Menu

Discussion Forum

New Discussion
Filtern

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.

Obtaining Different Results for same Simulation

Posted 31.10.2017, 15:44 GMT-4 Low-Frequency Electromagnetics, MEMS & Nanotechnology, Heat Transfer & Phase Change Version 5.3 3 Replies

Gabriel Papadatos
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

I have been using COMSOL to simulate an electrochemical cell using the AC/DC and Heat Transfer Modules. I had been using COMSOL version 4.4 and recently upgraded to COMSOL 5.3. For some reason, when I run simulations with the new version, I get very different results for the plots of temperature vs time. The results from COMSOL 4.4 match up better with the experimental results. I have not made any significant changes to the simulation. The only change I made is that I replaced the "Surface-to-ambient-Radiation" physics into "Diffuse Surface", and all that did was change the values, the shape of the graph still does not match the results achieved beforehand with version 4.4. Are there any significant changes between COMSOL 4.4 and COMSOL 5.3 that I have not accounted for?

Thank You, Gabriel Papadatos

3 Replies Last Post 01.11.2017, 17:13 GMT-4
Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 31.10.2017, 17:10 GMT-4
Updated: 7 years ago 01.11.2017, 08:44 GMT-4

Hello Gabriel,

Did you perform a mesh convergence study on your 4.4 model (https://www.comsol.com/multiphysics/mesh-refinement) ? If not, it is possible that your 4.4 file was sensitive to the choice of mesh. In that case, since meshing algorithms have changed between versions 4.4 and 5.3, you could get a noticeable change in your solution when running the identical file in 5.3.

Best,

Jeff

-------------------
Jeff Hiller
Hello Gabriel, Did you perform a mesh convergence study on your 4.4 model (https://www.comsol.com/multiphysics/mesh-refinement) ? If not, it is possible that your 4.4 file was sensitive to the choice of mesh. In that case, since meshing algorithms have changed between versions 4.4 and 5.3, you could get a noticeable change in your solution when running the identical file in 5.3. Best, Jeff
Gabriel Papadatos
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 01.11.2017, 16:23 GMT-4

To Jeff,

Thank you for the information. I used the physics based mesh for version 4.4 and did not modify it for version 5.3. Do you reccomend performing a mesh convergence study on the file in 5.3, as I no longer have access to 4.4, or is itbest to recreate the mesh manually?

Best, Gabriel Papadatos

To Jeff, Thank you for the information. I used the physics based mesh for version 4.4 and did not modify it for version 5.3. Do you reccomend performing a mesh convergence study on the file in 5.3, as I no longer have access to 4.4, or is itbest to recreate the mesh manually? Best, Gabriel Papadatos
Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 01.11.2017, 17:13 GMT-4

Whether you are using a "User Controled" mesh or a "Physics controled" one, it is good practice to perform a mesh convergence study to ascrtain that the solution is not mesh dependent. In your case, you will need to do it in 5.3 since 4.4 is no longer available to you.

BTW, there is an example showing how to do a mesh convergence study using a parametric sweep in the Introduction to COMSOL Multiphysics manual (Accessible through File > Help > Documentation or here: https://cdn.comsol.com/documentation/5.3.0.316/IntroductionToCOMSOLMultiphysics.pdf ).

Best,

Jeff

-------------------
Jeff Hiller
Whether you are using a "User Controled" mesh or a "Physics controled" one, it is good practice to perform a mesh convergence study to ascrtain that the solution is not mesh dependent. In your case, you will need to do it in 5.3 since 4.4 is no longer available to you. BTW, there is an example showing how to do a mesh convergence study using a parametric sweep in the Introduction to COMSOL Multiphysics manual (Accessible through File > Help > Documentation or here: https://cdn.comsol.com/documentation/5.3.0.316/IntroductionToCOMSOLMultiphysics.pdf ). Best, Jeff

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.

Suggested Content