On 5/29/18 1:50 PM, john@kk9a.com wrote:
By using monobanders the out of band signal is reduced by the antenna, for
example, the 15m beam does not receive a 20m TX well. Even closely spaced
monobanders have this "pre-filter". Using tribanders probably requires
better bandpass filtering and you need especially good filtering if you
are using a single tribander.
I'll try some other combinations - that was just an easy one to knock
out during lunch.
If anyone has any particular antenna model to try, let me know.
The parasitic elements on the 15 m antenna are far enough from resonance
that they probably don't couple well to the driven element, so I think
you could treat it as a dipole.
In the far field (not necessarily the case here), a 3 element 20m Yagi
has about 13.5 dBi gain at 22 degree elevation angle (Cebik design, over
ground, so it gets a "ground bounce" gain).
For other ham bands, assuming it's perfectly matched at 20m (and not so
perfect for other bands)..
40m -14.4 dB mismatch, 5.0 dBi - net of -9.4 dBi
30m -12.8 6.4 dBi - -6.5
20m 0 13.5 13.5
17m - 8.5 6.2 -2.3
15m - 9.9 4.6 -5.3
12m -12.0 0.1 -11.9
10m -12.8 -12.0 -24.8
Some interesting observations - the gain down about 6-8 dB when not at
resonance - so what we're seeing is basically just a dipole over ground.
10m is horrible because the antenna is exactly double size and
approximately resonant, so the coupling among the elements is all
screwed up and out of phase.
The mismatch loss is about 10dB for "adjacent bands" and only 15 dB for
"far away bands". This is the "frequency selectivity" of the antenna.
In the near field (basically any sort of field day setup within the rule
determined radius), at best, you pick up 10-15 dB, certainly not more
than 20 dB, of extra isolation for being out of band. You might get
another 10-15 dB for "orientation".
And this is the fundamental problem with multiple antennas at the same
site - they make terrible filters.
I suspect that *any* monoband antenna operated well away from its design
frequency will be like a dipole - that is, it's not much different than
the 3 element yagi I modeled here.
John KK9A
jimlux wrote:
On 5/29/18 10:39 AM, jimlux wrote:
On 5/29/18 10:06 AM, Ed Sawyer wrote:
I have 2 towers separated by 140 ft. Monobanders. The towers are
running
roughly N/S so I am not fully pointing at one when beaming EU but its not
"tip to tip" either.
Keeping this "antenna" related - 140 ft is 2 wavelengths on 20m. On the
lower bands, you're substantially less than wavelength apart, so the
interaction is probably predominantly "near field" - that might make the
pointing direction irrelevant. The distribution and direction of the
electric and magnetic fields (esp if there's significant directive gain)
may be substantially different from what's in the far field.
Running a quick NEC model using L.B. Cebik's 3 element 20m Yagi,
separated by 140 ft..
Here's the relative current induced in the victim antenna by the
transmit antenna, as a function of the angle (0 is parallel, 90 is
victim pointing at Tx)
angle dB
-90 -59.3 (pointing away)
0 -54.8 (pointing parallel)
15 -50.3
30 -46.9
45 -43.7 (pointing 45 crossing boresight of tx antenna)
60 -40.2
75 -37.2
90 -35.2 (pointing right at tx antenna)
this means that if you have 1 Amp flowing in the transmit antenna, at 90
degrees, you'll have 35.2 dB less (0.0174 Amps)..
Assuming constant impedances,etc.. if you put a kilowatt into one
antenna, you'll get a about 300mW into the receiver (25 dBm).
If they're pointing parallel, the coupling is a lot less (-55 dB) - For
a kilowatt Tx (60dBm), you get +5dBm into the victim receiver.
Interestingly, pointing away only knocks it down by 3-4 dB relative to
pointing parallel.
Note that if these were 40m antennas, keeping the 140 foot spacing, the
coupling would be a LOT more.
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|