langevin damping parameter

Dear Lammps users,

I have a practical simulation question. Is there a good rule of
thumb regarding the damping constant for a langevin thermostat
for solid materials? By what measure can you judge if a langevin
thermostat is overdamped? Also is there a problem establishing
a valid ensemble if you have a very small damping parameter
(25 time steps)?

I can quickly achieve a correct lattice parameter using velocity
rescaling and a berendsen barostat with very small fluctuations.
However, I understand that this approach does not achieve a
correct ensemble and will eventually result in a flying ice cube.
I would like to use the temp/rescaling + press/berendsen to get
the right box size and then switch over to NPH with a langevin
thermostat to get the correct ensemble.

Is there such a thing as a langevin damping parameter that is too
small, and how can you tell? If someone could suggest some
practical advice or suggest a reference, I would greatly appreciate it.

Thanks for the help,

Like most thermostats, 100 timesteps is a good
damping time for temperature. See the fix nvt
doc page for additional discussion.