SOS!!! HFSS Stacked patch antenna problems

[cbear]

I've had problem with simulation stacked multi-layer patch antenna.
It seems the upper patch just does nothing. I am afraid it's the problem with the feed. (i am using coax/probe feed). Could you anybody possibly tell me how this should be done?

Thanks a lot!!!!!!

 

[Lokter_Whatsen]

It might be best if you could post your HFSS file up so that we can know?

BR,
Watson

 

[antennas321]

SOS!!! HFSS Stacked fix antenna problems

you can use a wave port to feed the coaxial line. the inner pin should be directly connected to the lower p_a_t_c_h, and the parasitic p_a_t_c_h should be okay the way it is (floating above the fed p_a_t_c_h on a dielectric layer). Make sure the problem converges to a good delta-S value (such as 0.005 if you don't seed the radiation boundaries). I would seed the mesh on the radiation boundaries to make sure the gain pattern can be calculated correctly.

Using 2D surfaces for the metallization should be fine. In the impedance response, you should see two resonances (one for each p_a_t_c_h). Make sure you are looking at the Smith chart as well as VSWR curves.

Check the boundaries to make certain your groundplane is defined properly (PEC surface will work).

 

[cbear]

hfss file

 

[cbear]

Thank you guys so much for your replies!!! I really greatly appriecate that...

I am really frustrated right now, because I am the only one who is using HFSS in my whole college...and I have no person to seek advice whenever i have problems.

To antennas321,

Like you said, so I do not have to feed the upper fix? (What if in the real antenna, the coax feed is also extended through a via to the top fix? This is bascially what I did. so i just drew a hole on the ground as the out connector, and then have the small cylinder(material- metal) as the inner connector with the same height as the lower substrate so that extends to the lower fix. but when I tried to feed the upper fix the same way just by making this cylindal the height to the top fix, HFSS gave me some modeling error saying some kind of conflict with the upper substrate. Do you understand what I am trying to explain? So i can't do the same feed as what I did with the lower fix. So what I am doing now so that there would be no modeling error is that i have to take the feed apart, and draw other cylinders from the bottom of the upper sub to the upper fix. So there is no modeling error. But bascially, I never get the upper fix resonating...

Yes, the ground is PEC and actually as well as the patches.

Also could you please explain a little more about seeding the mesh? i just have a box with radBoundary and never tried to seed the mesh? Usually I just leave the mesh as default (delta-s=0.03, maximum pass =3), sometimes, the solution converges and sometimes it doesn't... so what does the solution mean if it doesn't converge? Does that matter a lot?

The last question, i actually never looked at Smith Chart, only vswr. I just assumed they would match. Are they?

Finally, thank you guys so much again for your replies!!! You are not only experts at HFSS but also my life savers!!! Do you mind if I ask how I should learn with this software. I've gone through the training manuals coming from Ansoft, but it does not seem help too much when it goes into details. could you please recommend any available material about HFSS?

Thanks a loooooooooooooooot!

I've also posted the porject file here, and hope to hear some ideas from you guys!!!

 

[bingjiang99]

Hello cbear,

I used to work on the stacked patch antenna, and had the similar problems. My problem is that I have a small distance between the top and bottom patches. You can try to increase it to see its effect. Another possibility is that the top patch is too small. I didn't check your HFSS file, so I am not quite sure.

Good luck,

Thank you guys so much for your replies!!! I really greatly appriecate that...

I am really frustrated right now, because I am the only one who is using HFSS in my whole college...and I have no person to seek advice whenever i have problems.

To antennas321,

Like you said, so I do not have to feed the upper fix? (What if in the real antenna, the coax feed is also extended through a via to the top fix? This is bascially what I did. so i just drew a hole on the ground as the out connector, and then have the small cylinder(material- metal) as the inner connector with the same height as the lower substrate so that extends to the lower fix. but when I tried to feed the upper fix the same way just by making this cylindal the height to the top fix, HFSS gave me some modeling error saying some kind of conflict with the upper substrate. Do you understand what I am trying to explain? So i can't do the same feed as what I did with the lower fix. So what I am doing now so that there would be no modeling error is that i have to take the feed apart, and draw other cylinders from the bottom of the upper sub to the upper fix. So there is no modeling error. But bascially, I never get the upper fix resonating...

Yes, the ground is PEC and actually as well as the patches.

Also could you please explain a little more about seeding the mesh? i just have a box with radBoundary and never tried to seed the mesh? Usually I just leave the mesh as default (delta-s=0.03, maximum pass =3), sometimes, the solution converges and sometimes it doesn't... so what does the solution mean if it doesn't converge? Does that matter a lot?

The last question, i actually never looked at Smith Chart, only vswr. I just assumed they would match. Are they?

Finally, thank you guys so much again for your replies!!! You are not only experts at HFSS but also my life savers!!! Do you mind if I ask how I should learn with this software. I've gone through the training manuals coming from Ansoft, but it does not seem help too much when it goes into details. could you please recommend any available material about HFSS?

Thanks a loooooooooooooooot!

I've also posted the porject file here, and hope to hear some ideas from you guys!!!

 

[antennas321]

In HFSS, one solid (3D) object cannot intersect another or it will produce a Boolean error. You should subtract one object from the other. So in your case, you can subtract the coax pin from the substrate that is below the upper p_atch. That should fix the error.

Seeding the mesh means adding more vertices (nodes) before the adaptive mesher begins to add them where they are needed (where the fields are rapidly varying). For antenna problems, seeding the radiation boundaries is almost always necessary to ensure that the far-field gain values are correct. Think of it as sampling a field distribution with a certain number of discrete locations. Seeding the mesh increases the number of samples. Using 10 per wavelength is good.

I would recommend talking with an Ansoft app engineer to make sure you are setting up the problem correctly. The maximum passes should be set much higher than 3, try 10 or 12. Delta-S should be set to 0.01 or 0.005. If the solution doesn't converge, that means that your S-parameters are still changing and the solution is not close enough to the real answer. It is extremely important to reach convergence if you are trying to compare with measured data.

You can find much info on the web about Smith charts and complex impedance. You really can't design antennas based on VSWR alone -- it helps to have a good understanding of the input impedance behavior (Smith chart).

Good luck.

 

[engineer nandhu]

I've had problem with simulation stacked multi-layer patch antenna.
It seems the upper patch just does nothing. I am afraid it's the problem with the feed. (i am using coax/probe feed). Could you anybody possibly tell me how this should be done?

Thanks a lot!!!!!!

i want to know how to attach the two patch antennas by screws in HFSS