PLL story

I recognize the problem! In the past I have experimented a lot with PLL filters and most of the time I ended with a solution obtained by 'trial and error'.
The biggest problem I got was 'oscillation' of the PLL circuit, caused by improper damping.
The frequency would go up...oops...to high, then the frequency would go down...oops...too low etc. The avarage frequency was correct, but the frequency was never stable and constant.
On Harry's website you will find a PLL filter with an 'accelerator', if the frequency is too far off, the loop filter has a short time constant, but if the frequency error is small, the time constant of the filter is much higher. A very nice idea IMHO.

thanks for your article about PLL
Have you a "system" to calculate the values of the loop filter
for a CD4046 ?
thanks