My Actiontec GT701-WG router pretty much obliterates the 17m band when its
radio is active. It's probably causing some interference on other bands,
but 17m is pretty much useless when it's active. Changing the router's WiFi
channel has no effect on the noise.


Seems a bit of a jump in frequency to make that much of a difference?

It may be power-supply-related noise, or noise from some of the
baseband digital circuitry.

I've run into (and helped diagnose) a couple of significant
interference problems involving consumer-grade networking gear:

- Significant "birdies" on the 2-meter band, appearing every few
tens of kHz, with no modulation apparent. One of them sat right
on our city ARES group's preferred 2-meter simplex channel, and
prevented my city EC from using that channel from his condo.
Snooping around with a spectrum analyzer and a hand-held beam
pinpointed the source as his neighbor's condo (second floor, in the
back). This turned out to the location of a Netgear WiFi fouter.

The noise seems to have been harmonics from the on-board switching
voltage regulator. It was apparently coming out directly from
the board itself as RF radiation (through the case) and not
as conducted noise on the power-supply cord to the "wall wart" or
on the Ethernet cables... those were all adequately choked or
filtered.

Swapping out the router for a slightly different Netgear model
cured the problem... different switching-regulator circuitry, I
suppose.

- A nasty "buzz" audible on the squelch tails of conversations on
several local 2-meter repeaters. The interfering carrier
"wandered" across the input frequency range of the repeaters, was
strong enough to trigger the repeater receivers' carrier-sense but
didn't have a PL tone on it.

The source of the signal had been DF'ed and localized to an
apartment complex (about 15 miles from one affected repeater!) but
the original hypothesis that it was coming from a defective
roof-mounted TV antenna preamplifier hadn't panned out.
The spectrum-analyzer-and-beam technique identified the specific
apartment it was coming from, and the apartment owner volunteered
to have us check out his computer system. We confirmed that the
pulsing interference signal occurred only when he was downloading
content from the Internet. When we unplugged the 10BaseT hub, the
interfering signal went away.

The noise appears to have been a harmonic of the hub's onboard
crystal oscillator frequency... a "fast edge" of the Ethernet
clock was somehow being amplified, and leaking out onto the
wiring (it seemed to be present on both the 10BaseT cables,
on the wall-wart power wiring, and in the AC mains wiring!).
Somehow, enough of this signal was being radiated by the building
wiring to affect repeaters several miles away!

The apartment owner agreed to trade out his defective hub for one I
provided from my pile of spares (he got a few extra ports out of
the deal), and the problem was gone.

So, yes, there are all sorts of ways that a defective network
device can emit spurious signals... often in frequency ranges which
seem far separated from the equipment's nominal frequencies-of-use.


--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!