[CST]A Probe Fed Patch Antenna

attched is a very simple patch antenna model in CST.
Actually the model maybe incorrectly constructed because of at least three abnormal results.
1. the resonant frequency is not around 1GHz since the width is 30mm and the permitivity of the substrate is 25.
2. the rad.efficiency is only 5% and tot.efficiency is 0.002~~(i exported the simulation results in archive, so only 1D results are included)
3. the characteristic impedance of the coaxial line should have been 50ohm, but 42~~ohm in the 2-D result.
I really don't know what's wrong with my model.
BTW, is it normal that the meshcells are as high as about 234600?
Could someone kindly help take a look at the model and offer some comments?
Thank you!

Hy
your feed is in the center of the patch ? so your imp has to be 0 ohm.
try to move it along the e-plane to get the right impedance.
you can cut the plane into half to reduce the simulation time .
Try to run adaptive mesh to get the right mesh density for farther simulations.
best regard

hi bbvv,
thanks for ur reply!
actually the paper doesn't specify where the feedpoint is so i simply choose the center as a start. i suppose the location of the feedpoint does not affect the resonant frequency much, though it indeed has a dramatic influence on the impedance matching.
So my anticipation is i could get a S11 curve, showing 0.87GHz is the resonant frequency, even if the returnloss is only -5~-6dB, for example, not as good as below -10dB.
In addition, i doubt if "cut the plane into half " you mentioned means the symmetry plane when specifying the "boundary conditions". If u mean this, i still don't know which option should be chosen, Et=0 or Ht=0?
i ever tried to use adaptive mesh in other model, but it would only add the mesh desity. But i'm wondering if the number of meshcells is too large as an initial value.
thank u!

If u mean this, i still don't know which option should be chosen, Et=0 or Ht=0?
quote]
Ht=0.