From: C. Kumar ([email protected])
Date: Mon Oct 23 2000 - 07:57:42 PDT
Small resistors can really introduce numerical errors
or fail to converge. The primary reason is small
resistors (=large conductances) will reduce diagonal
dominance property of the circuit matrix.
--- "Ingraham, Andrew" <[email protected]>
> Unfortunately, there is a tendency for circuit
> designers to think that a 0
> ohm resistor is more economical (in $, watts, board
> space, etc.) than a 0
> volt DC source, so many people do use the 0 (or
> 0.001 or ?) ohm resistor
> approach. I see it a lot.
> As I understand it, the reality of SPICE is that
> voltage sources are
> mathematically simpler / less troublesome than
> It requires a change of mindset from old ways that
> some people find
> difficult to overcome.
> I personally saw cases where small (~milohm or
> micro-ohm) resistors in
> circuits introduced significant errors. Years ago I
> was using a derivative
> of generic SPICE2 and simulating ECL logic. All the
> voltages "became
> unhinged" when those small resistors were included.
> SPICE converged, but I
> was seeing steady-state node voltages up around +4V
> rather than the normal
> ~-1.4V, even though there was no applied voltage
> greater than 0V. A clear
> departure from reality. With no better explanation
> and little knowledge
> about the inner workings of SPICE, I chalked it up
> to some sort of
> mathematical dynamic range problem.
> Because I haven't seen these kinds of obvious errors
> lately, I am less
> likely to insist on 0V voltage sources, even though
> it is probably the more
> correct thing to do.
> **** To unsubscribe from si-list or si-list-digest:
> send e-mail to
> [email protected] In the BODY of message
> put: UNSUBSCRIBE
> si-list or UNSUBSCRIBE si-list-digest, for more
> help, put HELP.
> si-list archives are accessible at
Do You Yahoo!?
Yahoo! Messenger - Talk while you surf! It's FREE.
**** To unsubscribe from si-list or si-list-digest: send e-mail to
[email protected] In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
This archive was generated by hypermail 2b29 : Tue May 08 2001 - 14:29:50 PDT