[gmx-users] energy contributions ...

[Michael Brunsteiner]

dear all,

I try to estimate the influence of a solute molecule
on the structure/energetics of the surrounding water
using the trjorder tool. I simulate the solvent in plenty of
water using SPC/E and PME. After ordering the water in a trajectory
with respect to the distance to the solute I rerun a simulation
with energy groups containing water Nr 1-40, 41-80, etc ...

The resulting energies look fine and if I divide not only by the 
number of molecules in each group (40) but additionally by a factor 
TWO (?) the energy seems to converge to a value close to -39.0 (looks
reasonable for bulk water when considering only the SR electrost. 
contributions)
HOWEVER if I do the SAME thing with groups of a different size, eg 80, 
the energy converges to a different value (-38.0).
This is reproducible and the energy seems to get higher (less negative)
with increasing size of the groups used.
Is this an intrinsic numerical inaccuracy or did I overlook something ?

thanks in advance for any help!
cheers,
mic
 





		
__________________________________
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail