Steve,
In the last mail, you said thaat i have to equilibrate it well to reduce the fluctuations. Do you mean i have to reduce the damping parameter in fix/nvt???
I also tried fix/langevin for an exactly symmetric structure. But then when i output the tempeature of a region at every timestep it’s still fluctuating from 200-450K. I also tried a different way by using nvt first for a certain time on the whole system.and then using fix/langvin over a couple of rgions where i want to control the temperature. But then later when i calculate the temp, it’s still fluctuating.
fix 14 all nvt 300.0 300.0 0.1 drag 1.0
timestep 0.00025
run 100000
unfix 14
fix 16 all nve
fix 15 all langevin 300.0 300.0 0.1 48279 tally yes
compute 1 low temp
compute 2 high temp (these are two regions of about 25 atoms each)
thermo_style custom temp pe ke etotal c_1 c_2
thermo 1
timestep 0.0005
run 1000000