TMS Meeting Minutes, September 10, 2020

Attendees: Andrew, Breann, Heejoo, John, Matthieu, Trenton, Yang (Zoom Meeting)

Meeting discussion summary:

  • Paper on ELT's referenced the TMS work at LBT!
  • Intra and extra focal images were taken on LBCr on 2020_09_06 (thank you, Olga!)
    • Intra focal images have extra spot on pupil (curvature anomaly on LBCr)
    • The rotator angle determines the orientation
    • SAOImage ds9 is useful for simply viewing FITS images
    • astropy has a phenomenal library for processing images using Python
    • Images are 64 pixels in diameter (15 micron pixels)
    • Standing question: How is the analysis affected by how far out of focus you go?
    • Key Point: How you fit each image (centroiding vs. ellipse fitting)
      • Subasa determined that centroiding works much better
      • For this analysis, just take an XYZ centroid
      • Obtain trial image, normalize it, then subtract from basis image
      • Breann suggests writing a routine for centroiding using intensity weighting
    • Japanese version of this system is going on sky in about one week
      • Subasa and Andrew achieved residuals of a couple of arcseconds for convergence times of only a few seconds
      • When seeing is convolved, slows down convergence, but still very accurate, doesn't change final direction
      • Final images aren't nice and symmetrical; wide-field imager, so residual aberrations balance one another
  • Using machine learning for image recognition (aberration recognition) was discussed
    • This could be compared against the SVD sensitivity approach
    • Many training sets may be generated with OpticStudio or with huge LBT data sets
  • Software progress and test scheduling
    • Yang thinks he should be able to produce a software prototype within the next week and wants to schedule a test for later next week (Wed., Thur., or Fri.)
  • Next mountain visit may be possible towards the end of Sept.
    • eTalon sent quote for the large, achromatic collimators (price marked up substantially, $480.00/unit)
  • Still discrepancies between MATLAB and Python Ciddor Calculation
    • When CO2 values deviate from 450 ppm, the refractivity varies substantially
    • Trenton currently digging into Yang's code and MATLAB code to find the problem
    • Yang offered to help find the root cause of the problem once the next TMS test is completed
  • Closed-dome test data (2020_07_16) analyzed
    • Started at Zenith and descended by 10 degrees
    • Funky behavior at specific measurement points
    • Discovered that we dropped all channels in the original XLS output at the 23:45 timestamp
    • Reached out to eTalon about this problem and we believe that we might have an internal consistency error, as the data dropped on all channels simultaneously
  • Suggest having passive TMS running any time the LBC is on

Action Items:

  • Yang will work on the software prototype and be ready for the upcoming TMS test
  • Heejoo to prepare writeup about the state of the splitter box and the dead fibers to obtain a replacement from eTalon
  • Trenton to continue with data analysis, building non-sequential LBT model, and writing image analysis code
Topic revision: r1 - 16 Sep 2020, TrentonBrendel
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback