[RE: VCOs]

[Bissell, Harry]

The 4011 has the "normal" threshold of about 1/2 Vcc and that comes from the
matching of the n-mos and pmos transistors. The Schmitts are not so polite.
And worse, if you study the data sheets from each maunfacturer, you may get
a big surprise. Some chips do not have the Schmitt trigger as the first
stage in the input, ie the signal goes through a gate first, and then gets
the trigger action. These parts work like shit. I have been burned by some
divider chips that had this problem.  The "A" suffix cmos ase unbuffered,
they have less propagation delay, but also have much lower output drive
capability. Using the CMOS threshold in a circuit where you control the
other parameters isn't too bad, just don't replace the chip or you
recalibrate everything (well y'd have to anyway...)   :-) Harry Bissell

> -----Original Message-----
> From:	Haible Juergen [SMTP:Juergen.Haible at nbgm.siemens.de]
> Sent:	Monday, February 08, 1999 9:06 AM
> To:	Martin Czech; synth-diy at mailhost.bpa.nl
> Subject:	RE: [RE: VCOs]
> 
> 	>As long as the threshold margins are constant, I don't care about
> 	>amplitude.  The schmitt's are quite fast, you have to pay for that
> with
> 	>temp and supply variation (via supply you could adjust the
> amplitude).
> 	>The temp error can be a problem for large temp variations
> (thermostat ?).
> 	>
> 	>If the drive capabilty is nut sufficient one could add a switching
> tranny
> 	>instead of diode, like I proposed.
> 
> There must be something to CMOS thresholds. ARP used a 4011 for the
> threshold in the Odyssey - can't be that bad, then. Lot's of discrete
> stuff
> around that, including a p-channel FET to buffer the saw wave and discrete
> monoflop and discharge transistor - but in the midst of all that a 4011A
> gate to detect the threshold !
> 
> JH.