# Boundary Conditions in HEAT - Simulation Object

Member Posts: 1

HI:

I tried to follow following example for simulating my structure:https://support.lumerical.com/hc/en-us/articles/360041686234

I want to calculate the temperature rise of two silicon disks with different diameters and heights of 60nm，The figure above is my FDTD modeling

The figure above is my HEAT modeling,My substrate is sio2，The boundary conditions of glass and air are convection, and the boundary conditions of Si and air are convection。In the example, the substrate is Al2O3, au is on the substrate, and no boundary conditions are set between the two interfaces. In my design, I want to know how to set the boundary conditions of Si and sio2.？ Or no boundary conditions are set as in the example

• Posts: 80Ansys Employee

Hi,

Good question. Here the boundary conditions between Air and the other solids represent the convection cooling through the fluid. This doesn't apply to the Si and SiO2 interface. In general convection and temperature are two of the more commonly used boundary conditions in HEAT (both are used in the mentioned example). When setting up the boundary conditions for your simulation, I always advise referring to your understanding of your structure’s heat dynamics or any references you might have for a similar structure. Given your description, the structure seems similar to the one published in the example, so using convection and temperature for substrate should be a good starting point.

Best regards

• Posts: 2Member

Hi,

I have another question, such as;

1) In the following example:https://support.lumerical.com/hc/en-us/articles/360041686234 How is the length scale (3.4e-07) in the boundary condition calculate?

As far as I know, lt is calculated like this

But my calculation based on the data given in the model is not correct

Thanks

zywang

• Posts: 80Ansys Employee

Hi Zywang,

Good question. I’m currently not sure how we have derived this value in this example. I will check with my colleagues and update this post. In the meantime, you can test working with the constant model of convection and examine how/if the results will change.

Best,

Khashayar