low-gain JFET overdrive fizzyness

[iefes]

Hi folks,

I'm trying to build a simple low-gain OD. I put together something which is basically the runoffgroove Peppermill but with a little more gain from the first stage. I plan to include some modifications and extensions to that circuit but I would like to have it sound right first.

With overdriven sounds there is always some harsh fizz on top of the otherwise nice overdriven tone. It does not sound natural to me and decays quite abrupt. It's not very loud but it's there and it's annoying. I've tried different FETs already and I also included the 8k2 / 2n2 high-cut at the end of the circuit thats included in the Peppermill schematic.
The fizz is also present at lower gain settings. I'm trying to get a smooth overdriven sound out of this circuit, with a smooth decay of the overdrive.

I've had this issue before and I'm wondering if this is somewhat normal with overdriven FETs but people don't worry about it because in a real life situation you would not really hear it (on stage etc.)...?

I'd be happy to get some tips what I could try to get rid of this fizzyness. (I hope Fizz is the right word)

Thanks!

schem for reference:

[Aph]

The 8K2 resistor and 2n2 capacitor at the output in the peppermill form a low pass filter. Try something like that in your design.

edit.. I see you've tried that. Try a larger capacitor there.

[marcelomd]

I've seen several designs with a pair of 10k+2n2 low pass filters in series at the end. Same logic as the one you tried, but moar =)

[PRR]

That's not really a "low gain" circuit. It's got enough gain to be quite harsh. Probably want several capacitors here or there to blunt the high frequencies.

[iefes]

Okay, so what I take home from your answers is that it's got most likely to do with the high frequency content. Thats something to work with, thank! :-)  I'll try that network as suggested and see what happens. Would it be a good idea to shunt those high frequencies before entering the clipping stage?

@arph: The 8k2 / 2n2 did help a bit but wasn't enough.
@marcelomd: Yep, I know those. I didn't want to dump too much signal, but if these networks make it sound smoother that's definitely what I'll try.
@PRR: You're right. I plan to swap the 1M gain pot by a 250k or 500k at some point but didn't have one handy for the breadboard. But the fizzyness was also there with the gain rolled back.

Thanks so far!

[antonis]

I plan to swap the 1M gain pot by a 250k or 500k at some point but didn't have one handy for the breadboard. But the fizzyness was also there with the gain rolled back.

That Gain pot is actually a voltage divider for J201 Gate incoming signal..
What matters is dividing ratio and NOT its absolute value..
(to be honest, its value DOES mind when considered as load in parallel with M1 Drain resistor & C3 High pass filter)

Lowering its value lowers M1 gain but to a negligible level..!!
(4k7//500k) / 110R equals practically to (4k7//1M) / 110R ..

Also, lowering its value results in raising C3/U1 HPF cut-off frequency..
(more "undesirable" highs in your case..)

I should suggest to try placing some tens of pF caps in NFB loop of M1 (in parallel with R2) and/or J1 (between Drain & Gate)