[Cara Gao]

Hi, everybody. I am new in Openfoam. Recently I am using interFoam to simulate a falling droplet under gravity. However, the droplet falls much more slower than expected.
I have been stuck here for one month; really hope some one can help me out.

Attached is the p_rgh and U file.

Thanks!

Screenshot from 2014-08-28 10:45:39.jpg

Screenshot from 2014-08-28 10:46:12.jpg

[vigneshTG]

Hi Cara !!
Did you switch on gravity ? There's a file named g in constant directory, Check inside the file whether g is set as

value ( 0 -9.81 0 );

[Cara Gao]

Originally Posted by vigneshTG

Hi Cara !!
Did you switch on gravity ? There's a file named g in constant directory, Check inside the file whether g is set as

value ( 0 -9.81 0 );

Hi, Vignesh TG,

Yeah, I believe I did edit the g file in constant. I am guessing that maybe my boundary conditions for the case is not right. I set all the surrounding boundaries to be patches, and set pressure and velocity boundary conditions accordingly.

Thanks,
Cara

[Googlestuff]

Hi Cara.
Can you specify more precisely what you mean by "slower than expected?".

[Googlestuff]

I think further information about the case is necessary to help you. What does your mesh look like? How did you set the alpha field? Maybe you can upload your dictionaries? As is saw you based your simulation on the damBreak tutorial, is that right? If so, what did you change and what is standard?

Regards,
Sebastian

[Cara Gao]

Yes, the case is based on the dambreak case. In the damBreak case, there is only one patch for the side to atmosphere, the others are walls; but in my case (3D), the droplet falls in the atmosphere, so all the boundaries are patches except for the groud it falls on. And the boundary conditions are the same with that in damBreak.

p_rgh: total pressure

U: pressureInletOutletVelocty

The mesh is not the directory (too large to upload successfully). But all the other stuff are in it.

Thanks a lot for your help.
By the way, I am also working on a case about channel flow with fixed heat flux into it, but I didn't find any similar case in the tutorial. I want to set the heat flux directly instead of set a temperature gradient. Do you have any suggestion??

Thanks,

Best,
Cara

[Googlestuff]

Okay. I see you have five areas you assign boundary conditions to. Inlet, Outlet and Walls 1-3. But if you have a 3D Case there must be a sixth face. Maybe you forgot to assign a BC to that? Your dictionaries look just fine to me.

I am also simulating a falling drop under similar conditions, but i assign a start velocity to the drop so i don't have to wait for too long for the splash. Maybe that's an option for you too?

Regards!

[Cara Gao]

Dear Sebastian,

Thanks for your quick reply.
I group the walls at the same level, and name it wall 1,2,3. And then set boundary condition to these walls.
I think it will be easier if the droplet is assigned a initial velocity, but I cannot see the deformation(might be very small) at the beginning.

Thanks,
Cara

[Googlestuff]

Okay maybe i missunderstand. If you did group the walls, why did you assign the BC's to them individually? Maybe a picture of your mesh might help me to understand your geometry.

boundaryField
{
    INLET
    {
        type            pressureInletOutletVelocity;
        value           uniform (0 0 0);
    }
    WALL_1
    {
        type            pressureInletOutletVelocity;
        value           uniform (0 0 0);
    }
    WALL_2
    {
        type            pressureInletOutletVelocity;
        value           uniform (0 0 0);
    }
    WALL_3
    {
        type            pressureInletOutletVelocity;
        value           uniform (0 0 0);
    }
    OUTLET
    {
        type            fixedValue;
        value           uniform (0 0 0);
    }
}

Another thing to look at: Did you investigate the pressure and velocity fields for your simulation? Is there anything unexpected that would explain this behaviour? Hope this hints help!

Best regards,
Sebastian

[Cara Gao]

Hi, Sebastian,

I have attatched the pictures of the geometry, each picture for each boundary, except the inlet surface(because the maximam number of files is 5).

I have ever checked the presssure and velocity fields in paraFoam, and then became more confused.

Best,
Cara

[Googlestuff]

Ah ok. Now I get it

So with that configuration your BC's look reasonable.

Why were you confused looking at p and U?

I see your outlet is also a patch. I would recommend to define the type as "wall" in the blockMeshDict. I think in the tutorial they also use wall, because with this type the wall-distance is treated in a special way.

[Cara Gao]

The outlet is wall; I think in paraView it set all boundaries to be patch.

Thanks,
Best,
Cara

[Googlestuff]

So what confuses you looking at the pressure and velocity fields?

[Cara Gao]

It's like the droplet is in the areospace, the gravity is 10 times smaller. And the the accelation happens really slow.

Best,
Cara

[Elham]

Hi,

I am also simulating a falling droplet in OpenFOAM. Although the droplet is just under the action of gravity and it is in the air, but it tends to right or left depending on the max Courant number when it is falling . Can anybody help me to solve it?