[sdiy] Filters

[harrybissell]

Hahaha.... your questions are er.... very non-trivial indeed :^P
inline

Linium wrote:

> On Vendredi 2 Janvier 2004 14:50, jbv wrote:
> > > >Is there a (easy) way to modify this VC-BPF design :
> > > >http://www.cgs.synth.net/modules/cgs35_syntha_vcf.html
> > > >to add VC-Q to it ?
> > >
> > > Replace the Q pot with a LDR+LED (and 1k resistor parallel to the LDR) ?
> >
> > Mmh... I guess there my be a slighly more hi-tech, OTA-based
> > solution...
>
> I have this filter and i am wishing the same feature. Unfortunatly i don't
> know much about OTAs, still deciphering the tuts at
> http://www.uni-bonn.de/~uzs159/index.html
>
> It would be interesting to join our interest to find a solution.
>
> What would be the principle of the operations ? I see that there are two 1K
> resistors on the feedback loop the SteinerParker, the second of them is the
> variable resistor that set the resonnance.
>
> The goal is to use an OTA to replace this 1k pot, will a CA3080 do the job ?

The OTA will need a low level at the input. My first guess would be to remove
the pot... make a resistor divider of maybe 100K and 1K (cut to 1/100th level)
and input this to the OTA non inverting input.  Make another divider lust like it
on
the inverting input...and hook that to a pot from V+ to V- so you can adjust
offset
voltage if you need to...

Remove the second 1K resistor, and tie the OTA ouput to the emitter of the 2N2222

Now the OTA is going to have a hell of a time driving that 2.2K resistor... but
you probably
can't get around that problem. It will either work, or not.

>
>
> How it will be used ? Can we do something as simple as connecting the path of
> the filter to Iabc of the OTA with the current ouput going back in the
> circuit ?
> There are 2 differential inputs on the OTA. I guess that in our case, one (-)
> will be set to 0V, the other would be the CV input (?).

The CV has to be a current... there are many ways to do this, one or more
transistors...
transistor and an opamp... over a limited range you could just put some voltage
higher
than V- in there through a resistor.  You MUST not allow more than 1mA to flow
into
that pin or the OTA goes BYE...

>
>
> Could we connect the filter straigth to Iabc, or does it need a resistor to
> stay in the limit of the OTA ? (how much bias can it get).
>
> What would be the voltage required at V+ to get a gain ranging from 1 to
> something equal to 1K if it were a resistor ?

You can put in 1mA max (safely)... and the 2.2K resistor at the 2N2222 is the
load...
Can you get 'enough' drive... either get out the data sheet and do the math or
build one
and try it.

> Comments appreciated...

I STRONGLY advise building a sucessful, stand alone VCA using the OTA to get the
hang
of it first... before you try and hook into the filter. Check out Joergen
Bergfors site... he has
a bunch of VCA circuits so you can see what would be involved.

H^) harry

>
>
> Linium
>
>