Learning Forum

[lgutt]

In some simulations I'm running, I need the mode profiles found with the FDE solver in MODE to be returned in V/m. According to the documentation, this is also how the field is returned, but all results I get are clearly self normalized such that all mode profiles have a peak field value of 1.

There is an option in MODE to choose between different types of normalization under Settings -> Normalization state where "no normalization" can be chosen, but this doesn't change what is returned from MODE.

So: does anyone know how to get MODE to return mode profiles in units of V/m (and preferably with the mode itself being normalized to a power of 1 W/m^2)

[Guilin Sun]

 

What is purpose for mode profile E not in 1 v/m?  FDE solves the eigen values of the discrete, linear Maxwell equations, and it is in a cross section without propagation. Therefore it is common pratice to have its max amplitude to be 1v/m. You can multiply a factor to all field components to get want you want.

In EME, the modes are normalized to have a power of 1w.

Please step back and think it over about your purpose. 

 

[lgutt]

Fair enough.

To be more specific, I'm calculating some coupling coefficients that depend on the overlap between different modes in a waveguide. In this case, the modes are assumed to be orthonormal to each other. The modes are only orthonormal if they are individually normalized to contain the same power (typically assumed to be 1 W/m^2). This isn't the case if, for instance, both the TE0 and TE1 mode have a peak electric field of 1 V/m.
The modes can be renormalized by taking the integral of the absolute square of the field of each mode individually and finding a factor that renormalizes them to 1 W/m^2. Due to that, it would be convenient for me if there was a way of returning the found modes where each of them contained 1 W/m^2.

I guess I should rephrase my question as:

Is it possible to get MODE to return found modes such that they all are normalized to 1 W/m^2 (or normalized to some arbitrary power) insted of being selfnormalized to 1 V/m at peak?

[Guilin Sun]

If you simply want to get overlap coupling, you do not need to normalize:

https://optics.ansys.com/hc/en-us/articles/360034405254-overlap-Script-command

as replied previously you can normalize to whatever you want.

[lgutt]

Thanks, but that's not what I'm trying to do. What I'm looking at is how the coupling efficient between two modes (TE0 and TE1)  behaves when a perturbation is introduced and TE0 can couple to TE1.

[Guilin Sun]

Coupling can be estimated thrugh overlap: https://optics.ansys.com/hc/en-us/articles/360034405254-overlap-Script-command  no need field normalization.

If both TE0 and TE1 are from the same waveguide without perturbation, then there will be no overlap or coupling since they are orthogonal.

[lgutt]

Well, as I just said above I'm interested in the case where there is a perturbation introduced which results in a nonzero overlap. If you can find the overlap between modes with a perturbation included, then I will happily use the overlap tool. If not, I still hope you can answer whether it's possible to return mode profiles normalized to power instead of to 1.

[Guilin Sun]

Those are two different issues:

You can modify the mode fields to a power of 1W: Simply integrate the current mode profile of its Poynting vector and the result is its power. Please refer https://optics.ansys.com/hc/en-us/articles/360034405354-transmission-Script-command

Suppose the power is p0. Then you can normalize the fields as

E,Hxyz=E,Hxyz/sqrt(p0).

This should be easy to do.

The other is the so called "perturbation".  As long as you can get the perturbed mode profile, you can do an overlap.

Please note that, FDE gives mode profile for a waveguide assuming infinitely long, if there is no bending.

So both are straight forward. Please try to do the script and let me know if you have questions on scripting. Now we are circling on theoretical desciption.

[lgutt]

Thank you, but I think we are misunderstanding each other. If you look at my earlier posts above you can see that I am renormalizing the fields in exactly the way you are mentioning (just by integrating the fields that are returned from MODE in python).

It's not so much that I don't know how to normalize the fields to p0, I was just wondering if there was a simple option to return the modes normalized in this way in MODE as a matter of convenience. In any case it seems that the easiest way is what you mention above, but in that case I might as well do it in python.

Thank you for your help though.

 

[Guilin Sun]

You are right that FDE does not normalize field profile with power. Instead it is normalized to max(E)==1v/m. For other normalization you have do to your self.