Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: Emissions from 8b10b XAUI/HARI

Hi Henning,

your question is more based on transmission line construction rather
than line coding.  If the true and complement signals are both
of the EXACT same magnitude, and remain EXACTLY 180 degrees out
of phase with each other at all times, the two lines of the
balanced transmission line are tightly coupled and exactly matched
in all routing, AND (this is the important/impossible one)
both lines consume the itentical piece of space, then yes there
will be no radiated emissions.

OK, now that the theoretical parts are out of the way, I'm a bit
surprised at seeing this question coming from Giga, because I
know that your company has some of the best transmission line
design capabilities in the world.  

Getting slightly more realistic, the faster your signalling gets, 
the more difficult it is to keep a signal balanced.  At each point 
that an imbalance is created, you wind up with differential to 
common mode conversion, and these common mode signals are the ones 
that will give you most of your trouble.  Since a ground plane is 
also present, that means that the tranamission lines have both odd 
and even mode impedances, and to terminate them correctly, you 
need to terminate BOTH modes/impedances.  This measn that even 
though you have a balanced transmission line, it takes more than 
just one resistor across the pair to terminate the line.

If you are talking about a cable instead of a circuit board, then
things get worse, because most cable manufactures are willing to 
document their balanced or differential impednace, but you'll
be very hard pressed to find ANY that will also document their
even mode impedance.


Ed Grivna
Cypress Semiconductor

> From: "Henning Lysdal" <hl@xxxxxxx>
> To: stds-802-3-hssg@xxxxxxxx
> Date: Wed, 15 Mar 2000 13:57:44 +0100
> MIME-Version: 1.0
> Content-transfer-encoding: 7BIT
> Subject: Emissions from 8b10b XAUI/HARI
> X-Resent-To: Multiple Recipients <stds-802-3-hssg@xxxxxxxxxxxxxxxxxx>
> X-Listname: stds-802-3-hssg
> X-Info: [Un]Subscribe requests to  majordomo@xxxxxxxxxxxxxxxxxx
> X-Moderator-Address: stds-802-3-hssg-approval@xxxxxxxxxxxxxxxxxx
> All,
> I hope you will have patience with me as I ask a very basic 
> electromagnetics question (and remember I'm a chip guy, not a 
> board guy):
> Assume that you have two transmission lines with a common ground 
> plane with a pseudo CML driver. The signal on one will be exactly 
> the same as on the other except for a very important sign inversal. 
> Wouldn't the emissions from one line cancel that of the other 
> regardless of the coding???
> Henning
> Mileparken 22, DK-2740 Skovlunde, Denmark
> Tel: +45 7010 1062, Fax: +45 7010 1063
> Direct: +45 44546 154
> e-mail: hl@xxxxxxx,