This is for the case of uniform, horizontal magnetic field, of strength 1 kG, advected into the computational domain by inflows at the bottom at a depth of 20 Mm.  The magnetic field was slowly increased (with an efolding time of 5 hours) until the input field at 20 Mm reached 1 kG and thereafter kept constant. The horizontal size is 48 Mm.  The domain
extends from about the temperature minimum down to 20 Mm below the optical depth unity surface.

The Y-AXIS is the VERTICAL direction, pointing downward.  The x-axis points east and the z-axis points north.  
The coordinate system is rotating at a rate corresponding to a lattidute of North 30 deg.
Runs have different dimensions:  504x500x504, 1008x500x1008 and 2016x500x2016, with resolution 96, 48 and 24 km respectively horizontally.  Vertical resolution is 12-80 km all.

Horizontal boundaries are periodic.  In the vertical direction there are 5 ghost zone boundary layers at top 
and bottom that are unphysical.  In IDL the vertical physical region is iy=5,494.
     
Horizontal slices are stored as IDL save files, with variables interpolated to 3 levels: tau_cont=1, 0.1, 0.01.
     Files are labeled by the time of the first slice (in units of 100 sec) since the magnetic field started to 
     be advected in at the bottom.
     There are slices of B_x,y,z (in bb*.save), V_x,y,z (in vv*.save) as well as the emergent continuum intensity (in int*.save).
     Variables are labeled as e.g. Bx1(tau=1),Bx01(tau=0.1),Bx001(tau=0.01).
     Slices are stored at 1 min intervals and the times are stored in the variable timet.
     Note: in some files timet[0] is incorrect. It should be that timet[0]=timet[1]-0.6.

The units are:
     length: Mm=10^8 cm
     density: 10^-7 g/cm^3
     time: 100 sec
     magnetic field: kG
Thus derived unts are:
     velocity 10 km/s
     pressure 10^5 dyne/cm^2
     energy per unit mass 10^12 erg/gm
     flux 10^11 erg/cm^2/s
The grid is staggered with scalar variables: density, energy, temperature, pressure defined at the integer cell 
     centers.
     The vectors: momenta, velocity, magnetic field are defined at the integer-1/2 faces.
     The vectores: electric field, current are defined at the cell edges.

Four hour averages (with 2 hour cadence) will be added as they are produced.

Raw data files (*.scr and *.dat), where available, are fortran direct access unformated files.
Each snapshot has 9 variables: density, x,y,z-momenta, einternal energy per unit volume, temperature, B-x,y,z.

They can be read using IDL as follows:
     openr,u,/get,'filename'
     a=assoc(u,fltarr(nx,500,nz))
     rho=a[0]
     px=a[1]
     py=a[2]
     pz=a[3]
     d=a[4]
     t=a[5]
     bx=a[6]
     by=a[7]
     bz=a[8]
Slices could be read as, e.g. r=a[*,34,*,0], where e.g. 34 is the depth index.
The grid can be read in with the idl procedure: readmesh.pro
     For each axis it contains: cell centered increment, face centered increment, cell centered coordinate, 
     face centered coordinate, 1/(derivative of cell centered coordinate), 1/(derivative face centered coordinate)
The pressure, opacity and electron density are stored in the table 'table.dat' as a function of log(density) 
and internal energy per unit mass (e/rho).
     In IDL first do IDL> eos_table,'table.dat' to initialize, 
     Pressure is IDL> p=exp(lookup(alog(rho),e/rho,iv=0))
     Opacity is IDL> kappa=exp(lookup(alog(rho),e/rho,iv=1))
     Electron density is IDL> ne=exp(lookup(alog(rho),e/rho,iv=3))