Hello everyone,
I am using the stable version of lammps which has been released on 30 Jul 2016. I am going to get the temperature profile of gas which is located between two parallel walls.
The lower wall is colder than gas and the upper is hotter so there will be a heat flux from upper to lower wall and this cause temperature variation through the gas. Actually I am going to compute the temperature profile with fix ave/chunk and it works fine. The code is like this:
Hello everyone,
I am using the stable version of lammps which has been released on 30 Jul
2016. I am going to get the temperature profile of gas which is located
between two parallel walls.The lower wall is colder than gas and the upper is hotter so there will be a
heat flux from upper to lower wall and this cause temperature variation
through the gas. Actually I am going to compute the temperature profile with
fix ave/chunk and it works fine. The code is like this:
#####################################################################################
dimension 3
boundary p s p
atom_style atomic
neighbor 0.3 bin
[...]
In theory the temperature profile for these two case should be the same but
unfortunately they are not. It seems that I can’t use the restart file
correctly in my case. I don’t know whether lammps can continue the fix
ave/chunk command after restart file, correctly or not.
it can, but you are not really specifying the same bins.
please note, that you are using shrinkwrap boundary conditions in
y-direction and that has an impact on your box boundaries and due to
the way how define your chunks, also on the binning.
if you plot your profiles by bin coordinate, the plots match within
the statistical errors. see the attached graph.
So in this way I can’t use the restart file at all and it’s a big problem in
large simulations.It was really strange that in the case B, even the number of bins has been
increased with no obvious reason. (Bins which is used for fix ave/chunk
command in the height of the gas between two walls)
the reason is obvious, if you look at the box dimensions!
in the first run you have:
create_box 3 box
Created orthogonal box = (0 0 0) to (50 15 5)
in the second run you have instead:
read_restart restart.500000
orthogonal box = (0 -0.449365 0) to (50 14.8181 5)
that is a different size and origin for the restart.
axel.
