Author Topic: Fluke calibration misconseption - worth knowing!  (Read 5446 times)

0 Members and 1 Guest are viewing this topic.

Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Fluke calibration misconseption - worth knowing!
« on: April 22, 2021, 04:24:54 pm »
I have so often read posts where people talk about the calibration of the Fluke DMM's where it clearly indicated that people think of the calibration as a procedure where Fluke adjusts either hardware or software to compensate for drift.

Here is how a Fluke calibration is performed:
Fluke measures the DMM with high precision gear and note every measured data down on a calibration certificate.
This is done on every calibration and after several calibrations is it possible to predict when a meter will be drifted too much for your needs.

Calibration is NOT a calibration but an advanced status check of your meter's performance.
Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 
The following users thanked this post: Marco1971

Offline helius

  • Super Contributor
  • ***
  • Posts: 3632
  • Country: us
Re: Fluke calibration misconseption - worth knowing!
« Reply #1 on: April 22, 2021, 05:08:05 pm »
Calibration may include adjustment if it is possible and necessary to have readings within spec. Calibration is fundamentally a procedure comparing an instrument to a traceable standard. When your instrument comes with a cal certificate, the differences (offsets, errors) betwen it and the standard are written on the certificate. In the days of wood-cased instruments, these would be written on the paper inside the front cover and used as a matter of course when reading off measurements.

Not all instruments can be adjusted, but anything that can be compared to a standard can be calibrated.
 
The following users thanked this post: newbrain

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 19281
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Fluke calibration misconseption - worth knowing!
« Reply #2 on: April 22, 2021, 05:09:59 pm »
Calibration is not adjustment. That is standard metrology terminology, and is common to all calibration labs and all equipment.

You can send a steel ruler off to a calibration lab for calibration. Do you think they would ever alter its length?

Usually in metrology, stability is more important than precision or accuracy. With calibration, you can see how stable an instrument is over time, and do simple arithmetic to convert a measure value to the "true" value. Adjustment destroys the history and makes stability calculations unreliable.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: Electro Fan, larsdenmark

Online J-R

  • Frequent Contributor
  • **
  • Posts: 952
  • Country: us
Re: Fluke calibration misconseption - worth knowing!
« Reply #3 on: April 22, 2021, 05:13:19 pm »
Some Fluke-specific info and terminology:

Typically a handheld Fluke DMM will come with a piece of paper that is an assurance of sorts from them to you stating the DMM met its basic accuracy specifications when it left the factory.  However, that is not really a "calibration certificate".  If you want a Fluke DMM with a calibration certificate it is typically about $150 more for that option.  You can see this by looking at the Fluke web site and checking out the Fluke 289 and 87V for example.  There is the regular DMM and then another entry with "CAL".

However, for a bench DMM like the Fluke 8845A you will notice that it does come with "17025 accredited report of calibration included", so in that case it is a true calibration certificate.

For the Fluke 287/289 and 80 Series V there is a "Calibration Manual" available for download.  Fluke uses the terms "accuracy tests" under "Performance Tests" to explain the procedure to verify the DMM is working correctly and in within accuracy specifications. 

If the DMM does not pass those tests, then the "Calibration Adjustment" procedure is to be followed and "calibration adjustment steps" are provided.


 

Offline AVGresponding

  • Super Contributor
  • ***
  • Posts: 4643
  • Country: england
  • Exploring Rabbit Holes Since The 1970s
Re: Fluke calibration misconseption - worth knowing!
« Reply #4 on: April 22, 2021, 05:15:14 pm »
Calibration is not adjustment. That is standard metrology terminology, and is common to all calibration labs and all equipment.

You can send a steel ruler off to a calibration lab for calibration. Do you think they would ever alter its length?

Usually in metrology, stability is more important than precision or accuracy. With calibration, you can see how stable an instrument is over time, and do simple arithmetic to convert a measure value to the "true" value. Adjustment destroys the history and makes stability calculations unreliable.

This ^. Ninja'd while I was fetching the definition...

http://www.websters1913.com/words/Calibration
nuqDaq yuch Dapol?
Addiction count: Agilent-AVO-BlackStar-Brymen-Chauvin Arnoux-Fluke-GenRad-Hameg-HP-Keithley-IsoTech-Mastech-Megger-Metrix-Micronta-Racal-RFL-Siglent-Solartron-Tektronix-Thurlby-Time Electronics-TTi-UniT
 

Offline threephase

  • Regular Contributor
  • *
  • Posts: 205
  • Country: gb
Re: Fluke calibration misconseption - worth knowing!
« Reply #5 on: April 22, 2021, 07:36:56 pm »
The official definition of calibration comes from the International Vocabulary of Metrology, generally known as VIM. Currently on version 3, but those who are interested are lucky as there is a consultation on going over the next version and you can download this for free from the BIPM website;

https://www.bipm.org/documents/20126/54295284/VIM4_CD_210111c.pdf/a57419b7-790f-2cca-f7c9-25d54d049bf6

The official definition (section 5.18 on page 39 of the document above), is quite wordy, but it basically means that the term 'calibration' is indeed comparison of an instrument to a known standard, traceable to an official reference.

Calibration labs and manufacturers will differ though. Some calibration labs will provide a form up front for you to select the services you require, which can be either 'calibration' or 'calibration and adjustment' with before and after values or just after adjustment values. Those that do not offer this, in my country, you can be certain that 'calibration' is just checking and no adjustment is made.

Some of the more expensive apparatus that I have, the manufacturers do automatically adjust the instrument to the best performance they can get, it is quite rare for that to be done automatically though and it is always stated up front that they do this. It also comes with a hefty price tag.

 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28141
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Fluke calibration misconseption - worth knowing!
« Reply #6 on: April 22, 2021, 07:58:03 pm »
Calibration is NOT a calibration but an advanced status check of your meter's performance.
Called Performance Verification and the first step in a calibration process that may only measure/record measurement drift if the instrument still meets manufacturer accuracy spec.

Adjustments to meet manufacturers spec are a more involved process as most modern instruments only support closed box automated adjustment processes.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline boggis the cat

  • Regular Contributor
  • *
  • Posts: 218
  • Country: nz
Re: Fluke calibration misconseption - worth knowing!
« Reply #7 on: April 23, 2021, 04:14:38 am »
Adjustments to meet manufacturers spec are a more involved process as most modern instruments only support closed box automated adjustment processes.
This is an area where Fluke and other 'big brands' can provide the user value.

You can download verification and adjustment processes for most Fluke and Agilent instruments (and some of the other big name manufacturer's).  In most modern multimeters and similar, Fluke typically have 'hidden' calibration commands for automated verification and adjustment (it's all unavailable / under NDA) but also allow the user to run through an in-built adjustment process if they wish to do so.

The really cheap stuff (from China and elsewhere) make it essentially impossible to maintain calibration accuracy as no process to adjust is available.  You can still monitor it, and if it is stable enough then correct for inaccuracy, but it is less usable due to no way to adjust the indications.  Claimed specifications often range from very optimistic to ludicrous.

It's always worth checking out whether something you are looking to buy has any documentation available to allow adjustment and other maintenance.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 7729
  • Country: us
Re: Fluke calibration misconseption - worth knowing!
« Reply #8 on: April 23, 2021, 04:32:44 am »
Here is how a Fluke calibration is performed:

You asked them?  Fluke offers a huge range of calibration services for a wide variety of instruments.  The metrologist's definition of the word 'calibration' is not binding on the calibration service provider, so if you care, you should ask first.

For a typical bench DMM, a typical cal lab might offer three or four different services and either ANSI Z540, ISO 17025 or no accreditation as you request.  The cheapest services are 'no data' or 'OOT data only', where they will verify that it is within specs (typically the one-year specs) and either not give you any 'as found' data or only give you the ranges that were out of spec and adjusted.  The most expensive service might be ISO 17025 accredited with 'as-found' and 'as-left' data and adjustment, hopefully to 24hr specs.  Policies and details vary (a lot) from lab to lab and from instrument to instrument. 

FWIW, on standard DMMs up to 6.5 digits, Flukes listed calibration services will all include adjustment AFAIK.
« Last Edit: April 23, 2021, 04:34:18 am by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline boggis the cat

  • Regular Contributor
  • *
  • Posts: 218
  • Country: nz
Re: Fluke calibration misconseption - worth knowing!
« Reply #9 on: April 23, 2021, 04:46:28 am »
FWIW, on standard DMMs up to 6.5 digits, Flukes listed calibration services will all include adjustment AFAIK.
Only if Fluke consider it necessary.  It isn't clear what criteria they use.

When dealing directly with Fluke it is difficult to get information on their process and criteria.  They have a 'take it or leave it' approach.

A note on the 'service levels' stuff: if you are going to pay for a calibration service, then you are almost certainly going to want / require the data.  Possibly you can get by without the accreditation logo stamped on the certificate, but a piece of paper claiming that the instrument meets specifications is worthless.  Providers know this, so they're likely just using this option to low-ball a price to potential customers.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28141
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Fluke calibration misconseption - worth knowing!
« Reply #10 on: April 23, 2021, 04:55:55 am »
When dealing directly with Fluke it is difficult to get information on their process and criteria.  They have a 'take it or leave it' approach.
Since their calibrators are used very widely in the test equipment business they actually offer more services than most might think.

Gold Support Program
To keep pace with the fast changing oscilloscope market, Fluke Calibration is continuously writing new calibration procedures. For a one-off, low-cost payment (less than it would cost you to have three or four DSO oscilloscope calibration procedures written), you can buy into our MET/SUPPORT Gold Software Support Program which gives you access to every new procedure written by Fluke’s software support team during the next 12 months. On current performance, that’s upwards of 100 new oscilloscope calibration procedures per year. And you won’t have to request updates or wait for delivery; every new procedure we write is available for download from our web site. In addition to the free procedures library, you will receive 60 days priority support to get you up and running with MET/SUPPORT Gold. You also have the option to buy into our MET/SUPPORT Gold program, which offers you various advantages, including the opportunity to download new procedures from our web site or even obtain custom written procedures. If the oscilloscope calibration procedure you require is not available and you need it quickly, we can write it for you at very competitive rates as part of our fast-track procedure writing scheme.

https://us.flukecal.com/products/electrical-calibration/electrical-calibrators/9500b-oscilloscope-calibrator?quicktabs_product_details=1
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 7729
  • Country: us
Re: Fluke calibration misconseption - worth knowing!
« Reply #11 on: April 23, 2021, 05:10:39 am »
A note on the 'service levels' stuff: if you are going to pay for a calibration service, then you are almost certainly going to want / require the data. 

Actually quite a swath of calibration customers want nothing of the sort.  The calibration certification is required for one reason or another (ISO, regulation, etc) and they just want the cert at the lowest cost and least inconvenience.

Quote
Only if Fluke consider it necessary.  It isn't clear what criteria they use.

That is true for devices that are not out of spec to begin with.  And without the data, I don't think you would ever know.  It's very hard to reach anyone knowledgeable, in my experience as a not-very-important customer.
« Last Edit: April 23, 2021, 05:16:52 am by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Fluke calibration misconseption - worth knowing!
« Reply #12 on: April 23, 2021, 05:47:33 am »
Actually quite a swath of calibration customers want nothing of the sort.  The calibration certification is required for one reason or another (ISO, regulation, etc) and they just want the cert at the lowest cost and least inconvenience.

That was my experience working at an ISO9001 certified company. All equipment had to have up to date calibration, half the time the stuff came back with faults it didn't have when it was sent out, and calibration seals on the housing so I couldn't easily get in there and fix them. We had one very expensive device that came back with a screw missing from the inside, I tried to plug a cable in and the whole input board came unplugged from the motherboard  :palm:

None of that mattered though, there was a valid calibration sticker on the thing so we were good to go as far as the suits were concerned.
 

Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Re: Fluke calibration misconseption - worth knowing!
« Reply #13 on: April 23, 2021, 06:49:18 am »
Sorry I was not specific enough, what I meant by DMM was the handheld DMM's.
And as tggzzz wrote Its fare more important to know how a meter is drifting between calibrations than resetting and correcting for errors. If Fluke was capable to calibrate a meter, would you expect each meter to be calibrated to nearly perfection and a deviation certification for each DMM instead of a general standard printed in the manual.

It's a bit like a crystal, you keep an eye on how it drifts and do everything to never bump / reset it because resetting a crystal does that you have no chance of knowing it's drift pattern.
« Last Edit: April 23, 2021, 06:51:56 am by FriedMule »
Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Fluke calibration misconseption - worth knowing!
« Reply #14 on: April 23, 2021, 07:22:10 am »
This is an area where Fluke and other 'big brands' can provide the user value.

Really? There's an endless stream of Fluke fans here who are proud that their gear "never drifts".

The really cheap stuff (from China and elsewhere) make it essentially impossible to maintain calibration accuracy as no process to adjust is available.  You can still monitor it, and if it is stable enough then correct for inaccuracy, but it is less usable due to no way to adjust the indications.

None of my cheap Chinese gear appears to be drifting.

Maybe "drifting" isn't a thing any more now we got rid of the trim pots and use digital calibration instead.  :-//

Claimed specifications often range from very optimistic to ludicrous.

My $25 Anengs agree with much more expensive meters to within a single digit, even the 22000 count ones (AN870). Maybe digital calibration took out the "worker in a sweatshop twiddling trim pots" factor.

If one ever goes wildly wrong then they cost less to replace than a single Fluke calibration.

« Last Edit: April 23, 2021, 08:27:19 am by Fungus »
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28141
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Fluke calibration misconseption - worth knowing!
« Reply #15 on: April 23, 2021, 07:23:08 am »
If Fluke was capable to calibrate a meter, would you expect each meter to be calibrated to nearly perfection and a deviation certification for each DMM instead of a general standard printed in the manual.
No.
Every instrument drifts from the instant it is turned ON until it meets equilibrium operating temp where it only needs to meet the manufacturers spec and if this % (level) of accuracy is insufficient for your needs you need a better instrument.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Fluke calibration misconseption - worth knowing!
« Reply #16 on: April 23, 2021, 07:33:26 am »
Sorry I was not specific enough, what I meant by DMM was the handheld DMM's.
And as tggzzz wrote Its fare more important to know how a meter is drifting between calibrations than resetting and correcting for errors.

Yep.

Purely anecdotally: Has anybody here had a digitally-calibrated handheld DMM that "drifts"?

(ie. with age, not with ambient temperature)
« Last Edit: April 23, 2021, 07:35:47 am by Fungus »
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 9889
  • Country: nz
Re: Fluke calibration misconseption - worth knowing!
« Reply #17 on: April 23, 2021, 07:39:11 am »
For a lot of devices, if they're outside of their spec that means the device is faulty and should never be trusted from that point on.
At least not trusted for any important readings that need some level of confidence.

Even if you did tweak the calibration adjustments to bring it back into spec, would you now trust it, knowing it has a history of going out of spec in the past?

Usually what you care about is confidence in your readings.
You can never be totally sure a device didn't have a fault and go out-of-spec 3 seconds before you used it.
The trendline of the drift over the last n cal checks gives you confidence it's still in spec or when it might drift out of spec.

If the device has been in-spec for its last 10 calibration checks and not drift much from center then you can be quite confident it is likely still in spec when you're taking a reading now. It has a history of being accurate

Of course there are some exceptions, some devices need to have their calibration adjusted because drift is unavoidable.
« Last Edit: April 23, 2021, 07:48:20 am by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Fluke calibration misconseption - worth knowing!
« Reply #18 on: April 23, 2021, 07:45:19 am »
If the device has been in-spec for its last 10 calibration checks and not drift much from center then you can be quite confident it is likely still in spec when taking a reading now. It has a history of being accurate.

This is one of the reasons for always owning multiple DMMs - you can check them against each other and get to know/trust them.

Of course there are some exceptions, some devices need to have their calibration adjusted because drift is unavoidable.

Devices with a simple voltage reference are very unlikely to go out of spec with time, eg. handheld DMMs

 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 9889
  • Country: nz
Re: Fluke calibration misconseption - worth knowing!
« Reply #19 on: April 23, 2021, 07:50:42 am »
Of course there are some exceptions, some devices need to have their calibration adjusted because drift is unavoidable.

Devices with a simple voltage reference are very unlikely to go out of spec with time, eg. handheld DMMs

Yeah.

Other things like torque/load sensors often need to be adjusted during calibration because the act of using them (applying a torque/load) will, over-time, physically twist or deform them and cause drift.

You can monitor how bad this is by looking at how much adjustment is needed and shorten the time between calibration so it's always being adjusted before it drifts too much.
« Last Edit: April 23, 2021, 07:56:41 am by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Fluke calibration misconseption - worth knowing!
« Reply #20 on: April 23, 2021, 08:04:04 am »
Devices with a simple voltage reference are very unlikely to go out of spec with time, eg. handheld DMMs

...and if they do, they need to be binned/repaired, not adjusted.

A voltage reference that goes out of spec over time isn't to be trusted at all.


 

Offline Arhigos

  • Regular Contributor
  • *
  • Posts: 84
  • Country: gr
Re: Fluke calibration misconseption - worth knowing!
« Reply #21 on: April 23, 2021, 11:28:11 am »
For a lot of devices, if they're outside of their spec that means the device is faulty and should never be trusted from that point on.
At least not trusted for any important readings that need some level of confidence.


No. Devices having guaranteed drift only for a specific period of time. Normally one year, some can be extended for 2 years, for example if you check fluke 754 specification: 1 year 3vdc is 0.02 %  and 2 year is 0.03 % . Nothing can have a lifetime calibration spec.  |O

If you need trusted device to make sure everything is ok for all period of calibration you should get ISO/IEC certificate 17025 with accreditation. If you just need to check perfomance then  ANSI/NCSL Z540.1 should be enough.  (i'm talking about US, not sure about other countries)
« Last Edit: April 23, 2021, 11:31:27 am by Arhigos »
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Fluke calibration misconseption - worth knowing!
« Reply #22 on: April 23, 2021, 11:56:31 am »
if you check fluke 754 specification: 1 year 3vdc is 0.02 %  and 2 year is 0.03 % .

That's just ass-covering. Does it happen in practice?

I've seen 30-year old meters that agree with brand new ones down to the last digit. This makes me suspect that most calibration certificates are just ass-covering for legal reasons.

(for devices that are based around an internal voltage reference)
 

Offline BeBuLamar

  • Super Contributor
  • ***
  • Posts: 1137
  • Country: us
Re: Fluke calibration misconseption - worth knowing!
« Reply #23 on: April 23, 2021, 02:46:39 pm »
I sent a number of Fluke meters to Fluke for other problems and they did the calibration check and they adjusted none of them. That makes me think I wouldn't need to send them my meters for calibration yearly as very likely no adjustments is needed.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 19281
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Fluke calibration misconseption - worth knowing!
« Reply #24 on: April 23, 2021, 03:01:40 pm »
"Need" must be in terms of "for the purpose of ...".

Often calibration is done so you could demonstrate in a court of law that our measurements were correct (within the stated limits).

If only one person/organisation is involved, adjustment is easy: use a screwdriver to make the values agree. If more than one, then open your wallet!
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf