Re: [Discuss-gnuradio] Power Line Communications - GNURadio

[FredWatrous]

Hi Paul,

I saw your post and thought I would respond.  I found the HackRF a while
back, it's a comparable instrument to one which my company sells for PLC,
just so you know, and I've developed the PC software for displaying the
data.  RF and metering is an area we've done a lot of work in, and SDR is
really improving diagnostics for field work.

By now you've learned that PRIME is a star network, that doesn't use mesh,
and for diagnostic purposes, the strobe beacon is a real pain.  It's
basically the same as token ring if you remember IBM's old network
technology.  It sends out a pulse every few milliseconds and thus tells the
meters when they can talk.

If you setup anything, you need to be able to look at buffers in between
different samples in order to see anything of value, (you need to have a FFT
data rate under 50-60 ms per buffer and be able to look at individual FFT
curves).  Otherwise the beacon will kill any detail that would be of
interest.

Different modes are interesting for lab work and checking performance, but
not really in the field, other than to point to general problem areas.  On
the AC communication side you need to be able to find a particular AC/DC
network adapter in a substation area of 3 - 400 homes / apartments.  (It's
something we also do for various utilities.)  Though this is less
interesting in your case with DC connections.  However, having done a little
TPC and FTT10 between substations, cables running in parallel, (signal
transfer from non-shielded cables or heavily loaded power cable proximity),
and splices can definitely create problems.

For PRIME 1.3.6 you need 1024 samples at 250 kHz, and the signal is between
41 and 89 kHz.  PRIME 1.4 has 4096 samples at 1.0 MHz and uses 41 to 472
kHz.  Good luck!

Sincerely,

Fred



--
View this message in context: 
http://gnuradio.4.n7.nabble.com/Power-Line-Communications-GNURadio-tp59330p59429.html
Sent from the GnuRadio mailing list archive at Nabble.com.