yade-users team mailing list archive

["SHIU wenjie" <wen-jie.shiu@xxxxxxxxxxx>]

Hi Vaclav,

I did get some pre-generated numerical specimens for uniaxial test and 
triaxial test. I can send you some files later!!! Or you can even make the 
specimen by yourself! I can explain the way that i use to do the "specimen 
generation".

I take the Triaxialtest as an example base, i just add some changes in this 
class.
You can start from the radii expansion process to get a sphere assembly with 
isotropic property. When the stability criteria is satisfied, saving the 
position and the radius of all the spheres. And the specimen generation is 
done!! Be care about the ratio between the stiffness of the spheres and the 
target confinement.


How many spheres should we use?? Well, you can take 10000 spheres if you get 
a powerful PC. Indeed, i did some uniaxial simulations, and i did not get 
the size effect (between 2400 to 8000 spheres)!!

And the problem about the dilatance when runing the triaxial test, I am 
still trying to find out a solution!! It could be caused by the constitutive 
law or many other factors!!!

Regards

wenjie

----- Original Message ----- 
From: "Václav Smilauer" <eudoxos@xxxxxxxx>
To: <yade-users@xxxxxxxxxxxxxxxx>
Sent: Wednesday, November 28, 2007 1:21 PM
Subject: [Yade-users] reference sphere assembly


>I would like to calibrate material parameters to match macroscopic
> properties using uniaxial traction and later triaxial compression tests.
> Is there some widely used, pre-generated sphere assmebly that has
> properties such as isotropy and "reasonable" radii variance that could
> be used? (Something like that should be part of Yade, I think).
>
> If no, what procedures to you recommend to create such assembly? How
> many sphere should I go for?
>
> I will work soon on triax test which will have "true" hydrostatic
> confinement (by using delaunay to detect the solid edge and applying
> forces at relevant spheres) - that will solve the problem of restricted
> dilatance for triax (right, Wenjie?).
>
> Regards, Vaclav
> _______________________________________________
> Yade-users mailing list
> Yade-users@xxxxxxxxxxxxxxxx
> https://lists.berlios.de/mailman/listinfo/yade-users 

_______________________________________________
Yade-users mailing list
Yade-users@xxxxxxxxxxxxxxxx
https://lists.berlios.de/mailman/listinfo/yade-users