posted 20 October 1998 07:35 AM               

Umm, well - if an independent lab says they're QS certified, ask to see a copy of their certification. See what it says. Get a copy and FAX it to me. I'd love to see it!

QS9000 was targeted at companies who manufacture items for Ford, GM and/or Chrysler. Labs provide a service, not a 'product' per se. To my knowledge an independent lab is not 'certifiable' (a word here - companies actually register to ISO9000 [and qs9000 for that matter since a qs registration is an ISO9001 registration with additional stuff]. Yes - everyone says they are 'Certified' but they are [call me Clinton for the use of definitions] really REGISTERED. That's why they're called Registrars - they register companies. They don't certify a thing.)

There are places to contact and you can be audited to their 'spec'. The A2LA folks are at: http://www.a2la.org/ if you're a lab (what does your company do?).

NO - you do not have to be 'certified' to do your own calibrations, but you have to have a documented system and calibration procedures. You must also be ready to show evidence that your standards masters are traceable, etc. Standard cal stuff.

posted 29 October 1998 04:15 AM               

BUT - if their standards are not traceable to NIST and/or they are not in compliance with generally accepted calibration laboratory standards (have cal procedures, etc.) you have a problem. These are two different issues.

By the way - as far as I know, there is no such thing as a Laboratory being NIST *certified*. I may be missing something but as far as I know there is no NIST certification for a cal lab - only for standards. Nor is there a QS registration (or certification) for calibration labs. Labs within QS registered firms are considered *qualified* but not for outside testing or calibrating other companies equipment.

An FYI:

-----snippo-----

Subject: RE: Q: Certificate of Calibration/Guerra/MacKenzie
Sender: "J. H. MACKENZIE"

> Guerra Asks:
"Some times, I must send the equipment calibrate in manufacturer, since I
can't find accredited laboratories with capacity to calibrate the
equipment. In these cases, I ask the calibration certificate to indicate
standards used and traceability, measured values and calibration
uncertainties.

Do you think this information in calibration certificates is enough, too
> much or too less?"
>- - - - - - - -

This information should be sufficient. There is no requirement to use accredited labs for calibration as long as measurements are traceable to national or international standards.In fact you can do the calibrations yourself if you have the facilities and equipment.

Regards,

Jim Mackenzie
Consultant, Medical Devices Lead Auditor.
QUALIMEDD LTD.
http://www.qualimed

posted 29 October 1998 04:52 AM               

-----snippo-----

Miguel -

I have seen many cases where the manufacturer's certificate that comes with an instrument is not compliant with the calibration requirements of the standard. It is really only a certificate of conformance, rather than a certificate of calibration, and certifies only that the instrument meets the manufacturer's published specifications. When you back track, this is based on a statistical sampling technique, and a true traceable calibration cannot be shown. This may be what your auditor was seeing.

On the other hand, an accredited lab is usually extreme overkill, for shop instruments especially, as well as being economically unjustifiable. You generally don't need the level of uncertainty available from an accredited lab for shop floor or general inspection instruments. You are better off putting in the time and calculations up front when setting up your system (or afterwards, as you make new purchases), to determine which instruments need what level of calibration, and basing your calibration services - both internal & external - on those requirements.

My experience in setting up and running calibration labs is that a large percentage of registrar auditors really do not understand calibration and metrology, and therefore self-impose requirements which have nothing to do with the standard or its practical implementation. Many of these requirements have everything to do with making the auditor's job easier & absolutely nothing to do with ensuring the quality and consistency of the product, which after all, is what this whole thing is about.

Hope this helps.