Calibration of measuring instruments
Contenido
El calibrado o calibración es el procedimiento de comparación entre lo que indica un instrumento de medición y lo que debería indicar de acuerdo con un unidad de medida patrón de referencia") con valor conocido. De esta definición se deduce que para calibrar un instrumento o patrón es necesario disponer de uno de mayor precisión que proporcione el valor convencionalmente verdadero, que es el que se empleará para compararlo con la indicación del instrumento sometido a calibrado. Esto se realiza mediante una cadena ininterrumpida y documentada de comparaciones hasta llegar al patrón primario, y que constituye lo que se llama trazabilidad. El objetivo del calibrado es mantener y verificar el buen funcionamiento de los equipos, responder a los requisitos establecidos en las norma de calidad normas de calidad") y garantizar la fiabilidad "Fiabilidad (psicometría)") y trazabilidad de las medidas.
Durante el calibrado, se contrasta el valor de salida del instrumento a calibrar frente a un patrón en diferentes puntos de calibración. Si el error de calibración —error puesto de manifiesto durante la calibración— es inferior al límite de rechazo, la calibración será aceptada. En caso contrario se requerirá ajuste del instrumento y una contrastación posterior, tantas veces como sea necesario hasta que se obtenga un error inferior al límite establecido. En equipos que no disponen de ajuste, como termopares, etc., en caso de no satisfacer las tolerancias "Tolerancia (fabricación)") marcadas deberían ser sustituidos por otros previamente calibrados.
En la calibración, los resultados deben documentarse con un certificado de calibración, en el cual se hacen constar los errores encontrados así como las correcciones empleadas y errores máximos permitidos. Además pueden incluir tablas, gráficos, etc.
Traceability
Traceability is the property of the result of measurements carried out by an instrument or by a standard, such that it can be related to national or international standards and through these to the fundamental units of the International System of Units through an uninterrupted chain of comparisons, with all uncertainties determined.
Thus we have a pyramidal structure at the base of which are the instruments used in the current measurement operations of a laboratory. Each step or intermediate step of the pyramid is obtained from the one that precedes it and gives rise to the next through a calibration operation, where the standard was previously calibrated by another standard, etc.
Calibration process
When calibrating an instrument, the following types of errors can be found:
To carry out the calibration of an instrument, follow the following steps:
Since the standard does not allow the true value to be measured, it also has an error, and since various sources of error are also involved in the comparison operation, it is not possible to characterize the measurement by a single value, which gives rise to the so-called uncertainty "Uncertainty (metrology)") of the measurement.
In very simple words, calibration is nothing more than the comparison of readings (data returned) between a standard instrument and the test instrument. Calibration should never be confused with adjustment, which is one of the calibration processes.
Measurement of results
The result of any measurement is only an approximation or estimate of the true value of the quantity being measured—the measurand. In this way, the expression of the result of a measurement is complete only if it is accompanied by the value of the uncertainty associated with said measurement. Uncertainty is therefore numerical information that completes a measurement result, indicating the amount of doubt about this result.
Measurement uncertainty generally includes several components:
Type A: Those that can be estimated from statistical calculations obtained from the samples collected in the measurement process. In most cases, the best available estimate of the expected value of a magnitude Xi, of which n observations have been obtained, under the same measurement conditions, is the arithmetic mean of the n observations.
Type B: Those that are only based on experience or other information. This type of evaluation is determined by the contributions to uncertainty, estimated using non-statistical methods, and which are characterized by terms, which can be considered as approximations of the corresponding variances.
Generally, the calibration of measuring equipment for industrial processes consists of comparing the output of the equipment against the output of a standard of known accuracy when the same input—measured quantity—is applied to both instruments. Every calibration procedure can be considered as a process of measuring the error that equipment makes.
Calibration and uncertainty
Since any measurement process has an associated uncertainty, calibrations must take into account all significant sources of uncertainty associated with the error measurement process that is carried out. In the industrial environment it is accepted that a source of uncertainty can be considered non-significant when its estimate is lower in absolute value than 4 times the largest of all the estimated sources.-.
The contributions to uncertainty are determined by the components of that uncertainty, together with their calculation and combination:.
Other sources of measurement uncertainty
Complete knowledge would require an infinite amount of information. The phenomena that contribute to uncertainty and, therefore, to the fact that the result of a measurement cannot be characterized with a single value. In practice, there may be many sources of uncertainty in a measurement, including the following:.