Learning Forum

[ensiye97]

I’ve been simulating a TiN heater placed on top of a PN waveguide to obtain the thermal time constant using transient simulations. I applied power to the heater and monitored the temperature response in the waveguide.

In my measurements, the thermal response is asymmetric with a rise time of 15 µs and a significantly longer fall time of 65 µs. However, in the simulation, both the rise and fall times are symmetric at around 15 µs.

I’ve been trying to understand the cause of this discrepancy between the simulation and the real-world measurement, but I’m having trouble figuring it out. What factors could be contributing to the symmetric response in the simulation, and how might I adjust the model to better reflect the asymmetry seen in the measurements?

[kghaffari]

Hi.

This is an interesting question. I don’t remember making this comparison in HEAT myself before (the rise time vs the fall time); however, I would like to clarify few items in case I can make recommendations:

1. How do you measure the fall time? Do you have the shutter step on and step off functions applied to the source, in the same simulation? I think this would be necessary to ensure that with the initial step on, the system reaches steady-state temperature before turning the shutter off to measure the fall time.
2. How do you make your measurements? Specifically, where in the device do you read the measurement from? As you it’s know it’s important to make sure the data from the simulation and measurement correspond to the same region. 
3. Can you share a high level schematic of your device (e.g. the geometry in HEAT)? And outline the boundary condition(s) applied? Boundary conditions in HEAT are critical to the setup and obtained results; I would like to see the setup in case I can make recommendations.
4. Finally, would you be able to share the screenshot of your transient tab in HEAT?

Best regards,

Khash

[ensiye97]

 

Hi Khash,

Thank you for your response. I’ve attached a PowerPoint and my simulation file for your reference. In my simulation, I don’t use a shutter step for on/off; instead, I use the transient simulation by adjusting the time on/off as shown in the figs in the PowerPoint https://drive.google.com/file/d/1LN_NG5NwpR3mZX89tWDZ4yIk4q4bE1Tp/view?usp=drive_link. For measurement, the photonic chip is mounted on a PCB with a gold exposed pad, but there’s no metal pad inside the chip for heat dissipation. The chip is wire-bonded with silver glue to the PCB and exposed pad. The PCB is on a temperature-controlled stage. I apply a pulse voltage with a 10 ns rise/fall time to the heater and measure the rise and fall time constants. Please let me know if you have any other questions and I look forward to your opinion as this is crucial for my analysis.