Content-Transfer-Encoding: 8bit
Message-ID: <5.1.0.14.0.20021104143711.00a8f908@gemini.herts.ac.uk>
Date: Mon, 04 Nov 2002 16:10:26 +0000
To: rsgb_lf_group@blacksheep.org
From: "James Moritz" <j.r.moritz@herts.ac.uk>
In-reply-to: <2D280BB697C7EE4C979C5C89925418EF368F39@TNS-FBU-2E-003.corp .telenor.no>
MIME-Version: 1.0
Subject: Re: LF: VS: Feeding an inverted L-antenna
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

At 13:49 04/11/2002 +0100, you wrote:

>Wonder why seems to be a habit use a series coil to tune out the 
>capacitance of the 136kHz aerial, instead of more normal sort of matching 
>- using T- or L-network

Dear Jan-Martin, LF Group,

I would agree with everything Andy says - but there are also practical 
difficulties in making the T-network scheme work:

The components in the tuning network have to be highly rated - for example, 
a 500W TX driving a 50R load has an output current of about 3.2A. The 142p 
input capacitor has a reactance of about -j8.2kohms, so with 3.2 amps 
flowing in it, the voltage across it's terminals will be 26kV! The 
capacitor plates would probably need at least 50mm spacing... The inductor 
will have to withstand this voltage, as well as a higher current - about 
8.3A instead of 5.1A, and since it also has to have a higher Q and about 3 
times the inductance of the simple series coil, it will be a much more 
massive component. The output 295pF capacitor will have around 20kV across 
it. The capacitive reactance of the antenna can be fairly accurately 
predicted, but the loss resistance is more unpredictable, so at least 2 of 
the 3 components must be variable to be sure of getting a reasonable match.

So as well as improving efficiency, the series coil is much easier to build 
and adjust - The resistive component can be matched using a small ferrite 
cored auto-transformer, with several taps to allow a wide range of antenna 
resistance - an ETD49 core is more than adequate up to 1kW or more, and all 
the tuning is done by small variations to the series coil of about 1.3mH. 
The maximum voltage (now the antenna voltage) will be under 6kV with 500W 
TX, so much more manageable.

All this begs the question "why do people use T networks at HF?" The answer 
is, I suppose, that at HF the degree of mismatch is usually smaller, 
requiring a much lower loaded-Q network which is more practical to realise. 
Also, being able to twiddle all 3 component values gives you a reasonable 
chance of getting a good match, without having the faintest idea what the 
antenna impedance actually is!

PS...Once you have got your TX antenna working, the PS3 in conjunction with 
a calibrated antenna, is the perfect tool for measuring field strength, 
etc. It is also extremely useful in conjunction with a bridge for impedance 
measurements. The selectivity is more than adequate with about 100-200Hz BW 
for use as an RX, which it will do fairly well when the sensitivity is 
increased with a suitable preselector. I would not be without mine. But the 
1.5kHz "IF" output is really a bit uncomfortable to listen to, so 
down-conversion of some sort would be a good idea.

Cheers, Jim Moritz
73 de M0BMU