Specification of the reference coordinate system for crystal plasticity models when apply euler angle rotations

[wnoh3]

Hello,

I am currently trying to understand from what reference/initial coordinate frame Warp3d applies given Kocks euler angles. To my understanding, usually a rolling direction, transverse direction, and normal direction are established before applying any rotation per information such as slide 9 in this CMU lecture: http://pajarito.materials.cmu.edu/lectures/L1-Introduction-12Jan16.pdf . However, I do not see any documentation of this in the Warp3d manual or examples (including the subdirectory examples). If someone would be able to point me to a CP example or a specific section of the warp3d manual where they explain where a reference/initial coordinate system for the euler angle rotations is set, I would be very grateful. Apologies in advance if this question is very simple.

[reverendbedford]

It's in the documentation on page 3.12-3 (the potato diagram). We use the passive convention where the rotation goes from the reference to the crystal frame.

[wnoh3]

Hello,

Thank you for your quick reply! I have read the section of the paper your referred to and still have additional questions. Per the second paragraph in section 3.12 (page 3.12-1), the warp3d manual states that “Euler angles defining initial orientation of the atomic (crystal) lattice relative to the (global) model coordinate system.” However, the manual do not specify which global axes corresponds to which rolling, transverse, and normal axes necessary for the euler angle rotations (see diagram below). So specifically, does the global X axis = rolling direction, global Y = transverse direction, and global Z = normal direction? Is it different? If so, what is it? I believe this is important because if I assume the incorrect convention, then I will have the incorrect euler angles orientations.

[reverendbedford]

Hey,

The crystal orientation (Euler angles) cannot and do not tell you anything about the two coordinate systems you are moving between. They only tell you the transformation, in our case a rotation, between the two frames of reference. This isn’t special to warp, any time you see orientations/Euler angles you have the same problem. Information about these coordinate systems is extraneous to the job of warp3d which just solves problems and relies on you to define and interpret the results. So there is no convention baked into crystal plasticity codes — it’s up to you as the user.

On the crystal side the coordinate system is set by convention. As it turns out this doesn’t really matter for cubic crystals because of symmetry — any reasonable choice gives the same results. We use 100 = x, 010 = y, and 001 = z, but again this doesn’t really matter.

On the other side in the lab/reference frame the choice is entirely up to you and your experimental friends. What I mean is that if you were given a texture you also need the experimentalist who measured it to tell you something about the material. For example they would have to tell you x=TD, y=RD, z=ND. Oftentimes they won’t tell you but that’s because you can generally figure it out yourself by staring at pole figures.

If you’re starting from an untextured material then this is entirely up to you — you define the coordinate system by how you apply the loading. So if your roller is parallel to x and you pull the material in y then you get the coordinates in my example above.

Hopefully that makes sense!

On Wed, Feb 28, 2024 at 15:20 William Noh @.***> wrote:

Hello,

Thank you for your quick reply! I have read the section of the paper your referred to and still have additional questions. Per the second paragraph in section 3.12 (page 3.12-1), the warp3d manual states that “Euler angles defining initial orientation of the atomic (crystal) lattice relative to the (global) model coordinate system.” However, the manual do not specify which global axes corresponds to which rolling, transverse, and normal axes necessary for the euler angle rotations (see diagram below). So specifically, does the global X axis = rolling direction, global Y = transverse direction, and global Z = normal direction? Is it different? If so, what is it? I believe this is important because if I assume the incorrect convention, then I will have the incorrect euler angles orientations. image.png (view on web) https://github.com/rhdodds/warp3d/assets/122050959/a8761264-bb5f-4f16-bda7-19445eb17148

— Reply to this email directly, view it on GitHub https://github.com/rhdodds/warp3d/issues/24#issuecomment-1969940005, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABZISQZVROXP6NL3P7JE5ZTYV6NSBAVCNFSM6AAAAABD47C2DOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNRZHE2DAMBQGU . You are receiving this because you commented.Message ID: @.***>