[BC] Dallas' 1190

[PeterH5322]

 >What I foind interesting is that the major lobe of that 12 tower 
site appears
 >to be broadside from the array, with a large minor lobe coming off the
 >"bottom".

On such an array, the pattern maximum is always "normal" (i.e. 90
degrees) to the axis of the larger of the two spacing units.

The back-side suppression is always effected on the axis of the smaller
of the two spacing units.

The minor lobe towards the SE is a consequence of asymmetrically driving
the elements of the array, IOW, this array is not a strict
"multiplication method" array, where the six-tower broadside sub-array is
multiplied by the two-tower sub-array.


 >It still amazes me how patterns like this can be "manufactured" with
 >such
 >precision these days. The wonders of computer technology and GOOD 
Consultants
 >to oversee the build and tune-up!

The technology to more-or-less automagically design arrays of any
complexity has been around since around 1970.

The technology to analyze and compute arrays of any complexity has been
around since around 1965.

But, it would take a few years for the traditional consultants to embrace
such CAE (computer-aided engineering) tools.

One of the better examples of CAE in array (re)design is an infamous
station in Detroit, which went from 12 towers and 5 kW (unstable, and a
"critical array", requiring 24/7 operators) to 9 towers and 3 kW (ditto)
to 9 towers and 10 kW (stable and no longer a "critical array", operable
by remote control) over the course of about a decade.