Hi, Steve. It has taken some time to get to this and do it carefully.
And sorry, some of the needed info WAS right there in your first post.
Some days I just read right over tops of things and can't see them to
save my life. But the specifics you returned in your second post are
significant beyond your first post. I still would have asked you for
additional information. You didn't say whether you were supporting
this with trees or a tower, significant because of issues with tower
supported 160 antennas. But I have mostly left that alone.
Short answer: If your 60 some-odd radials were uniform length and
spacing all around, your pushing up the current max to the top of the
vertical probably costs you. Since your 60 odd radials are
miscellaneous lengths, the ground field cancellation under the radials
is missing and it's a different ball game. See long answer.
The term counterpole is used as a general term for "radials buried or
elevated, or counterpoise."
Long answer:
The resistance of the ground medium you describe will probably vary a
LOT with moisture content, the decaying material, if wet, will be
pretty conductive, and if dry pretty lossy. As to the varying lengths
of your radials, the current in the radials is an
apportion-to-the-individual-radial exercise based on the impedance of
each radial one at a time versus the others, a parallel Z calculation
problem. This means generally long radials with their lower Z get the
lion's share of current and little ones with the higher Z are starved,
especially so in dry weather. During drier weather, the longer
radials will tend to carry yet more of the current because the ground
is less accepting of the current and it stays more on the wire.
Current imbalance among the radials will run up your ground losses,
because the imbalance prevents field cancellation under the radials.
The moisture-driven change in radiation due to weather will
significantly exceed the difference in radiation between the two
solutions you asked about. There are varying opinions whether raising
the current max helps or hurts over radials. There are contradictory
results on this and I think I know why.
My take:
-----
1) If the radials are uniform length and spacing all around (yours are
not), AND dense (yours are) then the cancellation of induced fields in
dirt under the radials is worth 1.5 to 2 dB. This cancellation
requires the current max down at the feedpoint for full effect.
Moving the current max up unbalances the cancellation.
2) if the radials are miscellaneous (not dense OR not uniform length
OR not uniform spacing all around) then the field cancellation
underneath is largely or entirely lost to start with. If there are
enough of these miscellaneous radials to make it largely dense, then
you WILL RETAIN the benefit of counterpole low effective series
resistance, but LOSE the benefit of vertical vs. radial ground field
cancellation.
In case 2), not case 1), moving the current max up the vertical
reduces close in ground induction loss by reducing the field. How
much and the exact rules is an on-going research project. There is a
lot of anecdotal material to suggest that this ground induction loss
is significantly underestimated by the models, including the high
priced versions. This means that models will underestimate the true
advantage of raising current max.
-----
There are a lot of people who disagree with these last paragraphs. But
there has to be fire where there is THAT much smoke, and we will nail
it down beyond mere opinions sooner or later. Personally I have seen
WAY more than enough, but some require federal court case level data
plus IEEE publishing for convincing. Getting this to the IEEE will
take a while. There is zero commercial interest in the answers, and
therefore no funds.
As the example in extremis, I offer the ancient 80 meter end-fed half
wave L (EFHW), which is still the best single wire 80 meter antenna I
have ever had, including one very good inverted vee with apex at 85
feet (EFHW consistently beat the V on DX, for sure). The current max
is 65 feet in the air up at the bend on the L, and the counterpole can
be as skimpy as an iron pipe in the ground and ten 20' buried bare
radials, less in good ground. With a feedpoint Z of one or two
thousand ohms, a simply terrible ground counterpole giving an
effective series resistance of 100 ohms still works very well for both
local and DX. 100 ohm counterpole resistance turns a regular 1/4 wave
vertical into a dummy load. The 80m EFHW is the ultimate
moving-the-current-max-up-the-wire example.
This would be a very popular antenna if it did not require a variable
tuning network at the base of the vertical wire, to match 2K ohms Z to
50 across 80/75. This last requirement is just too much of a monkey
wrench for most people, as no one makes an off the shelf item for
this. There aren't enough of the 80 EFHW's out there for the masses to
have an 80 EFHW trumps 80 inv vee every time instinct. Otherwise the
question of why it works better would have been hounded out of the
mavens a long time ago.
If you are using a folded counterpoise, you have given up on
radial-based ground field cancellation as unobtainable or
un-something-or-other at your property. After that decision, moving
up the current max in a model will show as much as a dB improvement.
This difference gets larger as the vertical portion is higher. It will
not really make any difference if your vertical run is only 40 or 50
feet, but if you get up 70 to 90 feet before any bends, then it shows
up.
Another thing, comparing a cancellation top like a T or Steve's "S"
top with the top of an inverted L does not appear to change the low
angle radiation to any great degree if you are out in the trees. Where
the vertical run is close to the tower or supported by it, the tower
and its conductors are less induced for loss by an L than a T. (Quite
ugly technical matter for a side discussion.) Operationally the L
fills in the high angles for local coverage and eliminates or
minimizes skip zones without degrading the DX. This is very useful for
holding a run frequency running low power in a contest.
A moral of this discussion is that changes to a 160 antenna cannot be
accurately evaluated unless ground loss interactions from wire
configurations and the counterpole are discussed as meaningfully as
radiator wires and patterns. Carefully evaluated one sees that ground
induction is the gorilla in the room and where the vertical goes and
what is on top of it is the pussycat...to the point that until the
former is known right down to the knubs, for the moment, forget the
sky wires.
I have a 3/8 wave inverted L over an FCP, vertical up 90, and
horizontal out 105 feet. The current max is bracketed by the top 1/16
wave of the vertical run. At QRO I get RBN hits everywhere in the US,
plus Hawaii, plus Europe, with the antenna in a spot with no
possibility whatsoever for radials. The models, in my case, won't
even give me a tenth of a dB at low angles for converting my 105 foot
L top to a T or Steve's S top. So I keep the high angles, and I can
then hold a run frequency all contest running 100 watts -- no "skip
zone intrusions".
73, Guy.
On Sun, Jul 8, 2012 at 9:59 PM, Steve Flood <kk7uv@bresnan.net> wrote:
>>>>What are you connecting the coax shield to? Describe the counterpoles
> for your two antennas in detail.
>
> Both antennas were built over the same radial system, just at different
> times. (see my original message).
>
> I have 60 + radials of 14awg aluminum high-tensile electric fence wire.
> They vary from 50 to 120 feet long and are buried an inch or two in acidic
> pine forest mineral soil. Ground quality/conductivity is probably quite
> poor since it is very rocky.
>
> The inner ends of the radials connect to a 10 x 10 inch, 0.1" thick
> aluminum plate. Each radial wire is wrapped twice around a sheet metal
> screw and then the screw is tighten into the plate.
>
> The L-network is enclosed in a large plastic box 6 inches above the radial
> plate. The coax connection is a standard panel-mount SO-239 with a wide
> copper strap six inches long connecting the SO-239 'shield' to the aluminum
> plate with a 1/4" st,.steel bolt and 1 inch fender washers - all dissimilar
> metals treated with Penetrox.
>
> Steve, KK7UV
>
>
> _______________________________________________
> UR RST IS ... ... ..9 QSB QSB - hw? BK
_______________________________________________
UR RST IS ... ... ..9 QSB QSB - hw? BK
|