Picking the correct exposure for Deep Sky

A place to post guides, instructions and HOWTO documents on the use of SharpCap and Astrophotography in general
Forum rules
One tutorial per topic please.
Please only post content that you have created in this form.
Tutorials should be either in PDF format as an attachment or written in the first post of the topic. Links to video tutorials you created are also acceptable.
Fabioibirru
Posts: 25
Joined: Fri Sep 21, 2018 2:49 pm

Re: Picking the correct exposure for Deep Sky

#21

Post by Fabioibirru »

Dear Robin,

I'm glad to you about this article and other technical information about this topic.

Then I spent little time to translate your article into my native language because I found lot of people that use the old "GrandMother recipe" and do not belive that short exposures could lead to good results.
I'm a chemist and for me is unbelivable that such information are not a collective domain.
Anyway this is my little contribute to spread it also to non technical people.

I add to original article just 3 pictures taken from Smart Histogram part of SharpCap manual.

Could I share it to Astrophotographic italian community?

Best Regards

Fabio
Attachments
Traduzione Picking the correct exposure for Deep Sky.pdf
(890.5 KiB) Downloaded 430 times
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Picking the correct exposure for Deep Sky

#22

Post by admin »

Hi Fabio,

thanks for taking the time to translate the article – I'm very grateful, and of course you are welcome to share it with the Italian astronomy community :-)

Cheers, and clear skies, Robin
John Jennings
Posts: 1
Joined: Fri Jan 08, 2021 9:30 pm

Re: Picking the correct exposure for Deep Sky

#23

Post by John Jennings »

I'm using the Sharpcap online Sky Noise calculator to compute optimal Sub times for my CMOS QHY268C camera with a couple of different OSC filters. One is the Optolong LProExtreme which has (2) 7nm band passes.

My questions are:

(1) Does that mean I can a use a 14nm bandpass setting for mono in the calculator and then divide by 3 for the final Skynoise calculation?

(2) I estimate my Hutech LPSP3 broadband light pollution filter has about a total bandpass of 215nm if I add all the half widths of the bandpass notches. Can I use this simple technique to estimate the Sky Noise with different filters and a color CMOS camera?

(3) Is it possible to place any imaging filter in front of my portable SQM meter to get a reading to use in the online Skynoise calculator for the MPAS without utilizing the filter settings in the calculator?

These filters and the new third generation CMOS color cameras are some of the latest tools for imaging in the suburbs and most everyone is experimenting with different filters and sub lengths. I created a simple spreadsheet which is extremely illuminating based on you tutorial, but the big unknown for me is how to account for these filters in the calculation. My imaging sky ranges from 8.2 MPSAS east of the meridian to 7.8 west of the meridian as measured by my SQM meter. My current strategy is to split my automated imaging runs into multiple segments.

John J.
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Picking the correct exposure for Deep Sky

#24

Post by admin »

Hi,

hmm, interesting questions :)

1) Assuming that the two narrow band windows are each within the range of one colour of the RGB sensor (but in different colours), then I think you would treat this as a mono camera with 7nm filter. Imagine there is a 7nm window in the red region - light that passes through that will (almost) all pass the red filter and be counted by the red pixels (and practically none on the G/B pixels). The same if the second window is in the G region.

2.) Here I would find the colour (R/G/B) with the least bandpass and use that in the mono setting - for instance if there is a bandpass of 40nm in the blue then use that figure in mono setting, since that is what the blue pixels will see (and they will be getting less light than the colours with higher bandpasses).

3) Yes, pretty close to equivalent in the case of a mono sensor or a single band filter. More complex with multi-band filters.

cheers,

Robin
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Picking the correct exposure for Deep Sky

#25

Post by oopfan »

May I suggest a Plan B?

If you know your Bias level (mine is 200 ADU) then increase the exposure until dead space begins reading higher than Bias. How much higher? It depends on your camera's Read Noise, but a good target would be 100 ADU above Bias for a CMOS camera, or 200+ ADU for CCD.

Brian
elizaaa
Posts: 1
Joined: Mon Jan 11, 2021 8:09 pm

Re: Picking the correct exposure for Deep Sky

#26

Post by elizaaa »

Simple constellation shots with a camera on a fixed tripod usually run from about 10 to 30 seconds. Piggyback shots of wide fields can run about 5 to 10 minutes. Prime focus shots through a telescope can run 30 to 90 minutes. This discussion will primarily deal with long exposures for deep-sky astrophotography.
vintage binoculars made in japan
zerolatitude
Posts: 156
Joined: Mon Mar 01, 2021 5:24 am

Re: Picking the correct exposure for Deep Sky

#27

Post by zerolatitude »

Thank you very much for an excellent tutorial (along with your videos on this plus selecting the right gain). They helped greatly in my mount purchase decision.

There are a couple of points I'd like to bounce off you.

1. The thermal noise has been removed from the equation early as it is considered insignificant. I reworked the equations including that and the final equation seems a straightforward change - just adding the dark current to sky electron rate in the denominator? Is that correct?

2. This became important because I was also considering a dedicated cooled camera, which in my location is 2.5x the price of a DSLR like Nikon D5600. Putting the numbers for my location (17.5 mag sky) and the CMOS camera (2e read, 0.15 e dark current at 20 degrees C, extrapolated to 0.6 e at 34 degrees C), shows that the thermal noise is an insignificant (<10%) component of total noise. At least unless the sky pollution improves to a level of 20mag - which is not going to happen. So there is no practical benefit in getting the camera for the cooling alone (read noise, dark current, QE etc of the sensors are reasonably similar) if the sky electron rate is swamping the dark current. Would this be a correct conclusion?

3. With this level of light pollution, narrowband filters come up. Your lecture recommends 25x the exposure time for narrowband, but I was wondering if there is a finetuning possible here with the SNR. Broadband filters will (I assume) affect target and sky signals more or less similarly. But with narrowband, for the right target, you could have 10% reduction in the target but (possibly) a more than 50% reduction in sky electron rates, pushing up the SNR. This then opens up the possibility of reducing total integration time while keeping the stack SNR the same. I put these assumptions into a model and came up with a 30% reduction in total integration time required for the same SNR. So that opens up another optimization between total integration time and SNR. Does this approach make sense?

Of course, if the sky electron rate does drop drastically with the NB filter, will have to recalculate to see if the thermal noise is still insignificant :-)

Thanks
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Picking the correct exposure for Deep Sky

#28

Post by admin »

Hi,

good questions :)

In order...

1) yes, you can add the thermal noise into the equations in that way. I tend to ignore it because if you measure the sky background using the camera (as the SharpCap Brain/Smart Histogram operation does) then you measure (sky noise + thermal noise) - this is exactly what you want for the later calculations, so no point in trying to hold them separate throughout the work.

2) It depends on the thermal noise profile of the camera. I think I personally would want a cooled camera for deep sky work more because you can keep the sensor at a constant temperature (which improves dark subtraction and allows dark frame re-use). Additionally, some cameras seem to have glow patterns at long exposures that are reduced by cooling.

3) Yes, narrowband will get you an SNR advantage - you cut the background down by a large factor, but emission lines are unaffected. Of course you will still need much longer exposures because the background brightness is reduced so much. I built a little calculator to help estimate background rates with various equipment and light pollution levels - https://tools.sharpcap.co.uk

cheers,

Robin
AndreVilhena
Posts: 7
Joined: Sun Nov 01, 2020 10:25 am

Re: Picking the correct exposure for Deep Sky

#29

Post by AndreVilhena »

Hi,

I keep coming back to the post in this thread because they are, in my opinion, an excellent piece of information, useful for a lot of things. However, it is somewhat counterintuitive to the usual practice and sometimes doubts arise.

If I understood correctly, if the minimum subframe exposition is 12 sec, for example, that mean that taking a 120 sec frame will not improve SNR very much on the subframe when compared with 12 sec on. Hence it would be better to take 100 x 12 sec than 10 x 12 sec, as stack SNR would improve. But then the 100 x 12 sec stack would be less brighter.
Being less bright can be compensated in post-processiong but couldn't that hurt image quality by overstreching?

Another question: it is usually said that if f ration increases, subframe exposition time should also increase. But let's suppose a practical example (from a friend of mine... :D): going from an f/5 to an f/7 entails doubling the time, from 9 to 18 sec (my sky is bad...). If I shoot at 120 sec, regardless of the telescope, would that affect visibly SNR for the same intregration time?

Finally, SC has a graph showing the faintest detail level you can get - is it possible to explain the math behind?

Thank you in advance and again, thank you very much for those informative posts.

Regards,
André
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Picking the correct exposure for Deep Sky

#30

Post by admin »

Hi André

The SNR argument is slightly different - if the recommendation is 12s then the SNR of the final stack will not be improved much by taking longer sub frames (ie 120s). The individual 120s frames will each have better SNR than the 12s frames, but you get 10x more 12s frames, which compensates for the reduced SNR in each frame, leading to very similar SNR in the final stack.

Shooting at exposures beyond the recommendation will not degrade your final stacked image (unless you start to run into problems due to poor guiding, satellite trails, oversaturation, etc), but it will not improve it either. Again here we are talking about the SNR in the final stacked image that uses all the sub frames captured during a set integration time.

The faintest detected object calculation is based on the nose level for the final stacked image (which is in the calculations in the first 8 or so posts) divided by the total integration time. This gives the noise (in electrons per pixel) divided by the time (in seconds), giving an e/pix/second equivalent for the noise in the final stacked image. This is used as an estimate of what can/can't be seen in the final image (on the assumption that you can't see stuff below the noise level). Actually, you probably *can* see hints extended objects that are below 1 sigma, but the point of that graph is really to show that it goes almost flat once you get to the recommended exposure - longer sub exposures don't let you see fainter things...

cheers,

Robin
Post Reply