[foadsf]

I'm looking into Ankush Gupta thesis:

Shock Tube with rhoCentralFoam, sonicFoam, Lax-Friedrichs and MacCormacks

and there it has been mentioned that unlike rhoCentralFoam solver, sonicFoam requires the magU file. meanwhile in this post it has been mentioned that you need magU for both solvers. i was not able to find any examples in my own OpenFOAM installation 5.x folder including a magU file but here I found it on the github repo for the older version. My questions:

• What is this file?
• what is the difference with U boundary/initial condition file?
• in what cases you must have it?
• why it is required for sonicFoam but not the rhoCentralFoam solver?
• non-relevant but what does it have to do with foamCalc? it just comes up in many search results!

[dviam]

Hi Foad,

For starters, to my knowledge, the magU file is not "a required" file in any of the solvers you mentioned. It is a result of postProcessing. In older versions of OpenFOAM there was a tool known as foamCalc which allowed users to extract relevant information from their simulations through the command line. One such pieces of information would be the magnitude of velocity at various boundaries or in the internal field. In newer versions of openFOAM, which you seem to be using, a lot of the tools used in post processing have been consolidated into the postProces utility.

If you type postProcess -help you will be able to see a summary of it's uses. In order to calculate the magnitude of velocity at all time-steps, one now needs to type:

postProcess -func "mag(U)"

This will create the magU file in each of the time directories. It will be a list of velocity magnitudes at all grid points in the internal field and at the boundaries of the domain.

Hope this helps.

- Dorien