[gmx-users] unstability of system, lincs problem

Justin A. Lemkul jalemkul at vt.edu
Thu Apr 29 18:24:02 CEST 2010



Moeed wrote:
> Dear Justin,
> 
> Thank you for your answer. Regarding the message below:
> 
> 1- How is it possible to have a reasonable max force but not a resonable 
> potential?

The potential arises from attractive and repulsive interactions.  Positive 
energy implies net repulsion, even if the forces are reasonably low.

> 2- To remove this high potential from system, What do you suggest?
>      2-1 Do I need to reduce the number of molecules in the box (reduce 
> density?)

No.

>      2-2 Dies the idea of changinf cut off radius (you mentioned 
> somewhere inthe archive) would work?

Be careful about taking quotes out of context.  I have frequently advised users 
against playing with cutoffs (rlist, rcoulomb, rvdw) if they simply don't know 
what they're doing (and there have been some bizarre choices).  Often, the 
values of the cutoffs are determined by the force field parameterization scheme 
and shouldn't be altered (unless there is convincing evidence that they can be 
tweaked without ill effect).  Your 1.0-nm cutoffs don't seem to be a problem to me.

> 3- How Can I view the potential energies arising form various 
> contributions? (electrostatics, vdw,..) to see which type is causing 
> problem?

g_energy

-Justin

> 
> 
> Thank you for your help.
> 
> 
> 
> Moeed wrote:
>  >
>  > Dear gmx experts,
>  >
>  > I am having problem doing MD run for a hydrocarbon system. The system
>  > contains a stack of Hexane molecules using editconf.
>  >
>  > The distance between molecuels in the box is more than 30 A. I am
>  > wondering why I get large forces (system is blowing up) with this
>  > distance!. (LINCS warning) with dt=0.002
>  > Program mdrun, VERSION 4.0.7
>  > Source code file: constr.c, line: 136
>  >
>  > Fatal error:
>  > Too many LINCS warnings (1053)
>  > If you know what you are doing you can adjust the lincs warning
>  > threshold in your mdp file
>  > or set the environment variable GMX_MAXCONSTRWARN to -1,
>  > but normally it is better to fix the problem
>  >
>  >
>  > It seems Fmax, Epot values are reasonable. . In the list archive I read
> 
> I would say they are not.
> 
> <snip>
> 
>  > Steepest Descents converged to Fmax < 1000 in 30 steps
>  > Potential Energy  =  4.76092783832156e+05
>  > Maximum force     =  9.86600079729483e+02 on atom 3115
>  > Norm of force     =  5.33725886931530e+02
> 
> You have a reasonable force, but your potential energy is large and 
> positive,
> indicative strong repulsive forces in your system.
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================



More information about the gromacs.org_gmx-users mailing list