Re: [ESPResSo-users] Command galileiTransformParticles

[Owen Hickey]

Hey Salvador,

I would suggest using galileiTransformsParticles within the for loop
as well although I suspect this isn't  what is causing your dumbbell
to rotate.  From your description it sounds as though it is rotating
in a vacuum and there is no interaction between your particles and the
DPD fluid (especially if the rate of rotation doesn't change much over
time).  I would try perhaps increasing the density of DPD particles,
the cutoff of the DPD thermostat, or the coupling parameter gamma_dpd.
 Have you tried looking at a vmd movie of your system?  This often can
be quite useful in determining what is happening when something
strange happens.

Hope that helps,
Owen

On Fri, Dec 16, 2011 at 1:47 AM, Salvador H-V <address@hidden> wrote:
> Hi Owen,
>
> Thanks a lot for your reply.
>
> I tried what you suggested, but even if I applied the command just one time
> before the production cycle, the single dumbbell just began to rotate with
> no translation motion!
>
> What I have in my code is the following:
>
> #Termalization
>
> thermostat langevin $temp_therm $gamma_ld
> => Expected behavior but no a good thermostat for dynamical properties
>
> After thermalization,
>
> thermostat off
> galileiTransformParticles
>
> thermostat dpd $temp_ave $gamma_dpd $dpd_cut
>
> for {set i 1} {$i <= $ncic_ave} {incr i} {
> integrate 20
> save positions
> & I don't apply anymore the command: galileiTransformsParticles
> }
>
> Any suggestion or point out of any possible source of error would be greatly
> appreciated.
>
> Thanks a lot,
>
> Salvador
>
> On Thu, Dec 15, 2011 at 5:12 AM, Owen Hickey <address@hidden> wrote:
>>
>> Hello,
>>
>> I am not familiar with system_com_vel although I would try:
>> puts [system_com_vel]
>>
>> and see what it does.  Presumably it returns a list which you could
>> store in a variable instead of printing to screen.  For
>> galileiTransformParticles one simply needs to write that command with
>> no arguments to adjust the center of mass velocity to zero.  I am not
>> surprised that the system seems to drift without this using DPD as
>> numerical errors will eventually give the fluid a net velocity which
>> the dumbbell will follow.  The rotation seems a bit stranger.  The
>> reason that this does not happen with the Langevin thermostat is that
>> it is an implicit fluid with zero velocity (this is the
>> -friction*velocity term).  If you use both thermostats you should fix
>> the drifting problem but the Langevin thermostat will also kill
>> hydrodynamic interactions and thus defeat the purpose of simulating a
>> DPD fluid.
>>
>> Hope that helps,
>>
>> Owen
>>
>> On Thu, Dec 15, 2011 at 2:17 AM, Salvador H-V <address@hidden> wrote:
>> > Dear All,
>> >
>> > I am trying to simulate a system using DPD thermostat.
>> >
>> > In the user guide i found the following sentence: "... before using dpd,
>> > you
>> > have to stop the center of mass motion of your system, which
>> > you can achieveby using the command galileiTransformParticles. This may
>> > be
>> > repeated once in a while for long runs due to round off errors (check
>> > this
>> > with the command system_com_vel) "
>> >
>> > I did no find more information about these two commands.
>> > I would appreciate any help of how to apply both commands inside an
>> > ESPResSo
>> > simulation. What is the syntaxis?
>> >
>> > By the way, using the 2.1.5 version of ESPResSo I observed some
>> > unexpected
>> > behavior for a single dumbbell system using DPD thermostat.  The
>> > dumbbell
>> > seems to translates and rotates only in one direction. When I use the
>> > Langevin thermostat it seems to perform both translation and rotation
>> > random/brownian movement.  I using the same system with both
>> > thermostats.
>> > Any comment will be greatly appreciated.
>> >
>> > Thans a lot,
>> >
>> > Salvador H-V
>> >
>> >
>> >
>> >
>> >
>> > --
>> > =o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o
>> >
>> > Dr. Salvador Herrera Velarde
>> > Division de Ciencias e Ingenierias
>> > Campus Leon
>> > Universidad de Guanajuato
>> >
>> > =o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o
>> >
>> > Este correo ha sido editado para evitar el uso de acentos y guardar
>> > compatibilidad entre
>> > diferentes distribuciones
>> >
>> >
>
>
>
>
> --
> =o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o
>
> Dr. Salvador Herrera Velarde
> Division de Ciencias e Ingenierias
> Campus Leon
> Universidad de Guanajuato
>
> =o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o
>
> Este correo ha sido editado para evitar el uso de acentos y guardar
> compatibilidad entre
> diferentes distribuciones
>
>