Menu

Discussion Forum

New Discussion
Filter

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.

Infinitely long cylinder as a simple harmonic oscillator

Posted May 6, 2012, 6:51 p.m. EDT Heat Transfer & Phase Change 5 Replies

ForkTuner
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

This is a very simple problem. An infinitely long rigid cylinder with a given diameter and density functions as a simple harmonic oscillator with a given spring constant (per unit length). So this is a 2D problem. How to find the eigenfrequency with Comsol? Later I intend to couple this oscillator to a fluid etc.
5 Replies Last Post Jul 1, 2015, 1:40 a.m. EDT
ForkTuner
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 14, 2012, 4:36 a.m. EDT
Anyone?

This is what I have tried:

In model wizard choose 2D, solid mechanics, eigenfrequency. Then I draw a circle to represent the infinite beam and I choose some material for it. In the solid mechanics (solid) I create a new rigid connector and in the rigid domain I choose the entire circle. In the rigid connector I check a prescribed displacement in the Y-direction and set it to 0. In the rigid connector I add an applied force, and in the x-direction I give a value -100*u[N/m]. Create the mesh and try to solve. The solver gives the following error:

The source selection for the operator intrig11 must be nonempty. Error in multiphysics compilation.
-Feature: Compile Equations: Eigenfrequency (sol1/st1)
-Error: The source selection for the operator intrig11 must be nonempty

Any ideas?
Anyone? This is what I have tried: In model wizard choose 2D, solid mechanics, eigenfrequency. Then I draw a circle to represent the infinite beam and I choose some material for it. In the solid mechanics (solid) I create a new rigid connector and in the rigid domain I choose the entire circle. In the rigid connector I check a prescribed displacement in the Y-direction and set it to 0. In the rigid connector I add an applied force, and in the x-direction I give a value -100*u[N/m]. Create the mesh and try to solve. The solver gives the following error: The source selection for the operator intrig11 must be nonempty. Error in multiphysics compilation. -Feature: Compile Equations: Eigenfrequency (sol1/st1) -Error: The source selection for the operator intrig11 must be nonempty Any ideas?
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 May 14, 2012, 4:46 a.m. EDT
Hi

First I would say its a 1D problem as you have a constant section, it solves even quicker.
Then, I'm nut sure you can truely represent "INF" in a FEM (Finite Elements ...). You need to find a way to define your BC i.e. by a periodic condition. Have you checked the model library about PDE examples ?

--
Good luck
Ivar
Hi First I would say its a 1D problem as you have a constant section, it solves even quicker. Then, I'm nut sure you can truely represent "INF" in a FEM (Finite Elements ...). You need to find a way to define your BC i.e. by a periodic condition. Have you checked the model library about PDE examples ? -- Good luck Ivar
ForkTuner
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 14, 2012, 1:59 p.m. EDT
Thanks for the reply. You're right, it is a 1D problem at the moment, but once I couple it with a fluid, it's 2D. The "infinity" here is just that the problem is 2D.

I can make everything work just fine if I won't make the circle (cylinder) a rigid domain (i.e. so that the cylinder is also allowed to deform). I wonder why this is.. I suppose I could use it this way as well, but I would probably save some computational effort if the rigid body would actually be rigid.
Thanks for the reply. You're right, it is a 1D problem at the moment, but once I couple it with a fluid, it's 2D. The "infinity" here is just that the problem is 2D. I can make everything work just fine if I won't make the circle (cylinder) a rigid domain (i.e. so that the cylinder is also allowed to deform). I wonder why this is.. I suppose I could use it this way as well, but I would probably save some computational effort if the rigid body would actually be rigid.
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 May 14, 2012, 3:10 p.m. EDT
Hi

Ok but then I missed something, if you are in 2D and you see along the axis, your cylinder is a circle indeed, but by default ist NOT infinite in Z (out of the paper) it's rather finite and by default COMSOL uses 1[m] depth, most values comes out as values "per meter" depth, up to you to multiply by the thrue depth.

This does not really apply to eigenfrequencies, as thes are restrained truely to 2D and you do not see any Z, Rx, Ry modes (only X,Y, & Rz modes) by symmetry, so in fact you are missing many modes.

Now, you say you have a fluid around, this will indeed interact, but that is rather tricky, you have the added mass for "small deformations" but also the visosity. I'm not sure what is the best to use Acoustics or FSI ... ? and for eigenfrequencies ?

--
Good luck
Ivar
Hi Ok but then I missed something, if you are in 2D and you see along the axis, your cylinder is a circle indeed, but by default ist NOT infinite in Z (out of the paper) it's rather finite and by default COMSOL uses 1[m] depth, most values comes out as values "per meter" depth, up to you to multiply by the thrue depth. This does not really apply to eigenfrequencies, as thes are restrained truely to 2D and you do not see any Z, Rx, Ry modes (only X,Y, & Rz modes) by symmetry, so in fact you are missing many modes. Now, you say you have a fluid around, this will indeed interact, but that is rather tricky, you have the added mass for "small deformations" but also the visosity. I'm not sure what is the best to use Acoustics or FSI ... ? and for eigenfrequencies ? -- Good luck Ivar
Manik Kumar
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 9 years ago Jul 1, 2015, 1:40 a.m. EDT
how to oscillate a heated cylinder in a channel with air cross flow for 2D case
how to oscillate a heated cylinder in a channel with air cross flow for 2D case

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