I think PEC worked for me

Discussion in 'General' started by Maneesh Yadav, Jun 7, 2017.

  1. Maneesh Yadav

    Maneesh Yadav Cyanogen Customer

    Joined:
    Jul 12, 2016
    Messages:
    31
    Finally a long clear night to get some testing done. Earlier I was having some trouble with streaky stars at modest exposure times: http://forum.diffractionlimited.com...-pec-something-else-entirely.3283/#post-18486 . Suggestions here helped guide me to the following actions:

    -Further secured cables as best as possible, there was a little play that could have been affecting things
    -Replaced GTOCP2 box with GTOCP3 box
    -Used PEMPro for polar alignment and PEM correction (GTOCP2 box wouldn't work with out-of-the-box playback, GTOCP3 was easy upload)
    -PEMPro suggested correction worked well (I believe *most* of the error was in the PE since the graphs looked similar between the GTOCP2/3 boxes).

    I think the autofocuser might have been off a little but I took multiple exposure including 2 5min exposure of m14 and a 10 minute (last digit of filename). The stars in all images seem round to me (there is no autoguiding). Would anyone mind spot checking the quality of these images? I think the small differences between the two 5 min exposure are not systematic errors in the scope setup.

    Next step will be getting the guiding working, a lot more hope now; but let me know if any obvious problems jump out to you with the images.
     

    Attached Files:

  2. Colin Haig

    Colin Haig Cyanogen Customer

    Joined:
    Oct 27, 2014
    Messages:
    786
    Hi Maneesh, 0015 and 00110 are really good, especially compared to the problems you were having previously. Stars are quite round, and none of the previous problems. I was surprised to see how good 110 is, as its a 10 minute exposure! Well done.

    Image 0025 is trailed slightly, not sure why. If you zoom in to 400% or place the Information window over a star in the upper left, you'll see the flatness is closer to 0 in the other two images.

    Get the guiding going next, it will help.

    After you fix the guiding, take a look at improving the focus.
    The focus could be a tiny bit sharper. Take a look at the FWHM numbers and improve those a bit. You could start imaging with this level of focus, and then come back and work on improving it some more later.

    You've made impressive improvements - congratulations!
    Cheers
    Colin
     
  3. Maneesh Yadav

    Maneesh Yadav Cyanogen Customer

    Joined:
    Jul 12, 2016
    Messages:
    31
    Thanks Colin!

    Yes, I noticed the autofocuser position had changed from what it usually finds....I still don't have a good feel for how much the focus should change day to day, but I think this is a matter of having autofocused on a star that was too dim or too bright. I'll see if being a little more optimistic for my target 1/2 flux changes how the routine works as well.

    In any case, grateful for your assessment!
     
  4. Colin Haig

    Colin Haig Cyanogen Customer

    Joined:
    Oct 27, 2014
    Messages:
    786
    The focus will change a little bit due to thermal effects and mechanical effects.
    As the metal tube and adapters cool down, they will shrink enough to affect focus a bit.
    Mechanically, things may flop around a bit, for example from the east side to the west side of the peer.
     
  5. Maneesh Yadav

    Maneesh Yadav Cyanogen Customer

    Joined:
    Jul 12, 2016
    Messages:
    31
    I think I found my problem with the focus, the target 1/2FD was about the best 1/2FD, when I set it to 2 higher (as it is clearly suggested) the focus changed significantly between sequential autofocus runs. I don't quite understand how the sampling changes the "well" of the V with the target 1/2FD but the minimum was more clearly visible and the star profile was clearly sharper. When target 1/2FD is near the best 1/2FD the well looks more like a wiggly line and it is harder to pick out the minimum.
     
  6. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    2,871
    If you set the 1/2FD to "about the best" then it will not work. The settings has to be above the best focus.

    The minimum is determined from the slopes on either side, not by simply looking for the lowest point, so the points right at the minimum are not important.
     
  7. Maneesh Yadav

    Maneesh Yadav Cyanogen Customer

    Joined:
    Jul 12, 2016
    Messages:
    31
    Indeed, but it still silently *a* best focus with 1/2FD set to "about best". Does autofocus just use target 1/2FD as a cutoff for the scan or does it somehow affect the sampling? My FLI DF2 moved from position 1000 to 1100 (approximately) from two subsequent autofocus runs on the same star in a clear sky. It seemed to be a result of changing the target 1/2D but I suppose it could be due to some other variation.
     
  8. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    2,871
    The Target Half Flux allows the software to reliably determine when the optimum focus region has been scanned. That is used to determine where to stop the scan. It does NOT affect anything else.
     
  9. Maneesh Yadav

    Maneesh Yadav Cyanogen Customer

    Joined:
    Jul 12, 2016
    Messages:
    31
    Ah, then there must be some other reason behind my variation. I'll have to take a closer look next session.

    EDIT: Ah, perhaps I missed your point in that it could be affecting the slope estimate. It is finding a minimum with either setting, but perhaps a poorer one when 1/2FD is too close to "about best".
     
    Last edited: Jun 9, 2017

Share This Page