[gmx-users] LINCS warning and segmentation fault for methane in 1-octanol simulation

[vivek sharma]

Hi There,
 I was trying to run a equilibrium run (npt) for system of "Methane in
1-octanol" (1 methane molecule with 200 1-octanol molecules with OPLS FF
parameters). Following is the series of steps I followed:
1. Energy minimization with steep integrator for 5000 steps
2. Energy minimization with l-bfgs integrator for 5000 steps
3. NVT run for 100 ps with sd integrator
4. NPT run for 100 ps with sd integrator (with Pcoupl = Parrinello-Rahman,
tau_p = 0.5, compressibility = 4.5e-05, ref_p = 1.0)

During the NPT run I am ending up with following error (too many LINCS
warning):
---------------------------------------------
Step 25406, time 50.812 (ps)  LINCS WARNING
relative constraint deviation after LINCS:
rms 11.778495, max 363.475445 (between atoms 2158 and 2157)
bonds that rotated more than 30 degrees:
 atom 1 atom 2  angle  previous, current, constraint length
   2152   2150   90.0    0.1090   1.3690      0.1090
   2152   2151   90.0    0.1090  36.5266      0.1090
   2158   2156   90.0    0.1091   0.5637      0.1090
   2158   2157   90.0    0.2349  39.7278      0.1090

Back Off! I just backed up step25406b_n0.pdb to ./#step25406b_n0.pdb.1#

Back Off! I just backed up step25406c_n0.pdb to ./#step25406c_n0.pdb.1#
Wrote pdb files with previous and current coordinates
Segmentation fault
------------------------------------------------------------

*.mdp files for above simulation are adapted from tutorial on free energy
calculation by Justin.

Please suggest me what could be wrong with above simulation.

Regards,
Vivek Sharma
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120209/6fb92494/attachment.html>