Staff Assigned: Brian Brown Beta Testing Aluma with AO

Discussion in 'Aluma CCD Series' started by Greg Crawford, Oct 25, 2019.

  1. Greg Crawford

    Greg Crawford Cyanogen Customer

    Joined:
    Oct 11, 2014
    Messages:
    101
    Thanks Colin! I've installed the firmware on the camera and updated MaxImDL. It now says "Beta 0". I began testing the AO tonight, but unfortunately it came over overcast before I could try it. I also think Doug posted saying the wireless situation would becoming later and would require an SD card to transfer the firmware to the camera. I'm sure I've read about the method in the manual. Once I get a clear night I'll send reports to you and Adam.
     
  2. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    9,932
    Appreciate any feedback you can provide. I haven’t been able to sky test the latest because of clouds, but it was getting very close to ready at the previous version so I’m hopeful it’s good to go.
     
  3. Greg Crawford

    Greg Crawford Cyanogen Customer

    Joined:
    Oct 11, 2014
    Messages:
    101
    I tried out AO8 guiding tonight with some widely varying results. Initially the results were good. However, as the night wore on things deteriorated. Some of this may have been the results of sky haze, so I don't want to report them as fact yet. One thing that did seem consistent was the appearance of a "ghost guide star" in the "Locate" window. When a potential guide star entered the field of view and I clicked on "Locate", the program would sometimes place a box around what appeared to be another star close to the border of the frame. I experimented with rotation of the camera, and movement to a different target. Frequently, although not every time, a guide star would always be identified by the program at position x=150, y=226. To my mind this was suspicious. If it had been a real star, I imagined it would at least move around on the chip with rotation and change of target. Consequently I am calling it the "ghost star" :). I suspect that problems that occurred as the night wore on may have been caused by the program jumping to track on the ghost star. The problem was that despite calibration working for both drive and AO calibration, while guiding, the star would eventually move away without correction. To show that my ghost star is not a Halloween hoax, I have uploaded a file showing what it looks like.
     

    Attached Files:

  4. Colin Haig

    Colin Haig Staff Member

    Joined:
    Oct 27, 2014
    Messages:
    7,379
    Location:
    Earth
    Thanks for the report. We appreciate it.

    It might be helpful if you save the autoguider images, and take a look to see if the "ghost star" is actually in any of the saved AG images.
    Also, enable the tracking log.
    They are both under the Camera Control, Guide tab, Options [>] button.
    Here's the tracking images option enabled. The track log is also on this menu.
    recordtracking.png
    Then we can have a look at the actual images downloaded by Camera 2, and see if the "ghost star" is there.
    The track log file will also help us evaluate what went on during the process.

    The guider tracking images will show up somewhere like:
    C:\Users\Greg\Documents\MaxIm DL 6\Settings\GuiderTrackingImages

    The Track.log file will be somewhere like:
    C:\Users\Greg\Documents\MaxIm DL 6

    It would be great if you could share that.
    Thanks
    Colin
     
  5. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    9,932
    That's a great report! I'll ask @Brian Brown to look into the ghost.

    Also I'll move this to a separate thread so it will be less confusing!
     
  6. Greg Crawford

    Greg Crawford Cyanogen Customer

    Joined:
    Oct 11, 2014
    Messages:
    101
    OK, following the directions Colin gave I turned on both the option to save tracking images and the option to save the tracking log. I am not practised at trying to coordinate the log with actions, so I'll upload a few of the log files and hope that they cover the time period. If not let me know and I'll try again. However, the snip of the screenshot should tell it all. (See Guiding 2019-10-28.png uploaded.) In that you can see both the Locate window and the tracking image for that window downloaded. The Locate window completely ignores the obvious star and fixates on a non-existent star at the position x=150, y=226. This was the same unchanging location on the chip as the "ghost star". I kept increasing the exposure time to 15 seconds and an even fainter star started to appear to the left of the obvious guide star. As you can see from the graphic uploaded, the ghost star has a brightness of 8. When I clicked on the real star it showed a brightness of zero.

    Now earlier in the night it had been possible to locate the real star in the guider image and to calibrate using it. However, when I initiated the command to track, the star did not appear in the tracking window. At this stage in the proceedings the real guide star gave a count of over 1,000 in the Brightness window. As the night wore on and became cloudier, I set the Locate command to "Continuous". Every so many downloads of the tracking image I would see the locate box rest on various positions in the image, but there were no stars in the downloaded image at that position. I wondered whether hot pixels were fooling the detection procedure, which brings me to the question of whether the tracking images were being dark-subtracted, an option which I had selected. The program went through the motions of creating a dark master for the guide chip, but whether they were applied, I'm not so sure. The last tracking log file, 20191028.log should give results towards the end of my experiment.

    I suppose this goes back to me finding saved tracking images and checking the FITs header. They were not in the location Colin suggested, so I'll keep looking on the observatory computer.
     

    Attached Files:

  7. Brian Brown

    Brian Brown Standard User

    Joined:
    May 8, 2019
    Messages:
    113
    Hi Greg,

    I'm sorry to say that the images are not going to be found in that location. The version you currently have is not yet configured for image saving but I will PM you an updated version along with some instructions.

    -Brian.
     
  8. Brian Brown

    Brian Brown Standard User

    Joined:
    May 8, 2019
    Messages:
    113
    Hi again,

    While looking into this phantom star phenomenon I had come to realize that there was a small graphics bug in the AO display that was filling in the red circle drawn on the screen which created a "fake star" in the event that it latches onto a hot pixel. This explains the discrepancy between MaxIm and the AO display. I am still running some tests on the plug-in but I will get you an updated beta version asap.

    -Brian.
     

Share This Page