We are using LIGGGHTS to generate particle flow streams in belt conveyor/transfer chute analyses. A few of our people have noticed an odd behavior with particles becoming "stuck" to the drive pulley, and as of yet, we've been unable to find what we've done incorrectly. Below you will find links to two videos, generated by our own post-processor. Forgive me for the partial results as a bug with STL files prevents the system components (belts, chutes, pulleys) from being displayed, but the particle flows are displayed correctly.
https://www.dropbox.com/s/8p7oa1p5fen0nco/WithPulleyTest.wmv
https://www.dropbox.com/s/vz2qwmnnllekk2g/IdealCascadeWPulleyNoR.avi
In the first video, a particle becomes stuck at the point where the belt comes into contact with the cylindrical component we've set up to behave like a pulley. This particle appears within a circle. The video shows the simulation again from the other side of the belt, where users can more clearly see a line of particles building up next to the initially "stuck" particle. And the last part of the video is is the system shown at approximately 2/3 speed. The line of particles breaks up at one point, among a burst of particle activity sending particles in all directions. We don't know if this burst of activity was due to the line-up or not.
The second video is the result of an experiment we performed. Our goal was to determine if pulley rotation was possibly oriented the wrong direction and how this might affect the results. So the pulley's rotation was set to 0. In this video, shown only from one direction, a build-up of particles seemingly adhering to many areas of the pulley can be seen. Admittedly, without the boundaries present, it may be difficult to visualize.
I also include links to the folders containing the input files.
https://www.dropbox.com/s/c0sioiw6kzjqa8a/WithRotation.zip
https://www.dropbox.com/s/ijfyaz0bz86mrt1/NoRotationExample.zip
This is actually a quite simple example, the files including pulley rotation are included below.
https://www.dropbox.com/s/ir215zcuzavt19q/ExampleFiles.zip
Thank you,
Clint
ckloss | Fri, 05/03/2013 - 01:03
Just a few comments after a
Just a few comments after a quick glance at the case setups:
With Rotation: The surface velocity of the pulley is 140 m/s, so far from any realistic value with this time-step size
Without rotation: The high friction factor clamps particles between pulley and deflection chute
So no bug as far as I can see, everything working as expected. Therefore I moved the thread to the general user section
Christoph