yade-users team mailing list archive

[Jerome Duriez <Jerome.Duriez@xxxxxxxxxxx>]

Dear Yade users,

Changes were introduced recently in the code of "Triaxial" engines (e.g. TriaxialStressController) : see code update [1] for details. Stress / strain-like attributes of such engines are now to be considered as positiv in tension.

We understand this could appear as unconvenient to the geomechanics guys, and it will require you update your local scripts, in case you update your yade version. But the aim was to have an unique framework throughout the whole code so that users do not have to check in the doc (or in the code...) each time they set or get a stress / strain value using Yade. We are much closer from this goal now.
Note that post-processing operations in python example scripts still reflect the geomechanics point of view, with curves designed on purpose to show positiv values.

Note also that the meaning of TriaxialStressController (and derived engines) stress attribute has slightly changed. There is now one unique convention for stress(id), whatever id value : stress(id) returns the sample stress, as measured along boundary 'id', with 'id' between 0 and 5. Hence there is no more any sign difference between stress(0) and stress(1) for example.


Second, for what concerns users of CapillaryPhys or MindlinCapillaryPhys classes, the meaning of fCap has been inverted in the code update [2]. For a given interaction, fCap corresponds now to the force sustained by the particle corresponding to interaction.id2 (it was the force sustained by interaction.id1 before). Here again, this could require some updates of local scripts on your side, but corresponds now to a global Yade framework.


We hope this will finally improve your Yade experience !

Jerome Duriez

[1] https://github.com/yade/trunk/commit/19e81a785b19a8eab2718ebb5ad5eb9456355818
[2] https://github.com/yade/trunk/commit/40df7e8944bfafa2d6aac19b837e4a091ea29b34