Owen Duffy wrote:
. . .
(To preempt Roy, the velocity factor of the outside surface of the shield
clad with a thin layer of vinyl will be close to 1, close enough for the
purpose at hand.)

I'll just add the reason -- the common mode current, which is what
causes feedline radiation and what we're trying to suppress, is on the
outside of the coax. The commonly specified velocity factor (around 0.66
for solid dielectric coax, a bit higher for foamed dielectric) applies
to the field inside the coax where the differential mode current flows,
not to the outside where the common mode current is. So you use a value
near one as Owen says. And it's not at all critical for this purpose.

A standing wave is present on an antenna or radiating feedline -- every
half wavelength there's a current null, and offset a quarter wavelength
from these are current maxima. For example, there are current nulls at
the ends of a half wavelength dipole (or, an even better example, a Yagi
parasitic element) and a maximum at the middle. When you insert a balun
in a transmission line, it causes a current null at that location, so
there'll be a maximum a quarter wavelength down and another minimum a
quarter wavelength below that if current is induced by coupling and the
length to ground supports that distribution. We want to make that
distribution impossible, and inserting the second balun does that.

This is easily observed by modeling, but you have to keep in mind that
the actual path the current takes to the Earth along the outside of the
coax can be considerably more complicated than most simple models
represent. So models can tell you what *can* happen although maybe not
necessarily what *is* happening in a given installation.

Roy Lewallen, W7EL