Reasons for calibration

I'm looking for some arguements in my corner to use against management concerning meters and calibration. We have an onsite person dedicated to calibration of scales, load cells, actuators, and meters. Our manager wants to cut costs by eliminating this person and having us do the calibrations. This is not a problem, but rather than calibrating the meters that we use every day, he doesn't see the need for calibrated meters for troubleshooting etc. Instead, he wants to equip us with the cheapest meters he can find. We have Biddle multi indictors and power sources in the plant, but our meters are much more convenient, plus, I have no clue when the Biddle instruments were calibrated last. I'm just looking for any good arguements to shoot down this initiative before it really takes hold. Any ideas are welcome. Thanks.

Russ

You arent using ISO9001 or some other quality assurance system by any chance ?
According to ISO9001 (and other systems), all measurement equipment (*) should be calibrated in accordance with certain directions.

*: OK, it is not ALL equipment that MUST be calibrated. I believe that if they are not important for assuring the quality or correct production time for the products (like meters for general troubleshooting), then it is OK not to calibrate them.
I think (but am not sure), that the non-calibrated equipment must then be marked with something like "does not need to be calibrated / is not calibrated".

Yes ISO is a good point. My facility is ISO certified. All quality critical instruments are calibrated at scheduled intervals. The calibration equipment is certified by a qualified outside source. As for multi-meters, the instrument calibration guy has a certified meter, but the rest of us don't. We can compare ours to the certified one and know if it's out of spec. It can then be sent off for calibration/repair.

So as a compromise, ask for one meter to be certified, then use this one to check all others and repair/calibrate as required.

However using the cheapest meter on the market will cost more in the long run, I/we use only Flukes that are true RMS.

Just a thought.

Very true, el-cheapo meters can cause probs in the long run especially when trying to set up voltages accurately, like for example on a LVDT interface. We only use Fluke true RMS meters too. Incidentally anyone know where i can get some narrow tipped leads for a Fluke 175 in the UK as the normal one's tips are too damn big.

From management's stand point, the loss in production or non-conforming product can be a lot more than the expense of keeping your instrumentation calibrated. And as JesperMP has pointed out, calibration of instruments has to occur according to your quality policy/manual to be compliant with ISO/QS, etc. Normally the period is recommended by the instrument manufacturer.

But then, maybe your organization is not concerned with compliance and then your choices are wide open. Why maintain when you can just address the issue when the equipment or process fails?

Cheap, and I mean real cheap meters scare me at times. Seeing people with a $15 volt meter testing the voltage on the incoming side of very large main switches makes me shudder. Regards Alan Case

We are ISO 14001 certified, at least that's what the flag out front says. My thinking is that if we are setting up machines with uncalibrated meters, how can we say that quality will not be affected. If my meter says I have an 18mA signal, when it's actually 16.5mA, I have amost a 10 percent error. I guess the "one calbirated meter" would work, but who really wants to waste time checking, beyond that, if our "calibrated meter" goes defunct, or is shipped out for calibration, what reference do we have? In essence, this is what our manager wants, and what I am against, if nothing else, purely for my inconvenience.

Russ

I worked in a foundry once where management looked at instrument calibration as a "nice to have" when we have nothing else to do. Well, we always had something else to do, if you know what I mean. That all changed when a representative of our insurance carrier toured the plant one day and noticed all of the cal stickers that were two years out of date (if they had a sticker at all). Management's position on the matter changed very quickly.

Alan Case said:

Cheap, and I mean real cheap meters scare me at times. Seeing people with a $15 volt meter testing the voltage on the incoming side of very large main switches makes me shudder. Regards Alan Case

Click to expand...

Aye. Even more so when the person doing the testing left the meter in Ohms or has the leads in the current rather than voltage jacks. Even if the meter doesn't blow up in his face, the surprise from the smoke and sparks may be enough to make him fall into some other dangerous piece od equipment.

Edit: And we tend to not be too concerned with meter calibration when used for strictly non critical trouble-shooting (24 volts vdc, 120 vac, 480, etc.) However we like others have mentioned buy only Fluke true RMS meters. Back in the old days we had a nice Simpson analog.

One thing to be careful of when buying cheap multi-meters is the overvoltage class that they are rated at. In an industrial environment you need class III meters. A lot of the cheaper meters for domestic equipment use are class II which isn't good enough on any industrial 400V - 480V supply.

So from an electrical safety point of view use a good quality class III rated meter.

Calibration

I have been in many plants, some have actual technicians that do monthly or weekly calibrations. Some rarely calibrate their equipment. Let say you have a flow meter that is out of calibration and putting 20% more chemicals on a product. That 20%
even in a month's time can add up to some major losses for the bean counters. Same thing if using load cells and adding to much ingredients to a product.
On the opposite side is if you are putting 20% less on the product. Then you get into inferior products.
If it is someones job to do the calibrations, it will get done.
If it is no ones job to do it, it will not get done.

russrmartin said:

I'm looking for some arguements in my corner to use against management ......

Click to expand...

Don't get in the habit of thinking of management as the enemy. That is rarely the case. They just have a different perspective and a different skill set.

First off, understand that things that are intuitively obvious to you, like cheap meters aren't as accurate, slow you down, and affect production, is a totally foriegn concept for them. You need to get your own thoughts organized, and then explain it to them in writing. If you can't write it down, you probably don't totally understand it yourself.

Second, management is supposed to be looking at the money and the bottom line - that is what the stockholders pay them for! You need to get a handle on the cost per hour of lost production. If your company sales are $5 million a year and you run 24/7 your cost for lost production is $600.00 per hour. If you are down one extra hour per month because of poor quality diagnostic tools you will save almost $7,000 per year - probably enough to pay for lots of instruments. If you loose 1/2% of production efficiency due to poor calibration that amounts to $25,000 per year - again tht is enough to buy lots of calibration service.

Take these concepts and apply them to your situation, and make your case on $ and ¢. That is what is going to count.

I'm in management's corner on this one. For most checks, an uncalibrated meter is sufficient, as long as it is reasonably accurate. Think about it this way, if you read 475V on a 480V circuit, are you going to troubleshoot the supply (480) or the load in a non-working circuit? The answer is obvious. Regardless of whether the meter is calibrated, the small voltage difference is a moot point. "I've got 480," one thinks.

Now, for more precise metering, such as process meters and critical levels, then calibration is not an option; it is imperative that all items be calibrated properly. If you're selling by weight, and the weight sensor is 10% off, then you're either cheating your customer (can you say, "lawsuit?") or you're cheating yourself (can you say, "profit margin?"), so that measurement will need a paper trail to protect your company in case of disputes.

You'll have to look at your process and determine which items require calibration, and which items do not, and make the call from there.

I worked in a calibration laboratory about 20 years ago (it seems I've done everything I ever did "about 20 years ago" as I get older), when the Fluke 77 was relatively new. I believe the cal interval on those was 18 months to 3 years. We never found a working 77 that was out of calibration in the two years I worked there. I'd expect similar results for Fluke 87's.

As far as inferior equipment... well, in a pinch, I guess I'd use it, but as a norm, I'd never use cheap equipment. Cheap equipment is cheap because it is not as good as the expensive equipment, and hospital visits tend to increase costs a lot faster than equipment costs.

In my previous company, calibrations are important, for those meters are used to monitor instruments that must have as little error as possible. Would not want a radar to go out of tune, and planes having to crash for a landing. =P
So if you are working in a company that deals with precision instrumentations, calibrations are a must.

russrmartin, maybe for some cases, calibrate a few of the meters, scales, etc, and used those calibrated equipment as the ideal condition; the other meters can then be adjust with reference to the tweaked ones. It could cut some cost and still stay in tuned.

regards
Sherine T.

ChuckM said:

Aye. Even more so when the person doing the testing left the meter in Ohms or has the leads in the current rather than voltage jacks. Even if the meter doesn't blow up in his face, the surprise from the smoke and sparks may be enough to make him fall into some other dangerous piece od equipment.

Click to expand...

Done that a few times myself lol (well i think it was about 2AM last time) but proves a point, went across 415V with a Fluke 73 set to volts but left the leads in the current points. Took out a trip, blew the fuse in the meter but nothing more. But worse has happened, would definately recommend a quality meter to anyone as i was using an el-cheapo one i borrowed a while back, didnt read any voltage when i went phase to phase with it, but when i got my hands stuck across two phases it just proved the meter was not working. Always a worthwhile point, if your doing work on any dangerous voltages then check that the meter is working by checking a known voltage first rather than relying that it is ok.