Looking for hard to simulate MOS circuit(s)

Thread View: comp.lsi.cad

3 messages

[T] Back to Threads [V] Tree View [G] Group Articles

3 total messages Started by Arrigo Benedetti Thu, 21 Jan 1993 16:12

#759

Author: Arrigo Benedetti
Date: Thu, 21 Jan 1993 16:12

27 lines
1741 bytes

I am looking for some simple MOS circuits that are difficult to simulate
with SPICE-like circuit simulators. I need this because I'm implementing
the infamous gmin stepping algorithm in a multilevel circuit simulator
that I'm developing for my thesis. In particular, I'm interested in MOS
circuits that will not achieve convergence with Newton-Raphson, but that
will converge with the gmin stepping algorithm. Our simulator at present
can only do .OP and .DC sweep analyses and handle SPICE LEVEL 1 mosfets,
so please submit only circuits exhibiting convergence problem with these
type of analyses and device model.
What I'm looking for is a simple circuit (<= 10 devices) for which it is 
known that it will get to convergence ONLY with gmin steppping.
I'm aware that circuits with positive feedbacks (e.g. static MOS SR
flip-flops) are hard to simulate with existing techniques.

Descriptions files in SPICE format are welcome.

I will summarize any reply to the Net.

Thanks,

-Arrigo Benedetti
-- 
Arrigo Benedetti             e-mail: arrigo@cube.sublink.org (preferred)
University of Bologna                benedett@deis01.cineca.it
undergraduate student (EECS)         
Via S. Agata, 11               Tel:  (home)   +39 59 224929
41100 Modena - Italy                 (office) +39 59 216688 (fax) +39 59 220727

Re: Looking for hard to simulate MOS circuit(s)

#760

Author: Michele Favalli
Date: Sat, 23 Jan 1993 09:22

6 lines
803 bytes

In the anonymous ftp site "mcnc.mcnc.org", in the directory 
/pub/benchmark/CircuitSim90 there are benchmarks giving 
problems to electrical level simulators (CaZm, SPICE ...) 
and some documentation.                                          

Ciao Michele

Re: Looking for hard to simulate MOS circuit(s)

#764

Author: J.A. Barby
Date: Mon, 25 Jan 1993 03:10

21 lines
1274 bytes

In article <1993Jan23.102232.681@cine88.cineca.it>, michele@deis21.cineca.it (Michele Favalli) writes:
> In the anonymous ftp site "mcnc.mcnc.org", in the directory 
> /pub/benchmark/CircuitSim90 there are benchmarks giving 
> problems to electrical level simulators (CaZm, SPICE ...) 
> and some documentation.                                          
> 
> Ciao Michele
> 

The following CircuitSim90 netlists can be solved using gmin ramping
only:
	bjt/loc
	mos2/e1480
	mos3/counter
	mos3/gm19
The others either do not need ramping or need things more powerful than
gmin ramping.
-- 
 Jim Barby  <jabarby@vlsi.UWaterloo.ca>
 VLSI Group, University of Waterloo, Waterloo, Ontario, Canada, N2L 3G1
 519-885-1211x3995, FAX 519-746-5195

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads