In general, I agree with those comments and the unwanted nulls can be
very large. I did exactly as G3TXQ recommends when I installed my tower
and chose antenna heights for each band, but I'd like to point out a few
qualifiers:
a. The ARRL statistics were generated using the TANGLE (takeoff angle)
variable in the VOACAP program. The primary author of that program has
stated that of the 20 or so variables that VOACAP handles, TANGLE is the
one that has the least rigorously supporting background and its accuracy
is probably the least dependable.
b. The ARRL statistics for each path represent the average of VOACAP
calculations generated for each month over a full eleven year sunspot
cycle. At any point in the cycle the optimum angle could be somewhat
different.
c. The angle which avoids a null to Europe may be different than the
angle which avoids a null to the Middle East, even though both may be on
the same beam heading.
d. The optimum arrival angle for any path changes as the opening
progresses. A null that hurts you when the band is wide open may not
have any effect when the band is just opening or closing. The is sort
of the "higher is almost always better" principle for snagging the tough
ones before the pileups get large.
e. The actual arrival angle for any DX signal can change significantly
on an almost instantaneous basis. I've fed 40m BC carrier signals
received from two different antennas (one of which was not resonant) at
different heights on the tower into the two phase locked receivers of my
K3 while in diversity mode, then fed the audio from each receiver into
the stereo sound card of my computer while viewing them with an audio
oscilloscope application set to trigger on one of the signals. I
estimated the resultant "phase jitter" between the two channels
represented arrival angle changes of as much as 15 degrees over
fractions of a second. A good friend of mine used to help with similar
measurements made by the U.S. government a few decades ago with
equipment and antennas designed just for that purpose, and he told me
that tens of degrees of shift could occur within 100 msec.
As I stated, I also used the ARRL statistics since they are the same as
the ones used to generate the Figure of Merit in HFTA. But there is a
reason that the folks who can afford it build stacks of antennas instead
of relying on a single antenna. Stacks not only can give you some
additional gain, they tend to fill in the nulls of the vertical pattern.
73,
Dave AB7E
On 12/1/2011 12:16 PM, Steve Hunt wrote:
> I think we sometimes concentrate too much on looking at antenna heights
> that will maximise gain at certain take-off angles, and forget about the
> nulls. Those deep elevation nulls can be "killers" if arrival angles
> over a wanted path happen to fall in them.
>
> So, before going firm on a height I'd encourage you to look at the ARRL
> Angle-of-Arrival statistics for various bands and paths, and make sure
> your selected height is not going to put a deep null at a high
> probability arrival angle for the bands/paths you are most interested
> in. It might be a better trade to be 1dB weak for 80% of the time rather
> than 20dB weak for 20% of the time, if you get my point.
>
> 73,
> Steve G3TXQ
>
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|