Stirring simulation

Support Particles and Physics Simulations
1

Hi! I am relatively new to Blender and during fluid simulations I have stumbled upon a problem with stirring modelation. Most likely I just don’t know something really obvious, but I have strugled with this for two days straight!!

My problem:

I want to animate liquid inside a reactor, which is stirred by the impellers.

What have I tried:

  1. I have created a domain (see attached picture).
  2. Inside the domain I placed a cylinder with an open top, defined it as Obstacle (Shell/Volume tried everything).
  3. Defined the axle, baffles, impellers and other parts, which come in contact with the fluid as obstacle.
  4. Then tried two things:
    a) Placed a smaller cylinder in the mentioned cylinder and defined it as fluid.
    b) Placed a UV Sphere in the cylinder and defined it as Inflow.
  5. Placed some location/rotation keyframes for rotation.
  6. Tried baking at various resolutions from 100 to 250.

After simulation:

  1. If the fluid was created according to a), It drops into the hollow cylinder and wobbles around during the simulation and it is clear that no strirring is present, but the impellers rotate.
  2. If the fluid was created according to b), It just doesen’t flow into the cylinder (tried turning up the -Z velocity nothing goes).

Please, If somebody has some ideas, please, let me know!!! Let me know, if additional info is needed! Using Blender 2.78.

scr_1.png1346×694 179 KB

2

I can see that the scale on your domain is not applied (transofrm panle 1,1,1.7) hit ctrl a choose scale. Do this for all object in your scene.

The cylinder meant to hold the fluid should be manifold with proper normals. You could just add a solidify modifier, or model it. As it is now the normals are pointing the wrong way, so the sim may get confused.

Your impellors should have thickness and preferably be manifold too.

Also change viewport display to final, otherwise all your high resolution baking can’t be seen in the viewport.

3

Thank you! I will try doing what you suggested. Will get back with the results! Hope everything is going to be well))

4

Here’s a basic fluid spinning with propeller setup for anyone interested. It is baked at a resolution of 160 for 100 frames which takes about 15 minutes. For testing you can lower resolution ~50 - 90 and rebake.
Fl-13-simplefluid.blend on dropbox

5

Thank you Photox! I have done what you suggested and everything worked out fine! The problem was in the non-manifold geometry.