This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles

for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.

Designation: E898 − 20

Standard Practice for

Calibration of Non-Automatic Weighing Instruments1
This standard is issued under the fixed designation E898; 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.

1. Scope 1.6 This standard does not purport to address all of the
1.1 This practice applies to the calibration of electronic safety concerns, if any, associated with its use. It is the
non-automatic weighing instruments. A non-automatic weigh- responsibility of the user of this standard to establish appro-
ing instrument is a measuring instrument that determines the priate safety, health, and environmental practices and deter-
mass of an object by measuring the gravitational force acting mine the applicability of regulatory limitations prior to use.
on the object. It requires the intervention of an operator during 1.7 This international standard was developed in accor-
the weighing process to decide whether the weighing result is dance with internationally recognized principles on standard-
acceptable. ization established in the Decision on Principles for the
Development of International Standards, Guides and Recom-
1.2 Non-automatic weighing instruments have capacities mendations issued by the World Trade Organization Technical
from a few grams up to several thousand kilograms, with a Barriers to Trade (TBT) Committee.
scale interval typically from 0.1 micrograms up to 1 kilogram.
Note that non-automatic weighing instruments are usually 2. Referenced Documents
referred to as either balances or scales. In this practice, for 2.1 ASTM Standards:2
iTeh Standards
brevity, non-automatic weighing instruments will be referred to
as balances; however, the scope of this practice also includes
E617 Specification for Laboratory Weights and Precision
Mass Standards
scales.
(https://standards.iteh.ai)
1.3 This practice only covers electronic non-automatic
2.2 ISO Standards:3
ISO 9001 Quality Management Systems – Requirements

Document Preview
weighing instruments where the indication is obtained from a
digital display. The measuring principle is usually based on the
force compensation principle. This principle is realized either
ISO/IEC 17025 General Requirements for the Competence
of Testing and Calibration Laboratories
2.3 OIML Documents and Recommendations:4
by elastic deformation, where the gravitational force of the D28 Conventional Value of the Result of Weighing in Air
object being weighed is measured by a strain gauge ASTMthatE898-20
R76-1 Non-automatic Weighing Instruments – Part 1: Met-
https://standards.iteh.ai/catalog/standards/sist/f1ef36c1-f5ca-47e7-a5b6-a52e7e6b092d/astm-e898-20
converts the deformation into electrical resistance, or by rological and Technical Requirements – Tests
electromagnetic force compensation, where the gravitational R111-1 Weights of classes E1, E2, F1, F2, M1-2, M2, M2-3 and
force is compensated for by an electromagnetic counterforce M3 – Part 1: Metrological and Technical Requirements
that holds the load cell in equilibrium. 2.4 EURAMET Guide:5
1.4 Units—The values stated in SI units are to be regarded Calibration Guide No. 18 Guidelines on the Calibration of
as standard. No other units of measurement are included in this Non-automatic Weighing Instruments
standard. 2.5 JCGM Guides:6
1.5 This standard does not purport to be suitable as the sole JCGM 100 Evaluation of Measurement Data – Guide to the
testing process for weighing systems designated for commer- Expression of Uncertainty in Measurement (GUM)
cial service under weights and measures regulation. The legal
requirements for such instruments vary from region to region, 2
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
and also depend on specific applications. To determine appli- contact ASTM Customer Service at [email protected]. For Annual Book of ASTM
cable legal requirements, contact the weights and measures Standards volume information, refer to the standard’s Document Summary page on
the ASTM website.
authority in the region where the device is located. 3
Available from International Organization for Standardization (ISO), ISO
Central Secretariat, BIBC II, Chemin de Blandonnet 8, CP 401, 1214 Vernier,
Geneva, Switzerland, http://www.iso.org.
1 4
This practice is under the jurisdiction of ASTM Committee E41 on Laboratory Available from the Organisation Internationale de Métrologie Légale (OIML),
Apparatus and is the direct responsibility of Subcommittee E41.06 on Laboratory 11 Rue Turgot, 75009 Paris, France, https://www.oiml.org/en.
5
Instruments and Equipment. Available from EURAMET e.V., Bundesallee 100, 38116 Braunschweig,
Current edition approved May 1, 2020. Published June 2020. Originally Germany, https://www.euramet.org.
6
approved in 1982. Last previous edition approved in 2013 as E898 – 88(2013). DOI: Available from the Bureau International des Poids et Mesures, Pavillon de
10.1520/E0898-20. Breteuil, 92312 Sèvres, France, https://www.bipm.org.

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

1
 E898 − 20
JCGM 200 International Vocabulary of Metrology – Basic 3.1.4 conventional mass value, n—mass value indicated in
and General Concepts and Associated Terms (VIM) the calibration certificate of the weight.
2.6 ILAC Guideline:7 3.1.4.1 Discussion—The conventional mass value of a body
ILAC-G24 Guidelines for the Determination of Calibration is equal to the mass of a standard that balances this body under
Intervals of Measuring Instruments, ILAC-G24 conventionally chosen conditions, that is, at 20 °C, a density of
2.7 USP General Chapters:8 the standard of 8000 kg ⁄m3 in air of density 1.2 kg ⁄m3.14
USP General Chapter 41 Balances 3.1.5 coverage factor (k), n—number larger than one by
USP General Chapter 1251 Weighing on an Analytical which a combined standard measurement uncertainty is mul-
Balance tiplied to obtain an expanded measurement uncertainty.
2.8 UKAS Guide:9 3.1.5.1 Discussion—The coverage factor (k) is typically
UKAS LAB 14, Edition 5 In-house Calibration and Use of calculated based on the applicable degrees of freedom (ν).13
Weighing Machines 3.1.6 eccentric loading error, n—error when the load is not
2.9 Code of Federal Regulations:10 placed in the center of the weighing platform.
CFR Part 58 Good Laboratory Practice for Nonclinical 3.1.6.1 Discussion—The deviation in the measurement
Laboratory Studies value caused by asymmetrical placement of the center of
CFR Part 211 Current Good Manufacturing Practice for gravity of the load relative to the load receptor.15
Finished Pharmaceuticals 3.1.7 error, n—difference between the indicated quantity
2.10 FDA Guidance:11 and the applied quantity.
FDA Questions and Answers on Current Good Manufactur- 3.1.7.1 Discussion—Error is the measured quantity value
ing Practices, Good Guidance Practices, Level 2 Guidance minus a reference quantity value.13
– Equipment 3.1.8 linearity, n—ability of a weighing instrument to follow
For dated references, only the dated version applies to this the linear relationship between a load and the indication.15
practice. For undated references, the latest edition (including 3.1.9 maximum capacity, n—maximum weighing capacity,
all amendments) applies to this practice. that is, the maximum capacity whose weight can be determined
3. Terminology iTeh Standards on a balance.16
3.1.10 maximum permissible error (mpe), n—limit by which
(https://standards.iteh.ai)
3.1 Definitions of Terms Specific to This Standard:
3.1.1 In this section, the official definitions of the terms are
the measured quantity indication can deviate from the nominal
applied value.

Document Preview
provided, as are, where appropriate, simplified definitions that 3.1.10.1 Discussion—The extreme value of measurement
relate the terms to the scope of this practice. error, with respect to a known reference quantity value,
3.1.2 accuracy class of weights, n—class designation of a permitted by specifications or regulations for a given
weight or weight set which meets certain metrological require- measurement, measuring instrument, or measuring system.13
ments intended to maintain the mass values withinASTM specified E898-20
limits. 12 3.1.11 measurement standard, n—realization of the defini-
https://standards.iteh.ai/catalog/standards/sist/f1ef36c1-f5ca-47e7-a5b6-a52e7e6b092d/astm-e898-20
tion of a given quantity, with stated quantity value and
3.1.3 calibration, n—operation that establishes a relation associated measurement uncertainty, used as a reference.13
between the indication of the weighing instrument and the
reference weights, including the associated measurement un- 3.1.12 measurement uncertainty, n—parameter that quanti-
certainties. fies how far a measurement value might be away from the true
3.1.3.1 Discussion—An operation that, under specified (unknown) value.
conditions, in a first step, establishes a relation between the 3.1.12.1 Discussion—The non-negative parameter charac-
quantity values with measurement uncertainties provided by terizing the dispersion of the quantity values being attributed to
measurement standards and corresponding indications with a measurand, based on the information used.13
associated measurement uncertainties and, in a second step, 3.1.13 minimum weight, n—smallest sample weight re-
uses this information to establish a relation for obtaining a quired for a weighment to just achieve a specified relative
measurement result from an indication.13 accuracy of weighing.15
3.1.14 nominal capacity, n—nominal value of the weighing
capacity, derived from the maximum capacity by rounding it
7
Available from the International Laboratory Accreditation Cooperation, ILAC down to a number with less significant digits.15
Secretariat, PO Box 7507, Silverwater NSW 2128, Australia, https://ilac.org.
8
Available from U.S. Pharmacopeial Convention (USP), 12601 Twinbrook 3.1.15 repeatability, n—ability of a weighing instrument to
Pkwy., Rockville, MD 20852-1790, http://www.usp.org. provide indications that are close together when weighing the
9
Available from the United Kingdom Accreditation Service (UKAS), 2 Pine same object several times under reasonably constant test
Trees, Chertsey Lane, Staines-upon-Thames, TW18 3HR, UK, https://
www.ukas.com. conditions.
10
Available from the United States Federal Register, https://www.archives.gov/
federal-register.
11 14
Available from U.S. Food and Drug Administration (FDA), 10903 New Definition from OIML D28.
15
Hampshire Ave., Silver Spring, MD 20993, http://www.fda.gov. Nater, R., Reichmuth, A., Schwartz, R., Borys, M., and Panagiotis, Z.,
12
Definition from OIML R111-1. Dictionary of Weighing Terms, Springer, ISBN 978-3-642-02013-1, 2009.
13 16
Definition from JCGM 200 (VIM). Definition from OIML R76-1.

2
 E898 − 20
3.1.15.1 Discussion—The ability of an instrument to pro- rep repeatability
vide results that agree one with the other when the same load 0 no load
is deposited several times and in a practically identical way on 4. Significance and Use
the load receptor under reasonably constant test conditions.16
4.1 This practice will enable calibration laboratories and the
3.1.16 scale interval, n—the difference between two con- user to calibrate electronic non-automatic weighing instru-
secutive indicated values.16 ments and quantify the error of the balance throughout the
3.1.17 sensitivity, n—change in indication divided by measurement range, usually from zero to maximum capacity.
change in the applied quantity. The error of indication is accompanied by a statement on
3.1.17.1 Discussion—The quotient of the change in an measurement uncertainty, which is individually estimated for
indication of a measuring system and the corresponding change every measurement point. This practice is based on the test
in a value of a quantity being measured.13 procedures and uncertainty estimation described in the
3.2 Symbols: EURAMET calibration guide cg-18. However, while
d—scale interval EURAMET cg-18 allows for a very flexible execution of the
D—drift, value change with time measurements, the test procedures described in this practice are
d1—smallest scale interval more fixed to enable a better comparability between calibra-
E—error of indication tions executed by different calibration laboratories or users.
I —indication This practice may also serve as basis for accreditation of
k—coverage factor calibration laboratories for calibration of electronic non-
L—load automatic weighing instruments.
m —object mass 4.2 This practice allows the user to decide whether the
Max—maximum capacity calibrated balance is fit for its intended purpose, based on the
Max1—upper limit of partial weighing range with smallest assessment of the calibration results. Usually, this assessment
scale interval is done by ensuring that the measurement uncertainty of all
mc—conventional value of mass weighings the user performs on the instrument is smaller than
mmin—minimum weight
mN—nominal mass iTeh Standards a specified relative tolerance established by the user. This
approach is commensurate to assuring that the smallest net

(https://standards.iteh.ai)
mpe—maximum permissible error of a measured quantity amount of substance that the user weighs on the instrument
mref—reference mass (so-called smallest net weight) is larger than the minimum
n—number of repeated weighings weight, which is derived from the calibration results.
R—reading
s—standard deviation
Document Preview 4.3 This practice, in Appendix X2, provides information on
the periodic performance verification on the balance that
SF—safety factor should be carried out by the user between the calibrations.
T—temperature (unit K) ASTM E898-20 Calibration together with periodic performance verification
u—standard uncertainty allows the user to ensure with a very high degree of probability
https://standards.iteh.ai/catalog/standards/sist/f1ef36c1-f5ca-47e7-a5b6-a52e7e6b092d/astm-e898-20
U—expanded uncertainty that the balance meets the user requirements during its day-to-
uc—combined standard uncertainty day usage. It helps users comply with requirements from other
α—intercept (for uncertainty in use) standards or regulations that stipulate periodic tests and cali-
β—slope (for uncertainty in use) brations of quality-relevant instruments.
ν—degrees of freedom
5. Calibration Conditions
νeff—effective degrees of freedom for combined standard
uncertainty 5.1 Standard Weights
ρ—density 5.1.1 General Requirements—Test loads shall consist of
ρ0—reference density of air, ρ0 = 1.2 kg/m3 standard weights that are traceable to the SI unit of mass, with
ρc—reference density of mass, ρc = 8000 kg/m3 the possible exception of test loads used for measurements of
ρS—density of standard weights a comparative nature—for example, measurement of eccentric
loading or repeatability.
3.3 Symbol Subscripts and Definitions 5.1.2 Standard Weight Traceability—The traceability of
Subscript Related weights to be used as standards shall be demonstrated by
B air buoyancy calibration consisting of:
conv convection 5.1.2.1 Determination of the conventional value of mass mc
D drift or the correction δmc to its nominal value m N :δm c 5m c 2m N , or
L load
N nominal value both, together with the expanded uncertainty of the calibration
dig digitization U, or
ecc load in different locations on the
weighing pan 5.1.2.2 Confirmation that mc is within specified maximum
i, j number permissible errors mpe: m N 2 ~ mpe2U ! #m c #m N 1 ~ mpe2U ! .
max maximum
min minimum 5.1.2.3 The standards should further satisfy the following
ref reference requirements to an extent appropriate to their accuracy:

3
 E898 − 20
kg tion. Non-acclimatized weights trigger a convection effect of
(1) Density ρs sufficiently close to ρ c 58000 m 3 ,
the surrounding air that may impact the indication of the
(2) Surface finish suitable to prevent a change in mass balance. Information on acclimatization times may be found,
through contamination by dirt or adhesion layers, and for example, in EURAMET cg-18, Specification E617, or
(3) Magnetic properties such that interaction with the OIML R111-1.
instrument to be calibrated is minimized.
5.1.2.4 Weights that comply with the relevant specifications 6. Calibration Practice
of OIML R111-1 or Specification E617 should satisfy all these 6.1 Measurement Parameters General—This practice in-
requirements. cludes the following measurement parameters:
5.1.3 Recommendation for the Selection of Weight Repeatability;
Classes—It is recommended selecting the weights in order to Eccentric loading; and
minimize their uncertainty contribution to the balance uncer- Error of indication.
tainty as far as reasonably possible so that user tolerance 6.1.1 These three measurements may be performed in any
requirements are met. The following selection criteria for order to accommodate efficient usage of weights and detect
weight classes could be considered: potential defects of the weighing instrument at an early stage.
5.1.3.1 1 000 000 < Max/d: Select an OIML E2 or ASTM
Class 1 weight, or a weight with uncertainty not worse than the 6.2 Calibration
permitted OIML E2 or ASTM Class 1 uncertainty, which is 6.2.1 Calibration Range—The calibration range is typically
one-third or less of the maximum permissible error; it is from the zero point of the balance to maximum capacity, or the
recommended to always use the conventional mass value as the weighing range specified by the customer.
reference value; 6.2.2 Calibration Location—Calibration shall be performed
5.1.3.2 150 000 < Max/d ≤ 1 000 000: For a weight where at the location of use of the balance. If calibration at the
the nominal value is used as the reference value, select an location of use is not possible, then this exception shall be
OIML F1 or ASTM Class 2 weight or better; for a weight where noted in the calibration certificate. If the balance is moved to
the conventional mass value is used as the reference value, another location after calibration, the following situations may

iTeh Standards
select an OIML F2 or ASTM Class 4 weight or better;
5.1.3.3 15 000 < Max/d ≤ 150 000: For a weight where the
occur:
(1) Difference in local gravitational acceleration;
(2) Changes in environmental conditions; and
(https://standards.iteh.ai)
nominal value is used as the reference value, select an OIML
F2 or ASTM Class 4 weight or better; for a weight where the
(3) Impact (damage) during transport.

Document Preview
conventional mass value is used as the reference value, select NOTE 1—Each of the above may cause the performance of the balance
an OIML M1 or ASTM Class 5 weight or better; and to change and can invalidate the calibration. Therefore, moving the
balance after calibration should be avoided. Recalibration of the balance
5.1.3.4 Max/d ≤ 15 000: For a weight where the nominal is recommended after it has been moved if evidence of altered perfor-
value is used as the reference value, select an OIML M1 or mance cannot be excluded.
ASTM Class 5 weight or better; for a weight ASTM where the E898-206.2.3 Preparation Before Calibration—Check the balance
conventional mass value is used as the reference value, select
https://standards.iteh.ai/catalog/standards/sist/f1ef36c1-f5ca-47e7-a5b6-a52e7e6b092d/astm-e898-20
nameplate or product identification. Identify the model, scale
an OIML M2 or ASTM Class 6 weight or better.
interval, maximum capacity, serial number, manufacturer, and
5.1.3.5 For multi-interval and multiple range balances, Max
other applicable information. The balance shall be energized at
and d refer to the interval/range with the smallest scale interval,
an appropriate time prior to calibration, such as the warm-up
abbreviated as Max1 and d1, respectively.
time specified in the balance instructions or the time set by the
5.2 Substitution Loads user.
5.2.1 Calibration requiring substitution loads is not within 6.2.3.1 If there is no such rule, the balance warm-up time
the scope of this practice. Substitution is discussed elsewhere should be no less than 30 min.
(see EURAMET cg-18). 6.2.3.2 If as-found data is to be collected, do not level prior
5.3 Other Measuring Instruments Used for Calibration to as-found calibration. The balance should be levelled before
5.3.1 Thermometer for monitoring air temperature with a as-left calibration.
resolution of 0.5 °C or better. 6.2.3.3 If the balance is equipped with an auto-zero
function, this function should be turned off before calibration,
5.3.2 Hygrometer for monitoring ambient relative humidity
if possible.
with a resolution of 5 % RH or better.
6.2.4 Measurement of Repeatability—Under repeatability
5.4 Calibration Environmental Conditions conditions, the same load is placed on the balance pan several
5.4.1 Environmental temperature and relative humidity times in a practical and consistent manner in a short time
shall be recorded and reported. period. This provides an indication of the balance ability to
5.4.2 Calibration shall be carried out under stable environ- produce consistent results, expressed as a standard deviation.
mental conditions, where possible. Calibration shall be carried Repeatability conditions include:
out within the normal working environmental conditions of the (1) The same measurement procedure;
installation site. (2) The same operator;
5.4.3 The test loads should be acclimatized to the environ- (3) The same measurement system; and
mental conditions in the vicinity of the balance before calibra- (4) The same operating conditions and the same location.

4
 E898 − 20
6.2.5 The test load should consist of a single weight location of measurement points can vary depending on the
whenever possible. shape of the pan.
6.2.6 For balances with only a single scale interval, the test 6.2.10 Depending on the mechanical construction of the
load is usually chosen between approximately 50 % of the balance, the eccentricity test might not be possible or required,
maximum capacity and approximately 100 % of the maximum for example, for balances with a hanging weighing pan or
capacity. If the customer requires a special measurement point, hopper scales. Care must be taken when performing this test on
the test load value can be adjusted. A typical application for balances with leveling pans to position weights so that they do
which the repeatability test load may be adjusted is weighing not tip off the tilted pan surface.
quantities on the balance with mass values significantly smaller 6.2.11 The measurement can be performed with the follow-
than maximum capacity. ing methods:
6.2.7 For multi-interval/multiple range balances, the test 6.2.11.1 Method 1 (Recommended for Single Range
loads are usually chosen between approximately 50 % to Balances)—Set the value to zero before measuring. Place the
approximately 100 % of the upper limit of each interval/range. test load at position 1 and then move to the other positions in
If the customer requires a special measurement point, the test sequence. The indication is recorded for each position.
load value can be adjusted. A typical application for which the 6.2.11.2 Method 2 (Recommended for Multi-interval/
repeatability test load may be adjusted is weighing quantities Multiple Range Balances)—First place the test load in position
on the balance with mass values significantly smaller than 1 and perform a tare operation on the balance. The test load is
maximum capacity. then moved sequentially to the other positions. The indication
6.2.8 Before each measurement, set the balance to zero, load is recorded for each position.
the test load in the middle of the balance weighing pan, and
6.2.11.3 Method 3 (Recommended for Single Range Bal-
record the indication after stabilization. For all balances with a
ances that Might Easily Show Drift)—Set the value to zero
scale interval of 0.1 mg and finer, complete a minimum of 10
before measuring. First place the test load in position 1, remove
measurements. For all balances with a scale interval larger than
it, then place it in the next position, remove it, and so on, until
0.1 mg, complete a minimum of 5 measurements. For multi-
it is removed from the last position. For each position, record
interval/multiple range balances, the number of measurements
iTeh Standards
the indication. The indication is set to zero each time the load
shall be the same in all intervals/ranges. The smallest scale
is removed.
interval shall be taken as reference for defining the minimum
6.2.11.4 Method 4 (Recommended for Multi-interval/
number of measurements.
(https://standards.iteh.ai)
6.2.9 Measurement of Eccentric Loading—The error of
indication of the same load at different positions on the
Multiple
place the
Range Balances that Might Easily Show Drift)—First
test load in position 1 and perform a tare operation on

Document Preview
weighing pan is expressed by the maximum difference between
the indication of the different positions and the indication of the
the
and
balance.
then move
Move the test load to the first off-center position
back to position 1. Record the indication for the
center position. A single weight should be used when possible. off-center position. When the load is back at position 1 and if
For multi-interval/multiple range as well as single range the display is not zero, then the indication is set to zero before
ASTM E898-20
continuing. Move the test load to the next off-center position,
balances, the test load is preferably approximately one-third of
https://standards.iteh.ai/catalog/standards/sist/f1ef36c1-f5ca-47e7-a5b6-a52e7e6b092d/astm-e898-20
the maximum capacity. Measurements include placing the test and back to position 1. Continue until the test load has been
load at different locations on the pan in such a way as to ensure removed from the last off-center and center positions.
that the center of gravity of the applied load is at the positions NOTE 2—If the weighing system auto-zero is activated, the eccentric
shown in Fig. 1, or as close as possible. The number and loading test results can be affected, resulting in a falsely small error. The

FIG. 1 Load Positions

5
 E898 − 20
calibration technician must be observant of such effects and disable the In the formula:
auto-zero, if necessary, to obtain correct results (see 6.2.3).
ILi = ith position indication (i = 2, 3, …), and
6.2.12 Measurement of Error of Indication—The measuring IL1 = center position indication.
points should be uniformly selected within the weighing range 7.3.2 The maximum error of eccentric loading is the value
to be calibrated, as far as reasonably possible. At least 5 reported and included in uncertainty calculations.
different test loads shall be required, including zero and
maximum capacity or close to the maximum capacity (for 8. Evaluation of Measurement Uncertainty
example, 0, 25, 50, 75, and 100 % of maximum capacity).
Measurement points may be added according to customer While the following sections document the likely uncer-
needs. Ideally, each test load should consist of a single weight; tainty for the balance calibration, error sources contributing to
however, multiple weights may be used when necessary to uncertainty may be added or removed as appropriate for the
achieve the appropriate test load. Before each measurement, situation. The error sources described in this practice are the
the indication value is set to zero. The measurement can be most likely to consider.
selected according to the following methods: 8.1 Standard Uncertainty of the Error of Indication During
Calibration—The measurement model of the error of indica-
6.2.12.1 Method 1—Increase from zero load to maximum
tion is:
capacity. Unload the weights at each step of the measurement
process. After unloading, check the zero point. If the zero point E 5 I 2 m ref (7)
is not zero, set the value to zero. The formula for calculating the combined standard uncer-
6.2.12.2 Method 2—Increase from zero load to maximum tainty is:
capacity; no need to unload the weights during the measure-
u 2c ~ E ! 5 u 2 ~ I ! 1u 2 ~ m ref ! (8)
ment. This method may produce creep effects in the results.
6.2.12.3 Method 3—If the customer requests, the measure- 8.1.1 Standard Uncertainty of the Indication u(I)—The
ment can also be performed from maximum capacity to zero measurement model for the indication is:
load, without unloading weights after each measurement. As in I 5 I L 1δI digL 1δI rep 1δI ecc 2 I 0 2δI dig0 (9)

iTeh Standards
Method 2, this may produce creep effects in the results.
8.1.1.1 Standard uncertainty caused by the rounding error of
7. Calibration Results the no-load indication u(δI ). δI represents the rounding
dig0 dig0

(https://standards.iteh.ai)
7.1 Error of Indication Measurement Results—For each test
load, the error of indication (E) is calculated as follows:
error of the no-load indication. The half width of the scale
interval is d / 2, a rectangular distribution is assumed:
0

Document Preview ~ ! ~ = !
E 5 I 2 m ref (1) u δI dig0 5
2
d0
3
(10)

7.1.1 For the value of the reference mass, the nominal mass 8.1.1.2 Standard uncertainty caused by the rounding error of
of the test load can be taken: ASTM E898-20
the indication at load u(δI ). δI represents the rounding
digL digL
https://standards.iteh.ai/catalog/standards/sist/f1ef36c1-f5ca-47e7-a5b6-a52e7e6b092d/astm-e898-20
m ref 5 m N (2) error of the indication at load. The half width of the scale
interval is dL / 2, a rectangular distribution is assumed:
7.1.2 Alternatively, the conventional mass value of the test
load can be taken: dL
u ~ δI digL ! 5 (11)
m ref 5 m c 5 m N 1δm c (3) ~ 2 =3 !
NOTE 3—On multiple range / multi-interval balances, dL can change
7.2 Repeatability Measurement Results—The standard with the indicated value.
deviation, s, is calculated from the load indications of the 8.1.1.3 Standard uncertainty caused by repeatability
repeatability test. u(δIrep). δIrep indicates the repeatability error of the balance, as

s5 Π1
Σ n ~ I 2 Ī !
n 2 1 i51 i
2
(4)
determined in 7.2, a normal distribution is assumed:
u ~ δI rep ! 5 s ~ I ! (12)
(1) For a single range balance, a repeatability measurement
where: may be taken at only one load, and this measurement may be
1 n used to represent the repeatability uncertainty of the entire
Ī 5 Σ i51 Ii (5)
n range of the balance. Multiple repeatability measurements at
different loads may be performed.
In the formulas:
(2) For multi-interval/multiple range balances, repeatabil-
s = standard deviation, ity measurements shall be made separately for each partial
Ii = indication when the ith load is applied, and weighing range, and the uncertainty of each interval/range will
Ī = the average of n indications.
be represented by the uncertainty of the repeatability test in the
7.3 Eccentric Loading Measurement Results respective interval/range.
7.3.1 The error of eccentric loading is calculated based on 8.1.1.4 Standard uncertainty caused by measurement of
the indication values obtained in different load positions ∆Iecc,i eccentric loading u(δIecc). δIecc indicates the error due to the
∆I ecc,i 5 I Li 2 I L1 (6) change of the center of gravity of the test load. This effect may

6
 E898 − 20
occur when the test load is composed of multiple weights and ρ0 mpe
u ~ δm B ! 5 =1.07 3 1024 11.33 3 1026 K 22 ∆T 2 ·m N · 1
can be based on the following assumptions: ρc ~
4 =3 !
(1) The difference ∆I ecc,i 5I Li 2I L1 is proportional to the (20)
distance from the center of gravity of the load to the center of where T is the maximum change in ambient temperature
the pan and is proportional to the load value. The maximum assumed for the site between two consecutive calibrations.
difference |∆I ecc,i | max is determined according to 7.3, and a
rectangular distribution is assumed. As the standard uncertainty 8.1.2.3 Standard Uncertainty Caused by Weight Drift
is expressed as a function of the indication, I, the proportion- u(δmD) (Optional)
ality constant—the load value—is with a very good approxi- (1) The value change with time (D) can be estimated from
mation substituted by the indication: the mass change after successive calibrations of the standard
weight. The difference between the mass of the weights in the
I ? ∆I ?ecc,i max last two calibration cycles or the difference in the mass of the
u ~ δ I ecc ! 5 (13)
~ 2L =3 !
ecc
weights in the recent consecutive multiple calibration cycles
may be used. At least three calibration values are recom-
where: mended to establish trends in the weight stability.
Lecc = test load used for the eccentricity test. (2) In the absence of standard weight drift information, D
8.1.1.5 The standard uncertainty of the indication is ob- may be, in the worst case, selected as the maximum permis-
tained by the following formula: sible error of the standard weight; however, this selection
might result in an overestimation of the uncertainty.
u 2 ~ I ! 5 u 2 ~ δ I dig0 ! 1u 2 ~ δ I digL ! 1u 2 ~ δ I rep ! 1u 2 ~ δ I ecc ! (14)
(3) The standard uncertainty caused by weight drift, assum-
8.1.2 Standard Uncertainty of Reference Mass u(mref)—The ing a rectangular distribution, is:
measurement model for the reference mass is: D
u ~ δ m D! 5 (21)
m ref 5 m N 1δm c 1δm B 1δm D 1δm conv (15) =3
8.1.2.1 Standard Uncertainty of Standard Weights u(δmc) 8.1.2.4 Standard Uncertainty Caused by Convection—If the

iTeh Standards
(1) If the conventional mass of the standard weight is used weights are not sufficiently acclimatized to the environmental
as the reference mass, where the calibration uncertainty, U, and temperature in the vicinity of the balance, the convection effect
the coverage factor are given in the weight calibration of the surrounding air on the weights should be considered. If
(https://standards.iteh.ai)
certificate, the standard uncertainty is:
U
not negligible, the convection effect shall be included in the
standard uncertainty of the reference mass. The uncertainty due
Document
weight is used as Preview
~ ! u δm c 5
k
(16)
to convection is most relevant for non-acclimatized weights of
(2) If the nominal mass of the standard classes OIML F1 or better, and ASTM 3 or better. Further
the reference mass, assuming a rectangular distribution, the information can be found in EURAMET cg-18, Appendix F.
standard uncertainty is: 8.1.2.5 The standard uncertainty of the reference mass is:
ASTM E898-20 2
mpe u ~ m ref ! 5 u 2 ~ δ m c ! 1u 2 ~ δ m B ! 1u 2 ~ δ m D ! 1u 2 ~ δ m conv ! (22)
https://standards.iteh.ai/catalog/standards/sist/f1ef36c1-f5ca-47e7-a5b6-a52e7e6b092d/astm-e898-20
u ~ δm c ! 5 (17)
=3 8.1.3 Combined Standard Uncertainty of the Error of Indi-
(3) If the test load consists of multiple standard weights, cation uc(E):
the standard uncertainty is the arithmetic sum of the standard
u 2c ~ E ! 5 u 2 ~ I ! 1u 2 ~ m ref ! (23)
uncertainty of each standard weight.
Then substitute using Eq 14 and Eq 22.
8.1.2.2 Standard Uncertainty Caused by Air Buoyancy
u(δmB) u 2c ~ E ! 5 u 2 ~ δ I dig0 ! 1u 2 ~ δ I digL ! 1u 2 ~ δ I rep ! 1u 2 ~ δ I eec ! 1u 2 ~ δ m c !
(1) If the sensitivity of the balance is adjusted prior to 1u 2 ~ δ m B ! 1u 2 ~ δ m D ! 1u 2 ~ δ m conv ! (24)
calibration, the standard uncertainty for air buoyancy is: 8.2 Expanded Uncertainty of the Error of Indication
mpe 8.2.1 The expanded uncertainty of the error of indication
u ~ δm B ! 5 (18)
~ 4 =3 ! error is:
(2) If the sensitivity of the balance is not adjusted prior to U ~ E ! 5 k·u c ~ E ! (25)
calibration, the standard uncertainty for air buoyancy is:
8.2.2 The coverage factor, k, shall be selected using 8.2.2.1,

u ~ δm B ! 5
S 0.1 · m N ·
ρ0
ρc
1
mpe
4 D (19)
8.2.2.2, or 8.2.2.3.
8.2.2.1 Option 1: Select the coverage factor, k, from Table 1
=3 based on the degrees of freedom associated with the repeat-
(3) If information on the temperature variation on the ability measurement in the calibration. Degrees of freedom are
balance calibration site is available, Eq 19 can be replaced by: the number of repeatability weighings minus one.

TABLE 1 Coverage Factor k for Different Effective Degrees of Freedom veff

veff 4 5 6 7 8 9 10 15 20 50 `
k 2.87 2.65 2.52 2.43 2.37 2.32 2.28 2.18 2.13 2.05 2.00