[tom.opt]

Hi All,
I'm currently trying to simulate NASA rotor 37 in rhoSimpleFoam with k-omega SST Model.

I've had several goes at it but every time my model crashes after a few thousand iterations.
I tried Mass flow inlet and static pressure outlet as well as total pressure inlet and static pressure outlet but both have failed.
Has anyone managed to successfully simulate rotor 37 or rotor 67 in OpenFOAM?

If yes, could you please share the boundary condition files you used or give me some hints

many thanks,
Tom

[hjasak]

Hi,

Have a look at:

Josip Žužul: Numerical evaluation of the performance curve for the NASA rotor 67, Faculty of Mechanical Engineering and Naval Architecture, University of Zagreb, 2017

on my CFD research group web site

https://foam-extend.fsb.hr/dissemina...tudent-thesis/

There is also a direct download link for the Thesis in English with an extended abstract in Croatian:

http://foam-extend.fsb.hr/wp-content...aster_2017.pdf

Hope this helps,

Hrv

[tom.opt]

Originally Posted by hjasak

Hi,

Have a look at:

Josip Žužul: Numerical evaluation of the performance curve for the NASA rotor 67, Faculty of Mechanical Engineering and Naval Architecture, University of Zagreb, 2017

on my CFD research group web site

https://foam-extend.fsb.hr/dissemina...tudent-thesis/

There is also a direct download link for the Thesis in English with an extended abstract in Croatian:

http://foam-extend.fsb.hr/wp-content...aster_2017.pdf

Hope this helps,

Hrv

This is great thanks.
I just managed to compile the extended version and am going through the tutorials.

[tom.opt]

has anyone ever managed to do rotor 37 in OpenFoam I wonder?
I've already been struggling for over a month with little luck.
Perhaps the solver can't cope with this geometry/flow.

[stephan_aubin]

Hi to all
I am also interested in running a rotor67 to learn compressible rotating openFOAM.
Tom > did you succeed ?
Stéphan

[tom.opt]

Hi Stephane,
I have not managed to get a satisfactory result with the current version of OpenFoam.
I'm afraid that for rotor 37 the problem was that mass doesn't conserve when using periodic boundaries. I tried with both the regular and extended edition.
I suggest trying to simulare a full annulus mesh if you want to do rotor 37 ,if you have the computational power.
The rotor 67 test case is slightly easier to mesh and simulate, I suggest you have a look at the above referenced paper. It is quite good

[stephan_aubin]

Dear Tom
Thanks for your inputs.

Frankly, Rotor37 is a must to validate compressible turbomachines. It is well documented and works on all the industrial solvers...

Wandering on the forum I found some info about the works done by Matvey Kraposhin, a member of this group.
I tried to contact him at it appears he has a good compressible MRF solution that also works for low Mach Numbers...

I keep you posted !

Stéphan

[tom.opt]

Originally Posted by stephan_aubin

Dear Tom
Thanks for your inputs.

Frankly, Rotor37 is a must to validate compressible turbomachines. It is well documented and works on all the industrial solvers...

Wandering on the forum I found some info about the works done by Matvey Kraposhin, a member of this group.
I tried to contact him at it appears he has a good compressible MRF solution that also works for low Mach Numbers...

I keep you posted !

Stéphan

Thanks,
if you managed to get rotor 37 validated in openfoam please let me know.
I need to get this testcase running so I can test some flow control models.

Tom

[stephan_aubin]

OK !!
I will no problem !
Stéphan

[tom.opt]

Originally Posted by stephan_aubin

OK !!
I will no problem !
Stéphan

Hi,
have you managed to make any progress on rotor 37 ?

Tom

[stephan_aubin]

Dear Tom
no but because I haven’t had the time to work on it.
thans for asking!
Stéphan

[ersoflow]

Dear all


Today I am starting to work on rotor 37 case in openfoam.
I will be also updating you about my progress.


Some initial questions:

How did you get the geometry/mesh?

If you got it from the tables of NASA technical reports (I assume by using .curve files), what tool did you use to generate geometry/mesh ?


Regards.


Ersin

[tom.opt]

Originally Posted by ersoflow

Dear all


Today I am starting to work on rotor 37 case in openfoam.
I will be also updating you about my progress.


Some initial questions:

How did you get the geometry/mesh?

If you got it from the tables of NASA technical reports (I assume by using .curve files), what tool did you use to generate geometry/mesh ?


Regards.


Ersin

HI Ersin,

I got the geometry/mesh from a colleagues but I think it should be readily available on the internet as a CAD file.
To mesh it I have tried several software packages:
PADRAM(which is an in house code that I got for turbomachinery)
GridPro
SnappyHex Mesh

The best mesh was obtained with GridPro after a considerable amount of work, however if you are good with SnappyHex mesh then that's free to use and can also get decent meshes.

To export into OpenFOAM I usuaully first save in fluent format the mesh then fluentToFoam.

[ersoflow]

Hello Tom.
Thanks for your reply.

I got the .curve files for hub, shroud and profile from the internet.
I realize that it is easy to generate mesh in Turbogrid by using.curve files.
Now I am try to export the mesh into openfoam, however Turbogrid does not export meshes in .msh format. Maybe I should first import it to fluent.

Regards.

Ersin

[tjuAeron]

Originally Posted by tom.opt

Hi,
have you managed to make any progress on rotor 37 ?

Tom

Hi Tom,

Have you successfully got results on rotor37? I'm working on the same problem but can't reach a converged simulation. The initial residual of pressure is as large as 0.2 after 5000 timesteps.

[ersoflow]

Hello all,

In this post, I would like to speak of my progress on Rotor 37 work.
Hopefully that helps someone.

- Geometry is obtained as curve files from the internet (can be found easily).

- TurboGrid is used to generate mesh. It reads curve files for hub, shroud and blade profile.

- To transform the mesh from TurboGrid into openfoam, I found my way by intertransforming to cfx and fluent, and lasty fluentMeshToFoam.

- Now, I am having several runs with steadyCompressibleMRFFoam in foam-extend 3.1.
Attention is required to have proper face ordering on cyclic boundaries.

I will keep this thread updated as I progress.

Regards.

Ersin.

[tom.opt]

Originally Posted by ersoflow

Hello all,

In this post, I would like to speak of my progress on Rotor 37 work.
Hopefully that helps someone.

- Geometry is obtained as curve files from the internet (can be found easily).

- TurboGrid is used to generate mesh. It reads curve files for hub, shroud and blade profile.

- To transform the mesh from TurboGrid into openfoam, I found my way by intertransforming to cfx and fluent, and lasty fluentMeshToFoam.

- Now, I am having several runs with steadyCompressibleMRFFoam in foam-extend 3.1.
Attention is required to have proper face ordering on cyclic boundaries.

I will keep this thread updated as I progress.

Regards.

Ersin.

Could you share details of how your boundary conditions are configured please?

[ersoflow]

Hi Tom,

The boundary condition that I used in my last run as follows: (which are very similar with the tutorials)
(since it would be too long I neglect reporting the conditions regarding kEpsilon turbulance components)

p:

internalField   uniform 101325;

boundaryField
{
    inlet
    {
        type            totalPressure;
        phi             phi;
        rho             none;
        psi             psi;
        U               U;
        gamma           1.4;
        p0              uniform 140000;
        value           $internalField;
    }
    outlet
    {
        type            fixedValue;
        value           $internalField;
    }


  "per1"
  {
    type            cyclic;
  }
  "blade|hub|shroud"
  {
    type            zeroGradient;
  }

U:

internalField   uniform (0 0 100);

boundaryField
{
    inlet
    {
        type            pressureInletVelocity;
        value           uniform (0 0 100);
    }
    outlet
    {
        type            inletOutlet;
        inletValue      uniform (0 0 0);
        value           $internalField;
    }

  "per1"
  {
    type            cyclic;
  }
  "blade|hub|shroud"
  {
        type            fixedValue;
        value           uniform (0 0 0);
  }

T:

internalField   uniform 300;

boundaryField
{
    inlet
    {
        type            totalTemperature;
        phi             phi;
        rho             none;
        psi             psi;
        U               U;
        gamma           1.4;
        T0              uniform 300;
        value           $internalField;
    }
    outlet
    {
        type            zeroGradient;
        value           $internalField;
    }

  "per1"
  {
    type            cyclic;
  }
  "blade|hub|shroud"
  {
    type            zeroGradient;
  }

The results seemed not really bad but my work is still in progress.

How is your work going?

Ersin.

[tjuAeron]

hi Ersin，
Currently I'm doing the same thing with foam-extend-4.0 but with little progress.
What do you mean by "The results seemed not really bad"? What about pressure ratio and isentropic efficiency you got?
And why you set p::inlet:0 1.4e5 pa rather than 101325 pa in experiment? And so as the Temperature condition of inlet.
Thanks for you sharing

[tjuAeron]

Originally Posted by ersoflow

The results seemed not really bad but my work is still in progress.

How is your work going?

Ersin.

hi Ersin，
Currently I'm doing the same thing with foam-extend-4.0 but with little progress.
What do you mean by "The results seemed not really bad"? What about pressure ratio and isentropic efficiency you got?
And why you set p::inlet:0 1.4e5 pa rather than 101325 pa in experiment? And so as the Temperature condition of inlet.
Thanks for you sharing