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.

point velocity condition in Incompressibel flow 2D model.

Please login with a confirmed email address before reporting spam

I am trying to solve for the flow field in the geometry as shown in the attached file. I have managed to fix all the boundary conditions except for one. At point 4 in the model I need to specify a velocity. But point settings in COMSOL allows me to only specify the point pressure. Can anyone please help as to how I can specify the point velocity???

Thanks


5 Replies Last Post 8 mar 2012, 04:52 GMT-5
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 19 nov 2009, 17:01 GMT-5
Hi

I'm not sure its possible like that, normally a "point" heritates some of its features from the edges arriving there, (just as edges heritate from the adjacent surfaces ...) and if the logic of the velocity along these edges do not match, the point values is undefined.

I.e. your point 3 belongs to a "noslip" edge, hence velocity is normally "0" by heritage, point 4 is higher up, do you expect a linear, parabolic or exponential velocity distributionalong edge 2 ? and probably also 1 by symmetry ?

Perhaps there is another way to define your velocity ? along an edge ?

Good luck
Ivar
Hi I'm not sure its possible like that, normally a "point" heritates some of its features from the edges arriving there, (just as edges heritate from the adjacent surfaces ...) and if the logic of the velocity along these edges do not match, the point values is undefined. I.e. your point 3 belongs to a "noslip" edge, hence velocity is normally "0" by heritage, point 4 is higher up, do you expect a linear, parabolic or exponential velocity distributionalong edge 2 ? and probably also 1 by symmetry ? Perhaps there is another way to define your velocity ? along an edge ? Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 19 nov 2009, 17:30 GMT-5
I get your point. But my outer boundary is not a solid body...its just a layer of fluid. Hence I have left that as an open boundary with no viscous stresses. But to evaluate it, I must have a the velocity at the point 4 ( which is not a solid bondary ). Could you suggest any technique I can implement this along an edge which would give me the same outcome?

Thanks

Reeshav
I get your point. But my outer boundary is not a solid body...its just a layer of fluid. Hence I have left that as an open boundary with no viscous stresses. But to evaluate it, I must have a the velocity at the point 4 ( which is not a solid bondary ). Could you suggest any technique I can implement this along an edge which would give me the same outcome? Thanks Reeshav

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 19 nov 2009, 18:23 GMT-5
I am not sure the kind of physics you are trying to model. I suppose you are not talking about free surface/interface modeling, right? (if you are, this is a much more complicated area- you should perhaps first familiarize yourself with kinematic and traction boundary conditions, then move on to implementation by a suitable method such as level set, or a variation of the ALE mesh to move the interface)

In any case, you cannot specify a velocity at a point as this doesn't seem to make physical sense - think about how can you get a certain mass or momentum flux out of a "point".

If you are simply modeling the flow in a C shaped segment, it should work just by setting inflow and outflow conditions on boundaries, not points. Perhaps you can explain more exactly what you are trying to model.



I get your point. But my outer boundary is not a solid body...its just a layer of fluid. Hence I have left that as an open boundary with no viscous stresses. But to evaluate it, I must have a the velocity at the point 4 ( which is not a solid bondary ). Could you suggest any technique I can implement this along an edge which would give me the same outcome?

Thanks

Reeshav


I am not sure the kind of physics you are trying to model. I suppose you are not talking about free surface/interface modeling, right? (if you are, this is a much more complicated area- you should perhaps first familiarize yourself with kinematic and traction boundary conditions, then move on to implementation by a suitable method such as level set, or a variation of the ALE mesh to move the interface) In any case, you cannot specify a velocity at a point as this doesn't seem to make physical sense - think about how can you get a certain mass or momentum flux out of a "point". If you are simply modeling the flow in a C shaped segment, it should work just by setting inflow and outflow conditions on boundaries, not points. Perhaps you can explain more exactly what you are trying to model. [QUOTE] I get your point. But my outer boundary is not a solid body...its just a layer of fluid. Hence I have left that as an open boundary with no viscous stresses. But to evaluate it, I must have a the velocity at the point 4 ( which is not a solid bondary ). Could you suggest any technique I can implement this along an edge which would give me the same outcome? Thanks Reeshav [/QUOTE]

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 20 nov 2009, 23:17 GMT-5
thanks fir your support. What i am trying to do is to find the concentration gradients in the annulus. Basically the inner half circle is a solid collector and the annulus is a layer of fluid which has suspended particles in it. So I am using something called a Happel cell model for my simulation which by definition states that the outer fluid boundary must have no vorticity. Which is why I have put boundary 6 and 3 as no viscous force open boundary. However i have a flow which is gravity assisted and has some value V at point 4. Is there any other way I can solve this problem?

Thanks
thanks fir your support. What i am trying to do is to find the concentration gradients in the annulus. Basically the inner half circle is a solid collector and the annulus is a layer of fluid which has suspended particles in it. So I am using something called a Happel cell model for my simulation which by definition states that the outer fluid boundary must have no vorticity. Which is why I have put boundary 6 and 3 as no viscous force open boundary. However i have a flow which is gravity assisted and has some value V at point 4. Is there any other way I can solve this problem? Thanks

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 8 mar 2012, 04:52 GMT-5
Hi!!
Have you resolved this problem? How? I have a similar one, I am simulationg the Happel's model too, but I can't implement the correct boundary conditions in the external circle.
Thanks a lot!
Francesca
Hi!! Have you resolved this problem? How? I have a similar one, I am simulationg the Happel's model too, but I can't implement the correct boundary conditions in the external circle. Thanks a lot! Francesca

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.