[gmx-users] Constraint Error

Pawan pwnbabel at chem.iitb.ac.in
Fri Nov 5 08:33:16 CET 2004


Hi Jair,
I do run the energy minimization before position restrainted. By changing
the tolerance and nsteps parameter does not helps me. Because in my case
enregy is converging  due to machine precesion.. Is there any way from
which i can set the option that the energy is converged only due to
spcecified option in md file and not due to the machine precesion ?.

Regards
Pawan

> I found an answer for this problem posted in 2003...
>
>>> Have you run energy minimization? If not, do. If so, try it again,
>>> perhaps
>>> with a lower tolerance and/or larger nsteps. Also, it could help to
>>> first
>>> run md with position restraints, but *without* pressure coupling. If
>>> that
>>> works, continue the run *with* pcoupl.
>
> I also have a doubt about this issue. How can I rerun a position restrined
> with pcoupl after a previous run with no pcoupl.
> Jair
> IQ-UFRJ
>
>
> On Wednesday 27 October 2004 08:21, Pawan Babel wrote:
>> Hello gmx-users,
>>
>> My simulation was truncated due to following error.
>>
>> -----------------------
>> Step 361846, time 723.692 (ps)  LINCS WARNING
>> relative constraint deviation after LINCS:
>> max 1.104418 (between atoms 230 and 231) rms 0.098450
>> bonds that rotated more than 30 degrees:
>>  atom 1 atom 2  angle  previous, current, constraint length
>>     218    219   90.0    0.6208   0.2209      0.1090
>>     220    222   90.2    0.1084   0.1254      0.1090
>>     230    231   90.0    0.2872   0.2125      0.1010
>> Constraint error in algorithm Lincs at step 361846
>>
>> Step 361847, time 723.694 (ps)  LINCS WARNING
>> relative constraint deviation after LINCS:
>> max 372413038592.000000 (between atoms 218 and 219) rms
>> 24367785984.000000
>> bonds that rotated more than 30 degrees:
>>
>> Constraint error in algorithm Lincs at step 361847
>> Large VCM(group rest):      4.59862,     32.04837,      7.38148,
>> ekin-cm:
>> 9.02012e+07
>>
>> ------------------------
>>
>> Do anyone have idea how to solve this problem.
>> thanks in advance
>> Pawan
>>
>
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
>




More information about the gromacs.org_gmx-users mailing list