[TowerTalk] NVIS antennas Re: dumbing down

[Jim Lux]

At 02:00 AM 7/22/2005, Tom Rauch wrote:

>At 1/8th to 1/4 wl high NVIS signal is maximized. Anyone who
>spent any time at all with antennas would know even with a
>perfect reflector efficiency drops rapidly when the antenna
>is spaced less than 1/10th wl from a reflector, and if the
>reflector (earth) was lossy things would go in the toilet
>fast as height was reduced.


Might well be, though, that a 18" or 2' high wire is much easier to deploy 
in a field situation, than trying to get a wire up 1/4 wavelength on 40 or 
80?  From a system standpoint, you might be willing to accept the lower 
efficiency (and the certain interactions with objects likely to be near it) 
in exchange for a faster deploy.

Certainly, if it were stormy weather, I'd much rather have an antenna 18" 
off the ground than try to put up a 60 foot piece of wire some tens of ft 
in the air, as long as the low antenna actually "worked" at all.

The real interesting question might be, how low can you go?  Is the 
efficiency dropoff a linear gradual decrease from 1/4 wavelength, or is 
there some point of diminishing returns where the loss starts to really 
increase quickly. Maybe 18" was chosen by the first person to try it 
because they happened to have 18" high supports, and that became 
legendary.  18" is a convenient height: you can step over it, it's high 
enough that you won't trip over it, it's low enough that you won't touch it 
with your hands unless you reach down, dogs can run under it.

As far as vehicular NVIS antennas go, there's actually been a fair amount 
of theoretical modeling and antenna range testing of various 
configurations.  Some of this work was to figure out "how do we use what we 
already have to best advantage" (i.e. using that big long whip... do we 
mount it in front and pull it back, mount in back and pull it forward, put 
a hinge in it, bite the bullet and replace it entirely)  For instance, 
there's a bunch of papers by Austin and his coworkers that are of interest.

(I note that Austin (1998) says:  Extensive tests have shown that the 
proper antenna for NVIS operation is some form of horizontal element, such 
as a lambda/2 dipole at a height of lambda/10 to lambda/4 above the 
ground.  However, at typical NVIS frequencies, both the length of the 
antenna and its height above ground make such systems feasible only in 
non-mobile applications.  For mobile use, or even in the static mode if 
space is severely restricted, non-resonant antennas (i.e. of length 
<<lambda/2) must be used.

Austin cites as a reference for the above dipole recommendation a paper by 
Hagn and Vincent:
"Comments on the performance of selected low power HF radio sets in the 
tropics", IEEE Trans on Veh. Tech." VT-23, 1974, pp 55-58.

I don't have a copy of this particular paper (but I'll get one).  However, 
I've talked to George Hagn on other topics, and he is a big believer in 
experimental methods, so I'll bet that if that paper recommends certain 
antenna heights, it was as the result of a lot of tedious testing of 
antennas at various heights.  However.. one has to be careful, it might be 
antennas in a jungle, which may not be representative of typical US NVIS 
use. An antenna in a jungle is essentially buried in a dielectric layer, 
which looks like a waveguide, and the propagation to a distant point, also 
within the jungle, might be carried entirely within the jungle, as opposed 
to up to the ionosphere and back down. (That would be all those papers by 
Cavalcante, etc. that were discussed on this list a year or so ago)

Whatever it is, the goal on practical antennas for field comm is not 
necessarily to make the "best performing antenna", but to make an 
"adequately performing antenna" that meets all the other requirements.  It 
might well be that they're willing to give up 10 dB of loss in exchange for 
adding a 500W SSPA.  They're not going for maximum miles per milliwatt QRPP 
contesting.