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.

Implementing Ogden's model for modeling compression response of Polyurethane foams in COMSOL

Balaji Ramanujakannan

Please login with a confirmed email address before reporting spam

Tech support team and other users,

I'm trying to implement Ogden's model for polyurethane elastic foams into COMSOL. Currently, I'm working with the structural mechanics module and trying to model the uniaxial compression stress-strain response of these foams. The tend to behave in a non-linear fashion and a typical stress-strain curve is attached herewith. As I'm new to COMSOL, can anyone suggest how to change the governing equations and boundary conditions and preferably some tutorial on the equation modification process.

Thanks in advance,
Balaji


4 Replies Last Post 4 juin 2010, 04:54 UTC−4
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: 1 decade ago 1 juin 2010, 01:19 UTC−4
Hi

I believe you could start with entering the non linear curve into your material properties (ie. make a custom material), I would also suggest to start with an isotropic material (even if wrong) and when your happy with your model , only then make it anisotropic.
I do not think you need to change anything else. In fact, you do not even need to make a "custom material", you can start to define your "function" and type it straight into the subdomain material properties, but for later its usefull to enter as a custom material.

Be sure you understand the specific naming of COMSOL fo he different variables.

Like this I believe it should be rather staightforward.

And do not forget to give us some feedback, either here or at a Comsol conference, so we can learn too ;)

have fun Comsoling
Ivar
Hi I believe you could start with entering the non linear curve into your material properties (ie. make a custom material), I would also suggest to start with an isotropic material (even if wrong) and when your happy with your model , only then make it anisotropic. I do not think you need to change anything else. In fact, you do not even need to make a "custom material", you can start to define your "function" and type it straight into the subdomain material properties, but for later its usefull to enter as a custom material. Be sure you understand the specific naming of COMSOL fo he different variables. Like this I believe it should be rather staightforward. And do not forget to give us some feedback, either here or at a Comsol conference, so we can learn too ;) have fun Comsoling Ivar

Balaji Ramanujakannan

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 1 juin 2010, 04:20 UTC−4
Hi Ivar,

Thanks for your efforts in replying. Yeah I will try to work in the way as you suggested. Actually, my material in not anisotrpic, but it behaves non-linear in any direction of load invariably.

My current understanding is that, I'll have to start with PDE application mode and then proceed with general mode as I realize from comsol blogspot (website link given below), that PDE in general form should be used for non-linear applications.

comsol-users.blogspot.com/2006/11/crash-course-on-comsolfemlab-for.html

To be even more specific, I can summarize my queries as follows:

1. Is my assumption to work with PDE general form correct . Some of my equations are not actually PDE and they relate the strain energy to extension ratio. What would be the governing equation in this case?

2. Is there a requirement for me input experimental data as an input parameter as in like other analysis softwares. Or can I simulate the compression of foam and view it's deformation by feeding in the necessary equations?

3. What all do I need to input in the constants under subdomain settings menu?

4. Can you pl. explain a bit about naming of COMSOL for different variables?

Sorry for shooting many questions. Thanks for co-operation. Sure, will keep posted in this thread on my development.

Regards,
Balaji

Hi Ivar, Thanks for your efforts in replying. Yeah I will try to work in the way as you suggested. Actually, my material in not anisotrpic, but it behaves non-linear in any direction of load invariably. My current understanding is that, I'll have to start with PDE application mode and then proceed with general mode as I realize from comsol blogspot (website link given below), that PDE in general form should be used for non-linear applications. http://comsol-users.blogspot.com/2006/11/crash-course-on-comsolfemlab-for.html To be even more specific, I can summarize my queries as follows: 1. Is my assumption to work with PDE general form correct . Some of my equations are not actually PDE and they relate the strain energy to extension ratio. What would be the governing equation in this case? 2. Is there a requirement for me input experimental data as an input parameter as in like other analysis softwares. Or can I simulate the compression of foam and view it's deformation by feeding in the necessary equations? 3. What all do I need to input in the constants under subdomain settings menu? 4. Can you pl. explain a bit about naming of COMSOL for different variables? Sorry for shooting many questions. Thanks for co-operation. Sure, will keep posted in this thread on my development. Regards, Balaji

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: 1 decade ago 1 juin 2010, 06:15 UTC−4
Hi

certainly if you have only the basic Comsol Muliphysics, or you do want to implement everything from the bare equations you can start this way, you will certainly learn a lot.
But if I had to reply, I would use the solid structural module, import/create your geometry, define the material variables on the subdomains (and for you define a function based on the material non-linear shape with corresponding comsol I/O variables), define the load/fixed etc BCs and run a solve.

but I beleve you must use the detailed matrices for the material properties as you will not have standard E and nu expressed in a simple form, but if your material is isotropic, you should not have too many items to fill in.

Really you should read through the doc, I know its long, (we have all passed by there) but it's worth it to be far quicker te next time.

Unfortunately, I do not have time to set up a full model for you (I'm a user as you, and just now sitting in a teleconf meeting, but pls do not say ;)

Take i.e. the rubber large deformation case from the model library, there are also several examples in the doc

Have fun Comsoling
Ivar
Hi certainly if you have only the basic Comsol Muliphysics, or you do want to implement everything from the bare equations you can start this way, you will certainly learn a lot. But if I had to reply, I would use the solid structural module, import/create your geometry, define the material variables on the subdomains (and for you define a function based on the material non-linear shape with corresponding comsol I/O variables), define the load/fixed etc BCs and run a solve. but I beleve you must use the detailed matrices for the material properties as you will not have standard E and nu expressed in a simple form, but if your material is isotropic, you should not have too many items to fill in. Really you should read through the doc, I know its long, (we have all passed by there) but it's worth it to be far quicker te next time. Unfortunately, I do not have time to set up a full model for you (I'm a user as you, and just now sitting in a teleconf meeting, but pls do not say ;) Take i.e. the rubber large deformation case from the model library, there are also several examples in the doc Have fun Comsoling Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 4 juin 2010, 04:54 UTC−4
Hi Balaji and Ivar,

I am not familiar with foam modeling, but I have been working on implementing ogden model in Comsol 3.5a.

I succeed in 2D, but 3D sometimes it works and sometimes doesn't.

The way I did was go modify the strain energy function in the menu 'physics-Equation system system-subdomain setting' to be the ogden energy function.

But since in the function, we have to calculate the eigenvalues of the strain tensor, then we need to define some global expressions there and define the ogden energy function.

I just read something about new comsol 4.0, seems it's rather easy to implement ogden now, but I am not sure, and I am trying to install the new version to see.

Hope this helps.
/Li
Hi Balaji and Ivar, I am not familiar with foam modeling, but I have been working on implementing ogden model in Comsol 3.5a. I succeed in 2D, but 3D sometimes it works and sometimes doesn't. The way I did was go modify the strain energy function in the menu 'physics-Equation system system-subdomain setting' to be the ogden energy function. But since in the function, we have to calculate the eigenvalues of the strain tensor, then we need to define some global expressions there and define the ogden energy function. I just read something about new comsol 4.0, seems it's rather easy to implement ogden now, but I am not sure, and I am trying to install the new version to see. Hope this helps. /Li

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.