Thanks again for the reply. Please don't think I'm trying to be a smart-aleck below - I know some theory but I do not have any practical experience working with these signals.
I am not familiar with the 8VSB modulation we use here in the States, so I do not know how to quantify the signal purity required against uncorrelated interference and noise.
I know my present TV was an early design so the equalizer performance is probably very much sub-par by today's standards.
I would think that a good equalizer would treat a weaker second copy of the same signal as multipath and make use of the power in both signals. If the equalizer could not resolve the delay difference then the second copy would become a definite interference. If the delay difference was too small the signals could aid or destroy the signal power before they get to the equalizer. This could be solved by changing the cable length on one of the antennas.
The relative power of the two copies of the same signal would be set by the gain of each antenna in the direction of the transmitter. For the two 8-bay designs this would be a 12-20 dB difference at this 100 degree radial spread. For a well-designed receiver I'd think that would be enough of a difference to prevent problems either way.
I have a natural aversion moving parts, which is primarily why I don't want a rotor setup in my attic. (I also hate insulation!)
Surely someone else has been dumb enough to try this.
Don