export mesh data for an explicit selection

May 15, 2013, 12:01 pm

≪ Previous: PML Error: Could not find any cartesian regions

Dear all,

I am a relatively new user COMSOL with a lot of Finite Element experience both on the programming and theory level. I started with COMSOL today which went actually quite fine and fast.

I would like to use that to interface assemled system matrices and mesh data to my own solvers. And it was successful for simple shell models.

At this point, I am also wondering if that is possible to export some selected node data or not. What I am directly interested is 'can I create node groups in COMSOL' which can be exported in some kind of text data file format or in binary file format?

Let me be more specific, lets say I have two fixed edges of a square plate and I would like know which nodes are associated with this selected fixed edges so that I can extract the constrained dof infornation myself. These kinds of information might give me more freedom for simple FRF analysis that I would like to conduct.

I learned that I can extract the matrices Kc and K, for instance, for the system stiffness matrix which represent the matrix on which the boundary conditions are applied and the full system matrix without any boundary conditions applied, respectively.

I would like to know or compute the fixed dof information by either getting it from COMSOL(which would be nice since no extra computations will be necessary.) or by iterating over the nodes myself.

Any ideas or pointers are appreciated on this matter.

Best regards,
Umut

↧

Weak Constraint thermal flux

May 16, 2013, 5:30 am

≫ Next: comsol element reference and theory manual

≪ Previous: export mesh data for an explicit selection

Hi every body

I am using Weak form to model strongly coupled thermal-structural equations.My model consists of two layers.
I used Weak constraint to ensure that my thermal flux will be continuous but
Comsol gives me an error that "Expression must be linear in test function"..and i dont know how to solve it.
What does it mean?.Do I have to change a shape function?if yes,how can I do that

I used (q22-q2)*test(u3), where u3 is dependent variable for temperature and q variable for heat flux of each layer.

Coud you please help me?

Thank you

↧

comsol element reference and theory manual

May 16, 2013, 8:10 am

≫ Next: Adding points and lines to 3D models with parametric surfaces

≪ Previous: Weak Constraint thermal flux

Dear all,

Depending on the module that is selected, the elements are selected automatically also I guess.

I have several questions related to this proposition if that is right, namely,

+ Is there a list of all the elements that are used in COMSOL? With the node ordering and related stuff? To interface the mesh to external software?

+ For instance, I conducted a modal analysis on a shell model and it uses 2nd order quad elements by default, which I could understand from the dof and node correspondence. Is there a way to select 1st order elements?

+ Also for the structural mechanics module, is there a way to do this selection to change the order?

+ There is also no theory manual for COMSOL, right, that would also be nice to show exactly what is implemented in the software.

Best

↧

Adding points and lines to 3D models with parametric surfaces

May 16, 2013, 11:35 am

≫ Next: System of PDEs with mixed order

≪ Previous: comsol element reference and theory manual

Hi all,

I was wondering if there is a way to add a point or line to a model in which I have a 3D block with the top surface replaced by a parametric surface derived from a DEM (digital elevation model).

I guess this can be done in either the geometry section so it is there for the entire model run, or in the results section under data sets.

The end result I am after is the displacements along a line that traverses the parametric surface or at a certain point on the parametric surface.

Any help in this matter would be greatly appreciated, and let me know if I haven't explained the problem sufficiently.

Thank in advance!

James.

↧

System of PDEs with mixed order

May 16, 2013, 7:03 pm

≫ Next: Creep - Plasticity

≪ Previous: Adding points and lines to 3D models with parametric surfaces

Hello,

I am trying to solve the attached system of PDEs using the general PDE interface. However, I always get a singular matrix error. What am I missing?

↧

Creep - Plasticity

May 15, 2013, 8:11 am

≫ Next: Electric potential and current density analysis

≪ Previous: System of PDEs with mixed order

I want to use the non-linear structural mechanics module for simulation of plastic deformation by yielding. There is a creep module and a plasticity module. In the creep module, I find more suitable approaches. Could someone explain to me the difference between creep and plastic deformation? Until now I thought that it is the same but I am not an expert in this field.

↧

Electric potential and current density analysis

May 16, 2013, 11:40 am

≫ Next: Capacitive sensor - coupling with circuit

≪ Previous: Creep - Plasticity

Dear friends,

I analyzed a model using Time-dependent solver and for the time: range (0, 0.016,1). The model is consisted of different layers with different conductivity and permittivity and a Terminal with an AC current (square waveform, Amplitude:4.4 mA, angular frequency:377Hz ) is applied on top of the layers. When I click on the slice plot the amount of Electric Potential is equal to 14V. But when I plot Line Graph the value of electric potential for the same edge is about KV!!!!? Another question is that why do we have such a current density the same as the attached figure.Could somebody help me to understand the difference? Is there any document about analyzing such the plots?

Regards,

↧

Capacitive sensor - coupling with circuit

May 17, 2013, 12:07 pm

≫ Next: Help for a better 3D mesh

≪ Previous: Electric potential and current density analysis

Hello,

I build a small model of a capacitive sensor (small conducting plate). An electric dipole serves as the source of an electric field.

The field induces a surface current density at the sensor plate. But i cannot figure out how i couple this current with an attached circuit (see attached figure of circuit).

I tried lumped ports (electrical circuit module) and terminals (electric currents module), and coupled those with an external I vs U node.

But since the sensor only has one contact (one node, see figure) I don't think lumped ports and external I vs Us are approriate.

Would be nice if someone could give me a hint. Thanks.

↧

Help for a better 3D mesh

May 18, 2013, 8:37 am

≫ Next: Unable to implement moment on the system!

≪ Previous: Capacitive sensor - coupling with circuit

Hello! I am a beginner in COMSOL and i need some help with a 3 D mesh!

In my model i have to focus on the cylinders and around them.

So i need finer mesh for this area but coarser mesh for the rest in order to have less elements.

Furthermore i try to have element quality bigger than 0.3.

I do appreciate any suggestions to fix my mesh!

Thank you for your help !

↧

Unable to implement moment on the system!

May 17, 2013, 5:09 am

≫ Next: Plot

≪ Previous: Help for a better 3D mesh

Hi,

I am simulating a model of a structure which has a central rotating shaft and other components attached to it. The shaft is rotating by a motor and hence the other components should rotate with it too. I am facing a couple of problems with this model:

1. When I carry out a Stationary Analysis, it returns the error:'The relative residual is greater than the relative tolerance.'
2. Then, when I try with Time dependent Analysis instead, it computes the results, but incorrectly, as the shaft doesn't rotate but the other components attached to it move under their own weight.

I am unable to understand where I am going wrong. I have attached the model . Any kind of help would be appreciated.

Thanks in advance!

Regards,
Amitosh

↧

Plot

May 18, 2013, 3:58 am

≫ Next: How to set Point dipole polarized in xy-plane in Comsol 3.5?

≪ Previous: Unable to implement moment on the system!

Hi everybody

I simulate the rib waveguide with mode analysis and in picture i attack it ,is a 3D plot.
I want to convert it to 1D plot in x and y direction.How can i do it?I think it is so easy but i dont know how can i do?please help me .

Thanks in advance for your responses and helping.

↧

How to set Point dipole polarized in xy-plane in Comsol 3.5?

May 19, 2013, 11:58 pm

≫ Next: Inlet outlet mass flux is not identical

≪ Previous: Plot

I have tried to set the power source of LED as polarized Point dipole in xy-plane in Comsol 3.5, but do not know the step, Could you help me please? or can you tell me which model library to refer to?

↧

Inlet outlet mass flux is not identical

May 13, 2013, 9:24 pm

≫ Next: Mass extraction and saturation of bulk solution

≪ Previous: How to set Point dipole polarized in xy-plane in Comsol 3.5?

Hi all,

I am trying to measure the bulk temperature, but initial testing on the conservation of mass failed as it gave me different results between the inlet and the outlet mass flux. This result makes me doubt all my simulation results so I would like to know what's going on. The model is a rectangular duct flow with 1) heat transfer on one side and 2) prescribed temperature at one of the walls and the channel inlet.

I used two 'general projection' planes to measure the mass flow rate over the entire channel length. As you can see in the graph, there is a sharp rise at x=0 (not to mention the fluctuations. I would say it's a numerical uncertainty due to the meshing). I don't understand why this is happening as this happens even if I refine the mesh, which is very concerning.

Any suggestions on why there is a discrepancy in the mass flux along the channel? Please let me know if there is any mistakes in my model.

Thanks!

Best,
Tae Jin Kim

ps: I also quantified the bulk temperature based on some discussions threads and extensive trial and error. At this point, I think it's almost correct, but please let me know if there is an error.

↧

Mass extraction and saturation of bulk solution

May 14, 2013, 2:22 am

≫ Next: Shallow water waves model in coefficient PDE or general PDE

≪ Previous: Inlet outlet mass flux is not identical

Hi all,

I am fairly new to COMSOL. I am after solving a non steady-state packed bed extractor with fluid flow convection in the axial direction of the extractor and diffusion of soluble material from the solid particles.

The first thing I am trying is to solve the diffusion equation in 2D with different boundary conditions, as if the particles were in a stirred vessel.

My question is:

When a forced convection boundary condition is imposed, how can be the saturation of the bulk fluid be accounted for?

Manually the process would be

1) Bulk concentration of the fluid is 0 at time 0
2) After the first time step has elapsed, the solution of the diffusion equation gives a total flow (mass released) out of the particle. That mass divided by the volume of the vessel gives the bulk concentration for the next time step.

In COMSOL, how can I update the bulk concentration by imposing that it would be the volume integral over the particle divided by the volume of the vessel for each time step? Is there a way to create a for loop of should I combine this problem with Matlab?

Thanks in advance

B

↧

Shallow water waves model in coefficient PDE or general PDE

May 14, 2013, 2:49 am

≫ Next: Particle Selection

≪ Previous: Mass extraction and saturation of bulk solution

Dear All,
I want to solve the Shallow water waves equations in PDE module in 2D. h is the water elevation, U and V the velocity components along x and y.

How can I implement an incoming pulse in the times domain. Coming from the left on the structure.

If I use the initial conditions, I can specify the shape at given time but I don t see any way of defining for example a incoming wave from the left to the right with a shape like:
h(x,y,t)=exp((t-t0)^2)sin(2*pi*freqt)

Best regards
Muamer

↧

Particle Selection

May 14, 2013, 3:48 am

≫ Next: PML Error: Could not find any cartesian regions

≪ Previous: Shallow water waves model in coefficient PDE or general PDE

Hello,

I running a simulation to release particle in an electric field and observe their tracks. I wish to find out how many particle and how fast they reach a certain boundary. I have been able to determine the drift time for each particle, but have not found out how to select particle based on the boundary they first collide with. Is there a simple way to do this in COMSOL?

Many thanks!

↧

PML Error: Could not find any cartesian regions

May 14, 2013, 7:19 am

≫ Next: Creep - Plasticity

≪ Previous: Particle Selection

Hi

I am new to PML. I am using Structural mechanics module to implement PML in a cantliver to determine the waves propagating to the PML. I am getting this error: Could not find any cartesian regions. My dimensions are in micro meter but using conversions to use it in meters. Please suggest what am I missing here.

Is there any model on PML using SM module?
Thank you.