Re: Parallel Port Interface / correction!

Hi,

A schmidt trigger device (an inverter for instace) might have a
logic "1" trip level at 4V and a logic "0" trip at 1V. This is
entirly diffrent from "normal" logic that may have a threshold of
2.5V.

Think of it this way. If you have a signal with +/-.5V noise riding
on it and you barely get above 2.5V, then the noise could take you
thru the thresold several times, giving false steps. For the same
reason it "squares up" slow rising or falling inputs.

On a schmidt triggered device noise is less of an issue since it
would have to drop the signal from 4V to 1V before it would be
noticed.

By analogy a schmidt trigerring is kind of like a microswitch. It
takes a certain pressure to trip the switch and a lot less pressure
to release it. That is to say, you can't "tease" the switch. Either
it is closed or it is open.

On opto-isolators: Never drive an opto with an opto. Optos have very
leisurely rise and fall times, meaning any noise on the line would
have ample time to take the signal thru the threshold several times
before signal would finally settle at a "1" or a "0".

Mariss

--- In CAD_CAM_EDM_DRO@y..., Gary.Rose@c... wrote:
> Doug,
>
> Can you explain a little why the SCHMITT trigger is significant
(like
> what is it)?
>
> TIA
> Gary
>
> --- In CAD_CAM_EDM_DRO@y..., Doug Fortune <pentam@h...> wrote:
> > Correction! Not a NAND (lost my mind apparently),
> > we need an EXCLUSIVE-OR gate as Alan suggests.
> >
> > But the '136 is not a SCHMITT trigger either..
> >
> > So here is an alternative (I can't remember the
> > chip I used last time, can't find my book...),
> > so the suggestion instead is a:
> >
> > Non-Inverted:
> > - 74'341 or 74'344 octal buffer/line driver with Schmitt
> > trigger
> >
> > Inverting:
> > - 74'340 octal buffer/line driver with Schmitt trigger
> >
> > Both have the same pinouts, so if you use a socket, you
> > can just switch chips to change between inverting and non-
> inverting.
> > All three devices are 3 state, but we will just enable them
> > permanently (using pin1 and pin19).
> >
> > Bonus points to Alan for keeping me honest....