urgent help in HFSS!!

[tw20053010]

solve inside hfss

Dear Sir,

I am doing simulation for a microstrip line filter by HFSS. I find the simulated results are totally different if I elect the solve inside when I define the metal. Could you tell when should I elect solve inside and what is it used for? Thank you very much!

Regards,

wei

 

[ghb]

hfss solve inside

solve inside - implies exactly what it means. for instance if your conductor is of the planar style, you would not be able to solve inside. also attached to that same idea is if your conductor is of finite thickness but you defined a perfect E boundary (whether boundary defined or PEC defined) there would be no need to solve 'inside'.

however if your conductor is of finite thickness and not PEC or perfect E (copper for instance), you may wish to solve inside to consider skin effects and so forth.

normally this is not needed, however in some cases that may be. i'm not sure of your particular design, ie frequency so forth where skin effect, losses etc may have to be accounted for.

i would pay attention to your meshing if you have differences between solving inside versus not. you may have a meshing issue if experiencing radically different solutions (IMO) and you do not expect them.

 

[tw20053010]

hfss solve inside

Thank you for your reply!! I only add manually skin depth mesh and elect solve inside on the metal conductor. Becuase for my structure, the thinckness of conductor is much thinner than the skin depth required in the low frequency. This leads a different result from metal conductor without solve inside and skin depth mesh.
Thus, I donot know which results should I believe, with skindepth mesh and solve inside one or without one?

 

[ghb]

i suggest redefining your metal as pec or e boundary. then compare results of all three..

i would lean towards the non skin depth approach, not because i know your application, but because i've used hfss to simulate many different types of antennas or feeding networks and have had good results when simply using pec.

if the results are comparable with pec material and non inside solving, i would still venture to guess your meshing interior to the material is incorrect.

gordon

 

[krishna410]

goood question .. but it doesnt req if its a simple structure

 

[tw20053010]

Thanks a lot for your helps!!!!!