first lm399 10v reference

[Svgeesus]

If you really want to use a trimmer (which I would avoid, see comment from TIN above), the next step is to replace one resistor by another resistor and a trimmer in series.

A series trimmer requires a very low value (the imbalance between a pair of 10k 0.01% resistors is just 1 ohm), which has a looser tolerance and depends more on the end-stop resistance. From what I have learned reading the Metrology section, a high value parallel trimmer is a much better arrangement.

I have seen the wirewound trimmers from Bourns recommended, but from the datasheet the bulk metal foil Vishay ones seem superior.

[cellularmitosis]

If you really want to use a trimmer (which I would avoid, see comment from TIN above), the next step is to replace one resistor by another resistor and a trimmer in series.

A series trimmer requires a very low value (the imbalance between a pair of 10k 0.01% resistors is just 1 ohm), which has a looser tolerance and depends more on the end-stop resistance. From what I have learned reading the Metrology section, a high value parallel trimmer is a much better arrangement.

I have seen the wirewound trimmers from Bourns recommended, but from the datasheet the bulk metal foil Vishay ones seem superior.

Here's a scheme used to trim a resistance standard (from https://www.eevblog.com/forum/metrology/teardown-standard-resistors/msg891673/#msg891673 ).

[kj7e]

The scheme above is very similar to what I used on my 7 -> 10v buffer.  The TC the ~20 Ohm trimmer section is attenuated by nearly 500 times and this gives a very precise narrow band of adjustment  19.052-19.719.

[try]

Hello Svgeesus,

sorry I did not express this precisely enough:

It's no rocket science to shunt one leg and approach the target output voltage.

If you really want to use a trimmer (which I would avoid, see comment from TIN above), the next step is to replace one resistor by another resistor and a trimmer in series.

Regards
try

Added: excel screenshot

You calculate yourself a divider (R1 and R2 in series). Next thing is to shunt one leg of the divider by installing R3.
Depending on the intended precision you could place other bigger resistors R3', R3'' ... in parallel to R3.
You could as well extend R3 by putting another resistor R4 in series to R3, maybe another.
Countless variations are possible.

At some point you exchange comfort against stability by replacing any resistor (neither R1 nor R2) in your resistor by a smaller resistor and a trimmer in line to each other.

Look at the scheme from cellularmitosis:
Imagine R3 was 1k Ohm before. Now it has been replaced by 400 Ohm and a trimmer with 1 k Ohm.
Around trimmer mid point (500 Ohm) you roughly hit the 1k (~900)

Tighter version:
Make R3 910 Ohm (1k||10k) and use a trimmer rated 200 Ohm.

[Svgeesus]

Thanks, sorry to have missed your point before. I see what you mean and understand the approach.

I do need some way to apply small amounts of trim over the coming years, after initial setup.

[niner_007]

I'm working on a new cutout for the reference

[niner_007]

Hello

I have made some updates. For now, I won't focus on getting an accurate 10.00000V, but rather a stable output. I'm using the 8 pin version of LTC2057, I'm unsure about this, there's not enough space for guard rings around the IN+ and IN- opamp inputs, as indicated in the datasheet.

[niner_007]

with the guard traces

[niner_007]

here is another update, back to LT1001, added bypass capacitors (ceramic for input and heater; film for zener output, amp supply, and feedback loop), now using a precision resistor network, with 0.5ppm tracking, for the feedback loop

[TiN]

I'd think you have more progress if you actually build the circuit and play with it .