CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Falling sphere (http://www.cfd-online.com/Forums/fluent/47951-falling-sphere.html)

Graham April 16, 2008 20:09

Falling sphere
 
Hi,

I'm a newbie to Fluent. I have a problem and I'm hoping somebody can give me some tips or point me to some relevant examples to get me started.

Very simply, I want to begin by modeling a glass sphere falling in a water tank (no flow) due to gravity, both away from the wall or near to a wall.

I can find how to model droplets or bubbles, or even dispersed particles, generated by nozzles and so on, but I do not understand how to place a single solid sphere at a position in the tank (x,y,z) at time t=0 and then model how the fluid interacts as it falls due to gravity.

Hopefully someone can show me a similar project.

This is not for a homework project.

CDE April 17, 2008 01:15

Re: Falling sphere
 
you can use the discrete phase model (DPM). Just set up an injection at a point and display the particle trajectory

Graham April 17, 2008 02:05

Re: Falling sphere
 
Right, but doesn't that assume the spheres are point-like and do not perturb the flow?

I want a fulling interacting solid sphere that is falling due to gravity and displacing the fluid. There is no flow, it's falling in an isolated tank, so I'm not sure how the particle tracking would work.

My spheres are not necessarily point-like.

Ultimately, I want to be able to model the passage of the fluid around the sphere, rather than the actual path followed by the sphere itself (although that would be a nice by-product).


CDE April 17, 2008 02:36

Re: Falling sphere
 
ok, thats quite a different problem. You are correct that the DPM assumes the volume fraction is negligible and will not perturb the fluid. Have a look at the VOF model. Maybe you can tweak the settings of the second fluid and surface tension to imitate a solid particle. This is just a suggestion and I'm not sure it will work!

Graham April 17, 2008 03:46

Re: Falling sphere
 
Yeah I was thinking of that, but it seems like something of a fudge.

I would have though a sinking sphere would be a standard kind of problem, with a well-worn solution.

I was thinking of using dynamic meshing, but I think that requires prior knowledge of the behavior of the sphere, which I might not have (for instance if it gets close to a wall).

Giles April 17, 2008 07:22

Re: Falling sphere
 
The newer versions of Fluent include a 6DOF solver, which you can set up with dynamic meshing so that you start with an initial mesh, and it will compute the flow and any extra forces you may wish to add such as gravity, and time step the mesh with dynamic meshing to model the spheres movement with time. Fluent use an example of a bomb falling from a bomb bay as an example of this is their tutorials, I'd recommend trying to find that as its very similar.

Graham April 17, 2008 18:26

Re: Falling sphere
 
Thanks.

Do you have a link for that particular example? I don't appear to have it with my installation (Fluent 6.3).

It sounds ideal. I think the 6DOF is the way forward for this problem.

Giles April 23, 2008 19:13

Re: Falling sphere
 
Sadly I've only got a a paper copy, its part of Fluent's training program for dynamic meshing. Possibly contact fluent and they may be able to supply it to you electronically.

Graham April 23, 2008 19:15

Re: Falling sphere
 
What's the name of the article? I should have access to it, but I can't see a link on their site at the moment.

Giles April 25, 2008 12:34

Re: Falling sphere
 
The one day course for dynamic meshing notes: Lecture 6: Coupled flow via the 6 DOF Solver of 1DOF UDF

and

Tutorial 5: Coupled Flow using 6DOF (store separation from a 3D delta wing)

Graham April 25, 2008 13:02

Re: Falling sphere
 
Thanks. I will look at this. I also discovered the falling box example that was enough to get me going (for now). Thanks!


All times are GMT -4. The time now is 20:14.