Following the ADIsim simulation you uploaded in this thread: ADF4007 (eval board) as phase detector and PLL
I had to change R2 to 81 ohm and C1,C4 to 5 nF, and C2 to 480 nF, but they are really close. M2, M1 at either 1,0 or 1,1 (for positive or negative polarity) - same result.
I'm using an op-amp: LM741CN - it has a lower slew rate and bandwidth than the OPs from Analog, but it's what is available. The result is a constant voltage output at about -14 VDC, regardless of what changes in the input signals. The circuit is on a breadboard - starting with the CP output (wire soldered to a pad on the ADF4007 and connected to a breadboard with an op-amp) following the schematic precisely, except the above changes.
Could the op-amp be the problem? It's the same one used before that resulted in pulsed output.