[Lorddrinkalot]

Hey guys,

I have a problem with my simulation, and I don't know any further.

At the moment, I'm just trying to simulate a normal flow through an L-shaped geometry (image attached). For now, it's isothermal and with a B field of 0.

The boundary condition for U

dimensions      [0 1 -1 0 0 0 0];

internalField   uniform (0 0 0);

boundaryField
{
    inlet
    {
        type            fixedValue;
        value           uniform (0.02 0 0);
    }
    outlet
    {
		type    zeroGradient;
        //type          pressureInletOutletVelocity;
		//value			uniform (0 0 0);
    }
    Wall_side
    {
        type            noSlip;
    }
    Wall_out
    {
        type            noSlip;
    }
    Wall_inside
    {
        type            noSlip;
    }

}

Conditions for the p_rgh

dimensions      [1 -1 -2 0 0 0 0];

internalField   uniform 0;

boundaryField
{
    inlet
    {
        type            fixedFluxPressure;
		value			uniform 0;
    }
    outlet
    {
	    type            fixedValue;
        value           uniform 0;
	    //type            fixedFluxPressure;
		//value			uniform 0;
        //type            inletOutlet;  //outletInlet
		//inletValue		uniform 0;
    }
    Wall_side
    {
        type            fixedFluxPressure;
		value			uniform 0;
    }
    Wall_out
    {
        type            fixedFluxPressure;
		value			uniform 0;
    }
    Wall_inside
    {
        type            fixedFluxPressure;
		value			uniform 0;
    }


}

I have tried several conditions with inlet/outlet, etc., but nothing helps.

For the first few time steps, everything seems good, but then after some time, the pressure flips or does something strange (see images). The inlet is at the bottom, and the outlet is on top of it. So, the flow comes in at the bottom half of the L and leaves at the top half.

I am new to OpenFOAM, and I can't figure out why this is happening.

I'm using the buoyantPimpleMhdFoam (MHD and heat transfer), but both are switched off for now to check and see if the case is running, which it is not.

Boundary condition: no heat transfer, so it's constant.

internalField   uniform 573;

boundaryField
{
    inlet
    {
        type            fixedValue;
		value			uniform 573;
    }
    outlet
    {
        type    zeroGradient;
    }
    Wall_side
    {
        type    zeroGradient;
    }
    Wall_out
    {
        type    zeroGradient;
    }
    Wall_inside
    {
        type    zeroGradient;
    }



}

The B field is set to 0 in the conditions (internalField uniform (0 0 0).

Please let me know if you need anything else.

Can someone help me with this kind of problem??


Thanks, Christoph

[Lorddrinkalot]

Hey,

I added a GitHub repository for the case to see the different files.
https://github.com/Lorddrinkalot/OpenFoam_Case.git

I should mention that I am using the Boussinesq approximation.
I also tried it with icoPolynomial, and in that case, the solver is working. However, for my case, I need the Boussinesq equation of state, and in that case, some weird stuff is happening.

[Roman1]

With the other solvers (ex. fireFoam) p_rgh must be approximately the same as normal pressure, not zero. This is a common pressure corrected with the height.

[wkernkamp]

Originally Posted by Roman1

With the other solvers (ex. fireFoam) p_rgh must be approximately the same as normal pressure, not zero. This is a common pressure corrected with the height.

Yes, and isn't that a compressible solver (to model buoyancy).

[Lorddrinkalot]

So, hey guys,

I fixed my problems, or rather, I let it run longer. My problem was that I didn't let the Simulation run long enough in the business approximation. After simulation for some hours, the Pressure P-rough looks rigth