Re: 8b/10b and EMI
Rick Walker wrote:
> Even at 1Gb/s, the various ground loops intrinsic to this fiddly game
> make it difficult for most designs to solidly meet EMI regulations
> under a wide range of construction practices.
> Combine this with the fact that radiation from a current loop
> is proportional to (Iloop*Area*frequency^2), and you have a problem
> that is 100x more difficult at 10G than 1G. Anyone who had a mild
> headache at 1G will definitely be doing some tough engineering at 10G.
> Rick Walker
I think that good engineering practices can reduce the EMI of spikey
codes like 8b/10b. However, the Maxwell equations still apply, as Rick
pointed it above. Hence, solutions will always be found -
they will only be more costly at higher symbol rates.
Therefore, when looking at the different proposals on the table,
the symbol rate in the fiber and on the Copper traces, and its impact
on EMI, and final yield and cost, can not be ignored.
Hence, the EMI factor will be more expensive in serial proposals
and less expensive in 4-WDM proposals.
And within the 4-WDM proposals, 3.125 Gbaud will be more
expensive, 2.5 Gbaud will be cheaper and 1.25 Gbaud will be
the cheapest one.
The symbol rate in the fiber and Copper traces has also a
similar impact on a separate item: packaging cost.
These considerations should be part of any valid comparison
of the various 10 GbE proposals.
Jaime E. Kardontchik
San Jose, CA 95131