TopBand: Elevated Radials

[John Mitchell]

At 09:32 AM 3/19/98 +0000, w8ji.tom@MCIONE.com wrote:

Hi Tom, thanx for the well-considered reply.
>
>> 1. Radiator portion electrical length and configuration. Obviously, 127 feet
>> or so perfectly straight up would be better ; )  Less vertical length means
>> more horizontal component, less low angle radiation. 
>
>Not unless the radials,  feedline, or other conductors other than the 
>vertical element are radiating. 

Oops, I wasn't clear enough.  I was referring to a reference 1/4 wave
radiator.  The point was that folding over portions of it to make an
Inverted L, which I presently have to use, provides some horizontal
radiation, which I feel sure you would agree with.

>> 4. Spacing and length of radials/counterpoise.  Like a dipole close to
>> earth, the coupling makes the resonant point lower. 
>
>If you are talking about the resonant frequency of the antenna 
>system, a better ground often makes the resonant frequency higher
>and more stable with moisture changes in the soil. 

No, I was actually talking about a dipole antenna.  But I see your point.  I
was trying to make an analogy between resonant 1/4 wave radials, and a
dipole antenna.  In fact, the first pair of elevated radials I installed
were salvaged from my former Inverted Vee! ; )

>
>For the radial itself, moving it close to earth makes the resonant 
>frequency lower. But resonance has little to do with loss or 
>efficiency.  

But resonance has a lot to do with current flow, right?  Revisit the diagram
I drew of antennas over salt water, if you would.  I had hoped to stimulate
some debate over this.  After all, if a 1/4 wave vertical with only 2
resonant radials suspended 66 feet above the ocean will work as well as one
with an infinite number of radials at the same height over the same perfect
ground, doesn't this have implications for elevated radials in general.  Am
I seeing this right or not?   
>
>> Improving the
>> conductive ground screen under a vertical usually results in narrowing the
>> bandwidth.
>
>That's a popular rumor, but factually bandwidth can increase, 
>decrease, or stay the same when system efficiency is improved. I can 
>site countless examples of that, I'll give two.

But, it <usually> results in decreased bandwidth, as when going from a
single ground rod in the mud to many radials,  Right?

>If I move that same radiator over a copper plate 1/2 wl in diameter , 
>BW will increase along with efficiency and feedpoint impedance.

Yes, but isn't this impractical?  I mean, that's the whole problem.  Few
people can afford to copper plate their near field, let alone the Fresnel
zone.  Don't elevated radials give some near field advantage over standard
ground radials when the ground is very lossy?
>
>Current in the radials will unbalance due to coupling to the 
>surroundings as well as antenna pattern. You can actually force the 
>system to have LESS efficiency by forcing equal currents in a 
>system.

I don't see how current in resonant radials will unbalance due to coupling
to surroundings.  Let's say you put two 127 ft radials out, 15 feet above
the ground at the feedpoint.  Currents flowing in those two radials are, of
course, within ground effect, that is some of the energy is coupled to the
lossy ground.  That's why I thought it important to improve ground
immediately under elevated radials by concentrating ground radials there.
But is seems like the elevated counterpoise has to offer some advantage in
terms of ground loss, compared to not using it at all.
>
>Balancing the currents does NOT guarantee minimum radiation from 
>the radials (unless the are in free space and equal lengths and 
>spacings), and it almost certainly will not improve efficiency 
>since it has little to no effect on nearfield coupling... and that is 
>the main source of loss. 
>
Well, what I thought I said was that balancing the currents enabled them to
cancel out, leaving the vertical radiator to do the radiating.  Is this not
correct?

>I believe the crux of the issue is:
>
>1.) There are NO comprehensive tests that show the NEC models of 
>low horizontal wires (like radials) are accurate.

I didn't go there, and I agree with you.  I tried to explain one view of
this issue, using as few technical terms as possible.
>
>2.) There are real world direct measurements from multiple sources 
>that show small elevated system, when close to earth, are less 
>efficient than a modest sized conventional system.

I feel this is where I stand to learn a lot from experimenting.  But we need
to define some terms here, i.e., "small" elevated system - how small;
"close" to earth, how close; also the conductivity and topography of earth,
and "modest" sized conventional system.  Let's take one definition and run
it thru.  A inverted L vertical, 1/4 L long, with four elevated radials from
a feedpoint 15 feet above lossy earth.  Add some radials (a dozen or so) at
the base under the counterpoise to help with coupling losses.  Now, which
would be better - using the elevated radials with ground radials, or
dispensing with the elevated radials and feeding at ground level with
limited radials?  This is important to me, since I could get more of the Inv
L to be vertical (15 feet more) by not elevating the feedpoint and radial
system.
>
>3.) A lot of stuff gets published without being reviewed or checked 
>out, and that starts rumors that live forever.

Agreed.
>
>4.) We all like to think our systems are best because we all can 
>make contacts. We stick 15 antennas within 300 feet of each other, 
>test them over the air where QSB alone can make 20 dB of change from 
>minute to minute against people hundreds or thousands of miles away, 
>postulate some very strange theories that are one step above pure 
>voo-doo, and reach a cast-in-stone conclusion.

I hope I didn't give the impression that I thought my system was best.  I am
painfully aware of its shortcomings, but believe I have optimized it given
my constraints.  If dispensing with elevated radials would improve it, I
would be the first to do so, but I'm not convinced of that.
>
>But hey, that's why it's so much fun!...

Nice note to end on, Tom.  I plan to take your advice and measure, and will
share data when I do.

Thanks and 73,
John K4IQ


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