Simulation of an IFA with ADS and HFSS

[Bou]

Hello and happy new year

I've simulated an Inverted F antenna with ADS and I designed the same antenna in HFSS but the problem is in aim to have the same result
(the best S11) I had to increase the distance between the feeding and the shorting point in HFSS by 3 mm!!!

plz have u any explaination? what design should I keep? is that normal?


many thanks

 

[WimRFP]

3mm is large when the size is some cm, but small when the size is 1m.

When the geometry is exactly the same (inclusive substrate layers), the problem should be in substrate/metal properties, or meshing. Check the dielectric properties and the conductivity of the metal you use. If you have an electrically very small (P)IFA, Ohmic loss may affect feedpoint position.

You may also reduce the meshing size (so more cells/lambda) to make sure your meshing is not too coarse. Did you try another port scheme or size of port?

 

[Bou]

I used the same properties of metal and substrate, for meshing I tried different frequencies bu I got always this difference .... I noticed that in HFSS we can draw substrate with a specific dimension but in ADS, the software doesn't give this possibility

 

[WimRFP]

Does this mean that your geometry (including dielectrics) is different between simulations? When ADS doesn't support finite size dielectrics, you can only compare them when you use infinite dielectrics in HFSS as well.

Is the difference only in the position of the feed, or also in the length of the antenna element? If you have low radiation efficiency, does it change when you change the metal conductivity? The reason for this is that how ohmic loss is implemented in the simulator may differ per simulator (sorry I don't know the details, check the documentation for that).

What is the actual size of the antenna (to figure out whether 3mm is significant or not)? Do the current density plots look the same, and as expected (also just to check whether there is something wrong in the geometry)?

 

[Bou]

I tried to increase the size of the substrate but I did't get the same result, there is a difference in the feed point and length too,I redisigned my antenna and I used the same dimensions like ADS and I noticed that when I add perfect E boundary to my metal conductor (copper) I got better S11 (-22db) any ideas?

Does this mean that your geometry (including dielectrics) is different between simulations? When ADS doesn't support finite size dielectrics, you can only compare them when you use infinite dielectrics in HFSS as well.

Is the difference only in the position of the feed, or also in the length of the antenna element? If you have low radiation efficiency, does it change when you change the metal conductivity? The reason for this is that how ohmic loss is implemented in the simulator may differ per simulator (sorry I don't know the details, check the documentation for that).

What is the actual size of the antenna (to figure out whether 3mm is significant or not)? Do the current density plots look the same, and as expected (also just to check whether there is something wrong in the geometry)?