Amps
[Top] [All Lists]

[AMPS] Parallel coaxial feedline For 1/2 Characteristic Impedance

To: <amps@contesting.com>
Subject: [AMPS] Parallel coaxial feedline For 1/2 Characteristic Impedance
From: W8JI@contesting.com (Tom Rauch)
Date: Sun, 17 Dec 2000 02:46:43 -0500
Hi Billy, 

> This is it:
> 
> The Idea was to use two 72-Ohm Belden cables in parallel to obtain 36-Ohms
> to match a 32-Ohm 1/4 wave ground-plane antenna.  In doing this, the I
> squared R losses would also be 1/2 unless there is something that I am not
> seeing. Since the tests show that it works,  if I qualified the tests
> properly, I would have no doubts except that I have been a Ham and a RF
> engineer for almost 40 years and have never heard of  running two cables
> in parallel.

Everything is fine, except you did not cut the cable loss in half 
since the line is 1/4 wl long and since the losses are not entirely 
copper losses.

If the line is very short, and if all losses are copper losses, then 
loss in each line is 1/4 because current is half in each line and 
copper area in each line is unchanged. Since each line has 1/4 the 
power loss, and since there are two lines that means total power 
loss is half.

This only applies to lines that are very short in terms of the 
wavelength, and only when copper losses dominate the line. 

If dielectric losses are significant, the dielectric loss doubles 
because you double the parallel loss of the shunting dielectric and 
voltage remains identical! 

In the case of HF with most transmission lines, copper losses 
dominate. But they are not the sole loss mechanism. Also, since 
the line is somewhat long, something different happens. The line, 
because it is mismatched, has standing waves. Since the longer 
line gives the waves "a place to stand" current increases at one 
end of the line and decreases at the other, in the case of a 1/4 wl 
long mismatched line.  

With longer lines, they start to follow transmission line loss specs. 
SWR loss specs do NOT apply to short lines, but they do to longer 
lines that have room for standing waves.

If the lines are long enough, you are really parallelling two 
transmission lines with x dB loss. Each line gets 1/2 the power, 
and of course has the same loss.

Let's say the line loss is 1 dB. The power through each line is -
3dB, and the loss -1 dB. The net power at the end of each line 
would be -4 dB, but it is combined for 3 dB addition. The result is 
you add 3 dB to the -4dB and you get -1dB!

You have the same exact loss, when two lines are parallelled, as 
you do with a single line as long as the SWR is same in either 
case. But again this only applies to long lines. Short lines behave 
differently, and loss decreases as SWR is increased as long as 
the line current is reduced by the SWR! 

Your case is in between the extremes in the above example, and 
you may have reduced loss somewhat but certainly not by half. 
The amount of loss depends on how the losses occur in the lines, 
and what the two parallel lines do to load SWR on each individual 
line.

It's actually a complex problem unless the lines are so short they 
have no room for standing waves or so long they have at least a 
complete standing wave cycle in the line!
73, Tom W8JI
w8ji@contesting.com

--
FAQ on WWW:               http://www.contesting.com/FAQ/amps
Submissions:              amps@contesting.com
Administrative requests:  amps-REQUEST@contesting.com
Problems:                 owner-amps@contesting.com


<Prev in Thread] Current Thread [Next in Thread>