Re: Energy drif in NVE ensemble

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Thu Jan 15 2009 - 11:56:33 CST

Are you applying external forces to the system? You appear to have IMD
turned on...
Peter

eprates_at_iqm.unicamp.br wrote:
> Dear namd users,
> I am simulating a system constituted of about 23000 atoms (water molecules
> and one very small polimer - PEO of 8 monomers ).
> After 4 ns of equilibration (NPT), I have run 250 ps in the NVE ensemble,
> at 298 K. The problem is that the energy increased much more than a
> resonable 0.5% of the initial value.
> I have already read the previous messages about this subject, but even so,
> I could not find out the cause to this problem.
> Just bellow I copied the input file... Can anyone see a parameter that
> could be the cause of this problem?
> I thank you all in advance.
> Érica T. Prates
>
> # Molecular system
> structure PEO-4_fim.psf
>
> ## Coordinates in PDB format
> coordinates peo4-term1.out.coor
> extendedsystem peo4-term1.out.xsc
> velocities peo4-term1.out.vel
>
> ## Constrain all bonds including hydrogens
> rigidbonds all
>
> wrapall on
>
> ## Force field
> paratypecharmm on
> parameters par_PEO-4
> exclude scaled1-4
> 1-4scaling 1.0
>
> ## Option short-range interaction approximations
> switching on
> switchdist 10.0
> cutoff 15.0
> pairlistdist 20.0
>
> margin 0
> stepspercycle 10
>
>
> ## Particle mesh ewald
> pme yes
> pmegridsizex 32
> pmegridsizey 32
> pmegridsizez 32
> #(the length of the cubic box is of about 33 A)
>
> ## Output
> outputenergies 100
> outputtiming 100
> outputPressure 100
> binaryoutput no
> imdon yes
> imdfreq 1
> imdport 3111
>
> restartname peo4-din2
> restartfreq 1000
>
> outputname peo4-din2
> dcdfreq 4
> XSTfile peo4-din2.xst
> XSTfreq 1000
>
> ## Integrator
> timestep 1
> run 250000
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:15 CST