/
Nightly Observing Sequence

Nightly Observing Sequence


OSELOTS operates relatively autonomously during a night.  About ten minutes of preparation are required, plus a brief stop to remove the lens cap after darks are done. 

If you have time the day before the observing run, you should consider following these steps through taking reference wavelength data. That way, if there is an issue with the slit-fiber alignment (the current most common problem with OSELOTS), repairs can be performed during the day pre-observing. 

The setup sequence is as follows, with pictures included for reference: 

Approximate locations of components are labeled in this photo: 

Setup Procedure

  • Make sure powerstrip for OSELOTS is turned on
    • (see image below)


  • Turn on AC-DC adapter for the camera
    • From O to | 
    • (see image below) 
    • AC-DC adapter should be turned off after each observing run 


  • Make sure that OSELOTS control computer is powered on.   
    • Under normal operating conditions, the computer is left on. 
    • (see image below) 

  • Log onto OSELOTS control computer
    • Username is: stubbslab, password is normal password for group computers
    • Can be done remotely via: 
      • ssh stubbslab@auxtel-sl01.cp.lsst.org
  • Open the terminal app on the OSELOTS control computer (shortcut on Desktop)  
  • move to the ~/Documents/OSELOTS/ directory 

    • cd ~/Documents/OSELOTS/
  • Enter superuser mode: 
    • $ su 
    • (password is same as accessing the computer over ssh)  
    • NOTE: All camera acquisition commands must be run as super user.  Simply running commands as sudo WILL NOT WORK.  You MUST enter su mode, and stay there to take your images. 
  • Run a quick execution of the core data taking script, doPixisImaging.bash  
    • bash doPixisImaging.bash -e 0 -o Cool -n 1 -h 1
    • OPTIONAL: you may also provide the -d command line option, to specify the date string (will be used to make the data repository and appended to file names) 
    • Example: 
      • bash doPixisImaging.bash -e 0 -o Cool -n 1 -h 1 -d 2022_06_28
    • If no -d option is provided, code will choose present date, rolling over at 1600h UTC. 
    • This will take one 0 s exposure (not a good bias, because the camera is not cooled).  This accomplishes the following:
      • starts cooling the camera
      • initializes stage with knowledge of home position 
      • Builds the directory for the night 
      • Generally makes sure all system components are working together 
  • Make sure the code executes without error 
  • Check that the directory was created, and that a file exists (replace the date below with the appropriate date, formatted YYYY_MM_DD): 
    • ls /home/sashab/Documents/PIXISData/2022_06_09/
    • If this is the first time you ran the code for this night, you should see two files: the new image file and a plaintext tally file: 
      • image_tally.txt                Cool_2022_06_09_1.fits         
  •   (OPTIONAL, BUT SHOULD DO IF POSSIBLE) While the camera cools, set up for wavelength calibration observations  
    • Find a reference wavelength source, stored in the OSELOTS toolkit (see photo above)
      • The default source is the HG-2 Ocean Optics arclamp (see photo below): 

    • Connect the OSELOTS optical fiber to the reference light source
      • First disconnect the fiber from the lens mounted to the telescope
      • Then, EITHER:
        • place the light source on a stand so that it can reach the fiber 
      • OR: 
        • feed the fiber through the cable ties that hold it in place until it can reach the light source (see photo below)
        • IF YOU EXTEND THE FIBER FEED, YOU MUST RETURN IT TO ITS PREVIOUS LENGTH.  
          • Black marks on the fiber next to the two cable ties in the photo below indicate where along the fiber length the cable ties should be positioned
            • If you cannot find the black marks, make them yourself with a black sharpie BEFORE extending the cable length

    • Now that the cable and calibration light source are connected, we're ready to take some reference exposures
      • Turn on the calibration light
      • Run the following command on the OSELOTS control computer (remember: you must be logged in as su): 
        • bash doPixisImaging.bash -e 5 -o HG2 -p HG2_f23p0 -n 1 -h 0 -f 23.0
        • This will take one 5 ms exposure at focus position 23.0 mm 

      • Check that the saved data looks as it should: 
        • Recommend using ds9, which is installed on the OSELOTS control laptop and can be invoked from the command line: 
          • cd /home/sashab/Documents/PIXISData/2022_06_09/
          • ds9 HG2_f23p0_2022_06_09_2.fits 
        • If working remotely, you will need to copy the data to your local machine: 
          • scp stubbslab@auxtel-sl01.cp.lsst.org:/home/sashab/Documents/PIXISData/2022_06_09/HG2_f23p0_2022_06_09_2.fits . 
        • If the system is working and the focus is good, you should see a series of lines, with those at the center most in focus, with the gaps of the fiber visible in the lines (see left panel in figure below; sample ds9 file also included for reference)
        • If the image appears out of focus (see middle panel in figure below; sample ds9 file also include), then repeat this measurement, moving the -f command line value until good focus is achieved. 
          • bash doPixisImaging.bash -e 5 -o HG2 -p HG2_f23p0 -n 1 -h 0 -f 22.9
          • bash doPixisImaging.bash -e 5 -o HG2 -p HG2_f23p0 -n 1 -h 0 -f 23.1
          • .... 
        • If the lines are ragged and well below full well of ~50k ADU (see right panel in figure below; sample ds9 file included for reference), then the slit and fiber have slipped out of alignment.  Fixing this takes ~ 1 hour.  See the process in the trouble shooting page: OSELOTS Troubleshooting

 


      • Once you are satisfied that the data looks as it should (ideally like panel on left), take 5 exposures at that focus position: 
        • bash doPixisImaging.bash -e 5 -o HG2 -p HG2_f23p0 -n 5 -h
          0 -f 23.0
    • Once the images are acquired, return the system to its normal state: 
      • Disconnect the fiber from the light source
      • Turn off and store the light source
      • Connect the fiber to the lens 
      • IF THE FIBER LENGTH WAS EXTENDED, retract fiber through cable ties until it is back in its base position (black marks adjacent to two photographed cable ties) 
  • We are now ready to take darks
    • (Good time to run these is while observers are at dinner)  
    • Turn off dome lights, cover OSELOTS with black sheet 
    • Light conditions should be as close to true observing conditions as possible
    • MAKE SURE LENS CAP IS STILL ON
    • Take five or six 300 s dark exposures, interwoven with bias images.  
      • Remember your best focus position  
      • for i in {1..5}; do     bash doPixisImaging.bash -e 0 -o Bias -n 1 -h 0 -f 23.0. ;     sleep 5;     bash doPixisImaging.bash -e 300000 -o Dark -n 1 -p Dark_f23p0 -f 23.0 -h 0 ;     sleep 5;      done
    • Dark images should look something like this, when a Gaussian smoothing is applied: 

  • We are now ready to go on sky! 
    • Remove the lens cap on the OSELOTS lens mounted on the telescope
    • Begin your 300 s observations, interleaving biases:
      • Approximate the number of images by the amount of time until dawn, divided by 5 minutes.  The program will stop at 1159h UTC, regardless. 

      • Example: starting exposures at 2000h CLT, sunrise at 0600h CLT, we take 60 X 10 / 5 = 120 exposures: 

        • for i in {1..120}; do     bash doPixisImaging.bash -e 0 -o Bias -n 1 -h 0 -f 23.0 ;     sleep 5;     bash doPixisImaging.bash -e 300000 -o Sky -n 1 -p Sky_f23p0 -f 23.0 -h 0 ;     sleep 5;       done
    • Check the first exposure (obviously dome needs to be open) to verify sky lines are visible (see image below; raw fits file also included for reference)
      • After this check, system can operate autonomously for the rest of the night 

 

  • End of night: 
    • At the end of the night, first thing tomorrow morning, the observers should: 
      • Recap the imaging lens
      • Turn off the camera power (the black AC-DC adapter imaged above)
      • The powerstrip for the other components can be left on


Copyright © 2024 The President and Fellows of Harvard College * Accessibility * Support * Request Access * Terms of Use