Grounding input or output when bypassed?

[DWBH]

I have a pedal that I'm rehousing that has this high pitched noise/oscillation when in bypass (true bypassed, now).
Should I ground the circuit's input or output (in bypass mode) to get rid of the noise?
Grounding the input seems common and obvious, but are there any differences in bypassing the circuit's output?

[Minion]

In true bypass the whole curcuit is allready totally bypassed so if done properly there is no possibly way any noise from the curcuit can enter your signal , that is pretty much the whole point of true bypass as opposed to transistor/fet switching which can cause signal loss ,noise and crosstalk.....

I would recheck that you have your switches and input/output jacks wired correctly .....

[DWBH]

In high gain circuits, it's good practice to switch the input of the circuit to ground when the pedal is bypassed.

[anchovie]

In true bypass the whole curcuit is allready totally bypassed so if done properly there is no possibly way any noise from the curcuit can enter your signal , that is pretty much the whole point of true bypass as opposed to transistor/fet switching which can cause signal loss ,noise and crosstalk.....

I would recheck that you have your switches and input/output jacks wired correctly .....

Not true. You can wire up for "true bypass" in a way that leaves the effect input disconnected and floating. My first ever build was a Dr Boogey and I wired the 3PDT footswitch the obvious simple way. The lack of an input connection to the board when switched off caused the effect circuit to go nuts and oscillate and the contacts on my "bypass" part of the footswitch were able to pick this up just like interference, seeing as this was all happening inside the box.

DWBH: I use the wiring that you posted all the time now. Grounding the input shuts the effect up perfectly. Grounding the output instead would be just like having a volume pot turned all the way down - the effect circuit is still doing stuff even if the final output gets muted, so there would still be the risk of oscillation being transmitted to the bypass signal path.

[DWBH]

Grounding the input shuts the effect up perfectly. Grounding the output instead would be just like having a volume pot turned all the way down - the effect circuit is still doing stuff even if the final output gets muted, so there would still be the risk of oscillation being transmitted to the bypass signal path.

Thanks!!
That's what I wanted to hear 

[WLS]

Good tutoral, Helped me wire my first box and is exactly what your talking about.

http://gaussmarkov.net/wordpress/thoughts/wiring-up-a-1590b/

Hope this helps... 

[Minion]

In true bypass the whole curcuit is allready totally bypassed so if done properly there is no possibly way any noise from the curcuit can enter your signal , that is pretty much the whole point of true bypass as opposed to transistor/fet switching which can cause signal loss ,noise and crosstalk.....

I would recheck that you have your switches and input/output jacks wired correctly .....

Not true. You can wire up for "true bypass" in a way that leaves the effect input disconnected and floating. My first ever build was a Dr Boogey and I wired the 3PDT footswitch the obvious simple way. The lack of an input connection to the board when switched off caused the effect circuit to go nuts and oscillate and the contacts on my "bypass" part of the footswitch were able to pick this up just like interference, seeing as this was all happening inside the box.

DWBH: I use the wiring that you posted all the time now. Grounding the input shuts the effect up perfectly. Grounding the output instead would be just like having a volume pot turned all the way down - the effect circuit is still doing stuff even if the final output gets muted, so there would still be the risk of oscillation being transmitted to the bypass signal path.

It doesn"t matter that the Input or output are floating or if the curcuit is oscilateing like a freight train because in proper true bypass mode the effect curcuit isn"t part of the signal path anymore it is truely bypassed .... the input jack and the output jack are directly connected via the switch so the curcuit should have about as much effect on the signal as your car parked outside has on the signal.... Unless the distortion is leaking through the ground but I have never had any problem like that....

[anchovie]

It doesn"t matter that the Input or output are floating or if the curcuit is oscilateing like a freight train because in proper true bypass mode the effect curcuit isn"t part of the signal path anymore it is truely bypassed .... the input jack and the output jack are directly connected via the switch so the curcuit should have about as much effect on the signal as your car parked outside has on the signal.... Unless the distortion is leaking through the ground but I have never had any problem like that....

Really? Funny that the example I gave was from personal experience then, and that I found the grounded input solution on this very forum when other people were building the Dr Boogey at the same time. An oscillating high-gain distortion will transmit in the same way that a radio station does - if you want to listen to the radio you don't run a long cable from the building with the DJ in it! We have a preference for housing our effects in metal boxes that provide shielding because of the very existence of wireless interference.

[WLS]

Really? Funny that the example I gave was from personal experience then, and that I found the grounded input solution on this very forum when other people were building the Dr Boogey at the same time. An oscillating high-gain distortion will transmit in the same way that a radio station does - if you want to listen to the radio you don't run a long cable from the building with the DJ in it! We have a preference for housing our effects in metal boxes that provide shielding because of the very existence of wireless interference.

Good point that's why on a clear day I get "FM 90 whatever" when I have a prject on the breadboard, but once I box it up it's as quiet as a mouse.

[MikeH]

Anchovie is right; some effects will oscillate when left hanging and these oscillations will be picked up by your signal chain the same way it picks up radio interference.

[DWBH]

Well, grounding the input fixed my problem, so there you go. Thanks guys