two-in-one pedals feedback noise

[gregtoolson]

So I've built several pedals that have two circuits in one box but often have the problem of a feedback kind of noise when both circuits are engaged. For example, I put a Dan Armstrong Green Ringer octave in with a Fuzz Face, shared power and in/outputs, but separate switches so that either one can run alone or both together. But for some reason when I put the octave first, run in series into the fuzz, I get an unbearable squeal when both are engaged.

Another example is a FET booster and treble booster I just wired up that annoyingly didn't have this issue when I breadboarded it. But now that they're in the box together: squeal (though only with single coil pickups in this case).

[GibsonGM]

Hi Greg, welcome.  Did you follow a 'known and verified'  2-in-1 switching scheme, like that found a General Guitar Gadgets?  First guess I'd make, barring a build error/ missing ground, is lead dress....too much wire flying around all over each other, coupling signals into each other.   

Both are pretty 'gainey' circuits, so you likely need to use shielded wire (shield grounded at one end only), shortest runs possible, and ins separated from outs.  You could try moving wires around while it's feeding back and see if anything changes (proving coupling is happening).

Can you post some gut shots of your build?

[gregtoolson]

Thanks for the response, Mike. Here's a gut shot of the boosters. I'm definitely guilty of messy wiring. Do you really think that's all it comes down to? I prodded them around while both circuits were engaged and didn't really notice any difference in the squeal's frequency.




Also, here's the wiring scheme I did with the 3pdts. This is the switch layout I got from Klaus at Das Musikding and is what I use on all my single-circuit pedals with no issue. For the two-circuit pedals I just run the output of the first switch into the input of the second switch and ground them together. The power is split to both circuits from the battery.

[antonis]

Switches wiring is OK but it works as: A or B or A+B (not B+A)

Wiring indeed is a mess and needs tidying up..
IMHO, the main suspect for squealing are transistors legs length..
You have to cut them short and place Miller caps (47πF to 100pF) between Collectors/Drains & Bases/Gates..

[gregtoolson]

Thanks for noticing that, Antonis!

I screwed up with the order of switching. I had the FET boost going into the treble booster, when I wanted the other way round. Just swapped them around and now no squeal.

As for the other two-in-ones that I've built it is more a microphonic feedback noise, so I'll try some heatshrink on the transistors' legs.

[antonis]

I'll try some heatshrink on the transistors' legs.

It will do no good..
Just cut the legs as short as it's physically possible and put a smear of solder on the leg-socket junction.

[GibsonGM]

Yes! The legs are acting like little antennas if they are long...

[gregtoolson]

Ok guys, so I amputated the transistors and my Fuzz Face already has 33pF from C-B on both transistors, (I usually do this on silicon FFs anyway to cut out their inherent highs). I don't know if the DA Green Ringer will work with caps on C-B, though.

Anyway, I'm starting to suspect that "feedback" is possibly not the best way to describe the issue. See, the fuzz+octave pedal works fine in that order. But when I switch it around so that the octave circuit runs into the fuzz (which is how I would like it because the octave is more pronounced that way) it makes a game-show buzzer kind of a noise, though a bit more high pitch, if you know what I mean. So why would the transistors' legs be microphonic only in one order and not the other?

Another two-in-one I've done with a similar problem is a lm386 mini amp with the Kay tremolo circuit in front. As soon as the volume goes more than halfway something overloads and it sounds like a goose honking to the trem's oscillation. Also, there's a volume drop when the trem is engaged, yet when I run an independent Kay trem into a mini amp not only does it sound fine cranked up but the trem actually gives a bit of a volume boost.

We've solved that the FET boost+treble booster pictured above was just me being stupid about the order. (I bet an independent treble booster pedal with an independent FET boost going into it via patch cable would freak out and squeal too). But a DA-GR octave patch-cabled into a FF works great in their own enclosures.

[GibsonGM]

Can you test the FF with a generic booster, without the GR?  And then try the GR with a boost in front, no FF.    This may reveal what's happening.   

Active devices amplify whatever they are fed...if the incoming signal has noise, it will be amplified (less evident with a small-size guitar signal).    If there is coupling or a cold solder joint etc, a higher input signal can make that worse.  Maybe a boosted input can 'reveal' some of what is happening here.

Another thing to try might be to audio probe when both are engaged.   Try to find out where this noise becomes audible, that might give a clue what is going on. Is it right at the FF input? Or after Q1? And so on.   Might be something off causing this...

[duck_arse]

show us the circuit diagrams for your both builds, please. do either have supply R//C filters? two in a box, single supply, you probably should have a supply filter for each, as perturbations of the DC from one circuit can be amplified by the other circuit, in ways you don't want.


[oops - spelling]

[GibsonGM]

Yes - no perturbations please! ^ 

[gregtoolson]

Right, so it's definitely the FF that is perturbed is this case. I ran our FET+treb. boosters into it and there's all sorts of noises. But I don't think that the FF is entirely culpable here.

I've actually been working on and off with this particular pedal for several months now and had all but given up. It started out when I built a second DA-GR using Ge diodes (this first has Si) that I liked better that's gone on my main board. So I had the (seemingly) brilliant idea of putting this old GR circuit to use by housing in with a FF as a two-in-one. I have half a dozen completed FF circuits floating around unenclosed that I've already swapped in/out, changed pots and switched switches, the only constant is the aluminum box and this DA-GR circuit.

Here's the Green Ringer schematic, Duck. And the other is just your run-of-the-mill FF circuit. I already had the idea of adding a simple power filter to the FF circuit: 9v through a 100ohms resistor with a fool's diode and 100uF electrocap to ground. Do you think the octave needs one too, or more robust filters? Keep in mind I only use battery.

[GibsonGM]

I'd do it for de-coupling, but it might not have much impact.   I wonder if the FF's input impedance is causing something?  (anyone?)

[duck_arse]

And the other is just your run-of-the-mill FF circuit.

no such thing. show circuits. everytime. you never know. and, yes, battery, but two circuits, no line R/C's, no isolation from the perturbs.