Learning Forum

[Aaron Schreyer-Miller]

Hello,

I am having an issue getting the ansys example script to work. I do not trust that the dispersion calculation is being done correctly because I get a different figure of merit than when I plot manually. 

When I manually click frequency sweep in MODE I get the correct dispersion curve, however using the script provided in the Ansys examples I get a completly different curve.

Thanks.

[Aaron Schreyer-Miller]

I believe this is an issue with the way the code is calculating the GVD. I am unsure why there is a difference between clicking the frequency sweep manually in the MODE GUI and running the example provided script.

[Aaron Schreyer-Miller]

Solution: Call the frequency sweep function
frequencysweep - Script command – Ansys Optics

switchtolayout;

findmodes;
selectmode(1);
setanalysis("track selected mode",1);
setanalysis("detailed dispersion calculation",1);

frequencysweep;
D=getdata("frequencysweep","D");
f=getdata("frequencysweep","f_D");

plot(c/f*1e6,D*1e6,"Wavelength (um)", "Dispersion (ps/nm/km)");


For some reason this is the only way to properly compute the dispersion in Lumerical. I hope this helps others that have the same issue.

[Guilin Sun]

Thank you for your effort on this topic. The script was developed many years ago which simply uses difference to get the 2nd order derivation. Later FDE has been improved a lot. For example, even at a single frequency point, it now can give dispersion result. You are right that it is better to use the built-in frequency sweep.