THE MAGAZINE FOR COMPUTER APPLICATIONS
Circuit Cellar Online offers articles illustrating creative solutions
and unique applications through complete projects, practical
tutorials, and useful design techniques.

DETERMINING MEASUREMENT ACCURACY

by Hristo Stefanov

Start ý Defining the Terms ý A Little Theory ý Sources of Error ý Error Budget ý Error Budget Example ý Offset and Gain Errors ý Sources and PDF

The vast invasion of embedded systems in the real world engenders the need of measuring different signals. Even a battery charger needs to measure the temperature of the cell. But if you measure a signal without knowing the error, you are doing nothing, because every measurement is done with some amount of error. Someone can tell you that the ambient humidity is 70% RH. From this message, you donýt get any information about the accuracy. Imagine taking a measurement with a horse filament!

Usually you would say that the ambient humidity is 70% RH with maximum error of ± 2% RH. Even the error cannot be specified without error. So the most correct way is to say that at 95% probability, the ambient humidity is in the range 70% RH ± 2% RH. This way, you get to know the uncertainty of the measurement.

The uncertainty is a figure of merit associated with the actual measured value, the boundary limits within which the true value lies. Contributors to this potential for inaccuracy include the performance of the equipment used to make the measurement, the test process or technique itself, and environmental effects. The assessment of uncertainties of measurement is a task more suited to a mathematician rather than the average engineer. That is why most of us use the term error. By definition, errors are known and can usually be taken into account by correcting measured values, whereas uncertain measurements merely define the limits of potential inaccuracy.

NEXT