How to simulate air flow in a pipe in a fluidised bed

Hello, I would like to simulate a fluidised bed heat exchanger with immersed tubes. However I always get the message that there are no standard fluid cells in the computation. I am using STL files for the geometry. Is there any easier way to simulate air flow in the pipe, maybe in 2D ? Any help is appreciated. I’ve included the file below for reference. I’m using the SMS meshing workflow based on tutorial 3.8.
cubalagi_2025-03-18T174857.186987.zip (8.4 MB)

Your mesh is too coarse. There is only one cell in the z-direction, which means the pipe diameter is smaller than the cell size and therefore cannot be resolved by the mesh. You need to refine the mesh in all directions so the mesh size is much smaller than the pipe diameter.

I am also not sure what is the purpose of the hx wall region, but since it is a rectangular box, it may not be needed.

I’m trying to make it as a 2 dimensional simulation so that’s why there is only 1 cell in the z-direction. Is it better to move the pipes into the other axes ? Thank you for your help regardless, appreciate it.

You could take a a thin slice in the z-direction, say z_min=0.07m and z_max=0.08m and keep 1 cell in the z-direction.

Now, you can’t have fluid on both sides of the wall though, zero thickness walls are not supported. You can model the flow inside the pipe, or outside the pipe, not both, unless you resolve the thickness of the wall. You would need a mesh finer than the wall thickness. Heat transfer across the wall is not supported, so MFiX may not be the right tool for this application.

I see, thank you for your help once again. :grin:

Does this mean it is possible to simulate flow inside and outside of the pipes if we create a pipe with thickness ?

Yes, but you will need a very fine mesh if the pipe wall is thin. You will need to apply boundary conditions along the inner and outer wall, i.e. the heat transfer within the pipe wall cannot be modeled.

Ah I see. Since the heat transfer within the pipe wall cannot be modeled, let’s say if we simply simulate the sand fluidised bed flowing outside of the pipe walls (no air flow inside the pipe), is it possible to observe the temperature change at the outer pipe walls due to the hot sand ? Because previously I managed to capture the change in temperature for the sand but no for the pipe wall.

I’ve attached a zip file here
indirectheatexchanger_2025-03-26T145844.748444.zip (4.3 MB)
of what I’m trying to do to help better illustrate it. Many thanks for your help Jeff.