Yade

Time Stepper for a new material model

Asked by Ricardo Pieralisi

Hello all,

My question is regarding the time stepper for different models. Now YADE has the time stepper for elastic and viscoelastic models [1]

My model is like viscEL but the spring (Kn and Ks) and the dashpot (Cn and Cs) are nonlinear. The problem is with the dashpot (the material is much more viscous than elastic), as the rheological's coefficient changes with the overlap volume I have to use time stepper. Is it possible to relate my model with the existent ones [1]? Is it possible to change the rheological's coefficient each time step inside [1]? . How do I load the values?

Thanks a lot,
Ricardo.

[1] https://github.com/yade/trunk/blob/master/pkg/dem/GlobalStiffnessTimeStepper.cpp

Question information

Language:
English Edit question
Status:
Answered
For:
Yade Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Jérôme Duriez (jduriez) said : 		#1

Hi Ricardo,

My point of view is that, as soon as your interactions physics derive from NormShearPhys, and if they store the appropriate (current, "tangent") value for the spring stifnesses, you should have no problem using GlobalStiffnessTimeStepper (GSTS) with non constant interaction properties. Each time GSTS executes, it loops over interactions to use stored properties values. (Then, no need to "load" changing values)

I am not familiar with dashpot, but I would think the same holds ?

Hope this helps a little

Revision history for this message
Chareyre (bruno-chareyre-9) said : 		#2

Hi Ricardo,
It will work if GSTS can find relevant values for kn, ks, cn and cs.
An approach is to keep them updated to reflect the tangent values of
stiffness/viscosity: dF/du and dF/dv (u: disp., v: velocity).
For instance, if your law reads: fn = E*un^2, you would define
phys->kn=2*E*un so that GSTS can use it.

I can see a regression in HertzMindlin on this aspect since, some time
ago, it was defining kn and ks in this way for the sole purpose of
getting results in GSTS.
Currently they are not defined (hence kn=ks=0).

Bruno

Can you help with this problem?

Provide an answer of your own, or ask Ricardo Pieralisi for more information if necessary.

To post a message you must log in.