I assume you mean by your reference to "...small physical separation between the two active elements..." that the 1/4 wave cage curves in toward the phasing coil at the top and thus is close to the radiator, right? I doubt you mean the separation allowed by the insulator is too close.
That said, I don't know what the full impacts is, but think I see this bottom section designed as a form of a cage radiator, for which the advantages are well established. Such designs have an affect that adds a positive affect to the bandwidth and the Q lowering feature noted with very large diameter radiators at a minor cost to weight, wind loading, and material costs.
IMO, the phasing either works or it doesn't. The top coil should determine that and if it didn't I think you would have seen a rather sorry operating antenna, that very likely showed a rather high angle radiation pattern just based on its length alone.
In the ARRL AHB page 7-7, 17th addition, this is discussed. There noting that the spacing of the sleeve elements could be closer and that doing so will help result in a more circular azimuthal radiation pattern rather that showing directivity.
I attach this info here:
[ATTACH]4574[/ATTACH]
Since this design seems to fit my understanding of what a vertical collinear should look like in general...I see little problem with the idea, unless you're right, that instead...the tune means everything in the Big Mac as in the Sigma4.
BTW, here is the formula I used to figure the length ratios as per the Sleeve Antenna section in the book noted above. This does not account for material diameter and I assume the numbers would, as a result, fall closer to the elements made of thin wire.
wavelength in feet x (ratio x .25) = length in feet
36.18 x (3.2 x .25) = 28.80'
36.18 x (3 x .25) = 27.14'
the text tells us that anything over 3.2:1 ratio of radiator to sleeve will produce very high sidelobes and less gain on the horizon. I know you based your work on this subject study and report, so what do you think accounts for the difference you saw with your length at or near 31'? The article goes on to describe some exceptions at several larger ratios 4:1, 6:1, 8:1, etc, but in those cases the very limiting result are noted in the text right below the sections I highlighted. It looks like 31' feet fall way to long for the minimum ratio of 3.2:1 and way too short for the next ratio of 4:1.
Since I'm talking of the top of my head and understand little of the subject, how can I reconcile this problem?
