PDI:Observing CHARIS FPDI

CHARIS and NIR (F)PDI

• Port from the google doc page "PDI Observing Procedure"
• See also for synchronization with VAMPIRES: VAMPIRES wait for CHARIS.
• Trying to distinguish FPDI with and without FLC (using FLC and nFLC below).

Get prepared

• Waveplate unit

  • Open the waveplate unit VNC (vncviewer -shared garde.sum.naoj.org:2).
  • Start the WPU controller with garde$ obsSTARTgen2sum.
  • Move the HWP in the beam.
  • (Cals) Follow the calibration procedure.
  • (Obs) Enable ADI sync.

• Coronagraphy

  • Identical to non-PDI modes.

• Astrogrid

  • Binning should be 3 or 4 or the satellite spots will not clear the field stop in K band.
  • TBC if binning 2 clears the field stop in H for NIR-only observing.

• Polarimetry

  • Field stop: field_stop 0; fine tune with field_stop <x|y> goto <value_in_mm>.
  • [CHARIS] charis_wollaston
  • [FPDI] ircam_wollaston
  • [FLC] ircam_flc in

• HWP Synchronization (more below)

  • Kill any pre-existing instance of the HWP sync script: pgrep -lf hwpsync.
  • Choose every how-many CHARIS frames we want to switch the HWP. Doesn't matter to CHARIS, but switching too often reduces slightly the VAMPIRES and FPDI duty cycle. 1 minute max.
  • Spin up hwpsync with the appropriate options -c, -f and -v.

• [FPDI] Adjust Apapane exposure / NDR / gain so it looks nice.

• [FLC] Switch Apapane to external trigger. The display will freeze.

• [FLC] Start the FLC with flc_trigger start <Apapane NDR>. The display should unfreeze.

Acquisition

• Start CHARIS frames in gen2
• Start CRED1 archiving with Ctrl+Shift+S.

Sanity checks.

• Is the HWP moving?: Is the offset position updated on garde?
• Am I getting the CHARIS syncs: hwpsync should be verbose on that.
• Is the FLC working?: Put in the polarizer, it should blip left and right.
• Is the HWP position reported?: the following results should change when the HWP moves.
  • scxkw-getsetpm x_BHWPAG (lowercase x)
  • scxkw-getsetpm RET-ANG1
  • (scexao6) ls $(ls -d /mnt/tier1/ARCHIVED_DATA/*/ | tail -n 1)/kcam/*.fits | tail -n 1 | xargs -I {} fitsheader {} | sort | grep RET-ANG1. This last one ensures the WP angle makes it all the way to archived FITS files.

HWP synchronization script

A single script handles HWP syncronization between CHARIS, NIR, and VAMPIRES. It should be started with the appropriate flags and will handle the rest.

Full path to script ~/src/software-main/bin/hwpsync.

Usage:
    hwpsync <num> [-v] [-c] [-f] [-p <p>] [-d]

Options:
    -h, --help  Print this
    <num>       Number of CHARIS frames per position
    -v          Vampires PDI
    -c          CHARIS PDI
    -f          Fast PDI
    -p <p>      Port number [default: 18448]
    -d          Debug mode - no commands sent to Vampires/IRCS

🔧 More technical:

• This will spin up a logfile parser on the CHARIS computer (and kill a pre-existing instance). The source code for this parser is in swmain/hwp/charis-feeder.
• charis-feeder lets hwpsync know CHARIS has finished an exposure.
• If it's been <num> frames:
  • If -f (NIR PDI): stop the FPDI FITS cube. Restart it after HWP movement. This ensures clean cubes per HWP position.
  • If -v: forward the trigger to VAMPIRES and don't move the HWP.
  • If not -c: CHARIS will not trigger for the HWP. This is replaced by a timer of 60 seconds.
• Re-fetch the HWP position from garde, upload it to Redis, mark it in the FPDI SHM so we don't have 10-sec Redis round-trip delay in the FITS files.

Get done

• Wollastons out; field_stop out
• [FLC] FLC out, don't forget flc_trigger stop.
• Retract the WPU on the garde computer.