HETGS Scattering Analysis using ACIS-2C0

• ACIS-2C in 114 row mode -- 2 pixel gap evident
  
• 5 observations: Test IDs G-H2C-SC-88.001 through 88.005
  
• FAM moved in y by 22 mm between observations for some overlap
  Actually required 22.25 mm moves in FAM Y and 0.016 mm in Z for
  consistency (from known W-M wavelength and 0.065 mm shift in Z
  after 4 FAM moves)
  
  

• Used mrdfits to read FITS event data
  
• Read from FITS header: test ID, chip ID (0/1), exposure time,
  TLMAX4 (max readout in chip x), DETNAME (w148c4), and
  window size
  
• Phase, grating, and detector parsed from test ID
  
  

• TG period, dispersion angle, Rowland distance
  Taken from Dan Dewey's (early) analysis of phase 1 data
  Update should have only a minor effect
  
• Mirror focus distance
  Values from in-tank measurements (TRW)
  
• Chip scale and DN to eV conversion constants assumed
  Scale set at 24 microns, estimated 0.32 eV/DN for w148c4
  
• Other parameters determined from 88.00x data:
  Beam FOA in chip coordinates, rotation
  Measured FOA position using 0th order: (584, 72.8)
  Measured rotation required for horizontal spectra: 0.6°
  
  

1. Set event PH to sum of 25 pixel PHs
   "ASCA" grading criterion computed from 3×3 but not used
   Many cases of 2nd event in 5×5 window frame
   This figure shows the pulse height distribution for the entire 5 image data set. The input is monochromatic at 1.775 keV and background is negligible. Thus, the pileup peaks are apparent and there is a tail of incomplete charge collection events below 1 keV. The ACIS-2C0 gain parameters vary significantly over the entire array, so peaks are broad.
2. Randomized events by ± 0.5 pixel before rotations applied
   
3. Offset by FOA loc, then rotated 0.6° from Chip XY to XRCF YZ
   
4. Offset by FAM shift; coordinates now in pixels from 0th order
   
5. Rotated by grating dispersion angle (HEG: -5.2°, MEG: 4.7°)
   New coordinates are dispersion and cross-dispersion distances
   NOTE: Rotation about (virtual) 0th order location, not chip FOA,
   so the rotations are distinguishable (as with HSI)
   
6. Coordinates are scaled to physical units (mm)
   
7. "Unwrapped" 2C to center dispersion line
   Depend only on cross-dispersion (xd) distance
   Result is modulo W (= 114*0.024 mm) so that -W/2 < xd < W/2
   

   
   Grayscale image showing all the data combined after unwrapping. The dispersed orders show up as bright vertical lines because the horizontal scale is compressed. Holes in the middle of the images result from event selection to avoid pileup. Click on the image for more complete annotation and a larger version which is stretched to show scattering better.
   
   

8. Set wavelength based on grating equation assuming sin phi = tan phi
   
9. Converted xd values to angles (in arcsec) using mirror focal length
   
10. Determined exposure for each wavelength bin
    Set up dummy events at corners
    Sent events through same transformations as photon events
    Added chip exposure time to allowed wavelength bins
    
11. Accumulated exposures, events into encompassing arrays
    Exposure function gives the resultant exposure time as a function of wavelength, adding all 5 chips.
    
    

• MEG shows no detectable scattering
  
• Fitted Gaussians with variable location, width, background, and normalization to cross dispersion profiles at each wavelength bin for HEG scattered light only. Some comments and results from the results :
  • The top curve gives the centroid location in arc seconds from a fiducial horizontal reference. Note that the centroid drifts downward by about 1.0 arcsec as the wavelength increases from 5 Å to 21 Å. This drift is corrected with a 0.016 mm Z shift for each 22 mm FAM motion. Could be caused by a diff. of FAM and XRCF Y axes of 0.04°
    
  • The middle curve gives the inferred total counts in the Gaussian; scattering peaks appear near 7 Å, 14 Å, and 21.2 Å. Bad fits occurred at the HEG and MEG orders. These should have been at 3.5 Å (or 7.0 Å for the MEG +1), 7.0 Å (HEG +1), 10.5 Å (MEG +3), 14.0 Å (HEG +2, MEG +4), etc. At HEG +3, the line is observed at 20.5 Å, compared to the expected value of 21.0 Å, thus wavelengths of orders showed systematic error. The wavelength errors are eliminated by
    assuming that the FAM moved 22.25 mm on each step, 1.1% too high. (The FAM was commanded to move 22.000 mm and FAM log data agree.)
    
  • The bottom curve gives the fitted Gaussian sigma parameter; good fits give values near 0.5 arc sec (about 2 ACIS pixels, as expected).
    
  • Fits have been eliminated where unreasonable centroids, widths, or intensities were obtained, causing gaps in the plots. Some bad data points remain, however, and have to be reviewed individually.
    
• Sawtooth location pattern revealed ACIS-2C rotation of 0.6°
  
• "Source" aperture defined as 2.5 arcsec (5 pix) wide
  
• Estimated background as mean of region 11-26 arcsec above src
  This figure shows the estimated background and the counts in the source region. Note that the "background" increases to 0 Å due to (mirror) scattering from 0th order. The orders show up in background due to mirror scattering and when the MEG image does not land on top of the HEG image; e.g., at 10.5 Å and 17.5 Å. (Note wavelengths are corrected.)
• Subtracted BG and normalized to expected rate in +1 of 1700 cps
  This figure shows the net scattered light after dividing by the estimated exposure time. There are still bad background subtraction errors near the MEG and HEG orders.
• Added points at +1, +2, +3 orders affected by pileup
  This figure shows the result. Some bad background subtraction results have been eliminated. Error bars (in positive direction only) have been added to show that the "detection limit" is about 1e-5 of total power per 0.05 Å bin.
  
• The integrated power in "scattered" light is about 1% of the total in first order.
  

• Still ongoing, but peaks off of orders are probably telltale features