Hi,
I’m doing MSST simulation of SPC/E flexible water using kspace_style is pppm. For higher strain rate case, after some steps (after ~2ps) the sample get higher kinetic energy along with a compression upto ~60%, I get the error “out of range atoms - cannot compute pppm”.
From the previous discussion in this forum, I have made some tricks to overcome this issue but failed.
- I have set smaller timestep but after that specific time ~2ps it throws error.
- have used neighbor_modify delay 0 every 1 check yes, but failed.
- Used “communicate” command with various cutoff from 5 - 60 A yet I got that error.
- But surprisingly when I used pppm 1.0e-4 in the place of 1.0e-6, it works for more steps but energy conservation suffers significantly.
- As the last when i change neighbor 1.0 in the place of 2.0 (real unit) it works for few more steps but the error is same.
How can I fix it ? Any idea will be appreciated.
Hi,
I'm doing MSST simulation of SPC/E flexible water using kspace_style is
pppm. For higher strain rate case, after some steps (after ~2ps) the sample
get higher kinetic energy along with a compression upto ~60%, I get the
error "out of range atoms - cannot compute pppm".
From the previous discussion in this forum, I have made some tricks to
overcome this issue but failed.
1. I have set smaller timestep but after that specific time ~2ps it throws
error.
how much smaller? perhaps it was not small enough.
Hi Rajdeep,
I know nothing about your MSST simulation, but I know one case
the “out of range atoms” error may occur. It is when your simulation
box volume changes drastically, hence atoms will move out of the
PPPM grid which is calculated at the beginning of your simulation.
If this is the case, you can overcome it by writing several lines of
“run ${shorterRunningSteps}” command in your input script. In this
way, the PPPM mesh will be recomputed everytime when each
small section starts, according to the current configuration.