Jump to content

sgl_imaging_challenge_banner_31.thumb.jpg.b7a41d6a0fa4e315f57ea3e240acf140.jpg

confused-- longer exposure in a darker sky???????


Recommended Posts

Hi all i have been searching the net for a rough guide on camera settings for gain now i think im going mad!!! 

i found a guide on the CN forum and it gives settins for different sky zones im bortal 5-6 and it gives me the gain settings for my scope f7.5 but what i cant get my head around is the fact of ...the better ( darker sky) you have the longer you have to expose a image so im i missing something here a example a 0 gain setting at f7 is 1.3 mins in a orange zone yet if in a green zone it would be 7 mins ???????? 

Link to post
Share on other sites

It's not 'The longer you have to expose' but 'the longer you can expose' without saturating your chip with light pollution. The advantage of long exposures is that they go deeper - ie capture the faintest signal. However, we must distinguish between CCD and CMOS now. Long exposres are great with CCD cameras because the camera has considerable read noise, so 'longer but fewer' subs have fewer doses of read noise. CMOS chips have very low read noise so a huge number of shorter subs provide perfectly good results.

In general in AP it's always best to experiment rather than take some else's word for what's best.

Olly

  • Like 3
  • Thanks 1
Link to post
Share on other sites

With a darker sky you have the option of doing longer exposures to pick up fainter detail. I don't kow of any hard and fast rules for setting exposure times. You also have to note that it's total integration time that counts for dso's.

Louise

Link to post
Share on other sites

Sorry Olly, as I understand it it's not about saturating your chip with orange glow. The general rule for a single exposure is expose until your read noise is not the limiting factor. That happens earlier in the exposure for low-read-noise cameras (modern CMOS, for instance). So what other noise sources are there that we should think about? There's dark noise which is what you get even with the shutter closed. And there's shot noise, which is the uneven dribble of photons from light polluted "background" sky. The more light pollution there is, the more photons come from it  and the associated noise is the square root of the average number of photons. So that increases as we expose and at some point becomes much larger than the one-time penalty of read noise we pay at the end of the exposure (the "sub" some people call it). At that point, there's little point in carrying on with that exposure and we might as well start a new one. Starting a new one has advantages because our tracking/guiding could fail at any moment and you don't want to ruin a half-hour exposure because of stray photons in the last minute. Also, we don't want to saturate the bright interesting objects! So, dark skies means longer exposures are optimal than bright skies.

 

Edited by themos
  • Like 3
Link to post
Share on other sites

Yes, saturation is one part of the story - but really, I have not been able to saturate my CMOS chip, although I use unity gain and therefore have "full well" of around 4000.

Main thing is about what is dominant component to the noise. Noise adds like linearly independent vectors - or maybe simplest way to explain it - look at rectangle - sides being individual noise components, while diagonal being resulting noise (length of each). Highest ratio of diagonal length to (longer) side is with square - diagonal is 2^0.5 x side (or around 1.41). If you take any rectangle where one side is much larger than other - diagonal will be "approaching" that longer side in length. So with high LP, in short exposure LP noise becomes dominant component (the longest side) and read noise contribution is small. My LP is such that I get around 1e per second from LP, so in 60 seconds sub, I get around 7.7e of LP noise. That is about 5 times more than read noise (slightly less at 4.55 bit lets make it nice and round).

Just for comparison if we take rectangle with sides 5:1 - diagonal will be ~5.1 (5.099...) - so although read noise is 1.7e, total noise went up by just 0.1 or in percent - 2.5% compared to LP noise. So longer subs are just going to be closer and closer to LP and there is not much point in longer subs .

Now if you are imaging in dark skies, around 21 mag - you are getting x10 less LP photons, so in 60s exposure you might get only 6e, or in terms of noise - ~2.45 - that is pretty close to read noise (in CCD it is even less) - that is why you have to go for long exposure to compensate.

 

  • Like 2
Link to post
Share on other sites

right ok i got that thanks for clearing that up!! 

i have been using the unity gain 139 at 120sec and even tried it at 300 gain no wonder my image looked awful!! 

i have to say this forum is amazing as the time you lot are saving me with correcting mistakes!!

my aim is to get my setup ready and running smoothly for solid imaging in the Autumn ready for orion cant wait to try my narrowband filters..

Link to post
Share on other sites
9 hours ago, themos said:

Sorry Olly, as I understand it it's not about saturating your chip with orange glow. The general rule for a single exposure is expose until your read noise is not the limiting factor. That happens earlier in the exposure for low-read-noise cameras (modern CMOS, for instance). So what other noise sources are there that we should think about? There's dark noise which is what you get even with the shutter closed. And there's shot noise, which is the uneven dribble of photons from light polluted "background" sky. The more light pollution there is, the more photons come from it  and the associated noise is the square root of the average number of photons. So that increases as we expose and at some point becomes much larger than the one-time penalty of read noise we pay at the end of the exposure (the "sub" some people call it). At that point, there's little point in carrying on with that exposure and we might as well start a new one. Starting a new one has advantages because our tracking/guiding could fail at any moment and you don't want to ruin a half-hour exposure because of stray photons in the last minute. Also, we don't want to saturate the bright interesting objects! So, dark skies means longer exposures are optimal than bright skies.

 

Indeed. I used the term 'saturation' carelessly and not in the technical sense. I agree with your more precise description and we agree regarding your conclusion: under dark skies longer exposures are optimal. However, not imaging under LP means that I don't know the answer to this question: what are the consequences in terms of available dynamic range if the background sky levels become very high?

Regarding the over-exposure risk in long subs, I generally find it over-stated. There are obvious cases where it's certainly a problem: M42 is the classic example and needs multiple exposure lengths combined using some kind of HDR method. With my amateur apertures it's rare that galaxy cores, for instance, become over exposed in 30 minute luminance subs. When they do I can almost invariably use my RGB layer as luminance to cover what's been lost. Stellar cores certainly reach 65000 ADU in Lum but that doesn't matter since, again, I have my RGB layer to hand and it's easy to draw star colour into cores in post processing. (This is not scientific imaging, of course.) On those occasions when I've been chasing very faint signal, experiment has confirmed that I do significantly better with, say, 10x30 minutes than with 20x15. But this is with a CCD of relatively low QE and with deep wells. So I ignore the over exposure of stellar cores, though some imagers use it to set maximum exposure times.

Long subs do require reliable guiding but acts of vandalism by aircraft and satellites are usually reparable. If the sigma routine is not sufficient a cosmetic repair of the affected sub before stacking is usually effective.

All of this is from the world of CCD. Astonishingly good ultra-short exposure, huge sub-count CMOS images are now appearing.

Olly

Link to post
Share on other sites

Under LP the dynamic range is not impacted. Indeed, for a given target at a given gain, saturation of a subframe is less likely as fewer target photons are captured in a shorter exposure and the sky background would be exposed to the same level.  So lets say you need 30 seconds under LP vs 300s in a dark sky to bury the read noise with sky background. Then you've captured 10x the target photons with the dark sky. What is heavily impacted under LP skies is SNR. 

  • Thanks 1
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.