Dear board,

I am trying to carry out MD simulations on liquids and have a couple of quick questions.

1) I built a cell using amorphous builder, then ran 1ns of NVT Discovery Dynamics (Anderson), DD,  to equilibrate it. Because I am interested in the motions within the liquid I want to do a production run using the NVE ensemble, using the same velocities as the last frame of the NVT equilibration stage.  Every time I try to change the ensemble to NVE, and to reuse the velocities I come across a major problem. To check if everything is going to run correctly I do a quick 200frame simulation with this new setup.  When I check the output file, instead of using (NVE, reusing the velocities) it either keeps (NVT, reuses the velocities) or uses (NVE, creates new velocities from a Boltzmann).

Can anyone tell me the right way to go about changing the ensemble and keep the velocities from the equilibration step?

2) When I tried to do the NVT equilibration with a Nose thermostat, instead of the temp tracking at 298K (like I set it at), it tracks at ~273K.  Why is this?

3) I am using the Compass forcefield in the DD, with partial charges estimated from a DFT calculation. The DD status says that H bond contribution to the energy is 0 during the simulation, but I know I should have H-bonds.  How do I "turn H bonds on"?

cheers,

Andrew.