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.

Eigenfrequencies of heated structure

Please login with a confirmed email address before reporting spam

Hallo

I am a relative new user of COMSOL and I have some problems with one of my models. I hope someone can help me.
I am trying to model how the eigenfrequency of a tensioned double clamped beam changes as the temperature of the structure changes. The beam and its support are made from different materials. The idea is that the tension of the beam changes as the temperature of the structure changes due to the difference in coefficient of thermal expansion of the beam and the support. Hence a change in tension gives a change in eigenfrequency.

My question is: How do I implement the thermal stress and strain when I calculated the eigenfrequency?

Best regards Tom

6 Replies Last Post Dec 8, 2010, 3:32 p.m. EST
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 Nov 29, 2010, 8:43 a.m. EST
Hi

it's the other way around:

1) thermal stress is set up via temperature increase from the HT physics, that one feed into the Structural ST physics to adapt the resulting stress level, all via a stationary solver process
2) the Structural eigenfrequency solver is called upon, and being asked to start to linearise from the previous stationary solved (and stored) solution

This works OK in v4 (call up both solvers sequentially, and check that you apply the correct physics for the correct solvers. But in v3.5a you must still go via matlab to read in and load back the first stationary case to have a stressed starting point for the eigenfrequency analysis

--
Good luck
Ivar
Hi it's the other way around: 1) thermal stress is set up via temperature increase from the HT physics, that one feed into the Structural ST physics to adapt the resulting stress level, all via a stationary solver process 2) the Structural eigenfrequency solver is called upon, and being asked to start to linearise from the previous stationary solved (and stored) solution This works OK in v4 (call up both solvers sequentially, and check that you apply the correct physics for the correct solvers. But in v3.5a you must still go via matlab to read in and load back the first stationary case to have a stressed starting point for the eigenfrequency analysis -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Nov 29, 2010, 10:47 a.m. EST
Hi

Thanks for your reply.

In my model I start by calculating the thermal stress and strain, as you suggested. However, I do not know how to use the calculated stress and strain in the frequency simulation.

I have tried to find a tutorial that explains this, but without any success. Are you aware of one?

Best regards Tom
Hi Thanks for your reply. In my model I start by calculating the thermal stress and strain, as you suggested. However, I do not know how to use the calculated stress and strain in the frequency simulation. I have tried to find a tutorial that explains this, but without any success. Are you aware of one? Best regards Tom

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 Nov 30, 2010, 1:34 a.m. EST
Hi

Store/save your stress calculation, add a new solver case eigenfrequency and tell the solver to start from the "stored solution", thenit will linearise the stress values.

Take a buckling axample there you ahve COMSOl setting up a stationary and an eigenfrequency calculation and be inspired ;)

--
Good luck
Ivar
Hi Store/save your stress calculation, add a new solver case eigenfrequency and tell the solver to start from the "stored solution", thenit will linearise the stress values. Take a buckling axample there you ahve COMSOl setting up a stationary and an eigenfrequency calculation and be inspired ;) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 8, 2010, 4:49 a.m. EST
Hi

Thanks for your help. I think I am making progress. Now I can change the temperature of the structure and then calculate the eigenfrequency. However, when I plot the deformed structure after calculating the stress and strain, it looks not right.
The idea it that a beam with a high coefficient of thermal expansion (AL) is support at both ends by a material which has a lower coefficient of thermal expansion (Si). When the temperature of the structure is lowered the beam will contract more then the supporting structure. The will generates a tensile stress in the beam. Due this I would expect a straight beam after the temperature has been lowered. This is not the case, see the attached model.

Do you know what I am doing wrong?

Best regards
Tom
Hi Thanks for your help. I think I am making progress. Now I can change the temperature of the structure and then calculate the eigenfrequency. However, when I plot the deformed structure after calculating the stress and strain, it looks not right. The idea it that a beam with a high coefficient of thermal expansion (AL) is support at both ends by a material which has a lower coefficient of thermal expansion (Si). When the temperature of the structure is lowered the beam will contract more then the supporting structure. The will generates a tensile stress in the beam. Due this I would expect a straight beam after the temperature has been lowered. This is not the case, see the attached model. Do you know what I am doing wrong? Best regards Tom


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 Dec 8, 2010, 12:14 p.m. EST
Hi
in the model you sent there is no HT or HS so I do not see how you stresss the beam thermally.

I would propose that you start very simple, just with yut Al beam, add some prestress pulling it and make a first eigenfrequency scan with a parameter loop.

Thereafter you can make it with several materials

--
Good luck
Ivar
Hi in the model you sent there is no HT or HS so I do not see how you stresss the beam thermally. I would propose that you start very simple, just with yut Al beam, add some prestress pulling it and make a first eigenfrequency scan with a parameter loop. Thereafter you can make it with several materials -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 8, 2010, 3:32 p.m. EST
Hi

Thanks

I stress the structure thermally by setting the strain reference temperature and the temperature of the structure in Solid Mechanics (solid) -> Linear Elastic Material Model 1 -> Thermal Expansion.

I will try to get ther thermal deformation correct before I look at the eigenfrequency.

Best regards
Tom
Hi Thanks I stress the structure thermally by setting the strain reference temperature and the temperature of the structure in Solid Mechanics (solid) -> Linear Elastic Material Model 1 -> Thermal Expansion. I will try to get ther thermal deformation correct before I look at the eigenfrequency. Best regards Tom

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.