[cctv]

Hi
I am wondering if there is a method to move the boundary at a specified speed and change the mesh automatically in openFoam. I mean the most side wall move, not a solid body in liquid. In that case, the whole simulation domain would change, not partial domain.(something like piston moving in the vessel). I know a solver named icoDymFoam could address body moving in liquid issues, but I don't know if it could be used to simulate the situation I described above. if so, how to. If not, if there exists any other methods to fullfill it.
Thank you.

[nlinder]

Hi,

do you know this tutorial?

OpenFOAM-2.2.x/tutorials/incompressible/pimpleDyMFoam/movingCone

Is that what you are looking for?

Greetings
Nicklas

[cctv]

Originally Posted by nlinder

Hi,

do you know this tutorial?

OpenFOAM-2.2.x/tutorials/incompressible/pimpleDyMFoam/movingCone

Is that what you are looking for?

Greetings
Nicklas

Yes, I know this,but this is a solide body in the liquid. Not what I want. I need, for example, the inlet boundary moving.

[nlinder]

Yes, but don't you just need to apply the pointMotionU to whatever boundary you want?

[cctv]

Originally Posted by nlinder

Yes, but don't you just need to apply the pointMotionU to whatever boundary you want?

Yes，though the boundary moved, the result is not correct. Have you ever tried before? e.g. I want no body in liquid and the most left boundary move at a specified speed? In addition, this case is dynamic mesh with cell compressed, how about with layer addition and removal?
Thanks.

[nlinder]

I use this procedure to move a boundary. Basically I compress the fluid between two plates. My moving boundary is a solid, but not embedded in fluid. I just thought that it would work the same way for any other boundary. I don't know what boundaries you set therefore I cannot say if it will work for your case. Also if you say the results are not correct it's difficult to help.. You need to be more precise..

Greetings

[cctv]

Originally Posted by nlinder

I use this procedure to move a boundary. Basically I compress the fluid between two plates. My moving boundary is a solid, but not embedded in fluid. I just thought that it would work the same way for any other boundary. I don't know what boundaries you set therefore I cannot say if it will work for your case. Also if you say the results are not correct it's difficult to help.. You need to be more precise..

Greetings

Would you please have a look at it and see anything wrong in it?
In addition, I am wondering how to specified the motion of the boundary and is there a way to address boundary wall motion with layer addition and removal way?
Thanks.

[nlinder]

Hi cctv,

I had a look at your case, but I do not understand what you try to do.

It is a 2D case. We have 3 boundaries:
- movingWall
- fixedWall
- farField

But if I got you right from your previous posts, you don't want to move a wall, correct? So maybe you can give me a hint what each boundary is plant to do?

Thanks
Nicklas

[cctv]

Originally Posted by nlinder

Hi cctv,

I had a look at your case, but I do not understand what you try to do.

It is a 2D case. We have 3 boundaries:
- movingWall
- fixedWall
- farField

But if I got you right from your previous posts, you don't want to move a wall, correct? So maybe you can give me a hint what each boundary is plant to do?

Thanks
Nicklas

Hi Nicklas, I do want to move a wall, I just want to simulate a case similar to a piston moving in a vessel. as the liquid is incompressible, the fixedWall's U boundary condition, used as the outlet, is inletOutlet. Thank you.

[nlinder]

First problem was a typo in the constant/dynamicMeshDict file

"libfvMotionSolvers.so"

correcting this, you can go on correcting the wrong/missing parts of your setup (fvSolution is missing some entries, fvSchemes as well and the pointMotion should be a vector, not a scalar). Make sure you set a totalPressure BC for the deforming walls and set the other BCs as desired (outlet or whatever).

Greetings
Nicklas

[cctv]

Originally Posted by nlinder

First problem was a typo in the constant/dynamicMeshDict file

"libfvMotionSolvers.so"

correcting this, you can go on correcting the wrong/missing parts of your setup (fvSolution is missing some entries, fvSchemes as well and the pointMotion should be a vector, not a scalar). Make sure you set a totalPressure BC for the deforming walls and set the other BCs as desired (outlet or whatever).

Greetings
Nicklas

Thank you for your reply. there is no typo in it because what I send you is a case run in 1.6-ext. what you said is how to run in 2.1.0 or even higher version. however ,I could try it in this version. What's more, do you know how to move the wall with layer addition and removal methods? just like movingbodysolver. Thanks.

[nlinder]

Would have been easier if you told that before...

This is what I got as a "result":
https://dl.dropboxusercontent.com/u/587565/ani.gif
(2.2.x)

About layer addition or removal I don't know much. I only used dynamicMeshRefinenemt for adding or removing cells. I don't know if the combination works as well..

Have fun

[cctv]

Originally Posted by nlinder

Would have been easier if you told that before...

This is what I got as a "result":
https://dl.dropboxusercontent.com/u/587565/ani.gif
(2.2.x)

About layer addition or removal I don't know much. I only used dynamicMeshRefinenemt for adding or removing cells. I don't know if the combination works as well..

Have fun

Thank you very much. Could you tell me how to add or remove cells? and where is the code they map the old field into the new field?
Good luck

[mturcios777]

This may be of to interest to you:

http://www.cfd-online.com/Forums/ope...n-removal.html

You may have an easier time making things work as you are using 1.6-ext (I assume from previous posts). There may be a movingConeTopo mesh class; if not you will need to make your own.

[Bodo1993]

Hello, I want to study the effect of moving a solid wall out of a 2D fluid domain. My solver does not have the dynamic meshing library. What could be a work around for the above?
Thanks

[VishJay]

Originally Posted by nlinder

I use this procedure to move a boundary. Basically I compress the fluid between two plates. My moving boundary is a solid, but not embedded in fluid. I just thought that it would work the same way for any other boundary. I don't know what boundaries you set therefore I cannot say if it will work for your case. Also if you say the results are not correct it's difficult to help.. You need to be more precise..

Greetings

Hi Nicklas,
I am trying for months to model something similar to what you have done “ compress the fluid between two plates"(FSI). My case is for cold metal forming and I need to analyse lubrication flow (film thickness analysis) between a die and a deformable (Plastic deformation) wall/body. So I need to simulate with topological changes of deforming solid and fluid domains. I have adapted solids4Foam tutorial case upsetBillet (elastoplasticity) for solid domains which is exactly what I need. To integrate with fluid I followed tutorial case 3dTube case setup, but I am not sure yet what boundary types/conditions I should apply for fluid domain. Could you kindly advise me to setup this case?
In order to setup moving boundaries I tried totalPressure BC for the deforming walls but yet not able to model deformable walls.
I have attached my geometry of the model and boundary conditions I have used. Please advise..
Thank you,