A set of criteria that provide meaningful description of quality of measurements under static conditions are called static characteristics. These criteria are
1. Accuracy
Accuracy is defined as the closeness with which the reading of the instrument approaches true value. The true value is just impossible to be determined experimentally. True value of a quantity can be defined as
the average of an infinite number of measured values when the average deviation due to the various contributing factors tends to zero.
2. Error
Errors are unavoidable in any measurement system. Attempts can be made to minimize them by suitably designing the system by taking care of all sources of errors. In order to have an overall estimate of all the sources is essential. The accuracy of any measurement system is measured in terms of its errors.
Static error is defined as the difference between the best measured value and the true value of the quantity.
Relative static error is the ratio of absolute static error to the true value of the quantity under measurement.
Relative static error =(Absolute error)/(Measured Value)
Also, True value = measured value × (1 – relative static error)
3. Repeatability
If an instrument is used to measure same or an identical input many times and at different time intervals, the output is not the same but shows a scatter. This scatter or deviation from the ideal static characteristics, in absolute units or a fraction of the full scale, is called repeatability error.
4. Reproducibility
It represents the degree of closeness with which a given input quantity be repeatedly measured within a close range. Reproducibility is a measure of closeness with which a given input may be measured over and over again. An instrument with high reproducibility has zero or negligible drift.
5. Drift
Drift causes the measurement result to vary for a given input quantity. It occurs due to wear and tear of parts, hysteresis effect in metals, changes in metals caused by contamination or other causes, environmental factors, stray electric and magnetic fields, thermal emf, changes in temperature and mechanical vibrations, high mechanical stresses in parts, etc. Thus drift is an undesirable quantity and it occurs very slowly. Drift can’t be easily compensated for but can be carefully guarded against by care, prevention, inspection and maintenance. It is drift which causes change in calibration of instrument.
6. Sensitivity
It is the ratio of the magnitude of output to the input signal or the response of measuring system to the quantity being measured.
For an instrument having linear calibration curve sensitivity is the slope of the calibration curve and is constant over the entire range of the instrument. If the calibration curve is non-linear, the sensitivity of an instrument should be high and thus the range should not be high in comparison to value being measured.
Scale factor is reverse of sensitivity.
7. Dead Zone
Dead zone is the largest change in input quantity for which a noticeable change in output is observed from zero reading. It may occur due to friction in the instrument which does not allow pointer to move till sufficient driving force is developed to overcome the friction forces. Dead zone is caused by backlash and hysteresis in the instrument.
8. Linearity
The ability of a test to obtain results that varies in a manner directly proportional to changes in the concentration of the substance under analysis, or by a well-defined mathematical transformation. Independent linearity and proportional linearity are the two forms of specifying linearity. If the output just remains within the full scale output without being proportional to it, then it is called as independent linearity. If the output remains proportional to the full scale output then it is called as proportional linearity. If the input-output relation is not linear for an instrument, it may still be approximated to a linear form when it is used over a very restricted range. However, an instrument which does not possess linearity can still be highly accurate.
9. Hysteresis
While testing an instrument for repeatability, it is often seen that input-output graphs do not coincide for continuously ascending and then descending values of the input. This non-coincidence of input-output graphs for increasing and decreasing inputs arises due to the phenomenon of hysteresis. Some causes for hysteresis effect in an instrument are internal friction, sliding or external friction, free play or looseness of a mechanism, etc. Hysteresis effects are best eliminated by taking readings corresponding to ascending and descending values of the input and then taking their arithmetic average.
10. Threshold
Threshold input of the instrument is the certain minimum value of the input below which no output change can be detected.
11. Resolution
If the instrument is being used for measurement, there is a minimum change in the input for which certain detectable change in the output is observed. This incremental change in input is referred as resolution.
12. Span
The range of variable that an instrument is designed to measure is sometimes called the span. Dynamic range is the ratio of the largest to the smallest dynamic input that the instrument will faithfully measure.
