[TowerTalk] Log Periodic Choice

[jimlux]

Ralph Matheny wrote:
> NO disrespect intended, but I frankly don't put a lot of stock in one set
> of antenna "measurement" over another set.  Indeed one might be able to
> make comparisions between several antennas and find a dud that is 10 db 
> below the rest, but to resolve 2-3 db over a dipole with accuracy just 
> won't happen to my satisfaction.  There are too many things that can 
> change and and go wrong in the worst way.
> 
> 

I will grant that getting 2-3 dB accuracy in an afternoon of testing in 
someone's back yard using the s-meter on a 30 year old rig is dreaming.

However, with some moderately careful technique, and simple equipment 
(calibrated step attenuator), it's very possible to get 1dB sorts of 
precision. There's a lot of work involved, though, in hoisting antennas 
into place, etc.

If the antennas a reasonably simple, modeling can easily give you 
relative performances to better than a dB, but, of course, can't readily 
account for the detailed effects of a specific ground/soil environment. 
   On the other hand, the impact of the ground will be quite similar for 
antennas that are generally similar (i.e. comparing a horizontal dipole 
against a horizontal yagi).

The other thing performance related that's quite valuable, but for which 
  fairly low precision measurements are useful, is understanding the 
rear and sidelobe behavior. A fairly crude measurement can distinguish 
between 3 dB F/R and 10dB or 20dB F/R