[gmx-users] group temperatures and kinetic energies

[Dmitri Dubov]

Dear gmx'ers,

I am somewhat confused relating temperatures and kinetic energies in gromacs output.
I simulate NVE system of N0 rigid molecules in vacuo, 'comm_mode' is set to 'None'. When I check the output 'Kinetic En.' and 'Temperature', the relation between them (see Eq. (3.13) of Manual) corresponds to 6*N0 degrees of freedom, that's OK.
Then the system is divided into two groups with N1 and N0-N1 molecules. While reruning the simulation (with 'energygrps' option), the energies of both groups are written to energy file, and the output includes also 'Temp_group1' and 'Temp_group2'. My wish is to obtain separate kinetic energies of the groups. I try to calculate them using for group degrees of freedom 6*N1 and 6*(N0-N1) values, resp. The problem is that the sum of instantaneous group kinetic energies is not equal to full 'Kinetic En.'! I could suppose that it is due to some relative motion of the groups, but the difference in energies through the simulation may be both positive and negative.

What is the real source of this difference?

Thanks in advance,
-- 
Regards,
 Dmitri                          mailto:ddubov at ngs.ru
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20101208/93dbe156/attachment.html>