Thanks. Everything is as expected then, always good to hear.
For motion engines, I prefer not to use them as we can simply
assign velocity directly. Soon, you will not have to modify
anything in Newton, since this behaviour of non-dynamic bodies is
currently being implemented.
Test scripts without Newton is not a good thing IMHO, you are
putting yourself in a special case wich will never happen in
routine simulations. It can hide existing problems or create
problems that would not exist with Newton.
I partially agree. I guess it depends on the task. I found quite
useful the use of motion engines to test contact laws.
Partially is good enough. ;)
I would have a question about the ratcheting. I have run the same test
also for HM and of course I do not get the same good result as in the
case of the linear formulation. The reason is that in HM the
tangential stiffness depends on the normal force (or say on the
overlapping alike), and this is different from the initial one once I
come back with the rotation (say at the last step). Hence shear force
is not equal to zero although granular ratcheting is avoided. Do you
think this is expected? I have not really investigated too much the
granular ratcheting in the literature. Is this phenomenon perhaps
addressing only the linear elastic case?
I don't know your model enough to give a precise answer. Based on what
you write, it seems you could proove irreversible behaviour in cycles
even the analytical way?
Variable stiffness makes me wonder if it is not something linked to the
well known issue in hypoelastic models, see e.g.
http://linkinghub.elsevier.com/retrieve/pii/S0749641903001347.
Quoting :
"it is demonstrated that all hypothetical materials with non-potential
finite elastic or hypo-elastic constitutive relations can create an
energy from nothing, i.e. work as perpetual motion machines. This gives
a ‘physical’ proof of necessity of potential conditions in general
finite elasticity and hypo-elasticity and their extensions to finite
viscoelasticity."
Cheers.
Bruno
_______________________________________________
Mailing list: https://launchpad.net/~yade-users
Post to : [email protected]
Unsubscribe : https://launchpad.net/~yade-users
More help : https://help.launchpad.net/ListHelp
