Fluke 732C

[manganin]

10V main output plus 1V and 100mV divided outputs. Release 2018.

Anyone else heard the rumor?

[dr.diesel]

I have not heard that, but hears to hoping!

[TiN]

I found that unlikely, but wouldn't bet last penny . Maybe the align new standard for the SI volt redefinition, but the uncertainty difference is much smaller than any room-temperature solid-state reference can realize, no?

P.S. hope they don't put the flashy LEDs all over it..

[CalMachine]

More info!

 

[Edwin G. Pettis]

Sorry guys but Fluke hasn't put out any such mention of a 'new' 732.....the question is why would they, there is no necessity for it.  And no, any slight change in the 'definition' of the volt isn't going to affect any of the 10 volt references, they are all adjustable.

[CalMachine]

.....the question is why would they, there is no necessity for it.  And no, any slight change in the 'definition' of the volt isn't going to affect any of the 10 volt references, they are all adjustable.

Why would they??  Easy.

$$$$$$

I agree, that any redefinition of the volt won't bring a need for a new reference...  but, I think it's about time they updated their tried+true reference.  All of their calibrators got a face lift... I don't see any particular reason to leave out the 732B?

[chuckb]

Maybe they will bring some of the Wavetek / Datron 7000 series features to the 732 series.
1. High isolation power supply, so you don't have to disconnect the 732b from power to calibrate it
2. What else?

[CalMachine]

Maybe they will bring some of the Wavetek / Datron 7000 series features to the 732 series.
1. High isolation power supply, so you don't have to disconnect the 732b from power to calibrate it
2. What else?

What Manganin mentioned would be fantastic.  The 1V and 100 mV out??  Yes plz.   Perhaps some kind of 10x amplifier after the 7=>10V boost to give 100V?  Or even implement a PWM topology instead?  I know there are improvements to be made on the 4910.

[Jay_Diddy_B]

[humour]

I understand the improved performance will come from the special construction:




Especially the 'Special Needs Soldering'.

[/humour]

Jay_Diddy_B

[CalMachine]

[humour]

I understand the improved performance will come from the special construction:




Especially the 'Special Needs Soldering'.

[/humour]

Jay_Diddy_B

You're looking at ppb stability with this quality of craftsmanship!   

[Edwin G. Pettis]

Hilarious, I laughed out loud when I seen it.....yeah, PPB, in a pig's valise!

[dr.diesel]

I might be out of touch with today's needs, but given how good the 732B already is, is there a market for a higher class unit?  < Too bad a pile of NDAs prevent a very interesting discussion here.

Or is the demand for the 732B still enough to consider a cost cutt/refined (uh hum cheapened probably) unit? 

And well Danaher, which I just realized changed their name to Fortive.   

[HighVoltage]

You're looking at ppb stability with this quality of craftsmanship!   

Wow, memories of the funny past with that guy.
LOL, may be Fluke hired him for the 732C

Well more seriously ... I would like to see a 732C.... Finally we would see hopefully more 732B on the secondary market and I could get one for a reasonable price.

[Jay_Diddy_B]

Hi,

I looked at 732B teardown. I don't know if it is late production, I suspect that it isn't. The teardown showed a thru-hole board. This is a target for redesign because the thru-hole parts will be getting hard to get.

The battery chemistry may be another target. VRLA or SLA is still a good for this application, but Li-ion may be considered sexier.

The reference section may be a target for value engineering. It may be that 732B are much better than the specification.

Regards,

Jay_Diddy_B

[TiN]

Li-ion may be considered sexier.

And welcome to all the shipping regulations/limitations/restrictions.
From my experience "sexier" and "prettyier" are next to the last priority items in metrology field.

[manganin]

the question is why would they, there is no necessity for it.

I might be out of touch with today's needs, but given how good the 732B already is, is there a market for a higher class unit?

I was told that problems meeting the RoHS directive were the main reason. My source didn't know if it is going to be a 732B based redesign or something totally new. Both approaces can be easily justified.

P.S. hope they don't put the flashy LEDs all over it..

So last season. Nowadays that would (unfortunately) be a color touchscreen.

[TiN]

color touchscreen.

Mhm, with linux running 4-core Cortex-A9. But I still believe in common sense of Fluke marketing given the battery power requirements?

[Kleinstein]

RoHS and the option to use a different battery chemistry would be good reasons for a new model. Li would be a problem to shipping, but could be a nice option in the lab or travel by car. Ideally one would have hot swap-able battery packs with Li / SLA / NiMH options.  I doubt they would redesign the oven part, excepts if they need to for RoHS or similar, but a more compact design here could make it work with lower power.

It would not make much sense to add more voltages (except 10 V and raw 7.xx V) - this is meant as a portable voltage standard only. So weight and power can be important.

It is unlikely to find a much higher class version, unless they get AD or someone else to design / make new ref. chips and get them tested over a few years. A pressure tight sealed version might be a crude idea but would be more like expensive / heavy.

[Edwin G. Pettis]

Updated news about 732 in the works.....coming to a theater not so soon.

[MisterDiodes]

To put the end to the rumor, I just called Fluke to get the story - and since I've got a 732a going thru cal cycle at their Everett facility right now I thought it was an appropriate time to get the facts - at least what little the product manager could tell me for sure:

732C (working name right now) is in development now, probably released for 2018 /19.  Redesign is focused on more efficient oven for longest possible battery life, they are looking at different battery pack chemistries that are still shippable without a hazmat certificate.  Main change is on the outputs:  There will be the usual 10V primary output (basically about same specs as 732a/b), and then a better quality 1V and 100mV secondary outputs, (possible built-in Hamon dividers 10:1 and 100:1, but exact specs unknown).  The old 1.018V output is replaced by the new 100mV output.  The specs on 1V and 100mV are unknown since the product is still in development and pricing is yet unknown, but these are planned to have better stability specs than the current secondary outputs.  Aside from the voltage outputs the box will use a "similar" architecture to 732b for Vref, as expected. 

So not a lot of exact specs at this time - they are still testing the new design - but the 732"c" is -not- a rumor and is in the certainly in pipeline for upcoming release, probably under the Fluke brand.  Or whatever company they've spun off to by then.

[CalMachine]

probably under the Fluke brand.  Or whatever company they've spun off to by then.

What else do you know?

 

[MisterDiodes]

Let's just say about the only thing consistent at Fluke is the brand logo and "change".

That being said it's nice they still cal 732a's, even if they don't repair them. 10V is 10V and a well aged, working 732a perfectly holds its own against a 732b with no issues and typically lower noise - but we're lucky enough to be able to drive our stuff in for cal.  Fluke will rent you a 732a shipping crate complete with 4ea battery packs installed for around $375, and then you pay freight shipping on 128Lbs total to send your unit in for cal.  That's the nice thing about 732b's if you're not within driving distance.

I was talking with the cal room manager at Fluke, and they still see a lot of in-service 732a's that come in for a cal cycle, probably still as many of those as 732b's, but eventually attrition will catch up at some point.  I know several of the cal facilities in Portland, Oregon typically use 732a's, but again they are only a a few hour's drive from Everett.

We use a 752a with the 732's for 1V and 100mV anyway (we never really use the secondary outputs on the 732's), so no advantage for us in "upgrading" to a 732C, but other labs would see an advantage in a lighter weight Vref for easier shipping.

[ManateeMafia]

Fluke's offerings have seemed to coincide with other products like the 5440A/B, 57x0A, and 3458A that have a need for that product. Unless they plan on releasing some other test gear that needs a different reference level, I wouldn't think there would be an advantage to a second or third output. If they wanted to improve battery life, then get rid of the 1.018 and be done with it.

[fcb]

Interesting (&refreshing) that Fluke have been so candid about the 732C. I don't imagine the Osborne Effect is that prevalent in metrology.

[VintageNut]

Fluke's offerings have seemed to coincide with other products like the 5440A/B, 57x0A, and 3458A that have a need for that product. Unless they plan on releasing some other test gear that needs a different reference level, I wouldn't think there would be an advantage to a second or third output. If they wanted to improve battery life, then get rid of the 1.018 and be done with it.

The primary lab that I visit has a bunch of 732B,A and Wavetek units. They keep 10V  and the ohm for a next-door cal lab that has a Fluke 57XX. I think that the 1V and 100mV outputs are not needed by everyone.

This primary lab has more than 10 years of data for their gaggle of 10v boxes. The value of these 10V boxes is the history and ability to predict the values between calibrations. A new sexier 732C would have zero value for an established cal lab. Only after a number of calibration cycles would the new 732C have utility.

[Awesome14]

Hi,

I looked at 732B teardown. I don't know if it is late production, I suspect that it isn't. The teardown showed a thru-hole board. This is a target for redesign because the thru-hole parts will be getting hard to get.

The battery chemistry may be another target. VRLA or SLA is still a good for this application, but Li-ion may be considered sexier.

The reference section may be a target for value engineering. It may be that 732B are much better than the specification.

Regards,

Jay_Diddy_B

One of my 732Bs drifts 0.1ppm/90 days. I think the spec is 0.3ppm/month.

[VintageNut]

Hi,

I looked at 732B teardown. I don't know if it is late production, I suspect that it isn't. The teardown showed a thru-hole board. This is a target for redesign because the thru-hole parts will be getting hard to get.

The battery chemistry may be another target. VRLA or SLA is still a good for this application, but Li-ion may be considered sexier.

The reference section may be a target for value engineering. It may be that 732B are much better than the specification.

Regards,

Jay_Diddy_B

One of my 732Bs drifts 0.1ppm/90 days. I think the spec is 0.3ppm/month.

I have seen the long-term data for a bank of 732Bs at a primary lab. 0.03 ppm / month is what a good 732B looks like at that lab.

[chuckb]

I have seen the long-term data for a bank of 732Bs at a primary lab. 0.03 ppm / month is what a good 732B looks like at that lab.
[/quote]

Do you know if the 0.03ppm / month drift was the actual drift rate or the uncertainty of the predicted drift rate?

[Awesome14]

NBS, now NIST, has a paper: https://www.gpo.gov/fdsys/pkg/GOVPUB-C13-2c2d0649085cc34834a8f42deaad328e/pdf/GOVPUB-C13-2c2d0649085cc34834a8f42deaad328e.pdf

At the very end in conclusions it mentions 10V can be transferred between labs with an accuracy of 0.08 ppm using zener standards. But it's from 1987. So, there may have been improvements. Fluke Calibration has a paper: http://download.flukecal.com/pub/literature/1260304D_734A_732B_Fract_Trace_AN_w.pdf

that makes claims superior to the NIST paper. But these are all based on characterization, not actual drift.

[VintageNut]

I have seen the long-term data for a bank of 732Bs at a primary lab. 0.03 ppm / month is what a good 732B looks like at that lab.

Do you know if the 0.03ppm / month drift was the actual drift rate or the uncertainty of the predicted drift rate?
[/quote]

Actual. They have their bank of qty 4 732Bs calibrated every year. They have all of the calibration history plotted as a long-term trend. The drift of one 732B in particular is about 0.36 ppm/year; very linear over a very long time.

[PE1RQF]

Hereby a picture of the new 732C.
https://ibb.co/jGaOfR

Specification not known yet, but will include retrace/ hysteresis specifications.

[Echo88]

Fluke will presumably present the Fluke 732C at the CPEM on 8-13.07.18:

http://www.cpem2018.com/newsletter/04/Newsletter_04_CPEM_2018.html

"We expect another excellent conference, where we hope to both share the research on our new products such as the Fluke 732C Zener reference standard, and to attend the presentations and poster sessions of other electrical metrologists from around the world."

[TiN]

Fluke released 732C/734C without any fuss.

The internal wire-wound-based resistors have been replaced by hermetic thin-film resistive networks, which are less prone to time- and temperatureinduced drift.

Not sure how I feel about this

They also give a glimpse on the guts. Good old Linear Technology LTFLU-1ACH on ceramic substrate with AD706 and OP07(?).
LTFLU chip manufactured in 2014 from date code, so I guess we can estimate when Fluke started testing for this project. 



797933 hermetic film resistor network likely used to generate additional 0.1V range. It is not actively driven, as we can see from max current spec for this channel (20pA).



DC stability spec for 732C is same as 732B, as expected, but now 1V output have specified stability for 90 days / 1 year.
New 0.1V output specified for 1.2ppm/30d or 9.8ppm/1y and TC +/-0.2 uV/V.  Hysteresis specified at 0.25ppm, same as 732B.

Now calibration procedure also recommends Keithley 2182A instead of old Fluke 845AB.

And as with modern trend  , no details on service information or schematics in manuals. I bet 734C box needed lot of development work to get that logo badge changed to new fancy yellow  .

Anybody getting DEMO unit? 

[quarks]

Fluke 732C with 1V and 0.1V and one year stability up to 9.8ppm, I wonder how this should be of any use for anyone with a 5720/5730 calibrator.

[dl1640]

I heard they will release new 8.5digit meter in the end of this year, with faster speed to beat 3458.

[HighVoltage]

Fluke released 732C/734C without any fuss.

May be now, we see a price drop of the 732A and 732B and they become affordable again.

How much is a 732C or 734C?
Anyone know?

[e61_phil]

Now calibration procedure also recommends Keithley 2182A instead of old Fluke 845AB.

Does anyone know why? I thought the main feature of a good null meter is very low leakage/good isolation. Is the 2182 special in this case or is it because it has very high resolution (100pV afaik) and Fluke and Keithley are now within the same company?

Should I use my Keithley 182 instead of my Fluke 845?

[dr.diesel]

May be now, we see a price drop of the 732A and 732B and they become affordable again.

How much is a 732C or 734C?
Anyone know?

I could see a few upgrading their 732As, but I doubt many will upgrade their Bs, eBay pricing will likely not change much, but here to hoping!!

However, if say Keithley released a K8510, that was a true competitor to the 3458A, then I could see tons of those showing up.

[TiN]

I see zero point for owners of _proven_ and _characterized aged_ 732 units , does not matter A or B-version to start selling them in favour of new refaced C.
Like quarks mentioned, benefit of having 0.1V likely overrated, any serious lab already have 752A for that  . It would make sense only for new customers and new labs to get C ref.
Metrology DC references market is not like fruitty phone, it takes years and years to get new product proven, even if it's a game changer, like 3458A or 57xx series were 4 decades ago.
I'm sure 3458's legacy will live, and slapping fancy TFT display wouldn't make up for ADC linearity

[dr.diesel]

Once new models are available, here are a few reasons some minor upgrades will happen:

 - 30 year old equipment is perceived as a production risk, especially when downtime is measured in the $10k + minute range  (this has personally happened to me as an engineer)
 - 732A's physical size, production rack space is valuable, shipping cal costs.  Logistical people complaining to no technical manages do win fights.
 - A half width, 2U, blank face ethernet capable 3458A replacement would capture some interest.
 - Capital expense equipment refresh/cycle planning.

A bunch of the above will directly impact 3458A's, as they are used in production and not just metrology, where their history isn't even tracked.

[splin]

Annual stability:
10V    2ppm
1V      3ppm
.1V   9.8ppm

If they have a 10:1 (or 7.2:1) divider which has < 1ppm additional drift above the 10V output why would the 1V -> .1V divider be so much worse?

[TiN]

Because 9.8ppm of 0.1V is 98 980 nanovolt. You sneeze on it and touch the connector, and those nanovolts will run like wild. Usually for this level special connectors are required, not just copper binding post, exposed to drafts.

[fonograph]

What in bloody hell is that! Please tell me its not really a Fluke product,I would be so disappointed in my favorite T&M manufacturer.Did Fluke ever sold product with that level of build "quality"?

[fonograph]

I heard they will release new 8.5digit meter in the end of this year, with faster speed to beat 3458.

Where did you hear that? It may be possible,Transmille just released their 8100 series that competes with 8508a.

[Echo88]

You are about a thousand years late to the party fonograph: The picture you posted is the soldering job of a voltage reference from a guy who is long banned here and who sells those DIY voltage references on ebay. Fluke isnt involved in this at all.

Faster 3458A, meh. Even better linearity, now that would be very impressive.

[splin]

Because 9.8ppm of 0.1V is 98 nanovolt. You sneeze on it and touch the connector, and those nanovolts will run like wild. Usually for this level special connectors are required, not just copper binding post, exposed to drafts.

980nV not 98nV. Thermal EMFs at the connectors could be several uV but they are outside of Fluke's control - they surely will be additional uncertainty that has to be accounted for by the user.

I've just noticed that the output impedance of the .1V output is specced at <= 100R. Someone said it is unbuffered which suggests a 111R/999R divider. So part of the difference may be due to it being more difficult to achieve low ratio tracking ratio drift with these relatively low resistance values.

The dividers absolute resistance drift will be somewhat greater than the tracking drift and its not inceivable that it could be as high as say 50ppm whilst keeping ratio drift at less than, say,  2ppm. A 50ppm change in divider current running through a 200m ohm connection/trace resistance difference between the top and bottom connections to the divider would create another 1ppm change in output.

It would be a bit of a stretch to increase those numbers much further though - a Vishay MORN thin film divider with 1ppm tracking TCR has a 100ppm absolute and 20ppm ratio 1 year shelf life stabilty specification and I would expect Fluke to do rather better unless they don't place much importance on the .1V output. So 9.8ppm stills seems rather a lot to to me. I can't see internal temperature changes could account for .5uV or so of thermal EMF *drift*.

[dl1640]

I heard they will release new 8.5digit meter in the end of this year, with faster speed to beat 3458.

Where did you hear that? It may be possible,Transmille just released their 8100 series that competes with 8508a.

From a local Fluke rep.
I have one 3458 at work, they came and promote me their 8508 to measure current upto 20A, but i really dont care about this feature, then they said a faster meter will come...