed by
Laurentian University pursuant to License Agreement. No further reproductions authorized.