[sdiy] S&H problems

[Raymond Wilson]

Oops. Forgive my poor reading of your original email. Your resistor is between the gate and the source not the source and the drain. So... then... the resistor is not the culprit. Sorry for the bum steer.

Ray



-----Original Message-----
From: Ray Wilson [mailto:rayw at csd.net]
Sent: Thursday, September 11, 2003 7:52 AM
To: David Reichert; synth-diy at dropmix.xs4all.nl
Subject: RE: [sdiy] S&H problems


The 1 meg across your switch is the main culprit for the droop. You want the
switch to be infinite resistance (ideally) when off. 1 meg is very low
resistance when your talking about a cap draining unless its a very large
cap.

Ray

Look here for ideas:

http://atlas.csd.net/~rjwsoft/sandh.html

http://atlas.csd.net/~rjwsoft/NewAugustNewSampleHold.html

Keep in mind that even commercial sample and hold circuits have a droop
specification that is typically millivolts per second. You will note that
the cap in my design is pretty big but it droops pretty slowly but droops
nevertheless.

Hope this helps

Ray

-----Original Message-----
From: owner-synth-diy at dropmix.xs4all.nl
[mailto:owner-synth-diy at dropmix.xs4all.nl]On Behalf Of David Reichert
Sent: Thursday, September 11, 2003 7:21 AM
To: synth-diy at dropmix.xs4all.nl
Subject: [sdiy] S&H problems


Hi everyone

I'm currently trying to put together a sample and hold circuit, I'm not
having any trouble getting it to sample, but I'm quite struggling with
the  holding end of things!

Seems to me that the trouble is in the switch, it's obviously leaking
back through the switch. I'm using a JFET, The drain is the output,
source as the input. I have a 1M resistor connected between the source
and gate. As well, I have a diode connected with the anode connected to
the JFET gate and the cathode is the controller input terminal.

Both the input and output are buffered, which leads me to suspect that
the leakage is through the gate, but I have not confirmed this. This is
backed up by the fact that when I tried removing the source connection
before closing the switch, the capacitor still discharged.

I've tried several different types of FETs, switching the roles of the
source and drain, adding a short delay before the gate switches off (in
case it's a timing problem) and none of these has helped the situation.

If anyone has some advice, it would be much appreciated!

Thanks in advance,
Dave