boletín informativo del sistema interamericano de metrología

boletín informativo del sistema interamericano de metrología

INFO IM
Enero 2008 / January 2008
BOLETÍN INFORMATIVO DEL
SISTEMA INTERAMERICANO DE METROLOGÍA - OEA
INFORMATIVE BULLETIN OF
THE INTERAMERICAN METROLOGY SYSTEM - OAS
S
AMERICANO
S
A
ANIZ CIÓN
G
R
O
D
LOS ESTA
D
E
IM
SISTEMA
INTERAMERICANO
DE METROLOGÍA
O
INFOSIM
ÍNDICE
CONTENTS
Introducción / Introduction
Humberto Brandi....................................................................………...……5
Metrología legal en la Argentina
Héctor Laiz.....................................................………………..……….……..7
Evaluation of stability in ethanol in
water certified reference material measurement uncertainty
under transport and storage conditions
Vanderléa de Souza, Janaina Marques R. Caixeiro,
Raquel D. C. Cunha, Laura A. N. Valente,
Renata M. H. Borges, Rafael A. L. da Silva……………...…………….…..10
Control of prepackages
Eberhard Seiler……………………………………………………….............15
Del software de los instrumentos de medición
sujetos a aprobación de modelo ¿Autenticarlo o validarlo?
C. Cajica Gómez.........…………………………………………………….….16
Comparison of methods for the weighing test in
calibration of high capacity non-automatic weighing intsruments
Luis O. Becerra, Eduardo González, Félix Pezet,
José Revuelta M. José Revuelta R. Sylvia Maeda........................…..….....18
NOTI-SIM…………………………………………………………………….....26
4
INTRODUCTION
INTRODUCCIÓN
Whenever a conflict of interests arises due to differences between fair and unfair
measures, legal metrology comes into place in order to solve controversies and
to contribute to the harmonization of the parties involved in the measurement.
Legal metrology helps to defend the citizens' interests by means of verifying the
necessary measurements to acquire the basic goods, fuel, energy, and preserve
the life, health and safety of the persons, and the environment as well. These are
the reasons with we have decided to devote this issue of INFOSIM, the SIM
Bulletin, to legal metrology.
This issue includes papers related to:
·
activities developed to apply and update some elements through the Instituto
Nacional de Tecnología Industrial -INTI- as the authority responsible for legal
metrology in Argentina;
·
the procedure to control of the stability characteristics of reference materials
required for testing ethanol contents in the breath of vehicle drivers in Brazil;
·
some easy actions to determine the contents in prepackaged products out of
the sight of the consumer, to verify the statements in the labels in economies
where no control of net contents of prepackaged products is in place;
·
precise the difference between authentication and validation of software used
in measuring instruments under model approval; and,
·
a comparative study of a number of calibration methods of large scales.
As an already necessary section, NOTISIM section provides information about
SIM activities.
We deeply appreciate the institutes that could contribute in providing the material
for this issue, and also to those who responded our call but could not deliver
information this time. The original language of the paper has been respected.
Those papers which had been previously been published somewhere else, have
the corresponding permission granted by the first publishers. We thank specially
IMEKO Secretariat for it.
We expect that this issue of INFOSIM promote the full development of legal
metrology in the SIM region, as a means to obtain fair measurements to protect
the interests of our societies, which we are part of and we serve to.
Humberto Brandi
President of SIM
5
64
METROLOGÍA LEGAL EN LA ARGENTINA
Héctor Laiz
Instituto Nacional de Tecnología Industrial (INTI), Argentina
[email protected]
RESUMEN
Este trabajo presenta la evolución reciente de la Metrología Legal en la Argentina, donde los cambios operados
fortalecieron la intervención del Instituto Nacional de Metrología (INTI) en el ámbito regulado. Se presentan las
acciones emprendidas para llevar los controles de los instrumentos reglamentados a todo el territorio y para
incorporar al ámbito regulado los instrumentos con impacto para la sociedad y el Estado. Se muestran cuales son
los desafíos tecnológicos a enfrentar.
1. INTRODUCCIÓN
En el año 2003, el Presidente de la Nación emitió
el Decreto 788 que modificó esta distribución de
funciones, asignando al INTI, además de sus
históricas tareas de metrología científica e
industria, la ejecución de todas las tareas técnicas
necesarias en Metrología Legal. La Tabla I resume
la división actual de funciones entre el INTI y la
Secretaría de Comercio de la Nación.
A la hora de analizar el diseño de un sistema
nacional de control de las mediciones en ámbitos
críticos, como el comercio, la salud o el ambiente,
surgen recurrentemente
interrogantes
polémicos:
- ¿cuál debe ser el grado de intervención del
Estado?
- ¿la supervisión estatal debe estar en manos de
un organismo administrativo o tecnológico?
- ¿debe aceptarse la autodeclaración del
fabricante como base de las aprobaciones?
- ¿la metrología legal es una actividad
técnicamente menor o demanda investigación
científica en metrología?
- ¿cuál debe ser el rol del Instituto Nacional de
Metrología?
La evolución reciente de la Metrología Legal en la
Argentina, aporta un ejemplo de abordaje a estas
cuestiones, en especial para países de un grado
de desarrollo intermedio.
3.EVOLUCIÓN DEL SISTEMA
2.1. Situación en el 2003
Tal como se expuso en el punto anterior, la
intervención directa del INTI en metrología legal
se inicia al fines del 2003. El sistema, en ese
momento, presentaba los siguientes problemas:
- pocos instrumentos reglamentados
- instrumentos aprobados con serias deficiencias
técnicas, como consecuencias del sistema de
aprobación basada en la declaración del
fabricante.
- casi nula supervisión de instrumentos en uso
(verificaciones periódicas y vigilancias de uso),
como consecuencia de los escasos recursos
técnicos del organismo a cargo hasta ese
momento.
2. MARCO LEGAL
La Metrología está regulada en la Argentina por la
Ley 19511 del año 1972. Hasta el año 2003 esta
Ley estaba reglamentada por dos Decretos, el
1157/72 y el 829/94. En el primero se asignaba la
responsabilidad de ejecución de la Metrología a
dos organismos: el INTI, en cuanto a la metrología
científica e industrial, es decir el mantenimiento y
diseminación de los Patrones Nacionales de
Medida y la Secretaría de Comercio de la Nación
en cuanto a las actividades concernientes a la
reglamentación de Instrumentos, aprobación de
modelo y verificaciones. El posterior Decreto 829
de 1994, estableció que las aprobaciones de
modelo y las verificaciones primitivas podían
aprobarse por declaración jurada del fabricante.
Cambios operados
Al iniciar las actividades del INTI en el área se
definieron tres objetivos:
a. extender la verificación periódica de
instrumentos reglamentados a todo el territorio
nacional.
b. establecer un sistema de aprobación de
modelos con la rigurosidad técnica necesaria
c. modernizar los reglamentos existentes y
elaborar los reglamentos faltantes y
demandados por la sociedad o el Estado
7
4. DESPLIEGUE DE LAS VERIFICACIONES
Tal como se señaló, se iniciaron las actividades en
el 2004 partiendo de una situación prácticamente
inexistente de verificaciones periódicas. La
estrategia adoptada se adaptó a las
particularidades de cada sector. Como ejemplos,
tomaremos los casos de balanzas de alta
capacidad, surtidores de combustibles líquidos y
tanques de almacenamiento.
4.1 Verificación de Balanzas
Se incorporaron dos camiones y las pesas
necesarias
para equipar 4 camiones
calibradores. Adicionalmente, se incorporaron al
sistema operadores privados, que actúan bajo la
supervisión del INTI. Su cantidad y requisitos
pueden verse en [1]. Estos prestadores actúan
siempre con la presencia de verificadores del
INTI. En al año 2006 se realizó la verificación
periódica de 3150 balanzas de alta capacidad, y
se inhabilitaron otras 240 en vigilancias de uso. La
Fig. 1 muestra la verificación de una balanza en un
establecimiento minero en la Provincia del Salta a
4200 m sobre el nivel del mar.
Fig. 2 – Sistema diseñado en INTI para verificar
surtidores de combustibles líquidos.
4.3 Verificación de tanques
La calibración de tanques de almacenamiento
representa un caso paradigmático. A pesar de
tener sus mediciones un alto impacto en la
determinación de las exportaciones argentinas,
no se encontraban dentro del ámbito de la
metrología legal. La Fig. 3 muestra a técnicos del
INTI realizando estas calibraciones. En el año
2006, se calibraron 337 de estos tanques.
Fig. 3 - Calibración de tanques de almacenamiento de
líquidos
Fig. 1 - Un camión calibrador del INTI verificando una
balanza de pesar camiones en una mina de ácido
bórico en la Provincia de Salta, a 4200 m sobre el nivel
del mar.
5. ENSAYOS DE APROBACIÓN DE MODELO
La efectiva realización de los ensayos de
aprobación de modelo conlleva dos desafíos: la
actualización tecnológica para adaptarse a las
innovaciones en el diseño de los instrumentos, y
la optimización de la capacidad de respuesta del
laboratorio para evitar tiempos de esperas
inadmisibles. Si bien el INTI contaba ya con
experiencia en los ensayos involucrados, el nuevo
requisito de obligatoriedad produjo un incremento
de la demanda que obliga a un permanente
replanteo de la organización del trabajo y al
análisis de las inversiones necesarias. Como
parte de las inversiones y desarrollos efectuados
vinculados con las aprobaciones de modelo de
4.2 Verificación de Surtidores
En cuanto a la verificación de surtidores de
combustibles líquidos, las verificaciones se
iniciaron en el 2005, con 10 equipos propios
distribuidos por todo el país. La Fig. 2 muestras los
equipos desarrollados para esta tarea. En el 2006
se verificaron 35150 manqueras de expendio. La
complejidad creciente de los sistemas de fraude
electrónico motiva la constante capacitación de
los verificadores.
8
metrología legal se destacan una nueva cámara
de ensayos climáticos de gran volumen, el
desarrollo de la escala primaria de fuerza y la
construcción de una cámara semianecoica para
compatibilidad electromagnética de 19 m x12 m x
10 m (ver Fig. 4).
- caudalímetros de oleoductos
- humedímetros de granos
- medidores de emisiones vehiculares
- sistemas de pesaje en cinta
El reglamento de tanques de almacenamiento, ya
aprobado puede consultarse en [2]. El resto de los
reglamentos propuestos pueden consultarse en [3].
7. DESAFÍOS TECNOLÓGICOS
Las mediciones en el ámbito regulado presenta
crecientes desafíos tecnológicos, básicamente
por dos motivos: la necesidad de abracar nuevas
áreas, como las mediciones ambientales y la
propia evolución de los instrumentos de medición
que incorporan nuevas tecnologías. Entre estos
desafíos, que motivan el desarrollo de nuevas
líneas de trabajo en el INTI, podemos destacar:
·
Seguridad en el software
·
Seguridad de la información
·
Telemedición
·
Compatibilidad electromagnética
·
Medición de gas en grandes caudales
·
Medición de energía eléctrica en alta tensión
·
Contenido de contaminantes (por ejemplo, PCB,
pilas y baterías)
·
Composición de comodities
Fig. 4 - Construcción de la cámara semianecoica de
19m x 12 m x 10 m para ensayos de compatibilidad
electromagnética
6. ASPECTOS REGLAMENTARIOS
El ámbito reglamentado no incluía, en el 2003,
instrumentos de alto impacto social, como los
medidores de servicios públicos (eléctricos, gas,
agua) o los surtidores de gas natural vehicular,
combustible utilizado por el 40% del parque
automotor. Por otro lado, alguno de los
reglamentos vigentes necesitaban urgente
modernización, por ejemplo, el reglamento de
balanzas databa de 1980 y el de surtidores de
combustibles líquidos de 1932. Para revertir el
atraso reglamentario se creó un equipo de
expertos específicamente con este objetivo. Se
elaboraron las propuestas de reglamentos de:
- tanques de almacenamiento
- medidores de energía eléctrica
- medidores de gas domiciliario
- medidores de agua domiciliaria
- medidores de pulsos telefónicos
- medidores de gas de alto caudal
- esfigmomanómetros
- medidores de gas natural vehicular
- alcoholímetros
8. CONCLUSIONES
La baja presencia del Estado en el ámbito
regulado de la Metrología sumada al sistema de
autodeclaración del fabricante resultó, en el caso
argentino, en notables perjuicios para la
protección del consumidor y para el interés fiscal.
La intervención directa del Instituto Nacional de
Metrología resulta clave para la óptima
supervisión del sistema, dada la creciente
complejidad técnica necesaria.
La evolución de la tecnología de medición, motiva
necesidades de investigación que hacen
necesaria la cooperación regional entre los
Institutos de deben proveer las soluciones a estos
desafíos.
REFERENCIAS
[1] Listado de reparadores de balanzas auditados por
el INTI.
www.inti.gov.ar/metrologia/list_balanzas.htm
[2] Reglamento sobre Tanques Fijos de
Almacenamiento,
www.inti.gov.ar/metrologia/199.htm
[3] Nuevos reglamentos en elaboración,
www.inti.gov.ar/metrologia/reglamentos.htm
Se modernizaron, o se encuentran en proceso de
modernización los reglamentos de:
- balanzas de pesaje no automático
- surtidores de combustibles líquidos
Actualmente se encuentran en proceso de
elaboración los reglamentos de:
- celdas de carga
9
EVALUATION OF STABILITY OF ETHANOL IN WATER CERTIFIED
REFERENCE MATERIAL: MEASUREMENT UNCERTAINTY UNDER
TRANSPORT AND STORAGE CONDITIONS
Vanderléa de Souza1, Janaína Marques R. Caixeiro 1, Raquel D. C. Cunha 1 , Laura A. N.
Valente 1, Renata M. H. Borges 2, Rafael A. L. da Silva 1
1
INMETRO, Directorate of Scientific and Industrial Metrology, Chemical Metrology Division, Duque de Caxias, Brazil
2
INMETRO, General Coordination of Accreditation, Rio de Janeiro, Brazil
[email protected]
ABSTRACT
This study simulated the transport and storage conditions of ethanol in water certified reference material (CRM)
produced by the Chemical Metrology Division of Inmetro – DQUIM with the purpose to estimate the measurement
uncertainty related to stability. The short-term stability study was performed on five different concentrations in
terms of g ethanol/100g solution (%) of the ethanol in water CRM: 0.0509%, 0.0890%, 0.1145%, 0.3820% and
0.4960%, at the temperatures of 4 ºC and 60 ºC. On the other hand, the long-term stability study was developed
on four different concentrations: 0.0509%, 0.0890%, 0.1145% and 0.4960%. In this paper will be shown data from
the long- term stability study concerning 20 weeks.
The method used complies with ISO GUIDE 35, BCR-Guideline for Feasibility Studies and ISO GUIDE 34.
According to the statistical parameters used in both studies the stability of the ethanol in water CRM was
confirmed for all concentrations studied.
Keywords: certified reference material, stability, ethanol in water, uncertainty.
1. INTRODUCTION
respectively [2]. In the case of ethanol in water
solutions the estimation of the uncertainty related
to stability is extremely important as ethanol is
volatile and the impact of its volatility in regarding
to storage and transport conditions is crucial for
quality assurance of the material.
Stability and homogeneity are key characteristics
in the certification of reference materials, as they
impact the validity of the certified values and their
uncertainties. Hence, in the certification process,
the reference materials shall be submitted to
stability and homogeneity tests [1].
The short-term stability study was performed in
five different concentrations in g ethanol/100g
solution (%) of the ethanol in water CRM:
0,0509%, 0,0890%, 0,1145%, 0,3820% and
0,4960%. The follow-up of the short-term stability
was performed at the temperatures of 4ºC and
60ºC.
The long-term stability study was
developed in four different concentrations:
0,0509%, 0,0890%, 0,1145% and 0,4960%. In this
paper it is shown data from the long-term stability
study concerning 20 weeks. However, this studies
continue under development at DQUIM/
DIMCI/INMETRO.
ISO GUIDE 35 [2] establishes that the certification
of a reference material requires a detailed study of
all sources of uncertainty that affect the certified
value. Among these sources are the uncertainties
of characterization, homogeneity, transport and
storage.
The Ethanol in Water certified reference material
(CRM) is produced by the Chemical Metrology
Division of Inmetro - DQUIM to test breath alcohol
analyzers used to determine the content of alcohol
(ethanol) present in the expired air of vehicle
drivers.
All the procedures used in this study are complied
with ISO GUIDE 35, ISO GUIDE 34 and the BCR
document- Guideline for Feasibility Studies [2-5].
The uncertainty concerning the storage and
transport of a CRM is estimated based on a
stability study of long-term and short-term,
* This paper was originally presented at the IMEKO XVIII World Congress, Rio de Janeiro, Brazil, September 2006.
10
2. PURPOSE
was performed, the respective “time-zero”
analyses (the first analyses before initiating the
cycles) were also performed. The internal
standard solution was added to the calibration
solutions and new calibration solutions and a new
internal standard solution were prepared.
The aim of this work is to demonstrate the
estimation of uncertainty related to the transport
(short-term stabilility study) and storage (longterm stability study) of the ethanol in water certified
reference material in five different concentrations.
In the second cycle the analyses were performed
once a month during a period of five months (20
weeks). All material was stored and protected
against light at the temperature range from (20
0,3) °C to (25 0,3) °C.
The short-term stability study was to assess the
influence of temperature on the stability of the five
different concentrations of the ethanol in water
CRM, by submitting them to temperatures above
and below room temperature, simulating extreme
conditions. The long-term stability study consisted
of following up quantitatively, for a period of 20
weeks to estimate the impact of the storage on the
concentration of ethanol at room temperature. To
the room temperature is been considered the
range from (20 0,3) °C to (25 0,3) °C.
The analysis was performed by gas
chromatography with flame ionization detector
(GC/FID),
on-column injector (65ºC, rate:
16ºC/min, 125ºC (3min)), and the quantification
was performed by the addition of 1-propanol as
internal standard. The column used was a DBFFAP of 60m (1,00m of phase thickness 0,53mm
of external diameter). The temperature program
was: 65ºC (10min), rate of 15ºC/min up to 120ºC
(6min), column flow rate 5,65mL/min. The
injection volume was 1L. The detector was kept at
220ºC. The calibration curves were made with
gravimetrically prepared standards by using two
curves, with 8 points each, in the concentration
range of 0,035% to 0,105% and in the
concentration range of 0,085% to 0,58%,
respectively.
3. METHODS
During the stability studies the ethanol in water
solutions were considered as reference materials
(RM's) under a certification process.
For both stability studies each RM was prepared in
a 5 L bottle subsequently subdivided into nine
500 mL bottles.
In short-term stability study were prepared two
different groups as follows: three bottles were
considered as reference samples and, after being
weighed, were stored at 4 ºC (reference
temperature). The other six bottles were weighed
and stored in an stove at 60 ºC. Each two days,
two bottles were removed from the stove and left in
the laboratory at room temperature for two hours.
Immediately after, they were weighed and stored
at the reference temperature (4 ºC).
The method also included checking the
homogeneity within each bottle, as two aliquots
were analyzed for each bottle, each one of them in
duplicate.
The statistical criteria adopted for assessing the
stability and uncertainty in both studies were
based on linear regression and residue analysis in
conjunction with analysis of variance (ANOVA).
Two criteria shall be simultaneously met in order to
determine the stability of the material: a p-level
greater than 0,05 and an angular coefficient (B)
near to zero [2].
At the end of eight days, when all bottles had
already been submitted to the reference
temperature, the group was exposed to the
laboratory room temperature (from (20 0,3) °C to
(25 0,3) °C), and weighed and analyzed by gas
chromatography.
To the short-term study two more criteria were
evaluated: the maximum percent differences
acceptable for sample mass of each bottle before
and after the study temperature (60oC) and the
ethanol concentration of each bottle submitted to
study temperature (600C) against ethanol
concentration of bottles submitted to reference
temperature (4oC). These two parameters are
presented at Table 1.
In long-term stability study, the monitoring of each
analyte concentration in the RM was performed in
two different cycles in time. The first cycle began
one week after the preparation of the calibration
curve and the samples.
The analyses were performed once a week,
during a period of four weeks. In the same week
the preparation of calibration curves and samples
11
Figure 1: Ratio between the ethanol concentration for each
bottle submitted to 60 ºC and the ethanol concentration for the
bottles submitted only to the temperature of 4 ºC (reference),
for each RM concentration.
Table 1: Parameters and maximum percent differences
for the short-term stability study
Maximum
Percent
difference (%)
Parameter
Sample mass before and after the study
temperature (60 °C)
Analyte concentration after study temperature
against analyte concentration at reference
temperature
Figure 2 shows, for each bottle in each RM
concentration, the mass percent difference of each
bottle before and after the study temperature (60 oC).
0.05
1.0
0,10
[1]
[2]
[3]
[4]
[5]
4. RESULTS
0,08
M ass Di fference (%)
According to the number of days in the stove the
ethanol concentrations obtained for each bottle in
short-term stability study after them being
submitted to the reference temperature are shown
in the Table 2.
Table 2: Ethanol concentration (g ethanol/100 g solution)
after bottles were submitted to the temperature of 4ºC
Number of days in stove
2
4
0,06
0,04
0,02
0,00
-0,02
Concentration
0
7
[1]
0.0507
0.0507
0.0508
0.0507
[1]
[2]
0.0507
0.0891
0.0508
0.0888
0.0506
0.0892
0.0507
0.0889
[2]
0.0891
0.0890
0.0893
0.0892
[3]
0.1155
0.1158
0.1155
0.1157
[3]
0.1155
0.1155
0.1154
0.1153
[4]
0.3887
0.3893
0.3898
0.3885
[4]
0.3887
0.3909
0.3905
0.3909
[5]
0.4576
0.4587
0.4591
0.4588
[5]
0.4576
0.4577
0.4583
0.4567
0
2
4
6
8
10
Bottles
Figure 2: Mass Percent Difference after reference
temperature for each bottle in each RM concentration .
In long-term stability study, the results obtained for
the [A], [B], [C] and [D] ethanol concentrations
during the 20 weeks are displayed in Table 3.
Table 3. Ethanol concentrations (g ethanol/100 g
solution) obtained during the period of study.
Figure 1 shows, for each RM concentration, the
ratio between the ethanol concentration found in
each bottle submitted to the temperature of 60°C
and the bottles considered as reference samples,
submitted only to the temperature of 4°C.
Period of Study (weeks)
Ethanol
0
2
3
4
16
20
[A]
0.05090 0.05098 0.05081 0.05076 0.05097 0.05088
[B]
0.08940 0.08966 0.08876 0.08888 0.09043 0.08993
[C]
0.11499 0.11508 0.11418 0.11461 0.11603 0.11531
[D]
0.45841 0.46276 0.45918 0.45820 0.46556 0.46100
Ratio concentration
1,02
[1]
[2]
[3]
[4]
[5]
1,01
Figures 3 and 4 show the variation of the RM
concentrations [A] and [B], [C] and [D] during the
period of study.
1,00
0,99
2
3
4
5
6
7
Days in Stove
12
Table 4 and Table 5 show respectively to the short
and long-term stability study, the uncertainty
related to the transport and storage to each CRM
concentration, the angular coefficient of the curve,
the p-level and standard deviation (STD) in
relation to time.
conc A
conc B
0,11
0,10
[E tha no l] (% )
0,09
0,08
0,07
0,06
Table 4:- Uncertainty for each concentration of CRM,
obtained by statistical treatment (linear regression)
0,05
0,04
Concentration
-2
0
2
4
6
8
10
12
14
16
18
20
Time (weeks)
Figure 3: RM concentrations [A] and [B], obtained during
the period of study.
conc C
0,60
*
Angular
p-level (p) Uncertainty*
Coefficient (B)
STD
[1]
-1.589E-06
0.8127
0.000041
5.89184E-06
[2]
5.293E-06
0.8875
0.00023
3.30646E-05
[3]
1.822E-05
0.4244
0.00013
1.83044E-05
[4]
1.216E-04
0.4713
0.00097
0.000138037
[5]
2.359E-05
0.8506
0.00077
0.00011042
The uncertainty is expressed in g ethanol/100g solution.
conc D
0,55
0,50
Table 5: Statistical results obtained by linear regression
for the respective ethanol concentrations.
0,45
[Eth anol] ( %)
0,40
0,35
Angular
Concentratio Coefficient
n
(B)
0,30
0,25
0,20
[A]
[B]
[C]
[D]
0,15
0,10
0,05
-2
0
2
4
6
8
10
12
14
16
18
2.370E-06
5.353E-05
4.978E-05
1.913E-04
*
p-level (p) Uncertainty
0.6641
0.1187
0.1536
0.2550
0.00010
0.00054
0.00057
0.0029
STD
5.0635E-06
2.7021E-05
2.83232E-05
0.000144097
20
*
Time (weeks)
Figure 4: RM concentrations [C] and [D], obtained during
the period of study.
The uncertainty is expressed in g ethanol/100g solution.
The uncertainty related to the transport and
storage (short and long term stability study) for
each CRM ethanol concentration is calculated by
the product of standard deviation and the time of
study as show in Equation 1.
ISO GUIDE 34 [3] establishes that the uncertainty
inherent to a CRM shall be calculated on the basis
of the combination of all sources involved in the
certification process, i.e., the combination of the
uncertainties taken from the characterization, the
homogeneity, the storage and the transport, being
the two latter the object under estimation in this
study.
To the short-term stability study
the long-term stability study
t = 7 days and to
t = 20weeks.
U Stability =STD* t
(1)
The uncertainties related to storage and transport,
long-term and short-term studies, respectively,
were calculated through an spreadsheet, that
provide the following parameters: the standard
deviation of ethanol concentration in relation to the
time of study, the angular coefficient of the curve
related to the variation of ethanol concentration
with the time and p-level.
where:
U Stability = Uncertainty Stability
STD= Standard Deviation
t= time
13
5. DISCUSSION
6. CONCLUSION
In short-term stability study, the ratio between the
ethanol concentrations found for each bottle at 60
ºC and the bottles submitted only to the
temperature of 4 ºC (reference samples) was
approximately 1, showing that no significant
variation of concentration occurred when
submitting the CRM to this range of temperature.
The short-term stability study characterized for all
concentrations studied the stability of the ethanol
in water CRM in the range of temperature from 4
ºC to 60 ºC at the period of 7 days, providing an
estimation, for each concentration, of
the
uncertainty inherent to the transport of this
material.
In relation to the study to evaluate the mass
difference of the bottles before and after they had
been submitted to the study temperature (60 oC) it
can be seen that there was no lost of mass
evidencing the stability of ethanol solutions for all
concentrations.
The long-term stability study of the ethanol in
water RM demonstrated the stability of the CRM
along the 20 weeks period of study (5 months).
Hence, this period corresponds to the period of
storage shown in the CRM certificate. This study
also facilitated the estimation of the uncertainty of
the ethanol in water CRM storage, in regard to the
certified values of ethanol concentration.
As the study evidenced the CRM stability in the
temperature range of 4 °C to 60 °C for 7 days, the
uncertainty obtained is deemed to be the
uncertainty inherent to the transport of CRM, since
during the transport a material may be submitted
to variable conditions and the range of
temperature studied simulates extreme
conditions.
In relation to long-term stability study, the Figures
3 and 4 show that for all ethanol concentrations no
significant difference was detected during the
period of study (20 weeks). The calculations
performed demonstrated that the variation of each
concentration, in each week, compared to time
zero (t=0) is not greater than 1.0%, indicating a
satisfactory concordance of data and evidencing
the stability of the material.
It is important to mention that the long-term
stability study of ethanol in water CRM is under
development at DQUIM/ DIMCI/ INMETRO with
the objective to reach the total time of 13 months.
7. REFERENCES
[1] PAUWELS, J.; LAMBERTY A.; SCHIMMEL, H.;
Homogeneity testing of reference materials,
Accred. Qual. Assur (3) 51-55, 1998.
[2] ISO GUIDE 35:2006 Reference materials –
General and statistical principles for certification.
The results of this study made possible the
estimative of the impact of the CRM storage on the
ethanol concentration, and also the subsequent
determination of the uncertainty taken from the
CRM storage during the period of study. This
uncertainty can be seen at the Table 5.
[3] ISO GUIDE 34:2000 General requirements for
the competence of material producers.
[4] BCR Guidelines for feasibility studies on
Certified Reference Materials, 2002.
The statistical parameters used in both stability
studies are those obtained from linear regression
and residues analysis. These parameters,
expressed by the angular coefficient (B) and the
p-level (p), indicate if a material is or not stable. A
material is deemed to be stable when the following
criteria are simultaneously met: B 0 and p > 0,05.
At the Tables 4 and 5 can be seen that these two
criteria were met, confirming the stability of the
ethanol in water CRM for all concentrations
studied.
These results meet the criteria established in ISO
GUIDE 35, ISO GUIDE 34 and BCR Guidelines for
Feasibility Studies, evidencing the certification of
the RM for the period of study.
14
CONTROL OF PREPACKAGES
Eberhard Seiler
Chairman of OIML´s Permanent Working Group on Developing Countries
[email protected]
Remark
Dear reader,
As chairman of the Permanent Working Group for Developing Countries of the International Organization of Legal
Metrology (OIML) I take this opportunity to draw your attention to a special Discussion Forum as part of OIML´s web site.
This Forum was created to facilitate the cooperation, the exchange of information and ideas especially among
Developing Countries.
There you will find my proposal for the control of prepackages as printed here.
If you carry out checks of prepackages already it is perhaps not necessary to read the following. If you are not yet active
in this field, you will find some arguments and ideas why and how to start even with very limited resources.
Why checking prepackages?
carried out. Products which are needed for the daily life
and which are fairly expensive should be chosen such as
edible oil or milk products. Results obtained in Germany
show significantly higher offences for these products than
the mean of 5%, namely 13.5% respectively 10.8% in
2006.
One disadvantage of this proposal is the destructive
nature of the check, you must open the container and pore
out the liquid. Usually, you have to pay for the product. If
money is not available you can try to argue with your local
producers to see their quality management systems and
their measures to guarantee the correct filling. Perhaps
you can check their products in their premises on their
account.
Even if that should not be possible you can do something.
Since we are all consumers we have to buy prepackages.
Why not using prepackages for private consumption for
the first survey? If you measure before you use the food
stuff and if you measure whenever you use the next item
you will get information about the degree of filling of this
special item and whether you get the mean value of the
declared quantity after a sufficient long period of time.
Another possibility to start activities is the check of empty
containers which can be easily collected and used for this
purpose. The containers should carry an information
about their nominal volume. There is an OIML
Recommendation R 96 Measuring container bottles
specifying the necessary requirements. Make the
experiment and determine the maximum volume of
containers! I am sure you will find some which are smaller
than indicated.
Checking the net content of prepackages is one important
and continuous task of legal metrology services. These
controls are carried out in Germany for more than 30 years
and the verification authorities aim at checking the
producers once a year. More than 20 000 checks are
carried out every year. Although the offences against the
relevant regulations decreased during the years the mean
value of offences is still around 5%.
The European Union established filling requirements
which producers of prepackages must comply with. If they
do so they can put an e on their packages which allows
them to put their products on the European market. The
legal metrology authorities of the packer`s country are
responsible for the supervision. Despite all these
regulations and checks the loss for the consumer in
Europe due to under filling sums up to many million Euro
per year. It is almost certain that losses in countries
without any control of the net content are even higher and
should be a challenge for the legal metrology services.
What can be done?
What can be done if there are no legal requirements for
prepackages in force? Does it mean that nothing can be
done? I don´t share this opinion. Checks can be carried
out without regulations in force but of course without legal
consequences. The OIML Recommendation 87 Net
content in packages can be used for the checks. The
results obtained will reflect the market situation. They
should be used as arguments in discussions with the
responsible ministry for drafting the necessary national
regulations.
But what can be done if there is no equipment? Of course,
this will be a handicap but here is my proposal for a first
survey with simple and cheap measuring instruments:
Since many prepackages are filled by volume the checks
require standard graduated glass flasks (R 43) or
graduated glass cylinders. If they are not among the
measuring instruments of the legal metrology service they
can be purchased for a small amount of money. If there is
no specialized shop for such measuring instruments they
may be available at shops for hospital equipment.
It is recommended to set priorities for the checks to be
Final remarks:
1.
2.
15
As shown above checks can be carried out even
with very little resources. I hope the proposals
will encourage you to start activities. If you have
comments or questions don´t hesitate to contact
the Forum (under OIML´s home page:
www.oiml.org you will find Developing Countries
and Discussion Forum. In case you need a pass
word contact one of the given reference
persons).
It would be of interest to learn about your
experience. Please let us know your results
which you can publish on OIML´s website.
DEL SOFTWARE DE LOS INSTRUMENTOS DE MEDICIÓN
SUJETOS A APROBACIÓN DE MODELO ¿AUTENTICARLO O
VALIDARLO?
C. Cajica Gómez
Centro Nacional de Metrología
[email protected]
Un instrumento de medición de operación automática,
entendiéndose por ello a un instrumento de medición
que incluye instrumentación electrónica basada en un
circuito integrado denominado microprocesador o un
microcontrolador para facilitar tanto su manejo como la
obtención de los resultados de sus mediciones, tendrá
características metrológicas tan buenas como su
programación lo permita, incluso los modelos
matemáticos programados en los circuitos integrados
podrían corregir la falta de linealidad en la operación de
los mecanismos, transductores y sensores; realizar
corrección por histéresis o señalizar la condiciones de
saturación, entre otros.
normas mexicanas obligatorias (NOM) y que sirvan de
base o se utilicen para una transacción comercial para
determinar el precio de un producto o servicio, deben
contar con aprobación de modelo o prototipo, artículo
10 de la Ley Federal Sobre Metrología y Normalización.
La aprobación de modelo es el procedimiento por el
cual se asegura que un instrumento de medición
satisface las características metrológicas,
especificaciones técnicas y de seguridad establecidas
en las normas antes referidas.
En México, los instrumentos de medición cuya
verificación inicial, periódica o extraordinaria es
obligatoria, así como las normas obligatorias que los
regulan son los siguientes:
Actualmente la legislación mexicana exige que los
instrumentos para medir que se encuentren sujetos a
Instrumentos
Norma Oficial Mexicana que los regulan
Instrumentos para pesar de
bajo, mediano y alto alcance de
medición
Sistemas para medición y
despacho de gasolina y otros
combustibles líquidos
NOM-010-SCFI-1994 “Instrumentos de medición-Instrumentos
para pesar de funcionamiento no automático-Requisitos
técnicos y metrológicos”
NOM-005-SCFI-2005 “Instrumentos de medición-Sistema para
medición y despacho de gasolina y otros combustibles
líquidos-Especificaciones, métodos de prueba y de
verificación”
NOM-014-SCFI-1997 “Medidores de desplazamiento positivo
tipo diafragma para gas natural o L. P. Con capacidad máxima
de 16 m3/h con caída de presión máxima de 200 Pa (20,40
mm de columna de agua)”
Medidores para gas natural o L.
P. con capacidad máxima de 16
m3/h con caída de presión
máxima de 200 Pa (20,40 mm
de columna de agua)
Relojes registradores de tiempo
Taxímetros
NOM-048-SCFI-1997 “Instrumentos de medición-Relojes
registradores de tiempo-Alimentados con diferentes fuentes de
energía”
NOM-007-SCFI-2003 “Instrumentos de medición-Taxímetros ”
Entrada en
Vigor
1999-08-08
2005-11-27
1998-12-22
1998-12-01
2003-09-08
naturaleza intrínseca de este tipo de software de poder
mejorarlo o adicionarlo con nuevas funcionalidades.
El avance tecnológico actual ha permitido que todos los
instrumentos de medición sujetos a normas obligatorias
incluyan instrumentación electrónica basada en
microcontrolador o microprocesador, permitiendo con
ello que los instrumentos sean programables y realicen
las mediciones sin la intervención de una persona,
operando de forma automática.
Por lo anterior se puede entender de manera lógica que
el comportamiento metrológico de cualquier
instrumento de medición podrá estar limitado por el
software que lo opere, y un punto importante a
considerar es que este software guardará la equidad en
las transacciones comerciales como la ética profesional
del fabricante lo permita. Así, considerando la situación
que se presenta para los instrumentos de medición de
operación automática, su verificación inicial, periódica y
extraordinaria debería de incluir un procedimiento de
revisión del software que permitiera comprobar
íntegramente la forma en que los resultados de las
mediciones realizadas y el establecimiento de precios
para el caso de transacciones comerciales, son
calculados y mostrados. Actualmente ninguna norma
oficial mexicana lo incluye.
Los fabricantes de instrumentos de medición que
incluyen estas nuevas tecnologías, realizan la
programación de los microcontroladores o
microprocesadores conformando un software que es
colocado dentro del propio circuito integrado, este
software que reside en los circuitos integrados es
conocido como Firmware.
La actualización del software que reside en los circuitos
integrado de los instrumentos de medición es una
actividad periódica que realizan los fabricantes, dada la
16
Los esfuerzos hasta ahora para realizar de alguna
manera algún tipo de verificación de software solo han
llegado a establecer un proceso de autenticación y solo
por una norma oficial mexicana: la NOM-005-SCFI2005 "Instrumentos de medición-Sistema para
medición y despacho de gasolina y otros combustibles
líquidos-Especificaciones, métodos de prueba y de
verificación". La autenticación en esta norma está
basada en una técnica de reducción criptográfica, lo
que permite conocer en efecto, si el origen de este
software es efectivamente el del fabricante. No
obstante, la operación del mismo no es factible de
conocer y solo el fabricante sabe como opera y como
controla el instrumento de medición, con ello la
autoridad que aprueba el modelo o prototipo del
instrumento realiza un acto de fe sobre la ética
profesional del fabricante respecto a la equidad en la
transacción comercial en la que interviene el
instrumento de medición.
la garantía de la equidad en las transacciones
comerciales en las que participa.
Existen fuertes dificultades técnicas para realizar una
validación de software en los términos antes descritos,
pues tanto el abanico de opciones y posibilidades
tecnológicas para programar circuitos integrados como
la disposición de tecnologías de las diferentes marcas
de los propios circuitos son muy grandes; implicaría la
necesidad de contar con varios especialistas en la
materia para realizar las verificaciones, aunado a que la
revisión de un programa de cómputo podría requerir
mucho tiempo. La participación del fabricante en esta
revisión sería indispensable así como la utilización de
normas nacionales o internacionales relacionadas con
la verificación de software; por otra parte el fabricante
tendría la necesidad de mostrar a la autoridad el Know
How de su tecnología, lo que implicaría manejar
esquemas de protección industrial y manejo de
información confidencial seguros.
La limitación de la autenticación puede compensarse
agregando esquemas de revisión de cada versión de
software de cada fabricante a nivel de los códigos de
operación del mismo, lo que implicaría conocer el
archivo del programa fuente en el lenguaje que
entienden los circuitos integrados, comprobando todo
subprograma, función, procedimiento, subrutina,
control de eventos, y manejo de interrupciones para lo
que fuera programado el microprocesador o el
microcontrolador correspondiente, este procedimiento
conformaría una validación cuyo dictamen de
verificación daría la certeza de lo que realmente realiza
la programación del instrumento de medición y con ello
Los resultados que se esperarían con el
establecimiento de un procedimiento de validación de
software bien valen la pena, la autoridad cumpliría
cabalmente con la responsabilidad de guardar la
protección al consumidor asegurando la relación precio
cantidad en las transacciones comerciales en las que
los instrumentos de medición de operación automática
participan, aunado al reforzamiento y mejor cobertura
de los esquemas de evaluación de la conformidad en el
procedimiento de la aprobación de modelo o prototipo.
Las siguientes dos figuras esquematizan los dos
procesos de verificación de software citados.
17
COMPARISON OF METHODS FOR THE WEIGHING TEST IN CALIBRATION
OF HIGH CAPACITY NON-AUTOMATIC WEIGHING INSTRUMENTS*
Luis O. Becerra 1, Eduardo González 1, Félix Pezet 1, José Revuelta M 2, José Revuelta R 2, Sylvia
Maeda 2
1
2
CENAM, Querétaro, México, [email protected]; [email protected], [email protected]
Básculas Revuelta Maza, S.A. de C.V., Torreón, México, [email protected], [email protected],
[email protected]
Abstract: This paper presents the results obtained from the comparison of different techniques for the weighing
tests for the calibration of high capacity weighing instruments, in order to evaluate their use as a function of the
amount of available weights and the required uncertainty of the instrument in their normal use.
Keywords: Mass, Weighing instruments, substitution weights.
1. INTRODUCTION
eccentric load indication neither the repeatability.
In calibration of high capacity non-automatic
weighing instruments a common problem is the
amount of available mass standards (or weights)
additionally to the difficult of handling and
transporting large weights.
It was decided to make the test methods for the
adjusted instrument and repeat them for the same
instrument misadjusted intentionally in order to
evaluate the sensitivity of the test methods for the
evaluation of the characteristic response of the high
capacity non-automatic weighing instruments.
This work presents the uncertainty comparison of
different procedures for the weighing tests in
calibration of a truck scale. The main difference of
these procedures is based on the amount of standard
weight used.
The weighing test methods evaluated were:
i. Mass standards
ii. Substitution loads
iii. Combinatorial Technique
There is not any standard, guideline or
recommendation in Mexico or at international level
about the required uncertainty for the calibration of
those instruments; therefore it is a choice of the user
which is the suitable uncertainty for the calibration of
his own weighing instrument.
3. WEIGHING INSTRUMENT TESTED AND
MASS STANDARDS
The truck scale tested is located in Rancho Monte
Carlo in Torreón Coahuila, Mexico.
The scale tested has the following characteristics
(see fig. 1):
Brand:
Revuelta
Model:
RCC-1880-VR
Serial number:
19645-C.780R
Range:
80 000 kg
Type:
Truck scale
Resolution:
10 kg
Accuracy Class:
OIML Medium III [2]
Load receptor
Dimensions:
18 m x 3 m
Points of support: 8
The calibration tests were done following the
guidelines of EA-10/18 [1].
2. PURPOSE
The purpose of this study is to estimate the
uncertainty of different calibration methods for high
capacity non-automatic weighing instruments as a
function of the amount of mass standards used and
the characteristics of the weighing instruments such
as repeatability and sensitivity.
For the calibration were used eighty mass
standards with the following characteristics,
Nominal value: 1 000 kg
Accuracy class: OIML M1 2
Density: 4 782 kg/m3
All tests were done in two phases:
·
Adjusted instrument
·
Instrument misadjusted intentionally on its
characteristic response (linearity) but not the
* This paper was originally presented at the IMEKO XVIII World Congress, Rio de Janeiro, Brazil, September 2006. It is reproduced with permission.
2
The Mexican Standard NOM-038-SCFI-2000 [5] allows the construction of weights of large nominal value made of steel box filled with metallic material.
18
eccentric loading is calculated from the following
formula,
D
I ecci =
Ii I1 +
d
res
(1)
where,
Ii
is the indication of the weighing
instrument when a the load is placed at
the end i of the load receptor
I1
is the indication of the weighing
instrument when a the load is placed at
the centre of the load receptor
is the correction due to the resolution of
d
res
the scale with zero as mean value.
4.2 Repeatability test
The repeatability test was done using a vehicle
with load (approx. 26 000 kg) and the same vehicle
with tow (approx. 51 400 kg). For this test each
load was placed three times at the centre of the
load receptor.
Fig. 1. Truck scale tested.
From the three indications for each load Ij the
standard deviation for the load j is calculated using
formulas (2) and (3). The repeatability of the scale
is calculated with formula (4), which considers the
contribution of the resolution.
2
1 n
s Ij = å
I ji Ij
n1 i=
1
(
)
Fig. 2. Mass standards used.
()
(2)
1 n
(3)
Ij =
I ji
å
n i=
1
4. TESTS
The tests were done for the adjusted and
misadjusted weighing instrument. The calibration
tests performed were the following:
· Eccentric loading test
· Repeatability test
· Weighing test
2
s Scale
d ö
æ
=
s Ij +
ç÷
12 ø
è
(
)
2
(4)
where,
n
is the number of repetitions, three for this
exercise
Iji
Is the repetition i of the indication with the
load j on the load receptor
d
is the resolution of the scale
The differences among test methods are focused
in the weighing tests which were done with
different amount of mass standards and
substitution weights and the number of
measurements done.
4.3 Weighing test
For the weighing test, three methods were
applied.
4.1 Eccentric loading test
The eccentric loading test was done using a
vehicle with load (approx. 26 000 kg). The vehicle
was placed in three positions along the load
receptor of the scale, at the center of the load
receptor and at the ends of the load receptor of the
truck scale.
4.3.1 Weighing test with mass standards
The weighing test was done with eighty mass
standards of 1 000 kg placed one by one in the
load receptor.
The results of this exercise were taken as
reference values in order to evaluate the others
weighing test methods.
The load was placed within the point of support of
the scale in order to avoid a malfunction or
damage of the instrument. The effect of the
19
4.3.2 Weighing test using substitution loads
For this exercise, there were used m = 16 000 kg in
mass standards (16 pieces of 1 000 kg each), and
loaded vehicles in order to have loads close to the
following values,
5.2 Buoyancy correction
Even when the calibration of the scale was done in
conventional mass, all indications of the scale
were corrected by the buoyancy effect by the
following formula,
æ
r
1, 2 ö
aCal ÷
d
mB =
mc ç
çr
÷
è m
ø
Q1 16 000 kg
Q2 32 000 kg
Q3 48 000 kg
Q4 64 000 kg
The weighing sequence was the following,
i. m
ii. Q1
iii. Q1 + m
iv. Q2
v. Q2 + m
vi. Q3
vii. Q3 + m
viii. Q4
ix. Q4 + m
where,
mc
is the conventional mass of the mass
standard placed in the load receptor
r
aCal is the air density during the calibration
r
is the density of the mass standard or the
m
substitution weight
5.3 Inverse of the sensitivity of the weighing
instrument
The inverse of the sensitivity of the scale was
evaluated for the weighing test in order to correct
the mass difference (given in indications of the
scale) between the mass standard (or known
mass) and the unknown mass or load.
4.3.3 Weighing test applying the combinatorial
technique [4]
This exercise was done using m = 8 000 kg as
mass standards, and four loaded vehicles with the
following approximate mass values,
R1 17 000 kg
R2 13 200 kg
R3 10 460 kg
R4 5 730 kg
The impact to the uncertainty due to this factor is
directly proportional to difference between the
indication of the load and the indication of the
mass standard.
The inverse of the sensitivity is evaluated several
times for the weighing test using the following
formula,
(7)
The goal of this method is to have the 2n
combinations of loads. For this exercise and
taking 5 individual loads (taking the mass
standards as one of them) there were 32
combinations of load.
æ
ö
ær
mc
1,2 ö
1
aCal ç
÷
ç
÷
S=
×
1+
d
ç
÷
Re p(
S )
ç
÷
(
I x+
Ix )
+
d
+
d
+
d
m
Ecc
Re p
Round ø
è r
m
ø
è
1
where,
Ix+m
5. ERRORS OF INDICATION
5.1 Rounding error of the indication
In order to reduce the uncertainty due to the
resolution of the scale, there were applying small
extra weights in steps of dT = d/10, for this exercise
dT =1 kg.
The corrected indication due to the rounding error
(IL) is calculated by the following formula,
d
IL =
I+
nd T
2
(6)
is the indication of the weighing instrument
when a load x and the mass standard are
on the load receptor
Ix
is the indication of the weighing instrument
when the load x is on the load receptor
d
Ecc
is the correction due to the eccentric load of
the scale with zero as mean value.
d
1
is the correction due to the repeatability
)
Re p (
Sof the scale with zero as mean value.
(5)
where,
I
is the indication of the scale
n
are the number of small extra weights
placed in the load receptor
dT is the value of the small extra weights
20
d
Round
is the correction due to the rounding error
of the indication with zero as mean value.
d
Re p
is the correction due to the repeatability
(or dispersion) of the sensitivity
evaluations. This correction has zero as
mean value.
5.4 Evaluation of the error of indication
The error indication for the weighing test is
calculated from the next formula,
2
Ei =
a0 +
a1 I (
a2I (
use)
i +
use)
i
(13)
where the indication in use has the following form,
(8)
E=
IL mc +
d
mB +
d
+
d
+
d
Ecc
Re p
Round
I(
=
I+
d
m 'B +
d
d
d
(14)
use )
rep +
ecc +
res
For the weighing test using substitution loads and
applying the combinatorial technique, it is
necessary to evaluate the conventional mass of
the load x, m Load x , see formula (9).
C
where
d
m' B is the buoyancy correction due to the
density difference between the mass
standards used in the calibration and the
unknown material.
æ
r
1,2 ö aCal
÷
mc ç
+
D
I×
S 1
çr
÷
(9)
èm
ø
mC (
Load x )
=
ær
ö
aCal ÷
ç
1çr
÷
Loadx )
è(
ø
D
I=
Ix Imc
The relation between the Indications and the
errors could be expressed as,
Xb
=
Ye
where,
(10)
is the matrix of the indications of the
weighing instrument
X
where,
is the vector of the fitting coefficients
r
is the mean density of the load x
(
Loadx )
Y
is the mass difference between the mass
standard and the load i reading in
indications of the weighing
I
Equation (15) could be solved by Gauss Markov
approach [7],
sensitivity of the scale
1
1
1
ˆ=
(15)
b
(
XTF
X)
XTF
Y
6. EVALUATION OF THE UNCERTAINTY OF
THE ERROR
The uncertainty of the error of the indication is
evaluated according GUM's method [6] applied to
formula (8) as the mathematical model,
u(
E)
=
[
ci ×
u(
xi )
]
å
2
where,
ˆ is the estimate of the fitting coefficients
b
is the covariance matrix, where was
introduced the combination of the
variance of the errors and the variance of
the fitting [7,8]
(11)
i
¶
E
ci =
¶
xi
The covariance matrix of the fitting coefficients,
(12)
ˆ
cov b
was calculated from the following
expression,
where,
ci
is the sensitivity coefficient due to the
input quantity i
u xi
is the standard uncertainty of the input
quantity i
is the vector of errors (related to the
indications calculated in chapter 5)
is the vector of errors (of fitting)
.1
is the mean value of the inverse of the
S-
n
(15)
1
1
ˆ)
cov(
b
=
(
X TF
X)
(15)
Last matrix has in its main diagonal the
variance of the fitting coefficients. The uncertainty
of the indications errors evaluated by the formula
(13) is evaluated by GUM method.
7. FORMULA TO DESCRIBE ERRORS IN
RELATION TO THE INDICATIONS IN USE
In order to derive a formula to describe errors in
relation to the indication in use I (
, with the
use )
following form,
The variance of the adjusting coefficients (a0, a1
and a2) and variance of the indication in use are the
contributions to the uncertainty for the indications
errors of the weighing instrument in use.
21
Characteristic response of the truck scale
1st experiment
90.0
In dicatio n error (kg)
70.0
50.0
30.0
10.0
-10.0
-30.0
-50.0
0
10000
20000
30000
40000
50000
60000
70000
80000
70000
80000
Indication (kg)
Mass standards
Substitution loads
Combinatorial Techinique
Fig. 3. Errors of the indication for the adjusted instrument.
Characteristic response of the truck scale
1st experiment
250.0
200.0
150.0
Indication error (kg)
100.0
50.0
0.0
-50.0
-100.0
-150.0
-200.0
-250.0
0
10000
20000
30000
40000
50000
60000
Indication (kg)
Mass standards
Substitution loads
Combinatorial Technique
Fig. 4. Graph of the errors of the indication of the truck scale calculated by second order formula. The
uncertainty of the indication in use was not considered for this evaluation. The correlations between adjusting
coefficients (a0, a1 and a2) were not considered.
22
Characteristic response of the truck scale
2nd experiment
50.0
0.0
Indication error (kg)
-50.0
-100.0
-150.0
-200.0
-250.0
-300.0
-350.0
-400.0
0
10000
20000
30000
40000
50000
60000
70000
80000
Indication (kg)
Mass standards
Substitution loads
Combinatorial Technique
Fig. 5. Errors of the indication for the misadjusted instrument.
Characteristic response of the truck scale
2nd experiment
100.0
0.0
Indication (kg)
-100.0
-200.0
-300.0
-400.0
-500.0
-600.0
0
10000
20000
30000
40000
50000
60000
70000
80000
Indication (kg)
Mass standards
Substitution loads
Combinatorial Technique
Fig. 6. Graph of the errors of the indication of the truck scale calculated by second order formula. The uncertainty
of the indication in use was not considered for this evaluation. The correlations between adjusting coefficients (a0, a1
and a2) were not considered.
23
Table 1. Errors of the indication evaluated by formula (13) for the adjusted instrument.
The uncertainty of the indication in use was not considered for this evaluation.
Indication
8 000
16 000
24 000
32 000
40 000
48 000
56 000
64 000
72 000
80 000
Mass standards
Errors
Uncertainty
2
3
6
3
10
3
14
4
18
4
22
5
26
5
30
6
34
7
38
7
Errors
-1
1
2
3
5
6
7
8
9
9
Substitution Loads
Uncertainty
32
41
54
69
86
106
128
152
179
209
En*
0,10
0,14
0,15
0,15
0,15
0,15
0,15
0,14
0,14
0,14
Combinatorial Technique
Errors
Uncertainty
En*
0
15
0,13
4
22
0,09
8
31
0,08
10
43
0,09
13
57
0,09
15
74
0,10
16
93
0,11
17
115
0,11
17
139
0,12
17
166
0,12
Table 2 Errors of the indication evaluated by formula (13) for the instrument intentionally misadjusted.
The uncertainty of the indication in use was not considered for this evaluation.
Indication
8 000
16 000
24 000
32 000
40 000
48 000
56 000
64 000
72 000
80 000
Mass standards
Errors
Uncertainty
-23
4
-44
4
-64
5
-85
5
-106
6
-126
7
-147
7
-167
8
-188
9
-208
10
Errors
-29
-56
-82
-107
-132
-156
-180
-203
-226
-248
Substitution Loads
Uncertainty
48
61
80
102
127
156
187
223
261
304
En*
0,12
0,19
0,22
0,22
0,21
0,19
0,18
0,16
0,15
0,13
Combinatorial Technique
Errors
Uncertainty
En*
-32
27
0,32
-57
38
0,33
-81
54
0,31
-106
74
0,29
-131
97
0,26
-156
124
0,24
-181
155
0,22
-206
191
0,20
-231
230
0,19
-256
275
0,17
* Normalized error [9]. The normalized errors were calculated taking as reference the errors evaluated by
mass standards method.
24
REFERENCES
9. CONCLUSIONS
[1] European co-operation for acreditation, “EA10/18 Guidelines on th4e calibration of
nonautomatic weghing instruments” June 2005
Metrological characteristics of the weighing
instruments such as repeatability, eccentricity and
sensitivity should be introduced in the evaluation
of the error's uncertainty, and it is very important to
have a good estimation of those values in order to
have a good estimation of the errors too.
[2] OIML “R 76 Non automatic weighing
instruments; Part 1: Metrological and technical
requirements –Test”. Edition 1992, Amendment
1 (1994)
When substitution loads are used, it is
recommended to follow as much as possible the
clause 3.7.3 of OIML R76 [2], in order to keep
under control the error's uncertainty.
[3] OIML “R 111-1 Weighits of classes E1, E2, F1,
F2, M1, M1-2, M2, M2-3 and M3; Part 1:
Metrological and technical requirements”.
Edition 2004.
Selection of the calibration method for truck scales
should be based on one hand on the comparison
of the required uncertainty (for the normal use of
the instrument) and the calibration uncertainty,
and on the other hand on the available mass
standards (weights), substitution mass and the
metrological characteristics of the weighing
instrument under calibration.
[4] Clarkson M., Collins T., “A combinatorial
technique for the scale verificacition”. OIML
Bulletin Volume XLIII, Number 2, April 2002.
[5] Secretaría de Economía, “NORMA OFICIAL
MEXICANA NOM-038-SCFI-2000, PESAS DE
CLASES DE EXACTITUD E1, E2, F1, F2, M1,
M2 y M3”
[6] BIPM, IEC, IFCC, ISO, IUPAC, IUPAP, OIML
“Guide to the expression of uncertainty in
measurement”, Reprinted in 1995.
Indication errors and their associated
uncertainties obtained from substitution loads
method and combinatorial technique methods
were calculated with the same formulas and under
the same assumptions.
[7] Bich W. “Variances, Covariances and
Restraints in Mass Metrology”, Metrologia 27,
111-116 (1990)
The significant difference between the substitution
loads method and combinatorial technique is the
number of indications taken which represents the
degrees of freedom for the characterization of the
characteristic response of the scale.
[8] Bich W., Cox M.G., Harris P. M. “Uncertainty
Modelling in mass Comparisons”, Metrologia
1993/1994, 30 495-502.
[9] Wolfang Wöger -Remarks on the En –
Criterion Used in Measurement Comparison,
PTB-Mitteilingen 109 1/99, Internationale
Zusammenarbeit
The goal of all calibration methods (specially the
weighing test) should be to reach, as much as
possible the maximum capacity of the weighing
instrument.
It was presented a Gauss Markov approach [7] for
calculating the adjusting coefficients of the formula
(13). The Gauss Markov approach gives a good
estimation of the variance values of the adjusting
coefficients of the formula (13).
The covariance values among the adjusting
coefficients were not taken into account in order to
have a conservative estimation of the uncertainty
of the indication errors calculated by formula (13).
25
NOTI-SIM
SIM News in 2007
Noticias en el SIM en 2007
The metrology in the Americas thanks you Steve!
SIM sends its regards on your retirement.
Comparaciones entre laboratorios
Interlaboratory comparisons
·
Flujo luminoso.
Comparación iniciada en abril de 2004, las mediciones
fueron terminadas en julio 2006. El primer borrador del
resultado de la comparación está por emitirse.
Participan: INMETRO, INTI, NIST, INMS-NRC y
CENAM como piloto.
·
Luminous flux
Started on April 2004, the measurements ended on
July, 2006. The first draft is to be issued soon.
Participants: INMETRO, INTI, NIST, INMS-NRC y
CENAM as pilot.
·
Density of liquids
The results of this comparison will be linked to those
from key comparison CCM.D-K4.
The circulation of the high accuracy hygrometers
used as comparison artifacts started on April 2007
and ithe end of the measurements is expected on
February 2008.
Participants: INMS-NRC, NIST, LACOMET,
CENAMEP, IBMETRO, SIC, INEN, INDECOPI,
INEMTRO, CESMEC, LATU, BSJ y CENAM as pilot.
·
Densidad de líquidos
Comparación cuyos resultados se ligarían con la
comparación clave CCM.D-K4 a llevarse a cabo
próximamente.
La circulación de hidrómetros de alta exactitud como
patrones viajeros para la comparación inició en abril del
2007, y las mediciones se finalizarán en febrero de
2008.
Participan: INMS-NRC, NIST, LACOMET, CENAMEP,
IBMETRO, SIC, INEN, INDECOPI, INEMTRO,
CESMEC, LATU, BSJ y CENAM como piloto.
·
Real time intercomparisons network on time
and frequency
On 2007 5 new SIM members have received a GPS
system to participate in the network: Colombia,
Guatemala, Jamaica, Brazil and Argentina, joining
those already in this SIM network: National
Research Council (NRC), National Institute of
Standards and Technology (NIST), Centro Nacional
de Metrología (CENAM), Centro Nacional de
Metrología de Panamá (CENAMEP), and Instituto
Costarricense de Electricidad (ICE). The results of
this intercomparison are publicly available in
http://tf.nist.gov/sim/. On February 2008, a course on
Time and Frequency Metrology will take place in the
facilities of the Instituto Nacional de Tecnología
Industrial, (INTI), Buenos Aires, Argentina, where
discussions about the network will be included. It is
expected that 15 SIM laboratories will be in the
network by 2009.
·
Red de comparación en tiempo real de
tiempo y frecuencia
Durante 2007 5 nuevos laboratorios de
países
miembros del SIM han recibido el Sistema GPS para la
comparación en tiempo real de patrones de tiempo y
frecuencia: Colombia, Guatemala, Jamaica, Brasil y
Argentina. Estos nuevos laboratorios se suman a los
que ya participan en la red de comparación desarrollada
por el SIM: National Research Council (NRC), National
Institute of Standards and Technology (NIST), Centro
Nacional de Metrología (CENAM), Centro Nacional de
Metrología de Panamá (CENAMEP), Instituto
Costarricense de Electricidad (ICE). Los resultados de
la comparación son públicos y pueden ser consultados
en la dirección electrónica http://tf.nist.gov/sim/. En
febrero del 2008, el SIM realizará un curso de
Metrología de Tiempo y Frecuencia en el Instituto
Nacional de Tecnología Industrial, (INTI), Buenos Aires,
Argentina, donde esta red será uno de los puntos a
tratar. Se estima que en 2009 habrá alrededor de 15
países participantes.
26
Training and related activities
Capacitación y actividades similares
·
SIM Guidelines on the calibration of non-automatic
weighing instruments
Con la intención de discutir el documento SIM Guidelines
on the calibration of non-automatic weighing instruments,
del 6 al 9 de noviembre se realizó una reunión en Costa
Rica, coordinada por LACOMET y el CENAM, en la cual
participaron 30 personas representando 19 países. Uno
de los principales aspectos abordados fue el
entendimiento de la diferencia entre calibración y
verificación en instrumentos para pesar no automáticos
·
SIM Guidelines on the calibration of nonautomatic weighing instruments
This meeting aimed to discuss the document SIM
Guidelines on the calibration of non-automatic weighing
instruments, was held on November 6-9 in Costa Rica,
coordinated by LACOMET and CENAM. There were 30
attendees from 19 countries. One of the main issues
was the common understanding of the difference
between calibration and verification of non-automatic
weighing instruments.
·
Workshop on the steps to achieve recognized
measurement capabilities
Del 12 al 14 de noviembre se llevó a cabo con la
participación de 16 personas provenientes de Antigua y
Barbuda, Brasil, Costa Rica, Chile, Dominica, Grenada,
Ecuador, México, Panamá, Paraguay, Perú, St, Lucia,
Trinidad y Tobago, y Uruguay, bajo la conducción de
personal proveniente de ABBS, CENAM, LACOMET y
LATU. Como parte de la capacitación se realizó un
ejercicio en algunos de los laboratorios de LACOMET.
·
Workshop on the steps to achieve recognized
measurement capabilities
On November 12 – 14, this Workshop was held in Costa
Rica with the attendance of 16 persons from Antigua y
Barbuda, Brazil, Costa Rica, Chile, Dominica, Grenada,
Ecuador, México, Panama, Paraguay, Peru, St, Lucia,
Trinidad and Tobago, and Uruguay. The conduction was
done by personnel from ABBS, CENAM, LACOMET y
LATU. A practice on laboratory assessment was done in
some of the LACOMET facilities.
·
II Advanced School on Evaluation of Uncertainty in
Measurement
A realizarse en Rio de Janeiro, Brasil, del 10 a 14 de
diciembre de 2007, bajo los auspícios de INMETRO.
·
II Advanced School on Evaluation of Uncertainty
in Measurement
To be held in Rio de Janeiro, Brazil, on December 10 –
14, 2007, organized by INMETRO.
·
Metrología de presión
Curso programado para llevarse a cabo del 10 al 14 de
diciembre en las instalaciones del INDECOPI, Perú, con
instructores del CENAM.
·
Pressure metrology
Course to be held on December 10 – 14, in the facilities
of INDECOPI, Peru, with trainers from CENAM.
·
IMEKO 20th TC3 and 3rd TC 16 International
Conference
Noviembre 26-30, 2007, en Merida Yucatán, México.
Incluye temas sobre fuerza, masa, par torsional,
densidad, presión y vacío.
·
IMEKO 20th TC3 and 3rd TC 16 International
Conference
Noviembre 26-30, 2007, in Merida Yucatan, Mexico.
Topics are on force, mass, torque, density, pressure and
vacuum.
·
IMEKO – TC 11 First International Symposium RMO
2008 – Regional Metrology Organisations 2008:
“Metrology, testing, and accreditation – breaking the
trading barriers”.
A realizarse en Cavtat – Dubrovnik, Croacia, Noviembre
12–15, 2008. Fecha límite para proponer trabajos: June
2.
·
IMEKO – TC 11 First International Symposium
RMO 2008 – Regional Metrology Organisations 2008:
“Metrology, testing, and accreditation – breaking the
trading barriers”.
To be held in Cavtat – Dubrovnik, Croatia, November
12–15, 2008. Deadline for submission of papers: June
2.
Documentos
Documents
·
VIM
Con la intención de disponer de una única versión del
International vocabulary of basic and general terms in
metrology (VIM) - Third edition en español para ser
utilizado por todos los países de habla hispana, 17 dentro
del SIM y España, se ha distribuido una primera versión
en español del borrador final del mismo. Se han recibido
comentarios de algunos países. Se está en espera de la
versión definitiva del VIM por parte del BIPM/ISO para la
revisión y publicación final.
·
VIM
Aiming to produce an unique version of the International
vocabulary of basic and general terms in metrology
(VIM) - Third edition in Spanish to be used by 17 Spanish
speaking countries in SIM and Spain, a draft of the
Spanish translation of the Final Draft of the document
has been circulated for comments. A number of them
have been received, awaiting the publication of the
definitive version by BIPM/ISO to be taken into account.
SIM Management
Gestión en el SIM
·
Meetings of the Council, the Technical Committee
and the Professional Development Committee.
Held on July, 2007, during the NCSLI conference in St.
Pail., Min, USA.
·
Reuniones de su Consejo, Comité Técnico y
Comité de Desarrollo Profesional.
Se realizaron durante el NCSLI en el mes de julio en la
ciudad de St. Paul, Min., EUA.
27
·
XIII Asamblea General del SIM, en Ottawa en
Septiembre, 2007.
·
XIII SIM General Assembly, in Ottawa on
September, 2007.
Teniendo como anfitrión el INMS/NRC, Canadá, contó con
la asistencia de 24 países miembros y representantes de
organizaciones como BIPM, CIPM, APMP, EUROMET,
SADCMET y OIML.
Hosted by INMS/NRC, Canada, it was attended by 24
member countries and other representatives of
organizations like BIPM, CIPM, APMP, EUROMET,
SADCMET and OIML.
Destacan los siguientes aspectos:
As highlights of the Assembly,
o Se aprobó la emisión de certificados por el QSTF a los
países cuyas CMC tengan un sistema de gestión de la
calidad revisado y aprobado, para fines del cumplimiento
de uno de los requisitos del CIPM-MRA. La vigencia de
estos certificados será de cinco años.
o It was approved that a certificate will be issued by the
QSTF to those countries with quality management
systems reviewed and approved, aimed to fulfill the
requirements of the CIPM MRA. The certificates will be
valid for five years.
o Un grupo ad hoc revisará la estructura actual del SIM y
propondrá revisiones de los estatutos y otras
disposiciones en la medida en que se requiera.
o An ad hoc Working Group will review the current SIM
structure and suggest revisions to the statutes and bylaws as required.
o Se requiere que todos los miembros del SIM preparen
un informe anual de sus actividades en metrología, así
como de las mejoras y desarrollos de sus infraestructuras
metrológicas.
o All SIM members are required to prepare an annual
report on metrology activities, improvements and
developments of metrology infrastructure in their
country.
o Se están promoviendo relaciones más estrechas entre
el SIM y la IAAC.
o Greater collaboration SIM-IAAC is being fostered.
o Felicitaciones al Grupo de Trabajo en Metrología
Química en su X aniversario.
o Congratulations to the Chemical Metrology Working
Group on their 10th anniversary.
o Se vertieron expresiones de gratitud al Dr. Stephen
Carpenter, anterior Consejero Técnico del SIM, por su
labor en beneficio de la metrología en el SIM.
o Warm words of appreciation to Dr. Stephen
Carpenter, former SIM Technical Advisor.
·
Reuniones del Quality Systems Task Force.
Se celebró una reunión en la ciudad de Miami, Fla., EUA,
en el mes de febrero de 2007, en la cual se aprobaron los
sistemas de calidad que soportan algunas de las
capacidades de medición y calibración del INMS-NRC,
Canadá; LACOMET e ICE, Costa Rica; y CENAM,
México.
·
Quality Systems Task Force meetings
On the February 2007 meeting, in Miami, Fla., USA, the
quality management systems supporting some CMC of
INMS-NRC, Canada; LACOMET and ICE, Costa Rica;
and CENAM, Mexico, were approved.
On the September 2007 meeting, in Ottawa, Canada,
the quality management systems supporting some CMC
of TCC-NRC, Canada; INTI, Argentina; CESMEC, Chile;
NIST, USA; and, ININ and CENAM, Mexico were
approved. Besides, LACOMET, CENAMEP and
CENAM reported on the updating of their quality
management systems.
Otra reunión se llevó a cabo en el mes de septiembre en la
ciudad de Ottawa, Canadá, en la cual fueron aprobados
los sistemas de calidad que soportan algunas de las
capacidades de medición y calibración del TCC-NRC,
Canadà; INTI, Argentina; CESMEC, Chile; NIST, USA; e,
ININ y CENAM, México. Adicionalmente se informó sobre
la actualización de los sistemas de calidad de LACOMET,
Costa Rica; CENAMEP, Panamá; y CENAM, México.
.
·
Reunión del Quality Systems Working Group.
Se llevó a cabo en Costa Rica el 15 de noviembre, en la
cual se delineó una propuesta de plan de trabajo, y fue
electa Annamaría Narizano, LATU, Uruguay, como su
coordinadora.
·
Meeting of the Quality Systems Working Group.
Held in San José, Costa Rica, on November 15, 2007,
resulting in a proposal for a plan of activities, and in the
election of Annamaría Narizano, LATU, Uruguay, to
chair it.
.
Important SIM dates on 2008
Fechas importantes para el SIM en 2008
Reunión del Consejo del SIM, Marzo 2
Reunión del SIM QSTF, Marzo 3-4
Día Mundial de la Metrología, Mayo 20
Reunión del Consejo del SIM,
Agosto 2*
NCSLI Conference, Agosto 3-7
XIV Asamblea General SIM,
Septiembre 29 – Octubre 3*
SIM Council Meeting. March 3*
SIM QSTF Meeting, March 3-7*
World Metrology Day, May 20
SIM Council meeting, August 2*
NCSLI Conference, August 3-7
SIM XIV General Assembly, September 29 – October 3*
* To be confirmed
* Por confirmar
Further information:
http://www.sim-metrologia.org.br/calendar_2008.php
Mayor información:
http://www.sim-metrologia.org.br/spanol/calendario_2008.php
28