I could not get the two types of 2 coil Polydoroff windings to be especially directional, or sensitive. I found a barefoot PL-450 was better completely separating two 5 KW stations on 560 AM in the evening and gave stronger signals with less noise. That is working with stranded #14 insulated wire for these experiments.
Because the 'new' ferrite requires more turns to reach the same uH, I was not able to assemble a 3 coil over ferrite to test in this smaller size. 1.6 X 3.5 inch stack just under each 3 inch wide coil.
A few years ago I found a 3 coil over my 2.125 X 16 inch ferrite stack was sharply directional. By email I asked both Everett and Steve R. and neither had any idea why that may be.
I plan to buy more ferrite to go under a third coil in a couple weeks. Although my 3 coil is not exactly from the 2 coil Polydoroff patents shown in files here repeatedly, it is some what similar as the cap or DSP chip is connected between the inner ends of the outer coils, and the contra coil in the center between the jumper from the outer ends of the side coils, so I think the cap is balanced between all three coils. In the files it says that would be the highest voltage point. I suppose that means the leads to the cap or DSP chip should not be close to each other, but that's just a guess.
I would welcome suggestions for things to try.
If the tests with stranded #14 and three Contra coils goes well, I plan to buy Litz hoping to improve it more. 660/46 Litz is slightly narrower than my #14, so I could add a few turns to each of the three coils in series for slightly more uH.
Steve R. found spacing the ferrites side to side under coils on an FSL increased Q, so I may try that on the toroid stack. With out test gear, it may not be obvious if Q increases some what.
If I can't make the separated 3 coil design to work well, I can just wind a single coil of Litz over the stack and experiment with spacing the ferrite toroids underneath. -FARMERIK