[BC] Skirt feeds.....

[Phil Alexander]

Dave,

I'm sure that will work although I've never tried it. 
It would make building the model a lot easier, but it
won't give a very good representation of what happens 
in the real world.

It is true that the NEC model is very "finicky" at the
bottom end near ground, but if you model a six wire
skirt with a girdle and drive it through a spider I have 
found it works very well. However, you have to feed the 
spider in a location that is inside the center of the 
tower and that geometry gets a little bit tricky. <g>

You can either join spider legs at the lower ends of 
the drapes or - what I like best - is use 12 wires in 
the girdle so there are mid-point joints between the 
drapes for connecting the spider. Remember that a well 
constructed skirt fed uni-pole has a high drive impedance 
which does cause unusual sensitivity to imbalances near 
ground. It is only when you see this in the model that 
you start to understand how important the feed can be.
At least that is when the light came on for me. <g>

Careful balance is the key. Fortunately I have a program
(EZNEC Pro/4 v.5.0) that displays currents in the wires as 
a visual line more or less parallel to the wire with current
proportional to displacement from the wire. This is much
better and easier than opening the window that lists
currents of segments and trying to visualize the result.

Phil Alexander, CSRE, AMD

-----Original Message-----
>From: Dave Dybas <dd92251 at aol.com>
>
>I've had pretty good results modelling folded mono-pole towers 
>by following this:
> 
>If a skirted mono-pole is modelled in NEC there are a few 
>tweaks necessary to obtain valid results.
> 
>A) The skirt wires should not be tied together at the 
>bottom...in the model.
> 
>B) Each skirt wire should be excited by its own source.
> 
>C) All sources should be the same value
> 
>D) The actual base impedance is the value of one of the skirt
>wires divided by the # of skirt wires. I'm certainly no expert 
>on the subject of NEC or folded-mono-poles but after a little
>research and hours of computer modeling, I think I've got a 
>better grasp of how to do it.