Issue with particles escaping out of the mesh wall

Submitted by erik2019 on Tue, 03/19/2019 - 10:50

Hi all LIGGGHTS users,

I've encountered an issue with particles escaping out of the nozzle mesh wall. My aim for the simulation is to create a steady state powder flow out of the nozzle which i've uploaded the CAD file. I used ovito as the visualization tool and i noticed the no. of particles started decreasing as the powder flow hits the entry of the nozzle. After zooming, i saw a few particles flying out of the nozzle walls. Any idea how to avoid this? I assumed there is something wrong with my 'fix wall/gran' command.

Thanks in advance.

AttachmentSize
PDF icon simlulation_of_nozzle_powder_flow.pdf72.31 KB
Image icon nozzle_.png40.84 KB

mardussa | Thu, 10/17/2019 - 00:40

Hi Erik

I have had similar issues in the past, as many people have also I believe. I think in many cases it can be traced back to a timestep that is a large portion of the Rayliegh time,dt_r = PI*r*sqrt(rho/G)/(0.1631*nu+0.8766). As you can see the Rayleigh time is dependent on the size, density and shear modulus (a function of Young's modulus and poisson's ration) of the particles.
I have seen some users say that having a timestep that is less than 20% the Rayleigh time is a good start, but smaller may be required.

So change the timestep, and try again.

Another option is to set an artificial Young's modulus to increase the Rayleigh time, however this may not be appropriate if it is unrealistic when considering the interactions of the particles you are looking at.

Robert