yade-users team mailing list archive

[Bruno Chareyre <bruno.chareyre@xxxxxxxxxxx>]

> I better mean, once looking at this line:
> Real Kr = Da*Db*Ks*2.0; // just like "2.0" above - it's an arbitrary
> parameter
> why the factor of 2 should always work in this case?

I'll not try and justitfy the 2, it's inherited from Plassiard who put 2 after some
calibration.
Think the other way : decide what you would like instead of 2, and let you implement it
somehow.

The best would be IMO to add the new factors in Bodies physics, the same as poisson for ks.
So the only dimensional input parameter for stiffnesses would be Young.

A consistent naming (suggestion) would be :
- rename poisson -> alphaKs
- alphaKr (rolling stiffness factor)
- alphaKtw (twist stiffness factor)

Then in Ip functor :
Kr = Da*Db*Ks*alphaKr;
Ktw = Da*Db*Ks*alphaKtw;

You are welcome to do that in the existing CohFrict functors, it would be good improvments.

Cheers.

Bruno