Dear Lammps experts,
I have classified my model domain to several chunk bins by compute chunk/atom command in order to calculate the temperature/chunk for each chunk. However, at the end I have noticed that it is not possible to show the chunk results with dump custom command!
I mean lammps can only print the chunk results with fix nve command which is not good enough for presenting the results. Actually, I expected to attribute the chunk properties to their atoms in order to plot them by dump custome and better observation…please let me know whether it is possible or not?!
with the best regards,
Dear Lammps experts,
I have classified my model domain to several chunk bins by compute chunk/atom command in order to calculate the temperature/chunk for each chunk. However, at the end I have noticed that it is not possible to show the chunk results with dump custom command!
that is not correct. please see below.
I mean lammps can only print the chunk results with fix nve command which is not good enough for presenting the results.
wrong. fix nve does not produce any output. the typical way to output per-chunk data would be fix ave/chunk, which i would consider perfectly suited to represent that kind of data.
Actually, I expected to attribute the chunk properties to their atoms in order to plot them by dump custome and better observation…please let me know whether it is possible or not?!
yes, it is possible. you need to pay better attention to the documentation. there is lots of useful information. check out compute chunk/spread/atom which does exactly what you seem to be looking for.
but i would question the usefulness of what you are asking to do. averaging over chunks and then projecting that information back to atoms is doing some rather arbitrary averaging and may be misrepresenting information significantly. i would rather output the per-atom kinetic energy, and if that is too noisy, use fix ave/atom to output time averaged numbers. that would rather be meaningful, as it will be correlated with each atom and not subject to some arbitrary partitioning.
axel.
Thanks for your explanation.