Reconstruction
- OIM HDF5 File - This is the name and location of the .h5ang file you wish to reconstruct
- a. The minimum and maximum slice numbers will be automatically displayed with the z resolution or spacing.
- Start Slice and End Slice - This defines the range of sections you wish to reconstruct
- a. Like you can choose to not import the entire dataset, you can choose to not reconstruct everything you imported.
- Phase Types - these are required descriptors to classify the ÔtypeÕ of phase each phase in the EBSD scan. These classifications are used when moving the data to the Synthetic Builder, but are not critical to the reconstruction itself only the statistics calculations.
- Options - These are all the options that can be set to control the reconstruction
- a. Merge Twins - this is an option that will indentify neighboring grains that have a twin relationship (3,5,9) and merge them together. Effectively, this is a second burn algorithm (performed on the identified grains, rather than the individual voxels) using the specific twin misorientations as a joining criterion rather than the misorientation tolerance used for forming the grains.
- b. Merge Colonies - this is an option that is similar to the Merge Twins option, but will reconstruct prior beta grains in an alpha-beta titanium microstructure. It will first identify grains (which would be alpha colonies in a transformed beta microstructure) and then perform a burn algorithm on neighboring grains/colonies, using the specific misorientations between alpha variants from the same beta grain as the joining criterion.
- c. Rectangularized Sample this option will fix the sample to look like a perfect rectangular sample by determining the smallest rectangular box that fits around the reconstructed volume and coarsening the structure to fill in the empty space within the rectangular box. Note that the box is created with axes along x, y and z. This option is generally not used unless the reconstruction needs to be used in a program that cannot handle having Ôempty spaceÕ outside of the sample (i.e. possibly a Finite Element simulation).
- d. Alignment Method this allows the user to choose the method used to align neighboring sections. Outer Boundary simply tries to align the area identified as sample on each section. Minimum Disorientation moves the sections until there is a minimum value of disorientation between all voxels in one section and the voxel directly above it in the next section. Mutual information segments each section independently and then performs a mutual information analysis on the grain ids.
- e. Ang Orientation this defines the origin of the ang file. Note that TSL and HKL define different locations on an EBSD scan as the origin. Additionally, the origin may be defined differently for the Euler angles and the morphological features. This setting allows the user to define what location in the 2D EBSD map they want to be the origin for both the Euler angles and morphological features. The Euler angles and x/y coordinates of the scan are all consistently modified by the code during reconstruction and not changed in the .h5ang file (which means the origin needs to be set every time the dataset is reconstructed.
- f. Min Grain Size - this a clean-up filter that can be set to merge grains below a defined size after grains have been formed. The voxels of the removed grains are reset to unassigned and then the region is filled in by a local grain growth algorithm.
- g. Disorient. Tol. - this is the orientation difference tolerance allowed between neighboring voxels before they are declared to belong to different grains. Neighboring voxels with misorientations lower than the tolerance are joined during a burn algorithm.
- h. Min. Image Quality - this is a threshold value that is used to define the outer border of the sample and terminate grain formation during segmentation. Image Quality is a metric used by TSL to define the quality or perfection of the orientation pattern at a data point. Generally when serial sections are collected, there is some extra area collected around the sample and this data usually has ÔmuchÕ lower values of Image Quality. This tolerance thresholds the outer area and also ends the grain formation burn algorithm, because it is not desirable to begin formation of a grain with a ÔveryÕ low value of pattern quality.
- i. Min. Confidence Index - this is threshold used the same as Min. Image Quaity, but using another TSL data metric that defines the confidence in the orientation determined from the orientation pattern.
- j. Down-sample Res. - this a factor used to down-sample the 3D volume if desired. If the value is set to 1, then nothing happens, but any other value will scale the resolution by that value. A new grid is created and laid over the original data grid. This option is usually only used if the dataset is too large to transfer into some other program(s) and needs to be ÔcoarsenedÕ.
- k. Bin Step Size - this is the step size used in correlating statistical descriptors to grain size. The value is set in microns, so for example, if the maximum grain size is 50 microns, a value of 5 would create 10 size bins in which grains would be segregated to create shape, no. of neighbors, etc. distributions as a function of grain size. This value should be selected by balancing each of the following: how quickly are other statistics likely to vary with size and how many grains are needed in each size bin to describe grains of that size. Creating size bins too small will result in not enough grains in a bin and creating size bins too large may miss identifying correlations.
- Output Directory - location that all output files will be written to.
- Output File Prefix - this option allows the user to define a prefix to place on the output data files (which have a fixed basename). The full names of the files will update in real-time on the list of output files.
- Write Binary VTK Files This option allows the user to choose whether the output visualization files are in binary or ASCII format.
- Optional Output Files these are all visualization files that can be created. The output files are .vtk files of structured grid type.
- a. Reconstruction Visualization - this file contains (at each voxel) grain ids, phase ids and the number of neighboring voxels that have grain ids different from the reference voxel.
- b. Reconstruction IPF Visualization - this file contains (at each voxel) grain ids and Inverse Pole Figure colors
- c. Reconstruction Disorientation Visualization - this file contains (at each voxel) grain ids and Kernel Average Misorientations and Grain Average Misorientations
- d. Reconstruction ImageQuality Visualization - this file contains (at each voxel) grain ids and Image Qualities
- e. Reconstruction SchmidFactor Visualization - this file contains (at each voxel) grain ids and Schmid Factors
- f Reconstruction DownSampled Visualization - this file contains (at each voxel) grain ids for the down-sampled volume
- g. Reconstruction Grains - this file contains each grain with properly defined voxels and their neighbor connectivities. This file is a binary HDF5 format of vtk poly data. Additionally, all the attributes listed in the other files are all contained in this one file.
- Load/Save Preset this allows the user to save input settings or load previously saved input settings