Author Topic: Fluke 732 calibration price list and procedure  (Read 16702 times)

0 Members and 1 Guest are viewing this topic.

Offline Bill158

  • Regular Contributor
  • *
  • Posts: 105
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #25 on: October 25, 2017, 07:57:00 pm »
Perhaps I am wrong in my understanding by what it means to be traceable to the SI through NIST, and directly traceable to NIST
I think an easier way to understand all of this is as follows.
1.  By an act of the US Congress NIST (NBS) maintains THE LEGAL VOLT for the United States of America!  NIST's JVS is right and FLUKE's JVS is somewhat different.  Probably not very much and well below 0.1 ppm.
2.  FLUKE sends NIST a bank of 732Bs with what they think is "THE VOLT", as assigned by FLUKE's JVS.
3.  NIST sends that bank back, after comparing that bank of cells with the NIST JVS, with the statement "THIS IS THE VOLT" end of discussion!
Therefore the 732A that CalMachine gets back from FLUKE is "traceable to NIST", even though there may be several layers of instruments and measurements between NIST and his 732A.  Also FLUKE can state the the biggest error between what NIST says that "THE VOLT" is and the volt that FLUKE assigns to CalMachine's 732A.  I don't know, maybe 0.3 ppm?
At least that is the way I understand all of this "traceability" stuff.
Bill
 
The following users thanked this post: CalMachine

Offline Edwin G. Pettis

  • Frequent Contributor
  • **
  • Posts: 451
  • Country: us
  • The plural of anecdote is not data.
Re: Fluke 732 calibration price list and procedure
« Reply #26 on: October 25, 2017, 08:17:53 pm »
What MisterDiodes is saying (and myself) is that customers often require traceability to NIST or they won't do business with you.  Our standards must be traceable through whatever cal lab we use to NIST, there are no exceptions that I know of.  All of the resistor manufacturers must maintain traceability to NIST of their standards or they can't sell to a wide range of customers including aerospace, industrial and government.  Anyone who makes resistor standards must have treaceability to NIST, no exceptions, an SR-104 must have a record of traceabilty to NIST to be accepted (just like mine does), that is the whole point of having a National Calibration Lab so that everybody can trace back to them and the same primary standards, everybody is on the same page.  I am sure that Cal Machine, et al are aware of this requirement.

I can't say I've heard of JJAs not having some kind of traceability back to NIST as you mentioned, they are compared to each other periodically and tweaked if necessary so that they are all in agreement, whether there is a paper trail for that or not, I don't know, I have no need for a voltage standard with such low uncertainty.  The problem with such low sub fractional uncertainties is that they start changing  once they are off the primary standard they were compared to, that is the real world, everything affects a standard, whether it is at .01 PPM or PPM levels, everything drifts, there is no getting around it and everything in the environment will affect to some degree or other.  Fluke states that once your voltage standard leaves their primary lab, that .06 PPM will start increasing as stated in note 4, it is open ended if you read it.  While a nice aged Vref may have very predictable drift, a mathematical prediction does not qualify as a calibration.  I know what the drift of my SR-104 is on a yearly average but that doesn't constitute a calibration, it still has to go back to the cal lab to verify it, no freebies unfortunately.

 
The following users thanked this post: try, CalMachine

Offline Moon Winx

  • Regular Contributor
  • *
  • Posts: 83
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #27 on: October 25, 2017, 08:20:09 pm »
Our A2LA/17025 accredited 732B test report has the boiler plate text that includes "...this certification is traceable to the internationally accepted representation of the SI Volt through the Josephson constant (KJ-90 = 483597.9 GHz/V) and the use of a Josephson voltage standard operated by [my lab]". We definitely don't claim NIST traceability for the Volt. If you didn't have a JVS you would need to, but if you can create the volt in your lab that's where the traceability ends (well, actually it ends wherever your frequency source traceability ends, but that's a whole 'nother domain). For intrinsic standards like this and the QHR, the auditors from the accrediting agency look at your inter-lab comparisons and uncertainty calculations to decide whether you can claim a certain uncertainty on your scope of accreditation. NIST or other NMI traceability only applies to non-intrinsic standards.
« Last Edit: October 25, 2017, 08:24:58 pm by Moon Winx »
 

Offline Moon Winx

  • Regular Contributor
  • *
  • Posts: 83
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #28 on: October 25, 2017, 08:32:11 pm »
Our A2LA/17025 accredited 732B test report has the boiler plate text that includes "...this certification is traceable to the internationally accepted representation of the SI Volt through the Josephson constant (KJ-90 = 483597.9 GHz/V) and the use of a Josephson voltage standard operated by [my lab]". We definitely don't claim NIST traceability for the Volt. If you didn't have a JVS you would need to, but if you can create the volt in your lab that's where the traceability ends (well, actually it ends wherever your frequency source traceability ends, but that's a whole 'nother domain). For intrinsic standards like this and the QHR, the auditors from the accrediting agency look at your inter-lab comparisons and uncertainty calculations to decide whether you can claim a certain uncertainty on your scope of accreditation. NIST or other NMI traceability only applies to non-intrinsic standards.

To clarify a little, 732Bs can be traceable to NIST's JVS or any lab's JVS. That's where the volt chain ends. This goes for other intrinsic standards also. A lot of parameters do not have an intrinsic standard and must be traceable to the US standard at NIST. Examples of this are capacitance, inductance, ac voltage (although that is about to change), kilograms, meters, and frequency.

 
The following users thanked this post: CalMachine

Offline CalMachine

  • Frequent Contributor
  • **
  • Posts: 477
  • Country: us
  • Metrology Nut
Re: Fluke 732 calibration price list and procedure
« Reply #29 on: October 25, 2017, 11:07:28 pm »
I went a little digging into the actual ISO 17025 standard, since I'm still fairly new to this stuff.  I found a little excerpt that pertains to this debacle and clears up some of the uncertainty in my head...   :palm:   

ILAC P10 states :

Quote
Metrological traceability requires an unbroken chain of calibrations to stated references, all having
stated uncertainties – refer VIM [1]. The persistent misconception that metrological traceability may be
linked to a particular organization (e.g., “traceable to a specific National Metrology Institute”) fosters
continued confusion with regard to its nature. Metrological traceability pertains to reference quantity
values of measurement standards and results, not the organization providing the results.

I see I went wrong by thinking that, since NIST is the highest tier on the pyramid in the US on the chain to the SI, you would in-turn have tracebility to them, inherently, by traceability to the SI. 

I've also gathered some other pertinent information.  NIST's current uncertainty is 0.19 PPM on a 10 V measurement.  During Fluke and NIST's last comparison, their reported values were in significantly greater agreement than that.  If NIST ever decided to revisit their budget, everyone could receive more certain measurements 'directly traceable to NIST', if you happen to need them
« Last Edit: October 26, 2017, 02:05:07 am by CalMachine »
All your volts are belong to me
 

Offline dacman

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #30 on: October 26, 2017, 02:22:12 am »
I went through an NVLAP audit with NIST auditors a few years ago and they had us change our traceability statement from traceable to NIST or other national labs to traceable to the international system of measure (SI) through NIST or other national labs (and other things such as ratio or reciprocal type measurements).  It doesn't matter what type of report it is.  We never state that we are traceable to NIST anymore.

NIST was sending out a PJVA (Programmable Josephson Voltage Array) for intercomparison with labs with a JVA unless they couldn't handle another JVA or didn't need the uncertainty, then they would send a bank of four 732Bs.  This allows the lab to claim traceability through NIST.  Fluke's Primary lab uncertainty statement with their JVA indicates to me that they got the PJVA.
Modify message
 

Offline Moon Winx

  • Regular Contributor
  • *
  • Posts: 83
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #31 on: October 26, 2017, 03:36:02 am »
The > 0.19 ppm typical measurement uncertainty given at NIST is kind of surprising. That means that they do not measure customer zeners with a JVS but with a bank of other characterized zeners. You would think with current automation that they would just use the JVS but I'm sure they have a good reason for not going that route. We measure all 732s against the JVS, just because the 10 V output uncertainty can easily climb out of our customers' required range during the 1 year calibration interval. Comparing it to our bank would max out the output uncertainty in a matter of weeks or months, depending on the unit.
 

Offline MisterDiodesTopic starter

  • Frequent Contributor
  • **
  • Posts: 457
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #32 on: October 26, 2017, 04:37:56 am »
OK, enough pissing contest.  Let's get back to the original topic, and that is calibration of 732a at Fluke Everett Service center, and why they will give you 0.3ppm or 0.2ppm uncertainty of measure traceable to NIST.

Part of it has to do with NIST's side of the equation, and NIST being able to verify manufacturer's claims...

I had a nice long chat with a PhD metrologust directly at NIST this afternoon to clarify some questions about NIST's involvement of the certification process.

1.  There is ONE legal volt in US, and that is the NIST version.  If you do business internationally then you would specify something along the lines of traceable to SI thru NIST...but none of our needs are met by that.  Fluke's volt is Fluke's Volt, and does not count as legal anything, at least for what we need.  JJ-Array at point A does not necessarily legally equal JJ-Array at point B, and that is why there still has to be a means to inter-compare the two, either by a fleet of 732b's or transferring compact JVS systems between labs.  A JJ-array is close to an intrinsic measure, but not quite...For instance in case of war and the GPS system goes down and we lose a common worldwide  time/ freq reference, there will be extra uncertainty if we have to use a different means to keep excitation frequency of the JJ-As matched, but that's a different story. 

2.  NIST uncertainty (95% confidence) is spec'd to 5nV/V on their JVS.  Fluke's uncertainty (95% confidence) their JVS is currently is 5nV/V PLUS extra 50nV if you want a traceable measure (That's where Fluke's measure spec uncertainty of 100nV or 0.01ppm on 10V scale comes from).  NIST has the luxury of being the golden standard, so they don't have the extra added transfer uncertainty.  They ARE the definition of the volt for US.  That being said they still intercompare with all the bigger labs - everyone is trying to keep the standards in close agreement.

3. For our needs and doing business with our customers, we MUST have have the words "traceable to NIST" on both absolute value and uncertainty.  "Traceable to Fluke" is worthless a lot of times.  You will see in several places in Fluke documentation that yes you can get a measure at Fluke to better than 0.1uV/V U95 BUT that is relative to Fluke's Volt.  The NIST traceability for uncertainty will be larger (that part appears in the fine print).

4.  YOU CAN cut out Fluke and send your Fluke 732a / b to NIST for Measure and Cal, and they will measure it against the NIST JVS for 30 days, and give you a characterized report.  The fee is $2311 and you pay for all shipping.

https://www.nist.gov/calibrations/voltage-measurements-calibrations

The official reports will use 1uV/V uncertainty.  For the rest of this post I will assume we are on a 10V scale so that 1uV/V is 1ppm, and 95% confidence.

NOW - As explained by NIST, for any known Zener Diode Reference, about the best anyone is going to do is about 0.2ppm uncertainty.  They have seen it go as low as 0.1ppm uncertainty but that is rare.  So the Calibration is quoted as the best you'll get is 0.19ppm, and the normal range of uncertainty on any Zener Diode Vref they've seen is around 0.1 to 0.5ppm

The limiting factor here is the noise of the Vref.  This includes Zener Noise, Temperature, Humidity, Pressure effects on the system.  So in the words of NIST:  They watch your Vref for 30 days at a measuring resolution of down to .005ppm uncertainty and then look at the overall drift and noise - and assign an absolute voltage value and 95% confident uncertainty to the Vref, which will be no better than 0.19ppm.

The other problem NIST pointed out:  You can measure a 732a/b to some ridiculous fine resolution on a JVS and come up with some value.  The problem is making use if that measured value:  Unless you're going to use that measured value within a very short time frame, trying to get a transfer measure out of your Vref much below 0.1ppm uncertainty is like trying to measure bowl of jello. And forget transporting the 732 and expecting that measure to still be valid at a very small uncertainty.

5.  Zener noise of 732a's is spec'd at 1uV RMS or less (typically way less than that when they age, ours are down around 0.35 or 0.50uV).  732b is spec'd at 0.60uV rms (.1 to 10Hz), and usually measure around about that.  You are never going to get a Cal certificate better than 0.2ppm uncertainty because anything below that uncertainty is going to be a measure of mostly the noise of the Zener Diode Vref - and not a very stable or usable measure.

So the bottom line is YES you could get your 732 measured at NIST - but all that happens is you might get 0.19ppm uncertainty.  Send it to Fluke and measure against their JVS and you've dropped all the way down to 0.2ppm uncertainty and saved a few bucks.  Plus Fluke will perform an adjustment if you need it.

NIST is only going to work with the "Traceable to NIST" uncertainties with values they can verify.  And that lower uncertainty limit - for now at least - is about 0.19ppm for a zener diode reference.

Have Fun!
« Last Edit: October 26, 2017, 03:19:56 pm by MisterDiodes »
 
The following users thanked this post: Edwin G. Pettis, wim.de.jong.59, Moon Winx, Laurent G

Offline dr.diesel

  • Super Contributor
  • ***
  • Posts: 2214
  • Country: us
  • Cramming the magic smoke back in...
Re: Fluke 732 calibration price list and procedure
« Reply #33 on: October 26, 2017, 11:46:34 am »
Very interesting.  Big thanks to all for taking the time to further discuss/elaborate. 

Offline Moon Winx

  • Regular Contributor
  • *
  • Posts: 83
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #34 on: October 26, 2017, 03:40:28 pm »
There may have been some slight miscommunication with the NIST physicist, and I don't expect anyone to take my word over NIST's, but I think some points need to be clarified.

2.  NIST uncertainty (95% confidence) is spec'd to 5nV/V on their JVS.  Fluke's uncertainty (95% confidence) their JVS is currently is 5nV/V PLUS extra 50nV if you want a traceable measure (That's where Fluke's measure spec uncertainty of 100nV or 0.01ppm on 10V scale comes from).  NIST has the luxury of being the golden standard, so they don't have the extra added transfer uncertainty.  They ARE the definition of the volt for US.  That being said they still intercompare will all the bigger labs - everyone is trying to keep the standards in close agreement.

Fluke's uncertainty from their JVS should match NIST's at around 40 - 50 nV. NIST doesn't have the "golden standard" JVS, they have, as you said in (1), the legal standard for the volt. But as I'm trying to convey, the representation of the Volt can be created in any lab with a proper operating JVS. There is no transfer uncertainty involved just due to the JVS being away from NIST. There is no additional uncertainty added and as long as the JVS systems are using the same null voltage measurement equipment/technique and thermal offsets are comparable, the uncertainty of a 732B measurement using the any JVS will roughly be the same (disregarding the short-term stability of the 732). As far as I can tell, this results in a measurement uncertainty of roughly 50 nV.

3. For our needs and doing business with our customers, we MUST have have the words "traceable to NIST" on both absolute value and uncertainty.  "Traceable to Fluke" is worthless a lot of times.  You will see in several places in Fluke documentation that yes you can get a measure at Fluke to better than 0.1uV/V U95 BUT that is relative to Fluke's Volt.  The NIST traceability for uncertainty will be larger (that part appears in the fine print).

Do you have a example of that? A picture of the fine print? This points to Fluke comparing against a bank of volt references rather than Fluke's JVS. There is a relative uncertainty associated with those measurements. The bank itself has an uncertainty associated with its 10 V value, and the reported measurement uncertainty should include this. The difference between the bank's uncertainty and the measurement's uncertainty is the transfer uncertainty. If Fluke is just reporting the transfer uncertainty, that doesn't do the customer much good.

4.  YOU CAN cut out Fluke and send your Fluke 732a / b to NIST for Measure and Cal, and they will measure it against the NIST JVS for 30 days, and give you a characterized report.  The fee is $2311 and you pay for all shipping.

https://www.nist.gov/calibrations/voltage-measurements-calibrations

The official reports will use 1uV/V uncertainty.  For the rest of this post I will assume we are on a 10V scale so that 1uV/V is 1ppm, and 95% confidence.

NOW - As explained by NIST, for any known Zener Diode Reference, about the best anyone is going to do is about 0.2ppm uncertainty.  They have seen it go as low as 0.1ppm uncertainty but that is rare.  So the Calibration is quoted as the best you'll get is 0.19ppm, and the normal range of uncertainty on any Zener Diode Vref they've seen is around 0.1 to 0.5ppm

The limiting factor here is the noise of the Vref.  This includes Zener Noise, Temperature, Humidity, Pressure effects on the system.  So in the words of NIST:  They watch your Vref for 30 days at a measuring resolution of down to .005ppm and then look at the overall drift and noise - and assign an absolute voltage value and 95% confident uncertainty, which will be no better than 0.19ppm.

The other problem NIST pointed out:  You can measure a 732a/b to some ridiculous fine resolution on a JVS and come up with some value.  The problem is making use if that measured value:  Unless you're going to use that measured value within a very short time frame, trying to get a transfer measure out of your Vref much below 0.1ppm is like trying to measure bowl of jello. And forget transporting the 732 and expecting that measure to still be valid at a very small uncertainty.

5.  Zener noise of 732a's is spec'd at 1uV RMS or less (typically way less than that when they age, ours are down around 35 or 50uV).  732b is spec'd at 60uV rms (.1 to 10Hz), and usually measure around about that.  You are never going to get a Cal certificate better than 0.2ppm uncertainty because anything below that uncertainty is going to be a measure of mostly the noise of the Zener Diode Vref - and not a very stable or usable measure.

So the bottom line is YES you could get your 732 measured at NIST - but all that happens is you might get 0.19ppm uncertainty.  Send it to Fluke and measure against their JVS and you've dropped all the way down to 0.2ppm uncertainty and saved a few bucks.  Plus Fluke will perform an adjustment if you need it.

NIST is only going to work with the "Traceable to NIST" uncertainties with values they can verify.  And that lower uncertainty limit - for now at least - is about 0.19ppm for a zener diode reference.

Have Fun!

If they will measure your 732 against the JVS they need to update the webpage at the link you provided. It just contains verbiage referring to measuring it against a bank of working standards.

I think there may be some confusion in what is reported on a JVS 732B measurement. The uncertainty listed on a test report is the uncertainty of the measurement at the time of measurement and has little to do with anything but that. So the owner of the 732 would never use just this value to state the uncertainty of the voltage output... he would need to combine the reported uncertainty with the stability/environmental coefficients of the 732 (from spec) to get the output voltage uncertainty. If you have enough historical data, you can predict the output voltage and pretty much remove the stability part of the equation.

I would raise a lot of questions with NIST or Fluke if I received a 732B test report with a 0.19 or 0.2 ppm uncertainty. That's huge.
 

Offline MisterDiodesTopic starter

  • Frequent Contributor
  • **
  • Posts: 457
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #35 on: October 26, 2017, 05:29:08 pm »
Good grief...I'm trying here and this is becoming a bigger time sink than I ever anticipated...Last post from me:

RE: In the US, Legal volt is US is the legal volt kept at NIST.  Period. NOT Fluke, Boeing, Sandia Labs or anyone else.  NIST doesn't have to  report a "uncertainty difference to NIST", and that doesn't mean their JVS is better than anyone elses.  You can take that argument up on a legal forum.  That's what I have to deliver to my customers on a calibration certificate and that's the terminology the auditors look for.  I can also use "traceable to the SI " - that would be proper -if- I use JVS documented standards all the way through.

There is much more to the conversation - I'm giving you the highlights here, trust me.  I think you're missing the point of calibration of a 732 TRANSFER STANDARD.  It has to be able to transfer a voltage measured from point A and deliver that measure to another time and location point B.  Yes you can can measure the 732 to a lower uncertainty than 0.2ppm.  BUT THE NOISE OF THE ZENER itself limits the usability of that measure.  Unless you expand the uncertainty limits to a practical value, that measure is only going to be valid on the order of minutes or hours.

In the words of the kind NIST Dr. - They could measure it down to 5nv/V uncertainty and be well into the noise of the 732, but then you wouldn't be able to move that 732 down to the other end of the workbench and expect that uncertainty to be valid 5 minutes later.  Why on Earth would you want to do that?  The measure and calibration limits are there as a practical guideline for how accurate your transfer standard is - and with the assumption that you will transport the Vref to another location and have the measure & uncertainties be valid at least for a practical time frame.

That is also why, if you look at NIST lab services - the uncertainties reported for different devices are different.  If you want to characterize 1.018V standard cell references, they will give you a minimum .04 ppm uncertainty on that measure.  That's because that technology is quieter than a Zener (but more work to maintain over time).

RE: Lab Intercomparison
For instance - NIST traveled very recently to Fluke Everett to do an inter comparison between NIST's compact JVS and Fluke's JVS (as well as many Zener standards) and the difference between labs was much less than expected.  I won't give you the results here (hasn't been published yet) but it is way, way less than 40~50nV difference.  That result quoted is from an older test.

RE: NIST Measurement method:  I am told by NIST that NIST has a new scanner system setup and your 732 is read directly against their newer JVS to 5nV/V uncertainty, for 30 days.  I can only go by what NIST tells me directly.  As a taxpayer I want to believe them over you at this point.  But that's just me.  If that's Fake News or not I don't know...

What I know for sure, and is agreed on by Fluke and NIST:
Right now, the only practical way to get better uncertainty than a traceable 0.2ppm uncertainty measure cal on on a Zener based reference is to have multiple Zeners Vrefs on a tight calibration schedule measured against  a JVS-backed reference group.

Otherwise you can purchase a JVS system which I'm sure MoonWinx is familiar with.  That is considered a near intrinsic standard, at least if you have an accurate frequency reference available  (i.e. GPS is verified working).

The push is to develop a more affordable JVS with a programmable Vref output up to 10V, and hopefully requires only ~20GHz excitation frequency, not ~75Ghz.  It would be nice if you could have it not require liquid helium also.  NIST tells me this is being worked on and improvements made:

https://www.nist.gov/news-events/news/2013/04/primary-voltage-standard-whole-world

The other clarification is:  If you reference directly to a JVS, and can show an accurate frequency measure (GPS time lock or your own local atomic clock reference) then that would be a good intrinsic reference, and then the statement on the calibration document would be "Traceable to the SI" which is what the SI volt is.

There ya go.  Have Fun!





 

« Last Edit: October 26, 2017, 06:02:56 pm by MisterDiodes »
 
The following users thanked this post: Edwin G. Pettis, Laurent G

Offline Moon Winx

  • Regular Contributor
  • *
  • Posts: 83
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #36 on: October 26, 2017, 07:16:32 pm »
Sorry MisterDiodes, I'm not trying to become a time sink for you! I promise I'm here to help clarify some points just so wrong information doesn't spread. I work and help develop these systems everyday and have been doing this for 15 years. I've been through 6 on-site audits for these JVS systems where our methods of performing and reporting zener voltage reference measurements have been scrutinized. We now have 3 of these systems in our lab, including a state-of-the-art pulse-driven AC Josephson system. I know all of the guys who do this work at both campuses of NIST. I talk to them quite often and they are some of the smartest people I've ever met, way smarter than me, so I don't doubt at all what they told you to be true. But I think your interpretation of that *may* be a little off. Remember, NIST is not 17025 accredited (and it makes perfect sense why they wouldn't be) so they don't necessarily have to know about or follow the ISO standard. Some of their metrologists perform audits for NVLAP or A2LA, but NIST is only accountable to their own quality standard.

Having said that, I doubt NIST goes against 17025 and reports a measurement uncertainty that takes into account future use of that item. I suspect the reported uncertainty is the measurement uncertainty at the time of measurement and nothing else. And when I use the term "measurement", in the case of a 732, it includes all voltage samples recorded over a span of several days or weeks. Usually the mean of the samples is reported along with the combined measurement uncertainty that takes into account the variation of the readings throughout the measurement (in NIST's case, the readings over the 30 day period). This variation is the largest component of uncertainty in these measurements. Typically we see something like < 100 nV (< 0.01 ppm) scatter over a few weeks of samples, and I suspect NIST sees about the same. So we combine this with other sources of uncertainty and it usually comes out to less than 0.05 ppm (@ k = 2).

So you can see why I would raise an eyebrow if NIST or Fluke is reporting something much higher. 0.05 ppm to 0.19 or 0.2 is a huge jump so I am really thinking they are comparing the customer's 732B to a JVS-characterized bank and not measuring it directly with the JVS. That large reported uncertainty makes more sense that way, but I could be wrong. It's also curious that NIST's own webpage with the larger-than-expected uncertainty states it uses the bank comparison method for this. If they measure it with their JVS they could report a much lower uncertainty is all I'm saying.




« Last Edit: October 26, 2017, 07:25:07 pm by Moon Winx »
 
The following users thanked this post: CalMachine

Offline MisterDiodesTopic starter

  • Frequent Contributor
  • **
  • Posts: 457
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #37 on: October 26, 2017, 10:56:33 pm »
MoonWinx - Double Checked at NIST and yes, when you send in a 732 it is checked -directly- against the JVS on their scanner system, and both Fluke and NIST JVS system have 5nV/V uncertainty while measuring against the JVS.  Fluke has a small additional uncertainty added to allow for any misalignment to NIST, according to Fluke.  For purposes of Calibration report the specified uncertainty can't be better than the noise floor of the device being tested.

I am looking at a live data sample from what is considered a very good 732b testing at NIST right now on their JVS and I'm seeing more like +- over 200nV spikes over just a couple days, not weeks.  It gets worse than that over 30 days.

So you might want to check with your buds and see what the real story is - otherwise you have some extraordinarily good 732's!

CalMachine:  Fluke did get back to me and -no- traceable Cal Cert will be issued from Everett Fluke (Service or Park) at better than 0.2ppm uncertainty for a 732.  Again, their measuring technique will be finer resolution than that while device is under test, but the inherent noise in the Zener Diode is the real limit on the real device uncertainty.

In other words taking a certified measure with an uncertainty that's below the usable noise floor of the instrument under test is futile - because it's just noise.

Their lab services do include other measures with less uncertainty than that of course - but that's not necessarily for Zener Diode Vrefs and not necessarily traceable.  For the purposes of this thread we're talking about a 732 calibration certificate only.

If you hear different do let us know, but that's what I found out from this end. 

Somebody else's turn to research now.
 
The following users thanked this post: Edwin G. Pettis, chuckb, CalMachine, Moon Winx

Offline Moon Winx

  • Regular Contributor
  • *
  • Posts: 83
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #38 on: October 27, 2017, 12:23:58 am »
I am looking at a live data sample from what is considered a very good 732b testing at NIST right now on their JVS and I'm seeing more like +- over 200nV spikes over just a couple days, not weeks.  It gets worse than that over 30 days.

So you might want to check with your buds and see what the real story is - otherwise you have some extraordinarily good 732's!

Now we get into the hairy details! That range you see seems pretty typical for the samples we take too, and our customers don't have especially good 732s and in fact, they are probably the most abused ones out there. ;)

As you might imagine there has been a ton of study on how these zener references behave and several years ago a paper was published that gives methods for breaking through the specified day-to-day noise floor specified by the manufacturer.

It was an empirical study based on 140 zener references and accurately predicts zener output values during the calibration interval.

The voltage samples during the weeks-long measurement are fitted to the known trend curve using a linear least squares fit and the scatter of the samples is computed using the curve fit residuals. This legitimately separates the zener noise from how its known value is changing over those days and so the variation contributor to the measurement uncertainty is lowered a bit. So the +/- 200 nV you see will usually become something (much) smaller after the predicted value curve fit.

If you want to get more into the weeds and care to read about this method: PROJECTING ZENER DC REFERENCE PERFORMANCE BETWEEN CALIBRATIONS  http://vmetrix.com/ZenerP.pdf. It's authored by NIST alum Clark Hamilton who pretty much invented the JVS and created the software that still runs most of the world's JVS systems.



 

Offline dacman

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #39 on: October 27, 2017, 12:37:42 am »
NIST was certifying JVAs to 0.022 uV/V using a bank of zeners.  That is to within 220 nV.

Since I get NVLAP certified reports traceable "through" NIST to the SI from Fluke, I'll explain one.

I never use the 0.06 uV/V uncertainty as an uncertainty of the Zener.  (The uncertainties I use are even well above 0.2 ppm.)  0.06 uV/V is used in an uncertainty analysis.

The number does vary.  I've seen 0.07 uV/V and years ago used to get 0.1 uV/V (and 0.4 uV/V from the Fluke A2LA lab).

The measurement uncertainty (0.06 or 0.07 uV/V) is at k=2 or about 95%.  (It is an NVLAP requirement that the approximate percentage be on the report.)

Later in the report, it mentions that other factors need to be considered, and if they are not known, Fluke gives some estimates at one standard deviation.
I think the estimates in uV/V (at k=1) went something like:

Temperature: 0.05
Pressure: 0
Humidity: 0.05
Seasonal Effects: 0.058

If the measurement uncertainty of 0.06 uV/V is converted to k=1, then it is 0.06/2 or 0.03.

To get the combined uncertainty at k=1, these terms are RSS'd together and the result is: 0.09625 uV/V k=1
For k=2, this is multiplied by two: 0.1925 uV/V k=2

I would round up to 0.20 uV/V.  If I had the 0.05 uV/V reported uncertainty from Moon Winx, the result would be 0.19 uV/V.

If this was the only reported value I had, I would need to use the uncertainty in the manual, which is 2 ppm/year.

But, I have several calibrations, and these are charted.  The value of the Zener is predicted.  A standard error of the prediction is calculated (the Excel STEYX function can be used for this).  A statistical multiplier for the standard error is used to increase the uncertainty over time.
 
The following users thanked this post: CalMachine, Moon Winx

Offline MisterDiodesTopic starter

  • Frequent Contributor
  • **
  • Posts: 457
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #40 on: October 27, 2017, 01:02:55 am »
Yes, those are good math gymnastic stunts, and then on a simpler level: consider the Zener's inherent noise...say .5uV to a 1uV RMS, .1 to 10Hz - at any given time.  These devices do have a noise floor at the moment you pull a measure from them. One measure is never enough.

Of course as dacman pointed out:  the very best predictor is to develop an un-adjusted long term drift history which is what I pointed out a long ways back in the first post.  A 732 with a long (many years) recorded calibration report history is much more valuable than a new unit, I say.

Back on topic:  How does this all relate to purchasing a 732 -Traceable Calibration Certificate- at Fluke with better than 0.2ppm uncertainty in the first place?  I'm still kind of wondering how that happens - or if that's something they just don't do anymore?  I've talked to 4 people at Fluke, from sales to lab people - all say that just isn't available for the 732.
 
 

Offline Moon Winx

  • Regular Contributor
  • *
  • Posts: 83
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #41 on: October 27, 2017, 01:30:22 am »
Yes, those are good math gymnastic stunts, and then on a simpler level: consider the Zener's inherent noise...say .5uV to a 1uV RMS, .1 to 10Hz - at any given time.  These devices do have a noise floor at the moment you pull a measure from them. One measure is never enough.

Of course as dacman pointed out:  the very best predictor is to develop an un-adjusted long term drift history which is what I pointed out a long ways back in the first post.  A 732 with a long (many years) recorded calibration report history is much more valuable than a new unit, I say.

C'mon man! Nobody takes one sample and calls it a day!  I think you are dismissing the mountains of research that have gone into the performance of these devices... that the noise floor you speak of is much, much lower in practice.

I don't get this: you dismiss curve fitting as a mathematical stunt but then go on to claim the best predictor is.... a curve fit?

Back on topic:  How does this all relate to purchasing a 732 -Traceable Calibration Certificate- at Fluke with better than 0.2ppm uncertainty in the first place?  I'm still kind of wondering how that happens - or if that's something they just don't do anymore?  I've talked to 4 people at Fluke, from sales to lab people - all say that just isn't available for the 732.
 

Looks as though dacman gets them. Maybe he can chime in on that.
 

Offline dacman

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #42 on: October 27, 2017, 02:18:20 am »
MisterDiodes, could you contact someone from the Fluke primary lab or possibly Jeff Gust (the head Fluke metrologist who took over from Dave Deaver).  I sat through a class he taught on calculating the uncertainty of the predicted value of standards (not just Zeners) and he stated that they were giving 0.07 uV/V on Zeners and I told him they were giving me 0.06 uV/V.

I've sat through at least a half-dozen classes on how to calculate uncertainty and two of them were taught by NIST mathematicians.  Before combining terms, they should be converted to one standard deviation.

If I combine all the terms on the report, I get 0.20 uV/V at k=2, just like you are being told, but the measurement uncertainty is stated to be 0.06 uV/V.  (I think pressure uncertainty is estimated to be zero because I'm at the same altitude as Fluke, although pressure is included in seasonal effects.)
 

Offline Edwin G. Pettis

  • Frequent Contributor
  • **
  • Posts: 451
  • Country: us
  • The plural of anecdote is not data.
Re: Fluke 732 calibration price list and procedure
« Reply #43 on: October 27, 2017, 02:21:37 am »
MisterDiodes did not say anything about taking one measurement and calling it a day, read it carefully, "One measure is never enough."

He is not talking about curve fitting either, he is talking about taking all of the actual readings over a period of years (even more than one reading a year if you're adventurous) and doing a simple average of it, that will give a rather accurate figure for the next drift measurement.  These references tend to drift in a general direction at a fairly constant amount every year once they are old enough.  No curve fitting or any other mathematical high jinks are needed.  It is also a fact that you simply cannot average out very low frequency noise over relatively short periods of time (we're talking days at least for an average), these low frequency components are indeed random and cannot be predicted or averaged by any math for short periods.  It is this noise that limits the absolute accuracy and for 732s, that is about 0.2 PPM, sometimes a little lower but that is the best you're going to get and no amount of math fiddling is going to change that.  If that was true, NIST and everybody else would be using such techniques to reduce the uncertainties further beyond the noise floor.  Ask the experts, they'll give you the same answer, 0.2 PPM for a 732 , maybe 0.1 PPM for an exceptional one.  That is all that needs be said, this was the intended subject of this thread in part, nothing to do with JVS as such although the information is indeed interesting.

For quite a few 732s, the noise is actually measured between about 0.5uV and 1uV RMS, I do not see where anybody can claim that noise floor is any lower than that, much lower....not for any zener based standards I've seen.
 
The following users thanked this post: MisterDiodes

Offline Moon Winx

  • Regular Contributor
  • *
  • Posts: 83
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #44 on: October 27, 2017, 03:12:44 am »
 
MisterDiodes did not say anything about taking one measurement and calling it a day, read it carefully, "One measure is never enough."

He is not talking about curve fitting either, he is talking about taking all of the actual readings over a period of years (even more than one reading a year if you're adventurous) and doing a simple average of it, that will give a rather accurate figure for the next drift measurement.  These references tend to drift in a general direction at a fairly constant amount every year once they are old enough.  No curve fitting or any other mathematical high jinks are needed.

Again, c'mon man! All of this is known. It's been studied. You don't need to believe me, you can do a search on scholar.google.com and find study after study on zener voltage reference measurements and behavior from all over the world. The output of the zener references can be predicted, and those predictions from years' past have been verified. This isn't woo or "mathematical high jinks".

It's not a simple average over a period of years, as that would actually give you a much worse prediction. If you would actually read the paper I linked to in a previous post, you would see there is a sinusoidal pattern on top of a linear drift for ALL zener references, just to different degrees, and you can use this evidence to predict the output (and it's been verified), but it's just a little more complicated than a trendline (as there is a sinusoidal overlay), so it is more appropriately called a curve fit.

 
It is also a fact that you simply cannot average out very low frequency noise over relatively short periods of time (we're talking days at least for an average), these low frequency components are indeed random and cannot be predicted or averaged by any math for short periods.  It is this noise that limits the absolute accuracy and for 732s, that is about 0.2 PPM, sometimes a little lower but that is the best you're going to get and no amount of math fiddling is going to change that.  If that was true, NIST and everybody else would be using such techniques to reduce the uncertainties further beyond the noise floor.  Ask the experts, they'll give you the same answer, 0.2 PPM for a 732 , maybe 0.1 PPM for an exceptional one.  That is all that needs be said, this was the intended subject of this thread in part, nothing to do with JVS as such although the information is indeed interesting.

I promise you the output voltage of the 732B can be known to way less than 0.2 ppm. I don't know what else to say other than I scrutinize this every day at work. And NIST knows this too, regardless of what they put on a customer's test report.

It has everything to do with the JVS as there really is no other way to measure the 732B output to a degree that can assign a value to it.
 

Offline MisterDiodesTopic starter

  • Frequent Contributor
  • **
  • Posts: 457
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #45 on: October 27, 2017, 03:55:52 am »
I'm asking if there was a way to purchase a 732 traceable fluke cal certificate better than 0.2ppm uncertainty.  It sounds like what dacman possibly got was a "measure" lab service product from Fluke, not a traceable cert. maybe?  Again: Fluke currently says "No" to any 732 traceable cal cert any better than 0.2ppm (Service center or Park Cal, at least to the four people I've talked to), which is exactly how I reported the process on the first page of this thread.

What Fluke tells me is that 0.2ppm is the spec'd usable noise floor of the 732, and remember:  They are only measuring for 15 days for the cal cert, even at Fluke's primary cal lab level, measured against their JVS - NIST tells me the same thing about the usable noise floor of the 732 and they measure the DUT over 30 days measured against NIST's JVS, and still the end result in reported measured uncertainty on the calibration cert doc is basically the same ==> 0.2ppm (0.19ppm for NIST).  So there is a pattern there, and it doesn't seem to go much below 0.2ppm uncertainty, as Edwin pointed out.  We know that 0.1ppm and below is going to be noisy territory for sure, inherent in the Zener itself, and yes I know it can be possible to extract something that might sorta resemble lower uncertainty if you squint hard enough somewhere down in the muddy slime of noise, but that takes time - the longer the better.  That's harder to qualify that as a traceable cert though - and that's what I'm after. 

I know MoonWinx extracts lower uncertainties, (I get it) but does that pass as a legal cal cert?

And one thing for sure - my customers are going by the uncertainty listed on the cal cert as my basis for my equipment cal cert audits.

I even asked the MoonWinx's smart guys at NIST if there is a "legal" documented way to even have a Vref Volt traceable, cal'd uncertainty Vref system of say 0.1ppm uncertainty available in the lab - and even -they- said that would be very difficult to do without either going to a -very- tight multiple 732 cal schedule (and probably custom characterized & aged 732's), or just purchasing a JVS.

 I've also looked back at a bunch of 732 docs for older cals, and it's never been anything better than 0.3ppm to 0.5ppm for the "standard" NIST traceable cal certs from Fluke, even from over 20 years ago.

So that's what I'm having to go with at this point - this is the information that I'm getting from Fluke -and- NIST  - 0.2ppm uncertainty is the practical best uncertainty limit of a documented, traceable, cal'd 732.  I know it could possible be lower, but as a cert?

I'll have to leave it at that for the moment - because that's what I can verify at this time.
 

Offline CalMachine

  • Frequent Contributor
  • **
  • Posts: 477
  • Country: us
  • Metrology Nut
Re: Fluke 732 calibration price list and procedure
« Reply #46 on: October 27, 2017, 04:52:37 am »
I'm asking if there was a way to purchase a 732 traceable fluke cal certificate better than 0.2ppm uncertainty.  It sounds like what dacman possibly got was a "measure" lab service product from Fluke, not a traceable cert. maybe?  Again: Fluke currently says "No" to any 732 traceable cal cert any better than 0.2ppm (Service center or Park Cal, at least to the four people I've talked to), which is exactly how I reported the process on the first page of this thread.

You've got 2 forum members right here who have recounted on several occasions within this thread the ~'Traceable'~ 17025 accredited certificates from Fluke with a reported measurement uncertainty of 0.06 PPM. Of course these are legal documents....  ISO 17025 accredited calibrations are 100% perfectly legal for the government, lots of the Biomedical industry, some of the consumer audio industry, geothermal industry... I could go on.  In fact, many businesses are now seeking out 17025 accredited cals, which has been made clear that we are 'Traceable to SI through NIST', and does not directly state 'Traceble to NIST'.

I've also looked back at a bunch of 732 docs for older cals, and it's never been anything better than 0.3ppm to 0.5ppm for the "standard" NIST traceable cal certs from Fluke, even from over 20 years ago.

I am not sure of the level of scrutiny on 'NIST Traceable' calibrations back then; but, these days they are one of the lowest scrutinized levels of 'traceable' calibrations.

Your customers' requirements to be 'Traceable to NIST' for the 'LEGAL US VOLT' is barring you from receiving any lower uncertainties.  You aren't going to get any better when your 'LEGAL' best is 0.19 PPM.  Thankfully, being accredited to an international standard, '~traceable~' to SI...  I have the availability to receive more certain measurements on my references.  Receiving a 0.3 PPM uncertainty could be the largest contributor in ones' budget.  Getting that 0.06 PPM (if everything is done right) could mean the world for a scope.
« Last Edit: October 27, 2017, 04:54:35 am by CalMachine »
All your volts are belong to me
 

Offline Moon Winx

  • Regular Contributor
  • *
  • Posts: 83
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #47 on: October 27, 2017, 01:04:46 pm »
According to an section on JVS traceability on Wikipedia (I know, I know)...

Quote
A Congressional act in 1904 established the U.S. Legal Volt to be a quantity defined by the National Bureau of Standards, now the National Institute of Standards and Technology (NIST). With the 1990 international agreement on the Josephson representation of the volt, NIST defined the U.S. Legal Volt to be the same as the international volt representation. Since the success of the first Josephson array voltage standards in 1984, their use has proliferated to more than 70 national measurement institutes (NMI's), military, and commercial laboratories around the world. This has resulted in some confusion about the traceability of non-NMI's that are in possession of a JVS that is, in principle, as good as the national standard. Some guidance on this question is provided in International Standards Organization (ISO) documents that state the general principle that intrinsic standards like the JVS, that have participated in a comparison with an NMI, can claim traceability.

..and another section that might be of interest regarding uncertainty and zener reference noise floor:

Quote
While the voltage appearing across the terminals of a Josephson device is, in principle, given exactly by V = n f /  K J, in any real measurement there are a variety of potential sources of error and uncertainty as listed in Table 2. In the case of a known error, such as a reference frequency offset or a known leakage resistance, a correction can be made. It is then the metrologist’s task to assign realistic numbers to all uncertainties including the uncertainty in the corrections. One method of doing this notes that only items 1 and 2 in Table 2 depend on the voltage across the Josephson array. All of the other components are about the same regardless of the array voltage. Therefore, the combined effect of items 3-8 can be quantitatively evaluated by making a set of measurements of a short circuit using exactly the same algorithm that is used for any other measurement. The standard error resulting from items 3-8 is just the root mean square (RMS) value of the set of short circuit measurements.[64] Additional experiments must be performed to estimate frequency and leakage uncertainty. Internationally accepted procedures for combining uncertainty and establishing confidence intervals is the subject of the BIPM’s Guide to the Evaluation of Uncertainty in Measurement.[65] Typically, the total uncertainty contribution of a Josephson system in a measurement averaging time of a few minutes is a few nanovolts. Since the most common use of these systems is the calibration of Zener standards with a noise level of 50-100 nV, the contribution of the Josephson system is negligible.

This jives with my experience in measuring the noise in the 732B outputs. 50 - 100 nV or around 0.01 ppm @ 10 V. Again, much less than the specifications state for the 732B.

I wonder if the 0.19 or 0.20 uncertainty given to you is because this item has no history with the calibration lab performing the measurement? Did they mention if you sent it in again within a year you could get a better uncertainty?

EDIT to add: I just checked on our zener reference measurement system that uses a 734B bank as a standard and our typical measurement uncertainty on the cal reports for this system is around 0.16 ppm, which is good enough for the most common use of these standards (artifact cals of the 5720 & 3458).

« Last Edit: October 27, 2017, 01:10:43 pm by Moon Winx »
 

Offline MisterDiodesTopic starter

  • Frequent Contributor
  • **
  • Posts: 457
  • Country: us
Re: Fluke 732 calibration price list and procedure
« Reply #48 on: October 27, 2017, 04:38:34 pm »
Finally the veil of bullshit is pulled away.  Verified info from Fluke (lab manager #5 I've talked to) and NIST (PhD metrologist).

First, in response to CalMachine:

Yes Fluke will do a raw measure (to ISO17025) of a 732 and get you to .06ppm uncertainty against Fluke's JVS.  As I have explained several times.  That is a certified "voltage measure", not a calibration certificate.  A Calibration Certificate (to ISO 17025 AND Z540-1-1994) from Fluke also includes a Test Uncertainty Ratio of 10:1 and the entire instrument is tested to be in spec.  A Calibration Certificate is what I have to supply to my customers, they would never, ever go for just a certified measure.   Yes an ISO17025 qualification includes traceable to SI through NIST, but you have to make the distinction between a calibration certificate vs. just a voltage measure.

The little tidbit you're leaving out from a Fluke .06ppm uncertainty measure is the warning from Fluke that uncertainty is valid for about 2 hours on a typical 732.  That's why they can do this as on on-site measure down to .06ppm with some multiple freshly cal'd 732's, do your measure on-site, and then get back to the JVS and recal those 732's right away.

To recap:  A Fluke Calibration Certificate on a Fluke 732 is performed to ISO 17025 and ANSI Z540-1, and is traceable to the SI through NIST, and if it is purchased thru Fluke that includes a -complete- instrument verification test and 10:1 Test Uncertainty Ratio (TUR).  This is the practical usable limit of a single 732 if it's being shipped anywhere where traceability back to SI / NIST volt is maintained.

NOTE:  You can also have a Calibration Certificate and more Certified Measures done at more frequent intervals to help characterize your 732 to lower uncertainty - but that will vary from unit to unit and has a drift history is developed.

Now for use in your own lab you can of course do whatever statistical math you like on your own 732 .  For actual commerce beyond your lab it's the Calibration Certificate measurement uncertainty that counts as the basis for any product measure we send to our customers.  At least that's only what my customers demand to see - If I tried to claim some measurement uncertainty based on some "certified measure" they'd just laugh at me.  I have to be able to show the complete third party traceable calibration certificate.

To MoonWinx:

Finally you've got a verifiable number from Fluke Cal.  Yes, on a 734a where four 732b's were cal'd at the same time or close to same time, you CAN get a ISO17025 Calibration Certificate for the entire 734a listed as 0.16ppm uncertainty as a group  - this number will vary slightly depending on the actual measured noise of each 732, but that is perfectly in the ballpark of a purchased product.

Thanks for double-checking that and coming up with a more realistic traceable absolute value uncertainty spec.

You have the luxery of having a JVS to play with so that's easy to see those numbers on a 732 even better than spec.  Most of us don't have that capability.

From NIST:
To clarify about a JVS - Yes it is an intrinsic standard ONLY if the operating lab can prove that they have an accurate basis for time and frequency measurement (i.e. GPS or local atomic clock).  So the JVS by itself is an intrinsic standard, be in practical terms it requires the minimum known uncertainty of the JJ-Array chip excitation frequency.  Also in practical terms for the US - Currently All 20 labs that have JVS's are NVLAP / ISO17025 and participate regularly with NIST to inter-compare JVS's and several 732bs - and NIST also inter-compares with all the other international labs as well. 

A little more info on the origin the 10:1 Test Uncertainty Ratio (TUR) or Gage Maker's Rule which is typical lab practice:

https://blog.alliancecalibration.com/tur-what-is-it

When you get other measures from Fluke they are typically list as a TUR of 10:1 or in some cases 4:1.  A lab measurement of 1:1 would be a bit suspect for lab practice.
« Last Edit: October 27, 2017, 04:46:48 pm by MisterDiodes »
 
The following users thanked this post: Edwin G. Pettis

Offline CalMachine

  • Frequent Contributor
  • **
  • Posts: 477
  • Country: us
  • Metrology Nut
Re: Fluke 732 calibration price list and procedure
« Reply #49 on: October 27, 2017, 05:30:56 pm »
Finally the veil of bullshit is pulled away.  Verified info from Fluke (lab manager #5 I've talked to) and NIST (PhD metrologist).

First, in response to CalMachine:

Yes Fluke will do a raw measure (to ISO17025) of a 732 and get you to .06ppm uncertainty against Fluke's JVS.  As I have explained several times.  That is a certified "voltage measure", not a calibration certificate.  A Calibration Certificate (to ISO 17025 AND Z540-1-1994) from Fluke also includes a Test Uncertainty Ratio of 10:1 and the entire instrument is tested to be in spec.  A Calibration Certificate is what I have to supply to my customers, they would never, ever go for just a certified measure.   Yes an ISO17025 qualification includes traceable to SI through NIST, but you have to make the distinction between a calibration certificate vs. just a voltage measure.

The little tidbit you're leaving out from a Fluke .06ppm uncertainty measure is the warning from Fluke that uncertainty is valid for about 2 hours on a typical 732.  That's why they can do this as on on-site measure down to .06ppm with some multiple freshly cal'd 732's, do your measure on-site, and then get back to the JVS and recal those 732's right away.

To recap:  A Fluke Calibration Certificate on a Fluke 732 is performed to ISO 17025 and ANSI Z540-1, and is traceable to the SI through NIST, and if it is purchased thru Fluke that includes a -complete- instrument verification test and 10:1 Test Uncertainty Ratio (TUR).  This is the practical usable limit of a single 732 if it's being shipped anywhere where traceability back to SI / NIST volt is maintained.

NOTE:  You can also have a Calibration Certificate and more Certified Measures done at more frequent intervals to help characterize your 732 to lower uncertainty - but that will vary from unit to unit and has a drift history is developed.

Now for use in your own lab you can of course do whatever statistical math you like on your own 732 .  For actual commerce beyond your lab it's the Calibration Certificate measurement uncertainty that counts as the basis for any product measure we send to our customers.  At least that's only what my customers demand to see - If I tried to claim some measurement uncertainty based on some "certified measure" they'd just laugh at me.  I have to be able to show the complete third party traceable calibration certificate.

 :palm:

What you say and what I've actually received, and experienced, do not seem to corroborate.

If you could be so kind as to interpret this 'CERTIFIED MEASURE' (according to you)  document I received back with the 732B?  I would greatly appreciate it.  I've got years of data from PSNA, FLUKE, and Process Instruments on this unit.

All your volts are belong to me
 
The following users thanked this post: Moon Winx


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf