[lammps-users] Pdamp


I wanted to know the effect of Pdamp, when I am doing a quench in the pressure.

I am quenching at say 0.001 atm/fs. I did two independent simulations with different
Pdamp values. First Pdamp value was 1000 fs and second Pdamp value was
59651221 fs (equivalent to 155 Mil MD steps).

Second value I choose because of my wrong understanding of what is given in manual

“a value of 1000.0 means to relax the pressure in a timespan of (roughly) 1000 time units”

I thought pressure relaxes from Pstart to Pstop in this timespan and hence I gave that value.

Outputted pressure is same from both simulations. But the density fluctuation in the second
simulation (very big Pdamp case) is almost zero.



The pressure changes from Pstart to Pstop over the course of your simulation independently of the Pdamp parameter. The Pdamp parameter governs how quickly the local oscillations are damped. The larger the constant, the “gentler” the correction. A value of 6e7 is for all practical purposes equal to no correction whatsoever.


Dear LAMMPS users,

I want to dump the stress tensor for each atom:

compute 1 all stress/atom virial
dump 1 all custom 250 dump.custom.* id type x y c_1

ERROR: Dump custom compute does not calculate per-atom vector

Am I doing something wrong within the dump custom command?

With kind regards,
Thomas Wagner.

Dear Thomas,

Try using
dump 1 all custom 250 dump.custom.* id type x y c_1[1] c_1[2] c_1[3] c_1[4] c_1[5] c_1[6]

for all the components of the stress. It could help.

If you're ramping the pressure from Pstart to Pstop, then
you need to do so over a timescale many times longer than Pdamp.