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.


Image Exposures in LP vs Dark Site

This topic has been archived. This means that you cannot reply to this topic.
3 replies to this topic

#1 Jeff2011



  • *****
  • topic starter
  • Posts: 3381
  • Joined: 01 Jan 2013

Posted 06 August 2014 - 10:05 PM

I have been playing around with the image exposure calculator in SkyTools3 Pro and it has enlightened me.  You enter in your gear and your conditions and it tells you based on a whole lot of factors the times to image, what your sub exposures should be and what your SNR will be.  So I picked the Helix for tonight and based on my conditions (Bortle 7 at zenith, 78 degF,  90% humidity, elevation 22 meters, seeing average), it suggested 293 exposures at 40 seconds for a whopping SNR of 6 with the luminance filter.  I don't have the detailed specs on the Astrodon filters yet, so I am using typical filter values that came with SkyTools.   


I then decided to see what it would be at Bortle 2 skies with my other conditions the same.  It suggested 4 exposures of 28 minutes with an SNR of 19. 


The interesting thing is that if I increased the length of the 40 second sub exposure for my conditions, it did not help my SNR. I have always been trying to maximize sub exposure lengths and had used the histrogram as a guide.  I don't know how accurately Sky Tools can simulate my actual conditions as the area in the sky that I would be imaging the Helix in is the darkest part of my sky. So I might have to fudge SkyTools results a bit, but it gives me a ballpark number.  


The other thing that has me confused is this that for the color filters it chooses different sub exposures for each filter.  I thought the idea was to get equal sub exposures and number of sub exposures with the RGB filters if they were balanced like the E-Gen2 Astrodons. If I use different sub exposure lengths and different number of subs, won't that throw off the colors during processing?





#2 Madratter


    Voyager 1

  • *****
  • Posts: 10793
  • Joined: 14 Jan 2013

Posted 07 August 2014 - 09:36 AM

It will give you the minimum time for best SNR. However, typically longer exposures do not hurt that SNR. And they are far more practical. You can set the minimum time it will use. Of far more interest is playing with the times of exposure. Tonight with the Moon is a good example. How much SNR ratio improvement you get/don't get imaging while the Moon is up is instructive. For example the best single period is from 3:45 to 4:25. Using just that gives a SNR of 11 for me. Imaging with the Moon up from 11 on improves it to 24.

#3 JoLo


    Ranger 4

  • -----
  • Posts: 353
  • Joined: 09 Feb 2010

Posted 07 August 2014 - 02:57 PM

I keep it simple.  5 minutes at home with pretty good sky conditions, 10 minutes at my dark site.  It probably is not the absolutely best, most accurate exposure for the highest SNR possible, but it produces nice pics for me.  It has also allowed me to build a dark master library with my new cooled CCD with various 5 and 10 minute masters with different temperatures.


If you use different sub exposure lengths, that means different sets of dark frames if they are far enough apart....me no do.  I will, however, vary the number of exposures between filters depending on the target.  For instance, I imaged the Iris Nebula this weekend and did 10x5 min on the luminance and blue channels, and 6x5 min on the green and red filters.  It really helped me to bring out the blue-ness of the Iris.  Another example might be M45 with a little extra blue, or the Eagle or Lagoon with a little extra red. 


For the majority of targets, however, I keep the RGB numbers the same, and do 1.5 to 2 times the number of luminance.



#4 Jon Rista

Jon Rista


  • *****
  • Posts: 21600
  • Joined: 10 Jan 2014

Posted 07 August 2014 - 06:20 PM

The problem with imaging under light polluted skies is they limit your background sky level. You can expose until all three color channels (for OSC) or the single luminance peak (for mono) have fully separated from the left-hand edge of the histogram...however any further exposure beyond that point is really not going to improve anything, since the minimum level of all pixels is also going to increase. That effectively keeps pushing up your noise floor. If you can fully separate your histogram under LP skies in three minutes (there should be a bit of a gap between the peak in the histogram and the left edge), then exposing for longer than that is not really going to do you any good.

Ironically, I actually experience a bit of a quirk when I go from imaging in my back yard, to imaging at a dark site. I usually expose for about four minutes at home, and my red channel when using an Astronomik CLS filter usually fully separates (sometimes I have to use 270 seconds). At a dark site, I have to expose for about double that, at the same aperture and ISO, to fully get that red channel above the read noise floor. That caught me off guard, but as it turned out, the LP was adding to the exposure of the actual DSO, and lifting it above the read noise floor quicker. My first few exposures (with a Canon 7D) at a dark site were actually pretty much duds, as the majority of the red channel was buried in the read noise, and I did not have enough subs to combat it (and I calibrated with darks, biases, and flats.)

Things may be different with a good CCD. If the camera has nice random read noise, you might not experience exactly the same problem I did, but there really is a LOT less light out at a dark site, so it is not surprising that you can take much longer exposures, but fewer of them.

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