ISO17025 for In-House Calibration Labs: Is it required?

[Charles Wathen]

I work for a company that has it's own internal calibration lab. We calibrate about 88% of the in house equipment, with the rest going out to calibration vendors.

During an external audit, the auditor asked if we are accredited to ISO17025. Our answer was no we are not. Here is what we submitted to the auditor:

Is our in-house cal lab accredited to ISO17025?

Answer: No.

Rationale: ISO/IEC 17025:2005 specifies the general requirements for the competence to carry out tests and/or calibrations, including sampling. It is applicable to calibration vendor performing tests and/or calibrations. AV quality system is certified to ISO 13485. The ISO 17025 accreditation is not required.

I'm not really excited about the verbiage that we used, as we are being audited against ISO13485:2003. It was my recollection that internal labs within a company do not have to be accredited to 17025. As another colleague point out, "ISO17205 is applicable to calibration vendors performing tests and/or calibrations". This may lead auditors to ask if each and every external vendor is accredited.

 

[Sidney Vianna]

I believe that, in the Automotive Sector, ISO/TS-16949 requires in-house calibration labs to be accredited to ISO/IEC 17025. But no other major QMS Standard (ISO 13485, AS9100, TL-9000, etc.) has a similar requirement.

So, a potential requirement for in-house cal labs to be accredited would have to be driven from another source; e.g., customers.

 

[JoShmo]

Clealry, your auditor was a) wasting your time with chit chat or b) has only half a clue about the meaning of their question.

It should need zero response in either way. Being accredited is totally different to implementing the ISO 17025 requirements - something many auditors seem to have never understood or never thought about (hence they lead with such questions). I bet they aren't even qualified to comprehend what the technical aspects of 17025 even mean.

Don't waste your time any further with responding to this auditor. They won't know what to do with the answer, anyways...

 

[WCHorn]

I believe that, in the Automotive Sector, ISO/TS-16949 requires in-house calibration labs to be accredited to ISO/IEC 17025.

In ISO/TS 16949:2009, accreditation to ISO 17025 can be used to show conformance to the internal laboratory requirements, but as stated in the note to 7.6.3.1, it is not mandatory.

 

[Sidney Vianna]

Don't waste your time any further with responding to this auditor. They won't know what to do with the answer, anyways...

I don't think we should be jumping to conclusions without understanding the context of the question. This could be a discovery question by an external auditor to ascertain if s/he should spend more or less time assessing the organization's internal cal lab. Maybe, using a basic risk-based logic, the auditor could be thinking that, if the lab in question has a valid, scope-relevant ISO 17025 certificate of accreditation, the auditor would not need to spend (much) time assessing the lab, him/herself.

Please note that the OP did not state that the auditor required the lab to be accredited. S/he just asked about the status.

 

[Project Man]

I think saying less is better. The last two sentences of your rationale are sufficient to answer the question.
Our company is ISO13485 and we also have in-house calibration. The question the auditor asked is a fairly common one but a simple answer is usually sufficient. So, unless you or your customers require you to be ISO17025 you're just fine because, as stated by Sidney Vianna, ISO13485 doesn't require you to be ISO17025 certified.

 

[JoShmo]

I don't think we should be jumping to conclusions without understanding the context of the question. This could be a discovery question by an external auditor to ascertain if s/he should spend more or less time assessing the organization's internal cal lab. Maybe, using a basic risk-based logic, the auditor could be thinking that, if the lab in question has a valid, scope-relevant ISO 17025 certificate of accreditation, the auditor would not need to spend (much) time assessing the lab, him/herself.

Please note that the OP did not state that the auditor required the lab to be accredited. S/he just asked about the status.

I don't agree the auditor should have even braoched the subject. Sure they were asking IF it was accredited and NOT requiring it - they have no grounds to ask for that. Why would they even ask the status? A lab is a lab. IS there provision in MD-5 etc to spend less time evaluating a lab of the comply to ISO 17025 which auditing the rest of the quality system?

 

[Sidney Vianna]

I don't agree the auditor should have even braoched the subject. Sure they were asking IF it was accredited and NOT requiring it - they have no grounds to ask for that. Why would they even ask the status? A lab is a lab. IS there provision in MD-5 etc to spend less time evaluating a lab of the comply to ISO 17025 which auditing the rest of the quality system?

Please note the OP was referring to an audit against ISO 13485. That being the case, IAF MD5 is irrelevant and the appropriate document is IAF MD9. Further, there is nothing wrong to adjust an audit plan based on verifiable, credible evidence. So, the existence of a valid 17025 accreditation for the lab could definitely be used, both to minimize the time spent by a CB or NB auditor of the lab, while maintaining the minimum audit time prescribed in Annex D.

Both IAF MD5 and 9 allow for adjustments (increase/decrease) of audit time. But, we are speculating about the reason for the auditor's question and it would be useful to know why the question was raised. In a world of assembly line approach to 3rd party audits it would be positive to see an auditor using risk based logic to adjust an audit plan.

 

[howste]

I don't agree the auditor should have even braoched the subject. Sure they were asking IF it was accredited and NOT requiring it - they have no grounds to ask for that. Why would they even ask the status? A lab is a lab. IS there provision in MD-5 etc to spend less time evaluating a lab of the comply to ISO 17025 which auditing the rest of the quality system?

Wow.

As an auditor, I know what the requirements are within the scope of each audit, and I will ask whatever questions I feel are appropriate to ask to get the information I need. I don't currently audit ISO 13485 for a CB, but I have in the past. I know that ISO 13485:2003 does not require the organization or their calibration suppliers to be accredited to ISO 17025. However, on many occasions I've asked about ISO 17025 (knowing it's not required) because I know that if it is being complied with that the calibration or testing being done is more likely to be done correctly. Whether the answer is yes or no, I will still evaluate what they have in their system - no harm no foul.

For the record in every audit I've conducted I've asked about things that were not specifically required, but if they did have it would help me to understand and effectively evaluate their system. I frequently ask if they have instructions, records, lists, certifications, etc. not required by the standard. If they chose to implement these things then they are part of their system and are included in the scope of the audit.

 

[Jen Kirley]

Here is what we submitted to the auditor:...

What sort of audit was this? Do you get visited by this auditor?