AW: [sdiy] Re: OTA nonlinearities

[harry]

Hi Martin:

Well I want to start with... sum all the channels together and then use one pot
as a master.

OK so I'm a dickhead today....   ;^)

Seriously:  The LED solution can be quite good if you use two stages
of attenuation... like a PI network...  A dual cell can do this easily.

NOT so easy for the FET or LDR is any kind of matching attenuation of
many channels at once, or forcing them all to a particular curve.

There are some really high performance VCAs that have appeared in app
notes... esp National Semiconductor, and also EDN magazine etc. They usually
use discretes for the most critical parts of the circuit.

The digital pots are usually zipper-noise magnets !!!
I'd likely start with a linearized CA3280 and work up from there.

H^) harry

Czech Martin wrote:

> FET and LDR solutions have usually low dynamic range.
> I don't know what is required here, but even 60dB attenuation
> may be too much crosstalk. You can hear this in silent parts
> of your programme, somebody or something talking into
> is very anoying. If the digital pot has equidistant taps (i.e. linear)
> this may be not so usefull for audio, the last steps are always too coarse.
>
> m.c.
>
> > -----Ursprüngliche Nachricht-----
> > Von:  Anderson, Robert O [SMTP:RobertOAnderson at eaton.com]
> > Gesendet am:  Mittwoch, 9. Mai 2001 14:53
> > An:   'synth-diy at node12b53.a2000.nl'
> > Betreff:      RE: [sdiy] Re: OTA nonlinearities
> >
> > The usually use a FET on one leg of a voltage divider.
> > The voltage divider has less distortion than OTAs and is suitable for VCA
> > and automated mixer applications.
> > Regards, Bob A
> >
> > -----Original Message-----
> > From: Nils Pipenbrinck [mailto:np at inverse-entertainment.de]
> > Sent: Wednesday, May 09, 2001 7:19 AM
> > To: synth-diy
> > Subject: [sdiy] Re: OTA nonlinearities
> >
> >
> > Thanks for your replies (especially thank you, Jim.. your OTA analysis was
> > very interesting to read!)
> >
> > 0.5% THD for a OTA based VCA is a bad performer (in my eyes).. It might be
> > ok to have such a distortion on mono synth-waves. I think the distortion
> > will be hearable as soon as I route a complete audio signal (read: music)
> > trough the OTA.
> >
> > I wonder what options I have to build a better VCA... I looked up the
> > datasheet of the SSM 2024, and it's not that much better.. the THAT-Chips
> > look like a perfect choice, but they are a lot more expensive.
> >
> > There are also a lot of 4-quadrant analog multiplier chips.. I never
> > dealed
> > with them, but it shouldn't be that hard to turn them into VCA's..
> >
> > I even considered dropping VCA's all together and go for digital
> > potentiometer chips.. Anyone has experiences with them? They are no VCA's,
> > but suitable for my application (some kind of mixer-matrix for hifi audio.
> > I
> > want to control more than 2 channels with one pot, so stereo-pots won't do
> > the job). Of cause I would need some digital stuff around my module if I
> > choose the digi-pot way, but I would get midi-support and presets for
> > free.
> > Stuff that's always nice to have, but not really required for me.
> >
> > Does anyone has another idea how to control lots of audio channels with
> > just
> > one pot?
> >
> >
> > Nils Pipenbrinck
> >
> > BTW: I'll try out the asm-1 VCA, just to get a feeling how bad 0.5% THD
> > really is.. I mean.. I'm listening to my music with a self-build tranny
> > headphone amp, and I doubt, that it's THD is below 0.5% or so. I don't
> > care
> > about that.. my amp somehow "flavours" the sound, and I like it that way..
> >