Short vs. Long Exposures [Deep Sky] Acquisition techniques · andrea tasselli · ... · 101 · 4852 · 34

Take a look at my M51 image.  A higher resolution version can be found  at  my gallery here on Astro Bin .  With the help of BlurXTerminator, I am getting a FWHM of ~ 1 arc-seconds.  The OTA was an 8 inch SCT.  This was created using 10 second exposures with a luminance filter.  A total of 2240 exposures were used in the final  image shown.  The camera was an ASI533MM Pro with at gain setting in ZWO speak of 200 (actually  it is really a linear gain of 10) The readout noise I would expect to be ~ 1.3 electrons RMS.  While the final SNR may not be great, the system is generating resolution close to what could be produced if the OTA was in space.  This is thanks to the benefits of DSO lucky imaging.  I only used 50% of the exposures collected during the real-time lucky imaging acquisition process.

Great example of optimizing the stack for a specific goal, and this is one of the areas where very short exposures can shine!

Jon Rista:

Tareq Abdulla:
I still enjoying your images with those old cameras, sounds i shouldn't give up my old cameras then, keep going, i will see what kind of exposures i will do with all my cameras old or new.

Thank you.

I am of the pretty strong opinion that camera technology, really, is less important than the quality of the photons you are capturing. By that, I mean, polluted vs. not. Polluted skies are devastating to image quality, and IMHO, getting way from the light pollution or eliminating the light pollution, is the single best thing any imager can do for their astrophotography.

This either means using narrow band filters, which is an option for imaging under light polluted skies. Or, finding and using a decent dark site (which are often FAR closer to people than gray/black zones...as cameras have STATIC sensitivity, compared to human eyesight which is DYNAMIC sensitivity. In years past, I think I determined that outside of some of the more densely populated areas in the eastern half of the US, and similar with the EU, most people probably live within an hour of a reasonably and sufficiently dark site for good quality astrophotography (for broadband, OSC or RGB, as well as narrow band). 

If people can find and use a decent dark site, it will be more transformative to their astrophotography than any camera. This is not to say that technology...cameras, scopes, don't play a role...they do. But, for most people, the difference between a light polluted back yard and a decent dark site is often 15-25x, which usually far outpaces any relative differences between Camera A or B, or telescope A or B. The differences between cameras and telescopes would still make a difference at a given imaging site...so once you have a dark site you can use, then depending on your specific goals, a bigger scope, or a better sensor, could then allow you to optimize your results for your specific goals. So cameras and scopes (and mounts) DO matter...just, IMO, not as much as the difference between light polluted skies and dark skies.

You are welcome!

That is very understandable and clear, as they all are saying, darkness skies is no match and no substitution, that is true, i followed you back in years and read your posts and comments, so we all know that sky quality matters more than gear we have, this gear thing has a factor to a degree if we have everything good as sky and conditions and environment, it is our nature as human to keep upgrading and developing, we see some differences, but as you said i can't say it is day and night difference, and it wasn't like so very long time changing, i started only in 2017, so only 6-7 to really seeing BIG huge difference to be honest, i just feel sad that in 1-2 years with technology people rush rapidly getting rid of their old gear, so i saved for months and years to afford QHY163M and ASI1600MM and didn't use them much yet at all only to give them up now because all are moving to new sensors, so i felt like what was wrong with old sensor if they were like kings even with issues back then, and i am not rich to keep buying new stuff everything they are out, for example i spent a lot to buy my Astrodon/Chroma SHO 5/3nm filters at 1.25" size, it will be crazy stupid for me to go for another brand lesser with larger size only that i have to.

Darker skies is a dream, so that most or all top known observatories around the world are placed in those dark skies, Chile and NM are two places well known for that, in my country there is no dark skies any nearby us, i mean even if i drive for about 2 hours the best i can get is Bortle 4 in the Bortle 5 dress, for my even 1 drive is not worth it for like 2-3 degree Bortle for me, unless it is Bortle 3 then i am not interested, and my financial situations just prevent me to spend for driving and instead i spend it for gear, at least in my yard i can do all bright targets including planetary and some nebulae with narrowbanding, so only because of OSC i won't sacrifice my time and money and life for it, i even didn't try to image OSC more and more nowadays since they placed LED in front of my house, it could be ok not completely bad yet, but i didn't have the gut yet to try as i spent/focused myself last 3 years only for shopping, and because last 1-2 years i got shocked about budget expected so i delayed to be back imaging, but this year i am making sure i buy like 1-2 very important main items then i will get back immediately, regardless how bad the sky is, after all i have to use what is there and get used to it, i have another issues in my life that i don't want to think about driving/traveling issues as well, i once went into a journey with astro academy in my country, went to a site that is good dark enough, it was like Bortle 4/5, was amazing really, different than city of course, but getting there was like mission impossible, even the mentor said it is not a good dark site yet, so for me it was like i won't try harder for a better sky then , their observatory with 17" CDK and 6"/7" TEC is actually in Bortle 8/9, so i won't try to go far.

Tareq Abdulla:

Jon Rista:

Tareq Abdulla:
I still enjoying your images with those old cameras, sounds i shouldn't give up my old cameras then, keep going, i will see what kind of exposures i will do with all my cameras old or new.

Thank you.

I am of the pretty strong opinion that camera technology, really, is less important than the quality of the photons you are capturing. By that, I mean, polluted vs. not. Polluted skies are devastating to image quality, and IMHO, getting way from the light pollution or eliminating the light pollution, is the single best thing any imager can do for their astrophotography.

This either means using narrow band filters, which is an option for imaging under light polluted skies. Or, finding and using a decent dark site (which are often FAR closer to people than gray/black zones...as cameras have STATIC sensitivity, compared to human eyesight which is DYNAMIC sensitivity. In years past, I think I determined that outside of some of the more densely populated areas in the eastern half of the US, and similar with the EU, most people probably live within an hour of a reasonably and sufficiently dark site for good quality astrophotography (for broadband, OSC or RGB, as well as narrow band). 

If people can find and use a decent dark site, it will be more transformative to their astrophotography than any camera. This is not to say that technology...cameras, scopes, don't play a role...they do. But, for most people, the difference between a light polluted back yard and a decent dark site is often 15-25x, which usually far outpaces any relative differences between Camera A or B, or telescope A or B. The differences between cameras and telescopes would still make a difference at a given imaging site...so once you have a dark site you can use, then depending on your specific goals, a bigger scope, or a better sensor, could then allow you to optimize your results for your specific goals. So cameras and scopes (and mounts) DO matter...just, IMO, not as much as the difference between light polluted skies and dark skies.

You are welcome!

That is very understandable and clear, as they all are saying, darkness skies is no match and no substitution, that is true, i followed you back in years and read your posts and comments, so we all know that sky quality matters more than gear we have, this gear thing has a factor to a degree if we have everything good as sky and conditions and environment, it is our nature as human to keep upgrading and developing, we see some differences, but as you said i can't say it is day and night difference, and it wasn't like so very long time changing, i started only in 2017, so only 6-7 to really seeing BIG huge difference to be honest, i just feel sad that in 1-2 years with technology people rush rapidly getting rid of their old gear, so i saved for months and years to afford QHY163M and ASI1600MM and didn't use them much yet at all only to give them up now because all are moving to new sensors, so i felt like what was wrong with old sensor if they were like kings even with issues back then, and i am not rich to keep buying new stuff everything they are out, for example i spent a lot to buy my Astrodon/Chroma SHO 5/3nm filters at 1.25" size, it will be crazy stupid for me to go for another brand lesser with larger size only that i have to.

Darker skies is a dream, so that most or all top known observatories around the world are placed in those dark skies, Chile and NM are two places well known for that, in my country there is no dark skies any nearby us, i mean even if i drive for about 2 hours the best i can get is Bortle 4 in the Bortle 5 dress, for my even 1 drive is not worth it for like 2-3 degree Bortle for me, unless it is Bortle 3 then i am not interested, and my financial situations just prevent me to spend for driving and instead i spend it for gear, at least in my yard i can do all bright targets including planetary and some nebulae with narrowbanding, so only because of OSC i won't sacrifice my time and money and life for it, i even didn't try to image OSC more and more nowadays since they placed LED in front of my house, it could be ok not completely bad yet, but i didn't have the gut yet to try as i spent/focused myself last 3 years only for shopping, and because last 1-2 years i got shocked about budget expected so i delayed to be back imaging, but this year i am making sure i buy like 1-2 very important main items then i will get back immediately, regardless how bad the sky is, after all i have to use what is there and get used to it, i have another issues in my life that i don't want to think about driving/traveling issues as well, i once went into a journey with astro academy in my country, went to a site that is good dark enough, it was like Bortle 4/5, was amazing really, different than city of course, but getting there was like mission impossible, even the mentor said it is not a good dark site yet, so for me it was like i won't try harder for a better sky then , their observatory with 17" CDK and 6"/7" TEC is actually in Bortle 8/9, so i won't try to go far.

I guess there is one fairly consistent advantage to CMOS sensors over most CCDs: Pixel size. The smaller pixels do allow more flexibility. The lower read noise is also a factor that lowers the barrier to entry and achieving success, as it doesn't take as much effort to acquire usable subs with lower read noise. I think these are two key things that resulted in a large and rapid shift from CCD to CMOS.

There are some sensors like the IMX455 and specifically the QHY600 camera that are from a specifications standpoint, quite superior to most of the commonly used CCD cameras. It has a large sensor, is 16-bit, has exceptional dynamic range, has a variety of viable gain settings and freely variable gain, and the camera itself is designed such that it can be configured in a very wide array of options to support just about any use case (it even supports water cooling if you live in a warmer climate!) Its a very reliable camera, and works well remotely. It sports all the other benefits of CMOS that make imaging easier. So, in that respect, it is not surprising to have seen a shift from large and very expensive CCD sensors like the KAF-16083, KAF-16200, and other similar larger sensors to the IMX455 cameras. I think in a lot of cases, though, the imagers who switched were already imaging under quality skies, so they would have been able to leverage benefits by switching to the IMX455 cameras. 

FWIW, anyone who says Bortle 4/5 skies are not dark enough, simply doesn't understand the nature of dark skies from an IMAGING standpoint. As I mentioend before, cameras have STATIC sensitivity. They don't have the dynamic response that the human eye does, which is why it is really unnecessary to drive 100+ miles away from any light pollution zone. Bortle is a human-relative scale, and it is designed to help humans figure out what they could see with their own eyes, and that is primarily because of our dynamic response. If we are still relatively near a light pollution center, then that LP will light the surrounding landscape, and if we look into the LP bubble, that will also affect our eyes...our irises will stop down, the natural response of our eyes will shift out of the purely scotopic vision mode, into mesopic vision (or perhaps even photopic, which is really detrimental to our ability to see in the dark!) So getting far away from light pollution is important for VISUAL observing. 

Cameras do not have these problems. If there is an LP bubble on the horizon, even if it is not all that far away, if a camera is pointed away from it, that LP bubble has no impact on the camera. My own dark site is Bortle 4/5, it is about 35 minutes from my home, and I've measured as dark as 21.6mag/sq" (this is about as dark as Colorado gets most of the time...this is due to natural factors, and its rare to measure 21.8 or 22mag/sq", and most of the time it isn't even 21.6). Most of the time my dark site is between 21 and 21.3mag/sq", and this is WILDLY better than my back yard (for broadband.) Its not contest, no comparison. I could literally spend weeks acquiring data from y red zone back yard, stack 50-100 hours of data, and still not produce an image of the same quality as a single night at my dark site would allow for. 

So, Bortle 4/5 is IMHO exceptional, compared to any city/urban/suburban light polluted zone. If you already live in say a yellow zone, bortle 5/6, then the difference betwen that and Bortle 4 is not going to be very large. But if you live in Bortle 6-9, which is orange through white, then the difference by going to a Bortle 4 or 5 zone is going to be huge. Bortle 5 is dark yellow, which is around 20.5-20.8mag/sq" and that is still two stellar orders of magnitude different from the average suburban zone. It could be 2.5-3 stellar orders of magnitude compared to an urban zone. Bortle 3 is pretty rare, but also unnecessary to get very good dark site imaging. Bortle 4 and 5 are very good for dark site imaging, and should give you considerable gains, especially if you live in a Bortle 8/9 zone. Most of my broadband imaging is from about midway between Bortle 4 and 5, skies that measure 21-21.3mag/sq" most of the time. Sometimes they measured as bright as 20.8. I'll testify that 20.8mag/sq" is also still very good for imaging. I had about 2/3rds of the sky that I could point to that had no LP on the horizon, and about a third of the sky that did have LP on the horizon. As long as I imaged away from the LP zones, then the quality was excellent.

If you can find a Bortle 4/5 zone within an hour from home, my honest opinion is that one night at that zone is worth weeks if not months of imaging from Bortle 8/9. ONE SINGLE NIGHT. Doesn't matter what equipment you have. Skill with setting up and acquiring your data without tracking issues is probably much more important than gear quality at a Bortle 4/5 dark site. As long as you can acquire good data, then MOST equipment, will fare very well at such a dark site, and allow you to acquire a lot of good quality data every night you visit. One night a month, could be sufficient to produce two to three images a night that are of very good quality, rich color, high contrast, even with less than ideal equipment. Most of my images were acquired with a Canon 5D III DSLR. Its actually a pretty darn noisy sensor, especially during the summer. The dark current noise during the summer was the biggest detriment to my images from my dark site, and even that, was still nothing compared to the devastating effects of light pollution. 

So, don't give up on Bortle 4/5, if you have such a site within an hour drive, it is IMHHO well worth the trip, and a vastly more efficient way to image than from a Bortle 8/9 light polluted zone (regardless of equipment.)

Tareq Abdulla:
I don't know about sky quality these days, and when i look at the light pollution map using the data of 2015 i can see that there is like 30-40 minutes drive away Bortle 4/5, but i can't tell now after almost 9-10 years if it stays as Bortle 4/5, and i will be honest, i hate to drive anywhere longer than 20 minutes, i go to some clubs which are like 40 minutes away and i am like exhausted or off mood, so i hope i can have the mood to go to darker skies even few times per month or per year, i have first to have good enough portable gear so i can use it while i am at that dark sites.

The only benefit from my location in Bortle 8/9 is that if i can sort out my life then i can keep imaging for so many nights, Doug imaging from Bortle 8 for long hours resulting amazing images mainly planetary nebulae, what a dedicated, and i see some collaboration images won here done at like 100-1000 hours, if they were under Bortle 1-3 i doubt they need that much time and to collaborate, so i assume most of them are in LP skies, and they produced nice results, i have like 10 scopes and it is growing, we all know that one trip to dark sky worth like weeks or even months in LP skies, but as long we are living under LP we can keep imaging for weeks and months long enough so we don't feel exhausted or off mood, and i have to test my sky again now to see what is the quality even with LED lights, if it is completely bad even for NB then sooner or later i will give up, and force myself going to dark skies.

For now i will try to take different exposures to put for comparison, if i am lucky i will try to point at direction away from LP as much i can and see what is the signal and data quality, i remember until like 20201 i tried short total time of different targets, i was able to get most of them within 1 hour up to 2, for example M51 and Andromeda were almost clear for less then 3-4 hours with one filter [Lum], even Ha got me something, and while i tested my OSC camera for the first time on M3 and M13 the clusters i had them in so short time, the only issue was i used not good filters which cut it a lot so processing was a headache, i can imagine if i take longer exposures or more frames then it will be different, but i stopped back then and they started the LED lights so i didn't dare to test again, but i think it is so long time now and i must try before i travel far for dark skies or jump to any wrong conclusion, for me if i manage to have like nice results in about 4-10 hours in my yard then i won't travel to dark skies for just 1-2 hours complete, i might try it but most likely it will be maybe 2-3 times in the year, i know that can be something, but this little times won't help me to keep going there, every year my situations getting worse, so if doing short exposure under light pollution can give me something then why not, i am always able to capture like 5-10 minutes xposure with narrowbanding just fine, and for broadbanding i think 1 minute was the longest i can with RGB and less with Lum, but that is with fast scopes or lenses, if i use like F6-F8 then i can have like 2-3 minutes with LRGB maybe and much longer with NB.

FWIW, if you use narrow band filters, you can image just fine under almost any skies. Narrow band imaging of planetary nebula under bortle 9 skies is just fine. Its not optimal, but still far better than broadband imaging under the same skies. Narrow band is a game changer for anyone who likes to image any kind of emission nebula, and it makes imaging under light polluted skies viable. 

I've done a bit of dual-band NB imaging with OSC cameras. It can work, but IMO if you really want to get some great images under light polluted skies, another area where the right hardware can help is a monochrome camera with distinct NB filters. You'll get lower noise with a mono+NB approach than osc+DB. If you have a mono camera already, then narrow band planetary imaging is ENTIRELY a very viable option and you can get great results. Most of my narrow band imaging is from my Bortle 8/9 back yard. NB imaging from a dark site does provide even better contrast, and for some objects you can have regions in a frame that are pure read noise (no object signal, lowests noise possible, effectively "true black"). The same areas from a light polluted site will have a small amount of signal in them from broadband LP (i.e. LED lighting or other forms of broadband emissions, of which there are actually a lot in urban and suburban areas these days). But, you still get very high contrast results with narrow band, despite this small passage of LP. 

So its not like imaging from a light polluted zone isn't possible, it is. Particularly with narrow band filters and mono cameras. The greatest detriment from LP occurs with broadband imaging. I find that OSC is the least viable in light polluted zones. Mono+RGB filters is better, but you are still going to suffer a lot from all the excess pollutant signal. Gradient extraction has gotten better these days, which is a very welcome development. That said, gradient extraction only removes the LP offset, and it cannot remove the additional noise...so that additional noise will still hurt your SNR and your color quality. Nothing you can do about that, outside of get TONS of data (tens of hours), and even then, such a broadband image from a Bortle 8/9 still wouldn't compare to 4-5 hours at a dark site.

If your goal is say planetary nebula, then I think you are just fine imaging under Bortle 8/9 skies with sufficiently narrow NB filters. I'd say 5nm or less and they will get the job done nicely. ;) Now, with narrow band, exposures are going to be longer, or at least longerish. You probably couldn't get away with any sub-minute exposures, as discussed in the OP. You would probably need several minutes per sub at least. In my case, with f/4 and f/5 scopes, I was using 10 minute subs! My pixel size is 2.4 microns, a little smaller than the more common 3.75 microns used in most CMOS astro cameras. So you might need 8 minutes or somewhere thereabouts. I guess it depends on whether you have an HCG mode with even less read noise...in my case, I had around 2-2.4e- read noise, but I know some cameras with HCG modes have as little as 1-1.5e-, so you might not need quite 10 minute subs with NB. Again, "long" is often relative, and compared to CCD imaging that often required 30 minutes or longer subs, 10 minutes is not "crazy" long. Its not short, but its certainly doable. If you have just 1.5e- read noise and a 5nm filter, you might only need 5 minute subs.

Jon Rista:

Tareq Abdulla:
I don't know about sky quality these days, and when i look at the light pollution map using the data of 2015 i can see that there is like 30-40 minutes drive away Bortle 4/5, but i can't tell now after almost 9-10 years if it stays as Bortle 4/5, and i will be honest, i hate to drive anywhere longer than 20 minutes, i go to some clubs which are like 40 minutes away and i am like exhausted or off mood, so i hope i can have the mood to go to darker skies even few times per month or per year, i have first to have good enough portable gear so i can use it while i am at that dark sites.

The only benefit from my location in Bortle 8/9 is that if i can sort out my life then i can keep imaging for so many nights, Doug imaging from Bortle 8 for long hours resulting amazing images mainly planetary nebulae, what a dedicated, and i see some collaboration images won here done at like 100-1000 hours, if they were under Bortle 1-3 i doubt they need that much time and to collaborate, so i assume most of them are in LP skies, and they produced nice results, i have like 10 scopes and it is growing, we all know that one trip to dark sky worth like weeks or even months in LP skies, but as long we are living under LP we can keep imaging for weeks and months long enough so we don't feel exhausted or off mood, and i have to test my sky again now to see what is the quality even with LED lights, if it is completely bad even for NB then sooner or later i will give up, and force myself going to dark skies.

For now i will try to take different exposures to put for comparison, if i am lucky i will try to point at direction away from LP as much i can and see what is the signal and data quality, i remember until like 20201 i tried short total time of different targets, i was able to get most of them within 1 hour up to 2, for example M51 and Andromeda were almost clear for less then 3-4 hours with one filter [Lum], even Ha got me something, and while i tested my OSC camera for the first time on M3 and M13 the clusters i had them in so short time, the only issue was i used not good filters which cut it a lot so processing was a headache, i can imagine if i take longer exposures or more frames then it will be different, but i stopped back then and they started the LED lights so i didn't dare to test again, but i think it is so long time now and i must try before i travel far for dark skies or jump to any wrong conclusion, for me if i manage to have like nice results in about 4-10 hours in my yard then i won't travel to dark skies for just 1-2 hours complete, i might try it but most likely it will be maybe 2-3 times in the year, i know that can be something, but this little times won't help me to keep going there, every year my situations getting worse, so if doing short exposure under light pollution can give me something then why not, i am always able to capture like 5-10 minutes xposure with narrowbanding just fine, and for broadbanding i think 1 minute was the longest i can with RGB and less with Lum, but that is with fast scopes or lenses, if i use like F6-F8 then i can have like 2-3 minutes with LRGB maybe and much longer with NB.

FWIW, if you use narrow band filters, you can image just fine under almost any skies. Narrow band imaging of planetary nebula under bortle 9 skies is just fine. Its not optimal, but still far better than broadband imaging under the same skies. Narrow band is a game changer for anyone who likes to image any kind of emission nebula, and it makes imaging under light polluted skies viable. 

I've done a bit of dual-band NB imaging with OSC cameras. It can work, but IMO if you really want to get some great images under light polluted skies, another area where the right hardware can help is a monochrome camera with distinct NB filters. You'll get lower noise with a mono+NB approach than osc+DB. If you have a mono camera already, then narrow band planetary imaging is ENTIRELY a very viable option and you can get great results. Most of my narrow band imaging is from my Bortle 8/9 back yard. NB imaging from a dark site does provide even better contrast, and for some objects you can have regions in a frame that are pure read noise (no object signal, lowests noise possible, effectively "true black"). The same areas from a light polluted site will have a small amount of signal in them from broadband LP (i.e. LED lighting or other forms of broadband emissions, of which there are actually a lot in urban and suburban areas these days). But, you still get very high contrast results with narrow band, despite this small passage of LP. 

So its not like imaging from a light polluted zone isn't possible, it is. Particularly with narrow band filters and mono cameras. The greatest detriment from LP occurs with broadband imaging. I find that OSC is the least viable in light polluted zones. Mono+RGB filters is better, but you are still going to suffer a lot from all the excess pollutant signal. Gradient extraction has gotten better these days, which is a very welcome development. That said, gradient extraction only removes the LP offset, and it cannot remove the additional noise...so that additional noise will still hurt your SNR and your color quality. Nothing you can do about that, outside of get TONS of data (tens of hours), and even then, such a broadband image from a Bortle 8/9 still wouldn't compare to 4-5 hours at a dark site.

If your goal is say planetary nebula, then I think you are just fine imaging under Bortle 8/9 skies with sufficiently narrow NB filters. I'd say 5nm or less and they will get the job done nicely. ;) Now, with narrow band, exposures are going to be longer, or at least longerish. You probably couldn't get away with any sub-minute exposures, as discussed in the OP. You would probably need several minutes per sub at least. In my case, with f/4 and f/5 scopes, I was using 10 minute subs! My pixel size is 2.4 microns, a little smaller than the more common 3.75 microns used in most CMOS astro cameras. So you might need 8 minutes or somewhere thereabouts. I guess it depends on whether you have an HCG mode with even less read noise...in my case, I had around 2-2.4e- read noise, but I know some cameras with HCG modes have as little as 1-1.5e-, so you might not need quite 10 minute subs with NB. Again, "long" is often relative, and compared to CCD imaging that often required 30 minutes or longer subs, 10 minutes is not "crazy" long. Its not short, but its certainly doable. If you have just 1.5e- read noise and a 5nm filter, you might only need 5 minute subs.

Excellent post and answer.

I do have mono and filters good enough, so i am all set fine with NB, but i won't give up broadbanding also even it is like very very difficult under Bortle 8/9, i wasn't planning to have APOD like RGB from Bortle 8/9, but wanted to have good results, because RGB is also stars signals, like i do NB with RGB stars, for that i do like 30sec up to 1min exposure for total nearly 1-2 hours, i don't think about doing galaxies from here, and clusters are like so so here but can be done, so my main targets are nebulae either emission or planetary and maybe i will try some reflected or dark nebulae too if possible.

I watch the light in my area of LED turned into slightly yellowish or warmish color cast, so i don't know if that will make it like less bad and less scattered across the visible spectrum, the lights aren't spread out all the way, but they are very slightly tilted to down like not fully parallel to street, almost like 10 degree tilt, for me it could be bad P with LED, but the dome is less than from old lights used, but i know LED is still worse anyway, and i really didn't try to do any DSO imaging either NB or broadband under those new lights to check out, i will do sooner or later and see.

I will try to choose the maximum longest exposure for RGB and even Lum under my sky and compare it with short exposures of my choice doing almost same total integration time, or even making the shorter exposures more time than longer ones and see if it is about integration time or exposure length itself regardless of sky quality, this is what this topic here, and i think under Bortle 8/9 we can't go so long exposure with LRGB unless we use like F10-F20 scopes maybe.

Take a look at my M51 image.  A higher resolution version can be found  at  my gallery here on Astro Bin .  With the help of BlurXTerminator, I am getting a FWHM of ~ 1 arc-seconds.  The OTA was an 8 inch SCT.  This was created using 10 second exposures with a luminance filter.  A total of 2240 exposures were used in the final  image shown.  The camera was an ASI533MM Pro with at gain setting in ZWO speak of 200 (actually  it is really a linear gain of 10) The readout noise I would expect to be ~ 1.3 electrons RMS.  While the final SNR may not be great, the system is generating resolution close to what could be produced if the OTA was in space.  This is thanks to the benefits of DSO lucky imaging.  I only used 50% of the exposures collected during the real-time lucky imaging acquisition process.

I concur --  it is amazing just what sort of effective FWHM and resolution is possible now by selecting short subs and following up with deconvolution .  A big fast scope --and if you are lucky enough -  dark skies to get more SNR into each sub definitely helps.   I did much the same as you and following with Blur Exterminator - details further below in the thread -  think that down to an effective resolution of ca 1.2  arcsec - and no obvious artifacts as I checked the detail off the NAS/ ESA HST pictures of the same. Tim

Take a look at my M51 image.  A higher resolution version can be found  at  my gallery here on Astro Bin .  With the help of BlurXTerminator, I am getting a FWHM of ~ 1 arc-seconds.  The OTA was an 8 inch SCT.  This was created using 10 second exposures with a luminance filter.  A total of 2240 exposures were used in the final  image shown.  The camera was an ASI533MM Pro with at gain setting in ZWO speak of 200 (actually  it is really a linear gain of 10) The readout noise I would expect to be ~ 1.3 electrons RMS.  While the final SNR may not be great, the system is generating resolution close to what could be produced if the OTA was in space.  This is thanks to the benefits of DSO lucky imaging.  I only used 50% of the exposures collected during the real-time lucky imaging acquisition process.

You are right Bob and that's a good  accomplishment with an 8" scope, but keep in mind that's a lot easier to do with a small telescope.  1"- 4" scopes can pretty easily achieve diffraction limited performance even under just "pretty good" skies.  You also have to keep in mind that outside of the iso-kinetic patch, seeing still causes the stars to translate differently so you have to be careful about simple angle and linear translation registration.  What you really need is to use a registration process that aligns the stars in patches--much as BXT does with sharpening.  Spline fitting in the PI Integration tool might work but it sometimes does weird things so you have to be careful with it.  

John

John Hayes:

Take a look at my M51 image.  A higher resolution version can be found  at  my gallery here on Astro Bin .  With the help of BlurXTerminator, I am getting a FWHM of ~ 1 arc-seconds.  The OTA was an 8 inch SCT.  This was created using 10 second exposures with a luminance filter.  A total of 2240 exposures were used in the final  image shown.  The camera was an ASI533MM Pro with at gain setting in ZWO speak of 200 (actually  it is really a linear gain of 10) The readout noise I would expect to be ~ 1.3 electrons RMS.  While the final SNR may not be great, the system is generating resolution close to what could be produced if the OTA was in space.  This is thanks to the benefits of DSO lucky imaging.  I only used 50% of the exposures collected during the real-time lucky imaging acquisition process.

You are right Bob and that's a good  accomplishment with an 8" scope, but keep in mind that's a lot easier to do with a small telescope.  1"- 4" scopes can pretty easily achieve diffraction limited performance even under just "pretty good" skies.  You also have to keep in mind that outside of the iso-kinetic patch, seeing still causes the stars to translate differently so you have to be careful about simple angle and linear translation registration.  What you really need is to use a registration process that aligns the stars in patches--much as BXT does with sharpening.  Spline fitting in the PI Integration tool might work but it sometimes does weird things so you have to be careful with it.  

John

Hey John, just curious, what is the "iso-kinetic patch"? I don't think I've heard of that before...

John Hayes:

Take a look at my M51 image.  A higher resolution version can be found  at  my gallery here on Astro Bin .  With the help of BlurXTerminator, I am getting a FWHM of ~ 1 arc-seconds.  The OTA was an 8 inch SCT.  This was created using 10 second exposures with a luminance filter.  A total of 2240 exposures were used in the final  image shown.  The camera was an ASI533MM Pro with at gain setting in ZWO speak of 200 (actually  it is really a linear gain of 10) The readout noise I would expect to be ~ 1.3 electrons RMS.  While the final SNR may not be great, the system is generating resolution close to what could be produced if the OTA was in space.  This is thanks to the benefits of DSO lucky imaging.  I only used 50% of the exposures collected during the real-time lucky imaging acquisition process.

You are right Bob and that's a good  accomplishment with an 8" scope, but keep in mind that's a lot easier to do with a small telescope.  1"- 4" scopes can pretty easily achieve diffraction limited performance even under just "pretty good" skies.  You also have to keep in mind that outside of the iso-kinetic patch, seeing still causes the stars to translate differently so you have to be careful about simple angle and linear translation registration.  What you really need is to use a registration process that aligns the stars in patches--much as BXT does with sharpening.  Spline fitting in the PI Integration tool might work but it sometimes does weird things so you have to be careful with it.  

John

Hi John.   I don't understand how such high resolution would really be easier with a 4 inch scope  ?  In theory a 4 inch might deliver a diffraction limited resolution of 1.3 arcsec but  to  adequately sample you would  need an image scale of  0.7 or preferably below.?  Compared to a big scope the little scope would obviously need  longer subs a  e.g. compared to my 12 inch -   some 90s rather than 10s  to get the same number of electrons into the pixels and to achieve a similar SNR per sub.  I am probably biased because of living under unsteady skies and not even being able to imagine  sky conditions holding steady  at < 2 arcsec.  So sub time is a factor as well as the aperture size?   Also, as Bob also said I think , part of the beauty of the short subs is the fact of being able to select out the better ones from a large number.  I guess that there is some sort of upper limit on useful aperture - and atmosphere effects worse across too big an aperture - but in as much as lucky imaging works at all I think it better to use a big scope -  get lots of the shortest but still decent enough SNR subs that you can select from - all the better under dark skies -and/ or just restricting  to the brighter objects?  In practice it does seem that most lucky imagers do follow the big scope, short sub philosophy and now combined with Blur Exterminator deconvolution  it seems not unusual  to see images with resolutions right down below 1.5 .  Tim   

PS Your comment about tweaking registration and alignment in patches was interesting  - might try spline fitting then.  I guessed that it was to do with the fact that a short window of steadiness across a small aperture is more probable than across  a  larger aperture - but am not sure that I fully understood or know at what point that really matters?

Tim Hawkes:
Hi John.   I don't understand how such high resolution would really be easier with a 4 inch scope  ?  In theory a 4 inch might deliver a diffraction limited resolution of 1.3 arcsec but  to  adequately sample you would  need an image scale of  0.7 or preferably below.?  Compared to a big scope the little scope would obviously need  longer subs a  e.g. compared to my 12 inch -   some 90s rather than 10s  to get the same number of electrons into the pixels and to achieve a similar SNR per sub.  I am probably biased because of living under unsteady skies and not even being able to imagine  sky conditions holding steady  at < 2 arcsec.  So sub time is a factor as well as the aperture size?   Also, as Bob also said I think , part of the beauty of the short subs is the fact of being able to select out the better ones from a large number.  I guess that there is some sort of upper limit on useful aperture - and atmosphere effects worse across too big an aperture - but in as much as lucky imaging works at all I think it better to use a big scope -  get lots of the shortest but still decent enough SNR subs that you can select from - all the better under dark skies -and/ or just restricting  to the brighter objects?  In practice it does seem that most lucky imagers do follow the big scope, short sub philosophy and now combined with Blur Exterminator deconvolution  it seems not unusual  to see images with resolutions right down below 1.5 .  Tim   

PS Your comment about tweaking registration and alignment in patches was interesting  - might try spline fitting then.  I guessed that it was to do with the fact that a short window of steadiness across a small aperture is more probable than across  a  larger aperture - but am not sure that I fully understood or know at what point that really matters?

Tim,
Remember that I'm not talking about resolution; I'm talking about diffraction limited performance.  In the extreme, consider a telescope with 1" aperture that has an Air Diameter of 2.44*lambda/D in object space.  That's a diameter of 10.9 arc-seconds.  You'd need pretty poor seeing to not be able to image that pattern and my guess is that you could do it just about anywhere.  The Fried parameter under somewhat average seeing conditions is typically quoted at around 100 mm (or about 4").   Remember that the Fried parameter describes the size of an imaginary telescope aperture for which the diffraction limited angular resolution is equal to the resolution limited by seeing.  So in most cases, the image quality of a telescope larger than roughly 100 mm is limited more by the atmosphere than by diffraction.  In locations where the seeing is as good as 0.4", the Fried parameters is as large as 300mm but that is quite rare.

For extended objects, irradiance in the image plane depends ONLY on the focal ratio; not the aperture.  So the exposure required for a 4" relative to a 12" if both have the same F/#, depends only on how the image is sampled.  If they are both sampled the same in object space, the exposures will be nearly identical.   The difference will be that the bigger telescope delivers a bigger image.  If you switch to stars and hold the focal ratios the same, then yes, the bigger telescope will pick up a lot more stars.

John

John Hayes

In the paper that you referenced the author claims that the isokinetic angle is ~ 0.3 D/h. where h is the height above the ground of the "main region" of the turbulence.  They showed typical values for h like 3 KM.  How do they know that?  I would think the h value would be site dependent.  For an 8 inch telescope this angle would be ~ 4 arc-seconds.  My guess is that this value of h is too high for most amateur  astronomy observing sites.  Maybe at mountain top sites where the major professional observatories are located 3 KM is about right.  

So I am planning to try to actually measure what value the isokinetic angle looks like at SRO.  I would think that if you are at a site surrounded by tall trees, the effect of the trees might create a turbulent layer just above them due to shear of the laminar flow.  This could be the region responsible for most of the seeing.  If that were the case, the isokinetic angle might be much larger.

This is of interest to me because I am guiding and image aligning on a 12 x 12 arc-min region, not a single guide star.  The size of the isokinetic angle will have an impact on the amount of averaging I am getting as compared with guiding on a guide star.

Bob

We are starting to get pretty far from the topic of this thread (and I'm certainly a contributor to the thread drift) but I suppose that the effects of turbulence are at the core of long v short exposure strategies.

A 4.7 arc second isokinetic patch sounds a bit small for an 8" scope.  I'll have to look into that number more.  The patch size must be a LOT smaller than I realized.  It's also interesting that the patch size scales with the diameter.  That's not immediately intuitively obvious to me.   I have another source on atmospheric optics that give a completely different relationship for the patch size so this requires more research.  The original papers are probably the best place to start.

I haven't read all of Fried and Kolmogorov's papers but for the most part, the ones that I have read analyze the effects of turbulence generated in the far field.  The far field is generally considered to be much, much greater than 10x the focal length so that approximations to the diffraction integral can be made as if the disturbance were effectively infinitely far away.  That means that the theory we are discussing is ineffective at describing local seeing effects, which includes ground seeing, dome seeing and turbulence within the telescope itself.   Local effects may range from inside the telescope up to a height of maybe ~ 100 m above the surface--depending on the location.   This source (again from ESO):

http://www.eso.org/genfac/pubs/astclim/papers/lzthesis/node11.html#:~:text=The%20seeing%20observed%20by%20a,tropopause%20at%20about%2012%20km

also includes mention of a lower atmospheric boundary layer but the Kolmogorov model relies on homogeneous small scale structure generated by wind shear in the upper atmosphere and that's what most discussions of the effects of seeing refer to.  Wind shear can occur from changes in wind velocity such as at the edges of the jet stream or from a change in direction of strong winds in the upper atmosphere.  The weather service used to rely on ballon soundings to determine winds and temperatures in the upper atmosphere but that has all been replaced by weather transponders attached to commercial (and some private) aircraft.  There are so many aircraft aloft at different altitudes all over the world that weather models contain very dense data about current atmospheric conditions.  This makes it pretty straightforward to see where in the atmosphere the winds are the strongest and where laminar and non-laminar flows are occurring, which improves the accuracy of seeing predictions.  As a pilot, my feeling is that the "typical height" of 3 km (roughly 10,000') for the main turbulence layer as stated in that paper is at the very low end of where it might actually occur.  The bigger winds are typically at an altitude of 30,000' to around 42,000 feet--at the top of the troposphere (which is latitude dependent).   The choice to put major observatories on high peaks is generally to get them above the ground turbulence layer, which often driven by solar convection and is often visually identifiable at the top of a scattered cloud layer.  This is why the Hawaiian observatories are located on such high peaks--both to get above the solar convection layer and to get into the dry high altitude air.  Even high on a mountain, the location should be selected to lie under a region where the high altitude winds are mostly laminar.  This is largely the case for many of the sites in Chile where many of the observatory sites are not on particularly high mountain peaks.

John

John Hayes:

Tim Hawkes:
Hi John.   I don't understand how such high resolution would really be easier with a 4 inch scope  ?  In theory a 4 inch might deliver a diffraction limited resolution of 1.3 arcsec but  to  adequately sample you would  need an image scale of  0.7 or preferably below.?  Compared to a big scope the little scope would obviously need  longer subs a  e.g. compared to my 12 inch -   some 90s rather than 10s  to get the same number of electrons into the pixels and to achieve a similar SNR per sub.  I am probably biased because of living under unsteady skies and not even being able to imagine  sky conditions holding steady  at < 2 arcsec.  So sub time is a factor as well as the aperture size?   Also, as Bob also said I think , part of the beauty of the short subs is the fact of being able to select out the better ones from a large number.  I guess that there is some sort of upper limit on useful aperture - and atmosphere effects worse across too big an aperture - but in as much as lucky imaging works at all I think it better to use a big scope -  get lots of the shortest but still decent enough SNR subs that you can select from - all the better under dark skies -and/ or just restricting  to the brighter objects?  In practice it does seem that most lucky imagers do follow the big scope, short sub philosophy and now combined with Blur Exterminator deconvolution  it seems not unusual  to see images with resolutions right down below 1.5 .  Tim   

PS Your comment about tweaking registration and alignment in patches was interesting  - might try spline fitting then.  I guessed that it was to do with the fact that a short window of steadiness across a small aperture is more probable than across  a  larger aperture - but am not sure that I fully understood or know at what point that really matters?

Tim,
Remember that I'm not talking about resolution; I'm talking about diffraction limited performance.  In the extreme, consider a telescope with 1" aperture that has an Air Diameter of 2.44*lambda/D in object space.  That's a diameter of 10.9 arc-seconds.  You'd need pretty poor seeing to not be able to image that pattern and my guess is that you could do it just about anywhere.  The Fried parameter under somewhat average seeing conditions is typically quoted at around 100 mm (or about 4").   Remember that the Fried parameter describes the size of an imaginary telescope aperture for which the diffraction limited angular resolution is equal to the resolution limited by seeing.  So in most cases, the image quality of a telescope larger than roughly 100 mm is limited more by the atmosphere than by diffraction.  In locations where the seeing is as good as 0.4", the Fried parameters is as large as 300mm but that is quite rare.

For extended objects, irradiance in the image plane depends ONLY on the focal ratio; not the aperture.  So the exposure required for a 4" relative to a 12" if both have the same F/#, depends only on how the image is sampled.  If they are both sampled the same in object space, the exposures will be nearly identical.   The difference will be that the bigger telescope delivers a bigger image.  If you switch to stars and hold the focal ratios the same, then yes, the bigger telescope will pick up a lot more stars.

John

John Hayes:
The Fried parameter under somewhat average seeing conditions is typically quoted at around 100 mm (or about 4").   Remember that the Fried parameter describes the size of an imaginary telescope aperture for which the diffraction limited angular resolution is equal to the resolution limited by seeing.  So in most cases, the image quality of a telescope larger than roughly 100 mm is limited more by the atmosphere than by diffraction.

Thanks for your reply and the learning.  I certainly didn't know most of that.  One further question though.  Surely part of the point of Blur Xt deconvolution is that it can beat the seeing - always provided that your sampling rate supports a higher resolution.   So in a short frame you get atmosphere -distorted star PSFs but distortions  that are at least distorted in a  reasonably consistent way within any given region of the frame and over a short time.  BlurXt (I presume) iteratively calculates the correct local cmpensatory correction  and then applies it.   So the question  is that while it is clearly always better to start from a near perfect image and to then apply deconvolution to -- in my experience at least -  BlurXt takes you a long way even when the star shapes are not perfect .  In my M51 picture above average Eccentricity was up at maybe 0.55 prior to correction and deconvolution.  Maybe consistency of blur is more important to the end product than lack of blur as a starting point to apply deconvolution to?  Tim

Tim Hawkes:
Thanks for your reply and the learning.  I certainly didn't know most of that.  One further question though.  Surely part of the point of Blur Xt deconvolution is that it can beat the seeing - always provided that your sampling rate supports a higher resolution.   So in a short frame you get atmosphere -distorted star PSFs but distortions  that are at least distorted in a  reasonably consistent way within any given region of the frame and over a short time.  BlurXt (I presume) iteratively calculates the correct local cmpensatory correction  and then applies it.   So the question  is that while it is clearly always better to start from a near perfect image and to then apply deconvolution to -- in my experience at least -  BlurXt takes you a long way even when the star shapes are not perfect .  In my M51 picture above average Eccentricity was up at maybe 0.55 prior to correction and deconvolution.  Maybe consistency of blur is more important to the end product than lack of blur as a starting point to apply deconvolution to?  Tim

Russ had a genius idea for BXT and he had to solve a lot of the details to make it work as well as it does.   At a high level, the concept is actually pretty straight forward.  I should add here that  Russ hasn't given me any inside information but here's my guess about how he might have implemented it.  It is simply a neural network that is loaded with NxN patches of Hubble images that have been mathematically blurred (probably with just a Gaussian blur function).  N might be a value that ranges from 32 to maybe 512--depending on how Russ chose to set it up.  There might be anywhere from 300,000 to 1,000,000 samples loaded into the training set, which is then trained using the original blurred data to find the best match out of all of the samples.  The training can include a lot of different parameters including the amount of blurring, asymmetry in the blurring (smear), and noise levels.  When you sharpen your own image, the data is subdivided into NxN patches so that each patch in your data can be identified with the "mostly likely" fit to a solution.  Once identified, the information in that patch is replaced with the original image data that created the best-fit blurred data.  Note that this is not the same as simply inserting Hubble images directly into your image.  The image patches are small enough that the Hubble data serves mostly as a way of supplying a nearly limitless source of "sharpened patterns" that can be used to show what your more blurry data might look like without the blurring mechanism.  I believe that the process for de-blurring the stars is similar but it may be different enough that it runs as a separate process from the structure sharpening.  That's something that Russ would have to address.  I could imagine that the star correction NN could be loaded with mathematically computed Moffat data that has been filtered through a range of aberrations as well as image translations.  One of the tricky parts to all of this is to get everything normalized properly so that the results all fit together seamlessly.

So nothing in BXT is like the traditional deconvolution process that requires a PSF kernel.  BXT has the ability to solve for the seeing conditions, but Russ didn't choose to work that into the solution.  Regardless, BXT doesn't have to know anything about the seeing to work well.   It just uses mathematically blurred data and since the process is applied patch-wise across the field, it can effectively correct for field aberrations (that vary with position) as well as for motion blur.

John