getting particle distribution to insert correctly

Submitted by msandli on Tue, 06/17/2014 - 06:49

I'm trying to simulate a particle flow where I roughly know the distribution of particle size (vendor supplies a table of mesh size and mass %). While the range of particle size is decently large - probably about a max of 4:1 between largest:smallest: - the mass % at those extremes is fairly small.

Anyways, I'm trying to use multiple fix particletemplates for the differing mesh sizes, but I either seem to encounter an error messege saying:

ERROR: Cannot use neighbor bins - box size << cutoff (../neighbor.cpp:1670)

or

ERROR: Please use style 'bin' in the 'neighbor' command together with triangular walls (../fix_neighlist_mesh.cpp:210)

The second one is quite frustrating since this is after applying the fix that the manual suggests for the first error.

Please look at my input script and see if there's something I'm still missing, or if I need to change my bin size to something else I haven't considered, or of there's another problem entirely. Please be aware that this script has lots of comments - it is named "test" after all

AttachmentSize
Plain text icon input script3.12 KB
Plain text icon stl geometry3.34 KB
Plain text icon stl geometry422.35 KB

Hobbygeneral | Tue, 06/17/2014 - 15:15

Hey,

i have tried your input script and could produce the same error. If i haven't missed an important part of your skript you are trying to insert particles of a radius between .0525 and .2975 m in a simulationbox which is much smaller. I don't know if this is possible, because i am a newbie to liggghts.

Anyway, if you increase the size of the domain the script is running without errors on my machine. Is this helping?

msandli | Tue, 06/17/2014 - 16:32

when i increase the size of my z dimension - it's the smallest one and is the one throwing errors - my simulation doesn't insert any particles at all. So, no error, but no data.

Hobbygeneral | Tue, 06/17/2014 - 17:16

your are using the particledistribution below?:

fix pdd1 all particledistribution/discrete 1. 5 40 .001 50 .374 70 .446 100 .164 140 .015

I think you will have to increase the size of the insert region to handle the big particles of radius 0.2 ... 0.2975.