Re: [ESPResSo-users] Question about the setting of agrid in L-B fluidics

[Peter Košovan]

Hey Joe,

sorry for responding with such a delay - I somehow missed this conversation.

My original comment was for a polymer in the Langevin thermostat, i.e. without hydrodynamics. There the rule of thumb is max 200-300 beads. I am not sure for LB fluid, here I only played with short polymers, and it also depends on other paramters of your simulation. But there is a simple way to estimate it: take a series of short polymers (e.g. N=20, 50, 70), measure the autocorrelation time of some slow variable, such as Re, plot it as a function of N in double-log scale and extrapolate to estimate autocorrelation times for higher N.

Note, however, that you need more than bring your system to equilibrium - that's just the initial stage. After that, you need to collect a suficient number of statistically independent samples of system configurations, given by the ratio of runtime (discarding equilibration) and the autocorrelation time.

As Axel pointed out in response to my earlier comment, if you are interested in short-time dynamics, you may try and use MC to generate a number of equilibrium configurations and then use LB fluid to follow dynamical evolution on a shorter time scale.

With best regards,

peter

2015-05-12 15:53 GMT+02:00 Joe Ada <address@hidden>:

Thanks Peter. I believe with LB fluid, a 16um long polymer with 50nm persistance length is even hopeless.

I thought if I started with generating them in a SAW way, I may reach the equilibrium status reasonably fast with the help of gpu p3m. 

So could I know the practical max number of beads people have done for time up to 100ms using gpu system with LB fluid? 

Joe

On Mon, May 11, 2015 at 11:43 AM, Georg Rempfer <address@hidden> wrote:

We have had a similar question recently:

    http://lists.nongnu.org/archive/html/espressomd-users/2015-04/msg00045.html

"simulating 4k long polymer with Langevin dynamics is hopeless. The dynamics of such a long chain is far too slow to obtain meaningful statistics in a reasonable time (human lifetime). The characteristic time scale increases roughly as N^{5/3}. If you need such long chains, you should use MC."

    - Peter Košovan 

On Mon, May 11, 2015 at 5:22 PM, Joe Ada <address@hidden> wrote:

Hi, 

I try to run a model with ions and polyelectrolyte in solution in external electric field. The polymer is a 20um long chain with 1600 beads and springs, and the total number of the coions and counterions are 3*1600. What I care about is the hydrodynamic interaction between different beads, beads-ions and ions from the LB part. 

What is the practical setting of this grid number? 

My understanding of the agrid is that it determines the spacial resolution of the fluid. In the user guide, it says "It should also be noted that the LB nodes are located at 0.5, 1.5, 2.5, etc (in terms of agrid).". May I assume the resolution is 0.5 agrid? Does that mean that I have to make this 0.5 *agrid smaller than the typical distance ion travels during time tau*MD time step? 

Thanks,

Joe

--

Dr. Peter Košovan

Department of Physical and Macromolecular Chemistry

Faculty of Science, Charles University in Prague, Czech Republic

Katedra fyzikální a makromolekulární chemie

Přírodovědecká fakulta Univerzity Karlovy v Praze

www.natur.cuni.cz/chemistry/fyzchem/
Tel. +420221951290

Fax +420224919752

---

Pokud je tento e-mail součástí obchodního jednání, Přírodovědecká fakulta Univerzity Karlovy v Praze:

a) si vyhrazuje právo jednání kdykoliv ukončit a to i bez uvedení důvodu,

b) stanovuje, že smlouva musí mít písemnou formu,

c) vylučuje přijetí nabídky s dodatkem či odchylkou,

d) stanovuje, že smlouva je uzavřena teprve výslovným dosažením shody na všech náležitostech smlouvy.