Guys,
I really don't know why these wildly varying multiple lobes is a big
surprise. Astronomers have been horizontally "stacking" antennas for
more than 50 years. Some of the baselines are as large as the earth's
orbit. Horizontal "stacking" produces an "inteferometer" effect-- that
is multiple lobes in the azimuthal plane. If the baseline is large, the
angular size of the main lobe gets smaller and smaller. Great if you're
interested in resolving small objects in space. However, there is the
problem of ambiguity. That is knowing which of the many lobes you're
listening to. Sure it's solvable for the astronomers. For amateurs,
the problem is one of putting too much power in side lobes and not into
the main lobe. It would take some kind of variable phasing network to
resolve. Radar system engineers have solved this problem many years ago
with passively rotated and tilted arrays.
You can see a similar inteference effect by running EZNEC for a 2 meter
horizontal antenna at 200' up. There are a gazillion elevation lobes.
73 de Brian/K3KO
On 3/17/2013 21:36, Dan Maguire wrote:
Although this topic has been beaten into the ground I'd like to post a
clarification. Stan Stockton and I have exchanged a few private emails and I
now see how I misinterpreted his remarks.
Stan proposed two towers separated by 1.5 WL. He then suggested looking at the
pattern when both antennas were rotated 45 deg off broadside. Turns out he was
thinking of *only* that particular rotation angle. *My mistake* was making the
assumption that he was interested in *all* rotation angles.
If the two towers are on a North-South line, with both antennas at 1.5 WL above
real/average ground (which puts the TOA at 9 deg elevation) and the towers are
separated by 1.5 WL, here's a comparison of the azimuth patterns at 9 deg
elevation. The blue trace shows both antennas facing East and the red trace
shows both antennas facing NE.
http://ac6la.com/adhoc/TwoTower0vs45.gif
Stan was correct in asserting that the NE facing antennas produce a pattern
which is comparable to the broadside pattern. One might even say that the
pattern is better. The max gain is down by only an insignificant 0.20 dB and
there is only a single large sidelobe.
To satisfy my own curiosity I then wanted to see what would happen if the spacing between
the towers was changed. In this animation the towers are still on a North-South line,
both antennas are pointed NE, but the spacing between the towers is varied from 0.75 WL
to 4.0 WL. Look in the lower right corner to see the spacing (variable "S") in
WL. With most browsers you can use Esc to stop the animation and F5 to restart.
http://ac6la.com/adhoc/Pattern_vs_Spacing.gif
Another way to look at things is to plot the max gain vs the tower spacing. In
the first frame of the animation above the tower spacing is 0.75 WL (S = 0.75).
The outer ring is frozen at 18.42 dBi and with S = 0.75 the max gain is down
3.56 dB from the outer ring, or 14.86 dBi. In the chart below, Max Gain vs
Spacing, the first data point is for S = 0.75 and the dBi value is 14.86.
Other data points show the max gain at different tower spacings.
http://ac6la.com/adhoc/Gain_vs_Spacing.gif
The S = 1.5 WL spacing is a good compromise between max gain and clean pattern.
I stand corrected.
Dan, AC6LA
http://ac6la.com/
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2012.0.2240 / Virus Database: 2641/5684 - Release Date: 03/17/13
-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2012.0.2240 / Virus Database: 2641/5684 - Release Date: 03/17/13
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|
