CST is not accurate!?

recently I simulate the rcs of dielectric resonator(very small 2cm*2cm,*0.898cm)
i found that the frequency of the peak of rcs is not the resonant frequency.in theory the frequency of the peak of rcs is the resonant frequency ,why?
the mesh value

i upload the simulate mode ,maybe you can help me!
any advice regards!


I want to simulate it with HFSS but I don"t how to set,who can give me an example
of simulating RCS of dielectric!

hi,try to increase lines per wavelength to 20,lower mesh length to 20,in specials,in refinements set refine at pec/lossy metal edges to 5

Thanks I am trying!

Added after 1 hours 3 minutes:

hi,abdoeng
I simulate it after your advice ,the result is not right.

you can download my simulating mode,I need your help?

I also have a question
I only concern rcs of the bandwidth of 2.40-2.50, when I simulate the fmin-fmax set to 2.40-2.45 is right or not?

yes,u right but i have cst 5.1 ,is its ur version,i could help u
regards

Dear gliet

RCS in CST is right only from 2006b before it. It was wrong. So which Ver. you have?


pl

hi
I use cst 5.0
RCS is right only in cst 20006?but there is three examples of RCS in cst 5.0.

they can give you 1000 of examples but you can change the fact that
the result is not right.
there is more but they fix it in 2006B.

pl

I've seen several cases where CST products give "wrong" results.

For example EM Studio cannot properly compute the inductance of a single turn circular wire loop, IMO, and it apparently gives significantly less accurate results than Infolytica MagNet V6 for other coil configurations.

IIRC, MWS failed when I tried to analyze a helical resonator; i.e., a coaxial resonator with a helical inner conductor.

So I'm not surprised at all to hear of other failures.

what can i do?
cst is not right,but what software is more accurate than cst,HFSS?

HI,

I would suggest u to try MicroStripes 7.5 - TLM(transmission Line Matrix) based.

I have been using it for component design (antenna/phased aray) and installed performance and their automatic meshing features (lumping ) is helping a lot to acheive greater accuracy and still ok for simulation speed
microstripes.com

hope this help
good luck
AW

Hi,

I would suggest you to try SEMCAD X.

Best wishes

Dear gliet

CST is accurate as long as you keep the rules in the pdf attached and check the mesh

that you don't have more then tree substrates in the same cell

pl

Gilet

Try:
1. the adaptive mesh refinement .
2.change the maximum number of pulses at steady state to 100.
3.the accuracy to 60.

pdf here for cst 2006b,any pdf of cst2006 will be useful
regrds