Oscillation Issues When Drive Pedal Bypassed

[josephfra]

Hi all,

I've built a pedal using the following schematic:


It works mostly as intended, except that there's a very prominent oscillation when bypassed. The frequency of the oscillation changes when the gain control is changed. Can anyone think why this would be happening, and how I could fix it? I took a video to show it clearly.

Another detail - unsure if it's relevant - but the clipping LED doesn't seem to light up when the switch is set to LED clipping, but it originally did. So maybe it broke somehow..?

Here's a couple of pictures of the internals too, incase they're helpful. (The censors aren't covering anything essential - just embarassing logos I had printed onto the boards )


Note the red cable running from the footswitch PCB (it's connected to a ground pin), which is taped onto the enclosure. I added this because I found that it reduced the oscillation from being 'deafeningly loud' to just 'fairly annoying'. I'm not sure why this is - the oscillation is 'deafeningly loud' when the circuit is completely outside of the enclosure too.

[R.G.]

1. How exactly does your bypassing connect to the active circuitry?
2. Does the circuit oscillate when you just open the wire to the input 1k?
3. Is your input wire shielded?

[antonis]

there's a very prominent oscillation when bypassed.

Sorry but, unless I got it wrongly, a bypassed effect should be free to behave howsever it likes to.. 

[josephfra]

1. How exactly does your bypassing connect to the active circuitry?

I have the boards wired like this, if this is what you're after. All the wiring is done with the JST connectors shown on the pictures (other than the indicator LED).

2. Does the circuit oscillate when you just open the wire to the input 1k?

Sorry - I'm not entirely sure what you mean by this. I'm not sure if this is what you were intending for me to try, but I detached the connector between the jacks and the footswitch, and the oscillation is still there despite the jacks not being connected to anything, so that's very interesting to me!

3. Is your input wire shielded?

If you're referring to the wire going between the jacks and the footswitch, then no it's not.

[josephfra]

there's a very prominent oscillation when bypassed.

Sorry but, unless I got it wrongly, a bypassed effect should be free to behave howsever it likes to.. 

Well I suppose? But it's affecting my clean signal, which I'm not much a fan of.

[josephfra]

So it seems this is the minimum amount of circuit I need to get oscillations. The input/output jacks are floating, the footswitch is completely detached, and the only connection between the main board and the jack board is the power. Of course if I detach that, there's no oscillation.


Edit: In another development, it turns out that if the pedal is completely detached from the amp, and especially when I'm touching the pedal, I can hear the oscillations though the amp anyway. Weird? Or maybe not so weird to someone who understands this better than I do?

[antonis]

But it's affecting my clean signal, which I'm not much a fan of.

If by "clean" signal you mean the bypassed one, I can't see no way other than via power supply..

[R.G.]

I checked your schematic and didn't see it. If you don't have one, connect a 0.01 to 0.1uF >>ceramic<< capacitor across the top of the opamp package from V+ to V-. Opamps are prone to oscillation if their power supply is not bypassed with low-ESR caps very near the package.
This may not be THE problem,  but if a ceramic bypass is not there, it could certainly be one problem. I use between one ceramic every opamp package to one every two packages, depending on layout.

[idy]

Many arrange bypass switch so when bypassed the circuit input is grounded.

[josephfra]

I checked your schematic and didn't see it. If you don't have one, connect a 0.01 to 0.1uF >>ceramic<< capacitor across the top of the opamp package from V+ to V-. Opamps are prone to oscillation if their power supply is not bypassed with low-ESR caps very near the package.
This may not be THE problem,  but if a ceramic bypass is not there, it could certainly be one problem. I use between one ceramic every opamp package to one every two packages, depending on layout.

Unfortunately I don't have one of these to hand, but next time I order parts I will absolutely do this. Many thanks for the advice! I'll keep this in mind for future builds.

Many arrange bypass switch so when bypassed the circuit input is grounded.

Aha! The oscillation indeed goes away if I hold a ground lead to the circuit input. Many thanks - I'll be rearranging the bypass switch.

[idy]

I do notice that the insulated jacks mean the box is not grounding, thus not shielding the circuit. Your red wire is your work-around. Many ways to make this solid: screw, toothed washer around stomp switch with a wire soldered to is. Some companies use a spring, which seems dodgy.

[bluelagoon]

Here is a depiction of a better way to route your bypass switch circuit, which grounds both the input and output when in bypass mode -

[bluelagoon]

Now why would a spring for ground contact with the enclosure be dodgy, if it does the required job with the required result.??

[duck_arse]

which grounds both the input and output when in bypass mode

this is on a per-pedal basis. some won't like having their output grounded.