rf probe coupling to cilindrical cavity resonator

[yefj]

hello, there is a cavity filled with dielectrics shown bellow,it was planned to resonate at 12.5GHz.
i want to couple it using an RF-proble(coax line) represented by a descrete port connected to a cylinder shown bellow.
I have very few parameters , probe length which controls the resonant frequency and inductance
probe width which controls the capacitance .
The only good resuld i got is when my rfproble radius is 0.01mm as shown bellow.
And still i got only S11=-5db at resonance.

is there a way i could have a probe diametre of 1mm and get a good couling to the coax probe?
when i put a 0.2mm diameter i get a smith chart as shown in the end backly open circuit .
is there a way for compensation of thicker coax probe to get the result as i got in 0.01 radius?
Thanks.

 

[volker@muehlhaus]

Your coupling line orientation/direction looks wrong. In the designs that I created or simulated, the coupling line is 90° from what you show here.

In the example show below the coupling line is curved, but it could also be straight.

 

[yefj]

Hello ,Thank you very much for the responce.
I am working with TE11 mode i need to put the probe in the direction of the maximum of the E-field as shown bellow.
first i put a descrete port a descrete port and as you can see bellow i get the mode perfectly.
but when i tried to put a dipole in the same direction as the descrete port i got almost zero matching.

As i see it i got a descent match for 0.01mm radius which is not practical when i increase the radius of the probe its just ruins the match totaly ,but the interesting thing is that the "resonance point" stays at the same frequenc no matter how wide my dipole.
am i wrong?
we are working on different modes ,and they are excited differently.
please help me understand how can imake my coax /dipole /probe be with normal radius and not ruining the matching?
Thanks.