LIGGGHTS® - User Forum

I am trying to perform an analysis with particles moving along a hot conveyor. My script works fine if I leave out the temperature in fix mesh/surface. But when I add in the temperature, the analysis crashes after 25000 timesteps. Could someone help me understand what I need to change in my input script?!?!

Here is the script;

#Conveyor (600C) and Chute
atom_style granular
atom_modify map array
boundary f f f
newton off

communicate single vel yes

units si

region reg block 0.0 3.3 -0.3 0.8 -0.5 3.0 units box
create_box 1 reg

neighbor 0.01 bin
neigh_modify delay 0

Hi all:
Can some one tell me what's the mesh wall method used in LIGGGHTS?
I saw a topic contains about two thesis before. But I can't find it now.
Appreciate any help in advance!

Hi Christoph,

May I know when LIGGGHTS will support GPU?

Cheers,
Chuncheng

Hi,
I am having trouble finding instructions on how to install LIGGGHTS on Ubuntu 10.4.

I seem to have the source code, but do not know how to compile it.

Any help would be appreciated.

Kind Regards, Robert

Hi,

I want to build particles of complex shapes through using the multisphere approach with LIGGGHTS. I use:

atom_style hybrid granular molecular
(...)
neigh_modify delay 0 exclude molecule all
(...)
fix freezing all rigid molecule.

So far I did simulations where the beads composing a particle of complex shape have the same size. Everything works nice! Now the beads composing the complex particle can have different sizes and here I face a problem.

I want to dump a velocity profile, averaged over pre-defined layers. To my surprise, there's a LIGGGHTS fix that does exactly this (ave/spatial).

I'm a little confused regarding the options: Nevery, Nrepeat & Nfreq. I've been experimenting with it but I'm still not 100% confident of what I'm doing.

What I think each option does:

Nevery: fetches new values every N steps
Nrepeat: not really sure.
Nfreq: compute average every N steps.

Hi,
I install colloid package by typing
make yes-colloid
make fedora_fpic

and I get no errors, ending with :

text data bss dec hex filename
2472063 93744 24 2565831 2726c7 ../lmp_fedora_fpic
make[1]: Leaving directory `/home/vajiheh/LIGGGHTS/liggghts_GIT/src/Obj_fedora_fpic'
vajiheh@vajiheh-desktop:~/LIGGGHTS/liggghts_GIT/src$

I'm using LIGGGHTS 2.0.3, and in my input script this line is giving me trouble:

fix xwall active wall/gran/hertz/history primitive type 2 xplane ${wall_xlo} ${wall_xhi}

where the wall_xlo and wall_xhi variables are defined earlier. This gives the error:

ERROR: Fix wall/gran/hertz/history (id xwall): invalid keyword or keyword(s) in wrong order (fix_wall_gran.cpp:224)

According to the documentation, this syntax should be correct.
A little bit of debugging brought me to this section of code beginning at line 56 of primitive_wall_definitions.h:

Dear friends,

I have installed "LIGGGHTS-PUBLIC" on my desktop computer and then run my code in the LIGGGHTS without any error with this command: liggghts < in.liggghts_1

But when I tried to run it in the parallel processing mode ( mpirun -np 4 liggghts < in.liggghts_1), it was not possiable.

I received the following message:

mpirun was unable to launch the specified application as it could not find an executable:
Executable: liggghts
Node: ubuntu
while attempting to start process rank 0.
--------------------------------------------------------------------------

Hi,
I have a problem one moth ago, and I tried to fix the problem, but is very difficult to understand.

I am trying to simulate the transport of grains in a single screw extrusion, for that i build two wall in STL format. One wall is rotating and another one does not move. I am using gran/herzt/history interaction and wall/gran/hertz/history for the wall and particles.

Firstly, I use particles of d=3mm, normally the message "WARNING: Dangerous build in triangle neighbor list." appears but the run does not stop, and the post-processing corresponds very well to the results expected.