Alt-BEAM Archive

Message #05653



To: beam@sgiblab.sgi.com
From: Sean Rigter rigter@cafe.net
Date: Wed, 11 Aug 1999 20:10:59 -0700
Subject: [alt-beam] Re: To Schmitt or no to Schmitt ...


Hi Bob,

You will notice that Nv cores of more than 2 stages (ie 4Nv microcore) always
use Schmitt triggers. The problem is recurrent saturation problems. Grounded
Bicores (resistor to ground) can use Schmitt triggers or linear inverters (ie
240) and suspended Bicores (resistor between inputs) invariably use inverters
(ie not Schmitt triggers) There are some exceptions to the rule (read Wouter
Brock's Bicore tutorial) and I believe that the 2 diode auto PNC circuit I
described a few weeks earlier would allow using a 240 to be used for a
microcore.I will test this idea out and let you know. BTW the reason why Bicores
can use linear inverters is that they are designed to operate in the saturated
mode. The capacitively coupled positive feedback which occurs in saturated Nv
cores actually acts like a Schmitt trigger not by shifting the input thresholds
but by shifting the input signal levels rapidly through the linear region. This
all has to do with feedback and the concepts of positive and negative feedback
are so important and so universal that they deserve to be written up in a
tutorial form.

Well I hope these examples help you understand some of the reasons for the use
of Schmitt triggers in BEAM circuits and a few other details that no doubt have
puzzled a lot of people before you.

regards

wilf

"Simpson, Bob" wrote:

> Many explanations of microcores claim the necessity of Schmitt inverters,
> but other similar circuits are built without them? In fact, some sites say
> that the Schmitt piece is vital, but then go on to show all the circuits
> built around the 74x240 chip, which doesn't have the Schmitt triggers. What
> gives?
>
> -Bob

Home