[aeromaik]

Hi everyone,
I'm quite new to this world, so I'm sorry if I forgot doing something else before this.

I have done a mesh of an object and run it using the SimpleFoam solver, and it definitely works after a lot of attempts.
The residue is correct and the solution is as expected.

I've surfed the site but I can't find (and I think is my fault) anything teaching me from ground level how to switch a good simulation working in simplefoam to rhosimplefoam.

Maybe I have to start again from zero level, but before doing this I would like to be sure I'm not loosing time.

The fact is that I would like to push the velocity up to early transonic regime (Mach = 0.6-0.7) and I would like to see the correct loads over the object in a steady case.

Anything included a link to a guide or similar would be really appreciated.

Sorry for my english and if I didn't do the correct job, I will do eveything for being correct again.
Thanks in advance !

aeromaik

[Krao]

Hi aeromik,

It's been long since you posted, but if you remember your findings, can you please share it here. It would be very useful.

Thank you,
Krao

[Carlo_P]

Hey,

I didi lots of time, since I normally initialize the rhosimplefoma with simplefoam.
You can try this step:


1) Change the p file. You have to change the dimension, look at one p file in tutorial. You have also to add the atmosferic pressure.


2) Add the T file


2) Add transonic yes (if it is transonic) in your FvOption - Simple


3) Add a small value for the relaxation factor of rho.


4) You have to change the numerics in fvSchemes. Try and read the error in openfoam.


Try this step. I'm not sure that they are all.
Cheers,
Carlo

[Krao]

Originally Posted by Carlo_P

Hey,

I didi lots of time, since I normally initialize the rhosimplefoma with simplefoam.
You can try this step:


1) Change the p file. You have to change the dimension, look at one p file in tutorial. You have also to add the atmosferic pressure.


2) Add the T file


2) Add transonic yes (if it is transonic) in your FvOption - Simple


3) Add a small value for the relaxation factor of rho.


4) You have to change the numerics in fvSchemes. Try and read the error in openfoam.


Try this step. I'm not sure that they are all.
Cheers,
Carlo

Thanks Carlos for your kind reply, I will work on with the given inputs from your side.

Regards

[Krao]

Hi Carlo,

I did as specified by you and somehow my simulation started running, thank you. But I need to understand the underlying physics, to analyse the results.

My operational range is always about Ma: 0.3, and I am using SimpleFoam with MRF for this, but I am curious to know the results in non-design operating conditions therefore I'm studying the behavior at Ma: 0.5. Therefore it would be nice if you can suggest any similar tutorial, which I can follow and implement those settings to my rhoSimpleFoam with MRF. Also, I would be grateful if you can suggest me related literature.

Regards,
Krao

[maike151]

Hi Krao,


I am trying to do the same thing. Do you remember how you got rhoSimpleFoam to run?
I did the same steps that are suggested above but when I run rhoSimpleFoam, I get the following error message:
[10] --> FOAM FATAL ERROR:
[10]
[U[0 1 -2 0 0 0 0] ] + [(rho*MRFZoneList:acceleration)[1 -2 -2 0 0 0 0] ]
[10]
[10] From function void Foam::checkMethod(const Foam::fvMatrix<Type>&, const Foam:imensionedField<Type, Foam::volMesh>&, const char*) [with Type = Foam::Vector<double>]
[10] in file /opt/OpenFOAM/OpenFOAM-7/src/finiteVolume/lnInclude/fvMatrix.C at line 1291.
[10]
FOAM parallel run aborting



I am not sure if this is about the dimension of p. Let's assume my simpleFoam solution converged in 1000 iterations. Thus I have the folder 0 and 1000. The p file in 1000 just contains some hieroglyphs with information about the field of p in the 1000th iteration. Should I just replace this with a file with the right dimension or is all the information for the initial solution lost then?


I would appreciate any help.
Regards,
Maike

[Carlo_P]

The hieroglyphs are creted because you are openging a binary file probably.
Check in the controlDict.
The pressure in simpleFoam is not the pressure, but the relative pressure/density.
In rhoSimpleFoam you need the absolute pressure.


So you have to add the atmosferic pressure at the p from simpleFoam and moltiply by the desnity (1.2 if air).


Later, you hae to change the line with the unit of measurement.

[maike151]

Hello Carlo,


thank you for the quick response!
I changed the writeFormat to ascii and the hieroglyphs are gone thank you!


I think I followed your instructions:


In SimpleFoam, I had as outlet BC for p

     outlet
    {
        type            fixedValue;
        value           uniform 0;

    }

and now changed it to

    outlet
    {
        type            fixedValue;
        value           uniform 0.4e5;

    }

(which is atmospheric pressure * 0.4, because my air is at 853,15 K and thus the density is about 0.4). Dimension [0 2 -2 0 0 0 0] and internalField also 0.4e5.



Then I ran the case, took the file of the last iteration, copied it into my rhoSimpleFoam case, changed the dimension of p to [1 -1 -2 0 0 0 0] and changed the outlet BC back to

outlet
    {
    //type            totalPressure;
        //p0              uniform 100000; //40000;
        //value           uniform 100000; //40000;
        type            fixedValue;
        value           uniform 100000; /40000;
    }

(Also the ones that are commented out)


Also changed everything else as mentioned above. Unfortunately I still get the same error


My inlet BC for p is zeroGradient, for U I have massFlowRate at inlet and inletOutlet at the outlet.