Learning Forum

[Sirui Tang]

Hello,

I am currently simulating a fiber-to-chip coupler with free space in the middle. To calculate coupling efficiency, my plan is to save the monitor d-card data from the FDTD simulation and then load it into MODE for overlapping with the fiber mode. I've chosen this approach because MODE allows for position optimization during the overlap, eliminating the need for an additional parameter of fiber position when using FDTD sweep with a mode expansion monitor or with ports and S-parameter. However, I found that the saved d-card data are in a non-normalized form, whereas the fiber mode is in normalized form, resulting in values differing by tens of magnitudes. Is there a way to save normalized data from the FDTD d-card, or alternatively, can I edit the results in the MODE d-card?

Thank you in advance for the help.

[Amrita Pati]

Hi Sirui,

I apologize for the late reply. It is correct that in FDE, the field results are normalized to the peak intensity (|E|^2). But I believe this normalization should not affect the overlap result. If you look at the equation used for the calculation of overlap,

Let's say E1 and H1 are the results from the FDE. If they are normalized by a constant (peak |E|^2), it will cancel out in the numerator and the denominator. I ran a test to double-check this. I collected the field results from a DFT monitor in FDTD simulation of a fiber. Then I calculated it's overlap with the fundamental TM mode of the fiber in FDE. As you can see, there is a large difference in the values of the field results due to the normalization. But there is still a pretty high overlap as it is the fundamental mode of the fiber. 

Are you sure that the position of the fiber in the FDTD and FDE simulation is accurate?

Regards,
Amrita