A weird solution to a problem, but is it legitimate ?

[Nuts]

Hello everyone !
I recently bought a flanger pedal - 'Donner Jet Convolution', which is supposed to be a clone of the Electro-Harmonix Electric Mistress.
The pedal sounds pretty good, but if I set the 'color' knob too high (past 3 o'clock) I hear a 'jet' sound even in Bypass mode, and even when the strings are muted.
I opened the pedal, and saw that it does have a 3pdt switch, so I tried to identify the terminals.
I connected what I believe to be the input of the effect to ground, but the noise still occurred.
Out of frustration (and stupidity) I started to randomly connect the effect input pin to other pins on the switch, and when I connected the effect input to the effect output the noise was gone.
So I'm thinking to rewire the switch in a way that connects the effect input to the effect output when bypass, but is it OK to do that ? can it cause any damage to anything ? I'm worried about the power supply the most, will it be fine ?
Thanks !

[swever]

You've probably rewired it to true bypass, as opposed to something else that they had there.

If I were you I would identify the board's input, output, ground, and +V pads/wires, and rewire the bypass 3pdt from scratch.

The next step would be to detemine how to disable all the bypass and bypass buffer circuitry in the board, but that may not be neccessary if the effect does not get powered from battery.

[swever]

Okay I am probably wrong because it says "true bypass" on the enclosure.

I would try to figure out if the 3pdt grounds the output when bypassed. If it does I'd try breaking that connection.

Take all I say with a grain of salt though because I don't really know what I'm saying 

[Nuts]

The 3pdt does not ground the effect output at all, thanks though.
What about connecting the effect input to the effect output, do you know if it can be a bad idea ?

[BetterOffShred]

That's called a feedback loop, usually when you hook the output of an effect up to the input of itself, you get crazy oscillations.   

[vigilante397]

Does the original switch wiring ground the effect input when bypassed? That's what I usually like to do and I don't generally get noise leakage. Just re-read and saw you said you tried that already. Whoops.

I wouldn't recommend shorting the effect input and output based on theory, but you have already said that worked and fixed the problem. And no, I don't see any possible way that could damage anything, least of all the power supply. So if your solution solves the problem successfully without creating more problems, who am I to tell you it's wrong?

[Nuts]

It does not ground the effect input when bypassed, but I did try to connect the effect input to the ground myself but the noise was still there.
I just want to make sure that connecting the effect input to the effect output can't cause damage to anything, because I don't really understand how and why it worked, it was only random luck.

OK I just read your last edit after I wrote this.
So thank you

[Danich_ivanov]

I would try output to the ground.

[digi2t]

Just for reference, here are some bypass wiring schemes. Depending on how much noise your circuit is putting out will determine which scheme you'll need. I've had an occasion or two where I had no choice but to ground the effect in and out to keep an effect quiet in bypass (screaming, screaming, screaming oscillation!  ).

[Kipper4]

I'd opt for output grounded.
Having said that it depends on what the problem is. If it's a signal bleed in the switching it might not help.

[Nuts]

I did try to connect the effect output to ground, and the effect output and effect input ground, but the noise was still there.
This is how they originally wired the switch :


Another thing that was weird about this pedal is that when I connected another pedal that has a buffer in it (any boss pedal), before the flanger, the noise was gone.
When I tried it with a truebypass pedal before the flanger, the noise was still there.
So a buffer before that flanger also eliminates that noise.

[Danich_ivanov]

Could be the switch, i had switches that leak. If not, then i would trace the circut to see what is going on there. Also, could be that it has nothing to do with the pedal, but something else in a chain, while this pedal reacts to a problem in a way it does. Hard to say.

[pinkjimiphoton]

being all SMD doesn't make for a very promising mod/repair

i would likely ground the output of the board in bypass. i'd open mine up and take a peek, but it lives on my live board in the truck