Jim Lux wrote:
. . .
You can measure the feedpoint Z with probably 10% accuracy, but it's the
Rrad that's the tricky thing to determine, and without that, you'll not
get the ability to infer Rloss from Rfeedpoint. Of course, if the rough
and ready approximation shows Rrad to be 0.5 ohms, and you measure 25
ohms at the feedpoint, then you can just ignore Rrad, and assume the
loss resistance is 25 ohms (since the uncertainty in the measurement is
2.5 ohms)
. . .

And of course that's just where modeling becomes valuable. Even a crude
model will probably predict Rrad better than a "rough and ready
approximation". And a decent model will very often give a value for the
feedpoint Z that's more accurate than a casual or, sometimes, even a
careful measurement.

Roy Lewallen, W7EL