Simulating a torsional load on a RVE

[Erfaaannn]

Hello everyone,
I'm trying to simulate a torsional load with the spectral solver of DAMASK 3. For this purpose, I generated my shear load with the deformation gradient rate tensor as follows:
...

• discretization:
  t: 10
  N: 50
  boundary_conditions:
  mechanical:
  dot_F: [[0, 0, 0],
  [1.e-2, 0, 0],
  [0, 0, 0]]
• discretization:
  t: 80
  N: 50
  boundary_conditions:
  mechanical:
  dot_F: [[0, 0, 0],
  [1.e-2, 0, 0],
  [0, 0, 0]]
• discretization:
  t: 10
  N: 50
  boundary_conditions:
  mechanical:
  dot_F: [[0, 1.e-2, 0],
  [0, 0, 0],
  [0, 0, 0]]

I want to investigate the initial and final pole figure during this simulation. In the initial texture the [100] crystallographic direction is parallel to the Y axis (as shown in the attached image). I expect the final pole figure of my simulation to rotate about 90 degrees just like the example image that I attached in this message. But the problem is that I don't know how to simulate such a torsional deformation with the load files of spectral solvers of DAMASK. Let me tell you that I tried a lot to increase the time of applying the deformation gradient rate. But it appears that there is a limitation in the simulations for very large deformations and usually the error of maximum cut back of DAMASK happens when I increase the time. I will be so thankful if anyone can help me with this problem.

[MarDiehl]

unfortunately, torsional load with the spectral/grid solver is not possible.

[eisenlohr]

The reason for this inability lies in the fact that the spectral/grid solver always operates with periodic boundary conditions, which are incompatible with torsion of a unit cell/RVE. In other words, a unit cell that underwent torsional deformation cannot tile space anymore.