As previously defined, refers to a calibration check and adjust its output of an instrument to correspond exactly (or proportional) to its input through a specific range. To calibrate an instrument, we must know the input quantities and / or output associated with the instrument under test. A device used as a benchmark to compare their response to the response of an instrument is called "calibration standard" or "pattern". In simple words, a pattern is something we can use to compare a calibrated instrument. Therefore, any calibration can only be as good as the standard we use.
Patterns or "calibration standards" can be divided into two categories: patterns used to "produce" an accurate physical quantity (eg pressure, temperature, voltage, current, etc..), And patterns used to simply "measure" a physical quantity with a high degree of accuracy. An example of the first category is the use of boiling water (at sea level) to "produce" a temperature of 100 degrees Celsius (212 degrees Fahrenheit) to calibrate a temperature indicator, while an example of the second category would be using an accurate thermometer to measure a laboratory arbitrary source temperature compared with the temperature indicator that we are calibrating.
Metrology laboratory, the latest standards are based on the fundamental constants of nature, and are called intrinsic rules. A modern example of an intrinsic standard for time is called the atomic clock uses cesium atoms to produce isolated frequencies which are inherently fixed and reproducible worldwide. Instrumentation workshops within the industries or factories could not really afford the costs associated with having intrinsic rules, so you must resort to other devices for calibration purposes. Ideally, it is a chain of calibration for any device from the instrument shop to an intrinsic standard or primary metrology laboratory in the country.
Standard instruments used for calibration of instrumentation workshops should be sent periodically to metrology laboratories for re-calibration or re-standardization, where its accuracy is checked by comparing it with another (better accuracy) pattern which are compared with other calibration standards much higher level and ultimately contrasted with the intrinsic rules. In house calibration step "chain", there is a progressive degree of uncertainty. Prose Standards fewer intrinsic uncertainty, while field instruments (eg pressure transmitters, temperature, etc..) Present the greatest uncertainty.
It is important that the degree of uncertainty in the accuracy of a standard instrument is significantly lower than the uncertainty we expect to have in that we are calibrating the instrument, otherwise no point calibration. This ratio is called Test Uncerainty uncertainties Ratio (test uncertainty ratio) or TUR. A good rule is to keep at least TUR 4:1 (ideally would be 10:1 or higher), where the standard instrument is many times more accurate (less uncertainty) field instruments are calibrated with the same.
I had personally witnessed the confusion and wasted time trying to calibrate instruments generated for a small field while the instrument uncertainty is an uncertainty similar pattern. In one case, the instrument attempts to calibrate a tire pressure with a tolerance of + -0.25% of span using a pattern that was only + -1% over the same span.
What we learn here is that we always make sure that the pattern is used to calibrate reliable accuracy (enough). No standard instrument is perfect, perfection is not what we need. Our goal is to be accurate enough for reliable calibration within the specified limits.
The following articles discuss more standard instruments used in calibrating instrumentation workshops.
