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

ISO and Exposer time question

  • Please log in to reply
26 replies to this topic

#1 droe

droe

    Viking 1

  • -----
  • topic starter
  • Posts: 603
  • Joined: 01 Jul 2015
  • Loc: Fenton, Mi

Posted 18 February 2018 - 12:05 AM

So last night I decided to shoot the Crab nebula. I started at 800 ISO and 5 minute exposers. After an hour I decide to try 400 ISO at 10 minute. I never shot 10 minute exposers before and I did not know what to expect. Now use BackyardNikon and I put the image on full screen mode so I can see the difference in the images when they change. What was surprising was after 10 minutes and the image changed, I could not see a different between the 5 minute exposer and the 10 minute exposer. Was the jump from ISO 800 and ISO 400 a perfect doubling of exposer time or what? Is there any worth in going from 5 to 10 minute exposer with different ISOs?

 

My 2 images, first one is the ISO 800/5min and the second one is ISO 400/10mins. (shot in raw but converted to jpg so they could go on the website)

Attached Thumbnails

  • CRAB_LIGHT_300s_800iso_127mm_20180216-22h48m26s463msSM.jpg

Edited by droe, 18 February 2018 - 12:08 AM.


#2 droe

droe

    Viking 1

  • -----
  • topic starter
  • Posts: 603
  • Joined: 01 Jul 2015
  • Loc: Fenton, Mi

Posted 18 February 2018 - 12:06 AM

So last night I decided to shoot the Crab nebula. I started at 800 ISO and 5 minute exposers. After an hour I decide to try 400 ISO at 10 minute. I never shot 10 minute exposers before and I did not know what to expect. Now use BackyardNikon and I put the image on full screen mode so I can see the difference in the images when they change. What was surprising was after 10 minutes and the image change I could not see a different between the 5 minute exposer and the 10 minute exposer. Was the jump from ISO 800 and ISO 400 a perfect doubling of exposer time or what? Is there any worth in going from 5 to 10 minute exposer with different ISOs?

 

My 2 images, first one is the ISO 800/5min and the second one is ISO 400/10mins. (shot in raw but converted to jpg so they could go on the website)

ISO400/10min exposer

Attached Thumbnails

  • CRAB_LIGHT_600s_400iso_127mm_20180216-22h54m17s525msSM.jpg

Edited by droe, 18 February 2018 - 12:07 AM.


#3 17.5Dob

17.5Dob

    Soyuz

  • *****
  • Posts: 3895
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 18 February 2018 - 01:11 AM

FWIW, both images are extremely over exposed......

Your histo is at 1/2 way !!  I like mine just but barely off the left edge.

I shoot ISO 200 and 6 min exclusively.

Single RAW sub in preview

38573726481_c18935f8d4_z.jpg

And after
25513309528_1466fbb5e0_c.jpg


  • gilbertgrape87 and Aduffy1976 like this

#4 SKYGZR

SKYGZR

    Vanguard

  • *****
  • Posts: 2015
  • Joined: 13 Aug 2009
  • Loc: Planet Earth, Milky Way Galaxy, Speeding towards the Virgo Supercluster

Posted 18 February 2018 - 01:23 AM

Yes, "play" with your exposure times....set BYNIK to take a "series" of different times & length's . Start low, and work high...or vice-versa....average multiples will have a better result when combined/stacked



#5 Samir Kharusi

Samir Kharusi

    Vanguard

  • *****
  • Posts: 2461
  • Joined: 14 Jun 2005
  • Loc: Oman

Posted 18 February 2018 - 03:13 AM

The ISO used is unimportant except for dynamic range (OK, also Read Noise, stay at 200 or above and you ought to be fine). What is important is one exposure was 5 minutes and the other was 10 minutes. Process the 2 frames individually, each to the best of your ability as astro images, and there should be an obvious difference in the Signal to Noise Ratio. Reason you do not see a difference is because those frames were processed like we do for daytime shots. A 1000th of a second exposure at ISO 1000 on a sunlit beach will look essentially identical to 1/100th second at ISO 100. Your 10 minute frame at ISO 400, after ASTRO processing, should look the same as two frames each 5 minutes long at ISO 800, but stacked. NB Stacking the same frame does NOT improve the SNR since you are not doubling the Integration Time. Here is an example of doubling Integration Time and SNR:
hist_1to56x2min.jpg

 

PS the comments made about over-exposure: The ISO setting does affect the dynamic range available but otherwise does not affect what SNR you end up with after astro processing your stacked frames. Rule-of-thumb is to expose at your chosen ISO so that the skyfog peak on the Back-of-Camera Histogram is at about one third along the X-axis. NB: the BoC Histogram is NOT the same as the Linear Histogram in post processing software. The latter shows a histogram that has not had a gamma-stretch and hence will show the skyfog peak very close to the origin. Easiest is to pay attention to the camera's BoC Histogram, instantly available in the field, no laptop required. 50% along the X-axis on the BoC Histogram still has plenty of headroom, but 50% on a Linear Histogram hardly has any headroom (only a factor 2x), so it's important to state what histogram is being described. From the appearance of the Jpeg versions posted, I would guess that the BoC Histogram is roughly at a third to a half (i.e. plenty of headroom). A Linear version of the same frames would appear almost black. Of course, I may be mistaken...


Edited by Samir Kharusi, 18 February 2018 - 03:14 AM.

  • timmbottoni, sharkmelley, bobzeq25 and 1 other like this

#6 gilbertgrape87

gilbertgrape87

    Vostok 1

  • -----
  • Posts: 137
  • Joined: 30 Sep 2017
  • Loc: Mesa, AZ

Posted 18 February 2018 - 06:12 AM

FWIW, both images are extremely over exposed......

Your histo is at 1/2 way !!  I like mine just but barely off the left edge.

I shoot ISO 200 and 6 min exclusively.

Single RAW sub in preview

38573726481_c18935f8d4_z.jpg

And after
25513309528_1466fbb5e0_c.jpg

As a beginner myself, this post has given me SO much more insight to exposure times than anything else I've read (and I've read a lot).

 

People always talk about using the stretched histogram on the back of the camera, but for those that use software to capture, you don't get that luxury

 

So seeing an example of a raw histogram that close to the floor being able to extract such fine filament detail is incredibly enlightening.

 

Thank you!



#7 droe

droe

    Viking 1

  • -----
  • topic starter
  • Posts: 603
  • Joined: 01 Jul 2015
  • Loc: Fenton, Mi

Posted 18 February 2018 - 04:16 PM

Thanks everyone for all the great info. It's a great help.

Samir - you were spot on with "From the appearance of the Jpeg versions posted, I would guess that the BoC Histogram is roughly at a third to a half (i.e. plenty of headroom)."

After stacking and processing this image it turned out good but not great. I need to get better at data acquisition and learning proper exposed times seem pretty important.

Thanks again.

#8 Samir Kharusi

Samir Kharusi

    Vanguard

  • *****
  • Posts: 2461
  • Joined: 14 Jun 2005
  • Loc: Oman

Posted 20 February 2018 - 03:29 AM

Your data acquisition is great. You are in great shape as long as you can track precisely and satisfy Rule 1. below. The only issue left is to collect a LOT of Integration Time. Most beginners under-estimate how much IT they need to get a great result. These rules-of-thumb will guarantee a nice result each time:

 

1. Your sub-exposure length has to be long enough. Minimal is 1-minute at f2.8, 2-minutes at f4, 4 minutes at f5.6, 8 minutes at f8, etc. If your mount is not precise enough for these lengths then you should use a shorter focal length/faster f-ratio lens.

2. Your Integration Time has to be a LOT! Not less than an hour at a dark site for an almost "satisfactory" result. You can NEVER have too much; and every time you double it, you will see an improvement.

3. If you are at a light polluted site you will need much, much more Integration Time. Typical outer suburbia already requires 15 hours just to equal the SNR from one hour at a dark site. Measure the light pollution at your home site so you know what to expect. It's trivially simple to do the measurement.

 

Less important, but unfortunately too many beginners keep worrying about:

 

4. ISO setting does not really matter much. Just stay at 200 or higher. For very minor optimisation (better quantisation in your camera's Analog/Digital Converter) use a high enough ISO that your BoC Histogram shows the skyfog peak at around 1/3rd along the X-axis.

 

Go forth and click away hundreds of subs on one target, much more rewarding than slewing around onto several targets each night.


  • timmbottoni and michele like this

#9 droe

droe

    Viking 1

  • -----
  • topic starter
  • Posts: 603
  • Joined: 01 Jul 2015
  • Loc: Fenton, Mi

Posted 20 February 2018 - 10:48 AM

Thanks Samir,

 

I fall into all those beginner traps. My biggest problem is IT and wanting to take an image of everything. I have now spent a year doing this and got some nice images of everything around the entire sky. This next year I will focus on just a few seasonal targets and put in place the suggestions you made about IT for some awesome pictures. I appreciate all your help.


  • timmbottoni likes this

#10 Melgabean

Melgabean

    Sputnik

  • -----
  • Posts: 42
  • Joined: 15 Jan 2018
  • Loc: Colorado

Posted 20 February 2018 - 08:12 PM

I was actually going to post something similar to this. So what I'm gathering from this post is that 1 10 min exposure at iso 400 will look the same as 2 5 min exposures at iso 800 stacked. So the key to capturing an excellent photo is just more integration time and not so much about the iso and exposure used.

So get the BOC histogram to 1/3 at whichever iso I want and then take tons of images to equal as many hours as I can handle of integration time?

What would be the point of doing longer exposures at a lower iso vs many short exposures at a higher iso? Fewer images to deal with?

#11 Samir Kharusi

Samir Kharusi

    Vanguard

  • *****
  • Posts: 2461
  • Joined: 14 Jun 2005
  • Loc: Oman

Posted 21 February 2018 - 01:25 AM

"What would be the point of doing longer exposures at a lower iso vs many short exposures at a higher iso? Fewer images to deal with?"

 

OK, we are now delving into the secondary intricacies. Rule No.1 above states that the subexposure has to be long enough. Long enough means that the skyfog peak has to be much higher (>>30x) than the Read Noise of your camera. Basically your skyfog mountain peak on the BoC Histogram has to be entirely separated from the X-origin. Now the strange part: The more your local sky is light polluted, the shorter the exposure it requires to achieve this separation between the Skyfog Peak and the X-origin (where the Read Noise resides). E.g. at a dark site and f2.8 you need a one-minute subexposure to separate entirely your Skyfog Peak from the X-origin. At your average suburb of a large city the skyfog is somewhere between 15x and 40x as bright as at a pristine dark site. Let's focus on the more optimistic 15x. So you need an exposure of one minute divided by 15 to effect the same separation at f2.8 as at the dark site. Four seconds grin.gif  Nice! There must be a catch... The catch is that your Integration Time also has to be 15x longer. To get the same SNR Signal to Noise Ratio) in your final, processed stack at the dark site (one hour IT) and your suburbian site you need 15 hours IT. You can achieve 15 hours IT at the suburb by shooting 13,500x4second subs at an ISO, say, of 1600, or 900x60second subs at an ISO of 200. Yes, indeed, fewer images to deal with. Minimal subs explained here and for those keen on the intricacies, more explained here.

 

What do most of us do in real life? We shoot 100x60second subs in suburbia. And then we complain that the SNR is poor. Of course it's lousy since it is the equivalent of less than 7 minutes at a dark site (= 100/15). Just follow the earlier mentioned rules-of-thumb and shoot all night on one target. Your images will be almost satisfying. For "great" images follow the same rules at a dark site + all night on one target.  


  • timmbottoni and Tele like this

#12 Melgabean

Melgabean

    Sputnik

  • -----
  • Posts: 42
  • Joined: 15 Jan 2018
  • Loc: Colorado

Posted 21 February 2018 - 03:18 AM

Samir you have been super helpful! That was an excellent explanation.
  • timmbottoni likes this

#13 gilbertgrape87

gilbertgrape87

    Vostok 1

  • -----
  • Posts: 137
  • Joined: 30 Sep 2017
  • Loc: Mesa, AZ

Posted 21 February 2018 - 05:36 AM

There is so much talk regarding the BoC histogram, but for those that regularly usr a computer to capture, that's not necessarily something that's available right?

Is the auto stretch that BackyardEOS performs the same level of stretch that Canon cameras do built in? Or SGP? In other words: does this rule of thumb apply to the auto stretches of popular capture software? The reason I ask is because if I stick to 1/3-1/2 of the histogram in BYE, I am regularly clipping stars (at least I am assuming that the reason why my stars have no color is because they are all getting clipped).
  • timmbottoni likes this

#14 Samir Kharusi

Samir Kharusi

    Vanguard

  • *****
  • Posts: 2461
  • Joined: 14 Jun 2005
  • Loc: Oman

Posted 22 February 2018 - 06:07 AM

BoC histogram does not involve much effort because you need only one frame to check it. Before you start your astro capture run, just take a frame and see what the BoC histogram looks like. Or a few if you are unsure what ISO is appropriate for your conditions. The beauty of BoC histogram is that it is set up more or less identically no matter what camera make/model you use. There is a gamma stretch applied that makes your image look similar to another from a totally different camera as long as f-stop, sub exposure length and ISO have been set the same. Auto stretch in astro software has little (nothing?) to do with producing nice daytime photos so auto-stretch depends on too many other factors. A suggestion: take a photo of a blank wall and set the exposure so that you get a one-third along the X-axis peak representing the blank wall, and a second photo with the peak at 50%. Open both frames in your astro software and see where those peaks are at in the software's histogram. Remember the two locations along the X-axis for all your future astro field activities.


  • Tonk and timmbottoni like this

#15 squreshi786

squreshi786

    Vostok 1

  • -----
  • Posts: 132
  • Joined: 26 Feb 2018

Posted 28 February 2018 - 03:37 PM

hi guys

 

some what related could I ask what should be the ADU of the flats or exposure settings for flats when taken with lights at ISO 200  or if taken at ISO 1600. Unfortunately this is an area where I don't think there is lot of sage advise and I would appreciate if canon users specially 6D users can share there expertise in under what Backyard EOS or SGPro settings do they take there FLATS.

 

Thanks

 

Saqib



#16 Samir Kharusi

Samir Kharusi

    Vanguard

  • *****
  • Posts: 2461
  • Joined: 14 Jun 2005
  • Loc: Oman

Posted 01 March 2018 - 01:47 AM

I simply take Flats at the same ISO as the Lights. Recall that the Flats are used in a division arithmetic process, hence my internal logic says that the up/down excursion with brighter/darker should be similar to what has been used in the Lights. I have NOT spent any time actually thinking about it undecided.gif . Also it would seem to be critical that the overall average exposure for the Flats ought to be somewhere the camera behaves linearly, i.e. somewhere in the middle of the BoC Histogram, not near over-exposure and saturation. Hence just use your camera's auto-exposure setting and flip to manual with the same setting, so that all your Flats receive the same exposure. Frankly, the only critical part is that the Flats are exposed to very even illumination. Sometimes this can be quite a challenge, e.g. Flats using a Hyperstar, because Hyperstars have severe vignetting. So, to work well, I found it best to use sky Flats when using a Hyperstar, usually dawn sky. But twilight would also be great. When using small OTAs such as refractors or camera lenses I found that a diffuser (e.g. paper or t-shirt) in front of the lens hood (dew shield) of a refractor is good enough. The reason is that, compared to Hyperstars, lenses generally show little vignetting.

 

PS. You can easily check if the process of taking your Flats is valid. Take a bunch of Flats as per your standard procedure, say, at ISO 200 plus one at ISO 1600. Stack the lot at ISO 200. Use this Stack to check how well it flattens any one of the Flats at ISO 200. Use an eyedropper read out on the flattened Flat to gauge the "flatness". I suspect that if you use a Flat taken at ISO 1600 the flattening may not work as well. If it does, no worry! Lights and Flats need not be at the same ISO. I expect that it does matter confused1.gif



#17 timmbottoni

timmbottoni

    Vanguard

  • *****
  • Posts: 2091
  • Joined: 25 Aug 2005
  • Loc: W Chicago suburbs, IL USA

Posted 01 March 2018 - 11:42 AM

I posted this question to Samir as a PM, but will share it here in case anyone else has opinions, and to see what others who are much smarter than me think.

 

I am very familiar with the basics of astrophotography but haven't really gotten serious about it just yet.

A General Question:  Have you seen this article, and what do you think of it?  http://dslr-astropho...rophotography/¬†

 

I have a Canon 60D, and have tried using the Astronomik CLS EOS Clip filter, and it seems to work but just OK, as it creates a very blue image. I live in very light polluted western suburbs of the Chicago area (Bortle 7). I can autoguide if needed, and can get decent short exposures without autoguiding as well. I have Pixinsight, and have the book "Inside Pixinsight" as well, but am still very much inexperienced at this, so here is the big question.

 

I'm using my William Optics FPL53 doublet APO 61mm ZS61 (focal length 360mm, focal ratio F/5.9) and my Canon 60D, and wanting to shoot something like the popular Rosette Nebula this weekend (and here are the detailed questions).

1. What ISO should I use? Seems like 800 is best or should I go 1600?
2. Should I use the CLS Astromik filter, or just try to process out the light pollution in Pixinsight?
3. How long of total integration will I need to get decent results?

 

Thanks 

 

Timm 



#18 gilbertgrape87

gilbertgrape87

    Vostok 1

  • -----
  • Posts: 137
  • Joined: 30 Sep 2017
  • Loc: Mesa, AZ

Posted 01 March 2018 - 12:20 PM

Is your 60D modified? If not, I haven't used a CLS for a while and I don't miss it. When I did use it, it gave the blue images, but that only made it marginally harder to fix (in my opinion). I would still probably use it on emission nebula to try and get longer exposures so you don't need a bajillion subs.

 

If you ARE modified (which I would strongly recommend if you want to use the camera for emission nebula), I would GUESS that you would need one (or a luminance filter) to cut out the IR emissions on those targets.

 

I'll let more experienced people address questions 1 and 3, but here's a crude example.  This was taken on 12/11/17 with a full spectrum modified Canon 1000D, CLS-CCD filter, and about 4 hours total exposure. The ONLY processing done on this was one pass automatic background extraction and STF just to give you an idea of the data underneath.

Attached Thumbnails

  • 15199267106930.jpg

Edited by gilbertgrape87, 01 March 2018 - 12:54 PM.

  • timmbottoni likes this

#19 timmbottoni

timmbottoni

    Vanguard

  • *****
  • Posts: 2091
  • Joined: 25 Aug 2005
  • Loc: W Chicago suburbs, IL USA

Posted 01 March 2018 - 12:56 PM

Is your 60D modified? If not, I haven't used a CLS for a while and I don't miss it. When I did use it, it gave the blue images, but that only made it marginally harder to fix (in my opinion). I would still probably use it on emission nebula to try and get longer exposures so you don't need a bajillion subs.

 

If you ARE modified (which I would strongly recommend if you want to use the camera for emission nebula), I would GUESS that you would need one (or a luminance filter) to cut out the IR emissions on those targets.

 

I'll let more experienced people address questions 1 and 3, but here's a crude example.  This was taken on 12/11/17 with a full spectrum modified Canon 1000D, CLS-CCD filter, and about 4 hours total exposure. The ONLY processing done on this was one pass automatic background extraction and STF just to give you an idea of the data underneath.

My Canon 60D is NOT modified.

 

Thanks for the reply!

 

Timm



#20 Samir Kharusi

Samir Kharusi

    Vanguard

  • *****
  • Posts: 2461
  • Joined: 14 Jun 2005
  • Loc: Oman

Posted 01 March 2018 - 07:57 PM

With unmodified cameras you are adding needless burdens trying to shoot Ha nebulae. Choose targets that have plenty of colors other than deep red Ha, e.g. M42, M31, etc. Color balancing the CLS and its effect on filtering available colors are explained here. Frankly, one can go on guessing and second guessing but a step by step approach will lead you to excellent results much more quickly. IMHO these are the essential steps:

 

A. Measure your local skyfog at zenith. Here's how.

B. Determine the length of the longest subs you can shoot with and without autoguiding (at least 60+% yield well-tracked). State what focal length and f-ratio OTA you want to use.

 

The two above results will let us have a meaningful conversation about all other details. Otherwise answering one query just brings up another... Now, as to Timm's queries:

 

It sounds like you have a lot of light pollution, so

 

1. Use ISO 800 or even as far down as 200. Does not matter. Just keep your BoC Histogram between 20 and 30% with your chosen length of subs.

2. Do not use the CLS since your camera does not record much Ha anyway. Rosette is the wrong target for unmodded cameras.

3. You can NEVER have enough Integration Time. 40 hours ought to be OK but you can start with 3... 

 

People will tell you that they have successfully shot Ha with unmodded cameras. Yes, with supreme endurance, you can. You can also shoot IR with unmodded cameras but why suffer?


  • timmbottoni likes this

#21 timmbottoni

timmbottoni

    Vanguard

  • *****
  • Posts: 2091
  • Joined: 25 Aug 2005
  • Loc: W Chicago suburbs, IL USA

Posted 01 March 2018 - 09:40 PM

With unmodified cameras you are adding needless burdens trying to shoot Ha nebulae. Choose targets that have plenty of colors other than deep red Ha, e.g. M42, M31, etc. Color balancing the CLS and its effect on filtering available colors are explained here. Frankly, one can go on guessing and second guessing but a step by step approach will lead you to excellent results much more quickly. IMHO these are the essential steps:

 

A. Measure your local skyfog at zenith. Here's how.

B. Determine the length of the longest subs you can shoot with and without autoguiding (at least 60+% yield well-tracked). State what focal length and f-ratio OTA you want to use.

 

The two above results will let us have a meaningful conversation about all other details. Otherwise answering one query just brings up another... Now, as to Timm's queries:

 

It sounds like you have a lot of light pollution, so

 

1. Use ISO 800 or even as far down as 200. Does not matter. Just keep your BoC Histogram between 20 and 30% with your chosen length of subs.

2. Do not use the CLS since your camera does not record much Ha anyway. Rosette is the wrong target for unmodded cameras.

3. You can NEVER have enough Integration Time. 40 hours ought to be OK but you can start with 3... 

 

People will tell you that they have successfully shot Ha with unmodded cameras. Yes, with supreme endurance, you can. You can also shoot IR with unmodded cameras but why suffer?

WOW  - thanks for so much good free advice.  I will search for better targets and measure my light pollution. 

 

Update:  It was clear out so I did the test following the instructions and hit mid point at 25 seconds, which isn't on your chart.  So, yeah, its really bad here.

 

Not exactly what I was hoping for, of course, but very very useful

 

Timm


Edited by timmbottoni, 01 March 2018 - 11:55 PM.


#22 Samir Kharusi

Samir Kharusi

    Vanguard

  • *****
  • Posts: 2461
  • Joined: 14 Jun 2005
  • Loc: Oman

Posted 02 March 2018 - 03:07 AM

OK, now we are off to a good analysis. 25 sec (=0.42 minutes) to mid BoC Histogram at f4 and ISO 800. That means that your skyfog is at Visual Limiting Magnitude 3.5 (you can just about make out the head of Orion and you cannot make out the Milky Way at all). That is the same as roughly Mag 17.5 /sq arc-sec, suburb of a large city. A dark site is at Mag 21.5/sq arc-sec. Your site has a sky that has 4 Stellar Magnitudes brighter sky fog. Each Stellar Mag = 2.512x so your sky is 2.512^4 or 40x brighter than at a dark site. I.e. 40 Hours Integration Time at your site will give similar SNR (Signal to Noise Ratio) as one hour at a dark site. Note that typically you need 3+ hours at a dark site IT to get a very nice DSO image, so you have a major battle ahead! 120 hours IT ! The good news is that you can use reasonably short sub exposures, even at f5.9 and your skyfog will still be well detached from the origin. I suggest that you set ISO to 200 and use the longest subs you can track unguided. With so much skyfog it's hardly worth the bother of autoguiding. 2-minute subs at ISO 200 should show some detachment from the origin. If 2-minutes is too long to track well, just use one minute at ISO 400. As long as you can see a gap between the origin and the skyfog peak your subs will stack satisfactorily and you can skip autoguiding. The main headache is to shoot an awful of of subs! Each night you can get a max of around 3 hours IT before the target gets low near the horizon, but 180x1 min or 90x2 min is a good start.

 

What can you do to improve your situation? Get the camera modded and then you can use Ha narrowband filters. Frankly, your skyfog is too much even for the usual broadband light pollution filters like the CLS. Read this. Go for 7 or 8nm Ha and the narrowest OIII filters  (12nm?) that are still reasonably priced. People are often scared to mod their cameras. No need. Get somebody commercial to mod it for you (with a rectangular bandpass UV-IR Blocker) and your camera will still work very well for daytime pics of Grandma. You just use Custom White Balance instead of Auto WB. If you go full-spectrum you will have the headache of sourcing a clip-in UV-IR Blocker for your camera model. Sort this out first, before choosing full-spectrum or you may later find out that you cannot use all your camera lenses with that camera. There are plenty of clip-in filters for APS-C cameras but more limited availability for full 35mm format.


  • timmbottoni likes this

#23 timmbottoni

timmbottoni

    Vanguard

  • *****
  • Posts: 2091
  • Joined: 25 Aug 2005
  • Loc: W Chicago suburbs, IL USA

Posted 02 March 2018 - 09:36 AM

OK, now we are off to a good analysis. 25 sec (=0.42 minutes) to mid BoC Histogram at f4 and ISO 800. That means that your skyfog is at Visual Limiting Magnitude 3.5 (you can just about make out the head of Orion and you cannot make out the Milky Way at all). That is the same as roughly Mag 17.5 /sq arc-sec, suburb of a large city. A dark site is at Mag 21.5/sq arc-sec. Your site has a sky that has 4 Stellar Magnitudes brighter sky fog. Each Stellar Mag = 2.512x so your sky is 2.512^4 or 40x brighter than at a dark site. I.e. 40 Hours Integration Time at your site will give similar SNR (Signal to Noise Ratio) as one hour at a dark site. Note that typically you need 3+ hours at a dark site IT to get a very nice DSO image, so you have a major battle ahead! 120 hours IT ! The good news is that you can use reasonably short sub exposures, even at f5.9 and your skyfog will still be well detached from the origin. I suggest that you set ISO to 200 and use the longest subs you can track unguided. With so much skyfog it's hardly worth the bother of autoguiding. 2-minute subs at ISO 200 should show some detachment from the origin. If 2-minutes is too long to track well, just use one minute at ISO 400. As long as you can see a gap between the origin and the skyfog peak your subs will stack satisfactorily and you can skip autoguiding. The main headache is to shoot an awful of of subs! Each night you can get a max of around 3 hours IT before the target gets low near the horizon, but 180x1 min or 90x2 min is a good start.

 

What can you do to improve your situation? Get the camera modded and then you can use Ha narrowband filters. Frankly, your skyfog is too much even for the usual broadband light pollution filters like the CLS. Read this. Go for 7 or 8nm Ha and the narrowest OIII filters  (12nm?) that are still reasonably priced. People are often scared to mod their cameras. No need. Get somebody commercial to mod it for you (with a rectangular bandpass UV-IR Blocker) and your camera will still work very well for daytime pics of Grandma. You just use Custom White Balance instead of Auto WB. If you go full-spectrum you will have the headache of sourcing a clip-in UV-IR Blocker for your camera model. Sort this out first, before choosing full-spectrum or you may later find out that you cannot use all your camera lenses with that camera. There are plenty of clip-in filters for APS-C cameras but more limited availability for full 35mm format.

Thank you very much Samir!!!!

Very interesting and useful. bow.gif waytogo.gif

 

Translation - I picked the worst possible place on earth for this hobby - foreheadslap.gif

Yes, I am in the suburbs of a large city.  Location is here ---

 https://www.lightpol...ers=B0FFFTFFFF¬†

 

Seriously, I am still having fun, and learning, so that when I do move somewhere dark, I will know what I'm doing.  If I start to get more serious, I think I can see the value in spending enough money to get into a dedicated camera and narrow band.

 

I hoping to try some AP with clear skies the next couple of nights, and with the ZS61 I can do a couple of minutes unguided so I will try some things you suggested if the clear sky forecast isn't wrong.  I've had zero chance to try anything this Winter due to cold, snow, wind, clouds, other things going on, etc. so just getting out and trying on something will be fun.

 

Clear skies!

 

Timm 



#24 bobzeq25

bobzeq25

    Voyager 1

  • *****
  • Posts: 12903
  • Joined: 27 Oct 2014

Posted 02 March 2018 - 10:40 AM

OK, now we are off to a good analysis. 25 sec (=0.42 minutes) to mid BoC Histogram at f4 and ISO 800. That means that your skyfog is at Visual Limiting Magnitude 3.5 (you can just about make out the head of Orion and you cannot make out the Milky Way at all). That is the same as roughly Mag 17.5 /sq arc-sec, suburb of a large city. A dark site is at Mag 21.5/sq arc-sec. Your site has a sky that has 4 Stellar Magnitudes brighter sky fog. Each Stellar Mag = 2.512x so your sky is 2.512^4 or 40x brighter than at a dark site. I.e. 40 Hours Integration Time at your site will give similar SNR (Signal to Noise Ratio) as one hour at a dark site. Note that typically you need 3+ hours at a dark site IT to get a very nice DSO image, so you have a major battle ahead! 120 hours IT ! The good news is that you can use reasonably short sub exposures, even at f5.9 and your skyfog will still be well detached from the origin. I suggest that you set ISO to 200 and use the longest subs you can track unguided. With so much skyfog it's hardly worth the bother of autoguiding. 2-minute subs at ISO 200 should show some detachment from the origin. If 2-minutes is too long to track well, just use one minute at ISO 400. As long as you can see a gap between the origin and the skyfog peak your subs will stack satisfactorily and you can skip autoguiding. The main headache is to shoot an awful of of subs! Each night you can get a max of around 3 hours IT before the target gets low near the horizon, but 180x1 min or 90x2 min is a good start.
 
What can you do to improve your situation? Get the camera modded and then you can use Ha narrowband filters. Frankly, your skyfog is too much even for the usual broadband light pollution filters like the CLS. Read this. Go for 7 or 8nm Ha and the narrowest OIII filters  (12nm?) that are still reasonably priced. People are often scared to mod their cameras. No need. Get somebody commercial to mod it for you (with a rectangular bandpass UV-IR Blocker) and your camera will still work very well for daytime pics of Grandma. You just use Custom White Balance instead of Auto WB. If you go full-spectrum you will have the headache of sourcing a clip-in UV-IR Blocker for your camera model. Sort this out first, before choosing full-spectrum or you may later find out that you cannot use all your camera lenses with that camera. There are plenty of clip-in filters for APS-C cameras but more limited availability for full 35mm format.

Thank you very much Samir!!!!
Very interesting and useful. bow.gif waytogo.gif
 
Translation - I picked the worst possible place on earth for this hobby - foreheadslap.gif
Yes, I am in the suburbs of a large city.  Location is here ---
 https://www.lightpol...ers=B0FFFTFFFF¬†
 
Seriously, I am still having fun, and learning, so that when I do move somewhere dark, I will know what I'm doing.  If I start to get more serious, I think I can see the value in spending enough money to get into a dedicated camera and narrow band.
 
I hoping to try some AP with clear skies the next couple of nights, and with the ZS61 I can do a couple of minutes unguided so I will try some things you suggested if the clear sky forecast isn't wrong.  I've had zero chance to try anything this Winter due to cold, snow, wind, clouds, other things going on, etc. so just getting out and trying on something will be fun.
 
Clear skies!
 
Timm

Samir's advice is good, but do not despair.  I have a good deal of experience imaging in light pollution.  Look at my astrobin, mostly done in skies only a bit better than yours.  Red Zone, Bortle 7, mag per arc sec squared low 18s.
 
There's some narrowband, especially more recently, but some broadband also, it's not impossible.  I'm not a big fan of broadband light pollution filters, some use them, some don't.
 
Good gradient reduction in processing is essential is those skies.  There are good tools in PixInsight, Star Tools, Astro Pixel Processor.  The Gradient xTerminator addon for Photoshop.  More total imaging time is also essential.  My rule of thumb is that one hour is minimal, two better, four is good.  Serious folks do ten or more, imaging over multiple nights.
 
A couple of things, less important.  LRGB works better than One Shot Color in light pollution.  It gathers data faster, and more data is key.   The filters have a notch where sodium vapor lights emit, that's useful.
 
One darn good idea is to go halfway.  Get just an H alpha filter (O(III) is substantially more difficult).  Either mod the camera or get one of the new CMOS (mono).  Shoot H alpha in black and white, which can be really nice.  You'll see some of that on my astrobin.  I do have a good mount (CEM60), I do 10-15 minute exposures.  You either need a low read noise CMOS camera or, a good mount to do longish exposures.

Edited by bobzeq25, 02 March 2018 - 11:25 AM.

  • timmbottoni likes this

#25 Endro32

Endro32

    Lift Off

  • -----
  • Posts: 8
  • Joined: 24 Apr 2017
  • Loc: Rives Junction, Michigan

Posted 02 March 2018 - 12:05 PM

Definitely would recommend shooting in RAW if your camera supports it. This will allow you to adjust the ISO in your image-editing software if you need to.




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