Compute the pressure and Stresses

LAMMPS LAMMPS Beginners
1

Dear All,

I am using Lammps (lammps/29sep2021) to compute the stresses and the pressure:

I did the following:

variable          pr    equal press/10000
variable          sxx   equal pxx/(10000)
variable          syy   equal pyy/(10000)
variable          szz   equal pzz/(10000)
variable          sxy   equal pxy/(10000)
variable          sxz   equal pxz/(10000)
variable          syz   equal pyz/(10000)


velocity all create 300.0 1972
thermo          1     #output thermodynamics every 1000 timesteps
thermo_style    custom step temp pe press lx ly lz   density

fix statdump all print 1 "${st} ${tm} ${Et} ${Ep} ${Ek} ${tmp} ${pr} ${sxx} ${syy} ${szz} ${sxy} ${sxz} ${syz}" screen no file dump.stat
run 0

The result obtained at the end of the simulation reading the dump.stat is the following:

              average pressure (GPa): 27.46786
   average stresstensor xx xy xz (GPa):      31.0318      -0.0100      -0.0095 
   average stresstensor yx yy yz (GPa):      -0.0100      23.3890       0.0095 
   average stresstensor zx zy zz (GPa):      -0.0095       0.0095      27.9827

I was expecting (31.0318 + 23.3890+27.9827 )/3=27.467833333333335 should be of the opposite sign with the average pressure.
It that correct? If yes, please, what could be the source of the mistake?
Thank you.

2

Why? The result you get is consistent with what you compute.
You define sxx as pxx/10000 and the other tensor elements similar and pr as press/10000. Since press is (pxx+pyy+pzz)/3.0 your averaged output is consistent with what you define. If you want sxx be the negative of pxx/10000 you have to change your variable definition accordingly.

3

Thank you for your answer.

Because of compute this :
peratom all stress/atom NULL
compute p all reduce sum c_peratom[1] c_peratom[2] c_peratom[3]
variable press equal -(c_p[1]+c_p[2]+c_p[3])/(3*vol)
thermo_style custom step temp etotal press v_press.

I have found it following the link compute stress/atom command — LAMMPS documentation

4

But there is no compute stress/atom in your input.