From: "Peter O. Brackett"
Subject: Re: Ebers-Moll equation
Date: Sat, 30 Nov 2002 12:02:06 -0500
Organization: MindSpring Enterprises
X-Server-Date: 30 Nov 2002 17:06:38 GMT
X-Newsreader: Microsoft Outlook Express 5.00.2919.6600
"Paul Burridge" wrote in message
> On Fri, 29 Nov 2002 22:18:52 -0500, Phil Hobbs
> opined thusly:
> >I certainly wouldn't want to rely on Ebers-Moll SPICE models, but then I
> >wouldn't want to rely on Gummel-Poon for dreaming up new circuits, or
> >figuring out how they work. Would you?
> Which model does SPICE utilize?
Since Gummel-Poon is actually a superset of Ebers-Moll.
Gummel and Poon [1970's] "merely" elaborated on the very useful and
isightful model of Ebers
and Moll [circa early 1950's], if you leave out some of the G-P parameters
you get the E-M model.
Anyone who espouses to know how BJT's operate had better commit Ebers-Moll
to memory, and
be able to run it through it's paces with real numbers mentally. There are
only a very few parameters
and relationships to memorize and it only takes a Sunday afternoon to do so,
it's not too heavy to
carry around and you will amaze everyone with your command of BJT
operations, and then you are
on your way to joining the elite society of the "lighning empiricists".
Try that with Gummel-Poon.
Bob Widlar [The king of BJT linear design, RIP], Barrie Gilbert [The
translinear guy.] and many others
have show us all "the way" using no more than Ebers-Moll.
Indialantic By-the-Sea, FL.