Ultra Precision Reference LTZ1000

[CaptnYellowShirt]

Oh-- and he also said that the LM399 should be run at about 1mA of Zener current for best stability.  The more stable you can make the Zener current, the more stable will be the output voltage.  He said that there is about 1uV of voltage change for 1uA of current change.

This is something that's been on my mind since you had your conversation with Bob. We talk a lot on this forum about the zener side of the circuit, but I haven't read a great deal about the current sources for these devices.

I picked up a copy of "Current Sources and Voltage References" (a pdf of which can be found though an easy google search). I'm only half way though it now, but its clear the authors want the reader to understand that (from a circuit design stand point) voltage and current standards are really two sides of the same coin -- its a coupled problem. In fact it might even be proper to say that a Fluke 732 is more a thermodynamic standard than an electrical standard (as the circuit in its totality is comparing concentrations, diffusions, drifts, etc).

I bring the topic up because I am planing a long term drift experiment where a number of zeners (40'ish) are going to be energized for several months -- their voltages being compared every few hours.  And the current source in my design is still an unsolved problem.

Does anyone have any thoughts on this? Do any of the LTZ1000 based references stand out in terms of their current sources?

[Dr. Frank]

Oh-- and he also said that the LM399 should be run at about 1mA of Zener current for best stability.  The more stable you can make the Zener current, the more stable will be the output voltage.  He said that there is about 1uV of voltage change for 1uA of current change.

This is something that's been on my mind since you had your conversation with Bob. We talk a lot on this forum about the zener side of the circuit, but I haven't read a great deal about the current sources for these devices.

I picked up a copy of "Current Sources and Voltage References" (a pdf of which can be found though an easy google search). I'm only half way though it now, but its clear the authors want the reader to understand that (from a circuit design stand point) voltage and current standards are really two sides of the same coin -- its a coupled problem. In fact it might even be proper to say that a Fluke 732 is more a thermodynamic standard than an electrical standard (as the circuit in its totality is comparing concentrations, diffusions, drifts, etc).

I bring the topic up because I am planing a long term drift experiment where a number of zeners (40'ish) are going to be energized for several months -- their voltages being compared every few hours.  And the current source in my design is still an unsolved problem.

Does anyone have any thoughts on this? Do any of the LTZ1000 based references stand out in terms of their current sources?

If you supply the zéner current for the LM399 from its own amplified output (e.g. 10V), by a stable resistor (i.e. 3k, ww., <5ppm/K), this already is a stable current source.

The LTZ1000 circuitry also contains two such intrinsic current sources, one for the zener diode, set up by the stable UBE voltage and the 120 Ohm resistor, and the other one for the collector current, built by the zener voltage and the 70k resistor.
The stability of those sources is already included in the stability calculation in the datasheet, stating 1/100 .. 1/500 of influence from the stability of the resistors.
Similar calculation can easily be done for the above described current source for the LM399H.

So everything is fine, it is not required to use external current sources.

Simply build your circuit that way, there are other, more prominent drift mechanisms than those currents.

If you try to monitor your references, question arises, against which other superior or equally stable source you want to measure?

Frank

[CaptnYellowShirt]

Simply build your circuit that way, there are other, more prominent drift mechanisms than those currents.

If you try to monitor your references, questions comes up, against which other superior or equally stable source you want to measure?

In my case I'm using the LM329 which is just a buried zener. Effects of temperature is what I'm after (77K, 200k, 273k, and 333k). So I'm going to compare them to themselves -- relative drift wrt temp.

My initial plan was to wire them in series using a single current source (~1mA). But that's a fairly non-standard source (280V @ 1mA).

[TiN]

Time to join club, first LTZ online.

More details and bunch of photos in my thread.

[bingo600]

I'd like to join the LTZ club , but would need some finished PCB's (2..3)
Does any posters have some surplus boards ie. Branadic or ... ?

TIA
Bingo

[babysitter]

I might have a leftover possibly, if I didnt give everything to Quarks (It was stuffing a half-used pcb panel and had some overproduction)

EDIT: No, sorry, no leftovers.

[branadic]

I'm not in the LTZ but in the LMx99 club, that's fine for me up to know.

[CaptnYellowShirt]

I want to see the LTZ design that uses hot glue, bailing wire, and tar paper. 

[bingo600]

I might have a leftover possibly, if I didnt give everything to Quarks (It was stuffing a half-used pcb panel and had some overproduction)

EDIT: No, sorry, no leftovers.

@babysitter
Damm 

Any other ?

/Bingo

[CaptnYellowShirt]

Anyone see anything fishy here:

http://r.ebay.com/FY1E33

http://r.ebay.com/mZh1JL


I feel bad for the guy that bought the second unit.

[codeboy2k]

Anyone see anything fishy here:

http://r.ebay.com/FY1E33

http://r.ebay.com/mZh1JL


I feel bad for the guy that bought the second unit.

Batteries dead.  No In-Cal light ...
an expensive recalibration in the future

[CaptnYellowShirt]

Anyone see anything fishy here:

http://r.ebay.com/FY1E33

http://r.ebay.com/mZh1JL


I feel bad for the guy that bought the second unit.

Batteries dead.  No In-Cal light ...
an expensive recalibration in the future

I'm talking about the fact its the exact same auction: copy and pasted. One seller is a surpluser, the other is a brand new ebay member with zero feedback. 

Note to self: make thousands running fake ebay sales of precision references.

[codeboy2k]

I'm talking about the fact its the exact same auction: copy and pasted. One seller is a surpluser, the other is a brand new ebay member with zero feedback. 

Note to self: make thousands running fake ebay sales of precision references.

yes, that does seem rather fishy..  Although I didn't notice it at first as you did.

I think those final prices are a little high for an out of cal unit, though. More like I want to pay just $300.

[CaptnYellowShirt]

  The calibrations of this thing are very overpriced by Fluke-- so much so that I can't afford it, and am looking for a hobbyist friendly way to get it done.

Besides Fluke uses the "Canadian Volt"... which I think is 2/3rds English and 1/3rd French?

[ManateeMafia]

Anyone see anything fishy here:

http://r.ebay.com/FY1E33

http://r.ebay.com/mZh1JL



I feel bad for the guy that bought the second unit.

Batteries dead.  No In-Cal light ...
an expensive recalibration in the future

I'm talking about the fact its the exact same auction: copy and pasted. One seller is a surpluser, the other is a brand new ebay member with zero feedback. 

Note to self: make thousands running fake ebay sales of precision references.

It's possible that eBay [or PayPal] security will stop this transaction before any money changes hands-- since they are liable to restore the buyer's losses [well, most of them anyway].

If something seems too good to be true, it probably is.  I bought my 732B from a dealer.  The calibrations of this thing are very overpriced by Fluke-- so much so that I can't afford it, and am looking for a hobbyist friendly way to get it done.

Good luck to the buyer. I noticed it the day it showed up and reported it to eBay as a potential fraud.
eBay does not care and they will not allow you to state why it is a potential fraud. The fact they do not reply and ask for further information proves they only care about making their quick buck.

[CaptnYellowShirt]

Good luck to the buyer. I noticed it the day it showed up and reported it to eBay as a potential fraud.
eBay does not care and they will not allow you to state why it is a potential fraud. The fact they do not reply and ask for further information proves they only care about making their quick buck.

I did the same thing. As you point out... a lot of good it did us.

[Rigby]

Caveat emptor.

Also if you're paying with PayPal you have some protection there, if you're quick enough.

[CaptnYellowShirt]

http://www.sciencedirect.com/science/article/pii/S0026271401002517

In this paper, an accelerated life test and noise measurements during the test for subsurface Zener diodes are carried out. The correlation between the reference voltage degradation and the 1/f noise in the devices is analyzed on the test data, and the physical mechanism leading to the correlation is discussed. In addition, a reliability screening approach using 1/f noise measurement is proposed.

PM if you're interested.

[JBeale]

Right now there are 4 sellers of LTZ1000 on eBay. Two of them, Polida and yankee_electronic claim new and use the same generic pic from web. Polida is just bulk part seller, I've seen few comments that people got fake parts but can't say for sure, I'd say both of them are questionable as they don't even have the real photo. Other two, hifi-szjxic and bbshonic, sell used and they claim 100% tested. bbshonic even "Gurantee exchange if it is fault" and have some history of selling them.

FWIW, I purchased four used 0.01% 10k resistors from hifi-szjxic in 2012 and as far as I can tell, they are as claimed. The values are consistent but I don't have an absolute reference standard to compare with. I assembled the four into a full bridge and with 10 V excitation, measured an imbalance of 0.1 mV so the ratios of those two pairs are mismatched by 0.001% or 10 ppm.

(edit: oops- confused about the date; didn't realize that post was more than a year old)

[CaptnYellowShirt]

Right now there are 4 sellers of LTZ1000 on eBay. Two of them, Polida and yankee_electronic claim new and use the same generic pic from web. Polida is just bulk part seller, I've seen few comments that people got fake parts but can't say for sure, I'd say both of them are questionable as they don't even have the real photo. Other two, hifi-szjxic and bbshonic, sell used and they claim 100% tested. bbshonic even "Gurantee exchange if it is fault" and have some history of selling them.

FWIW, I purchased four used 0.01% 10k resistors from hifi-szjxic in 2012 and as far as I can tell, they are as claimed. The values are consistent but I don't have an absolute reference standard to compare with.

(edit: oops- confused about the date; didn't realize that post was more than a year old)

Got a link?

[cyr]

http://stores.ebay.com/HIFI-AUDIO-IC

I bought some used 0.01% vishay resistors from the same seller a year or so ago as an experiment. I mounted them om dual banana plugs and measured them on the 8508A at work:

1k - 1.000101k
10k - 9.99986k
100k - 100.0074k

[Blackart]

Thermocouple accuracy
After reading that Datron ageing paper ( thanks CaptYellowShirt ) Ive embarked on a oven to thermally work my refs & resistors for use ( and final ref boards ). To that end I've decided to build a automatic oven to do the c110 C ageing. The first step was to get some reliable temperature readings. I had 5 off new type K thermocouple made with "nude" i.e. welded and unprotected ends. Then hooked them up to a HP 34970A mix data acquisition unit. Then mounted the thermocouples in contact with a large block of copper. The first surprise was just how temperature sensitive they were even walking around workshop would show up as temperature disturbances. So I wrapped up all the sensors in a sealed plastic bag and further insulated them. The result is a on the graphs below called RAW data. Yo can see 100 readings at 30s intervals. Note almost 0.3C variance of the thermocouple all off the same real of with made in the same batch. I then applied calibration factors and re ran the test. As you can see in the next graph they are all pretty much in aggreance. I might further tweak TC#5 down abit. NB I have not calibrated these absolutely just to each other. I wanted to be sure they all drift at a similar rate.

Note the rise at 550s is me closing the workshop door by about 200mm ! Allowing the room to heat a bit.

I think I/we can be fairly confident that the thermocouples are good enough to use in the oven and for further a characterising of the Refs after I build them. I what to have consistency in the equipment through this process.

Lucas

[CaptnYellowShirt]

Does anyone have experience with this kind of 'oil': http://en.wikipedia.org/wiki/Krytox

In some of my reading I've come across the use of this stuff in precision references as a thermal transfer fluid. Its Fluorocarbon nature made me wonder if it was oxygen or water vapor resistant?

Hey, good find!  If I had to hazard a guess, I would have to say that it is at least water-vapor resistant.  The fluorocarbon family of compounds is the only thing I have found that is 100% water-vapor proof, but I don't know if that extends to Krytox.  It's always best to go right to the source-- ask your Dupont FAE directly, [and post the answer here please!]

They didn't have data on water vapor, but they did have it for O2 and N2 solubility at 20degC....

O2: 218 ppm
N2: 123 ppm

The rep said hydrocarbon oils are typically in the range of 1000-2000ppm for O2.

No permittivity data was available.

[Dr. Frank]

Thermocouple accuracy
After reading that Datron ageing paper ( thanks CaptYellowShirt ) Ive embarked on a oven to thermally work my refs & resistors for use ( and final ref boards ).

...

Lucas

Thanks to Brandon, I also read this interesting paper.

Anyhow, I still disagree to try to pre-age the LTZ1000, or hermetically sealed, oil filled metal foil resistors, if you use them for the reference.

The paper describes the fast ageing of 1N829 zener diodes, which are used in the Datron 4000, but definitely NOT the LTZ1000, which sits in the Datron 4910!

The big difference is, that the LTZ1000 already has such a small typical drift (1ppm/yr.) from the beginning, that further ageing makes no sense.

The 1N829 have typically 20 ppm/yr, and may be stabilized by that method to  a drift level where the LTZ begins.

The LTZ on the other hand shows strong hysteresis of several ppm, if its chip is heated to high temperatures as 110°C, which may lead to a strong creeping drift behaviour in the following years.

Therefore, a heat shock leads to a less stable reference.

Same goes for metal foil resistors, the oil filled types have such low annual drifts, (and also hysteresis) that any heating will have negative influence on stability.

Frank

[babysitter]

Excuse me for being not a expert isn't comparing those physically quite different devices a bit unfair, subsurface Zener Diode in the LTZ vs. 1N829 which I consider a stacked-cylinder-Design with surface effects ?