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.

solid deformation and heat transfer

Please login with a confirmed email address before reporting spam


First off before I ask my question is state that I have gone through the forums and the book in great depth and have not been able to figure this out.

I'm using version 4.3 and do not have the solid mechanics add on package, but do have the heat transfer package.

I have spent days trying to figure out how I take a simple 2-D square and apply mechanical load to deform the shape then apply heat transfer.

I have tried performing this operation a number of ways. I have tried combining them, two separate studies, two separate studies with ALE, two different models, and import the deformed mesh, and more.

All I want to do is this simple operation, and every time I get the results as if they are split. The solid mechanics solution shows deformed shape, the heat transfer shows the original shape with the applied physics.

Even when I try to remesh with a deformed mesh (following forum and instruction manual) it gives me the same shape.

I don't know if there is a setting I do not have checked, I'm missing a simple step, or there is a bug in the program.

These are the forum articles I that where similar.
www.comsol.com/community/forums/general/thread/15851
www.comsol.com/community/forums/general/thread/15185/
www.comsol.com/community/forums/general/thread/29475/

I do not have any terms coupled due to strictly looking at this as a boundary value problem which I was lead to believe Comsol can solve. I have seen much more involved examples and video with the coupled physics and which all be nice for later on, and if I had the packages.

So again how would mechanically deform a 2-D square and have heat transfer modeled with the deformed geometry.

A bonus would be to undergo this operation one or 2 more times.

Thank you;

6 Replies Last Post Aug 17, 2015, 6:08 a.m. EDT
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Apr 9, 2015, 2:11 a.m. EDT
Hi

moving mesh (ALE) is typically for fluid flow or representing the air/vacuum in a couples structural and electrostatic ES or EC model.

So do not use ALE for thermal stress, you need to add the thermal strain into the structural strain. Well its part of the structural + thermal modules, so without the structral module you have to write quite some eqautions into COMSOl to get everything up.

But you could start in 2D by adding the thermal strain e_t= alpha*(T-Tref) into your structral model, but I believe you cannot avoid digging into the udnerlaying COMSOL equations (turn on the equation view)

At least you will learn how COMSOL is working "under the hood" ;)

--
Good luck
Ivar
Hi moving mesh (ALE) is typically for fluid flow or representing the air/vacuum in a couples structural and electrostatic ES or EC model. So do not use ALE for thermal stress, you need to add the thermal strain into the structural strain. Well its part of the structural + thermal modules, so without the structral module you have to write quite some eqautions into COMSOl to get everything up. But you could start in 2D by adding the thermal strain e_t= alpha*(T-Tref) into your structral model, but I believe you cannot avoid digging into the udnerlaying COMSOL equations (turn on the equation view) At least you will learn how COMSOL is working "under the hood" ;) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Apr 9, 2015, 11:36 a.m. EDT
Ivar

Thank you for your reply. It helps put my mind at ease to know it was not something silly I was missing, and that it is possible. I wanted to start digging into the equation more anyway so this gives me an excuse to start early.

Is there an issue with version 4.3 or do I need an add on from a different package to deform a geometry and use it in another study with similar or different physics? I followed all the forum advice from the past and it does not seem to work. You mentioned in one to make sure the nonlinear solver is turned on in 4.3; however I checked the manual and that solver is only available with certain packages. Every time I try to resuse geometry I get the undeformed/original shape. I'm attaching the instructions I used, plus I followed the instruction manual step-by-step and get the same results.

I very grateful for your help and time.

Thank you;
Ivar Thank you for your reply. It helps put my mind at ease to know it was not something silly I was missing, and that it is possible. I wanted to start digging into the equation more anyway so this gives me an excuse to start early. Is there an issue with version 4.3 or do I need an add on from a different package to deform a geometry and use it in another study with similar or different physics? I followed all the forum advice from the past and it does not seem to work. You mentioned in one to make sure the nonlinear solver is turned on in 4.3; however I checked the manual and that solver is only available with certain packages. Every time I try to resuse geometry I get the undeformed/original shape. I'm attaching the instructions I used, plus I followed the instruction manual step-by-step and get the same results. I very grateful for your help and time. Thank you;

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Apr 9, 2015, 11:47 a.m. EDT
Hi
Indeed by default you are in the Material Frame so you must save the X+u,Y+v,Z+w results to get a deformed geometry. Now I'm not used to COMSOL without the structural physics, where all you want is already implemented. But I believe you can save the deformed geometry to reuse it, also without the full structural physics.

--
Good luck
Ivar
Hi Indeed by default you are in the Material Frame so you must save the X+u,Y+v,Z+w results to get a deformed geometry. Now I'm not used to COMSOL without the structural physics, where all you want is already implemented. But I believe you can save the deformed geometry to reuse it, also without the full structural physics. -- Good luck Ivar

Walter Frei COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Apr 9, 2015, 2:14 p.m. EDT
Dear Sepideh,

As my colleague Ivar correctly points out, the ability to compute the geometric nonlinear deformation is already available in the Structural Mechanics Module (as well as the MEMS Module) and you would need to use this functionality if you wish to model heat transfer on a deformed body. The Solid Mechanics interface in the core package addresses purely infinitesimal strain problems.

Also keep in mind that it is not only the deformed shape that you will need to consider, but also the change in the material properties. (As you compress or stretch a material, the thermal conductivity, specific heat, and of course the density change.)

These changes of the material properties are described on pp. 687-691 of the COMSOL Multiphysics Reference Manual version 5.0. When you have the Structural Mechanics or MEMS Module the large structural deformations will be correctly computed and the thermal properties will be correct.

For the modeling you are describing you will want to use either the Structural Mechanics or MEMS Module, and if you do not have these products with your license please contact your local COMSOL Sales Office:
www.comsol.com/contact
Dear Sepideh, As my colleague Ivar correctly points out, the ability to compute the geometric nonlinear deformation is already available in the Structural Mechanics Module (as well as the MEMS Module) and you would need to use this functionality if you wish to model heat transfer on a deformed body. The Solid Mechanics interface in the core package addresses purely infinitesimal strain problems. Also keep in mind that it is not only the deformed shape that you will need to consider, but also the change in the material properties. (As you compress or stretch a material, the thermal conductivity, specific heat, and of course the density change.) These changes of the material properties are described on pp. 687-691 of the COMSOL Multiphysics Reference Manual version 5.0. When you have the Structural Mechanics or MEMS Module the large structural deformations will be correctly computed and the thermal properties will be correct. For the modeling you are describing you will want to use either the Structural Mechanics or MEMS Module, and if you do not have these products with your license please contact your local COMSOL Sales Office: http://www.comsol.com/contact

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Apr 9, 2015, 6:42 p.m. EDT
Ivar and Walter

Thank you both for answering my questions. Because of your help I now know how to move forward.

Thank you;
Ivar and Walter Thank you both for answering my questions. Because of your help I now know how to move forward. Thank you;

Please login with a confirmed email address before reporting spam

Posted: 9 years ago Aug 17, 2015, 6:08 a.m. EDT
I would like to thanks Ivar and comsol team for their support, as new I am struggling to the modelling of a laser heating a layered materials with the second layer transparent to the laser wavelength.

How can obtain a general temperature profile which take the difference in materials properties:
By solving 2 diff equations ? or just one equations with a weak boundary condition at the interface for heat flux and temp continuity.

I do not know how to integrate the interface boundary in my one equation,
please an Hint

Merci beaucoup.
I would like to thanks Ivar and comsol team for their support, as new I am struggling to the modelling of a laser heating a layered materials with the second layer transparent to the laser wavelength. How can obtain a general temperature profile which take the difference in materials properties: By solving 2 diff equations ? or just one equations with a weak boundary condition at the interface for heat flux and temp continuity. I do not know how to integrate the interface boundary in my one equation, please an Hint Merci beaucoup.

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.