Solution for two frequencies in HFSS

[soaringeagle]

Hi,

In HFSS , we normally specify one solution frequency and we run the sweep for some range of frequencies that includes the solution frequency ( normally in the centre). Like if we run the sweep from 500 MHz to 1000MHz, the solution frequency is chosen to be 750MHz.

But if there are two separate ports and they are exciting two different antennas on the same substrate, what should we put in the solution frequency? Like one antenna is designed to resonate at 2.2GHz and other antenna is designed to resonate at 2.4GHz, what should we put in the solution frequency field of HFSS ?

Urgent help is needed please.

 

[PlanarMetamaterials]

Hi soaringeagle,

Your convergence (i.e., solution) frequency should always be equal to, if not greater than, the highest (frequency) point in your sweeps, otherwise convergence is not guaranteed at the higher frequencies.

 

[volker@muehlhaus]

Your convergence (i.e., solution) frequency should always be equal to, if not greater than, the highest (frequency) point in your sweeps

Are you sure that this is a good idea here?

If you set the solution frequency to the maximum sweep frequency, instead of the antenna resonance frequency, you optimize the mesh for a frequency where the antenna reflects most energy. This might not give the desired mesh refinement in those volume areas that are relevant for antenna operation (at resonance).

I think it would be wise to refine the mesh (also) at the resonance frequency (or resonance frequencies), to ensure we have a good mesh there.

 

[PlanarMetamaterials]

It is true that it is a bit of a tradeoff, but if you need to know about frequencies above resonance, then meshing at the higher frequency is really the only valid option. If you care mostly about resonance, then yes, you should mesh there.

In this case, I would think that meshing at 2.4GHz + would seem most appropriate.

 

[volker@muehlhaus]

then meshing at the higher frequency is really the only valid option.

I haven't used HFSS, but in other FEM solvers you can specify multiple frequencies for adaptive mesh refinement.

 

[PlanarMetamaterials]

Its true that you could have multiple solution setups if you wanted to have multiple convergence (meshed) frequency points, however the frequency sweeps therein would probably not correspond. To check this, I would do one solution at 2.2 GHz, one at 2.4 GHz, and one at 2.6 GHz with a frequency sweep down to 2 GHz. The correlation between the sweeps and the convergence points should give you a good idea of which setup is most valid.

 

[volker@muehlhaus]

Its true that you could have multiple solution setups if you wanted to have multiple convergence (meshed) frequency points, however the frequency sweeps therein would probably not correspond.

I don't get what you mean? For my Agilent FEM solver, I specify the different solution frequencies for mesh refinement, and then it generates a mesh that is good (for deltaS) at all these solution frequencies. After having found that mesh, it runs the frequency sweep.

 

[streamlet]

This is an interesting feature. Do u mean EMPro?

I don't get what you mean? For my Agilent FEM solver, I specify the different solution frequencies for mesh refinement, and then it generates a mesh that is good (for deltaS) at all these solution frequencies. After having found that mesh, it runs the frequency sweep.