[gmx-users] Large fluctuation of protein temperature with nose hoover thermostat

[Stephane Abel]

Hi gromacs experts,

For my simulations of peptide (8 AA) in water (5100, SPC/E) and with 1 
Cl- i use the Nose-Hoover Thermostat to simulate in NVT ensemble. Since 
i am a beginner with GROMACS, i have a question with the tau_t parameters

Parameters i used

I ; Temperature coupling is on
tcoupl = nose-hoover ; nose-hoover thermostat
tc-grps = Protein Non-Protein ; two coupling groups - more accurate
tau_t = 0.4 0.4 ; time constant, in ps
ref_t = 310 310 ; reference temperature, one for each group, in K

When i compute temperature (15, 40 and 41) of the system with g_energy 
for 30 ns, i obtain the following values.

Energy Average RMSD Fluct. Drift Tot-Drift
-------------------------------------------------------------------------------
Temperature 310 2.46886 2.46886 6.12135e-07 0.0171398
T-Protein 310 32.1134 32.1117 -4.10352e-05 -1.14898
T-Non-Protein 310 2.47549 2.47548 8.55869e-07 0.0239643
Heat Capacity Cv: 12.473 J/mol K (factor = 6.34265e-05)

gcq#109: "Ease Myself Into the Body Bag" (P.J. Harvey)

As you can see the temp fluctuation is larger ( > 10 %) than the values 
for system and the solvent It is normal for you ? > If not how to 
correct it

Thank in advance for your help