[Top][All Lists]

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

Re: [ESPResSo-users] Polymer between Lennard-Jones walls

From: Axel Arnold
Subject: Re: [ESPResSo-users] Polymer between Lennard-Jones walls
Date: Thu, 15 Nov 2012 16:53:05 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:16.0) Gecko/20121026 Thunderbird/16.0.2

On 11/15/2012 12:20 AM, Heng ZHANG wrote:
Hello, Axel,

Thank you for the reply. As you said, I added the second constraint, so it is like:
#constraints setup
constraint wall normal 0 0 -1 dist [expr -$box_length] type 2
constraint wall normal 0 0 1 dist 0 type 2

Also I used the minimum distance as the criteria to tell whether the warm up session is enough. However it reports with:
background_errors 0 {086 particle coordinate out of range, pos = 1.767e+12, image box = -2147483487} {086 particle coordinate out of range, pos = -5.26831e+15, image box = -2147483544} {061 wall constraint 1 violated by particle 1586} {061 wall constraint 1 violated by particle 3467} 
I assume the criteria is not strong enough, so I increased it but still there are violations.

So, you still have somewhere particles on bad distances. During warmup, write out the energy at relative short intervals (like 10 steps), to monitor which interaction is overshooting. Are you using any force capping? Then be careful not to cap the interactions with the walls.

About the simulation without PBC, I added the walls in x and y directions, but in blockfile the x and y components are still out of the range [0, $box_length]. To emphasize, I didn't use LJ walls in xy directions but pure constraints with reflections, e.g.
constraint wall normal 0 1 0 dist 0 type 2 reflecting 1

Do you use periodic boundary conditions or not? If you do, your walls should be a bit off the boundaries for reflecting bc, since a wall at 0 or box_length will never be violated thanks to folding.

It might be easier if you post a minimal working example script for your problem, then I can directly check rather than asking you to do something.


JP Dr. Axel Arnold
ICP, Universität Stuttgart
Pfaffenwaldring 27
70569 Stuttgart, Germany
Email: address@hidden
Tel: +49 711 685 67609

reply via email to

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