From: "James Moritz" <j.r.moritz@herts.ac.uk>
Organization: University of Hertfordshire
To: rsgb_lf_group@blacksheep.org
Date: Mon, 15 Oct 2001 20:10:15 +0000
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 8bit
Subject: <Tech>Re: LF: Tree Losses
In-reply-to: <5.0.2.1.2.20011015140639.00a2c4a0@mail.pncl.co.uk>
Message-ID: <E15tD4s-0006o0-00@gemini.herts.ac.uk>
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Dear Walter, LF Group,

As far as the practical question "how far away do things have to be 
to make no difference to the antenna loss" goes, my answer would 
be "when they are much further away than the other things which 
are causing losses". Everybody's antenna has ground under it, and 
it is known that LF fields penetrate some distance into the ground. 
Most soil is not a great dielectric, therefore, the losses in the 
ground will be considerable, and if there is nothing else around, the 
ground losses will dominate. This certainly seemed to be the case 
at Puckeridge, where the "small" antenna had only moderately less 
loss in open fields than when set up in my garden, surrounded by 
trees.

Therefore Rik's comments of about twice the antenna height seem 
a reasonable rule of thumb. I found moving my antenna so it was 
several metres, rather than a few metres from trees, buildings, etc, 
made a significant improvement - over a few incarnations it has 
gone from about 60ohms to 30ohms loss resistance. Bear in mind 
the feed point has just as much voltage on it as the rest of the 
antenna, so keeping this (and obviously the loading coil as well) 
well off the ground will help too.

A tuned TX antenna will obviously suffer losses when used for 
receive in just the same way as when it is used for transmit, 
although because of the high noise levels, this probably will not 
make a noticeable difference to the overall SNR. The situation with 
small loops and whips will be a bit different, because the external 
signal/noise to internal preamp noise margins are smaller. I did 
some experiments with my FS measuring gear on the DCF39 
signal - the ferrite rod showed variations of only about 3dB in a 
wide variety of surroundings - trees seemed to make things no 
better or worse. Tuned loops do not seem to suffer significant Q 
reduction so long as they are about a loop diameter or so from 
surrounding objects. I have not yet done quantitative 
measurements with an E-field whip, but data from EMC sources 
suggests that this will suffer much greater screening effects - the E-
field is more seriously affected than the H field, and 10dB or more 
variations can be expected.

An object of a few wavelengths or less between TX and RX in size 
will not cast  a radio "shadow" extending many wavelengths 
beyond the object because the wavefront will diffract around  it, like 
waves passing a ship in the sea. Close up to the screening object, 
there may well be a screening effect - the only place this is likely to 
happen on LF is in a narrow valley, or in a gap between large 
buildings.

Cheers, Jim Moritz
73 de M0BMU