[gmx-users] program crashed

[Dongsheng Zhang]

Dear GMX users,

After I edit vdwradii.dat and run genbox, my program still crashes when
I run position restrained MD with heating the system from 100K to 300K.
When I check the log file. I find the kinetic energy doubles (step 38880
compare to step 38870, by the way, the re_t is the same in these two
steps) when the constraint problem starts to take place. Does anyone
know why KE doubles? 


The following message is from my log file:


           Step           Time         Lambda
          38870       77.74001        0.00000

   Rel. Constraint Deviation:  Max    between atoms     RMS
       Before LINCS         0.029985   1381   1382   0.004488
        After LINCS         0.002836    360    390   0.002298

Current ref_t for group Protein:    269.4
Current ref_t for group SOL:    269.4
   Energies (kJ/mol)
          Angle       G96Angle Ryckaert-Bell.          LJ-14
Coulomb-14
    1.86369e+03    8.21120e+02    3.90504e+03    9.13784e+04    1.63718e
+03
        LJ (SR)   Coulomb (SR) Position Rest.      Potential    Kinetic
En.
    2.63152e+04   -2.08977e+05    1.31793e+03   -8.17386e+04    3.00352e
+04
   Total Energy    Temperature Pressure (bar)
   -5.17034e+04    2.67248e+02    1.12117e+02

           Step           Time         Lambda
          38880       77.76000        0.00000

   Rel. Constraint Deviation:  Max    between atoms     RMS
       Before LINCS         8.028940   1182   1184   0.184534
        After LINCS         6.697157   1182   1184   0.153625


Step 38880, time 77.76 (ps)  LINCS WARNING
relative constraint deviation after LINCS:
max 6.697157 (between atoms 1182 and 1184) rms 0.153625
bonds that rotated more than 30 degrees:
 atom 1 atom 2  angle  previous, current, constraint length
   1182   1184   90.0    0.1090   0.8390      0.1090
   1182   1185   50.4    0.1090   0.1093      0.1090
Constraint error in algorithm Lincs at step 38880
Current ref_t for group Protein:    269.4
Current ref_t for group SOL:    269.4
   Energies (kJ/mol)
          Angle       G96Angle Ryckaert-Bell.          LJ-14
Coulomb-14
    1.92677e+03    8.35797e+02    3.89187e+03    9.12880e+04    1.63733e
+03
        LJ (SR)   Coulomb (SR) Position Rest.      Potential    Kinetic
En.
    3.17595e+04   -2.09746e+05    1.31354e+03   -7.70932e+04    8.35900e
+04
   Total Energy    Temperature Pressure (bar)
    6.49684e+03    7.43768e+02    4.87090e+03


Step 38881, time 77.762 (ps)  LINCS WARNING
relative constraint deviation after LINCS:
max 0.385035 (between atoms 1182 and 1184) rms 0.009398
bonds that rotated more than 30 degrees:
 atom 1 atom 2  angle  previous, current, constraint length
   1182   1183   35.1    0.1078   0.1122      0.1090
   1182   1184   89.9    0.8390   0.1510      0.1090
   1182   1185   57.3    0.1093   0.1129      0.1090
   1242   1245   61.1    0.1090   0.1092      0.1090


Thank you for your help!

Dongsheng

On Tue, 2006-07-04 at 09:52 -0400, Dongsheng Zhang wrote:
> mark,
> 
> That is a good idea. Thank you very much!
> 
> 
> Dongsheng
> 
> 
> On Tue, 2006-07-04 at 15:38 +1000, Mark Abraham wrote:
> > Dongsheng Zhang wrote:
> > > Mark,
> > > 
> > > Thank you for your reply. EM works fine before position-restrained MD.
> > > Do you have any other suggestion besides to reduce the number of water
> > > inside the nanotube? Thanks!
> > 
> > You could edit vdwradii.dat in the gromacs/share/top directory to give 
> > either the water atoms or nanotube atoms more bulk for genbox - see man 
> > genbox for a few details.
> > 
> > Mark
> > _______________________________________________
> > 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.
> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> _______________________________________________
> 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.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php