-
We just received a newly calibrated 732a back from Fluke Everett, WA labs - and I thought I'd share the current price list and customer procedure here. Prices are subject to change and there may be adjustments depending on what you ask for. These are just baseline prices we got from Fluke.
Before you send in an older 732a (or 732b), I recommend doing (as required) a complete battery replacement, adjust battery charger, replace power supply capacitors, etc. Let the 732 stabilize for a few months at least and make sure noise and stability is in spec. Also note that when your 732 gets returned all the adjust pots will be sticker-ed over as well as the Battery ON-OFF switch, power module, battery module and top cover. Breaking any sticker invalidates the calibration - so if that's important to your business make sure the 732a/b is up and running well -before- sending it in.
The price of calibration listed below is just for the 10V main output. On a 732a if you want the 1V and 1.018V outputs adjusted to ppm accuracy, that's an extra $450. The cost to do the 1.018V output on a 732b is an extra $300. The uncertainty on these secondary outputs is only going to be 0.7ppm.
We use a 752a to get an accurate 1V and 100mV derived from the 732's, so we don't use the secondary outs from the 732's anyway.
BY DEFAULT the 732 main 10V output is NOT adjusted to 10V +- 0.5ppm during cal. Most of time you want an accurate drift history of your unit and for that you don't really want intervening adjustments; as long as your 732 is within a reasonable distance of 10V, Fluke does NOT recommend adjustment, they will just do a measure only for Z540 data report - that way you know exactly the absolute value of your 732, which is what you're after anyway. They will perform an adjustment if requested. Keep this in mind though: If you do request an adjustment on an older 732a that you're not sure about where the pot adjustment range ends, you might be out of luck. Fluke will not repair 732a's, nor will they change the 732a's coarse adjust jumper board - so it's important to have a rough idea of the absolute value of the 732a before sending it in for cal and requesting adjustment.
The cost to EVALUATE the repair cost of a 732b is $300. In other words it'll be $300 minimum to just find out what's wrong with a 732b. The actual repair cost will go up from there.
If you can't drive in your 732 to Everett, for an extra $375 + shipping they will rent you a shipping case for your 732a complete with 4 extra battery packs for longer hot run time. This is how they will ship your 732a back to you as an express freight shipping packing weight of 128 lbs. If the "Cal" light goes out (meaning the unit went cold) the calibration is no longer valid. So make sure to coordinate shipping of your unit.
Fluke Labs has a measuring uncertainty ratio of 10 to 1...In other words they are measuring and adjusting your 732 to several of their 732's which are on a tight calibration rotation (30, 60 or 90 day) with their JJ-Array, and their measuring resolution is 10X finer than the measured value printed on your report. When your unit leaves the facility your measured absolute voltage value will be to 0.3ppm uncertainty traceable to NIST with 95% confidence. If you had an adjustment done it is typically to 10V +- 0.5ppm or to whatever you've requested and paid for, and then the actual measured value is also on the report.
Cost of calibration for 732's:
ANSI Z-540 (This is most common) You get the 1 yr cal certificate and measured value of 10V main output: 732a is $789, For 732b: $738
A2LA Accredited Calibration - If you need this higher level of legal paperwork you'll know - for 732a: $1126, for 732b: $1075
Primary Standard Lab Calibration: If you need to get down to 0.2ppm uncertainty relative to NIST, you can calibrate your unit directly against the JJ-Array. For 732a $1,885. For 732b: $1780
The procedure is as follows, and accurate as of October 2017.
1. Call in to Fluke Calibration, and get setup with your contact info. If you're a business you send in your business info also, for instance sales tax exemption records, etc. You tell them the serial number of the 732, and which calibration service you want, and setup payment account, Credit Card, or whatever payment method. They will get back to you within a day or so with an RMA number.
NOTE: There is an online calibration system on the web, but for whatever reason it does not yet work with 732's. You have to call in to get setup with RMA. At least as of Oct 2017.
2. Once you have RMA number, and know how you're going to get your 732 to Fluke (Ship it in or drive it in) you call in to schedule about when your unit can arrive. They want to make sure there is space on the racks available for your unit as soon as it arrives. So you get setup with an appointment for your 732 arrival day.
3. Now get your 732 to Fluke - either ship it or drive it in. They will give you the proper address if you're driving it in - and you need to get that from them. They prefer to receive 732's for calibration on Monday / Tuesday / Weds if possible.
4. Once they get the 732 it is plugged into mains power. The CAL light should still be on when they get the unit. If the CAL light is off then that will require a longer stay to make sure the unit stabilizes before running any tests.
5. The unit is tested for noise, battery pack operation and initial voltage measure is taken.
6. A few days later you will get notice that your unit passed initial functional checks and will go in for full calibration cycle, usually about 2 weeks. You will get an estimate of Ship Day. During this time your unit is being scanned and inter-compared with several Fluke 732a/b units, and they look for any signs of TC problems, oven issues, output noise, etc.
7. At any time during this period if the 732 fails to meet spec it is marked as a Non Functional unit. If it is a 732b they will discuss repair options. 732a's are just returned to owner.
8. On the morning of Ship Day they will call or email you to let you know the unit is ready. They pull the last measure voltage value at just before 8am (at least for us) and run the math and print the reports. Then your unit gets stickered: All the adjust pots are covered, the cover screws, the rear module access screws and Battery On/ OFF switch. It is ready to travel by about 10am.
They will keep the unit plugged in until you arrive to pick it up in person, or they will get the shipping container prepped and ready to go - when the shipping company truck pulls up they will unplug the unit, seal it up in it's box and place it on the courier truck within a few minutes. Either way they will keep those batteries charged up until just before it leaves the facility.
After all that, this is what a working - calibrated - 732a looks like, ready to go to work (cert number removed for EEVblog). You hardly ever see this on eBay
-Have Fun!
-
Thank you for detailed walk-thru. I have a question regarding tempco. You mention they will check for TC problems, how is that going to happen?
Is there a additional service available to get tempco measured as well? -
RE TC - In a simplified sense the racks with devices under cal is measured / controlled for temperature separately from the rack with the reference units. As the rack with DUT's varies very slightly over temperature (within a degree or so) they are looking for problems during the test period, and to make sure each unit's oven temp sensor / Vref output is staying solid-stable. That's how it is described to me, and that's what I can relay to you here. It's more they aren't measuring actual TC, it is part of a Go / No Go check - and they are making sure that the 732 meets all published specs as tested over the test period.
Their absolute value measuring resolution internal to Fluke labs is at .01ppm, so they'd be able to detect an out-of bounds shift on Vref output relativity easily even if the DUT ambient temp shifted less than a degree. The final measure you get on your certificate is 7.5 digits (0.1ppm resolution), and again: Uncertainty to NIST is 0.3ppm for the Z540 cal @ 95% confidence (For the lower-rated cals - the Primary Labs cal is 0.2ppm uncertainty to NIST). At least up to the moment your device leaves the back room and the final measure is taken - that is the recorded absolute voltage value of your unit. You will get the official time stamp of the last measure taken so you can start calculating your extrapolated expected drift compensation down the minute Fluke took it's last reading.
Beyond that: You're more than welcome to pay the $1,500 "Observation Fee" where you can go in the magical back room for up to 4 hours and watch your unit get calibrated - just leave your cell phone and any other personal electronics at the security guard desk. For some calibration requirements it's not good enough that you have the paper certificate - if you're in Aerospace especially you'll need to be physically present for the last few hours of calibration and sign off (aka "Witnessed Calibration") and then you'd be free to examine the exact process in detail - and probably still not be allowed to post observed process details online. Fluke has a lot of proprietary processes in that room they wish to keep proprietary, and the Observation Fee setup also includes a pretty strict NDA and background check.
As it is when you get to Fluke you're allowed only into the shipping / receiving desk area, and you must show your ID RMA and paperwork for equipment drop off and pickup - and that's as far as you're allowed into the building.
That's all I can describe online at least for standard cal procedure. If you need to have a unit specifically measured for TC I think they can do that but you'd have to contact Fluke for a "Custom Lab Service" procedure.
-
They've given us 0.06 PPM. Go to page 20 in this document and look at the top: "http://download.flukecal.com/pub/literature/NVLAP%20Labcode%20105016_FPLAccreditationScope_Exp%2020180630.pdf"
-
...Check Note 4 - That .06ppm might be misleading and generally more relative to Fluke Labs - and sort of leaves the exact spec up in the air. If that's what you're after, that's OK but all of our cals are spec'd to be relative and traceable to NIST only - which is the real landmark to measure against most of the time here in the US - at least if you have to pass an ISO-type cal audit. 0.3ppm uncertainty to NIST is the best you'll get from Fluke for standard z540 cal, and 0.2ppm relative & traceable to NIST if you cough up the extra $1200 for the Primary Standards cal. on a 732. I just double-checked that with their lab services sales department. If you're after a cal that's relative only to Fluke labs (and done onsite, maybe?) yes then it can be reduced uncertainty. Most certificate audits we need won't accept that though.
At that point it's sort of bragging rights anyway, really. How long will it take your 732 to drift .06ppm?
EDIT: Also - you got that document for Fluke Park Labs NVLAP (voluntary), not ANSI scope at their Cal lab location. That 10V reference they talk about in that NVLAP scope is their own J-array, not NIST's. NIST traceability is what counts, legally, for a lot of doc requirements. -
Here is the same document on the NIST website:
"https://www-s.nist.gov/niws/index.cfm?event=directory.retrieveDocument&documentId=EFA2EE38EC04463227D2986EC2B09EF1CC011DE7E3B3199D634405632CA6E57FABB392A73B80A93783019527580D5CDAC5E2364510C496ED90E740F1D012CE63&csrfToken=509FCCC06F4FFBBE9FA912B752984623C169A760"
Here is the page it's on:
"https://www-s.nist.gov/niws/index.cfm?event=directory.detail&labid=248&programId=4&csrfToken=509FCCC06F4FFBBE9FA912B752984623C169A760"
If you look at the first page, on the right side near the top, it mentions Z540. We have gotten more than one NVLAP certified 0.06 PPM on-site calibration. -
...Check Note 4 - That .06ppm might be misleading and generally more relative to Fluke Labs - and sort of leaves the exact spec up in the air. If that's what you're after, that's OK but all of our cals are spec'd to be relative and traceable to NIST only - which is the real landmark to measure against most of the time here in the US - at least if you have to pass an ISO-type cal audit. 0.3ppm uncertainty to NIST is the best you'll get from Fluke for standard z540 cal, and 0.2ppm relative & traceable to NIST if you cough up the extra $1200 for the Primary Standards cal. on a 732. I just double-checked that with their lab services sales department. If you're after a cal that's relative only to Fluke labs yes then it can be reduced uncertainty. Most certificate audits we need won't accept that though.
At that point it's sort of bragging rights anyway, really. How long will it take your 732 to drift .06ppm?
I think you might have your information a little off, or perhaps I am wrong. But, the measurement uncertainty I've received from Fluke in 2014,15,16 on their 'Service Lab' 17025 accredited cal was 0.3 PPM. The service lab is where a majority of their calibrations occur. This most recent year I spring for the Fluke's 'Standards Lab' 17025 accredited cal. This is where I received a reported measurement uncertainty of 0.06 PPM.
Looking at their scope and the ~ J Array CMC of 0.01 PPM, they can very easily report a 0.06 PPM measurement uncertainty (depending on how stable the DUT 732 in question is)
Note 4 specifically refers to 'On-site' calibrations. Which would be relevant if you sprung for the calibration package where they send you their characterized 732B. I think it's just a coincidence that their reported measurement uncertainty on these few 732Bs happens to be the same as their onsite CMC.EDIT: Also - you got that document for Fluke Park Labs NVLAP (voluntary), not ANSI scope at their Cal lab location. That 10V reference they talk about in that NVLAP scope is their own J-array, not NIST's. NIST traceability is what counts, legally, for a lot of doc requirements.
Are you under the assumption that Z540 or 17025 certificates from labs that are not NIST, are not NIST traceable? -
0.06 PPM is the uncertainty of the measurement at the time of calibration. It is not the accuracy of the 732. Reports from Fluke on these types of calibrations have a section on adders such as temperature, humidity, pressure, and seasonal effects, which if combined are greater than the 0.06 PPM measurement uncertainty. Accredited laboratories are supposed to trend standards like these and standard resistors and gauge blocks, and most any fixed accredited standard, and they are supposed to determine the uncertainties themselves using the accredited reported values, which can include a trend prediction.
-
See Page 2, Calibration DC Reference - this is what we're using for Z540 scope - and what Fluke Everett Service Center Cal lab (not Fluke Park) will give you when you send in a 732- and it is indeed 0.3ppm, and printed on my cert I received last week:
https://www.a2la.org/scopepdf/2166-01.pdf
That's all I can describe to you, and that matched up to my reality.
When I requested Z540, relative and traceable to NIST, this is what I got: 0.3ppm, and that is what I was told was the best uncertainty spec available for that request - outside of going to Primary Lab cal. My cal documentation includes the words "traceable to NIST" which is what I have to have to pass our audit, and it also includes all the equipment and traceable cert numbers from Fluke's equipment.
Now they will MEASURE to a finer resolution of course relative to Fluke internal references (10X better than printed on the report) - but the UNCERTAINTY relative to NIST is no better than 0.3ppm. You have to remember that there are transfers between NIST and FLuke JJ-Array, and then from Fluke JJ-Array to cal lab 732's, and then from cal lab 732's to your 732 under test. There is no such thing as a "perfect" transfer - just low, non-zero uncertainty at every step.
NOW: The sales person I double checked with implied that onsite 0.06ppm seemed to believe that is relative to Fluke's own JJ-Array - and not directly traceable to NIST due to Fluke traveling to onsite location and introducing additional unknown uncertainty <maybe>. If you look at Note 4 on the Fluke Park doc that shows that 0.06 ppm uncertainty is -variable-, and that's where the difference lies. Now maybe our wires got crossed somehow - but all I know is I got what I needed. I'll let someone else research the finer points of bragging rights documentation on their time.
After all it is mostly all documentation only differences on uncertainty. Unless you have a JJ-Array handy it'll be hard (impossible?) to accurately measure any difference in the end to verify 0.3ppm vs. 0.06ppm uncertainty - especially once the equipment has left the building. Once any drift sets in you're probably outside those limits anyway, and back to the bounds of published operating specs and known drift history extrapolation estimates.
Have fun!
-
Now they will MEASURE to a finer resolution of course relative to Fluke internal references (10X better than printed on the report) - but the UNCERTAINTY relative to NIST is no better than 0.3ppm. You have to remember that there are transfers between NIST and FLuke JJ-Array, and then from Fluke JJ-Array to cal lab 732's, and then from cal lab 732's to your 732 under test. There is no such thing as a "perfect" transfer - just low, non-zero uncertainty at every step.
NOW: The sales person I double checked with implied that onsite 0.06ppm seemed to believe that is relative to Fluke's own JJ-Array - and not directly traceable to NIST due to Fluke traveling to onsite location and introducing additional unknown uncertainty <maybe>. If you look at Note 4 on the Fluke Park doc that shows that 0.06 ppm uncertainty is -variable-, and that's where the difference lies. Now maybe our wires got crossed somehow - but all I know is I got what I needed. I'll let someone else research the finer points of bragging rights documentation on their time.
After all it is mostly all documentation only differences on uncertainty. Unless you have a JJ-Array handy it'll be hard (impossible?) to accurately measure any difference in the end to verify 0.3ppm vs. 0.06ppm uncertainty - especially once the equipment has left the building. Once any drift sets in you're probably outside those limits anyway, and back to the bounds of published operating specs and known drift history extrapolation estimates.
Have fun!
I am an engineer over several JVS systems at my work, and I don't know how Fluke does things but it doesn't make any sense to me that there would be traceability to NIST from Fluke's JVS. It's an intrinsic standard with the traceability coming from the frequency source (whatever that may be, for us it's a NIST-corrected GPS signal) and the uncertainties for that are WAY below the uncertainty for the JVS volt (somewhere in the order of parts in 10^11 or 12. Every few years the big US primaries do a JVS intercomparison, usually led by NIST, to see how well each lab compares to one another. The results are anonymous, of course, but you know your lab's measurement so you can see how your system is performing in comparison to other labs. Labs may find they are a significant amount away from the weighted mean and adjust their measurement uncertainty accordingly.
We tend to give a measurement uncertainty of 0.05 ppm for the 732Bs we measure with the JVS, and we'll give a printout of the predicted values w/ uncertainty of that value for each day for a little over a year. If we have enough historical data, the uncertainty stays pretty low throughout the interval because the 10V output can be predicted to a great degree. If we have no historical data, the predicted value stays constant while the measurement uncertainty rises throughout the interval, usually surpassing 2 ppm by the end.
EDIT: To add, the uncertainty of the JVS volt is REALLY small, I think even sub-ppb, but I'll have to double-check. The 50 nV (k = 2) uncertainty we give comes from the transfer of that volt to the 732B along with all the sources of error that entails. -
I work for a gov't lab, so we can't compete for or accept work that can be done commercially. Sorry, because I personally would love to help out. I just discovered this forum several months ago and I wish our workers had 1/10th of the curiosity/interest in metrology that I see posters here have! Metrology as a hobby?? I never knew such existed!
-
OK, so keeping a calibrated 732 at home is 150 dollar a month. OK.
You could also buy an extra bedroom or a car from this money. -
That's why Joe Geller's SVR-Ts were such good value.
-
so what would maintaining a bank of 8 saturated standard cells cost?
regards.
-
OK, so keeping a calibrated 732 at home is 150 dollar a month. OK.
You could also buy an extra bedroom or a car from this money.
This is the cost of doing precision business and chasing PPM's, so if you're doing this you probably want to be seeing a profit somewhere. That's why nobody gives away calibrations.
As a business, AND if you do work for any customer who deals with the US Govt. or at least who worries about cal cert documentation - you really have to be able to show traceability to NIST most of the time, (or whatever governing body is in your country). Fluke isn't a government entity, so no matter how good and fancy and stable their reference is - unless they can show traceability to NIST's volt as an unbroken measure chain - their measure is not worth snot for a certificate audit. At least not for what we do or our customers.
By the way - you CAN retrieve every cal certificate on every piece of equipment Fluke used between your device and the actual NIST transfer to Fluke. That's an extra documentation fee of $150 per hour, they said it runs around 4 or 6 hours depending - they will give you a closer estimate if you need to order that.
-
But come on. I am talking about the volt nut, who wants to have the absolute accuracy at home. Just becuase it is a hobby.
-
...I get it. Hence the "nut" in the label, and I respect that. Whatever floats your boat, Have Fun! - all I'm doing is reporting on the current price list and sharing the real experience of how it worked as of last week.
-
See Page 2, Calibration DC Reference - this is what we're using for Z540 scope - and what Fluke Everett Service Center Cal lab (not Fluke Park) will give you when you send in a 732- and it is indeed 0.3ppm, and printed on my cert I received last week:
https://www.a2la.org/scopepdf/2166-01.pdf
That's all I can describe to you, and that matched up to my reality.
When I requested Z540, relative and traceable to NIST, this is what I got: 0.3ppm, and that is what I was told was the best uncertainty spec available for that request - outside of going to Primary Lab cal. My cal documentation includes the words "traceable to NIST" which is what I have to have to pass our audit, and it also includes all the equipment and traceable cert numbers from Fluke's equipment.
Now they will MEASURE to a finer resolution of course relative to Fluke internal references (10X better than printed on the report) - but the UNCERTAINTY relative to NIST is no better than 0.3ppm. You have to remember that there are transfers between NIST and FLuke JJ-Array, and then from Fluke JJ-Array to cal lab 732's, and then from cal lab 732's to your 732 under test. There is no such thing as a "perfect" transfer - just low, non-zero uncertainty at every step.
NOW: The sales person I double checked with implied that onsite 0.06ppm seemed to believe that is relative to Fluke's own JJ-Array - and not directly traceable to NIST due to Fluke traveling to onsite location and introducing additional unknown uncertainty <maybe>. If you look at Note 4 on the Fluke Park doc that shows that 0.06 ppm uncertainty is -variable-, and that's where the difference lies. Now maybe our wires got crossed somehow - but all I know is I got what I needed. I'll let someone else research the finer points of bragging rights documentation on their time.
After all it is mostly all documentation only differences on uncertainty. Unless you have a JJ-Array handy it'll be hard (impossible?) to accurately measure any difference in the end to verify 0.3ppm vs. 0.06ppm uncertainty - especially once the equipment has left the building. Once any drift sets in you're probably outside those limits anyway, and back to the bounds of published operating specs and known drift history extrapolation estimates.
Have fun!
Ofcourse that matched up to your reality. You went for a Z540-1 calibration at their service lab. You won't be getting more than 0.3 PPM uncertainty traceable to NIST, there.
But to go and make the claim that Fluke, as an entity, cannot report a measurement (Traceable to NIST) that is below 0.3 PPM, is hogwash. What do you think their NVLAP scope of ISO 17025 accreditation is for? It explicitly states how 'uncertain' their best measurements are traceable to NIST. See notes 1 and specifically note 2.
Note 4 for is specifically referring to 'Field' or on-site measurements that do not take place in their laboratory....I get it. Hence the "nut" in the label, and I respect that. Whatever floats your boat, Have Fun! - all I'm doing is reporting on the current price list and sharing the real experience of how it worked as of last week.
I very much appreciate the information and content you provide! I hope you don't feel I am being combative. I just want to make sure the right information gets propagated out to the people.
-
Sheesh you're a lot of work
Hang tight - I'm on the phone with Fluke now, as I would like to see the difference as well. They are researching that answer. Right now, the best -NIST traceable- uncertainty cert they offer is 0.2uV/V (.2ppm) for primary standards cal, .3ppm uncertainty traceable to NIST for certificate. That's it. Quote -un Quote.
There is no difference between uncertainty between Z540 vs A2LA cert levels, but with primary cal there is one less transfer involved, and so slightly better uncertainty.
As I noted before - OF COURSE they can do a lower uncertainty measure - but traceable chain to NIST or not for legal certification is the question.
This is what the sales dept. is telling me. So I am asking the difference between that and the Fluke Park labs scope that shows onsite 10V cal to .06ppm uncertainty, but notice that includes Note 4, which leaves that completely open ended. Otherwise Fluke Park will give you 0.4uV/V traceable uncertainty.
I am currently waiting on hold... I'll let you know when I know. -
Sorry CalMachine - right now it's only 0.3ppm or 0.2ppm uncertainty traceable to NIST certificate for a 732, that's what they are telling me. They are going to have someone else get back to me on clarification of a custom lab service quote for anything lower uncertainty for a legal -traceable- NIST cert on a 732.
So at the moment, no hogwash involved. but I'm moving up the sales dept chain. And presumably moving into even more betterer bragging rights documentation, which should blow even more money on a piece of paper...
-
CalMachine - I just got a call back from sales dept. - they wanted to let me know they are referring my question to someone in the metrology group for more advanced scope of accreditation questions on custom lab services.
But right now, the official word from their sales dept. on 732 cals for Z540, A2LA is 0.3ppm uncertainty traceable to NIST on the cert, and 0.2ppm for primary labs cal. This is at Everett Service center, and that's all they could sell me right now for NIST traceable uncertainty on a 732 - which is all I'm interested in, because that's what I have to have to pass a cert audit.
It could be that custom lab services will offer even more exciting ways to burn through the cal budget, we'll see. -
But right now, the official word from their sales dept. on 732 cals for Z540, A2LA is 0.3ppm uncertainty traceable to NIST on the cert, and 0.2ppm for primary labs cal. This is at Everett Service center, and that's all they could sell me right now for NIST traceable uncertainty on a 732 - which is all I'm interested in, because that's what I have to have to pass a cert audit.
It could be that custom lab services will offer even more exciting ways to burn through the cal budget, we'll see.
I think you should re-read my last comment.
I was agreeing with you, that at the service lab you won't be getting better than 0.3 PPM. Their service lab is A2LA accredited.
I specifically stated their 'Standards Lab' (Park Lab) and their NVLAP scope of 17025 accreditation. You are thinking their service lab scope applies to their standard lab, for some odd reason.
I wouldn't seek out, concrete, technical information from sales staff. They have technical support numbers for a reason. -
I still thinks it's very odd that Fluke would state NIST traceability for their 732B cals. I'm pretty sure they have a Josephson volt so I wonder why they would spend the extra shit-ton of money to have their own 732s calibrated at NIST instead of sending to their primary lab for a JVS measurement.
-
I still thinks it's very odd that Fluke would state NIST traceability for their 732B cals. I'm pretty sure they have a Josephson volt so I wonder why they would spend the extra shit-ton of money to have their own 732s calibrated at NIST instead of sending to their primary lab for a JVS measurement.
That is what they have to do... All 17025 accredited calibrations are traceable to NIST. All Z540 calibrations are traceable to NIST. Even lower, 'NIST Traceable' cals, which are sometimes colloquially called 'as found cals', are traceable to NIST. Perhaps there is a misunderstanding of what 'traceable to NIST' actually means?
The very first sentence on their calibration certificate states 'This calibration is traceable to the SI through recognized national measurement instututes (NIST, PTB, NPL, NIM, NRC, etc.)'. -
Perhaps I am wrong in my understanding by what it means to be traceable to the SI through NIST, and directly traceable to NIST