Integrator used for granular simulations


I am trying to analyze the effect of a high speed projectile impact on a granular spherical bed bounded by walls. I have already carried out this simulation in YADE. But we are observing a growth in the total energy of the system during initial timesteps. This energy growth stops after sometime and stays at that higher energy level. We are suspecting this phenomenon due to the integrator used in that software. I want to ask the users of YADE whether moving to LAMMPS will mitigate this energy growth effect ??

LAMMPS uses a simple time integrator (velocity Verlet).
Granular systems typically have “hard” contact potentials
meaning you need to be careful to use a small enough
timestep, esp if particles are moving fast. LAMMPS has
a couple options (fix nve/limit, fix dt/reset) to try to insure
your timestep is not too large or adapt it as needed.

But neither of these are very sophisticated either.