Can I use the CMM after the Calibration has expired?

[jleond]

The CMM calibration expires 7-19-08 and the service provider will start the new 2008-09 calibration on the week of 7-28-08. My question is: how long can the CMM be used to measure parts after the due date? I heard that we can use it (legally) 30 days after due date. Is this true? Is there a standard that states this or something related to this?

Thanks in advance for your help.

[Coury Ferguson]

Originally Posted by jleond

The CMM calibration expires 7-19-08 and the service provider will start the new 2008-09 calibration on the week of 7-28-08. My question is: how long can the CMM be used to measure parts after the due date? I heard that we can use it (legally) 30 days after due date. Is this true? Is there a standard that states this or something related to this?

Thanks in advance for your help.

In my opinion, I don't see a problem with it. I can't address the 30 days after rule.

A Suggestion: Change you Calibration Procedures, stickers and such to only reflect Month/Year only (example: 07/2008). This will help you in the long run.

[BradM]

Hello there!

In general, calibration intervals are much more of an Art than a Science. The perfect calibration interval is that which find error 50-75% of the unacceptable point, at least that is my opinion.

To your point, it would all depend on what your quality management system states. Do you have procedures that states when calibration is past due?

Without anything documented in your system or anything in writing, it does not look good to use any instrument beyond its due date.

The best thing for you do do is establish a calibration procedure for when instruments are recalled for calibration (like month/year), using an instrument past its due date, tagging, etc. Like in your example, I would make the due date 07/08. That way you have the month to get the work done.

Hopefully that helps.

[BradM]

Originally Posted by Coury Ferguson

In my opinion, I don't see a problem with it. I can't address the 30 days after rule.

Actually, I agree with you. But... there are many auditors who are not as processed-oriented as they should. They would see 7/19/2008, look at their watch and see 7/22/2008, the absence of any procedural exception, and they start writing.

The best thing to do is write up some exception/ some report stating that your organization says it's OK to use, IMHO.

As far as Rule of Thumbs (30 days, etc.), I have seen them all over the place, and again depends on the industry and the people involved.

[Coury Ferguson]

Originally Posted by BradM

Actually, I agree with you. But... there are many auditors who are not as processed-oriented as they should. They would see 7/19/2008, look at their watch and see 7/22/2008, the absence of any procedural exception, and they start writing.

The best thing to do is write up some exception/ some report stating that your organization says it's OK to use, IMHO.

As far as Rule of Thumbs (30 days, etc.), I have seen them all over the place, and again depends on the industry and the people involved.

If an Auditor is choosing to be that picky, they need to re-evaluate their techniques. We are talking a couple of days, still within the same month of calibration due date.

I would challenge any Auditor that would write this up as an NC.

[Stijloor]

Originally Posted by Coury Ferguson

If an Auditor is choosing to be that picky, they need to re-evaluate their techniques. We are talking a couple of days, still within the same month of calibration due date.

I would challenge any Auditor that would write this up as an NC.

If I were the auditor, I would look at other instances where devices exceed the stated calibration due date (systemic issue). If this is indeed the case, I would write an NC.

Now, having said that, if you allow deviations from an established requirement, even if it's only one device, where does one "draw the line?"

What do you think?

Stijloor.

[Coury Ferguson]

Originally Posted by Stijloor

If I were the auditor, I would look at other instances where devices exceed the stated calibration due date (systemic issue). If this is indeed the case, I would write an NC.

Now, having said that, if you allow deviations from an established requirement, even if it's only one device, where does one "draw the line?"

What do you think?

Stijloor.

Alright I see you point. I don't totally agree with, but I understand where this is coming from. (added: if it is a systematic problem then I would identify it as a Major NC, but I think we are talking about one CMM)

My question back is: Is there not a master diametrical sphere to check the accuracy of the CMM? And this diametrical sphere is calibrated to be accurate within millionths or even tenths.

That is why I suggested only identifying the stickers with Month/Year would help in the long run.

[AndyN]

Firstly, there's nothing 'legal' about it - literally or metaphorically. You should have past data that shows for each calibration, what the 'as found' condition was. Compared to the specification limits and the tolerance of the features measured on the cmm, you should be able to write yourself an 'extension' to the recall, based on DATA and the decision you made, based on that.

If your external auditor gives you grief, roll up the deviation, coat the end in Vaseline, and.........

Of course, make sure your extension form is a) mentioned in your procedures etc and b) is a controlled document.

Go for it!!