[sdiy] digital control of CV

jure zitnik kokoon at gmail.com
Mon Sep 3 08:15:33 CEST 2007


do DACs generally need to be followed by a lag circuit? wouldn't a
simple low pass filter (or even just a smoothing cap) on each channel
take care of the clock noise? are there any other noises present when
dealing with multiple DACs on a chip?

jure

On 9/3/07, Robert Shanks <synthlab at nc.rr.com> wrote:
> Hi all  -
>
> > Using Octal DACs rather than a single DAC and sample & hold buffers
> > _might_ be more expensive, but not hugely.
>
> Won't you still need sample and holds or lag circuits/filters for
> smoothingthe DAC output?
>
> Robert
>
> On Sep 2, 2007, at 7:37 PM, Tom Wiltshire wrote:
>
> > Jure,
> >
> >> tom - wouldn't that be much more expensive?
> >
> > Using Octal DACs rather than a single DAC and sample & hold buffers
> > _might_ be more expensive, but not hugely. Farnell Online stock the
> > AD5346 for £3.70 in the UK. This deals with eight channels.
> >
> >> how many CV channels do you plan on having?
> >
> > I'd been working with 64 channels as an upper limit, Eg. 8 chips.
> > This costs (roughly) 8x£4 = £32 or about $60US. Whether I require
> > all of those depends on the final number of parameters in the synth
> > voice, which I haven't decided for sure.
> >
> >> also, you'd probably need more accuracy than 8-bit
> >> for VCO control, right?
> >
> > The Prophet 5 used 7-bit parameter control for most of its
> > parameters, I'm told, with 14-bit resolution for the oscillator CV.
> > I'd thought to use 8-bit control of most parameter CVs, but was
> > intending to use a digital oscillator which would be mostly
> > controlled digitally. For a monosynth, this would probably be a
> > direct MIDI link to the oscillator(s). 8-bit parameter CVs to the
> > oscillator would only control things like Envelope or LFO Depths,
> > fine tuning (+/-100 cents maybe in one cent intervals - smallest
> > perceptible difference is around 6 cents.) or semitone offset.
> >
> >> well, there's also 12-bit version
> >
> > There is, which means you could upgrade the whole thing to 12-bits
> > without having to alter anything much, except make sure you had
> > enough pins on the processor that stores patches. I thought 8-bits
> > would be enough. Certainly for the first try!
> >
> >>  if i understand correctly... this "octal" DACs are
> >> essentialy really nothing more than a single DAC followed by... a
> >> s&h?
> >> do i read it correctly?
> >
> > Well, I don't know exactly what's on the silicon, but the datasheet
> > block diagram shows eight separate 8-bit resistor string DACs, so
> > there's no S&H's of any kind, and hence no droop. That was one of
> > the advantages, together with the low chip count and consequently
> > simple circuit.
> >
> > Regards,
> > Tom
> >
> > _______________________________________________
> > Synth-diy mailing list
> > Synth-diy at dropmix.xs4all.nl
> > http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>




More information about the Synth-diy mailing list