Dear Ali, Generally length if is considered to be lambda, waveguide gives virtually infinite length.

Try.

Best wishes 

Hi Ali,

To simulate an infinite cylinder in HFSS you should draw it such that it touches both upper and lower boundaries. Depending on the polarization of interest you should set the boundary condition on both upper and lower faces to be either PEC (E field parallel to axis) or PMC (H field parallel to axis). On another note, HFSS only handles 3D structures but not 2D. To avoid using too much computer ressources, you should make your cylinder as thin as possible (one mesh element along height).

As for your unit cell simulation, I'm not sure I understand your problem. Could you provide a detailed explanation of what you're trying to do?

Good luck,

Thanks all for your contribution

Mr Thomas:

I have a cylinder coated with a dielectric and array of patches. I want to plot RCS of this structure when illuminated with plane wave. Theoretically, the structure is infinite in length along z axis but in practice it is limited. If I want to simulate the finite structure, I should choose the height of cylinder to be at least 5*lambda. This results in large number of meshes since the radius is one lambda. Thus, if I can simulate infinite structure, it would result in less number of meshes and lower simulation time. 

Master slave boundary conditions in HFSS are for simulating a unit cell but infinite cylinder is periodic just along z axis. I wanted to use these boundary conditions for infinite cylinder.

I think using PEC on the top and bottom  might be useful. I will test it

Thanks

Ali,

Let me summarize your problem to make sure I understand what you're trying to do. You want to compute the RCS of a single metallic cylinder coated with a dielectric and some metallic patches. These patches are arranged periodically along the z-axis. If that's right, then you have to do different simulations for the infinite case and the finite case.

For the infinite case, you should simulate only one unit cell and put master/slave boundaries along the z-axis and PML along all others axes. The excitation should be a plane wave. If your unit cell has a plane of symmetry normal to the z-axis and your plane wave comes at normal incidence, you can replace master/slave boundaries by PEC or PMC depending on the polarization. I've attached a pdf that explains this (see figure on page 8). It is from CST not HFSS but the physics is the same.

For the finite case, you should simulate the complete cylinder and all boundary conditions should be PMLs. The excitation should still be a plane wave. To avoid long simulation time, you might want to use HFSS's hybrid FEM-IE if you have access to it. I've attached a pdf that explains this (starts on page 9). Essentially, it will use the finite-element method (FEM) to mesh the dielectric and the method of moments (IE) to propagate the fields in free space all the way to infinity.

Good luck,

Dear Thomas Lepetit

I am also trying to do the same thing with Mr.Ali.If it is possible,Could you please share the pdf file that you mentioned in your comment with me.Because I am not be able to see it.

Best Regards

Ali,

Were you able to do the simulation? I came across this conversation when I was trying to find out how to create an infinite structure (a strip in my case) in HFSS.