hfss frequency shift mesh option
Did I get you right? You have 1 GByte of Ram and you can not solve more the 27K tets? I don't think this is correct. I'm normaly a CST "Fan" but im pretty sure the HFSS can handle more then 27000 tets within 1GByte of Ram.
F.
nadnerb
0 mode (All E fields in the same direction) and a pi mode (opposing directions)
??? - "0-mode" means 0 degree phase shift, "pi-mode" - 180 degree phase shift per cell.
My task is to tune the full cavity
What does it mean "to tune" ? Do you want to match the input wg with cavity ?
The left port (the larger of the two remaining) is not a port
Yep! Please write clearly how many ports you defined and how many modes are in each port.
Please, DO NOT MIX UP port and boundary, because its completely different things.
The piece was modeled successfully in SUPERFISH
SUPERFISH is eigenmode solver, but you talking about driven solution.
What do you want to compare ?
Anyway, please send a full geometry to have a qualified answer.
RFSimulator
im pretty sure the HFSS can handle more then 27000 tets within 1GByte of Ram
Its true for most cases but sometimes, if you defined port with many modes and
have a high lossy material, then 1Gb might be not enough even for 30K tets only ...
Hi
thanks for your comments. I did know that HFSS is this memory consuming for multimode high lossy structures. Interesting?..
Please post the gemometry (*sat file) here.
F.
nadnerb
Ups, I can see the geometry now ...
One remark - try to exclude radiation boundary if don't worry about radiated pattern.
Just change it to E-plane (or to impedance plane in case to be more precise)
This will help HFSS a lot :)
Note that a radiation boundary can be modeled as lossy if the
environment variable ZERO_ORDER_ABC_ON_PORT is set = 1
well you are good on the number of tets, i'll recommend having a look at the boundary definitions and specially your solutions setup.. look for examples from ansoft website..