[gmx-users] Pressure average problem

[Mark Abraham]

On 31/03/2010 5:19 AM, nishap.patel at utoronto.ca wrote:
> Hello,
>
> I am simulating a box of water (TIP3P ) in a box of 3nm*3*3nm.My
> simulations are 100ns each. I am using different cut-off. i.e
> vdw-type = switch
> rvdw-switch = 0.8
> rvdw = 1.0, 1.2 and 1.3
>
> and these are my parameters for Pressure coupling":
> Pcoupl = Parrinello-Rahman
> Pcoupltype = isotropic
> tau_p = 2.0
> compressibility = 4.5e-5
> ref_p = 1.0
>
> Now after my simulations, when I use g_energy to extract average
> pressure I get different values for all three cut-offs.
> For the cut-off of 1.0nm, the average value doesn't converge to 1.0atm
> but it is rather ~2.1atm, for 1.2cut off the avg. pressure is ~1.7 and
> for 1.3cut off avg. pressure is ~1.03. I don't understand why the
> pressure is not converging to 1.0atm for the cut-off of 1.0nm.. All the
> parameters are exactly the same, the only difference between these
> simulations are the cut-off values. Could anyone please give me some
> insight on how the pressure coupling works and if this behaviour seems
> normal?

You've not told us critical information, like the rest of your .mdp 
file, or how you equilibrated before running your simulation from which 
you collected data. You could be doing all sorts of weird things and we 
simply couldn't know. Your electrostatic treatment is far more 
significant than your use of rvdw.

Haphazardly changing simulation parameters is not a desirable thing to 
do. Find a suitable model of physics in a published calculation, 
evaluate its applicability to your situation, test its performance 
suitably, and then apply it.

Mark