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

Re: New thread on EMI


I just realized you had a lot of bad experience with your 100 tx, a twisted 
pair Fast Ethernet.
This is totally different from the comments I have provided, which are 
optical link.

I also had bad experiences with the twisted pair.  Often, I had to redo the 
EMI design to pass the EMI test.  I did not particularly enjoy it.

All the shielded copper (or twisted pair) cables leak to certain extent.  
They are not perfect shield.  Therefore, for a copper based link, the whole 
EMI headache is shifted to the cable which carries the EMI emission outside 
of the cabinet.  The good cabinet shield does not help at all.  Often the EMI 
reading changes very significantly depending on how you lay the cable on the 
ground.  As you may have noticed, the EMI emission levels of a cable which is 
randomly scattered on the ground, or which is left on a reel are 
significantly different.

We mostly spent our effort in dealing with the cable EMI problem, but not 

I am not sure, in a copper link, how much the scrambled IDLE will provide a 
difference or relief?  I believe the general leaking from the copper cable 
and imperfect connector-shield will dominate your EMI design but not IDLE.

In the 10GbE application, I believe the copper connections will be a short 
one to interconnect stacked switches.  Probably the cable length is a foot, 
or two.  The whole thing can be inside a big cabinet.  I believe those 
mechanical engineers will find some way to solve the EMI problem, if there is 


 Look at 100tx.  There are five components that escape via the port that are 
 clock or clock harmonic related.  These are very nasty and climb quite 
nicely with
 port count.  So I don't agree that the IDLE is not a problem.  We can agree 
 Take care
 Joel Goergen >>


Ed Chang