Balance Calibration SOP for a Pharmaceutical Environment

Hello,
First time post so be gentle.

I work for a third party calibration company that calibrates lab equipment for pharma companies. We have a new Operations Manager who is reviewing our SOP's. I should note that his experience is from working in the research side and this is his first time working for a 3rd party vendor.


In meeting with him today, he pointed out that our calibration SOP for balances is wrong because ISO, GMP/GLP and the FDA require that we do an internal calibration before capturing the "As Found" data. This is not how our SOP is written or has ever been performed before. Our SOP states that we are to measure the unit to capture the "As Found" data, perform the internal cal and measure the unit again for the "As Left" data.


Now I'm fairly certain that our ISO 17025 does not call out any specifics to this, as I just went through our audit and it was never mentioned. I also am a little wary to write a new SOP that states we perform an internal calibration prior to documenting the "As Found" data. The unit would always pass and the data would not be accurate if this is the case, in my opinion.


I guess my question and reason for posting is to see if anyone is aware of a regulation that requires this to be performed?

Thank you!

Hello there! Thank you for coming to the Cove!

I moved your thread to the general calibration forum, so that it may get better responses.

I would be very intrigued to know where your operations manager gets that information from. I agree with you in general principle. However, consider:

1. Are the users procedurally and consistently performing internal calibration prior to use? If so, then performing that same service prior to calibration would make sense.
2. Do the balances (for the newer ones) have auto-calibration function? If so, then it would be OK to allow auto-calibration to occur.

I guess it depends too, on what the purpose of the calibration is. If a balance is having routine performance checks conducted for drift/variance and the scheduled calibration is to verify corner load and the internal weight accuracy, then performing the internal calibration would be less of a factor than if the scheduled calibration is to assess the drift/accuracy of the balance (as it is not part of a routine performance check system).

My opinion: I would be very leery in conducting an internal calibration prior to the scheduled calibration, unless it is consistently utilized by the users prior to use.

Of course, that also assumes the balance is being zeroed, level checked, etc.

I work for a third party cal. lab with 17025 accreditation for mass.. as far as calibration is concern the as found data is the result before you conduct any calibration adjustment regardles internal or external adjustment was done.. and as BradM says given the balance is being cleaned, zeroed, level checked, before capturing the as found reading...

maybe your boss is trying to be boss... you know what i mean...

We are an ISO 17025 Cal lab as well. Our understanding of it is: As Found is values obtained from the condition of the unit "as you find it"; As left is its condition should you find it out of tolerance and make adjustments or to make improvements to unit "to make it better". So I agree with everyone above.

However, there is the rare case as mentioned by BradM where the customer uses it a certain way and wants the calibration done the same. We have a "note" in one of our calibration procedures where this has occurred that allows the technician to "deviate" at the customers request - but it must be documented who at the customer's facility approved the deviation.

Maybe your way out is to include a similar note in the SOP and that way it covers you and your boss's way of thinking.

Hope I helped a little.

1)
I would agree with Brad M; would like to add following...

If understood your post correctly, you are performing the 'as-is'(x) and 'as left data' and performing internal calibration in between.
general purpose of the internal calibration is to adjust any internal variations - fluctuations, as an internal control within the instruments to be adjusted automatically. (ref. instrument.manual)

hence, this practice could add value of providing some insights about that level of variations/fluctuations within the operations, if you have set an acceptance criteria for the difference between both of them x-y.
this may be good practice, but is it required or not is an assessment you need to make ...specific to your own organization / stringency required for your operations.

2) from a pharma perspective, the primary source of references is to look at relevant pharmacopoeia. and then follow the instrument manufacturer's recommendations. (of course, all of them within purview of the requirements of specific operations being carried out...URS or risk etc.,)

i may not be able to provide link/text of pharmacopoeia (u must find them in elsmar threads!!!... + find relevant for the market you are operating) but heres quick references from net...
http://irmm.jrc.ec.europa.eu/EURLs/EURL_PAHs/whats_new/Documents/faq_on_calibration_final__2_.pdf

http://pharmaceuticalvalidation.blogspot.in/2009/01/sop-calibration-of-balances.html
http://www.radwag.com/pliki/artykuly/good_weighing_practice_in_pharmaceutical_industry.pdf
http://www.pharmaguideline.com/2011/01/analytical-balance-calibration.html
hope this helps...