I recall the problem with my original V4k, I called the Qv4k.
In the second prototype, which was the first that had the cone radials grounded to the mast at the bottom, I attached the radials in such a way that they curved nearly parallel to the vertical for several inches instead of angling away at around the proper 30°.
It was a time when I was trying to sort out the right size for the gamma match, and the real dimensions of the antenna. I wasn't certain at that point about many things, but the discussion raised the question regarding the potential significance of the angle of the radials. I now realize I may have been concerned with capacitance between the radials and the vertical, but at the time, it was just trying to get the dimensions right in order to get the experience right. I wondered whether the antenna would tune out properly and match where it should if the radials skirted too closely along the vertical.
I believe you are correct in saying when once the question was answered that the proper angle away from the vertical from the bottom upward was about 30° the discussion moved beyond any other deeper point that we seem to have just now revisited.
I have now begun to suspect that having the radials emerge directly off the vertical in this fashion
may actually impact the true length of the radials due to the way your model has reacted to this kind of setup. How much, or whether at all or not, I hope to find out.
I will build this connection for the next one, but with four stand-offs for a four radials cone:
