Pete,
How interesting - I was certainly unaware of your prior work.
As you can see from my web site I came at this in the context of Hexbeam
- I was continually being asked "what's the best height". So I felt I'd
better try and spell out some of the issues. I decided early on that a
logarithmic figure of merit was the right approach because most of us
think in those terms, and it seemed clear that any "averaging" would be
invalid unless I converted to linear units.
I was vaguely aware of the debate over the antennas used to generate the
ARRL stats. Pragmatically, all I would say is that if there are no other
stats freely available at this stage the argument is interesting but
academic :)
I was pleased, but surprised, that QST accepted the article for
publication. Surprised because they seem to steer clear of anything as
frightening as a mathematical formula :)
Steve G3TXQ
Pete Smith wrote:
>
> Steve, I published substantially the same method in the US National Contest
> Journal in January 2001, with a correction in the September/October issue
> to incorporate the point about applying the weighting to linear rather than
> logarithmic relationships. N6BV incorporated a Figure of Merit using the
> same method in the first version of HFTA, which has been out since around
> 2003 or 2004.
>
> A rather more interesting question to me, at this point, is what antennas
> should be specified for the two ends of the path to generate the arrival
> angle statistics. There is a discussion of this on the ARRL TIS web page,
> but as far as I'm concerned the jury is still out on that subject; in the
> past, I had argued for isotropic antennas on each end of the path, so that
> the statistics would reflect the actual behavior of the ionosphere rather
> than the selection of antennas. I'm now coming around to a somewhat
> different view, if what you're interested in is trying to open up the most
> "layers" of stations to work in a contest, for example. At some point, the
> lowest layers are guys with low dipoles or lossy ground-planes.
>
> With luck maybe we can smoke Dean out and restart the discussion.
>
> 73, Pete N4ZR
>
>
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk@contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>
>
>
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|