[raunakbardia]

Hello,

I came across the release of IsoAdvector code and have recently started testing this code.

I had a question about the UEqn.H file in the code. In this file, there is a reference to UEqnSolved.H.

...
        phi = uFactor*phi0;
        U = uFactor*U0;
    }
}
else
{
    #include "UEqnSolved.H"
}

I am unable to find this header file anywhere in the entire directory of isoAdvector and am wondering where this is invoked from.

I tried to change the isoAdvector code for a test case and rightfully, an error is thrown by the build due to its inability to find the UEqnSolved.H file.

In file included from kappaFlow.C:109:0:
UEqn.H:38:28: fatal error: UEqnSolved.H: No such file or directory
#include "UEqnSolved.H"
^
compilation terminated.
make: *** [Make/linux64GccDPInt32Opt/kappaFlow.o] Error 1

Given that I cannot find this header file in my original isoAdvector download either is interesting; because I have been able to compile the original code and use it for some other test cases.

If someone involved in the development of isoAdvector code can answer this question, it will really speed up my testing of the code.

Thank you.

[akidess]

My guess is you are working on a broken version.

[raunakbardia]

This is what I first thought. But, even on the git repository of isoAdvector I see the same thing. The file, UEqn.H includes a header file UEqnSolved.H, if we don't use a prescribed velocity.
This header file does not exist in the repository, nor in the entire OpenFOAM v1612+ directory, which is the version I am using.

Work Around:
Because I want to solve for U & P in my dynamic test cases, I simply copied the original interFoam header file for velocity and the solver compiled with the isoAdvector advection scheme without a problem.

However, I am still at a loss on how the UEqn.H that comes from IsoAdvector git repository, works?

Thank you.

[akidess]

There is no include for UEqnSolved.H in the foam-extend tree, and neither in the v1712 release, and neither in previous versions of the github repository you are checking. Thus I still believe it's a bug in the latest repository commit.

[raunakbardia]

Thank you for your response.
I did not realize that the new release has got the interIsoFoam solver built-in.
Until I shift to this version of OpenFOAM, which is dependent on my system admin, I will look for the older repositories of the code.

[akidess]

If you don't want to wait just get the app from the repository develop.openfoam.com and compile it yourself.

[roenby]

Hi Raunak

If you install isoAdvector using the Allwmake script from github.com/isoAdvector it will copy the interFoam solver from OpenFOAM-x.y.z/applications/solvers/multiphase/interFoam to the isoAdvector installation directory and modify the files relating to the interface advection step before compiling it into the interFlow solver. In other words the UEqn.H file included in the github.com/isoAdvector source code is never used. I will remove it to avoid confusion.

Have a look in the README.md file for installation and usage. Also feel free to report an issue in the github repo. This may save yourself and others a lot of time :-)

Best,
Johan

[raunakbardia]

Thank you, Johan.

I started to do that a while back but forgot to update this thread about that.

You have further clarified it for all of us.

[Arghavani]

Hello everyone,

I cannot get how we can have isoAdvector while the OpenFoam 5.x is installed?

and what would happen to the tutorials in interFoam folder, all will be deleted?

kind regards,
Arghavan

[roenby]

Hi Arghavani


Go to https://github.com/isoAdvector/isoAdvector and follow the installation instructions in the README file.


Installation will not touch or delete interFoam (the new solver is called interFlow).



Best,
Johan

[Arghavani]

Hi Johan,

Thank you for your reply.
I did it once but It damaged my OpenFoam and I had to reinstall it.
But I will ask my boss to help me at this time.

kind regards,
Arghavan

[roenby]

That sounds terrible and should not be possible.
I would be very interested to get more details on what you did.

Do you remember to which directory you cloned isoAdvector?
Do you know what your $FOAM_USER_APPBIN and $FOAM_USER_LIBBIN variables are set to?



A general advice: Once you have installed your OpenFOAM version and confirmed that the installation works, you should always remove write access for anyone to the installation. You can for instance do this by going to the installation directory and type:


chmod -R ugo-w


Which means "change permissions (-R)ecursively removing write access (-w) to for User, Group and Others (ugo)".


Then if you at a later time accidentally try to write to the installation directory you will get a warning (note that you can still accidentally remove files with rm -f). If you want to be even safer (and you have permissions), you could transfer ownership of the installation files to the root using chown.

[Arghavani]

thank you for additional explanation.
actually I did it before Chrismas holiday and I only remember that I downloaded the IsoAdvector package and I coyed to my OpenFOAM folder and then since I needed a master version of the OpenFOAM I downloaded that one and in the middle of the installation I got an error and after that, I couldn't run the OpenFoam. I don't remember more details. I have never noticed these things you mentioned (Do you remember to which directory you cloned isoAdvector?
Do you know what your $FOAM_USER_APPBIN and $FOAM_USER_LIBBIN variables are set to?) and I don't have enough knowledge with the installation part. since that time I was busy with another project, I stopped dealing with that problem and now I want to continue it and then it is my problem. I will ask my expert boss to help me. and Thank you again.

Best,
Arghavan

[jobin2600]

Hello Everyone

I recently found this excellent method for surface reconstruction. Since I am dealing with mass transfer, I need to add a source term to the alpha equation. I see that MULES take the source terms Su, Sp from alphaSuSp.H . But couldn't figure out if isoAdvector handles those terms as well. I had installed isoAdvector from Github on OpenFOAM 6(and it works perfect).

How can I add the source terms to the alpha equation and make use of isoAdvector to solve the equation ?

Any help is highly appreciated

Best regards

Jobin Raju

[roenby]

Hi Jobin Raju


Adding source terms for compressibility was done by Henning Scheufler here:
https://github.com/DLR-RY/VoFLibrary


In particular, in alphaEqn.H of the compressibleInterFlow solver we now have:


#include "alphaSuSp.H"
advector->advect(Sp,(Su + divU*min(alpha1(), scalar(1)))());

To see how the advect function of isoAdvector handles the source terms see here.


Note that hopefully this work will be integrated in the upcoming OpenFOAM-v2006.


Kind regards,
Johan

[jobin2600]

Hi Johan

Is there any reference material or publications available from which I can understand the theory behind the interface reconstruction part(iso-method) in the isoAdvector library. I went through your papers and couldn't find the mathematical details of the algorithm used.

Any help is highly appreciated.

Best,
Jobin

[roenby]

The reconstruction step of isoAdvector is for instance described here (section 3.1):https://doi.org/10.1016/j.jcp.2019.01.009


Or here (also section 3.1):
https://royalsocietypublishing.org/d...98/rsos.160405