Re: [ESPResSo-users] Sistematically decreasing of Temperature with DPD thermostat

[Ulf Schiller]

On 01/15/2014 09:23 PM, Axel Arnold wrote:
> Hi!
>
> As far as I understand, RATTLE checks for convergence really that the
> constraints are fulfilled by Euclidean distance, so that would still
> work with constraining to a plane.
>
> However, DPD and RATTLE are probably a problem. At best you loose the
> moment conservation, but RATTLE will eat up most of the kicks into the
> dumpbells main axis to maintain the geometry. So you basically loose the
> fluctuating part, while the drag part is nearly unchanged. So, should
> give you very low temperatures.

Hi,

This is what I thought at first, too, but then I stumbled across the
call to  correct_vel_shake(). Even though it is not called after the
first velocity update, would it not more or less (depending on dt)
remove the velocity along the bond? In that case it is a bit puzzling
that the energy drain seems to be so strong. Anyhow, the combination of
RATTLE and DPD is definitely an issue.

On a related note, one could maybe use predictor velocities to calculate
the DPD forces. That might take the constraint forces into account
implicitly, in the sense that one uses the velocities that were obtained
with the velocity verlet scheme if one knew the constraint forces a
priori. If I understand correctly, but without knowing in detail,
something like that is done in Q-DPD. It's just an idea how to eliminate
the dissipation along the bond.

Cheers,
Ulf

--
Dr. Ulf D. Schiller                        Building 04.16, Room 3006
Institute of Complex Systems (ICS-2)       Phone:   +49 2461 61-6144
Forschungszentrum Jülich, Germany          Fax:     +49 2461 61-3180


------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich
Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzender des Aufsichtsrats: MinDir Dr. Karl Eugen Huthmacher
Geschaeftsfuehrung: Prof. Dr. Achim Bachem (Vorsitzender),
Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt,
Prof. Dr. Sebastian M. Schmidt
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------