Verification instead of calibration?(ISO 9001:2008)

[Kchnwtch]

Hi! First, not trying to open up old definition wounds (is it calibration if you can't adjust, etc). I promise I've read through all the posts on this, but if I've missed one, feel free to refer me to it.

We're coming up on our first surveillance audit ISO 9001:2008. 7.6(a) says "measuring equipment shall be calibrated or verified or both....the basis used for calibration or verification shall be recorded."

Rather than get our CMM arm re-calibrated (lots of money) we want to do an in-house verification of its calibration. (It was calibrated last year.)

According to 7.6(a) we can do this, if we have recorded the basis we are using for our verification, right? So if we can prove that the arm gives accurate measurements for the parts we need it to measure, and we've written up how we verified that, then we're good...right?

But we still won't be good with the folks who CALIBRATED the arm, right? The certificate will still say "calibration expires on xx/xx/xx." Will that make a difference to the auditor if we can verify the calibration is still accurate? Thanks for your help!

 

[howste]

What will you use to verify the CMM? You need to use calibrated standards, traceable to NIST (or other national standards).

Standards do exist to calibrate a CMM. To me it sounds like you're trying to "weasel" out of using calibrated standards by leaving out some of the important text in the requirement:

...calibrated or verified, or both, at specified intervals, or prior to use, against measurement standards traceable to international or national measurement standards; where no such standards exist, the basis used for calibration or verification shall be recorded...

You can choose an appropriate calibration/verification frequency that is different from the calibration lab's certificate. It should be based on risks and frequency/usage at your facility. There is no one-size-fits-all frequency for calibration of CMMs.

 

[Kchnwtch]

We do have traceable to NIST standards we can use, so the only weaseling I'm trying to do is weasel out of $2K's worth of calibration of a tool that has probably maintained its calibration, and to figure out how the auditor would see the verification as opposed to the calibration. Thanks!

 

[howste]

If you have the standards and the knowledge of how to properly do it, then you can do it. There is no requirement that an outside lab do it. If you verify it with traceable standards and you maintain records of that, then you've met the requirements.

 

[Howard Atkins]

The certificate from the lab if it ISO 17025 should only show an expiration date if you gave them one.
They are not qualified to give such a date
see
NCSL International

Calibration Due Dates

The 17025 is more stringent than the Z540-1 on the placement of calibration due dates on calibration certificates and calibration labels. Although manufacturers may designate time periods over which products should perform without degradation, the 17025 makes clear that it is the responsibility of the end-user organization to determine the appropriate calibration interval under the requirements of its own quality system.

 

[JoShmo]

What will you use to verify the CMM? You need to use calibrated standards, traceable to NIST (or other national standards).

This has been an industry recgnized methid for 30 years http://www.renishaw.com/en/mcg-systems--6674

Not calibrated...

 

[JoShmo]

W

Standards do exist to calibrate a CMM. To me it sounds like you're trying to "weasel" out of using calibrated standards by leaving out some of the important text in the requirement:

How can you tell that? Fact is a cmm can probably go 5 years before recal. Yearly is rediculous. A granite table and arms/scales are going to go out of cal in a year? Get an MCG and use that to verify the volumetric accuracy - job done!

 

[Mike S.]

Get an MCG and use that to verify the volumetric accuracy - job done!

Our CMM guys do that and it is fine with the auditors of all stripes.

 

[howste]

How can you tell that?

Obviously I couldn't - because Kchnwtch already confirmed this was not the case.

The rest of your post agrees with what I already posted.

 

[Eredhel]

You can also choose how often to have it done. I have large surface plate that hadn't been checked in years. We checked it and it was spot on. So I bumped it to two years. If it's still good then I'll bump it back to three.

It's up to you how often you do it. Unless there is new language in the 2015 version I've missed.