PLUTO code for plasma wind interaction with a spherical obstacle (2D, aligned B field)
If you would like to run the below simulations you will first need to install the PLUTO simulation framework. The below were run using PLUTO v4.2 with CHOMBO v3.2 for adaptive mesh refinement, in order to simulate on a 2048x2048 equivalent grid. However, similar results can be obtained even on a 256x256 uniform grid without CHOMBO. In the following I will assume you have already read the very good documentation available at http://plutocode.ph.unito.it/files/userguide.pdf, and have the test problems running and working.
VenusLike_cs200.gif: An animation of the Venus-like scenario with cs=200km/s and RE/RDM=2; the black corresponds to Gran Sasso and the red to Stawell.
Herein lies the DarkSphere PLUTO simulation code files.
definitions.h: Sets up the 2D MHD simulation parameters. We declare two user input parameters, MACH and BFIELD, which are defined as the plasma wind mach number and the (aligned) B field strength respectively [their values are NOT set here]. We also set some UNIT_VALUES, but thanks to the scaling properties of the MHD equations (and the fact we are interested in only the steady state solutions) it doesn't really matter what these are except to avoid numerical small/large number problems.
pluto.ini: Once the pluto simulation has been made, this is the file you can feed in to change parameters for the run. [Grid] We run with 128x128 initial uniform grid (with 4 levels of AMR this becomes 2048x2048). [Time] Here tstop is set to 2000, which is overkill for the supersonic simulations, but necessary for subsonic, since matter can propagate upwind and interact with the inflow boundary. The user should run several test runs to make sure the system has come to an approximately stable configuration by time tstop. [Boundary] Note here that (X1,X2) are the (radial,theta) coordinates [NOT cartesian]. The X1-BEG boundary should be set to userdef for the Moon-like case and reflective for the Venus-like case [see the init.c file]. [Chombo HDF5 output] The output is set only to write the last frame to file; obviously this can be changed for testing purposes. [Parameters] Here one can set the input MACH and BFIELD.
init.c: The plasma is initialised with flat density, pressure, and velocity. Most of the action is in the UserDefBoundary function. We first have an INTERNAL_BOUNDARY condition which puts a floor on the density; this doesn't come into play for the unmagnetised simulations but can be necessary for highly magnetised simulations (such as Moon-like with BFIELD = 1). The X1_END boundary condition is the inflow/outflow conditions for theta above/below pi/2. The X1_BEG boundary is the dark sphere surface, and should be called only for the Moon-like case; it is absorbing for the windward side and reflective for the leeward side. If the user would like, the X2_END boundary condition can be used for the supersonic Moon-like simulations as the inflow boundary with theta=[0,pi/2] in pluto.ini (since why bother simulating the upwind conditions when there is no upwind activity!).
data.0130.hdf5: Output data type for CHOMBO simulations. This particular file is our output for Venus-like with MACH = 1.3.
xxxM1.30.dbl: Output data type for uniform simulations. These files are actually created from the .hdf5 file using the python script MakeDataFiles.py.
MakeDataFiles.py: Takes in the .hdf5 files and spits out .dbl files where only every Sk cells are read. With e.g. Sk=8 this reduces the 2048x2048 tables to 256x256, for testing and/or to speed up the downstream analysis. It requires a working implementation of h5py and pyPLUTO (see the PLUTO documentation).
ChangeFileNames.sh: A shell script to change the data output file names from the job scripts [in /jobMaker] into a format data.xxxx.hdf5 which can be read in using pyPLUTO.
MakePlot.py: Uses pyPLUTO to read in .hdf5 files and make a plot of e.g. density.
Plot 130.png: Example output from MakePlot.py of the density for MACH = 1.3 Venus-like simulation.
jobMakerMoon.sh: If you have a cluster system here is a script which will make job submit files for each MACH number you would like to run (here we have Moon-like simulations with MACH = 0.74:1.77 in steps of 0.01). Each job navigates to the location above where you already have a compiled and working version of the DarkSphere PLUTO code, copies that folder to a Temp folder, moves inside it and overwrites the pluto.ini file for the appropriate MACH, runs pluto, copies the output data file to some location, and then deletes the Temp folder. You can submit all the jobs (here with qsub) by running the JobSubmitter.sh file created alongside all the jobs.