Strange thing with CST

Hi,
I'm simulating some microstrip antennas in CST. I'm trying to put ground plane to all my designs from the same metal of patch.
I know the boundry condition Et=0 do the same. I prefer to put as well.
Anyway, I tried to delete the ground plane and re-draw it today.
Only delete and re-draw with same dimensions and same material.
I discovered that mesh cells gone to half and there is new results. The important thing that there is shift in s11 notch about 0.3 GHz also.
Any idea about why ??

there might have been a problem in your port definition.

I'm a bit confused at what you're trying to describe, a few pictures might help.
Have you changed the e-wall boundary condition to open or open add space once you have drawn the ground plane? this might be the problem, but I'm not sure I'm really understanding what you've done.

In fact, it is very basic abd simple thing what I'm doing.
Anyway, let me explain it again with simple words. Please forget what I have written before.
I have microstrip patch antenna with inset. I think it is so famous. I did many simulations changing different parameters for it.
Last days, I tried to just change the boundry conditions. I have discovered that CST re-define the mesh when I'm just click on boundry conditions button and make even no change for the settings.
It is so strange for me I can say. The structure is same, the frequency band is the same, and the boundry conditions are the same.
Is there any reason or explanation for that ?
I can send you the CST project if you like.
Regards