natural convection in closed cavity with chtMultiRegionFoam and variable properties
 

Hallo everyone,

since this is my first post in this forum I'd like to thank all of you first, who make this forum so helpful for beginners in OpenFOAM, like me.

My problem:
My aim is to simulate a geometry, which resembles a closed cavity with water in it and relatively thick walls. The walls are cooled from all sides. Flow shall be laminar and i want to use variable thermophysicalProperties for water.

I think the best solver for this is the chtMultiRegionFoam. But it doesn't work. So i build a simplified case geometry with the tutorial fitting to my case to identify the problem (see attached file). The problem seems to be with the pressure (p and p_rgh) initiated by the use of variable thermophysicalProperties for water (solid ones are kept constant) as the solver breaks after one or two time Steps. But with const. variables it works better.

To test my polynomial thermophysicalProperties I also created a case from the tutorial of buoyantPimpleFoam/hotRoom (see attachments) but it has the same problem. Since my temperature range is between 237 and 300 K the Boussinesq approximation would not fit to the problem.

I also found this thread: http://www.cfd-online.com/Forums/ope...roperties.html
about problems with the variable properties.

Does anyone have experiences with natural convection in closed cavities for fluids with variable properties in OpenFoam and has some hints for me how i can modify it to receive a converging solution?

I'm using OF 2.3.1

I hope this is enough information about my problem. Otherwise feel free to ask. I would be thankful for any help. Thank you in advance!

Mike