fix rigid Vs body style

Hello LAMMPS users,

We have been running simulations with the following body input commands
(see attached the whole input data file). However we have noticed that
the rotation of sub-particles is wrong. This could be because a) we are
not using BODY as we should, b) we should use "fix rigid" instead c) the
BODY package does not take care of the sub-particles rotation. Does
anyone have some sort of experience with similar configurations?

Regards,
Luis

The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

input_lammps (999 Bytes)

Please use fix rigid or fix rigid/small.
Axel

So, should we use “fix blocca big rigid/small molecule” instead of “fix blocca big rigid molecule” ?
Would that fix the subparticles rotation within the bodies?

Thanks

I don’t undestand why you are using body particles and pair body/nparticle.
And trying to use rigid bodies as well. Just do one or the other.
Probably just fix rigid/small is what you want. That command has nothing
to do with body particles, so just use atom_style atomic.

Steve

Hi Steve,

The idea is to create bodies with sub-particles inside. Bodies and sub-particles are “fix rigid”. However the orientation/rotation and forces would only be calculated when two bodies are within the cut-off distance. This would considerably reduce unnecessary calculations of the sub-particles interactions.

Regards,
Luis

Hi Steve,

The idea is to create bodies with sub-particles inside. Bodies and sub-particles are “fix rigid”. However the orientation/rotation and forces would only be calculated when two bodies are within the cut-off distance. This would considerably reduce unnecessary calculations of the sub-particles interactions.

sorry, but that logic makes no sense to me. cutoffs apply to individual particles as well and for large bodies would actually reduce the amount of work better than when the cutoff would be applied to entire bodies instead of individual particles. …and with using neigh_modify exclude, you can skip over intrabody forces computations with fix rigid or fix rigid/small in addition to that.

the BODY package is designed to be used with integrators that haven’t been added to LAMMPS yet. as explained in the description of the package, the rigid body integrator is just a demonstration of the principle. it does offer no advantages over using fix rigid / fix rigid/small, but is lacking several features.

axel.