Error in Identifying Maximum Texture Components in ODF Section Plots

[Alec Saville]

Hello everyone,

I've been trying to identify the "most important" texture components (eg: maximum volume fraction) using MTEX 5.11 in ODF section plots and have run into a slight issue. The data I'm working with comes from beartex files imported after initial processing using MAUD 2.33.

Whenever I run the below code, MTEX will correctly identify the first maximum in the 30 degree section, but miss identifying the second maximum in the 0 degree section and instead mark an orientation in the 23 degree section (this can also be seen in the attached .jpg). Upgrading to MTEX 5.2 after the official release did not solve the issue.

Code in Question:

odf = calcODF(pf_rotated,'resolution',5*degree);  % Rotation is to align data with specific reference frame and completed in prior step.
[m,ori] = max(odf,3);
plotSection(odf,'phi2',[0 7.5 15 22.5 30 37.5 45 52.5]*degree)
annotate(ori);

I found one previous question on here where a similar issue was reported for an ODF calculated pole figure, but no definitive answer came out of that discussion (to my understanding).

Does anyone have an idea where I may be going wrong or an alternative command to use?

If anyone also has insight on how to automatically document each of the "most important" texture components (Bunge angles pre-adjusted for MTEX's inverse definition of orientation angles) and their respective volume fractions (both ODF and Uniform ODF) after MTEX identifies them I would be very grateful. My previous efforts were less than successful and any calculated volume fractions were questionable at best.

Of course, let me know if extra information would help.

Thanks in advance,

Alec

[Rüdiger Kilian]

Hi,
did you also try with the latest release (5.2.4)?
Cheers,
Rüdiger

[Alec Saville]

Hi,

I mistakenly thought my installation of 5.2.4 was 5.2, but it was 5.2.4 instead.  Both my 5.2.4 and 5.1.1 installations show this issue.

Cheers,

Alec

[Ralf Hielscher]

I would guess that the peak in the 0 degree section belongs to the same component as the detected maximum in the 30 degree section. 

How can you be sure that this is not the case?

Ralf.

[Alec Saville]

Hi Ralf,

I've already accounted for potential multiplicity in this material by restricting phi2 to 53 degrees maximum (hexagonal material), so my original suspicion was this is an independent texture component. I'm also assuming a triclinic specimen symmetry given my pole figures, and didn't see any symmetrical texture components in my 3D ODF's.

Evaluating the coordinates of both texture components by eyeball also shows both sets of values are different. Is there some logic I'm missing here?

Cheers,

Alec

[Ralf Hielscher]

Ok, in the case of hexagonal symmetry this should be indeed be a different component. Could you provide me the data in some way?

Ralf.

[Ralf Hielscher]

Hi Alec,

this appears to be a plotting bug. In your code you computed the most prominent 3 components. The first and the third one get plotted but the second one is missing.

This means you can rely in your further analysis on this second one as well - it just does not show up in the section plot.

In a pole figure plot it will be plotted correctly. In order to resolve this issue you may copy the file at

https://github.com/mtex-toolbox/mtex/blob/develop/plotting/ODFSections/%40ODFSections/ODFSections.m

into the path mtex/plotting/ODFSections/@ODFSections/ODFSections.m

Thank you very much for reporting this issue.

A happy new year,

Ralf.

On Wednesday, December 4, 2019 at 8:53:56 PM UTC+1, Alec Saville wrote:

[Alec Saville]

Hi Ralf,

Thank you so much for the help! I'll download the file and proceed with my analysis accordingly.

Have a great New Year,

Alec