RE: [SI-LIST] : Question on propagation delay........

About this list Date view Thread view Subject view Author view

From: Ingraham, Andrew (Andrew.Ingraham@compaq.com)
Date: Mon Apr 17 2000 - 12:37:50 PDT


>Here is what I thought was a simple question but
>has turned into a discussion of some debate. Three
...
>as different widths. That being said, what is the flight
>time of these topologies as a function of length. Is it
>strictly a function of the dielectric?
 
There is propagation delay along the traces, and then there is the effective
signal delay from point A to point B, measured at any arbitrary point (say
50%) on the waveform. Don't confuse the two.

The propagation delay (or velocity, if you will) along the traces is
determined by the dielectric. Also to a much lesser degree by the
conductors, but we usually assume "perfect" conductors for this calculation
which causes little error.

The effect of trace impedance (i.e., trace width), source impedance, load
impedance, and discontinuities (vias) may cause the effective signal delay
to be slightly or very different. This is obvious if you consider a short
trace with a big 1000 pF capacitor on the end of it; the effective delay of
the net would be much greater than that of the trace alone. (But one may
ask, how much of this delay is flight time and how much is source/load
interaction? However, if you consider a line driven by a risetime that's
much smaller than the electrical length of the line, then if you measured
the delay between 50% points, the long delay would really appear to be from
the input end to the output end of the trace.)

The effective signal delay is a function of dielectric alone, only in the
ideal case of a matched impedance at one or both ends and no
discontinuities.

Along any uniform section of transmission line, we can break any waveform
into two components: one moving forward and the other reflected back. Each
component itself moves along at a velocity determined by the dielectric.
How the two components add up at discontinuities, and at the load end, will
determine the effective signal delay.

Andy

**** To unsubscribe from si-list or si-list-digest: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Apr 20 2000 - 11:36:12 PDT