Which one is better, HFSS or CST ?

[plasma]

Re: H F S S vs C S T

Hi costox

IE3D: MoM-Based EM Simulator
--------------------------------------------------------------------------------
IE3D is a full-wave, method-of-moments based electromagnetic simulator solving the current distribution on 3D and multilayer structures of general shape. It has been widely used in the design of MMICs, RFICs, LTCC circuits, microwave/millimeter-wave circuits, IC interconnects and packages, HTS circuits, patch antennas, wire antennas, and other RF/wireless antennas.

look at http://www.zeland.com/

PL

 

[yuanzy97]

Re: H F S S vs C S T

Does one know where is many macro file for CST? THank you!

 

[Catalyst]

Re: H F S S vs C S T

Does Microwave studio support low frequencies below 50 MHz?

 

[RFSimulator]

Re: H F S S vs C S T

Hi,

in general yes! the question is, which solver is the best choice for your problem.
If the strcutre is electrical small (much less then a wavelength) you can use the freq. domain solver of MWS. For low freq. problems it might be more effctive then the time domian solver. What kind of problems you are looking at.

F.

 

[blue icon]

Re: H F S S vs C S T

About this forum I have to say that it is very interesting to discuss advantages and disadvantages about these two masters on electromagnetics simulation.

I would like to share one experience:

At the moment I'm studying an electromagnetic bandgap structure (a structure that exibits a region of microwave frquencies with high resistance to the propagation of electromagnetics waves) and it can be seen that it is a very strong resonant structure. It's jus a microstrip line with some holes in ground plane equaly spaced.

I have tested in MWS and HFSS and the results look more or less aproximate. My question is the "more or less": in MWS it appears a ripple in the more attenuated zone (propably due to truncation errors). MWS need to filter this (with the AR filter inside) and it should be necessary to put a filter to cut the low ripple, in the opposite to HFSS that doesn't exibhit this ripple.

As a broad-broadband simulation HFSS gets more unstable in the high frequencies (discrete solution - 5-50GHz range sweep, step 100MHz). MWS don't get so unstable and it can predict better the solution in high-frequencies.

They seem to be very powerful but there are still large steps of evolution to be expected.

Thank to all who read my post.

 

[costox]

H F S S vs C S T

Many of the 2D and 3D functions of MWS annoyed me so much this week.
Example: when you sweep, why do you need to create then select 2 curves? The way HFSS does it is much more intuitive: you select the curve, you click sweep, then you enter how much and which direction you want to sweep... Done!

Many more annoyances like that happens with MWS.

 

[xcyyx]

H F S S vs C S T

usually ,I use IE3D to simulate printed circuit , it's simple and accurate,both HFSS and CST are not good chioce for this kind simulation . HFSS is accurate for simulating antenna radiation.If you need design coaxial structure, CST will be a good choice.

 

[auguste_com]

Re: H F S S vs C S T

Well, CST is better than HFSS in my opinion. It is faster.

 

[mpero]

Re: H F S S vs C S T

Hi Folks,

In regard to the VBA code: I think CSTs VBA is much simpler to use than that of HFSS. To learn the VBA code stuff, just open the "History List" from your CST MWS workspace. Double click in whichever command, what you see is actually the VB code that does the function. Copy and paste this stuff on the VB Editor (by clicking in MACROS/OPEN VBA Macro Editor).

In regard to speed: I think HFSS eats Sh*t in comparison to MWS. I tried once to simulate a simple dumb horn antenna, I was using HFSS normally by then. It turned out to be impossible. Then I tried MWS, even without simmetry planes it could get it done, the results agreed perfectly with those from Ballanis book.

But to say that HFSS is a piece of junk is not true. It is a good software, but to my point of view CST is much better.

peace

 

[Mansour_M]

Re: H F S S vs C S T

Hi Everyone!

I don't use CST so much instead I use HFSS and it is really awful. It's not fast and accurate. The mesh generation part is not able to work with connected shapes and the calculator is not friendly.

Thanks!

 

[rfmw]

Re: H F S S vs C S T

here are some results comparing three different designs (3D structures) in both, cst and hfss. i find the results very close and similar, but the difference should be even smaller with better meshes in both simulators.

Zodd ... cst, Zeven ... hfss

the main problem with both simulators i find the memory appetite.

sorry for the blurry images of the structure.

 

[SAJ25]

H F S S vs C S T

hi rfmw

whould you please upload the cst and hfss files of your simulation. i would like to work on your simulation
regards

 

[rfmew]

Re: H F S S vs C S T

I'm using both.

I think HFSS is very expensive.
Normal CST includes some options of HFSS.

 

[3dandyw]

Re: H F S S vs C S T

hi,

back on memory issues, MicroStripes employs the Transmission Line Matrix (TLM) method to solve Maxwell’s equations in the time domain.

on several case i found out to be as accurate as CST but far more attractive for memory requirement and simulaiton spped (i guess their mesh technology is very advanced)

i suggest u give a try

actually i would b interested to run your model on MS
feel free to post it

regards
AW

 

[pengboy]

H F S S vs C S T

i dont know CST a lot, just tried a couple of times. i heard CST is faster for narrow band designs.

 

[SAJ25]

H F S S vs C S T

no penboy!!! hfss is ideal for narrowband simulation and cst is perfect for broadband structures!

 

[drasko]

Re: H F S S vs C S T

I have never tried CST, I must admit.

But when I hear people are saying something like: HFSS is fast for narrow band it sounds so untrue.

HFSS was very slow for what ever I tried it, and I am dealing with pretty simple models, and not wideband. I didn't dare using it on complex problems. Is it faster than CST I don't know, but it's very slow, that I know for sure.

As far as SCT, I don't know, although I believe they are surely faster than HFSS for wideband due to the nature of time-domain simulation, and low speed of HFSS.

 

[cisconur]

Re: H F S S vs C S T

hfss use whole memory and cpu and sometimes solving period takes hours or days , i dont know is it normal or not; and i didnt use cst but i am sure it is better than hfss

 

[nadabro]

Re: H F S S vs C S T

I think we should make a benchmark for CST and HFFS..to compare different computer system´s(CPU/RAM)/different OS/diferent versions of CST and HFFS/simulation times/results/etc..

The same design for CST and HFFS...if all of us contribute giving the results, we can make a "database" that would help many people.

 

[prafuldeo]

H F S S vs C S T

hi.. i've been usig cst from long time now and i worked with hfss and ads too.. but since i got hands on cst.. no turning back. its so easy and fast.. fully recommended..