On Sun, 2 May 1999 17:55:44 -0400, you wrote:
>
>A tribander 20' above the guy station should not interact enough to cause
>a problem except on a computer.
Modelling getting good enough to handle a lot of things that can
barely be measured. Unbroken wires 20' below a tribander can be either
trivial or totally screw up the pattern, depending on lengths. If the
SWR changes a couple of tenths, pattern is being messed with.
>This was supposed to be an excersise in reality, not
>perfection...remember.
The modelling has it's bad moments, it's weaknesses, it's occasional
not-to-be-trusted results. HOWEVER, working within the limits, it
gives valuable insight into the behavior of RF on conductors that is
nearly impossible to measure with the equipment available to most of
us.
I modelled a 40m 45 degree sloping folded dipole, a matching section
of 300 ohm twin lead to get a wideband match to the 450 ohm ladderline
in use, so I could use a 9:1 transformer at the house end without
worrying about the stuff that goes wrong with transformers operating
away from design impedances.
I cut the stuff to lengths specified by the modelling (EZNEC2 for this
one), strung it up. 1.5:1 at the band edges, 1.2:1 in the middle,
works great.
THAT's an exercise in reality. Putting stuff up without trying to
validate design with modelling is an exercise in denial of reality,
unless you have an antenna test range and many K bux worth of
measuring equipment.
73, Guy
--. .-..
Guy Olinger, K2AV
k2av@qsl.net
Apex, NC, USA
--
FAQ on WWW: http://www.contesting.com/towertalkfaq.html
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: owner-towertalk@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|