ie3d sonnet
I know that IE3D uses MOM, which is also used by Agilent Momemtum and Sonnet. But someone told me that IE3D is a 3D EM simulator instead of 2.5D like Momemtum and Sonnet. I am little confused. Is IE3D doing the same thing as HFSS does? What makes IE3D different than Momemtum and Sonnet?
And what about cost? What is cheaper?
Rainleo:
IE3D is not a full 3D EM simulator. Full 3D EM simulators mesh volume, and IE3D only meshes surfaces. They claim to be able to do 3D bodies, but they are limited to rather simple bodies. Sonnet also does simple embedded 3D bodies as well. Momentum cannot model embedded bodies (dielectrics), so it is the most restricted in this respect.
If you really need to accurately model full 3D structures (i.e., non-planar) then I recommend that you look to full 3D EM solutions.
In terms of technical approach, IE3D and Momentum are most similar--they use more similar mathematics. They model (by default) planar structures in an unshielded environment. Sonnet models planar structures in a shielded environment.
--Max
Full 3D EM simulators based on Method Of Moments (EMC Studio, FEKO, WIPL-D) do not mesh volume and mesh surface only too.
Good point. But you would realistically use these 3D MoM tools for scattering problems and full 3D structures (like radar cross-section analysis of a full airframe) that are electrically too large for volume meshing approaches, I think. I may be overly general with that statement, but that's usually the way I think of it.
thx, maxwellian and EMC for the answer.
I am only an EM simulator user, not an EM specialist. To my understanding, IE3D can mesh the vertical plane rather than of horizontal ones. I don't know Momemtum can do it or not. But I know for sure that sonnet can not do it.
In our case, a lot of structure is simple. But most of the metal trace, for example, an inductor, have higher thickness than its width. I am not sure if Momemtum and Sonnet can still give the accuract results.
Rainleo:
I'm not sure what you are referring to when you say "mesh the vertical planes," but I do know that Momentum, Sonnet and IE3D all have ways of modeling true metal thickness so that coupling between metal sidewalls of adjacent tracks is considered. Any of these vendors have references to successful modeling of thick metal spiral inductors, baluns and transformers.
Momentum is limited to planar structures only. Sonnet can model embedded dielectric or metal bodies that have a constant cross-section through any dielectric layer, like holes, gaps or truncated capacitor dielectrics. (I have seen a Sonnet model for a microstrip-coupled dielectric puck resonator that showed a fundamental resonance within 0.1% of the measured resonance frequency.) IE3D can mesh simple bodies with non-uniform cross section, so I think IE3D may be the most flexible from the point of view of what you can actually try to model.
But if your bodies are even moderately complex, you will be better off with a full 3D EM simulator, because these planar simulators become very inefficient when you need accurate modeling of 3D bodies.