[SI-LIST] Re: Diff Z for asymmetric coupled striplines

The only public-domain 2D field solver (as opposed to demo versions of 
commercial tools) I know of is TNT/MMTL:

http://mmtl.sourceforge.net/


It's not the most comprehensive tool (e.g. can't calculate AC loss, and 
assumes frequency independent parameters), but it's pretty easy to use and 
can do you asymmetric stripline case and much more. If you have only 2 
conductors (other than the ground planes) it will give you differential 
impedance directly. It has a graphical frontend so design entry is 
straightforward. I found it to be quite accurate, usually within about 1% 
compared to other tools as far as impedance is regarded. There are a few 
minor bugs that are easy to work around:

- it does not know that far-end crosstalk (FEXT) can't go beyond -6dB (= 
50%). I.e. if you increase the line length, the reported FEXT grows 
linearly without bounds. Easy to solve, simply truncate any result larger 
than 50% to just 50%.

- ground wires are defined by assigning them a name starting with "gr" 
(not "g" as the manual indicates).

- never use negative y-offsets for conductors. It confuses the solver 
completely even though the graphical display will show the conductors 
where they are supposed to be. Instead, drag and drop them with the mouse 
to the correct layer in the graphical display.

- never ever use a name with a space in it. It will make the file 
uneditable and the solver won't run the file.


Not bugs, but easy to get wrong:

- don't forget to choose the right unit. Default is meters (upper right 
corner of the screen). Not important for impedance, but very important for 
crosstalk.

- You always need to define at least one ground plane. Thickness is not 
important since the solver assumes surface current anyway.

- Don't use too many segments. It will make calculation slow and at some 
limit I found it can lead to numerical instability (i.e. results become 
inaccurate). Typically  between 10 and 50 is a good choice. Always try to 
change the number (e.g. double it), re-run the simulation, and make sure 
the change in the calculated result is negligible. (That's a good approach 
for ANY field solver!)

Hope that helps

Wolfgang







otter30@xxxxxxxxxxxxx 
Sent by: si-list-bounce@xxxxxxxxxxxxx
10/20/2008 09:15 AM

To
si-list@xxxxxxxxxxxxx
cc

Subject
[SI-LIST] Diff Z for asymmetric coupled striplines






Hello,
I need to determine the differential impedance of pair of coupled 
stripline traces. They are edge coupled and are asymmetric in the sense 
that the distances to the 2 reference planes are different - both traces 
are
4 mils from the upper plane and 7mils from the lower.

Our pcb fabricator told us that this configuration would have a diff Z of 
100 ohms, but I have gotten a value from the cross section calculator in 
Cadence Allegro of 79 ohms. I'd like to find another method to settle 
this, as our board does have SI problems.

The diff Z formula for this configuration isn't in any of the well known 
SI or TL texts, although the symmetric case (H1 = H2) is. Can anyone 
either:
 1) Point me toward the correct equation 
 2) Suggest a free field solver that would let me calculate this value?

I've downloaded the student version of Ansoft Maxwell, and am trying it 
out. If it can extract the mutual 
capacitance and inductance, is there a general method of getting the diff 
Z from these?

Any help is greatly appreciated,
Jim


------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field


List technical documents are available at:
                http://www.si-list.net

List archives are viewable at: 
                                 http://www.freelists.org/archives/si-list
or at our remote archives:
                                 
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
                                 http://www.qsl.net/wb6tpu
 




------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field


List technical documents are available at:
                http://www.si-list.net

List archives are viewable at:     
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu

Other related posts: