John - KD5YI wrote:
....

Okay, Reg, then go read the material referenced by footnote 13.
That's one
reason I included it. Maybe that way we won't need to rely on your
memory.

The confusion comes (in the quoting of the texts) because of a failure
to consider that there are two different mechanisms that can limit the
power handling capability of the line. One is power dissipation
(temperature rise), and the other is voltage breakdown. Clearly the
minimum power dissipation for a given input power and matched line
occurs where the line attenuation is minimum. But if you make the
inner conductor slightly larger, it may be able to get rid of heat
enough better (for a given line construction) that the inner conductor
temperature rise is slightly lower, even though the power dissipation
is slightly higher. I would expect, though, that the optimal
construction in most circumstances would result in an impedance only
marginally lower than the minimum attenuation case, and the improvement
would be a very small one. You'd have to convince me it was really
important to get me to worry about it beyond just minimizing
attenuation.

If it's voltage breakdown that limits the line power handling
capability, the air-insulated impedance of the line will be at a D/d
that results in about 30 ohms impedance for air-dielectric line, and
..66 times as much for solid polyethylene line.

And if it's maximum voltage-handling you want, the D/d results in
somewhere around 50 ohms with air-insulated line, about 33 ohms with
solid poly, if memory serves. I could look it up if it's really
important.

Generally in ham applications, (reasonably well matched) lines will be
power dissipation limited, not voltage limited.

Cheers,
Tom