EBSD data slices to 3d microstructure

[Zephan D.]

Hello All,

I am trying to make a microstructure from ebsd data.

Using the prebuilt pipeline I can get dream3d structures from the .ctf files I have.

However, I do not know how to piece them together properly.

I think I would need to use the append Z slice filter, but when i generate my microstructures, I do not have "CellData."

What can I do?

I wish to put three different slices together with varying thicknesses for each one.

Thank you!

Z

[Zephan D.]

I solved my problem.

However, when I place my slices together, one of the slices jumps to the bottom, and I dont know why.

I've included a shot of the structure in paraview with a slice applied

I had to use a rather confounded method to attain layering of repeated slices, so the dream3d file is wrapped as a zip so I may share it.

What could be the reason of one of my appended slices jumping to the bottom instead of going on top?

Does the Append Z filter reset after a certain amount of slices?

Thanks!

Z

[Michael Jackson]

If you have a set of CTF files that are supposed to be stacked in 3 dimensions (versus a 2D Montage) then there is an example of doing this in the Prebuilt Pipelines->Workshop->EBSD Reconstruction ->(01) SmallIN100 Archive. That first pipeline will create an .h5ebsd file from your CTF files. Then you run the 2^nd pipeline (02) SmallIN100 Initial Visualiztion pipeline to read in the .h5ebsd file and start processing it.

 

--

Mike Jackson

--
You received this message because you are subscribed to the Google Groups "dream3d-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dream3d-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dream3d-users/f0814754-78b5-4ac7-8a66-634afffcececn%40googlegroups.com.

[Zephan D.]

I think I did not hit reply all in my response. I will repeat it here so it can exist in the forum.

These prebuilt pipelines are useful in introducing my data to dream3d. 

However, my current problems are different, mainly when putting the slices together.

I would like to

1. Orient my ebsd slices to line up with each other (I know they are sequential, but I believe I must rotate them to line up by euler angles better)
2. Assign proper thickness to the grains
3. Combine the ebsd data slices together

What I'm doing now:

1. I am looking into softwares to orient my data properly
2. I read in the forum that setting slice thickness is impossible in dream3d, but perhaps something has changed
3. Trying Append Z in older version of dream3d to see if they will not make my slices go out of order. If there is another way to "glue" slices together, I would like to try that as well

Thank You!

Zephan

[Michael Jackson]

The current way that DREAM3D uses to stack slices of EBSD data we assume that all the slices are generally oriented in the same direction. They can have shifts in the XY plane (which filters can correct for), but you will need to ensure that all slices are oriented in the same direction.

 

The current DREAM3D reconstruction filters do not operate on a Rectilinear grid, only on a strict regular grid. This means that along the same axis the slice to slice thickness will remain constant. You can attempt to work around this by stacking multiple copies of the same slice to make that a “thicker” slice. For instance if the second set of data that you acquired is 3x thicker than the first set of data then you could create the following input files:

 

Slice_01.ang

Slice_02.ang

Slice_03.ang

Slice_04.ang

 

Where Slice_02, Slice_03 and Slice_4 are the exact same data file just duplicated. Use the filter “Import Orientation Files to H5EBSD” to convert the EBSD files into an HDF5 archive file FIRST. Then create another pipeline that uses the “Import H5Ebsd File” to import the data into DREAM3D and start processing.

 

This _should_ work if performed correctly. Do let us know if you are still having issues after these steps are performed.

 

--

Michael Jackson | Owner, President

      BlueQuartz Software

[e] mike.j...@bluequartz.net

[w] www.bluequartz.net

 

From: <dream3...@googlegroups.com> on behalf of "Zephan D." <zepha...@gmail.com>
Date: Tuesday, October 11, 2022 at 1:39 AM
To: dream3d-users <dream3...@googlegroups.com>
Subject: Re: EBSD data slices to 3d microstructure

 

I think I did not hit reply all in my response. I will repeat it here so it can exist in the forum.

These prebuilt pipelines are useful in introducing my data to dream3d. 

However, my current problems are different, mainly when putting the slices together.

 

I would like to

1.       Orient my ebsd slices to line up with each other (I know they are sequential, but I believe I must rotate them to line up by euler angles better)

2.       Assign proper thickness to the grains

3.       Combine the ebsd data slices together

What I'm doing now:

1.       I am looking into softwares to orient my data properly

2.       I read in the forum that setting slice thickness is impossible in dream3d, but perhaps something has changed

3.       Trying Append Z in older version of dream3d to see if they will not make my slices go out of order. If there is another way to "glue" slices together, I would like to try that as well

Thank You!

Zephan

 

On Wednesday, October 5, 2022 at 8:56:47 AM UTC-4 mike.j...@bluequartz.net wrote:

If you have a set of CTF files that are supposed to be stacked in 3 dimensions (versus a 2D Montage) then there is an example of doing this in the Prebuilt Pipelines->Workshop->EBSD Reconstruction ->(01) SmallIN100 Archive. That first pipeline will create an .h5ebsd file from your CTF files. Then you run the 2^nd pipeline (02) SmallIN100 Initial Visualiztion pipeline to read in the .h5ebsd file and start processing it.

 

--

Mike Jackson

 

On 10/5/22, 00:39, "Zephan D." <dream3...@googlegroups.com on behalf of zepha...@gmail.com> wrote:

 

Hello All,

 

I am trying to make a microstructure from ebsd data.

Using the prebuilt pipeline I can get dream3d structures from the .ctf files I have.

 

However, I do not know how to piece them together properly.

 

I think I would need to use the append Z slice filter, but when i generate my microstructures, I do not have "CellData."

 

What can I do?

I wish to put three different slices together with varying thicknesses for each one.

 

Thank you!

Z

--
You received this message because you are subscribed to the Google Groups "dream3d-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dream3d-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dream3d-users/f0814754-78b5-4ac7-8a66-634afffcececn%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "dream3d-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dream3d-user...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/dream3d-users/20ed7470-a0e9-464f-9bf9-27a87d18a634n%40googlegroups.com.

[Zephan D.]

Hello Michael,

I did end up copying the .ang files several times to get a "thickness."

However, this is not as clean as simply using only 3 unique slices.

The microstructure I found with 3 slices seems to flow very well.

The microstructure I found with all slices does not seem to flow at all.

I have included a video here to explain more.

I focus on a large blue crystal easily visible through Euler Angles.

This crystal is noticeable throughout the structure made from 3 slices.

This does not happen when the structure is made from all slices.

This crystal, from "slice 1" probably shows up better when only 3 slices are used, because it carries the same "weight" as all other slices.

When using all slices, slice 1 is left with only 1 slice. However, none of the crystals mesh together the way they do when the structure is thinner.

Best,

Z