Yahoo Groups archive

Cgs synth

Index last updated: 2026-04-28 23:16 UTC

Thread

cgs30 band pass filter

cgs30 band pass filter

2006-02-15 by Al

Hi there,

New to the forum and taking my first steps cascading some of the
cgs30 filters (the PCBs are pristine btw).

A few questions, hopefully you can put me on the right tracks.
Hope it's not too much to swallow in one post, but here goes

The notes on Ken's site mention that the Rev1 cct had the resonance
pot wired for 'bandwidth' and the Rev2 is altered to give resonance
adjustment. I'd like to try wiring a board with the pot for
bandwidth. Is this a question of swapping over the links on the board
to the pot, i.e. swap the outer connections over or is the track
different to/from the pot?
If it's wired for resonance, does it effectively act to
increase/decrease the resonance with or without affecting the
bandwithand? And if wired for bandwidth, would the resonance be
less/unaltered in any way other than by the Q? (not sure how clear
that question is to anyone else)
Perhaps it's better to ask, can it be wired so the bandwidth can be
adjusted and keep the resonance damped in any way? Some one suggested
to me that changing the 6K8 would help to achieve this, any thoughts?

btw If I'm asking is it 6, or half a dozen? please tell me :)

Also looked at the difference between Reverse Audio Taper (Neg Log)
pots and linear response for the frequency adjustment. The curves for
the two types are very different, how percievable is the difference
and is it worth going the extra mile to track down these elusive
pots? I'm aware of the use of the log curve in relation to how we
hear through studying acoustics, just not, practically, how the
filter would differ.

Oh dear there's more, hope that's ok to keep firing questions like
this - please let me know if not.

Ken shows the cascaded filters wired to the DC mixer cct and I'd like
to know if you could put a pot between the filter and input to the
mixer - the other inputs if used show a 50K pot at their inputs (for
attenuation/gain I figure?). Is it worthwhile to have one from the
filter if you're using them on other inputs? Or would it be
better/possible to link the cascaded filters to a socket and patch
this into the mixer, treating them as discrete modules?

nearly there :?

finally (fer now) would there be any benefit in using neg log pots
for the 50K pots on the mixer inputs (I ask because, looking into the
neg log thing, they seemed to use them in old equipment for volume
pots) and shielded wire through the filter boards (tinned copper bus
wire's the given) and to the mixer to cut down on interferance or is
that *pointless/undoable/over the top (*delete as applicable)

And
Phew, that'll do (I got some PSU q's too but I'm trying to figure
that meself first)

I've been saving 'em up as you can see but I hope you don't mind too
much

Any comments, directions to old threads, or advice on any of the
above are welcome.

Thanks in advance

Al

Re: cgs30 band pass filter

2006-02-17 by sasami@hotkey.net.au

>The notes on Ken's site mention that the Rev1 cct had the resonance
>pot wired for 'bandwidth' and the Rev2 is altered to give resonance
>adjustment. I'd like to try wiring a board with the pot for
>bandwidth. Is this a question of swapping over the links on the board
>to the pot, i.e. swap the outer connections over or is the track
>different to/from the pot?

Simply a case of swapping the outer wires. It was not a true bandwidth
function, instead just labelled so due to the apparent change of bandwidth
when adjusting the resonance.

>Perhaps it's better to ask, can it be wired so the bandwidth can be
>adjusted and keep the resonance damped in any way? Some one suggested
>to me that changing the 6K8 would help to achieve this, any thoughts?

put a single gang pot in series with one of the 100k stages (one will be
backwards from what you want), or use two separate 100k pots instead.

>Also looked at the difference between Reverse Audio Taper (Neg Log)
>pots and linear response for the frequency adjustment. The curves for
>the two types are very different, how percievable is the difference
>and is it worth going the extra mile to track down these elusive
>pots?

not in my oppinion, no.

>Ken shows the cascaded filters wired to the DC mixer cct and I'd like
>to know if you could put a pot between the filter and input to the
>mixer - the other inputs if used show a 50K pot at their inputs (for
>attenuation/gain I figure?). Is it worthwhile to have one from the
>filter if you're using them on other inputs? Or would it be
>better/possible to link the cascaded filters to a socket and patch
>this into the mixer, treating them as discrete modules?

The latter would be more versatile, but if you don't mind the cost of doing
a dedicated mixer for the filters, it would be easier to use. Wire the pots
as per the DC mixer diagram.


>finally (fer now) would there be any benefit in using neg log pots
>for the 50K pots on the mixer inputs (I ask because, looking into the
>neg log thing, they seemed to use them in old equipment for volume
>pots) and shielded wire through the filter boards (tinned copper bus
>wire's the given) and to the mixer to cut down on interferance or is
>that *pointless/undoable/over the top (*delete as applicable)

no. In this case regular log would be more appropriate.



Ken
_______________________________________________________________________
Ken Stone sasami@... or sasami@...
Modular Synth PCBs for sale <http://www.blaze.net.au/~sasami/synth/>
Australian Miniature Horses & Ponies <http://www.blaze.net.au/~sasami/>

Re: cgs30 band pass filter

2006-02-17 by Al

Hi Ken

Thanks for the prompt reply,

one or two points that I'm a bit vague on still, perhaps you can confirm I understand your comments correct.
Show quoted textHide quoted text
>Simply a case of swapping the outer wires. It was not a true bandwidth function, instead just labelled so due to the apparent change of bandwidth when adjusting the resonance.

ok, but if I...

>put a single gang pot in series with one of the 100k stages (one will be backwards from what you want), or use two separate 100k pots instead.

this'll give a 'true' bandwidth function ?

and to clarify...
>put a single gang pot in series with one of the 100k stages

between pin7, from the 1st opamp, and the 100K, or the other side (does it matter)?

and this'd be either that / or, one between the 680R and the 100K, or on the other side (node from pin7 of the 2nd amp)?

>(one will be backwards from what you want),

not quite sure what you mean here

>or use two separate 100k pots instead.

instead of the dual gang on the cct ?

>>worth going the extra mile to track down these elusive pots?
>not in my oppinion, no.

that's good enough for me, phew : )

>The latter would be more versatile, but if you don't mind the cost of doing a dedicated mixer for the filters, it would be easier to use.

cost's not a problem (yet) and I might order another dc mixer board still

>Wire the pots as per the DC mixer diagram.

out of the filter via 50K pot and 47K onto B (below),



or just use one of the input pads on the mixer with a pot right?
>>benefit in using neg log pots
>regular log would be more appropriate.

ideal.


Well, thanks again for the input (no pun intneded), really very helpfull.

Cheers

Al



[Non-text portions of this message have been removed]

Re: cgs30 band pass filter

2006-02-17 by sasami@hotkey.net.au

>ok, but if I...
>
>>put a single gang pot in series with one of the 100k stages (one will be
backwards from what you want), or use two separate 100k pots instead.
>
>this'll give a 'true' bandwidth function ?

It should. I haven't tried it, but essentially a band pass filter is a high
pass and a low pass in series. Moving one of them will change the bandwidth.

>and to clarify...
>>put a single gang pot in series with one of the 100k stages
>
>between pin7, from the 1st opamp, and the 100K, or the other side (does it
matter)?

It does not matter shich side of the pot you connect to, though it will
matter which of the 100k sections you connect to. I think the one mentioned
here is the correct one, but if you find the band getting narrower to the
point where nothing will pass, try putting it in series with the other 100k
instead.

>and this'd be either that / or, one between the 680R and the 100K,

Yes, this is the correct position for the second possible location.


>>or use two separate 100k pots instead.
>
>instead of the dual gang on the cct ?

yes.


>
>>Wire the pots as per the DC mixer diagram.
>
>out of the filter via 50K pot and 47K onto B (below),

there was no "below". art isn't transmitted in this group.

Look at the CGS04 page. Each bandpass output goes to a pot as per the
example pot shown on the CGS04 circuit diagram.


Ken
_______________________________________________________________________
Ken Stone sasami@... or sasami@...
Modular Synth PCBs for sale <http://www.blaze.net.au/~sasami/synth/>
Australian Miniature Horses & Ponies <http://www.blaze.net.au/~sasami/>

Re: cgs30 band pass filter

2006-04-20 by Al

Hi there Ken,

been a while I know - and remote viewer of the forums - but thought
I'd let you know I finally got round to getting the bandpass and
matrix boards up and running with no probs (bar the everyday
distractions and sorting out a power supply) and it's all good : )

trying to get my head round doing some tests (and then maybe some
mods). I'd like to suss the slope and 'Q' out and think I'm on the
right track, but was wondering, if I'm not totally lost, can you
confirm the CGS is a '2nd order' filter?
From reading up a bit about these kind of state variable filters
(yours is essentially one o' these right?), I'm a bit foxed on
whether I should get the slope of 16 or 20dB that I seem to be
reading from my frequency plots. D'you know if that's about right? or
is it more like 12dB and I just can't read a bode plotter right :?

Q looks about 6 by my reckoning - does that sound about right?

Any suggestions on how to figured these if I'm a country mile out
would be gratefully recieved

Thanks in advance

Al

p.s. having admired it all from a far for years, now I'm hooked -
thou the family think I'm gone mad :) - ! I'm enjoying finally
getting my hands dirty - a convert to the cause, who knows maybe I'll
even summon up the courage to start building a whole synth one day ;)

Chrz

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.