yade-users team mailing list archive

[Bruno Chareyre <bruno.chareyre@xxxxxxxxxxx>]

> question on damping used for instance in the Triaxial Test. Is it right 
> that the solution is affected by the damping? Cause is what happens. 
>   
Yes in general. No, in special cases (quasistatic process).
You can obtain the "same"(*) stress-strain behaviour with different 
dampings.
(*) In fact "almost" the same, which can make a big difference in 
chaotic systems...

> The 
> damping included in the NewIntegrator class acts as a global damping, right? 
>
>   
Yes.
> If I want to use the local damping I see that I should use the 
> ViscousForceDamping class. In such case whatever else should I adopt for the 
> part related to the integration algorithm and Newton force law?
Put damping=0 in the latest.

>  Cause I see 
> that NewtonIntegrator includes everything on this side.
> Last question: Should 
> the damping helping to reach the equilibrium at each time step?
>   
It can.
>  When I include 
> the damping should I expect a decreasing in the total number of iterations?
>
>   
It's all a question of numerical experimentation. In the triaxial test 
in general, you will not really reduce the number of iterations with a 
strong damping, except perhaps in the first iterations when you compress 
the gaz cloud.

Bruno


--

_______________
Chareyre Bruno
Maître de Conférences

Grenoble INP
Laboratoire 3SR - bureau E145
BP 53 - 38041, Grenoble cedex 9 - France
Tél : 33 4 56 52 86 21
Fax : 33 4 76 82 70 43
________________