EMsoft-org / EMsoft

Public EMsoft repository
Other
61 stars 94 forks source link

EMTKD - inconsistent pattern rotation #130

Open Habakuk224 opened 3 years ago

Habakuk224 commented 3 years ago

Hi,

I have found an issue in EMTKD - incorrect rotations of the beam intensity.

At first, I was not sure what is the correct convention for the sample tilt - I would use -20°, Wiki tutorial on TKD says +20°. Anyway, I produced both masterfiles for alpha-Ti with EMMCfoil.txt, changing only the sig to +/-20°, processed through EMTKDmaster and plotted using EMTKD with EMTKD.txt (for simplicity - pattern centered on detector) as a series with detector tilt from -70° to 110°. The (0001) basal plane of Ti is parallel with the foil, Euler angles (0.0, 0.0, 0.0). Orientation of the patterns is the same as when viewed by HDFview.

Now, the Kikuchi pattern seems to be rotated correctly, while the dark band, which I assume is an intensity singularity in the foil plane, is moving in an opposite direction. (It could actually be the opposite way, if I got the pattern orientation upside-down - intensity would be rotated correctly and master pattern not.)

Source code version : 5_020201010 (Win64 build)

m20_m p20_m

elena-pascal commented 3 years ago

Hi,

I think this is partly related to this discussion. The black band indicates where the intensity cuts off on the detector (due to geometry) and it should just be just black on the other side. This is something that had been fixed in the EBSD code but perhaps not in the TKD code.

I think you are right, the intensity rotation seems off. The MC code is independent of the diffraction code, so it could be that frames are not consistent. I'll have a look.