[adrieno]

Hi foamers,

I'm trying to run rhoPimpleFoam adapting the angleDuct tutorial to my case.
The thing is, when I run it, I'm facing this fatal error :

/*---------------------------------------------------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.2.2                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
Build  : 2.2.2-9240f8b967db
Exec   : rhoPimpleFoam
Date   : Apr 01 2016
Time   : 10:28:20
Host   : "SAFSV4199"
PID    : 36914
Case   : /home/saf128648/OpenFOAM/saf128648-2.1.1/2016_03/31_03_Storage_23
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //


Create time

Create mesh for time = 0


PIMPLE: max iterations = 50
    field "(U|k|epsilon)"    : relTol 0, tolerance 0.0001

Reading thermophysical properties

Selecting thermodynamics package 
{
    type            hePsiThermo;
    mixture         pureMixture;
    transport       const;
    thermo          hConst;
    equationOfState perfectGas;
    specie          specie;
    energy          sensibleEnthalpy;
}

Reading field U

Reading/calculating face flux field phi

Creating turbulence model

Selecting turbulence model type RASModel
Selecting RAS turbulence model kEpsilon


--> FOAM FATAL ERROR: 
Different dimensions for =
     dimensions : [1 -1 -1 0 0 0 0] = [0 2 -1 0 0 0 0]


    From function dimensionSet::operator=(const dimensionSet&) const
    in file dimensionSet/dimensionSet.C at line 165.

FOAM aborting

• FIRST

This dimension correspond to the viscosity (kinematic and dynamic), but I didn't change it in any file from the tutorial (which run fine) so I can't see where it comes from...
So if you have any idea on this problem of dimension, let me know.

• SECOND

As I said, I'm trying to adapt my case to the angleDuct exemple which is one of the three examples that are running rhoPimpleFoam. I'm doing that in order to know, what file to specify in folders 0, constant and system... But I've seen that in another example ( the mixerVessel2D one) the files where not exactly the same...
Indeed, there is a file named "transportProperties" in the folder constant of the mixerVessel2D case which is not on the angleDuct case... How is this possible ? I mean ,why, in one, is it necessary to specify it but not in another (while they are running the same solver and same turbulence model) ?


Thanks,
Adrien

[adrieno]

Hi again,

In fact, my problem of dimension was du to my pressure which where still define as the kinematic pressure (like in an incompressible case , so dummy I am... but that was my first compressible simulation ).

If someone has any idea for my second question , I'm still interested.

I'm now facing a new problem that I can't fix for few hours now .
When I run rhoPimpleFoam, it tolds me that my volScalarField kEpsilon isn't avaible :

 

/*---------------------------------------------------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.2.2                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
Build  : 2.2.2-9240f8b967db
Exec   : rhoPimpleFoam
Date   : Apr 01 2016
Time   : 17:47:17
Host   : "SAFSV4199"
PID    : 3420
Case   : /home/saf128648/OpenFOAM/saf128648-2.1.1/2016_04/01_04_Storage_24
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Create mesh for time = 0


PIMPLE: max iterations = 50
    field "(U|k|epsilon)"    : relTol 0, tolerance 0.0001

Reading thermophysical properties

Selecting thermodynamics package 
{
    type            hePsiThermo;
    mixture         pureMixture;
    transport       sutherland;
    thermo          hConst;
    equationOfState perfectGas;
    specie          specie;
    energy          sensibleEnthalpy;
}

Reading field U

Reading/calculating face flux field phi

Creating turbulence model

Selecting turbulence model type RASModel
Selecting RAS turbulence model kEpsilon


--> FOAM FATAL ERROR: 

    request for volScalarField kEpsilon:G from objectRegistry region0 failed
    available objects of type volScalarField are

12
(
thermo:mu
thermo:psi
h
rho
k
thermo:psi_0
alphat
p
T
mut
epsilon
thermo:alpha
)


    From function objectRegistry::lookupObject<Type>(const word&) const
    in file /home/opencfd/OpenFOAM/OpenFOAM-2.2.2/src/OpenFOAM/lnInclude/objectRegistryTemplates.C at line 164.

FOAM aborting
...

while when I run it from the angleDuct OpenFoam Tutorial, all is fine :

/*---------------------------------------------------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.2.2                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
Build  : 2.2.2-9240f8b967db
Exec   : rhoPimpleFoam
Date   : Apr 01 2016
Time   : 17:38:32
Host   : "SAFSV4199"
PID    : 3010
Case   : /home/saf128648/OpenFOAM/saf128648-2.2.2/run/tutorials/compressible/rhoPimpleFoam/ras/angledDuct
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Create mesh for time = 0


PIMPLE: max iterations = 50
    field "(U|k|epsilon)"    : relTol 0, tolerance 0.0001

Reading thermophysical properties

Selecting thermodynamics package 
{
    type            hePsiThermo;
    mixture         pureMixture;
    transport       sutherland;
    thermo          hConst;
    equationOfState perfectGas;
    specie          specie;
    energy          sensibleEnthalpy;
}

Reading field U

Reading/calculating face flux field phi

Creating turbulence model

Selecting turbulence model type RASModel
Selecting RAS turbulence model kEpsilon
kEpsilonCoeffs
{
    Cmu             0.09;
    C1              1.44;
    C2              1.92;
    C3              -0.33;
    sigmak          1;
    sigmaEps        1.3;
    Prt             1;
}

Creating field dpdt

Creating field kinetic energy K

Creating fintite volume options from fvOptions

Selecting finite volume options model type explicitPorositySource
    Source: porosity1
    - applying source for all time
    - selecting cells using cellZone porosity
    - selected 8000 cell(s) with volume 0.00025

Porosity region porosity1:
    selecting model: DarcyForchheimer
    creating porous zone: porosity

Starting time loop

Courant Number mean: 0 max: 0
Time = 1

PIMPLE: iteration 1
DILUPBiCG:  Solving for Ux, Initial residual = 1, Final residual = 0.0368343, No Iterations 1

First:
From where does this kEpsilonCoeffs are resulting when all is fine ? because there are not specify in any file of 0 , constant or system either.

Secondly:
Have you any idea of the meaning of my error message ?
I cant fix it for a moment now, it's making me crazy.

[fabian_roesler]

Hi

first of all to your question from your first post:
The transportProperties dictionary is not necessary for most compressible simulations as you're using thermophysicalProperties which defines all properties for your simulation. However, for some postProcessing tools transportProperties is still required. This could be the reason for the file still existing in one of the tutorials.

kEpsilonCoeffs are defined in RASProperties dictionary. However, most of the time this properties do not have to or even should not be touched. That's the reason why they are to there you're able to define them there but they are defined as default if not.

I can't answer your last question concerning your error by now but if I have some time left I will dig into it and give you some feedback.

Cheers

Fabian

[adrieno]

Hi foamers,

Thank you very much fabian_roesler for your help.
Unfortunately I couldn't find the origin of the error message, but I've been simplifying my case at the maximum (I have now a beautiful box) in order to be as close as possible to the exemple angleDuct. Hopefully now the solver is running, I'll now try to complicate my case step by step.

Thank you again for all your response fabian_roesler.

Just last questions everyone concerning rhoPimpleFoam that I can't explain myself.
I fact I'm running my case with the rhoPimpleFoam solver because according to the openFoam guide's description that's the one who's matching the most with my case. But I've been wondering why there isn't a simple Piso solver for turbulent and compressible case ? Because, pimpleFoam isn't really a good transient solver because of its large time-step isn't it ?

[fabian_roesler]

Hi adrieno

Actually the PIMPLE Solver is a merged PISO SIMPLE method.
For only one outerCorrector the PIMPLE method is a PSIO as it does one walk through all conservation equations. Thus, under relaxation is not allowed for PIMPLE with one outerCorrector. Increasing the outer correctors allows a more stable solution for higher Courant numbers. In this case under relaxation is allowed.

Cheers

Fabian

[adrieno]

Thank you again,

Is there any difference between using PISO (I mean pimple with one outercorrector) and a max courant number less than 1 instead of using Pimple with many outercorrector and a much higher courant number ?

I guess this is just a question of objective ? I mean if someone is interested by the transition (with a little deltaT) or not ? Am I wrong ?

[anishtain4]

Usually this:

From function objectRegistry::lookupObject<Type>(const word&) const

means the dictionary has not been provided, or the initial file might be missing from 0 folder. can you post a list of your files in 0 directory and content of your turbulentProperties file?

[adrieno]

Thank you anishtain4,

I've been able finally to run the solver. I don't know why this message error appeared because I've made my 7 files in my 0 folder (T, p, U, k, epsilon, mut, alphat) and correctly define the kEpsilon for turbulence model. But well... finally it's working now.

[adrieno]

By the way what does the "alphat" file correspond to exactly ? I can't understand its utility.

[anishtain4]

Glad to hear it's working. Did you find the issue? it might help others if they face the same problem later if you post the reason.

Not sure about the alphat, I use LES rather than RANS models. But if you look into the source files it should be easy to spot it. You can use "grep -nr" command to spot it fast

[adrieno]

Hi anishtain4,

I'm sorry but as I said, unfortunately I can't find the error. I would post the solution for sure otherwise. I had to run the solver on a case much simple that the one I used to have... And after modify it slowly.

There's 4 weeks now that I'm working on OpenFOAM, I don't know much, but I have now a golden rule :

I start from something that's working, then I try to adapt it to my case POINT by POINT.

[anishtain4]

it's a good rule. Also when I first started it helped me a lot to look into doxygen and check source codes too. It may look a little scare at first, but in time it will be super efficient.
Also if you don't have a background in c++ you might want to start reading it because as a OF user a time comes soon that you realize you need to know how to do basic c++.