Re: [ESPResSo-users] constraint violated by a particle

[Georg Rempfer]

Hello Soheila,

this error means that at least one particle is located inside the 
constraint. This can have several reasons. The most common one is that 
your time step it too big, which allows a particle that travels at a 
high enough velocity to jump over the singularity of the LJ potential in 
one time step. Another reason could be that there are particle types 
present in the system, which do not interact with the constraint due to 
a missing wall-particle interaction. Such particles could freely travel 
into the constraint and cause this error.

The information you gave is not sufficient for us to determine the 
actual cause. If you can not track down the mistake after this, please 
show us your whole simulation script, or even better, a simplified, 
cleaned up version which reproduces the error.

Greetings,
Georg Rempfer


Am 16.05.2012 11:19, schrieb address@hidden:
> Hello
> everybody,
>
> I am
> trying to simulate a system of
> several polymers by using Espresso package. I have two constraints in my
> system which I have constructed
> them by the following commands (with box_l=100):
>
>
>
> 1:    constraint wall
> normal 0 0 1 dist 5 type 5
> penetrable 0
>
> 2:    constraint wall
> normal 0 0 -1 dist -80 type 6
> penetrable 0
>
>
>
> I know that all of the polymers are in Z>5 and Z<80
> at the beginning of the simulation. When I start the simulation, all polymers
> aggregate on the first (type=5) constraint (Lj-Epsilon of that is 3) after
> about 5000 time steps.
>
> The simulation stops with a background error (below) after
> about 10000-20000 time steps.
>
>
>
> Background
> error :{061: wall constraint 0 violated by particle …}
>
>
>
> I have performed this simulation many times with
> different parameters, but I have not got any good result. I must consider a
> special point (about constraints) in my simulation?
>
> I would be thankful if you guide me.
>
>
>
> Best Wishes,
>
> Soheila