Message-ID: <24606.200101041733@gemini>
From: "James Moritz" <j.r.moritz@herts.ac.uk>
Organization: University of Hertfordshire
To: rsgb_lf_group@blacksheep.org
Date: Thu, 4 Jan 2001 17:37:05 +0000
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 8bit
Subject: LF: Stable frequencies / CFH Plots
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Dear LF group, 

If you have a receiver and signal source which can be tuned to 
198kHz or one of the other stable broadcast frequencies, you can 
compare the frequencies with parts-per-billion resolution quite 
easily using spectrogram. Couple a sample of the signal source 
into the antenna circuit when the receiver is tuned to the broadcast 
frequency, and juggle the level of source and/or broadcast signals 
so they are approximately equal. Hopefully both signals will be 
within a fraction of a hertz of one another, so will appear 
superimposed as a single line on the spectrogram. However, as 
they drift in and out of phase, there will be peaks and nulls in the 
combined signal level. The frequency difference is the reciprocal of 
the period in seconds between 2 successive peaks or nulls, which 
can be measured with the cursor. If you choose an appropriate 
spectrogram time-base, and leave the set-up running for 1/2 an 
hour or so, it is easy to measure difference frequencies of the 
order of 0.001Hz, which is 5 parts in 10e9 at 198kHz. The only 
slight problem is that it does not tell you whether the difference is 
positive or negative, but that can be found by trial and error 
adjustment.

I have tried this with 198kHz; also the Loran "carrier" at exactly 
100kHz works, and might be more convenient if your reference 
signal is at a fixed frequency like 10MHz. Another possibility is to 
use a decade divider chain to obtain 1kHz from the reference; the 
harmonics in the LF range will be strong enough to make the 
comparison to an "odd" broadcast frequency.

I was interested to see Rik's plots of the results from the CFH 
logging session (http://www.qsl.net/on7yd/t181200.htm) - as 
G3NYK suggested, deep fades do occur at different times at the 
different locations. The relative signal levels between different 
receive locations  vary rapidly, so it is very likely that one person 
will be seeing nothing whilst their neigbor is getting good copy of a 
transatlantic signal. This suggests to me that, with marginal signals, 
there is not much point in going to very long dot lengths, or using 
very long "grab" sequences with BPSK software, since the signal 
will probably fade in and out in the time it takes to send a callsign. 
The propagation on LF seems to get more and more volatile as the 
distance increases

It might be interesting if Alan were to make some plots with his 
logging gear while CFH is off-air (or on a different frequency when 
it comes back on again) of the noise level for comparison; after all, 
it is the signal to noise ratio which determines whether a signal can 
be received or not.

Cheers, Jim Moritz
73 de M0BMU