Looking for sample hold circuit (not for audio)

[Kevin Mitchell]

I'm trying to figure out the best analog solution for grabbing and saving a voltage as a reverence with a push button. Ideally each button press would refresh the saved/held voltage.

I have a few ideas but not without taking circuits I'm familiar with and drastically altering them. Thought there might be a more common way to achieve what I'm trying to do.

Any ideas? Help? 

-KM

[strungout]

Revere the All Mighty Voltage! 

[Kevin Mitchell]

Lol!

Maybe I could use a Jfet, incoming voltage on the source and a latching circuit on the gate?

Haven't really experimented with jfets yet.

-KM

[PRR]

https://www.analog.com/media/en/training-seminars/tutorials/MT-090.pdf

[Mark Hammer]

How long do you need that voltage to be held?

Would what you need to accomplish lend itself to digital encoding?

[Kevin Mitchell]

Thanks, PRR! I'll dig in and see if anything sticks 

Mark, I'm trying to avoid digital for this. I know I can easily do it through programming but I'm trying to come up with a very simple circuit - for tinkering reasons for now.

I'd prefer the sampled voltage to exist indefinitely - or until the source changes or the circuit disconnects from power. But anything above a minute shouldn't be necessary so I'll take what I can get for now.

Sorry I'm trying to not explain my intentions with this one as I have a few
But I will say it's a single supply circuit and I need to refer back to a voltage that only existed temporarily or intermittently.

-KM

[anotherjim]

A digital thing doesn't have to require a CPU. Standalone parallel ADC and DAC with a latch between them make an infinite sample & hold (so long as the power stays on).
But I suppose a FET - CAP - HiZ buffer is simpler if you can get the drift good enough.
It might be more mechanical...

Those switches could be a single sprung centre-off SPDT. A high Z amp (TL072) and done?
https://www.eecs.tufts.edu/~dsculley/tutorial/opamps/opamps7.html

[Mark Hammer]

Which begs the question: what is the least leaky capacitor material/type?

[anotherjim]

Which begs the question: what is the least leaky capacitor material/type?

It can be difficult to understand how any of the non-polarized dielectrics can be any worse than the surroundings and buffer input leakage. This article might help...
https://www.sciencedirect.com/topics/engineering/hold-capacitor

Not sure if it would need a reset/clear function, but the ability to force it to some preset voltage other than 0v might be useful.

[Eb7+9]

you might be able to get away with just a few passives ...



... not for audio ?!

I think we can fix that ...

[ElectricDruid]

There was a conversation on the Synth-DIY mailing list recently about the MiniMoog's keyboard circuit, and how it held the voltage when you released the key. The answer, as you've probably guessed by now, is S&H! So then there was a discussion about good circuits to use and how much droop the voltage suffers over time with various designs.

You can go simpler if you're not so fussy about the voltage drooping, but if you want something that really stays put, this is the one:

http://ijfritz.byethost4.com/super_sh_rand_web.pdf

Ian reckoned this circuit loses about 3 cents/hr!

The Music From Outer Space circuit got some mentions too:

http://musicfromouterspace.com/index.php?MAINTAB=SYNTHDIY&PROJARG=SINGLEBUSSKEYBOARD2007/SINGLEBUSSKEYBOARD2007.php&VPW=1325&VPH=1199

It's a different standard though - droop is seconds per cent, rather than cents per hour.

Another thing that came out of the discussion is that once the input impedance of the op-amp is beyond a certain point, the significant losses are from leakage in the storage capacitor itself, so you choose caps to minimise that.

The whole thread is archived here if you're interested:

https://synth-diy.org/pipermail/synth-diy/2020-June/173171.html