Learning Forum

[APat]

Dear Tech Support Team,

I have an intensity beam profile from a microspectrophotometer that I would like to replicate for running optical simulations using FDTD. Can I simply input that beam profile using import source tab or would I require more detailed information. The intensity beam profile that I have is a simple XY plot of intensity as a function of wavelength from 400 to 800 nm. I don't have spatial E field or H field numbers. I greatly appreciate your help in this regards.

Thanking you,

Best regards,

Anvay.

[Greg Baethge]

Thank you for posting your question on the forum. The import source requires you to load the field information, in the form of a dataset:
EM = rectilineardataset("EM fields",x,y,z);
EM.addparameter("lambda",c/f,"f",f); # Optional
EM.addattribute("E",Ex,Ey,Ez);
EM.addattribute("H",Hx,Hy,Hz);# Optional
You need to specify at least the electric field.
In your case, since you only know the beam intensity profile, you will have to make some assumption. For instance, you can consider that the source is polarized along x. If we call B the beam profile, a matrix with the beam vs x, y, z and lambda, we can define:
Ex = sqrt(B/max(B));
Ey = 0;
Ez = 0;
Note I normalize the field amplitude so abs(E)^2=1.

[APat]

Dear Thank you for your prompt response. This was helpful. I had a follow-up on my previous question. The lamp profile of the microspectrophotometer that I am referring to is unpolarized. In such a scenario, how do I consider the polarization of the source along the X, Y and Z? Will the assumption that you made earlier still be used? For your reference, I have attached the beam profile below. Do you think I can still use the procedural steps that you mentioned previously for this dataset?
Thanking you once again Best regards Anvay.

[Greg Baethge]

You're welcome, The thing is, FDTD is intrinsically polarized, so you have to define a polarization for your source. That said, it is possible to get the results for an unpolarized source. You will have to run 2 simulations, with each orthogonal polarizations. So, for a propagation along z, you can run 1 simulation with a source polarized in x, and 1 simulation with the source polarized in y.
Then you can sum incoherently (eg. you sum the field intensities) to get the unpolarized, incoherent results. See this post for more information.

[APat]

Dear Thank you again! I believe I understand the point you are trying to make. I can define the input source once as s-polarized and other as p-polarized, just like we do for a Plane Wave/Gaussian source. This helps a lot.
Best regards Anvay.

[APat]

Dear In the command script "EM = rectilineardataset("EM fields",x,y,z)", what should I define the values of x, y, and z. As I mentioned in the previous correspondences, I only have a lamp beam intensity profile as a function of wavelength. I do not have any spatial data for the same. What's the simplistic assumption I can work with in this case?
Best regards Anvay.

[APat]

Dearand others I tried creating a custom source using the lamp beam profile from the microspectrophotometer. Apparently, I am getting the following error and I am not sure where the issue is. When I do this exercise for a single wavelength, I do not have any problem. However, once I include all the wavelengths from 400 to 800 nm, I get the error message. I have also attached the .txt file below from where I am reading the data. Could you please help me in this regards?
Thanking you in advance Best regards Anvay.

[Greg Baethge]

When creating an import source, you need to specify a spatial field profile. Depending on the propagation axis, the field profile (that is 2D) will be defined on the xy, xz or yz plane. In a rectilinear dataset, x,y,z correspond to the coordinate vectors for x, y and z where the data is recorded. For a source defined in xy plane, x and y will be vectors and z a single value.
In your case, you don't really need a import source, actually. You can use the source the most relevant for your application (plane wave or Gaussian beam, for instance), set the bandwidth to match the spectral data you have. Then you can apply this spectrum in a post-processing step. SInce FDTD will give the impulse response of the system, you can simply apply the spectrum to the results (we use this to apply the solar spectrum in solar cell simulations).
I hope this will help.