Mail Archives: geda-user/2012/11/19/13:38:00
John Doty:
> On Nov 19, 2012, at 7:57 AM, Karl Hammar wrote:
>
> >> On Nov 19, 2012, at 6:50 AM, Evan Foss wrote:
> >>> Oh when I used Omega on resistors the old ones would fail. SDB might
> >>> have fixed it but there was some reason that I can not remember for
> >>> not using it.
> >> How should "mΩ" be translated?
> > milli Ohm.
> SPICE won't understand that (can't have that space).
Sorry, I jumped in the middle of thread.
My gut feeling is then that 10 mΩ should be translated to 0.01 and
and 10MΩ to 10000000, and to leave out the multiplier in the output.
But from what I understand that would be a change in behaviour,
and they are never fun.
I'd say, keep current unless you specify a e.g. -SI option to get
the above.
> In SPICE, either "m" or "M" is "milli".
> The actual unit is a comment, so SPICE will accept 1000MHz as the
> value of a one ohm resistor.
> But Unicode is troublesome in a format based on Hollerith encoding ;-).
Are you saying that we need to put a "H" i col. 6 also?
> >> What about "MΩ"?
> > mega Ohm.
> That would be "meg" in SPICE. The challenge here is to avoid
> translating "M" spuriously.
We are talking syntax of units/values, so given attribute
"value=10mΩ", accept whitespace before or after the number but not
between multiplier and unit. If something else than a number is last
it must be an unit. Accept Ohm as an alias for Ω, u for µ, else keep it
strictly SI-style, that's my take. Shall I write a prototype parser
for testing?
> > Some people write KΩ, and what is a Kelvin Ohm ?
> It's wrong. Some people write mHz for megahertz, that's wrong too.
> I'm sometimes a pulsar astronomer, so I actually use mHz for
> millihertz.
Yes, and I recently bought a signal gen. and one req. was mHz range.
> > Don't do that is the answer.
> >> Should "m" by itself represent "meter" (SI) or "0.001" (SPICE).
> > m by itself is just a lone m, ask the author what he means by it.
> > A number plus m as in 10 m is 10 meter.
> > If spice treat 10m as 0.01, I would call that a local anomaly.
> We're talking about SPICE netlisting here, and under what
> circumstances (if any) the netlister should attempt to translate
> standard SI notation into its SPICE equivalent.
Don't break current behaviour, add new options to get any changed
behaviour.
> Asking the author is not a practical option in an automated process.
Maybe one shouldn't put a lone m there then.
Regards,
/Karl Hammar
-----------------------------------------------------------------------
Aspö Data
Lilla Aspö 148
S-742 94 Östhammar
Sweden
+46 173 140 57
- Raw text -