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

Difficulties Imaging

  • Please log in to reply
42 replies to this topic

#1 Karlp295

Karlp295

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 22 Sep 2020

Posted 23 September 2020 - 04:08 AM

I have been trying astrophotography for a couple of months and am having some success. The last three weeks I have been trying to image Andromeda and have been disappointed by what I have managed to get. My latest efforts have included taking approx. 200 frames of 60 second exposures at 1600ISO with mu Canon DSLR 600D unmodified at prime focus through my telescope Celestron 130slt. My backyard conditions are not ideal as there are lights nearby and it is Bortle 5. I also use a SVBONY CLS clip in filter.

 

Some of the frames were rejected and after stacking 167 frames in Siril the result was not much better than I last got with significantly fewer short exposures. I took bias, dark and flat frames about 20 each.

 

Sorry having difficulty uploading images. I will try again.

 

Is Andromeda a difficult target? I seem to be getting good images of M33.

 

 



#2 Tapio

Tapio

    Gemini

  • -----
  • Posts: 3,147
  • Joined: 24 Sep 2006
  • Loc: Tampere, Finland

Posted 23 September 2020 - 04:57 AM

A couple of hours exposure of M31 should mean pretty nice image.

Have you stretched the stacked image ?


  • limeyx likes this

#3 SilverLitz

SilverLitz

    Apollo

  • -----
  • Posts: 1,172
  • Joined: 17 Feb 2018
  • Loc: Louisville, KY

Posted 23 September 2020 - 10:04 AM

M31 has more dynamic range than most DSOs, with a very bright core.  I would expose M31 at 1/2 the exposure time (per sub) than M33.


  • BinoGuy likes this

#4 Karlp295

Karlp295

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 22 Sep 2020

Posted 24 September 2020 - 03:00 AM

I have stretched the image. Apollo, do you mean I should shoot M31 at 30 secs and what ISO level do you suggest?

 

Here are my latest two images:

 

M31 3050
M31 40 x 25s ISO 800.
 
M31 4025

M31 30 x 50s ISO 800

 

The latest version with 167 x 60s ISO 1600 is no better than this, even worse in fact!

 


Edited by Karlp295, 24 September 2020 - 03:11 AM.


#5 RJF-Astro

RJF-Astro

    Viking 1

  • -----
  • Posts: 670
  • Joined: 13 Aug 2018
  • Loc: Zeist, Netherlands

Posted 24 September 2020 - 03:24 AM

Hi Karlp, 3 hours should get you a nice image of M31. It is actually one of the easier targets, although you should keep an eye on the center. But yours seems fine.

 

To give a more meaningfull response we need to know your processing workflow. Also if you can post the stacked, unprocessed image we can give it a go and see what is there in the data. The advice could then be to either focus on better data collection, or to improve either calibration or processing.

 

One thing I do notice in the top image is that focus is off. This will have a negative impact on image quality, as the stars are more prominent and details are blurred. This is something to fix in data collection: make sure you are properly focused, maybe add a bathinov mask if you don't use one already.


Edited by RJF-Astro, 24 September 2020 - 03:25 AM.


#6 Tapio

Tapio

    Gemini

  • -----
  • Posts: 3,147
  • Joined: 24 Sep 2006
  • Loc: Tampere, Finland

Posted 24 September 2020 - 05:15 AM

Can you put the unstretched image somewhere ?



#7 SilverLitz

SilverLitz

    Apollo

  • -----
  • Posts: 1,172
  • Joined: 17 Feb 2018
  • Loc: Louisville, KY

Posted 24 September 2020 - 02:49 PM

 

I have stretched the image. Apollo, do you mean I should shoot M31 at 30 secs and what ISO level do you suggest?

 

Here are my latest two images:

 

 
M31 40 x 25s ISO 800.
 
 

M31 30 x 50s ISO 800

 

The latest version with 167 x 60s ISO 1600 is no better than this, even worse in fact!

 

 

(My handle is SilverLitz; Apollo is CN's classification based on how many posts I have made.)

 

The 25s bracket looks quite a bit better than the 50s set, even though this is not fair as the 25s set has only 1000s of total integration vs. 1500s on the 50s set.  The shorter exposures do not blow out the core as much and have much better color.  

 

I am not specific regarding 25s vs 30s, as your specifics are determined by YOUR skies, camera, scope etc...  But I think you get the idea, not all targets need the same exposure lengths, and M31 and M42 need shorter exposures than most DSOs. 

 

With your Canon Rebel (?), ISO 800 is recommended, as it is the ISO Invariant setting for Rebels or 80D.  Canon 7Dmk2, 60D/70D, and 5Dmk4 the recommended setting is ISO 1600.

 

Make your life much easier by selecting just a few exposure lengths and a single ISO setting.  I normally only exposure in full stop increments, such as 15s, 30s, 60s, 120s, 240s.  You will need a dark library for EVERY combination of exposure and ISO, as well as sensor temperature.  Getting a temperature matched dark library for non-cooled/regulated cameras is a major PITA; limit the number of combinations!

 

You need many more subs however for good images, MUCH more than you expect.  SNR doubles for each 4x increase in total integration time.  My 1st attempt at M31 last year had 346min of total exposures, shot with a Canon 70-200mm f/2.8L lens (at 200mm and f/2.8), using my ASI183mm-Pro at Unity gain and LRGB filters plus more exposures using Ha filter (to enhance the red star forming knots).   This was from my Bortle 5+ backyard.  Worse skies (higher Bortle) or slower scope (higher f/) require more time.

 

M31_200mm_LHaRGB_90_10 (LoRes).JPG



#8 Karlp295

Karlp295

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 22 Sep 2020

Posted 25 September 2020 - 02:22 AM

Thanks for your advice Silverlitz. I was disappointed because I had seen online some good pictures that stated 120 x 60s exposures. I think that was taken from a lens on a DSLR rather than through a scope. My scope is f/5 so I guess if a camera uses a lens of F/2.8 my total exposure needs to be 3-4 times longer doesn't it?

 

As you say it obviously exposure time is not long enough, although I also need to be more experienced at bringing out the detail in post processing. 

 

My workflow has been as follows for these:

 

I have used Siril for stacking and for stretching, calibrating and removing gradient in the picture.

I also have played around in Raw Therapee to get the picture as best I could.

 

Now I am trying Star Tools and only just beginning to get how that works.

 

I have uploaded a Tif file of the unedited stack of 167 frames at 60secs and 1600 ISO including Bias, Dark and Flat frames. Flat frames are also new for me and I'm not sure if these were absolutely as they should be.

 

The file is here:

https://1drv.ms/u/s!...n7c0x?e=2Xt9QO 

 

on my google drive. You should be able to download the file.


Edited by Karlp295, 25 September 2020 - 05:49 AM.


#9 RJF-Astro

RJF-Astro

    Viking 1

  • -----
  • Posts: 670
  • Joined: 13 Aug 2018
  • Loc: Zeist, Netherlands

Posted 25 September 2020 - 04:36 AM

Thanks! It is difficult to tell. But the stacked image is not what I would expect with 2 hours of data. There is definitely not much signal. There are complex gradients, also tracking/focusing errors.

 

I would start with sky quality: was the moon out? And were there high clouds? Did the optics suffer from dew?

 

Then: was the scope properly collimated? Did you check focus with a bathinov? And how was the tracking perfomace/polar aligment?

 

After that comes calibration. Calibration errors can degrade the stacked image quality. How did you make the calibration frames (darks, bias and flats)? I do not know Siril for stacking, but I am assuming you do not have to configure like PixInsight for instance, so if you did not touch any settings it should be fine.

 

The problem does not lie in post processing. The data just is not there yet, so your focus should be on data collection and calibration.



#10 Karlp295

Karlp295

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 22 Sep 2020

Posted 25 September 2020 - 05:53 AM

I did not use tracking. No dew, 26 degrees C. No moon. Collimated scope. Could the flat frames be a problem?

 

I just don't see any detail. using the same setup I also imaged M33 and all looks good.

 

Confused!



#11 venyix

venyix

    Explorer 1

  • -----
  • Posts: 89
  • Joined: 22 Jul 2020

Posted 25 September 2020 - 09:25 AM

I did not use tracking. No dew, 26 degrees C. No moon. Collimated scope. Could the flat frames be a problem?

 

I just don't see any detail. using the same setup I also imaged M33 and all looks good.

 

Confused!

 

Is it possible that some of your dark frames aren't very dark? I would use a program like Adobe Bridge or something and go through EACH image that you are stacking, Flat, Bias, Dark, Lights. Toss any darks that seem to vary in brightness, make sure your images are fairly uniform, drop any light frames with blur or clouds in them. This is three hours of exposure on my Sony A6500 with around 45 second subs. This is the raw stacked image, and if you stretch it a little bit, you'll see how much more data there is in this image. Also, maybe try imaging for about an hour without the clip in filter and see if you get some better results.

 

https://drive.google...iew?usp=sharing

 

Here is the edited image...

 

The Andromeda Galaxy

Edited by venyix, 25 September 2020 - 09:26 AM.

  • BinoGuy and xanadu30 like this

#12 RJF-Astro

RJF-Astro

    Viking 1

  • -----
  • Posts: 670
  • Joined: 13 Aug 2018
  • Loc: Zeist, Netherlands

Posted 25 September 2020 - 10:18 AM

I did not use tracking. No dew, 26 degrees C. No moon. Collimated scope. Could the flat frames be a problem?

 

I just don't see any detail. using the same setup I also imaged M33 and all looks good.

 

Confused!

Flats can certainly degrade image quality, as can bias and darks. I would check your calibration frames like venyix says. Start with the masters and then the individual frames.

 

If you used the same calibration frames on M33 and the result was good, well that can be a pointer that the problem lies elsewhere. But try that first, even though it might be a different night, as long as the temperature is roughly the same the darks should still do their job.



#13 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 21,466
  • Joined: 27 Oct 2014

Posted 25 September 2020 - 10:56 AM

Here's what I could get from your data.

 

The CLS is killing you on this target.  It doesn't magically filer "light pollution", it whacks out hunks of the spectrum, hoping that what it whacks is mostly light pollution, what it passes is mostly signal.  For emission nebulae, that can work some.  For Andromeda, not so much.

 

It reduces the signal from Andromeda too much.  It makes color processing very difficult, I had to pull out all my tricks to get something halfway decent.

 

You don't need it.  Bortle 5 is way darker than my Bortle 7 skies.  Below your image is one of mine of Andromeda.  Nikon D5500.  70mm telescope. 1.2 hours.  No (not so) magic light pollution filter used.  I did work hard on doing good gradient reduction in processing.

 

Better version with more details here.

 

https://www.astrobin.com/263253/J/

 

Important bottom line.  This is more complicated than many beginners think.  You'll never get enough information from short posts here.  I recommend this book for starters.  It discusses gradient reduction.

 

https://www.amazon.c.../dp/0999470906/

 

result_ABE.jpg

 

M31 SV70T smaller.jpg


Edited by bobzeq25, 25 September 2020 - 11:04 AM.

  • nimitz69 likes this

#14 Karlp295

Karlp295

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 22 Sep 2020

Posted 25 September 2020 - 12:25 PM

I used the same darks and bias frames for M33. I'm going to try without the filter which is SVBony CLS. Let's see what happens. I will take 30s frames at ISO 800. 

 

I'll post the results.


  • bobzeq25 likes this

#15 RJF-Astro

RJF-Astro

    Viking 1

  • -----
  • Posts: 670
  • Joined: 13 Aug 2018
  • Loc: Zeist, Netherlands

Posted 25 September 2020 - 12:41 PM

Bob has a point about the filter, which still cuts out a big deal of light comming from galaxies. However it cannot explain the difference between your M33 and M31. If calibration is the same except flats, you could try stacking without flats. But taking another shot at M31 is also a good plan. Keep us posted!



#16 idclimber

idclimber

    Viking 1

  • -----
  • Posts: 579
  • Joined: 08 Apr 2016
  • Loc: McCall Idaho

Posted 25 September 2020 - 01:15 PM

Take a single photo and look at the histogram before proceeding. If possible you want to reduce the exposure time so the core is not overexposed. On my Nikon I can set it so it shows what areas of the image are at the clipping limit. Any areas that are at that limit will loose all color and detail and will not recover from any curve or stretch. 



#17 GoldSpider

GoldSpider

    Apollo

  • *****
  • Posts: 1,088
  • Joined: 20 Apr 2015
  • Loc: Pennsylvania

Posted 25 September 2020 - 04:11 PM

I doubt the CLS filter accounts for all of what's going on there, but I agree it's not going to help you on broad spectrum subjects like galaxies.  I'm more inclined to believe that signal is being subtracted by improper calibration frames.



#18 Karlp295

Karlp295

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 22 Sep 2020

Posted 26 September 2020 - 10:33 AM

So here is the latest image stacked with darks and flats. One hour as a test of 30 sec exposures without the CLS filter.

I can see a bit more detail and it gives me some hope but I also notice the difference in the number of stars in my photo compared to yours.

 

If I go for two more hours will my image be much better? 

Perhaps I should try this with no flats before proceeding.

 

M31Sequatorsirilproc1hr


#19 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 21,466
  • Joined: 27 Oct 2014

Posted 26 September 2020 - 11:27 AM

So here is the latest image stacked with darks and flats. One hour as a test of 30 sec exposures without the CLS filter.

I can see a bit more detail and it gives me some hope but I also notice the difference in the number of stars in my photo compared to yours.

 

If I go for two more hours will my image be much better? 

Perhaps I should try this with no flats before proceeding.

 

Flats help.  It may be necessary to do them better, but omitting them will not make your images better.  No one does that.



#20 Karlp295

Karlp295

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 22 Sep 2020

Posted 26 September 2020 - 12:05 PM

No the image is not much different just using darks. The darks were taken immediately after the light frames and with the cap on the telescope. All settings the same. 

 

This latest batch were taken at ISO 800 which is recommended for my camera. Previously I had used 1600 ISO.

 

After 1 hour I can see indication of the dust lanes is this normal after this many frames?

 

Note my camera is unmodded.



#21 GoldSpider

GoldSpider

    Apollo

  • *****
  • Posts: 1,088
  • Joined: 20 Apr 2015
  • Loc: Pennsylvania

Posted 26 September 2020 - 12:16 PM

No the image is not much different just using darks. The darks were taken immediately after the light frames and with the cap on the telescope. All settings the same. 

 

This latest batch were taken at ISO 800 which is recommended for my camera. Previously I had used 1600 ISO.

 

After 1 hour I can see indication of the dust lanes is this normal after this many frames?

 

Note my camera is unmodded.

Any way you could share your data with us?  Even if it's just a dozen or two of your light frames and a half-dozen calibration frames.  Just so we can see what you're working with.


Edited by GoldSpider, 26 September 2020 - 12:33 PM.

  • limeyx likes this

#22 Karlp295

Karlp295

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 22 Sep 2020

Posted 27 September 2020 - 03:24 AM

I've just uploaded 6 light frames and 4 dark frames so you can check what they look like. If you need to see more I can upload them later but I think this is enough to get an idea?

 

Thanks.

 

Here's the link to the files:

https://drive.google...iew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...iew?usp=sharing



#23 idclimber

idclimber

    Viking 1

  • -----
  • Posts: 579
  • Joined: 08 Apr 2016
  • Loc: McCall Idaho

Posted 27 September 2020 - 07:44 AM

You have significant star trailing in the light frame I opened. If most of your stacked frames have the same movement, then your fine details will be lost. 



#24 idclimber

idclimber

    Viking 1

  • -----
  • Posts: 579
  • Joined: 08 Apr 2016
  • Loc: McCall Idaho

Posted 27 September 2020 - 07:50 AM

This frame clearly show there was movement of the mount during your exposure. 

Attached Thumbnails

  • IMG_4487.jpg


#25 RJF-Astro

RJF-Astro

    Viking 1

  • -----
  • Posts: 670
  • Joined: 13 Aug 2018
  • Loc: Zeist, Netherlands

Posted 27 September 2020 - 09:33 AM

That looks like a good explanation. Reading back I noticed this:

 

I did not use tracking.

I assume you mean guiding? Because no tracking on a 60 second sub woud be very difficult. In any case tracking seems to be the issue. Check your polar alignment, preferably the last thing you do after all mechanical adjustments. If you have no go-to you should check PA after moving to a target.


  • limeyx likes 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






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics