Hey
I know that currently LIGGGHTS only allows particle shapes to be either spherical or ellipsoid. I saw that there was discussion about putting in other granular shapes back in Aug 2010, and wondered if this is being worked on at the moment or if more pressing issues have taken precedence? It certainly has started to come up in alot of computational papers of recent - quite intensively investigated in papers by PW Cleary (2006;2010). In this paper he introduces a 'blockiness factor' in his DEM code where N = 0 is the case of a sphere, and as this factor increases, it becomes more and more cuboid. This is pseudo-cuboids are in alot of cases closer to physical shapes of granular media used experimentally. What do you think Chris? I assume that it is quite a large overhaul of some of the integration methods. It would be nice to see.
By the way, thanks again for your help previously (and to others including okeles and rageulmoon). My simulations are working nicely now :)
Jay
ckloss | Wed, 04/27/2011 - 20:49
Hi Jay, the ellipsoid shape
Hi Jay,
the ellipsoid shape is an MD potential (gayberne). For granular, LIGGGHTS supports spherical particles, and multisphere is being worked on. Regarding other shapes such as superquadric: Yes, of course that would be nice.
>>I assume that it is quite a large overhaul of some of the integration methods
There are already integrators that could be used - see ASPHERE package. What would be needed is the pair style, i.e. efficient contact detection, overlap calculation etc... Cleary has (I guess intentionally) not published anything about that, but nevertheless of course it would be doable.
If someone is eager to start working on that, I can point you to a couple of papers that could serve as starting point
Another thing that we would need to think about is how to viz the superquadrics.
Cheers,
Christoph
jwarnett | Thu, 04/28/2011 - 11:03
OK. Well I am probably going
OK. Well I am probably going to be another 9 months - a year on what I am working on now, but after that I would certainly consider working on this if someone on your team isn't already.
If you would be able to point me to the papers it would be at least interesting to read about in the mean time.
By the way are you holding another conference on LIGGGHTS in Linz anytime soon? I missed the last one as I was away
Jay
ckloss | Thu, 04/28/2011 - 13:42
E.g. those papers might be
E.g. those papers might be interesting
http://web.univ-ubs.fr/limatb/EG2M/Disc_Seminaire/ESMC2009/papers/MS-14/...
http://web.univ-ubs.fr/limatb/EG2M/Disc_Seminaire/ESMC2009/papers/MS-14/...
or just google "superquadric contact detection"
>>By the way are you holding another conference on LIGGGHTS in Linz anytime soon?
We will be holding one more short course this year, and are thinking about a CFDEM/LIGGGHTS user meeting in 2012.
Christoph
marc | Mon, 08/22/2011 - 17:44
Multisphere example
Hi Christoph,
I saw the multisphere example video and got really excited about it.
I tried to find something in the doc, but it is not available yet. Would be possible to have access to some simple example with multisphere, where I could look at the set up structure?
Thanks for the help.
Best Regards,
Marc
junchen00 | Fri, 08/26/2011 - 09:25
I have an idea about vizing
I have an idea about vizing the superquadrics using paraview.
first, creat a superquadric source;
then, read your vtk files including particles postions;
the last, use glyph with custom sources. The following image shows what I have done.
sorry, I have no idea on how to attach images.
Chen
reza-ss | Sat, 09/05/2015 - 10:58
Manual
Hi,
Sorry for disturbing you,
I'm looking for a pdf manual for liggghts, but I cant find, because:
1-In web versions I got something like this :
Page not found
The requested page "/media/DEM/docu/fix_property_atom.html" could not be found.
2- I want to use ellipsoid atom_style but I can't find out how to use it, because there isn't any help in examples or in tutorials
(How can I use "particletemplate/sphere" for ellipsoids??)
Best regards,
Reza
aaigner | Thu, 09/24/2015 - 15:20
Hi Reza!
Hi Reza!
Bests
Andreas