Message-ID: <3799079F.172A@xtra.co.nz>
Date: Sat, 24 Jul 1999 12:23:59 +1200
From: "vernall" <vernall@xtra.co.nz>
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
Subject: Re: LF: Ground systems
References: <m117igN-0002SpC@fwd07.btx.dtag.de> <3798E3DD.2907@xtra.co.nz>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Hi all,

I missed one other piece of information, which is now added:  Kevin
ZL4MD uses link coupling from his LF transmitter, which isolates the RF
output, so there is no path for the mains (safety) earth to travel via
his RF feeder to the LF antenna.  That is how he keeps the RF and mains
earth separate.  Unless the two are electrically separate, and known to
be safe, the experiment of connecting or removing a link between the two
earth systems is not available. 

Bob ZL2CA

vernall wrote:
> 
> The comments on ground systems are very interesting and I have some
> contributions:
> 
> For a single vertical LF transmitting antenna, the most natural path for
> a radiating current is perfectly radial to the insulated base of the
> vertical, across the surface of the ground.  Placing bare wire on or
> just under the ground, as so-called "radials" is a proven way of
> improving radiation efficiency.  The problem is in the number and length
> of radials, which generally gets beyond the scope of an amateur antenna.
> 
> For a typical amateur LF vertical antenna, the first wires in are the
> "best value" and they generally improve efficiency no matter how ideal
> or non-ideal they are placed.  Metal stakes at the far ends are also
> likely to "earn their keep".  On adding more and more "ground
> connections" there is a situation of diminishing returns (probably
> logarithmic?) and also situations can arise where natural current paths
> (perfectly radial) may conflict with forced current paths by wire
> radials installed in practical ways that involve corners, bends, etc,
> such as dodging a house or concrete yard!
> 
> I have a practical situation to report where actual measurements were
> witnessed by myself and Kevin ZL4MD, when I was visiting his Central
> Otago QTH in February.  Kevin has a farm, so has opportunity to have a
> bigger T antenna and longer radials than many others who experiment on
> LF.  He had run lengths of surplus power line out as radials, from a
> central earth node at the base of the vertical.  The radials were on the
> surface or slightly below ground.  He also could connect to the mains
> earth via a low loss large value polypropylene capacitor, with the
> intention to connect the electricity mains earth in parallel with the RF
> earth.  We did tests for antenna resistance (with current and voltage
> probes on a directional coupler, as well as line current checks) and the
> result we got was not what we expected.  Connecting the mains earth to
> the RF earth INCREASED THE RF RESISTANCE OF THE ANTENNA.  We re-checked
> several times, and also checked that the voltage across the capacitor
> was low (which it was).  So there was no mistake in the experimental
> result.  In that case, it was not a good move to connect the mains
> earth.
> 
> The explanation is most likely that the path of the mains earth is back
> to Kevin's house (the shack is some 50 metres away), with driven earth
> there, then taking an oblique path back via the powerline to the
> distibution power line, far from being in a radial direction.  This
> means that RF forced to flow via that mains earth is far from the ideal
> radial direction, and conflicts with the current departing via the
> intended RF earth with radially laid bare wire.  Nature's answer of the
> current conflict is to "increase the system resistance".  This could
> also be thought of as a sort of "shorted turn" in the ground system.
> This type of event arises only when the radial system is getting fairly
> good anyway, but serves as a reminder that it is not always best to bond
> all grounds together.
> 
> Regards,
> 
> Bob ZL2CA