← Back to team overview

yade-users team mailing list archive

Re: [Question #693568]: Stresses drop to zero in periodic triaxial shear simulation

 

Question #693568 on Yade changed:
https://answers.launchpad.net/yade/+question/693568

    Status: Open => Answered

Jan Stránský proposed the following answer:
Hello,

In your code
###
def prepareShear(rate,confinement):
    ...
    triax.stressMask = 7
    triax.goal[1] = -1e3*confinement
    triax.goal[0] = triax.goal[2] = -triax.goal[1]/2
    O.cell.velGrad=Matrix3(rate/2,0,0,0,-rate,0,0,0,rate/2)
   ...
###

> triax.stressMask = 7
> triax.goal[1] = -1e3*confinement

you set stress-controll, so strain/volume is NOT kept constant (print
also strain/volume, not only stress) but is the result of simulation
(trying to reach defined stress).

> triax.goal[0] = triax.goal[2] = -triax.goal[1]/2

you require **tensile stress** in two directions. As the triax engine
tries to reach this, it enlarges the cell, probably causing volume
enlargement and "particle separation". I suspect this also causes the
final zero stress.

> O.cell.velGrad=Matrix3(rate/2,0,0,0,-rate,0,0,0,rate/2)

this is overwritten by PeriTriaxController.

Proposed solution:
- set triax.dead = True (not to overwrite cell deformation)
- set O.cell.velGrad as it is now (without triax, the volume should be kept constant)

cheers
Jan

-- 
You received this question notification because your team yade-users is
an answer contact for Yade.