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.
Marangoni Convection on a surface
Posted Oct 4, 2011, 5:30 p.m. EDT Wave Optics, Heat Transfer & Phase Change, Studies & Solvers 3 Replies
Please login with a confirmed email address before reporting spam
Hi,
i am trying to simulate the Marangoni convection on the surface of a metal sheet, which is heated by a laser. The model is threedimensional and the Incompressible Navier-Stokes and the Convection and Conduction modules are used in a stationary analysis in COMSOL 3.5.
According to the "Marangoni Convection" guidance in the model library the surface tension is modeled by a PDE Mode Weak Form, but in 3D.
In 2D the equation in the weak form is: lm_test*(eta1*uy-gamma*Tx)+u_test*lm
For the 3D Model i tried it with 2 dependent variables lm1 and lm2 for the surface tension in the XY-plane in this way:
lm1_test*(eta1*uz-gamma*Tx)+u_test*lm1
lm2_test*(eta1*vz-gamma*Ty)+v_test*lm2
Unfortunately it is not possible to find a solution. I suppose the problem deals with the boundary conditions. In the 2D model the Dirichlet Boundary Condition is omitted at the vertex between the boundary with surface tension and the boundary with fixed temperature by changing the equation system: (s<1)*(T0_cc-T)
How can i adapt this to the 3D model? I tried (z<0.02)*(T0_cc-T), but it doesn't work.
What other problems should cause the problem?
Best regards and thanking you in anticipation,
Clemens
i am trying to simulate the Marangoni convection on the surface of a metal sheet, which is heated by a laser. The model is threedimensional and the Incompressible Navier-Stokes and the Convection and Conduction modules are used in a stationary analysis in COMSOL 3.5.
According to the "Marangoni Convection" guidance in the model library the surface tension is modeled by a PDE Mode Weak Form, but in 3D.
In 2D the equation in the weak form is: lm_test*(eta1*uy-gamma*Tx)+u_test*lm
For the 3D Model i tried it with 2 dependent variables lm1 and lm2 for the surface tension in the XY-plane in this way:
lm1_test*(eta1*uz-gamma*Tx)+u_test*lm1
lm2_test*(eta1*vz-gamma*Ty)+v_test*lm2
Unfortunately it is not possible to find a solution. I suppose the problem deals with the boundary conditions. In the 2D model the Dirichlet Boundary Condition is omitted at the vertex between the boundary with surface tension and the boundary with fixed temperature by changing the equation system: (s<1)*(T0_cc-T)
How can i adapt this to the 3D model? I tried (z<0.02)*(T0_cc-T), but it doesn't work.
What other problems should cause the problem?
Best regards and thanking you in anticipation,
Clemens
3 Replies Last Post Oct 7, 2011, 9:52 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi,
sincerely I wasn't able to solve the problem yet.
Is it the right approach to model the surface tension by 2 lagrange multipliers in 3D or is there a more suitable solution?
Best regards and thanking you in advance,
Clemens
sincerely I wasn't able to solve the problem yet.
Is it the right approach to model the surface tension by 2 lagrange multipliers in 3D or is there a more suitable solution?
Best regards and thanking you in advance,
Clemens
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi,
there is a well-documented example for the Marangoni convection in the Model Library:
www.comsol.com/showroom/downloadfile/model/110/version/comsol35a/file/marangoni.mph
www.comsol.com/showroom/downloadfile/model/110/version/comsol35a/file/marangoni_sbs.pdf
www.comsol.com/showroom/downloadfile/model/110/version/comsol35a/file/marangoni.pdf
Best wishes, Marcel
there is a well-documented example for the Marangoni convection in the Model Library:
www.comsol.com/showroom/downloadfile/model/110/version/comsol35a/file/marangoni.mph
www.comsol.com/showroom/downloadfile/model/110/version/comsol35a/file/marangoni_sbs.pdf
www.comsol.com/showroom/downloadfile/model/110/version/comsol35a/file/marangoni.pdf
Best wishes, Marcel
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi,
I started with the Maragoni model from the model library, but it is two-dimensional. In the 2D model the boundary conditions at the vertices have to be changed.
Afterwards I tried to adapt this to a my 3D model. Sincerely I don't know how to change the boundary conditions in the 3D model.
My second request is about the formulation of the lagrange multipliers:
lm1_test*(eta1*uz-gamma*Tx)+u_test*lm1
lm2_test*(eta1*vz-gamma*Ty)+v_test*lm2
instead of lm_test*(eta1*uy-gamma*Tx)+u_test*lm in the 2D model.
Is this the right way to formulate the lagrange multipliers?
I have read the chapter about weak contributions in the reference guide, but there is no further information about how lagrange multipliers work. Are there any restrictions for lagrange multipliers, which could cause the problem?
Best regards,
Clemens
I started with the Maragoni model from the model library, but it is two-dimensional. In the 2D model the boundary conditions at the vertices have to be changed.
Afterwards I tried to adapt this to a my 3D model. Sincerely I don't know how to change the boundary conditions in the 3D model.
My second request is about the formulation of the lagrange multipliers:
lm1_test*(eta1*uz-gamma*Tx)+u_test*lm1
lm2_test*(eta1*vz-gamma*Ty)+v_test*lm2
instead of lm_test*(eta1*uy-gamma*Tx)+u_test*lm in the 2D model.
Is this the right way to formulate the lagrange multipliers?
I have read the chapter about weak contributions in the reference guide, but there is no further information about how lagrange multipliers work. Are there any restrictions for lagrange multipliers, which could cause the problem?
Best regards,
Clemens