Multiregion Heat Transfer + natural convection (water) with chtMultiRegionFoam
 

Dear OpenFoamers,

For my case study, I want to simulate Heat transfer in a multi region solid+fluid (water) (see attached picture). The case is as follow: In a box there are water and a solid phase, which is located in the middle of the box. The right wall of the box has higher temperature compared to the others (365K compared to 300K, internal temperature 300K).

I expect the natural convection within the box due to temperature difference. And somehow the solid part will be heated. I use the chtMultiRegionFoam to solve my case. I modified the tutorial case multiRegionLiquidHeater. In the thermophysicalProperties I define water with IcoPoly3Thermo, temperature dependent datas for Cp, mu, kappa and rho. But my case is not working.

My question is: is chtMultiRegionFoam suitable for this case? Or any suggestion, solver or boundary conditions ?

Thank you in advance,
H.P.

continuity Error?
 

Hi H.P.

what do you mean with the case does not run? Are there any messages in the log? Or do you have continuity problems? That's what I would suggest. As you use a temperature dependant rho-function for incompressible water, the volume of the water changes, violating continuity. You can stick to the chtMultiRegionFoam. May be you can use a Boussinesq approximation or add an outlet?

Regards Fabian

Dear Fabian,

Thank you for your reply.

I also think the problem lies in the calculation of pressure and density, which caused to the change of volume. As u can see from the log file, the temperature was not calculated correctly. I also tried with either inletOutlet or outletInlet boundary, but it didn't help much. The same problem occurred.

I am not so sure, how the solver couples pressure, temperature and density in this case, for water as temperature dependent substance. Do u have any idea?

H.P.

Courant number
 

Hi H.P.

from what I can see in your log file, there are no continuity errors. But at the end of the run, the Courant number increases. Have you tried variable time steps? You can limit the courant number to 1 or below in the controlDict and let the solver adapt the time steps. Have a try.

Regards

Fabian

Hi Fabian,

Now I can run the case, but the heat capacity Cp is kept as constant. Within 20°C to 90°C, there is a big change in Cp. So I tried to run the case with this constant Cp. Its not so correct, but its running.

I will update if any new result comes out. Thank for your replies, very helpful. :D

Regards,
H.P.

Hi... I am interested in setting up a chtMultiRegionFoam-case.. could you upload yours or send it to me via e-mail?

Thanks!