Content-Transfer-Encoding: 8bit
Message-ID: <5.1.0.14.0.20030417135556.00b007a8@gemini.herts.ac.uk>
Date: Thu, 17 Apr 2003 16:20:54 +0100
To: rsgb_lf_group@blacksheep.org
From: "James Moritz" <j.r.moritz@herts.ac.uk>
In-reply-to: <001201c304ce$a6b723a0$54cdfc3e@ian>
References: <5.1.0.14.0.20030416120614.024e6ea0@gemini.herts.ac.uk>
MIME-Version: 1.0
Subject: Re: LF: The Mystery - continued. Part the second.
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 Ian, LF Group

You wrote:
>The antenna current is now 2A plus or minus  the accuracy of the system...

I'm not sure which band this applies to, but in an earlier mail you said 
the TX puts out 205W at 73k and 100W at 136k. For Iant = 2A, this works out 
to a total loss resistance, including that of the loading coil, of  51 ohm 
at 73k, and 25 ohm at 136k, which in either case is very respectable - so 
you are probably getting near optimal efficiency for a fairly small antenna 
already, and I would be surprised if a major improvement were possible on 
what you already have.

Withe the substantial ground system you describe, experience shows that 
even quite large additions or subtractions from the grounding won't make a 
major difference to antenna efficiency. I'm not sure why it should make a 
huge difference if the antenna end is earthed directly, or via a torturous 
route back along the coax, but it seems to me the main effect would be to 
add inductance in series with the earth connection, which, depending on how 
the ground connections are made in the antenna circuit as a whole, might 
affect the impedance transformation produced by the tapped loading coil - 
since you are matching to a very low impedance, it would only take a few uH 
of distributed inductance added into the circuit to significantly modify 
the matching network behaviour. If this is true, it should be possible to 
achieve the same antenna current with or without the direct ground 
connection by changing the tapping point on the loading coil to suit each 
connection. Another related possibility is that, with the ground connected 
at the loading coil end, part of the amplifier output current returning to 
the amplifier earth terminal will flow through the ground, rather than via 
the coax braid. It may be that this path is subject to greater resistive 
losses than if the current flows exclusively  in the coax braid. In either 
case, transforming the PA output impedance to 50ohm as others have 
suggested would reduce the problem. Or you could leave it as it is with the 
ground wire disconnected...

As regards the fuses blowing when a matching transformer is used, does your 
PA have a DC blocking capacitor at the output? Normally, a loudspeaker 
connects directly to the output so that no bulky audio coupling capacitor 
is required. However, with no input signal, the PA is bound to have a DC 
offset at the output of up to a few 10s of mV. When the load is a speaker 
on the end of long leads, this is not much of an issue, but if an RF 
transformer with a few turns of heavy wire as the primary is connected 
directly across the output of the amplifier, several amps of DC will flow, 
which may be what is causing the problem. The DC offset at the output may 
also be increased if the amplifier is driven into non-linearity, 
which  probably will be the case when trying to get a reasonable RF output. 
A DC blocking capacitor of several uF worth of polyester capacitors in 
series with the output will prevent this being a problem.

Cheers, Jim Moritz
73 de M0BMU