- **OpenFOAM Running, Solving & CFD**
(*https://www.cfd-online.com/Forums/openfoam-solving/*)

- - **Solver for an incompressible, turbulent flow with heat transfer**
(*https://www.cfd-online.com/Forums/openfoam-solving/86043-solver-incompressible-turbulent-flow-heat-transfer.html*)

Solver for an incompressible, turbulent flow with heat transferHi there,
I think I have posted my first thread (read here) in the wrong Forum. That's the reason of this post here :-) I search a solver for an incompressible, turbulent flow with heat transfer. I still have read through the list of solvers in the User Guide), but can I use a solver from the group "Heat transfer and buoyancy-driven flows", if I have no buoyancy-driven flows? The solver i refer to is: "buoyantPimpleFoam Transient solver for buoyant, turbulent flow of compressible fluids for ventilation and heat-transfer" What would be your suggestion of solvers I should use? Can I use the settings specified in controlDict fvschemes fvsolutions etc in the tutorials of the solver for my run with the same solver? Regards tH3f0rC3 |

well maybe it would be quite good to describe what do you want to model?
What is you medium to transfer? Is it water or other incompressible fluid, or is it just air? Do you need the incompressibility assumption? To cut off the dependence of density on pressure and rely only on temperature change was idea of making the solution easier and more stable - meaning making another assumption with a negligible error. If you look into the solver files like UEqn.C, pEqn.C and hEqn.C you can (a) directly see what is solved, (b) see that it should be OK to use the solver for your case. The buoyant flows can be very turbulent if you are concerned with this. You may try to set you gravitational acceleration of zero to see the effect, and still the solver should do what you want. The thing you should be worry about is the mesh at the wall and appropriate BCs if you are interested in the heat transfer at the wall. good luck matej |

Hi matejfor,
thank you for your answer! I want to model an oven with an integrated fan. The air will be drawn in on one area an blown out directly over and under the surface which shall be heated. The aim is to describe the temperature on this surface. The medium of the flow will be air at round about 300°C. We can calculate with an incompressible medium. We can also model that there is no dependence of density on temperature. The flow will probably be turbulent. So which solver should be the best to use in your opinion? Regards, tH3f0rC3 |

your density doe not depend even on temperature and you are interested in heat transfer? well - to be honest - you a ripping off the physics significantly. If you want to really want the toothless physics you should write your own solver in foam. I think it is necessary mainly to make sure you understand what you are doing. If you do not understand all the consequences, just use the buoyantPimpleFoam without being scared of the buoyant word at the start.
Just got an idea, but I'm not sure it would work: If you really want the density to be constant with your heat-transfer computation (thus - to my knowledge- computing garbage) you may try to use buoyantBoussineqPimpleFoam with the expansion coefficient set to zero - turning the density variation to zero. have fun matej |

Hi matej,
that is the way I wanted to do it. To search a solver which represents nearly the situation I want to simulate and than modify the Solver to my criterias. I will try with your suggested modification. Regards, tH3f0rC3 |

well then another approach would be to take the pisoFoam - incompressible and turbulent and add the enthalpy equation to it.
there are more ways to play around. According to my experience it's the best to start with one and see where you got stuck and what is going to happen. With every trial and with each error you are faster coder in foam :). matej |

Quote:
where can i set the expansion coefficient set to zero and turning the density variation to zero? Is it in the 0-files? |

All times are GMT -4. The time now is 13:36. |