Author Topic: Ultra Precision Reference LTZ1000  (Read 632610 times)

cellularmitosis, montemcguire and 5 Guests are viewing this topic.

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 741
  • Country: de
Ultra Precision Reference LTZ1000
« on: January 13, 2013, 03:42:41 AM »
While waiting for the ESI gear I ordered for my resistance measuring stuff, I started to look for the best possible voltage reference.
I already have a few good DC Sources (Valhalla 2701C, Burster 6406, Knick SJ300 and a few Geller SVRs), but I do not have a "real" Voltage Reference Standard (like Fluke 732B or a Datron or any similar) to compare them to.

Besides "Josephson-Voltage-Standard" the next best still seems to be LTZ1000 and this is the reference I would like to give a try.

I read a lot of datasheets, Application Notes, Nuts&Volts posts, Bop Pease stuff, Joe Gellers info's (incl. his Patent), discussions in different blogs and many more sources.

Here are a few links I found good and worth reading (and translating):

http://www.linear.com/product/LTZ1000
http://www.maxmcarter.com/vref/
http://www.febo.com/pipermail/time-nuts/2008-November/034723.html
http://www.hellocq.net/forum/read.php?tid=139719
http://www.amobbs.com/thread-3593996-1-1.html
http://www.eefocus.com/lymexbg2vo/blog/09-12/181353_9b971.html

The last one shows a lot of boards, so I think of trying to get one or more, if that is possible.

Now I would be interested if anyone here in the forum has already done a LTZ1000 DIY project and is willing to share the outcome?
Or maybe one of you know where to get a well designed board or plans to do one and can offer to buy it?

thx
quarks
« Last Edit: January 13, 2013, 05:13:11 AM by quarks »
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 24752
  • Country: au
    • EEVblog
Re: Ultra Precision Reference LTZ1000
« Reply #1 on: January 13, 2013, 06:29:22 PM »
The LTZ1000 is the 7V reference used in industry standard HP3458A 8.5 digit multimeter.
Part number 03458-66509
No oven, it just sits inside the meter case on it's own board with it's circuitry.
If it's good enough for the worlds best reference multimeter used in practically every cal lab, then it's good enough for any DIY project!

Dave.
 

Offline (In)Sanity

  • Regular Contributor
  • *
  • Posts: 229
  • Country: us
Re: Ultra Precision Reference LTZ1000
« Reply #2 on: January 13, 2013, 07:01:34 PM »
The LTZ1000 is the 7V reference used in industry standard HP3458A 8.5 digit multimeter.
Part number 03458-66509
No oven, it just sits inside the meter case on it's own board with it's circuitry.
If it's good enough for the worlds best reference multimeter used in practically every cal lab, then it's good enough for any DIY project!

Dave.

This is a very interesting piece of information.  I had no idea the LTZ1000 reference was used in the 3458A.   I have a very nice condition 3457A which I believe uses the LM399 as a reference.   I have to wonder if their is any advantage to upgrading the 3457A to use an LTZ1000 or is it's 6 1/2 (7 1/2) digit resolution would see no real advantage?  It would of course need to be re-calibrated which I'm due for anyway.  I'm always looking to improve something just for the sake of doing so.   

On another note an interesting video might be one on temperature coefficients and techniques used to stabilize and/or compensate for temperature drift.   

Jeff
 

Offline PA4TIM

  • Frequent Contributor
  • **
  • Posts: 931
  • Country: nl
  • instruments are like rabbits, they multiply fast
    • PA4TIMs shelter for orphan measurement stuff
Re: Ultra Precision Reference LTZ1000
« Reply #3 on: January 13, 2013, 07:15:04 PM »
Dave, it depends on how you look at it, like the LM399 the LTZ1000 has a heater resistor so it is inside his own little oven.

A member of the voltnuts list has just modded his solartron 7081 and replaced the zeners with a LTZ1000 board. He also did measurements on noise and stability. He is really good on this stuff. 

Voltnuts is a "sister" of timenuts. Several members made LTZ1000 references. Jim Williams has an appnote about a sub-ppm standard and he used a LTZ1000 too. I made his design but then using a LM399 because I have no source for a LTZ1000.

I have a Fluke 332, a guildline 4 Standardcell cabinet ( will all calibration reports since production) and several other Fluke, Philips and homebuild calibrators/references. (a gift from a company that closed down theirs calibration lab) Besides that ESI resistors, two Flukes and a ESI KV devider, standard capacitors and inductors ect. And the famous, still made GR-1620 capacitance bridge that was first used by the GenRad lab self, then when it closed moved together with a staff member to a national brittisch Cal-lab and later sold to a dutch call lab and finaly ended in my lab.

Network analyses, capacitors anf voltnutting is my thing. Nice to meet an other voltnut. Blackdog, here on the forum also has a lot of knowledge about voltreferences. If i remember well also with the LTZ1000 but i think we will see him here. LTZ1000 works like a magnet too voltnuts .

I'm very interested to see your design and the result. I read the hardest part is something with setting the heater current. But some say that is not, others say it is. As far as I found out it only was a problem on some old 3458 reference boards

The things that make it hard are the parts around it. My 10V LM399 bases reference is now powered on 24/7 for months. I monitored it for months and it becomes more stable every month but still it varies 2-4 ppm and that is a reaction on temperature and humidity. So one of my next projects Will be an oven for the reference and a relative humidity and temp meter that will serve to give me those values and work as a controller for the oven. ( as soon as i find out if it is nessaserry and possible to keep humidity in the oven constant) for temp I will use one of Williams early designs when he was stil working for MIT.
www.pa4tim.nl my collection measurement gear and experiments Also lots of info about network analyse
www.schneiderelectronicsrepair.nl  repair of test and calibration equipment
 

Offline (In)Sanity

  • Regular Contributor
  • *
  • Posts: 229
  • Country: us
Re: Ultra Precision Reference LTZ1000
« Reply #4 on: January 13, 2013, 07:23:46 PM »
The things that make it hard are the parts around it. My 10V LM399 bases reference is now powered on 24/7 for months. I monitored it for months and it becomes more stable every month but still it varies 2-4 ppm and that is a reaction on temperature and humidity. So one of my next projects Will be an oven for the reference and a relative humidity and temp meter that will serve to give me those values and work as a controller for the oven. ( as soon as i find out if it is nessaserry and possible to keep humidity in the oven constant) for temp I will use one of Williams early designs when he was stil working for MIT.

So does this mean that my 1993 3457A should be far more stable then a fresh off the shelf LM399,  or do they degrade over time?   I fear becoming a volt nut,  I already fuss about stuff being out of tolerance by 10 uV,  let alone a fraction of that. 
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 7546
  • Country: my
  • reassessing directives...
if something can select, how cant it be intelligent? if something is intelligent, how cant it exist?
 

Offline PA4TIM

  • Frequent Contributor
  • **
  • Posts: 931
  • Country: nl
  • instruments are like rabbits, they multiply fast
    • PA4TIMs shelter for orphan measurement stuff
Re: Ultra Precision Reference LTZ1000
« Reply #6 on: January 13, 2013, 08:16:02 PM »
40 dollar is rather cheap, i would not trust that. They also have grades. I bought 1027 from Farnell, they only had C grade, then I asked a sample from LT, the sended me one, B grade. Allready a few ppm better but they have A grade too. But no clue how to get one if you are not named Agilent or so.

Every reference needs to be "burned" in a good brand meter uses burned in references, it takes about 1000 hours. The yearly calibration will show you the behaviour. Then there will be a moment the changes between calibration become very small and you can extend the time between them. So for hobby use a nice aged calibrated meter with call history is most times good for many years of fun without the need for calibration. A friend bought a 3457, had it calibrated ( in the states that is very affortable, here is is allmost the pricec of a new meter) and it was still spot on. My Prema came with a cal report ( just cal, no adjustments made) . All measurement were spot on. 10V was 10.000,000V so i know it is allready stable and little chance in changing much in time. My new Keitley 2000 now changed around 10 uV in two years. But most of that happend the first months.
www.pa4tim.nl my collection measurement gear and experiments Also lots of info about network analyse
www.schneiderelectronicsrepair.nl  repair of test and calibration equipment
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 14373
  • Country: za
Re: Ultra Precision Reference LTZ1000
« Reply #7 on: January 13, 2013, 08:29:11 PM »
Look for used ones, and buy a dozen or so. Then you can sort them into 3 groups: dead, working but drift, and good.
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 741
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #8 on: January 13, 2013, 09:06:06 PM »
The LTZ1000 is the 7V reference used in industry standard HP3458A 8.5 digit multimeter.
Part number 03458-66509

Hi Dave,
thanks for your reply. I knew that and I can add that Fluke also had a special version of it, called HP 3458A/HFL, where I believe they only changed one resistor on the LTZ1000 board to lower the heater temp but resulted in even better specs than the original HP meter.

About the part number, do you know or think that it is possible to order this as a spare part when you do not have the 3458 DMM?

thx
quarks
« Last Edit: January 13, 2013, 09:52:26 PM by quarks »
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 741
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #9 on: January 13, 2013, 09:37:32 PM »
Dave, it depends on how you look at it, like the LM399 the LTZ1000 has a heater resistor so it is inside his own little oven.

A member of the voltnuts list has just modded his solartron 7081 and replaced the zeners with a LTZ1000 board. He also did measurements on noise and stability. He is really good on this stuff. 

Voltnuts is a "sister" of timenuts. Several members made LTZ1000 references. Jim Williams has an appnote about a sub-ppm standard and he used a LTZ1000 too. I made his design but then using a LM399 because I have no source for a LTZ1000.

I have a Fluke 332, a guildline 4 Standardcell cabinet ( will all calibration reports since production) and several other Fluke, Philips and homebuild calibrators/references. (a gift from a company that closed down theirs calibration lab) Besides that ESI resistors, two Flukes and a ESI KV devider, standard capacitors and inductors ect. And the famous, still made GR-1620 capacitance bridge that was first used by the GenRad lab self, then when it closed moved together with a staff member to a national brittisch Cal-lab and later sold to a dutch call lab and finaly ended in my lab.

Network analyses, capacitors anf voltnutting is my thing. Nice to meet an other voltnut. Blackdog, here on the forum also has a lot of knowledge about voltreferences. If i remember well also with the LTZ1000 but i think we will see him here. LTZ1000 works like a magnet too voltnuts .

I'm very interested to see your design and the result. I read the hardest part is something with setting the heater current. But some say that is not, others say it is. As far as I found out it only was a problem on some old 3458 reference boards

The things that make it hard are the parts around it. My 10V LM399 bases reference is now powered on 24/7 for months. I monitored it for months and it becomes more stable every month but still it varies 2-4 ppm and that is a reaction on temperature and humidity. So one of my next projects Will be an oven for the reference and a relative humidity and temp meter that will serve to give me those values and work as a controller for the oven. ( as soon as i find out if it is nessaserry and possible to keep humidity in the oven constant) for temp I will use one of Williams early designs when he was stil working for MIT.

Do you remember the name of the voltsnuts guy or can give a hint how to identify him? I also read there from people planing transplating LTZ1000 into Valhalla, so that could be the next interesting project.

About the Jim Williams article, was it AN86 or the designfeature at edn you mean?  I think I have read most (if not all) of the relevant info's. It is so much theory that I feel I now have go practical to learn and understand it better.   

About your gear, you are a lucky man and now I can proof to my wife that I am not the craziest nerd on the planet  :-+ That helps a lot to get even more goodies.

thx
quarks

« Last Edit: January 14, 2013, 04:07:56 AM by quarks »
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 741
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #10 on: January 13, 2013, 09:48:11 PM »
$40 a pop hmmm...
http://www.ebay.com.my/itm/IC-LINEAR-TO-99-LTZ1000ACH-/250886390149?pt=LH_DefaultDomain_0&hash=item3a69fe8185

thx for the link. Have you tried this source? My feeling is unless one can confirm it is no fake I tend to stay away.

The min. order of 250$ and the price of 54.50$ (QTY 1-99) at Linear Technology is ok. But before that, I try to find a local (German/Europe) source to avoid shipping + tax + custom (which will easily double the sellers price). If anyone knows where to order, please share.
« Last Edit: January 13, 2013, 09:54:37 PM by quarks »
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 741
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #11 on: January 13, 2013, 09:51:35 PM »
Look for used ones, and buy a dozen or so. Then you can sort them into 3 groups: dead, working but drift, and good.

do you know of a trustworthy source to buy from?

thx
quarks
 

Offline muvideo

  • Frequent Contributor
  • **
  • Posts: 390
  • Country: it
Re: Ultra Precision Reference LTZ1000
« Reply #12 on: January 14, 2013, 12:40:22 AM »
A member of the voltnuts list has just modded his solartron 7081 and replaced the zeners with a LTZ1000 board. He also did measurements on noise and stability. He is really good on this stuff. 

Do you remember the name of the voltsnuts guy or can give a hint how to identify him? I also read there from people planing transplating LTZ1000 into Valhalla, so that could be the next interesting project.

I think Fred is referring to Mickle T., see:
http://www.febo.com/pipermail/volt-nuts/2013-January/002322.html
He posted also other articles before,
for Mickle the 7081 is a buggy imperfect machine,
so he decided to make it better  :o
Hats off to him!

For the little I undertand, the reference itself
is only a part of the things needed to have e
stable voltage reference, that is an assembly
that has to be characterized as a whole.
Just see all the crazyness that goes to the
selection of the right resistors,
Fred can give many info about these things.

My Prema came with a cal report ( just cal, no adjustments made) . All measurement were spot on. 10V was 10.000,000V so i know it is allready stable and little chance in changing much in time.

I'm so curious about these meters, what model
do you have? Do you have some docs like spec
sheer and or manuals? How does it perform?
If you can post some pics ( a teardown maybe?)
I think it will be the first I've seen on the net.

Thank you
Fabio.
Fabio Eboli.
 

Offline jnd

  • Contributor
  • Posts: 49
  • Country: cz
    • Personal stuff
Re: Ultra Precision Reference LTZ1000
« Reply #13 on: January 14, 2013, 12:51:24 AM »
40 dollar is rather cheap, i would not trust that. They also have grades. I bought 1027 from Farnell, they only had C grade, then I asked a sample from LT, the sended me one, B grade. Allready a few ppm better but they have A grade too. But no clue how to get one if you are not named Agilent or so.

Every reference needs to be "burned" in a good brand meter uses burned in references, it takes about 1000 hours. The yearly calibration will show you the behaviour. Then there will be a moment the changes between calibration become very small and you can extend the time between them. So for hobby use a nice aged calibrated meter with call history is most times good for many years of fun without the need for calibration. A friend bought a 3457, had it calibrated ( in the states that is very affortable, here is is allmost the pricec of a new meter) and it was still spot on. My Prema came with a cal report ( just cal, no adjustments made) . All measurement were spot on. 10V was 10.000,000V so i know it is allready stable and little chance in changing much in time. My new Keitley 2000 now changed around 10 uV in two years. But most of that happend the first months.
I wouldn't say $40 is cheap enough to be not trustworthy, it's comparable to new prices from Linear and Chinese always get them for better price. I just checked Taobao and there are lot of them for around $20.

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. However I didn't find any feedback for any of those sales.

For myself I'd rather buy used and burned in, working parts since I don't have any instrument which could be used to track any changes in the reference. I guess someone will have to try and report back what they got, so who's first? >:D

For the board, I haven't seen any for sale, only the ridiculously priced meter boards like Dave showed. Ideally I want some kit or finished board like what's been on the Chinese forums because getting all the precision parts around is single quantities must be PITA. So if anyone wants to make small run, I'm interested for one or two.
Wannabe volt-nut, slowly hunting solid meters with low budget.
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 741
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #14 on: January 14, 2013, 01:41:17 AM »
I think Fred is referring to Mickle T., see:
http://www.febo.com/pipermail/volt-nuts/2013-January/002322.html
He posted also other articles before,
for Mickle the 7081 is a buggy imperfect machine,
so he decided to make it better  :o
Hats off to him!
thx for the link.

About Prema, it is a German company located in Mainz (nearby where I live) but unfortunately no longer produce any DMMs.

I don't know the model you are talking about, but here are some hopefully interesting (but German) links:
http://www.amplifier.cd/Test_Equipment/other/Prema-DMM.html
http://www.amplifier.cd/Test_Equipment/other/Prema-5000.html
http://www.amplifier.cd/Test_Equipment/other/Prema-5017.html
http://ohh.de/homepage.htm
http://ohh.de/5017.htm
http://ohh.de/6048.htm
http://ohh.de/8017.htm

I just checked my archive I only have found Prema 5000, 5017 and 6040 documents (all in German).
« Last Edit: January 14, 2013, 02:22:47 AM by quarks »
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 741
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #15 on: January 14, 2013, 02:04:18 AM »
For myself I'd rather buy used and burned in, working parts since I don't have any instrument which could be used to track any changes in the reference. I guess someone will have to try and report back what they got, so who's first? >:D

For the board, I haven't seen any for sale, only the ridiculously priced meter boards like Dave showed. Ideally I want some kit or finished board like what's been on the Chinese forums because getting all the precision parts around is single quantities must be PITA. So if anyone wants to make small run, I'm interested for one or two.

I still hope someone here in the forum has already gone through this and is willing to share. For myself I would buy all the needed components (also more than I need for myself, to later on share) but I will probably not even try to do my own board, because my last layout is maybe from 30 years ago and was not as professional as I want this one to be.
« Last Edit: January 14, 2013, 02:06:10 AM by quarks »
 

Offline TorqueRanger

  • Frequent Contributor
  • **
  • Posts: 323
  • Country: us
Re: Ultra Precision Reference LTZ1000
« Reply #16 on: January 14, 2013, 02:19:33 AM »
For myself I'd rather buy used and burned in, working parts since I don't have any instrument which could be used to track any changes in the reference. I guess someone will have to try and report back what they got, so who's first? >:D

For the board, I haven't seen any for sale, only the ridiculously priced meter boards like Dave showed. Ideally I want some kit or finished board like what's been on the Chinese forums because getting all the precision parts around is single quantities must be PITA. So if anyone wants to make small run, I'm interested for one or two.

I still hope someone here in the forum has already gone through this and is willing to share. For myself I would buy all the needed components (also more than I need for myself, to later on share) but I will probably not even try to do my own board, because my last layout is maybe from 30 years ago and was not as professional as I want this one to be.

Do me a favor and just google LTZ1000 and look at all the pics...
 

alm

  • Guest
Re: Ultra Precision Reference LTZ1000
« Reply #17 on: January 14, 2013, 02:25:45 AM »
This is a very interesting piece of information.  I had no idea the LTZ1000 reference was used in the 3458A.   I have a very nice condition 3457A which I believe uses the LM399 as a reference.   I have to wonder if their is any advantage to upgrading the 3457A to use an LTZ1000 or is it's 6 1/2 (7 1/2) digit resolution would see no real advantage?  It would of course need to be re-calibrated which I'm due for anyway.  I'm always looking to improve something just for the sake of doing so.   
The LTZ1000 is also used in various voltage references like the Fluke 732B and some Datron references. Note that HP ran the LTZ1000 at a very high temperature in the HP 3458A, which made the drift much higher than in voltage reference applications. Much lower drift could be achieved by lowering this temperature. Downside is smaller operating temperature range.

The accuracy of a a DMM is not just determined by the internal reference. It also depends on other factors like ADC linearity and the input resistor network drift. Note that the HP 3456a has better 90 day and 1y accuracy specs than the HP 3457a, even though they share the same selected LMx99 reference.

So does this mean that my 1993 3457A should be far more stable then a fresh off the shelf LM399,  or do they degrade over time?   I fear becoming a volt nut,  I already fuss about stuff being out of tolerance by 10 uV,  let alone a fraction of that. 
As some people already mentioned, the majority of the drift is in the first year or so. The reference should be much more stable than when it was produced. Note that HP would also have done a far amount of burn in and selection.

For the board, I haven't seen any for sale, only the ridiculously priced meter boards like Dave showed. Ideally I want some kit or finished board like what's been on the Chinese forums because getting all the precision parts around is single quantities must be PITA. So if anyone wants to make small run, I'm interested for one or two.
Someone used to be selling 3458a reference boards for around the $100 mark, probably from some surplus source. Haven't seen these for a while, though. PCB design (eg. guarding) is going to be critical, leakage currents can easily induce errors at the single ppm level that these circuits are operating.
 

Offline muvideo

  • Frequent Contributor
  • **
  • Posts: 390
  • Country: it
Re: Ultra Precision Reference LTZ1000
« Reply #18 on: January 14, 2013, 03:11:16 AM »
About Prema, it is a German company located in Mainz (nearby where I live) but unfortunately no longer produce any DMMs.

I don't know the model you are talking about, but here are some hopefully interesting (but German) links:
http://www.amplifier.cd/Test_Equipment/other/Prema-DMM.html
http://www.amplifier.cd/Test_Equipment/other/Prema-5000.html
http://www.amplifier.cd/Test_Equipment/other/Prema-5017.html
http://ohh.de/homepage.htm
http://ohh.de/5017.htm
http://ohh.de/6048.htm
http://ohh.de/8017.htm

I just checked my archive I only have found Prema 5000, 5017 and 6040 documents (all in German).

Thank you for the links, I dont read German, but the web pages
can be translated by online translator, I previously have stumbled
in amplifier.cd  page about prema multimeters, but I forgot about that.
I found also PA4TIM's excellent page about his meter:
http://www.pa4tim.nl/?p=3624

Fabio.
Fabio Eboli.
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4768
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: Ultra Precision Reference LTZ1000
« Reply #19 on: January 14, 2013, 03:16:40 AM »
There's a lot of discussion about the LTZ1000 in the archives.  One summarized some the work of this Chinese group, making, analyzing and dissecting the LTZ in an attempt to create a low cost improved reference, here one photo of an ambitious multiboard reference with an LTZ each.  The Chinese site also analyzed the 3458a reference board.



It spans 2006-2010, and the thread is 50+ pages.  Lots of photos and data sheets.

http://www.hellocq.net/forum/read.php?tid=139719&page=6
Best Wishes,

 Saturation
 

Offline PA4TIM

  • Frequent Contributor
  • **
  • Posts: 931
  • Country: nl
  • instruments are like rabbits, they multiply fast
    • PA4TIMs shelter for orphan measurement stuff
Re: Ultra Precision Reference LTZ1000
« Reply #20 on: January 14, 2013, 03:57:40 AM »
http://www.pa4tim.nl/?p=3624. Prema 5017 7.5 digit benchmeter. Very solid build. The meter does not switch of totally uing the front powerbutton, the reference stays powered.
Teardowns not on this kind a stuff, only if i have to open it because of repairs. I do not like more dirt and dust inside just for pictures ;-)

I see the link to the modified 7081 allready is given by fellow Voltnut Fabio.

I have a Solartron 7061. I made some modifications too. I have not replaced the reference diode, if I remember well Fabio has done some experimenting in biasing it for better tempco. I just did a check up according the manual. And then I mounted a heatsinks on the outside with a fan on it. I made a temperature controller to drive the fan. The sensor is mounted inside de meter and the fan spins slow around 18 degrees roomtemp, speeding up if roomtemperature changes. It keeps the inside of the meter ( at the measure point) at 39 degrees.

The biggest problem with voltnutting is environment. My lab changes temp from about 17 to 33 degrees over a year and relative humidity from 20 to 40 %
If you want to keep everything tracking you need climate control. But that is not easy and i think costs a lot. Those of you in a all year long hot country probably have airco so maybe it then is possible.

I did not know the appnote number from the Williams reference , it is in my book Analog circuit design part 1, but i looked it up for you, according the book It should be number 86 a standards lab grade 20 bit DAC with 0.1 ppm/C drift. Look at the picture of the setup, a bunch of 3458, two KV dividers and his reference, that is a very expensive table.

My LM399 is made this way. I also used two KV's and the bufferamp. But I made a 10 turn potentiometer on the front that gives me about 800 uV. This for easy adjusting abd keep my KVs free for use with the 332. ( one for deviding, the second for checking)  This pot is the weakest part. The tempco is probably horrible. And mist of the drigpft at this time probably comes from this. If i am going to make the oven for it, the portentiometer will be changed.  I will make is so that 10 turns give me about  100 uV in that case tempco will not do much because one full turn (1K) will give 10 uV so if the tempco is 1000 ppm the resistance of the 10K pot will change 10 Ohm in total. So maybe this gives a 0.1uV fault / degree ( if i do the math correct, i have dyscalculus and no calculator at hand right now)
On the other hand if I  ditch the pot I maybe do not need the oven. Resistors I use are 0.05 to 0.01 % from dead a Fluke 8500 benchmeter.  There is guarding, i avoided mechanical stress ( the lm399 anf LTC1052 are mounted on sockets but without the ( leaking and dielectric) plastic. So no thermal stress to the components. After this everything is cleaned with ipa.
Only problem i have to solve is EMC ( if i power down everything except the prema or solartron and the LM or 332 or whatever and use a KV i get a nice linear behavour. So 1uV is 1uV on the meter and 10V is 10V. But if I leave the lights on ( halogene on a rail) everything goed wrong. There is about 150 uV deviation over 10V  in that case. And the strange thing the readings go down. It dies not matter if I use the 332, LM or whatever. The meter straight on the 332 and using its own divider gives no problem so it must be the cables from source to KV and from KV to meter ( but shielded, twisted, coaxial, triax ect all give the same problem.

Welcom to the darkside of voltnutting ;-)

www.pa4tim.nl my collection measurement gear and experiments Also lots of info about network analyse
www.schneiderelectronicsrepair.nl  repair of test and calibration equipment
 

Offline (In)Sanity

  • Regular Contributor
  • *
  • Posts: 229
  • Country: us
Re: Ultra Precision Reference LTZ1000
« Reply #21 on: January 14, 2013, 04:17:03 AM »
Thanks Alm for the replies.

This is a game I would love to get in to.   I thought about just getting a high grade LM399,  aging it for a long time and taking a measurement with my 3457A.   The idea being that the vref alone without any other components in play should in theory drift less then the combined circuit in my 3457.  I then of course run in to the issue of which one really drifted or did both,  etc.   It seams like you need 3 or 4 references to really figure out what's going on.  Also as I mentioned before a video on how all of this works and techniques to help compensate for it might be interesting.   Items like adding semiconductors in the right place to compensate for the drift of the resistors..etc.  I have a home brew power supply right now that's fairly stable,  it's only stable because the cheap op-amp just so happens to drift in the opposite direction as the cheap resistors,  etc.   I guess you could also say I just got lucky. 

So is a low grade LTZ1000 better then a high grade LM399 ?   

Jeff
 

Offline muvideo

  • Frequent Contributor
  • **
  • Posts: 390
  • Country: it
Re: Ultra Precision Reference LTZ1000
« Reply #22 on: January 14, 2013, 04:43:05 AM »
I see the link to the modified 7081 allready is given by fellow Voltnut Fabio.

I have a Solartron 7061. I made some modifications too. I have not replaced the reference diode, if I remember well Fabio has done some experimenting in biasing it for better tempco. I just did a check up according the manual. And then I

Thank you for the onorific title, but I'm only a VoltNut
follower :) and unfortunately nothing to say about the
7061 stability, my lab is worse than yours and I'd like to
have at least another meter with similar or better specs
before trying anything on reference.

The biggest problem with voltnutting is environment. My lab changes temp from about 17 to 33 degrees over a year and relative humidity from 20 to 40 %
If you want to keep everything tracking you need climate control. But that is not easy and i think costs a lot. Those of you in a all year long hot country probably have airco so maybe it then is possible.

I throw in another element :)
I still dont have experience in building precision references,
it's a time (and money) consuming task, but one can learn
a lot. My question is if it's reasonable to build a temperature
and humidity controlled enclosure for the reference.
Once chosen the right components, and defined the layout
(quite a result, not easy ), the next thing to do should be
the charachterization, i.e. log the aging behaviour over time.
AFAIK the worse offenders remaining that screw up the
stability are temperature and humidity. Build an ovenized
enclosure can be done (and it's been done). I have never
seen a humidity controlled enclosure yet, but I've seen
around some solid state dehumidifiers, nothing cheap
unfortunately (google SPE dehumidifer).
Maybe that can compensate for poor lab enviroments :)

Fabio.
Fabio Eboli.
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 741
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #23 on: January 14, 2013, 04:45:26 AM »
There's a lot of discussion about the LTZ1000 in the archives.  One summarized some the work of this Chinese group, making, analyzing and dissecting the LTZ in an attempt to create a low cost improved reference, here one photo of an ambitious multiboard reference with an LTZ each.  The Chinese site also analyzed the 3458a reference board.



It spans 2006-2010, and the thread is 50+ pages.  Lots of photos and data sheets.

http://www.hellocq.net/forum/read.php?tid=139719&page=6

thx saturation,
your picture is missing but think I know which one you mean.
I have read all of this quite a while ago and I think you where the one who posted the link back then.
 

Offline (In)Sanity

  • Regular Contributor
  • *
  • Posts: 229
  • Country: us
Re: Ultra Precision Reference LTZ1000
« Reply #24 on: January 14, 2013, 04:50:55 AM »
Stupid question perhaps,   but is it possible to run one of these references inside of a vacuum to perhaps eliminate humidity from the picture?   Again it might be a stupid question.

Jeff
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf