walt wrote in
:

Owen, I don't understand where you found a 5% drop in power when you
say it should have been 0%. A 5% drop from what value to what value?

Hello Walt,

I wrote an article with the mathematical development for the fact that Pf
(as indicated by a properly calibrated directional wattmeter such as the
Bird 43) is independent of load impedance if and only if the Thevenin
source impedance is 50+j0 ohms. The article is at http://vk1od.net/blog/?
p=1028 .

This provides the basis for a simple go/nogo test for the hypothesis that
a particular transmitter, under particular conditions, exibits Zs=50+j0.

Where did you find that data? If you're referring to 100w Pf with the
50 + j0 load and then the 95w Pf the mismatched load, you've got to
understand that the source was now delivering only 75w, and the 20w of
reflected power added to the 75w = 95w. These are two separate
measurements--the first a stand-alone value and the second the sum of
two values. Is this a point you overlooked?

The source is Ch 19a where you report Pf=100, Pr=0 in Step 1; and Pf=95
and Pr=20 in Step 8. I am only interested in the relative values of Pf in
Step 1 (VSWR(50)=1) and Step 8 (VSWR(50)=3).

I conceded in another post that within reasonable expectations of error,
and considering the load case (VSWR(50)=3), that Zs under those
conditions and subject to only one load test is probably very close to 50
+j0, but probably not exactly.

In one test that I conducted with a range of loads, the drop in Pf was up
to 21% with VSWR(50)=1.5, but it varied with different VSWR(50)=1.5 load
angles. Whilst some values supported the proposition that Zs=50+j0,
others indicated that it was significantly different. Zs did not appear
to be constant, it appeared to be load dependent to some extent.

Going back to your case, if the source was 50+j0, we should expect Pf=
100, and if we trust your readings from the Bird 43 as characterising the
load to have rho^2=21%, we would expect the power delivered to the load
to be 78.9W rather than 75W, and error of 0.2dB. Not a large error, but
hard to account for entirely as instrument error.

In the measurements of an IC7000 that I made, the measured output power
on one VSWR(50)=1.5 load was 82.5W when it would have been 104.6W had the
source been 50+j0, an error of 0.8dB. I opined that this test did not
support the proposition that Zs was not 50+j0

Owen