yade-dev team mailing list archive

[Chiara Modenese <c.modenese@xxxxxxxxx>]

Right, reading Bruno and Vaclav's comments I agree the bug is invalid, sorry
to have reported it. I can see now that the way the periodic boundary works
in Yade, that is using positions in the absolute space, is different than
Cundall/Thornton implementation; hence, also the algorithm presented in
their papers has to be treated accordingly. I will mark the bug as invalid.
(Thanks for explaining me.)

Chiara


2010/11/12 Václav Šmilauer <eu@xxxxxxxx>

> I was thinking about this a bit and I think the bug is not valid.
>
> The Cell.homoDeform==1 updates positions and does not touch velocities
> at all. Hence incident velocity (computed from difference of velocities)
> will always correspond to the fluctuation velocity, regardless of
> positions in absolute space -- the meanfield velocity only acts at
> positions, and does not affect velocities.
>
> If you look at Cell.bodyFluctuationVel, it simply returns state->vel in
> cases homoDeform∈{0,1}.
>
> Yes, meaning of the number is shifted, but if you assign the old
> parameter to NewtonIntegrator, it will shift it before assigning to
> Cell.homoDeform (and gives a very clear warning about that). The
> ordering is according to the increasing smoothness of the adjustment
> (none, positions, velocities, velocities+convection).
>
> --
> homoDeform=1 does not handle the periodic case correctly.
> https://bugs.launchpad.net/bugs/674000
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in Yet Another Dynamic Engine: New
>
> Bug description:
> Currently, in the periodic case, if homoDeform=1 (alias Cundall way) we do
> not adjust the values of the relative velocities as according to the
> velocity gradient. This should be done, since, in such a case, we only
> update particle positions and not particle velocities.
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/yade/+bug/674000/+subscribe
>

-- 
homoDeform=1 does not handle the periodic case correctly.
https://bugs.launchpad.net/bugs/674000
You received this bug notification because you are a member of Yade
developers, which is subscribed to Yade.

Status in Yet Another Dynamic Engine: New

Bug description:
Currently, in the periodic case, if homoDeform=1 (alias Cundall way) we do not adjust the values of the relative velocities as according to the velocity gradient. This should be done, since, in such a case, we only update particle positions and not particle velocities.