Liquid in a cyclone

[JuhaHo]

Is there any development on this issue?

I started this all over again and got the results as seen on the attachment. Now, what I need to change to get this behave more according to laws of physics?

[Svetlin.Nikolov]

Hello,

How can we help?

Any additional information you could share would be more than welcome, otherwise it's like us looking at our crystal ball...

Cheers!

[JuhaHo]

Thanks,

I don't know what additional information is required. The liquid source is on right side of the cyclone and the output is on the top. Small amount of liquid falls down and there's another source that gives additional speed to liquid going down. In cyclone, the down stream should be circulating the walls and upstream in the middle. The liquid is non-newtonia.

I know the speed of the incoming liquid. Is it possible to calculate correct outgoing velocity value for the liquid?

[JuhaHo]

The simulation is advancing, so here's updated screenshot with stream lines. I hopes this gives better idea and helps to give me instructions how to make Phoenix imitate natural physics. As well, the recommendation for outgoing velocity is needed..

[Svetlin.Nikolov]

I don't know what additional information is required. The liquid source is on right side of the cyclone and the output is on the top. Small amount of liquid falls down and there's another source that gives additional speed to liquid going down. In cyclone, the down stream should be circulating the walls and upstream in the middle. The liquid is non-newtonia.

1. Please share the settings of the Simulator and the Sources.
2. What emitters do the Sources use? Please show me how they look in the scene.
3. Are there any forces in the scene now or you removed them?
4. Is there any tyFlow simulation or other particle systems except for Phoenix in the scene?

As for the outgoing velocity, the idea you shared with support is exactly what I would advise you.

Cheers!

[JuhaHo]

Thanks Svetlin, Let's start with simulator. Here's the latest settings. I started with Paint preset. The material is pulp having consistency of 5% (95%being water).

[JuhaHo]

I realized, it's totally different thing to see the liquid filling the volume in free air compared to the situation where the liquid is flowing through the filled system. Obviously it has to behave a lot differently, since the liquid doesn't fall freely once filled It's not easy to observe how the liquid is moving when it's filled. This goes to rendering as well as the liquid model on UI. Here for instance, I can't see what's happening in the narrow gap, because there is no streamlines going through.

[Svetlin.Nikolov]

If there is no streamline showing, are you sure the Detail Reduction at the top of the preview rollout is not increased? Try lowering it down to 0 and temporarily turning off Auto Reduction so it doesn't automatically increase again (see more info here https://docs.chaos.com/display/PHX4MAX/Liquid+Preview).

In addition to the free air test you did, try running the simulation initially with 0 viscosity, 0 non-newtonian, 0 surface tension and disabled wetting. Verify that it's doing what you expect. Then start enabling these options one by one and simulate after each one, so you can see at what point the simulation starts doing something you don't expect.

Cheers!

[JuhaHo]

Thanks,

Here's screenshot with more dense streamlines. I wonder what is happening on red lines section.

The blue cylinder on the left is the lower source, while the box blocks liquid from going up.

Regarding my remark on free air simulation. It's not necessary to see filling the system - only the flow instead. What I meant, is that when looking filling simulation one might get wrong idea of how this works. Since the fluid is non-newtonian, I have increased the value from zero. Basically the material is water, that is used for transferring other solid particles. This cyclone is used for separating heavy particles such as stones from the fluid.

[Svetlin.Nikolov]

Hmm, does the geometry really get narrow where the red part of the streamlines is?

Could you please run an STL check modifier over all involved geometries to make sure there are no errors, or they would break the simulation.

Cheers!

[JuhaHo]

Yes, the geometry gets narrow. Only small portion liquid goes through that gap along with stones, while the rest - the cleaned liquid - travels through the upmost connection. That's why there's additional flushing water source below the narrow gap.

The geometry is not CAD-geometry. I have modeled it by myself in Max according original CAD-geometry. Then I have voxelised the geometry to make sure it's completely closed, watertight and has proper thickness according Phoenix requirements (2x voxel size).

So what are the red streamlines in the small gap telling me?

[Svetlin.Nikolov]

It's expected - fluid would flow faster if the pipe gets narrower: https://en.wikipedia.org/wiki/Bernoulli%27s_principle

[JuhaHo]

It's expected - fluid would flow faster if the pipe gets narrower: https://en.wikipedia.org/wiki/Bernoulli%27s_principle

So, the red streamlines mean that the fluid is going faster at that point? Do I know in which direction it's flowing just by looking the streamlines?

[Svetlin.Nikolov]

Yess, red is the fastest, green is medium speed and blue is the slowest.

Right now there are no arrows on the streamline preview, we have a note to make such. For now you can turn off the Streamlines and show the Velocity arrow preview.

[JuhaHo]

It's expected - fluid would flow faster if the pipe gets narrower: https://en.wikipedia.org/wiki/Bernoulli%27s_principle

I think, the Bernoulli principle doesn't work with Non-Newtonian liquids.