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

NINA platesolving accuracy

Astrometry Astrophotography Imaging Software
  • Please log in to reply
8 replies to this topic

#1 Dovakun

Dovakun

    Explorer 1

  • *****
  • topic starter
  • Posts: 57
  • Joined: 16 Aug 2018

Posted 02 March 2021 - 06:38 AM

I recently started using NINA for image capture. So far I really like the functionality. Autofocus works great once dialed in and the mosaic planner is awesome! But platesolving seems to be falling short for me. I'm using M82 for my testing ground. After slewing and focusing I first tried to run ASTAP. No stars were found. Tried increasing my exposure time and nothing. So I tried using astrometry.net. This did find stars and it did get M82 close but only within 1 arcmin and it took forever due to having to upload an image every time to solve. Next I tried Platesolve2. This was pretty much the same accuracy as astrometry.net but was much faster. The problem still remained though, I can't actually get M82 centered. I dropped the pointer tolerance to .5 arcmin and it just continuously loops. Telescope is not within tolerance, re-slew, take exposure, platesolve, etc.  I'm imaging at 1960mm @F7 with an ASI1600mm Pro so being that far off is not great for multi-night imaging. 

 

Current settings are:

ASI1600mm @139 gain

10s Exposures

.5 pointer tolerance

Platesolve2 with 5000 regions using APM catalog

 

Am I missing something? Has anyone gotten NINA to solve within .5 arcmin?

 

Thanks in advance. Clear skies.


Edited by Dovakun, 02 March 2021 - 06:39 AM.


#2 ButterFly

ButterFly

    Surveyor 1

  • *****
  • Posts: 1,900
  • Joined: 07 Sep 2018

Posted 02 March 2021 - 06:53 AM

NINA isn't a plate solver.  It just asks your plate solver to solve.  NINA does the syncing and slewing.

 

Turn up the gain.  Your are pointing, not imaging.  ASTAP is by far the quickest platesolver when you get working.  Any one will handle that focal length with the right indexes.



#3 dghent

dghent

    Apollo

  • *****
  • Vendors
  • Posts: 1,055
  • Joined: 10 Jun 2007

Posted 02 March 2021 - 07:19 AM

At 1960mm you’re likely using the wrong database in ASTAP if you’re getting the “no stars” error from it. If you’ve downloaded the H17 database, then replace that with the H18 one. ASTAP’s website and documentation explains this necessity for those running at longer focal lengths.

 

The Tolerance setting in NINA is for the Centering function. If a plate solve says you're outside the specified distance around the desired coordinates, then another solve/sync/slew operation will be done until it gets you within that tolerance. In ability to get inside 0.5' can come down to any number of issues. Foremost is your mount's sense of time could be off and the position it calculates for the coordinates puts it slightly off actual in the RA. You could also be suffering from backlash .Finally, you could also have sync turned off in NINA, so plate solves aren't updating the mount with their findings.


Edited by dghent, 02 March 2021 - 08:39 AM.

  • Dovakun, limeyx and AstroVagabond like this

#4 sbharrat

sbharrat

    Viking 1

  • -----
  • Posts: 501
  • Joined: 28 Nov 2020
  • Loc: NJ, USA

Posted 02 March 2021 - 08:22 AM

 

Current settings are:

ASI1600mm @139 gain

10s Exposures

.5 pointer tolerance

Platesolve2 with 5000 regions using APM catalog

 

Am I missing something? Has anyone gotten NINA to solve within .5 arcmin?

 

Thanks in advance. Clear skies.

I have my setting set to 1' tolerance. Usually the final result ends up in the 20" area.

 

But... my FL is only 600 and imaging scale is about 2"/px

 

Cheers



#5 Dovakun

Dovakun

    Explorer 1

  • *****
  • topic starter
  • Posts: 57
  • Joined: 16 Aug 2018

Posted 02 March 2021 - 09:14 AM

At 1960mm you’re likely using the wrong database in ASTAP if you’re getting the “no stars” error from it. If you’ve downloaded the H17 database, then replace that with the H18 one. ASTAP’s website and documentation explains this necessity for those running at longer focal lengths.

 

The Tolerance setting in NINA is for the Centering function. If a plate solve says you're outside the specified distance around the desired coordinates, then another solve/sync/slew operation will be done until it gets you within that tolerance. In ability to get inside 0.5' can come down to any number of issues. Foremost is your mount's sense of time could be off and the position it calculates for the coordinates puts it slightly off actual in the RA. You could also be suffering from backlash .Finally, you could also have sync turned off in NINA, so plate solves aren't updating the mount with their findings.

I didn't even realize there were 2 DBs! I'll definitely give that a try. As far as the lack of centering. I can rule out sync because when it goes to center the object the first slew will move the mount, then it will get it to the same spot every single time but slightly outside of center when using the crosshairs. I first thought backlash as well but I would think it would be unlikely as I let it run for 20min and the object didn't move. I would think with no guiding the backlash would move the object around a little or cause trailing in the stars. The last thing I will need to check is the time and coordinates. The CEM60 has a built in GPS and time is usually spot on but that could have slipped at some point.

 

Thanks!



#6 Dovakun

Dovakun

    Explorer 1

  • *****
  • topic starter
  • Posts: 57
  • Joined: 16 Aug 2018

Posted 02 March 2021 - 09:17 AM

NINA isn't a plate solver.  It just asks your plate solver to solve.  NINA does the syncing and slewing.

 

Turn up the gain.  Your are pointing, not imaging.  ASTAP is by far the quickest platesolver when you get working.  Any one will handle that focal length with the right indexes.

Correct, but the function in NINA is called platesolve. Since the other platesolvers were able to find matching stars it is likely due to me downloading the wrong database for ASTAP. The accuracy is something I'm going to play with as far as validating time, coordinates, and backlash. It should be clear again tonight so I will give it a try.



#7 dghent

dghent

    Apollo

  • *****
  • Vendors
  • Posts: 1,055
  • Joined: 10 Jun 2007

Posted 02 March 2021 - 09:41 AM

Correct, but the function in NINA is called platesolve. Since the other platesolvers were able to find matching stars it is likely due to me downloading the wrong database for ASTAP. The accuracy is something I'm going to play with as far as validating time, coordinates, and backlash. It should be clear again tonight so I will give it a try.

NINA has a generic interface for talking to multiple plate solvers, as you found out (Astrometry.NET, PS2, ASTAP, and so on). If these underlying solve apps produce an error, NINA will try to relay it to the user. It might look like NINA itself is producing the error, but it is in reality coming from the separate solving app. NINA itself does not "do" the actual plate solving. It just passes the data needed to solve to the solver, and digests its response.


Edited by dghent, 02 March 2021 - 09:59 AM.

  • sn2006gy likes this

#8 Phil Sherman

Phil Sherman

    Gemini

  • *****
  • Posts: 3,355
  • Joined: 07 Dec 2010
  • Loc: Cleveland, Ohio

Posted 02 March 2021 - 11:40 AM

Plate solves should give you the mount's position accurate to an arcsec. When this information is used to move the mount to your target, the mount's sky map and movement controls also come into play. My mount, controlled with EQMOD, has two pointing models. The more accurate one, which uses a triangle of sync points surrounding the target, rarely doesn't center a target within 150 pixels of the center of the frame. Unfortunately, using this model yields poor results when using plate solve to move the scope a short distance. The other pointing model, nearest point, always has the target in the camera's fov. Plate solve controlled mount movement rarely doesn't accurately move the mount to the desired location.

 

I suspect that something in your sky map / sync points is causing your mount to not accurately position when plate solve is moving the mount.


Edited by Phil Sherman, 02 March 2021 - 11:45 AM.


#9 Dovakun

Dovakun

    Explorer 1

  • *****
  • topic starter
  • Posts: 57
  • Joined: 16 Aug 2018

Posted 03 March 2021 - 12:11 PM

Thanks for the responses everyone. Downloading the H18 DB and bumping up the scope settle time to 15sec worked like a charm. Solves within .5 arcmin very quickly!


  • dghent and ButterFly like this


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





Also tagged with one or more of these keywords: Astrometry, Astrophotography, Imaging, Software



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics