yade-dev team mailing list archive

[Luc Sibille <luc.sibille@xxxxxxxxxxxxxx>]

Hi all,

I am still investigating the different moment-rotation contact laws in 
Yade (version 0.50 and last update version).

I have two interrogations:

1/ First in every case relative particle rotations is totaly recomputed 
at each time step from the initial orientation with a code like that:

Quaternionr delta((rbp1.ori * (initialOrientation1.conjugate())) * 
(initialOrientation2 * (rbp2.ori.conjugate())));

is it safe to make such a computation, where after a long time 
simulation, relative rotation can become very large more than 2*pi 
whereas there is after a correction for the computed relative rotation 
if it is greater than 2*pi.
In addition:
> Just to note that this formula will "fail" if the relative rotation of a
> particle will be more than 180 degrees (the quaternion will take it the
> other way around, as it goes always the shortest way). 


I imagined a priori to compute this relative rotation incrementally (as 
the shear force), to avoid to deal with great relative rotation. What is 
your opinion? why did you choose this formula and not an incremental 
formula?


2/ Then I rediscovered that aa.angle() can return nan, as commented in 
source file, it happens when the quaternion is close to identity. Please 
apologise for my lack of culture, but what does represent physicaly a 
quaternion close to identity?

Thank you in advance for your replies,

Luc




--
Luc Sibille

Université de Nantes - Laboratoire GeM UMR CNRS

IUT de Saint Nazaire
58, rue Michel-Ange - BP 420
44606 Saint-Nazaire Cedex, France

Tel: +33 (0)2 40 17 81 78