Heat conduction through Sand or pulverised materials
 

Hi, everybody

Currently I should simulate the heat conduction through the pulverised materials (like sand) in CFX. I don´t think it´s a good idea to model it as a continous solid because of the gap between the sand grains. So my opinion now is to simulate it as a multiphase flows (model the sand the dispersed solid or transport particle solid and set the air as continous fluid).
Because I am a new user in the CFD area, so I would like to ask, does any of you have any experiences of simulating such problems?

Thanks a lot,
Dawei

What are you actually trying to do? There are many different approaches and you have not provided enough information to make an informed decision. Is the sand in air? Or water? Is it moving? What is driving the heat transfer? What is the important outcome of the analysis?

hi, ghorrocks

What I want to do is actually, I set up a stationary container made of steel. There are sand in the container. It is under the atmosphere condition (1 bar, 20 C). And the sand inside can not be modelled just as a continuous solid due to the fact that it is actually millions of small sand grains in the reality, there are gaps between each grain. So we assume it is air, which fills these gaps. As boundary condition I set a heat flow to the outer wall of the container, the sand inside will be heated up. As outputs the temperature profile in the sand should be simulated.
So I´d like to discuss, which possibilities are there in CFX to simulate this. What I have found is model the sand as dispersed solid or particle transport solids, then it will be taken as multiphase flow, which sometimes leads to convergence problems

Thanks!!

as a first stab at this, multiphase setup and for initial conditions implement a porosity factor based on your sand grain shape ( if assumed spherical particles porosity is somewhere 0.65 )

I don't think a multiphase approach to this is wise.

Personally I would either do experiments or look in the literature for thermal conductivity values for sand at the appropriate density and grain size. Then model it as a simple solid using the thermal conductivity and specific heat which represents a sand grain/air pile. This is very easy to do.

quite true, to be honest I was with the impression that the sand/air was a fluidized bed setup but reading the problem description again I agree with glenn.

Yes, that is actually what I have been doing until now. Now I also think it is not necessary to use multiphase flow to simulate this.
Thanks for the answer, Glenn and George