[gmx-users] Re: Non-equilibrium MD: simulating a moving plate

[Vitaly V. Chaban]

 Hi Zhongqiao,

I think it would work notwithstanding it's quite tricky to implement...

My current task is to test a sieving membrane. So some external action is
required to force the mixture go towards the place where separation occurs.
In papers, a rigid wall moving with constant velocity is sometimes used for
this. If one just accelerates the mixture particles with acc-grps, it seems
not realistically as for me, so I'm in guesses what's the best way to manage
such tasks with gromacs.

Thank you for the useful remarks,
Vitaly


>I am not sure if it works. You can try to put your system into two
>frozen layers which will exert a resisting force on the wall. You still
>need to put an acceleration on the wall. But after a steady state is
>reached, you will see the wall will move approximately at a constant
>velocity.

>Hi and thank you! This cleared the situation. I didn't care about
comm-mode but only set comm-grps...

>And let me one more question. Is there any possibility to define a
constant velocity of my rigid wall, not an acceleration?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090324/d78cb1b7/attachment.html>