Patch antenna simulation - Sonnet Software

[jmga9]

Hi everyone,

I´m trying to simulate a patch antenna, but i´m having problems because i dont know how to make the ground plane. I have done 3 dielectric layers: air, dielectric, air. But i´m not sure if this is the right way to do a finite ground plane! I will be really gratefull if anyone could help!!
The software simulation is Sonnet!

Best regards,

jmga9

 

[volker_muehlhaus]

You can model the antenna this way, with air above and below, if the size of the ground plane matters and you want to explicitely draw the ground plane. In this case, the top and bottom cover of the analysis box must both be set to "Free Space".

However, for a "normal" patch antenna with large ground, just use the bottom of the analysis box as ground. In this case, you only define two dielectric layers: substrate and air above. Set the bottom cover to "Lossless" or whatever metal you want to define for the ground plane, and set the top cover to "Free Space".

Have you seen the patch antenna examples that come with the software?
From the Sonnet taskbar, go to Help > Examples.

 

[jmga9]

Hi,
thanks for your quick response.
Yes i saw the examples, but I was in doubt about ground plane!! because i've done in practice this antenna and i had approximately 268 + j3,9, for ressonant frequency (2.42Ghz) so i was waiting have similar values with sonnet! since i´m not having i supose that i was doing something wrong!
I send my patch antenna, this is with the normal ground plane!

best regards,

jmga9

 

[volker_muehlhaus]

Yes, there are mistakes. The important mistake is the box size: you must keep much more distance between the antenna and the box side walls. Have a look at the User's Guide, chapter on "Antennas and Radiation".

Also, the port type is not correct. You have defined a auto grounded port at the box wall. The software detects this mistake and changes this to box wall port. Auto grounded ports are only used inside the analysis area, when you can not extend the line to the box wall.

I have corrected the model and also changed the substrate epsilon to 4.7, which is what I measured for "my" FR4 material at 2GHz.

 

[jmga9]

ok, i will see that chapter!!
But theres's a big problem i'm using sonnet lite, so its limitated to 14MB. So i have to see how resolve this problem!!
thanks a lot for your help
best regards,

jmga9

 

[volker_muehlhaus]

Try the attached model with bigger cell size and coarser mesh. This should work in Sonnet Lite with <15MB.

 

[jmga9]

Yes that works! i'm having 109.3 - j37.42. Do you know the Sonnet eficiency?

 

[volker_muehlhaus]

Do you know the Sonnet eficiency?

I do not understand your question. What do you mean?

 

[jmga9]

The comparision between the simulation results and practice results? are they similar?

 

[volker_muehlhaus]

The comparision between the simulation results and practice results? are they similar?

Yes, from more than 10 years of experience with Sonnet I really trust the results if the material data is well known.

For your antenna, the big unknown are the properties of the substrate. It loooks like you have FR4, and there are big tolerances in the epsilon between different FR4 suppliers. The typical microwave substrates are quality controlled for consistent epsilon, but FR4 is not.

 

[jmga9]

Great!! yes youre rigth i have seen that, because in the box where substrate came said that epsilon is 5, FR4, but i have seen in sonnet library 4.9 and another places 4.7. I think the best should be Duroid which have epsilon 2.2, and lower loss, at least is what i have seen that is more used! By the way all this is for my master project! i´m searching a simulator to help with the antennas!
I request a sample of Duroid and i received last week, i will search a document where they used epsilon 2.2, do the simulation, the antenna, and see the results.

Best Regards,

jmga9