[lammps-users] lammps example, peptide, error

Here is the error info,

Memory usage per processor = 22.6512 Mbytes
---------------- Step 0 ----- CPU = 0.0000 (sec) ----------------
TotEng = nan KinEng = 1134.9186 Temp = 282.1005
PotEng = nan E_bond = 16.5572 E_angle = 36.3726
E_dihed = 15.5190 E_impro = 1.9426 E_vdwl = 692.8945
E_coul = 26771.8055 E_long = nan Press = nan
ERROR: Out of range atoms - cannot compute PPPM

Energy is nan…

Atoms exceed 2^31 may occur this kind error as shown:

__*Out of range atoms - cannot compute PPPM*__
One or more atoms are attempting to map their charge to a PPPM grid point that is not owned by a processor. This is likely for one of two reasons, both of them bad. First, it may mean that an atom near the boundary of a processor's sub-domain has moved more than 1/2 the [neighbor skin distance](http://lammps.sandia.gov/doc/neighbor.html) without neighbor lists being rebuilt and atoms being migrated to new processors. This also means you may be missing pairwise interactions that need to be computed. The solution is to change the re-neighboring criteria via the [neigh_modify](http://lammps.sandia.gov/doc/neigh_modify) command. The safest settings are "delay 0 every 1 check yes". Second, it may mean that an atom has moved far outside a processor's sub-domain or even the entire simulation box. This indicates bad physics, e.g. due to highly overlapping atoms, too large a timestep, etc.
I try to fix this problem through modify lmptype.h. but nothing happen.

I see README tells

If you get bogus, large energies on timestep 0 when you run this
example in.peptide, you likely have a machine/compiler problem with
the pair_style “long” potentials which use Coulombic tabling by
default.

See the “Additional build tips” sub-section of the manual in
Section_start.html in the “Making LAMMPS” section for details and
suggestions on how to work around this issue.

but I did not find any solution.

you probably linked to the wrong FFTW compile.

you need a double precision(!) compile of 2.1.5.

axel.

If you are running examples/peptide/in.peptide, then I don't
see what the 2^31 atoms comment is about. You also
should (almost) never need to edit lmptype.h.

You are not getting the Coulombinc tabling error, which would
be a bad E_coul value. My guess is something with
the FFTs, as Axel indicated. I don't know how you can
build FFTW wrong, but you can link to the wrong library
I suppose, e.g. the single precision version as Axel
indicated.

Steve