[lammps-users] MEAM problem


I have builded lammps(16Mar07) with meam in serial mode by linking ifort.
But When I run a test job using the input script of ~/examples/meam/in.meam,
I got the message:

Step Temp E_pair E_mol TotEng Press
0 0 -636.93784 0 -636.93784 -278470.97

ERROR: Lost atoms: original 128 current 123

It can give the correct results of E_pair and Tot_eng for the first(initial) step,
but the calculated pressure is different from the value in the log file included in the directories.
The older version(23Feb07) runs well for the same input file.

Hou Huaiyu
Nanjing University of Sci. & Tech., China

I have this problem as well.

-Mike Sellers
Chemical engineering
UB Grad Student

Hou Huaiyu wrote:

Just posted a patch for this.


Hi ,

      In the lammps documentation, it says each bond, pair, angle etc
coefficients can be overridden with such syntax:
       pair * * 1.0 1.0 2.5
       pair 2 3 1.0 1.25 3.0

   This is quite reasonable. However, if " * * " comes second like below:

       pair 2 3 1.0 1.25 3.0
       pair * * 1.0 1.0 2.5

   Does this mean all pair coeff are set to be "1.0 1.0 2.5" ( pair 2 3 are
overridden) ?

   In particular, if I have:

   read_restart my.restart
   pair * * 1.0 1.0 2.5

   After this, can all pair coeff be set to "1.0 1.0 2.5" ?



In all your examples, the last pair_coeff command will
override anything before it. There is nothing magic about
* * for pair_coeff args. It is as if they were expanded
to be N^2 pair coeff commands, where N = atom types.

Hope that makes sense,