[vatant]

Hello,


I compiled lesInterfoam and started my case. I go t the following error:

--> FOAM FATAL IO ERROR : cannot open file

file: ../tutorials/lesInterFoam/jetbreak/2.18818e-09/pd at line 0.

Function: regIOobject::readStream(const word&)
in file: db/regIOobject/regIOobjectRead.C at line: 68.

FOAM exiting


What is going wrong here ? I copied the available case file nozzleflow2d into another case file and ran the simulation.

Also, I want to know if the currently existing LES simulation can be used for compressible liquids ?


Thanks

Vatant

[henry]

Is 2.18818e-09 your start time or did this error occur after a while?

Try switching off runTimeModifiable in controlDict.

lesInterfoam is written based on interFoam which is an incompressible VOF code.

[vatant]

My start Time is zero and after the first iteration I got this error.(i started with default initial time step of 1e-08).

Also, given a volScalarField, is it possible to convert it to a dimensionedScalar at the given cell?


Thanks

Vatant

[henry]

Did switching off runTimeModifiable work?

The cell value returned by [] is a scalar but you can convert it into a dimensionedScalar:

dimensionedScalar(field.name(), field.dimensions(), field[celli])

[vatant]

No, i tried to switch off my runtimeModifiable in controlDict.

Also, i reduced my time step size and got the same problem.

--> FOAM FATAL IO ERROR : cannot open file

file: ../run/tutorials/lesInterFoam/jetbreak/2.18818e-09/pd at line 0.

Function: regIOobject::readStream(const word&)
in file: db/regIOobject/regIOobjectRead.C at line: 68.

FOAM exiting



Is there a way to test whats wrong with writing file options ?


Thanks



Vatant

[henry]

How often are you writing results? Every time step?

Are you writing to the disk on the machine you are running on?

[vatant]

I am writing every 1e-06 time step as default indicated in Foam. I tried to decrease time steps, clear data storage etc..still had the same problem.

I might want to add some modication that I inserted and compiled in the multiphase/lesInterFoam.

In the pressure equation. pEqn.H i added an extra term computing ddt(pd)

fvm::laplacian(rUAf, pd) == fvc::div(phi)+comp*fvm::ddt(pd)

where comp = a volScalarField type

Do you think this formulation is creating any problem ?


Thanks

Vatant

[henry]

I doubt it but I guess you have tried with out it to check.

[mattijs]

Hi Vatant,

about your writing problems: have you tried increasing the writePrecision? (6 by default)

You say you are writing every 1E-6 but your error lists 2.18818e-09 so I guess your time step gets adapted and becomes very small. There might also be problems when you hit timesteps on the order of 1E-15.

[vatant]

I set the write precision to 8. My minimal dt = 1e-09 and i enforce the courant condition ~ 0.5 actually decreased to 0.2 to check the difference.

Do you think, it has anything to do with files mismanagement ..im sure I have no problem but if you think I should look into the files system again.

I was thinking that this error was due to error in reading dictionary or objects like in correctPhi.H ,,where pd.objects() comes into calculations.

Any thoughts about this Dr.Weller ?

[henry]

Do you get the same problem when running standard lesInterFoam on the same case?

[vatant]

I dint make a backup copy ..i however tried to recompile lesInterFoam to see what is the cause of the problem.

I got it compiled but had the following warnings:

createFields.H: In function `int main(int, char**)':
createFields.H:2: warning: unused variable 'nGammaCorr'
createFields.H:7: warning: unused variable 'nGammaSubCycles'
createFields.H:2: warning: unused variable 'nGammaCorr'
createFields.H:7: warning: unused variable 'nGammaSubCycles'
../OpenFOAM/OpenFOAM-1.0/src/cfdTools/lnInclude/readPISOControls.H:3: warning: unused variable 'nCorr'


Could you help me fixing these warnings ? or is it ok to have these warnings during compilation.



Regards,



Vatant

[vatant]

I made some changes and removed some errors except one with the 'ncorr'. In the createFields.H, i inserted a volScalarField...

volScalarField scalar
(
IOobject
(
"scalar"
runTime.timeName(),
runTime,
IOobject::NO_READ
IOobject::NO_WRITE
)
(..)
)
I compiled the code and used this volScalarfield in calculation.

The code ran but gave the following error:

--->FOAM Warning : ObjectRegistry::checkIn(regIOobject&):
Registering object '..' with objectRegistry 'time'
This is only appropriate for registering the regions of a multi-region computation or in other special circumstances.
otherwise please register this object with it's region (mesh)


If we create a volScalarField how should it be registered to Mesh type ?


Vatant

[henry]

This issue is covered in the release notes:

OpenFOAM-1.1/ReleaseNotes-1.1

[vatant]

Thanks Dr.Weller, I was able to Fix the changes. I have another question on fixing patches.

I converted a fluentMesh to foam and wanted a liquid inlet for simulating a jet using lesInterFoam. In gamma BC, I set inlet value = 1
and then started the code.

In the beginning, i got a message

No Target Patch Present.

My code ran for some time and diverged. Same code that I ran for the test case converged.

In my case, I have flow travelling a small distance inside nozzle and then ejecting into air, while in the les test case, you have a patch field set up.

Do I have to set any patch field ,for a case like simulating a flow coming out of a nozzle ?

I modified all the variables : the inlet was a pressure inlet not an velocity inlet.

Also, Dr.weller, How do you initialize K with given U value in the inlets ? (relation between K and U ?? ) If it is a pressure inlet , how to initialize K ?

In my case, I modified K to a very low value , since the pressure chamber has total pressure associated with it.


Thanks


Vatant