[sdiy] Maxim analog switch ICs

[john mahoney]

Proving once again that there are few new ideas. I should have known that
you have "been there, done that," Juergen!

> But if the OB-X uses larger signals, you need a switch that
> can handle these voltages, *or* some special signal handling
> or level shifting.

Yes, the levels are different. As you know, the OB-X has a discrete filter
much like the SEM (the X has a reduced parts count through the use of FET op
amps) while the OB-8 uses the CEM3320. I suppose this accounts for the
signal level differences. Level shifting was discussed by Tim and Jeff, and
they ruled it out -- it would require at least 2 more op amps per voice
card, we don't know what effect the reduced amplitude would have on the
filter's sound, and it could also add noise and distortion.

> ... Fortunately, today there are
> plenty CMOS switches which can handle 30V. So I'd expect
> no problem if you use one of these.

That's our quest! What do you suggest instead of this Maxim chip?
--
john


----- Original Message ----- 
From: <jhaible at debitel.net>


> I did this modification on my OB-8 some years ago:
> http://home.debitel.net/user/jhaible/hjob8mod.html
> Hey, even sacrificing the "Noise" function sounds
> familiar! (;->)
>
> In the OB-8, you really can use standard 15V CMOS
> switches (positive-only signal voltages),
> because the CEM chips *also* work with positive signals.
> (They only have a small auxiliary negative voltagefor
> the biasing / expo / current mirror stuff).
>
> But if the OB-X uses larger signals, you need a switch that
> can handle these voltages, *or* some special signal handling
> or level shifting. Fortunately, today there are
> plenty CMOS switches which can handle 30V. So I'd expect
> no problem if you use one of these.
>
> Speaking of the Maxim DG409, I just received two samples
> which will be used in a slightly changed version of
> Don Tillman's brilliant trapezoid QVCO. I need the 30V
> feature there, too.
>
> JH.