Calibration of inspection, measuring and test equipment

[Angela-2007]

Help!

I have been auditing our inspection and measuring equipment. I went to look to see if the calibration co. we were using, were Guide 25. Then I remembered there had been something in the July interpretations. When I looked into it I found that it said that "The requirements of cl. 4.10.7 and 4.11.2.b.1 apply only to production or service parts or production materials released by the customer for purchase or manufacture, including all testing for PPAP requirements. this excludes testing for parts or materials under design or development" What the heck does this mean. We utilize a humidity cabinet, peel strength, oven for researching materials for some of the labels we make. Do these need to be calibrated? Or do they not need to be calibrated if we use them for simply testing the materials?

Someone please help me with this requirement!!!!

Angela

 

[Jerry Eldred]

I can't give you an official interpretation of how Guide 25 (ISO 17025) applies in your case. I can tell you how I would handle it (as I am in a QS9000 context).

If the accuracy of any setting or reading on any of them have potential impact on a deliverable to the customer, I would calibrate them. The only reason I would not calibrate any of them would be if setting accuracy had no impact. If the oven (for example) was set for 100 degrees C, and had a tolerance of +/-5 degrees C, and I set the oven for 80 degrees C or even 120 degrees C (for example) and it had virtually no impact that could in any way end up as a customer dissatisfier, I would consider not calibrating.

I am a big fan of relating calibration practices to what a VP level person I highly respect said once -- "does it pass the common sense test?" I am even proactive in explaining to my customers that we don't calibrate instruments to meet QS9000 or Guide 25 or ISO17025. We calibrate instruments to be sure they make the measurements we expect them to. I further have said that if QS9000 went away, and we made that political connection based on that standard, management could easily think we exist to fulfill the requirements of a standard. And when the standard goes away, that we don't need to calibrate any more.

My reommendation would be to ask some questions about the setting or measurements on the instruments. On a peel force tester, if the peel force matters, then it needs to be calibrated. On the humidity cabinet, if the humidity settings matter, then it needs to be calibrated. And the same criteria for the rest.

I find myself regularly falling back to my calibration roots. I don't calibrate anything to meet a standard. My job is to make sure any quantitative setting or measurement made is within expected limits, to an expected confidence level. And as the various requirements come and go....

MIL-C-45662
MIL-STD-45662
MIL-STD-45662A
ISO-10012
Guide 25
ISO17025
ISO9000
QS9000
TS16949
TS16949 2nd Ed.
ISO9000-2000
and so forth...

The basic needs for calibration have not changed. The real changes have been at a documentation level, and standardization of how methods, skills, etc. are established and documented. I'm still making the same decisions based on the same criteria as to what should be calibrated, how often, using what method, using what standard.

Review each of your items as I described above, and revert back to the basics. I tend to lean toward calibrating if there is the possibility of quality or reliability issue. I see at least the potential in your circumstance. If you are selling labels to external customers, or your labels are going on deliverable product to an external customer, and there are any performance specifications or criteria for those labels, I would definitely lean toward calibrating those instruments. I also strongly lean toward calibrating instruments used in design. That is sometimes as critical as production instruments, as engineers make design decisions based on their readings.

 

[Angela-2007]

Jerry;

Thanks for your help. One more questions. We use outside services to calibrate. Those companies need to be Guide 25 certified, QS compliant or certified. Is this correct?

Angela

 

[Atul Khandekar]

Jerry Eldred said:
I am a big fan of relating calibration practices to what a VP level person I highly respect said once -- "does it pass the common sense test?" I am even proactive in explaining to my customers that we don't calibrate instruments to meet QS9000 or Guide 25 or ISO17025. We calibrate instruments to be sure they make the measurements we expect them to. I further have said that if QS9000 went away, and we made that political connection based on that standard, management could easily think we exist to fulfill the requirements of a standard. And when the standard goes away, that we don't need to calibrate any more.

..And I am a big fan of yours Jerry and always look forward to your responses in this forum! Could you please explain what exactly do you mean by the 'common sense test' please?
Thanx,
-Atul.

 

[Jerry Eldred]

To be clear, my background is 25 years in the calibration world, including military, starting up a couple of labs, working for a third party lab, and working for four different Fortune 100 companies, currently as a metrology engineer. My expert skill is not strict interpretation of the standards documents (although I am lately spending far too much time studying in that area)..

Depending on which standard your company has to comply with (I'll assume it is QS9000 (soon to be replaced by TS16949), your calibration subcontractors fall into a narrower category than general subcontractors. A third party calibration lab should be accredited to ISO17025 (replacement for Guide 25 - which is now obsolete, I believe). I would not even use "QS compliant" as a criteria for a calibration subcontractor. Both QS9000 and TS16949 refer back to ISO17025. As long as your calibration subcontractor is accredited to ISO17025, that is normally considered acceptable.

I'm going to have to bite my lip and refrain from skeptical comments. I'll just say that in the use of third party calibration subcontractors, make sure they are accredited, not just compliant. I have quite a lot of experience in that area, and try to make sure they are adequately capable of the calibrations I need done. I recommend looking at the scope of accreditation used by a given lab to be sure they have scrutinized in the area in question. I've gone so far as to have the same instrument calibrated by two different vendors and compared the results to provide confidence in the results they delivered.

 

[Ken K]

Couple of things Angela:

Guide 25 has been replaced by ISO 17025.

Outside calibration labs need to be ISO 17025 certified.

They cannot be QS.


AS far as your original question which Jerry pretty well covered for you, ask yourself this question...

Can I really rely on test data off equipment that is not calibrated?

I think the answer is obvious. The risks are not really worth the headaches that will follow.

 

[Angela-2007]

You guys are wonderful! Thanks for the help! I'm sure I will be picking your brains again, soon!

Angela

 

[Jerry Eldred]

ATUL: I'll be happy to explain that.

I have worked with many different people. Some of them not very good, and some of them very good people, with very good understanding of what matters in business.

When you work in a big company, and are at a high position in the company, there are many people who present proposals to you for new ideas. This person is a Vice President. Since there is a company confidentiality implication, I won't say any more details about the person.

As a Vice President, sometimes people present things to you just to try to impress you. You may have seen people who will come up with ideas, that the main reason for the idea is to try to impress a higher manager. This Vice President is very intelligent. If you present something to him, he looks at all of the details, and you have to convince him that it is an intelligent idea. You have to convince him that it isn't something just to produce nice looking numbers on a chart, but that it really makes a positive difference.

I have been in meetings with this Vice President when he looked at a proposal that looked pretty, but didn't actually make a real difference for making better products for the customer. He has asked the question, "does it pass the common sense test?" sometimes when it wasn't clear whether an idea was really valuable, or just something to make nice numbers on a chart.

So in the case of the calibration question in this thread, when deciding whether or not to calibrate something, if I am not sure how it complies with QS9000 or TS16949, or other standards, I ask questions about whether it complies with common sense. A multimeter that is only used by maintenance technicians to check facilities equipment may not need to be calibrated to meet QS9000, but common sense says that the maintenance technician may use that meter to check whether there is electricity flowing in a circuit. If the meter has not been calibrated, it may comply with QS9000, but common sense says that it should be calibrated periodically so that the technician can trust his life in using it.

There are so many people in business who spend a lot of energy trying to do what management wants. Sometimes management does not know the right answers. And lower management tries to come up with answers that upper management will like, without thinking about whether it passes the common sense test.

This same Vice President was watching a presentation by a manager. The manager presented some not very intelligent material. The Vice President asked questions about the material that the manager couldn't answer. The manager made the remark, "I thought that was what you wanted." The Vice President (because he is an intelligent manager) wanted most to hear the truth, rather than what the manager thought would make him happy. That presentation didn't pass the common sense test. I often tell my management that my job is to do my best to present you the truth whether you like it or not. Then your job is to decide what to do with it.

I hope that gives you some idea of what I meant.

 

[Atul Khandekar]

Jerry Eldred said:
...I hope that gives you some idea of what I meant.

Yes Jerry, Thank you very much.
Rgds,
- Atul.

 

[Graeme]

Jerry Eldred said:

... A third party calibration lab should be accredited to ISO17025 (replacement for Guide 25 - which is now obsolete, I believe). ... Both QS9000 and TS16949 refer back to ISO17025. As long as your calibration subcontractor is accredited to ISO17025, that is normally considered acceptable.

The Foreword of ISO/IEC 17025:1999 says specifically that it cancels and replaces ISO/IEC Guide 25:1990.

I would also strengthen what Jerry said -- the calibration subcontractor should be accredited to ISO/IEC 17025, AND their accredited scope must include the measurement area and range of interest.

Another test of the need to calibrate is the one I saw put forward by Phil Stein a few years ago: "Does it matter if the measurement result is wrong? If it does, then calibrate the instrument. If it doesn't matter, then why are you making the measurement?"

Graeme