7/29/2019 Why Calibration Fluke

http://slidepdf.com/reader/full/why-calibration-fluke 1/2

 Why calibrate

test equipment?

What is Calibration?

Many people do a field comparison check of twometers, and call them “calibrated” if they give thesame reading. This isn’t calibration. It’s simply afield check. It can show you if there’s a problem,but it can’t show you which meter is right. If bothmeters are out of calibration by the same amount

and in the same direction, it won’t show you any-thing. Nor will it show you any trending — youwon’t know your instrument is headed for an “outof cal” condition.

For an effective calibration, the calibration stan-dard must be more accurate than the instrumentunder test. Most of us have a microwave oven orother appliance that displays the time in hours andminutes. Most of us live in places where wechange the clocks at least twice a year, plus againafter a power outage. When you set the time onthat appliance, what do you use as your referencetimepiece? Do you use a clock that displays sec-onds? You probably set the time on the “digits-challenged” appliance when the reference clock is

at the “top” of a minute (e.g., zero seconds). Ametrology lab follows the same philosophy. Theysee how closely your “whole minutes” track thecorrect number of seconds. And they do this atmultiple points on the measurement scales.

Calibration typically requires a standard thathas at least 10 times the accuracy of the instru-ment under test. Otherwise, you are calibratingwithin overlapping tolerances and thetolerances of your standard render an “in cal”instrument “out of cal” or vice-versa. Let’s look athow that works.

Two instruments, A and B, measure100 V within 1 %. At 480 V, both arewithin tolerance. At 100 V input, A reads99.1 V and B reads 100.9 V. But if youuse B as your standard, A will appear tobe out of tolerance. However, if B isaccurate to 0.1 %, then the most B willread at 100 V is 100.1 V. Now if youcompare A to B, A is in tolerance. Youcan also see that A is at the low end of the tolerance range. Modifying A to bringthat reading up will presumably keep Afrom giving a false reading as it experi-ences normal drift between calibrations.

F r o m t h e F l u k e D i g i t a l L i b r a r y @ w w w . f l u k e . c o m / l i b r a r y  

 Application Note

You’re serious about your electri-cal test instruments. You buy topbrands, and you expect them to beaccurate. You know some peoplesend their digital instruments to ametrology lab for calibration, andyou wonder why. After all, theseare all electronic — there’s nometer movement to go out of bal-ance. What do those calibrationfolks do, anyhow — just changethe battery?

These are valid concerns,especially since you can’t use your

instrument while it’s out for cali-bration. But, let’s consider someother valid concerns. For example,what if an event rendered yourinstrument less accurate, ormaybe even unsafe? What if youare working with tight tolerancesand accurate measurement is keyto proper operation of expensiveprocesses or safety systems? Whatif you are trending data for main-tenance purposes, and two metersused for the same measurementsignificantly disagree?

7/29/2019 Why Calibration Fluke

http://slidepdf.com/reader/full/why-calibration-fluke 2/2

2 Fluke Corporation Why calibrate test equipment?

Calibration, in its purestsense, is the comparison of aninstrument to a known standard.Proper calibration involves use of a NIST-traceable standard — one

that has paperwork showing itcompares correctly to a chain of standards going back to a masterstandard maintained by theNational Institute of Standardsand Technology.

In practice, calibrationincludes correction. Usuallywhen you send an instrument forcalibration, you authorize repairto bring the instrument back intocalibration if it was “out of cal.” You’ll get a report showing howfar out of calibration the instru-ment was before, and how far

out it is after. In the minutes andseconds scenario, you’d find thecalibration error required a cor-rection to keep the device “deadon,” but the error was wellwithin the tolerances requiredfor the measurements you madesince the last calibration.

If the report shows gross cali-bration errors, you may need togo back to the work you did withthat instrument and take newmeasurements until no errors areevident. You would start with thelatest measurements and work

your way toward the earliestones. In nuclear safety-relatedwork, you would have to redo allthe measurements made sincethe previous calibration.

Causes of calibrationproblems

What knocks a digital instrument“out of cal?” First, the majorcomponents of test instruments(e.g., voltage references, inputdividers, current shunts) cansimply shift over time. This shift-

ing is minor and usually harmlessif you keep a good calibrationschedule, and this shifting is typi-cally what calibration finds andcorrects.

But, suppose you drop a cur-rent clamp — hard. How do youknow that clamp will accuratelymeasure, now? You don’t. It maywell have gross calibrationerrors. Similarly, exposing a DMMto an overload can throw it off.Some people think this has little

effect, because the inputs arefused or breaker-protected. But,those protection devices may nottrip on a transient. Also, a largeenough voltage input can jump

across the input protectiondevice entirely. This is far lesslikely with higher quality DMMs,which is one reason they aremore cost-effective than the lessexpensive imports.

Calibration frequency

The question isn’t whether tocalibrate — we can see that’s agiven. The question is when tocalibrate. There is no “one sizefits all” answer. Consider thesecalibration frequencies:

•Manufacturer-recommendedcalibration interval. Manufac-turers’ specifications willindicate how often to calibratetheir tools, but critical meas-urements may require differentintervals.

• Before a major critical meas-uring project. Suppose you aretaking a plant down for testingthat requires highly accuratemeasurements. Decide whichinstruments you will use forthat testing. Send them out forcalibration, then “lock themdown” in storage so they areunused before that test.

• After a major critical measur-ing project. If you reservedcalibrated test instruments fora particular testing operation,send that same equipment forcalibration after the testing.When the calibration resultscome back, you will knowwhether you can consider thattesting complete and reliable.

• After an event. If your instru-ment took a hit — somethingknocked out the internal over-

load or the unit absorbed aparticularly sharp impact —send it out for calibration andhave the safety integritychecked, as well.

• Per requirements. Somemeasurement jobs requirecalibrated, certified test equip-ment — regardless of theproject size. Note that thisrequirement may not beexplicitly stated but simplyexpected — review the specsbefore the test.

Fluke CorporationPO Box 9090, Everett, WA USA 98206

Fluke Europe B.V.PO Box 1186, 5602 BDEindhoven, The Netherlands

For more information call:In the U.S.A. (800) 443-5853 orFax (425) 446-5116In Europe/M-East/Africa (31 40) 2 675 200 orFax (31 40) 2 675 222In Canada (800) 36-FLUKE orFax (905) 890-6866From other countries +1 (425) 446-5500 orFax +1 (425) 446-5116Web access: http://www.fluke.com

©2004 Fluke Corporation. All rights reserved.Printed in U.S.A. 4/2004 2153519 A-ENG-N Rev A

Fluke.Keeping your world 

up and running.

One final note

While this article focuses on calibratingDMMs, the same reasoning applies to yourother handheld test tools, including processcalibrators.

Calibration isn’t a matter of “fine-tuning”your test instruments. Rather, it ensures youcan safely and reliably use instruments to getthe accurate test results you need. It’s a formof quality assurance. You know the value of testing electrical equipment, or you wouldn’thave test instrumentation to begin with. Justas electrical equipment needs testing, so doyour test instruments.

• Monthly, quarterly, or semi-annually. If you do mostlycritical measurements and dothem often, a shorter time spanbetween calibrations means

less chance of questionable testresults.• Annually. If you do a mix of 

critical and non-critical meas-urements, annual calibrationtends to strike the right balancebetween prudence and cost.

• Biannually. If you seldom docritical measurements and don’texpose your meter to an event,calibration at long frequenciescan be cost-effective.

• Never. If your work requires just gross voltage checks (e.g.,“Yep, that’s 480V”), calibration

seems like overkill. But what if your instrument is exposed toan event? Calibration allowsyou to use the instrument withconfidence.