Twitter   Patreon WIGGLING 'LITE' IN GUEST MODE

Author fine-tuning a self oscillating VCF
colourloss
 I have the Qu-bit Tone and have been experimenting with using at as a VCO; however, it is quite difficult to tune to exact frequencies. For example, I always seem to be either slightly sharp or slight flat when trying to tune to C. Is there a way I can use one of my my maths channels to fine tune this? If so, how would I set it up?
wechard
 Don’t know about that particular VCF, but in general for fine-tuning CVs it helps to chain them through extra attenuation. If you have Maths, try taking an offset from the channel two output, inputting that to channel three, and taking your CV from the channel three output. That way channel two will set a maximum and channel three allows small adjustments within that limit. If channel two is set very low to start with, that can give a lot of fine control.
colourloss
 So I'd plug the CV out from my sequencer into ch.2 of maths, then take that out into ch.3, then from 3 into the cv on the filter?
wechard
 No, sorry, what I was describing was for fine-tuning a static voltage to get a single frequency. If the problem is that the VCF doesn’t seem to be following a sequencer or keyboard output in tune, that would be something else. Are you sure that this particular VCF has reliable 1V/oct tracking while self-resonating? If the filter itself is a bit off in tracking, there’s probably no way to improve it with external modules.
colourloss
 It tracks reliably it's just that I can't seem to tune it to a precise frequency. I think the cut off knob is just a bit too sensitive.
wechard
 OK, well here's something that might work -- i just tried it with an SQ-1 sequencer and the 1v/oct input on my DPO and it was fine, so it should translate to your situation. Take the sequencer output and plug that into channel 1 of maths, set with 0 rise and 0 fall and linear response. Take the channel 2 output of maths and patch it to the channel 3 input. To be on the safe side, also patch a dummy cable into the channel 4 output. Take the SUM output as the CV to the VCF. That should let you dial in a very small offset to sum with the sequencer CV, which might be more sensitive than the pot on the VCF for fine-tuning after the VCF is as close as it can get on its own. To dial it in, first turn the channel 3 attenuverter on Maths all the way to the right, and set the channel 2 attenuverter so that it just barely shifts the pitch of the filter with channel 3 all the way open. On my Maths, that takes just an extremely tiny setting on channel 2. Then you can use the channel 3 attenuverter to dial that back further, to get an even smaller offset as needed.
diophantine
 Generally you'd make a fine tune knob by sending 1V (or something similarly small) though an attenuator/attenuverter to a secondary input on the filter. The attenuator knob becomes the fine tune knob. If you only have 1 CV input in your filter, just send the 1V and your 1V/Oct signal through a mixer or voltage processor (I think there is one in Maths?), with the 1V attenuator knob as the fine tune, and the knob for the 1V/Oct signal set to fully open.
sduck
 VCF's are never really good at tracking 1v/oct accurately - you probably don't want to use it for pitch accurate purposes unless you're really desperate. The 1v/oct input is best used for effects purposes, which they're really good for.
colourloss
 I think I'm realising that now. The above tips work for fine tuning but they have the effect of throwing of the 1v/oct tracking. What sort of effects? Other than filter sweeps are possible?
sduck
 The Tomita whistle patch is one example - you can usually get a decent octave or two of reasonably pitched tracking. Like the X-Files theme.
PrimateSynthesis
 I don't know of a simple way to externally make the CV tracking more logarithmic. But you can tune it to your other gear, and get it to track better, by using a combination of summing an offset voltage with a scaling amp. You pick a note (eg. the lowest note you are planning on playing). Adjust the offset so that the frequency matches the CV. Then you pick a note octaves away from the first note, then adjust the scaling until the frequency matches that CV. Then go back to the first note, and repeat.
JohnLRice
 Trying to accurately tune can even be a problem with VCO's that only have a coarse tuning pot (like the Erica Synths Pico VCO etc) If you have the skills, knowledge, and "guts" to do serious modifications that may or may not work and may or may not be reversible . . .you could try to either replace the single turn pots with multi-turn ones OR add additional pots for fine tuning.
wechard
 colourloss wrote: The above tips work for fine tuning but they have the effect of throwing of the 1v/oct tracking.

This shouldn't be the case if your mixer/processor is accurate. In the patch I described above, which is one implementation of the same patch diophantine mentioned (although I forgot the part about leaving the channel 1 attenuverter fully open), you are just offsetting the sequence up or down by a fixed amount. So the frequency ratios among sequence notes will be the same and if your filter is tracking accurately the sequence should sound identical, just transposed. It shouldn't make the 1v/oct tracking any worse (or better) than it was without the offset.
colourloss
 JohnLRice wrote: If you have the skills, knowledge, and "guts" to do serious modifications that may or may not work and may or may not be reversible . . .you could try to either replace the single turn pots with multi-turn ones OR add additional pots for fine tuning.

I have the guts but think I lack the skills and knowledge so will leave it alone for now!

I'll have another play around with the patch suggestions above though and try to get it working. Thanks everyone.
 Page 1 of 1