Validation of Calibration Methods and Traceability - ISO 17025

[AaronL]

We are trying to improve a calibration method to meet 17025 standards, and given that the only standard describing the device we want to calibrate (a gyroscope) is very extensive and way beyond the scope of what we want to measure we're thinking of developing/modifying our own method.

Reading through 7.2 "Selection, Verification and Validation of Methods" as well as earlier parts of section 7, it seems that developing non standard methods is allowed as long as you correctly verify/validate them through means such as the systematic consideration of errors or interlaboratory comparison. However section 6.5 on metrological traceability states that a method should relate to some reference material (not relevant in our case) or some existing standard (like the one we would rather not follow) and be traceable through an unbroken chain of calibrations.

So my question is, would developing a method and validating it ourselves in the case where standards already exist still allow us to conform to 6.5?

 

[dwperron]

You are correct in your interpretation of 7.2. You can develop your own method as long as you can prove that you are producing valid results.

As for section 6.5, all of your calibrations will need to demonstrate traceability to a national laboratory standard. In your case, that will mean that all of the calibration standards that you use to calibrate the gyroscopes need to have calibrations that are traceable to a national lab. This is separate from validating your method.

 

[AaronL]

all of the calibration standards that you use to calibrate the gyroscopes need to have calibrations that are traceable to a national lab

The plan is to develop our own calibration table, for which there are no national laboratories who would offer a calibration (there's only one company in the country who is even accredited to calibrate gyroscopes at all), so would this just be a case of ensuring that all of the measuring equipment we use (voltmeters, inclinometers, etc) are calibrated by an accredited organisation?

 

[dwperron]

Correct.
By using traceable instruments in your calibration you have demonstrated traceability of your measurements to a national lab.
Be sure to account for the measurement uncertainties of your standards.