high frequency emag, Ansys vs. Ansoft 1

[coolsti]

After struggling for over 2 years trying to perform 3D modelling of high frequency electromagnetics with Ansys in order to get S-parameters, etc., I am at the point of giving up. Producing an appropriate 3D model and meshing it to get full E and H fields is not an easy task, but to then extract the needed information as a function of frequency and having confidence in the results seems impossible.

Now we are looking at Ansoft's HFSS, but the price tag is the show stopper here, after having spent so many bucks on the Ansys license (which we continue needing for mechanical and thermal sims).

Has anyone similar experience here? Am I truely dreaming when I think that I can use Ansys for high freq emag within only one lifetime? Is the high price tag for Ansoft worth it/necessary for us?

Best regards
Denmark

 

[S21]

Hi, Denmark:

Ansoft's HFSS is very expensive (about $50,000). We also have ANSYS lincese and have solved some HF problem using ANSYS. But when the shape of model become complicated, ANSYS can not go through because of meshing problem. Trust me, if we can totally understnad ANSYS, it can solve problem.

Best Regards.

S21

 

[coolsti]

Hi S21

thanks for the response. Yes, we know about the HFSS price tag, and are trying to decide what to do about it. Additionally to really make full use of it, we would need a few extra modules, so the price tag will be well over 50K $.

It would be interesting to be able to share some experiences and tips regarding HF simulations with Ansys. I have been using Ansys to calculated S parameters for models involving coplanar waveguide lines going around bends and through various "disturbances". I still am learning what is correct and not correct regarding mesh coarsesness, boundary conditions, etc. etc. But lately I find I am getting better results, although time consuming.

Regarding your comment about complicated structures, we have not yet tried to import 3D models but have built them ourselves within Ansys using APDL scripting. I am trying more and more to make use of auto-meshing but find Ansys' capabilities very very poor in this regards. Therefore I find I am usually meshing my models withing my scripts by setting a lot of size controls using lesize or esize. Usually I find which plane in the model is the best candidate for producing a 2D floor plan and then sweep the mesh into 3D with extrusion.

Regards,

/Steve
Denmark

 

[caviac]

I used to make HF analysis with these 2 softwares, but just simple waveguide application with ANSYS and really complex studies with HFSS.
Yes it has a cost, but it gives payback so make some payback analysis on the buying of the software...sometimes it can be cleared in less than 1 year...

caviac

 

[mueschen]

Hello caviac,

from your experience,
would you be able to state if ANSYS,
or HFSS if not the former, provides
the following Modeling capabilities :

- Loading options:
antenna conductor fed with a voltage or Power exitation
As to avoid nessesity of definition of a impressed current load distribution
(needs be known, which is not always the case for other then standard geometries)
If yes, please indicate if PEC or explicit conductor model.

- analysis features:
the capability to analize receiving performance of antenna
i.e. behaviour under incoming field for hertz/marconi, horn/dish type antennas

- analysis capability:
antenna AND waveguide/feed line(e.g. coax ) in the same model:
to analyze effect of feedline/wg_structure on antenne behaviour,
i.e. BEYOND isolated/segregated modal and s-parameter studies of waveguide structures
or isolated/segregated antenna analysiswrt near/farfield/radiation_pattern etc.

evtl. much thanks,
for sound feedback