Hi Madam ( Aparna Sathya Murthy )!

I am using HFSS and found this youtube video to plot the dispersion diagram for a unit cell (https://www.youtube.com/watch?v=AXOpFgJcI38).

But, my antenna is an IFA antenna and has a lumped port for excitation.

My doubt is: can I use both lumped port and periodic boundary conditions at the same time to plot the dispersion diagram? Or is it enough to just simulate the antenna using eigen mode analysis using periodic boundary conditions to get the dispersion diagram (without the lumped port)? How does the software (HFSS) recognise where to excite the antenna (as I have a gap for capacitor and another gap for lumped port within the same IFA antenna)?

With Regards,

Suhas D

Suhas Dinesh unfortunately, you cannot. The S11 response of the antenna most likely is (and should be!) dominated by the type of antenna itself, and not a metamaterial cell that you have embedded inside it.

You will need to simulate the metamaterial unit cell on its own to obtain its dispersion. Embedding it in a transmission-line and extracting the scattering parameters is a common technique.

Thank you Madam Aparna Sathya Murthy and Sir Stuart Barth for your help.

I was going through a paper titled "Small Antennas Based on CRLH Structures: Concept, Design, and Applications."

In this paper the authors have come up with the dispersion diagram of a 3 cell metamaterial antenna (please find attached the image). But I don't understand how they got it. This is the kind of dispersion diagram I need for my metamaterial IFA antenna to observe the Right Handed and Left Handed regions.

For a 2 port leaky wave antenna, there is this formula acos((1-S11*S22+S12*S21)/2/S21), which can be used to plot the dispersion diagram. But what about a device with a single port, such as an antenna?

Any ideas? Please do help!

With Regards,

Suhas D

Suhas Dinesh the authors state in the paper: "In Figure 3a, we plot the dispersion diagram associated with the unit cell shown in Figure 2" -- i.e., they simulated one unit-cell on its own.

Based on the quality of the dispersion data, I would assume they obtained it from an eigenmode solver, not from extraction via scattering parameters.