[Lowfer] HIFER POWER?

[Ed Phillips]

Roger Thompson wrote:
> 
> Ed,
> 
> I made up a quick model and don't get quite the same results.  I find
> max gain of 1.78 dBi away from the tower at 45 degrees and a F/B of 1.76
> dB.  Gain is down almost 8 dB at 15 degrees elevation.  This is using
> the EZNEC real high accuracy ground model with medium conductivity.  I
> remember MININEC having some limitations regards ground, but my
> documentation is in storage.  I ran the model with MININEC ground
> assumptions and had higher gain, but pretty much the same elevated
> pattern.  Even with perfect ground, I still get a gain peak near 45
> degrees.
> 
> I added 4 slightly elevated radials to the slant wire and assumed a feed
> point near the bottom end of the wire and get 37.9 + j119.3 as the feed
> point impedance.  The real part seems pretty much in the correct range
> for the length of radiator.  The unconnected tower will reflect an
> unknown reactance back into the fed antenna and I'm not completely
> confident of the overall impedance results as I don't know how well the
> program handles close coupled elements.  If I were building the antenna,
> I'd just measure the actual impedance and build a matching network based
> on the results.
> 
> Max gain on the horizon will only occur with perfectly conducting
> infinite ground and certain types of antennas, so that MININEC result is
> suspect for application to any real antennas.
> 
> In summary, I'd say the overall antenna system is still dominated by the
> tower, therefore the higher angle of radiation still exists.  As to the
> gain of 7.9 dB, even if it were dBi, that value would be in the range of
> a fairly well optimized 3 element yagi, which I just can't see happening
> with this particular arrangement.  I conclude you might as well feed the
> tower.
> 
> Roger
> AD5T

Roger Thompson wrote:
> 
> Ed,
> 
> I made up a quick model and don't get quite the same results.  I find
> max gain of 1.78 dBi away from the tower at 45 degrees and a F/B of 1.76
> dB.  Gain is down almost 8 dB at 15 degrees elevation.  This is using
> the EZNEC real high accuracy ground model with medium conductivity.  I
> remember MININEC having some limitations regards ground, but my
> documentation is in storage.  I ran the model with MININEC ground
> assumptions and had higher gain, but pretty much the same elevated
> pattern.  Even with perfect ground, I still get a gain peak near 45
> degrees.
> 
> I added 4 slightly elevated radials to the slant wire and assumed a feed
> point near the bottom end of the wire and get 37.9 + j119.3 as the feed
> point impedance.  The real part seems pretty much in the correct range
> for the length of radiator.  The unconnected tower will reflect an
> unknown reactance back into the fed antenna and I'm not completely
> confident of the overall impedance results as I don't know how well the
> program handles close coupled elements.  If I were building the antenna,
> I'd just measure the actual impedance and build a matching network based
> on the results.
> 
> Max gain on the horizon will only occur with perfectly conducting
> infinite ground and certain types of antennas, so that MININEC result is
> suspect for application to any real antennas.
> 
> In summary, I'd say the overall antenna system is still dominated by the
> tower, therefore the higher angle of radiation still exists.  As to the
> gain of 7.9 dB, even if it were dBi, that value would be in the range of
> a fairly well optimized 3 element yagi, which I just can't see happening
> with this particular arrangement.  I conclude you might as well feed the
> tower.
> 
> Roger
> AD5T

Roger:

	Thanks for taking the effort to check out my answers; I certainly
suspected them or I wouldn't have written.  I expected the ground
conductivity to be important, particularly around here where the soil is
bone dry most of the year(600 ohm resistance between 6' ground rods 18"
apart!!!). but didn't expect it to be that much.  I have looked at
various NEC software and see that some of it can accomodate finite
ground conductivity (and dielectric constant?).  That must be important
to guys designing BC systems, where the revenue depends on the coverage
area.  As for the gain, I know that value is bogus and don't even know
what it is intended to mean.  Dipole in air comes out OK.  This is a
very simple NEC to which I know how to input data, so started with it.

	Feeding the base of the tower is the easiest way to go, but measuring
the impedance directly is another matter.  I have a couple of 10 kW +/-
BC stations about a mile away and they put so much signal into the tower
that I can't use my Twin-T GR bridge as there is too much BC output to
permit a reading, at least with the simple receiver input filter I've
used.  I've done measurements with the tower as a lowfer antenna by the
substitution method.  Tuned up the system and then replaced the antenna
with a variable capacitor and resistor in series and adjusted for the
same current at a constant input power.  The capacitance is about 450
uufd and the resistance (ground and insulators mainly) is about 45
ohms.  No need for a high-Q loading coil although I use one.  Problem
here will be that the current will be pretty hard to measure.

	Your patterns sound very similar to the ones I get for the tower alone
with "perfect" ground, with major lobe up around 45� and signal way down
toward the horizon.  Probably not coincidence but doesn't necessarily
validate the program.

	What to do?  I'm getting the urge but not enough ambition yet......

Ed