Air flow through a cylinder with UDF issues

MFiX How to
#1

Hello,

I have a model that is intended to use air flow through the bottom to push particles upwards through the cylinder. However, I’ve encountered a few issues and was hoping they could be resolved. Here is a video of the run.

  1. Some of these particles tend to group around the middle of the cylinder at high velocities, which caused many “Velocity exceeds limit: 2250.0” warnings. Why are they grouping together like this, and how can I fix it? I was thinking it may be due to overlap, but at the 0.06s mark, there is a single particle in the center that randomly increases in velocity without overlap. It is as though the particles are drawn to the same point. I have tried changing the mesh, changing the geometry to a primitive and altering the resolution, changing the geometry to a box, changing the time step, and changing the residuals and linear equation parameters, but none of those seem to fix it.

  2. I have a boundary condition for mass outflow set at the top of the cylinder. All of the particles reaching the top face should be leaving, but it appears that many of the particles don’t leave and fall back down. Why is that?

  3. I have set up a user defined function based on the pulsating fluid bed tutorial where the inlet velocity should be gradually increasing to reach 22.2504 m/s. However, it appears that the velocity is always 22.2504. Is there a mistake in how I defined the function, or do I have to disable the mass inflow boundary condition? Edit: removing the y-axial velocity from the inlet BC and redefining each BC so that the inlet ID was 1 fixed this particular issue.

entrainment_velocity_2023-07-19T150529.092579.zip (40.4 MB)

Thanks,

Ryan

#2

Here is a more evident example of the first problem I have listed. The particles start grouping at the center around the 0.35s mark. Changing the drag model does not appear to help either.

#3

Thanks for the detailed report. Since it’s summer, some of our team are on vacation, so it might be a little while until we can investigate this case.

1 Like
#4

I see 2 issues:

  1. There is a conflict with your geometry, where the wall cylinder is closed at the y=ymin and y=ymax planes. try to extend the height of the cylinder (say a height of 0.06m instead of 0.05m). You will also get a cleaner cylinder if you use the procedural geometry instead of the implicit shape.

  2. Instead of the Mass outflow BC, use a Pressure Outflow BC.

2 Likes
#5

Thank you for the help! Implementing your suggestions fixed the issue.