Simulation can not be completed for using PML Boundaries

[rifat.nazneen]

Thank you for contacting Lumerical support. Before proceeding, please note that this is a public forum and anything you post will be visible to the general public. For users residing in the United States, please ensure that the files and information contained within your posts are not International Traffic in Arms Regulations (ITAR) or export controlled.

To help us solve your issue as quickly as possible, please provide the following information:

1. What device are you trying to simulate? Include diagrams if available.
I am trying to simulate Single bus Silicon nitride ring with 80um radius and width 1.3um. The thickness of the waveguide 400nm.

2. What results are you trying to obtain? Be as specific as possible.
I am trying to sweep different parameters (gap, width) to obtain optimum FSR, high Q-factor from transmission result. I used power and frequency monitor to observe the result.

3. Description of the problem or issue.
Once I simulated the design by using PML boundaries in varFDTD, it has completed up to 71% and then automatically closed the Job Manager box. Even, the simulation completed for 5points (sweep parameters) up to highest 65%. I didn’t get the desired transmission result.

4. Lumerical product and software version.
Mode Solver and Ansys Lumerical 2020 R2.2

5. Please attach your simulation files, including any logs.
80um_SiN_ring.lms (1.0 MB)

[kjohnson]

Hello @rifat.nazneen,

The bus waveguide needs to extend to the PML boundary, other wise the light will be reflected back at the end of the waveguide instead of being absorbed by the PML.

Also, I think this ring is too large to simulate the entire structure in one varFDTD simulation. This is especially true for a Q factor calculation, which can require a very long simulation time. Other approaches for this calculation are discussed in this topic:

Ring resonator in SOI using varFDTD is diverging for longer simulation times, for a standard component

Si_ring_varFDTD_Cband_RLG-30100.3.lms.zip (387.6 KB) Hello, Using Lumerical 2020 R2.2 MODE I am simulating a simple silicon ring resonator for SOI. . Si rib: 0.5 um X 0.22 um, Radius =30um, Gap=0.3um, Coupling length=10 um. For 5ps the simulation finished with status 1. The field monitors on the resonators the field inside cavity hasn’t decayed muchin 5000 fs. The transmission spectrum and Q values of resonances don’t look very realistic too. For longer simulation times it diverges. Pleas…

I would recommend that you consider one of these other approaches for your simulation, I think they will be much more efficient for this type of device. Please let me know if you have any questions.

[rifat.nazneen]

Hi @kjohnson ,

Thank you for your feedback. According to your feedback, I extended the bus waveguide to the PML boundary and it’s working now. I reduced the radius of ring for 50um to get the transmission response in varFDTD simulation. But unfortunately, I am getting ripples on the transmission spectrum.

1. May I know why I am getting these ripples on the spectrum? (See the attachment of the response).
2. How can I fix this problem?
   I followed several threads on increasing simulation time for getting smooth transmission waveform. But when I increased the simulation time, it showed more ripples.
   

I will greatly appreciate if you kindly provide me some effective feedback on this matter.

Thanks.

[kjohnson]

Hello @rifat.nazneen ,

I believe that this is caused by the simulation time being too short. When you decrease the simulation time there will be more ripples, but the amplitude of the ripples will decrease as well. You can see this in this plot of the transmission spectra from the ring resonator example with different simulation times:

This plot is taken from this topic, which discusses this effect: Troubleshooting FDTD Simulations - #3 by kjohnson. The reason for this phenomenon are discussed in this topic: Simulation time and Frequency domain monitors

Because of the large radius, your ring will require a much longer simulation than the times in this plot. I would recommend you continue increasing the simulation time until the simulation runs to the autoshutoff threshold. You can tell if it runs to the autoshutoff threshold by looking at the “status” result of the FDTD solver object. A “status” result of 2 indicates that the simulation ran until the autoshutoff threshold was reached, a “status” of 1 indicates that the simulation ran until the simulation time was reached.

I hope this helps. Let me know if you have any questions.

[rifat.nazneen]

Hi @kjohnson,

Thank you for the feedback. I increased the simulation time upto 160kfs and reached the autoshutoff threshold with narrowband source. When I used broadband, it didn’t reach the autoshutoff level. But the result seems unexpected. I am attaching the response here.

I want to know why the resonance dips are not reaching zero? It shows the transmission power is small and the ring is not providing any resonance. Please have a look and give me your valuable suggestions.
Thanks.

[kjohnson]

Hello @rifat.nazneen ,

I’m not sure why this would be happening. Could you please share some screenshots or your updated file so I could take a look?

[rifat.nazneen]

Hi @kjohnson ,

I am attaching the file here. Please have a look and let me know how can I get the proper resonance peak.
ANT_50um_Narrowband_SiN_ring_0.16umgap.lms (530.6 KB)

[kjohnson]

Hello @rifat.nazneen ,

I don’t see any major issues with the simulation file. What is the expected transmission spectrum? For a single bus ring resonator with low losses I would expect the power transmission to be approximately one. In this simulation the only loss mechanism is the bend loss, which will be low for a radius this large. If this were a double bus ring then I would expect to see clear transmission peaks, but for a single bus ring I believe that this is roughly the expected response. Can you give some more details on the results you are expecting?