High Q-factor resonator in CST

[supernano]

I am trying to simulate CPW resonators in CST. When I place a floating stub of line next to a feedline, and simulate the s-parameters of the feedline, it shows the resonance as expected, but as I move the stub away from the feedline to try to get the intrinsic Q-factor of the resonator, the simulation results start to become very inconsistent (resonance frequencies jump around or don't show up at all), for Q-factors larger than ~1000.
I am using the Frequency Solver and have so far only tried minor tweaks, such as narrowing down the frequency range or adding several equidistant frequencies around the expected resonance frequency, but this hasn't helped much.
Does anyone have experience running similar simulations and could give me some tips for getting CST to simulate very high q structures?

 

[JohnBG]

Hi supernano this is JohnBG2021, could you please post a view of the part of the circuit where you intend to place the resonator?

please include a visual or side notation scale reference so readers know how wide/long is the track and how close or far from track have you tried so far.

--- Updated Feb 28, 2021 ---

Also important, would you please be so kind to mention the following:
1.- at least nominal frequency for your circuit: On what freqs do you want the circuit on
2.- what is the working BW: [f1 f2], or is it working in multiple bandwidths?
3.- what is the target Q? what value achieved so far?