Understanding cfdemSolverPisoScalar and packedBedTemp tutorial

Hello,

I'm new to OpenFOAM/LIGGGHTS and I'd like to use CFDEM for simulations of fluid-particle interaction with both momentum and temperature coupling between fluid/particle phases.

I am starting with the packedBedTemp tutorial with the cfdemSolverPisoScalar solver to understand CFDEM's out-of-the-box capabilities. However, I don't fully understand what is happening in this tutorial problem, and would be grateful for some insight on the following questions:

1) To isolate the heat transfer phenomenon, I set the CFD inlet velocity and initial velocity conditions to zero, such that the CFD domain is just a fluid at rest. When I run the problem like this, the hot particles heat up the fluid, but only where the DEM domain intersects the CFD domain. The temperature of the CFD domain ahead of and behind the particle domain remains at the initial temperature value, so it seems as if the conduction term in the fluid is not active. Upon looking at the source code for the solver, it seems that all the standard heat equation terms are there. So I was wondering what is happening with this behavior, and why am I only seeing heat transfer between particles and fluid, and not within the fluid itself via conduction? Note that when I run the problem with original settings, the CFD domain behind the DEM domain remains at initial temperature while the CFD domain ahead of the DEM domain heats up, indicating that heat transfer via convection due to fluid flow seems to be working normally.

2) In the same example, run with the original settings of the tutorial, the particles do not seem to move at all. Are the particles too heavy to be influenced by the fluid flow in this case, or is momentum coupling between particles and fluid somehow turned off?

3) What is the difference between the solvers cfdemSolverPisoScalar and cfdemSolverPisoSTM? They seem to give nearly identical results in the two packedBedTemp tutorial cases associated with them.

Any insights would be greatly appreciated. Thank you.

Forums:

CFDEM®coupling - User Forum

bmercer | Wed, 01/11/2017 - 23:35

I ramped up the inlet

I ramped up the inlet velocity and got particle motion, so the momentum coupling is on and item (2) above is resolved. But I would still be interested in hearing anyone's thoughts on items (1) and (3). Thank you!

bmercer | Sat, 01/14/2017 - 01:01

Ok I figured out that the

Ok I figured out that the thermal diffusivity (therefore conductivity) was low due to the fluid representing air, thus causing a perceived lack of conduction. Increasing the thermal diffusivity lead to heat conduction in the fluid like I was expecting. So item (1) is resolved.

I now have a good handle on what the solver is doing, but I am still a little confused about item (3), the difference between cfdemSolverPisoScalar and cfdemSolverPisoSTM. Does anyone have any insight on this? As far as I can tell, *STM has 'species transport' but I don't really understand how this influences the problem.

Thank you!

cgoniva | Thu, 01/19/2017 - 12:55