Multimeter calibration

[cornbread]

A bit off topic but I know this esteem group would have opinions on this subject. We are having a debate concerning multimeter calibrations. We have specific calibrated meters we use to comply with ISO. The question is: should multimeters used for daily troubleshooting be calibrated and if so how often.

The argument against calibration...cost.. and you basically test your meter each time you use it via "verify..test..verify".

The argument for claibration is a false reading that gets someone injured.

As always... your replies are greatly appreciated.​

 

[iggy]

I can't say I calibrate my meter unless I see a problem with it. You can always check low voltage if you think you are having errors in your readings, and always have another meter working with highvoltage. Thats just me, always double check.

 

[Tom Jenkins]

Many test standards require instrumentation to be calibrated annually. If you need good accuracy then that is a reasonable frequency. For normal troubleshooting and field service I've generally used Iggy's method - if it ain't broke, don't fix it.

One exception is probes and leads. They should be inspected weekly and replaced every few months, depending on frequency of use. This is especially true if you are using them on anything over 120 VAC.

 

[Mike_RH]

For day to day troubleshooting, I have my meters labeled "For indication only - Calibration not required". This seems to satisfy the ISO auditors.
Lab equipement is calibrated every year.

A more interesting topic is how often (ever?) do PLC analog IO cards get calibrated?

 

[JHarbin]

A more interesting topic is how often (ever?) do PLC analog IO cards get calibrated?

As often as our customers ask us (and pay us) to. Of course, they are never off enough to need calibration, but they are verified and everyone is happy.

 

[Timbert]

Add another for "if it ain't broke, don't fix it."

I used to work in a calibration lab. The newer (anything more recent than a Simpson 260) solid-state devices seldom failed and when they did it tended to be catastrophic. We would usually just verify they were still within spec and send 'em back.

I've seen that leads and low batteries tend to be the Achilles' heal of modern multimeters. I can't remember the last time a meter I used had a problem that wasn't the battery or the leads.

The argument for claibration is a false reading that gets someone injured.

Again they are much more likely to be hurt by a bad lead when measuring high voltage.

Calibration isn't really required to see if a high voltage circuit is "hot." Do you really care if the voltage is 1000 VAC or only 998 VAC? Either voltage is just as dangerous.

If this issue is of great concern implementing an operational check before use would increase confidence that a meter isn't broken. I'd feel much more confident in a piece of test equipment if I tested a 120 VAC wall socket just prior to testing that 1000 VAC circuit than if I saw a nine-month old calibration sticker on the side.

 

[George Graziano]

What we use to do when I worked in an ISO plant, send out a meter that was "Our Stanard", once a year. We then used that to calibrate the rest.
George

 

[Brijm]

Add another for "if it ain't broke, don't fix it."


Calibration isn't really required to see if a high voltage circuit is "hot." Do you really care if the voltage is 1000 VAC or only 998 VAC? Either voltage is just as dangerous.

If this issue is of great concern implementing an operational check before use would increase confidence that a meter isn't broken. I'd feel much more confident in a piece of test equipment if I tested a 120 VAC wall socket just prior to testing that 1000 VAC circuit than if I saw a nine-month old calibration sticker on the side.

That would be my recommendation as well. For safety purposes, the meter should be checked on a known live circuit, before it is trusted...

 

[rdrast]

Normal day-to-day test instruments here aren't calibrated unless they have a problem, in which case they are sent back to the manufacturer (mostly Fluke).

On the other hand, the Process Calibrators we use (also mostly Fluke) require a re-certification / calibration check every 12 months at the most.

When calibrating process equipment, it is a general practice to use the process calibrators, and then use our personal test instruments just to check their accuracy.

 

[Manglemender]

When I started my first job in maintenance, in 1985, I bough a Fluke 77 multimeter. I still use it to this day and it is as accurate as the day I bought it although it has gone through a number of sets of leads. Due to customer requirments, other (calibrated) meters have come and gone but my old Fluke77 remains constant.

Nick

 

[iant]

i use a Fluke 12 forr basic use and a Fluke 87 MIII for all testing that requires to be fairly accurate.
when specific calibration is required we can Hire meters with current calibration certificates.
after 10 years I Know my meter needs calibration but for now it is OK