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

Long exposure question.

  • Please log in to reply
8 replies to this topic

#1 Ghostnotes

Ghostnotes

    Lift Off

  • -----
  • topic starter
  • Posts: 8
  • Joined: 15 Nov 2021

Posted 11 August 2022 - 10:20 PM

One thing I love seeing in S&T are the pictures sent in by their readers.They always include the equipment information exposure details etc... However one thing I never really got is how some of these exposures range out to 20/30/40hrs. Is it a matter of tracking all night then picking up where you left off the next night by means of requiring the object, which I'm guessing is so.Is there a particular technique? I've really never seen an article to where it goes into detail about doing this. Some of the details describe multiple photos and stacking which I understand, but are some of these actually 5-10 hour single exposures? Not that I'm going to try in the immediate future. It's more for just understanding.


Edited by Ghostnotes, 11 August 2022 - 10:21 PM.


#2 Jeffmar

Jeffmar

    Surveyor 1

  • -----
  • Posts: 1,955
  • Joined: 18 Mar 2012
  • Loc: salt lake city, utah

Posted 11 August 2022 - 10:48 PM

There is a limit on how long the exposure can be for most objects. If your exposure is too long the bright objects get blown out and you get white blobs with no detail. This happens because the pixels on the camera sensor get overloaded. You can decrease noise and increase the signal by either taking some very long exposures or taking many more shorter exposures. Since there is an exposure limit where you start to lose detail on bright objects the best way seems to be doing many shorter exposures instead of one or a few very long exposures.

 

The exception is something like the days or weeks long exposures that the Hubble space telescope took of a supposedly empty section of sky and got thousands of galaxies. It worked because the galaxies were so dim it took many hours to gather enough photons to get images of them.

 

It all depends on how bright the target is. Bright targets such as the Orion Nebula, the Andromeda Galaxy, Pleiades, and M13  can be imaged with exposure times like 30 seconds or less, but the more images you get the better the final image will be. Some people even shorter exposures but get a huge number of images. Other targets such as faint nebulae, and galaxies that are relatively far away can require 5 to 10 minute exposures and you still need to get a lot of them for a quality image.

 

The more images you can get to stack the better your signal to noise ratio will be. Signal to noise ratio can be put on a curve where you get a lot of improvement doing the first 10 images, less doing the next 10, and even less improvement for the 10 images after that. A lot of people will limit their number of exposures to about 40 but other folks will go night after night imaging the same target and get hundreds of photos to stack. I suppose it all depends on how OCD your want to be.


Edited by Jeffmar, 11 August 2022 - 11:41 PM.


#3 D_talley

D_talley

    Aurora

  • *****
  • Posts: 4,527
  • Joined: 07 Jul 2005
  • Loc: Albuquerque, New Mexico

Posted 11 August 2022 - 11:03 PM

I think the OP is misunderstanding how long a single exposure is. No you don't take a 5 -10 hour single exposure of an object. As Jeffmar stated, you take numerous exposures to provide the best signal to noise ratio for that object.  The total time of all the exposures is what the people are reporting in S&T. 

I normally take 30 minute exposures on very dim object and try to get at least 30 or more exposures over a period of time.  that would total up to 15 hours. 

Another reason you would not want to take an exposure of 5-10 hours is what happens at 9 hours 30 minutes when your neighbor turns their backyard light on.. you wasted an entire exposure and night.  With shorter exposures you can just throw out the bad exposure and keep the others. 


  • rgsalinger, Jeffmar and Skysmacker like this

#4 Dynan

Dynan

    Fly Me to the Moon

  • *****
  • Posts: 6,310
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 11 August 2022 - 11:04 PM

Here is a great explanation of exposure length effectiveness:

 

https://www.youtube....h?v=3RH93UvP358

 

And remember, the longer your exposures are, the more satellites and airplanes you are inviting into your image. Although these can usually be processed out if not too bright.



#5 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 29,820
  • Joined: 27 Oct 2014

Posted 11 August 2022 - 11:07 PM

One thing I love seeing in S&T are the pictures sent in by their readers.They always include the equipment information exposure details etc... However one thing I never really got is how some of these exposures range out to 20/30/40hrs. Is it a matter of tracking all night then picking up where you left off the next night by means of requiring the object, which I'm guessing is so.Is there a particular technique? I've really never seen an article to where it goes into detail about doing this. Some of the details describe multiple photos and stacking which I understand, but are some of these actually 5-10 hour single exposures? Not that I'm going to try in the immediate future. It's more for just understanding.

Here's what you're missing.  People often shoot _hundreds_ of subexposures.  So, (just an illustration) 600 5 minute subexposures would be 50 hours.  What really counts here is the total time, NOT the subexposure.  Because we stack subs and the important thing is how many photons we captured, total.  How you divide the total into subexposures is just a tweak.

 

Here's an image I did of the Pleadies.  662 ten second exposures.  They needed to be short because it's an F2 scope, and I'm in light polluted skies.  Longer exposure, and my pixels would have been overloaded by light pollution.

 

https://www.astrobin.com/t5173s/

 

A rite of passage for an imager is shooting one target over multiple nights.  It used to be pretty difficult to get lined up on target really well.  So typically you'd have to crop the edges some, where you only had data for a single night.  No more.

 

There's a technique called platesolving.  You take a sub or an image.  Feed it into a computer which does pattern recognition on the stars, and tells you _exactly_ where you're pointed.  So you can go back to the precise spot.  It's amazingly simple.  Picture of the input/output screen for a platesolving program below.  Note that the error is less than an arc second, which is less than 1 pixel.

 

PlateSolve2-Screenshotv2.jpg

 

More important.  I saw your post in the beginners forum where you referenced AP and software.  Have you done any Deep Space photography?  If not, realize that the second biggest beginner error (after an inadequate mount, you dodged that one <smile> ) is too big a scope.  Your 8 inch SCT defines "too big a scope".  It's not impossible to learn on one.  Just a very hard very long process.  Not much fun.  The following are not outliers, they're the standard experience.

 

"After months of learning and overcoming challenges <with the SCT>, and finally buying a shorter FL APO refractor, I really really really wish I had listened to everyone on here and started learning the imaging basics on THAT frac instead of on the SCT.   Trust me"

 

"I regret spending the first 6 months trying to learn imaging with an 8" Edge, with that scope it was a losing effort. Fortunately got a nice little refractor, and not only have the quality of my images improved but I'm actually enjoying the process of learning how to do it!"

 

I have many more of those.

 

The scope would work fine for the planets and the Moon.  Completely different thing, there's a forum called Solar System Imaging.


Edited by bobzeq25, 11 August 2022 - 11:44 PM.

  • Dynan and Skysmacker like this

#6 Ghostnotes

Ghostnotes

    Lift Off

  • -----
  • topic starter
  • Posts: 8
  • Joined: 15 Nov 2021

Posted 11 August 2022 - 11:20 PM

Okay that makes sense to me. I did not know if the high end CCD/CMOS image cameras had some sort of metering to them that allowed super long exposures without blowing out the bright areas.

 

Here is a great explanation of exposure length effectiveness:

 

https://www.youtube....h?v=3RH93UvP358

 

And remember, the longer your exposures are, the more satellites and airplanes you are inviting into your image. Although these can usually be processed out if not too bright.

That is one of my issues. Depending on wind conditions and flight plan rotations, on a busy weekend when the flight plans call for inbound 270* L/R It's a plane every 2 minutes until 2am sometimes. Positioning puts them at about 30* and 70* north of my position. So anything north between those viewing angles will have nice streamers..



#7 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 29,820
  • Joined: 27 Oct 2014

Posted 12 August 2022 - 12:06 AM

Okay that makes sense to me. I did not know if the high end CCD/CMOS image cameras had some sort of metering to them that allowed super long exposures without blowing out the bright areas.

 

That is one of my issues. Depending on wind conditions and flight plan rotations, on a busy weekend when the flight plans call for inbound 270* L/R It's a plane every 2 minutes until 2am sometimes. Positioning puts them at about 30* and 70* north of my position. So anything north between those viewing angles will have nice streamers..

Take enough subs, and most processing programs can take out airplane trails (for most of us anyway, your situation is extreme).  They recognize airplane trails as outliers, and fill in those pixels with an average of the pixels not contaminated.  Doing it at the pixel level means that small deviations of the planes uncover uncontaminated pixels.

 

PixInsight has all the standard methods plus a special "Large Scale Pixel Rejection" tool which specifically seeks out things like airplane trails.



#8 Ghostnotes

Ghostnotes

    Lift Off

  • -----
  • topic starter
  • Posts: 8
  • Joined: 15 Nov 2021

Posted 12 August 2022 - 12:29 AM

Here's what you're missing.  People often shoot _hundreds_ of subexposures.  So, (just an illustration) 600 5 minute subexposures would be 50 hours.  What really counts here is the total time, NOT the subexposure.  Because we stack subs and the important thing is how many photons we captured, total.  How you divide the total into subexposures is just a tweak.

 

Here's an image I did of the Pleadies.  662 ten second exposures.  They needed to be short because it's an F2 scope, and I'm in light polluted skies.  Longer exposure, and my pixels would have been overloaded by light pollution.

 

https://www.astrobin.com/t5173s/

 

A rite of passage for an imager is shooting one target over multiple nights.  It used to be pretty difficult to get lined up on target really well.  So typically you'd have to crop the edges some, where you only had data for a single night.  No more.

 

There's a technique called platesolving.  You take a sub or an image.  Feed it into a computer which does pattern recognition on the stars, and tells you _exactly_ where you're pointed.  So you can go back to the precise spot.  It's amazingly simple.  Picture of the input/output screen for a platesolving program below.  Note that the error is less than an arc second, which is less than 1 pixel.

 

attachicon.gifPlateSolve2-Screenshotv2.jpg

 

More important.  I saw your post in the beginners forum where you referenced AP and software.  Have you done any Deep Space photography?  If not, realize that the second biggest beginner error (after an inadequate mount, you dodged that one <smile> ) is too big a scope.  Your 8 inch SCT defines "too big a scope".  It's not impossible to learn on one.  Just a very hard very long process.  Not much fun.  The following are not outliers, they're the standard experience.

 

"After months of learning and overcoming challenges <with the SCT>, and finally buying a shorter FL APO refractor, I really really really wish I had listened to everyone on here and started learning the imaging basics on THAT frac instead of on the SCT.   Trust me"

 

"I regret spending the first 6 months trying to learn imaging with an 8" Edge, with that scope it was a losing effort. Fortunately got a nice little refractor, and not only have the quality of my images improved but I'm actually enjoying the process of learning how to do it!"

 

I have many more of those.

 

The scope would work fine for the planets and the Moon.  Completely different thing, there's a forum called Solar System Imaging.

You would think by now I should know how long exposures are done, but up until recently it was a passing thought.

For sure, I plan on getting a faster/ smaller scope in the near future. Not to mention guiding because of my focal length will be problematic. I will still buy a guidescope/camera.Aaand I live in a Bortle 7 area. For now I'm just going to focus( no pun intended) on the planets and lunar/solar to get my technique down...setup, comms, alignment, plate solving. I have in the past used a wedge for polar alignment and got good results. The wedge was painstakingly set up and aligned to the base. But after borrowing it for a few weeks.......nope...

I was thinking of selling the nextstar mount but have decided to keep it for later use.

 

As far as software goes, programs like PHD2,SharpCap, Photoshop etc. Ive had them on my PC for quite some time, just never really had a use for them until now. Except for Photoshop,

I do very well with it. I understand the how's and why's, execution will probably be a different story. Im sure I will still have plenty of questions. I do think I should be able to get started with it. I know it's apples to oranges but when I first got the nextstar, I immediately started looking for control software. I think I used the hand controller maybe 2-3 times since I bought it. I had Stellarium up and running in about 20 minutes.


Edited by Ghostnotes, 12 August 2022 - 12:53 AM.


#9 Ghostnotes

Ghostnotes

    Lift Off

  • -----
  • topic starter
  • Posts: 8
  • Joined: 15 Nov 2021

Posted 22 August 2022 - 06:39 AM

So it's official I pulled the trigger last night.

I also decided to get Celestrons OAG and ZWO ASI174. I know it doesn't have the FOV as a stand alone, but I got a good deal on it so I figured what the heck. I can always sell it. From the spec's and affordability, it seems the ZWO sensitivity should help.




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