Re: Constraint violated by particle

[Kai Szuttor]

Hi Ahmad,

as I have mentioned changing the direction of the normal vectors fixes the 
issue.
What you maybe did not do is also changing the sign of the distance. The idea 
is that
the product of normal vector *and* distance gives a point on the wall surface. 
Therefor,
to have a wall at z = 50 parallel to the x-y plane you would have to use 
normal=[0,0,-1] 
and dist=-50.

Best,

Kai

On Wed, Oct 28, 2020 at 09:32:20AM +0000, Ahmad Reza Motezakker wrote:
>    ​Dear Kai,
> 
>    Attached please find the minimal version. Thank you again.
> 
>    Best Regards,
> 
>    Ahmad Reza
> 
>      ----------------------------------------------------------------------
> 
>    From: Kai Szuttor <kai@icp.uni-stuttgart.de>
>    Sent: Wednesday, October 28, 2020 10:17 AM
>    To: Ahmad Reza Motezakker
>    Cc: espressomd-users@nongnu.org List
>    Subject: Re: Constraint violated by particle
>     
>    please attach a minimal not working example reproducing your issue
> 
>      Am 28.10.2020 um 09:50 schrieb Ahmad Reza Motezakker <armot@kth.se>:
> 
>      
> 
>      ​​Dear Kai,
> 
>      Before sending my email, I tried to change the directions of the normal
>      vectors but it gave me the same error. To clarify it, Attached please
>      find the schematic of the way that i have defined the constraints and LB
>      boundaries and their normal vectors.
> 
>      Thank you very much.
> 
>      Best Regards,
> 
>      Ahmad Reza
> 
>      ----------------------------------------------------------------------
> 
>      From: Kai Szuttor <kai@icp.uni-stuttgart.de>
>      Sent: Wednesday, October 28, 2020 9:28 AM
>      To: Ahmad Reza Motezakker
>      Cc: espressomd-users@nongnu.org List
>      Subject: Re: Constraint violated by particle
>       
>      Hi Ahmad,
>      i guess your normal vectors are pointing in the wrong direction?
>      Best,
>      Kai
> 
>        Am 28.10.2020 um 09:23 schrieb Ahmad Reza Motezakker <armot@kth.se>:
> 
>        
> 
>        Dear All, 
> 
>        I have a system of a polymer chain, LB fluid and wall constraints (a
>        small chain of polymer, Two wall for fluid boundary and at the same
>        place two wall constraints for setting up the wall-particle
>        interaction). I make the system in the order of creating the polymer
>        chain, fluid and its LB boundaries (Wall), and constraints. I do not
>        face any error during polymer warming up, equilibration, and
>        fluid-polymer warming up. I define the polymer far from the
>        constraints and add lennard-jones repulsion between the polymer and
>        constraints to avoid any constraint violation. I define the LB
>        boundaries and constraints as follows:
> 
>        -----------------------------------------------------
> 
>        logging.info("Setup LB boundaries.")
>        top_wall =
>        
> espressomd.lbboundaries.LBBoundary(shape=espressomd.shapes.Wall(normal=[0,
>        0, -1], dist=0))
>        bottom_wall =
>        
> espressomd.lbboundaries.LBBoundary(shape=espressomd.shapes.Wall(normal=[0,
>        0, 1], dist=50))
>        system.lbboundaries.add(top_wall)
>        system.lbboundaries.add(bottom_wall)
>        #for constraints only
>        TopWall = Wall(dist=0, normal = (0,0,-1))
>        BottomWall = Wall(dist = 50, normal=(0,0,1))
>        system.constraints.add(shape = TopWall, particle_type = 10,
>        penetrable=False)
>        system.constraints.add(shape = BottomWall, particle_type = 10,
>        penetrable=False)
>        -----------------------------------------------------
>        But when I run the integration, in the very very first step (while the
>        polymer is the middle of the box and far from the walls) I face the
>        error : Constraint violated by particle.
>        Could you please give me some hints on that? I really appreciate your
>        help.
>        ​Best Regards,
>        Ahmad Reza
> 
>      <boundaries.pdf>

signature.asc
Description: PGP signature