Weird signal bleeding issue

Hey so recently I don’t know if this is something that anyone has had an issue with or similar before, but I just encountered something that kind of got me wondering and concerned. So today I was playing with my homemade modular synth when I noticed the oscillators sounded a little funny when I had them patched to my mixer, and began checking to see if they were fine and working properly. Somehow I soon found out that even though it wasn’t patched, my lfo at the other end of the synth was affecting the signal even though it wasn’t patched into anything. And to give you somewhat of an idea of my setup, my oscillators are on a different shelf than the mixer and the lfo being powered by a different, separate power supply, while the lfo and mixer are on another shelf but also being powered by a completely different power supply. I’m not sure if this is because of the module itself, being the hagiwo lfo, the setup I have, or the metal I am using for the panels. I’m really stumped about this and so far this is the only issue I’ve had but I also noticed that when I press down on a key to control the cv of an oscillator, the pitch of it sort of just tunes up or down even when playing a different note, but again I don’t know if it’s related. Can anyone help?

1 Like

Hi there, first of all my view is that if it is a DIY synth we can always expect minors issues, the next question is can you actually survive with them.
That being said, my best guess is that your units, modules are on different groung level. It would mean that, as an exemple, the “0 Volt” for your lfo translate to 0.2 Volt for your mixer and that could create noise.
As I mention noise I am also thinking that it could be the wires not being insulated properly and bleeding into one another. :thinking:

4 Likes

That is a good point, like it’s not bad to the point where my entire signal is messed up because of it, but at the same time I don’t know what other modules might be doing this too. But now that you bring up the fact of the ground levels, how would I determine that’s the case? Would I use a multimeter or something and how would I go about fixing it if there is a way?

1 Like

You can measure it as you would measure any voltages, a difference of potentials between 2 points. Switch on both psu, and measure the voltage between the 2 grounds connections using one of the ground as “ground” and the other one as your actual voltage. As always be careful when playing with the psu, there is a lot of current.
You can then try again with a jack socket connecting your lfo and your vco to see how this is affecting it.
Good luck

Edit: you could try to power the modules one at a time and see at which point it becomes an issue.

1 Like

Thank you so much for the advice!

2 Likes

One thing i have heard of and seen post on is joining all the grounds together from the PSU’s efectivly making one common ground plane. Not sure how effective it may be in this situation and not tried it myself yet.

Rob

2 Likes

I cant see why you couldn’t do that. This is actually what you do when connect jacks between modules anyway.

I had lots of issues myself when connecting the keystep37 to my kosmo. I was hearing noise and click every time a key was pressed. I realised that it was only happening when powered from the USB. Using a dedicated psu for the keystep sorted it. I assumed that it was probably due to a “ground” problem.

2 Likes

Apart from blocking capacitors placed close to components which have highly varying currents (caused by lots of voltage changes etc.), a common ground is definitely a good idea to reduce noise.

If you use aluminium panels which are connected to the ground through e.g. the jack sockets, then patching across your cases/shelves already pulls the grounds together. It’s however good practice to join them. That’s what you also often see on turn-tables for example a screw at the back to attach a ground wire that is connected to the the audio system and mixer.

Another source of noise are stripboards whose traces can act as antennas. Op-amps are fast switching circuits and can generate a lot of electromagnetic noise which are transmitted through the traces of electronic circuits. Especially the inputs of op-amps are very sensitive to such noise since those are the two inputs of a differential amplifier. As you may already know, the potential (voltage) difference between those inputs is amplified by a huge number (usually >100000) so if you have a difference of 1 uV (micro volt, or millionth of a volt), it will be amplified to 0.1V which is definitely in the ballpark of audio signal levels.

That’s btw. also the reason why you see so many metal boxes (shielding) around such components or modules in audio or high-precision equipment.

That being said, always make sure to cut the traces as short as possible, especially around delicate components like opamp- or FET-inputs in general, or shield them.

Last but not least, also the mounting direction of the circuit can play a big role. Traces which are aligned in parallel, will like “couple” to each other in the electromagnetic field.

5 Likes

I had a similar thing with my KeyStep. Not only if powered from USB but if connected to USB, when the synth was going into a USB mixer. I ended up getting a ground loop isolator which fixed it.

8 Likes

I have kind of fixed the issue for the most part using a different lfo, but I am still experiencing the weird oscillator detuning. Basically I have 4 oscillators being controlled by the midimuso 12 and for the most part they work find. But after sometime, or sometimes, when I would play the fourth note, all the oscillators would detune at the same time and no matter what I did I still had the problem. The lfo doesn’t seem to be the issue, but I’m wondering if anyone has had a similar problem like that