Richard Clark wrote in
:

On Sat, 12 Jun 2010 19:52:15 GMT, Owen Duffy wrote:
....
My recollection of Walt's tests were that they tested at points other
than Zl=0 and Zl=infinity.

Steps 1 and 2 are quite explicit.

I have reviewed Walt's additional material at
http://www.w2du.com/r3ch19a.pdf .

Walt's load at step 1 is stated as 50+j0 without tolerance, but I accept
that Walt is thorough and would have confidence that it is low in error.
He reports Pf=100W and Pr=0, again accepted as indicating that the load
and calibration impedance of the '43 are similar, and probably quite
close to the nominal impedance, close enough for the purpose at hand.

At Step 8, he substitutes a load with measured impedance and calculated
rho^2 as 0.235, VSWR(50) is approximately 3. Reported Pf=95W, Pr=25W, and
on that basis rho^2=0.263. The discrepency in rho is not discussed in the
paper.

A key observation between Step 1 and Step 8 is that changing from a VSWR
(50)=1 load to a VSWR(50)=3 load resulted in a 5% change in observed Pf.
This observation depends on the difference between two direct readings
from a single instrument (the Bird 43), and the accuracy of calibration
of both the Bird 43 and the 50 ohm load to their nominal 50+j0 impedance.

I have given a mathematical development at http://vk1od.net/blog/?p=1028
of why Pf should be independent of load impedance.

The observed change in Pf at 5% is rather small considering the load
change, and suggests that Zs in this case is probably quite close to 50
+j0 although the experiment was limited to just one load case.

I make the "just one load case" condition because in my experience, a
transmitter might exhibit insignificant change in Pf for some load
variations, but for others it is signficant. The case I documented
earlier is one such example.

Owen