[lammps-users] bug (or feature) regarding Nose-Hoover thermostatting and fix rigid

Dear all,

I'm using LAMMPS 16 April08 build. This is just a report on an issue
that took me a while to figure out.

It seems important to define rigid body fixes AFTER the nvt
thermostat/integrator has been defined.
If it is done the other way round the translational velocities all go
to zero but and the system develops a huge angular momentum.

Best wishes,

Fix rigid and fix nvt both do time integration. So you should
not be integrating the same atoms with both fixes.

You didn't get this warning?
WARNING: One or more atoms are time integrated more than once