[TowerTalk] Another Ground Radial Question

[Jim Lux]

On 5/24/11 7:08 AM, pehaire wrote:
> I agree with you.
>
> In Beer's article on using different verticals in a 2 element phased array
> by varying the lenght of the feed line to each vertical, he uses radials and
> he pointed out that when radials crossed he bonded them together.
>
> It looks like to me that a half square could be thought of as 2 verticals
>   each with one radial and the 2 radials attached or bonded together at their
> ends.  Thus 2 verticals spaced a halfwave apart.
>
> Noll in his book describes 2 verticals spaced almost a half wave and fed
> with  300 ohm line and has a switch in the center of the 300 ohm line that
> he can go from end-fire to broadside.  The close spacing I guess is to
> reflect the velosity factor of the 300 ohm line. The close spacing causes a
> slight loss  in gain. But who cares about plus or minus one db.  That is one
> sixth of an S unit.  Or you could increase power by one fourth and more than
> make up for the loss.
>


In most phased array schemes, you're not looking for forward gain as 
much as reverse null depth.  You can have huge phasing errors with small 
effect on gain: The gain (in dB) is 20*log10(1+cos(theta)) where theta 
is the phase error.

10 degrees is 0.05 dB gain decrement
45 degrees (an 1/8th wavelength!) is only 1.35 dB decrement.

You can do better than 45 degrees without a tape measure by just pacing 
the distance between antennas.


But that same 10 degree error in phase changes an infinite null to -36 
dB, and the 45 degree error changes the null to only -10dB.