next up previous contents
Next: Equal-mass Particles Up: Running a Calculation: Initial Previous: Automatically Generated Input Files:   Contents


Single Stars

In theory, there are an infinite number of ways to set up a star in SPH. We offer built in routines to cover two important cases. First, a run where all SPH particles are given the same mass. In order to do this, we need to use a variable interparticle spacing to get the correct density profile. Alternately, we can use evenly-spaced particles of varying mass to reproduce a given density profile. It is easiest to do this by setting up a cubic lattice, but unfortunately, such configurations are unstable against deformations, and thus take a longer time to relax. Instead, we use a hexagonal close-packed lattice, the configuration found, among other places, in the carbon atoms in a diamond, since these lattices are absolutely stable.

For all single star models, we recommend you relax your configuration long enough for all the vibration in the lattice to damp away significantly. Typically, the stellar model you compute will end up slightly smaller than the radius you requested. This is to be expected, since the extent of the density distribution extends two smoothing lengths beyond the edge of the particles. As a rule of thumb, don't worry too much abut small deviations from what you expect in the very outer edges of a star, since SPH cannot interpolate the density profile at an edge exactly (although it still does a better job than grid based codes).



Subsections
next up previous contents
Next: Equal-mass Particles Up: Running a Calculation: Initial Previous: Automatically Generated Input Files:   Contents
Joshua Faber 2003-06-28