[TowerTalk] Funniest thing I've seen in weeks

[Jim Lux]

At 08:41 AM 6/30/2004 -0700, Al Williams wrote:

> >----- Original Message -----
> >From: "W0UN -- John Brosnahan" <shr at swtexas.net>
>
>snip
>
> >What struck me as funny is the title of the symbol:
> >"Folded dipole aerial with 3 executives and 1 reflector sim. VDE"
> >Guess something got lost in the translation and 3 directors became
> >3 executives.
>
>Why is the longer element called a reflector while the shorter
>element called a director>  What is the origin of the naming?
>Basically, don't they both re-radiate because of the same reason?
>
>What is the relationship, if any, between the Yagi reflector and
>ionosphere and ground reflections?
>
>k7puc

I'm going to wing it here...

Probably it derives from earlier antenna design techniques relying on 
empiricism and intuition.  Consider the classic curtain array with a 
reflecting screen behind it.

Then, you'd go to two element arrays with a passive reflector behind the 
driven element (just because, conceptually, it doesn't look like there 
would be any "blockage" from having a shorter element in front).  I'm 
pretty sure that a dipole in front of a single reflector was developed 
empirically, with the optimum length for the reflector determined by trial 
and error as opposed to analysis.

Yagi and Uda did their work long before all the basic analytical techniques 
we use now were developed. (the Yagi paper is 1928, Uda's in 1927)  I'm 
sure that their design was an outgrowth of experimental methods, and with 
the term reflector already established, the addition of the term director 
was a natural extension.

A fascinating paper by G. Sato, in IEEE Ant & Prop Magazine, June 1991, 
talks about Yagi antennas and their history.

Carter published his paper on circuit relations in 1932 (which started the 
idea of considering multielement antennas as coupled radiators in a circuit)

The Hallen paper (upon which most of the modern analytical techniques for 
mutual impedance are based) was in 1938.

Hansen and Woodyard did their paper on superdirective arrays in 1938 (Most 
all amateur beams are a case of a superdirective array), but they were 
concerned more with "all driven" arrays (such as used in broadcast or radar)

Schellkunoff published his paper on arrays in 1943.
Walkinshaw  did one of the first analytical papers on Yagis in 1946 (for 
special cases 4 directors), Green published a paper in 1966 giving 
experimental designs for some cases; and even as late as 1969, Thiele 
published his paper on Analysis of Yagi-Uda-Type Antennas, which probably 
was the first real "computer optimization" of these sorts of antennas (on a 
7094, which had enough memory for 27 elements!)


You're right, though, they all reradiate because of mutual coupling, and 
there's no particular reason why you couldn't for instance, make an antenna 
with 3 reflectors and a driven element.  Of course, you have to consider 
other aspects of optimization: minimizing mass and size (and cost of 
aluminum, and structural support, etc.).. this tends to drive you to one 
element behind and many in front.

There may be advantages to multidirector designs in multiband antennas too.

You've also got to consider marketing. People expect to see one reflector 
and many directors, and if you were to design and market something 
different, you'd have to do a lot of explaining to sell it.  Consider, for 
instance, a design like the Force 12 C3SS, where the driven element is 
somewhat tricky, with the 10 and 15 meter driven elements excited by the 20 
meter element, rather than a direct wired connection.

I'll comment that designing (and optimizing) an antenna design like the 
C3SS would probably be impractical without a computer to do the analysis, 
which is why you don't see similar designs from the 40s and 50s or even 
60s.  In that era, you'd be doing your analysis with analytical means, and 
you'd naturally gravitate to a trapped design for multiband use, because 
the analysis is at least tractable.  Even an "interleaved" design where 
you've essentially got multiple monoband arrays on the same boom would be a 
nightmare to analyze without a computer.