Learning Forum

[russ44]

Hello, I am studying hybridized modes between slab and buried waveguides.

During the process of setting the FDE simulation region, I noticed that the loss changes drastically depending on the simulation region, even when all other conditions remain the same.

Since most of the loss is caused by slab modes (due to coupling with slab modes), I believe this phenomenon is plausible.

However, at the same time, I suspect it may be an inaccurate result.

Do you have any suggestions on how to resolve this issue?

I would really appreciate it if you could help me.

• I set my simulation region with PML boundaries (default condition). (for hybridized mode with upper slab and lower buried waveguide) 
• If I span simulation region(along y axis) to 5*core width, 6*core width, 7*core width -> I found that the loss changes to 6.3051dB/cm, 4.36dB/cm, 31.9dB/cm.  

[Amrita Pati]

Hi Lee,

The simulation span can have a large impact on the losses. This is because if the simulation span is too close to the core, modal interactions with PML will create artificial losses/gain. So, we always recommend performing convergence testing to confirm the accuray of the simulation. In convergence testing, a series of simulations are run with increasing simulation span. You will see that beyond a certain point (simulation span) the losses start to converge. But your point about the coupling to slab mode also makes sense. Would you be able to tell me what is the wavelength and what is the overall simulation span for let's say 7*core width?

Regards,
Amrita