Different temperatures at same timestep

Sir,
Do you need a working version of the input script?
I am using the default temperature compute.
Following is a stripped down version of the script:

read_data “data.singleTIP4P” add append
replicate 16 32 ${zr}

include “settings”

velocity all create ${Tdown} 102486 dist uniform loop geom
compute temp_F water temp/com

fix nvt1 water nvt temp 300 300 100
dump trj all custom 500 wat.lammpstrjnvt.${time_period} id type x y z vx vy vz
run 5000
unfix nvt1
undump trj

fix nve1 water nve
dump trj all custom 500 wat.lammpstrjnve2.${time_period} id type x y z vx vy vz
run 5000
unfix nve1
undump trj

#end

Thank you

Sritay Mistry

Graduate Student, Aerospace engineering
Indian Institute of Technology, Kanpur

A small, but perhaps significant correction in the input script I previously provided is that I have a fix efield before the nve run, but the magnitude of this fix is numerically zero. The input script now reads as

read_data “data.singleTIP4P” add append
replicate 16 32 ${zr}

include “settings”

velocity all create ${Tdown} 102486 dist uniform loop geom
compute temp_F water temp/com

fix nvt1 water nvt temp 300 300 100
dump trj all custom 500 wat.lammpstrjnvt.${time_period} id type x y z vx vy vz
run 5000
unfix nvt1
undump trj

fix Efield water efield 0 0 0

fix nve1 water nve
dump trj all custom 500 wat.lammpstrjnve2.${time_period} id type x y z vx vy vz
run 5000
unfix nve1
undump trj

#end

Thank you

Sritay Mistry

Graduate Student, Aerospace engineering
Indian Institute of Technology, Kanpur

A small, but perhaps significant correction in the input script I
previously provided is that I have a fix efield before the nve run, but the
magnitude of this fix is numerically zero. The input script now reads as

​this is not useful. it is missing important information. from other files
and from this input.
for example, there would have to be a fix shake, fix rattle or fix rigid
somewhere.
it also looks like an overly and needless complex input for debugging. so
please try to make it a​s simple and as small as possible (e.g. without
variables, where they can be avoided) also also so that it reproduces the
issue at hand as quickly as possible.

ideally, you should be doing your debugging yourself. that is by starting
from an empty input and then adding only the minimum amount of lines from
the old input in order to build a meaningful simulation. you should be
seeing yourself which statement, when added, is causing the issue. LAMMPS
input scripts are not so different from other (generic) script languages
(e.g. python or matlab), so the debugging isn't that much different either.

axel.