dear all
I want to simulate separation of heavy and light particles in Knelson concentrator using DEM by LIGGGHTS 3.0.1
this concentrator works with mixture of particles and water but here I ignore the water. heavy and light mineral particles are separated based on differences in gravity and centrifugal forces while the concentrator rotates very fast.
there are some problems I encounter during my simulations:
1) I can not insert particles with specific features like size, density, percent ... uniformly inside the concentrator (look at the input attached file) ?
2) in Knelson concentrator heavy particles separates from light particles and must sit in rings (look at the geometry and shape of concentrator in ZIP file in attachment) and light particles must draw out of separator, but for my case no real separation occurs !? actually both particles move through the separator.
I do not where is the problem? should i change the commands for insertion?
any help regard the above problems is highly appreciated.
Mohammad
| Attachment | Size |
|---|---|
 input.txt | 2.83 KB |
 screenshot.jpg | 45.63 KB |
| no_centerifugal_force_on_particle.jpg | 66.96 KB |
ckloss | Thu, 10/30/2014 - 15:02
Hi m.fatahi ,
Hi m.fatahi ,
>>1) I can not insert particles with specific features like size, density, percent ... uniformly inside the concentrator (look at the input attached file) ?
I see no reason why this should not be possible
Can you please describe the problem in detail? e.g. with a picture?
>>I do not where is the problem?
Try to decribe your problem as detailed as possible - I do not have the time to debug your simulation starting from first principles
Christoph
m.fatahi | Sat, 11/01/2014 - 08:30
HI Christoph
Hi Christoph
thank you for your response. I put a screenshot of particle insertion. you can see that particles insert randomly and not in a group.
in the input file, I defined a region box for insertion but this box is not seen in simulation. I changed parameters like time step, neighbor and particle rate severally to adjust particle insertion in a group but no difference happen just the speed of randomly insertion.
about my second problem, the STL file is the rotating space in Knelson concentrator. I want to show the track of heavy and light particles motion. instead of being in the rings, heavy particles move through STL walls. what is the problem? is there any command that I miss to consider?
do you have any idea about the parameters i should set?
Regards
Mohammad
aaigner | Tue, 11/04/2014 - 14:43
Some remarks
Hello Mohammad,
I took a short look at your input script and found some mistakes ...
neighbor 1.5e-6 bin) should be in the range of your particle radius. Please increase it.Best regards
Andreas
m.fatahi | Mon, 11/10/2014 - 16:21
no centrifugal force on particles
Hello Andreas
thanks for your comments. I made some corrections in skin factor and time step. I used fix insert/rate/region command instead of fix insert/stream and now it is OK for particle insertion.
but when particles enters the rotating bowl, there should be different forces acting on them . the main is centrifugal force on particles. but here I see no force except gravity. all particles after insertion go down of the bowl and pass it. you can have a look at the picture. in real separation heavy particles must remain in the rings and light ones get out of bowl. where is the problem?
should I use a fix addforce command ?
do you have any idea?
thanks for your great help.
Mohammad
richti83 | Mon, 11/10/2014 - 17:43
rolling_friction ?
I think with a rolling friction of 0.01 you can not expect any influence from the mesh-motion to the particles ... and a PERIOD of 0.06 is very very fast (T=1/f), did you mean omega instead ?
m.fatahi | Tue, 11/11/2014 - 06:59
rolling_friction ?
hi richti
I am not sure about rolling friction value. because I tested with 0.3 but no difference. my particles are quartz (light) and gold (heavy). the mesh material is polyurethane. do you have any idea about rolling friction value?
about the PERIOD actually, mesh must rotate with a speed up to 1500 rpm. I assume 1000 rpm for this case that is equal to a period of 0.06 and I used fix move/mesh command to rotate the bowl.