Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
First, what do you really mean by 'almost forever'? 1 hour, 1 day, 1 year, 1 millennium?
An ideal capacitor will fold a charge forever if it is perfectly isolated, but ideal capacitors and perfect isolation only exist in mathematics.
A real capacitor has two inherent ways to make this hard:
1...Insultaion resistance. It is not infinate, charge will gradually flow across it, discharging the capacitor
2...Dielectric Absorption. This is a recovery of charge in a discharged capacitor. Quite interesting, search Google for more on this.
You can isolate the capacitor with FET's, or even relays, and pot it in something with high resistance to negate leakage to air. You could even carefully remove the charged capacitor and ship it into space on the shuttle. Even then, you can't get away from the capacitor's own leakage.
I suppose you could seperate the plates and put them into hard vacuum. This may keep the charge for quite a while.
Really, the most sensible answer if you really need long-term, accurate storage (more than a few minutes or hours depending on accuracy required) has been given:
Digitise it quickly and store it in digital memory. Even then, it won't last for ever!
OK, digitise it and make a physical model. Stone Henge has been around a while - maybe this is what it was for :wink:
I think you can probably do a real infinite Sample and Hold ciruit if you are using Switched Capacitor Technique: Each Cycle Sampled back the output of the sample-and hold and then fold back the sampling capacitor as a holding capacitor in the next clock phase. Since your output is sampled and recovered in every clock cycle, charge will not lost provided that you have the power supply and clock supply.
You can probably try this method to see if it can help. Many books discussed the SC sampled-and-hold circuits.
I think your sampling rate is slow, so
you can use simple circuit technique to do that:
signal --> ADC ---> Registers --> DAC --> reconstructed signal
The ADC can be a SAR type with an counter to trigger it, whenever
the (S/H value - Signal) < 1LSB ===> stop the counter ,
then the value in the counter is the code of the signal.
then you can use a resister strings DAC with the counter values
to controll the switchs to reconstruct it.
The code in counter can be infinite as long as the power is given.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
