| Author |
Message |
xxgcatz
Joined: 19 Feb 2004 Posts: 53
|
17 Nov 2004 1:53 electrical large cst |
|
|
|
|
Now electric large antenna simulation is a difficult disk. I want to know whch one of cst hfss and feko is good at in the electric large antenna simulation! thanks a lot!
cheers!
|
|
| Back to top |
|
 |
KMPA
Joined: 04 Dec 2003 Posts: 89 Helped: 2
|
17 Nov 2004 7:47 large antenna hfss to cst |
|
|
|
|
Feko is OK for large problems
KMPA
|
|
| Back to top |
|
 |
RFSimulator
Joined: 24 Jan 2004 Posts: 176 Helped: 15 Location: Europe
|
17 Nov 2004 15:14 compare hfss cst feko |
|
|
|
|
Hi,
for electrical large structure (several wavelength in all directions) HFSS is out of the game. This leaves CST (or time domain in general) or FEKO (MoM)
Which software is more suitable depends on the structure itself. If you only have a small amount of metal surfaces and dielectric parts in your structure MoM will be faster because you only have to mesh the surfaces. A limited number of metal parts will lead to very few mesh cells.
If you have electrical large structures with a large amount of metal parts and dielectrics (like a complicated corrugated horn with a length of 15-20 wavelength) time domain will be better.
F.
|
|
| Back to top |
|
 |
George137
Joined: 29 Feb 2004 Posts: 27 Helped: 1
|
18 Nov 2004 17:55 hfss feko |
|
|
|
|
| What is FEKO special at?
|
|
| Back to top |
|
 |
Element7k
Joined: 07 May 2003 Posts: 390 Helped: 8
|
24 Nov 2004 0:44 feko price |
|
|
|
|
Hi George137,
feko as compared to CST Microwave Studio has included in its package asymptotic high frequency techniques (e.g. Uniform theory of Diffraction; UTD and Physical Optics; PO) for use with electrically large objects. You can also use hybrid methods e.g. MoM/UTD or MoM/PO for problem solving. It is a especailly good tool for antenna placement and EMC applications. Of course it can perform many other tasks. See http://www.feko.info. Memory requirements and calculation speed for solving problems is reduced from normal Method of moments by using the MLFMM (Multiple level fast mulitpole method). It is however frequency domain so it uses surface mesh as compared to cells in time domain methods. Every different solver has its special appplications.
Regards,
Element7k
| George137 wrote: |
| What is FEKO special at? |
|
|
| Back to top |
|
 |
rugbyfun
Joined: 21 Oct 2004 Posts: 22 Helped: 1
|
24 Nov 2004 12:25 feko vs hfss |
|
|
|
|
I would not say that HFSS cannot handle
large radiation problem
I could show a simulation results of very large waveguide array
of course HFSS being a professional tool giving accurate results
require a professional machine with professional characteristics
|
|
| Back to top |
|
 |
Google AdSense

|
24 Nov 2004 12:25 Ads |
|
|
|
|
|
|
| Back to top |
|
 |
aw
Joined: 09 Jun 2001 Posts: 58 Helped: 2
|
28 Nov 2004 0:48 |
|
|
|
|
but how long did the hfss model take to simulate and on what system with how many processors and how much RAM? - not everone can afford expensive workstation solutions..... the method simply implies this regardless of what anybody might say...
hfss = high frequency small structure simulator ?
|
|
| Back to top |
|
 |
RFSimulator
Joined: 24 Jan 2004 Posts: 176 Helped: 15 Location: Europe
|
28 Nov 2004 12:33 Re: electric large antenna simulation with CST HFSS or FEKO |
|
|
|
|
Hi,
I have seen an Ansoft presentation were they showed electrical large structures like a 12 arm power divider with 600 000 tets. Solution time was about 1.5 H (for each freq. point as far as I remember)
Computer was a Sun Enterprise 4500 with 8 Processors and 8 Gb Ram.
I check the hardware price after this presentation. At this point is was much more then 100 000 US$ !! Pretty expensive if you consider that one could use any time domain code to solve the same structure on a "normal" PC.
F.
|
|
| Back to top |
|
 |