On Oct 4, 11:33 pm, Roy Lewallen wrote:


Here's a fun experiment with EZNEC.

1. Open the VHFGP.EZ example file. Click View Ant to open the View
Antenna display. In the View Antenna display control section, click
Center Ant Image so you can see the antenna better.
2. Add the following wi

End 1: 0, 0, 5 (wavelengths)
End 2: 0, 0, 4.727 (wavelengths)
Diameter: 0.25 (inches)
Segments: 6

This represents the outside of a feedline connected to the feedpoint.
3. Click the Currents button. Look at the display and, in the Currents
box, compare the current on the outside of the "feedline" (Wire 6,
Segment 1) to the main radiator current (Wire 5, Segment 1).
4. Change Plot Type to Elevation. Click FF Plot to see the 2D elevation
pattern.
5. Reconsider the statement about decoupling. . .

This is admittedly contrived to show a particularly extreme case. But
try different lengths of "feedline" either open or connected to ground
and you'll find other cases where the feedline current is high and the
pattern distorted. You'll also find cases where inserting a "balun"
(high impedance load) in the "feedline" will actually increase the
feedline current due to changing the current distribution to a value
more favorable for the particular feedline length.

Roy Lewallen, W7EL

I'd be curious to see what you get doing the same with a 1/2 wave
whip.
IE: usual j-pole, 1/2 wave whip...
I tried doing a test using the demo version. Being it was limited
segments,
I tried to keep it even, and used twice the segments for the 1/2 wave
radiator vs the 1/4 wave, but used the same amount of segments for
the "feedline" .
I also used "real ground", and the elevation plot so I could more
easily see the appx real world plots.
With the GP, I notice heavy current when the feedline is appx 1/4
wave, but not so bad when it's longer. In some cases I saw a gain
where the feed currents seem to be in phase with the antenna
currents.
When trying the 1/2 wave, I didn't see the problem too much using a
short 1/4 wave feed, but the longer lengths were much worse than
the plots for the GP.

1/4 WL GP

6.58 dbi at 3 degrees -no feedline
5.26 dbi at 42 degrees -feedline .28 wl
7.39 dbi at 3 degrees -feedline 1 wl
6.09 dbi at 3 degrees - feedline 2 wl
6.77 dbi at 3 degrees - feedline 3 wl
7.05 dbi at 3 degrees - feedline 4 wl
7.87 dbi at 3 degrees- feedline 5 wl and grounded at "0"

1/2 WL whip

6.43 dbi at 3 degrees -no feedline
6.84 dbi at 3 degrees -feedline .28 wl
6.23 dbi at 3 degrees -feedline 1 wl
6.67 dbi at 55 degrees - feedline 2 wl
14.82 dbi at 68 degrees - feedline 3 wl
21.42 dbi at 67 degrees - feedline 4 wl
6.68 dbi at 3 degrees- feedline 5 wl and grounded at "0"

Anyway, I may have had problems running this
test with the limited segments, but it seems to
show the 1/2 wave as having the worse problems
of the two overall. Really bad at 3-4 waves length
of line.
So while the decoupling for the GP is not always the
greatest, I think it's still probably less a problem
than the usual 1/2 wave whip with no decoupling.
I've never used chokes or baluns per say to decouple
a VHF vertical. I always use radial sets, cones, sleeves,
etc..
The usual ground plane really needs two radial sets to
decouple the line well. Most good sleeve dipoles will
use an extra sleeve for decoupling the line.
You might get a bit different results using unlimited segments.
But I betting the trend will still be fairly close, with the
non-decoupled 1/2 wave being the worst overall at the longer
line lengths.
MK