›

Fundamentals - LDN 183 Astrometric Solution Keeps Failing

Stickfarm8
in PixInsight
Adam,

Happy New Year sir!

I'm finally getting back thru the Fundamentals training and ran into a snag on the LDN 183 problem. The data is somewhat different than what you used in the WBPP Setup and Interactive Normalization video - but I recognized that instead of the bias frames, you supplied Calibration Masters.

I was able to work thru all the issues, including the interactive local normalization, selecting the best frames as you did in the video. Everything seemed to continue well thru Integration. But no matter what I tried, I could not get the Astrometric solution to work. I tried using the Image Solver script on the Master Lights that were generated - with the same no result.

Any suggestions as to what I could be doing wrong would be most appreciated.

Thanks,

Rich
 

Comments

  • Stickfarm8
    Just to clarify - not sure it makes a difference, but the data I was using was from the Fundamentals section, NOT the Fundamental Workflow Examples.

    Rich
  • BlockHeadBlockHead
    Actually that doesn't clarify... please remind me (or show me links) did I make data available in two different places? (I did not think so.)

    Regarding LDN 183... can you show a screenshot of your ImageSolver screen? (What parameters are there for the coordinates, platescale...etc etc?

    -the Blockhead
  • Stickfarm8
    Adam,

    Good morning! 

    The data I was working with initially was from Video 54 in the Fundamentals List - 

    "54. LDN 183 WORKFLOW (You now know enough to follow along in a *real* Workflow session.) Screen shot is attached."

    I did get the Astrometric solution to solve this morning (fresh attack!!). I had remembered in one of the earlier video where someone had a similar problem and you pointed out how the the focal length was incorrect preventing a solution because the calculated plate scale was incorrect. You showed how to use the Aladin site to estimate it. When I look at the FITS header the focal length is 0 (screenshot attached). In PI, there was a value of 700 (default I believe) in the Astrometric solver information that popped up when WBPP ran the Astrometric solver interactively after it could not solve. When I changed the focal length to 500 - everything solved!

    After looking closer at the data FITS header in the second spot, it is the same data Adam. That was my mistake - but it certainly is the same data. But there are two different links to it. Apologies for the confusion Adam.

    But I am still struggling to understand why the data did not populate the information correctly for Astrometric solver to work from the FITS header. 

    With the focal length in the FITS header showing as zero, I thought the other way the information on plate scale could be carried by the FITS header was thru the CDELT1 and CDELT2 values, along with pixel width. But in looking at the FITS headers on the subs, I don't see that entry in the FITS header either.

    I do truly appreciate your emphasis on looking at all the data closely before doing any processing. When I initially ran WBPP on this problem I did chose to do the LN interactively. I now appreciate how valuable that extra effort is.

    Adam, I would appreciate if you would point out what I am missing on the FITS data that should have enabled plate scale to be calculated.

    As always thank you!
     
    Rich
    Screenshot - Fundamentals Workflow Data Links.jpg
    1936 x 1467 - 386K
    Screenshot - Workflow Example - 2nd set of data for LDN183.jpg
    2032 x 1906 - 776K
    Screenshot 3 - Fits Info From Blue sub - PI.jpg
    3321 x 1802 - 853K
    Plate Solver Result on a Gree sub starting window.jpg
    2456 x 1643 - 682K
    Plate Solver Result on a Gree sub.jpg
    2019 x 1386 - 581K
  • BlockHeadBlockHead
    That is the point. You will only find the plate scale in the FITs header IF the person who took the data was using software that saves this information to the images (and furthermore, the person needs to input the correct value). Not everyone has that kind of setup. Part of my lesson is that you should not assume at anytime that the information is correct in the header or that it populated correctly. You should always LOOK at the values in ImageSolver (and any other process) and make certain they pass the sniff test. If you are not certain... you need to think about what the parameters mean. 

    This is why my content stresses the explanation of parameters and concepts over monkey-see-monkey-does approaches. I know people like to  "fail fast" and just try stuff- but there is also merit to having a good think about stuff as well.

    Anyway... looking for some steps to get off my high horse and avoid stepping in the ...well you know what.

    -the Blockhead

  • Stickfarm8
    Adam,

    This is why my content stresses the explanation of parameters and concepts over monkey-see-monkey-does approaches. 

    Amen!
    When I hit it fresh this morning and started digging into the data deeper as I should have done- it all fell into place. 

    I love the training videos Adams. The challenge is with all the details that you present its a LOT of information to absorb in one pass. But that is what makes this training you've put together so valuable. 

    FWIW - I LOVE it when you get on your high horse. It brings some of us back to where we need to be focusing.

    Now I'll stop sucking up....... lol!

    Thanks!
Sign In or Register to comment.