I have done some calibration tests and wanted to do some calibration calculations. So can anyone suggest me how to do uncertainty measurements for calibration?
This is the procedure I recommended. (1) Write down your measurement (calibration) model, i.e. the relationship between the calibration quantity and the influence quantities (error sources). In many calibrations, the measurement model is just a sum of several influence quantities. (2) Determine the standard uncertainty (SU) of each influence quantities, which may be Type A (estimated using the data obtained from repeated observations) or Type B (estimated with other means than data). (3) Calculate the combined SU according to the law of propagation of uncertainty (LPU). (4) Calculate the expanded uncertainty (EU) as EU=kxSU, where k is the coverage factor. Usually k=1.96 or 2 is used at the 95% coverage (or confidence), assuming that the normality holds according to the Central Limit Theorem. For more discussion about uncertainty estimation, please refer to Article A unified theory of measurement errors and uncertainties
To obtain the uncertainty, a typical rule of thumb is to take one-half the unit of the final decimal point in a measurement. Uncertainties Must Be Stated Using This Rule - The experimental uncertainties should be expressed to a one-separable figure.
Have a look: https://www.isobudgets.com/7-steps-to-calculate-measurement-uncertainty/
The uncertainty analysis should follow specific standards, such as "ISO/IEC Guide 98-3:2008 - Uncertainty of measurement".
For a brief analysis, some statistical parameters, such as the Min-Max Range, the Mean Value, the Standard Deviation, and the Type A Standard Uncertainty (Standard Error of the Mean) can be used. More details can be found my below paper:
Article Improved Calibration Technique for Two-Probe Setup to Enhanc...
Your question depends on the type of instrument used in the measurement, as the uncertainty value is affected by the conditions surrounding the measuring device.
Are they standard conditions in a laboratory to which ISO 17025 procedures apply, or is the measurement process performed in non-standard conditions as these variables are taken into account (such as temperature, humidity and atmospheric pressure).
Also, the uncertainty value depends on the repeatability readings at a certain point, and the resolution value of the instrument.
And, do not forget the uncertainty value of the reference instrument used in the calibration process.
Therefore, please clarify the type of measuring device used in the calibration process so that we can assist in the process of calculating the uncertainty
The uncertainty analysis should follow specific standards, such as "ISO/IEC Guide 98-3:2008 - Uncertainty of measurement" available for free from the BIPM website. Other useful documents include UKAS document M3003 which is a free download.
NPL yhe UK's NMI produces a free good practice guide on measurement uncertainty (www.npl.co.uk) and there is some e-learning that was offered for free during COVID and may still be available.
You should review the Measurement Uncertainty Guide (GUM) to have the procedure on how to obtain the measurement uncertainty. If you need something more specific about certain magnitudes, you can check the EURAMET guides or some INM guides.
Probably, the best reference document you can read is the GUM: "GUM: Guide to the Expression of Uncertainty in Measurement" as it has been said before. But if you prefer to read a shorter document, an alternative could be :
EA-4/02 M - European Accreditation (european-accreditation.org)
"Evaluation of the Uncertainty of Measurement in Calibration"