Measurement Uncertainty (ISO 17025) - Seeking Lay Person Understanding

[cpine]

I am a little confused. With respect to measurement uncertainty, if an outside lab is performing calibrations, is it necessary to perform them internally.

The reason I ask is I am not quite sure that I completely understand the requirement.

Can someone offer an explanaton that offers the lay person a better understanding.

We are looking at ISO 17025. The company is a distributor and as such has a inspection lab of which once again we source our calibration.


Sincerely,

C. Pine

[Miner]

Could you clarify your situation? Is your company the gauge user that is outsourcing calibration to a company that subcontracts the actual calibration? Or receiving gauges for calibration that you then subcontract? Or something totally different?

[Randy Stewart]

You do not have to calibrate again, however the Uncertainty given by the calibration lab must be reported.

[cpine]

The situation is that I am sending test equimpent out to be calibrated.

This is a practice that is followed by all corporate business units, and what concerns is that those that are ISO 17025 registered clearly define measurement as being performed. They actually include the formula in the quality manual (which is what is confusing me). If one sources calibration, why include a procedure?

I believe that your answers are correct however I'm looking for ammunition for corporate.

Thank you so much for your time and help


Chuck

[Jim Wynne]

Originally Posted by cpine

The situation is that I am sending test equimpent out to be calibrated.

This is a practice that is followed by all corporate business units, and what concerns is that those that are ISO 17025 registered clearly define measurement as being performed.

You're saying that your company has business units that are 17025 accredited, and "clearly define measurement as being performed?" Did you mean to say "calibration" is being performed?

Originally Posted by cpine

They actually include the formula in the quality manual (which is what is confusing me). If one sources calibration, why include a procedure?

The formula for what?

Originally Posted by cpine

I believe that your answers are correct however I'm looking for ammunition for corporate.

You need to help us to help you by clarifying the issue that's concerning you. It sounds like what you're saying is that calibration is being outsourced by business units that are perhaps qualified to do the calibrations internally. What is corporate doing that you need to correct?

[cpine]

I do apologize for being less that clear.

Allow me to start over.

Even tho all business units are required to source calibration, the majority of those units state that measurement uncertainty is being performed by each unit in addition to the accrediated lab they are currently using for calibration.

Once again and as always, Thanks for all of your support and guidence

Chuck

[Jim Wynne]

Originally Posted by cpine

I do apologize for being less that clear.

Allow me to start over.

Even tho all business units are required to source calibration, the majority of those units state that measurement uncertainty is being performed by each unit in addition to the accrediated lab they are currently using for calibration.

Once again and as always, Thanks for all of your support and guidence

Chuck

OK, thanks for the clarification. Do they actually calculate uncertainty, and you feel the effort is redundant? If so, and you want to convince the bosses that they should stop, the best way to do that is to estimate how much it costs for them to do it and demonstrate that it's unnecessary because the outside sources are already doing it. Most bosses hate wasted money, but they might just get a glazed, far-away look in their eyes if you start using Greek phrases like "measurement uncertainty."

[Miner]

Originally Posted by cpine

I do apologize for being less that clear.

Allow me to start over.

Even tho all business units are required to source calibration, the majority of those units state that measurement uncertainty is being performed by each unit in addition to the accrediated lab they are currently using for calibration.

Once again and as always, Thanks for all of your support and guidence

Chuck

I am still a little unclear about your situation, but hopefully I am on-target with the response.

You appear to be outsourcing the calibration of your gauges, and your corporate requires that you calculate the measurement uncertainty of the gauges.

If this is correct, the following should answer your question.

Measurement uncertainty is a combination of uncertainties from many sources. One part of this would be provided by the external calibration provider. This part is that uncertainty inherent to the gauge itself and how it was calibrated.

However, there are other factors outside of the control of the calibration provider that you must quantify yourself, such as repeatability, reproducibility, effect of temperature deviation from standard conditions, etc. This is probably where your corporate formula comes in.

I have attached an article that addresses the general concept.