Page 1 of 4 1234 LastLast
Results 1 to 10 of 37

Hybrid View

  1. #1
    Join Date
    Apr 2008
    Location
    Central Virginia, USA
    Posts
    753

    Default Plate solve failed

    Tonight, a very clear night with very good seeing, was an excellent opportunity to image the Tadpoles again. I started my plan as normal, but after slewing to the target, ACP could not plate solve the pointing image. It slewed to an offset and took another pointing image, but that plate solve failed as well, so ACP parked the scope and closed the roof. I waited a while, then started everything again, and again ACP could not plate solve the pointing image.

    Does this have anything to to with the time change back to Eastern Standard Time? I double-checked my PC clock, and it's correct, and I used the ASCOM driver to set the AP1200 mount's clock to the PC clock, so everything looks correct. The ASCOM driver shows the time as 21:50 (correct) and the time zone as -5 (correct). ACP shows the UTC time as 02:58, (also correct ).

    The coordinates in my plan are 05:22:29.11 33° 28' 24.8", and ACP displays these coordinates. The ASCOM driver's coordinates are very close, at 05:23:50 33 30 08.

    Everything worked perfectly three nights ago on the Cocoon Nebula. The only thing that's changed since then is the time change. I thought I didn't have to worry about that once I set up ACP to honor automatic time changes. Did i miss something?

    I've attached a screen shot of the PC just after ACP downloaded the pointing image on the second run, and the log from that same run.

    Please help! Thanks.

    --- Mike
    Attached Images Attached Images
    Attached Files Attached Files

  2. #2
    Join Date
    Nov 2015
    Location
    Christchurch, Dorset, United Kingdom
    Posts
    362

    Default Verifying plate solve images after the event

    Hi Mike.

    Is (are) the failed plate solve image(s) from that run still in the pointing exposures folder of your image directory?

    You might be able to glean more information by submitting those to visual Pinpoint or online at astrometry.net and then compare the solved plate(s) to where the telescope thinks it is pointing.

    William.

  3. #3
    Join Date
    Oct 2005
    Location
    Mesa, AZ
    Posts
    33,368

    Default

    William is right, the most likely culprit for "all of sudden it won't plate solve" is pointing. See if you can find the pointing exposure and solve it in Visual PinPoint with the All-Sky option. The resulting VPP will log the pointing error.
    -- Bob

  4. #4
    Join Date
    Apr 2008
    Location
    Central Virginia, USA
    Posts
    753

    Default

    Quote Originally Posted by Bob Denny View Post
    William is right, the most likely culprit for "all of sudden it won't plate solve" is pointing. See if you can find the pointing exposure and solve it in Visual PinPoint with the All-Sky option. The resulting VPP will log the pointing error.
    There are no pointing exposures in my images directory. How do I tell ACP to save them and where to put them? That would help for future plate-solve failures. For now, I suppose I can slew somewhere and take a 10s bin4 exposure, then use Visual PinPoint to solve it.

    The All-Sky option in ACP's PinPoint Preferences tab is unchecked.

    What would cause a pointing error? Pointing was fine three nights ago; the maximum pointing error in that night's log is 7.4 arcmin. Nothing changed in the observatory since them. Last night I connected APJog (see *** note below) to the mount, and unparked from the previous run. I then connected SkyX to the mount, and slewed to a star near my target focus in FocuxMax before ACP's first plate solve. The star (or some star) was centered in the camera's FOV. Finally, I connected ACP to the mount.

    *** APJog explanation: I connect the APJog utility to the mount to keep the ASCOM driver connected even if other apps disconnect the mount. This is because I use ASCOM Direct guiding in MaxIm. Sometimes during setup I need to disconnect the cameras, and that also disconnects the mount and closes the ASCOM driver. The APJog utility connection avoids this.

    I just checked the ASCOM driver, and the mount's coordinates are precisely what they should be. Plus the local time and time zone are correct, as they were last night.

    It might be clear enough tonight to check pointing on M33. I've imaged this often enough to know how it's framed after a slew to it.

    I surely don't understand how/why pointing should have changed three nights after a successful run. No Windows 10 updates were installed since then, and nobody has been in the observatory.

    --- Mike

  5. #5
    Join Date
    Oct 2005
    Location
    Mesa, AZ
    Posts
    33,368

    Default

    I was only guessing why solving started failing. Or is this a ‘sample of one’? You may notice that ACP tried to solve the pointing exposure, then it stepped off 2 fields of view and tried again. So it failed to solve two completely different fields. It does this because the area around extended objects (galaxies, star clusters) often lacks catalog stars and thus makes solving difficult.

    I’d check pointing as you suggested. ACP has a FindLostScope script that is harmless to run and will make it the process automatic.
    -- Bob

  6. #6
    Join Date
    Apr 2008
    Location
    Central Virginia, USA
    Posts
    753

    Default

    Quote Originally Posted by Bob Denny View Post
    I was only guessing why solving started failing. Or is this a ‘sample of one’? You may notice that ACP tried to solve the pointing exposure, then it stepped off 2 fields of view and tried again. So it failed to solve two completely different fields. It does this because the area around extended objects (galaxies, star clusters) often lacks catalog stars and thus makes solving difficult.
    It's not a sample of one. The same thing happened twice. Pointing failed on the first run, so I shut down ACP and everything else, then started over. The plate solve failed on the second run as well. The target was IC410, and there seem to be plenty of stars in the region. ACP's log shows:
    02:58:49 642 image stars found
    02:58:49 1029 catalog stars found


    This seems about right. ACP logs for other targets in the same area show similar numbers.

    Quote Originally Posted by Bob Denny View Post
    I’d check pointing as you suggested. ACP has a FindLostScope script that is harmless to run and will make it the process automatic.
    I'll see what I can do tonight.

    Maybe I should run my known-good M33 plan, and see if plate solving works or fails.

    Regarding ACP saving pointing images, how do I make that happen?

    --- Mike

  7. #7
    Join Date
    Oct 2005
    Location
    Mesa, AZ
    Posts
    33,368

    Default

    The pointing exposures are saved in [My] Documents\ACP Astronomy\Images\PointingExps with names of p-targetname.fts
    -- Bob

  8. #8
    Join Date
    Apr 2008
    Location
    Central Virginia, USA
    Posts
    753

    Default

    Something clearly is wrong here. Slewing to M33 didn't put it on the sensor, so I started the Tadpoles plan again, and captured the coordinates while ACP was exposing the pointing image. I've attached a screen shot of that. The red-bordered rectangle is the plan's specified coordinates. Those, ACP's coordinates, and the ASCOM driver's coordinates match fairly well.

    But after the plate solve failed and ACP shut down the observatory, I looked at the FITS header of the pointing image. The RA was good but the Dec was off by 2 degrees! From the header: DEC = '+35 41 55.2'

    How can this be? Doesn't ACP supply coordinates to MaxIm to write into the FITS header? How can ACP display a correct DEC of 33 degrees, but the FITS header has 35 degrees?

    I don't understand how the coordinates in the plan, the ASCOM driver, and ACP can agree so closely, while the FITS header is so wrong. I have attached a text file with the full contents of the FITS header.

    Thanks for any help.

    --- Mike
    Attached Images Attached Images
    Attached Files Attached Files

  9. #9
    Join Date
    Nov 2015
    Location
    Christchurch, Dorset, United Kingdom
    Posts
    362

    Default Plate solved coordinates don’t match requested target coordinates.

    Quote Originally Posted by Mike Dodd View Post
    Something clearly is wrong here.

    I don't understand how the coordinates in the plan, the ASCOM driver, and ACP can agree so closely, while the FITS header is so wrong. I have attached a text file with the full contents of the FITS header.

    Thanks for any help.

    --- Mike
    Hello Mike.

    If I understand things correctly….

    The RA and DEC coordinates that ACP and the mount controller display panel show you are the assumed coordinates that the telescope is pointing at and these are the coordinates specified by your plan. ACP told the mount to point there and that is what it expected the mount to do.

    The .fits header of the solved pointing exposure shows the computed coordinates derived from processing the pointing exposure in Pinpoint and then ACP writes that information into the header of that pointing image, the header shows the solved coordinates not the requested coordinates.

    Assuming that the Pinpoint solve was correct and there is not an obscure error in the star catalogue used for the solve computation then it would appear that the mount is not pointing where it should.

    If your mount uses an internal pointing model it may have become corrupted, or, if it has a homing sensor that provides an absolute mechanical reference angle that sensor may have moved or become defective and although unlikely with a high end mount the OTA might have moved and no longer be perpendicular to the RA and DEC axis.

    I think if I were looking at this I would double check the pointing exposure computation by manually stripping out the .fits coordinates from a copy of the pointing exposure header, or converting it to a .tiff format, and upload it to astrometry.net for a blind solve and then compare the computed results.

    If ACP/Pinpoint and astrometry.net both compute the same coordinates but ACP’s control panel and the mount control panel both show the same requested coordinates were somewhere different to the solved pointing image then that has got to be a mount/telescope, mechanical, software, electronic or optical problem.

    If ACP’s and astrometry.net’s solved coordinates differ in DEC by the the error you have discovered then that might be a catalogue error or detection parameter setting that is a incorrect in Pinpoint but you are reporting a failure on multiple targets? which makes this unlikely and points to a problem with the mount or OTA in my mind.

    My interpretation may be wrong, I’m nowhere near experienced enough with ACP but that is how I read the issue.

    William.

  10. #10
    Join Date
    Apr 2008
    Location
    Central Virginia, USA
    Posts
    753

    Default

    William.... ACP did not solve the pointing image. The log says:
    03:27:27 Plate-solve pointing image.
    03:27:27 709 image stars found
    03:27:27 669 catalog stars found
    03:28:06 ** No matching stars found. Check your estimated center-point RA/Dec, and your image scaling and quality.


    I don't know when the FITS header was written, but ACP says it didn't solve the plate.

    --- Mike

 

 

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Failed to Solve Plate Issues.
    By David Chisholm in forum Pre-Sales Technical Questions and Help
    Replies: 4
    Last Post: Jun 22, 2017, 17:13

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •