Y axis in observation point – reg

[Sridarshini Thirumaran]

can anyone help me regarding this? i have given an input with 1V/m amplitude (gaussian wave) as input and the output i obtained is shown in fig below… can you please help mne in identifying the unit in Y- axis…

Attachments:

1. 

   A.png
   

[Scott Newman]

Hello Sridarshini,

It looks like you are plotting the spectrum from either the main Analyzer window or through the Detector Analysis option under the tools menu. Regardless these are plotting DFT results and as such have the same y-axis units as the time domain which is V/m.

Scott Newman

[Sridarshini Thirumaran]

oh okay. but how come its being 1.1 v/m for an input amplitude of just 1v/m Newman?

[Scott Newman]

First impression would be that the results are not converged. However, without knowing what the simulation is or where that observation is being taken from I am limited in what I can say. Can you please attache the design file? Remember you must zip it before attaching.

[Sridarshini Thirumaran]

yes i will attach the layout. This is the basic layout.

Attachments:

1. test.zip

[Manpreet Singh Dhaliwal]

Dear Sridarshini,

According to my opinion first you will change the min. and max. frequency range at detector side. in optiFDTD Software the output are always show in four different terms i.e. amplitude, phase, real and imaginary. Which option you was choose that was show at the Y axis of graph.

[Scott Newman]

Sridarshini,

Which observation point did you use in presenting the original graph? Neither observation points 1 or 2 should come anywhere near an amplitude of 1.

I stripped out the photonic crystal and added a full XZ observation plane in order to specifically measure the input plane.

What you are looking at is the effect of how the input plane is initialized and your used of a 2D transverse Gaussian profile with a width on the order of the wavelength. If you increase the width of your Gaussian by a factor of 2 or 3 you will see that the input plane’s amplitude will be one.

Scott

[Sridarshini Thirumaran]

Newman, as I said that the output is greater than input provided, here is an example. i have attached the layout , where amplitude is 1V/m and output which i am getting is greater than 1. Kindly help me out in sorting this issue.

Attachments:

1. no-of-coupling-rods.zip

[Sridarshini Thirumaran]

Thank you manpreet singh. But thats what iam not aware of…

[Scott Newman]

Manpreet, the y-axis is not dependent upon what frequency range you choose. While you are correct that Observation Areas give you the ability to see the results 4 different ways, the Observation Points do not. What you see when you are looking at an observation point is the amplitude as a function of time or frequency depending on what you ask for.

[Scott Newman]

Sridarshini,

You did not understand my original response as your new design has effectively all the same issues as the first in terms of testing for an amplitude of 1 from your input plane.

1) You cannot check the amplitude of the signal at an observation point on the output of a photonic crystal and expect it to be the same as the input, therefore you are not testing to see if the input plane has an amplitude of 1. Photonic crystals have loss, they can also focus a beam (one of the possible reasons you are getting an amplitude greater than one). The best way to check is to delete your photonic crystal, create an Observation Area and see the amplitude, see ObservationAreaNotParaxial.png. See point 2 why you are seeing 1.08 in that figure.

2) You did not correct the issue with the Gaussian source. As stated previously the Gaussian source operates under the paraxial approximation which requires that the width of the Gaussian distribution be much greater than your source. Your source is 1.55um and your width is set to 1.44um, this does not satisfy that condition which results in an amplitude higher than one, see ObservationAReaNotParaxial.png.

3) You are not running with a high enough resolution. While your spatial mesh is acceptable for your wavelength in the material, you did not take into consideration the sizes of your features. Your rods have a radius of 0.1um but your resolution is 0.044um, this means that some of your rods will only be defined by 4, maybe 5 grid points. In designer take a look at your 2D Refr_Idx-Re(y) tab and see how irregular your rods are, see attachment discretizedStructure.png. With higher resolutions the rods are better defined and the values you are looking at drop off.

Scott

Attachments:

1. 

   discretizedStructure.png
   

2. 

   ObservationAreaNotParaxial.png
   

[Sridarshini Thirumaran]

Oh thanks a lot Newman I will look into that again.

[Author]

Posts