Menu

Discussion Forum

New Discussion
Filtrer

Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

Out of Memory while creating mesh!!

Posted 10 janv. 2012, 22:37 UTC−5 Interfacing, Geometry, Mesh, Structural Mechanics Version 4.0a, Version 4.2a 7 Replies

Rohan Repale
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi,
I am trying to create a 2D model, wherein I import geometry from AutoCad.
The Problem is that the geometry is so huge, that I cannot form mesh on it, even with a very high element size.

Now, I cannot use symmetry BC, because it isnt symmetric bout any axis and I cannot scale it or redesign it.

So is there any way, I can solve this?

Is there a way to increase the memory allotment of comsol?

Will running the program on a supercomputer cluster help solve the problem?

Thank you,

-Rohan.
7 Replies Last Post 29 avr. 2012, 02:58 UTC−4
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 11 janv. 2012, 01:31 UTC−5
Hi

the only way I know about, and you will not be pleased, is to simplify your geometry, that is suppress small features that are not really relevant, replace complex struts by bars etc, it's quite some work

PS : forget the older versions, use only the latest one ;)

--
Good luck
Ivar
Hi the only way I know about, and you will not be pleased, is to simplify your geometry, that is suppress small features that are not really relevant, replace complex struts by bars etc, it's quite some work PS : forget the older versions, use only the latest one ;) -- Good luck Ivar
Sally Ahmed
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 avr. 2012, 07:23 UTC−4
Hi

I ran out of memory while creating the mesh
My structure is so simple..It consists of four layers and the thicknesses are in the micro or nano region
However, the width and length in cm... it took soo much time and could not create the mesh :(

i tried to make the length in the micro region and it created the mesh easily
but i need to to be in the cm region..is there any suggestion to solve this problem?

Thanks alot
Regards
Sally
Hi I ran out of memory while creating the mesh My structure is so simple..It consists of four layers and the thicknesses are in the micro or nano region However, the width and length in cm... it took soo much time and could not create the mesh :( i tried to make the length in the micro region and it created the mesh easily but i need to to be in the cm region..is there any suggestion to solve this problem? Thanks alot Regards Sally
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 avr. 2012, 07:31 UTC−4
Hi there are a few ways:

one is to consider the thinnest (or a few of) and replace them by boundary physics

another is to use a distorted (scaled-stretched) mesh, I often use mapped mesh of a middle surface then sweep the mesh outwards with a given distribution (at least 3 elements per layer thickness as else you annot resolve a 2nd order equations nor gradient

Meshing is an art ;) training makes champion here too, but still start coarse and then check carefully by refining that your solution is mesh independent (more or less)

--
Good luck
Ivar
Hi there are a few ways: one is to consider the thinnest (or a few of) and replace them by boundary physics another is to use a distorted (scaled-stretched) mesh, I often use mapped mesh of a middle surface then sweep the mesh outwards with a given distribution (at least 3 elements per layer thickness as else you annot resolve a 2nd order equations nor gradient Meshing is an art ;) training makes champion here too, but still start coarse and then check carefully by refining that your solution is mesh independent (more or less) -- Good luck Ivar
Sally Ahmed
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 avr. 2012, 08:17 UTC−4
Thanks alot lvar for your prompt reply

Actually I am new to Comsol and i do not have experience with meshing and how to set the parameters for the distribution..etc
Do you have any document or tutorials of meshing thin films?

I want to try your second suggestion but i do not know how to do this

Thanks again
Sally
Thanks alot lvar for your prompt reply Actually I am new to Comsol and i do not have experience with meshing and how to set the parameters for the distribution..etc Do you have any document or tutorials of meshing thin films? I want to try your second suggestion but i do not know how to do this Thanks again Sally
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 avr. 2012, 10:01 UTC−4
Hi

check the main web site with the videos, then do the exercices ... Also if you can make it take a few of the excellent COMSOL courses, there is one specifically for meshing but it's already advanced yo you should master the meshing well before going to it

--
Good luck
Ivar
Hi check the main web site with the videos, then do the exercices ... Also if you can make it take a few of the excellent COMSOL courses, there is one specifically for meshing but it's already advanced yo you should master the meshing well before going to it -- Good luck Ivar
Sally Ahmed
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28 avr. 2012, 11:59 UTC−4
Hi Ivar

I am getting this error when i try to compute a stationary study..Do you know what does this error mean?

FYI: I am simulating an energy harvester on a flexible substrate and i used distribution for meshing

Error:
Failed to find a solution.
The relative error (18) is greater than the relative tolerance.
Returned solution is not converged.

Thanks alot
Regards
Sally
Hi Ivar I am getting this error when i try to compute a stationary study..Do you know what does this error mean? FYI: I am simulating an energy harvester on a flexible substrate and i used distribution for meshing Error: Failed to find a solution. The relative error (18) is greater than the relative tolerance. Returned solution is not converged. Thanks alot Regards Sally
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 29 avr. 2012, 02:58 UTC−4
Hi

check the KB (knowledge base) ;)

--
Good luck
Ivar
Hi check the KB (knowledge base) ;) -- Good luck Ivar

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