Hello everybody,
I performed a research on the mailing list and actually didn’t find a similar topic.
Problem: I have N spheres on a 2D layer, I would like to add springs between them. Let us say: for each sphere [of diameter sigma] I draw a circle of diameter D [D > sigma] whose centre coincides with the centre of the sphere itself. All the spheres whose centres lie inside the circle are pair-connected to each other through a spring [of rest-length equal to the distance at the moment of linking, constant equal for each one].
The commands “fix/spring” and “atom style bond” seem to me not to be adequate for what I am willing to do. More specifically, I understood that the first rather works with groups of particles; the latter is not compatible with the style (“sphere”) I already defined.
As a LAMMPS beginner, I might be wrong: please correct me, I am looking forward to your advice.
I attach my input script.
Thank you in advance,
Best regards
Mario
in-pour-N-10-T-1-p-1.0 (1.49 KB)
Hello everybody,
I performed a research on the mailing list and actually didn't find a
similar topic.
perhaps spending more time reading the documentation might help as well.
Problem: I have N spheres on a 2D layer, I would like to add springs between
them. Let us say: for each sphere [of diameter sigma] I draw a circle of
diameter D [D > sigma] whose centre coincides with the centre of the sphere
itself. All the spheres whose centres lie inside the circle are
pair-connected to each other through a spring [of rest-length equal to the
distance at the moment of linking, constant equal for each one].
this sounds very much like adding (harmonic) "bonds".
The commands "fix/spring" and "atom style bond" seem to me not to be
adequate for what I am willing to do. More specifically, I understood that
the first rather works with groups of particles; the latter is not
compatible with the style ("sphere") I already defined.
atom style bond *can* be made compatible with style sphere through
using a hybrid atom style.
you'll have to pay attention to the special_bonds settings, if you
want to just add the harmonic interactions, you'll have to use an 1.0
1.0 1.0 setting to avoid any exclusions.
axel.
Try and simplify these explanations.
This sounds like normal component of the gran/hooke/history to me. That is if you mean that particles never give a tension force - only compression by “All the spheres whose centres lie inside the circle are pair-connected to each other through a spring [of rest-length equal to the distance at the moment of linking, constant equal for each one]”
Otherwise it just sounds like you’re looking for a simply connected network of springs (or higher connectivity if the distance that determines whether particles are neighbors is arbitrary). In that case, I don’t think there’s anything ‘out the box.’
> Hello everybody,
>
> I performed a research on the mailing list and actually didn't find a
> similar topic.
perhaps spending more time reading the documentation might help as well.
I definitely agree.
>
> Problem: I have N spheres on a 2D layer, I would like to add springs
between
> them. Let us say: for each sphere [of diameter sigma] I draw a circle of
> diameter D [D > sigma] whose centre coincides with the centre of the
sphere
> itself. All the spheres whose centres lie inside the circle are
> pair-connected to each other through a spring [of rest-length equal to
the
> distance at the moment of linking, constant equal for each one].
this sounds very much like adding (harmonic) "bonds".
Exactly.
> The commands "fix/spring" and "atom style bond" seem to me not to be
> adequate for what I am willing to do. More specifically, I understood
that
> the first rather works with groups of particles; the latter is not
> compatible with the style ("sphere") I already defined.
atom style bond *can* be made compatible with style sphere through
using a hybrid atom style.
you'll have to pay attention to the special_bonds settings, if you
want to just add the harmonic interactions, you'll have to use an 1.0
1.0 1.0 setting to avoid any exclusions.
axel.
Many thanks, Axel, for your precious hint.
Best regards
Mario
It is actually the second case you mentioned.
Thank you too, Eric!
Regards,
M
Axel’s comment was a good one. And more appropriate since I never do bonds. But I’d expect difficulties if your initial state is anything but a lattice configuration; there are at least no obvious options to me. In general I don’t know that LAMMPS is all that great at representing interactions that are fundamentally stochastic.