CST convergence problem

EDAboard.com Forum Index -> Electromagnetic Design and Simulation -> CST convergence problem

Author

Message

winglj

Joined: 21 Feb 2005
Posts: 191

05 Jul 2006 1:09 reducing the pba fill limit

Hi, recently I am doing the simulation of patch antenna by using CST. But when I using the waveguide port for coaxial cable, there are some convergence problems arising:

The CST time-domain solver shows me the messege:
" The calcuation seems to be unstable! Consider reducing the value for the stability factor in the solver specials options or reduce the PBA fill limit in the automesh properties."

And the output signal will oscillate at the end of the simulation.

I am wondering how can I solve this problem without affect the accuracy?

It seems the output signal oscillation is due to numerical error. But what kind of setting will generate this type of errors?

And what's the meaning of the PBA fill limit? How can it affect the simulation/ mesh generation?

Thanks in advance, and any recomendations will be welcome.

wave-maniac

Joined: 05 May 2001
Posts: 279
Helped: 3
Location: USA

05 Jul 2006 1:56 cst pba fill limit

One of the reasons could be the too short length of the transmission line between the port and the closest discontinuity. Try increasing its length.

Good luck

plasma

Joined: 20 Dec 2001
Posts: 577
Helped: 40

05 Jul 2006 23:00 cst convergence

if the input not parallel lines or not perpendicular to the port that's one of the reasons

pl

marruto

Joined: 22 Mar 2007
Posts: 1

05 Dec 2007 18:05 pba meshfill limit

Hello, I have the same error message: " The calculation seems to be unstable! Consider reducing the value for the stability factor in the solver specials options or reduce the PBA fill limit in the automesh properties."

And I send my simulation picture,It's a Rf-coil and I was readying your answers and I made your suggestions but I could not make the simulation,Someone can help me?

Thank you

Pushhead

Joined: 19 Jan 2005
Posts: 195
Helped: 8

08 Dec 2007 15:46 cst mesh settings

Hi,

1. Check if the port plane is aligned with the coax face and that there is no gap between them.
2. On the transient solver "specials" menu check the box called: "Add electric shielding around port region".
3. Another reason could be your mesh settings. Try to do a little mesh study and see what are the optimal mesh settings in which you get both reasonable results and convergence.

good luck,
P.

Google
AdSense
Google Adsense

08 Dec 2007 15:46 Ads

loucy

Joined: 26 Aug 2001
Posts: 432
Helped: 10

09 Dec 2007 5:01 Re: CST convergence problem

most likely it is due to some elementary error in setting some material or boundary.

mpero

Joined: 09 Feb 2006
Posts: 38
Helped: 4

16 Feb 2008 14:02 Re: CST convergence problem

Other solution, try to increase the mesh on the objects near the port only. Select the object and by right clicking of the mouse choose the "Local Mesh Properties" option.

Fill in the "maximum mesh step width" with nice values, especially along the axis which is normal to the WG port.

good luck

waseem44

Joined: 06 Jul 2007
Posts: 30

30 Apr 2009 16:46 Using CST for Shielding

Hi guys

I recently started using CST microwave 2008. I used it first for SE of a box wit a plane wave and installed a waveguide 1 cm. I am facing problem in getting some correct results i used a probe inside. creating a seam shows no difference, idont know where is something wrong. how can i use some port instead a plane wave
someone plz help me in using cst
thanks
waseem