Internal calibrations - Part of an ISO 17025 accredited testing laboratory (Automotive)

[Sam Fisher]

Hi all.
Looking for some enlightening.
I'm working for a automotive giant in a testing lab as a quality engineer.
This lab provides testing for the plant. To be more precise, we are simulating real life conditions for different car parts. The lab is 17025 accredited with a declared scope of Testing, by the national accreditation organization. As you all may know, as an accredited lab you have to use traceable measuring instruments.
In a continuous search for optimization and cost efficiency I was thing that we might be able to perform some calibrations internally. Now comes the interesting part. We are already performing those type of calibrations, with traceable reference standards, working procedure and uncertainty calculation, all of this not as a official calibration, more as intermediate check in-between third-party calibrations.
Our national accreditation institution states that we have to acreditate separtly in order to able perform those calibrations. But I have different opinion on that.
Since we have already all of the procedures and experience and we already have those audits each year, why we can not incorporate the calibrations in our base qms? Of course with all special requirements:
Working procedure
Qulified personnel
Measurent uncertainty budget
Use of traceable working standards.
Audits on techinal part
Interlabs
Etc.
Dose anybody here have any kind of experience on this subject? Or even a hint.

Thank you all!

 

[dwperron]

This used to be an easier question to answer.

In 17025:2005 it is crystal clear:
"5.4.6.1 A calibration laboratory, or a testing laboratory performing its own calibrations, shall have and
shall apply a procedure to estimate the uncertainty of measurement for all calibrations and types of
calibrations."
So it was obvious that a testing lab was allowed to calibrate their own equipment. Of course, it would be necessary to meet all of the requirements mandated for a 17025 calibration laboratory. It's just that calibration would not be part of your accreditation, and you thus could not calibrate for an outside party.

In 17025:2017 it gets a bit murky:

"7.6.2 A laboratory performing calibrations, including of its own equipment, shall evaluate the
measurement uncertainty for all calibrations."
Unfortunately, this verbiage is not as clear regarding a testing lab performing its own calibrations, but the intent is still there. After all, why would they mention "including of it own equipment" if they we only referring to calibration laboratories?

So I would agree with your position.
Good luck!

 

[Sam Fisher]

This used to be an easier question to answer.

In 17025:2005 it is crystal clear:
"5.4.6.1 A calibration laboratory, or a testing laboratory performing its own calibrations, shall have and
shall apply a procedure to estimate the uncertainty of measurement for all calibrations and types of
calibrations."
So it was obvious that a testing lab was allowed to calibrate their own equipment. Of course, it would be necessary to meet all of the requirements mandated for a 17025 calibration laboratory. It's just that calibration would not be part of your accreditation, and you thus could not calibrate for an outside party.

In 17025:2017 it gets a bit murky:

"7.6.2 A laboratory performing calibrations, including of its own equipment, shall evaluate the
measurement uncertainty for all calibrations."
Unfortunately, this verbiage is not as clear regarding a testing lab performing its own calibrations, but the intent is still there. After all, why would they mention "including of it own equipment" if they we only referring to calibration laboratories?

So I would agree with your position.
Good luck!

Thank you so much. That is very encouraging. I mean, we acknowledge that we have to be evaluated for this part to. But why the need for a separate accreditation? From my perspective this is a birocratic thing.
Do you have a any references from other accreditation institutions that are addressing this issue?

 

[dwperron]

The folks at Eurolab say "This clause has remained almost unchanged "
https://www.aphl.org/programs/food_...ments/EUROLAB Handbook ISO IEC 17025 2017.pdf

Canadian Association For Laboratory Accreditation Inc. (CALA) , an organization that accredits testing, not calibration, labs says:
"7.6.2 Internal Calibrations If a laboratory performs internal calibrations on its balances, thermometers and pipettes, it must have a documented procedure that includes the estimation of uncertainty and it must use staff that is appropriately trained. CALA reserves the right to include a calibration expert on assessment teams to laboratories doing their own calibrations. "
https://cala.ca/wp-content/uploads/P07-CALA_Application-2017.pdf
So they speak about testing labs doing their own calibrations.

UKAS states:
"•7.6.2 requires evaluation of measurement uncertainty for all calibrations, including those a laboratory performs on its own equipment (i.e. “in-house” calibrations)"
https://www.ukas.com/download/publi...ditation/ISO-IEC-17025-2017-Presentation.pptx

 

[AllTheThings]

Hey all, I thought I would resurrect this discussion. I appreciate Sam starting this, and the cites from CALA and UKAS.

I am currently starting to get my lab in shape for ISO 17025 accreditation as a test-only lab. We intend to calibrate our own environmental chambers (temperature and humidity) using traceable accredited calibration humidity and temperature probes.

Usually, we send stuff out, but chambers are special. They don't move easily, if at all. They end up on different schedules, given trying to calibrate around multi-month tests. We are far more familiar with the controllers, sensors, and uncertainties than the cal houses tend to be. We also have enough of them that beyond the convenience, the costs start to break even. When we acquire new to us (but used) chambers or conduct heavy maintenance/sensor replacements, we already run a calibration as a check, before having an external provider in for an accredited calibration.

This all leads in the direction of us doing the calibration work in-house.

We have started putting together uncertainty budgets and more formal procedures for the activity.

My questions (edit, plural ): What will an auditor want to see to accept this path? Do we need to note anything special in our documents regarding the scope of the calibrations we do, and the uncertainty approach? Do we need to cite 7.6.2 as justification for not outsourcing the cal? Any impartiality concerns we need to address?

 

[dwperron]

My questions (edit, plural ): What will an auditor want to see to accept this path? Do we need to note anything special in our documents regarding the scope of the calibrations we do, and the uncertainty approach? Do we need to cite 7.6.2 as justification for not outsourcing the cal? Any impartiality concerns we need to address?

You will need to notify your accreditation body that you perform calibrations on some of your equipment. That heads up will allow them to send an assessor familiar with calibration, and with the proper documentation to audit your calibration program.
Yes, you will need to set up a 17025 compliant calibration program. That will include everything that a calibration lab is required to do under 17025, such as is covered in 17025 Sections 6.4 through 6.6. You will need to produce and validate (with records) your calibration procedures as per Section 7.2. Section 7.7 would normally require you to participate in a proficiency testing program, I would check with your accreditation body and see if they would require that or if you can set up your own internal method to comply with Section 7.7. You will need to keep calibration records as per Section 7.8.4.

That's a start
Good luck!

 

[AllTheThings]

dwperron,

Thanks for the insights. That is really helpful for me to consider.

Luckily, we have the structure in place for much of those specifics under our ISO 9001 certification...But still have some gaps. With respect to 7.7...Well, 7.7.2, I have been giving that some thought. I may see if the AB lets us satisfy it under 7.7.2b by comparing the calibrated references we send out to in-house working standards. Otherwise, PT for just temperature and humidity isn't too expensive. I will share the experience towards the end of the year, when we are past the audit.

 

[Jim Wynne]

You will need to notify your accreditation body that you perform calibrations on some of your equipment. That heads up will allow them to send an assessor familiar with calibration, and with the proper documentation to audit your calibration program.

I think there might be some confusion. The OP is preparing for 17025 accreditation.

 

[dwperron]

I think there might be some confusion. The OP is preparing for 17025 accreditation.

He is with a 17025 accredited test laboratory that is looking into calibrating their own equipment. That means he will be assessed for his ability to calibrate his equipment to 17025 as well. He was asking what was involved in that process.

 

[Jim Wynne]

He is with a 17025 accredited test laboratory that is looking into calibrating their own equipment. That means he will be assessed for his ability to calibrate his equipment to 17025 as well. He was asking what was involved in that process.

Please read the 2nd paragraph of the post you quoted.