Content-Transfer-Encoding: 8bit
Message-ID: <5.1.0.14.0.20030519161314.00acb0d8@gemini.herts.ac.uk>
Date: Mon, 19 May 2003 17:53:52 +0100
To: rsgb_lf_group@blacksheep.org
From: "James Moritz" <j.r.moritz@herts.ac.uk>
MIME-Version: 1.0
Subject: LF: Earth experiments
Content-Type: text/plain; charset=us-ascii; format=flowed
Sender: <owner-rsgb_lf_group@blacksheep.org>
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

Dear LF Group,

In the discussion over transmission lines, G3LDO said:

 >Just a thought. Where should you connect the system to ground if the
 >two components are some distance apart? Conventional wisdom suggests
 >it should be under the loading coil.

It seems to me the earth connection should be where the return current 
flowing through the earth has the shortest, lowest resistance path, ie. 
directly under the antenna where the field is greatest. In an effort to 
test this, I did some experiments over the weekend.

The antenna arrangement here has an inverted L antenna, with a loading coil 
and matching transformer about 10m from the shack, fed through about 15m of 
50ohm coax. The ground consists of 6 x 1m copper pipe ground rods, 
distributed within about 3m radius of the loading coil. The shack end of 
the coax was also grounded through various ground rods around the house, 
plus the mains earth and gas and water pipes. With this arrangement, the 
total loss resistance of the antenna/earth combination was 36ohms. 
Disconnecting the antenna ground, so that all the ground current had to 
flow via the shack ground resulted in 37 ohms loss resistance. Isolating 
the shack ground using a common-mode choke in the coax feeder, so all the 
current flowed in the antenna ground, resulted in 37 ohms loss resistance 
too. With both grounds connected, an RF ammeter showed that about 90% of 
the ground current flowed through the antenna ground rather than via the 
coax braid.

The shack and antenna grounds are both at one end of the 40m long wire 
antenna, so most of the current has to flow through the ground for some 
distance before reaching either earth connection. But if more ground rods 
are added under the length of the antenna further from the feed point, very 
little RF current flows in them. I think this is because the distributed 
inductance of the wire connecting the more distant ground rods to the 
common grounding point has a reactance that is large compared to the 
resistance of the ground connection. So I tried a "star" ground system 
consisting of 11 ground rods distributed around the centre of the antenna 
span with a radius of roughly 13m. Each rod was connected using 13m of wire 
to a central point insulated from ground, which was then connected back to 
the low-potential end of the matching transformer with a bus wire. In this 
way, all the ground current had to flow through an equal impedance, 
whichever ground rod it passed through, and checks with an RF ammeter 
confirmed the current in each ground rod was indeed roughly equal. With the 
shack ground isolated as before, the loss resistance was now 33 ohms, so a 
modest 11% improvement in antenna efficiency had been achieved. The loading 
coil inductance also had to be reduced slightly due to the added inductance 
in the ground connection. I then tried disconnecting some of the rods, to 
see how many were actually needed - the results were:

11 rods - 33R
8 rods - 34R
5 rods - 36R
3 rods - 41R
2 rods - 44R
1 rod - 55R

So the resistance associated with a single rod is about 20 - 25 ohms; there 
is not much to be gained by having more than 5 rods. The best that was 
achieved was only about 2dB better than using a single ground rod.

The ground at my QTH is a thick layer of clay, which seems to be 
waterlogged most of the time, so I imagine the ground conductivity is quite 
high. I expect grounding arrangements would have more effect where the 
ground conductivity was lower, and the loss resistance was higher to start 
with, or the antenna was very large and had a lower loss resistance - does 
anyone have any experience of this? If you have similar ground to mine, the 
moral of the story would seem to be that the location of the ground system 
is not very critical provided it is reasonably close to the antenna, and 
that there is little to be gained by having more than about 5 ground rods, 
unless the loss resistance of the antenna is very low to begin with.

Cheers, Jim Moritz
73 de M0BMU