Limiting in Verilog-A


Subject: Limiting in Verilog-A
From: Geoffrey.Coram (Geoffrey.Coram@analog.com)
Date: Mon Feb 16 2004 - 11:44:39 PST


Hi -
This message will serve as a test of the new mailing list
for the Verilog-AMS Device Modeling subcommittee.

During our last conference call, it seemed that we had
gotten a little beyond the existing technology: no one
really knows what limiting is necessary or useful. I
proposed creating $fetlim to access the simulator's
built-in fet limiting algorithm, if it exists. What I
don't know is if I also need to provide an initial guess
for $fetlim to be useful at all. (In BSIM3, there are a
few lines of code that check on the first iteration if
vgs=vds=vbs=0, in which case vgs=vth+0.1, vds=0.1 is used.)

Conveniently, Jim Barby has a set of benchmarking circuits.

What I hope to do is write Verilog-A models for the
devices used (diode, bjt, mos2, mos3 -- Tiburon already
contributed diode to the designers-guide web site, perhaps
they can also add bjt; I'm expecting to write mos2 and/or
mos3 with Jim's help since I don't think Tiburon bothered
with such old models).

Then, I'd like to get $fetlim implemented somewhere so I can
compare dc convergence between Spice native models and
Verilog-A. This comes in two pieces: a Verilog-A
compiler/interpreter and a Spice-like simulator.

On the Verilog-A side, we have ADMS which is open-source,
as well as proprietary solutions from Tiburon or elsewhere.

On the Spice side, it would be a better comparison if we
could connect up with Berkeley Spice so we can use the
exact same limiting algorithms. I think Tiburon was
working on this, but I don't know where they are or how
we can use a proprietary solution as proof of concept.
ADMS connects to zspice, which (a) runs under Win32 and
(b) probably doesn't have limiting. Is it easier to fix
these two points or write a connection from ADMS to
Berkeley Spice?

Comments would be appreciated.

-Geoffrey

-- 
Geoffrey J. Coram, Ph.D.    Senior CAD Engineer     
Analog Devices, Inc.        Geoffrey.Coram@analog.com 
804 Woburn St., MS-422,     Tel (781) 937-1924
Wilmington, MA 01887        Fax (781) 937-1014



This archive was generated by hypermail 2b28 : Mon Feb 16 2004 - 11:48:16 PST