[gmx-users] Re: Strong egative energy drift (losing energy) in explicit water AMBER protein simulation

[ms]

Looking deeper, it seems most of the energy loss is concentrated in the 
SR Coulomb interactions:

Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
Bond                        2771.28         17    92.5698   -120.382 
(kJ/mol)
Angle                        3796.4         25    106.584   -169.737 
(kJ/mol)
Proper Dih.                 5106.24         10    54.0963   -66.3269 
(kJ/mol)
Improper Dih.               261.421        1.7    21.1174   -11.8454 
(kJ/mol)
LJ-14                       1865.58       0.59    36.7259   -2.92646 
(kJ/mol)
Coulomb-14                    16174         14      118.8   -58.9593 
(kJ/mol)
LJ (SR)                     94061.9        560    1333.14    3958.04 
(kJ/mol)
Coulomb (SR)                -578731       1900    4045.78   -13704.5 
(kJ/mol)
Coul. recip.               -74923.8         23    81.0255   -145.854 
(kJ/mol)
Potential                   -529618       1500    3004.68   -10322.5 
(kJ/mol)
Kinetic En.                 86141.8        610     1295.7   -4294.67 
(kJ/mol)
Total Energy                -443476       2100    4220.26   -14617.1 
(kJ/mol)
Temperature                  292.49        2.1    4.39949   -14.5823  (K)
Pressure                    1.02119     0.0046    133.601 -0.0134421  (bar)
Volume                      343.801       0.32   0.955516    -2.2844  (nm^3)
Density                     1012.34       0.94    2.81349    6.72636 
(kg/m^3)



Also, the drift is almost perfectly linear (apart from the noise).


-- 
Massimo Sandal, Ph.D.
http://devicerandom.org