Re: [SI-LIST] : RE: PCI crosstalk on Dual stripline

About this list Date view Thread view Subject view Author view

From: Scott McMorrow ([email protected])
Date: Tue Nov 21 2000 - 12:04:39 PST


Doug,

The answer is that you are going to need to perform a
real crosstalk and even/odd mode timing pushout analysis
on all the possible configurations to know the answer. Because
this is a reflected wave bus, there are some extremely complex
wave interactions which occur. I would suggest 5 or 6 coupled
line simuations to be absolutely sure.

There are not any rules of thumb, except don't allow z-axis
coupling. Since you cannot enforce that, you will need to
perform some extensive analysis of this system, which may
include package coupling, board coupling, topological
variations, and other coupling concerns.

When you say PCI, I assume you mean 64bit/66MHz PCI
or possibly PCI-X. In these cases, be very, very careful.
It is extremely easy to break these busses in real systems
with improper add-in card design.

regards,

scott

--
Scott McMorrow
Principal Engineer
SiQual, Signal Quality Engineering
18735 SW Boones Ferry Road
Tualatin, OR  97062-3090
(503) 885-1231
http://www.siqual.com

Doug Hopperstad wrote:

> Here is a question that should generate a lot of opinions: When routing PCI > devices, what rules or guidelines are used for parallel and tandem crosstalk > for the given stackup: (The board material is set for FR408 with a > dielectric constant of 3.8) > > Dual stripline: > ----------------------- Ground Plane (0.5 ounce) > ------ ------- 5 mil trace layer (0.5 ounce) > ------ ------- 5 mil trace layer (0.5 ounce) > ---------------------- Ground Plane (0.5 ounce) > > * The Plane-to-Layer spacing is 8 mils > * The Layer-to-Layer spacing is 10 mils > * The trace width is set for 5 mils (which will provide a 70 ohm trace > impedance) > * The trace-to-trace spacing is set for 5 mils (10 mil pitch) > > The traces are routed orthogonal for the most part. The major problem is > that the routing out of the ASIC is hindering the orthogonal routing and > making the tandem crosstalk a big concern. > > Doug Hopperstad > QLogic Corporation > [email protected] > > **** To unsubscribe from si-list or si-list-digest: send e-mail to > [email protected]. 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 > ****

**** To unsubscribe from si-list or si-list-digest: send e-mail to [email protected]. 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 : Tue May 08 2001 - 14:30:12 PDT