Yes indeed. I hope no one has interpreted all this as meaning that I
believe it has any direct relevance to typical amateur antenna
applications. It doesn't. As Bill and quite a few others have stated,
the output Z of the PA isn't important at all for our applications. And
for nearly any calculation you care to do at HF, the assumption that Z0
is purely real is entirely adequate. The precise Z0 might possibly be
important if very precise measurements are being made, but that's not
something done by most amateurs.

But there was information posted that's incorrect, even if it's not
directly relevant to most of us, and that's what prompted my posting in
response.

Roy Lewallen, W7EL

William E. Sabin wrote:

Roger to that. In the special case of conjugate matching generator to
load, via a Z0 line, if we know the generator impedance we can do that.
But for PAs the generator impedance is "who knows what?" so the best we
can do is make the load equal to the complex Z0. Then forward power is
all there is and reflected power is zero. My Bird meter then tells me
that the calculated VSWR is 1.0:1.0. which is what my PA is designed for.

If my coax gets so lossy that I have to worry about stuff like this, I
will buy new (better) coax.

Bill W0IYH