New LTZ1000 10V Reference build (WAS: Agilent Keysight 03458-66509...)

[Rax]

I decided to radically modify this first post and organize in it the information collected during this pursuit. I'm hopeful this will help others put together a functional and accurate LTZ1000/A 10V standard. Of course there are other designs out there (and here), but the designer of this solution was very kind and helpful, and without that there would be no project.

Below are some of the more significant posts, links and other information necessary to build this.

• Dr. Frank's kit link: https://www.eevblog.com/forum/metrology/ultra-precision-reference-ltz1000/msg3886166/#msg3886166
• OSHPark PCB (design by "pepaslabs") can be ordered here: https://oshpark.com/projects/uwqf5gCo/view_design[/url
• Bulleted post with directions #1: https://www.eevblog.com/forum/metrology/agilent-keysight-03458-66509-dcv-reference-pca-for-3458a/msg5211921/#msg5211921
• Bulleted post with directions #2:  https://www.eevblog.com/forum/metrology/agilent-keysight-03458-66509-dcv-reference-pca-for-3458a/msg5258568/#msg5258568
• Attenuation factors for the critical resistors (TC-wise) in the circuit: https://www.eevblog.com/forum/metrology/agilent-keysight-03458-66509-dcv-reference-pca-for-3458a/msg5212140/#msg5212140
• Critical TC resistors can be ordered (in the US) at https://webdirect.texascomponents.com/SearchResults.asp?Cat=1857. No minimum quantity, excellent performance, reasonably priced (YMMV), ready in a few days in fully custom values and specifications. Frankly, it's pretty mind blowing this option is possible. I recommend the .25% Z201s for all these critical positions. You money should go to superior TC, not tolerance. Another options would be PRC (long term drift specified: https://www.eevblog.com/forum/metrology/agilent-keysight-03458-66509-dcv-reference-pca-for-3458a/msg5214996/#msg5214996), but I found this path difficult to source and therefore not recommended.

ORIGINAL POST BELOW:
In trying to put together more references in my lab, I've also been looking at these.

How much stability is to be expected from these? I am assuming one positive aspect is those available on the marketplaces out there are probably well aged.

That said, the PCB design is very vanilla, with no cutouts, etc.

In case these are worthy to chase, there's then the issue of cleanly supplying them, putting them in an optimal case, etc. Has anyone taken on a project like this before?

Thank you.

[alm]

How much stability is to be expected from these? I am assuming one positive aspect is those available on the marketplaces out there are probably well aged.

They might be well aged if they were continuously powered. If they were laying on a shelf for a long time, that might well start a new aging cycle. See the long term drift part of 3458A service note 18. If they have been powered for a long time, then I'd expect them to have settled below 4 uV/V/year in drift.

That said, the PCB design is very vanilla, with no cutouts, etc.

No voodoo slots of unclear function? I think quite a lot has been written about them in the long LTZ1000 thread.

In case these are worthy to chase, there's then the issue of cleanly supplying them, putting them in an optimal case, etc. Has anyone taken on a project like this before?

There was a supply of relatively cheap (~$100) 3458A reference boards a number of years ago. For example see this volt-nuts thread. If you search for posts from that time on volt-nuts or eevblog you might find people who repurposed them as stand-alone reference. Note back then there were some reports of having popcorn noise on some of those boards suggesting they were factory rejects, but I don't know if that applies to the currently available boards.

I would be concerned about the high temperature set point accelerating aging for a voltage reference.

[Rax]

No voodoo slots of unclear function? I think quite a lot has been written about them in the long LTZ1000 thread.

Well, good to learn an informed opinion on the value of that. As far as I gathered, its purpose is releasing mechanical tensions/vibrations/whatnot such that the reference would be pampered for immaculate Vref out.

But as much as I tried - multiple times - I've been unable to go through that entire LTZ1000 thread. I don't expect to live quite as long as that requires!... Not with my current lifestyle.

I would be concerned about the high temperature set point accrelating aging for a voltage reference.

Not sure at all what you mean with this. Is "accrelate" a GRE word?...

[Dr. Frank]

In trying to put together more references in my lab, I've also been looking at these.

How much stability is to be expected from these? I am assuming one positive aspect is those available on the marketplaces out there are probably well aged.

That said, the PCB design is very vanilla, with no cutouts, etc.

In case these are worthy to chase, there's then the issue of cleanly supplying them, putting them in an optimal case, etc. Has anyone taken on a project like this before?

Thank you.

Hello Rax,
to my knowledge, nobody has used these expensive HP3458A boards as an external, standalone reference.
That makes no sense at all, as these have many disadvantages, and even design flaws.

First, the LTZ1000A is misused by hp engineers, as it's running on about 90..95°C oven temperature. That's necessary because the 3458A is intended as an industry  environment DMM, not as a metrology instrument, i.e. it is specified to run up to 55°C.
As a proper reference, the oven temperature must not be higher than 60°C; the Datron / Wavetek / Fluke 7001 (designed by John R. Pickering) uses 45..50°C instead, like all other oven based references from Fluke, i.e. the 732's. I also have designed my references like this, i.e. @ 45 ... 55°C.
That gives a drift rate from scratch (w/o elaborate pre-aging) of < 1ppm/year, typically -0.5ppm/y.
References boards from hp commonly age -4 .. 8ppm/y, see also the threads about the 34470A and its reference board.
Only the high stability option is tested to -2ppm/y. Waste of money, to my opinion.

As a stand-alone reference, you urgently need a buffer circuit, to protect the LTZ1000 reference from short circuits, which would shift the reference voltage.
As well you can use this buffer to create a reasonably stable 10V output.

HP designed several other errors into the circuit, especially they still assemble this 200k resistor , which should NOT be used on the A version of the LTZ1000, as it creates an unwanted T.C. of about 0.3ppm/K. W/o that resistor, it probably would have near zero T.C.
The whole board is not covered, so the delicate solder joints might create thermal noise from the continuous air draught. In the LT datasheet it is strictly recommended to protect ALL solder joints of the PCB from air draughts.   

There are several other improvements to this standard circuit, like the adding of several blocking capacitors by Andreas, which definitely gives better suppression of external e.m.c., and to my opinion better noise performance.

I have designed the most simple PCB, single sided, non -A version, no voodoo slits, no fancy 4 layers with mystical, squiggled tracks.
Simply thermally balanced tracks, where needed, and least expensive components, i.e. no ultra stable Vishay hermetically sealed, oil filled VHP something resistors.

I have put it in a double shielded enclosure, i.e. tuner box with thermal insulation, and an outer aluminum case, and used a simple 12V supply.
Andreas and branadic have added a low noise LDO on their boards, so that's the best solution, I guess, especially when you add a battery backup.

These circuits simply run ultra stable over time, and are nearly free of noise and "popcorn" disturbances.

So I recommend to build one or two of those reference boards:

https://www.eevblog.com/forum/metrology/ultra-precision-reference-ltz1000/msg1336573/#msg1336573

https://www.eevblog.com/forum/metrology/ultra-precision-reference-ltz1000/msg3886166/#msg3886166

Frank         

[Rax]

This is great, Frank, thank you very much for guiding me through the weeds of the gianormous LTZ1000 thread and pointing a couple of posts with very substantial input.

This gives me a lot to go on. Thanks!!

[Rax]

Not to push my luck too much on this, but does anyone still have any excess boards lying around in the aftermath of their "LTZ1000 reference design adventure," or maybe gerbers or KiCad files, or whatever the design files were?

PM works as well, if you prefer.

[Rax]

So I recommend to build one or two of those reference boards:

https://www.eevblog.com/forum/metrology/ultra-precision-reference-ltz1000/msg1336573/#msg1336573

https://www.eevblog.com/forum/metrology/ultra-precision-reference-ltz1000/msg3886166/#msg3886166

Frank         

OMW, Frank!... Order from OSH Park pending (thanks to Andreas/Jason!), I should be up and running in a couple of weeks.

Meantime... Is there a BOM available? Or at least, which parts are more critical? I am aware some of the resistors are.

[Rax]

It looks like R1 through R5 have enhanced requirements for tempco. I think Frank used 3ppm types, and I'm combing the internet for affordable ones in the US. Anything under 5ppm is an arm and a leg though... 

[alm]

Have a look at the Sfernice resistors from: this Polish seller. Shipping to the US shouldn't be too bad. Initial tolerance doesn't matter much. I don't know how many values they still have in stock, but note that for most positions it's the ratio, not absolute value, that matters. Details are on the forum, but I don't have a link handy. I wrote a script for that a long time ago that would find suitable resistor pairs. I could probably dig that up if you're interested.

[Kleinstein]

In the normal design the most critical resistors are R4 and R5 setting the oven temperature. Here the ratio is what matters and an array with 8 equal resistors (e.g. TDP type or NOMC) with resistor combinations is an option for a few ratios.

The next critical ones are R1 and R2. R3 is already less critical and does not have to match R2 - it is just convenient to have a shorter BOM. Also the value for R2/R3 does not have to be exactly 70 K some 47 K or 75 K would be OK too.
There is a table on how much the drift of the resistors is attenuated.

By far the most critical resistors will be the one for the 7 to 10 V step, if one only has a 10 V output. With both a 7 and 10 V out one may be able to measure this separately and thus correct for this.

For the resistors a 5 ppm/K and even 10 ppm/K would be OK. The point is more with the long term stability, but this often not specified very well. A low TC is more used as an idnicator for good quality, though the TC does not directly correlate with long term drift. It is more that high excess noise may indicate a tendence for long term drift. For this reason, I no longer like the PTF56 series very much. Similar the NiCr NOMC version is perferred over the TaN NOMCA version.
If SMD is OK, I would consider the Susumu  RG series as a relative afforable option, possibly even the 10 ppm/K ones.

The op-amp is usually still a LT1013 - keep in mind the unusual pinout of the SO8 version. I personally see not need for one of the better versions here.

[3roomlab]

OSOP 12x array ?
it allows the 12.5:1 ratio
eg digikey : OSOPTC2001AT0

In the normal design the most critical resistors are R4 and R5 setting the oven temperature. Here the ratio is what matters and an array with 8 equal resistors (e.g. TDP type or NOMC) with resistor combinations is an option for a few ratios.

The next critical ones are R1 and R2. R3 is already less critical and does not have to match R2 - it is just convenient to have a shorter BOM. Also the value for R2/R3 does not have to be exactly 70 K some 47 K or 75 K would be OK too.
There is a table on how much the drift of the resistors is attenuated.

By far the most critical resistors will be the one for the 7 to 10 V step, if one only has a 10 V output. With both a 7 and 10 V out one may be able to measure this separately and thus correct for this.

For the resistors a 5 ppm/K and even 10 ppm/K would be OK. The point is more with the long term stability, but this often not specified very well. A low TC is more used as an idnicator for good quality, though the TC does not directly correlate with long term drift. It is more that high excess noise may indicate a tendence for long term drift. For this reason, I no longer like the PTF56 series very much. Similar the NiCr NOMC version is perferred over the TaN NOMCA version.
If SMD is OK, I would consider the Susumu  RG series as a relative afforable option, possibly even the 10 ppm/K ones.

The op-amp is usually still a LT1013 - keep in mind the unusual pinout of the SO8 version. I personally see not need for one of the better versions here.

[Rax]

I could probably dig that up if you're interested.

Yes, please. Thank you!

[Rax]

In the normal design the most critical resistors are R4 and R5 setting the oven temperature. Here the ratio is what matters and an array with 8 equal resistors (e.g. TDP type or NOMC) with resistor combinations is an option for a few ratios.

The next critical ones are R1 and R2. R3 is already less critical and does not have to match R2 - it is just convenient to have a shorter BOM. Also the value for R2/R3 does not have to be exactly 70 K some 47 K or 75 K would be OK too.
There is a table on how much the drift of the resistors is attenuated.

By far the most critical resistors will be the one for the 7 to 10 V step, if one only has a 10 V output. With both a 7 and 10 V out one may be able to measure this separately and thus correct for this.

For the resistors a 5 ppm/K and even 10 ppm/K would be OK. The point is more with the long term stability, but this often not specified very well. A low TC is more used as an idnicator for good quality, though the TC does not directly correlate with long term drift. It is more that high excess noise may indicate a tendence for long term drift. For this reason, I no longer like the PTF56 series very much. Similar the NiCr NOMC version is perferred over the TaN NOMCA version.
If SMD is OK, I would consider the Susumu  RG series as a relative afforable option, possibly even the 10 ppm/K ones.

The op-amp is usually still a LT1013 - keep in mind the unusual pinout of the SO8 version. I personally see not need for one of the better versions here.

Well, I did some forensics in my parts bin (and several projects on the back burner) and found some stuff:

• A bunch of PTF56 and 65s, such as: 1k (.05%/10ppm); 100R (.1%/5ppm); 20k (.01%/10ppm); 200k (.1%/5ppm); 100k (.05%/10ppm); 10k (.05%/5ppm); 50k (.1%/10ppm); 100k (.1%/10ppm); 600R (.1%/5ppm)
• A TT RC55Y: 100R (.1%/15ppm)
• Z201: 100R (.005%/.2ppm)
• These are in a loan of sorts, but may be able to use them: Z102k(?): 4k (.02%/2.5ppm?); 1k (.05%/2.5ppm); VHS102: 40k (.1%/2ppm); Dale WWA-23: 30.1k (?/?)

I'll research the topic of ratios I need to be seeking, but that's what I have at hand (list for myself, which I thought won't hurt posting here).

[alm]

I could probably dig that up if you're interested.

Yes, please. Thank you!

I couldn't find what I used for the resistance ratios, but I did find what I used to extract the list of values. if you open the developer console (right click -> inspect, then go to console tab) on an eBay page with a list of resistance values like this, and paste this line will print out a list of all in-stock values on that page as CSV. Do this on every page in the shop, paste the results in a CSV file and then do some spreadsheet manipulation (probably pivot table) to make it into a neat list with one value per row.

Code: [Select]

options = []; for (option of document.getElementById("x-msku__select-box-1000").childNodes) { if ((option.attributes.disabled?.nodeValue != "disabled") && (option.childNodes.length > 1)) { options.push(option.childNodes[1].data) } }; console.log(document.location.href, ',', document.getElementsByClassName('x-price-primary')[0].innerText, ',', options.join(','));


[Rax]

this Polish seller[/url]. Shipping to the US shouldn't be too bad.

Wow, that's a really good source (particularly assuming all those are genuine, which I take as the case due to you recommending it - not that I assume any guarantees, but at a minimum good experience from an exigent customer ). With the exposure here, I wonder how long they'll have what I assume is a limited stock.

Shipping charges are not a buck more, nor less than what I pay to my usual domestic large distributor (where I purchase 90% of my parts). Of course, transit will be distinctly different, but at least the overall cost of the project can be kept at bay.

[Rax]

In the normal design the most critical resistors are R4 and R5 setting the oven temperature. Here the ratio is what matters and an array with 8 equal resistors (e.g. TDP type or NOMC) with resistor combinations is an option for a few ratios.

From the precision resistors I already have here, the only combo with a 12.5x ratio I can see is the 50k/4k combo. But I assume being different types (different TCs, etc.), that's actually a very poor fit.

[Rax]

By far the most critical resistors will be the one for the 7 to 10 V step, if one only has a 10 V output. With both a 7 and 10 V out one may be able to measure this separately and thus correct for this.

The R12-R15+R18 is a very confusing set. I assume this is for fine setting the voltage out for 10.00000V. Not sure which value for each resistor (for instance, R12 vs. R12A, as they're not supposed to be populated at the same time, correct?) is supposed to be used. Particularly as I'm not looking to get 7V out. I'm only looking for 10V out of this.

[Kleinstein]

In frank's schematics the resistors with the A are alternative footprints to allow different resistor types on the same PCB. Usually they overlap and thus no parallel use sensible. The value of R14 depends on the individual reference voltage and maybe the exact values of some other resistors. The resistor combination is used to get most of the gain from 2 very stable resistors, than a small part  (e.g. 1-3%) for a coarse trim (R14) and finally the fine trim from the trimmer.

There are different ways to combine resistors. With THT parts chances are one could also fit a slightly different configuration (e.g. one more trim resistor) on the same PCB.

I would consider it good to have not only acess to the 10 V output, but also the 7 V - ideally with a 2nd buffer amplifier. This would allow to check for drift of the 7 to 10 V stage.

P.s.: the 7 V output from the FB divider may be good enough, as this is essentially only for testing either the 7 or 10 V, but usually no both at the same time.

[Rax]

In frank's schematics the resistors with the A are alternative footprints to allow different resistor types on the same PCB. Usually they overlap and thus no parallel use sensible.

That's what I thought, but in some cases (R12, R13) the main and the alternate seem to have different specific values. Are those there just to point out a couple of options for the same resistor?

[Rax]

P.s.: the 7 V output from the FB divider may be good enough, as this is essentially only for testing either the 7 or 10 V, but usually no both at the same time.

I assume this means JP9, correct? It seems almost a "free" 7V output, and I assume no trimming is really necessary for it (other than gain considerations). I guess relative examination of JP8 and JP9 would allow for monitoring the drift of the "7V to 10V"stage.

For reference, I am adding here the v2 schematic (I think this is the latest and greatest), and the PCB "parts side."

[Kleinstein]

The different values for R12 / R12A and so one are different example combinations, not bond to the form factors. It is mainly about the ratio and staying in a reasonable range.

I would ideally consider an added resistor (could be a simple one) between the 7 V raw reference and where C9 and the LTC1052 meet, so to really keep away switching spikes, though the LTC1052 is a low current noise one and thus only relatively small spikes. The resistor would also avoid capacitive loading the LT1013.

Not much trimming for the 7 V - if the non A version is used one may want a selected R10 to fine trim the TC. The suitable value depends on the units and details like the thermal setup and set temperature. The 400 K in the DS are more like an order of magnitude number. The actual best value may be from some 100 K to 1 M.

[Grandchuck]

Hi Rax.  Regarding your Reply #19, I have three of those.  That board and circuit are a good choice.  My three are better than my Agilent 3458A, which has a measurement uncertainty of 3.64 ppm on the 10 volt dc range (I have the Agilent calibration report).

[alm]

My three are better than my Agilent 3458A, which has a measurement uncertainty of 3.64 ppm on the 10 volt dc range (I have the Agilent calibration report).

How can you tell measurement uncertainty from a calibration report? The calibration report will state the deviation at the time of calibration (which is not an uncertainty but an actual measurement) and the uncertainty of the calibration, which is more a statement about the calibration procedure and the lab, than the meter. If you sent your LTZ-based reference to the same lab and it got calibrated using the same equipment, you'd likely get a very similar uncertainty.

Of course the specifications of your 3458A will state an uncertainty, but so will the LTZ datasheet. In either case you can develop a better estimate using multiple calibrations over a period of time.

[Grandchuck]

Thanks alm.  After years of measuring and expermenting, it is clear that the Dr. Frank references are more stable than the DMM.  I have some collected data that, to me, support my assumption.  I could be wrong, of course.  I am not in a position to buy more stuff or employ professional calibration services.  I have sealed off that rabbit hole 
I just might be a recovering voltnut.

[alm]

Thanks alm.  After years of measuring and expermenting, it is clear that the Dr. Frank references are more stable than the DMM.  I have some collected data that, to me, support my assumption.  I could be wrong, of course.

I would agree that's likely, and I'm not challenging your assertion, but I'm confused how you can look at a cal certificate and derive the current measurement uncertainty of the instrument from that. And in particular how you can learn anything about the stability of an instrument unless you have multiple calibration certificates (which I agree is an expensive hobby).

[Grandchuck]

Not claiming to "derive the current measurement uncertainty of the instrument from that".   I feel that my 3 references are more stable than the DMM.  That is all.

[Rax]

In looking at low-TC resistors, I came across these (see enclosed datasheet). To me, they look like naked z-foils (meaning - the enclosure is stripped off), which I can't imagine being good for their TC... What's any air movement doing to their exposed resistive core? Supposedly though, they have the same TC as their "dressed" siblings (.2ppm/C), but to me that sounds like at best wishful, at worst a sprinkle of fairy dust or snake oil.

Has anyone tried these by any chance? If the observed performance is as stated on the datasheet, I may be aware of a relatively affordable source (vs. performance point).

[Vgkid]

A TC test on the VAR
https://www.eevblog.com/forum/metrology/t-c-measurements-on-precision-resistors/msg879467/#msg879467

[Rax]

A TC test on the VAR
https://www.eevblog.com/forum/metrology/t-c-measurements-on-precision-resistors/msg879467/#msg879467

So not very good. It looks like an order of magnitude higher than the Vishay claims on TC specifically.

[TUMEMBER]

A TC test on the VAR
https://www.eevblog.com/forum/metrology/t-c-measurements-on-precision-resistors/msg879467/#msg879467

So not very good. It looks like an order of magnitude higher than the Vishay claims on TC specifically.

https://www.eevblog.com/forum/metrology/t-c-measurements-on-precision-resistors/msg879820/#msg879820
"Re. Z-foil humidity drifts, I had initially used standard (non hermetic) as references, and have seen drifts of about 5-10ppm over the year. That was the end of these references... The hermetic ones were up to Vishays spec (2 ppm in 6 years, low load, iirc).
So if you want to use standard z-foils in the voltage divider (multiplier) of an LTZ design to step up to 10V, they are probably not worth it. I have not done any tests with dissicators or putting them into oil, which my be worth a try, so cannot comment.
The other option /besides hermetic resistors/dividers) that may be more stable is dividers in one package. This way the humidity impact affects both resistors on the substrate (made from the same material) the same way and one would expect a more symmetrical behaviour."

[Rax]

The point is more with the long term stability, but this often not specified very well.

One US manufacturer that seems to provide this data is PRC. Their HR/HVA series - their highest precision and stability series - show a 10ppm/year "greater stability." TCR can be as good as 1ppm/C, or "matched" to .5ppm/C.

Datasheet enclosed.

[Rax]

I'll continue posting here as I find more similar data.

My intention with this "new" LTZ1000 10V reference build thread is to make it as useful and as simple and as I can, organize as much as possible information accrued trough the efforts here (which should be for the benefit of others down the line from me), and to introduce new information where possible.

For instance, I don't think information on these "long term rated" PRC resistors was presented here before (anywhere on the Metrology section).

[Rax]

Research results this far below. Please note this is relative to Frank's design.

• PCB and schematic - see https://www.eevblog.com/forum/metrology/agilent-keysight-03458-66509-dcv-reference-pca-for-3458a/msg5200593/#msg5200593
• R4/R5 set the oven temperature. These should be under 2.5ppm/K TCR. The recommended value is:
  
  • LTZ1000 - 45-50C: 12k/1k
  • LTZ1000A - 55-60C: 12.5k/1k or 10k/800R, depending on resistor available values for a certain type
• R1 (120R) also needs to be under 2.5ppm/K
• R10 is only needed for the LT1000 (per datasheet). Tweaking the value is subject to experimentation. I don't yet have this nailed down.
• R12/R13 - the schematic at the link above shows two values available for two types of resistors. Also under 2.5ppm/K. The relevant part is the ratio - 1.4 - as this sets 10V out from the Vref actual output. For instance:
  
  • FLCY: 10k/4k
  • econistors: 15k/5.6k

In the US, these resistors are pretty pricey, but the only reasonable solution I was able to find is Texas Component Corporation. They sell S102s on the C and K variety, and also Z102s. Some of the values above will need to be custom, some others are available stock.

[Kleinstein]

There is no hard 2.5 pm/K limit for the resistors. The resistor drift is attenuated by a certain factor, that varies for the different resistors in the circuit.
The attenuation ratio is on the order of 100 and if the demand is not too high one may well get away with 10 ppm/K resistors. This is especially with the non A version, when R10 is used to trim the resudual TC anyway. The relevant point is anyway more long term drift and not actually the TC. A precision reference is usually used near room temperature.

It is only the R12/R13 ratio that has much less attenuation (around 3 fold) and here even 2.5 ppm/K may not be good enough.

R4/R5 and also R12/R13 are only relevant as a ratio, so the relative TC / dirft is what matters. This can make quite some difference with resistor arrays.
For both ratios it is possible to use resistor arrays like TDP, e.g. as  6S:2P or 2S:6P for a 12:1 ratio. The specs for the relative TC are not that great (5 ppm/K), but this includes the more tricky low resistance versions. With the relevant 2K or 5 K per resistor the matching is typical much better and averaging also helps a little.

[Dr. Frank]

Rax,
I found my  table 3_, where I collected the measurements from other volt-nuts concerning these attenuation factors.
That means, that the T.C, and the timely drift is attenuated by the sum of these factors for the completed circuit.
The numbering R1 .. R5 refers to the original  LT datasheet.
I have found similar results, and selected my econistor resistors, after I measured their individual T.C.s.
As you can see from table 4_, the calculated overall T.C. of the circuits differ from the measured ones, in case that you don't use this compensation resistor R9 (it's R10 in my schematic).
I think, that the residual T.C. is dominantly determined by the individual LTZ1000.
Andreas made detailed experiments with longer or shorter legs, and was able to show, that the T.C. also depends heavily on that.
Therefore, the ultimate quest for ultra low T.C. of these resistors is not necessary.
Medium stability, lower cost resistors like these econistors (compared to those ultra stable Vishay VHPxxx, hermetically sealed, oild filled ones) do the job perfectly.

I also can tell, that all these references perform very stable and predictable over time (since 2016), as all of those show an initial, annual drift of < 1ppm/y. after a short run-in, and w/o any pre-ageing / baking of the LTZ.  Meanwhile, all of them show drifts between -0.25ppm/y .. -0.5ppm/y. only.

Frank

[TUMEMBER]

There is no hard 2.5 pm/K limit for the resistors. The resistor drift is attenuated by a certain factor, that varies for the different resistors in the circuit.
The attenuation ratio is on the order of 100 and if the demand is not too high one may well get away with 10 ppm/K resistors. This is especially with the non A version, when R10 is used to trim the resudual TC anyway. The relevant point is anyway more long term drift and not actually the TC. A precision reference is usually used near room temperature.

It is only the R12/R13 ratio that has much less attenuation (around 3 fold) and here even 2.5 ppm/K may not be good enough.

R4/R5 and also R12/R13 are only relevant as a ratio, so the relative TC / dirft is what matters. This can make quite some difference with resistor arrays.
For both ratios it is possible to use resistor arrays like TDP, e.g. as  6S:2P or 2S:6P for a 12:1 ratio. The specs for the relative TC are not that great (5 ppm/K), but this includes the more tricky low resistance versions. With the relevant 2K or 5 K per resistor the matching is typical much better and averaging also helps a little.

Very similar conclusions can be drawn from this post
https://www.eevblog.com/forum/metrology/ultra-precision-reference-ltz1000/msg1051957/#msg1051957

[Rax]

In the US, these resistors are pretty pricey, but the only reasonable solution I was able to find is Texas Component Corporation. They sell S102s on the C and K variety, and also Z102s. Some of the values above will need to be custom, some others are available stock.

Not to advertise too strongly, but the other advantage to TCC is they have differentiated pricing for resistors from a series, where less precision is cheaper. For instance, the Z201s (.2ppm/C) are available at decreasing cost from .005% to .25%. Hence, where relative TC is critical (ratio of two resistors across temperature differentials), this is very advantageous as one can order less precision but very high TC for a relatively low price.

[julian1]

I can report good experiences ordering from texas components also.

Turnaround time, from order placement, to *custom* manufacture, to having the parts in my hands was ten days.

For the product lines they carry z201, s102, vsmp, smnz etc - and unless somebody tells me otherwise - I suspect they are the original and sole manufacturer of these parts that are more commonly known,sold, and marketed under the Vishay/VPG name.
Presumably it was decided there was still a good business-case to carry on trading under their own name, even after they were acquired/folded into the Vishay group.

Edit. clarify potentially misleading point.

[Rax]

courier options limited and expensive.

I am not quite seeing that - the shipping options I'm presented with are two (FedEx 2nd day and "domestic 2nd day," which I assume is USPS Priority) and they're roughly at the price I'd expect them to be retail. Though maybe julian1 is looking at international shipping which may be an entirely different business.

[Rax]

But turnaround time, from order placement, to *custom* manufacture, to having the parts in my hands was ten days - which is what actually matters.

The other positive aspect to them is this custom resistor option. So one can ask for any value they want/need and they'll make it as ordered. If the turnaround on that is 10 days, that's pretty dope. And at a price that's lower than the regular, large, off-the-shelf retailers in the US (for the same exact product).

[julian1]

I fear I've given the wrong impression from my comment.
Ordering is very simple and streamlined, and international courier pricing is not unreasonable. 
The point I wanted to convey was not to be put off, if the UI experience isn't as slick as digikey or mouser.
They are a *very* handy vendor to know about.
hopefully they will continue taking small-lot custom orders from the general public, even if they probably only just break even.

[Rax]

The relevant point is anyway more long term drift and not actually the TC. A precision reference is usually used near room temperature.

I found and provided some spec data on the PRCs, and found some reports of 1-2ppm/yr drift for the Z201s (which I can pretty easily source in the US, alongside the S102s). So the Z201s seem to fare pretty well.

I'd be interested to learn what other resistors may have returned in similar tests. Namely, whether there are some standouts in a world where this data is seldom specified.

[Kleinstein]

Data on the long term drift are rare and even if one finds something this is often for accelerated tests at higher temperature or similar and the question is if all units behave the same. Most data-sheets are written when the parts are new and at that time there are naturally no long time data available.
There are some long term dirft data for the Susumu RP SMD resistors, though it is hard to read the graphs. I consider these resistors more at the lower end for what makes sense with the LTZ1000.

Anecdotal data from more normal users of the resistors can be tricky as they can be biased (getting more of the extremes) and the conditions are more real life and often less comparable. There is more than just time and temperature to effect the long term drift. Temperature and humidity cycles and maybe PCB stress can have an effect.

For the resistors directly at the LTZ1000 the Z201s are overkill . For the 7 to 10 V gain part they can be justified.

[Dr. Frank]

 
I can confirm what Kleinstein stated: long term drift is rarely specified, and if done, it's often not reliable, or the drift specification is barn-door wide.

Especially all claims from Vishay have to be treated with utter care!

I purchased 5 EA, 10kOhm, 0.0005%, with measurement protocol, oil filled, hermetically sealed VHP202Z from Vishay, which were "specified" to age < 2ppm/ 6 years.

I don't know until today, what went wrong at Vishay, but those resistors, which I wanted to use as secondary standards, already had already drifted out of specification on arrival.
The annual drift, until today, was always -1ppm/year only. The promised low T.C. of the Z - technology, as of typical (!) 0.05ppm/K, or so, was not met at all, in reality it is between 0.3 .. 1pppm/K, which is following their more realistic maximum specification.

On the other hand, in my hp3458A from 2000, they already used those VHP101 for the internal 40kOhm standard, which is hermetically sealed, probably oil filled, definitely rock stable, as seemingly it is not drifting at all / any more, or not more than +/- 0.5ppm over at least 6 years. I precisely monitor my resistors over many years by aid of ab precision, btw.

This VHP101  consists of two resistor chips in series, with C and K characteristics, i.e. their T.C.s nearly cancel, to give a very low overall T.C., which my monitoring data also reveal. If I remember correctly, the T.C. seems to be below 0.2ppm/K.. I'll check that
I always recommend to order VHP101 if intended for such high stability  applications.

Those are not necessary for the LTZ1000 circuit.
Frank

[Rax]

Thank you both.

From my research this far, other than TCC, every other place I talked to asks for volume and/or takes weeks to source (if you meet the first criteria). Pricing is not available until you meet the first two criteria. Honestly, my time is more valuable than the price difference from average cost to great resistors exceeding the requirements that cost a bit more. When looking at off the shelf resistors, they are not available at needed value, and they're typically more expensive (in some cases far more expensive - we're talking some getting close to $200, cause why not?  ) than the custom ones from TCC (which are not in all cases higher cost than their "off the shelf" ones?... go figure).

Then at TCC specifically, the price difference between Z201s and any of the S102s is either small or even at the advantage of the Z201 in some cases. I ended up getting a set of Z201s for about $20 each.

It's true, Rhopoint in UK has 2ppm resistors for about $7-8 (though not all values in stock, and those not in stock need 20+ volume and have a lead time of 28 weeks...), but by the time I factor in the cost of shipping I'm just about breaking it even with TCC.

PRC sent me around to their distributors and that led to more useless email exchanges and that just piles more wasted time.

Just the quirkiness of today's markets and availability of products and supply chain shortages and who knows what else.

[Rax]

Also, btw, I couldn't find the combination of values needed at the recommended ePay sellers either. Being that the relative TC is critical, I'm keen on making sure any pairs are of the exact the same kind of resistors, including manufacturing batch.

[Andreas]

hmm,

please read:
https://www.eevblog.com/forum/metrology/t-c-measurements-on-precision-resistors/

many resistors of the same batch with different outcome.
in the single digit / sub ppm T.C. range every resistor looks individual.
The amount of cement to bond the metal to the substrate is never the same nor equally distributed.
And small differences in packaging have large effects.

with best regards

Andreas

[Andreas]

Hello,

another recommended reading:
https://www.eevblog.com/forum/projects/vishay-bulk-foil-drift-after-soldering/msg445297/#msg445297
so solder with care.

with best regards

Andreas

[Jendas]

Hello,

with inquiry about precision resistors I tried to contact this manufacturer: http://www.direct-token.com/en/resistors-ppm/index.html. They make a wide variety of various resistors. I send them a email about UPCS resistor type (similar to Vishay S102 or Z201 etc.. ) with this response on specific values:

UPSC 70K T C10 P, US$6.5/Pc for 30Pcs/MOQ.
UPSC 12K T C10 P, US$6.5/Pc for 30Pcs/MOQ.
UPSC 1K T C10 P, US$6.5/Pc for 30Pcs/MOQ.
UPSC 120R T C10 P, US$6.5/Pc for 30Pcs/MOQ.
Lead Time is 5~6 weeks after order and remit payment confirmed.

All resistor are C10 = ±2 ppm/℃,  and T = ±0.01%
Datasheet: http://www.direct-token.com/en/pdf/resistors-ppm/ultra_precision_resistor_upsc.pdf

with best regards
Jendas

[alligatorblues]

Regulator ICs like the LTZ1000 make possible very stable voltage standards. But after only a short time futzing with 2 of them, I realized a precision oven is required, because the external components add a lot of temp drift. I completed one, fully enclosed, very stable at 10VDC. but temp drift of 1.0ppm/C. Even though the regulator is temp regulated, and even if the heater control is well implemented, the other components have to be inside a precision oven to obtain the results experimenters want.

 

[Dr. Frank]

Regulator ICs like the LTZ1000 make possible very stable voltage standards. But after only a short time futzing with 2 of them, I realized a precision oven is required, because the external components add a lot of temp drift. I completed one, fully enclosed, very stable at 10VDC. but temp drift of 1.0ppm/C. Even though the regulator is temp regulated, and even if the heater control is well implemented, the other components have to be inside a precision oven to obtain the results experimenters want.

An outer oven is not required at all!
Maybe you have used low quality resistors only, or you made another design error.
Please provide your Bill of Material, and your schematic.

Please do not make such misleading statements in this forum, that for the LTZ1000, a precision oven is required!

Frank

[dietert1]

I think a temperature gradient of 1 ppm/K is a lot and easy to have if the 10 V to 7 V divider is made from 2 ppm/K resistors or any other low TC resistor. Dr. Frank did not understand that statement.
Of course the statement that an outer oven is required is a bit strong. Usually people get away with compensating the residual TC, which is fairly easy to do and has been decribed in many places.
As the residual TC contains nonlinear terms, the best solution is tuning the residual TC to zero at a certain temperature and run the circuit inside an oven of that temperature. Others attempted to compensate the nonlinear residual TC, too.

Regards, Dieter

[Dr. Frank]

I think a temperature gradient of 1 ppm/K is a lot and easy to have if the 10 V to 7 V divider is made from 2 ppm/K resistors or any other low TC resistor. Dr. Frank did not understand that statement.
Of course the statement that an outer oven is required is a bit strong. Usually people get away with compensating the residual TC, which is fairly easy to do and has been decribed in many places.
As the residual TC contains nonlinear terms, the best solution is tuning the residual TC to zero at a certain temperature and run the circuit inside an oven of that temperature. Others attempted to compensate the nonlinear residual TC, tool.

Regards, Dieter

Hello Dieter,
no, I understood his various statements very well.
Please read his first two sentences.

There he stated, that the LTZ1000 circuit (the "regulator IC", as he calls it  ) requires an outer oven, because the external components add a lot of T.C.

This statement is ultimately wrong, as all the volt-nuts, who really contributed with a T.C. characterization of their LTZ circuit have demonstrated, that the overall T.C. is well below 0.1ppm/K when using appropriate external resistors, and especially when T.C. compensation is used to bring that T.C. down to near zero .. INSTEAD of using an outer oven, of course.
Especially due to the attenuation factors of the resistors instability parameters, those external component in contrary do NOT contribute a lot to the overall T.C.
 
I also remember that we already had this big discussion, that an outer oven would degrade the performance of the LTZ circuit, and is counter-productive.

Then he obviously mixes the following gain stage, to 10V, into his consideration.
But that's a completely new aspect.
Sorry Dieter, I fully understood that as well!

Sure, if he uses not T.C. - matched resistors, or ones which have too high a T.C., this 10V output will have too a high T.C. as well.
You can of course "heal" that design fault by using an oven... that's called over-engineering,I'd guess.

In his last sentence, he concludes, that only due to the mediocre T.C. of his 10V gain stage, that the WHOLE circuit must be inside an oven.

And that simply shows, that he has not understood the basic LTZ circuit at all.

Dieter, anyhow, I hope you and your family are doing fine now in Brazil!

Frank

[dietert1]

When someone designs a 10 V reference, there will always be the problem to make a 10 to 7 V divider with low TC. One solution is a precision PWM. When using a resistive divider, even the best available resistors will likely produce a residual TC of 1 ppm or more. Plus amplifier temperature drift, thermal EMF and the like. I think that is what alligatorblues meant.
In general it is easy to tune the linear TC to zero or near zero at a certain temperature. Due to nonlinear terms, the usable temperature range will still be limiited, e.g. to about +/- 4 K around the zero TC temperature for less than 1 ppm error. Using an oven is a valid method to get rid of that last ppm. Of course if you have a 23 +/- 1 °C lab everything becomes easier.
A peltier oven is fairly easy to make and does not require a higher than ambient temperature.
Another consideration is whether it makes sense to TC adjust the 7 V reference and the 10 V gain stage separately. I`d guess most of us would try to save some work and put everything onto one PCB that fits inside the oven, with one TC adjustment for the whole circuit. With some luck one can even save the TC adjustment and instead adjust the oven temperature to the zero TC temperature.
As far as i understand the ADR1001 implements these concepts to a certain level. Similar to other AD products it compensates TC to some level and then relies on oven temperature stability.

Regards, Dieter

[Rax]

[...] the usable temperature range will still be limited, e.g. to about +/- 4 K around the zero TC temperature for less than 1 ppm error. Using an oven is a valid method to get rid of that last ppm. Of course if you have a 23 +/- 1 °C lab everything becomes easier.
[...]
Regards, Dieter

I have not participated in all the original pursuits and research - obviously, hence this thread - but I am getting close to putting together at least a couple of kits (probably at least one LTZ1000 and one LTZ1000A), though could build several with different bits I put together. My environment doesn't fit this description, so I'll likely have some data to report along these lines.

[Dr. Frank]

When someone designs a 10 V reference, there will always be the problem to make a 10 to 7 V divider with low TC.
...
 When using a resistive divider, even the best available resistors will likely produce a residual TC of 1 ppm or more. Plus amplifier temperature drift, thermal EMF and the like.....

In general it is easy to tune the linear TC to zero or near zero at a certain temperature. Due to nonlinear terms, the usable temperature range will still be limiited, e.g. to about +/- 4 K around the zero TC temperature for less than 1 ppm error. Using an oven is a valid method to get rid of that last ppm. ...

Regards, Dieter

Hello Dieter,
I still do not agree at all to your assumptions / proposals.

Here's my LTZ #3 which I use as a traveling standard. Recently ab precision compared both reference outputs against his 732B with about 0.3ppm uncertainty, and on return, it agreed within 0.3ppm with my reference group as well.

I used 2 FLCY from AE (those two blue components, I think 10k and 4k) for the 10V gain stage.
Per datasheet, they have a T.C. of 2.5ppm/K, but I already knew from another characterization, that they are good for about 1ppm/K.
I have not measured these in advance, only in situ, i.e. the overall T.C.
The T.C. of the direct LTZ output is about 0.02ppm/K.

The diagram shows a box - T.C. of the 10V output of about 0.7ppm/11k = 0.06ppm/K.
The chopper amplifier, either an LT1050, or an ADA4522, have of course no influence, and emf's are very low by design, i.e. in the 0.02ppm ballpark.

For lab temperatures, any non linearities can be neglected.

I also see no big timely drift of this 10V reference, I will review my monitored data and will report back.

Frank

[Kleinstein]

The 7 V reference part does not need an extra oven, though a stable temperature would not hurt, unless it gets rather hot. With the trim resistor (R10 in Franks circuit) one can trim the linear TC, though this can add some 2nd order term. It is still good enough for a normal indoor range like +-10 K.

For the 7 to 10 V step a oven for a stabilized temperature is a possible way to go, especially if the temperature variations are large.
However even than one should still have rather good resistors as the resistors also effect the long term drift. With a small temperature range one may than get away without an oven.  If really needed one can trim the TC by adding a little copper wire / trace to contribute to one of the resistors. This may still be easier than building on oven.

[Andreas]

When someone designs a 10 V reference, there will always be the problem to make a 10 to 7 V divider with low TC.

Mhm,

with a statistical divider I got 0.026 ppm/K difference between 6.6V and 10V output on my ADR1000A#1
https://www.eevblog.com/forum/metrology/lowest-drift-lowest-noise-voltage-reference/msg3931196/#msg3931196

The only problem is the ageing drift of this divider (but now stabilizing after ~2 years)

with best regards

Andreas

[Kleinstein]

Getting 0.03 ppm/K is likely some luck, but a resistor array (e.g. 8 equal resistors) is a real option. With my ADC I got some 0.5-1 ppm/K for a circuit with 2 skale factors  (ref. amplification 7 to 28 V) and the ADC itself (more like an -1 gain amplifier).
The 2:1 ratio for the ADR1001 is a little easier, but one can still get a 10:4 = 5:2 ratio (for a 7.1 V ref) from 7 equal resistors.

[iMo]

I think one cannot generalize on the 7->10V stage based on a single (or couple) of successful builds with such small TCs as above, imho. Those are almost outliers

What Dieter indicates is that the version with the ovenized 7->10V stage is a safe and least expensive solution than an elaboration with statistical arrays (randomness) or manual fine-tuning of TC (time, labor costs). It means ie. with an 0.1C oven and 5ppm/C resistors you will always get the total TC<1ppm/C with no effort or risk.

[Dr. Frank]

I think one cannot generalize on the 7->10V stage based on a single (or couple) of successful builds with such small TCs as above, imho. Those are almost outliers

What Dieter indicates is that the version with the ovenized 7->10V stage is a safe and least expensive solution than an elaboration with statistical arrays (randomness) or manual fine-tuning of TC (time, labor costs). It means ie. with an 0.1C oven and 5ppm/C resistors you will always get the total TC<1ppm/C with no effort or risk.

Well, this technique of T.C. matching of resistors and T.C. characterization / trimming of the complete circuit has been done by FLUKE for all of their old reference instruments, like 730, 731, 332/335, and this statistical array plus T.C. trimming has been done by John R. Pickering in his Datron / Fluke 7000 references, very successfully.
I don't regard an ovenized 7=> 10V stage neither as a least expensive, nor as a simple solution, especially in regards of volt-nuts, who mostly strive for most simple and cost effective solutions, which also work practically. That's always my approach, to invest some good ideas and a bit of measurements for simple solutions, instead of over-engineering things. Btw., I work in R&D of the Automotive Electronics industry, where you have to save every tenth of a penny.

I can't remember that anybody has published his fully ovenized, practically working and fully characterized (10V) LTZ reference here. So, please think again about "no effort, no risk".

It's clear that my low T.C. 10V reference is achieved by chance, but I never made any statement, how probable or improbable this is. I also did not state any generalization of my result.
Therefore, I don't value both your statements, either that the likelihood for a pure resistive divider is rather high to end up with > 1ppm/K, or this is almost (always) an outlier result. None of us has a statistical idea, or can present a test series, how the probability for a good T.C. matching is, for the different suppliers of such PWW or BMF resistors.

Anyhow, John Pickering used 2 EA of those TiN (?) resistor arrays, in an entangled schematic, to create a stable statistical divider. It's obviously very successful in series production, as they could guarantee a very good timely stability .. the T.C. is trimmed internally, but I assume, that the overall T.C. is already quite low as determined by the arrays themselves.

branadic made similar experiments with a single array, and trimmed the T.C. by means of a short piece of copper wire. The timely variation was quite high, afaik.
If I understood Andreas correctly, he achieved a very low T.C. with a single (?) array as well.
Therefore, this method is not that elaborated, but already approved for volt-nuts.

In the end, why I insist on simple, already validated solutions is the following:

I know Andreas lab on-site, and RAX also described his man's cave in detail. Both labs suffer from big temperature variations (+15°C / - 5°C, or so), therefore, to anyhow do ppm/sub ppm metrology, one needs ultra low T.C.s for the references. I mentioned my exchange with AB Precision @ 0.3ppm transfer level; both our basement labs might have different room temperatures, (in fact, exactly the same), therefore this requirement is as well mandatory.

Even with such very constant temperature conditions (+/-0.2°C) in my lab, the residual T.C. of any precision instrument can be quite annoying .. like the T.C. of a transfer instrument like the HP3458A. I promise to soon write an article about my recent T.C. measurements on the 03458-66509 reference board, and the overall T.C., and how this might disturb sub ppm transfers.

Frank   

[Kleinstein]

A small oven for a constant temperatre is not that difficult to build. It is quite common for very stable quartz oscillators.
However the oven can only solve the TC part and not the long term drift, that is often also important and maybe even the more important part.

Though no guaranteed by the specs, using a combination of resistors from a single resistor array can give quite good TC and also long term stability.
Combining resistors in series / parallel is a way to get the wanted ratio from of the shelf parts (e.g. 8 equal resistors as TDP, or NOMC). Statistical averaging to get better matching is a wellcome bonus and if needed a 2nd array is still affordable.

For the voltnuts some trim (e.g. with the extra copper wire) and try, test and if needed rework is not so bad as it is for commercial production.

[Rax]

RAX also described his man's cave in detail. Both labs suffer from big temperature variations (+15°C / - 5°C, or so)

Frank   

Frank - that must be Andreas', but in my case, there's nowhere near such wild variations. As I said before, during the spring through fall, my bench variations have been at worst +-2-3C (around a midpoint of around 25C). In a typical/moderate daily cycle (so no heatwaves, etc.), there's less than 1.5-2C delta. Southern California (at least where I'm at) doesn't typically get frost at all the entire winter. Our normal yearly variations are very small, comparatively, and therefore, we almost never need to use our HVAC system, year round.

Interestingly, the garage (my bench) is always the warmest room of the house (meaning, it never gets cold in there, even if the larger rooms in the house may be a little chill). This space always stays very even and comfortable. The one exception is late in the afternoon when the sun hits the garage door directly AND it's during a heatwave, then I have to turn on the heat pump I have in the garage. Not ideal (a portable unit), but it can shave off the additional heat when otherwise the garage would go beyond 27-28C.

I'll be able to extend my observations on my bench temperature into this winter, which is a season I haven't had such close observations as I did during the past spring through fall.

[Rax]

I am able to start populating a Frank board, sourced from OSH Park (Jason designed).

What would be some of the recommendations for building it? I am aware of the warnings on using too long a time on soldering leads of the LTZ1000 and the Z201s. I've been soldering from childhood, and also having (hand)made and built PCBs for about as long, am obsessively attentive to keep the soldering time as short as possible. I never do over 4-5 seconds at a time, and that's plenty enough for perfect soldering joints/points. I'm also considering using dissipating aluminum clips for each leg removing some heat during soldering to protect the part.

I'm also going to be careful with touching the board (I actually always solder with vinyl gloves). I'll probably wash the board to be populated with IPA before starting. Post soldering I always wash the board off rosin and all crude with IPA.

There's also a 3D print cover for the LTZ1000 on xDevs - have people here found that useful? I assume it'd remove any residual sensitivity to air drafts due to convection and whatnot.

[Jacques]

There's also a 3D print cover for the LTZ1000 on xDevs - have people here found that useful? I assume it'd remove any residual sensitivity to air drafts due to convection and whatnot.

In my experience adding a cover already made a measurable difference for the LM399, so I'd certainly add one with the LTZ1000 as well. I normally use 3D printed ones (e.g. nylon); maybe others here have also experimented with metal covers?

Edit: note that this was within a larger instrument, however.

[Dr. Frank]

I am able to start populating a Frank board, sourced from OSH Park (Jason designed).

What would be some of the recommendations for building it? I am aware of the warnings on using too long a time on soldering leads of the LTZ1000 and the Z201s. I've been soldering from childhood, and also having (hand)made and built PCBs for about as long, am obsessively attentive to keep the soldering time as short as possible. I never do over 4-5 seconds at a time, and that's plenty enough for perfect soldering joints/points. I'm also considering using dissipating aluminum clips for each leg removing some heat during soldering to protect the part.

I'm also going to be careful with touching the board (I actually always solder with vinyl gloves). I'll probably wash the board to be populated with IPA before starting. Post soldering I always wash the board off rosin and all crude with IPA.

There's also a 3D print cover for the LTZ1000 on xDevs - have people here found that useful? I assume it'd remove any residual sensitivity to air drafts due to convection and whatnot.

RAX, do you have self-clamping tweezers with copper plating?
It's most important to not heat the LTZ1000 via its legs, so I clamp such a tweezers on the top side, while I do a very short soldering on the bottom side, probably 1..2 sec. only. I also use these tweezers for the precision resistors.
Both the LTZ, as well bulk metal foil resistors show permanent hysteresis of several ppm, when getting heated or cooled down too much.

I still use leaded tin, btw.

I also do not shorten the legs of the LTZ, that helps to keep the chip cool, and the length of the legs goes into the T.C. budget.
My assembly is w/o a cap on the top side, as displayed in the pictures. I assume, this also goes into the T.C. budget, but I never experimented, what the impact of an additional cap would have on the overall T.C. W/o a cap, the whole thermally insulated assembly / circuit gets warm, and therefore .. the whole circuit is sort of thermalized.

There's also no cap on the soldering side, as my PCB design and mechanical assembly is intended to create an iso-thermal plane on the bottom side, as recommended in the datasheet.
That means, if you orientate the PCB horizontally, add a thin thermal insulation around the whole PCB, like in my pictures, then all solder junctions should remain on the very same temperature.
The assembly / PCB should never be tilted when in operation, as this would create temperature differences on the junctions, as well the LTZ chip itself is prone to tilting, which might create shifts of several ppm. This effect was found by Andreas. I even tilted my HP3458A, and have seen this effect, which is not mentioned at all in the specification, in contrary to specifications for HP OCXOs.

Frank

[antintedo]

I found that BU-34C copper alligators work well, very convenient to use on small legs. When designing a board from scratch it's worth elongating the pads, heavily pretinning them and adding flux before assembly, soldering time can be reduced to just half a second

[Rax]

I started putting this together.

Wanted to collect some input on using sockets for the op amps ("IC1" and "IC3"). I'd highly prefer them (honestly, I always use sockets... for ease of repair or upgrading), but maybe there's the risk of getting some parasitic EMF created between dissimilar alloys on the leads vs. socket etc. But maybe that just gets added to the overall voltages budgets, internal to the circuit.

[deepfryed]

You can get gold plated dual wipe DIP sockets, they're a bit more pricier but I've seen them used in some test jigs. Alternative is gold plated ZIF sockets, but I reckon everything additional would add some TE offset / noise.

[deepfryed]

heavily pretinning them and adding flux before assembly, soldering time can be reduced to just half a second

I've seen mentions of lower temperature cadmium tin solder in literature, anyone used it before ?

[Andreas]

Hello,

I've seen mentions of lower temperature cadmium tin solder in literature, anyone used it before ?

a) its highly toxic. (even the damps during soldering).
b) you have only advantage when soldering copper to copper. No advantage when soldering Kovar (LTZ1000) leads to copper (PCB).

with best regards

Andreas

[aronake]

heavily pretinning them and adding flux before assembly, soldering time can be reduced to just half a second

I've seen mentions of lower temperature cadmium tin solder in literature, anyone used it before ?

I got hold of a roll of 60% cadmium and 40% tin solder. Also found and bought a small bag each of laboratory grade pure cadmium and tin. Still not used it, and well aware it is toxic. Plan to do some experimentation with this on TEMF.

[MK]

heavily pretinning them and adding flux before assembly, soldering time can be reduced to just half a second

I've seen mentions of lower temperature cadmium tin solder in literature, anyone used it before ?

I got hold of a roll of 60% cadmium and 40% tin solder. Also found and bought a small bag each of laboratory grade pure cadmium and tin. Still not used it, and well aware it is toxic. Plan to do some experimentation with this on TEMF.

The fumes when you try and melt the cadmium into the tin are meant to be quite severe. Needs a LOT of care to do safely from what I have read.

[Rax]

Interesting progress here, and I think I figured out some useful steps through the process (thank you Frank!). I'll likely update the original post to include a simplified, clear step by step guide to put this together. Again, my goal here is to help others working on this after me so they'd have a streamlined guideline available (and stop bugging the veterans!). The below assumes the builder is located in the US.

Once one has the PCB at hand:

• Populate the board, excluding IC3 and associated parts
• Dial in R10 to provide temperature compensation
• Measure the exact output off the particular LTZ part used
• Calculate the gain ratio necessary for that particular value
• Order appropriate resistors from TCC. You can be extremely close to 10V due to the capacity to customize very high quality resistors.
• Kick back and relax for the few days it takes them to fabricate your custom resistors. Single malt recommended for the evening completing the steps above. Make sure your LTZ kit is running to start aging
• Populate the rest of the kit and slightly adjust the gain ratio ladder, if necessary, to hit the exact target value. I wager that, if R12/R13 are judiciously chosen per LTZ output, R18/R15 can do all the rest of the work to sub-ppm levels.

Example: My LTZ outputs 7.159V. I started with a 10k/4k pair for R12/R13, but this landed me too high (10.0226V). If I instead get an 10.1k Z201, that'll land me at Vout = 9.99425V without any additional resistance on top of that "ladder." With a 20 ohm pot, I get a top of adjustable range of 10.02263V. An 10 ohm R18 makes that adjustable down to 9.99897.

So, with the values above, one should be able to adjust between 9.99897V and 10.02263V. A more narrow selection of R12 and R13 should allow for an extremely narrow trim range (low ppm or sub-ppm).

Now I have to figure out dialing in R10 - welcome recommendations - but I assume it involves running days-long logs of the LTZ output with R10 at different values between 200k and 1M and the one that returns the flattest curve on the largest delta T wins the big prize.

[Rax]

I'd be very interested in hearing about some methodologies employed by others to tweak the TC resistor (R10, 400k) when using the LTZ1000 version of the reference.

I've started by letting the kit run - so I'd also age it as much as I can - and logging 8.5 digit measurements (20s integration) over 24hrs (this far) on my Prema 6048. I guess 6.5 digits would probably be enough - and maybe a different type of reference would provide more reassurance of dissimilar tempco for relevant measurements - but why not have the additional granularity. I'm planning to do the same with different values for R10, and see which value seems to exhibit the lowest Vout variation per 1C over random deltas during the days each individual value is examined.

I've searched the relevant older threads to nausea but didn't come up with a clear and simple methodology. I've also looked at Frank's recent "3458A-related" series of articles (which obviously have an "A-version" at their center) in the main LTZ thread, but I'm looking for a simpler approach and practical points to set it up. If anyone has a link to such, please share. Many thanks in advance.

[Rax]

Here's a roughly 18hrs plot with R10 = 400.1k. I seem to be getting .694ppm/C for this R10 value. I not quite ure what is to be expected from this at this level of build, but running different values will certainly tell me.

Please note the Prema is currently running with upgrades that negate its calibration (so measurement values are probably about 7ppm high). The kit is not in an enclosure, but in free air. I think I'm OK with both those determinations, as they stretch a bit the environmental conditions which is conducive to the test.

[Smokey]

I'll continue posting here as I find more similar data.

My intention with this "new" LTZ1000 10V reference build thread is to make it as useful and as simple and as I can, organize as much as possible information accrued trough the efforts here (which should be for the benefit of others down the line from me), and to introduce new information where possible.

For instance, I don't think information on these "long term rated" PRC resistors was presented here before (anywhere on the Metrology section).

It would be great if you could keep an "index" in the first post of where important/interesting information is in this thread.  As you already found in the original LTZ1000 thread, it's a pain to try to find anything once a thread gets over a certain size.

[Rax]

It would be great if you could keep an "index" in the first post of where important/interesting information is in this thread.

That's a great point. I was planning to add the findings to the initial post - didn't think to have a second placeholder post as I maybe should have - or if I really get motivated write it up in a separate document, but links would probably work better (more true to the form of a forum and how these threads grow).

[Kleinstein]

The curve shown does not look like it can make a good test for the TC. The variations look rather fast and thus unsure if the temperature is stable.
Anyway the reference should be mounted in a case, as the TC correcton with R10 depends on the thermal setup. It uses the heater current to correct for the residual temp effect. Additional insulation can change things and the most suitable value for R10.

[Rax]

As you already found in the original LTZ1000 thread, it's a pain to try to find anything once a thread gets over a certain size.

I'm finding it just about impossible to find anything using the forum's search capability (within some limits). There was a recommendation somewhere to use google to search inside the forum instead (to bypass its own lackluster search engine), which some people have had success with. But I'm not having much more luck with that either.

The more massive the thread, the less chances awesome nuggets harbored deep inside the thread will emerge, is my experience. Or, really, any (useful) nuggets.

[Rax]

Before I put together a suitable tempco plotting rig (involving "beer boxes," per Frank... Which - yeah, go ahead and judge me!  - I have plenty of in different sizes. The only difficulty is which will I sacrifice?...) I'm going ahead with a LTZ1000A variety build.

There's one pick I'd appreciate some input on. R2 and R3 only have in the other of 1/1000 to 1/3000 contribution to the overall tempco, so not nearly as demanding as R1, R4, R5, R12, R13. It also seems the value is not critical, but I'm not sure where things get stretched too much. The datasheet seems to show in different places 30k, in (most others) 70k.

What I have at hand are:

• 50k PTF56 (10ppm TC)
• 40k VHS102 (2ppm TC, I think)

I didn't find any examples in the original thread of anyone going below 50k, but maybe there is and/or I missed it (but if the case, I'll be happy to use those VHS102s, which I have plenty of at hand).

[Kleinstein]

The value for the 70 K resistors is not that critical. However there is still a balance: a lower value gives marginally less noise (more current at the transistor), but also a slight larger TC without regulation and thus more importance for the set point. I both cases the difference should not be very large.

The 50 K PTF50 should be OK, good enough from the TC and drift, though there may be better choice in SMD form factor.
The VHS102 would be overkill and no real need for these in this application.

[Rax]

Please see modified first post - I restructured it to index/organize the information contained in this thread. A wiki of sorts.

[Rax]

After some aging of my current two kits:

• LTZ1000 (7.159V) kit: R12 = 10k; R13 = 3.95k
• LTZ1000A (7.187V) kit: R12 = 10.25k; R13 = 4k

Hence assuming the builder uses TCC to source custom fully specified resistors, given some calculations - once the stable output of the LTZ is known - ordering R12 and R13 discriminately removes the need for R14. Both the kits above can comfortably adjust to less than 1ppm of 10V without any need for a trim by R14. For the current modest price of about $40 for the two resistors, one can almost dial their LTZ kit to 8.5 digits worth of 10V, precisely. Honestly, one less anal than I am would be perfectly happy with just populating R12/R13, shunting everything else, and calling it a day.

I am using 20 ohms for R18 and a 20 ohm adjustable R15. The last fraction of the ppm becomes uber fine 1/4 or 1/16th of a rotation on R15.

[Rax]

For the current modest price of about $40 for the two resistors, one can almost dial their LTZ kit to 8.5 digits worth of 10V, precisely.

This is in Z201 flavor, which - hedging some criticism this spec is actually real - should provide .2ppm of TCC, which is obviously top notch. Third best to Fluke's legendary 000 TCC resistors and an "arm and a leg" VHP.

[Rax]

Slowly progressing with this while I let the two kits I built already age a bit.

I should probably put the kit using the LTZ1000A - which is my "primary" - in an insulated box (see pics for my following points). I'm planning to use copper tape for shielding. Not quite sure what would be the correct connecting of this shield - GND_PWR? Then the 10V output would use the gold-plated bananas, and the power supply would connect to the basic Pomona bananas.

Welcoming comments on these choices.

[Kleinstein]

Thermal isolation is a 2 sided thing.  It can attenuate fast temperature variations and with the active heater also reduce the overall temperature variations also for outside the regulated part. However it also leads to an high overall temperature and thus way more aging for parts like the resistors. The oven temperature at the LTZ1000 also needs a slightly higher value to still get regulation.  A temperature somewhat (e.g. 10 K to 20 K)  higher than room temperature can be good as this also reduces the humidity (the RH goes down by a factor or 2 for about every 10 K of temperature rise).

For the referenes the variable heat sources like the voltage regulators should ideally be outside the termal isolation, at least well away from the actual reference circuit to not effect gradients very much.

[Andreas]

Not quite sure what would be the correct connecting of this shield - GND_PWR?

Hello,

ask 3 different people and you will get 4 different opinions.
Effectively also the power supply and the wiring play a big role.
And without additional measures (filtering) the foil will not bring much improvement.

It would be better to have a thick walled metal housing which would improve the thermal equalisation around the reference.

In my references (with batteries within the housing) the outer metal shield can be used as guard.

For the referenes the variable heat sources like the voltage regulators should ideally be outside the termal isolation, at least well away from the actual reference circuit to not effect gradients very much.

This is true (at least for battery supplied equipment where the input voltage changes with charge state).

But I would not fear falling out of regulation.
I only had one LTZ out of 9 (with extremely low output voltage of 7.11V) where above 32 deg C environment there was a non linearity in T.C.
I simply set the temperature setpoint somewhat higher to fix it for my standard temperature range of 10-40 deg C

with best regards

Andreas

[Rax]

ask 3 different people and you will get 4 different opinions.
Effectively also the power supply and the wiring play a big role
Andreas

Good point, Andreas. I haven't explained my plan for supply this with power.

The second, nickel-plated set of banana posts would be for a simple two wire connection to one of my HP power supplies. I don't think I want this to have its own provisions (regulator inside its case, etc.), given it will be mostly off and tucked away, traveling occasionally, and if I wanted on for a long time, the HP supplies will be fine doing it.

Unless there a strong recommendation to provide regulation (and maybe further filtering, etc.) inside the case.

And without additional measures (filtering) the foil will not bring much improvement.

It would be better to have a thick walled metal housing which would improve the thermal equalisation around the reference.

Andreas

If the only advantage to the thick walled metal housing is thermal equalization (as it sounds like you're suggesting shielding is difficult to achieve without additional measures), I assume a similar result would be achieved by insulating the plastic box I was planning for this - wouldn't it?

I simply set the temperature setpoint somewhat higher to fix it for my standard temperature range of 10-40 deg C

with best regards

Andreas

This is something I also have to account for, Andreas, as though I don't quite have your fluctuations, I don't have metrology lab conditions either.

[Rax]

After about 3.5 months of aging, it seems the LTZ1000 kit (not properly temp compensated - btw, I see no useful path here, given the difficulties, additional work, and investment in additional gear to accomplish the tweaking of R10) has gone up by about 1.5ppm, while the LTZ1000A kit by about 4ppm. This doesn't factor in temperature conditions data, etc. - it's essentially ballpark. I don't think I care for logging this "pre-aging" phase of these kits; I think the useful logging starts about now.

So I'm readjusting them to as close to 10V as I can bring them (per Prema 6048), and then they soon may travel to a bench that can "17025 them."

[Dr. Frank]

The effort to trim R10 to T.C. is surprisingly low, but worth the time, and in your case of a lab with strong temperature changes, urgently necessary.
Otherwise, you will - again - get no fix point in your lab.

You only need a short term stable other reference, which you compare differentially.
The temperature on the DUT has to be changed rapidly, relative to the slow and lower temperature change of the room temperature for your other reference.

Then you can easily separate the T.C. of DUT vs. the other reference.

Frank

[Rax]

Maybe there's a finer point you're making here, Frank, but I thought the LTZ1000A equals or exceeds the thermal performance of the compensated LTZ1000, because it doesn't need the TC trimming resistor?

[...] 200k resistor , which should NOT be used on the A version of the LTZ1000, as it creates an unwanted T.C. of about 0.3ppm/K. W/o that resistor, it probably would have near zero T.C.

Or your samples and tests showed that not to be the case?

To state that differently, is the LTZ1000 w/ proper temperature compensation (by selected R10) more temperature-stable than the LTZ1000A (obviously) without?

You made the point elsewhere that one of the advantages of using the basic LTZ1000 is it can be run at lower temperatures, and therefore has lesser long term drift. I took note of that, but that's completely separate from this tempco consideration.

[Andreas]

Hello,

its all in the LTZ1000 thread.
T.C. depends on the 400K* resistor and lead length of the reference.
And most probably also on thermal isolation and temperature setpoint.

with best regards

Andreas

[Dr. Frank]

Maybe there's a finer point you're making here, Frank, but I thought the LTZ1000A equals or exceeds the thermal performance of the compensated LTZ1000, because it doesn't need the TC trimming resistor?

[...] 200k resistor , which should NOT be used on the A version of the LTZ1000, as it creates an unwanted T.C. of about 0.3ppm/K. W/o that resistor, it probably would have near zero T.C.

Or your samples and tests showed that not to be the case?

To state that differently, is the LTZ1000 w/ proper temperature compensation (by selected R10) more temperature-stable than the LTZ1000A (obviously) without?

You made the point elsewhere that one of the advantages of using the basic LTZ1000 is it can be run at lower temperatures, and therefore has lesser long term drift. I took note of that, but that's completely separate from this tempco consideration.

I only have two samples of A version references, and both were near zero TC w/o that R10.
I have no clue, why or if it is really a general characteristics of the A version.. no such detailed info from others, to my knowledge.

Anyhow, you need to check the individual TC, to make sure that you really got a reliable reference, over temperature.

As the LTZ1000A heats itself 10°C more than the non-A version, oven temperature has to be 10°C higher, under same environmental boundary conditions.
That means, if you run your reference between 18...28°C normal lab range, then you can as well use 50..55°C oven temperature for your A version.
I have tested my LTZ1000A assembly to work up to 32°C (or so), until the oven regulation freaks out.

The calculation is: max. environmental temperature + 7°C (enclosure heating) +10°C (self heating A version) +5°C (regulation margin) = minimum oven temperature.

This R10 does not affect the regulation stability, nor the overall stability, to my experience.

This check for zero TC can easily be done with a 2 temperature point measurement, i.e. you might cool, or heat the DUT by +/- 5°C, and then track the behaviour over the slow temperature relaxation.

Frank

[Rax]

Hello,

its all in the LTZ1000 thread.
T.C. depends on the 400K* resistor and lead length of the reference.
And most probably also on thermal isolation and temperature setpoint.

with best regards

Andreas

Andreas,
Thank you, but that wasn't at all my question.

I am not planning to use LTZ1000 parts, only LTZ1000A. So the whole conversation over the 400k resistor is N/A, except to the extent of whether the properly compensated part (LTZ1000) is more temperature-stable than the un-compensated (because it doesn't need temperature compensation) LTZ1000A.

[Rax]

As the LTZ1000A heats itself 10°C more than the non-A version, oven temperature has to be 10°C higher, under same environmental boundary conditions.
That means, if you run your reference between 18...28°C normal lab range, then you can as well use 50..55°C oven temperature for your A version.

This is another reason I prefer the LTZ1000A. Having it working at a higher temperature than the other part I think should make it a bit more tolerant to the occasionally higher temperature of my environment. Even so, keep in mind that, even if my lab is not temperature controlled 24/7, I do have heating and cooling in it which I can perform for sensitive measurements. These not being kits needing to be run 24/7 - such as a 732A/B would be - I can regulate my lab at a certain temperature, then turn on the kits and use them. Probably just make sure to build enough time in doing that for the kits to acclimate themselves.

I am using a 1k/12.5k ratio for the oven resistors, which should make the temperature about 55-60C per Frank. Per formula presented, that should provide sufficient margin for environmental conditions up to 33-38C, which is plenty more than I need.

[Kleinstein]

Well trimmed with R10 the non A version can reach near zero linear TC. Over a small temperature range it can this way be more temperature stable than the A version. However this can only trim the linear TC and not higher order terms (specially T² part). The compensation also adds additional 2nd oder TC as the compensation adds something proportion to about the square root of the heater power.

[dietert1]

Once more: The A-Version has less temperature margin for its oven due to more self heating of its chip inside the metal case. For a given ambient temperature span its oven temperature must be configured higher - a disadvantage, when low long term aging is desired. Then again, in this case ambient includes some plastic covers and so on.

At a given more or less constant heat resistance defined by the mechanical setup and LTZ variant selected the final oven power is about linear with ambient temperature (with a negative slope). So R10 is a way to add a nonlinear compensation (square root). The way to add a near linear compensation is a small resistor in series with the zener. This tunes the Ube temperature compensation by adding some gain. Usually a 15 to 20 Ohm resistor serves well. One can combine both methods and tune linear and first nonlinear term of TC(T). As far as i remember the F7000 schematic demonstrates that.

Regards, Dieter

[Kleinstein]

The resistor in series to the Zener adds some liner TC, however this is linear TC for the part inside the oven, not for external temperature changes.
The compensation with R10 is compensation for changes in the external temperature  or heater power that is related.

For a 10 V reference there can also be temperature effect from the 7 to 10 V gain stage, if this part is not termally stabilized. This would also apply the LTZ1000A version of cause.
One may get away with a single compensation for the 10 V TC, not necessay via R10 at the LTZ circuit, but with something like selecting suitable TC resistors with a fine trim.

[Rax]

For a given ambient temperature span its oven temperature must be configured higher - a disadvantage, when low long term aging is desired.
Regards, Dieter

I've started figuring out, I think, some ways to have yearly, seasonally, or more frequent traceable checks, so I think low long terms aging is less of a concern. I'm looking for a feasible, reasonable project with this, and so I think I just have to strike the LTZ1000 temp comp side project. In the "grand" scheme of things, that one alone is a pretty invested little project.

[Rax]

The way to add a near linear compensation is a small resistor in series with the zener.

I get it, but that's not a part available in the kit/PCB I'm using (Frank's), so it's more of a distraction from my vantage point.

[dietert1]

If one gets/makes an oven for TC determination one can later use it to run the reference(s) inside - at a constant temperature. So the investment isn't lost. Also: Using an outer oven at constant temperature may be as good and easier than the TC adjustment.

Regards, Dieter

[Rax]

If one gets/makes an oven for TC determination one can later use it to run the reference(s) inside - at a constant temperature. So the investment isn't lost. Also: Using an outer oven at constant temperature may be as good and easier than the TC adjustment.

Regards, Dieter

This perspective is indeed very pertinent (and in my mind, and in some ways, the opposite of dialing in by "400k resistor tweaking" an LTZ1000 kit, which is intended to make it nearly agnostic of environmental variations). As I was researching how to do the latter, I've came to consider getting a small incubator. But beyond that use, this would also allow me to strictly control the temperature the kit is "seeing," and so running all LTZ1000(A) references inside such an environment whenever they need to deliver their "output of calibration" I think is a very good idea.

And yet another option I am still considering is having some sort of heating provisions for the inside of the kit(s) enclosure(s). There's already a thermistor provided on the board, and running some sort of heater at constant/controlled temperature may not add a lot to the project (but rather back burner on my docket).