CFD Online Discussion Forums

CFD Online Discussion Forums (
-   OpenFOAM Pre-Processing (
-   -   Speeds of 100 m / s in Oven - pimpleFoam + Energy Equation (

jrsilvio_ver February 22, 2014 01:01

Speeds of 100 m / s in Oven - pimpleFoam + Energy Equation
1 Attachment(s)
Dear, good night.
I am new to OpenFOAM and I'm simulating the air flow within a continuous furnace.
I changed the pimpleFoam by adding the energy equation, resulting in my_pimpleFoam. Is continuing with the establishment of the boundary conditions for the entry, exit, adiabatic walls (up, down, left side and right side) and burner (central tube at 1000 K). At the end, I ran the command and checked the presence of extremely high speeds, on the order of 100 m / s. What is absurd for configuring an oven, it should work in natural convection.
Could you please check the attached files? Point me any errors, wherever they are.
Attached solver my_pimpleFoam and information about the design of the oven.

Finally, it has to be air enters the furnace at room temperature and atmospheric pressure and exits with a suction pressure of 10 cmH2O.
I appreciate everyone's attention and await response.

jrsilvio_ver February 22, 2014 01:06

1 Attachment(s)
Now follows the solver my_pimpleFoam.

jrsilvio_ver February 22, 2014 03:08

Images I
3 Attachment(s)
Attached pictures of the simulation.

jrsilvio_ver February 22, 2014 03:09

3 Attachment(s)
Attached pictures of the simulation.

jrsilvio_ver February 22, 2014 03:11

2 Attachment(s)
Attached pictures of the simulation..

jrsilvio_ver February 22, 2014 03:13

2 Attachment(s)
Mesh geometry.

alexeym February 22, 2014 09:48


1. If you'd like to simulate natural convection maybe it'll be better to start with buoyantBoussinesqPimpleFoam. The solver is more or less what you are trying to do in my_pimpleFoam but it also has buoyancy terms in pressure equation.

2. Are you sure that one nOuterCorrector step is enough for your simulation to converge? I'd increase this parameter up to 50 and add residual controls for termination of outer corrector loop. Something like this:


    nOuterCorrectors 50;
            tolerance 1e-4;
            relTol 0;

jrsilvio_ver February 22, 2014 18:06

Using buoyantFoam I have built a model of heat transfer by radiation?

jrsilvio_ver February 22, 2014 21:03

Guys, another detail.
Could anyone detail me the boundary condition "totalPressure"?
In my project, for example, I know that the inlet pressure is atmospheric and do not know the speed and output'm assuming a suction pressure of 10cmH2O, but in any case I know the speeds. How can I implement this boundary condition?
I am considering the density of air at the average temperature ((1000 +300) / 2 = 650 K) equal to 0.5356 kg / m.
Someone could solve my doubts?

jrsilvio_ver February 22, 2014 22:12

Nobody has an interest in the subject in question?

alexeym February 23, 2014 07:13


Originally Posted by jrsilvio_ver (Post 476255)
Using buoyantFoam I have built a model of heat transfer by radiation?

You'd like to have radiative heat transfer in your simulation? With buoyantBoussinesqPimpleFoam you can have it.

As I can guess from your case files totalPressure BC will calculate pressure with:


    if (psiName_ == "none" && rhoName_ == "none")
        operator==(p0p - 0.5*(1.0 - pos(phip))*magSqr(Up));

I still wasn't able to understand your last question about boundary conditions.

jrsilvio_ver February 23, 2014 14:44

The buoyantBoussinesqPimpleFoam would have included heat exchange by radiation, the energy transport and the transport of momentum? I would not need to make any changes in solver?
And as the boundary conditions used in my my_pimpleFoam, correct? The issue is that I do not quite understand how to use the boundary condition totalPressure, explain to me? Preferably with an example.
I greatly appreciate your attention.

jrsilvio_ver February 23, 2014 14:46

Another question regarding buoyantBoussinesqPimpleFoam. How could I totally remove the terms involving turbulence? In the case of turbulence model, since it will be working with laminar flow.

alexeym February 23, 2014 14:58


If you just take a look at the sources in $FOAM_APP/solvers/heatTransfer/buoyantBoussinesqPimpleFoam, you will find answers to your questions (concerning equations solved by the solver and if there is radiative heat transfer). As far as I understand the problem, you do not need to modify anything.

I'd suggest you forget about totalPressure for a moment and thoroughly describe physical conditions at the inlet and outlet; maybe you need completely different set of BCs. From the previous posts I wasn't able to figure out what's happening at the inlet and outlet boundaries.

About turbulence: put laminar RASModel in constant/RASProperties. In addition you can change "turbulence on" to "turbulence off" there.

jrsilvio_ver February 23, 2014 16:59

After disabling the turbulence model, tried to run the command and got:

__________________________________________________ _________________________
Create time

Create mesh for time = 0

Reading g
Reading thermophysical properties

Reading field T

Reading field p_rgh

Reading field U

Reading/calculating face flux field phi

Selecting incompressible transport model Newtonian
Creating turbulence model

Selecting RAS turbulence model laminar
Reading field alphat

cannot find file

file: /home/silvio/Mestrado/buoyantBoussinesqPimpleFoam/hotRoom/0/alphat at line 0.

From function regIOobject::readStream()
in file db/regIOobject/regIOobjectRead.C at line 73.

FOAM exiting

__________________________________________________ _________________________

If the turbulence model is disabled, as it is reading the alphat?

alexeym February 24, 2014 02:43


Reading code can be a source of endless insights. Here is the part of TEqn.H:


    alphat = turbulence->nut()/Prt;

    volScalarField alphaEff("alphaEff", turbulence->nu()/Pr + alphat);

    fvScalarMatrix TEqn
      + fvm::div(phi, T)
      - fvm::laplacian(alphaEff, T)
        radiation->ST(rhoCpRef, T)
      + fvOptions(T)

Surely it needs alpha cause the solver was made universal (i.e. to run in laminar and turbulent cases). If you disable turbulence nut == 0 => alphat == 0, so your alphaEff is just a constant.

If you so unhappy with alphat, you can modify solver (createFields.H, TEqn.H, and buoyantBoussinesqPimpleFoam.C files) and remove any mentions of turbulence from there. Though I really doubt you really need this ;)

All times are GMT -4. The time now is 12:01.