Bond energy become Zero!!!

From: Ming (ebullience_at_emails.bjut.edu.cn)
Date: Thu Jun 05 2008 - 04:17:01 CDT

Dear all,
I was trying to parameterize a novel ligand and apply it to NAMD. I've done some
tests and I can get the parameters for CH4 from Gaussian calculation. I use the
paratool plugin of VMD 1.8.6 to extract force constants and then integrate them
into a Charmm standard parameter file. After that, I solvated it with a water box
(18 X 18 X 18) and then minimized the system for 10000 steps with PBC. The
calculation ran well. However, when I try to MD it, a strange problem emerged: The
bond energy item equaled ZERO from the very beginning (see below), and the angle
energy became very small comparing with those in the minimization steps. By the
way, the bond energy in the minimization step was not zero.

############### In the MIN step ##########################
ETITLE: TS BOND ANGLE DIHED IMPRP
       ELECT VDW BOUNDARY MISC KINETIC
    TOTAL TEMP TOTAL2 TOTAL3 TEMPAVG
PRESSURE GPRESSURE VOLUME PRESSAVG GPRESSAVG

ENERGY: 500 51.7171 37.1374 0.0000 0.0000
  -1800.1506 250.0243 0.0000 0.0000 0.0000
-1461.2718 0.0000 -1461.2718 -1461.2718 0.0000
-1585.4762 -1307.9887 8000.0000 -1585.4762 -1307.9887

############# IN the MD step ###########################
ETITLE: TS BOND ANGLE DIHED IMPRP
       ELECT VDW BOUNDARY MISC KINETIC
    TOTAL TEMP TOTAL2 TOTAL3 TEMPAVG
PRESSURE GPRESSURE VOLUME PRESSAVG GPRESSAVG

ENERGY: 3000 0.0000 0.4894 0.0000 0.0000
  -1506.9501 145.7842 0.0000 0.0000 252.1150
-1108.5616 298.1669 -1108.3788 -1107.1115 308.1840
-49.1478 -32.1334 4300.4393 -70.3853 -69.2148

After several steps, the simulation crashed!
These are the Error message:

#####################################################
FATAL ERROR: Periodic cell has become too small for original patch grid!
Possible solutions are to restart from a recent checkpoint,
increase margin, or disable useFlexibleCell for liquid simulation.
------------- Processor 0 Exiting: Called CmiAbort ------------
Reason: FATAL ERROR: Periodic cell has become too small for original patch grid!
Possible solutions are to restart from a recent checkpoint,
increase margin, or disable useFlexibleCell for liquid simulation.

Charm++ fatal error:
FATAL ERROR: Periodic cell has become too small for original patch grid!
Possible solutions are to restart from a recent checkpoint,
increase margin, or disable useFlexibleCell for liquid simulation.

Stack Traceback:
  [0] _ZN9HomePatch13doMarginCheckEv+0x1d7 [0x825041f]
  [1] _ZN9HomePatch14positionsReadyEi+0x6e [0x824ec6a]
  [2] _ZN9Sequencer17runComputeObjectsEii+0xcc [0x82bbb8c]
  [3] _ZN9Sequencer9integrateEv+0xb04 [0x82c06c4]
  [4] _ZN9Sequencer9algorithmEv+0x334 [0x82bb490]
  [5] _ZN9Sequencer9threadRunEPS_+0xc [0x82ccb4c]
  [6] namd2 [0x831a849]
  [7] Charm++ Runtime: Converse thread (qt_args+0x72 [0x8394b92])

############################################################

I've tried to increase the PBC and margin, however, these didn't work for me.
Any suggestions? I thought the problem caused by the Zero bond energy but I am not
sure...This reallu confused me!! Anyone can save me???? Many thanks!!!!!!

PS: the topology and parameter calculated by mysely for CH4 are attached,the psf
and pdb I used in the test were also included.


This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:49:32 CST