MU (Measurement Uncertainty) in an Internal Calibration Laboratory

Charles Wathen

Involved - Posts
Hi everyone,
I work in an internal calibration lab, where we perform about 85% of the calibrations in house on instruments (calipers, mics, etc.) and equipment (ovens, water baths, etc.). We currently do not deal with MU on our calibrations that we perform in house. So this lead me to think about an approach for when it may be required.

Going back 50 years when the 4:1 rule was king, this is what we still go by at our company. I was thinking why would you want to waste time and money on a MU evaluation for a calibration that is at 20:1 ratio? So this started me thinking that maybe a possible solution would be to target those calibrations that are 5:1 or less for a MU evaluation?
 

dwperron

Trusted Information Resource
The idea behind measurement uncertainty is to maintain a direct traceabilty of your measurements to national standards. If you do not know what the measurement uncertainties are in each step of the traceability process then you do not have a traceable measurement.
You can have a 4:1 or 5:1 test accuracy ratio (TAR), but if your traceability chain steps have considerable uncertainty components you could end up with a low test uncertainty ratio (TUR). The only way to know is if each step in the calibration chain includes the measurement uncertainty. And I have surely seen some really bad uncertainty contributors introduced, including ones that are greater than the tolerance of the instrument.
As to whether or not it is worth the time and money.... what is the risk of a bad measurement worth to you?
 
Top Bottom