Coupled transmission lines using Sonnet

[TH22]

Hello,

I need some help with stimulating coupled striplines using Sonnet.
The coupled line consists of two strips with 4 ports. The lines have the same dimension and shared ref plates. I need to calculate the coupling coefficient of the lines for various spacings between the lines. How can I extract the Self Inductance of each of the strip transmission line as well as the mutual inductance using Sonnet?
Is there any equation for the coupling coefficient between lines in Sonnet?

Thank you

 

[volker@muehlhaus]

If you analyze the lines at low frequency, where lines are short compared to the wavelength, you can can use Pi model extraction. This gives you a SPICE model where you can read self inductance and k factor.

Pi model can be created from the Sonnet response viewer. It is a standard feature for all Sonnet products.

Or did you mean coupling S21?

 

[TH22]

Hi,
Thank you for your help, much appreciated. However, I am still facing some issues with the design. In my extracted Pi model, there is no k factor between the inductances of my design (L2 and L5 on the attached file). How do I extract the k factor between the two inductances in the
Pi model?
Can I use the RLGC model of the N-coupled model to determine the self-inductance and mutual inductance?
What is the L2:1 in the RLGC model? is it mutual inductance or k factor between inductances?

Thank you!

 

[BigBoss]

PI Equivalent Model is valid for electrically small structures only. ( 1/20 wavelength )

 

[TH22]

Thanks for the help.
I'm using a small wavelength for the design and the length of lines is half of the wavelength.
For this design, How can I extract the k factor between lines in Sonnet?

Thank you!

 

[FvM]

I'm using a small wavelength for the design and the length of lines is half of the wavelength.
For this design, How can I extract the k factor between lines in Sonnet?

Mutual inductance respectively k are parameters of lumped inductors. Line structures with length > \( \lambda \)/10 can't be analyzed as lumped elements. Parameters of coupled transmission lines are even/odd impedance and propagation speed and length. For small frequency range, an equivalent lumped circuit can be derived.

 

[volker@muehlhaus]

and the length of lines is half of the wavelength.

You need to analyze this in shorter segments.

A long line can be represented accurately as cascaded short segments. That is a standard approach for transmission lines, not limited to Sonnet.

 

[TH22]

You need to analyze this in shorter segments.

A long line can be represented accurately as cascaded short segments. That is a standard approach for transmission lines, not limited to Sonnet.

Thanks for your help.
I tried with short lines. But there is no k factor between the inductances of lines in the extracted Pi model.
I am using superconducting lines and material, same as the superconductor resonator model and material in Sonnet.
Should I consider anything else for the lines? Is this possible to find the k factor of these lines?
Can I use the RLGC model to calculate the k factor?

 

[volker@muehlhaus]

I tried with short lines. But there is no k factor between the inductances of lines in the extracted Pi model.

I would assume your k is small then.

In the PI model extraction window, there is an"Options" button on the top right. That defines a minimum coupling factor, which is 0.01 by default. Values below that limit are supressed in the output file. But you can lower the limit, if you wish.

~~

Looking at your model, I wonder what the analyzed line length and spacing was. Maybe it is better to extract coupling factor at a lower frequency, instead of making the line segments too short. We don't know enought details to give good advice here.