-
Whi
te P
aper Influences on Mass Calibration
Minimizing uncertainties
Contents
1 Introduction 2
2 Balances vs. comparators 2
3 Thermal stability 3
4 Vibrations 3
5 Air drafts 3
6 Magnetism 4
7 Air buoyancy 4
8 Density measurement 6
9 Conclusion 6
This white paper will be of interest to anyone involved in mass
calibration activities, from National Metrology Institutes (NMI's)
to private calibration laboratories of any size.
In mass calibration, making accurate measurements and minimizing
the uncertainty of such measurements is paramount. However, the
extent to which physical influences can influence daily mass
calibration are often under-estimated. This can lead to
uncertainties which do not fulfil the requirements of the applied
regulations. By analyzing multiple influence sources, the main
driver for performance and thus correction steps to improve the
combined measurement uncertainty can be defined.
-
2 White Paper METTLER TOLEDO
Influ
ence
s on
Mas
s Ca
libra
tion 1. Introduction
In mass calibration, minimizing uncertainties is critical and
multiple physical effects can affect the measurement. The main
influencing factors are the comparator balance itself, thermal
stability, vibrations, air drafts and magnetism. The impact of
static factors can be reduced by making appropriate choices with
regards to equip-ment and environment.The variable influence of air
buoyancy effects must be corrected, if limitations are exceeded.
However, it is recommended to apply corrections at all times, to
ensure the real conventional mass is reported and not the
uncorrected conventional mass, which could generate balance
calibration errors at a later stage.To enable these calculations to
be made, air and artifacts densities must be known to sufficient
accuracy, and then applied in mathematical calculations to correct
for these physical effects and thus generate the mass in
conventional mass values. To specify the ideal selection of
equipment, an uncertainty analysis should be carried out. This
analysis should determine the main influencing source of
uncertainties and therefore the correction required.
2. Balances vs. ComparatorsBalances and comparators are
influenced by the same physical effects, but these effects are
often undetectable on a typical balance. This is because the
influence of the physical effects is often lower than the
resolution of the balance. However, with comparators, the smallest
effects can be detected due to their very high resolution and
stability. Examples of manual and robotic mass comparators (figure
1).
Figure 1: Examples of manual and robotic mass com-parators. From
left to right: XPE56C (52 g / 0.1 g); XPE26003LC (26.1 kg / 1 mg);
a5XL (6.1 g / 0.1 g)
-
3White Paper METTLER TOLEDO
3. Thermal StabilityConvection caused by temperature differences
between the artifact and the environment, such as sun irradiation
or sub-optimal room heating, generates force changes applied to the
artifacts. Therefore, this force influences the mass readings,
which results in instability of the reading or incorrect
measurements.The extent of the convection influence can exceed the
limitation of regulations resulting in invalid mass values being
used in subsequent calculations. By enhancing the temperature
stability of artifacts and environment, these effects can be
reduced.
4. VibrationsVibrations generated by a multitude of external
sources can increase the stabilization times of high resolution
balances or comparators and prevent reliable readings from being
obtained. By installing sensitive instruments in an appropriate
environment, for example on rigid stone tables and in low vibration
areas of the building, such as the basement, comparator performance
is significantly improved and stabilization times are reduced.
Specific settings on the balance models can also be adapted to
improve the performance.
5. Air DraftsAir, with its significant matter of 1.2 kg/m3,
generates forces by interfering with surfaces. These forces applied
on a weighing pan generate mass changes of a significant extent.As
air drafts do not have a continuous stable flow, the variation of
forces generates instability in the balance readings. To reduce
this influence, mechanical protection of the balance weighing pan
is typically employed so that the air draft is decreased to a
non-influencing extent. The occur-rence of the air currents within
a laboratory should be considered such that no direct air flow is
directed on top of a high accuracy balance. It is preferable to
have an air flow pillar in front of the balance to reduce the
influence of the operators body heat. Balances should not be
installed in
-
4 White Paper METTLER TOLEDO
Influ
ence
s on
Mas
s Ca
libra
tion close proximity to swinging doors, as air turbulences and
drafts will have
an influence on the measurement. Sliding doors offer the
advantage of preventing air pulses and drafts. In addition, it is
recommended to avoid placing balances or sensitive measuring
devices in an area of the labora-tory which experiences a high
volume of foot traffic, in other words many people passing by
frequently.
6. MagnetismArtifacts containing ferrites may interact with
other magnetic sources. To eliminate this risk, it is recommended
to install balances and compara-tors in locations without ferrites
in the near vicinity. In weights handling, ferrites are also not
allowed, as interaction is very likely and cannot be eliminated and
damage of weights is possible. Minimum distances to fer-rites are
recommended to reduce the influences of magnetic interference on
mass measurements.
7. Air BuoyancyIn mass calibration, the effects of air buoyancy
must be also considered. In addition, the assumption that different
materials have different densi-ties must/should be made, as the
tested artifacts are commonly made of stainless steel or lower
grade steels. Different densities at identical mass lead to
different weights volumes, which are affected differently by the
ambient air.
For weight calibration, the mass has to be defined according to
the weights density convention of 8000 kg/m3, air density of 1.2
kg/m3, and temperature of 20C.
Weights densities differing from the conventional values in
combination with air density differing from the conventional value
of 1.2 kg/m3 must be corrected, to result in the conventional mass
(see Fig. 2).
-
5White Paper METTLER TOLEDO
7.1. Factors influencing air buoyancy
Air density is one of the factors which influence air buoyancy.
Air density reduces with altitude by 1.2 % per 100 m. This results
in the maximum air density deviation of 10% being exceeded at 830 m
above sea level. A weight's conventional mass is directly affected
by this air density change, if the weights density deviates from
the conventional value of 8000 kg/m3. The more the density of the
weight deviates from 8000 kg/m3, the stron-ger the influence on the
air buoyancy correction.
7.2. Extent of influences
According OIML R111, the air buoyancy influence shall not be
more than 25% of the maximum permissible error. [2]Two weights are
compared at an altitude of 840 m according to OIML R111: E1 1 Kg
weight, density 8048.25 kg/m3 E2 1 Kg weight, density 7840.65
kg/m3
This altitude results in an air density of approximately 1.070
kg/m3, and an air buoyancy effect of 0.425 mg, which is 26% of the
maximum permissible error (MPE) of 1.6 mg of the E2 1kg weight.As
this exceeds the 25% limit imposed by OIML R111, a correction must
be applied.
Commonly, it is strongly recommended to apply air buoyancy
correction each time, as influences are significant. This also
maintains standardized procedures for intercomparable measurement
results.
Figure 2: Effect of different densities on weight volume and air
buoyancy
Density Volume Buoyancy
113.63 cm38800 kg/m3
7200 kg/m3 138.88 cm3
Displaced air
volume
Displaced air
volume
Figure 3: Example calculation of the air buoyancy influence
7.3. Air buoyancy uncertainties
To calculate the air buoyancy effects, the air density and
artifacts densities must be known at high accuracy to enable low
measurement uncertainties.The accuracy of the artifacts density
contributes more to the air buoyancy correction uncertainty than
the ac-curacy of the air density. Therefore, it is necessary to
know the density of artifacts accurately to allow an air buoyancy
correction with sufficient accuracy. The combination of the
uncertainty influences must be calculated individually for each
artifact to find the specific uncertainties.
-
6 White Paper METTLER TOLEDO
Influ
ence
s on
Mas
s Ca
libra
tion 8. Density Measurement
To enable low air buoyancy correction uncertainties, the
densities of air and weights must be known to a certain level of
accuracy. For low accuracy mass calibrations in lower altitudes,
the density is estimated according to materials or defined by the
supplier material specifications. The air density is measured with
medium accuracy climate sensors. For higher accuracy measurements,
or where the combination of air density and weights densities
becomes more significant, the density of the weight must be
measured as well, to reduce the overall air buoyancy correc-tion
uncertainty. Weights density is measured according OIML R111 [2]
methods A3 or A2, both based on buoy-ancy force measurement of
volumes in known liquids. The weights are immersed in well-known
liquids and the buoyancy force is measured and deducted from the
mass. The mass difference in combination with the liquid density
results in the volume and density of the weight. With this
measurement technology, density is measured highly accurate and
reducing the air buoyancy correc-tion uncertainty to lowest levels.
To even improve the accuracy, the buoyancy force is measured
differentially to mass references to reduce the mass measurement
uncertainty.
9. ConclusionIn mass calibration, various external influences
can have an effect on weighing performance and therefore accu-racy
of measurement. This can lead to a disruption of measurement
quality of an unacceptable extent. In order to achieve reliable and
accurate mass measurement, environmental influences should be
minimized. To fulfill the requirement of all mass values being
stated in conventional mass, environmental influences and physical
effects must be corrected with appropriate measurement technology.
The weights densities and the air density must be known to a
sufficient level of accuracy to enable lower uncertainties in the
uncertainty of air buoyancy correction. This improves the overall
measurement accuracy as a result. By taking into consideration all
the physical influences which can affect weighing accuracy,
improving laboratory climatic conditions and applying air buoyancy
correction, the mass calibration capabilities of a specific
labo-ratory can be improved. This leads to generation of more
reliable calibration results, and full compliance with industry
regulations.
References
1. "Weights of classes E1, E2, F1, F2, M1, M12, M2, M23 and M3",
Weights, T. 9. Paris, France (2004).2. OIML R111 - TC 9/SC 3,
(2004)
-
7White Paper METTLER TOLEDO
DisclaimerMETTLER TOLEDO provides this White Paper as a service
to its customers. In reading or making any use of this document,
you acknowledge and agree to the following:This document may
contain inaccuracies and errors of both a substantive and/or
typographical nature. METTLER TOLEDO does not guarantee the
accuracy or completeness of the information or the reliability of
any advice, opinion or statement in this document. If you rely on
the information or any advice, opinion or statement, you are doing
so at your sole risk. METTLER TOLEDO does not guarantee that this
document or its contents are accurate, complete, reliable,
truthful, current or error-free.METTLER TOLEDO will not be liable
for any decision made or action taken by you or others in reliance
on the in-formation in this document. METTLER TOLEDO and its
affiliates are not liable for any Damages based on claims arising
out of or in connection with your use of this document.METTLER
TOLEDO DOES NOT ASSUME ANY RESPONSIBILITY OR RISK FOR YOUR USE OF
THE INFORMATION PROVIDED IN THIS DOCUMENT. THIS INFORMATION IS
PROVIDED WITHOUT ANY REPRESENTATIONS, ENDORSE-MENTS, OR WARRANTIES
OF ANY KIND WHATSOEVER, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT
NOT LIM-ITED TO, ANY WARRANTIES OF TITLE OR ACCURACY AND ANY
IMPLIED WARRANTIES OF MERCHANTABILITY, FIT-NESS FOR A PARTICULAR
PURPOSE, OR NON-INFRINGEMENT, WITH THE SOLE EXCEPTION BEING
WARRANTIES (IF ANY) WHICH CANNOT BE EXPRESSLY EXCLUDED UNDER
APPLICABLE LAW. In no event will METTLER TOLEDO or its affiliates
be liable for any Damages, even if METTLER TOLEDO is aware of the
possibility of such Damages, arising in connection with the
information provided herein. Damages includes but is not limited to
all losses and all direct, indirect, incidental, special,
consequential and punitive damages arising under a contract, tort
or other theory of liability (including reasonable legal and
accounting fees and expenses).
No part of this publication may be reproduced or distributed for
any purpose without written permission from METTLER TOLEDO.
2014 METTLER TOLEDO. All rights reserved.
-
Mettler-Toledo AGLaboratory WeighingIm Langacher 44P.O. Box
LabTec, GD712CH-8606 Greifensee, Switzerland
Subject to technical changes12/2014 Mettler-Toledo GmbHPrinted
in Switzerland 30228568
www.mt.comFor more information