Content-Transfer-Encoding: 8bit
Message-ID: <5.1.0.14.0.20030502115241.025564e8@gemini.herts.ac.uk>
Date: Fri, 02 May 2003 16:13:15 +0100
To: rsgb_lf_group@blacksheep.org
From: "James Moritz" <j.r.moritz@herts.ac.uk>
In-reply-to: <001701c31029$0888ea40$33ce28c3@erica>
References: <5.1.0.14.0.20030430120518.00ade070@gemini.herts.ac.uk> <001501c31001$6a079780$bee5fc3e@l8p8y6>
MIME-Version: 1.0
Subject: Re: LF: Mains Cable and other transmission lines
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 Peter, LF Group


>50ohm systems were chosen by equipment, test gear, coax cable and antenna
>manufacturers in the interests of standardisation - there is nothing 'magic'
>about 50ohms systems.
>There is no point in having an interim 50ohms section between a transmitter
>(that is other than 50ohms) and the antenna matching/loading system unless
>you wish to use it with a piece of existing or proposed 50ohm test
>equipment.


The choice is not entirely arbitrary - as I recall, 50ohm coax gives the 
best power handling capability for a given size of cable, while 75ohm gives 
the lowest loss. This is probably academic in amateur practice because the 
cost is usually a more significant consideration than the size!

It is true that there is no particular need to design an LF system to use 
50ohm impedances throughout. Provided the transmitter  is provided with the 
load impedance for which it is designed, and you have some way of verifying 
that the load actually is properly matched to the TX, there is no problem. 
Standardisation is not an insignificant advantage however - designing 
everything to match to 50ohms means any transmitter, antenna tuner, power 
meter, tuning aid, low-pass filter, dummy load etc., etc., can be 
interconnected with predictable results, rather than having to devise 
suitable matching for each piece of equipment, or conversion factors for 
measurements. I take the view that you have to do some sort of impedance 
matching whether you want to or not, so designing the system to be 50 ohm 
throughout is rather like designing everything to run from 230V, 50Hz - it 
saves a lot of headaches!

50ohms is also a very practical value - manageable voltage and current, 
reasonable number of turns on a transformer, sensible values  of L and C 
and so on. For example, 1200W from a Decca PA is 245V, 4.9A into 50ohms, 
which is quite manageable with ordinary coax (or mains...) cable and 
fittings, and other components. At 4 ohms, the current would be 17A, while 
at 600 ohms the voltage would be 850V, either of which would be more awkward.

If the connection between TX and antenna is very short, the characteristic 
impedance of the transmission line connecting them is of little 
consequence. But the transmission line does not need to be wavelengths long 
before it starts to make a difference - 100m of RG58 has only about 0.5dB 
loss at 136kHz when in a matched system. while it should, for example, be 
possible to produce a 4ohms impedance at the TX end of the cable with a 
suitable matching network at the antenna end, the losses would be much 
higher - more like 10dB, mostly because of I^2R loss. But the losses would 
be only be slightly increased if the load at the TX had to be 75 ohms.

Cheers, Jim Moritz
73 de M0BMU