We all see this happen all too often: A plant gets dinged on an audit because some auditors find test equipment in a tool room where there is no record of calibration. You know what happens next: Pressured by time, and wanting to satisfy auditors, the manager orders a "sweep" of the tool room and gathers up anything that has a meter needle, scale or digital display and sends them in to be calibrated.

So we end up with a steady stream of el-cheapo test equipment from the employee's tool boxes. You know, things like multimeters bought at Radio Shack, Sears, Wal-Mart, etc. If it were up to me, I'd slap a reject sticker on everything that ain't Fluke or Biddle or Simpson or other professional equipment that has published specifications and calibration procedures, but on the other hand, sure, we don't mind charging a customer more than triple what was paid for a $10-15 meter. But then finding specifications/accuacies is near impossible because one manufacturer often makes equipment for several stores, and nobody knows anything about what they're selling ("You got questions? We have no clue.")

So how best to treat the cheap stuff? Does anyone know where to find the specs for consumer grade test equipment? Or is it in our best interest not to even bother with them and advise our customers to place "calibration not required" or "not for quality measurements" or other similar stickers on the equipment and leave them behind?

There are numerous potential options......

First, if you use an accredited cal lab (or you are one) and have this issue, the military procedures (NA 17-20 series; T.O. 33K series) often have some generic specs for just such an instance, where the brand is not a Fluke or whatever.

Second, in absense of generic procedures, assign a higher value to the tolerance (not MU, it is different) and cal to a broader tolerance.

Third, if in fact the calibration can be ignored because the checks of any importance come after, then a "For Reference Only" sticker will likely work.....but the obvious question becomes.....if this is a measurement that means the product will move to the next step, then doesn't the measurement matter? If yes, then cal the equipment. If no, then why measure? I am paraphrasing Phil Stein only because I don't have the quote right in front of me.

Hope this helps.

Hershal

I've been through this myself numerous times. I don't disagree with anything Hershal has stated. But I'll present some of my experiences.

This is a very ambiguous topic. I have advised users many times to stick with reliable brand names. Since this is a public forum, I won't name any of the bad brands. I remember one VERY off brand DMM that wouldn't meet spec, we sent it to the mfr, it was returned repaired and cal'd and it still wouldn't meet spec. Bottom line on that one - scrap it. period. In another case, a customer bought three brand new off-brand lookalikes to compete with a Fluke 87. They were all brand new, and none of them worked correctly on AC volts, nor would they align into tolerance.

So answer one is to definitely buy brands that you have reasonable confidence will perform as specified.

Next.... I am an avid researcher for specs. I've been able to find specs on nearly every DMM brand I've come across. Radio Shack is actually not too difficult if you use a little finesse in searching. They actually have spec sheets for many on their website. GIDEP is very useful (if you are able to get access), and I have a number of bookmarks from Taiwan for companies that make some private label meters. All that said, I am not a big fan of off brand meters, as they often do not meet specs.

I firmly believe in test equipment in uncalibrated condition being properly labeled and documented. I would do this before spending a lot of money on unnecessary calibrations to inactive instruments.

Many companies have policies about personal test equipment. If they are bringing the instrument in to work to be used on anything that has any impact on product quality (directly or indirectly), it should be calibrated. Additionally, if it is used where safety could be impacted, it should also be calibrated. If the personally owned test equipment is inadequate for the required measurement, it should not be at work. Unless there are well understood rules to assure the right, calibrated instrument is used where it should be; and inadequate and/or uncalibrated test equipment should be precluded from improper use.

My experience is a good policy for personally owned test equipment is to treat it just like company owned test equipment - adequate accuracy, properly calibrated, documented in the system.

The one thing I would be careful about is placing "calibration not required" on "el-cheapo" test equipment. That could make a bad thing even worse. "Not for use on quality measurements" could be a better avenue, along with assuring everyone implicated understands the ground rules.

To follow up on one of Hershal's points, I have also long believed that if the person needed to use a piece of test equipment to make a measurement, it is an oxymoron to say that accuracy doesn't matter. For example, to measure a 115 VAC outlet: If accuracy doesn't matter, a lightbulb with pigtails is sufficient, and there is no need to a DMM. To measure a thickness where accuracy doesn't matter, a ruler will suffice, and there is no need for micrometers. I'm being a little tongue-in-cheek with the examples, but I hope they made the point. And that is, test equipment is quantitative by nature. Test equipment that is not calibrated does not have a known accuracy, and may lie to you. A defective AC Volts function on a DMM used to check a circuit to see if it is energized can cost lives, and any uncalibrated test instrument has the potential of telling a lie at the wrong time.

The way I deal with customers who bring me an "el-cheapo" instrument is to find the specs, attempt to calibrate it, and use it as a "teachable moment" regarding how well we can depend on our test equipment.

I've rambled on quite long enough. Hope these words are of some help.

We all see this happen all too often: A plant gets dinged on an audit because some auditors find test equipment in a tool room where there is no record of calibration. You know what happens next: Pressured by time, and wanting to satisfy auditors, the manager orders a "sweep" of the tool room and gathers up anything that has a meter needle, scale or digital display and sends them in to be calibrated.

So we end up with a steady stream of el-cheapo test equipment from the employee's tool boxes. You know, things like multimeters bought at Radio Shack, Sears, Wal-Mart, etc. If it were up to me, I'd slap a reject sticker on everything that ain't Fluke or Biddle or Simpson or other professional equipment that has published specifications and calibration procedures, but on the other hand, sure, we don't mind charging a customer more than triple what was paid for a $10-15 meter. But then finding specifications/accuacies is near impossible because one manufacturer often makes equipment for several stores, and nobody knows anything about what they're selling ("You got questions? We have no clue.")

So how best to treat the cheap stuff? Does anyone know where to find the specs for consumer grade test equipment? Or is it in our best interest not to even bother with them and advise our customers to place "calibration not required" or "not for quality measurements" or other similar stickers on the equipment and leave them behind?

Sack the auditor and the manager...?

..."teachable moment"...
I have a couple of friends who are teachers - I love this phrase - and take advantage of teachable moments myself with customers, suppliers, my children, etc.

BTW - "a light bulb with pigtails" - LMAO - good one!

"a light bulb with pigtails" - LMAO - good one!"

No joke Cari, I use one all the time. You can buy one at the local hardware store.

Dave

:topic: No kidding - it's a real thing? What are the pigtails?

Sorry for the jargon in my post. It is a lightbulb in a socket with two unterminated leads connected to the bottom. They are normally referred to as "pigtails."

Sorry for the jargon in my post. It is a lightbulb in a socket with two unterminated leads connected to the bottom. They are normally referred to as "pigtails."
Don't be - I learned something new today! :agree:

Having gone through this also, here's my 2 cents.

Why was the equipment there in the first place?
If it is being used what is it measuring?
Have there been any rejects that can be traced back to the equipment?
I'm sure you get the picture. If the equipment isn't being used - scrap it or have the person take it home. If it is for other projects and has nothing to do with QA then mark it. If it is being used for a QA measurement and no reject can be traced back to it, then what purpose does the measurment have?
Or is the equipement adequate enough.

Having gone through this also, here's my 2 cents.

Why was the equipment there in the first place?
If it is being used what is it measuring?
Have there been any rejects that can be traced back to the equipment?
I'm sure you get the picture. If the equipment isn't being used - scrap it or have the person take it home. If it is for other projects and has nothing to do with QA then mark it. If it is being used for a QA measurement and no reject can be traced back to it, then what purpose does the measurment have?
Or is the equipement adequate enough.

I agree, and to take it a step further, if those tools are not being use for acceptance measurements, why not label them "Reference use only" and not worry about them?

A lot of times (and we do it where I work), cheap meters are used for troubleshooting purposes only, e.g. checking fuses, batteries, continuity from point A to point B, etc., as opposed to taking "official" readings that are entered on a data sheet, or used to perform a calibration. For ours, we put "calibration not required" stickers on them and in the comments section write in "for troubleshooting only".

And quite the opposite happens too... Despite having a number of good meters (e.g. Fluke 87s), some people label them cal. not required, despite the valid reasons why they should have them calibrated, simply because they don't want to pay to have every meter calibrated by an outside vendor. Instead, they compare them in-house to a calibrated meter on an as-needed basis.

I have access to GIDEP, and usually reference a USAF T.O. or Navy NAVAIR procedure for a meter that has very similar functions, or we have our own in-house procedure that has loose tolerances for the stuff we just can't find any info on, no matter how hard we look.

There are many alternatives to "how", but the bottom line is assuring (by what ever means works for your company) that adequately calibrated instruments are used where process or product quality can by impacted.

I won't get legalistic on the following point, but I feel duty bound to at least voice a view (and see what some others think).

I have long believed that any measuring instrument should be calibrated. This is above and beyond the minimal "pass an audit" philosophy. My reasons are that there are other implications.

For example: A multimeter that is used to check AC line voltage (in a GO-NO-GO context). Perhaps you don't care whether the voltage is 5% or even 10% off in this situation. However, what are the implications to the incorrect reading? They may not be product defects, but more importantly, they may be a decision that the circuit is de-energized so someone can perform maintenance (or other reason).

For troubleshooting (and much of my background is in electronics repair): if I am using an uncalibrated meter to check transistors or resistors in a curcuit. The uncalibrated meter is of unknown accuracy. What are the implications to my troubleshooting? Do I end up misdiagnosing because a range (or ranges) on my DMM are either way out of tolerance or inoperative?

I'll also refer back to my "lightbulb with pigtails" illustration. If I didn't care what the reading was on the meter, I could have use a lightbulb with pigtails. If I am using a meter and reading the displayed value, I am by definition interested in a quantitative measurand.

I have been involved in many management meetings over the years about this issue. One factory at an company I am related to had a population of a few hundred Fluke 77's and 87's (and those families). They wanted to take them all off from the calibration program because they felt the meters didn't need to be calibrated due to their use. The real reason was cost savings.

Granted, the above meters are good quality and tend to remain in-tolerance for a long time. But no meter stays in-tolerance forever. My recommendation was to lengthen the calibration intervals. They had a 12 month interval. My input was the 2 or even 3 year interval would be better than never calibrating. As high a regard as I have for the above example meters, I have over the years had to align members of both model families.

I don't intend anything derogatory toward anyone's method of determining calibration intervals or whether or not to calibrate troubleshooting type multimeters; only to go on record about this philosophy. So please don't anyone take any offense.

I'll be interested in hearing some of the other views on this (or if anyone agrees).

Jerry,

I agree with the troubleshooting meters you mention. I have worked on many transformers and rectifiers during my years on submarines.
I think what Scott was mentioning was a "non-reading" needle deflection type check. Similar to wetting your fingers and quickly touching 2 leads :mg: .
Or the 9 volt battery on the tongue test.
What I refer to most is the person with calipers where the depth portion becomes bent and all he is verifying is that the hole is deeper than 4".
I have no problem with exchanging personnel safety with a quality measurement. All you need to do is rely on a broken meter to tell you that the capacitors in a, 1980s technology built, 14kw recitifier are discharged and then use a shorting rod (for backup) and see a beautiful blue fire ball bounce off your face shield to agree with you!!!:yes:

I have to wonder why some one would send you a meter that cost three times the original price to calibrate. If they were accurate enough to suit the task when I purchased them I always reguarded them as disposable at calibration time. Buying another new one and doing some product revarification to make sure nothing was askew. I also found that the cheaper equipment didn't do much worse that the expensive stuff when someone dropped it from five feet or managed to roll it over in a two-thousand ton press.

It is amazing how things "get rolled over" in a press. I've seen people leave the can of prussian blue in the press. We're talking BIG MESS. Not to mention grinding wheels (does wonders on a class 1 surface!), pens, union buttons, etc.

Think about it folks......Phil Stein was on track.........

If the measurement matters.....you need to calibrate.....

If the measurement does not matter, why do it?

That applies whether certifying equipment, or as Jerry has so correctly pointed out, even for troubleshooting.....

Hershal