Measurement Good Practice Guide: The Measurement of Mass and Weight

No.
71
Measurement
Good Practice Guide
The Measurement of
Mass and Weight
S Davidson, M Perkin, M Buckley
The National Physical Laboratory is operated on behalf of the DTI by NPL Management Limited, a wholly owned subsidiary of Serco Group plc
Measurement Good Practice Guide No. 71
The Measurement of Mass and Weight
Stuart Davidson, Michael Perkin

Division of Engineering and Process Control
National Physical Laboratory
Mike Buckley
South Yorkshire Trading Standards Unit
Abstract: This Good Practice Guide is intended as a useful reference for those
involved in the practical measurement of mass and weight.
ã Crown Copyright 2004
Reproduced by permission of the Controller of HMSO
ISSN 1368-6550
June 2004
National Physical Laboratory

Teddington, Middlesex, United Kingdom, TW11 0LW
Website: www.npl.co.uk
Acknowledgements
The authors acknowledge the financial support of the National Measurement System
Directorate of the UK Department of Trade and Industry.
The Measurement of Mass and Weight
Contents
1 Introduction ........................................................................................................................1
2 Weights................................................................................................................................2
2.1 Material.................................................................................................................2
2.2 Surface condition..................................................................................................2
2.3 Magnetism ............................................................................................................2
2.4 Weight cleaning....................................................................................................2
2.5 Weight handling ...................................................................................................3
2.5.1 Gloves.......................................................................................................3
2.5.2 Tweezers ...................................................................................................3
2.6 Clean surfaces.......................................................................................................4
2.7 Weight storage......................................................................................................4
3 Weighing techniques ..........................................................................................................4
3.1 Direct reading measurements ...............................................................................4
3.2 Weighing by differences.......................................................................................5
3.3 Substitution weighing ...........................................................................................6
3.4 ABA calibration....................................................................................................6
3.5 ABBA calibration .................................................................................................7
3.6 Cyclic weighing....................................................................................................7
3.7 Weighing by sub-division.....................................................................................8
3.8 Make-weights .......................................................................................................9
4 Air density measurement and buoyancy correction .......................................................9
4.1 True mass..............................................................................................................9
4.2 Conventional mass................................................................................................10
4.3 Buoyancy correction.............................................................................................10
4.4 The application of buoyancy corrections..............................................................11
4.5 Determination of air density from parametric measurements ..............................12
5 Balance assessment ............................................................................................................12
5.1 Balance location ...................................................................................................13
5.2 Measurement equipment.......................................................................................13
5.3 Two-pan balances .................................................................................................13
5.4 Single-pan mechanical balances ...........................................................................14
5.5 Single-pan electronic balances .............................................................................14
5.6 Assessment of two pan balances ..........................................................................14
5.6.1 Routine periodic checks............................................................................14
5.6.2 Rider weight .............................................................................................14
5.6.3 Internal balance weights ...........................................................................15
5.6.4 Linearity of scale ......................................................................................15
5.6.5 Sensitivity .................................................................................................15
5.6.6 Repeatability of reading............................................................................15
5.6.7 Repeatability of measurement ..................................................................15
5.7 Assessment of mechanical single pan balances....................................................16
5.7.1 Visual inspection and mechanical check ..................................................16
5.7.2 Drift ..........................................................................................................16
5.7.3 Calibration of internal weights .................................................................16
5.7.4 Effect of off centre loading (eccentricity) ................................................16
5.7.5 Hysteresis..................................................................................................16
5.7.6 Scale sensitivity (scale value)...................................................................17
5.7.7 Scale linearity ...........................................................................................17
5.8 Assessment of electronic balances .......................................................................17
5.8.1 Hysteresis..................................................................................................18
5.8.2 Effect of off centre loading (eccentricity) ................................................18
5.8.3 Scale error and linearity............................................................................18
6 Uncertainty in mass calibration........................................................................................19
6.1 Introduction ..........................................................................................................19
6.2 The measurement..................................................................................................19
6.3 The uncertainty budget .........................................................................................20
6.3.1 Symbol......................................................................................................20
6.3.2 Sources of uncertainty ..............................................................................20
6.3.3 Values .......................................................................................................21
6.3.4 Probability distributions ...........................................................................21
6.3.5 Divisor ......................................................................................................23
6.3.6 Sensitivity coefficient (ci).........................................................................23
6.3.7 Standard uncertainty in units of measurand, ui(Wi)..................................23
6.3.8 Degrees of freedom vi ...............................................................................23
6.3.9 Adding it all up .........................................................................................24
6.4 Air buoyancy uncertainty budget .........................................................................25
6.5 Reporting the results.............................................................................................27
7 References ...........................................................................................................................28
1 Introduction
The unit of mass, the kilogram, remains the only base unit in the International System
of Units (SI), which is still defined in terms of a physical artefact. Its definition is:
“The kilogram (kg) is the unit of mass; it is equal to the

mass of the international prototype of the kilogram.”
The difference between ‘mass’ and ‘weight’ is that mass is a measure of the amount
of material in an object, weight is the gravitational force acting on a body. However,
for trading purposes weight is often taken to mean the same as mass.
The international prototype of the kilogram is kept at BIPM, the International Bureau
of Weights and Measures in Sèvres, Paris. It consists of an alloy of 90% platinum and
10% iridium in the form of a cylinder, 39 mm high and 39 mm in diameter. It is stored
at atmospheric pressure in a specially designed triple bell-jar.
About 60 countries hold platinum-iridium alloy copies of the BIPM kilogram (K),
whose values have been determined directly from K. The National Physical
Laboratory (NPL) holds the UK copy (No. 18), which is referred to as the national
prototype kilogram, or simply kilogram 18 and is the basis of the entire mass scale in
the UK. The NPL participates in a wide range of international comparisons to ensure
that measurements made in the UK are equivalent to those made elsewhere in the
world. In the past there have been some problems with organisations based in one
country not accepting traceability to any NMI other than their own. This situation has
been addressed with the advent of a structured approach to international equivalence
via a Mutual Recognition Agreement (MRA) and regular international measurement
comparisons.
The use of appropriate mass standards

and their correct treatment at all times
is essential to mass metrology. Weights
are divided into classes from high
quality reference standards (Class E1)
to those used in industrial settings
(Class M3). These classes were
originally specified for legal metrology
purposes, but they are now in common
usage throughout mass measurement.
The International Organisation for
Legal Metrology (OIML) document
OIML R111 [1] specifies the properties
of weights of each class and specifies
the tests that are necessary prior to
carrying out a mass calibration.
Figure 1: The UK national standard kilogram
1
2 Weights
2.1 Material
Weights should be made of a material that is chemically inert, non-magnetic, hard

enough to resist scratching and of a density that meets the OIML R111
recommendations for its class. Austenitic stainless steel is generally used in the
construction of Class E1 and E2 weights. Lower accuracy weights may be
manufactured from plated brass, iron or other suitable materials. Unplated brass
should be avoided due to its susceptibility to atmospheric induced surface instability.
Class E1 and E2 weights

must be integral in
construction, i.e. be made of
a single piece of material.
Other weights classes can
be made up of multiple
pieces with a sealed cavity
to allow for adjustment. As
with the other properties of
weights the shape of
weights for particular
classes is defined in OIML
Recommendation R111.
Figure 2: A stainless steel secondary standard weight set
2.2 Surface condition
Prior to use weights should be inspected for surface damage such as scratching or
contamination. It is almost inevitable that weight surfaces will become slightly
scratched in use, but weights with gross scratching can potentially become unstable
due to contamination filling the scratch mark.
2.3 Magnetism
There are two magnetic properties that must be measured to characterise weights,
permanent magnetisation and magnetic susceptibility. Magnetic susceptibility is a
measure of whether the weight can become magnetised by being placed in a magnetic
field (magnetism of this sort is transient) while permanent magnetism is a feature of a
weight that cannot be altered. It may be possible to de-Gauss magnetically susceptible
weights using a commercial de-Gausser, but such treatment has no effect on
permanent magnetisation. OIML Recommendation R111 sets out permissible limits
for these two properties for various classes of weights.
2.4 Weight cleaning
Stainless steel weights (OIML E and F Classes) should routinely be dusted before use
using a clean soft haired brush. With the exception of this, the cleaning of weights
should be avoided unless it is absolutely unavoidable as it will affect their calibration
2
history. Where cleaning is unavoidable, it is recommended that weights should be

calibrated both before and after cleaning. Rubbing with a soft clean cloth is often
enough to remove marks. Solvent cleaning should be used as a last resort. In general
this will take the form of wiping the weight with a clean cloth that has been soaked in
solvent.
After cleaning it is necessary to allow weights to stabilise before calibrating them.

The stabilisation time will vary, from a few hours to several days, according to the
class of weight and the extent of cleaning that has been undertaken [6].
Cast iron weights, if in good condition, may be cleaned by brushing with a stiff brush.
Rust can be removed with a wire brush.
2.5 Weight handling
Weights must always be handled with the greatest care. It is important that they are
never:
a) touched with bare hands

b) handled with sharp or abrasive tools and materials
c) in contact with tools or surfaces that are not scrupulously clean
d) slid across surfaces
e) knocked together
f) breathed on or spoken over by the operator
g) cleaned by inappropriate means
The following handling methods are recommended in order to avoid the problems
mentioned above:
2.5.1 Gloves
Gloves should be worn whenever practicable. Chamois leather is an ideal material for
such gloves as it has good thermal insulation properties (so reducing thermal
influences on the balance during the weighing process) and affords a good grip on
large weights when manipulating them.
2.5.2 Tweezers
Tweezers, forceps or other specialist lifting devices should be used whenever

practicable to pick up weights and manipulate them inside the balance. It is important
that these tools have no sharp edges and should, when possible, not have metal to
metal contact with the weights.
3
Ungloved hand, potential heat Users hand in balance enclosure Gloved hand outside enclosure
transfer to weight and enclosure. affecting the thermal stability. Recommended practice.
Figure 3: Good practice in weight handling
2.6 Clean surfaces
Surfaces in a mass laboratory should be kept clean and dust free at all times.
However, it is advisable to place acid free tissue paper, or something similar, on any
surface prior to putting weights on it.
2.7 Weight storage
Weights should be stored in a manner that protects them from damage or

contamination and allows them to be easily identified. Manufacturers generally supply
weights in felt lined wooden boxes. Such boxes normally have a recess for each
weight and a lid which touches the weights when closed to hold then in place. Smaller
weights tend to be loosely confined within a small aperture in their box.
The major disadvantage with using boxes for storage is that they contact virtually the
entire surface of the weight so potentially causing contamination problems. It is
important that the interiors of weight boxes are kept clean to prevent a build up of
dust and other particles.
It is preferable for weights to be stored on acid free tissue paper under glass domes.
This is an ideal situation as only the base of the weight is in contact with other
surfaces. Regardless of how the weight is stored it is important to ensure that it is
clearly identified at all times so that it cannot get confused with similar weights.
3 Weighing techniques
There are many weighing techniques currently employed in mass metrology. This is
indicative of the wide range of processes that rely on weighing and the uncertainty
demands that are required by different industrial sectors and end-users.
3.1 Direct reading measurements
The most simple method of weighing is to simply place a test piece on a mass balance
and take the displayed reading as its weight. This type of measurement is only suitable
for low accuracy applications but even in this most straight-forward application it is
essential to follow good practice.
4
As with any form of mass measurement it is

essential to have the balance calibrated on a
regular basis. It is recommended that the
balance undergoes a full assessment and
calibration, by a suitably accredited body, on
a periodic basis which will be influenced by
the application and frequency of use. It is
important that the balance scale is tared
before use. If there is an internal balance
calibration feature this should be used prior to
making measurements on the balance. A
reading with zero load should be taken (z1)
followed by the reading with the load on the
pan (r1) and a final zero reading (z2). This
allows the user to compensate for any drift in
the instrument. The drift corrected reading (rd)
is given by:
z1 + z 2
Figure 4: Electronic balance rd = r1 −
2
3.2 Weighing by differences
This technique is particularly common in analytical chemistry. In general it involves

placing a container on the balance pan, noting the reading, and then adding a
substance to it. The final balance reading is noted and the difference between the two
readings is taken to be the amount of material in the container. This approach is fine
for relatively low accuracy requirements but is not ideal for more demanding
measurements. This type of measurement is susceptible to problems with temporal
balance drift (which is generally more of a problem with modern electronic analytical
balances than was the case with mechanical balances).
A more robust method of carrying out measurements of this type is to have two
similar containers, one to fill with the test substance and one to act as a reference. The
weighing should then be carried out as follows:
a) The reference container is placed on the balance pan and the reading noted
(ref1).
b) The empty container (to be loaded with the test substance) is placed on the
balance (test1)
c) The container is filled with the test substance and the balance reading
taken (test2)
d) The reference container is put back on the pan and the reading taken (ref2).
The weight of substance added to the container (wt) may then be calculated as
follows:
wt = (test 2 − test1 ) − (ref 2 − ref1 )
5
This weighing scheme eliminates drift and should allow better weighing uncertainties
to be achieved.
3.3 Substitution weighing
High accuracy weighing is generally carried out by comparing the test-weight with
mass standards of similar nominal value. This is known as substitution weighing.
When weight A (the first weight to be place on the balance pan) is compared with
weight B (the second weight used) there will generally be a difference (∆m) between
the readings with each of the weights on the pan. This may be expressed in terms of
the following equation:
A = B + ∆m
It is important to follow a consistent nomenclature when carrying out weighings and

expressing the comparison in terms of a simple equation is perhaps the best way to do
this. Throughout this section the concept of an equation is used to describe a
comparison weighing.
If the test-weight is of an unusual nominal value it will be necessary to use a standard

made up of several calibrated weights, or to use additional weights to make the test-
weight up to a standard’s nominal mass. In these cases it is recommended that a direct
reading be taken as a coarse check to ensure all the weights have been noted down.
The number of comparisons carried out in order to calibrate a weight, and the number
of independent mass standards used will depend on the uncertainty requirements of a
particular calibration.
There are two popular techniques for comparative calibration: ABA… and ABBA
calibrations.
3.4 ABA… calibration
This calibration involves the comparison of two weights, normally an unknown and a
standard, by placing each one in turn on the balance pan and noting the reading. The
process is symmetrical in that the weight that is placed on the pan first is also on the
pan for the final reading. It is possible to vary the number of applications of each
weight according to the application, but three applications of the test weight and two
of the standard would be a typical regime. This leads to the following measurement
scheme:
A1 B1 A2 B2 A3
where Ai and Bi represent the balance reading for the ith application of weights A and
B respectively
The difference between weights A and B is then calculated from the equation:
A1 + A2 + A3 B1 + B2
∆m = −
3 2
6
This may be generalised for an ABA comparison involving n applications of weight B

to become:
n +1 n
å Ai åB i
∆m = i =1
− i =1
n +1 n
3.5 ABBA calibration
The ABBA calibration method is potentially more efficient than the ABA method in
terms of the number of weight applications required to produce a calibration result of
a particular uncertainty. In this case the weights are applied to the balance pan in the
following order:
A1 B1 B2 A2
Where: An and Bn represent the balance reading for the nth application of
weights A and B respectively.
It is important in this weighing method that weighings B1 and B2 must be independent

of each other. To achieve this on a typical electronic comparator the weight must be
removed from the pan between readings. If a balance with an arrestment mechanism
is being used it is usually sufficient to arrest the balance and open and close the
balance case between the measurements. The mass difference (∆m) is calculated from
the following equation:
A + A2 B1 + B2
∆m = 1 −
2 2
3.6 Cyclic weighing
High accuracy (OIML Class E and F) mass calibration usually involves calibration
against two or more mass standards. Often it is necessary to calibrate several weights
of the same nominal value. In this case it is inefficient to calibrate each weight against
two standards. A technique known as cyclic weighing is used to rationalise the
number of measurements necessary.
Comparison Measured Corrected Using S1 value Using S2 value Measured

Difference Difference value
S1 v T1 d1 d1 - C T1= S1-d1+C T1= S1*-d1+C
T1 v T2 d2 d2 - C T2=T1-d2+C T2=T1-d2+C The average
T2 v T3 d3 d3 - C T3=T2-d3+C T3=T2-d3+C of the values
T3 v S2 d4 d4 - C S2*=T3-d4+C S2=T3-d4+C for T1 to T6
S2 v T4 d5 d5 - C T4=S2*-d5+C T4=S2-d5+C calculated in
T4 v T5 d6 d6 - C T5=T4-d6+C T5=T4-d6+C the previous
T5 v T6 d7 d6 - C T6=T5-d7+C T6=T5-d7+C columns
T6 v S1 d8 d8 - C S1=T6-d8+C S1*=T6-d8+C
Sum d1+ d2+...+ d8 0
Correction C = sum/8
* The value for S2 calculated using S1 (and vice versa) is checked against the true value of S2
The table shows a typical weighing scheme for a cyclic weighing. Although two
standards are used the number of weighings required is reduced from 12 to 8 through
the use of this scheme. The weighing scheme is symmetrical and, as each weight is
7
used the same number of times, the sum of the measured weight differences should be
zero. In practice this is not usually the case and a correction, C, is calculated to allow
for temporal changes. The weighings must be carried out in one series to allow this
correction to be made. The values for the test weights T can be calculated in turn from
each of the two standards and an average taken. The scheme contains two checks on
the goodness of the weighings, the size of the correction C and the agreement between
the values of T calculated from the two standards.
3.7 Weighing by sub-division
Sub-division is used for the most demanding applications. It involves the use of
standards of one or more values to assign values to weights across a wide range of
mass values. A typical example of this would be to use two or three 1 kg standards to
calibrate a 20 kg to 1 mg weight set. Equally it would be possible to use a 1 kg and a
100 g standard for such a calibration. Typically this is used for OIML E Class
calibrations.
This is most easily illustrated by considering how values would be assigned to a

weight set using a single standard. In reality the weighing scheme would be extended
to involve at least two standards. The standard is compared with any weights from the
set of the same nominal value and also with various combinations of weights from the
set that sum to the same nominal value. A check-weight, which is a standard treated in
the same manner as any of the test-weights, is added in each decade of the calibration
so that it is possible to verify the values assigned to the weight set. In the case of a
1 kg to 100 g weight set the following minimal weighing scheme may be used:
1000 =1000S
1000 = 500 + 200 + 200D +100C
1000S = 500 + 200 + 200D +100
500 = 200+ 200D+100
500 = 200+ 200D+100C
200 = 200D
200 = 100 +100C
200D = 100 +100C
100 = 100C
In a simple case such as this it is possible to calculate the values of the test weights
manually, but in a realistic situation when several standards are used and information
is required about the weighing scheme uncertainty it is necessary to undertake a least
squares analysis using a computer.
The sub-division weighing scheme has the following advantages:

a) it minimises use on (and hence wear on) standards
b) it produces a set of data which provides important statistical information
about the measurements and the day to day performance of the individual
balances
c) there is a redundancy of data (ie more measurements than unknowns).
It has the following disadvantages:

a) it requires a relatively complex algorithm to analyse the data
8
b) it necessitates placing groups of weights on balance pans (this can cause

problems for instruments with poor eccentricity characteristics or
automatic comparators designed to compare single weights.
3.8 Periodicity of Calibration
The period between the calibration depends on the amount and type of use the weights
experience and the accuracy class of the weights. As a general rule new weights
should be calibrated annually until a reasonable calibration history (at least three
calibrations) has been build up. After this, depending on the stability of the weights,
the calibration period can be extended to two or even four years.
3.9 Scale errors and the use of make-weights
Make-weights are weights that are added to a load in order to make it approximately
equal in weight to the object it is being compared with. Make-weights are added so
that only a small part of the comparator’s scale is used during a comparison. For
example if a 100 g scale had a 1 % error it would equate to a 0.2 g error if there were
a 20 g difference between the loads under test, but this could be reduced to a 0.01 g
error if a make-weight is used to balance the loads to within 1 g.
If a make-weight of value Mw is added to load A during a comparison the weighing

equation becomes:
A + M w = B + ∆m
so in order to calculate the value of A from B it is necessary to subtract the value of

any make-weights used in association with A from the sum of the value of B and the
calculated mass difference. Similarly, if make-weights had been used with weight B
the value of the make-weights would be subtracted from the value of A before taking
the difference in the balance readings into account.
4 Air density measurement and buoyancy correction

The measurement of air density is necessary in the field of mass measurement to allow
buoyancy corrections to be made when comparing weights of different volume in air. It
is particularly important when comparing weights of different materials or when making
mass measurements to the highest accuracy.
4.1 True mass
The mass of a body relates to the amount of material of which it consists. In terms of
the calibration of weight it is referred to as true mass in order to differentiate it from
conventional mass, which is generally used to specify the value of weights (see
below). The international prototype of the kilogram for which the mass scale
throughout the world is realised is defined as a true mass of exactly 1 kilogram. All
high accuracy comparisons should be performed on a true mass basis (including class
E1 and E2 calibrations) although values may be converted to conventional mass when
quoted on a certificate.
9
4.2 Conventional mass
This is the value normally quoted on a certificate and is the conventional value of the
result of a weighing in air, in accordance with International Recommendation OIML
R 33. For a weight at 20 °C, the conventional mass is the mass of a reference weight
of a density of 8000 kg/m3, which it balances in air of a density 1.2 kg/m3.
A conventional mass value for an artefact can be calculated from a true mass value
using the following equation:
æ æ 1 1ö ö
M c = M t çç 1 + çç − ÷÷ 1.2 ÷
÷
è è 8000 ρ ø ø
where: Mc is the conventional mass of the artefact in grams

Mt is the true mass of the artefact in grams
ρ is the density of the artefact in kg/m3
4.3 Buoyancy correction
This is a correction made when comparing the mass of artefacts of different volumes. It
is equal to the difference in the volumes of the artefacts multiplied by the density of the
medium in which they are compared (usually air). When comparing true mass values the
buoyancy correction to be applied between two artefacts can be given by the following
equation:
BC = (V1 − V2 ) × ρair
Where: BC is the buoyancy correction to be applied

Vn is the volume of the artefact n (ie Mn / ρn)
ρair is the density of the air at the time of comparison
The calculated buoyancy correction should be applied in the form:
M t 1 = M t 2 + BC
Where: Mt n is the true mass of artefact n
When calibrating standard weights comparisons are normally performed on a

conventional mass basis. For such comparisons the buoyancy correction depends on the
difference in air density from the conventional value of 1.2 kg/m3. Because of the way
the conventional mass is specified, comparisons made in air of density exactly 1.2 kg/m3
require no buoyancy correction no matter what the volume difference of the weights
being compared. For comparisons performed on a conventional mass basis the buoyancy
correction is given by the following equation:
. × 10 −3 )
BC = (V1 − V2 ) × ( ρair − 12
The buoyancy correction is applied with the same convention as for the true mass
correction, ie:
10
M c1 = M c 2 + BC
The equation for conventional mass buoyancy correction is more complicated than for
true mass and care must be taken with the sign of the correction. If weight 1 has a greater
volume than weight 2 and is compared in air of density greater than 1.2 kg/m3 the
correction (BC) will be positive.
4.4 The application of buoyancy corrections
The table below shows the magnitude of the buoyancy correction when comparing
weights of stainless steel with those of another material in air of standard density
(1.2 kg/m3) on a true mass basis.
Material compared with Buoyancy correction (ppm)

Stainless steel
Platinum Iridium 94
Tungsten 88
Brass 8
Stainless Steel 7.5*
Cast Iron 24
Aluminium 294
Silicon 365
Water 875
*This is the result of comparing two types of stainless steel, with densities 7.8 and 8.2 g/cm3
Table 1: Buoyancy Corrections when Comparing Dissimilar Materials in Air
The table shows that even when

comparing weights of nominally the same
material (such as stainless steel) attention
must be paid to buoyancy effects when the
best uncertainty is required.
When comparing weights of dissimilar

materials the effect of air buoyancy
becomes more significant and must be
applied even for routine calibrations when
true mass values are being measured.
Figure 5: stainless steel and platinum kilograms
When working on a conventional mass basis the buoyancy corrections become smaller.
International Recommendation OIML R 33 use a range for air density of 1.1 to 1.3
kg/m3 (ie. approximately ± 10% of standard air density) meaning the corrections are
about one tenth of the true mass corrections. This together with the limits specified by
OIML R 33 for the density of weights of Classes E1 to M3 mean that the maximum
correction for any weight is one quarter of its tolerance. This is generally not significant
for weights of Class F1 and below (although allowance should be made for the
11
uncertainty contribution of the un-applied correction) but for Class E1 and E2 weights
buoyancy corrections need to be applied to achieve the uncertainty values required for
weights of these Classes.
4.5 Determination of air density from parametric measurements
The standard method for determining air density involves the measurement of
temperature, pressure and humidity. From these measurements, and taking into account
carbon dioxide concentration for the best accuracy, the density of the air can be
calculated. The empirical formula for the calculation of air density recommended by the
Comité International des Poids et Mesures (CIPM) was derived by Giacomo [2] and
modified by Davis [3]. Table 2 shows typical and best achievable uncertainties for the
calculation of air density from the above parameters using the CIPM formula.
Routine Measurement Best Capability

Uncertainty ppm Uncertainty ppm
Temperature (oC) 0.1 360 0.01 36
Pressure (mbar) 0.5 500 0.05 50
Humidity (% rh/oC dew pt.) 5% 350 0.25oC 58
CO2 content (ppm) - - 50 21
CIPM Equation 100 100
Total (x 10-3 kg/m3) 0.86 720 0.16 133
Table 2: Routine and Best Achievable Realisation of Air Density using the CIPM Formula
5 Balance assessment
The need for regular and appropriate assessment of balances is vital both in providing
traceability of results and in providing accurate and reliable results. Balances can be
used either as comparators or as direct reading devices. The assessment of the balance
will depend on the mode in which it is used. When used as a comparator traceability
for the calibration will be provided by the mass standard against which the unknown
is compared and the performance of the balance is effectively given by its
repeatability. When used as a direct reading device traceability for the measurement is
in effect provided by the balance itself and a number of other factors such as the
linearity, stability and temperature sensitivity of the balance become important. In all
cases it is vital that:
• the balance assessment reflects the way in which the balance is used in practice
• it takes place in the same conditions as the balance would normally be used
It is essential that both these conditions are met if the results of the balance
assessment are to give an accurate reflection of the performance of the balance and
therefore provide a valid contribution to an uncertainty budget for measurements
performed on the balance.
12
5.1 Balance location
To perform at its best the balance should be located in a position as free from
vibration, thermal instability, direct sunlight and draughts as possible. Care should
also be taken to avoid external influences such as magnetic fields and radio frequency
interference. The performance of the balance depends to a great extent on the
environment in which it is located so some thought should be given to optimising its
position with respect to the above parameters while allowing practical use of the
balance. As previously stated the balance should be assessed in the location in which
it will be used. If a balance is used in a shop floor environment it should not be
assessed in a temperature controlled laboratory environment.
5.2 Measurement equipment
A basic assessment of the performance of a balance for repeatability, hysteresis and

eccentric loading can be performed with two un-calibrated weights of similar nominal
values. A more comprehensive assessment, however, requires a set of calibrated
weights and a temperature probe to monitor the temperature of the balance. Since the
sensing elements of most balances are temperature sensitive it is vital to keep a record
of the balance temperature while it is being assessed as unusual variations in
temperature may result in poor performance. A calibrated set of weights allow the
scale error of the balance to be assessed.
5.3 Two-pan balances

These balances consist of a symmetrical beam
and three knife edges. The two terminal
knives support the pans and a central knife
edge acts as a pivot about which the beam
swings. Two-pan balances are generally un-
damped with a “rest point” being calculated
from a series of “turning points”. Some
balances incorporate a damping mechanism
(usually mechanical or magnetic) to allow the
direct reading of a “rest point”.
Readings from this type of balance tend to be

made using a simple pointer and scale
although some use more complicated optical
displays. In all cases the reading in terms of
scale units needs to be converted into a
measured mass difference.
Figure 6: Two-pan mechanical balance
Due to the construction of the balance they are always used as comparators
(comparing the load on one pan with the other) but the way in which the comparison
is achieved can vary - double-double, double, double-substitution and substitution
modes can all be used. It is vital that the balance is assessed using the same mode of
comparison as is used for the normal operation of the balance.
13
5.4 Single-pan mechanical balances
These generally consist of a beam with two knife

edges, one to support the weighing pan and one
acting as a pivot. A fixed counterweight balances the
load on the pan.
This type of balance is usually damped and a series

of built in weights allow the calibration of a range of
mass values by maintaining a constant load on the
balance beam.
The displays on these balances tend to be of the

optical variety and the sensitivity of the balance can
usually be adjusted by the user.
Figure 7: Single-pan mechanical balance
5.5 Single-pan electronic balances
These are usually top loading balances with the applied load being measured by an
electro-magnetic force compensation unit or a strain gauged load cell. Single pan
electronic balances give a direct reading of the weight applied whereas the other two
mechanical balance types rely on the comparison of two forces (an unknown weight
with either an external or internal weight). Despite the possibility of using these
balances as direct reading devices (applying an unknown weight and taking the
balance reading as a measure of its mass) single pan electronic balances will always
perform better when used as comparators, comparing a standard and an unknown in
an ABA or ABBA sequence. As with all the balance types discussed the method of
assessment should reflect the way in which the balance is used in practice.
5.6 Assessment of two pan balances
This type of balance is less frequently used mainly due to the amount of time it takes
to make a weighing compared with electronic balances. For this reason only a brief
outline of the tests that should be performed on this type of balance is given. Further
details of these tests may be found in NPL Notes on Applied Science No.7, 'Balances,
Weights and Precise Laboratory Weighing', 1954.
5.6.1 Routine periodic checks
The following tests should be carried out on a regular basis and are essential to the
routine operation of the balance.
5.6.2 Rider weight
The effective value (mass value corrected for air buoyancy based on the actual density
of the weight) of the rider weight should be measured against suitable mass standards.
14
5.6.3 Internal balance weights
Instead of (occasionally as well as) a rider bar some two-pan balances have built in
poising weights. These should be removed and calibrated in the same way as the rider
weight.
5.6.4 Linearity of scale
This can be checked using either fractional weights or with the rider and rider bar.
5.6.5 Sensitivity
Measuring the sensitivity of the balance allows a measured difference in terms of

scale divisions to be converted into a mass difference and is therefore vital to the use
of the balance. A suitable sensitivity weight, which can be easily transferred between
the balance pans without arresting the balance, is required. The value of this weight
should be enough to cause the pointer to move along the scale by between one quarter
and one third of the total scale length. The sensitivity weight should be calibrated
against suitable mass standards. The weight is then swapped between the balance pans
(left to right and then right to left) without arresting the balance. Arresting the balance
would change the effective rest point slightly and compromise the accuracy of the
sensitivity measurement. The average effect of swapping the sensitivity weight (dav) is
calculated (the two figures should agree to better than 5%) and this value used to
calculated the sensitivity as follows:
2 × WS
S=
d av
Where: dav is the average scale difference

S is the measured sensitivity
WS is the effective value of the sensitivity weight
The sensitivity should be checked at at least four points across the weighing range of
the balance.
5.6.6 Repeatability of reading
The repeatability of reading can be checked by loading the balance and performing a
series of releases calculating a rest point for each series. Apart from arrestment the
balance remains undisturbed between releases making this measurement relatively
insensitive to the user. This makes this test useful in providing a temporal
measurement of the balance performance which is insensitive to external influences.
5.6.7 Repeatability of measurement
This provides a more accurate assessment of the balance's performance "in use". A
series of repeated weighings are performed using the balance in its normal
comparison mode (double weighing or substitution). The series of measured mass
differences can be statistically analysed to give a measure of the balance performance.
15
The average measured mass difference can also be compared with the certified mass
difference between the weights taking into account the uncertainties on these certified
values.
5.7 Assessment of mechanical single pan balances
Single-pan mechanical balances have generally been replaced by electronic balances,

which often offer better resolution and are easier to use. The range of tests that should
be performed on these balances is only briefly described.
5.7.1 Visual inspection and mechanical check
§ Check knife-edges and planes

§ Ensure the released balance does not foul
§ Adjust the zero of the balance
5.7.2 Drift
This will give an indication of the balance's stability and its sensitivity to changes in
environmental conditions. It will also give an indication of how long the balance
should be left to give a stable reading (the effectiveness of the damping). If the drift is
linear it may be eliminated by the use of a suitable symmetrical weighing method (eg.
ABA weighing).
5.7.3 Calibration of internal weights
Ideally the weights built into the balance should be removed and calibrated externally.
If this is not possible they can be left in the balance and calibrated by dialling them up
in combinations.
5.7.4 Effect of off centre loading (eccentricity)
The effect of off centre loading on the balance reading is assessed using a weight of
nominal value equal to about half the balance capacity (to avoid mechanical damage
to the balance). The weight is placed at the extremes of the pan and the results
compared with the reading when the weight is placed centrally. A significant
difference between the reading at the centre and extremes of the pan may indicate the
balance is fouling. This test is identical to that described in more detail for a single
pan electronic balance.
5.7.5 Hysteresis
This is checked by taking readings of increasing and decreasing load. Significant

difference may indicate mechanical problems within the balance. This test is identical
to that described in more detail for a single pan electronic balance.
16
5.7.6 Scale sensitivity (scale value)
This is checked by placing a weight on the pan equal to the full range value of the
(optical) scale. For most single-pan mechanical balances adjustment of the scale
sensitivity is possible. This test is identical to that described in more detail for a single
pan electronic balance.
5.7.7 Scale linearity
The (optical) scale should be checked for linearity at regular intervals along its range
using a calibrated set of (fractional) weights. This test is identical to that described in
more detail for a single pan electronic balance.
The repeatability of reading should be assessed by loading the balance (usually to

maximum load) and taking a number of readings. The balance should be arrested and
released between readings without otherwise disturbing it. This provides a measure of
the balance performance, independent of external variables, and can thus be used as a
repeatable check of the balance performance over time.
This represents a series of actual comparisons using the balance in its normal
weighing mode. Statistical analysis of the repeatability of measurement provides a
practical assessment of the balance performance under normal weighing conditions.
5.8 Assessment of electronic balances
This type of balance is by far the most

widely used and the simple principle of
operation means that a full assessment takes
the form of relatively few tests.
As with the other types of balance the

manner in which the balance is assessed will
depend on the way it is to be used in
practice. Before starting the assessment of an
electronic balance it should have been left
switched on for a least an hour and
preferably overnight (check the
manufacturer's literature for details). Where
appropriate the balance calibration facility
should be used before starting the
assessment.
Figure 8: Single-pan electronic balance
17
5.8.1 Hysteresis
Hysteresis occurs when, for a given weight, the balance displays a different reading
depending on whether the load is increasing or decreasing. It can be checked by:
§ Incrementally increasing the load by adding weights one at a time, noting the
balance reading after each addition
§ Remove the weights in reverse order, noting the reading after each removal
§ Compare the readings for the same load being applied to the balance
Significant deviation between the increasing and decreasing readings indicates a
poorly adjusted or dirty balance.
5.8.2 Effect of off centre loading (eccentricity)
Electronic balances are generally more sensitive to off centre loading than either of
the other types, as there is often little de-coupling between the balance pan and the
sensing element. This effect is assessed by placing a weight, of value about half the
capacity of the balance, at the extremes of the pan. Compare the results with the
reading when the weight is placed centrally. Typical results are shown below.
Reading sequence Centre Right Front Left Back

C>R>F>L>B 0.86 g 0.81 g 0.83 g 0.87 g 0.82 g
B>L>F>R>C 0.85 g 0.87 g 0.80 g 0.86 g 0.83 g
Mean reading 0.855 g 0.84 g 0.815 g 0.865 g 0.825 g
Reading relative to centre -0.015 g -0.04 g +0.01 g -0.03 g
The weight is placed at each position on the pan twice and the weighing scheme is
symmetrical with time so any balance drift is calculated out.
5.8.3 Scale error and linearity
Scale error and linearity can be checked with a suitable set of calibrated mass
standards. Measurements should be made at about ten equal steps across the range of
the balance. If the balance is usually used at a particular load or an area of the scale is
thought to be non-linear, additional linearity checks can be performed at that load.
The repeatability of an electronic top pan balance can be assessed by repeated

application of a mass standard. A set of ten weight applications should be measured at
at least two loads (usually half and full load) and at any other load where the balance
will be normally used.
18
A more relevant test for a balance that is always used as a comparator is to carry out
repeated comparisons of a pair of weights using the same weighing scheme that would
normally be used on the balance.
Generally the following data should be calculated for a repeatability assessment:

§ Maximum difference between consecutive weighings
§ Maximum difference between any weighings
§ Standard deviation of the series (σn-1)
§ Standard deviation of the series accounting for drift
5.9 Periodicity of calibration
The period between balance calibrations depends on the intensity of use of the balance
and on the type of measurements which are being performed. In general a full balance
assessment should be undertaken annually and if any changes to the balance set up or
location have occurred. If a balance is fitted with an internal calibration facility, this
should be checked on a weekly basis or if changes are made to the balance set up.
6 Uncertainty in mass calibration

6.1 Introduction
When making measurements there is always an element of uncertainty in the result.

We cannot know ‘true’ values – there are limitations in our knowledge and in the
performance of the instruments we are using. Therefore a measurement is not
complete without an estimate of the doubt that surrounds it (the uncertainty) and the
confidence we have in that estimate. This chapter gives an introduction to the
terminology and the main sources of uncertainty in the calibration of a weight of
nominally 100 g, when using a comparator, plus a brief summary of the process for
calculating the uncertainty and associated confidence level. All calculations are
performed in accordance with the ISO Guide to the Expression of Uncertainty in
Measurement [8] also known as the GUM.
6.2 The measurement
Although perhaps an obvious point, before starting it is worth confirming precisely

what the measurements are aimed at determining. In this example it is the
conventional mass of a weight and the following need to be considered:
• Which measurements and calculations will be required to enable you to establish

the mass value and the uncertainty in its determination? For example, will you
need to determine air density?
• How many measurements do you need to take? (The more measurements you take
the more representative the mean (average) value becomes, although there is a
reduction in benefit as the number of measurements increase beyond a certain
point. Ten measurements is a common choice and statistically valid but not always
19
practical – eg for economic reasons. One way of dealing with this problem is to
use some data from previous measurements to determine the performance of the
balance and then take a smaller number of measurements for the particular
calibration in hand. This is dealt with in paragraph 6.3.3 under repeatability.)
• How to take your measurements and calculate a mass value.
6.3 The uncertainty budget
Once you have determined the mass value you are ready to start calculating the
uncertainty in the measurement. The following table is a typical layout for an
uncertainty budget. It can be in the form of a computer spreadsheet – to make
repeated calculations easier - or it can be a paper table completed by hand using a
calculator. Each column in the table is dealt with separately below.
Symbol Source of Value Probability Divisor Sensitivity Standard vi or veff

uncertainty distribution coefficient uncertainty
6.3.1 Symbol
The symbol used in to denote the input quantity or the influence factor.
6.3.2 Sources of uncertainty
The sources of the uncertainty are dependent on the measurement process and
equation used - as defined in your procedures and by your laboratory environment.
Here we consider the most common sources.
Each of the input quantities in the measurement equation used to calculate the mass
value has an uncertainty. For example if the equation you are using is:
Wx = Ws + ∆W + Ab
where Wx is the unknown mass

Ws is the mass of the standard
∆W is the difference in the balance readings
Ab is the correction for air buoyancy
there is an uncertainty associated with each of the input quantities Ws, ∆W and
Ab. The uncertainty in ∆W depends on uncertainty due to other influence factors:
• Rounding errors in the comparator readings (δId)

• The uncertainty due to less-than-perfect repeatability of the readings (WR)
• The value associated with comparator linearity (δC)
Another influence factor to be considered is
• The drift of the standard with time (Ds)
20
Symbol Source of uncertainty Value Probability Divisor Sensitivity Standard vi or veff

distribution coefficient uncertainty
Ws Calibration of mass
standard
δC Comparator linearity
Ab Air buoyancy
Ds Uncorrected drift of the
standard
δId Digital rounding error
WR Repeatability
6.3.3 Values
The values associated with the sources of uncertainty are either measured, calculated
or come from a priori (previous) knowledge.
In our example:
Symbol Uncertainty source The value

Ws Calibration of mass The uncertainty in the mass standard is taken from its calibration
standard certificate.
δC Comparator linearity Estimated from previous measurements according to your
procedures.
Ab Air buoyancy A calculated uncertainty based on the air buoyancy correction
equation ie (Vs – Vx)(ρa – 1.2) (see paragraph 6.4) In this example
the volumes have not been measured and the uncertainty in the
volume difference is based on the values given in the OIML
recommendation R111 (if the volumes are measured the uncertainty
can be less) but the value of air density has been measured.
Ds Uncorrected drift of The uncertainty quoted on the mass standard’s calibration certificate
the standard will not include any contribution for drift in its mass value. The
evaluation of this effect is normally the responsibility of the weight’s
owner as they are best placed to evaluate how much its mass changes
between calibrations. Drift is usually determined by considering how
much a particular artefact has changed its mass value over a recent
period and extrapolating the figure to cover the period up to its next
calibration. In this example no previous calibration knowledge is
assumed and the uncertainty in the current mass value calibration is
also used to estimate the limits of drift.
δId Digital rounding error Each reading is subject to a rounding error. It is taken to be ± half the
resolution of the comparator. Such errors occur in the comparator
reading of the standard mass and the unknown mass.
WR Repeatability This is an uncertainty component which is a measure of the ‘spread’
of the repeated readings. It is estimated by determining the
experimental standard deviation of the mean (see Further reading).
In this example a previous evaluation of repeatability of the
measurement process (from ten comparisons between a mass
standard and an unknown mass) were used to establish a standard
deviation of 0.00017 mg which was then divided by √n, where n is
the number of readings in the current measurement – in this case
three).
6.3.4 Probability distributions
A probability distribution is a statistical description of how results behave. There are

three distributions commonly used in mass uncertainty budgets: normal (sometimes
21
called Gaussian), rectangular (sometimes called uniform) and triangular. The

following graphs illustrate these distributions where the zero value on the horizontal
axis represents the mean value of a number of readings, the vertical axis is the
probability of a particular value occurring and the broken lines represent plus and
minus one standard deviation (k=1) - encompassing approximately 68% of the
measurement values (ie 68% of the area under each curve).
6.3.4.1 Normal distribution

Normal distribution
This represents a group of measurements 0.45
where the values are more likely to fall 0.4
closer to the mean value than further 0.35
away from it. Repeated measurements 0.3
are an example of this type of 0.25

0.2
distribution. The graph shows normally
0.15
distributed data with a mean 0.1
value of zero and a standard deviation 0.05
of ±1. 0
-4 -2 0 2 4
6.3.4.2 Rectangular distribution
Rectangular distribution This represents a group of measurements

0.12 where the values are evenly spread
0.1
between two limits and never fall outside
these limits. An example is when using
0.08
an assumed air buoyancy correction (as
0.06 opposed to a measured or calculated
0.04 value). The graph shows a rectangular
0.02
distribution, again with a mean value of
zero, but limits of ±5. In this case one
0
-6 -4 -2 0 2 4 6 standard deviation is ±2.89.
6.3.4.3 Triangular distribution
Triangular distribution
This is the distribution you get when 0.25
adding two rectangular distributions.

0.2
An example is the uncertainty due to
rounding errors. The graph shows a 0.15
triangular distribution, again with a
mean value of zero and limits ±5. In 0.1
this case one standard

0.05
deviation is ±2.04.
0
-6 -4 -2 0 2 4 6
22
6.3.5 Divisor
In order to eventually sum all the individual input quantities they must be quoted with
the same confidence level. This is done by establishing a number by which the input
quantity uncertainty value is divided to convert it to one standard deviation and is
dependent on the distribution as shown in the table below.
Distribution Divisor
Normal 1 or 2
Rectangular √3
Triangular √6
The normal distribution has a divisor of either 1 or 2 depending on the confidence

level of the value quoted. For example on a certificate of calibration the uncertainty
might be quoted as ‘k = 1’ (~68%) or ‘k = 2’ (~95%) in which case the divisor is the
‘k’ number. Other values are sometimes used, for example k = 3.
6.3.6 Sensitivity coefficient (ci)
This is a multiplication factor which converts the uncertainty in the value of an input
quantity to a corresponding uncertainty in the output quantity (it sometimes has to
convert both quantity – such as temperature or pressure - to mass and also the right
units). In the example being discussed all the input quantities are already expressed in
the quantity mass and using the sub-unit milligram so the sensitivity coefficient is 1.
An explanation of the air buoyancy correction calculation is in paragraph 6.4.
6.3.7 Standard uncertainty in units of measurand, ui(Wi)
In order to add all the components together we need them in the same units and the
values for this column are simply calculated from
Value ÷ Divisor x Sensitivity coefficient (ci)
6.3.8 Degrees of freedom vi
The number of degrees of freedom is “…in general the number of terms in a sum
minus the number of constraints on the terms of the sum”[4]. Before considering this
further it is necessary to first appreciate that measurement uncertainties are considered
to fall into one of two categories, known as Type A and Type B.
• Type A uncertainties are those that are evaluated by statistical methods. For
example, uncertainty due to less-than-perfect repeatability of a measurement can
be reduced by calculating a mean value from several measurements.
• Type B uncertainties are evaluated by other means. They cannot be reduced by

taking more measurements – the uncertainty quoted on a certificate of calibration
cannot be made lower by repeatedly reading the certificate for example!
23
The degrees of freedom, vi, for individual uncertainty contributions are given by:
Type A vi = n-1 where n is the number of measurements used to evaluate the type A contribution
Type B vi is usually taken to be infinite
6.3.9 Adding it all up
6.3.9.1 Combined standard uncertainty u(Wx)

To obtain an uncertainty in Wx, the mass value of the unknown weight, the
components have to be added to obtain a combined standard uncertainty. As it is
unlikely that all the errors will have been at their maximum value in any one
measurement it is inappropriate to add them in arithmetically. The recognised way to
address this issue is to arithmetically add the squares of the standard uncertainties and
then take the square root of the result – this process is know variously as taking the
root sum of the squares (RSS) or quadrature summation.
In the example we are considering the summation is:
u (Wx) = 0.0250 2 + 0.0115 2 + 0.0216 2 + 0.0289 2 + 0.0020 2 + 0.00012
= 0.0454 mg
The resulting probability distribution will be a normal distribution unless one

rectangular distribution is much larger than the other components.
6.3.9.2 Effective degrees of freedom veff
In general the effective degrees of freedom, veff, will not need to be calculated if the
type A uncertainty is less than half of the combined standard uncertainty, there is only
one type A component and at least three measurements have been taken. Otherwise
the effective degrees of freedom will have to be calculated to ensure that the k-factor
of 2 will indeed give a confidence level of ~95%.
The effective degrees of freedom for the combined standard uncertainty will depend
on the magnitude of the degrees of freedom for the type A contributions in relation to
the type B. If the type B uncertainties are all taken to have infinite degrees of freedom
the relationship is shown using the simplified Welch-Satterthwaite equation:
u c4 (y )
v eff = where uc(y) is the combined standard uncertainty
æ u i4 (y ) ö ui(y) is the individual type A uncertainty contribution
çç ÷÷ vi is the degree of freedom in ui(y)
v
è i ø
therefore in our example:

0.0454 4 thus veff = 3.8E+11
v eff =
æ 0.00014 ö
çç ÷
è 9 ÷ø
24
In this example veff is a very large number which can be taken to be infinity. If this
value had been less than 100 a k-factor would have been calculated from a
distribution other than a normal distribution. More information about this can be
found in the further reading list but for our example the k-factor is two.
6.3.9.3 Expanded uncertainty
The expanded uncertainty U(Wx) is the combined standard uncertainty u(Wx)

multiplied by a k-factor which will give an uncertainty value with a confidence level
of approximately 95%, in this case 2.
The final uncertainty budget for our example is shown below.
Symbol Uncertainty source Value Probability Divisor Sensitivity Std. uncertainty vi or

±mg distribution coefficient ± mg veff
Ws Calibration of 0.0500 Normal 2 1 0.0250 ∞
standard weight
δC Comparator linearity 0.0200 Rectangular √3 1 0.0115 ∞
Ab Air buoyancy 0.0216 Normal 1 1 0.0216 ∞
Ds Uncorrected drift of 0.0500 Rectangular √3 1 0.0289 ∞
the standard
δId Digital rounding error 0.0050 Triangular √6 1 0.0020 ∞
WR Repeatability 0.0001 Normal 1 1 0.0001 9
u(Wx) Combined standard Normal 0.0454 >500
uncertainty
U(Wx) Expanded uncertainty Normal k=2 0.0908 >500
6.4 Air buoyancy uncertainty budget
In order to calculate the uncertainty in the air buoyancy correction for entry into the
main uncertainty budget an additional uncertainty budget has to be completed. Air
buoyancy (Ab) is dependent on the volumes of the weights and the air density with
the following relationship:
Ab = (Vs – Vx)(ρa – 1.2)
where (Vs - Vx) is the difference in volume between the standard and the unknown weight
(ρa - 1.2) is the difference between measured density of the air and the standard air density
There are two ways to calculate an uncertainty value for Ab, either working in relative
values or calculating the sensitivity coefficient directly. In our example the volumes
have not been measured so the value of (Vs - Vx) is taken to be the largest difference
possible according to the OIML recommendations [1] when comparing E2 and F1
weights – that is 1.3 cm3 with an uncertainty of ±1.3 cm3. This uncertainty is treated
as a rectangular distribution because the real value may lie anywhere between these
limits and thus the standard uncertainty (u(V)) is equal to ±(1.3 ÷ √3). In this example
the air density, ρa, has been measured as being 1.22 kg/m3 with an uncertainty of
±10% - thus the uncertainty in (ρa – 1.2), that is u(ρ), is ±0.012 kg/m3. Thus:
Ab = (1.3)(1.22 – 1.2) = 0.026
25
6.4.1.1 Relative uncertainty
The relative uncertainty in the air buoyancy, u(Ab)/Ab, is given by:
2 2
u ( Ab) æ u (V ) ö æ u ( ρ ) ö
= çç ÷÷ + çç ÷÷
Ab è (Vx − Vs) ø è ( ρ a − 1.2) ø
inserting the values gives:
2
æ 1.3 ÷ 3 ö æ 0.012 ö 2
u ( Ab) = Ab çç ÷ +ç
÷ è 1.22 − 1.2 ÷ø
è 1 .3 ø
u ( Ab) = 0.0216
This method of calculation works well for an equation where the only operators are
multiplication or division.
6.4.1.2 Partial differentiation
The other method, partial differentiation, sounds more complicated but is actually
quite straightforward for this type of equation.
Using the same equation, Ab = (Vs - Vx)(ρa - 1.2), and the same values as above, we
calculate the sensitivity coefficients - the numbers by which values of u(V), expressed
here in cm3, and u(ρ), expressed here in kg/m3, should be multiplied to calculate their
effect on the output quantity expressed in grams.
The partial derivative of a simple equation, such as the one we are looking at, is
simply the multiplier for the term for which we wish to calculate the partial derivative.
For example, if our equation is A = B × C then the partial derivative of B is C and the
partial derivative C is B.
In our air density problem the partial derivative of (Vs - Vx) is (ρa - 1.2) and the
partial derivative of (ρa - 1.2) is (Vs - Vx). In the correct terminology this is
expressed as :
∂ (Ab)
= ( ρ a − 1.2) = 0.02
∂ ( Vs − Vx )
∂ (Ab)
= (Vs − Vx ) = 1.3
∂ ( ρ a − 1 .2 )
The values 0.02 and 1.3 are entered directly into the sensitivity coefficient column of
the uncertainty budget and the remaining calculations are the same as for the main
uncertainty budget.
26
Symbol Source of uncertainty Value Probability Divisor Sensitivity Std. uncertainty vi or

±mg distribution coefficient ± mg veff
Vs – Vx Difference in volumes 1.3 Rectangular √3 0.02 0.0150 ∞
ρa – 1.2 Diff. in air density 0.012 Normal 1 1.3 0.0156 ∞
Combined standard uncertainty Normal 0.0216 ∞
The combined standard uncertainty is the same as in the previous method.
Partial differentiation is more difficult when the equation is more complex.
6.5 Reporting the results
In general calculations should not be rounded until the final result is calculated. The
uncertainty should be quoted to 2 significant figures and the result quoted to the same
number of decimal places. In the worked example the result would be reported in the
form:
100.000 71 g ±0.10 mg
and would be accompanied by a statement explaining how the uncertainty and its
confidence level are calculated such as:
The reported expanded uncertainty is based on a standard uncertainty

multiplied by a coverage factor k = 2, providing a level of confidence of
approximately 95%. The uncertainty evaluation has been carried out in
accordance with UKAS requirements.
This statement was taken from the UKAS document [5].
The uncertainty has been rounded to 0.10 mg; uncertainties should always be rounded
up rather than down to ensure that the value remains within the 95% confidence limit.
27
7 References
[1] International Organisation of Legal Metrology (OIML), International
Recommendation No 111:1994 Weights of classes E1, E2, F1, M1, M2, M3.
[2] Giacomo P., Metrologia, 18, 33-40, 1982
[3] Davis R.S., Metrologia, 29, 67-70, 1992
[4] BIPM, IEC, IFCC, ISO, IUPAC, IUPAP, OIML. Guide to the Expression of
Uncertainty in Measurement, International Organisation for Standardisation,
Geneva. ISBN 92-67-10188-9
[5] UKAS publication M3003. The Expression of Uncertainty and Confidence in
Measurement, 1997.
[6] Clarkson M.T., May B.J Metrologia, 38(2), 161-171, 2001
[7] Guide to the Expression of Uncertainty in Measurement, International
Organization for Standardization, (Geneva, Switzerland), 1993
28
5369/AAR/1K/0704
Recommended sale price: £25.00

Measurement Good Practice Guide: The Measurement of Mass and Weight

Uploaded by

Copyright:

Available Formats

Measurement Good Practice Guide: The Measurement of Mass and Weight

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Measurement Good Practice Guide: The Measurement of Mass and Weight

Uploaded by

Copyright:

Available Formats

No.

S Davidson, M Perkin, M Buckley

The Measurement of Mass and Weight

Stuart Davidson, Michael Perkin

National Physical Laboratory

“The kilogram (kg) is the unit of mass; it is equal to the

The use of appropriate mass standards

Figure 1: The UK national standard kilogram

Weights should be made of a material that is chemically inert, non-magnetic, hard

Class E1 and E2 weights

2.2 Surface condition

2.4 Weight cleaning

history. Where cleaning is unavoidable, it is recommended that weights should be

After cleaning it is necessary to allow weights to stabilise before calibrating them.

2.5 Weight handling

a) touched with bare hands

Tweezers, forceps or other specialist lifting devices should be used whenever

Figure 3: Good practice in weight handling

2.6 Clean surfaces

2.7 Weight storage

Weights should be stored in a manner that protects them from damage or

3.1 Direct reading measurements

As with any form of mass measurement it is

3.2 Weighing by differences

This technique is particularly common in analytical chemistry. In general it involves

3.3 Substitution weighing

It is important to follow a consistent nomenclature when carrying out weighings and

If the test-weight is of an unusual nominal value it will be necessary to use a standard

3.4 ABA… calibration

This may be generalised for an ABA comparison involving n applications of weight B

3.5 ABBA calibration

It is important in this weighing method that weighings B1 and B2 must be independent

3.6 Cyclic weighing

Comparison Measured Corrected Using S1 value Using S2 value Measured

3.7 Weighing by sub-division

This is most easily illustrated by considering how values would be assigned to a

The sub-division weighing scheme has the following advantages:

It has the following disadvantages:

b) it necessitates placing groups of weights on balance pans (this can cause

3.8 Periodicity of Calibration

3.9 Scale errors and the use of make-weights

If a make-weight of value Mw is added to load A during a comparison the weighing

so in order to calculate the value of A from B it is necessary to subtract the value of

4 Air density measurement and buoyancy correction

4.1 True mass

4.2 Conventional mass

where: Mc is the conventional mass of the artefact in grams

4.3 Buoyancy correction

Where: BC is the buoyancy correction to be applied

The calculated buoyancy correction should be applied in the form:

Where: Mt n is the true mass of artefact n

When calibrating standard weights comparisons are normally performed on a

4.4 The application of buoyancy corrections

Material compared with Buoyancy correction (ppm)

Table 1: Buoyancy Corrections when Comparing Dissimilar Materials in Air

The table shows that even when

When comparing weights of dissimilar

Figure 5: stainless steel and platinum kilograms

4.5 Determination of air density from parametric measurements