Learning Forum

[notyourphoton]

Hello,
I am new to Lumerical, and I am using FDTD to simulate a FP cavity. Can someone please guide me how can I plot transmission as a function of source power? Source can be a plane wave or a Gaussian wave. (I don't want to use interconnect, I am using FDTD/STACK package of Lumerical).
I have tried to use parametric sweep, but I couldn't figure out a way to plot transmission as a function of input power. Any help would be highly appreciated.

[Guilin Sun]

For linear optics Lumerical FDTD gives normalized result to remove the source-frequency dependency: Understanding frequency domain CW normalization
This is because the resulting output power is linearly proportional to the input source power. Therefore it is not necessary to plot such relation. It is only useful when simulating non-linear physics.
Transmission is usually given by frequency dependency, eg, different frequency can have different transmission due to the device performance, or light-matter interaction.

[notyourphoton]

Thank you for getting back to me on this. The FP cavity that I am trying to simulate has a nonlinear layer embedded between two mirrors. So, it is a nonlinear problem. Can you please suggest if there is a way to plot transmission as input source power. I do have plots of transmission as a function of frequency. Thank you.

[Guilin Sun]

In this case, you can use the frequency-domain unnormalized transmission: transmission*sourcepower (with the nonlinear frequency ) since transmission is always normalized transmission - Script command
You can also calculate the output power by integrating the Poynying vector from the monitor, which is equivalent to the above method.
However, such power is not the actual physical power that you can measure, as the frequency-domain result is calculated by Fourier transformation as shown in the previous link. A convenient method might be to compare the relative output power related to the first output power from your sweep. You may also get the max power in the time domain from the time domain monitor, with proper filtering to remove the portion originated from the fundamental frequency or initial injection. This will need you to do the math with scripting.

[notyourphoton]

Thank you so much for the detailed response. The problem is that both power and transmission are a function of frequency. And yes we can take those two and plot against each other. But we want to plot transmission as a function of input(output) power at a single frequency, then for another different frequency. In 2D this will be a plot of transmission against frequency and power (input/output). Can you pleaser share any example, if possible? Thank you.

[Guilin Sun]

You can have an image: two variables as frequency and input power, and the color of the image shows the strength of the output power. image - Script command ÔÇô Lumerical Support
For 2D plot, you just choose the index of the frequency to be plotted, and then plot the curve. You can plot several frequency points at the same time, eg, plot(x,z(:,1:2)); to plot two y points in the same figure, where z is a 2D function of x and y. Please try, and modify your script if there is error massage. The dimensions should match in plot.