cst surface plasmon

Hi all,
I have a question about surface plasmon simulation with gratings.
The basic idea is using gratings to couple electromagnetic wave (light) to a metallic grating surface. When the incident angle of the light satisfy the coupling angle, a reflection minimum will occur cause part of the light energy will get coupled into surface plasmon (in our case the light wavelength is 10um(30THz), the incident angle is 30 degree (angle to the structure normal), the period of grating is 20um, and the height of the grating is around 0.2~2um) I use a unit cell structure, in x y, and free space(add) in z. I could sweep the angle and see how the reflection changes.
But the problem is although I do see a dip at 30 degree but when I change the height of the air (free space above the metallic gratings) the reflectivity changes dramatically.(It should not happen because no grating settings have been changed) I tried to modify the mesh settings but it seems not working. I doubt whether it is the problem of my port setup. When I change the angles and look the ZminTM(0,0) ZminTM(0,0) 1 D results, I don't know whether the port will look correspondingly at the reflection angle? Also I don't know whether is the problem of the position of the port. I would be very appreciate If someone could take a look and give me some advice on this!
Cody

Can anybody help me?
Thanks!

Is there a bloch boundary condition that you can use? Also, did you ever solve this problem? Please post your finished CST file here. Thanks.
PS:
Is there a bloch boundary condition that you can use? Also, did you ever solve this problem? Please post your finished CST file here. Thanks.