A simple package for generating uniform SPH glass in a periodic box written by Isaac Backus.

Download and install depedencies:

• ChaNGa
• pynbody
• diskpy Note that your ChaNGa presets must be set up in diskpy

Download:

git clone https://github.com/ibackus/sphglass.git

Then add the sphglass/ directory to your PYTHONPATH.

import sphglass
import matplotlib.pyplot as plt

# Generate a glass in a box shape 1 by 2 by 3 (in x,y,z).  f is a SimSnap
# (see pynbody).  The only relevant quantites are the position and density
f = sphglass.glassBox(1000, [1,2,3])
# A good glass should have nearly uniform density, although there
# will always be some scatter
plt.plot(f['z'], f['rho'], 'x')
# If you think this is not uniform enough, you try to make it more glassy:
f = sphglass.reglassify()
plt.plot(f['z'], f['rho'], 'x')

The algorithm used here for generating a glass is pretty simple. The procedure is:

1. Generate random particle positions in a box -or optionally- generate them on a grid.
2. Create a tipsy snapshot with only gas particles
3. Time evolve in a periodic box with no gravity and a damping force. A random distribution (Poisson distribution) of particles is in a higher energy state (higher density) and will gradually relax to a lower energy, glass-like, uniform density state given enough time. The trick is to just estimate how much time is required for that and to have a dissipative force.

The default parameters used for this are stored in glassdefaults.param and can be edited.