yade-users team mailing list archive

[Chareyre <question159394@xxxxxxxxxxxxxxxxxxxxx>]

Question #159394 on Yade changed:
https://answers.launchpad.net/yade/+question/159394

    Status: Open => Answered

Chareyre proposed the following answer:
1/ This equation is formally correct, I don't see how a gain could be
introduced. Actually, I'd keep calling 1/mass the "inverse inertia"
rather than "gain", since it explains the two methods: one is static,
the other is dynamic. It is better to have two methods, since different
users will prefer one or the other. The only problem in dry granular
materials is that there is no simple derivation of stiffness (unlike
cemented materials with 12 interactions per body or more => no rolling).

2/ The comment is about this cycle :
resetForces->integrateAndUpdate->contactLaw->PeriCompressor->NewtonsLaw.
The simplest way to understand what id does is to rewrite it differently
(it's a cycle, so I can start anywhere) :
- compute forces at time "t" (contactLaw)
- define a new velGrad on the basis of computed forces
- integrate bodies motion based on forces at time "t" and corresponding
velGrad
- integrate cell deformation using the same velGrad (we reseted forces
in between, it doesn't matter)

It should be correct if I don't miss something: forces(t) are used to
define velGrad(t+dt/2) and body.vel(t+dt/2). I was not 100% sure when I
wrote the code, so I added this note. The "else" would be using forces
at time t to define velGrad at t+3dt/2.

Bruno

-- 
You received this question notification because you are a member of
yade-users, which is an answer contact for Yade.