Error using Berendsen thermostat and barostat

LAMMPS LAMMPS Mailing List Mirror
1

Hello,

I am attempting to do MD simulation of a glycine zwitterion in water using the C++ Sept 2010 version of Lammps. When I use “fix NPT” then the MD simulation runs fine, but instead when I use “fix NVE” along with Berendsen thermostat and barostat, then I get the following error:
“Out of range atoms - cannot compute PPPM”

I am using AMBER force field, PPPM for electrostatics and shake for hydrogen atoms. Timestep is 2fs.

I am unable to figure out the problem. I was wondering if someone could help me. Below is the segment of the script that works (using fix NPT) and the script that does not (fix NVE with Berendsen thermostat and barostat)

(A) the script that works:

units real
atom_style full
pair_style lj/cut/coul/long 10.0 9.0
bond_style harmonic
angle_style harmonic
dihedral_style charmm
improper_style cvff
special_bonds amber
read_data data.add1_800
pair_modify mix arithmetic
kspace_style pppm 1.0e-4
neigh_modify delay 0 every 1 check yes
group hydro type 2 4 8
timestep 2.0
fix 2 hydro shake 0.0001 20 0 b 1 3 6 7 a 9
group water type 7 8
fix 3 all npt temp 300.0 300.0 500.0 iso 1.0 1.0 500
run 1000000
timestep 2.0
fix 5 all momentum 1000 linear 1 1 1

(B) the script that does not work

units real
atom_style full
pair_style lj/cut/coul/long 10.0 9.0
bond_style harmonic
angle_style harmonic
dihedral_style charmm
improper_style cvff
special_bonds amber
read_data data.add1_800
pair_modify mix arithmetic
kspace_style pppm 1.0e-4
neigh_modify delay 0 every 1 check yes
group hydro type 2 4 8
timestep 2.0
fix 2 hydro shake 0.0001 20 0 b 1 3 6 7 a 9
group water type 7 8
fix 1 all nve
fix 6 all temp/berendsen 300.0 300.0 500
fix 7 all press/berendsen iso 1.0 1.0 500
run 1000000
timestep 2.0
fix 5 all momentum 1000 linear 1 1 1

Also, the script still gives the same error when I use Berendsen barostat and Nose Hoover thermostat.

Thanks for the help

Sumit Sharma

2

Hello,

I am attempting to do MD simulation of a glycine zwitterion in water using
the C++ Sept 2010 version of Lammps. When I use "fix NPT" then the MD

the first step should always be, that you try to reproduce
the issue with the current version of the cope. lammps
is continuously updated and nobody will try to track down
issues with an obsolete version. thanks.

simulation runs fine, but instead when I use "fix NVE" along with Berendsen
thermostat and barostat, then I get the following error:
"Out of range atoms - cannot compute PPPM"

this is usually a sign of some forces being overly high
or atoms moving too fast as a result of that.

I am using AMBER force field, PPPM for electrostatics and shake for hydrogen
atoms. Timestep is 2fs.

have you tried a shorter time step?
is this with an equilibrated system
or at the beginning of the MD?
did you follow the suggestions that
the manual make in case of lost atoms
or out-of-range atoms?
have you visualized your system?
does the error happen at the first step
or during the MD? and if during after
how many steps? how large are the
energies and are they reasonable?

I am unable to figure out the problem. I was wondering if someone could help

there is far too little information so that anybody
can help you with any specific help. you either
need to provide more information, provide an
(easy!) way to reproduce your problem or
try what the manual (and discussions in the
mailing list archive) suggest to do in cases
like yours.

me. Below is the segment of the script that works (using fix NPT) and the
script that does not (fix NVE with Berendsen thermostat and barostat)

this is pointless. if you want somebody to
track down the issue, you have to provide
a *complete* input. it doesn't have to be
your exact system, but something that
exactly reproduces the problem that you see.

(A) the script that works:

units real
atom_style full
pair_style lj/cut/coul/long 10.0 9.0
bond_style harmonic
angle_style harmonic
dihedral_style charmm
improper_style cvff
special_bonds amber
read_data data.add1_800
pair_modify mix arithmetic
kspace_style pppm 1.0e-4
neigh_modify delay 0 every 1 check yes
group hydro type 2 4 8
timestep 2.0
fix 2 hydro shake 0.0001 20 0 b 1 3 6 7 a 9
group water type 7 8
fix 3 all npt temp 300.0 300.0 500.0 iso 1.0 1.0 500
run 1000000
timestep 2.0
fix 5 all momentum 1000 linear 1 1 1

(B) the script that does not work

units real
atom_style full
pair_style lj/cut/coul/long 10.0 9.0
bond_style harmonic
angle_style harmonic
dihedral_style charmm
improper_style cvff
special_bonds amber
read_data data.add1_800
pair_modify mix arithmetic
kspace_style pppm 1.0e-4
neigh_modify delay 0 every 1 check yes
group hydro type 2 4 8
timestep 2.0
fix 2 hydro shake 0.0001 20 0 b 1 3 6 7 a 9
group water type 7 8
fix 1 all nve
fix 6 all temp/berendsen 300.0 300.0 500
fix 7 all press/berendsen iso 1.0 1.0 500
run 1000000
timestep 2.0
fix 5 all momentum 1000 linear 1 1 1

Also, the script still gives the same error when I use Berendsen barostat
and Nose Hoover thermostat.

...and it will also give the same error message
when you do hundreds of bad things. the error
message is not very specific and the real cause
is elsewhere.

cheers,
    axel.