NVT system implosion!!!

From: Marco Preto <mcpreto.fc.up.pt>
Date: Thu, 31 Oct 2002 00:29:24 +0000

Dear amber users:

I presently doing some simulations of small peptides (8 aa) in a water
box (around 3850 water molecules, PME implementation for long range NB
interactions), using sander (from AMBER 6).
Firtly I performed a minimization of the solvent (fixing the positions
of the the solute and Cl- counter ion). The geometry output from the
minimization is OK.
Secondly, I was planning to perform a NVT equilibration on the same
system (temp gradient from 0 to 300K and aplying harmonic constrains to
the solute ( peptide and counter ion). The molecular dynamic run to the
end without any error message, but when I checked the last geometry,
EVERY atom was possitioned in the Origin (0,0,0 coordinates)!!!???!!! I
tryed the same simulation without the constrains (I could have donne
some mistake in the group definition!) but got the same result! This
implosion ocours very quickly in the first 100 some fs! Have anyone
encoutered any problem like this?
Here are the inputfile for the minimization and the NVT dynamic run:
imin=1, maxcyc=2000, cut=12, igb=0,
ntb=1, ntpr=500, ncyc=200,
ibelly=1, ntc=2, ntf=2,
Group input for moving atoms
RES 10 3846

NVT Dyn:
       imin=0, irest=0, ntx=1, tempi=0,
       nstlim=50000, dt=0.001,
       ntt=1, temp0=300.0, tautp=1,
       ntpr=200, ntwe=200, ntwx=200,
       ntb=1, ntc=2, ntf=2, ntr=1,
Group input for restrained atoms
RES 1 9

Thanks for all the help


 Marco A. C. Preto               |"Quantum temporis lucratur qui non
 PH.D. student                   |aspicit quod vicinus dicit, agit
 Theoretical Chemistry group     |vel cogitat, sed tantum quod ipse
 Chemistry Deparmant - FCUP      |agit ad id iustum et sanctum 
 Porto, Portugal                 |reddendum"
                                 |   from my homonimous
email: mcpreto.fc.up.pt          |  
                                 |                 Marcus Aurelius
Received on Wed May 28 2003 - 14:53:02 PDT
Custom Search