Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Simulating 30,300 holes

Please login with a confirmed email address before reporting spam

Hello COMSOL Community,

I am looking for a way to simulate a MEMS mirror designed in AutoCad. The mirror is 4 mm and saturated with 5 micron holes all 15 microns apart for a total of 30,300 holes. I am hoping through simulation to produce a frequency response plot, electrostatic force measurment, deformation (concave & convex) computation, and a graph of deflection v. voltage. Any council or advice is appreciated in this endeavor.

Designing these in AutoCad took a tole on the best computer my engineering buidling had to offer so i assume it will need similiar if not better specs to simulating them.

Specs: Windows 10 GPU: Intel (R) HD Graphics 630 CPU: Intel i7-7700 CPU @3.60GHz RAM: 16 Gb



2 Replies Last Post 20 juil. 2020, 16:30 UTC−4

Please login with a confirmed email address before reporting spam

Posted: 4 years ago 20 juil. 2020, 05:18 UTC−4
Updated: 4 years ago 20 juil. 2020, 05:19 UTC−4

Hi German,

Have you considered using symmetries of your geometry to reduce the model's size?

It seems that the holes follow a regular pattern, so depending on your equations you might be able to divide the model size by a factor of 4.

Best wishes, Alexis

Hi German, Have you considered using symmetries of your geometry to reduce the model's size? It seems that the holes follow a regular pattern, so depending on your equations you might be able to divide the model size by a factor of 4. Best wishes, Alexis

Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 4 years ago 20 juil. 2020, 16:30 UTC−4

Suggestions/thoughts: 1. For some of the things you want to calculate, might it be possible to model the mirror without the holes? If so, I strongly recommend that you consider doing that. 2. Do you have any reason to believe that any of the phenomena of interest to you are periodic, and tied to the periodicity of the holes, and/or the wavelengths of illuminating radiation? If so, consider modeling a "unit cell" containing (for example) one hole, and leveraging periodic boundary conditions. Also, take a good look at Floquet boundary conditions, which may be appropriate to employ (depending on the physics and periodicity involved, if any) here. 3. Even if there is no well-defined periodicity you can use, have you considered combining bulk-level approximations (e.g., physical optics) with separately-modeled/understood detailed local interactions, with only the latter being computed using the finite elements in Comsol Multiphysics? For example, once you understand how a piece of surface with a single hole responds to various conditions of illumination, you may then be able to represent (such as via an integral) how a surface with an array of such holes responds, via superposition of the responses of individual patches/holes.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Suggestions/thoughts: 1. For *some* of the things you want to calculate, might it be possible to model the mirror without the holes? If so, I strongly recommend that you consider doing that. 2. Do you have any reason to believe that any of the phenomena of interest to you are periodic, and tied to the periodicity of the holes, and/or the wavelengths of illuminating radiation? If so, consider modeling a "unit cell" containing (for example) one hole, and leveraging periodic boundary conditions. Also, take a good look at *Floquet* boundary conditions, which may be appropriate to employ (depending on the physics and periodicity involved, if any) here. 3. Even if there is no well-defined periodicity you can use, have you considered combining bulk-level approximations (e.g., physical optics) with separately-modeled/understood detailed local interactions, with only the latter being computed using the finite elements in Comsol Multiphysics? For example, once you understand how a piece of surface with a single hole responds to various conditions of illumination, you may then be able to represent (such as via an integral) how a surface with an array of such holes responds, via superposition of the responses of individual patches/holes.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.