[francois]

Hi Foamers,

I'm playing with the simpleFoam solver for a simple open rotor/stator laminar flow configuration but I'm not able to reach the steady state of the flow.

The angular velocity on the rotor is around 300 tr/min so a Reynolds number (based on the rotor radius) around 2.10^5 is expected. According to the literature the flow is still laminar.

So I try simpleFoam with no turbulence models but even if convergence seems ok I'm not able to reach this steady state. Perhaps the flow is locally turbulent or in transition and a RANS model could be the way to go. Some experiences on those kind of flows ?

Here is one of my last iteration output in the log file:

Time = 95000

DILUPBiCG: Solving for Ux, Initial residual = 0.0114524, Final residual = 0.000374143, No Iterations 2
DILUPBiCG: Solving for Uy, Initial residual = 0.00827667, Final residual = 0.000238708, No Iterations 2
DILUPBiCG: Solving for Uz, Initial residual = 0.00574431, Final residual = 0.000142418, No Iterations 2
DICPCG: Solving for p, Initial residual = 0.0246974, Final residual = 0.000246175, No Iterations 54
time step continuity errors : sum local = 7.81603e-05, global = -2.14943e-06, cumulative = 0.07646
ExecutionTime = 182884 s ClockTime = 183024 s

Any advices to get better convergence to reach the steady state of the flow ?

My setup is as follow:

===========
:fvSchemes:
===========

ddtSchemes
{
default steadyState;
}

gradSchemes
{
default Gauss linear;
grad(p) Gauss linear;
grad(U) Gauss linear;
}

divSchemes
{
default none;
div(phi,U) Gauss upwind;
div(phi,k) Gauss upwind;
div(phi,epsilon) Gauss upwind;
div(phi,R) Gauss upwind;
div(R) Gauss linear;
div(phi,nuTilda) Gauss upwind;
div((nuEff*dev(grad(U).T()))) Gauss linear;
}

laplacianSchemes
{
default none;
laplacian(nuEff,U) Gauss linear corrected;
laplacian((1|A(U)),p) Gauss linear corrected;
laplacian(DkEff,k) Gauss linear corrected;
laplacian(DepsilonEff,epsilon) Gauss linear corrected;
laplacian(DREff,R) Gauss linear corrected;
laplacian(DnuTildaEff,nuTilda) Gauss linear corrected;
}

interpolationSchemes
{
default linear;
interpolate(U) linear;
}

snGradSchemes
{
default corrected;
}

fluxRequired
{
default no;
p;
}

============
:fvSolution:
============

solvers
{
p PCG
{
preconditioner DIC;
tolerance 1e-06;
relTol 0.01;
};
U PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0.1;
};
k PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0.1;
};
epsilon PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0.1;
};
R PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0.1;
};
nuTilda PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0.1;
};
}

SIMPLE
{
nNonOrthogonalCorrectors 0;
}

relaxationFactors
{
p 0.3;
U 0.7;
k 0.7;
epsilon 0.7;
R 0.7;
nuTilda 0.7;
}


===============
:RASProperties:
===============

RASModel laminar;

turbulence off;

printCoeffs on;

laminarCoeffs
{
}


Thank you very much.
Have a nice day

Francois

[francois]

Hi all,

Hopefully I have some time to go back on this case ...

My runs don't reach the steady state and convergence is still not achieved with residuals not so low.

I've attached some pictures to better describe my case. It's a simple rotor/stator configuration with a hole in the stator.

rotor_stator.jpg

I've also attached the distribution of the u,v,w velocity components on a radial line between the rotor and stator.

velocity_distribution.jpg

You can see that steady state is not reach here: one can observe large variation of u,v,w profiles between iteration 80000 and 90000. The behaviour is the same for a greater number of iteration (currently I'm at 120000 iterations).

Below you can see isocontours of the u component around the rotor/stator and one can note a "shear layer" near the rotor which oscillates with iterations (this trend is also observed on the uvw distribution figure just above)

rotor_stator_Ux.jpg

And to conclude below are the evolution of residuals with the iterations for Ux and Uy for example:

UxFinalRes.jpg

UyFinalRes.jpg

So I still have a few questions:

* what is the best procedure to obtain a steady state

* is my computational domain too small (Ddomain = 4rotor and Hdomain = 3Drotor)for this problem with a "feedback" of the outlet boundary condition
?

I will be very pleased to have some advices/ideas frome somone on those points and forgive me for this long post.

Thanks you very much
Regards

Francois

[francois]

Oup's sorry for the images
Will try again here :

Computationaldomain:



distribution of the u,v,w velocity components on a radial line between the rotor and stator:



u v w contours:





Residuals:

[francois]

Humm ... sorry
Don't work here with firefox 3.06 on ubuntu 8.10

What the best way to send/attach picture or files on the message board ?
I use:

\ image{foo} or \ attach{foo} with no space between \ and the after word

It seems that I always have to struggle until death to be able to send some files or images here.

[andersking]

I think your syntax is probably fine, but there's a very small limit on image sizes, 50K I think. It doesn't stop you from posting, but won't show the images.

Less than 50K is fine.

Cheers
Andrew

[francois]

Thanks Andrew for your advice

But my images sizes are always under 50k and I never saw the dialog for file upload

So the problem seems to be elsewhere ...

Cheers
Francois

[andersking]

Hmmm, hopefully it will all get betted with the new forum software.

Cheers
Andrew

[francois]

I've just modified the first post and was able to include some of the images which could not be posted in the last forum...

Yeah ... the new forum is better

[novyno]

I have the same convergence problem... Can someone help us?