[TowerTalk] Arriving signal interference pattern

[David Gilbert]

Hi, Jim.

The orientation used in the video was two horizontal dipoles both 18 
feet above "ground" (the surface of the earth, not necessarily the true 
RF ground plane) spaced 12 feet apart and positioned broadside to WWV.  
The frequency for the drone flyover was 28.26 MHz and the WWV frequency 
was 15 MHz (30 meters).  Clearly the spacing was better at 28 MHz but 
the result was still massively affected by ground reflections.

But you were correct in your earlier post that antennas stacked 
vertically will not give accurate results to them having different 
ground effects.  Both AC6LA and I proved that several weeks ago using 
EZNEC, with that discussion being located on the EZNEC groups.io forum.  
Like you, I also thought that maybe vertical antennas would be less 
affected, but EZNEC told me that over/under stacked verticals were just 
as affected by ground reflections.

I did, however, rotate my horizontal dipoles to be vertical just to test 
azimuth, but since an arriving signal not perpendicular to the plane of 
the two antennas has a phase determined by both its elevation and 
azimuth that wasn't useful either.

In any case I have improved the interference pattern visualization. The 
SVG version had some limitations and some errors, so just a few minutes 
ago I posted a link to a much better HTML version.

http://www.ab7e.com/Interference Pattern.html

In that post I made this point .... what we measure from two antennas is 
the gradient of the interference pattern caused by multiple paths.  And 
there will always at least be multiple paths caused by ground 
reflections even if propagation is stable and giving us only one signal.

I have pretty much convinced myself that what I was trying to do is 
futile because the results I got with my ArrivalAngle application and 
test setup correspond almost exactly to what I would expect if all I was 
measuring was a diffraction gradient.

73,
Dave   AB7E


On 4/14/2026 12:45 PM, Jim Lux wrote:
> 	
>
>
>   
> what was the relative positions of the two antennas (ht above ground?), and the wavelength
>
> On Mon, 13 Apr 2026 23:05:07 -0700, David Gilbert via TowerTalk <towertalk at contesting.com> wrote:
>
> I recently linked a YouTube video describing an application I created to
> try to measure incoming elevation arrival angles using the measured
> phase difference between two antennas spaced some distance apart.  The
> application is functionally accurate, but the results I got were pretty
> disappointing because the measured phase difference and subsequent
> arrival angle display was heavily affected by what I believe to be
> multipath effects from propagation, ground reflections, and nearby
> terrain.  When things were stable with no obvious propagation-related
> multipath effects the displayed phase was stable but it still seemed to
> be affected in magnitude by what I believe were ground reflections.
>
> I have created an illustration of what I believe happens under multipath
> conditions.  The red and green waves in the background (you have to look
> kind of closely) represent two incoming waves ... same signal but
> arriving at slightly different angles, with both their angles and their
> relative phase slowly changing over time. The brighter interference
> pattern in the foreground includes the effect of ground wave reflections
> and represents the combination of the four waves ... both of the
> incoming waves and both of the ground reflections.
>
> The is basically a 2D representation of different arrival angles and
> relative phase.  The real world would be 3D and look even more complex
> over time.
>
> http://www.ab7e.com/Interference%20Pattern.svg
>
> The peaks and nulls in the combined wave don't simply represent
> variations in amplitude.  They also represent variations in phase, and
> those phase variations can be extreme.
>
> The interference pattern is the result of actual calculations (built by
> Codex), but of course it uses arbitrary values.  An actual situation
> would likely vary significantly, but I think the illustration is
> informative.  At least it was for me.
>
> 73,
> Dave   AB7E
>
>
>
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>   
>
>
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk