Hi, I tried to apply the bonding model to my LIGGGHTS.
I downloaded the LIGGGHTS-WITH-BONDS at github (https://github.com/richti83/LIGGGHTS-WITH-BONDS), and also found the bondsfalloncone example at github (https://github.com/Polyun/LIGGGHTS-PUBLIC/tree/master/bondspackage/bonds...).
I just change few lines to work with my version of LIGGGHTS (LIGGGHTS-PUBLIC 3.3.0), and ran in.template.
I attached the changed in.template file here.
I got the results, but they are different with the examples. In the example, the particles are still bonded after they touched the cone, but in mine the bonds seem to be broken just after they touched the cone.
I also enclosing the video of my result and the video from the example folder.
I would appreciate any help.
Thanks,
Soo-Min
| Attachment | Size |
|---|---|
 bondsfalloncone.tar_.gz | 4.99 MB |
mschramm | Wed, 11/11/2020 - 20:35
New Versions
Hello,
This is a very old version of the bond package and I would recommend the following
https://github.com/schrummy14/LIGGGHTS_Flexible_Fibers
(I am very biased in the above. The only thing that would need to be adjusted would be the bond_coeff command as I have added functionality).
Looking at your example (which is missing the stl file), The biggest thing jumping out is your lack of a timestep command... The bond code requires lower timesteps than the hertz contact model.
Looking at your properties, 1.0e-6 should work (my branch has a timecheck command... {shameless plug}). Another thing that jumps out is how high the bond Young's modulus is compared to your break stress. To make sure things are working, I would set the break stress values at 1.0e32 (just for testing then dial them back).
I do not know why the special bond command is present as the bond code does not use special bonds. This may be old code from LAMMPS but I am not sure.
Soo-Min Ham | Thu, 11/12/2020 - 02:00
Paralle bonding
Thank you for your comment!
I tried to use the new version of the bonding model, but I found that those are based on the cylindrical bonding which has bonding radius, length, damping, and modulus.
However, I'd like to adjust the parallel bonding model with bonding radius, normal/shear stiffness, and normal/shear strength.
As following your comment, I changed the timestep and bond stiffness, but it still do not work.
If you get an idea, please let me know. It will be very helpful :-)
Thanks,
Soo-Min
mschramm | Thu, 11/12/2020 - 05:01
Time step
Hello,
What time step did you use?
Looking at the github that you posted (https://github.com/Polyun/LIGGGHTS-PUBLIC),
it appears to be based on the same bond model as described by D.O. Potyondy and P.A. Cundall.
This differs from mine by the factor of the bond length which my version automatically calculates.
For this linear bond, the critical time step is dt < 1/w
where
w = sqrt(k/m)
where k is the normal bond stiffness and m is the mass of a particle.
k = Y*A/Lb = Sn*A