Casing the Environment

We have completed our move to Evergreen Woods, at least to the point of unpacking many of our boxes of "stuff".  Time to check out the Electromagnetic Environment. 

It's not all good or all bad. The fact is that we're inside a fairly large complex of 249 apartments and commons buildings.  We are in a rural/suburban setting, but the buildings have complex systems for HVAC, data networks, and power distribution.  So we expect a somewhat challenging RFI situation for HF.  You see one example above.  All the peaks (separation ~106 kHz) seem to come from one source that probably gets here via the AC power line.

I am using a Flex 6500 SDR as an RFI receiver with an EMCO 6502 active loop antenna.  The antenna sits at my new second-floor operating position, not far from some metal objects that probably affect the measurements.  The proposed antenna site is on the roof above the third floor, which is probably (hopefully) a somewhat better location!

Zooming in to get some finer detail, we see three of these peaks in the 40 M band:

A working hypothesis is that these may come from the building elevator system, with the glitches occurring during elevator motion.

This 106 kHz frequency comb is not quite fatal for ham radio, because in most cases you can choose an operating frequency away from the RFI.

There seem to be several sorts of RFI that become visible at various operating frequencies and various times of day.  Some of them are quite broad (no narrow spectral peaks), and some are impulsive, like ignition interference or PLC-type digital signaling.  The Flex 6500 has several noise blanking options that are at least partially effective.

The antenna is probably going to be placed above the flat roof shown below:

A diagonal run of about 100 ft looks possible, using 20 ft masts.

Stay tuned for further developments!

So *you* have radio interference problems?

If you want to work at the National Radio Astronomy Observatory at Green Bank, WV (my stomping ground in ancient times), you better be prepared to leave your technology behind.  It's a little ironic, because they have some of the most powerful computers, receivers, and electronics you would ever hope to use, but all the local equipment is rigorously RFI-suppressed and tested in anechoic chambers before it is installed.

Check out "The Enemy Is Us" (pdf).

Added: NRAO Green Bank is near the center of the National Radio Quiet Zone.  This is a much larger area (13,000 sq. miles) where all radio emitters by law have to be coordinated to minimize impact at NRAO.

U-verse Downs & Ups

Image via WikipediaThe AT&T U-verse saga continues.  I was sad to find that, after my initial tests that showed I could run 80 M CW at 800 W, my latest round of tests showed really bad sensitivity.  I couldn't run over 20 or 30 W without causing the U-verse Residential Gateway (RG) to lose sync.  (That can mean a 1- or 2-minute TV and phone outage.)

After a bit of probing, it became clear that the RFI was entering on the internal home phone wiring.  Disconnecting it from the RG quieted the problem, but alas we are now very dependent on the U-verse VOIP service.  Disconnection is not an option, normally.  (Cell coverage is poor here - a story for another day.)

The AT&T installer had not fully appreciated my phone entrance system, which uses a special DSL surge suppressor that is bolted onto my Single-Point Ground system.  He ran his CAT5 cable from the wall plate straight into the RG. (That's OK for RFI, but not great for lightning surges.)  The house phones were fed via a circuitous route passing (backwards!) through the surge supressor, and generally making a nice RF pickup loop.

This afternoon, I pulled out half my station (the heavy part!) so I could get access, and rewired the phone system, adding ferrite chokes, etc.  Result: no trouble now with 800 W at 3600 kHz.  I'll do more tests later, when the contest goes away.

U-verse RFI tip:  You can use the ubiquitous DSL telephone line filters as an RFI choke.  These are transparent to telephone audio, but block the DSL frequencies.  (It's not clear whether they are effective at up to ~8 MHz that VDSL = U-verse uses, but it seems likely they are.)

More complete info will be at http://aa6e.net/wiki/RF_Compatibility.

More U-verse work

I did some more testing on U-verse vs HF QRO operation today.  The main results are reported at http://aa6e.net/wiki/Uverse, which is going to be where the data from this work is accumulated.  We only give the late breaking news here.

As expected, winding the incoming DSL line on a ferrite core (8 turns, type 77) has helped a fair bit -- allowing a doubling of amplifier power on a good range of frequencies.  We also see that the U-verse DSL does not show any problems for QRO at 20 meters and up.  80 and 40 (and 160) are going the be the troublesome bands.

But we have more countermeasures we can try.  That will be for another day.

I also have trouble with my carbon monoxide detector going off.  (It's very loud!)  And pickup in my computer loudspeakers is annoying on 40 M.  All this will yield to filtering.