[TowerTalk] NVIS (not exactly towers, but HF)

[jimlux]

On 6/10/20 6:44 PM, Dan Maguire wrote:
> John Simmons wrote:
>>>> I noted that your two links are modeled using NEC2. Let's see the results using NEC4.
> 
> See  https://i.imgur.com/aXqmRJm.png
> 
> As Jim Lux mentioned there is a slight difference in the calculated
> feedpoint impedance at resonance (jX=0) but the Gain (dBi) values are
> almost identical.
> 
> Also see this follow-up to the qrz forum post referenced earlier:
> 
> https://forums.qrz.com/index.php?threads/single-ground-radial-wire-placed-beneath-a-dipole-how-effective-is-it.709471/page-10#post-5500865
> 
>

And this is a subtle details that only an antenna metrologist would be 
concerned with.

When you have a series of plots, is it the *same antenna* at various 
heights? Or has the antenna length been adjusted for resonance at the 
same frequency at each height?

Or in practical terms, if you have a tuner or matching network that is 
readjusted.

If you were doing A/B testing those kinds of things might make a 
difference, but then you have to get into stuff like the tuner's 
resolution, whether the transmitter exhibits any "load pull" behavior 
(the output impedance varies with the connected load and/or output 
power). As well as a host of other factors that you need to at least 
"know" to reduce the measurement uncertainty.

I think one of the take homes is "there is a best height" and "if you 
can't put the antenna at the "best height" it will still work.

It has to be pretty low before it's "horrible" (i.e. more than 10 dB 
down from "best height")