[nazmulislam]

Hi,


I am a new user in OpenFoam. I am trying to investigate laminar flow through a axisymmetric periodic tube. Since this is a periodic tube, I am investigating the flow in one wavesection, the flow will be the same for other wavesection.



I have set the noSlip condition on the tube wall and wedge condition on the two wedges (I am studying 2D flow). However, I am not sure how to set periodic condition at the inlet and outlet. Since, this is a periodic tube, the velocity will be the same at the inlet and outlet.



I have searched online and found that I need to use "cyclic" boundary condition at the inlet and outlet. But I am not sure how to impose it at the inlet and outlet with a specified velocity there (inlet and outlet), say the velocity along the pipe axis is 1 m/s.


Any suggestion will be very helpful for me at the beginning of my OpenFOAM journey.


Thanks.

[nazmulislam]

I have set the u file in the 0 folder as follows:


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

internalField uniform (0 0 0);

boundaryField
{
inlet
{
type cyclic;
value uniform (1 0 0);
}

outlet
{

type cyclic;
}

pipewall
{

type noSlip;
}

wedge0
{
type wedge;
}



wedge1
{
type wedge;
}

}



and the p file in the same directory as follows:


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

internalField uniform 0;

boundaryField

{

inlet
{
type cyclic;
}

outlet
{
type cyclic;
value uniform 0;
}

pipewall
{
type zeroGradient;
}

wedge0
{
type wedge;
}

wedge1
{
type wedge;
}

}

[nazmulislam]

Also, I have setup the surface patches for different surfaces in the boundary file in the polyMesh folder as follows:



5

(

wedge0
{
type wedge;
physicalType wedge;
nFaces 1008;
startFace 1465;
}



outlet
{
type cyclic;
neighbourPatch inlet;
nFaces 7;
startFace 2562;
matchTolerance 0.0001;
}

inlet
{
type cyclic;
neighbourPatch outlet;
nFaces 7;
startFace 2569;
matchTolerance 0.0001;
}

wedge1
{
type wedge;
physicalType wedge;
nFaces 1008;
startFace 2487;
}

pipewall
{
type wall;
physicalType wall;
nFaces 46;
startFace 3495;
}

)

[nazmulislam]

When I run the solver 'simpleFoam' in the terminal, I get the following message:


--> FOAM FATAL IO ERROR:

Unable to set reference cell for field p

Please supply either pRefCell or pRefPoint


file: /home/nazmul/OpenFOAM/nazmul-10/run/symmetricPipe/system/fvSolution/SIMPLE from line 46 to line 53.

From function bool Foam::setRefCell(const volScalarField&, const volScalarField&, const Foam::dictionary&, Foam::label&, Foam::scalar&, bool)

in file cfdTools/general/findRefCell/findRefCell.C at line 102.


FOAM exiting

[nazmulislam]

I added the two lines in the "SIMPLE" block in fvsolution file:


pRefCell 0;
pRefValue 0;

[nazmulislam]

However, I received that "Sorry, the application simpleFoam has stopped unexpectedly"


Any suggestion will be very helpful for me.

[nazmulislam]

Hi,


I have applied the cyclicAMI boundary conditions (since I have unstructured mesh) for the velocity and pressure at the inlet and outlet for the periodic tube, which is now working.


But I had to put uniform velocity inside the tube domain initially, because if I set zero velocity inside the tube (internal velocity) initially then the solver stop after 1 execution with zero velocity and pressure everywhere.


However, I tried to set uniform velocity in the x-direction (along the tube axis) at the inlet and outlet with zero internal velocity initially, it seemed to me it was not working. That is, it was stopped after 1 execution with zero velocity everywhere.


internalField uniform (0 0 0);


inlet
{
type cyclicAMI;
value uniform (1 0 0);
}

outlet
{
type cyclicAMI;
value uniform (1 0 0);
}



Is there any way to implement the above conditions with cyclicAMI boundary conditions?


I really appreciate your help.

[mAlletto]

This function object ensures a given mean velocity. Hope it helps

https://www.openfoam.com/documentati...ity-force.html

[mfrl]

please remember that you are solving a mixed problem (Initial + boundary value problem), therefore the initial conditions must satisfy your BC's and in your case it will also drive the problem. Please try to start the flow with the desired uniform velocity. After a while it should converge to a fully developed turbulent/laminar velocity profile. In the Schlichting book you also can find the theoretical solution.
hope it helps

mfrl

Originally Posted by nazmulislam

Hi,


I have applied the cyclicAMI boundary conditions (since I have unstructured mesh) for the velocity and pressure at the inlet and outlet for the periodic tube, which is now working.


But I had to put uniform velocity inside the tube domain initially, because if I set zero velocity inside the tube (internal velocity) initially then the solver stop after 1 execution with zero velocity and pressure everywhere.


However, I tried to set uniform velocity in the x-direction (along the tube axis) at the inlet and outlet with zero internal velocity initially, it seemed to me it was not working. That is, it was stopped after 1 execution with zero velocity everywhere.


internalField uniform (0 0 0);


inlet
{
type cyclicAMI;
value uniform (1 0 0);
}

outlet
{
type cyclicAMI;
value uniform (1 0 0);
}



Is there any way to implement the above conditions with cyclicAMI boundary conditions?


I really appreciate your help.

[nazmulislam]

Originally Posted by mfrl

please remember that you are solving a mixed problem (Initial + boundary value problem), therefore the initial conditions must satisfy your BC's and in your case it will also drive the problem. Please try to start the flow with the desired uniform velocity. After a while it should converge to a fully developed turbulent/laminar velocity profile. In the Schlichting book you also can find the theoretical solution.
hope it helps

mfrl

Hello,


Thanks for your reply with your suggestion. I think, it will help me a lot to solve my flow problem.


However, could you please tell me how can I identify my "desired uniform velocity"?


Also, did you mention the book "Boundary Layer Theory" by Schlichting? If so, could you please specifically let me know where can I get the theoretical solution in this book relevant to my flow problem?


Cheers

[nazmulislam]

Originally Posted by mAlletto

This function object ensures a given mean velocity. Hope it helps

https://www.openfoam.com/documentati...ity-force.html

Hello,


Many thanks for providing me the specific suggestion regarding my problem along with the link for obtaining the detailed information.
However, I tried to add the following lines for the internal velocity field but I got some error message:


internalField
{
type meanVelocityForce;
selectionMode all;
fields (U);
Ubar (1 0 0);
}


It seems to me I could not use this object correctly. Can you help me a little bit further.


Cheers

[piu58]

I don't see a reason for an unstructured mesh. Your case can be meshed easily with blockMesh. I added the theory for your case from Hermann Schlichting, 1951.

[nazmulislam]

Originally Posted by piu58

I don't see a reason for an unstructured mesh. Your case can be meshed easily with blockMesh. I added the theory for your case from Hermann Schlichting, 1951.

Hello,


Thanks for your suggestion. I will try to create mesh with blockMesh.
Also, thanks for providing me the documents of equations for Poiseiuile flow.


Cheers