I am trying to understand how to use an HP11729C in its phase detector configuration to measure phase noise of several 10 MHz oscillators. One issue that has contradictory guidance in the HP11729 literature is the power level to use for the DUT input to the Microwave Test Signal input.
?
On page 21 of HP product note 11729B-1 (https://www.keysight.com/upload/cmc_upload/All/5952-8286E.pdf), it states that the CW microwave input level must be between 7 and 18 dBm. However, on page 15 of HP product note 11729C-2 (http://hpmemoryproject.org/an/pdf/pn11729C-2.pdf), it states the mixer inputs reach compression at 3 dBm. The operators manual (http://hparchive.com/Manuals/HP-11729C_ops_svc_11729_90017.pdf) does not mention any constraints on the power level of the DUT. This confusing guidance leads me to believe the 11729B and 11729C have different front-end hardware.
?
To get some idea what is the correct power level, I ran a test varying the input power to the Microwave Test Signal input of the HP11729C. I connected a Reference Oscillator through a directional coupler and attenuation pad to a swept-input SA (SSA3032X) and set the input power to 10.85 dBm. I then reattached it (Reference Oscillator->directional coupler->attenuation pad) to the Microwave Test Signal input of the HP11729C. I connected the coupler port of the directional coupler to a Rigol DS1104Z. Figure 1 shows the resultant signal shape displayed on the oscilliscope.
?
It is obvious that some of the power of the signal at 10.85 dBm (specifically, some of the higher harmonics of the 10MHz input) is reflecting back through the directional coupler and corrupting the signal going into the coupler port.
?
I then set the input power to 2.86 dBm and connected it to the Microwave Test Signal input. The result is shown in Figure 2
?
The oscillator trace of the input signal is nice and clean, indicating that very little (if any) reflected power is corrupting the coupler port output.
?
This suggests to me that the DUT input to the HP11729C should be kept below 3 dBm. However, I am interested in hearing from others who have used the HP11729C (John Miles?) if this is the correct strategy.
