75 to 50 ohm

[Bob Bruhns]

Hi Alex,

I am looking for my book with dimension equations for various
impedance coaxial lines with various dielectrics.  The signal sees
the space between the center conductor and the inside of the shield,
so the outer surface of the center conductor and the inner surface
of the shield should appear as smooth as possible to the wave.  A
linear taper from one set of dimensions to the other set of
dimensions is considered almost as good as the ultimate taper
design, which I gather would resemble some sort of a sinusoidal
curve.  This is starting to sound like windowing for an FFT!

Terman (McGraw-Hill, 1948) says linear taper is acceptable for
relatively small impedance transformations like this.  He says the
difference in loss compared to a more sophisticated taper is not
significant.  He suggests at least 1 wavelength of tapered section.
He says down to a certain frequency it is really a good transformer,
but below that it looks like an impedance bump.

  Bob, WA3WDR