1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
Dismiss Notice
You must be a registered member in order to post messages and view/download attached files in this forum.
Click here to register.

Multimeter verification

Discussion in 'ISO 9001:2015 - Quality Management Systems' started by Adam J, Apr 21, 2024.

  1. Adam J

    Adam J New Member

    Joined:
    Mar 2, 2024
    Messages:
    2
    Likes Received:
    3
    Trophy Points:
    1
    Hello! I’d like some feedback about monitoring and measurement.

    My employer manufactures gas detection instruments and is audited against ISO 9001:2015 and ISO/IEC 80079-34 (which covers quality management systems for manufacturers of products for explosive atmospheres). ISO 9001:2015 clause 7.1.5 and ISO 80079 clause 7.1.5 states, “When monitoring or measuring is used to verify the conformity of Ex Products, the measuring equipment shall be calibrated and a valid calibration certificate shall exist. Verification of measuring equipment against calibrated equipment is also permitted as long as it is properly documented.”

    Our products are calibrated in production using gas mixtures that include calibration certificates with traceability information and expiration dates, after which the gas must not be used. In practice, gasses in production are consumed long before they expire. That may not be true in other departments.

    Our company also has Fluke 87V multimeters used (1) by production operators to measure current consumption of handheld products and determine whether battery life will meet specification, (2) by repair technicians for general troubleshooting, and (3) by engineers during design work.

    We have a procedure saying that multimeters used for production or repair need to be annually verified for accuracy by checking 10 mADC, 10 VDC, 100 Ω, 1.0 kΩ, and 1.8 kΩ using a Fluke 743b Documenting Process Calibrator, which is itself calibrated annually by an external lab, and we retain the calibration certificate.

    Multimeters used for design engineering are marked “for reference only.” They are maintained at the engineers’ discretion.

    This leads to some less than ideal situations:
    • Auditors (and managers) take a detailed look at our multimeter calibration records, while ignoring our calibration gasses. I don't think an auditor has ever checked a gas bottle.
    • We have little reason to expect gas cylinders will be used past their expiration date, but there are no procedures or records to enforce that.
    • Multimeters in our repair department are regularly calibrated while multimeters in our engineering department are not, even though they have similar uses.
    • Multimeters in production are verified to a much greater accuracy (and with greater frequency) than is needed to maintain product quality. Loss of quality would require a measurement error of at least 10 mA.
    • Maintaining and/or replacing the Fluke Process Calibrator is expensive, considering the above point.
    How can I improve our procedures to reduce these non-ideal circumstances?
    • Should I require operators to check the expiration dates of the cylinders they're using? Would I also need to maintain any new records?
    • If I didn't need to abide by ISO/IEC, I'd calibrate one Fluke 87V, then use that to “sanity check” the others periodically. And I’d not check resistance. Is that acceptable?
    • Can multimeters used for general troubleshooting be marked as “for reference only”?
    • Is there a “cheaper” alternative to a process calibrator I can use when high accuracy isn't needed.
    • Since the multimeters aren't used in a way that affects products’ measurement accuracy, are they subject to clause 7.1.5?
    • If you do internal verifications of multimeters, what equipment do you use?
    If you read all that, thanks for your time. Insights are truly appreciated!
     
  2. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,285
    Likes Received:
    2,631
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    Welcome, Adam! Let's see if we can understand.

    For starters:

    ISO 9001:2015 doesn't state this - I think you've (only quoted) 80079.

    What's the basis of verifying annually? Do you have data to support this is effective?

    That's not good. Nothing should be "for reference only". It's a bogus statement to make. There are either verified/calibrated or not.

    Which auditor? If you are meaning an external auditor, you cannot rely on them fully understanding the science.

    Why give this to operators to do? Do they check if their work instructions are correct? Do they check their tools etc? I hope not. It's up to management to operate a system such that operators get on with the work they're given and paid for - not to go around checking on stuff.

    Can I ask how you came to this? We can't confirm.

    This makes sense to me. It's called "verification" and is best done at the start of a shift or other, similar times.

    We can't tell you this. What do your engineering people say about what checks are needed to prove functionality?

    Why are you worried about "cheaper"? Worry about effective, first. In my experience, you're already throwing away money without looking for cheaper methods.

    Who is responsible for deciding this? How can you show this to them? Asking us to tell you won't win any arguments...

    We can't say. Why are they being used if not to verify product quality? Your answer lies there...

    You can use anything you like to verify a multi-meter, as long as you can trust its accuracy and you can demonstrate it is suitable (4:1 resolution, stable, linear etc)

    From your statements and your questions it sounds very much to me that you don't have a "single point of truth" for matters of measurement in your organization. Maybe that's the real issue...
     
    Last edited: Apr 21, 2024
    tony s, Atul Khandekar and Adam J like this.
  3. Adam J

    Adam J New Member

    Joined:
    Mar 2, 2024
    Messages:
    2
    Likes Received:
    3
    Trophy Points:
    1
    Thanks for your response!

    You're correct...I quoted 80079 assuming that ISO 9001 clause 7.1.5 would be common knowledge here, but 80079's extension of 7.1.5 might not be.

    Good question! Initially, my employer hired a consulting firm to suggest procedures necessary for ISO 9001 and 80079 compliance. They suggested a 1-year verification cycle based on the Fluke 87V User's Manual, which states that accuracy is valid "for a period of one year after calibration." Since then, our historic data from these yearly checks shows that our Fluke 87Vs maintain acceptable accuracy in our application for at least five years.

    No one has ever suggested this to me, but your statement makes sense. Our internal auditor has instructed us to mark equipment as "for reference only" if it is not used for inspection measurements. For example, we label our power supplies this way, since they have current and voltage indicators, but our work instructions specify that voltage and current are measured by a multimeter, thus the power supply's internal meters are no more than indicators and we have no need to calibrate/verify them. Similarly, our design engineers' multimeters, which are not used in production or repair, have no effect on product quality, so our internal auditor suggested we label those instruments similarly.

    Would you recommend placing "Not Calibrated" stickers on these instruments instead? Or something else?

    By the way, your statement prompted some research during which I found this article: Why You Should Never Use "For Reference Only" In Your QMS. Are you the author?

    Yes, all our auditors (including for our internal audits) are external contractors. Agreed. That's why it seems like I should address this in our QMS, rather than rely on historic practice of using up gasses before they expire.

    You've given me something to think about! Our QMS has a "check and balance" system for multimeter calibration; quality and/or engineering staff verify equipment based on due date, and operators to check for expired verification stickers at the start of their work. This reduces the risk that equipment will be neglected. For gas cylinders, I figured an additional instruction to check gas cylinder expiration would be a minimal ask...but perhaps there's a way to verify gasses without burdening our operators.

    What I meant to say is that in the absence of being told what we do when I accepted my role as quality manager, this is what I'd consider acceptable to maintain product quality. I recently inherited the position of quality manager after a career as an electrical engineer, so it's quite possible that what makes sense to me may not be acceptable in an audit.

    Thank you; yes, verification is what I'm aiming for.

    My employer considers my expertise adequate, so long as our auditors agree. I ask in case someone here has relevant experience with multimeter calibration.

    Fair point! I say "cheaper" because my experience (which is in electronics, not quality management!) tells me that we're maintaining instruments at a level of precision which is far greater than we actually need, and I suspect that's an artifact of a historic consultant's advice.

    I have authority to decide this, but I do need to make evidence-based decisions. I know our products' battery life specifications (and therefore maximum current consumption) and nominal current consumption. The Fluke 87V specifications are far greater than is necessary to detect excess current consumption in production. I don't have an issue explaining that to another manager or an auditor, but I also haven't tried.

    Ok, I see. I was thinking in terms of measurement traceability, but since they're being used for product quality, it seems I can't escape applying some type of verification schedule to our multimeters.

    Thank you for that statement. I believe the Fluke 87V is overkill for what we need and the Process Calibrator even more so, so I'll likely modify our procedure to verify with a calibrated Fluke 87V instead.

    You've read between the lines pretty well. I inherited this position when my predecessor retired. When my employer originally decided to pursue ISO compliance, they did so with the help of a consulting firm, and the goal was to pass an audit, not improve quality. There are artifacts of this in our QMS, and some of that has become tribal knowledge that's hard to overcome. As my employer's trust in me grows, and I gain experience with more audits, I feel more confident to try and improve our QMS, and a forum like this I hope can help reduce my risk as I try to make things better.
     
  4. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,285
    Likes Received:
    2,631
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    OK, in that case you'd have perfect justification for extending the recall to 5 years and saving $$$. If anyone says "But, but, what if it gets damaged?" Tell them, calibration recall doesn't help that. It could be damaged the day after it gets calibrated...
    You might want to ensure they have a clue in future...
    Well, Adam, if it makes sense to you - and you can demonstrate it's effective, the auditors (based on previous comments) can "pack sand"...
    Yup, I've seen a lot of these folks who don't know the basics. Always good to test a theory with someone else. Our colleague @BradM is a calibration/measurement wizard (he's got a PhD too) so post here for confirmation.
    That would be me - not employed there anymore, however...
    Good deal. We'll help you...
     
    BradM and tony s like this.
  5. BradM

    BradM Moderator Staff Member

    Joined:
    Jul 31, 2015
    Messages:
    399
    Likes Received:
    317
    Trophy Points:
    62
    Location:
    Arlington, TX
    Goodness! I am just seeing your tag, Andy! I'm so sorry for not responding. Is there anything I can help out with?
     
    Andy Nichols likes this.