Breakdown Filter Module

[TekniK]

yes its a lot of stuff to do one thing if u build it 'closed' in one box,but hes filters HAS all the patchpoints that could be bring out on the frontpanel,so u could insert fx or delay at different steps,and an elaborate sequenced vca morpher pcb that could be used also separately invites other posibilities,u can throw in what u want.

[fonik]

very good points, you both.

since i wanted to build the "one trick pony" without any modular approach intended i think it is oky, this way.

BTW the filter boards provide additional inputs, so that i could use them as single filters (resonators ). these single inputs and the step switch that limits the counters steps are not on the test pannel though.

i think i will have to take a closer look into the simple VCA thread (single transistor VCAs?).
and hasn't there been a thread about using FETs as variable resistors? i did not want to use 8 LM13700 for the ARs...

[TekniK]

don't know ,could a waveshaper replace an ar function ?

[fonik]

[Uncle Krunkus]

Yeah, I reckon the single transistor VCAs sound good.
Maybe just 1 dual gang AR pot on each step, with one switch to shut off the Attack or Release or neither for that step/AR/filter.
That way you've just got 8 pots and 8 switches.
But then you still need to control trig lengths hey?

[fonik]

i actually would have wanted the ARs and VCAs on the control board, with one common attack and one common release control. i'd liked to keep it simple, not too much access on each single filters parameters. i fear the feature creep and i want to use my already etched filter PCBs. and maybe even fixed attack. i am actually happy with it as it is, but i don't like the pops!

[TekniK]

[TekniK]

[Funky40]

[TekniK]

Offtopic:funky40, i have a dual SEM multimode with VCQ pcb layout ready.

Its single sided ,allthought not tested it should be fully functional.

[fonik]

i get you points, funky. however a SEM-style with all bells and whistles would be another project, and IMHO strictly private DIY, because:
http://www.retrothing.com/2009/06/tom-oberheim-new-sem.html

in fact it is the switching that gets my attention now. it shall be driven by clock (which is NOT overlapping), but it should cross-fade. that's the trick i want to solve here, i think. a sequential switch, that produces no pops. therefor i want it to be on the controller board, and therefor i want it to have just one or two controls (Attack/Release).
i get it resolved it would be very cool to use it with 8 filters, just as teknik mentioned: x-fades between the filters with different cutoff, resonance, mode, and pan settings...

[Funky40]

yup, i get your point too.

[diablojoy]

I might not be getting my head round this correctly so please ignore this post if it is of no help but could you allow selective outputs through with vactrols AKA auto level to smooth the response that way it may be possible to get overlap[a held voltage on the vactrol control circuits ,same for all, controllable via one pot], however the trade off is speed of course . it may not be quick enough for somethings and you would lose separate A/R per step, but you do gain the ability to use CV to decide which filter , when and for how long. it could be run off a simple sequencer 0 -1 volts= filter 1, 1-2 volts= filter 2, 2-3volts=filter 3 and so on. the voltage quantized to 1 volt steps and run through comparators and logic gates and then to control ccts for the vactrols to do this.
the sequencer could supply a single A/R with the gate signals then or even another gate source could be used. a single VCA is then only
required as well. if the A/R is also voltage controlled you can gain back the A/R per step from the same or a different sequencer
yes i know it does mean at least 8 fast vactrols at more than $5.00 each
however i think that would be simpler and cheaper than 8 A/R's and
8 VCA's and less introduced noise and a lot less knobs
anyway may be something to consider

[fonik]

sounds very interesting, indeed. however my filters outputs are stereo, so this would require 16 vactrols - unless i would place them right before the pan control, but then i would touch the filter boards again. i did not wanted to do that.

i just tried a 2n5486 FET as a voltage controlled resistor in a spice model simulation. i did not get it working right. i tried to follow the description in the Siliconix AN105, but my approach is more trial&error

[diablojoy]

yes FETS would be quicker and cheaper to implement but like you
I would be doing trial and error. even more so as i know nothing on spice
modeling. I seem to remember reading about FET''s replacing vactrol's
for CV control in an echo module in a post on this site that may offer a helpfull clue in that direction
16 vactrols and associated circuitry would be getting a bit much perhaps
missed the stereo bit .
ah yes there try looking up scott bernardi's
pt2399 vc delay post

[fonik]

thank you very much for the hint. i will look into this thread.

thank you all for your kind suggestions. you all are very helpful. e-m is an incredible place, isn't it?

[haxster]

Fonik did you notice the floppy action when you used the Single Pot grabbing technique with the alpha pots? It makes a grab but the pcb will kinda swing side to side. Perhaps if you ever do that approach again make yourself a template/library for the Dual Alpha Pot even though your not really using the other side of the pot. When strategically located even (1) dual pot makes a sick stable grab, no more swing. But I guess when it is all done and setup no one is really going to swing it on the back, unless you have mice or giant centipedes.

Those ribbon cables accross blow me away! Mad respect.

[mono-poly]

Mathias this is really sounds great man i hope you come up with a solution.

[fonik]

[mono-poly]

SDS3!