How to calculate the drift of an instrument
Drift analysis in measuring instruments is essential in metrology. Before calculating the drift it is important to know what the meaning of drift in an instrument is. Every reference instrument is calibrated periodically as per the frequency that is decided by a laboratory.
Calibration establishes the metrological state of an instrument on the particular date it was carried out. But what is the state of the instrument the next day or a week after or after a month after calibration or when the validity of the calibration period completes?
With certain types of instruments – in particular “fixed” instruments, for example physical standards and limit gauges, the situation is relatively clear-cut; but it is not so plain with instruments that take measurements. There is a clear distinction between the two types of device, those where drift can be observed (since calibration uncertainties tend to be minor) and the others (measuring instruments).
However every time an instrument is calibrated there could be a change in the achieved values due to any one of the below reasons:
A) Change in traceability values because of the references used.
B) Error induced in the UUC due to its wear & tear.
C) Human error
D) Environmental conditions
E) Damage to the instrument.
Hence it is imperative to keep a record of the change in the error values that occur between consecutive calibrations. This gives stability to the instrument with respect to time. The values to check for the difference are usually taken from the calibration certificate.
So drift is the difference between the errors of two consecutive calibrations. Drift analysis in measuring instruments is essential in metrology.
A drift analysis is further prepared with the values achieved during the last few calibrations. This trend gives confidence to the user about the instrument.
Below is an example of an analysis chart for Masses of 2kg, 5kg, 10kg & 20kg.
Every reference instrument is periodically calibrated by an accredited laboratory. The accredited laboratory issues a calibration certificate in respect to the particular instrument. On certain occasions there could be an error induced in one or more of the readings during calibration. Therefore in such a case it becomes the responsibility of the Calibration Lab to check whether the results are acceptable to the user for their application. For this reason the lab has to decide beforehand the acceptance limits for the results.
This can be created in a separate document called as the acceptance criteria. Therefore acceptance criteria can be created for each parameter. And finally against each instrument the parameter is mentioned along with its acceptance limits. If by any chance the result goes over the limits as mentioned in the acceptance criteria, the lab has to either service this instrument or remove it from regular usage.
E.g. for a force lab calibrating class 1 testing machines the reference load cells must have a minimum of class 1 accuracy. However if it has a class 2 accuracy, then the particular instrument cannot be used for this kind of work.