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 The information provided may be out of date.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.

Modelling Compression of Material at a Rate

Please login with a confirmed email address before reporting spam

Hello,

I am a beginner to COMSOL 4.2a and I would like to build a simple model where a material (a cylindrical block of polymer) is compressed at a rate of 10 mm/min.

Using Solid Mechanics, I was able to construct the cylinder and apply a boundary load. However, this is just an instantaneous load.

My question is how do I go about implementing a compression of the material at a rate of 10mm/min?

Thanks for your help!

-Adrian

5 Replies Last Post 18.01.2013, 10:17 GMT-5
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 17.01.2013, 01:40 GMT-5
Hi

you can work in th time domain, and say the displacement is 10[mm/s]*t, but try to smoothen the displacement velocity as you will get many transient at the beginning around t=0, one way to improve is to remove the inertial terms (2nd order derivatives) but then you remove all wave propagations. But you can also, if the time behaviour is not the essential use a stationary parametric sweep and increase either the force load with a parameter or impose a displacement with a parameter. You loose the time response but you see the displacement force response, interesting in particular if your material is nonlinear, or for large deformations

--
Good luck
Ivar
Hi you can work in th time domain, and say the displacement is 10[mm/s]*t, but try to smoothen the displacement velocity as you will get many transient at the beginning around t=0, one way to improve is to remove the inertial terms (2nd order derivatives) but then you remove all wave propagations. But you can also, if the time behaviour is not the essential use a stationary parametric sweep and increase either the force load with a parameter or impose a displacement with a parameter. You loose the time response but you see the displacement force response, interesting in particular if your material is nonlinear, or for large deformations -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 17.01.2013, 12:41 GMT-5

Hi

you can work in th time domain, and say the displacement is 10[mm/s]*t, but try to smoothen the displacement velocity as you will get many transient at the beginning around t=0, one way to improve is to remove the inertial terms (2nd order derivatives) but then you remove all wave propagations. But you can also, if the time behaviour is not the essential use a stationary parametric sweep and increase either the force load with a parameter or impose a displacement with a parameter. You loose the time response but you see the displacement force response, interesting in particular if your material is nonlinear, or for large deformations

--
Good luck
Ivar


Thank you Ivar. I will try your suggestions and see where I can get.

Thanks,
-Adrian

[QUOTE] Hi you can work in th time domain, and say the displacement is 10[mm/s]*t, but try to smoothen the displacement velocity as you will get many transient at the beginning around t=0, one way to improve is to remove the inertial terms (2nd order derivatives) but then you remove all wave propagations. But you can also, if the time behaviour is not the essential use a stationary parametric sweep and increase either the force load with a parameter or impose a displacement with a parameter. You loose the time response but you see the displacement force response, interesting in particular if your material is nonlinear, or for large deformations -- Good luck Ivar [/QUOTE] Thank you Ivar. I will try your suggestions and see where I can get. Thanks, -Adrian

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 17.01.2013, 17:17 GMT-5
Update:

I tried to apply a prescribed velocity and conducted a time-dependent study. However, I am getting a

"Failed to find a solution" error message.

It says that there are void equations (empty rows in matrix) and returned solution is not converged.

Looking up previous discussions, it seems that this error could be tied to the size of the mesh. However, I have tried different mesh sizes with no avail.

Any suggestions would be great!

Thanks!
-Adrian
Update: I tried to apply a prescribed velocity and conducted a time-dependent study. However, I am getting a "Failed to find a solution" error message. It says that there are void equations (empty rows in matrix) and returned solution is not converged. Looking up previous discussions, it seems that this error could be tied to the size of the mesh. However, I have tried different mesh sizes with no avail. Any suggestions would be great! Thanks! -Adrian

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 18.01.2013, 01:53 GMT-5
Hi

are you using a segregated iterative solver or a direct one, perhaps the direct one would work (if you have enough RAM)

--
Good luck
Ivar
Hi are you using a segregated iterative solver or a direct one, perhaps the direct one would work (if you have enough RAM) -- Good luck Ivar

Nagi Elabbasi Facebook Reality Labs

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 18.01.2013, 10:17 GMT-5
Hi Adrian,

Are you getting this error message right at the start of the analysis or do you get a partial solution? If at the start then try ramping up the velocity gradually starting at zero as Ivar suggested earlier. Getting an initial solution should be easy since there are no applied loads. Consider also changing the time stepping setting in the transient solver from Generalized Alpha to BDF. I found the BDF to be easier to converge for some problems.

Nagi Elabbasi
Veryst Engineering
Hi Adrian, Are you getting this error message right at the start of the analysis or do you get a partial solution? If at the start then try ramping up the velocity gradually starting at zero as Ivar suggested earlier. Getting an initial solution should be easy since there are no applied loads. Consider also changing the time stepping setting in the transient solver from Generalized Alpha to BDF. I found the BDF to be easier to converge for some problems. Nagi Elabbasi Veryst Engineering

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.