parentGrainReconstructor

[ghost]

Hi Ralf,

In the step compute parent orientations, the parent phase was not reconstructed using the job.calcParentFromGraph function.

I don't have the FCC parent phase in my data, conssequently, in my CSList. So, I'm having trouble in the initial guess of orientation relationship.

Please, how can I deal with this problem?

[ralfHielscher]

Hi Ariane,
in case you are asking how to use parentGrainReconstructor - you have to download MTEX 5.6.beta.

A happy new year,

Ralf.

[ghost]

Hi Ralf,

In the step compute parent orientations, the parent phase was not reconstructed using the job.calcParentFromGraph function.

I don't have the FCC parent phase in my data, conssequently, in my CSList. So, I'm having trouble in the initial guess of orientation relationship.

Please, how can I deal with this problem?

[ralfHielscher]

The documentation says

The parentGrainReconstructor guesses from the EBSD data what is the parent and what is the child phase. If this guess is not correct it might be specified explicitely by defining an initial guess for the parent to child orientation relationship first and passing it as a third argument to parentGrainReconstructor.

This means you need to define the parent phase yourself by

csParent = crystalSymmetry('432','mineral','FCC') % whatever your parent phase is
p2c = orientation.KurdjumovSachs(csParent, ebsd('childPhase').CS)
job = parentGrainReconstructor(ebsd,grains,p2c)

Ralf.

[ghost]

Ralf,
I tried it before. The parent grains were not reconstructed.

job.calcParentFromGraph
computing parent grain orientations: 100%
 
ans = parentGrainReconstructor (show methods, plot)
 
 phase   mineral       symmetry  grains  area   reconstructed
 parent  Iron (Gamma)  "432"     17596   0%     0%           
 child   Iron (Alpha)  "432"     660     0.49%               
 
 parent to child OR: (340.1°,9°,63.6°)
   fit with child to child misorientations: 4.9°, 9.2°, 14°, 17° (quintiles)
   closest ideal OR: (31-2) || (12-1)   [112] || [012]   error: 2°

[ralfHielscher]

The reason is that you child 2 child misorientations do not fit the parent to child orientation relationship. what is the output of `job.calcParent2child` Ralf.

[ralfHielscher]

it seems you mixed the parent and the child phase. What es the output of

csParent = crystalSymmetry('432','mineral','FCC') % whatever your parent phase is
p2c = orientation.KurdjumovSachs(csParent, ebsd('childPhase').CS)
job = parentGrainReconstructor(ebsd,grains,p2c)

Ralf

[ghost]

Half,
I agree with you that parent and the child phases were mixed.

` csChild = ebsd('Iron (Alpha)').CS
csParent = crystalSymmetry('432', [3.6 3.6 3.6], 'mineral', 'Iron (Gamma)', 'color', [0.56 0.74 0.56])

p2c = orientation.KurdjumovSachs(csParent,csChild)

job = parentGrainReconstructor(ebsd,grains,p2c) `

job = parentGrainReconstructor (show methods, plot)

phase mineral symmetry grains area reconstructed
parent Iron (Gamma) "432" 0 0% 0%
child Iron (Alpha) "432" 18256 100%

parent to child OR: (335.8°,10.5°,65.8°)
fit with child to child misorientations: 3°, 4°, 4.8°, 5.6° (quintiles)
closest ideal OR: (10-1) || (11-1) [111] || [011] error: 0°

job.calcParent2Child

ans = parentGrainReconstructor (show methods, plot)

phase mineral symmetry grains area reconstructed
parent Iron (Gamma) "432" 0 0% 0%
child Iron (Alpha) "432" 18256 100%

parent to child OR: (340.1°,9°,63.6°)
fit with child to child misorientations: 1.6°, 2.3°, 2.9°, 3.9° (quintiles)
closest ideal OR: (31-2) || (12-1) [112] || [012] error: 2°

I did the intermediate steps according to documentation.

job.clusterGraph('inflationPower',1.6)

ans = parentGrainReconstructor (show methods, plot)

phase mineral symmetry grains area reconstructed
parent Iron (Gamma) "432" 0 0% 0%
child Iron (Alpha) "432" 18256 100%

parent to child OR: (340.1°,9°,63.6°)
fit with child to child misorientations: 1.6°, 2.3°, 2.9°, 3.9° (quintiles)
closest ideal OR: (31-2) || (12-1) [112] || [012] error: 2°

graph: 17596 grains in 600 clusters + 673 single grain clusters

job.calcParentFromGraph

computing parent grain orientations: 100%

ans = parentGrainReconstructor (show methods, plot)

phase mineral symmetry grains area reconstructed
parent Iron (Gamma) "432" 17596 0% 0%
child Iron (Alpha) "432" 660 0.49%

parent to child OR: (340.1°,9°,63.6°)
fit with child to child misorientations: 4.9°, 9.2°, 14°, 17° (quintiles)
closest ideal OR: (31-2) || (12-1) [112] || [012] error: 2°

Please, did I miss a step?

Is there any chance that parentPhaseId has not been identified correctly?
I don't have this problem when the data has an FCC phase (Pixels = 0). I believe that the phases are not mixed when the phaseId is extracted from ebsd.CSList. But, I don't know if I understood the code correctly.

Sorry if I couldn't explain the problem clearly.

[ralfHielscher]

Hi Arianne,

there is indeed a bug when the parent phase was not part of the EBSD data. I fixed it here 696ca1e. Are you able to apply this patch? You need to change only 3 lines of code within MTEX.

Except from this. Your grains have already been reconstructed. You can access them by job.grains('Iron (Gamma)')

Ralf.

[ghost]

Ralf,
I changed the code. Thank you!