Dual Drive Mach Zehnder Modulator

interconnect
needattention

#1

Hi all,
I am trying to design a dual drive MZM using some measured data. I am trying to get simulation result using the following attached file. I have a couple of questions regarding the simulation.

Question 1
As per the measurement data, if I select the device length near 550 um, the dual drive MZM should provide approximately 10 to 12 dB of ER, which is not shown by the simulation result. Can you please let me know where the error in the simulation is?

Question 2
Similarly for the two-tone test to calculate the nonlinearity of the MZM, how can I achieve the result output IFoutput vs RF drive voltage for fundamental, second-harmonic and IMD? What should be the criteria to choose the frequency of the two RF signals?

Note: For the two-tone test refer to the papers:

  1. B. Kolner, D. W. Dolfi, Intermodulation distortion and compression in an integrated electrooptic modulator, Applied Optics, vol. 26, no.1 pp.3676-3680, September 1987.

  2. Y. Zhou et al., Linearity Measurement and Pulse Amplitude Modulation in a Silicon Single-Drive Push-Pull Mach-Zehnder Modulator, Journal of Lightwave Technology, vol. 34, no. 14, pp. 3323-3329, July 2016.


#2

Single_Drive_Push_Pull_MZM.icp (1.4 MB)

Sorry for not attaching the simulation file previously.


#3

Hi @subhradeep.pee13,

Sorry for the late reply. For this MZM design, are you comparing its simulation result to the experiment result? Please let me know how you get the approximate ER for this design.

For the second question, we don’t have access to the provided reference. Is it possible to upload the PDFs here?


#4

Hi @gwang,

No I am not considering the experimental result. Static ER of the modulator can be estimated using the analytical formulation. For a single drive push-pull configuration, the transfer function can be written as,
CodeCogsEqn%20(1)

Normalized TF can be estimated as,
CodeCogsEqn%20(2)

So ER can be estimated using the relation if the values of loss, length and additional phase shift is known. By this method if the device length is chosen to be near 550 um, the modulator should provide a ER of 10 dB approximately which is not obtained from the Interconnect solution.

Regarding the dual-tone test, I am uploading some papers which I had mentioned earlier.

Thanks in advance.
ao-26-17-3676.pdf (616.7 KB)
07468539.pdf (677.2 KB)