fx loop blend circuit

[emosms]

I am trying to make a dry-wet blend mixer for an fx-loop.
It should have gain ~1 mid to end possition and attenuate to shutoff the opposite direction.



























But it oscilates at a very low frequency...

The send-return module is a seporate board with 6.3'' plugs and op-amp buffers. It is not connected at all.
Now the main signal (from U2:A) should have gain ~1 mid to end position and attenuate to shutoff in the opposite direction (works on simulator).

[Chuck H]

I'll open by saying that I don't think I have the chops to help at any real depth but I'll offer what I can. So...

I don't see anything about the circuit that should obviously cause a LFO. Perhaps there's something physical about the circuit like layout, lead dress or ground scheme or even a miswire that is causing the problem. For the purposes of analysis it will help to know WHEN in operation the LFO is happening. What position of the blend pot?

[Helmholtz]

With the blend pot in some middle position you have a feedback path between the input and output of the connected unit via R15/R16.
Does the blend pot setting affect the oscillation?

I might try wiring U2B as inverting amplifier providing a virtual ground input.

[Chuck H]

With the blend pot in some middle position you have a feedback path between the input and output of the connected unit via R15/R16.
Does the blend pot setting affect the oscillation?

I didn't see this as a feedback path but I want to understand. I see that the main signal and the effects output are summed here but not where anything is "fed back" in any way I recognize. If these two signals are out phase there would be some cancellation but if they are in phase aren't they just parallel signals to the U2B input?

I might try wiring U2B as inverting amplifier providing a virtual ground input.

Agree.

[Helmholtz]

I didn't see this as a feedback path but I want to understand. I see that the main signal and the effects signal are summed here but not where anything is "fed back" in any way I recognize.

It's a signal path between the SEND and RETURN terminals. As these will be connected to the input and output of the effects unit, it might cause the effects unit to oscillate.

[Chuck H]

It's a signal path between the SEND and RETURN terminals. As these will be connected to the input and output of the effects unit, it might cause the effects unit to oscillate.

Ah. I hadn't considered that. I was only looking at why the circuit shown might oscillate and failed to consider the units in the loop.

[emosms]

Oscillates regardless of the pot position.
Send return module (with own op-amp buffers and phase shift switch) is not connected at all.

As seen here, and other circuits on the net I don't know what else to do but rewire the second op-amp as inverting amp
- https://modwiggler.com/forum/viewtopic.php?t=258947

I made it non-inverting based on such schematics but I haven't seen non inverting circuit or an explanation if inverting circuit is a must (or just skimmed and missed that, looking fast through various sources)

[Mick Bailey]

Have you thoroughly checked for miswiring or solder bridges? When troubleshooting LF oscillation it's sometimes useful to parallel a much larger value cap across each of caps in turn to determine what affects the frequency. It can help in tracking down whether it's signal path or supply.

[emosms]

Motorboating was probably cause of loose solder (spoted a loose leg on the pcb).
Reworked it as inverting:

Now there is NO SOUND AT ALL, DEAD QUIET, no reaction to pots whatsoever.

Removed the module, bypassed it, the rest works as expected.

[emosms]

The problem...

[x-pro]

The problem...

The whole great beau monde has been misled.