Free surface liquid wont move

Free surface liquid wont move

Anonymous
Not applicable
2,380 Views
16 Replies
Message 1 of 17

Free surface liquid wont move

Anonymous
Not applicable

Hi i'm having problems with the free surface on my model, specifically with HOF, in the cfz file, the fluid is supposed to move from the cube that is a little higher into the tube and exit in the other end. Even when the tube has no HOF the fluid wont enter unless a big pressure is applied, as a boundary conditions, in one of the faces of the higher cube. Can someone help me see why is this happening?

0 Likes
Accepted solutions (1)
2,381 Views
16 Replies
Replies (16)
Message 2 of 17

Jon.Wilde
Alumni
Alumni

Hi,

 

A few points to make to get this working:

  1. You need to enable Free Surface (Solve -> Physics -> Free Surface)
  2. Set Gravity (I assume this will be 0,-1,0)
  3. Refine the mesh - check out the video here on VOF analyses for more details. We need to be sure to have a really fine mesh anywhere where the free surface can exist

Jon

0 Likes
Message 3 of 17

Anonymous
Not applicable
Hi, tank you for the reply.
I tried refining the mesh and setting up the free surface but the results
are more or less the same, do you think that the pressure configuration can
be the problem? The cube that is higher should be at atmospheric pressure,
is an open container, and the HOF should be enough to make the liquid flow.
0 Likes
Message 4 of 17

Jon.Wilde
Alumni
Alumni

Hi,

 

Take a quick look at your geometry - hide the main parts and you will see some strange surfaces/volumes floating inside. These ought to be removed in CAD. (Saying that though, they do not seem to appear in my results so it might be OK as is).


Have you considered running a 1/2 model? It would run in half the time 🙂

 

I made the changes I described before and refined the mesh. I gave it all a uniform mesh and gave the tubes a finer mesh still and it appears to be working fine. I just noticed that the top of the central section is not filled with water - so it probably could be meshed far coarser - we really need the finer mesh where the free surfaces can exist.

 

I think that really this is all down to mesh - even this is not fine enough to accurately capture the free surface.

 

Mesh.png

 

This is the beginning of the analysis:

Initialised.png

 

And this is it after running a few minutes (I'm stopping it now):

Water moving.png

 

Kind regards,

Jon

 

Message 5 of 17

Anonymous
Not applicable

Hi,
Thanks, I'll try refining the mesh even more. By running half a model do you mean without the top half of the central region? Or making a transversal cut of the model?
Thank You for your help.

0 Likes
Message 6 of 17

Jon.Wilde
Alumni
Alumni
Hi,

No problem.
By a half model I mean a vertical slice, so you're left with something useable 🙂
Apply a slip/symmetry boundary condition to the exposed face.

Thanks,
Jon

________________________________

Autodesk Limited
Registered Office: One Discovery Place, Columbus Drive, Farnborough, Hampshire GU14 0NZ
Registered in England and Wales, No. 1839239
0 Likes
Message 7 of 17

Anonymous
Not applicable

Hi,

I've tried with a finer mesh but still the liquid won't move. The entrance container must be open, can I set a boundary condition for atmospheric pressure in the upper face? maybe it help the liquid move.

Thank you for your time.

0 Likes
Message 8 of 17

Jon.Wilde
Alumni
Alumni

Hi,

 

I do not have the file to refer back to now (I only keep these forum models a month) but I do not think I had the top open. We should be able to drain fluid out regardless. It would not act like it was pulling a vacuum or anything, the fluid should just leave a void.

Are you changing the free surface default solver/physics settings at all?

 

Kind regards,

Jon

0 Likes
Message 9 of 17

Anonymous
Not applicable

Halfie.cfz is the configuration that i ran before, and Halfie 2.cfz is the one that i'm running now, with the static pressure in the upper side of the entrance. Both start filling the exit cube but don't empty the entrance cube. And i don't know if its a problem with the mesh or with the geometry file.

 

Thank you for your help

0 Likes
Message 10 of 17

Jon.Wilde
Alumni
Alumni

I would think this is the same issue as before - the mesh needs to be much finer where there is a free surface. Uniform and fine or CFD cannot capture the free surface. We should not really need that P=0.

 

Also - I highly recommend you halve the model, it would run in half the time and the mesh that will be required almost makes this a necessity.

 

Saying this - I am testing it here to comfirm my thoughts.

 

 

0 Likes
Message 11 of 17

Anonymous
Not applicable

Ok, i'll try that. 

Thank you

0 Likes
Message 12 of 17

Jon.Wilde
Alumni
Alumni

Hi - could you also try having a small air gap above the water in the full tank?

0 Likes
Message 13 of 17

Anonymous
Not applicable

Hi, the volume should be just above the liquid level or at the top of the tank? which conditions should i set?

0 Likes
Message 14 of 17

Jon.Wilde
Alumni
Alumni
Accepted solution

Hi, have a new volume, above the liquid volume that is not moving. This would have no initial condition assigned to it.

Message 15 of 17

Anonymous
Not applicable

Ok, i'll try that.

Tanks

0 Likes
Message 16 of 17

Anonymous
Not applicable

That solved the problem Smiley Happy it even runs with a coarser mesh.

 

Really thank you for your help, i've been stuck there for a couple of months, its my first project but is an important one.

0 Likes
Message 17 of 17

Jon.Wilde
Alumni
Alumni
No problem at all, very pleased to hear this is now working! 🙂

________________________________

Autodesk Limited
Registered Office: One Discovery Place, Columbus Drive, Farnborough, Hampshire GU14 0NZ
Registered in England and Wales, No. 1839239
0 Likes