LMX2594_9750MHz.zip I am using the LMX2594 in a design and was hoping you could help me out with a small problem.
I have been doing some phase noise measurements and comparing them with the predictions in PLLatinum Sim but there is some discrepancy between the two results. The real world measurement is up to 15dB worse at some offsets and I was wondering if you knew what might be causing this?
My PLLatinum simulation is in the attached folder:
- Data for the 10MHz reference is loaded into the sim (Wenzel 501-22578-05, datasheet attached).
- 10MHz level is measured at ~1.12V pk-pk on OSCINP pin.
- VCO noise and PLL noise set to use metrics as default.
- I have manually set the loop filter to be the same as on the actual board (schematic is attached).
- The red ‘comparison’ plot is my phase noise measurement (.dat file is attached).
- This was captured using R&S FSW, setup information is in the .dat file as well if needed.
As far as I can tell all of the inputs match the real world setup but when comparing the results of the simulation with the measurement there is a significant difference, especially between 100Hz and 1MHz offset.
I understand that this is a very sensitive measurement which can be affected by many small factors but I expected the result to be a little closer to the predicted than this. I was hoping you would have some theories on what could be causing the degradation I’m seeing here?
I have also attached the register set I am using in case that is of use to you. Any help will be much appreciated.
