›

Photometric Colour Calibration failure to Plate solve with M51

Nihal Mirza
in PixInsight
Hi Adam,

I wondered if you could shed some light on this error (see attached file for error log).

Basically I'm following your workflow example for M51 (however I am using my data, see below link to RGB file), PCC fails to plate solve (as far as I can tell).

I'm using the local catalogs, for GAIA and APASS. 

Having watched the video on youtube (where you mentioned about M51 being on the edge in one of the MAPS) as well as the bug you found while selecting Gaia, I'm not sure how to proceed, tbh they were about photometry rather than the plate solve issue.

I haven't used the data provided as part of the workflow but might give that a shot.

ps: I've tried changing limit magnitude as well as checked  the "force plate solve", even reduced back ground neutralization to 0.02. No luck.

M51_RGB.xisf

txt
txt
PCC_error.txt
13K

Comments

  • BlockHeadBlockHead
    You need to give me access to the file (share with anyone that has the link).
    Typically when people have this problem it is because their acquisition software is inserting garbage information in to the header. You should try to ignore the metadata option and input the information correctly.


    -the Blockhead
  • Nihal Mirza
    Hi Adam,

    Thanks for the hint.

    After doing some research on the ZWO acquisition software (ASIAIR), it does seem that ASIAIR is including some additional info in the FITS header that causes PCC to fail.


    How do I tell PCC to ignore metadata, I don't see any check box. Unless manual input for Focal length and pixel size essentially means that PCC will ignore meta data.

    ps: I also tried ImageSolver, so surely it doesn't plate solve (was too tired to try last night).

    Files shared via Google Drive (hopefully the link works):
    1. M51_RGB.xisf
    2. Single frame Luminance Fits file
    3. Same Lum as 2) but after nsg (xisf)


  • Nihal Mirza
    Forgot to mention, I'm reading through the PI thread to understand how to overcome this issue without going back to the original FITs file and changing FITs header :(
  • Nihal Mirza
    So I tried a few things and got success.

    Process I followed:

    WBPP > NSG > Master R/G/B file > ColorCombination > ImageSolver

    In Image solver rather than using the object co-ordinates from metadata, I requested Image solver to get Object data by selecting "Search" (some database in Germany)

    RGB - ImageSolver - OK
    PCC > OK

    Happy days.... :)

    However I don't know why when I do the same "Search Co-ordinates" for M51 in PCC (uses the same database) and manually add Focal length and pixel size PCC fails to plate solve.

    Maybe ImageSolver is doing things slightly differently?

  • BlockHeadBlockHead
    Yes, Image Solver will put a correct astrometric solution in the header- this is why it works.
    It is the astrometric solution which is a NON solution that the ASIAIR is putting in the header. This is absolutely unforgivable. The company should be taken to task for this.

    Note the value that ImagerSolver calculated for your platescale.It is good to know.

    I circled the checkbox for you for ignoring the metadata- but I think that PixInsight does not view an astrometric solution as metadata- i.e. it will not ignore it. So it takes in the garbage that ASI gives it.As you saw in the thread- people are advocating for a warning. Using ImageSolver to calculate a new and correct astrometric solution and overwrite the garbage is fine. It is also easy enough to write a small script to remove the garbage that ASI is putting in the header... but that is beyond non-programmers.

    -the Blockhead
    Capture.JPG
    538 x 1010 - 108K
  • Nihal Mirza
    Hi Adam, 

    I think I need another eye test, I totally missed that..i.e. what you circled.

    Imho ZWO is quite good in fixing issues but it takes a whole lot of shouting by a number of people before they act on an issue.

    Hopefully with their new beta version of ASIAIR, the FITS header issue will be fixed.
  • BlockHeadBlockHead
    Indeed, that is what the word on the street is...

Sign In or Register to comment.