Geller voltage reference schematic ??

[Overspeed]

Hello

As Mister Geller is out of business

Does that possible to locate a schematic of the Geller Voltage Reference ? I have a picture from this forum ( linked ) but not the BOM

I dont want to copy just to copy I just want to progress and get new PCB to built some .

Question : As the original is through-holes components I will upgrade with surface mounted when possible

Regards
OS

[BU508A]

BOM:

Parts List (from the website):

AD587 chip, e.g. AD587KN DIGIKEY AD587KN-ND (the AD587LN was discontinued), or AD586 series chip for a 5 V board
C062K104K1X5CA MOUSER 80-CK06BX104K
ECQ-V1H104JL Digi-Key P4525-ND
3299W-1-201 MOUSER 652-3299W-1-201 (200 ohms, use 1 k ohms when a wide selection of select and test fixed trim R's are not available)
TRIM R IRC 0.1% MOUSER 66-RC55LF-D-3.24K (for some 587 production runs: only if using a 200 ohm trimmer, 2.87K or 3.01K, see manual)
TRIM R IRC 0.1% MOUSER 66-RC55LF-D-7.32K (see manual)
COPPER LOOPS #18 SOLID COPPER WIRE
2 PIN IDC HEADER
IDC SHORTING PLUG
Mill-Max SOCKT MOUSER 575-193308
solder

Please see attached .pdf for the schematics

[Andreas]

Question : As the original is through-holes components I will upgrade with surface mounted when possible

Dont do that: the wires give a certain decoupling from PCB (and swelling of PCB by humidity)
For the same reason you should use a CERDIP8-Device for the AD587 (e.g. AD587JQ/AD587KQ)
otherwise the drift will be in the range of 0.2-0.5 ppm/% rH.

The SVR(-T) are missing a 100 nF capacitor between the output pins to reduce EMI influences.

The enhanced SVR-T Version from Lars is included in my AD587LW design.

https://www.eevblog.com/forum/metrology/ad587lw-10v-precision-travel-standard/msg1449488/#msg1449488

with best regards

Andreas

[HighVoltage]

A few years ago, we had some black edition boards produced.
I might still have a few ...

https://www.eevblog.com/forum/metrology/geller-labs-svr-lars-walenius-black-edition-with-587lq/

[Overspeed]

Hello

Thanks for your quick and detailed answer

My idea is to fit the PCB underneath the box cover as carry cot , no wire but studs / through the cover with ring seal / potting ( see picture  as possible other solution  rod are in thread brass so no leak between insulation and wire core ) and use a sealed enclosure with a Nitrogen valve very close to an ATEX enclosure like that

https://www.atexdelvalle.com/fr/coffrets-antideflagrants-atex/junction-boxes

I can make one easy in the right dims as 100 x 75 x 50 or make it easy in 6062 / 5083 alloy or even with a 4 inch square pipe a welded plate and a cover with a quality seal I can also make them from a block of  PEHD plastic

Question : The missing cap 100 nF on the schematic is between 5 and 6 pin ? a ceramic XR7 is enough ?

Regards
OS


 

[Conrad Hoffman]

Test your epoxy on that feedthru. Some filled epoxies are more conductive than you might think. A fun example is J.B. Weld. It's "non-conductive" in bulk, but a thin layer between two plates will be very conductive because the iron particles bridge the gap. Some conductive epoxies use the same principle to connect flex circuits to card fingers and not short out the fingers. Other epoxies have carbon and don't give the very high ohm readings you might like.

[Overspeed]

Test your epoxy on that feedthru. Some filled epoxies are more conductive than you might think. A fun example is J.B. Weld. It's "non-conductive" in bulk, but a thin layer between two plates will be very conductive because the iron particles bridge the gap. Some conductive epoxies use the same principle to connect flex circuits to card fingers and not short out the fingers. Other epoxies have carbon and don't give the very high ohm readings you might like.

Hello

Thanks for the advice , I agree for JB weld that not a true miracle product and most 2 component repair putty are loaded with micro aluminium powder or stainless steel 

I use this one

https://www.tedpella.com/technote_html/891-60%20TN.pdf

for low pressure / inert atmosphere I use also high temp cement as

https://www.final-materials.com/gb/232-high-temperature-ceramic-adhesive

for copper housing or brass housing / copper wire with reduced gag ( 1 mm ) Loctite 648 is wonderful as copper react with Loctite to generate a very strong glue able to work at 80 / 100 deg C safe only concern is that quite liquid so to request a plug / center to quick the Loctite in the hole

Avoid any silicone product as silicone react with copper under current to generate dark pollution on any contacts that hell . even some wire succeed to pollute enclosure

I will post a cutaway of my proposed design before Monday I have some files to send tomorrow I am late on a customer project but built a clever true precision and stable Voltage Reference with easily available components is very interesting .

Regards
OS

[CalibrationGuy]

Hello, I've engineered many projects, and sometimes a simple answer to a problem can get you 95% of the way. I have not tested this idea in this particular application, but maybe it will work for you. There is a great (and inexpensive) product, molecular seive dessicants. Basically, a metallic oxide that removes humidity from the air. We use this all the time to dry out our dewpoint sensors. Anyway, back to my idea.

You take the board with the reference and put it in a small waterproof plastic enclosure. Place a few beads (maybe a teaspoon) wrapped in some plastic mesh (so they don't bounce around) inside the enclosure with the electronics. Then, feed your wires through a couple of holes in the enclosure and seal the holes with epoxy. Voila, the beads will absorb the moisture and the reference will stay very dry for a very long time. Put the whole thing in a larger box with your low emf binding posts and power supply and you're good to go.

Maybe this idea is idiotic, who knows. I know that it will be cheap to make and test, so why not give it a shot?

Thoughts anyone?

TomG.

[Conrad Hoffman]

Molecular sieve works great, but there are several types. I'm thinking 4A, but do some research on which works well with water-sized molecules. Bake the stuff out at high temperature before using and it should keep things dry for a long time.

[CalibrationGuy]

Molecular sieve works great, but there are several types. I'm thinking 4A, but do some research on which works well with water-sized molecules. Bake the stuff out at high temperature before using and it should keep things dry for a long time.

Yes, we've used this product and were able to get -67 DegC dryness with our dewpoint sensors (one teaspoon dessicant in small 3/4" diameter tube) without a bakeout. Here's a plain link: https://www.amazon.com/dp/B09QGM9X7G (no affiliation with Amazon OR seller).

TomG.

[Overspeed]

Hello, I've engineered many projects, and sometimes a simple answer to a problem can get you 95% of the way. I have not tested this idea in this particular application, but maybe it will work for you. There is a great (and inexpensive) product, molecular seive desiccants. Basically, a metallic oxide that removes humidity from the air. We use this all the time to dry out our dewpoint sensors. Anyway, back to my idea.

You take the board with the reference and put it in a small waterproof plastic enclosure. Place a few beads (maybe a teaspoon) wrapped in some plastic mesh (so they don't bounce around) inside the enclosure with the electronics. Then, feed your wires through a couple of holes in the enclosure and seal the holes with epoxy. Voila, the beads will absorb the moisture and the reference will stay very dry for a very long time. Put the whole thing in a larger box with your low emf binding posts and power supply and you're good to go.

Maybe this idea is idiotic, who knows. I know that it will be cheap to make and test, so why not give it a shot?

Thoughts anyone?

TomG.

Hello

Thanks for the product link
To limit humidity contamination a sealed enclosure is a need OR to install a humidity gage

I will post a first design with an optional inside container for desiccants

Regards
OS

[CalibrationGuy]

Hello, I am very eager to see your results.

TomG.

[CatalinaWOW]

Desiccant works, but "very long time" is a fuzzy thing.  Achieving times measured in years requires very good seals and an appropriate ratio of desiccant weight to protected volume. 

The sealing part is hard and includes thinking about diffusion rates through the sealing materials.

Assuring that the contents of your box are initially dry also helps (baking, vacuum etc.) to avoid using up all of the desiccant capacity doing the initial dry out.

[Overspeed]

Desiccant works, but "very long time" is a fuzzy thing.  Achieving times measured in years requires very good seals and an appropriate ratio of desiccant weight to protected volume. 

The sealing part is hard and includes thinking about diffusion rates through the sealing materials.

Assuring that the contents of your box are initially dry also helps (baking, vacuum etc.) to avoid using up all of the desiccant capacity doing the initial dry out.

Hello

I agree that the primary condition is to be 100 % that the enclosure the PCB and all the parts are dry , that not so easy , and after enclosure closing to be sure that the seals are perfect , that why on military equipment the is humidity indicator

https://brownell.co.uk/products/humidity-indicators/r-f-screened-for-equipment-and-enclosures.html

Regards
OS

[CalibrationGuy]

Desiccant works, but "very long time" is a fuzzy thing.  Achieving times measured in years requires very good seals and an appropriate ratio of desiccant weight to protected volume. 

The sealing part is hard and includes thinking about diffusion rates through the sealing materials.

Assuring that the contents of your box are initially dry also helps (baking, vacuum etc.) to avoid using up all of the desiccant capacity doing the initial dry out.

A valid point, yes. Keep in mind that this dessicant is high performance, and you need just a little to accomplish your goal. Also, even with parts/enclosures that are not dry, humidity stabilization will occur within a few days. We have tested this, it's not just conjecture or speculation. Of course, the only way to see what happens is to try it out.

TomG.

[CalibrationGuy]

Maybe this will help. Here is an inexpensive humidity sensor that you can use to validate your prototype:

https://www.adafruit.com/product/4566

It will monitor humidity and it's only $5!

Have fun!

TomG.

[Gyro]

It's a shame the supply of NOS former Soviet hollow hermetic glass seals has dried up. They were ideal for running copper wires into sealed enclosures with no interruptions.

[Overspeed]

It's a shame the supply of NOS former Soviet hollow hermetic glass seals has dried up. They were ideal for running copper wires into sealed enclosures with no interruptions.

Hello

This kind of products is made  in USA from companies as https://www.completehermetics.com/hermetic-products-and-services/glass-to-metal-seals-and-feedthroughs

problem is the price $$$$$ as new and even on Ebay

Can be done with SS304 1/8 or 1/4 made in china and epoxy ... no so pretty but OK for a DIY or hobby purpose even  for a professional use as that a part for a show ...

I link my in process design with two valve one for flush on for nitrogen filling and a '' housing '' for desiccant 

Regards
OS

[Conrad Hoffman]

Tap blind holes from the inside- no reason to pass the screws to the outside. Use a single valve. Either evacuate and back-fill or run a small tube through the valve and blow the nitrogen through that, with the outflow around the tube. Or, ditch the valve entirely and just quickly install a pipe plug with Teflon tape. Put it in a plastic bag if you're really paranoid about getting the plug in place. The problem is you don't know if the finished device is leak free. High tech is you helium leak check it, then fill and solder off the fill tube. Another crazy thought, if you can do some glass-blowing, is to make a sealed large diameter glass tube with feedthrus, install everything, then neck it down and do the fill. Makes repairs a bit of a PITA though!

[Alex Nikitin]

Isn't it much simpler just to use a hermetically packaged reference if a good long term stability and insensitivity to humidity changes are needed?

Cheers

Alex

[CalibrationGuy]

Isn't it much simpler just to use a hermetically packaged reference if a good long term stability and insensitivity to humidity changes are needed?

Cheers

Alex

Yes. Just not for the rest of the circuit.

[Overspeed]

Isn't it much simpler just to use a hermetically packaged reference if a good long term stability and insensitivity to humidity changes are needed?

Cheers

Alex

Hello
Yes and no
as
Yes for the IC
No for all other connections contacts or components as copper and humidity is a very bad contact
I plan to make tests so I will use first plastic IC as the quite cheap and easily available 

I am not paranoiac but I usually overkill problem when bomb proof solutions are cheap or available in the warehouse as surplus
I have no problem to machine a sealed container and flush it with Nitrogen
I am very upset regarding humidity trapped and that very difficult to get full dried parts and install them in an enclosure
Nitrogen is nor always Nitrogen
https://nigen.com/nitrogen-gas-purity-grade-specification-industrial-medical-food/#SnippetTab
So if that not OFN that does solve 100 % of the problem

I will post a detailed drawing of my DIY epoxy sealed pass trough in 1/4 inch with BOM

I agree for blind hole to avoid to pass through the cover

Regards
OS

[dietert1]

Others have solved that before. For example https://www.eevblog.com/forum/metrology/experiments-with-vref-ovens/msg5486191/#msg5486191. That is a flat cable with 12 signals and with an epoxy seal. The seal is thin and covered with aluminum at least 8 mm in all directions. The aluminum cover also serves as a heat spreader to reduce thermal gradients. Capacitance of each signal line to the aluminum enclosure is about 40 pF. That aluminum enclosure is a standard trade product. It has a rubber seal to make it (almost) hermetic. With a little bag of desiccant inside, relative humidity will be 20 % or less inside if you wait a little.

When i built a voltage reference based on the Geller board in 2009, i put it into a simple thermostat as i did not have the patience for TC correction with thermistor networks.

Regards, Dieter

[Gyro]

...
I am not paranoiac but I usually overkill problem when bomb proof solutions are cheap or available in the warehouse as surplus
I have no problem to machine a sealed container and flush it with Nitrogen
I am very upset regarding humidity trapped and that very difficult to get full dried parts and install them in an enclosure
Nitrogen is nor always Nitrogen
https://nigen.com/nitrogen-gas-purity-grade-specification-industrial-medical-food/#SnippetTab
So if that not OFN that does solve 100 % of the problem

If you're going to this trouble, why not use something completely non-reactive like Argon.

[Conrad Hoffman]

Argon has different thermal properties, which you may like better or worse. Just FYI.