MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Clone of YuSynth Minimoog filter clone odd noise
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author Clone of YuSynth Minimoog filter clone odd noise
Randy
Happy new year all, wondering if someone has any idea what would cause this sort of noise:

http://www.youtube.com/watch?v=QMAFtW-l41w

This is a clone of the YuSynth Minimoog filter clone, and has always done this. You'll need to turn up the volume to hear it. The filter sounds fine but this noise bleeding through makes it quite a bit useful. I also have an actual YuSynth Minimoog filter clone, and it also has this noise a bit, although much lower in level.

So for you people who develop filters, do you have any idea what this noise is, and how I could get rid of it?

Thanks

Randy
nikmis
I built a yusynth minimoog filter on perf-board years ago and have similar noise. Its more like power supply hum on mine, when the resonance is high and signal input low volume especially. My filter is still very usable but it's still annoying.

Ive wondered if it's something like it needs better power decoupling and filtering


What PCB did you use when you built it?
Randy
The PCB I used was done by a guy on the Mutable Instruments forum. It is smaller than the YuSynth version but based on it apparently.

As with yours, my YuSynth version is usable but the clone of the clone is barely usable.

Randy
nikmis
Ok well I was inspired to fix mine so I tried replacing c13 in the yusynth schematics with two 100n, one from pin 8 of the tl072 to ground the other pin 4 to ground. It completely cut out the noise. Maybe you could try that? Yours seems much worse than mine though


Oh wait I just read on the mutable forums that you did exactly that back in January 2015... So nevermind


And oh wait, that was on my bench power supply it sounded great. Plugged in the supply with all the other modules it's still a bit noisy. But maybe less noisy
Randy
Thanks for the feedback nikmis, appreciate it. I also noticed it's less noisy in my smaller rack.

Randy
nikmis
Anyway the capacitors are to prevent parasitic oscillation, right? Our problem is more like the filter is amplifying power supply noise. This is my only noisy module, too. What power supply are you using?

I don't know a lot about circuit design but my guess is that the noise is in the opamps before and after the filter

Here are the schematics
http://www.yusynth.net/Modular/Commun/MOOGVCF/Moogfilter-sch.jpg

Im going to hook up my scope and see if the power supply noise is present before the final amplifier. If not, then I'll try putting 47pf caps from the input to the output of each opamps see if that helps. That's my only idea, maybe some expert will post by and tell us what to do.

If you google around, there are other people with the same issue, but I think mostly it's barely noticeable so people just live with it
Randy
You know more than I do. I use RT-65Bs, one in the small rack and 3 in the large rack. This is the only module that has an issue, but it is alot more quiet in the small rack.

Randy
Synthiq
The output from pin 1 & 11 on U2 is referred to the +15V supply while both the resonance feedback signal at pin 2 on U1 and pin 2 on the output amplifier U3 are referred to ground, so any noise on the +15V supply will be amplified as much as the signal itself so there is 0dB PSRR in the circuit. Noise on the -15V supply will modulate the current in R26 and half of that current will be coupled to pin 1 & 11 on U2 so there will only be 6dB PSRR from the negative supply.

If there is noise on the supplies, it must be removed by additional filtering or adding voltage regulators to isolate the circuit from the supply noise. The latter will probably require some component changes for a lower voltage from the new regulators.
Randy
Thanks Synthiq, wish I understood most of that. I have no doubt you are correct about the power supply noise, which might be why it seems okay in the smaller rack with one supply and not in the larger rack with three, although I'm not sure how all that noise is flowing around. I'm using cheap supplies 'though.

This is the only module exhibiting this noise. It's Eurorack as well, so +-12V.

Randy
Synthiq
The output from U2 is a current that varies with the signal. Since the load resistors R28 and R29 in series are connected to the positive supply, the absolute voltage at the same output can be written as V+ - R*I, where V+ is the positive supply, R is the sum of R28 and R29 and I is the sum of the currents from pin 1 and 11 of U2. What the next stage, U3a, sees is the variation of that output voltage from U2 since a capacitor blocks the dc signal. The problem is that if U3a sees a 1mV change, it is impossible to say if it is caused by V+ changing 1 mV or if the current I changes so the voltage across the resistor changes 1mV so it is not possible suppress the supply noise.

The previous stage, the ladder filter, has the same problem in that the output voltages follows the positive supply in a similar manner, but here we measure the difference between the two outputs so it doesn't matter if both moves up or down since the difference is still the same.

An opamp like TL074 has a typical 100dB power supply rejection ratio (PSRR) which means it can suppress power supply noise by a factor 100,000 so this is probably why you don't have problems with supply noise in many other modules.
Randy
Thanks for the explanation. What I really should do is compare the schematics of the clone clone to the actual YuSynth clone. The YuSynth version does exhibit the same noise but it is much lower amplitude.

So either there is a shortcut the second person took (the PCB is quite a bit smaller) or something different in the actual design.

Randy
nikmis
I replaced the electrolytic capacitors on the power input with 47uf and it made the noise quieter. Its still there but quieter. I feel like just replacing with larger caps isn't the best idea though. Maybe using 7812 and 7912 voltage regulators would work? My power supply is +-15
Randy
Tried 47uf, no difference on mine.

Randy
Synthiq
With a 47uF capacitor, the corner frequency of the lowpass filter with the 10ohm resistor is 340Hz so the fundamental and the first few harmonics of the mains frequency can still pass through without attenuation. You may have to increase the capacitance by another factor 5-10 to get significant attenuation of the fundamental.
Randy
So you're thinking up to 470uf?
Synthiq
Yes, but I realize a 220uF to 470uF capacitor is much larger than the current ones so I don't know if it will fit on the pcb. But at least it can be used as quick test to see if this is the problem before testing with added voltage regulators.
Randy
No difference with 220uf. Not sure I have a 470uf.

Randy
Synthiq
If you see no difference at all with 220uF capacitors, I doubt you will see much change with 470uF either. I assume you changed the capacitors on both the positive and negative supplies.

If you have a DMM, you can also try to measure the ac voltage on both supplies as well as the output, maybe both before and after you remove the 220uF capacitors to see if they make any difference. If the output level is much higher than the level on the supplies, the noise you see probably originates somewhere else.
Randy
I'm certain the noise originates somewhere else, I'm just not sure why the Yusynth filter is the only module that seems to pass it along. My actual Yusynth filter also passes this noise along, just at a much lower level than the clone.
Synthiq
This is just a wild speculation, but I read somewhere that if you have a high frequency oscillation in the circuit, it can increase the noise in the audio band. If this is the case, you can try to add a ~33pF or larger capacitor across the feedback resistors R5 and R32 for the two opams to reduce the phase shift in the feedback networks to help stabilize the amplifiers.
Randy
Thanks again. I'm going to take a little break, DIY troubleshooting is making me miss my acoustic piano.

Randy
nikmis
I tried the 47pf on the feedback resistors and 470uf on the power input. Now I have almost no noise. Thanks synthiq!
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group