[lammps-users] Error: Per atom virial was not tallied on needed timestep

LAMMPS LAMMPS Mailing List Mirror
1

Hello everyone,
I have been getting the error " Per-atom virial was not tallied on needed timestep" or sometimes its “energy” instead of “virial.”
I had seen the previous mails on this subject earlier and did not find anything. Moreover I cannot find those emails in the archive now. Can someone please explain why we get this error?

I have seen that I get the error when I change the dump frequency of either a energy/atom or stress/atom compute.
Written below, is a small sample input file where I get the error during the 3rd run after the 7000th timestep. I have tried to reduce the size of the problem while keeping it as similar as possible to my original system, hence the 3 runs.

Basic settings

units metal
dimension 3
boundary p p p
atom_style atomic

neighbor settings

neighbor 2 bin
neigh_modify delay 1

create geometry, unit 3.524 A

lattice fcc 3.524
region box block -20 20 -20 20 -2.5 2.5

#create_atoms
create_box 1 box
create_atoms 1 box
mass 2 58.71

EAM potentials

pair_style eam
pair_coeff * * //Ni_u3.eam

********** Define groups ************

Boundary atoms

region btop block INF INF 19 INF INF INF
region bbot block INF INF INF -19 INF INF
group btop region btop
group bbot region bbot

group boundary union btop bbot
group mobile subtract all boundary

A region in the middle to dump atoms

region dumpregion block -10 10 -10 10 INF INF
group dumpregion region dumpregion

initial velocities

compute new mobile temp/deform
velocity mobile create 100 887723 temp new

compute PE and stress

compute PE all pe/atom
compute stress dumpregion stress/atom pair

fixes

fix 1 all nve
fix 2 mobile temp/rescale 100 1E-3 1E-3 0.05 0.5
fix 3 boundary setforce 0.0 0.0 0.0

run

timestep 0.001
thermo 200
thermo_modify temp new

dump 1 all cfg 1000 all_.cfg id type xs ys zs type c_PE c_stress[2]
dump 2 dumpregion cfg 1000 cfgdata_.cfg id type xs ys zs id type c_PE c_stress[2]
dump 3 dumpregion custom 20000 energy.dat c_PE c_stress[2]

run 2000

#****************** 2nd run **********************#

Fixes

unfix 1
unfix 2
fix 4 boundary setforce 0.0 0.0 0.0
fix 5 all nvt 1E-3 1E-3 5 drag 0.1

Apply fix deform to apply a strain

fix deform1 all deform 1 y delta 0 0.1 remap x

dump_modify 1 every 500
dump_modify 2 every 500
dump_modify 3 every 500

timestep 0.001
run 5000

#****************** 3rd run **********************#

strain rate: 3.524 A / 10000 timestep

unfix deform1
fix deform2 all deform 1 y delta 0 0.1 remap x

dump_modify 1 every 200
dump_modify 2 every 200
dump_modify 3 every 200

timestep 0.001
run 5000

2

It has been fixed I think. See the 18th May patch for this or install the latest version and give your simulation a test run.

3

Hi Vikas,
Thanks for your mail. I tried it with the latest version of LAMMPS. I still get the same error.
Santosh

4

fix 5 all nvt 1E-3 1E-3 5 drag 0.1

This command is not the syntax for the most current version
of LAMMPS. When I change it (add "temp"), then your
script runs fine with the current version, without the per-atom virial
error you indicated.

Steve

5

Thanks for your mail Steve!
I was away for the last few days, so just checked the input file with new LAMMPS.

Sadly, I still get the error with the May 10th distribution. I am attaching the log file for you to see.
The error comes in the last part of the run when I change the dump frequency to 200, but again there was no error when the dump frequency was changed to 500 before that.
Santosh

log.lammps (7.06 KB)

6

May 10 is not the current version of the code.
This script (below) from bench/in.lj works fine for me. It seems to have
all the dump properties of your much more complex script.

Steve

# 3d Lennard-Jones melt

units lj
atom_style atomic

lattice fcc 0.8442
region box block 0 5 0 5 0 5
create_box 1 box
create_atoms 1 box
mass 1 1.0

velocity all create 1.44 87287 loop geom

pair_style lj/cut 2.5
pair_coeff 1 1 1.0 1.0 2.5

neighbor 0.3 bin
neigh_modify delay 0 every 20 check no

fix 1 all nve

thermo 200

compute 1 all pe/atom
dump 1 all cfg 1000 tmp1.*.dump id type xs ys zs c_1
dump 2 all cfg 1000 tmp2.*.dump id type xs ys zs c_1
dump 3 all cfg 20000 tmp3.*.dump id type xs ys zs c_1

run 2000
dump_modify 1 every 500
dump_modify 2 every 500
dump_modify 3 every 500
run 5000
dump_modify 1 every 200
dump_modify 2 every 200
dump_modify 3 every 200
run 5000