Learning Forum

[Anish_Mondal]

Hello,

I have applied the surface roughness suggested by Lumerical in the link given below. I want to measure the propagation loss in the waveguide due to surface roughness for a wavelength of 1.55um.[sigma = 2nm, correlation length (Lc) = 50nm.]

https://support.lumerical.com/hc/en-us/articles/360034382374-Tips-for-adding-surface-roughness-to-structures

I believe that I am using sufficient meshing for the simulation. I have placed 5 DFT monitors at uniform intervals to measure the power in the core of the waveguide passing through them.

[Greg Baethge]

Thank you for posting your question on the forum. Modelling loss due to surface roughness is quite challenging: it requires a mesh fine enough to resolve the roughness, and a propagation length long enough so its effects are noticeable (the loss should be larger than the numerical noise in the simulation).
First, I would run some convergence testing to make sure the simulation settings are ok. The other thing to look at, the DFT monitors will record the power flowing through them. This will not necessarily reflect the actual loss, as part of the power will not be in the waveguide's mode. For that reason, I would use a mode expansion monitor so you can get the actual coupling to the mode supported by the waveguide.