[Top] [All Lists]

[antennaware] Feeding cable lenght

To: <>
Subject: [antennaware] Feeding cable lenght
From: (Eric Gustafson)
Date: Tue, 3 Aug 1999 17:16:22 -0700

Hi Ed,

I'm sorry.  But I must take issue with one of your statements.
Please note I'm not taking issue with the advice specific to this
particular case.  Just the one explanatory sataement.

To: <>
>Date: Tue, 03 Aug 1999 22:25:31 +0100
>From: K4SB <>

Snip... (discussion about feedline lengths for feeding inverted V)

>This may be a case of my not understanding Mauri's reply, but to
>clarify a little...
>1. An electrical 1/2 wave, or multiples thereof, of any type and
>impedence will ALWAYS repeat the load impedence at the opposite
>end. This occurs regardless of the charateristic impedence of
>the line. For instance, if you have a 50 ohm load and a 50 ohm
>Tx output, AND the coaxial line is an electircal multiple of 1/2
>wave, it matters NOT what the impedence of the coax ( or
>whatever else you use ) is.

So far, so good.  This part is OK.

>You can use any impedence line, and the losses incurred will
>ONLY be that which is ordinary from the coax itself. If you use
>600 ohm line, and load impedence is 50 ohms to 50 ohms, the only
>loss will be the normal attenuation of a line of that
>length. SWR is and will indicate 1:1

This one depends (as the Commander in Chief says) on what the
meaning of "ordinary" is.  I admit that for practical purposes,
at 7 MHz, using ordinarily available coax, this is kinda nit
picky but...  If ordinary means the coax losses for that length
of line when operating in a matched condition, then this is not
strictly true.  There will be some additional loss due to
operating the line in a mismatched condition.

>As others have pointed out, a 70 to 50 ohm (or 30 to 50)
>mismatch does not introduce enough loss to even worry about.

This is basically true.  

But we should be careful how far we takes the concept.  For
example, suppose we have a system using a full wavelength of
RG-58 to feed some antenna at 7 MHz.  Since this is an integer
number of halfwaves, the desired 1:1 end to end impedance
transformation property is present.  In this case, the matched
loss would be about 1.113 dB.  At the SWR sugested above (1.5:1 -
halfway between the two examples), the additional loss due to
mismatch is only 0.072 dB.  A truly insignificant amount.

But if we let the SWR get up to 3:1, the additional loss due to
the mismatch would be 0.545 dB.  Some (not me) would conider this

If we let the SWR get to 5:1, the additional loss due to mismatch
is 1.118 dB.  Now, the loss due to mismatch is slightly larger
than the loss inherent in the line when matched (total loss is
2.3dB).  Now, even I consider this significant and would be
looking for a mitigation strategy.  I could select a lower loss
line.  Or I could take steps to match the antenna to the line at
the antenna.  This last option would also let me run only the
length of line physically required for the distance.

I realize the original question didn't include an SWR this high.
But your 600 ohm line comment above represents a 12:1 SWR.  Thank
goodness we can't really get 600 ohm coax!  If we could, and it
was as lossy as RG-58, the additional loss just due to the
mismatch would be over 3 dB.  More than half the applied power
would be lost just due to the mismatch between the line and the

73, Eric  N7CL

FAQ on WWW:     
Administrative requests:

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