On Fri, Jan 24, 2014 at 7:07 AM, Richard Fry <rfry@adams.net> wrote:
> But that wasn't necessary to make the point that the system with elevated
> radials installed over poor earth having no buried radials in it has
> essentially the same performance as the system using 120 x 1/4-wave radials
> (only), buried in that same earth.
"...essentially the same..."
A scientific definition of that would be interesting. My impression is that
I am highly "unappreciated" by certain folks when I use such terms. An
additional half dB loss in the antenna system is a permanent 12% increase
in the power bill for the transmitter if one is required to maintain
specified field strength. To a station manager that's probably an entirely
unsatisfactory increase in moolah drain. With a 10 or 50 kW station, that's
a *killer* difference. He would not use the term "essentially the same."
For some restricted ham situations being down only 1.5 dB from a commercial
method reference could easily be as good as is possible, and only 0.5 dB
would be wonderful. But for the science, 0.5 dB is simply 0.5 dB, and
worthy of accounting. A 0.5 dB here and a 0.5 dB there, and 0.5 dB
elsewhere starts to add up to barely catching some DX otherwise not had.
And a confirmation is a confirmation.
Run the four elevated over the radial field. The buzz I hear on the
grapevine about 4 elevated is that's the remedy for a radial field gone
sour, of course leaving the old radials *retired in place*.
Trying to disprove experience in non-commercial small lot situations by
referring to NEC 4.x model runs from the the middle of the commercial MF BC
paradigm just won't cut it. I have already stipulated many times that NEC
4 is calibrated for that paradigm and seemingly delivers well documented
good results. In the same manner, you can't prove distant low and sky-wave
patterns by measuring at the ground or only locally.
The small lot experience is that losses to ground are underestimated by NEC
4.x, the underestimation an opinion notably shared by Roy Lewallen, W7EL,
author of the respected EZNEC series of NEC shells, who has spent some
serious time chasing the issue.
I have gone to using .0002,1 (ghastly ground) as ground characteristics for
testing model changes for ground sensitivity, giving the worst possible
results for non-dense-radial counterpoise designs. This is used for
comparing a "before" model to an "after" model, with wire or counterpoise
changes. This accentuates changes in results due to the way ground is
induced. Picking solutions with the least sensitivity to ghastly earth,
again and again brings "happy results" in real world applications which are
borne out in RBN changes.
While one might think "ghastly earth" is too severe, difficult urban and
concrete-asphalt-ish small lot situations sometimes seem even worse,
pointing to some mechanism in play that is unbookept in NEC ground
treatment, or possibly undiscovered. Since the NEC ground method is both
monolithic characteristics and for essential reasons confined to ground
*approximations*, this is really not surprising.
73, Guy.
_________________
Topband Reflector Archives - http://www.contesting.com/_topband
|