Topband: Tensioning of RG58 coax for 2-way Beverage

[Tom W8JI]

----- Original Message ----- 
From: "Roger D Johnson" <n1rj at roadrunner.com>
Cc: <topband at contesting.com>
Sent: Sunday, February 08, 2015 5:20 PM
Subject: Re: Topband: Tensioning of RG58 coax for 2-way Beverage


>I can't believe that an unterminated wire hundreds of feet long in parallel 
>with the
> Beverage wire and, in close proximity to it, would not have a deleterious 
> effect.
>
> 73, Roger

You win, Roger!!

You can bet there would be some effects, but not what we might imagine.

The outside wall of the shield is the actual antenna, not the center 
conductor. As a matter of fact, the center conductor does nothing but convey 
the signal from the far end. It is pretty much invisible, as is the inside 
wall of the shield invisible, to signals.

Obviously the messenger is a close-spaced wire to the actual "thing" being 
used as the antenna, and shield effectiveness has nothing at all to do with 
mitigating that effect.

The effect of the messenger would not be so much because it is unterminated, 
but because it can form a long open stub with the antenna. This **will** 
cause the coax shield common mode  impedance to have ripples, and the 
shield's common mode is what forms the actual antenna. I've actually 
measured this effect, although it was years ago, and I'm 100% sure it 
happens. There is no doubt.

If you model (or better measure) a Beverage, and if you add a second 
conductor very close to the Beverage but not connected to the Beverage, you 
will see periodic SWR spikes or ripple. This is because of the stub effect.

If the second conductor is connected to the shield at each end (or to a 
single wire Beverage at each end if it is a twin lead with an unconnected 
wire) the spikes go away.

The model effect of that second unconnected wire is to make F/B  go almost 
away on the ripple frequencies. Since a model shows an effect, since a 
measured antenna (actually several) showed this effect, since logic tells us 
it should have an effect, and since it is incredibly easy to connect the 
shield to the messenger, I don't know why anyone would ever not connect it.

Now in fairness, if someone is lucky, the ripples can fall outside a band. 
In that case they would not matter. No one would notice.

But there absolutely is an effect, and the effect can cause SWR ripple and 
loss of F/B on frequencies where the parallel open stub happens to be a bad 
length. I never took the time to figure out if that is where the "stub" is a 
multiple of 1/2 wave or 1/4 wave, but intuition tells me it is where the 
stub is multiples of 1/2 wave. Intuition also tells me if one end is shorted 
and the other open, the ripples would occur at frequencies where the stub 
was an odd 1/4 wave or multiples.

73 Tom