volca keys aliasing.... what could be causing this?

kamalmanzukie · Post by **kamalmanzukie** » Wed Apr 13, 2016 11:47 pm

i noticed this with a slow lfo to cutoff on full blast and full resonance. if the regular cutoff is turned down, everything is fine. but turn the cutoff up and it pushes the filter cutoff's self oscillation past nyquist and some pretty gnarly aliasing rears is ugly head pretty quickly

what i don't get, though, is why an analog synth should have a nyquist limit. is this maybe from the delay? i made sure to plug directly into my receiver to rule out my audio interface being the problem

the keys has a constant 20khz hum, and i figured this might be some kind of filter with high resonance just to make sure nothing made it past 20khz, but if thats the case, obviously it isn't doing much good

JiggyWig · Post by **JiggyWig** » Thu Apr 14, 2016 7:51 am

Nyquist is a frequency of half the sampling frequency, if it's not sampling there is no Nyquist. I'm not sure what effect you're found here, but I think we can be pretty sure it's not aliasing even if it sounds like it.

If you're over your audio interface Nyquist limit your audio box filters should already be removing the out of band signal.

Presumably some sort of weird nonlinearity in the filter. Can't say as I've experienced this, but I'll try my unit later.

kamalmanzukie · Post by **kamalmanzukie** » Thu Apr 14, 2016 6:00 pm

i am in complete agreement, but looking in a spectrograph i see the doubling back of frequencies that is the dead giveaway of digital aliasing. and as stated earlier the behavior is still there when going through my stereo receiver, so any complications caused by digital sampling/ antialiasing filters after the output qnnot be factors.

i've established that the digital delay can definitely cause aliasing on its own. turn the delay on full blast and play notes at the very top of volcas range. the delayed notes are lower with no relation to the original frequeny. especially with the envelope modulating the vco frequency, you'll hear the delayed version will sweep frequencies down to up instead of up to down, which to me this is a dead giveaway.

the only problem is that for the original problem, the delay is turned all the way down, so i don't understand why i am seeing this behavior. i'd post a file here if i could to illustrate what i am seeing

tsutek · Post by **tsutek** » Fri Apr 29, 2016 10:57 am

even with analog, some systems exhibit artefacts that resemble aliasing IIRC

Spheric El · Post by **Spheric El** » Sat Apr 30, 2016 5:24 pm

The delay is digi I think ,so with no feedback may still be the cause.
Not noticed this though and ,unless messing up recordings, i'd put away visual analysis and forget about it.
Interesting never the less.

grenert · Post by **grenert** » Tue May 03, 2016 3:01 pm

It looks like you can add a jack to the VCA OUT pad to bypass the digital delay:
https://technicalnonsense.wordpress.com ... olca-keys/

tsutek · Post by **tsutek** » Sun May 08, 2016 9:06 pm

Thanks for the mod link! Has anyone tested if using this mod gets rid of the 18kHz noise in the singalpath?

AustinBru · Post by **AustinBru** » Wed Jul 06, 2016 9:01 pm

I would also be curious to see if this mod fixes any noise floor/hum issues