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

Astroberry Problems Redux. Please help.

  • Please log in to reply
58 replies to this topic

#1 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 14 May 2021 - 10:00 PM

Well I thought I had my plate solve issues worked out. The few plate solves I did last night worked perfectly.

 

However tonight has been   "Please check that you have sufficient stars in the image. Enable alignment logging in the setup tab."

 

No matter how much I change the exposure or the iso this error repeats. However in the image it does produce there are at least 30-50 stars that I can count.

 

I believe I have all the correct focal length and camera perimeters entered.

 

In profile editor I have the correct aperture and focal length for the telescope I'm using. It is selected as primary scope. 

 

For the camera in image setting I have the necessary info  updated via the "set" box to the right.

 

However in image info, the set box to the right WILL NOT let me update the parameters.

 

The other thing that seems off is in the Solution Coordinates (JNow) box it has the correct focal length, but it says 431.8' x 183.2' for calculated field of view which doesnt make sense

 

If I use an online field of view calculator  it has my field of view as 1.4 degrees by 56.3'.  I've tried a couple online calculators and the spit out the same info.

 

Camera is a Canon 60da. Link is to the important info: https://www.the-digi...aspx?Camera=736

 

Scope is a TMB130SS. 130mm with 910mm focal length.

 

I have the most recent version of Astroberry and I am using StellarSolver to plate solve. Note that the online solve has the same error message.

 

I tried to log it but I cannot find the log files.

 

 

thanks for your help!

 

 

 

Attached Thumbnails

  • IMG_7660.jpg
  • IMG_7661.jpg


#2 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 14 May 2021 - 10:22 PM

I should note too. I just took a 15 second image at iso 6400 in the image capture panel and there were a bunch of stars.

 

Did the same exposure in the plate solve panel and got not stars. doesnt make sense.

 

I just want it to work. I've wasted 6 or 7 clear nights so far messing with this system.



#3 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 15 May 2021 - 06:32 AM

So when I went out to the scope last the scope had slid down the mounting rings. Throwing the balance off. This could be part of the reason but still doesn’t explain why no sure showed up in the plate solver.

#4 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 15 May 2021 - 01:44 PM

Anyone? Lol

#5 Kevin Ross

Kevin Ross

    Apollo

  • *****
  • Posts: 1,233
  • Joined: 28 Mar 2018
  • Loc: Traverse City, MI, USA

Posted 15 May 2021 - 09:49 PM

I use ASTAP with Ekos, it's fast and has been pretty reliable for me. I don't know if Astroberry comes preinstalled with ASTAP or not, but it's easy to install if it doesn't (don't forget to get the catalogs too, they don't use the same catalogs as StellarSolver or astrometry.net).

 

Also, one tip I picked up from the INDI forums was to go to the alignment options, go to Scale and Position, and uncheck Use Scale.

 

You can also try longer or shorter exposures for plate solving, maybe it will like one better than another.

 

If all else fails, go the the official forums at indilib.org and ask for help there.


  • zjc26138 likes this

#6 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 16 May 2021 - 08:19 AM

Alright I’ll try those two things. Thanks!

I’ll pass on the indi forums for now. I’ve posted 3 topics so far and have barely gotten a response.

Cloudynights has been better so far.

#7 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 16 May 2021 - 08:40 PM

Still  no luck plate solving. I must not be picking areas with a lot of stars.

 

Tried an exposure on regulus of 20 seconds at iso 800 and only regulus shows up.

 

It's not the best night out, but the other night the same issue is happening.

 

Luckily I don't need plate solve tonight, as my one star alignment from the hand controller worked



#8 han.k

han.k

    ASTAP Software

  • *****
  • Vendors
  • Posts: 551
  • Joined: 09 Mar 2015
  • Loc: the Netherlands

Posted 17 May 2021 - 01:27 AM

I checked your image at Astrobin and the numbers, focal length  FOV and pixelsize and they match.

 

>>it says 431.8' x 183.2' for calculated field of view which doesn't make sense

This FOV is clearly wrong. Your FOV  is 84 x 53'.  So pixelsize X and Y should be corrected.  If you have problems to set it as indicated, you could increase the focal length to compensate. But if X and Y are different you have to change the focal length to either 910*22/4.3 or  910*15/4.3. Better is to update the X and Y settings or at least get them equal.

 

It could help if you upload an FITS image used for the solver to nova.astrometry.net and share the resulting link. Then we can check the parameters in the FITS header and the image quality. The image to solve is normally saved in path /tmp.  But any FITS file image will allow checking the settings in the header.

 

Han


Edited by han.k, 17 May 2021 - 01:33 AM.

  • zjc26138 likes this

#9 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 17 May 2021 - 06:44 AM

I checked your image at Astrobin and the numbers, focal length  FOV and pixelsize and they match.

 

>>it says 431.8' x 183.2' for calculated field of view which doesn't make sense

This FOV is clearly wrong. Your FOV  is 84 x 53'.  So pixelsize X and Y should be corrected.  If you have problems to set it as indicated, you could increase the focal length to compensate. But if X and Y are different you have to change the focal length to either 910*22/4.3 or  910*15/4.3. Better is to update the X and Y settings or at least get them equal.

 

It could help if you upload an FITS image used for the solver to nova.astrometry.net and share the resulting link. Then we can check the parameters in the FITS header and the image quality. The image to solve is normally saved in path /tmp.  But any FITS file image will allow checking the settings in the header.

 

Han

Thanks for your help!

 

Once I get home from work I will upload a fits file.



#10 Der_Pit

Der_Pit

    Vanguard

  • *****
  • Posts: 2,242
  • Joined: 07 Jul 2018
  • Loc: La Palma

Posted 17 May 2021 - 09:24 AM

The pixel sizes are wrong.  Size X and Size Y should not be the sensor dimensions (guess that is what you tried to enter there?).  The 60Da has 4.3μ square pixels.  So all three entries should be 4.3.

With the numbers that far off, it's no wonder the solver cannot make sense from the data....

 

I'd assume if you enter those,  you'll get the correct FOV displayed in the align tab.  Then solving should work nicely, and disabling the hinting (as Kevin Ross suggested) should not be needed.


  • zjc26138 likes this

#11 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 17 May 2021 - 10:13 AM

But the thing is the solver did work correctly several times with it being wonky

 

The more interesting issue now not being able to find stars in the field of view. I've adjusted my exposure as high as 60s at iso 1600. More than enough to get stars.

 

 

My guess is my transparency was crap last night. Tomorrow night the forecast is calling for really nice skies. I will try again with the above parameters fixed.

 

Thanks for all your help!

 

I do appreciate it!



#12 acommonsoul

acommonsoul

    Ranger 4

  • -----
  • Posts: 397
  • Joined: 08 Jan 2014
  • Loc: Hacienda Heights, Ca

Posted 17 May 2021 - 02:30 PM

Sometimes you can get it to solve with wonky settings, but its better to get it set right. What @Der_Pit mentioned above is correct. The pixel size should be 4.3 for both X and Y and Pixel size. When you first loaded your camera driver and it asked you for the sensor info what did you put? Should have been the sensor dimensions and then the pixel size (4.3 x4.3). Then double check and make sure the focal length is correct for your scope.  



#13 Der_Pit

Der_Pit

    Vanguard

  • *****
  • Posts: 2,242
  • Joined: 07 Jul 2018
  • Loc: La Palma

Posted 17 May 2021 - 04:33 PM

But the thing is the solver did work correctly several times with it being wonky

Can be a coincidence. It was looking for highly stretched (in one dimension) star patterns.  Likely it found a matching geometry for that - the completely off FOV suggests that, too.
 

The more interesting issue now not being able to find stars in the field of view. I've adjusted my exposure as high as 60s at iso 1600. More than enough to get stars.

It might be related to the wrong pixel size: With that, stars get elongated to ellipses, maybe the detection algorithm bails out on that.  You might also try different options for the solver configuration (Options in the Align tab gives you a window where you can adjust many aspects of the solver operation.  Hover the mouse on them, the tooltips give some helpful info).

 

My guess is my transparency was crap last night. Tomorrow night the forecast is calling for really nice skies. I will try again with the above parameters fixed.

I'm quite confident it's the parameters.  My solver usually finds a match within seconds(*) even for images in twilight when I can only see < 10 stars in the image....

 

(*) that is running on a laptop, not on a Pi....

 

Edit: You can also try putting the camera in greyscale mode.  At least my experience from guiding with a color camera was also that it's smoother in mono....


Edited by Der_Pit, 17 May 2021 - 04:35 PM.


#14 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 17 May 2021 - 09:20 PM

How does one change the camera settings to get the 4.3 number?

I know when you first hook up the camera it asks you.

But after that I can seem to edit that number. Is there a way?

#15 Kevin Ross

Kevin Ross

    Apollo

  • *****
  • Posts: 1,233
  • Joined: 28 Mar 2018
  • Loc: Traverse City, MI, USA

Posted 17 May 2021 - 09:23 PM

In the INDI control panel you can set it. On the Image Settings tab (not the Image Info tab). Your second screen shot up above is where you would go. Make sure to hit Save on the options tab. (I'm doing this from memory, I no longer have a DSLR hooked up to Ekos)


Edited by Kevin Ross, 17 May 2021 - 09:25 PM.

  • zjc26138 likes this

#16 dx_ron

dx_ron

    Ranger 4

  • *****
  • Posts: 341
  • Joined: 10 Sep 2020
  • Loc: SW Ohio

Posted 17 May 2021 - 09:38 PM

 Make sure to hit Save on the options tab.

Just emphasizing Kevin's reminder. That has bit me several times. The 'Save this configuration change' function is stashed away in an obscure place in the middle of Options.


  • zjc26138 and Hank Scorpio like this

#17 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 18 May 2021 - 11:44 AM

Thanks for your help everyone!

 

Hopefully I can get it squares away.

 

 

Another issue I had the other night was Ekos crashing. I had a run of 60 70s images and I went to check on it and noticed ekos had crashed and only took 10 frames.

 

 

Is it normal for ekos to crash?  I would hate to have an imaging run setup only to find out in the morning that ekos crashed.



#18 dx_ron

dx_ron

    Ranger 4

  • *****
  • Posts: 341
  • Joined: 10 Sep 2020
  • Loc: SW Ohio

Posted 18 May 2021 - 12:06 PM

Seems to crash more for some people than for others. I run Ekos on a 4GB pi4 that sits on the tripod and I connect to it via wifi VNC Viewer. I find crashes to be quite rare.


  • zjc26138 likes this

#19 corpusjonsey

corpusjonsey

    Sputnik

  • -----
  • Posts: 25
  • Joined: 11 Dec 2020

Posted 18 May 2021 - 12:09 PM

All of the crashes I have had were due to plugging/unplugging usb cables.


  • zjc26138 likes this

#20 Kevin Ross

Kevin Ross

    Apollo

  • *****
  • Posts: 1,233
  • Joined: 28 Mar 2018
  • Loc: Traverse City, MI, USA

Posted 18 May 2021 - 04:14 PM

Another issue I had the other night was Ekos crashing. I had a run of 60 70s images and I went to check on it and noticed ekos had crashed and only took 10 frames.

 

 

Is it normal for ekos to crash?  I would hate to have an imaging run setup only to find out in the morning that ekos crashed.

Unfortunately, that's not too terribly uncommon. I often would get crashes in StellarSolver, which prompted me to switch back to ASTAP for plate solving. No more crashes there. I also occasionally get a crash while autofocusing. It's not too often, and I haven't been able to identify the cause yet. Otherwise KStars/Ekos is pretty solid.

 

If it becomes too much of a problem, you might want to have a look at CCDciel. It's basically a KStars/Ekos replacement that will run on the Pi, using INDI as the driver backend.


  • zjc26138 likes this

#21 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 18 May 2021 - 08:31 PM

Still no luck. I changed width and height to 4.3 and 4.3 and top to 5184 and left to 3456. I click set and it spits out 0 for left and 0 for top. And 5202 for width and 3465 height. I also changed the calculated field of view to

84'x53'. Still get insufficient stars.

 

Also  have use scale and auto update unchecked in the plate solve options.

 

 

Maybe I should start all over again when it first ask you about your camera settings?  Is there an easy way to do this?

 

 

 

Attached Thumbnails

  • IMG_7737.jpg
  • IMG_7736.jpg


#22 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 18 May 2021 - 09:39 PM

Doesn't make any sense.

 

Take a 15s exposure and have tons of stars.

 

Do the same thing with plate solve and usually 1 to 0 stars.

 

I tried uploading a previous image and that did not work either.



#23 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 19 May 2021 - 06:37 AM

I ultimately gave up last night. This makes 7 or so wasted nights of imaging now. :lol:



#24 DubbelDerp

DubbelDerp

    Surveyor 1

  • *****
  • Posts: 1,694
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 19 May 2021 - 08:47 AM

Can you upload an image and share a link to it here? If one of us can figure out how to platesolve it, we might be able to get you on the right track.


  • zjc26138 likes this

#25 zjc26138

zjc26138

    Loved By All

  • *****
  • topic starter
  • Posts: 9,585
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 19 May 2021 - 09:16 AM

Yeah I can do this when I get home today.




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