KoT Breadboard Squeal, grounding issue?

[akmenpi]

I bread boarded a King Of Tone circuit. Each half of the circuit sounds great on their own, but when I cascade one into the other, I can't go past 9'oclock on the gain controls without a horrendous squeal.

I found that if lift the ground from the feedback loop in the first gain stages (the 10n cap after the 27k, 33k branch), then the squeal goes away. This leads me to think that I'm shaking the ground causing oscillation in the feedback loop (just a guess).

Is this something that someone else has experienced?

What are possible solutions?

(I'm not pumping up to 18v, don't know if that matters to this or not.)

schematic I'm using http://fuzzypedals.blogspot.com/2015/11/king-of-tone-overdrive-queen-of-bone.html

[duck_arse]

hello and welcome, akmenpi.

one thing you can try on the BB is to return that cap leg to VB instead of ground. this might help, might make it worse. you may also have a lead dress problem, causing feedback. poke the wires with a stick, see if the squeel moves.

we don't mind seeing photos of breadboards - don't be shy.

[akmenpi]

Thank you for the reply

I tried moving it to Vb, but the that didn't change the squeal. Next time I get a chance I'll check out the wires and I'll upload pics.

[duck_arse]

just out of interest, what opamps did you use?

[antonis]

I found that if lift the ground from the feedback loop in the first gain stages (the 10n cap after the 27k, 33k branch), then the squeal goes away.

Welcome also..

By lifting GND from the NFB loop you turn first stage into a buffer (voltage follower of unity gain..)
This tells you that you should cascade high gain stages with caution..

With both Gain pot set to max, overall cascaded gain is 90dB (8.4 X 22 X 8.4 X 22) at high frequencies (C3 & C14 considered shorts..)

[iainpunk]

well, 90dB is enough that a tiny bit of capacitive coupling between somewhat close wires will make things squeal.
i helped a friend with a squealing KoT clone and it was just to much gain and wires to close together. we replaced his wires with shielded wires, and that killed the squeal.

cheers

[akmenpi]

just out of interest, what opamps did you use?

JRC 4558D

[akmenpi]

I found that if lift the ground from the feedback loop in the first gain stages (the 10n cap after the 27k, 33k branch), then the squeal goes away.

Welcome also..

By lifting GND from the NFB loop you turn first stage into a buffer (voltage follower of unity gain..)
This tells you that you should cascade high gain stages with caution..

With both Gain pot set to max, overall cascaded gain is 90dB (8.4 X 22 X 8.4 X 22) at high frequencies (C3 & C14 considered shorts..)

There's still the 10k as well as the 100K from the GAIN pot in the loop, and there's no audible gain reduction (just more muddy)

I'm fairly certain it's not unity. It's still clipping just as hard in the next stage.

[antonis]

I'm fairly certain it's not unity.

I'm fairly certain for exactly the opposite.. 

It's still clipping just as hard in the next stage.

Any signal of slightly over 140mV Vpeak amplitude (considering JRC4558 output range about 1.5V less than supply voltage) is clipped..
(unclipped signal headroom ranges between +1.5V and + 7.5V => 6V where 280mVp-p X 22 => 6.16V..)

[akmenpi]

I'm fairly certain it's not unity.

I'm fairly certain for exactly the opposite.. 

It's still clipping just as hard in the next stage.

Any signal of slightly over 140mV Vpeak amplitude (considering JRC4558 output range about 1.5V less than supply voltage) is clipped..
(unclipped signal headroom ranges between +1.5V and + 7.5V => 6V where 280mVp-p X 22 => 6.16V..)

Perhaps I wasn't clear enough. I disconnected the feedback loop's communication to ground. I removed the network comprised of the 2 10n caps, the 27k, and 33K is just filtering.


The squeal is also goes away if I simply open the ground leg of the 10n cap.

[antonis]

Perhaps I wasn't clear enough. I disconnected the feedback loop's communication to ground. I removed the network comprised of the 2 10n caps, the 27k, and 33K is just filtering.
The squeal is also goes away if I simply open the ground leg of the 10n cap.

That's exactly what I'm saying..!! 
(by simply opening 10nF cap ground leg you open the entire NFB loop - no need for extra disconnections..)
Same happens when you "open" ground path by cutting connection both through 27k and 33k resistors..

[antonis]

More clear..