You need two airbox. Airbox 1 on the top of the EBG should be x6 ~ x8 of the EBG's thickness. After you placed the PML on the airbox 1, airbox 2 is need to include everything from top to bottom.

http://www.emtalk.com/tut_3.htm

thank u sir for reply but I want to calculate permittivity and permeability of metamaterial unit cell so in that case I want to know the size of Air box.

I think that is due to near-field. To calculate the permittivity and permeability of metamaterial unit cell does not need to know the size of Air box.I propose to simulate cst, which takes values ​​in software.

I think that Wei Wong is right, you need two airbox. As far as the PML can use the wizard in this version of hfss 14, while 20 steps to converge are too many. I do not know what precision you wish, but I would recommend you to fix target, of S, to 0.01 about 40 db.

Dear Sachin Kalraiya?

You use some retrieval technique to obtain the your parameters (permittivity and permeability) or use the software resource? You use the CST-MWS too, or only the HFSS? Regards.

I agree with Reza Scheilifar, please have a look at the paper from Z. Szabo "A Unique Extraction of Metamaterial Parameters Based on Kramers–Kronig Relationship" available from IEEE Xplore. They use CST to compute S parameters and provide a Matlab tool, available online, to extract epsilon and mu. The extraction results are independant of the thickness of the simulated sample.

I have done S parameters computation of metamaterial cells with periodic boundary conditions on left/right and up/down faces of the cell, and open boundaries on front/back faces. I used CST Microwave Studio, not HFSS. In this software, you can add airboxes between the front and back faces and the 2 ports where S11 and S21 are computed. But additionaly, in this software, it is possible to de-embed the S parameters to do as if they where calculated on the cell faces (De-embedding is the act of taking data that is measured in a test fixture and removing the effects of the fixture so that the data is accurate to reference planes that are more useful). Basically, the soft does the computation of the airbox transmission, removes it from the S parameters and provides S parameters at the cell faces, which is what you want if you want to extract effective constitutive parameters.

Maybe a similar option exists in HFSS ?

Otherwise, try to goole: "de-embedding s-parameters matlab"

I think Mr.  Kalraiya is asking what is the height of the unit cell box, so the de-embedding distance on the port side has nothing to do with this.On the top and bottom sides of the unit cell you apply PMC conditions to simulate the perpendicular H field to the SRR plane necessary to excite the structure. So if you visualize that this height determines the distance that the SRR is repeated in the  axis parallel to the SRR plane you get an idea what you need. There is a different explanation that the SRR is in a waveguide and the height  of the unit cell is a dimension of the waveguide and this has to be taken into account in order the port to excite the correct mode. In both cases, the S-parameters are different for different height dimensions but you get an indication where the resonance of the structure exists.Personally, I use for the height a value of a little more than 1/2*substrate side.So if you have a square substrate of 8.8x8.8 mm2 and on top a SRR you take the height on top and bottom to be e.g.4.6 mm, so the total height of the unit cell will be (2*4.6mm)+substrate height. Hope I helped.

plz help me to design MIMO metamaterial Antennas in HFSS

Please help to design spiral resonator as a building unit cell to build metamaterial superlens in HFSS. please please help me.

Usually, the airbox size is recommended to be at least 1/4 of the max wavelength.

If there are strong interactions between remote parts of the structure and only little interactions between nearby parts, then you may need to increase the airbox size to accommodate for those strong interactions. This would mean to accommodate for the field lines going over the structure - the fewer of these lines are disrupted by the boundary condition at the airbox, the more accurate and stable your model will run.

Hi Sachin

HFSS is just a calculator.It gives proper simulations only if you setup the right boundary conditions.

The usual method I use in simulating an antenna in HFSS is to build a air box that is with  length of side equals lambda/2.The lambda equals 300mm/f0.f0 is the center frequncy,f0=2.4 for 2.4GHz.

Remember to add a '-lambda/4' to the structure position to make sure the antenna is right at the center of the air box with radiation boundary.Hope it can be of help.

Best regards

Hi,

For a metamateial unit cell, normally it will be a square substrate. so you can take the dimension of the square as the dimension for the radiation box, and then you can assign PMC on the face perpendicular to the substrate. This will work good.