[RFI] Ethernet gear

[Martin Ewing]

Interesting post by KC2MMI over at eHam -- 
http://www.eham.net/articles/8302 -- about RFI troubles from Linksys 
Ethernet gear.  (Good luck getting any FCC action!)

I've had trouble from some Ethernet hubs and routers, and I've tried to 
minimize 100baseT Ethernet RFI by using shielded Cat5, ferrites, etc.  
Some thoughts -

-There's a lot of variation between computer and network products as to 
RFI generation.  The user has no way of knowing which are the good ones 
in advance.  Part 15 is little help! Is there a source of consumer 
information?  Anyone have real data?

-It's hard to characterize the problem, because there is such a wide 
range of frequency and modulation characteristics.  I have carriers at 
frequencies harmonically related to 100 MHz (harmonics of subharmonics), 
but these are much easier to live with than raspy stuff that wanders up 
and down the 20m band.  10baseT seems to be a lot better than 100baseT 
for RFI, but it's not easy to tell all your equipment to run at 10 MHz 
these days.

-Good antenna and feedline practice helps.  If you have RF in the shack, 
you've also got RFI sensitivity.

-WiFi (2.4 GHz) is a good alternative to Ethernet.  No RFI, no resonant 
wiring, ground loops, etc.  Just polluting a microwave ham band!

Is there any published data?  The ARRL RFI book has a lot about 
"computers" but not much about networks, buses, or all the little boxes 
we hang on them.  For example, does anyone have quantitative spectra of 
the signals flowing in 100bT and 10bT wires?  (I've searched far and 
wide.)  [Not to mention spectra for ADSL, CATV, and -- ulp -- BPL.]

73,
Martin