Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Effective index calculation

Please login with a confirmed email address before reporting spam

Hi All,

I am trying to simulate a structure in a paper where I need to use hybrid mode eigenvalue calculations for neff (RF Module). For a series of values, each time I calculate effective indices, then taking one of the solution I need to integrate some areas. But I cannot obtain the same figure in the paper. The problem is how can I know which solution (which neff) give me the right results. What are the criteria for selecting the right effective index?

Any help would be greatly appreciated.

Thanks in advance,
Elif

4 Replies Last Post 8 mag 2010, 13:15 GMT-4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 7 mag 2010, 13:51 GMT-4
What do you mean by taking figure on the paper?

Anyhow, there is no criteria in general to perfectly find out the right mode just by the numerical value of its effective index. Depending upon your circumstances, special rules might be made up though.

Sometimes when I have to compute the dispersion relation of a particular mode, I spot out the effective index of desired mode at two close/consecutive wavelengths and then use a linear prediction scheme to estimate effective index value for the following/preceding wavelength. Using the fact that dispersion curves are continuous, I compute the difference between the estimated effective index with effective indices of computed modes and pick out the one which minimizes the absolute error. However, it might give wrong answers sometime again depending upon the geometry you are using so manual verification might be desirable to ensure you have what you need.
What do you mean by taking figure on the paper? Anyhow, there is no criteria in general to perfectly find out the right mode just by the numerical value of its effective index. Depending upon your circumstances, special rules might be made up though. Sometimes when I have to compute the dispersion relation of a particular mode, I spot out the effective index of desired mode at two close/consecutive wavelengths and then use a linear prediction scheme to estimate effective index value for the following/preceding wavelength. Using the fact that dispersion curves are continuous, I compute the difference between the estimated effective index with effective indices of computed modes and pick out the one which minimizes the absolute error. However, it might give wrong answers sometime again depending upon the geometry you are using so manual verification might be desirable to ensure you have what you need.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 7 mag 2010, 17:26 GMT-4
Hi,

Thanks for the answer. Sorry that I didn't make it clear. What I am trying to do is re simulating a structure from one of the paper. I try to do the same thing what they did in the paper and get the same figure. For that I use eigenmode analysis in the hybrid-mode. I've set the ''search effective mode indices around' part, calculated number of neffs. By changing the some parameters in the structure, each time after calculations I have another set of effective indices. My question starts here how can I guess at each calculation which effective index is the right one. For my case (as far as I know) there is no other way such as dispersion curve to compare and estimate the next one.





Hi, Thanks for the answer. Sorry that I didn't make it clear. What I am trying to do is re simulating a structure from one of the paper. I try to do the same thing what they did in the paper and get the same figure. For that I use eigenmode analysis in the hybrid-mode. I've set the ''search effective mode indices around' part, calculated number of neffs. By changing the some parameters in the structure, each time after calculations I have another set of effective indices. My question starts here how can I guess at each calculation which effective index is the right one. For my case (as far as I know) there is no other way such as dispersion curve to compare and estimate the next one.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 8 mag 2010, 01:39 GMT-4
If I may guess, changing only one parameter at a time by a small margin would produce only small difference in effective index. This is then similar to my dispersion thing, I change one parameter (lambda) by a small difference and based on the previous two results make a guess using linear extrapolation that the newly computed mode would also have its index somewhere close.

If you are, however, changing multiple parameters simulataneously then I am afraid this trick cannot help your cause. May be looking at the exact problem from a theoretical angle would reveal some useful insights on the pattern of change.

All the best!
If I may guess, changing only one parameter at a time by a small margin would produce only small difference in effective index. This is then similar to my dispersion thing, I change one parameter (lambda) by a small difference and based on the previous two results make a guess using linear extrapolation that the newly computed mode would also have its index somewhere close. If you are, however, changing multiple parameters simulataneously then I am afraid this trick cannot help your cause. May be looking at the exact problem from a theoretical angle would reveal some useful insights on the pattern of change. All the best!

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 8 mag 2010, 13:15 GMT-4
Thanks very much for the answer Shakeeb. I'll dig for answers from a theoretical point of view as you suggested. Maybe I can find a way for estimation.

By the way, I would be grateful for any other comments and ideas.

Cheers,
Elif
Thanks very much for the answer Shakeeb. I'll dig for answers from a theoretical point of view as you suggested. Maybe I can find a way for estimation. By the way, I would be grateful for any other comments and ideas. Cheers, Elif

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.