CAI,
I have solved a similar problem using the technique suggested by  Steve Weir <[log in to unmask]>  in his message dated Thu, 01 Jul 1999 04:33:39. The way to do this in Ansoft's tool would be to re-label YOUR GND leads as GMETAL and solve for ALL your leads as signals very far from the solver's ground plane.  The tool allows you to "Zero" or "ground out" certain leads after it has completed the field solution.  Grounding out the GMETAL leads will then cause the proper transformation of the mutual coupling values to the values you need.

Aubrey Sparkman
SparkRight Solutions
 
 

CAI ZhongPing wrote:

Dear experts,

Sorry to disturb you again. :)  The following figure explains the case more clearly:

   ________    ________    ________    ________
  |________|  |________|  |________|  |________|
     GND        Signal       GND        Signal

So they are not driven differentially. General tools calculate differential impedance, right? Can just setting the distance to the plane very high get correct result?

Thanks for your help and look forward to your reply.

Cai

**** To unsubscribe from si-list: send e-mail to [log in to unmask] In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP.  si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****