[gmx-users] Dihedral angle autocorrelation function

[atanu das]

Hi All,
I am trying to calculate the dihedral angle autocorrelation function by using the following command ----
g_chi -s md.tpr -f md.xtc -corr -phi -psi

I get multiple files with autocorrelation functions for individual phi and psi dihedral angles which are really easy to identify as they get saved as corrphiARG2.xvg for phi angle of ARG-2 and so on.
Now looking at the individual autocorrelation functions I see that most of them don't get to zero and reach a plateau ~0.7 or 0.8 or 0.5. My system is a 11-residue peptide and I simulated the peptide for 1 micro-second in explicit water at 300K. I could have believed the data if the system was simulated for very short time (~1-2 ns). But, I can't convince myself that the autocorrelation function does not reach zero even after simulating for microsecond timescale (this means that the dihedral angle relaxation time is infinity!).
Could anybody tell me what am I doing wrong?
ThanksAtanu