When you do problems like this with modeling, be careful. While the wire
moments are specifically calculated segment by segment, what gets
calculated around ground is not literal calculation like the wires. Ground
is done with a GROUND APPROXIMATION METHOD (e.g. Norton/Sommerfeld). That
means a computational method that renders some accepted degree of
conformity with measurements from standing-man-with-meter.
Ground approximation methods are used because if the literal moments are
used to sufficient accuracy to use ground literally as a conductor, program
run times extend into days and even weeks. Further the instant data to
support those calculations at a specific site are not available, and a
monolithic, uniform soil characteristic is assumed. Without digging holes
everywhere, the instant site data is not available and we are left with the
assumptions and the approximations.
You take your risks with how far, and into which dark corners, ground
approximation methods generate results close enough to reality. Models
are not reality. We just hope that they are close enough to minimize the
actual repetitive task of prune and tune and measure on an antenna range.
Over time CERTAIN kinds of results have been programatically adjusted to
rather keen accuracy. Extrapolating the success of these carefully
cultivated areas to the universe of problems thrown at the models can tell
you things that just aren't true.
There are plenty of measured situations that NEC2/4 cannot generate
correctly. Most I've seen have to do with dirt.
NEC2/4 ain't a bible. It's a method.
73, Guy.
On Tue, Oct 2, 2012 at 2:17 PM, Dan Zimmerman N3OX <n3ox@n3ox.net> wrote:
> > Richard Fry, you know anything about this? A 5:1 change in soil seems way
> > out of line with what I recall from fields at WSPD on 1370 kHz and WOHO
> on
> > 1470 kHz. Were those stations exceptions?
>
> I don't know about further out where you'd have to proof an AM station
> but I just ran some models and there's kind of a weird relationship
> between vertical electric field strength at 1 wavelength out and far
> field efficiency as ground conductivity is varied for fixed
> permittivity.
>
> http://n3ox.net/files/1500kHz_monopole_cond.png
>
> This was a #12 copper wire monopole fed over 64 0.5wl radials 1 foot
> high. Power is held fixed at 1000W. The reference "0dB" case for both
> the vertical field measurement and the far field efficiency was their
> equivalent values over perfect ground, but with copper wire loss
> turned on, so the reference far-field case was about 82% efficient.
>
> I want to try different distances and a higher frequency like 40m
> because I think the results will be quite different. Would also be
> good to, instead of looking at the total radiated power, compare the
> far field gain at some elevation angle vs. total field at that
> elevation angle nearer the antenna. But at any rate, seems this
> plot is relevant to what you can deduce comparing FS measurements over
> different soils.
> _______________________________________________
> UR RST IS ... ... ..9 QSB QSB - hw? BK
>
_______________________________________________
UR RST IS ... ... ..9 QSB QSB - hw? BK
|
