[gmx-users] large pressure

[Shuangxing Dai]

Hi, all,
    I was wondering why I got large pressure after energy minimization. In MD, the pressure and pressure tensor is defined like ( equation 4.151 in manual):
                P=2*(E_kinetic  +  W)/(3*V), V is volume and W is Virial.
I run a simple system, fcc gold with LJ potential to check the pressure after energy minimization.  I got:

 Pressure (bar)
    8.75158e+03

Steepest Descents converged to Fmax < 1 in 1 steps
Potential Energy  = -1.4870625e+03
Maximum force     =  2.2925207e-05 on atom 1606
Norm of force     =  7.9803776e-06  

The force is very small and the energy minimization finished in 1 steps, since all the atoms are at equilibrium. So I was wondering why I got this huge pressure. And in a lot of simulations, I found huge pressure after energy minimization. Even in free energy tutorial on Wiki, I found large pressure as larege as 1 GPa. After energy minimization, the forces are very small and no velocity, so the pressure should be close to zero.
Anyone has any idea about this?
PS: Why the forces in traj.trr are all zeros? 

This is the  mdp file:

; RUN CONTROL PARAMETERS =
integrator               = steep
nsteps                   = 500
; Output frequency for coords (x), velocities (v) and forces (f) =
nstxout                  = 1
nstvout                  = 1
nstfout                  = 1
; Output frequency for energies to log file and energy file =
nstlog                   = 1
nstenergy                = 1
; ENERGY MINIMIZATION OPTIONS =
; Force tolerance and initial step-size =
emtol                    = 1
emstep                   = 0.01
; NEIGHBORSEARCHING PARAMETERS =
; nblist update frequency =
nstlist                  = 1
; ns algorithm (simple or grid) =
ns_type                  = simple
; Periodic boundary conditions: xyz or none =
pbc                      = xyz
; nblist cut-off         =
rlist                    = 0.9
; OPTIONS FOR ELECTROSTATICS AND VDW =
; Method for doing electrostatics =
coulombtype              = PME
rcoulomb                 = 0.9
; Dielectric constant (DC) for cut-off or DC of reaction field =
epsilon-r                = 1
; Method for doing Van der Waals =
vdw-type                 = Cut-off
; cut-off lengths        =
rvdw                     = 0.9

Thanks in advance,
Shuangxing Dai
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090709/3b27c290/attachment.html>