yade-users team mailing list archive

[chiara modenese <c.modenese@xxxxxxxxx>]

> ________________________
> Cundall way (as you rightly pointed out) is the following:
>
> *** update of velocities (he does not, but as you already said this would
> be the solution):
> vel+=velGrad*vel*dt
> *** update of positions:
> pos+=(velGrad*vel*dt)*dt
> ________________________
> Currently in Yade (NewtonIntegrator):
>
> *** update of velocities:
> vel+=(velGrad-prevVelGrad)*pos
> *** update of positions:
> pos+=(velGrad-prevVelGrad)*pos*dt
> ________________________
> There is a term of difference between these two formulations. Basically, it
> is like writing:
> pos_Yade = pos_Cundall - prevVelGrad*pos*dt
>
> Hello Bruno,

thanks for the update in the periodic. I was thinking again about the two
formulations. To the end of update positions, Cundall uses the gradient of
velocity which has the same unit of measure as the strain rate.

That said, why would you use (velGrad-prevVelGrad) instead of velGrad? Which
is the unit of measure of the rate of the gradient of velocity? Is this
formulation consistent? Taking the rate of velGrad instead of velGrad is not
really the same.

One more question. Why do not we move the spheres through the periodic
boundaries instead of updating their positions? Would this be possible? I
mean, like treating the periodic boundaries as walls. ATM, we use the rule
of continuum mechanics to updated positions of discrete particles, but is
this correct dealing with a particulate system? Would not be better to apply
the strain rate to the moving periodic boundaries and as a consequence
moving the balls?

Thanks a lot for your answers.

Chiara