Vernier TLD with interconnect

ansyslumericalidansyslumericalid Member Posts: 1


I have tried to calculate the Vernier TLD(Ring Vernier laser – Lumerical Support), have suffered from some counter-intuitive results.

I have focused to the effects of the racetrack condition to the Vernier tuning of TLD.

I'm sorry not to show my screenshot of TLD circuit here, because our workplace prohibits any non-permitted upload to internet.

Instead, I'll briefly explain my some modification.

  1. center wavelength = 1.31um
  2. left ring circumference : 2*157.4 = 2*(1.31*120)
  3. right ring circumference : 2*162.44 = 2*(1.31*124)
  4. half length of the racetrack (exclude ring) : 800+228.8 = 1.31*960
  5. length of soa : 800
  6. ALL neff and ng, including twlm elements : neff 3, ng 3
  7. all coupling coefficients : 0.01

My critical question is,

  1. Is setting delay compensation to waveguides on racetrack necessary?
  2. May it be critical? what number is correct?

As I see, delay compensations are applied to the waveguides in ring mirrors, but are not applied to the waveguides in racetrack, in above provided example.

And I also know, it's known well that fractional delay compensation is necessary to make the results between time-independent calculation and time dependent calculation of, at least, passive circuits coincide.(FSR with cavity delay compensation in Transient Sample Mode simulator – Lumerical Support)

The critical problem suffered me is, TLD peak intensity at 1.31um(the combs of two rings peaks coincide at exactly 1.31um) with in-phase racetrack condition(above racetrack length) is rather less than it with out-of phase racetrack condition.

(in-phase condition : <0dBm. out-of phase condition : ~20dBm, ???)

Please, please, please help me.

Thanks, thanks, thank you.


  • GWANGGWANG Posts: 83Ansys Employee

    Hi ansyslumericalid,

    Thank you for posting here. If the waveguide elements are not in feedback loops, then the delay compensation usually is not critical, but still good to have. The total number of delay compensation should add up to the total number of connections in the loop. It is hard to say what's the problem without the file, but I'd recommend testing the external feedback link with an Optical Network Analyzer using the "impulse response" in time domain to make sure it mataches the result in frequency domain. Please give this a try and let me know what you find.

Sign In or Register to comment.