Two poles at origin causing negative feedback to become positive. Now what?

[amsdesign]

A closed loop negative feedback system has a two poles at the origin on the feedforward path and this is causing its oscillations - cause its now a positive feedback system.


One pole : Inherent VCO pole at origin
Second pole : Very large capacitor as part of a charge pump (this cap's contribution to the pole is almost at the origin)

So pole1+pole 2=180 + negative feedback (180) = 360 = positive feedback

To correct this I added a zero to bring down the total phase shift back to 270.
But I will now need to add another pole to correct for voltage disturbances caused by the zero. And adding this new pole will cancel out the effect of the zero and bring me back to the same problem as before (negative feedback becoming positive) . Is there a work around?

 

[BradtheRad]

This might be a circumstance where a buffer is required. Its purpose is to prevent undesired interaction of one section of a circuit with another section.

 

[FvM]

Any PLL loop filter need to overcome this problem. Look at existing PLL implementations. They hardly have a pure C loop filter.