SOLVED LED noise from signal indicator bleeding into audio signal …

[LukeFRC]

Hello

Need help …

I built an Aion Lumitron - build report here

Aion Lumitron (MuTron iii) - with some mods - a bass player solders

So I got a kit from Musikding and it all went together super quickly! It’s the Lumitron design of a Mutron 3 which was just super simple build - the sister board for the 3pdt footswitch is a bit different to the ppcb approach but makes sense I guess… While they make them nice and easy to build...

forum.pedalpcb.com

Schematic is here https://aionfx.com/app/files/docs/lumitron_documentation.pdf
And a screen grab for your ease is here…



And have a problem …

Basically if the volume on the output is anywhere but fully open there is a crackling noise when there is signal coming through - but before the sense indicator LED after R23 turns on, once sense Led is lit no noise.
Volume on full no noise.
Remove the LED no noise.
Try a different colour LED, thr noise sounds different…
Try a low current LED, the point it turns on reduces, but still noise.

Something about it means google is not helping, as it isn’t the on/off switch popping, nor is it the indicators on a car…

Reddit question with similar suggests two ways to do it

https://www.reddit.com/r/diypedals/comments/100x5lf

“Use a small signal diode like a 1n4148 in series with the led to avoid excessive tinkering. Or add an extra resistor and 470/1000uf electrolytic inline with the power supply for noise decoupling.”

I kind of think this is so basic I must be doing something daft wrong - is Reddit right and how would you approach?

 

[Stickman393]

Hmm.

So: to reiterate: volume full up, no noise.

Anytime before sense LED lights with volume lower: noise.

Removing the sense LED results in no noise.

Hmm.

I gotta admit: that ones a little strange. Looking at the circuit: that sense LED is in parallel with your LEDS in those xvive optocouplers. That part of your circuit is nowhere near the "audio path". The lighting of the LEDS in your optocouplers effects the LDRs in your optocouplers, though, and the varying resistance in those things when they're exposed to light is the actual mechanism that the circuit is designed around.

I'm of two minds here: first off, pull the board out and check your solder connections on the opposite side. Specifically, check out R9 and R10: those are in parallel with your LDRs. The LDRS likely go low when exposed to light: that means that your signal will mostly be bypassing R9 and R10 when the optocouplers LEDS are lit.

Also: double check your charge pump and power supply connections. Make sure all the ICs are firmly seated in their sockets. I recently had a build that went super crackly on me, then I realized that pin 8 on an op amp had bent a teeny bit and was barely touching the socket cup.

 

[LukeFRC]

Thanks @Stickman393 - I’ll give them a check…

I was confused too about noise - it seems that it’s current but not enough to turn on noise… If it is getting to ground then when the volume is down there’s a path? Though that doesn’t make total sense as there’s just less resistance to ground and more to the signal …

 

[Stickman393]

You've got good instincts here, in that you're looking for clues and trying to connect them to the cause of the fault.

I learned long ago in my career that one has to take on a systemic view of things when troubleshooting. Symptoms don't always point you directly to the problem: sometimes you gotta follow a trail of breadcrumbs to find the root cause.

In the same manner, I would say that the presence of noise when that LED isn't lit is indictive less of the LED being the problem, but something associated with the operational state of the circuit *before* the LED is lit being the problem.

The LED on an unlit state isn't actually impacting the circuit: no current flows through the LED before the voltage across its legs reaches the LED'S Vf, similar to how a check valve will not allow fluid to flow through it until the pressure applied across it can overcome the spring mechanisms in the valve.

As for what the case is for the LED being removed from the circuit and no noise being present: I can't account for that. It could be that something else was done while you were unsoldering the LED that changed the way the circuit operates. Like, perhaps, an IC'S seating in its socket.

One last thing...try putting your finger on the plastic part of the ICs as you play. If one of those is the problem, you should notice some crackling noise and then an improvement.

I'm personally not a big fan of those types of IC sockets. The reason being that those sockets have a tendency of losing their springy-ness and making poor contact with an inserted IC. I find that the milled, round pin sockets tend to work much better, fwiw.

If that's the case, you could try to re-tension the springs with a pick or tiny screwdriver.

 

[LukeFRC]

You've got good instincts here, in that you're looking for clues and trying to connect them to the cause of the fault.

I learned long ago in my career that one has to take on a systemic view of things when troubleshooting. Symptoms don't always point you directly to the problem: sometimes you gotta follow a trail of breadcrumbs to find the root cause.

In the same manner, I would say that the presence of noise when that LED isn't lit is indictive less of the LED being the problem, but something associated with the operational state of the circuit *before* the LED is lit being the problem.

The LED on an unlit state isn't actually impacting the circuit: no current flows through the LED before the voltage across its legs reaches the LED'S Vf, similar to how a check valve will not allow fluid to flow through it until the pressure applied across it can overcome the spring mechanisms in the valve.

As for what the case is for the LED being removed from the circuit and no noise being present: I can't account for that. It could be that something else was done while you were unsoldering the LED that changed the way the circuit operates. Like, perhaps, an IC'S seating in its socket.

One last thing...try putting your finger on the plastic part of the ICs as you play. If one of those is the problem, you should notice some crackling noise and then an improvement.

I'm personally not a big fan of those types of IC sockets. The reason being that those sockets have a tendency of losing their springy-ness and making poor contact with an inserted IC. I find that the milled, round pin sockets tend to work much better, fwiw.

If that's the case, you could try to re-tension the springs with a pick or tiny screwdriver.

You were right - it’s not the LED at all
I thought it was one of the electrolytic caps but switched it with no improvement

Yes the sockets aren’t great, I’ve got some of similar in the garage that seem higher quality

 

[LukeFRC]

It’s a wierd static phase when there’s some signal going in, but before any audible change

 

[LukeFRC]

I wasn’t the only one… there’s a mod using a 100pF cap here …

Lumitron (MuTron III) low volume noise issue

Lumitron (MuTron III) low volume noise issue

www.diystompboxes.com