Circuits that respond well to voltage-sag?

[earthtonesaudio]

I built a negistor-driven charge pump that puts out about 30 volts at microamps of current. 
(thread: http://www.diystompboxes.com/smfforum/index.php?topic=68035.msg544896#msg544896 )
Drawing a little current makes a big drop in output voltage.  If I power a transistor gain stage with it, playing harder should cause the supply voltage to drop radically.  Like from 30V down to single digits of volts. 

So I have a tool without a use, and naturally, can't find what I'm looking for.

Can anyone help me find a circuit to use that responds "dynamically" to changes in supply voltage?  One that is stingy with the current consumption would be about right.  I was (not very creatively) thinking of using some fuzz or distortion circuit, but I don't even know where to start!

Thanks for the help in advance, I know this is a kind of obscure question.

[earthtonesaudio]

Shameless bump.

[John Lyons]

just build a test rig out of the charge pump circuit and try it with a few pedal circuits to see what you get.
Tube screamer circuits get a little sweeter when you limit the voltage and Fuzz Face circuits do neat things as well getting really cunky and zippy sounding. Jordan bosstone gets the same misiased yet totally usable thing going on...

Try it out and report back.

john

[breather-resistor]

I would recommend sticking to discrete circuits for starters and preferably npn or pnp based rather than J-fet.

I have spent a bit of time with voltage sag and while it certainly works with other types of circuits.  I have noticed the most interesting effects with these types of circuits.

breather.

[brett]

Hi
CMOS circuits should work quite well, but you'll want to "soak" the power into a parallel transistor circuit (e.g. the LBP-1 or a BMP output stage driving a small impedance) to get the sag. 
Try a tube-sound fuzz, maybe.
cheers

[earthtonesaudio]

Thanks everyone. 

I should clarify that my charge pump supplies microamps of current (even a 10M load drops power by >50%!) at it's "high-ish" voltages.  That makes TS type circuits pretty much off-limits for it, but I agree they do sound cool at lower voltages.

I was also toying around with the idea of using a low-power boost circuit with a constant-current source (like some Tim Escobedo designs)  but I don't know if that will defeat the purpose or not.

Brett, that's a good idea to use a parallel load-circuit.  I don't know if it will need it, but I'll give it a try.

[d95err]

You need a circuit where the current demand varies with the signal level. Most stompboxes are class A circuits, so current demand is more or less constant regardless of the input.

To simulate sag in a stompbox, I'd suggest using a voltage driven side-chain (like in a compressor), but having the side-chain control voltage to a FET gain stage rather than attenuation/gain.