yade-users team mailing list archive

[Cyndie <question249830@xxxxxxxxxxxxxxxxxxxxx>]

New question #249830 on Yade:
https://answers.launchpad.net/yade/+question/249830

HI all,

I did some simulations using kinemCNDEngine, but I have got a strange result concerning the stress tensor shear components. Since I presume that this engine performs a symetrical loading with respect to vertical and horizontal walls, I was expecting a symetrical stress tensor with equal values of shear components (tho12 and tho21).

To comupte the stress tensor within the sample (2D sample), I used the getStress function (Love-Weber formula).

It returns something like this:
[s1  tho  0
  ~0    s2    0
 0    0      0  ]

I don't understand why tho21 is equal or tends to zero?

In order to avoid that, I have applied the same engine twice  on both left upper corner horizontally and to the right lower corner vertically upwards. 
Should it be done like this on this case? I'm affraid this may not be a simple shear test anymore. Am I right?


I am using kinemCNDEngine with rigid frictionless walls- does it still working under such conditions or should I have periodic boundaries or add friction to my rigid walls?

The same problem persists  considering a square specimen because I was thinking that the difference in shear components values may be related to the rectangular geometry of my specimen. Is it impacted by the geometry or walls friction?

Besides I need some clarifications to understand exactly how does this engine work:
the upper plate moves horizontally and the vertical plates follow, but:
  - Can the upper and bottom plate move vertically?
  - Is the point (2D)/the line (3D) joining the vertical and the upper plate still the same during the test?

Thank you 

Cyndie

-- 
You received this question notification because you are a member of
yade-users, which is an answer contact for Yade.