yade-dev team mailing list archive

[François <francois.kneib@xxxxxxxxx>]

Thanks for your answers.
So it's clear that the input parameter kn has not the same signification
than the kn we obtain by doing kn=2*E*r with the input parameter E.
Thus, there is one more little thing I would like to discuss :
Imagine that there is contact between a FrictMat and a ViscElMat. So the
Ip2_FrictMat_FrictMat_FrictPhys will be used, and for the moment it needs
young modulus rather than stiffnesses. If the user have set kn,ks for his
ViscElMat
, and (required) young modulus for his FrictMat. It's probably a weird
user I agree, but what can we do for him, knowing that his ViscElMat.kn has
not the same signification than kn=2*E*r ? Two options (or more if you have
any suggestion) :

- Translate viscElMat.kn to a modulus using E=k/(2*r), and ignoring
massMultiply. That's bad I think, but just like setting stiffnesses and
modulus in the same simulation.
- Translate FrictMat.young into a stiffness using k=2*E*r, and multiply by
the mass if massMultiply is set true for the ViscElMat. Then compute
resulting Kn directly with stiffnesses. This looks even worse.

You understood, I vote for the first option. But I'm not sure that there
isn't an other way to do.

François


2013/12/19 Bruno Chareyre <bruno.chareyre@xxxxxxxxxxx>

> Sounds good.
> B
>
> On 19/12/13 12:15, Anton Gladky wrote:
> > Hi Francois,
> >
> > I am ok with these changes.
> >
> > Anton
> >
> >
> > 2013/12/19 François <francois.kneib@xxxxxxxxx>:
> >> Hi,
> >>
> >> I'm currently dealing with handling the contact between a ViscElMat and
> a
> >> FrictMat.
> >> To do this, and as I said some days ago, ViscElMat will have a Young
> Modulus
> >> and Stiffnesses. If both are set by the user, young modulus will be
> ignored
> >> (so the actual scripts will still work).
> >>
> >> I'm wondering about massMultiply. By default it is True and the
> resulting
> >> stiffness of the contact is K=k1*m1*k2*m2/(k1*m1+k2*m2), and the viscous
> >> factors are multiplied by the mass too. If the user sets stiffnesses, I
> >> understand that the masses comes into consideration when computing the
> >> resulting stiffness. But to my mind, if the user sets Young/Poisson,
> then
> >> ki=2*Ei*ri and the "particles shapes" are already taken into account,
> so we
> >> should not use massMultiply. That's not true for the viscous coefficient
> >> that should be computed the same way if the user sets stiffnesses or
> >> young/poisson.
> >>
> >> What I propose in the case the user sets young/poisson and
> massMultiply=true
> >> is : don't multiply for the stiffness calculation, but multiply for the
> >> viscous coefficients. In addition, by doing this, the behaviour will be
> the
> >> same than for the FrictPhys, which seems essential.
> >>
> >> Do you agree with this ? Of course, all this modifications will be
> described
> >> in the documentation.
> >>
> >> François
> > _______________________________________________
> > 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
> >
> >
> >
>
>
> --
> _______________
> Bruno Chareyre
> Associate Professor
> ENSE³ - Grenoble INP
> Lab. 3SR
> BP 53
> 38041 Grenoble cedex 9
> Tél : +33 4 56 52 86 21
> Fax : +33 4 76 82 70 43
> ________________
>
>
> _______________________________________________
> 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
>