Learning Forum

[krishnanunnira.20]

I have tried transient analysis in charge solver and using the data to plot refractive index change as a function of time in MODE solver. But  I am not able to obtain the index variation correctly. It will be helpful if someone can share the script file for obtaining refractive index variation from MODE solver for transient simulation.

[Guilin Sun]

When you mentioned "MODE solver", do you mean varFDTD? it is the only solver in MODE in time-domain. But it takes the steady-state charge induced index change in simulation since it is a deterministic solver. Only the refractive change with time in non-linear materials can be included but right not the index variation with time is not accessible. If it is this solver, and you want to access the index variation data, please file a feature result by referring to this post: 

New Feature vote:   Vote new features, and file your feature request

MODE also has FDE solver. Usually FDE takes charge data at CHARGE steady state. Thus it will  not have time-varying capability. However FDE can sweep the voltages from CHARGE data. https://optics.ansys.com/hc/en-us/articles/360042328674-PN-depletion-phase-shifter

Please elaborate more if the above replies do not solve the issue.

[krishnanunnira.20]

MODE solver, I mean FDE solver. In FDE solver if time varying capability is not possible it will be difficult to obtain the speed of the modulator. Is it possible to obtain charge carriers as a function of time from CHARGE solver itself, then I can calculate the corresponding refractive index change through direct formula. Can you share me a script to obtain charge carrier data from CHARGE solver as a function of time (since transient simulation will work in CHARGE).

[krishnanunnira.20]

Thank you for your earlier reply. Is it possible to obtain carrier injection data at the centre point of the waveguide after simulation. Normally charge monitor save the data as .mat file and we are using the same in FDE solver. If the carrier data is directly obtained (centre point of waveguide) then I can calculate the refractive index change using formulas.Can you please share a script file to get carrier injection data from CHARGE solver. 

[Guilin Sun]

Unfortunately it is not like that. It is the applied voltage change that causes the index variation. At certain voltage, the changes can only induce certain change of the index. What you need it to sweep voltage, not time. Please refer to examples online:
https://optics.ansys.com/hc/en-us/articles/360042327854-PIN-Mach-Zehnder-modulator

https://optics.ansys.com/hc/en-us/articles/360042322794-Ring-Modulator

https://optics.ansys.com/hc/en-us/articles/360042328774-Traveling-Wave-Mach-Zehnder-Modulator

[krishnanunnira.20]

But in a transient simulation for each time period there is a corresponding voltage change. Since voltage is changing there will be refractive index change, indirectly it is measuring effective index change as a function of voltage (even though we are give time variation as sweep). Through steady state we can measure the maximum index change but speed of the device can be measured only by using transient simulation.

[Guilin Sun]

"in a transient simulation for each time period there is a corresponding voltage change"

Do you mean the voltage applied is a continuous function of time? it is rare. Usually it is a shuter and the voltage does not continuously change.

We only have examples for CHARGE attribute as a function of voltage, not time. Transient simulation is mostly for bandwidth and other dynamic calculations, not for charge distribution. You may need to step back, and think over what exactly do you want. 

For modulator  it does not use continuous voltage with time. Please refer to the related examples provided previously.

[krishnanunnira.20]

Hi,

 Thank you for our reply. In my case I need to calculate the rise time required for the device. So if voltage is not continuous then I cannot use the transient analysis in CHARGE solver for my application. 

Is there any method to calculate rise time and fall time of a PIN device in Lumerical other than interconnect.

[Guilin Sun]

It is only possible to use CHARGE if INTERCONNECT is not available to you. I would strongly suggest you to follow the online examples. You can request your boss purchasing a license for the tools you need as this is the correct way to progress your project.

[Guilin Sun]

A little more info:

Optical frequency is much faster than the modulator for electrical signal.

transient simulation in CHARGE is to simulate the dynamic behaviors. Thus the rise time can be calculated in CHARGE, not in FDE, as optically there is no delay from the voltage to the index change.

Modulator speed, bandwidth and rise time can be done in CHARGE with some analysis.

Please step back and think it over. In addition, please search for some references and see how did they do such simulations to get the desired results.

[krishnanunnira.20]

Hi,

  Actually you didn't understand my problem correctly. 

In charge solver when we give voltage a sudden change in carrier concentration occurs and the carrier concentration will reach the peak value only after a certain period (rise time) and inorder to measure rise time we need to give a step voltage which is a function of time and in charge solver the only option available to give step voltage in such a manner is through transient simulation. The purpose of transient simulation is to study the systems behaviour over time.

In charge, transient method is available that is why i asked doubt regarding that and you told using the transient option  in charge it is not possible to get continuously varying voltage.

Thank you for your reply.

Also we have a licence for all the ansys solvers including interconnect and from interconnect it is only possible to get the eye diagram and not the rise time that is why i request you to suggest any other method.

 

[Guilin Sun]

Please check the transient simulation in CHARGE where it usually applies a step function to get the rise time. Even if you have a triangle pulse voltage applied, at each value of voltage, the index change induced is the same as at the steady state of the same voltage, since from voltage change to index change it is instant, compared with the electric response. therefore no need to get the index change at each voltage, and it is not the optics that affects the modulator speed. it is the electric signal.

If you want to know the index change at each voltage, you can simulate the steady state.

Your feature reuqest has been discussed already.