Towertalk
[Top] [All Lists]

Re: [TowerTalk] 80 M 4 SQUARE MODELING

To: towertalk@contesting.com
Subject: Re: [TowerTalk] 80 M 4 SQUARE MODELING
From: jimlux <jimlux@earthlink.net>
Date: Fri, 21 Jul 2017 12:19:57 -0700
List-post: <mailto:towertalk@contesting.com>
On 7/21/17 10:18 AM, Kenneth Goodwin wrote:
Whether it works or doesn't work, what's your alternative?  Biggest problem
with antenna modeling software is that the software can't take into
consideration the physical constraints without extensive sensitivity
analysis.

There are tools now that let you do that kind of sensitivity analysis in an automated way - since it doesn't take hours to run the NEC model for each try.

And, from a sensitivity analysis standpoint, you can use a "reduced complexity" model - you're interested in whether there are major changes in VSWR or gain or null depth (mostly the latter) as things move in the wind or relative to surroundings.

The surroundings can be modeled fairly simply for this kind of thing (sure, if you have a really good model of your house and surrounding trees, then load it in as numerical Green's function, but you probably don't need that).



And if you move your baseline model by 6 inches, the whole thing
changes and you don't have the means to control the physical parameters to
that degree.  Here is what you are supposed to have learned in numerical
analysis, the optimum solution is not necessarily the best solution. Ken
K5RG

Very much so.
Or another way - HF Yagi-Uda designs which have performance that falls apart if the end of the element moves an inch are probably not "robust"



_______________________________________________



_______________________________________________
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

<Prev in Thread] Current Thread [Next in Thread>