Designation: E77 − 14

Standard Test Method for

Inspection and Verification of Thermometers1
This standard is issued under the fixed designation E77; 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 (´) indicates an editorial change since the last revision or reapproval.
This standard has been approved for use by agencies of the U.S. Department of Defense.

1. Scope E2251 Specification for Liquid-in-Glass ASTM Thermom-

1.1 This test method covers visual and dimensional inspec- eters with Low-Hazard Precision Liquids
tion and test for scale accuracy to be used in the verification of 3. Terminology
liquid-in-glass thermometers as specified in Specifications E1
3.1 Definitions:
and E2251. However, these procedures may be applied to other
3.1.1 The definitions given in Terminology E344 apply.
liquid-in-glass thermometers.2
Some that are considered essential to this standard are given
NOTE 1—The use of NIST SP250-232 is recommended. below.
1.2 Warning—Mercury has been designated by EPA and 3.1.2 calibration, n—of a thermometer or thermometric
many state agencies as a hazardous material that can cause system, the set of operations that establish, under specified
central nervous system, kidney and liver damage. Mercury, or conditions, the relationship between the values of a thermo-
its vapor, may be hazardous to health and corrosive to metric quantity indicated by a thermometer or thermometric
materials. Caution should be taken when handling mercury and system and the corresponding values of temperature realized
mercury containing products. See the applicable product Ma- by standards.
terial Safety Data Sheet (MSDS) for details and EPA’s website- 3.1.2.1 Discussion—(1) The result of a calibration permits
http://www.epa.gov/mercury/faq.htm - for additional informa- either the assignment of values of temperature to indicated
tion. Users should be aware that selling mercury and/or values of thermometric quantity or determination of correc-
mercury containing products into your state may be prohibited tions with respect to indications. (2) A calibration may also
by state law.- determine other metrological properties such as the effect of
1.3 This standard does not purport to address all of the influence quantities. (3) The result of a calibration may be
safety problems, if any, associated with its use. It is the communicated in a document such as a calibration certificate or
responsibility of the user of this standard to establish appro- a calibration report. (4) The term calibration has also been used
priate safety and health practices and determine the applica- to refer to the result of the operations, to representations of the
bility of regulatory limitations prior to use. result, and to the actual relationship between values of the
thermometric quantity and temperature.
2. Referenced Documents 3.1.3 complete-immersion thermometer, n— a liquid-in-
glass thermometer, not specified in ASTM documents, de-
2.1 ASTM Standards:3
signed to indicate temperature correctly when the entire
E1 Specification for ASTM Liquid-in-Glass Thermometers
thermometer is exposed to the temperature being measured.
E344 Terminology Relating to Thermometry and Hydrom-
etry 3.1.4 partial-immersion thermometer, n— a liquid-in-glass
thermometer designed to indicate temperature correctly when
the bulb and a specified part of the stem are exposed to the
1
This test method is under the jurisdiction of ASTM Committee E20 on
temperature being measured.
Temperature Measurement and is the direct responsibility of Subcommittee E20.05 3.1.5 total-immersion thermometer, n—a liquid-in-glass
on Liquid-in-Glass Thermometers and Hydrometers. thermometer designed to indicate temperature correctly when
Current edition approved May 1, 2014. Published September 2014. Originally
approved in 1949. Last previous edition approved in 2007 as E77 – 07. DOI: just that portion of the thermometer containing the liquid is
10.1520/E0077-14. exposed to the temperature being measured.
2
“Liquid-in-Glass Thermometer Calibration Service,” NIST Special Publication
250-23, 1988, Superintendent of Documents, U.S. Government Printing Office, 3.2 Definitions of Terms Specific to This Standard:
Washington, DC 20402-9325. 3.2.1 reference point, n—a temperature at which a thermom-
3
For referenced ASTM standards, visit the ASTM website, www.astm.org, or eter is checked for changes in the bulb volume.
contact ASTM Customer Service at [email protected]. For Annual Book of ASTM
Standards volume information, refer to the standard’s Document Summary page on 3.2.2 verification, n—the process of testing a thermometer
the ASTM website. for compliance with specifications.

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

1
 E77 − 14
3.2.3 verification temperatures, n—the specified tempera- 5.5 Secondary Standard Thermometers— Secondary stan-
tures at which thermometers are tested for compliance with dard thermometers are more suitable for routine work, and may
scale error limits. be of various types as described below. They are simpler to use
3.2.4 Other descriptions of terms relating to thermometers than a primary standard thermometer with its accessory
are included in Sections 3 and 17 of Specification E1. equipment, the latter being capable of an order of precision and
accuracy far in excess of that attainable with liquid-in-glass
4. Significance and Use thermometers. The choice of a secondary standard will be
4.1 The test method described in this standard will ensure governed by various factors. The following criteria should, in
that the thermometers listed in Specifications E1 and E2251 so far as possible, be satisfied: The standard should be a
will indicate temperatures within the maximum scale errors calibrated thermometer of equal or preferably higher sensitivity
listed, be compatible with the apparatus, and serve the purpose than the thermometer to be verified, and it should be capable of
for which they were designed. giving results of an equal or preferably higher order of
accuracy and also of an equal or preferably higher order of
4.2 Thermometers that do not pass the visual and dimen- reproducibility or precision. Scale corrections should always
sional inspection tests may give erroneously high or low be applied in the use of these standards. Secondary standards
temperature readings, or may not fit into existing equipment may be of the following types.
used in ASTM methods. For accurate temperature measure- 5.5.1 Direct-Reading Resistance Thermometers—Direct-
ments the scale readings of the thermometer should be verified reading resistance thermometers are available commercially,
as described in this test method. are very convenient to use, and have the advantage over the
primary type that temperature indications are given directly in
5. Apparatus
the instrument reading. They should be completely recalibrated
5.1 Graduated Metal Scales or Templates—Maximum and every 6 to 12 months, depending upon the temperatures of
minimum specified linear dimensions are measured with usage. Ice points should be taken every 3 months.
graduated metal sales and templates on which lines are ruled at 5.5.2 Liquid-in-Glass Thermometers—Liquid-in-glass
suitable distances from reference points corresponding to the thermometers, when used as secondary standards, may be
maximum and minimum values of the several specified dimen- classified into two groups, those intended for testing general
sions. purpose total or partial-immersion thermometers, and those for
5.2 Micrometers and Ring Gages—Specified diameters of testing special use partial-immersion thermometers.
ASTM thermometers are checked using micrometers, or more 5.5.2.1 Total-Immersion Thermometers—In the case of gen-
conveniently with ring gages consisting of metal plates in eral purpose total-immersion thermometers, the sensitivity of
which holes have been formed corresponding to the maximum the thermometers to be tested will govern the choice of
and minimum values of the several specified dimensions. The standard. For thermometers graduated in 1, 2, or 5° divisions,
thickness of such gages should approximate the diameters of a set of well-made thermometers will be adequate when
the holes to minimize errors resulting from the axis of the calibrated and used with applicable corrections. For fraction-
thermometer stem being other than normal to the plane of the ally graduated thermometers a calibrated set of the following
gage. When specified, diameters may also be checked with thermometers is recommended. Specifications for these ASTM
conventional snap gages having plane parallel working faces. Precision Thermometers appear in Specification E1.
ASTM
5.3 Comparators—Comparators are required for verifica- Ther-
tion of scale accuracy of liquid-in-glass thermometers. Suitable mometer Length,
Celsius
types are described in Appendix X1. Number Range Divisions mm

5.4 Primary Standard Thermometer—The primary standard 62C −38 to +2°C 0.1°C 380
thermometer in the range from −183 to 630 °C (−297 to 1166 63C −8 to +32°C 0.1°C 380
64C 25 to 55°C 0.1°C 380
°F) is the platinum-resistance thermometer. Temperatures are 65C 50 to 80°C 0.1°C 380
not measured directly with this instrument. Its electrical 66C 75 to 105°C 0.1°C 380
resistance is determined by comparison with a standard 67C 95 to 155°C 0.2°C 380
68C 145 to 205°C 0.2°C 380
resistor, using a potentiometer, a Kelvin-type double bridge, or 69C 195 to 305°C 0.5°C 380
a Wheatstone bridge, (preferably of the Mueller type) or an AC 70C 295 to 405°C 0.5°C 380
resistance bridge. Temperatures may then be calculated using ASTM
suitable resistance-temperature equations. In order that it shall Ther-
mometer Length,
be satisfactory for such use, the thermometer should meet the Number Range
Fahrenheit
Divisions mm
requirement that the ratio of resistances at the steam and ice
points shall be greater than 1.3925. More complete information 62F −36 to +35°F 0.2°F 380
63F 18 to 89°F 0.2°F 380
on the construction and use of primary standard thermometers 64F 77 to 131°F 0.2°F 380
may be obtained from NIST SP250-22.4 65F 122 to 176°F 0.2°F 380
66F 167 to 221°F 0.2°F 380
67F 203 to 311°F 0.5°F 380
4
“Platinum Resistance Thermometer Calibrations,” NIST Special Publication 68F 293 to 401°F 0.5°F 380
250-22, Superintendent of Documents, U.S. Government Printing Office, 69F 383 to 581°F 1.0°F 380
Washington, DC 20402-9325. 70F 563 to 761°F 1.0°F 380

2
 E77 − 14
The foregoing set is calibrated for total immersion. With the °C or 500 °F), because the heating of the bulb, which is
exception of the first two, each thermometer is provided with necessary to drive the liquid into the expansion chamber, may
an auxiliary scale including 0 °C (32 °F), thus providing means overheat the glass and either break the bulb, because of the
for checking at a fixed point, which should be done each time pressure of the gas, or destroy the accuracy of the thermometer
the thermometer is used. The change in ice-point reading by expanding the bulb. Thermometers that have a contraction
should then be applied to all readings. It is only necessary to chamber below the lowest graduation are likely to develop
have a liquid-in-glass thermometer completely calibrated one separations either in the chamber or above it. It is frequently
time. Recalibration is performed as described in 6.3.8. possible to join such separations by cooling the thermometer so
5.5.2.2 Partial-Immersion Thermometers— General pur- that the separated portion as well as the main column both
pose partial-immersion thermometers, as commonly listed in stand in the chamber. Tapping the tube against the hand or the
manufacturers’ catalogs according to their own specifications, bulb on a soft spongy material, such as a rubber stopper,
are normally bought and sold without specification of the
usually will bring the liquid together. For more stubborn
temperatures of the emergent column for the various tempera-
separations it may be necessary to cool the bulb in dry ice to a
ture indications of the thermometers. In such cases, verification
point low enough to bring all of the liquid into the bulb itself.
is usually carried out for the emergent column temperatures
By softly tapping on a soft spongy material or against the hand
prevailing with the verification equipment being employed.
it usually is possible to bring the liquid together in the bulb.
5.5.2.3 Special Use Partial-Immersion Thermometers—
Special use partial-immersion thermometers, such as those The bulb should be allowed to warm up slowly. The liquid
covered in Specification E1, have specified emergent mercury should emerge into the bore with no separation.
columns or stem temperatures. These thermometers can be 6.1.1.2 In organic-liquid-filled thermometers distillation
used as standards to calibrate other thermometers similar in all may occur, with subsequent condensation of the colorless
details of construction above the immersion point, but may parent liquid in the upper part of the thermometer. Such
differ below the immersion point to the extent of including an thermometers should always be inspected for these separations,
auxiliary ice point scale. which can be repaired by the procedures described above. If
5.6 Engraving Date on ASTM Thermometers—If a ther- the thermometer has an expansion chamber that is observed to
mometer’s specification was changed, the year that it was be filled with liquid, the column can be reunited by very careful
changed is engraved on the back of the thermometer after the heating of the chamber to drive the liquid into the bore where
ASTM designation. For example, “12C-98.” it can be rejoined to the main body as described above.
6.1.1.3 Organic liquids as used in thermometers, in contrast
6. Procedure to mercury, wet the glass. Sufficient time should always be
6.1 Visual Inspection: allowed for drainage to occur, particularly when using or
6.1.1 Gas Bubbles and Separations—Gas bubbles are read- verifying such thermometers below 0 °C (32 °F). It is fre-
ily detected and are more likely to occur in shipment than quently a good practice to immerse only the bulb of the
during service. No method has been discovered that will thermometer. This keeps the viscosity of the liquid in the
entirely prevent such displacement of the gas. If bubbles are capillary low and aids in hastening drainage.
observed in the bulb, they can generally be removed by cooling 6.1.2 Globules of Liquid—Globules of liquid in the stem,
the bulb with dry ice or other convenient coolant until all the which result from mechanical separation, can normally be
liquid is drawn into the bulb. Gentle tapping of the thermom- rejoined by heating the bulb until the liquid column merges
eter while held upright will cause the bubbles to rise to the with the globules. If such globules appear to unite and then
surface. It is very important that, if the bulb is cooled in this reappear on cooling the bulb, they are indicative of oxidation
process below the freezing point of the liquid, care should be of the mercury or the presence of obstructions in the bore, and
exercised to warm the stem sufficiently during the melting should result in rejection of the thermometer.
process so that no solidification occurs in the stem; otherwise 6.1.3 Foreign Matter—Foreign matter in the bore can some-
the bulb may burst or the capillary may split internally because times be detected with the unaided eye, but it is generally
of the expansion forces generated in the bulb. convenient to use a magnifying glass of low power for this
6.1.1.1 If a mercury separation is observed in the stem,
examination. A magnifier of 10× is recommended for visual
several different ways are suggested for joining the columns,
examination. The most common types of foreign matter which
depending on the construction of the thermometer and the type
should be cause for rejection are glass chips, particles of dirt or
of separation. If a small portion of the liquid has separated at
lint, oxide of mercury (either red, yellow, or black), products of
the top of the column and the thermometer is provided with an
expansion chamber, the liquid usually can be joined by glass weathering commonly called white deposit, and stones or
carefully and slowly heating the bulb until the separated iron spots traceable to faulty glass fabrication.
portion is driven into the expansion chamber. Never heat the 6.1.3.1 Where a specific gas is specified to be used as filling
bulb in an open flame. When the column itself follows into the above the liquid, any other gas present may be treated as
chamber, the separated portion usually will join onto the main foreign matter. The most common example is the use of air
column. A slight tapping of the thermometer against the palm instead of nitrogen in mercury-in-glass thermometers, which
of the hand will facilitate this joining. This method should not may have been introduced by accident or in violation of the
be employed for high-temperature thermometers (above 260 specifications. The presence of air can readily be detected by

3
 E77 − 14
visual inspection under slight magnification (2× to 5×). The 6.3.3 Depth of Immersion for Partial-Immersion Thermom-
mercury will oxidize and will show as dull, moist, or will show eters —Thermometers of this type shall be verified by one of
a red oxide. the methods described in 6.3.7.2, 6.3.7.3, and 6.3.7.4 using the
6.1.4 Glass Faults—Glass faults may be of various types. appropriate comparators. If the thermometer is of the general-
Any stones or striae that distort the bore or its appearance purpose type, it should be verified by immersion to the
should be cause for rejection. Strains in the glass as observed specified depth and the readings compared with those of the
with a polarized light strain gage near enlargements in the stem secondary standard.
or bore, or at the top of the thermometer, are detrimental. If so, 6.3.4 Verification at the Ice Point:
severe fire cracks may later occur. Strains near the bulb are 6.3.4.1 Select clear pieces of ice or ice made from distilled
indicative of incomplete glass stabilization and are particularly or pure water. Discard any cloudy or unsound portions. Rinse
objectionable in thermometers for use above 150 °C (302 °F). the ice with distilled water and shave or crush into small
6.2 Dimensional Inspection: pieces, avoiding direct contact with the hands or any chemi-
6.2.1 Determine compliance with lineal dimensional speci- cally unclean objects. Fill the Dewar vessel with the crushed
fications by comparison of the thermometer with the appropri- ice and add sufficient distilled, preferably precooled, water to
ate scale or template described in 5.1. In the case of diameter form a slush, but not enough to float the ice. As the ice melts
measurements, the bulb or other portions of the thermometer it will be necessary to drain off some of the water and add more
being tested should not enter the smaller of the holes in the ring crushed ice. Insert the thermometer, packing the ice gently
gage (see 5.2), or fail to enter the larger of the holes. In the case about the stem, to a depth approximately one scale division
of ASTM thermometers, the diameter limits apply over the below the 0 °C (32 °F) gradation.
entire length of the section of the thermometer being tested. 6.3.4.2 After at least 3 min have elapsed, tap the stem gently
6.2.2 Inspect the thermometers for uniformity of graduation and observe the reading. Successive readings taken at least 1
spacing. Normally, the effect of taper in the capillary or min apart should agree within one tenth of a division. It may be
nonlinearity in the expansion characteristics of organic liquids necessary to repack the ice around the thermometer because of
is manifested by a gradual change in the spacing of the melting.
graduations. Discontinuities in the spacing are evidence of
6.3.5 Verification at the Steam Point:
faulty graduating. Inspection for uniformity of spacing can
generally be accomplished by eye, but for greater accuracy a 6.3.5.1 If the range of the thermometer is such that the ice
pair of dividers may be used to advantage. Adjust the dividers point is not included, the steam point at 100 °C (212 °F) can be
to extend over 10 or 20 graduations, and successive intervals used. For a description of the equipment and techniques best
can be intercompared rapidly. Where marked changes in suited to the purpose, reference may be made to NIST
spacing are noted, it is well to include in the verification SP250-23.2
several temperatures in that region. 6.3.6 Verification of Total-Immersion and General Use
Partial-Immersion Thermometers at Temperatures Other Than
6.3 Verification of Scale Accuracy:
Fixed Points:
6.3.1 Reading Thermometers to Avoid Parallax—The error
due to parallax may be eliminated by taking care that the 6.3.6.1 Determine the ice point of the primary or secondary
reflection of the scale can be seen in the mercury thread, and by standard and the thermometer to be tested as described in 6.3.4.
adjusting the line of sight so that the graduation of the scale If the secondary standard is of the liquid-in-glass type, it
nearest the meniscus exactly hides its own image; the line of should be held at room temperature for at least 72 h before this
sight will then be normal to the stem at that point. In reading determination unless the ice point was originally determined
thermometers, account must be taken of the fact that the lines immediately after heating to a specified temperature. The
are of appreciable width. The best practice is to consider the thermometer to be tested should be treated in similar manner.
position of the lines as defined by their middle parts. 6.3.6.2 Insert the thermometer to be tested and the standard
6.3.2 Depth of Immersion for Total-Immersion Thermom- in the thermometer holder, and adjust the temperature of the
eters —Although by definition total-immersion thermometers comparator to a value approximately 5 °C (10 °F) below the
should have the bulb and stem containing the mercury im- verification temperature. It may be advantageous to use two
mersed both in use and in verification, it is frequently incon- secondary standard liquid-in-glass thermometers, since obser-
venient to do so. If any portion of the stem containing mercury vational errors of the standard may then be readily detected.
is exposed, the emergent mercury column correction should be Apply sufficient heat to raise the temperature at a slow uniform
determined. No correction need be applied if it is found to be rate. At the verification point the rate should not exceed one
less than one fifth of the verification error; otherwise, the scale division in 3 to 10 min. Fulfillment of this requirement
proper correction should be applied. In some instances total- will ensure that any exposed portion of the stem will have
immersion thermometers may be used under conditions of attained thermal equilibrium before readings are taken.
complete immersion. With thermometers such as the 6.3.6.3 When the proper rate of temperature rise has been
kinematic-viscosity thermometer 30F, a significant error will established, read the thermometers in the following order at
be introduced unless the thermometer is verified under the equal time intervals: standard, thermometer or thermometers to
same conditions as those in use. This is due to the effect of be tested, standard, thermometers to be tested in reverse order,
increased gas pressure above the column producing a distortion standard. The average of the first and third readings of the
of the bulb. standard should agree with the second. A comparison of

4
 E77 − 14
differences in successive readings will also indicate if the rate Paraffin Wax Melting Point: 14C, 14F
of rise has been uniform. High Softening Point: 16C, 16F
6.3.6.4 Calculate the average readings for all thermometers. Saybolt Viscosity: 17C, 17F, 18C, 18F, 19C, 19F, 20C, 20F,
Apply the appropriate corrections to the reading of the stan- 21C, 21F, 22C, 22F, 77F to 81F, 108F, 109F
dard. Calculate the corrections to be applied to the thermom- Engler Viscosity: 23C, 24C, 25C
eters under test. Stability Test of Soluble Nitrocellulose: 26C
6.3.6.5 For verification purposes, one series of readings is Turpentine Distillation: 27C
usually sufficient. If the thermometer is being calibrated, check Aniline Point: 34C, 34F, 35C, 35F
determinations should be made. The results of each of at least Solvents Distillation: 38C, 39C, 40C, 41C, 42C, 102C,
three series should agree with the mean of the series within 103C, 104C, 105C, 106C, 107C
plus or minus one tenth of a division.
Stormer Viscosity: 49C
6.3.6.6 Unless otherwise specified, tests should be made not
Gas Calorimeter Inlet: 50F
less than 40 nor more than 100 divisions apart throughout the
Gas Calorimeter Outlet: 51F
range of the thermometer.
Congealing Point: 54F
6.3.6.7 Comparisons may be made at constant temperature
in preference to the use of slowly rising temperatures. Such a Petrolatum Melting Point: 61C, 61F
procedure is satisfactory if the bath temperature does not vary Antifreeze Freezing Point: 76F
more than the precision of reading and if the thermometer is Solidification Point: 91C to 96C, 100C, 101C
jarred, as by tapping, before taking a reading in order to Tar Acids Distillation: 111C
overcome any sticking of the mercury to the glass. Such Brookfield 122C, 123C, 124C, 125C
tapping is particularly important with thermometers having a The ice point was omitted from the above listed thermom-
capillary diameter of the order of 0.1 mm or less and may be eters as a matter of necessity. In order to supply an ice point,
employed to advantage in the rising temperature method of test when not included in the range of the thermometer, it is
as well. Sufficient time should elapse before taking readings to necessary to blow an enlargement or contraction chamber
ensure that all portions of the thermometer have attained below the first graduation to take care of the volume of
thermal equilibrium. This is particularly important at test mercury between the ice point and this first graduation. This
temperatures remote from room temperatures. contraction chamber may introduce an error in the readings
6.3.7 Verification of Partial-Immersion Thermometers With unless it is entirely immersed in or is at the same temperature
Specified Stem Temperatures: as the medium the temperature of which is to be measured.
6.3.7.1 Partial-immersion thermometers having specified This immersion requirement may be difficult if not impossible
stem temperatures, such as those in Specification E1, can be to attain in the apparatus in which the thermometers are used.
verified in one of three ways. Such limitations of space prevented the inclusion of ice points
6.3.7.2 One method involves comparison of the thermom- in the above group of thermometers. However, by increasing
eters at total immersion with total-immersion standards. The the lengths of the thermometers, ice-point scales can be
number of degrees of scale, which will be in the emergent incorporated in the thermometers to be used as standards. For
column when in actual use, shall then be measured. From these those in the group that are partial immersion, the dimensions
data the corrections under the specified emergent column above the immersion point should be as prescribed in the
temperatures may then be calculated (7.2). In the case of specifications for the corresponding ASTM thermometer. For
organic liquid-filled thermometers such as ASTM Low Cloud the total-immersion thermometers this requirement may be
and Pour, 6 °C and 6 °F, the coefficient of expansion of the waived.
liquid should be obtained by experiment or from the manufac- 6.3.7.4 A third method uses total-immersion standards,
turer in order to perform these computations. This method has measuring stem temperature of the emergent liquid column in
the advantage that the standard may be selected to have greater the partial-immersion thermometer with a Faden thermometer
sensitivity than the thermometer being tested, thus increasing or small auxiliary thermometers. With this data, stem tempera-
the accuracy of measurement. ture adjustments can be made. See NIST SP250-232 for
6.3.7.3 A second method, which is the one best suited to additional details.
large-quantity testing, involves comparison of the thermom- 6.3.8 Recalibrate by Checking Changes in Bulb Volume:
eters with standards similar in all details of construction above 6.3.8.1 The verification procedure described above can be
the immersion point, but may differ below the immersion point used to obtain corrections for the calibration of thermometers.
to the extent of including an auxiliary ice-point scale. Such 6.3.8.2 Small changes in the volume of the thermometer
thermometers, when completely calibrated, as by the National bulb during use, especially significant for large bulb and
Institute of Standards and Technology, may then be employed relatively short range thermometers graduated in fractional
indefinitely for verification purposes if periodic ice-point degrees, may occur, even though the thermometer has been
checks are made. The particular field of application of this carefully annealed or aged, and will result in changes in the
second method is evidenced by the fact that many of the ASTM indications as well as the corrections. Usually the bulb con-
thermometers include the ice point in the specified range, but tracts slowly, and, as a result, the thermometer will read higher
the following fractionally graduated thermometers are not after it has been placed in service. Less frequently the bulb
provided with an ice-point scale: expands and the thermometer will read lower after use.

5
 E77 − 14
6.3.8.3 Two procedures are followed for using the ice point k = differential expansion coefficient of the liquid and the
to determine the change in bulb volume and to recalibrate a glass of which the thermometer is made; for Celsius
liquid-in-glass thermometer. One method requires that the mercury-in-glass thermometers the value of k is 0.00016,
thermometer remain at approximately room temperature for 72 for Fahrenheit mercury-in-glass thermometers it is
h before the ice point is taken. The other method stipulates that 0.00009, for Celsius organic liquid-in-glass thermom-
the ice point reading be taken immediately or within a eters it is 0.001, and for Fahrenheit organic liquid-in-
definitely stated short period of time after heating to the test glass thermometers it is 0.0006.
temperature. Changes in the corrections of the thermometer n = number of degrees emergent from the bath,
may be observed by determining the changes in the ice point by T = temperature of the bath, and
either method. t = average temperature of the liquid column of n degrees.
6.3.8.4 Record the ice-point reading and compare it with the
7.2.1 The use of more accurate values of k is not justified,
previous reading. If the ice-point reading is found to be higher
because of the uncertainty of the true average temperature of
or lower than the ice-point reading of the previous calibration,
the liquid thread not exposed to the temperature being mea-
readings at all other temperatures will be correspondingly
sured.
increased or decreased. If the method involving the taking of
the ice point immediately after heating to a specified tempera- 7.2.2 For emergent columns of partial-immersion
ture is used, such as with most kinematic viscosity thermometers, where the column temperature is other than
thermometers, a note should appear in the table of corrections specified in Table 2 of Specification E1, calculate the correc-
for the thermometer as follows: tion to the observed reading as follows:
Emergent stem correction 5 kn ~ t s 2 t o ! (2)
NOTE 2—The tabulated corrections apply for the condition of immer-
sion indicated, provided the ice-point reading taken after heating to __ for where:
not fewer than 3 min is __. If the ice-point reading (taken in not fewer than
2 min and not more than 1 h after removal of the thermometer from the k = same as in 7.2,
heated bath) is found to be higher (or lower) than stated, all other readings n = the number of scale degrees from the immersion mark
will be higher (or lower) to the same extent. to the top of the mercury column. The ungraduated
6.3.8.5 When the thermometers are first received, make a length between the immersion mark and the first gradu-
monthly check of the ice point: later these time intervals may ation on the scale shall be evaluated in terms of scale
be lengthened if the changes are found to be insignificant. degrees and be included,
6.3.8.6 High-quality liquid-in-glass thermometers require ts = temperature specified in Table 2 of Specification E1,
to = observed average temperature of the liquid column of n
only one complete calibration in their lifetime, and it is
degrees.
possible to avoid the usual requirement for complete recalibra-
tion of the instrument by the recalibration of a single previ- 7.3 The limit of precision to which the corrections at the
ously calibrated temperature. The need for recalibration of calibration temperatures shall be recorded, to which correction
properly manufactured liquid-in-lass thermometers is due to tables or charts shall be made, or to which any temperature
the gradual relaxation of residual mechanical strains in the measured by the thermometer shall be reported, is a measure of
glass that have a significant effect on the volume of the bulb. the sensitivity of the thermometer and is generally described by
The recalibration of a single point provides a reliable indication the phrase, corrections stated to.
of the effect of this change in volume and provides a means for
7.4 The limit of reliability of any temperature measurement
the accurate adjustment of the remainder of the scale.5
made with a calibrated thermometer after all corrections,
determined as carefully as possible, have been applied is
7. Treatment of Data
termed the uncertainty in degree of the thermometer.
7.1 From the corrections at the ice point and the other
verification temperatures, a table of corrections can be pre- 7.5 For compliance with specifications, the errors in the
pared for each thermometer. If the thermometer is to be used temperature indications of ASTM thermometers being verified
frequently at temperatures other than those of calibration, a shall not exceed the scale errors as expressed in Table 1 of
correction curve will be found more convenient than interpo- Specification E1. Unless otherwise expressed, it is understood
lation from a table. In such a curve the corrections are plotted that such thermometers are to be used without application of
against the temperature readings. corrections.
7.2 Where corrections are to be calculated for emergent 8. Precision and Bias
columns, the following equation shall be used:
8.1 No statement is made about either precision or bias of
Emergent stem correction 5 kn ~ T 2 t ! (1)
Test Method E77 for visual and dimensional inspection since
where: the results merely state whether there is conformance to the
criteria for success or failure specified in the procedure.
8.2 Precision of Verification of Scale Accuracy
5
Wise, Jacquelyn, “A Procedure for the Effective Recalibration of Liquid-in-
8.2.1 It is not practicable to specify the precision of the
Glass Thermometers,” NIST Special Publication 819, 1991. Available from Super-
intendent of Documents, U.S. Government Printing Office, Washington, DC verification of the scale accuracy because of the wide variety of
20402–9325. thermometers listed in Specification E1. Of the 129 different

6
 E77 − 14
kinds of liquid-in-glass thermometers, they are all either for glass thermometers using the above information as a guideline.
use at total immersion or partial immersion. A statistical analysis of data collected on the various types of
8.2.2 The verification of total-immersion thermometers will thermometers verified could be used to obtain a value for the
produce more repeatable results because the temperature of the random error.
stem containing the mercury is at the same temperature as the 8.3 Bias
bulb, when the thermometer is immersed correctly. The tem-
8.3.1 Like precision, it is very difficult to determine a bias
perature of the emergent stem of partial-immersion thermom-
statement for the verification of the scale accuracy because of
eters can vary depending on the temperature of the environ-
the systematic error from the choice of standard used and the
ment and this will affect the repeatability of the data obtained.
gradients in the comparators in each laboratory.
The method used for measuring the temperature of the emer-
8.3.2 If a standard platinum resistance thermometer is used
gent mercury column, when stem-temperature measurements
as the standard, it can be capable of reproducing temperatures
are made, has an effect on the repeatability of the data.
to approximately 0.001 to 0.003 °C. If total-immersion liquid-
Generally, total-immersion thermometers have an uncertainty
in-glass thermometers, such as the ASTM set of 62 °C (62 °F)
of approximately ten times better than partial-immersion ther-
through 70 °C (70 °F), are used as standards, the uncertainty at
mometers.
the calibration points will range from 0.03 °C (0.05 °F) to 0.5
8.2.3 Most thermometers contain mercury as the actuating
°C (1.0 °F). The uncertainty of partial-immersion thermom-
liquid. Some thermometers, especially those for use at tem-
eters will range from approximately 0.3 °C (0.5 °F) to 2.0 °C
peratures below −56 °C (−69 °F), use an organic liquid. Unlike
(3.0 °F). This uncertainty will be less if special use partial-
mercury, the organic liquid will wet the wall of the capillary.
immersion thermometers are used as standards where the
Adequate time must be given for complete drainage to occur.
readings are adjusted for a specified stem temperature. In all
This can take up to 1 h and can have an influence on the
cases the average of two standard liquid-in-glass thermometers
precision of the measurement. Thermometers containing or-
will give a more accurate value of the temperature of the
ganic fluids are less accurate than those containing mercury or
medium of the comparator.
mercury thallium.
8.3.3 The error due to gradients in the comparators must be
8.2.4 Thermometers are graduated in intervals of 0.01°,
determined in each laboratory. These gradient errors will
0.02°, 0.05°, 0.1°, 0.2°, 0.5°, 1.0°, 2.0°, and 5.0°. The ability to
usually increase where comparisons are made in comparators
read these thermometers will have an influence on the precision
at higher temperatures.
of the measurements. If the thermometers are read with a
telescope, thermometers graduated in intervals of 0.01° or 8.3.4 The precision and bias for the verification of scale
0.02° can be read to the nearest 0.001°; in 0.05° or 0.1° to the accuracy must be calculated for each laboratory. An example of
nearest 0.005°; in 0.2° to the nearest 0.01°; in 0.5° or 1.0° to how this can be done is shown in NIST SP250-23.2
the nearest 0.05°; in 2.0° to the nearest 0.1°; and in 5.0° to the
nearest 0.5°. 9. Keywords
8.2.5 Each laboratory can determine the precision of the 9.1 calibration; liquid-in-glass thermometers; mercury sepa-
verification of scale accuracy or the calibration of liquid-in- ration; organic liquids; verification

APPENDIX

(Nonmandatory Information)

X1. DISCUSSION OF APPARATUS FOR VERIFICATION OF LIQUID-IN-GLASS

THERMOMETERS

X1.1 Comparators, General Requirements—Comparators

for use in verification are of two types, fixed-point or variable.
Certain general requirements should be met by all such
comparators and their accessory equipment. All readings
should be observed, using an optical device suitable for
observing vertical motions. Focusing range should start from
not more than 20 cm, magnification should be about 5 to 10
diameters, and the field about 1.5 cm. The eyepiece should be
provided with 90° crosshairs and may include an erecting
prism. Vertical movement should total approximately 30 cm
and may be obtained with both rough and delicate adjustment.
A suitable instrument is illustrated in Fig. X1.1; component
parts are available commercially.
FIG. X1.1 Magnifier for Reading Thermometers

7
 E77 − 14
X1.1.1 Heat input should be capable of accurate continuous sive characteristics in the various temperature ranges in which
control to meet the requirement of a very slowly rising they are to be used.
temperature at the test point. For best results an ac power
supply is recommended in conjunction with variable trans- X1.4 Comparators for the Range from −160 to −75 °C
formers. Such transformers are available commercially in (−256 to −103 °F)—For comparison in this range,
various capacities. Other types of voltage regulators, such as which is below the sublimation point of solid carbon dioxide,
the so-called induction type, may be used. While generally liquid nitrogen is used as the cooling medium for an isolated
more expensive, they are capable of finer adjustment. bath filled with some low-boiling hydrocarbon such as isopen-
tane. The dimensions given below are satisfactory but are not
X1.1.2 In all test baths a properly located well or other mandatory. The following individual items are necessary.
suitable provision should be made for using a platinum
resistance thermometer as ultimate primary standard. X1.4.1 Aluminum Block, 28 mm in diameter by 150 mm in
length. Four holes 10 to 11 mm in diameter may be drilled in
X1.2 Comparators, Fixed-Point: the center of the block, with the holes overlapping slightly so
X1.2.1 The most common and also most useful of the that the final pattern resembles a cloverleaf. The holes shall be
fixed-point type comparators is the ice-point apparatus. Fig. deep enough so that the liquid index of the thermometers under
X1.2 illustrates a typical setup, consisting of a Dewar flask, a test may be completely immersed. This block rests on a
thermometer holder, a viewing telescope of 10× magnification, cushion of glass wool on the bottom of the test tube, the
and the necessary supports and siphon tube for withdrawing cushion being of convenient height.
excess water. A suitable ice-shaving machine is a desirable X1.4.2 Test Tube, 33 mm in diameter by 235 mm in length
adjunct. For readings to one tenth of a division, the viewing with a resistance coil (165 Ω) on the outside. This is an
telescope is necessary, but it can be omitted if lower precision ordinary test tube with a spiral of resistance wire wrapped
is desired. around the outside. The wire may be held in place by glass
X1.2.2 For the steam point see NIST SP250-23.2 studs or by cellulose tape. The resistance shall be such that the
wire will not become dangerously hot if the full voltage of a
Comparators, Variable-Temperature variable transformer is accidentally sent through. The test tube
is in turn supported in the evacuable jar by a cushion of glass
X1.3 For verification at other than fixed points, the variable- wool.
type comparators are used. Figs. X1.3-X1.7 illustrate various
types of such units for use in the range from −160 to +621 °C X1.4.3 Vacuum Pump.
(−256 to +1150 °F). All of these units basically consist of a X1.4.4 Evacuable Vacuum Jar, unsilvered, with attached
well-stirred, insulated liquid bath provided with suitable con- side arm containing a stopcock, 50 mm in inside diameter by
trols for maintaining the temperature either constant or uni- 300 mm in inside depth. An ordinary unsilvered Dewar flask is
formly increasing. As with the ice-point equipment, a viewing fitted with a side arm containing a tightly fitting stopcock.
telescope is necessary for precision of one tenth of a scale
X1.4.5 Variable Transformer, 0.3 kVA.
division. The liquids used in the comparators are chosen to
fulfill the following requirements: low viscosity, X1.4.6 Vacuum Jar, silvered, without side arm, 80 mm in
nonflammability, nontoxicity, and freedom from other offen- inside diameter by 350 mm in inside depth.
X1.4.7 Platinum Resistance Thermometer and accessories,
or suitable liquid-in-glass secondary standard.
X1.4.8 Cardboard Cover: (not shown) to fit over the top of
the test tube.
X1.4.9 Glass Wool.
X1.4.10 Liquid Nitrogen.
X1.4.11 Isopentane.
X1.4.11.1 In use the apparatus is assembled as shown in
Fig. X1.3. The evacuable flask is positioned in the larger
container by three cork wedges strung on a wire. It must extend
above the nitrogen container enough so that nitrogen may be
added without spilling into the isopentane. The aluminum
block is completely covered with isopentane and the space
between the two Dewar flasks filled with liquid nitrogen. A
loose fitting cardboard shield with proper holes is placed over
the top to reduce condensation and convection. When the
system has cooled to a few degrees above the desired
temperature, the inner flask is evacuated by the pump to
stabilize the temperature equilibrium in the test tube. (When
FIG. X1.2 Apparatus for Calibration at Ice Point testing in the range of −157 °C (−251 °F), this evacuation is not

8
 E77 − 14

FIG. X1.3 Comparator for Temperature Range from −160 to −75 °C (−256 to 103 °F)

necessary). At the desired test temperature, heat transfer is other for short immersion. In either case the bath shall consist
exactly balanced by means of the heating coil. It will be of a Dewar-type vacuum flask, the walls of which may or may
necessary to agitate the isopentane by raising and lowering the not be silvered as required. Normally the unsilvered is pre-
thermometer, but any vigorous stirring at low temperatures will ferred. Metal vacuum flasks may also be employed instead of
generate enough heat to make difficult the maintaining of the glass type. The bath tops are not available commercially
proper equilibrium. and must be made of some nonheat-conducting material such
X1.4.11.2 Certain safety precautions must be carefully ob- as phenolic plastic. The small holes should be approximately
served. Some of the more important are as follows: 9.0 mm in diameter. The thermometers under test may then be
(1) Since liquid nitrogen has a lower absolute boiling point held in place by means of small sections cut from a length of
than liquid air, substantial quantities of liquid air or atmo- rubber tubing. It is important that the blades of the propeller
spheric moisture, or both, will condense within the walls of the shall give a downward thrust. If desired, the propeller may be
evacuable flask. Therefore, the flask should never be warmed mounted within a hollow tube so as to minimize swirling and
without being vented to the atmosphere or under vacuum and improve the agitation of the bath liquid. The individual items
should always be vented through a drying tube. needed are as follows:
(2) Liquid air or liquid oxygen, or a mixture of them,
X1.5.1 Materials for Long Bath:
should never be used as the refrigerant.
X1.5.1.1 Vacuum insulated flask 12 in. (305 mm) in depth
(3) At all times the level of the pentane should be slightly
by
above the surface of the aluminum block. Whenever fresh
chilled pentane is added, care must be taken not to spill any on X1.5.1.2 23⁄4 in. (70 mm) in inside diameter,
the heating coil. X1.5.1.3 Phenolic plastic cover, as shown in Fig. X1.4, and
(4) The heating coil should be attached to the variable X1.5.1.4 Glass stirrer shaft with two propellers.
transformer by means of thoroughly insulated leads. X1.5.2 Materials for Short Bath:
X1.5 Comparators for the Range from −80 to +5 °C (−112 X1.5.2.1 Vacuum flask 73⁄4 in. (197 mm) in depth by 45⁄8 in.
to +41 °F)—For temperature testing in this range, as X1.5.2.2 (117 mm) in inside diameter,
indicated in Fig. X1.4, two baths are recommended: one for X1.5.2.3 Phenolic plastic cover, as shown in Fig. X1.4, and
thermometers requiring a long depth of immersion and the X1.5.2.4 Glass stirrer with one propeller.

9
 E77 − 14

FIG. X1.4 Comparator for Temperature Range from −80 to +5 °C (−112 to +41 °F)

X1.5.3 Accessories (for either bath) : a number of ball bearings so that the plate may easily be spun
X1.5.3.1 Stirring motor, to bring any particular thermometer into view. Positive heat
X1.5.3.2 Sections of rubber tubing, distribution to all parts of the bath and thorough circulation are
X1.5.3.3 Solid CO2 and a means of crushing it finely, and obtained by a unique assembly of cooling and heating coils in
X1.5.3.4 Bath medium. This may be either alcohol, light a cylindrical housing which also surrounds the stirring propel-
hydrocarbon, or other organic solvent. If a water-miscible ler. When testing in the range of 5 to 95 °C (41 to 203 °F), the
solvent is used, the water content must be not more than 5 %. bath should be filled with distilled water. When testing in the
X1.5.3.5 In use the proper bath is partly filled with the bath range of 90 to 200 °C (194 to 392 °F), the bath should be filled
medium. Dry ice is added slowly to prevent bubbling over. At with a suitable high-flash-point oil. The motor should then be
the same time another portion of the liquid, sufficient to fill the mounted high enough to protect it from the heat of the bath.
bath and an extra amount to serve as a precooling bath, is
similarly chilled in a second container. When the two are near X1.6.1 An alternative design is illustrated in Fig. X1.6(b).
the proper temperature, the testing bath is filled and brought to This type may be used with water or a series of oils of suitable
temperature. The thermometers to be tested and the standard characteristics in the range of 5 to 315 °C (41 to 599 °F). At the
may be precooled in the standby bath to approximately the higher temperatures great care must be taken to avoid danger-
desired temperature before testing. If large batches of ther- ous flash fires which may occur particularly on removal of
mometers are to be tested, this precooling will materially thermometers or thermometer holder. A solid cover plate to
shorten the time required. replace the thermometer holder, as well as adequate fire
protection equipment of the carbon-dioxide type, should be
X1.6 Comparators for the Range from 5 to 375 °C (41 to provided.
707 °F)— In the range from 5 to 200 °C (41 to 392 °F)
a comparator as illustrated in Fig. X1.5, or its equivalent, may X1.7 Comparator for the Range from 200 to 621 °C (392 to
be used. It consists of a heavy wall borosilicate-glass jar 1150 °F):
suitably supported and equipped with heaters, cooling coil, and
stirring equipment. A bath provided with controls to maintain X1.7.1 A satisfactory comparator for this range is a salt
constant or very slowly rising temperature throughout the bath. One type is illustrated in Fig. X1.7. However, a suitable
testing range and with a top cover having a large opening is apparatus may be assembled using a pressed-steel pot as the
shown in Fig. X1.5. For thermometer testing an insert equipped salt container which is provided with external heaters,
with two rotary holders made as shown in Fig. X1.5 is insulation, stirrer, and thermometer holder. Welded or riveted
provided. Each of these holders is simply a flat plate resting on pots should under no circumstances be used. External electrical

10
 E77 − 14

in. mm
2 50.8
4 101.6
18 457.2

FIG. X1.5 Comparator for Temperature Range from 5 to 200 °C

(41 to 392 °F)

heating is commonly practiced, although gas fired units have attack of the glass may occur. This can be avoided by using
been used successfully. steel wells immersed in the molten salt into which the
X1.7.2 Care should be exercised, particularly above 400 °C thermometers are placed. Thin-walled steel tubes closed at one
(752 °F), to avoid bringing any organic matter or low-boiling end are suitable for this purpose.
liquid, such as water, in contact with the molten salt, since X1.7.4 Comparators with molten tin as the bath liquid, as
dangerous fires or explosions may occur. The bath should be shown in Fig. X1.6(a), have been used successfully in this
heated up slowly at the start to avoid the formation of pockets temperature range. Details of design, construction, and use
of molten salt which have a tendency to blow out the solid
may be obtained from the National Institute of Standards and
mass with disastrous results.
Technology.
X1.7.3 At the lower temperatures the thermometers may be
immersed directly in the salt, but at the higher temperatures

11
 E77 − 14

FIG. X1.6 Alternative Designs of Comparators for Temperature Ranges from 300 to 538 °C (572 to 1000 °F) and from 5 to 315 °C (41 to
599 °F)

12
 E77 − 14

FIG. X1.7 Salt Bath Comparator for Temperature Range from 200 to 621 °C (392 to 1150 °F)

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 [email protected] (e-mail); or through the ASTM website
(www.astm.org). Permission rights to photocopy the standard may also be secured from the Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923, Tel: (978) 646-2600; http://www.copyright.com/

13