Calibration Certificate Contents: How to report Measurement Uncertainty?

[howste]

ISO 17025 experts: Does the attached calibration certificate meet the ISO 17025:2005 uncertainty reporting requirements? If yes, please explain how. If no, please explain why not.

In Reply to Parent Post by ISO 17025:2005

5.10.4 Calibration certificates

5.10.4.1 In addition to the requirements listed in 5.10.2, calibration certificates shall include the following, where necessary for the interpretation of calibration results:

b) the uncertainty of measurement and/or a statement of compliance with an identified metrological specification or clauses thereof;

[BradM]

Well, reluctantly I marked yes. Reluctant due to:

1. They list an explanation of k factor and such but don't give an uncertainty
2. They reference a MIL spec that has been dead and gone for years
3. They could have easily listed the tolerances

I don't find where they are non-compliant.

The basis of my opinion is that they say they work against Handbook 44. The handbook lists all the applicable test points/tolerances. In addition the handbook specifies what tolerance weights they should be using (from Handbook 44):

N.2. Verification (Testing) Standards.

- Field standard weights used in verifying weighing devices shall comply with

requirements of NIST Handbook 105-1 (Class F) or the tolerances expressed in Fundamental Considerations, paragraph

3.2. (i.e., one-third of the smallest tolerance applied).

I really think they should list the uncertainty of their weight/weighing process, but I cannot state it as a requirement.

Outside of the compliance issue, I would return this form back as unacceptable. Regardless of 17025, I think a commercial calibration lab can do a little better.

[howste]

In Reply to Parent Post by BradM

Well, reluctantly I marked yes. Reluctant due to:

1. They list an explanation of k factor and such but don't give an uncertainty
2. They reference a MIL spec that has been dead and gone for years
3. They could have easily listed the tolerances

I don't find where they are non-compliant.

The basis of my opinion is that they say they work against Handbook 44. The handbook lists all the applicable test points/tolerances. In addition the handbook specifies what tolerance weights they should be using (from Handbook 44):

I really think they should list the uncertainty of their weight/weighing process, but I cannot state it as a requirement.

Thanks for your response.

So they are not required to state the uncertainty value on the certificate. Are they required to have the uncertainty values recorded somewhere in the company records? Does the k factor statement on the certificate have any purpose without the uncertainty value?

In Reply to Parent Post by BradM

Outside of the compliance issue, I would return this form back as unacceptable. Regardless of 17025, I think a commercial calibration lab can do a little better.

By the way, I'm in no way connected with the lab that produced this certificate. I'm looking for education about what's really required/not required. What do you find unacceptable?

[Hershal]

There is neither uncertainty provided nor an inclusion of an accepted metrological specification, and no reference to the specifications necessary to determine whether the item is in tolerance even though the statement is made that it is.

Not compliant.

[howste]

In Reply to Parent Post by Hershal

There is neither uncertainty provided nor an inclusion of an accepted metrological specification,

Thanks Hershal.

What would be an accepted metrological specification?

In Reply to Parent Post by Hershal

and no reference to the specifications necessary to determine whether the item is in tolerance even though the statement is made that it is.

Not compliant.

I believe the tolerance specifications are listed in NIST Handbook 44. Is that unacceptable?

[BradM]

In Reply to Parent Post by Hershal

There is neither uncertainty provided nor an inclusion of an accepted metrological specification, and no reference to the specifications necessary to determine whether the item is in tolerance even though the statement is made that it is.

Not compliant.

I was hoping you would chime in.

The 17025 does not mandate an uncertainty must be listed. Does it?

The metrological specification is listed in Handbook 44, that states the acceptable tolerances, test points, and other.

The one thing that I don't see that I think Handbook 44 specifies is corner load tests. I don't see any evidence of it.

[howste]

In Reply to Parent Post by Hershal

There is neither uncertainty provided nor an inclusion of an accepted metrological specification

I found this in a post you made in 2004. Is it what you're talking about?

In Reply to Parent Post by Hershal

There is one basic rule. It is that uncertainty must be reported unless there is an accepted Metrological equivalent. In the U.S. the equivalent is the 4:1 rules, specified in ANSI/NCSL Z540-1-1994, section 10.2.b. A U.S. based calibration lab can avoid reporting specific uncertainties IF a 4:1 can be maintained for the affected calibration area, AND the lab can put the evidence on the table for the accrediting body.

[BradM]

In Reply to Parent Post by howste

So they are not required to state the uncertainty value on the certificate.

Thus my question to Hershal. I think you should, but I don't see the must. Seeing how this is being verified to a known, validated standard (Handbook 44), I don't see how uncertainties are required.

In Reply to Parent Post by howste

Does the k factor statement on the certificate have any purpose without the uncertainty value?

The k factor statement is a explanation of the given uncertainty value and it's range; given the stated probability. It is essentially gibberish, without the stated uncertainty.

In Reply to Parent Post by howste

What do you find unacceptable?

Pretty much the entire document. Honestly, I'm answering the question surgically as I read 17025. However, my first response is the same as Hershal's. Reject the document.

Unacceptable? The MIL spec. reference, a k factor with no uncertainty listed, no tolerances given, no statement of accuracy or uncertainty for standards.

This document is better than most I've seen. However, with the 17025 on the certification sheet, they should be turning in a Lexus; not a Kia, to the customer.