[gmx-users] Non Equilibrium MD

chris.neale at utoronto.ca chris.neale at utoronto.ca
Sat Nov 6 06:01:57 CET 2010


> I have two questions regarding non equilibrium MD simulations in GROMACS.
> 1) If we use a thermostat (with T=300 K), will it subtract the non zero
> streaming velocity to maintain the temperature or will it reset the total
> velocity.

Different thermostats work differently. Read the equations to see how  
they work. I'm pretty sure that they will all work with overall  
velocities, but my guess is that you'd have to have some seriously  
unrealistic streaming velocity for it to make much of a difference.

> 2) How do we remove the COM motion only due to thermal noise and keep the
> motion due to external force as it is.

Not possible, as far as I know, in the way that you mean it. But  
remember the equal and opposite thing so as long as your flow is  
moving past your object, does it really matter which one is "moving"?  
(anecdote: "does New York stop at this train?")


More information about the gromacs.org_gmx-users mailing list