I have some questions and comments, naturally. ;-)
> > It seems likely that my 20 elevated radials aren't all equally sharing the
> > current. This article (sorry - I left it at home and forget who
> > to credit as writing it) simply shows that if any one of the radials
> > is near zero impedance (electrically a quarter wave) that the lion's
> > share of the current will go into that radial.
How on earth (no pun intended) do you get a "radial near zero
impedance"? There is no such thing unless it is lossless.
If the radial is lossless, that means it is not radiating or
dissipating power. More current in a radial like that would be good,
not bad. So if I had a zero impedance quarter wave long radial, I'd
WANT it to have all my ground system's current.
A radial with loss (either due to radiation or ohmic losses) would
have HIGHER impedance at the common point end than one with no loss.
> > The advantages are a better pattern because of more uniform current
> > flow. The other advantage is elimination of a possible horizontal
> > polarized component because of unequal distribution
I don't think we could say "eliminate". We might not even be able to
say equal current "reduces" the horizontal radiation!
Let's look at the problem's possible cause. Let's assume the radials
are all equal length and height, and the antenna is an
omni-directional antenna located symmetrically at the center of the
radials.
The only reason one radial would have more more current
than another is if the earth located directly below some radials
had greatly different characteristics than the earth below other
radials.
Now let's assume we balance the currents.
We still are stuck with different earth near the radial, and the
absorption and re-radiation of earth near each radial will remain
different. To truly cancel FAR-FIELD radiation from the radial system
(that system INCLUDES the earth, whether we like it or not) current
would have to be adjusted so the COMBINED effect of the earth's
"reflection" and the radial's current produced a far field null.
Of course there is NO null near the radial, only at a large distance
away (where the radials all appear to an equal distance to the
distant point).
Common sense tells us if non-uniform earth caused the unbalance,
equal current in each wire will guarantee the system will have some
radiation because the media close to each radial has different
absorption and "reflection".
> > So - I guess I will cut my radials down in length (perhaps to 3/16 wave)
> > and put an inductor in series with them and adjust it for minimum
> > impedance on 1825 kHz. Perhaps I will make some current measurements
> > before and after to verify the results. I could tune each radial
> > individually to make them all the same impedance at 1825 before
> > connecting them in parallel.
Of course impedance at the common point has little to do with loss in
the system. As a matter of fact a VERY low impedance resonant
elevated system can have terrible loss. The reason is simple.
Imagine the electric field and magnetic field near the radial. If the
radial is long with modest impedance current is distributed more
evenly along the wire. So is the voltage and electric field. Now if
the radial is shortened and loaded, current near the common
point increases. Loss is current squared (even if it is magnetically
induced loss), so if we double current or magnetic field near the
antennas base we quadruple loss. We also have the same problem at the
radial ends, because we increase voltage. If we double voltage we
quadruple E-field losses.
Yet at the very same time this occurs, the IMPEDANCE we measure at
the base can be low.
Take the system to the extreme, imagine four eight foot base loaded
mobile whips with high Q coils being used for radials. We might have
four twenty ohm systems in parallel, for 5 ohms of base
impedance. But ALL of the magnetic field will be near the center of
the antenna and coupled to loss earth quite well in a small area, and
the voltage and electric field losses will also be very high. No way
will this system have the loss of I*I*5 ohms.
The effect is current bunches up in the base of the vertical, and is
reduced at the top. K8BBI (W8XO) measured this effect. He had larger
base current and lower base impedance. He thought he had winner until
he did some FS tests with locals. When he removed his short
resonant radials and used longer conventional radials FS and base
impedance increased! All with no change in the radiator.
I was so interested in Dave's results I built a ten meter
vertical model. I duplicated his results.
That's why a large conventional system works so well; radiation,
magnetic and electric fields are spread around. The more radials and
the longer the radials are, the more the fields (the radiation and
both induction fields) are spread around. Make no mistake about it,
the radials are coupled very tightly to lossy earth even if they are
30 feet above it on 160 meters.
Near field effects are complex. We might think we are improving
things while we are doing the opposite. Only far field FS meters tell
the whole story.
73, Tom W8JI
--
FAQ on WWW: http://www.contesting.com/topband.html
Submissions: topband@contesting.com
Administrative requests: topband-REQUEST@contesting.com
Problems: owner-topband@contesting.com
|