[SOLVED] HFSS: Simulation taking too much time

Status
Not open for further replies.

anvc

Newbie level 6
Joined
Apr 18, 2013
Messages
12
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Visit site
Activity points
1,367
Hello all,

I am trying to simulate a simple UHF printed dipole antenna. Due its size (lambda/2=17cm), after some minutes of simulation the number of tets go to more than 4 million!

Even if after adding a mesh operation to limit the number of elements, the simulation still uses million of tets...

Could anyone here give me an advice on this issue?

Thanks!
 

Hi there
I had like your problem, simulation by HFSS taking a long time sth horrible, I know, but I decided to switch to ADS. While analysis a structure takes 40min in HFSS same structure takes only 5min...So I advice you use ADS.
BTW my structure was microwave filter.
I hope it helps you.
Good luck
 

@anvc - This sounds like a problem with mixed units or improper setup. A simple dipole should not take more than a couple of minutes to solve. Even though the structure may be large in physical size (17 cm), it should be reasonable in electrical size (lambda/2). Check to make sure your units are correct, and that your analysis frequency is appropriate. One more thing to check is if you have any high permittivity materials surrounding your antenna. If this is the case, like with a human head (er ~ 80), then you might want to enable the checkbox next to Lambda Refinement on the options tab of the Analysis setup, that specifies to use "free space wavelength." I don't recommend using this in general, but if you have a very high permittivity object, it can sometimes help.

@sayyedjamal - I'm not sure how ADS is going to solve a 3D antenna problem. I can see where you might be able to use Momentum instead of HFSS for a filter, but the antenna is inherently 3D. Also, HFSS should do a fine job with microwave filters. If you have any problems, I suggest you contact your local ANSYS Applications Engineer for assistance.
 

A printed dipole should not take long to run. Without seeing your setup it is hard to identify the problem .

Cesar
 


Since my structure is Microstrip and ADS Momentum compute S-Parameters I'd prefer to use ADS, I consult with my supervisor and he let me use ADS, like I said I had long time simulation problem, and honestly I find using ADS is perfect.
BTW I just saw a dipole example by ADS sth like far field can be seen:
 

View attachment Dipole_on_FR4_3.zip
A printed dipole should not take long to run. Without seeing your setup it is hard to identify the problem .

Cesar

Hello ! Thank you for your replies!

I send in attachment the HFSS file. After waiting 22min the simulation was over, but with an error:
"[error] Mesh leak is found for (2) of bodies: (Au_Feed) (Au_Feed_1) (5:02:52 PM Apr 20, 2013)"

This can be due to the thickness of the top Gold layer which is only 0.1um. However, as I will fabricate this antenna using Cu,Ni and Au as top-layer I would like to keep the real dimensions. Could you guys give me a hand here?

Thanks again!
 

Hello,

I went ahead and looked at your setup. The major error you had is that you had 3 dipoles on top of one another. You can only simulate ONE at a time. I went ahead and deleted two of the dipoles, leaving only one, and redrew the lumped port to match the current model. Took about 1 min to run.

Cesar
 

Attachments

  • Dipole_on_FR4_3_crg.zip
    32.5 KB · Views: 135

Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…