16 patch array antenna simulation using ie3d

EDAboard.com Forum Index -> Electromagnetic Design and Simulation -> 16 patch array antenna simulation using ie3d

Author

Message

xeneixe

Joined: 31 Mar 2004
Posts: 9

10 May 2004 18:34 16 patch array antenna simulation using ie3d

hello people. Actually i´m working in a 16 patch array antenna but i have some problems. firts of all, i´m using a 8 patches array (because the simulation time is too much in my Pc...)which is the half of the final array which i want to get. The resonant frecuency will be the sames for a 16 patch array? . second question: i used an infinite ground plane in the simulation, is it a big mistake? , and finaly, i´m using a complete simulation, could i try with other type? the accuracy is good?
data: Er=2,5 (polystyrene), h=5 mm.

if someone have some advice... thank you very much!!

xeneixe

mamali

Joined: 29 Apr 2004
Posts: 387
Helped: 4
Location: between hell and heaven

10 May 2004 21:29 Re: 16 fix array antenna simulation using ie3d

xeneixe wrote:

hello people. Actually i´m working in a 16 fix array antenna but i have some problems. firts of all, i´m using a 8 patches array (because the simulation time is too much in my Pc...)which is the half of the final array which i want to get. The resonant frecuency will be the sames for a 16 fix array? . second question: i used an infinite ground plane in the simulation, is it a big mistake? , and finaly, i´m using a complete simulation, could i try with other type? the accuracy is good?
data: Er=2,5 (polystyrene), h=5 mm.

if someone have some advice... thank you very much!!

xeneixe

dear xeneixe,
array does not influence resonance frequency (and if, very low).
infinite ground plane is increasing gain and bandwidth. u should take care of it, you will have lower gain and bandwidth with finite GP. it is mainly becouse of the radiation from GP edges, and becouse of no air gap, becouse of surface wave excitation.
finaly, you should first try other methods, they are p\retty good, i tried them, the best is separate simulating of feed net and p(at)tches. you could use periodic boundary too. it is well-explained in IE3D documentation.
++try other matrix solvers than SMSi, use AIMSIII.

ur friend, murti

plasma

Joined: 20 Dec 2001
Posts: 577
Helped: 40

11 May 2004 0:15 Re: 16 patch array antenna simulation using ie3d

Hi xeneixe

Look at the example directory you will find there good
examples of array. 16 element and more. also 64 element antenna
you can learn from the parameters how to build your antenna.
and print the help it's word documents.

Google
AdSense
Google Adsense

11 May 2004 0:15 Ads

divan

Joined: 01 Mar 2002
Posts: 91
Helped: 2

11 May 2004 4:34 16 fix array antenna simulation using ie3d

Sometimes to devide a problem to feeding network and patch array is preferrable. In this case you can simulate feeding by another tool, for example with MWOffice or Serenada and this will take a less time than with IE3D and allow real time tuning.

indiaco

Joined: 10 Apr 2004
Posts: 39
Location: Down South...

11 May 2004 4:46 Re: 16 patch array antenna simulation using ie3d

dear mamali :
why do you suggest to use AIMS ?
This i found in the ie3d help file :
"The PMS II, AIMS, Adaptive AIMS are not finalized. They do not have any advantage compared to other matrix solvers and they should not be used. "
i did simulate the same proyect with both solvers and ended up with higher vswr using AIMS III compared to SMSi(from 1.2 to 1.4). Which one should I finally set ?

thanks in advance.
NICO-

mamali

Joined: 29 Apr 2004
Posts: 387
Helped: 4
Location: between hell and heaven

11 May 2004 8:00 Re: 16 fix array antenna simulation using ie3d

indiaco wrote:

dear indiaco,
AIMS supposed that mesh elements with a certain distance and more, do not influence eachother's current, so do not comput the mutual impedance between them, resulting in a very sparse impedance matrix. and there is a lot of methods to efficiently solve the sparse matrix invers. thus, AIMS speed is so better than SMSi, if the strusture is big(about 3-4 lambda or more, more structures, its better) especially, in array designs, sometimes it is imposible to use SMSi. however, it is obvious that SMSi is more precise than AIMS. but the problem is speed.
AIMSIII is relatively robust. bytheway, look at the "matrix fullness" in the simulation dialog box, if it is about 5%, then you are using AIMS properly. even for 10% its speed is better than SMSi. but if it is more dense, then you have to use SMSi.

murti