Vetting a potential calibration vendor (supplier evaluation)

[TJM05]

Hi,
I am new to the Cove. Glad to be here...
In our process of ISO registration we are looking into using an outside source for gage calibrations. Are there any specific things I should be looking for in a potential vendor (other than accreditation)? Any specific questions to ask? I'm new to ISO and the science of calibration, so any assistance is appreciated.

Thanks,
TJM

 

[Hershal]

Re: Vetting a potential calibration vendor

The move is completed to obtain further input from other cal users and labs themselves, to hopefully help the poster avoid pitfalls, and to obtain the best solution.

 

[Hershal]

Re: Vetting a potential calibration vendor

Welcome to the Cove.

For a cal lab, besides the accreditation certificate, you need to look at the scope of accreditation. This is the part of the document that describes what calibrations the laboratory is accredited for, and therefore considered competent for, to provide to customers.

A laboratory may be able to provide excellent service on other calibrations also that are not under the scope, but those have not been established formally.

The accreditation may be provided by any of six U.S. based accrediting bodies (ABs), and since you are in NY you may also see our Canadian colleagues (SCC/CLAS). The six in the U.S. are IAS, NVLAP, A2LA, PJLA, L-A-B, and ACLASS.

You might want to also obtain uncertainty with the calibrations. However, under the American National Standards refering to the 4:1 Test Uncertainty Ratio (TUR) is also allowed.

Hope this helps a bit.

 

[Joe222]

Re: Vetting a potential calibration vendor

Welcome to the Cove.

For a cal lab, besides the accreditation certificate, you need to look at the scope of accreditation. This is the part of the document that describes what calibrations the laboratory is accredited for, and therefore considered competent for, to provide to customers.

A laboratory may be able to provide excellent service on other calibrations also that are not under the scope, but those have not been established formally.

The accreditation may be provided by any of six U.S. based accrediting bodies (ABs), and since you are in NY you may also see our Canadian colleagues (SCC/CLAS). The six in the U.S. are IAS, NVLAP, A2LA, PJLA, L-A-B, and ACLASS.

You might want to also obtain uncertainty with the calibrations. However, under the American National Standards refering to the 4:1 Test Uncertainty Ratio (TUR) is also allowed.

Hope this helps a bit.

Did PJLA ever get recognized for Calibration? I know they were recognized for accrediting Testing labs, but were still working on getting recognized to accredit Calibration. This was a few months ago, maybe they are now.

 

[Joe222]

Re: Vetting a potential calibration vendor

I just looked it up and they are now recognized for calibration as of May 31.

 

[TJM05]

Re: Vetting a potential calibration vendor

You might want to also obtain uncertainty with the calibrations. However, under the American National Standards refering to the 4:1 Test Uncertainty Ratio (TUR) is also allowed.

.

Please explain "refering to the 4:1 Test Uncertainty ratio". I'm a newbie to this stuff, so please forgive my over inquisitiveness, lack of understanding of terminology, etc.
TJM

 

[Joe222]

Re: Vetting a potential calibration vendor (supplier)

TUR is taken from the NCSL/ANSI Z540. The 4:1 ratio is that the standard used should have an uncertainty 4 times better than the unit under test.

For 17025 Accredited calibrations, you need the measurement uncertainty.

 

[Hershal]

Re: Vetting a potential calibration vendor (supplier)

TMJ05, no sweat, we are here to help.

Joe222 is right in that it is a ratio. This used to be known as Test Accuracy Ratio or TAR.

The definitions have changed as well. Joe is also right about Z540, which has three versions.

ANSI/NCSL Z540-1-1994 (now retired) in Paragraph 10.2.b describes, allows, and for some applications requires use of the 4:1 TAR. This Standard is still in common use and is likely to remain so for the foreseeable future. This Standard is literally the combination of two previous Standards. ISO Guide 25 takes care of Part 1 and MIL-STD-45662A takes care of part 2, both long since retired.

ANSI/NCSL Z540-2-1997 is also known as the "U.S. GUM" or Guide to the Expression of Uncertainty in Measurement, and while exceptionally informative and technically correct, it is also the most boring document in the history on Mankind in my opinion. But for Metrology professionals, Z540-2 or the ISO GUM is indispensable.

ANSI/NCSL Z540.3-2006 is the current replacement for Z540-1 and Clause 5.3 (and sub-Clauses) describe the Test Uncertainty Ratio known as TUR.

Joe is perfectly correct for TAR, but the TUR is now defined as "The ratio of the accuracy tolerance of the unit under calibration to the uncertainty of the calibration standard used." (where: Uncertainty is expressed as expanded uncertainty at approximately 95% confidence) And this is a two-sided approach to account for various ways of expressing tolerances.

Now, if this is confusing, don't stress, you are not even remotely close to being alone, many are.

If your organization is open to assuring ou have the training that you will find useful, on a short notice, there is training in uncertainty at NCSLI in about 12 days in San Antonio, website: www.ncsli.org

If like most they prefer to plan a bit, then other organizations also provide training in ANS/ISO/IEC 17025:2005 and uncertainty among other topics. There are calendar entries and notes describing upcoming training.

Hope this helps.