BAP wrote:

Where is the empirical or engineering data to support this conclusion?

Collins did do the frequency sweeps -- and it out performed many of the
alternatives at that time --
if it didn't Art Collins would have never used the connector (BNC was
available at the time).

Look at it as a simple tube... look at the outside diameter of the shield
and the inside diameter of the center conductor and plug and chug using
the formulae in the ITT Radio Engineer's Handbook for coaxial conductors.
You get something around 95 ohms characteristic for the front part of the
connector (from the insulator forward). The back part of the connector
is less important but ou have an additional discontinuity from the ring
around the insulator.

Say, didn't the Heathkit "lunchbox" AM transceivers use RCA phono
connectors as an antenna conection? Those lunchboxes operated into low
impedance 50 ohm loads all the way up to 148 mcs.

Yup! And the truth is, you can away with pretty lousy connectors in
most antenna applications. But put a pulse generator and a scope in
place of the transmitter and you'll see discontinuities at the connector
points.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."