yade-dev team mailing list archive

[Chiara Modenese <c.modenese@xxxxxxxxx>]

On 25 February 2011 08:06, Vincent Richefeu <richefeu@xxxxxxxxx> wrote:

>
> Le 23 févr. 2011 à 17:43, Bruno Chareyre a écrit :
>
> > p.s. Vincent, you defined viscosity in order to keep critical time-step
> below a given value IIRC. What was the reasoning behind?
>
> No. I (and other people like Sergei) defined the viscosity so that it is
> lower than the critical viscosity. If the viscosity is bigger than this
> critical viscosity the second order differential equation leads to an
> overdamped solution (which means exponential decays of the normal distance
> and thus the bodies never touch one another!). For practical reasons, one
> introduces a parameter which is a rate of the critical viscosity.
>

In PFC, as well as in many other codes, the viscous damping is introduce in
a system where the spring has linear behaviour. In such a case, one can
easily know about the viscous coefficient which would led to an overdamped
solution. When you talk about critical viscosity, is that what you mean?
Further, still in this case, it is analytical also the definition of
coefficient of restitution with respect to viscous damping. Anyhow,
according to the equations cited in PFC manual, the critical time step would
be smaller for viscous contacts even if the solution is underdamped (at
least for the linear dashpot model). Am I overlooking something? Why would
be the contrary to you? Sorry, I admit I have no great expertise on this
precise topic.

In my problem I have non-linear contact stiffnesses and an analytical
formulation cannot be found to identify the corresponding coefficient of
restitution given a certain level of damping. Though some people proposed a
numerical solution to the problem to physically relate viscous damping to a
corresponding coefficient of restitution (and in HM code there is this
option too). If in a quasi-static condition a different form of viscous
damping can make the difference, has to be checked, I guess. So, in which
cases to you the viscous damping leads to unphysical situations? I would be
happy if you could just give an example without going into much details.

Chiara


>
> Sergei went further in the law he implemented in yade...
>
> For some applications, the viscous damping leads to unphysical
> situations... I wil not dwell on the topic.
>
> Vincent
> _______________________________________________
> Mailing list: https://launchpad.net/~yade-dev
> Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~yade-dev
> More help   : https://help.launchpad.net/ListHelp
>