File nomenclature enhancements and other suggestions

[StopkaKris]

Hi Hemant,

Below is a list of suggested nomenclature clarifications to assist users correlating information between some of the MIDAS output files such as SpotMatrix.csv and Radius_StartNr_1_EndNr_1439.csv. I realize some of these changes may affect the flow of logic and be difficult to implement but I nonetheless wanted to compile a list here.

• SpotMatrix.csv and Radius…csv both have a header entitled “SpotID”. However, users need to examine IDRings.csv to correlate between “NewID(RingsMerge)” and “OriginalID”, which correspond to “SpotID” values in SpotMatrix.csv and Radius…csv, respectively. Can these files just specify “NewSpotID” and “OriginalSpotID”?

• The same parameters have different column headers in different files, e.g., “DetectorHor” and “DetectorVert” seem to correspond to “YCen(px)” and “ZCen(px)” between SpotMatrix.csv and Radius…csv, respectively. Can these column headers be standardized? Also: “OmeRaw” in SpotMatrix.csv and “Omega(degrees)” in Radius…csv. What does raw indicate? These omega values are the same, whereas “Eta” and “Eta(degrees)” differ slightly; is this because one value is before peak refinement and one is after?

• Ensuring all column headers have units (e.g., pixels, microns, or degrees; for SigmaEta and SigmaR, are these total spot width or FWHM?).

• The following values are written as floats instead of integers, so this change would decrease generated file sizes:

  • RingNr in SpotMatrix.csv
  • SpotID, NImgs, RingNr, NrPx, and NrPxTot in Radius…csv
  • NrPx and NrPxTot in Result…csv

• This is very minor: there is a double space in the header column between “Eta(degrees)” and “DeltaOmega” in the Radius…csv file. This trips up reading this file in pandas using python, and while one can get around it, it would be terrific if this can be fixed as the file is generated.

Also, I have a request for a new file with the following information. It would be very useful if the Radius…csv file could be modified such that it exports the spot data associated with each grain written to Grains.csv. So this file would be a modification of Radius…csv in which spots not associated with reconstructed grains are removed and a column is added that contains the grain ID corresponding to the first column of Grains.csv. This would allow users to easily examine spots that comprise each reconstructed grain and eliminate the need for custom scripts. I see that SpotMatrix.csv has the grain IDs of interest, but Radius…csv contains more valuable information.

Best regards,
Krzysztof Stopka

[StopkaKris]

To follow up on this, I have an email that states that the units for SigmaR and SigmaEta are both pixels. However, several of my reconstructed files (for an unloaded specimen as well as after several thousands of fatigue cycles) show a range for SigmaR of 0.01 to ~15-17, whereas SigmaEta ranges from 0.005 to ~4. That leads me to believe that the units used within MIDAS are probably microns/length for SigmaR and degrees for SigmaEta, because I would not expect spots to be significantly wider in the radial direction than the azimuthal direction. Furthermore, majority of SigmaEta values fall below 1, so I am unsure how this could be resolved as a width below 1 pixel. Below is a scatter plot of all ~65,500 spots corresponding to a 300 micron tall FF scans of an unloaded and heat treated IN718 specimen.