[prezza]

Hi guys

I have had a question for quite some time now. I posted it in Fluent forum, haven't got any help so far. So I really appreciate if somebody can help.

[ghorrocks]

I cannot say for other codes (eg fluent), but I can make suggestions for CFX.

But before I do I need some details:
Does the particle hit the walls? Interact with other particles? Leave the boundary? Start inside the domain? Start with an initial velocity? Any other forces on the body?

[prezza]

Originally Posted by ghorrocks

I cannot say for other codes (eg fluent), but I can make suggestions for CFX.

But before I do I need some details:
Does the particle hit the walls? Interact with other particles? Leave the boundary? Start inside the domain? Start with an initial velocity? Any other forces on the body?

Thanks for the reply.

I have a single particle not interacting with the wall. The particle shouldn't normally hit the wall. The location of the particle depends on its diameter and the lift force on it. So in the worst case it will sediment on the wall but that is not the goal of the simulation.
I have to put the particle somewhere at the beginning of simulation. Let's say very close to the wall. So as it moves along the channel it changes its height. It will not leave the boundary. Its initial velocity is zero or the velocity of the fluid. The forces on the body are drag and lift and the shear forces which cause rotation.

[ghorrocks]

What is the fluid it is in?

[prezza]

Water for the initial stage, but I intend to investigate more complex fluids later.

[ghorrocks]

OK, in that case the immersed solid approach should work fine - up until you have a variable density flow anyway.

Use the rigid body solver with a immersed solid.

[prezza]

Thanks a lot.

Just wondering why you asked about the fluid? Why is it important for the method selection and what other options are there for other fluids?

[ghorrocks]

Immersed solids are not supported for variable density flows. There are some other restirctions as well, read the documentation to check it applies to your case. The immersed solid approach is by far the easiest way to do this sort of thing so hopefully it will work for you.

[prezza]

I fortunately don't have a variable density flow. So I hope I can get my simulation done with it.

Thanks for the help.

[ebd]

Hi,

I have a similar question. I am modeling a deterministic lateral displacement device, 'bumping array', in which particles with variable diameter interact with each other as well as with periodically shifted rows of cylindrical posts.

(here is an image if clarification is needed: http://www.pnas.org/content/105/21/7434.figures-only)

I would like to be able to model particles suspended in a multiphase flow (although it is not absolutely essential) but I would appreciate any insight about which solvers will take into account drag between fluids/particles, particle-particle collisions, and particle-wall collisions.

Thanks.

[ghorrocks]

All these physics models are included in lagrangian particle tracking, but this assumes the particle is a point. If your particle size is a significant compared to the flow obstacles then this might not be an accurayc model to use.

The immersed solid approach will not work too well here either, it cannot handle collisons with walls or other particles or variable density flows.

You might need to use a DEM solver (such as EDEM) and couple it with CFX to do this model.

[ebd]

Thanks so much for input. I have been doing a bit of background research into the theory behind the immersed body model, but hadn't found any definitive answer, so I appreciate the clarification.

I have done some research into the EDEM software package, and it seems ideal for the system I am looking to model. Thanks for your help