@Don Bitters:
I'm not sure what you mean by the Uwave port. There are two RF input ports on the HP11729C, the 5-1280MHz port (used, in the phase detector configuration for the Reference Oscillator) and the Microwave Test Signal input port (used to connect the DUT - by Uwave are you using a "U" to represent Micro?). The Reference Oscillator is a Wenzel HF-ONYX-IV, which has an EFC pin allowing the fundamental frequency to vary between 10 MHz +/- 10 Hz. For the simple input power test I described, I used the Reference Oscillator with its signal running through a directional coupler and adjustable attenuation pad (adjustable in units of 1 dB). I haven't checked the Reference Oscillator signal for harmonics or purity, but it is a new unit and has pretty good harmonic specs < -30 dB (see . Phase noise is ultra-low for this device (e.g. 10Hz <= -136 dBc/Hz). My Rigol (1104Z) is an entry level unit without a phase noise measurement app.
I think John Miles has figured out the discrepancy between the recommended input level (7-18 dBm) and the statement that the mixer compression point is 3 dB. The first seems to reference the input level when using the downconverter hardware, which as you point out is not used for a 10 MHz signal. The second is the compression point on the phase detection mixer (not the Microwave downconverter mixer - ?see figure 4.4 on page 21 of?).
@Leo Bodnar:
I don't, as yet, have any phase noise plots, since I am trying to figure out the proper test setup procedure (including input power levels).
