Antenna excitation in HFSS

[Avia]

Hi, everyone,

I'm confused by a problem these days and I hope that you can help me.

I want to investigate the return loss of a TEM horn by using HFSS, while the impedance of the feeding line is supposed to be 50Ω.

Do I need to describe the concrete feeding structure in HFSS when doing simulation? If so, how to make sure that the feeding structure is just 50Ω when the feeding is complicated?

I wonder whether I can only define a lumped port with impedance of 50Ω at the throat of the horn instead of describing the feeding strucure .


Thanks a lot

 

[h_cet]

i have a microstrip antenna. and this antenna is finished by 50 ohm. i finish the antenna lumped port in hfss. reel : 50, 0 is entered in other box (i dont remember another box's name) you define the line from ground to patch.

i reached a very good solutions this way.

bye

 

[Avia]

Thanks a lot.

I have another question, you know that when simulating antenna in HFSS, we need to set a radiation boundary(e.g. A air box enclosing the radiation objects). The lumped port set in the air box will contact the air on one side and contact the radiation object(e.g. TEM antenna) on the other.

Therefore there are reflections on both sides of the lumped port, then the S11 we get represents what, the reflection between the lumped port and the antenna, or the one between the lumped port and the air?

Looking forward to your reply!

 

[costox]

Avia,

I think you didn't set up your lump port correctly.

I can have a look at your port if you post it here.

 

[Avia]

Thank you very much, costox.

It's not a pdf but a hfss file. So before opening it, pls delete the suffix".pdf".

 

[jallem]

I did not see any thing wrong with your design

 

[costox]

Avia,
are you sure of this?

There should be a pin in the middle not touching the TWO flanges (that you call horns).

The ground is BOTH OF THE FLANGES, not ONE flange.

 

[jallem]

In RF there is not concept of ground.
If the horn is feed by a waveguide there is not ground..
The aproximation by a lumped port is valid but he
will not be able to see any mode creation.
Granted, the best model he can go with is a waveport,
since HFSS will be able to analyze the modes in the
waveguide and come up with the real distribution of
currents and voltages at the cross section of
the waveguide.
This is a good example to explain the difference
between a lumped port and waveport.

 

[Avia]

Thank you very much, everyone. I got it.

 

[marchz]

Thank you very much, everyone.