Re: [SI-LIST] : Transmission Line Thy Applied to short interconnects

greg kimball (gkimball@indra.com)
Wed, 24 Mar 1999 23:17:03 -0800

MikonCons@aol.com wrote:
>
> Greg:
>
> Since your original inquiry was on the 18th, you may have dispensed with this
> problem; however, here's my two cents.... and don't take it negatively, as
> it's meant as constructive criticism.
>
> If you don't have a circuit simulator, GET ONE. The questions you posed could
> be answered with confidence in a couple of hours. Go ahead and generate the
> model(s) to include transmission lines, as that's what they are. The T-lines
> will simulate properly for low- and high-frequency AC simulations as well as
> time domain (transient) simulations. You could easily alter the driver
> rise/fall times, driving impedance, and voltage swings to bound worst-case
> conditions. Current modeling programs allow you to step one or more
> parameters simultaneously, so you can quickly optimize the design using a set
> of preferred (or perhaps dictated) parts by trial and error simulations.
>
> You may find no terminations are required, or that perhaps diode terminations
> are superior to R (or RC) terminations for the intended application. But all
> of these options could be evaluated in short order, and confident decisions
> made to speed the design. Plots of stepped parameter runs could be used to
> document (and justify) your design selections.
>
> If your management has refused funding for a simulation package, their head is
> in the sand. With the posture that requires, a swift kick is in order.
>
> Mike Conn
> Owner/Principal Consultant
> Mikon Consulting
>
> PS: If I seem a little testy, the stock market is doing badly today.
>
> **** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at

Mike-
You were the only respondant! Was my question to much on the fringe of
SI - i hope not. Mayby not so mainstream though?

The simulator approach is valid and constructive, and i will do that,
even though the actual modeling procedure is truely involved, unless
i make some wild approximations, with some loss of accuracy. I can do
this part.

I was actually looking for some supportive theory one way or the other,
to confirm or negate my thought process.

Thanks for the inputs Mike, and would appreciate some more feedback from
anyone.
Regards,
Greg

**** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. 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 ****