RE: [ng-spice-devel] An Historical note on spice netlist format


To "'ng-spice-devel@ieee.ing.uniroma1.it'" <ng-spice-devel@ieee.ing.uniroma1.it>
From "Gillespie, Alan" <Alan.Gillespie@analog.com>
Date Wed, 7 Feb 2001 16:36:07 -0000
Delivered-To mailing list ng-spice-devel@ieee.ing.uniroma1.it
Mailing-List contact ng-spice-devel-help@ieee.ing.uniroma1.it; run by ezmlm
Reply-To ng-spice-devel@ieee.ing.uniroma1.it


> 
> Can you explain better, what do you mean for a submodel ?
> 

As an example, some bipolar processes have an option for
an extra diffusion in the collector of an npn which
dramatically reduces the collector resistance. This
diffusion will usually have no significant effect on
any other device parameters, except it makes for a slighty
larger collector area.

So the standard bipolar model for the process might define
all the relevant parameters for the standard device (without
the extra diffusion) and the submodel would modify only the
RC ans the CJS parameters.

Possibly this doesn't need to be mentioned on the device
line. The device line could just call a different model,
and the .model card itself would say that most of
it's paramaters are from the standard device, and just those
two are different.


> I think that regularization will help to make a simpler parser. On the
> other end, we may allow for only one unlabeled parameter, this will be 
> not hard to do, since every unalbeled parameter is a number ({1-9}*)
> and a keyword must not start with a number.

Not when we start to allow expressions for values. PSPICE surrounds
all expressions with {} which might be necessary. I'd prefer to allow
expressions wherever you want them without the need for special
delimiters (gobbledigook :-), but I've seen that cause confusion for
some simulators. That's one of the reasons I say we need a language
expert to define a good general unambiguous syntax that doesn't involve
too much gobbledigook.

> > devicetype name nodelist paramlist
>   
>   bsim1      mybjt  3 2 4 1 emitter_area=1uM ...
> 
> Am I right ?
> 

That's what I meant at that stage in the email.

I changed my mind by the end of it, though ;-)

Cheers,

Alan

Partial thread listing: