[sdiy] Chopper circuits for audio and CV?

cheater cheater cheater00social at gmail.com
Thu Jun 3 00:45:57 CEST 2021


Oops, typo - s/this/the

On Thu, Jun 3, 2021 at 12:42 AM cheater cheater
<cheater00social at gmail.com> wrote:
>
> This 4067 is roughly $0.7.... that's much less expensive than a
> digital solution! I understand it would need some form of
> anti-aliasing filter before the signal gets multiplexed, and a
> reconstruction filter after the signal gets demultiplexed...
>
> I just wonder how good the 4067 switches are. For example, the
> Nexperia datasheet promises a feed through at 20 kHz of about -20 dB,
> which is bad (I guess it should be at least -60 dB). The OnSemi part
> promises -5dB which is just terrible! The TI part promises -95 dB. I
> wonder how true that is - and whether there might be other problems
> when using them for audio.
>
> Bear in mind that when using a switch in a monosynth, all the switched
> signals are at the same pitch, so feed through isn't a big problem. In
> a polysynth like this, each signal would be at a different pitch, so
> feed through would be immediately noticeable, especially after some
> non-linearities or multiplication (which creates sidebands).
>
> On Wed, Jun 2, 2021 at 11:09 AM Mike Bryant <mbryant at futurehorizons.com> wrote:
> >
> > Tom is right, you'll end up mostly digital.  I built my first polysynth in 1982 using telephone CODECs which I found would sample at twice the 8kHz rate used in telephones.  These multiplexed onto several 2.048Mbps busses to move data back and forth between analogue modules, but those gradually got replaced with Z80s running crude DSP code.
> >
> > And within two years I'd given up on analogue totally for synths other than for input and output ADCs and DACs.
> >
> >
> > -----Original Message-----
> > From: Synth-diy [mailto:synth-diy-bounces at synth-diy.org] On Behalf Of Tom Wiltshire
> > Sent: 02 June 2021 09:36
> > To: cheater cheater
> > Cc: synth-diy
> > Subject: Re: [sdiy] Chopper circuits for audio and CV?
> >
> > It was certainly commonly done with CVs in the internals of many polysynths. Usually one single DAC would be used (the expensive bit) to produce a time-sliced multiplexed CV which then got demultiplexed out to a lot of sample-and-hold buffers.
> >
> > Doing a similar thing with audio would be possible, but the clock would need to be a LOT higher. Take your safe sample rate, and then multiply by the number of channels.
> >
> > It’s a problem that screams “digital solution” to be honest! If you’re going to chop the signal into time slices and then reconstruct it, you’re half way there already.
> >
> > ==================
> >        Electric Druid
> > Synth & Stompbox DIY
> > ==================
> >
> >
> >
> > > On 2 Jun 2021, at 08:58, cheater cheater via Synth-diy <synth-diy at synth-diy.org> wrote:
> > >
> > > Hi all, I was wondering if anyone ever tried using a chopper circuit
> > > to put 16 voices of full bandwidth audio over one 1/4" cable. Also,
> > > the same thing goes for CV. Is that a feasible use case? What are the
> > > biggest problems that could be encountered here? I assume that the
> > > chopper clock can be distributed separately using a 50 ohm terminated
> > > distribution.
> > >
> > > Thanks
> > > _______________________________________________
> > > Synth-diy mailing list
> > > Synth-diy at synth-diy.org
> > > http://synth-diy.org/mailman/listinfo/synth-diy
> > > Selling or trading? Use marketplace at synth-diy.org
> >
> >
> > _______________________________________________
> > Synth-diy mailing list
> > Synth-diy at synth-diy.org
> > http://synth-diy.org/mailman/listinfo/synth-diy
> > Selling or trading? Use marketplace at synth-diy.org



More information about the Synth-diy mailing list