Re: gEDA-user: Compiling MMI (electromagnetic simulator)

["Daniel Nilsson" <daniel@xxxxxxxxxxxxxxx>]

Karel Kulhavy (clock@xxxxxxxxxxxxx) wrote*:
>
>On Sun, Dec 12, 2004 at 09:29:10AM -0500, Daniel Nilsson wrote:
>
>[...]
>
>> Here's a good collection of links (including fastcap):
>>
>> http://www.fastfieldsolvers.com/links.htm
>>
>> What problem are you trying to solve ?
>
>I am trying to determine how wide a trace must be to have impedance of
>50, 75, 93, 100 ohm and also how much capacitance per length my ordinary traces
>will have.

To calculate the impedance (with reasonable accuracy for slower speed applications)
you can just use impedance formulas such as the ones found here:

http://www.emclab.umr.edu/pcbtlc/

I get asked the question about total capacitance on a trace rather often but I have
still not heard anyone explain to me how that number can be of any practical use. If
you are trying to determine delay on a trace, you have to treat the line in a
distributed fashion. The total capacitance from end to end doesn't help in that
case. Only if the risetime is so slow that the whole trace can be treated as one
lumped element there would be some use of the total trace capacitance and
inductance, but in that case the delay of the trace is so low anyway that it usually
doesn't matter...

For stripline traces, the delay per length is only a function of the dielectric
constant of FR-4 (which is between 3.8 and 4.4 depending on the manufacturer). The
delay for inch is 85*sqrt(Er). For microstrip, the delay depends on the geometry. As
a rule of thumb though, you can assume around 140ps/inch for typical 50 ohm geometries.


/Daniel

http://www.emclab.umr.edu/pcbtlc/