[Bdew8556]

Hi Foamers.

I've got this error, any thoughts?




--> FOAM FATAL ERROR:

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

5
(
nut
k
nu
p
epsilon
)


From function const Type& Foam:bjectRegistry::lookupObject(const Foam::word&) const [with Type = Foam:imensionedField<double, Foam::volMesh>]
in file /opt/CFDSupportFOAM4.0/beta/OpenFOAM-dev/src/OpenFOAM/lnInclude/objectRegistryTemplates.C at line 193.

FOAM aborting


This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.

[alexeym]

Hi,

Since you post almost no context and also using CFDSupport's version of the software, people can only guess what is going wrong.

Usually wall functions try to lookup G field through object registry (see source of epsilon or omega wall functions). Field G is created by turbulence model in correct method.

The questions is: why field G is requested before it is created? Answers could be:
- CFDSupport forgot to import turbulence models changes but pulled wall functions changes.
- CFDSupport changed wall functions, so they simply do not work.
- Well, I can continue, but think, it is rather useless until you provide additional information.

What OpenFOAM(R) flavour do you utilise? What are your boundary conditions?

[Bdew8556]

Hi Alex,

Thanks for your reply.

I must admit this is my second week of using OpenFoam and I'm still very very green.

I don't actually know what the G value is?

I'm using OpenFoam 18.10

Boundary conditions are in this file. If that helps?

[wyldckat]

Originally Posted by Bdew8556

Boundary conditions are in this file. If that helps?

Quick answer: Still not enough information. Please follow the guidelines written here: How to give enough info to get help

Suggestion: Run one of the tutorial cases from the installation that you have got, that is the most similar to the case you are trying to run. If that tutorial case does not work, contact the people at the website from which you downloaded your version.

Furthermore:

Originally Posted by Bdew8556

I'm using OpenFoam 18.10

the standard versions of OpenFOAM are only provided by the following producers:

• OpenFOAM Foundation: OpenFOAM.org
• ESI-OpenCFD: OpenFOAM.com

You can see a complete list of known versions here: https://openfoamwiki.net/index.php/Main_Page - on the top right corner.

[alexeym]

Hi,

I agree with wyldckat, there is still not enough information.

Or I would put it another way: I do not want to figure out what CFDSupport has broken in OpenFOAM(R). I was not able to find a link to source code on their web-site (they say sources are provided as a part of installation packages), so it is practically impossible to figure out in what way they have messed up turbulence models.

[Bdew8556]

Thanks guys,

I'll see what the deal is with CFD support.

Do you know what "G" is though? and whether you need a separate file called "G" in the o/ directory?

[wyldckat]

Quick answer @Bdew8556: The G field is an intermediate field created by most k-epsilon turbulence models and is not meant to be created manually. It's created right here: https://github.com/OpenFOAM/OpenFOAM...Epsilon.C#L244

This means that this field only exists if the turbulence model is loaded in and is ongoing the updating of its coefficients, namely inside this method. See a few lines below that one, in #252? That's how the boundaries were updated and triggered the error with the infamous "G" field.

And my apologies, but I have to poke a little bit of fun here... I went back to your posts and my imagination wondered into the following possible workflow:

1. You started using OpenFOAM to do something in this particular case...
2. Then you suddenly sneezed, so you turned your head to not sneeze onto the screen...
3. And when you turned your head back to the screen, you saw the error message and forgot how you had gotten there.
4. And so you came to the forum looking for suggestions

What I mean by this is that in your descriptions there is no clear indication of how exactly you got the error message in the first place... so it could be due to running any number of applications, either solvers or utilities...


Either way, I'm guessing here that the fields were manipulated first by some utility, without properly loading the turbulence model. Perhaps you are trying to use and old solver with a more modern OpenFOAM version? If so, you should have said so.
For example, this is a recent feature: https://github.com/OpenFOAM/OpenFOAM...mpleFoam.C#L52 and is how the G field would be properly operational.

[Bdew8556]

Thanks guys.

No idea what was wrong with CFD support.

I had my nozzle simulation working fine yesterday and now it's not. I can't figure out the problem. Basically I'm just trying to simulate air at atmospheric pressure going through a nozzle, so pressure inlet mass flow rate outlet.

Can you see anything wrong with my p and U files?

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  dev                                   |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       volVectorField;
    object      U;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //



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

internalField   uniform (0 0 0);

boundaryField
{

    Inlet
    {

        type            pressureInletOutletVelocity;
    }



    Outlet
    {
        type            flowRateOutletVelocity
        value           uniform 2;

    }

    
    Walls
    {
        type            noSlip;
    }


}

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

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  dev                                   |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
    object      p;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

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

internalField   uniform 0;

boundaryField
{

    Inlet
    {
        type            fixedValue;
        value           uniform 101300;
    }



    Outlet
    {
       type             inletOutlet;

    }

 
    Walls
    {
        type            noSlip;
                 
    }


}

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

[LuckyTran]

Try posting the error message? What you are doing is like sending us a photo of your hair before you got a haircut and now you're asking what are all the possible illnesses you might have.

Give you tell us what the last doctor (error message) has already diagnosed?

The funny thing about coded software is, if you don't do anything to it, it doesn't change. What did you do to it?

[wyldckat]

Quick answer:

Originally Posted by Bdew8556

Can you see anything wrong with my p and U files?

[...]

    object      p;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

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

internalField   uniform 0;

boundaryField
{

    Inlet
    {
        type            fixedValue;
        value           uniform 101300;
    }

By accident, I did notice something very strange: the initial internal field is set to 0 and the inlet is set to 1 atmosphere... which means that you're starting the case as a vacuum, therefore it's an extremely complicated simulation, hence whichever crash you have gotten.