langevin tally command clarification

LAMMPS LAMMPS Mailing List Mirror
1

Dear users,

I am trying to compare the work done by thermostat in a steady couette flow and couette flow with oscillating wall. I know that the cumulative heat added/subtracted by the thermostat can be known by using the keyword tally along with thermostat. I needed clarifications with two things, as I am a bit confused:

1) the sign associated with the values displayed does +ve mean heat added to the system and -ve mean heat removed from the system by the thermostat or vice versa. (have provided the output for the two cases in the end)

2) the manual says that the "fix is 'extensive'", so if we are comparing values of systems with two different number of atoms do we have to scale it with the number of atoms.

Appreciate all the help.

Thank you.
BR/Joseph

steady couette flow:
.
fix 3 all langevin 1.10 1.10 1.0 699483 tally yes
.
velocity upper set 0.785 0.0 0.0 units box
.

   200000 0.93560932 63.198176 265.11403 796491.57 -0.41558024
   201000 0.93447718 63.196203 265.11055 796491.57 -0.41360458
   202000 0.93531263 63.196136 265.1083 796491.57 -0.41353773
   203000 0.93460914 63.196951 265.11295 796491.57 -0.41435216
   204000 0.9356674 63.197918 265.1153 796491.57 -0.41532026
   205000 0.93510296 63.197421 265.11197 796491.57 -0.41482273
   206000 0.93610267 63.197686 265.11063 796491.57 -0.41508595
   207000 0.93600379 63.197968 265.10929 796491.57 -0.41536791
   208000 0.93555026 63.198098 265.11453 796491.57 -0.41549651
   209000 0.93643791 63.198055 265.10914 796491.57 -0.41545272
   210000 0.93618822 63.196797 265.10408 796491.57 -0.41419364

unsteady couette flow:
.
fix 3 all langevin 1.10 1.10 1.0 699483 tally yes
.
fix 8 upper move wiggle 10 0.0 0.0 80.0 units box
.

   500000 0.93797867 63.202638 265.12212 796491.57 0.17071563
   501000 0.93828836 63.201902 265.11547 796491.57 0.17412465
   502000 0.93723008 63.200442 265.11639 796491.57 0.1767773
   503000 0.9355509 63.19922 265.12717 796491.57 0.17776939
   504000 0.93457048 63.196656 265.11791 796491.57 0.17873685
   505000 0.93155843 63.193505 265.12437 796491.57 0.17942446
   506000 0.92802622 63.187035 265.11753 796491.57 0.18298577
   507000 0.92501309 63.182848 265.11756 796491.57 0.18414718
   508000 0.92333305 63.178822 265.10969 796491.57 0.18548086
   509000 0.92118918 63.176077 265.11277 796491.57 0.18637509
   510000 0.91937183 63.174654 265.1236 796491.57 0.18710791

2

yes to both Qs. Note that thermo output can normalize values
by the number of atoms, depending on your settings.

Steve

3

Steve,

pertaining to the 1st question) Since I am performing a shear couette flow experiment, was expecting heat to be generated in the system because of shearing which would be removed from the system by the thermostat in order to maintain fixed temperature. Hence was expecting the sign of the output from "fix langevin tally" to be -ve, but instead it is +ve. I am bit confused and not sure if I am understanding it correctly.

Any help would be appreciated.

Thank you.
BR/Joseph

fix 3 flow langevin 1.10 1.10 1.0 699483 tally yes

4

I suggest you do a simple test where you start with an equilibrated
model at high T. Then use fix langevin tally to cool it to a low T,
and monitor the tally output.

Steve