Firstly, please avoid posting the same question twice (see Minimization curve ). Justin has replied there. In that thread, please provide the output and logs from the crashing nvt equilibration. That is where the problem occurs. The energy minimization looks fine.