compare: HFSS or Microwave CST

[chandregowda]

hfss cst

hi,

how do i make comparision between HfSS and Microwave CST (Performance-wise)


thanks in advance

 

[james cook]

cst vs hfss

For small variation of structutre like simulating temperature drift, I prefer CST over HFSS.

 

[chandregowda]

cst hfss

hi,

how are the results when compared to measured one?
(we are working on RF components and Antenna)
i am getting similar results in HFSS and measurements.

which one will take more simulation time?

thanks in advance

 

[rfmw]

hfss vs cst

I done extensive comparison between the HFSS (FEM) and MWS (FDTD). My antenna model was relatively (electrically) large. HFSS used approx. 30GB of memory when solving (direct solver) and the MWS used approx 200M cells (50 hours of solving time) for the same problem.

My feeling is that the MWS required for the calculation less time, since discrete frequency sweep in the HFSS takes a long time. Which one is faster is difficult to say, you have to consider convergence you're satisfied with, computer you're going to use for the 3D solvers, are you interested in broadband or narrowband, etc. Other users might add some more...

 

[demian1977]

cst hfss

If you need only few frequency point, and you have enough RAM, HFSS is the best. It's faster and more accurate. Else MWS can be usefull, but only if you use FIT and not tetraedral grid in frequency domain.

 

[madengr]

which is better in discontinuously hfss or cst

A couple of years ago I posted stripline benchmark results to compare 2.5D and 3D tools. I never did post the updated version with CST so here it is:

**broken link removed**

If you skip to the last page in the spreadsheet you will see a comparison between HFSS and CST (both FD and TD solvers). The error for HFSS looks good at 100 seconds, however as mesh density (and analysis time) increase the error diverges. The error curves for the CST solvers are smooth and monotonically decreasing.

This is only one type of problem so you can't conclude one is better than the other. I prefer CST (we recently purchased it) since I can choose between TD and FD solvers.

When evaluating CST for purchase I gave them a patch antenna design that I simulated in HFSS and measured. All three agreed well but the CST TD solver was faster with less memory; I don't have specific numbers.

Now that I have access to both I want to try a log-periodic dipole array I designed in HFSS and measured. Since it's broadband it will make a good comparison between FD and TD solvers.

 

[alen_ansoft]

I been using ansoft Ansys HFSS for long time and IT is pain to get it going (it does take for ever ) simple Diff pair extraction takes 10+ hours to simulate and the reults are not the final .. they are asking outrages prices for annual licensing . And it does not make sence to get this tool at all there are planty of other tools in the market MWOffice (AWR) or the best one is CST I do love it and it is 5-7X faster it does use a math engine like matlab . You can use cadence plus matlab or CST AWR Hfss is way over rated and SSSLLLLOWWW

 

[ay_be_why]

CST takes MUCH less time than HFSS for simulations, and the results have always matched so far....

Off topic, but can someone help me how to use current density monitors in CST? I'm simulating a patch antenna in CST and when I run the simulation with these monitors, i get a message that they are valid only for dielectric/magnetic losses and will be ignored....

 

[AmirMohammad]

CST is more efficient in handling memory and more powerful in meshing sophisticated structures. Considerng their solution methods : for broadband problems CST TD solver is faster (experiences of order 40times) than HFSS, but for narrowband problems CST FD solver or HFSS is suggested. Also CST can simulate large scale structures RCS such as sheep using its IE solver.

 

[katko]

Along the lines of the previous post, CST gives you more control over meshing. It's easier to do finer/variable meshing for particular parts of the model than in HFSS. Of course, you have to know what you're doing sometimes - from experience, using a stripline feed not quite aligned with other geometric features can take a while to get meshed correctly, often done manually. HFSS is much easier to mesh with the automatic refinement, but can generate huge meshes.

Narrow vs broadband is huge also. If you have a high-Q structure, don't use CST's time domain solver. If you have a wideband structure, don't use HFSS or CST's frequency domain solver. If you're only interested in a few frequencies, use frequency domain/HFSS.

Both are very accurate and can be just as good in different situations, but sometimes you have to know what you're doing in setting up the simulations. CST can also handle electrically larger structures than HFSS on the same computer, but it'll take a long time to simulate (CST puts a lot of demand on processor, HFSS puts a lot on RAM). A 12G RAM PC was able to simulate one large model in CST (but took days) but unable at all in HFSS.

 

[hisa]

in frequency domain, especially in resonant structure, I give different responses from CST and HFSS! which one is correct?

 

[covcst]

Both are good....for wideband structures CST TD is better, for narrowband structures, HFSS or FD solver of CST is preferred....But both requires understanding of the situation to set up a simulation; otherwise undesired results may appear...........

 

[drkirkby]

Re: hfss vs cst

I done extensive comparison between the HFSS (FEM) and MWS (FDTD). My antenna model was relatively (electrically) large. HFSS used approx. 30GB of memory when solving (direct solver) and the MWS used approx 200M cells (50 hours of solving time) for the same problem.

My feeling is that the MWS required for the calculation less time, since discrete frequency sweep in the HFSS takes a long time. Which one is faster is difficult to say, you have to consider convergence you're satisfied with, computer you're going to use for the 3D solvers, are you interested in broadband or narrowband, etc. Other users might add some more...

As a matter of interst, did you manage to solve it in HFSS using 30 GB RAM? I wonder if HFSS might have actually been faster, despite its use of a lot more RAM.

Dave

 

[tallface65]

I been using ansoft Ansys HFSS for long time and IT is pain to get it going (it does take for ever ) simple Diff pair extraction takes 10+ hours to simulate and the reults are not the final .. they are asking outrages prices for annual licensing . And it does not make sence to get this tool at all there are planty of other tools in the market MWOffice (AWR) or the best one is CST I do love it and it is 5-7X faster it does use a math engine like matlab . You can use cadence plus matlab or CST AWR Hfss is way over rated and SSSLLLLOWWW

This seems to be more of a trolling post (note the 2 posts of this user and the user name). No useful information here.

 

[angela18]

It depends on the structure, but both CST and HFSS has TD and FD. HFSS converges faster for complex structures.
See this paper for comparison: https://www.piers.org/piersonline/pdf/Vol6No4Page395to399.pdf