Message-ID: <14126.200101191713@gemini>
From: "James Moritz" <j.r.moritz@herts.ac.uk>
Organization: University of Hertfordshire
To: rsgb_lf_group@blacksheep.org
Date: Fri, 19 Jan 2001 17:17:20 +0000
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 8bit
Subject: LF: VA3LK on 137.7894
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Dear Larry, LF Group,

Getting the "wrong frequency right", and being able to identify the 
dot length with reasonable accuracy is a pretty tall order to be just 
a coincidence - so I think that must have been VA3LK last night. I 
will dig out some screen shots when I get home.

The most important thing regarding readability of the signal is the 
frequency - if the two carriers on almost exactly 137.790 turn out 
to be a permanent fixture, it is obviously important not to use this 
exact frequency; 0.6Hz low seemed to be in the clear. It is clearly 
neccessary to check how this compares with what people see at 
different QTH's before making great pronouncements on the 
subject, though. 

I will have a look at the frequency calibration again - I had to allow 
for around 2Hz error in the indicated frequency on the spectrogram, 
which is much larger than that due to the receiver, and clearly very 
significant for this type of use. However, drift appears to be 
insignificant. Measuring frequency to +/-0.1Hz in this way seems to 
be perfectly realistic.

As far as the dot length goes, anything over about 60s should be 
fine for me, since the 42mHz resolution is as low as DL4YHF's 
Spectrum Lab goes, as far as I can tell. I think Spectran has 
narrower resolutions, but my aged PC doesn't seem to be able to 
cope with it. Spectrum Lab also has nice image recording facilities. 
I don't currently have any specialised DSP hardware. Some sort of 
distinctive pattern would be nice; but the overall pattern duration 
should be less than, say, 15 minutes in order to take advantage of 
the shorter lived propagation peaks. 

As far as loading coil shelters go, if made of metal then the 
individual parts should be connected together and to ground; but 
there is probably something to be said for not having a continuous 
screen around the coil in order to minimise eddy currents, shorted-
turn effects, etc. The text books say that the distance from the 
surface of the coil to the screen should be at least the coil 
diameter, if Q is not to be significantly reduced. Steel is to be 
avoided due to high losses. Shielding would probably be only a 
benefit where the coil is located near objects with high dielectric 
loss - eg hedges, fences, reinforced concrete bunkers, the ground. 
If the loading coil is well clear of such things, screening is probably 
unneccessary.

I will set the gear up to record tonight, using the same settings as 
before. let's see if it's more than a one-off!

Cheers, Jim Moritz
73 de M0BMU