Learning Forum

[Jiang Xiong]

Esteemed Ansys technicians,

We are currently studying the wave propagation behavior in a dielectric slab, when a monochromatic wave (a sinasoidal wave with a single frequency) is incident from the left (input port). It seems in Lumerical, only Gaussian pulse with a finite spectrum, no matter how narrow, is allowed? Could you please kindly let us know the correct setting, if there is? I guess there is not any theoretical conflict between a monochromatic source and the FDTD algorithm itself. Otherwise, is there any alternative solution, if the monochromatic properties are exclusively concerned in our specific case.

We are looking forward to your suggestions and many thanks!

[Guilin Sun]

In general, FDTD uses pulse as source signal, which is more efficient than a sinusoidal wave. Please refer this post: 

Ansys Insight: FDTD 为什么采用脉冲光源激励而不采用正弦波光源

you can set a relatively narrow spectrum (longer pulse), and only record the single frequency in the frequency-domain monitor.

If you really want a single wavelengh injection, you can set a cw source: 

1.   FDTD CW source

but you will not be able to use frequency domain monitor, since the Fourier transform requires the signal to decay completely.

In summary, you can simply use a relatively narrow spectrum to get the monochromatic result in CW, steady state from DFT monitor.

[Jiang Xiong]

Thanks a lot for the suggestion!

We are now trying the narrow spetrum to perform the simulation.

Meanwhile, when I was checking the CW setting via the website link you provided, it seems the "CW source page" in the article entitled "1D cavity laser using 4-level 2-electron material" would direct to a wrong page:

https://forum.ansys.com/?p=/categories/photonicst/single-frequency-cw-source/34426

Would you please help check and let us see the correct page?

Thanks again for your patience!

[Guilin Sun]

I am not sure why it happens but I can successfully land to that page: https://optics.ansys.com/hc/en-us/articles/360042192813-1D-cavity-laser-using-4-level-2-electron-material

Please note that narrow spectrum means longer pulse, which will increase the simulation time. To get accurate DFT result from the frequency domain monitor, longer simulation time is required.

The CW source cannot give you accurate DFT result, since the signal lasts.

[Jiang Xiong]

Yes, the website you mentioned is indeed available, but the "CW source" page it suggests to refer is not (please see the blue link in that page in the following screen shot) .

Thanks also for your reminding on the simulation time and I quite agree with your comment. In fact, currently we are more concerned about the time-domain behavior of the signal.

[Guilin Sun]

Thank you for pointing out that link. I thought it was the above "FDTD cw source" which was my file link. the "cw source" page is here https://optics.ansys.com/hc/en-us/articles/360042192813-1D-cavity-laser-using-4-level-2-electron-material.

[Jiang Xiong]

Thanks Guilin, but I guess you accidentally copied the wrong website address of the "CW source" page. The address is identical to that in the previous reply, which still directs to the article "1D cavity laser using 4-level 2-electron material".

So would you please help offer the correct "CW source" page again? Many thanks and I'm really sorry for the inconvenience.

[Guilin Sun]

You are right. The forum post is here  https://forum.ansys.com/forums/topic/ansys-insight-single-frequency-cw-source/

[Jiang Xiong]

Yes, we see the post now. It would help us a lot.

Thanks so much for your patience and quick responses all the time!

[xudong Zhang]

I'm also interested in this setup and I have learned lots about how to create a custom source time signal.But it seems that my "users input Setting"is unavailable.Could you please provide more detailed steps on how to import custom signals?

[Guilin Sun]

Please refer the script file in the example. You need to use script to load the signal.

In many many cases it is absolutely no need to use monochromatic light source. Please think it again, and read above replies.

The only cases are the listed examples for few nonlinear simulation, or you only need to know time domain result, without separating the amplitude and phase, as the simulation is at steady state which records instanteneous fields from a time monitor. You cannot use frequency domain monitors, as it violates the Fourier tranform theorem.

 

[xudong Zhang]

OK, I'll try again. Thank you for your timely reply！