Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

ASTAP fails plate solving to 1arcmin

  • Please log in to reply
20 replies to this topic

#1 MikeECha

MikeECha

    Ranger 4

  • *****
  • topic starter
  • Posts: 358
  • Joined: 15 Sep 2018
  • Loc: Charlotte, NC

Posted 17 June 2021 - 10:44 PM

Hello there

 

I hope to find help on this issue that is driving me crazy. I am trying to image the Iris Nebula and NINA plate solves until the bright star in the nebula is almost at the center of the image but it then fails to plate solve from there on.

 

Fl 1365 f9 at  .36 arcsec/pix with 2.4micron pixels

H18 installed in ASTAP

Long and short exposures (15 to 240sec)  - fails

 

Any idea what to try next?

 



#2 Jim Waters

Jim Waters

    Fly Me to the Moon

  • *****
  • Posts: 5,877
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 17 June 2021 - 11:07 PM

Any error messages?  Are you at a light polluted site?  Is the F.L. and pixel size correct?  Are ASTAP updates installed?


Edited by Jim Waters, 17 June 2021 - 11:18 PM.


#3 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 25,071
  • Joined: 27 Oct 2014

Posted 17 June 2021 - 11:34 PM

Diagnosis suggestion.  Manual platesolving with PlateSolve2 and hand correction.  If that works, the problem is NINA, not the platesolving program.

 

I think it's better to get things going "manually", it makes troubleshooting the fancy image suite easier.

 

Minor point, may or may not be related.  The 183 is a niche camera, and the niche is short fast scopes.  I tried mine (purchased for a C8 RASA) with my 130mm F7.  The results were bad.



#4 Jim Waters

Jim Waters

    Fly Me to the Moon

  • *****
  • Posts: 5,877
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 17 June 2021 - 11:51 PM

Manually use ASTAP to plate solve.



#5 TelescopeGreg

TelescopeGreg

    Gemini

  • -----
  • Posts: 3,246
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 18 June 2021 - 01:16 AM

I wonder if your field of view might be too narrow to capture enough stars for the plate solve.  It is a bright nebula in the middle of a dark cloud. 

 

If your guider has a wider field of view, you might try using it instead (will need a separate profile), as a test. 



#6 han.k

han.k

    ASTAP Software

  • *****
  • Vendors
  • Posts: 579
  • Joined: 09 Mar 2015
  • Loc: the Netherlands

Posted 18 June 2021 - 02:37 AM

The star density around Iris nebula/NGC7023 is low which could result in a solve failure. I have compared the Gaia database with the Deep sky survey and this confirm this real and no an incomplete database.

 

If you have a small field of view, the only thing you could do is first point and sync the mount at a position near the Iris nebula and then slew to the nebula.

This should result in accurate pointing.

 

Still it would be interesting if you could share an image of the Iris nebula for more testing, e.g via https://ufile.io/ or nova.astrometry.net

 

Han

 

White is database, red in deepsky survey:

ngc7023_1.png

 

 

 


Edited by han.k, 18 June 2021 - 02:38 AM.


#7 MikeECha

MikeECha

    Ranger 4

  • *****
  • topic starter
  • Posts: 358
  • Joined: 15 Sep 2018
  • Loc: Charlotte, NC

Posted 18 June 2021 - 06:11 AM

Any error messages?  Are you at a light polluted site?  Is the F.L. and pixel size correct?  Are ASTAP updates installed?

Jim

 

Thanks for the answer. I have not seen the errors because when I see ASTAP first pup up after 2-3 seconds I know it will fail to plate solve so I stop it .I keep all my astro software up to date including device drivers. Part of my work flow is to check for updates.

 

I am in a Bortle 6-7 



#8 MikeECha

MikeECha

    Ranger 4

  • *****
  • topic starter
  • Posts: 358
  • Joined: 15 Sep 2018
  • Loc: Charlotte, NC

Posted 18 June 2021 - 06:13 AM

Diagnosis suggestion.  Manual platesolving with PlateSolve2 and hand correction.  If that works, the problem is NINA, not the platesolving program.

 

I think it's better to get things going "manually", it makes troubleshooting the fancy image suite easier.

 

Minor point, may or may not be related.  The 183 is a niche camera, and the niche is short fast scopes.  I tried mine (purchased for a C8 RASA) with my 130mm F7.  The results were bad.

Thanks Bob

 

I will check that out



#9 MikeECha

MikeECha

    Ranger 4

  • *****
  • topic starter
  • Posts: 358
  • Joined: 15 Sep 2018
  • Loc: Charlotte, NC

Posted 18 June 2021 - 06:19 AM

I wonder if your field of view might be too narrow to capture enough stars for the plate solve. It is a bright nebula in the middle of a dark cloud.

If your guider has a wider field of view, you might try using it instead (will need a separate profile), as a test.

I just started using Celestron OAG using OAG. A test yesterday showed RMS in the ~.32 arcsec. For a test Iast night I increased the plate solving exposure to 240s and at that exposure NINA detected more than 50 stars and still failed.

Edited by MikeECha, 18 June 2021 - 09:39 AM.


#10 MikeECha

MikeECha

    Ranger 4

  • *****
  • topic starter
  • Posts: 358
  • Joined: 15 Sep 2018
  • Loc: Charlotte, NC

Posted 18 June 2021 - 06:35 AM

The star density around Iris nebula/NGC7023 is low which could result in a solve failure. I have compared the Gaia database with the Deep sky survey and this confirm this real and no an incomplete database.

 

If you have a small field of view, the only thing you could do is first point and sync the mount at a position near the Iris nebula and then slew to the nebula.

This should result in accurate pointing.

 

Still it would be interesting if you could share an image of the Iris nebula for more testing, e.g via https://ufile.io/ or nova.astrometry.net

 

Han

 

White is database, red in deepsky survey:

attachicon.gifngc7023_1.png

Han thank you for your response

 

Yes my field of view height is .37.

 

The problem is only with this target. ASTAP works great with others. I wonder if it has to do with the dominant brightness of the star on the center of the nebula. The problem occurs very consistently. First try in NINA seams like a "ah I am here". Then the second is to gest close but I get "telescope not within pointing tolerance ( default 1 arcmin, also tried 2 arcmin). Then the third plate solve fails, the second ASTAP pop up and it goes to a 180deg search that also fails.

 

About the image. Sorry, I had NINA set to save the images but I can not find them.

 

Your suggestion to plate solve near and slew is great. I might end up doing that tonight. When I get images I will send you one. 



#11 han.k

han.k

    ASTAP Software

  • *****
  • Vendors
  • Posts: 579
  • Joined: 09 Mar 2015
  • Loc: the Netherlands

Posted 18 June 2021 - 06:44 AM

IMikeECha,

 

Yes, if  you have an image next time, it could be interesting to further investigate. These problem images could help to improve the solver.

 

A dominant star should not be a problem.

 

In some cases it could help to activate option "slow" in ASTAP. The solver will make smaller steps through the database. This will give a greater overlap between the search steps.

 

Other options to explore are there many hotpixels? If so hotpixel-supression by either using a dark or by minimum star size could help. Options you can only set in ASTAP.

 

Furthermore having the correct focal-length/sensor size specified helps. But Nina should give you a warning message if the solver thinks it is more then 5% off.

 

A RMS of 32" is not good for solving.

 

Han


  • MikeECha likes this

#12 MikeECha

MikeECha

    Ranger 4

  • *****
  • topic starter
  • Posts: 358
  • Joined: 15 Sep 2018
  • Loc: Charlotte, NC

Posted 18 June 2021 - 09:37 AM

Oops sorry that no is .32 RMS. I must have not had my readers on.

#13 vio

vio

    Mariner 2

  • -----
  • Posts: 270
  • Joined: 27 Apr 2011
  • Loc: Naples, FL

Posted 18 June 2021 - 01:40 PM

If I remember correctly, depending on the catalog used, you may only have the star/data files for wider field, you may need to install a few more data files to cover your field of view.

#14 MikeECha

MikeECha

    Ranger 4

  • *****
  • topic starter
  • Posts: 358
  • Joined: 15 Sep 2018
  • Loc: Charlotte, NC

Posted 18 June 2021 - 04:21 PM

If I remember correctly, depending on the catalog used, you may only have the star/data files for wider field, you may need to install a few more data files to cover your field of view.

I have H17 and H18 installed. Is there anything else I have to install?



#15 han.k

han.k

    ASTAP Software

  • *****
  • Vendors
  • Posts: 579
  • Joined: 09 Mar 2015
  • Loc: the Netherlands

Posted 19 June 2021 - 03:08 AM

The H17 goes down to magnitude 17. The H18 goes down to magnitude 18. You better delete the H17 (H17*.1476) to avoid ASTAP is set to use the H17.

 

Since your FOV is small, you could use online nova.astrometry.net as a backup just in case it is required. For FOV (field of view) equal larger then 0.4 degrees ASTAP will work very reliable. Below I think you could have some rare glitches like you experienced. It is all depending on star density. ASTAP will work fine for FOV as small as 0.2 degrees but only if sufficient stars are available. But outside the MilkyWay plane this is not always the case. Some time ago I have analysed the sky star density  and created the table below. This indicates that the H18 database goes deep enough.

 

That's why a new image of Iris nebula is interesting. Is the star density too low for solving or is it a setting?

 

Han

 

astap_database_range.png



#16 MikeECha

MikeECha

    Ranger 4

  • *****
  • topic starter
  • Posts: 358
  • Joined: 15 Sep 2018
  • Loc: Charlotte, NC

Posted 19 June 2021 - 04:15 PM

The H17 goes down to magnitude 17. The H18 goes down to magnitude 18. You better delete the H17 (H17*.1476) to avoid ASTAP is set to use the H17.

 

Since your FOV is small, you could use online nova.astrometry.net as a backup just in case it is required. For FOV (field of view) equal larger then 0.4 degrees ASTAP will work very reliable. Below I think you could have some rare glitches like you experienced. It is all depending on star density. ASTAP will work fine for FOV as small as 0.2 degrees but only if sufficient stars are available. But outside the MilkyWay plane this is not always the case. Some time ago I have analysed the sky star density  and created the table below. This indicates that the H18 database goes deep enough.

 

That's why a new image of Iris nebula is interesting. Is the star density too low for solving or is it a setting?

 

Han

 

attachicon.gifastap_database_range.png

I just sent you a link to a couple of images. They are not from last night as clouds came in and I did nothing. The fits image will plate solves but it is not at the center the next move to center will fail. That is why I asked before if brightness has anything to do with it.

 

I installed H17 after the fact to see it it would work. ASTAP still solves with H17 selected and just 15 sec exposure just not at the center. I never made anything of it until the other day Pixinsight Photometric CC plate solve would fail. That is why I set out to figure out what the problem is.

 

Let me know if you need more images. I have some I forgot I had in storage.



#17 han.k

han.k

    ASTAP Software

  • *****
  • Vendors
  • Posts: 579
  • Joined: 09 Mar 2015
  • Loc: the Netherlands

Posted 20 June 2021 - 08:34 AM

Thanks for the files. Both images are 120 seconds exposed and solve without problems and enough detection's. Brightness doesn't disturb. Even the center star is used for the solving. Hot pixels are ignored See detection's below. The limiting magnitudes are 15.6 and 16.3. Then there are enough stars detected.  I also looked to some Iris nebula images I could retrieve from nova.astrometry.net and cropped them to about 0.55x0.37 degrees. Some did solve some not. The one which solve did go deeper and showed detectable stars up to magnitude 17.3.

 

So since it is a little dark corner of the sky (for 0.55x0.37 degrees), the images should go to at least up to magnitude 15.6. For this the images should be reasonable in focus and exposure time should be long enough.

 

In general nova.astrometry.net requires less stars to solve. ASTAP will require about 30 stars.

 

So it is unclear why in previously it didn't work for exposure up to 240 seconds. So if possible I would suggest to try next time again and save some images of shorter exposure. If your image goes deep enough it should work.

 

 

Han

 

Below the detection result of the XISF file up to magnitude 16.3.  Red squares are star detection's. Yellow markers are stars used for the solution. You can generate this image by check-mark "Show extended log" in tab alignment.

 

L_L-PRO_2021-05-22_04-51-53_120s_G125_OFS10__-10C compressed.jpg



#18 MikeECha

MikeECha

    Ranger 4

  • *****
  • topic starter
  • Posts: 358
  • Joined: 15 Sep 2018
  • Loc: Charlotte, NC

Posted 20 June 2021 - 10:40 AM

Thanks for the files. Both images are 120 seconds exposed and solve without problems and enough detection's. Brightness doesn't disturb. Even the center star is used for the solving. Hot pixels are ignored See detection's below. The limiting magnitudes are 15.6 and 16.3. Then there are enough stars detected.  I also looked to some Iris nebula images I could retrieve from nova.astrometry.net and cropped them to about 0.55x0.37 degrees. Some did solve some not. The one which solve did go deeper and showed detectable stars up to magnitude 17.3.

 

So since it is a little dark corner of the sky (for 0.55x0.37 degrees), the images should go to at least up to magnitude 15.6. For this the images should be reasonable in focus and exposure time should be long enough.

 

In general nova.astrometry.net requires less stars to solve. ASTAP will require about 30 stars.

 

So it is unclear why in previously it didn't work for exposure up to 240 seconds. So if possible I would suggest to try next time again and save some images of shorter exposure. If your image goes deep enough it should work.

 

 

Han

 

Below the detection result of the XISF file up to magnitude 16.3.  Red squares are star detection's. Yellow markers are stars used for the solution. You can generate this image by check-mark "Show extended log" in tab alignment.

 

attachicon.gifL_L-PRO_2021-05-22_04-51-53_120s_G125_OFS10__-10C compressed.jpg

Hello Han

 

Ye I am waiting for a chance with the weather. It looks like I have another week of rain coming up.

 

The position of the images I sent you would plate solve at 15 sec exposure with H17 with no problem. It would be the next plate solve to the center of the nebula that would fail for both NINA and APT using ASTAP (Thus my thought about brightness "blinding" ASTAP). Those were part of a session where  the star was close enough to the center of the image so I ignore the final centering and went on with it.

 

I think that is why Pixinsight PCC plate solve fails and if I change the fit header with the right center coordinates then it solves. Just a speculation as I do not know much about this. 

 

I know that about a year ago or so, I would have to lower the exposure very much for plate solve otherwise it would fail. I was using APT in case that matters. IIRC that was discussed around the forums and fixed.

 

I have also seen this happen when trying to center bright stars like Sirius or Vega with my AT6RC at .36 arcsec/pix. But that was a while ago and I do not remember the exact conditions.

 

I will send you images when the weather let get some.

 

Thank you for your explanations.



#19 Bob Denny

Bob Denny

    Vendor (DC-3 Dreams)

  • -----
  • Vendors
  • Posts: 283
  • Joined: 17 Mar 2009
  • Loc: Mesa AZ USA

Posted 22 June 2021 - 05:10 PM

I should mention that most stellar reference catalogs have "holes" in the vicinity of extended objects like nebulae, galaxies, and especially star clusters. The automated pipelines used to produce the star positions and magnitudes struggle in dense or bright regions. Therefore, plate solves with an extended object near the center of the field, and no "room" around it, may fail more often.  Here's a (synthesized) 15x15 minute field of the M3 area using the USNO A2.0 catalog, down to 17 mag or so:

 

M3 Hole.jpg



#20 han.k

han.k

    ASTAP Software

  • *****
  • Vendors
  • Posts: 579
  • Joined: 09 Mar 2015
  • Loc: the Netherlands

Posted 23 June 2021 - 03:07 AM

Hello Bob,

 

Yes the USNO catalogue could even show square holes for global clusters. The newer Gaia catalogue is much better in that. But star density could result in new image solving problems. The Gaia catalogue can dissolve stars inside a cluster with a resolution of at least 0.6 arcseconds, much better then most Earth based catalogues. This difference in resolution could hamper solving and I had to fix that some time ago.

 

Han

 

This is an artifical M3 image for the same 16 x16 arcminutes as yours down to magnitude 18:

 

M3_gaia.png



#21 Bob Denny

Bob Denny

    Vendor (DC-3 Dreams)

  • -----
  • Vendors
  • Posts: 283
  • Joined: 17 Mar 2009
  • Loc: Mesa AZ USA

Posted 23 June 2021 - 10:47 AM

Yes the Gaia, and ATLAS (Gaia-derived), and even UCAC4 have smaller or no holes. I was trying to graphically illustrate the concept and that’s why I included the source as the old A2.0.


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics