Mini-nec does have severe limitations for complex arrays such as quads... You
need to retry the new modeling software packages, using nec, as they have
improved considerably... I currently use eznec, which has a subroutine for
intersecting wires such as quads... it generates wire quad designs that are
physically just a bit long when actually built... this is understandable as
it does not allow for velocity factor changes due to insulation, capacity of
the corner insulators, coupling to the house wiring, telephone poles, the sap
in the trees, etc... if I were able to more accurately reflect the
environment for a half dozen wavelengths around, it would generate models
that much closer...
The generated element spacings are right on as far as I can tell... I use the
calculated element lengths as a starting point... then, with the required
resonant frequency and impedence known for each element in the array, I use
the RF-1 Analyst to trim each element to these known numbers... this results
in a quad array that works as modeled...
This trim to the numbers approach is probably much faster (and more accurate)
than attempting to model the plumbing/wiring/towers/tree sap, etc. so that
the generated dimensions are correct to the inch...
Cheers ... Denny
--
FAQ on WWW: http://www.contesting.com/towertalkfaq.html
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: K7LXC@contesting.com
Sponsored by Akorn Access, Inc & KM9P
|