Re: [ESPResSo-users] terminal velocity for LB fluid

[Owen Hickey]

Hey all,

For calculating the constant g I would recommend applying the counter force to the fluid as suggested by Stefan and done in the original sample script. I would also perhaps lower the time step to 0.05 as their is a slight dependence on the time step if my memory serves me correctly. In addition you should be aware of the 1/L dependence on the box length and extrapolate to big box sizes (or use a much larger box size than was used in the original script).

Owen

On Fri, Sep 27, 2013 at 6:11 PM, Vincent Ustach <address@hidden> wrote:

Thanks Ulf,

I have noticed the node dependence but a simple averaging like in the sample script should solve the problem.

--Vincent Ustach
  University of California, Davis

On Fri, Sep 27, 2013 at 8:59 AM, Ulf Schiller <address@hidden> wrote:

On 09/27/2013 05:18 PM, Vincent Ustach wrote:

These are all great answers, thanks everyone. It is definitely more of a
physics question than a programming one.

Ulf,
My objective is to determine the constant g in the offset between the
bare friction and effective friction of the system in reference to the
above paper (Alrichs and Dunweg 1999, and again in Usta et al 2005). The
velocity of a particle under an external force was measured and I can
only assume a terminal velocity was somehow determined.

Owen,
I think simulating flow between two plates could give the desired
effect, as long as the system was large enough in the direction normal
to the plates. I'm going to try it.

Stefan,
This is a very interesting point but I need to look into the
implications of it to see if it satisfies the my problem. My hunch is
that it does.

Hi Vincent,

I think Stefan's solution is exactly what you can do to measure the settling velocity and the mobility/friction. In addition, you may want to fix the particle on the grid to capture the position dependence of the effective friction. Since ESPResSo uses linear interpolation, the effect is quite pronounced.

Flow between plates has a different Stokeslet of the point force, so that is more tricky but possible too.

Best,
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