Hi J, I asked a similar question when I started to get interested in LF, but
I found that the "advice" was often comflicting. It seemed that what suited
one person, one site, did not necessarily suit a different location. Much of
the accepted "wisdom" was based on the construction of commercial LW and BC
transmitting sites. Whilst this is beyond criticism it seems that it often
does not bear much relevance to amateur situations. How many amateurs can
put one quarter wave "radial" (1800 feet long)  out at 136 or even 190kHz
never mind 120?  Another factor comes to bear...most commercial antennas are
verticals....in this case there needs to be a good ground near the base of
the mast and out a distance at least equal to the height ofthe mast. This is
where the capacity of the antenna is "generated", and you want a low loss
capacitor. The situation is different on HF where radials are a more
significant portion of a wavelength. To get it in perspective a 100 foot
"radial" on 136 is like laying a 4 foot long radial on 80m and expecting to
see an improvement.

The Marconi is completely different and not used much by commercial LF
stations after about 1920. A paper by Meissner of Telfunken (?) copied in
the Appendix of the LF Experimenters Handbook (by Peter Dodd G3LDO),
describes experiments at the Geman VLF station at Nauen. The Germans had a
problem in 1920....lack of cash....and could not afford enormous copper
ground systems. Meissner investigated an antenna which a circulat "top-hat"
...like a Marconi spun about its vertical section. He made some important
discoveries that were used at Nauen and promptly forgotten about 10 years
later when long distance comms moved to the HF area. The major finding was
that the maximum "return" current, which after passing through the antenna
capacitance flows back to the transmitter "negative terminal", is not
collected under the feed point but under the remote end of the top-load.
This is not so strange when you remember that this is the maximum voltage
point. It turns out that with a Marconi it is better to "collect" the
current with a star of relatively shallow ground spikes from under the
remote end of the top load. It is also best if the collecting wire runs back
to the transmitter under the topload. There is some evidence that "tuning"
the return with a capacitor can provide a further improvement but I have not
tried this. I am working from memory now but half of the experimental site
at Nauen was boggy and I think he also found that wet ground was a worse
situation than "dry" ground !!.

The amateur situation is different again from this because there is almost
never an antenna in total isolation from lossy surrounding such as trees and
buildings. This to some extent is why there are location differences. I
became convinced that the only way to develope an antenna with minimum
ground loss was to build a simple bridge and measure it. (Details on my web
site) This way you can quickly evaluate changes without confusion due to
power dependent effects like corona when trying to maximise feed current.
You can even avoid having to allow for tuning component losses because you
just measure the antenna a capacity and resistance in series. The radiation
resistance of most practical amateur scale antennas is so small that
effectively the major component of resistance measured is the ground loss.

There is another factor in Marconi antennas that is ot modelled in NEC or
its offshoots. That is the effect of "flat-tops" Most of the "accepted
wisdom" on the Marconi will tell you that once the length of the top-wire is
about equal to the height of the vertical segment there is no advantage in
stringing any more top wire. THIS IS NOT CORRECT. It is true that there is
no appreciable increase in effective height or radiation resistance from
increasing the length of the top wire beyond this value. However in practice
the capacity of the antenna continues to increase at about 5 to 6 pF per
metre. This turns out to have two useful effects. First is reduces the size
of the loading coil required to resonate the antenna, and hence reduces
tuning losses. Secondly it reduces the current density flowing into the
ground. Measurements show, and they are supported by some theory from Alex,
that if you double the antenna capacity you half the ground loss. Since ERP
is current squared, halving the ground loss doubles the antenna current,
giving a four-fold,  6dB, more ERP ( a cheap "amplifier"). The rider on this
is that the top wire must be over "open ground" and not over lossy
buildings, trees or bushes. The extra capacity can be added as a star or
meander configuration it does not seem to matter. Using this technique
several stations over here have reduced their ground loss to about 10 ohms
(Laurie G3AQC, and Finbar EI0CF) but it requires usually in excess of 1200pF
or a total of 240m or top wire spaced at least 1m apart

Once you have cleared this one and you go above 100 watts or so you need to
be looking to reduce "insulator" losses usually corona discharges

That is a bit of a load but I copied it to the group as it may be of general
interest and there really is no alternative to measuring the changes you
make, rather than relying on what works elsewhere.

Sorry to be a bore for those who have seen me exercise my "hobby horse"
before

Regards de Alan G3NYK
www.alan.melia.btinternet.co.uk

----- Original Message -----
From: "J. B. Weazle McCreath" <[email protected]>
To: <[email protected]>
Sent: 30 July 2005 23:53
Subject: LF: Suitable ground?


Hello LFer's,

The soil at my QTH is composed mostly of gravel, which makes getting a
decent ground for RF purposes rather difficult for two reasons, driving in
ground rods, and
poor soil conductivity.  However, about 150 feet from my house is a water
well in
the form of an 8 inch diameter steel pipe that goes down 275 feet!  Has
anyone
on the list had any experience using such as a ground for a Marconi antenna?

Cheers, J.B.
VE3EAR - VE3WZL
EN93dr