Rotation Convention in Documentation for ComputeElasticityTensorCP

[simopier]

This issue (idaholab/salamander#130) was created in SALAMANDER by @dannypanduro with a question regarding the naming convention in the ComputeElasticityTensorCP Documentation. I am reproducing below since this is a question about MOOSE's solid mechanics module.

Hello, I am reviewing orientation matrices, so I am scouring online, and thought some of the text was unclear on the ComputeElasticityTensorCP page. https://mooseframework.inl.gov/salamander/source/materials/crystal_plasticity/ComputeElasticityTensorCP.html

The formula given for the Bunge Euler rotation matrix is the active form in equation 2, and in equation 3, you use R transposed, which would give you a passive rotation of the stiffness tensor. In the paragraph below equation 3, you state the function "employ an active rotation: ", but I think you meant a passive rotation?

I apologize for being verbose, (I was also using this note as a sounding board for my understand), or if I am mistaken.

Thank you,
Danny Panduro

[GiudGiud]

@sapitts @jiangwen84 do you know?

[sapitts]

Certainly the many, and sometimes conflicting, definitions of Euler angles across different fields complicate the use of the Euler angles. For this, and other numerical stability reasons, other codes use Rodrigues or modified Rodrigues formulations.

In this documentation, we use the nomenclature 'active rotation' to ensure consistency with the MOOSE parent class documentation, https://mooseframework.inl.gov/source/materials/ComputeElasticityTensor.html , which uses the 'passive' designation. The key point is how the equations are implemented in the code, which does use the transpose as you have noted.

Hope this helps!