Hi Hermann,
Regarding the tools, it would be convenient to use Hyperlynx Touchstone Viewer.
To see how the signal propagates from port 1 to 2, you can exclude ports 3-5 by
letting them be "non-connected". No circuit simulation is needed. In the
resulting s2p file you'll get the "usual" insertion loss, with resonances
caused by the stubs to existing but already invisible ports 3-5. Then, repeat
this operation by eliminating ports 2, 4 and 5 as "not connected" and so on. If
needed, you can also re-normalize the remaining ports to any convenient
impedance value, or let them be 50 Ohm.
We have used to think about connectors in terms of "insertion loss". However,
this logic works only when you have a conductor that connected two ports, and
other ports being relatively isolated (e.g. with their respective conductor
coupled to the first). If we add more ports at either side of the conductor, or
even split any port into several connected to each other, all the dependencies
in S-parameter matrix are drastically affected and the "insertion loss"
thinking doesn't apply.
Perhaps, this is what caused difficulty in interpretation.
Of course, passivity and causality should be in any case, regardless of the
port number.
Vladimir
HI Scott,
yes, this is an option, but I think very cumbersome.
I setup this once in ADS with automatic switching .. but also this
requires a lot of effort.
I would still hope that tool vendors would see this as important task
and bring in some automatism for this ..
Hermann
EKH - EyeKnowHow
Signal Quality - Made in Bavaria
Hermann Ruckerbauer
www.EyeKnowHow.de
Hermann.Ruckerbauer@xxxxxxxxxxxxx
Itzlinger Strasse 21a
94469 Deggendorf
Tel.: +49 (0)991 / 29 69 29 05
Mobile: +49 (0)176 / 787 787 77
Fax: +49 (0)3212 / 121 9008
