Several doubts about circuit model

I have several doubts about circuit model.
1): the txt file exported from Mode solution(FDE) has “loss” item (1.png). while the “Mode Waveguide” element also has loss and excess loss items (2.png) , where the “loss” item refers to propagation loss and “excess loss” refers to the loss caused by fabrication (for example, roughness of the sidewall).
The exported txt file contain loss, does it means the exported ldf file also contain loss item? If so, what is the relationship between the “loss” in ldf file and “loss” in “Mode Waveguide” element ?
Which kind of loss does the loss measured by experiment belong to?


2): when a mode profile of a bend waveguide (R=10um,angle=90) is imported to Interconnect, its length is a fixed value, which is equal to pi100.5, right?

3): For mmi, is it posible to build compact model by a way of combining the simulation result (S matrix )of Mode or FDTD and performance (insertion loss and power imbalance) by experimenat measure?

Hi @1683324,

Thank you for your questions.

  1. Yes, the .ldf file will contain the “loss” term, which corresponds to the “loss” in the MODE element. The “excess loss” of the MODE Waveguide element in INTERCONNECT corresponds to loss that is not measured by the FDE solver in MODE Solutions, for example, as you mentioned, scattering loss from sidewall roughness. An experiment would include all sources of loss, so both the “loss” and “excess loss” would be measured.

  2. If you import data for a mode from MODE Solutions into a MODE Waveguide element in INTERCONNECT, the length of the waveguide is not fixed and can be chosen by the user. If you create a bent waveguide element in INTERCONNECT, however, then the length will be determined by the angle and the radius.

  3. I discuss using experimental results to create compact models in your other post: How to build scripted element of MMI?

I hope that helps. Let me know if you have any more questions.

Thank you for respond. I have some other questions.
1.you mean, the “loss” item in ldf file corresponds to the “loss” in MODE Waveguide element in INTERCONNECT?

Then, as the experiment loss included both the “loss” and “excess loss”, which category should I put it? “loss” or “excess loss” ?
2. thank you for your respond. You verified my guess.
3. thank you for your respond and kind reminder.

Note that “loss” is measured in units of dB/m, and “excess loss” is measured in dB. This means that the change in signal amplitude due to “loss” will depend on the “length” of the element, but the change in signal amplitude due to “excess loss” will be constant. This is the only real difference between them.

It is up to you how to include your experimentally measured loss. You can keep it as a the measured value and just put it as “excess loss”. You could also divide the total measured loss by waveguide length and put it as “loss”, which would allow you to take into account the change in loss as you vary the length of the waveguide in your simulations. Of course, some types of loss (coupling loss, for example) will not depend on waveguide length and should be kept as “excess loss” so they don’t change when you vary the waveguide length. How you include the loss will depend on what you believe the sources of loss are and what you plan on varying in your simulations.

Please let me know if you have any questions.

Hi, kjohnson, thank you for your detailed explanation. However, I am confused about this loss of the ldf file. If I replace the “loss” in Mode waveguide element by experiment loss, where is the place of the “loss” of ldf file, where is the position of this loss in interconnect?

The loss from the .ldf file data will not appear in the Property View of the MODE waveguide element, but it will be included in the simulation. If you put a value as the “loss” in the Property View (shown below) of a MODE waveguide element, then this loss will be added to the loss in the .ldf file during the simulation.


The same is true for the “excess loss” property, but note that the excess loss is a fixed value with units of dB. The decrease in amplitude due to the “excess loss” will not depend on the “length” property.

Basically, to get the total loss, the formula is: \textrm{total loss} = (\textrm{"loss" property + loss from .ldf file})*\textrm{length} + \textrm{"excess loss" property}

When saving the mode data as an .ldf file in MODE, you also get the chance to overwrite the loss saved in the .ldf file. Where you include any experimental loss is up to you.

Hi,kjohnson, thank you for your detailed explanation. I got the relationship of different loss and how to calculate the total loss.

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.