Agreed. This was a good read and a great outcome. Well done to all involved!
?
If it helps, I can suggest a few performance tests that may be relevant now.
?
To test that the loop gain switch (at Q4A and Q4B) is performing OK, try posting up screen shots of the 8566B measuring its 100MHz cal output on a 20kHz span, a 50kHz span and a 100kHz span.
?
I think that this should show if the loop gain is changing correctly as an increase in loop gain should deliver an increase in loop bandwidth on the narrow span. It should be possible to spot this by looking at the phase noise response on the three different spans.
?
Also, increase the span to 6MHz and have a look to see how stable the response is at the centre of the display. There is usually a small amount of wobble in the swept trace and this is fairly normal as the analyser switches to lock and roll mode for spans over 5MHz. If the wobble is excessive, then maybe some other aspect of the loop has a minor problem.
?
