How is it that I have a problem I don't know how to fix...

[Silver Blues]

...with a pedal I designed?

Earlier today I remembered I hadn't taken any pictures of my Poinsettia build for the Circuit Designs thread even though I replaced the LED like two weeks ago. I was going to do so, but decided I better run it through the rig one more time to make sure the circuit is really what I wanted. I'm super glad I did too, because I discovered the oddest issue with it. I first suspected something was amiss when the tone of the pedal struck me as thinner and treblier than I remember it being on the breadboard. Then, I was just noodling and found that I got whistling intruding into my signal when I turned up the tone control on my bass, which went away as I rolled back (it whistled only when I wasn't playing). Eventually what I discovered was that if I left the pedal in bypass, output of the pedal plugged in to the amp and power connected with nothing connected to the pedal input, the pedal whistles faintly. Stranger still, if I touch one specific wire (namely the input wire coming from the relay bypass module to the effects board) or anything connected to the pedal's input (input cap, bias resistor, and even the clipping MOSFET that isn't connected to the input at all but is in close proximity) it changes (lowers) the pitch of the whistle.  Additionally, turning the gain knob (in bypass) alters the whistle's pitch as well. If I try to engage the pedal in this state, it produces a hellish screeching. If I plug something in to the input, it all goes away. The circuit and bypass module otherwise function fine and sound mostly all right.

I actually have no idea what's going on. I thought it might be some parasitic capacitance issue but I tried prodding the wires and the best I could do was lower the pitch. I even thought I might be going crazy and tested all of my other pedals to make sure they didn't do the same thing (they did not, of course). The reason I ask here first instead of diving in and trying to examine it is because I didn't notice this problem when I initially boxed it, and I have both boards mounted with some pretty strong tape; I don't want to risk damaging the boards if this issue is something that can be fixed in some other manner. My dad suggested there might be a micro-bridge somewhere bridging the inverting and non-inverting inputs and the theory sounds logical, but in that case I don't understand how it could still work mostly properly when set up normally. I can post pictures if this description isn't sufficient to ring any obvious bells.

Murphy's law always gets the better of you...

[Chris60601]

Occam's Razor?

[tabbycat]

i'm no great pedal debugger, but i've had my own share of trial and error and learned a few things there.

it sounds to me as if you have a feedback loop where there shouldn't be one, creating self-oscillation, possibly caused by a electro cap that is being a bit bipolar and looping via ground, or maybe two closely placed components that are touching and creating an unintended loop, or the bridge (as your father suggested). the feedback loop principle being the common theme to all.
this being because you say that touching bits around the input reduces the effect, which makes me think it is because you are acting as an offboard grounding agent for the rogue loop.

have you tried disconnecting the electro caps to ground in that area and grounding them on a ground that is nothing to do with that circuit? (direct to ground or ground on another pedal) or replacing the electrolytics around the troublesome area with ones from a different batch to those you originally used?

checking no components are touching and creating a loop also worth a thought. and the bridge as already suggested.

these are the easiest debug start points (requiring minimum intervention) i can think of for this one.

i hate debugging too. when you've put in the hours you just want the toy, not a headache.

good luck with it, sb.

[Frank_NH]

The Poinsettia is still on my to-build list and I hope to get to it in the not too distant future.  I have a whole bunch of CA3130s to use up!  

One suggestion I can offer is to try changing the chip.  I have an OCD build that I "fixed" by changing the IC.  Originally it was oscillating badly, something which went away when I put a buffer in front of it.  Changing the IC to a new TL082 fixed it.  Don't know why...

Your case may be different, and I would check all of your ground wires.  I would wire your 3PDT such that the board input and outputs are grounded when in bypass mode.

Good luck and let us know what you find out.

[Silver Blues]

Thanks for the suggestions guys, I'm away this weekend but I'll try all of the above when I get back and let you know. I'm not actually using a 3PDT but rather one of AMZ's DPDT swtiching boards, which is preset to ground the input during bypass (I think. I've used this circuit once before and had no problems with it).