[TowerTalk] Verticals on a Hill

Sun Jan 20 14:57:22 EST 2019

On 1/20/19 8:59 AM, VE6WZ_Steve wrote:
> Bill, I agree with your analysis and like you I wish there was a program like HFTA out there that could handle a vertical radiator.
> My remote QTH is on a gently sloping hill in all direction and is about 100m above the surrounding farmland.
> 
> Here is a very rudimentary and non-scientific powerpoint which includes a copy of the “vertical over sloping terrain analysis” from ON4UN’s book.  I don't know where he got this analysis or who did it.  Does anyone know??
> 
> https://drive.google.com/file/d/1UP45c5MWaWvA0T9no4DHW060FSgC-3Pk/view?usp=sharing <https://drive.google.com/file/d/1UP45c5MWaWvA0T9no4DHW060FSgC-3Pk/view?usp=sharing>
> 

That looks like the "tilt the radiator over horizontal surface, then 
tilt the pattern to make the radiator vertical" analysis which NEC will 
do nicely (at least for a uniform surface).

> This analysis of an 80m vertical over sloping ground certainly shows that a  significant low-angle gain benefit of up to 11dB over a flat terrain vertical.  I include a comparison to other plots from Johns book showing how a vertical over saltwater shows a similar outstanding low-angle response, which we have all heard in action from various well place DX-peds on the saltwater.  I extrapolate this to 160m.
> 
> I can only assume that building a HFTA like program for a vertical is a bit more complex than a horizontal radiator.

Very much more so.

There's two big problems:

Problem #1

For Vpol, the reflection from the soil is very much affected by 
incidence angle and soil properties. There's a Brewster Angle where 
there is no reflection. For Hpol, it's always almost a total reflection, 
independent of soil properties.  Well, the reflection goes from total at 
low elevation angles (grazing) down to something like 0.8 at vertical 
incidence, and more importantly, the phase of the reflection is 
basically constant.
Not so with Vpol.

Polarized sunglasses make use of this.. they're Vpol, because the Hpol 
constitutes most of the glare in a reflection)

Problem #2
In H-pol, you can use a simple terrain model of a piecewise linear (or 
interpolated spline) to represent the terrain.. The waves are parallel 
to the surface, so the calculation is easy.
In V pol, the waves are at an angle to the surface, so the calculation 
is more complex.

---
And then getting to the other problem - most antennas are not surrounded 
by circularly symmetric terrain (like a eroded lunar crater).. That's a 
problem for both H and V pol modeling. In real life rays are not 
constrained to a vertical plane in the direction of propagation.

All of this is solvable - the equations are fairly simple, but it's one 
of those things where there's not much economic incentive to develop 
*and validate* the modeling code.  It would be computationally 
intensive, but these days, with inexpensive or free cloud computing 
resources you can get a lot of computing for very little money.

Get up to VHF, UHF, and above, and there's tons of software out there to 
model land mobile radio, cellphones, etc. with all manner of modeling 
approaches that handles diffraction, arbitrary surfaces, buildings, etc. 
  One could probably use one of those codes for HF, by scaling (instead 
of running your model for 1.8 MHz, run it for 1.8 GHz, and load your 
terrain model scaled by 10.

The soil dielectric parameters have a big effect on the model - and that 
is really, really hard to know over a wide area.  You could probably 
look at data from microwave satellite observations and come up with some 
estimates at km or larger scale, but getting data at 10 or 100 meter 
resolution would be challenging.  So you've got the problem of "antenna 
on a hill, next to a farmer's field, with a forest, and a salt water 
beach...

At some point, though, for HF propagation, it just becomes easier to try 
it and see what happens.

I will confess, though, that it is fun and instructive to sit and run 
through models when the rain is coming down outside.

> 
> 73, de steve ve6wz.
> 
>> On Jan 20, 2019, at 6:06 AM, Bill via TowerTalk <towertalk at contesting.com> wrote:
>>
>> Several people have mentioned the effects of a hill on a vertical or vertical system.
>> I also know of no program that allows terrain to be considered for a vertical.  Several people who I consider informed came to the same conclusion I did, it enhances the signal if it is a downhill slope and is detrimental if is an uphill slope.  For example, if one looks at a Yagi at one wavelength over flat ground and then at one over sloped terrain, you can see how the angle has dropped.  Likewise, if you look at the flat results and then the print out of the uphill slope, it is easy to see the low angle has been attenuated.
>> It's my belief and some antenna gurus, that the steeper the slope, the lower the angle if down slope, and the more attenuated if up slope..  At my QTH I have a huge downward slope to EU and NA and a huge up slope to the top of Mauna Kea (14000 feet) which is LP for EU and SP for VK/ZL.  On 80 I have a 4 sq and on 160 a bent vertical and a very good symmetrical ground system.  For the same conditions and distances I can tell you the signal is much better in the direction of the down slope  than the up slope.  Is it a scientific test?  No.  However, it does indicate that if one has a big slope (in my case a 1:6 slope uphill and downhill.) it does affect how a vertical performs.  Mine and several others educated guess?  Going downhill at my place probably lowers the angel at least several degrees.  No, I'm not sure if the angle is elevated in the up hill direction.  No doubt though it is attenuated.
>> Those who have heard  me on 80 and 160 in EU and NA will agree that the down slope helps the signal.
>> Bill KH7XS/K4XS
>> _______________________________________________
>>
>>
>>
>> _______________________________________________
>> 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
> 