Wrong time delay of optical fibre in INTERCONNECT

hcunmphcunmp Member Posts: 2

I use a 1.7km optical fibre to propagate light whose wavelength is 1546nm using INTERCONNECT. This fibre can provide about 8us delay theoretically, but the practical time delay is only around 0.1ns in my project. And a warning is: "FIBER_1: Warning: propagation delay at frequency '1.931e+14' Hz is '-3.03032e-10' s, this value should be greater than or equal to zero."

This is a circuit for checking.


Why did this error happen and how can I resolve it?

Thanks!

Best Answer

  • GWANGGWANG Posts: 81Ansys Employee
    Accepted Answer

    Hi hcunmp,


    For that I think a Waveguide model would be more proper to use. The delay the fiber model adds is based on the number of taps used for the digital filter in the model but not the actual physical delay in the fiber. The waveguide element however adds the actual delay to the simulation when setup properly.

Answers

  • GWANGGWANG Posts: 81Ansys Employee

    Hi hcunmp,


    Thank you for the post. Since we cannot share the file here, it is a bit hard to pin point the problem. One thing you can try is to revert the sign of the dispersion in the fiber to get rid of the negative delay. The way we define dispersion in the fiber might be the reverse of the way you define it. Please give this a try and let me know whether you see more reasonable result this time.

  • hcunmphcunmp Posts: 6Member

    Thanks for your answer. This is the property of optical fibre. The delay time did not change whenever the dispersion is 16 or -16. Are there any other properties will influence that?

    The pictures below show a new checking circuit, some of its settings and the result. Could you tell me where I did wrong?



  • GWANGGWANG Posts: 81Ansys Employee

    Hi hcunmp,


    Thanks for the details. The first circuit is the right one to be used for the test of the pulse broadening, and here are my settings for the time domain simulation parameters and fiber parameters:

    Note that the fiber model only captures the dispersion effects in the fiber but not the actual time delay, since the effective/group indices values are not required for this model, and it is usually used for long haul communication (which would have long delay than simulation time window). Is what you want to measure the pulse broadening effect?

  • hcunmphcunmp Posts: 6Member

    Actually, what I want to see is only the time delay of signal.

    I want to confirm if you mean that I cannot see an obvious time delay by comparing the input signal and output signal of the fibre just like this graph (this is a result that I replaced the optical fiber with the element "optical delay" in the second circuit).


  • hcunmphcunmp Posts: 6Member

    I understand. Thank you!

Sign In or Register to comment.