I have a design with 2 dipoles working at different frequencies joined in one system and separated by Ferrite. I think, Ferrite was used to minimize the interference between 2 antennas as much as possible. I need to simulate this design in HFSS. Can I simulate ferrite in hfss or do I need to take permeability, permittivity numbers from the manufacturer (which are not available with the manufacturer).
All ferrites should be similar,,,right? does it matter different companies sell different ferrites with different permeability and permittivity?
Please advise so that I can see if I am thinking correct or not.
Thanks
Yes. Only in DrivenModal solver. Without GPU acceleration (I do not know is it possible to make domain decomposition with accelerated and not accelerated domains)
The solver applies saturated conditions with unknown to user surface pinning condition, and capable of solving even some magnetostatic(non-exchange spin-wave) modes. At least I was getting a surface spin-wave mode in an electromagnetic band-gap.
The nonuniform internal magnetic field can be supplied by Maxwell (data link setup is in biasing field configuration), but it is not perfect, because only Hi is supplied, other magnetic parameters are taken as uniform.
See: Vinod V K Thalakkatukalathil. Electromagnetic modeling and characterization of anisotropic ferrite materials for microwave Isolators/Circulators. Electronics. Université de Bretagne occidentale - Brest, 2017. English. NNT : 2017BRES0134.
> "Do I need to take permeability, permittivity numbers from the manufacturer"
Usually, absence of these parameters is a stopping factor for R&D. If the sample is big, you can order a torus made from it for measuring magnetic hysteresis and saturation. Saturation only can also be measured from drilled long slabs, ask for the technique in the lab. Epsilon and delta are measured using capacitive techniques. Guessing Ms is also possible by embedding the material in EM waveguide, or a spin-wave guide, and observing a pass-band. The latter relieves the measurement from influence of permittivity. DeltaH, the essential parameter for an absorber, can be guessed from wave-guide loss. But the correct way is a resonant measurement requiring a torus, or better a spin-wave resonance method requiring a very thin ferrite plate.
>" All ferrites should be similar,,,right? does it matter different companies sell different ferrites with different permeability and permittivity? "
No. Not even close. Ferrites differ significantly depending on material, and application. For RF waveguides garnets and spinels are used, where spinels are much more lossy. Hexaferrites are used for (sub)millimeter waves, where garnets do not offer strong enough magnetization, but they are lossy. There are many other ferrite types used in transformer cores not related to RF application, because of high loss. However those, and cheap spinels can be used as absorbers.
Between them, only RF garnets have some degree of similarity, but engineering samples still differ in loss and saturated magnetization, which are traded for thermostability.
A starting point for modelling would be to calculate the permeability of the ferrite using the Polder tensor (ref: R.F. Soohoo, Microwave Magnetics, Harper & Row, 1985, or other references). I take is an expensive package like HSFF will be abe to handle permeability, in a similar way to handling permittivity. You'll also need to know the permittivity of ferrites, which is also know from publications. To calculate the Polder tensor you'll need to know things like the applied magnetisation fields H, the magnetisation M of the ferrite, and the resonance frequency. So i don't see any reason why HFSS could not be used for this purpose. I'm sure this has been done by many companies in the design of microwave ferrite based isolators and circulators.
Good luck, Neil
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Electromagnetics
- Electromagnetic Fields