wasp (was: Re: [sdiy] Comparator = Opamp ??)

[jh.]

> Well, didn't the Wasp filter kinda do that? I mean, by using 4069UBs to
act
> as opamps?

Not quite. One 4069 inverter is no more than two MOSFET transistors, which
could be configured as anything: Digital inverter, linear amplifier,
comparator
(several ones in series), voltage controlled resistor (leave vcc pin
floating).
We're close to discrete design here.


> BTW, anyone ever build the Wasp filter with opamps instead of 4069s? Just
> wondering what effect is has..

Would be just an ordinary State Variable Filter. I'm convinced that Chris
Hugett
has used unbuffered inverters for the amplifiers because of the low cost,
not
because of any sonical feature, in the Wasp. Linear application of
unbuffered
inverters should not have been unknown by then - think of crystal
oscillators.
For audio, if you compare the open loop gain of a CMOS buffer with a
cheap 0.5V/us opamp, at DC the CMOS buffer will loose, but at 15kHz
things are different. Think of it as a discrete design, a one stage
amplifier,
no compensation needed for stability -> fast !

> I realize that a good portion of the unique sound of the 4069ubs is what
the
> Wasp filter is all about,

I'm not so sure if the Wasp filter inside the Wasp got so much coloration
from the 4069's. There's a lot of other unusual things in the wasp, just
look
at the staircase type saw waves (which are also used for PWM). This
could have a bigger effect on the sound than the 4069's nonlinearity.
If you take that filter *out* of the Wasp, however, to use it for complex
signals (rather than two raw VCOs), the unique distortion character
becomes prominent. Especially when you're adding some extra distortion
stages from the remaining buffers of a 4069 package.
(There were some guitar fuzz boxes designed around CMOS chips
as well, but I don't remember which ones.)

JH.