[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ESPResSo] how to specify normal wall?

From: Axel Arnold
Subject: Re: [ESPResSo] how to specify normal wall?
Date: Thu, 18 Jun 2009 10:29:35 +0200
User-agent: KMail/1.9.4

On Thursday 18 June 2009 00:25, Salvador Herrera Velarde wrote:
> Dear All,
> I am trying to to implement the wall normal constraints in one of my
> scripts,  however I run into some troubles.
> What I did was:
> # Particle potential ( type 0 )
> inter coulomb $lb dh $dk $rc
> set dc1 [expr {0.8* $bl } ]
> constraint wall normal 0 0 -1 dist $dc1 type 1
> set dc2 [expr {0.2* $bl } ]
> constraint wall normal 0 0 1 dist $dc2 type 2

Are you sure this is what you want? The "dist" parameter gives the distance of 
the wall along the normal to 0. The first wall is therefore located at z=-0.8 
bl, the second at z=0.2 bl. I guess you rather want dc1=-0.8 bl.

> inter 0 1 wall normal
> inter 0 2 wall normal

No, the constraints are active the moment you specify them, however, you need 
the inter command to specify how the constraints interaction with the 
particles. We do not have hard constraints (i.e. reflecting walls), but the 
walls can interact using any short ranged potential. One common choice would 
be a purely repulsive Lennard-Jones or WCA potential, which you would specify 
using e.g.

set sig 1; set eps 1
set cut [expr $sig*pow(2,1./6.)]
inter 0 1 lennard-jones $eps $sig $cut [calc_lj_shift $sig $cut] 0
inter 0 2 lennard-jones $eps $sig $cut [calc_lj_shift $sig $cut] 0

I.e., you specify the interaction between type 0 (your particles) and type 1 
(the first wall) and type 2. You could also use type 1 for both walls, then 
you don't need to specify the interaction twice, because I don't think you 
want your walls to have different interactions with the particles.

Many regards,

Dr. Axel Arnold 
Fraunhofer SCAI
Schloss Birlinghoven, 53754 Sankt Augustin, Germany
Tel: +49 2241 14 2575

reply via email to

[Prev in Thread] Current Thread [Next in Thread]