DG02816 said:
In the hope of minimizing confusion, I'm going to try describing the WALE configuration as having two COLUMNS of three towers each with the columns roughly 180 degrees apart and the Langer 1140 configuration as having three columns of two towers each, also with the columns roughly 180 degrees apart. This nomenclature leads to WALE having three ROWS of two towers each with the rows separated by approximately 90 degrees and the Langer 1140 configuration having two rows of three towers each, also with the rows separated by approximately 90 degrees. In each case, the columns are roughly parallel to the azimuth of the radiation maximum and the rows are roughly parallel to the azimuth of a line joining the radiation minima.
Among six-tower arrays, the Langer 1140 configuration is by far the more popular of the two, which is not to say that the WALE configuration is unique. Although both configurations are capable of generating narrow teardrop patterns, examination of a bunch of examples of both types suggests that the Langer 1140 configuration more readily produces deeper minima. OTOH, in situations where the rows are 90 degrees apart and the columns are 180 degrees apart, the WALE configuration requires slightly less land than the Langer 1140 configuration--even when you account for the land occupied by the ground radials: 540 degrees by 270 degrees (145,800 degrees^2) for the Langer configuration vs. 360 degrees square (129,600 Degrees^2) for the WALE configuration. Most likely, though, the shape of available plots is more important than the area. Moreover many WALE configurations space the columns somewhat further apart than 180 degrees and many arrays of both types space the rows closer than 90 degrees--sometimes as close as 60 degrees. Furthermore most of these arrays are not rectangular but are parallelograms. That is, the angle between the rows and columns is not 90 degrees.
The result is that if someone wanted to diplex the 1140 station into the WALE array, it wouldn't make sense to use the design that Langer proposed for 1140. A new 1140 design optimized for two columns and three rows would likely be the way to go.
