I would suggest dropping the value of RPD1.

That would lower the input impedance below 1M, wouldn't it. Maybe to around 667K?

 

I don't see any supply decoupling capacitors on your circuit diagram.

 

Try adding 10kΩ in series with the input followed by a 1nF capacitor to ground.

That'd create a LPF at 16kHz, wouldn't it? Not sure we want that per-se. Are you thinking to rid the audio of the oscillation noise with that?

 

I don't see any supply decoupling capacitors on your circuit diagram.

Are you referring to the power supply? It's there. 100nF right at Pin 8 of the op amp. You're right though, I didn't show it in the schematic. Didn't want to throw in more variables, but I can post a full schematic if that helps.


R11 and C18 form a nice LPF keeping digital noise from the ESP32 connected to the circuit from adding its special noise to the opamp (and that part seems to be working wonderfully in normal operation).

 

That would lower the input impedance below 1M, wouldn't it. Maybe to around 667K?

It would but curious if the noise is eliminated or reduced, start at 1M.
I have that circuit on my breadboard and see no noise even with RPD1 removed.
I see a path from IC6 pins 9 and 3 to pin 4 of IC5 along with the input jack. How long are those traces?
Do you have scope?

 

It would but curious if the noise is eliminated or reduced, start at 1M.
I have that circuit on my breadboard and see no noise even with RPD1 removed.
I see a path from IC6 pins 9 and 3 to pin 4 of IC5 along with the input jack. How long are those traces?
Do you have scope?

I do have a scope. I'm guessing it has something to do with our layout. Not sure if this is too hard to see what's going on, but here's the circuit board:

 

Possibly the layout.
Try this:
Set the relays for true bypass and check the output at RPD0 (10K) with the scope.

 

This is the scope at the "Output Pull Down Resistor" (RPDO):



I should have disconnected it from my amp to get this reading. I'm now hearing an 11K phantom ring in my ears.

 

I think it might be the layout.
Lift one side of C1 from the pc board and check again.

 

I think it might be the layout.
Lift one side of C1 from the pc board and check again.

That made it worse. I didn’t plug it into the amp this time but I had to change my scope to 20mV to get it to fit vertically on the screen.

 

Now that you have found the signal on the output, can you find it on the input pins of the op-amp? If so , which pin? And is there some remnants of it on the supply?

 

Now that you have found the signal on the output, can you find it on the input pins of the op-amp? If so , which pin? And is there some remnants of it on the supply?

I only see that on pins 1, 2, 6, and 7. Not seeing that noise on the opamp's input pins (3 and 5).

I'm wondering if, as soon as I connect something to the input pins (aka, my scope), that does the same thing that a guitar being plugged in does (makes it go away).

 

That made it worse.

Seems like the same noise witness earlier from the ECU.

 

Did you try lowering the value of RPD1?

 

I don't see a track connecting to pin 4 (negative supply) of the op-amp.

 

I don't see a track connecting to pin 4 (negative supply) of the op-amp.

That's connected to the Ground Plane

 

I don't see a track connecting to pin 4 (negative supply) of the op-amp.

Post #24. It's there.

 

Did you try lowering the value of RPD1?

Haven't yet. What's your thinking/theory on doing that? It'd lower the AC impedance seen by the guitar to about 667K instead of 1M.

 

I can't duplicate the problem your having on my breadboard regardless the value of RPD1.
I just think it's worth a try.
Other suggestion, can you power the board with a 9 volt battery?
Remove the GPIO lines from the board?

 

Here is the existing board build that we are currently testing.

1. Mom- is not connected (airwire) to GND, we had to connect the momentary switch to one of the grounds in the JST connector. That's really the only problem we've had, except this current issue.