yade-dev team mailing list archive

[Bruno Chareyre <bruno.chareyre@xxxxxxxxxxx>]

>> the main difference is that the contact stiffness kn and the ratio kt/kn are 
>> defined in the IP2 functor as indicated in the commit massage via MatchMaker
MatchMakers is right, yet I'm not sure it is incompatible with defining
parameters in materials as usual.

>> IMO, there is a potential of merging FrictViscoMat with ViscElMat.
I vote for this!


>>  However, I 
>> wanted to commit this law so that we can see the advantage of using MatchMaker 
>> and IP2 functors between different materials.
Ok.

>>  Last but not least I think we 
>> could probably split visco-elastic and capillary parameters.
I didn't know they were all merged. A result of recent changes by Anton?
It is easy to fix by inheritance, as in CapillaryPhys.

>> Anyway, I am glad you asked because I see potential of improving on both 
>> sides. What do you think?

My main concern is to avoid the increasing number of almost equivalent
classes, thus it rang a bell when I saw a viscous contact law commited
(which was obviously a duplicate for the most part). Thank you for
suggesting improvements, now they have to be made ;), since we don't
want duplicates in trunk.
Something to discuss with Anton and Raphaël, it seems.

Note that by starting an independant inheritance branch you lost the
advantage of automatic time-step determination with
GlobalStiffnessTimeStepper, which accounts for viscous effects only
after YADE_CAST<ViscElPhys*>.

Bruno