espressomd-users
[Top][All Lists]
Advanced

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

Re: [ESPResSo-users] Fwd: COMFIXED does not work


From: Axel Arnold
Subject: Re: [ESPResSo-users] Fwd: COMFIXED does not work
Date: Thu, 23 Aug 2012 18:57:16 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:14.0) Gecko/20120713 Thunderbird/14.0

On 08/17/2012 10:48 AM, Farnoosh Farahpoor wrote:
Dear Axel
I tried to run the  test suite script of COMFIXED.
It works as it is in the package source but when I add charge to the particle and use p3m method then it doesn't.
Is it possible in principal to use COMFIXED with charged system?
With est regards
Farnoush

Hi Farnoush,

sorry that it took a bit longer. COMFIXED requires nonperiodic boundary conditions - it removes the COM force, which does not work in PBC, and therefore not with P3M. In PBC, one has to do a Galilei transform instead, which of course has the disadvantage of interfering with the dynamics.

In general, COMFIXED is supposed to complain when you try to start it with periodic BC, but the error output was broken. It actually should also complain if you use more than one CPU, that is why the testsuite code bails out if there is more than one node. I have fixed the error output in SVN, should be upstream in a day.

For your simulations, if you do not need periodic bc, then you should use debye-hueckel with zero screening to get a direct Coulomb. That is slower, but P3M has necessarily periodic bc. If you need PBC, then you have to resort to periodically galilei-transform the particles of interest and use e.g. DPD so the moment is conserved. There is a routine "galileiTransformParticles" in the scripts subfolder, which you can use. If you want only parts of the particles, you can write a correspondingly modified version. However, I am not sure whether under periodic bc, that is a good idea...

Axel

On Fri, Aug 10, 2012 at 9:33 PM, Axel Arnold <address@hidden> wrote:
Pushing the same message twice won't help, probably no one using comfixed is
around at present.

In any case, the test suite script still seems to work, and it does the same
as you. So you might want to send your script in order to check what is the
mistake.

Axel


On Friday 10 August 2012 21:07:25 Farnoosh Farahpoor wrote:
> ---------- Forwarded message ----------
> From: Farnoosh Farahpoor <address@hidden>
> Date: Thu, Aug 9, 2012 at 1:10 AM
> Subject: COMFIXED does not work
> To: address@hidden
>
>
> Dear Espresso developers
> I am trying to use the "comfixed" interaction.
> COMFIXED feature is compiled in to the espresso package.
>
> code_info
> ESPResSo-3.1.0-64-gaf5c2bb-dirty-git
> { Compilation status { FFTW } { CUDA } { CONSTRAINTS } { LB_GPU } {
> COMFIXED } { PARTIAL_PERIODIC } { LENNARD_JONES } { EXTERNAL_FORCES } {
> ELECTROSTATICS } { LB_BOUNDARIES_GPU } }
>
> I use command "inter 0 0 comfixed 1". Type 0 are the monomers of my
> polymer. It has counterions of type 1.
> But during the simulation my polymer moves around and the COM is not fixed.
> Should I do anything else to use this command properly?
>
> With best regards
> Thank you in advance
> Farnoush
--
JP Dr. Axel Arnold
ICP, Universität Stuttgart
Pfaffenwaldring 27
70569 Stuttgart, Germany
Email: address@hidden
Tel: +49 711 685 67609




-- 
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]