Re: [ESPResSo-users] Lattice Boltzmann

[Dominic Roehm]

hi,


you should set the boundaries after you introduced the lb fluid (I not
sure if this is a must but definitely saver)
The viscosity of 33.22 is too high, the frictions is ok.
The problem with LB is that is only reproduces the correct HIs in a certain
parameter range and you have to rescale the necessary parameters to be in
that range. This includes the density of the fluid as well. I would
recommend to use a friction between 1-50, viscosity 0.5-5 and density
0.1-5 (for the density range I'm not 100% sure but 0.042 looks too small
for me). You should check your parameter set by investigating the motion
of a single particle in the fluid. This gives you the best way to check
for suitable parameters.

Dominic

ps: be aware that the timestep tau (which seems rather small) and the
agrid (1 is recommended) also affect the parameters. So if you change tau
you might have to adjust your parameter set.


> Hi espresso users,
> At first my simulation didn't include LB, but now I am going to see the
> Hydrodynamic effects, so I turned on the LB like this:#SETTING UP
> HYDRODYNAMICS
> kill_particle_motion
> thermostat off
> lbboundary cylinder center $c $c $c axis 0 0 1 radius $r length [expr
> $box_length/2] direction -1 penetrable thermostat lb 1.0
> lbfluid dens 0.042Â  visc 33.22Â  agrid 1.0Â  tau 0.005
> lbfluid friction 13.036integrate 100
>
> The parameters are water's, and I am almost sure that I calculated them
> correctly,
> I want to ask some questions:1. is it important to introduce lbboundary
> first, then lbfluid, or vice versa, or there is no difference?2.Since I
> got some errors for cylinder violation, should I use reflecting 1 for
> lbboundary cylinder?3. Without LB, the energy of my system is about
> -30000KT, though using  LB increases the system's energy greaty to 
> +3000KT, is it normal? Do you know what I did wrong?4.Is the
> electrohydrodynamics used only when we have external electric field?5.Do
> you think using remove_momentum will solve my problem, how can I use it in
> version 3.2.0?
> I would really appreciate if someone can help me solving the
> problem.Best,Roya
>
>


-- 
Dominic Röhm (Dipl. Phys)
Institute for Computational Physics
Universität Stuttgart
Allmandring 3
70569 Stuttgart, Germany
TEl: +49(0)711 / 685 - 67705