Menu

Discussion Forum

New Discussion
Filter

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 The information provided may be out of date.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.

No eigenpair solution selected

Posted Dec 16, 2012, 3:04 a.m. EST Studies & Solvers, Structural Mechanics Version 4.0 7 Replies

Behnam Z
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi

I'm new in COMSOL. I want to find out eigenfrequencies of a pinned-pinned rectangular concrete beam.
this beam has 5X5 cm cross section. It is a simple problem but I have some problems in getting good results.

Boundary conditions are assigned to edges at two ends.
I selected frequency domain model & eigenfrequency as study.
I used swept mesh from edges that is simply supported.

earlier, I was selected eigenfrequency as study and I encountered with this message:
Failed to evaluate variable Jacobian.
- Variable: mat.rho
- Geometry: 1
- Edge: 3 5 7 8

these edges are longitudinal edges (with no support). So, I used frequency domain model, too. But I have problem yet.

Can everybody help me?
Thanks
7 Replies Last Post Dec 17, 2012, 10:32 a.m. EST
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 Dec 16, 2012, 3:13 a.m. EST
Hi

looks like you have not defined materials for all your domains, check again the material node, and then the physics node

--
Good luck
Ivar
Hi looks like you have not defined materials for all your domains, check again the material node, and then the physics node -- Good luck Ivar
Behnam Z
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 16, 2012, 3:23 a.m. EST
Dear Ivar,

Thanks for your soon reply. But I defined Concrete from library with density, young's modulus and poisson ratio.
In geometry scope, I selected All domains. Is it true?

Dear Ivar, Thanks for your soon reply. But I defined Concrete from library with density, young's modulus and poisson ratio. In geometry scope, I selected All domains. Is it true?
Behnam Z
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 16, 2012, 4:24 a.m. EST
Dear Ivar;

this is a screenshot from my model.
material properties and some other features are shown.
Dear Ivar; this is a screenshot from my model. material properties and some other features are shown.

Attachments:

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 Dec 16, 2012, 9:18 a.m. EST
Hi

I missed the "edge" word in your error message, I stopped at "rho". The issue is that you use a "beam" phyiscs that is defined on edges only, while you are in 3D with a "domain" = volume, but the "domain" for "beams" are lines = edges.

So first of all you must define the material on the EDGES in the material node, as these are by default set to the highest order of the geometry = 3D volume = DOMAIN as you can see on your image too.

I agree this is confusing, the COMSOl developers could perhaps improve this, but it's far from obvious how, as "beam" phyiscs is special physics using geometrical entities 1 or 2 dimensions lower the the default geometrical domain entitiy.

So start to select edges for the material, as beams are lines=edges in 3D (or 2D if you have defined highest order = 2D)

And read through carefully the doc about geoemtrical objects, fem entities, physics on domains (but domains size depends on physics too) ...

--
Good luck
Ivar
Hi I missed the "edge" word in your error message, I stopped at "rho". The issue is that you use a "beam" phyiscs that is defined on edges only, while you are in 3D with a "domain" = volume, but the "domain" for "beams" are lines = edges. So first of all you must define the material on the EDGES in the material node, as these are by default set to the highest order of the geometry = 3D volume = DOMAIN as you can see on your image too. I agree this is confusing, the COMSOl developers could perhaps improve this, but it's far from obvious how, as "beam" phyiscs is special physics using geometrical entities 1 or 2 dimensions lower the the default geometrical domain entitiy. So start to select edges for the material, as beams are lines=edges in 3D (or 2D if you have defined highest order = 2D) And read through carefully the doc about geoemtrical objects, fem entities, physics on domains (but domains size depends on physics too) ... -- Good luck Ivar
Behnam Z
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 17, 2012, 7:06 a.m. EST
Thanks a lot Mr. Kjelberg
It is worked correctly. I selected all edges instead of all domain.

Now, natural frequencies is very greater than values expected. I calculated natural frequency of first mode about 274 Hz by means of formulas, but COMSOL results are shown 1382 Hz as fundamental frequency. units are same in both methods.

I think, if I choose "Solid mechanics" instead of "Beam", better results would be extracted. Is this true?

Thanks very much
Thanks a lot Mr. Kjelberg It is worked correctly. I selected all edges instead of all domain. Now, natural frequencies is very greater than values expected. I calculated natural frequency of first mode about 274 Hz by means of formulas, but COMSOL results are shown 1382 Hz as fundamental frequency. units are same in both methods. I think, if I choose "Solid mechanics" instead of "Beam", better results would be extracted. Is this true? Thanks very much
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 Dec 17, 2012, 7:16 a.m. EST
Hi

Beam are slightly tricky in the beginning, but normally quite OK and solves so much faster. A common error is getting the inertia wrong as there is a coordinate mapping for beam, check the doc, where X,Y,Z are pointing in "beam Space"

It's true that solid is tempting for many cases, due to the power of our computer today

--
Good luck
Ivar
Hi Beam are slightly tricky in the beginning, but normally quite OK and solves so much faster. A common error is getting the inertia wrong as there is a coordinate mapping for beam, check the doc, where X,Y,Z are pointing in "beam Space" It's true that solid is tempting for many cases, due to the power of our computer today -- Good luck Ivar
Magnus Ringh COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 17, 2012, 10:32 a.m. EST
Hi,

If the image shows a cantilever beam, then it is modeled as a 3D solid, and you should use a Solid Mechanics physics and define the material properties in the solid. If you use the Beam physics, COMSOL uses special beam elements, and you model the cantilever beam as a simple straight edge and define the cross-section properties in the physics settings (and the material properties for the single edge). Both approaches work and give the same eigenfrequencies, but a Beam model in general gives accurate results using much fewer elements than a full 3D model.

Best regards,
Magnus Ringh, COMSOL
Hi, If the image shows a cantilever beam, then it is modeled as a 3D solid, and you should use a Solid Mechanics physics and define the material properties in the solid. If you use the Beam physics, COMSOL uses special beam elements, and you model the cantilever beam as a simple straight edge and define the cross-section properties in the physics settings (and the material properties for the single edge). Both approaches work and give the same eigenfrequencies, but a Beam model in general gives accurate results using much fewer elements than a full 3D model. Best regards, Magnus Ringh, COMSOL

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