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.

Minimum digits in COMSOL simulation

Please login with a confirmed email address before reporting spam

Hello,

I am simulating a 3D structure and doing some eigen frequency analysis. This is a periodic structure and I am using the Floquet periodicity. The change in the wavelength should be around $10^-6$ nm. But I did not able to obtain this high precision. I am wondering about the minimum comsol digits that can be stored or simulated. I will be grateful if someone helps.

Thanks in advance!


1 Reply Last Post 9 mar 2021, 10:28 GMT-5
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 4 years ago 9 mar 2021, 10:28 GMT-5
Updated: 4 years ago 9 mar 2021, 10:32 GMT-5

Am I reading that number correctly? One millionth of one nanometer?? Are you trying to develop some kind of periodic model of the interior of a proton or other sub-atomic particle? Or perhaps some crystallographic model of matter in a neutron star? Anyway, although finite elements can indeed effectively resolve, at a bulk-level, physical effects that correspond to dimensions that are smaller than the individual element sizes, there are still limits. How large is your mesh size? Note: Comsol uses floating point numbers and (I think) double precision is standard, so that's plenty of computational digits of precision. But it also represents continuum space using a discrete mesh. So in many cases, only a handful of digits will actually correspond to physical reality, and the rest will change if you refine the mesh. It all very much depends on the (many) details of your model.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Am I reading that number correctly? One *millionth* of one *nanometer*?? Are you trying to develop some kind of periodic model of the interior of a proton or other sub-atomic particle? Or perhaps some crystallographic model of matter in a neutron star? Anyway, although finite elements can indeed effectively resolve, at a bulk-level, physical effects that correspond to dimensions that are smaller than the individual element sizes, there are still limits. How large is your mesh size? Note: Comsol uses floating point numbers and (I think) double precision is standard, so that's plenty of computational digits of precision. But it also represents continuum space using a discrete mesh. So in many cases, only a handful of digits will actually correspond to physical reality, and the rest will change if you refine the mesh. It all very much depends on the (many) details of your model.

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.