Topband: Tuning elevated radials

[Ford Peterson]

Gary-K9AY wrote:

> You hit one of my "hot buttons" -- I have not seen any convincing evidence 
> that equalizing currents in elevated radials is a valid exercise.  Even if 
> it is, the "solution" of changing radial lengths has much greater potential 
> for introducing pattern skew.
> 
> What is more important is to make sure that each radial is actually an 
> electrical 1/4-wave, with maximum current at the common connection.  The 
> easiest way to to this is to grid-dip or measure the impedance of pairs of 
> radials as dipoles, adjusting the lengths until the dip is at the desired 
> frequency and/or reactance is zero.  (Actually, you just want them of equal 
> electrical length so they have similar current distribution -- this is by 
> far the easiest to accomplish when they are 1/4-wavelength.)
> 
> Layout symmetry is also important for omnidirectional radiation (if that's 
> what you want).  Belrose has written quite a bit on this topic.

Gary, and others...

My thinking has been that the characteristic null that is supposed to occur at zenith is a good thing.  If nearby QRN arrives at a high angle, then the deep null at zenith should be a good first line of defense against noise.  

I have been playing around with models of the system and can see that even a slight change in electrical imbalance of the various radials causes a somewhat insignificant skew to the pattern at the horizon, but a significant skew at zenith.  Intentional distortion of the radials in an attempt to get some directivity is a waste of time, providing only 10dB or so of pattern skew at 23 degrees and no real increase in gain in the favored direction.  However, at zenith, the pattern changes by 30-40dB (that is if you can believe the perfectly balanced models over a perfect ground geometry).  

When I take your recommendation to the field and attempt to implement, the process is confounded by my use of an electrically massive structure at the feedpoint.  The radials do not all terminate at a single point.  The radials attach to a bulky bus wire wrapped around the tower, with the feedline attaching at a single point on one side of this loop.  Each radial then has multiple paths back to the feedline.  The feedpoint itself then becomes unbalanced by many feet.  So my thinking was that measuring and balancing the currents on individual radial wires would be the way to go.  Rather than tuning for 0j, tune for maximum (balanced) fire-in-the-wires!

I think I'm chasing a holy grail.  That's the problem with doing antenna work at 74 F in beautiful sunshine.  Antennas seem to play better when you build them at 30 F in blowing MN sleet!  When the pain becomes great, antennas are "good enough" if they take power and you get to head-back-to-the-shack.

Thanks for the help.  I think I'll press on and forget about radial tuning minutia...

73

Ford-N0FP
ford at cmgate.com