[TowerTalk] C-31XR vs. 4 el. SteppIR

[Jim Lux]

At 04:24 PM 3/16/2005, Bill wrote:
>Dan,
>
>I'm not trying to pick a fight and I know that I am going to regret asking
>this question but...

But hey.. that's why it's a "discussion list"... from discussion comes 
understanding and enlightenment.

>How can a 4 element multiband yagi with a compromised boom length deliver
>gain approximating that of a well designed 5 element monobander? This
>doesn't seem to make sense to me. (Regardless of who made the antennas.)

The well designed 5 el monobander is, by definition, a compromise design of 
some sort.. either feedpoint impedance bandwidth, or F/B, or forward 
gain.  An enormous amount of effort has gone into coming up with "useful" 
combinations of these that are still implementable by fixed elements.

It's entirely possible to get very high directivity (and possibly gain, 
depending on losses) with a physically small antenna, at the expense of 
narrow impedance bandwidth.  That's the whole Chu-Harrington tradeoff in a 
nutshell.  What the SteppIR allows you to do is to create an antenna design 
that has higher Q (and therefore, potentially higher gain, assuming the 
higher Q doesn't kill you with dissipative loss), because the bandwidth 
only needs to be 10-20 kHz, as opposed to a few hundred kHz.  I can't find 
my reference that relates the increase in Q to gain, so I can't tell you 
what a factor of 10 will do for you.


I'll note that an antenna fitting in a 24 foot diameter sphere at 14.1 MHz 
has a "Chu Q" of something like 1.7, which is quite low.  Such an antenna 
would have a "normal gain" (i.e. non-superdirective) of about 5 
dBi.  Clearly, there's a lot of room for jacking up the Q, considering that 
typical Yagis have Qs of 70 or so, and this is why the typical gain is up 
around 8-9 dBi for such an antenna.  Harrington says in his paper that 
doubling the gain (3 dB) is possible for antennas on the order of 2 
wavlengths long, and that as the antenna gets shorter, the potential 
increase is even greater (assuming the losses don't bite you first).

Harrington also comments that the maximum gain will be achieved with an 
endfire array with elements spaced a quarterwave apart (but not uniformly 
weighted in either phase or amplitude).


>Bill, W5VX