search

MechMicroMan / DefDAP

A python library for correlating EBSD and HRDIC data
Apache License 2.0
36 stars 18 forks source link

Check slip trace predictions #2

Closed JQFonseca closed 5 years ago

JQFonseca commented 7 years ago

We could compare them with with Channel 5, CDT excel spreadsheet etc.

mikesmic commented 7 years ago

Slip traces for FFC systems have been checked against Brad's excel spreadsheet and all agree.

JQFonseca commented 7 years ago

Are we doing this for HCP as well?

mikesmic commented 7 years ago

HCP slip traces are still a work in progress. I wrote code to transform HCP crystal vectors to an orthonormal basis that should then be able to be used in the same way as cubic crystal vectors. However @rhysgt checked with his maps and directions do not seem correct. I think it is related to alignment of the EBSD spatial coordinates and the coordinates the orientations are defined in. This would not be an issue for cubic structures due to symmetry.

JQFonseca commented 7 years ago

It would be good to get an update from @rhysgt. Maybe he could upload a notebook show how it doesn't work and we can all have a think about it?

JQFonseca commented 7 years ago

@mikesmic Could this be why we couldn't get the Mg grain to break up in the right way?

If the axis definitions are wrong, couldn't it still be a problem in cubic when we decompose the misorientation into the different components, even for cubic?

mikesmic commented 7 years ago

Yes this will most likely be the reason for that too.

mikesmic commented 7 years ago

The problem was with how I was orthonormalising the planes. The miller indices for planes are covariant so need to be transformed using the inverse L matrix where I was using just the L matrix. I have fixed this and the orthonormalised crystal vectors and slip trace vectors all agree with Brad's excel book. @rhysgt has tested with his data and gets decent agreement using a pre deformation EBSD map. He is going to take a post deformation map and check those soon.

JQFonseca commented 7 years ago

What does covariant mean? I don't understand it - you must explain it to me.

mikesmic commented 7 years ago

The procedure is given in section II of this paper http://aip.scitation.org/doi/10.1063/1.1661333 But from what I have read for the miller indices of a plane represent it's normal in reciprocal space, so it has to be transformed as a reciprocal lattice vector. For cubic it is simpler and the miller indices of the plane are its normal vector.

JQFonseca commented 7 years ago

Thanks. That I understand but what does that have to do with it being covariant?

On 7 Jun 2017, at 10:20, Michael Atkinson notifications@github.com wrote:

The procedure is given in section II of this paper http://aip.scitation.org/doi/10.1063/1.1661333 But from what I have read for the miller indices of a plane represent it's normal in reciprocal space, so it has to be transformed as a reciprocal lattice vector. For cubic it is simpler and the miller indices of the plane are its normal vector.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

mikesmic commented 7 years ago

Sorry I probably just confused things with that. I was just relating it to maths I've dealt with before to help me understand. For non-orthogonal coordinates you define a contra and co varying basis to deal with the unit vectors not being orthogonal to each other. Direct and reciprocal lattices use the same concept I think.

JQFonseca commented 7 years ago

No worries. I would like to understand what that means. I have read about this when using generalised coordinates. I hadn't understood the link between the reciprocal lattice and co- and contra-variance.

rhysgt commented 5 years ago

Slip trace predictions and Schmid factor calculation now checked for Zr and Ti