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

Best Resolution

  • Please log in to reply
11 replies to this topic

#1 Stargazer3236

Stargazer3236

    Soyuz

  • ***--
  • topic starter
  • Posts: 3762
  • Joined: 07 Aug 2010
  • Loc: Waltham, MA

Posted 14 September 2019 - 12:10 PM

Primarily for imaging purposes, what telescope at around 8" aperture, will I get better Resolution: An 8" RC or an 8" SCT? What about a Meade 8" ACF? Based on using my ASI294MC camera at 2x2 binning?



#2 vehnae

vehnae

    Explorer 1

  • -----
  • Posts: 93
  • Joined: 17 May 2013
  • Loc: Finland

Posted 14 September 2019 - 02:12 PM

All these scopes are essentially "diffraction limited" when properly adjusted. If the goal is the highest resolution, there are many other factors that need to be top-notch first, such as:

 

- Collimation

- Your local seeing

- Guiding accuracy

- Thermal management (proper cooling)

- Tracking accuracy

- etc

 

  ++ Jari



#3 Stelios

Stelios

    Cosmos

  • *****
  • Moderators
  • Posts: 7741
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 14 September 2019 - 03:55 PM

Primarily for imaging purposes, what telescope at around 8" aperture, will I get better Resolution: An 8" RC or an 8" SCT? What about a Meade 8" ACF? Based on using my ASI294MC camera at 2x2 binning?

I thought binning didn't benefit CMOS cameras? Might as well zoom out.

 

I would take an Edge HD over a Meade ACF (and take the Meade ACF over a regular C8). The Edges are superb scopes. They also can be used at three focal lengths, native F/10, F/7 with a reducer and F/2 with Hyperstar. That makes it superior IMO to the less flexible 8" RC. 

 

With your ASI294, the image scale at F/7 is a very reasonable 0.67. 

 

I know that I've got some very nice images with the Edge8 at F/7 (most yet not posted). I would definitely recommend that scope. 



#4 Stargazer3236

Stargazer3236

    Soyuz

  • ***--
  • topic starter
  • Posts: 3762
  • Joined: 07 Aug 2010
  • Loc: Waltham, MA

Posted 14 September 2019 - 04:05 PM

However, I just learned from another topic poster that the Meade ACF can use a regular Reducer Corrector, whereas the EDGE has to use a specifically manufactured reducer which costs up to 10X that price ($80 vs $800).



#5 bigjy989

bigjy989

    Explorer 1

  • -----
  • Posts: 99
  • Joined: 23 Nov 2010
  • Loc: Saline, MI

Posted 14 September 2019 - 08:05 PM

Initial question:  How experienced are you at dso imaging?    

 

In general - almost everywhere the best you can hope for is about 1.5-2.0 arc second seeing, short of being on the windward side of the coast or at high elevation.  Combined with tracking errors (1 arc second plus with a large scope on the ZEQ), collimation (essential to be actually diffraction limited but realistically 0.7 ish).   If your sticking with the ZEQ25 it will be hard at the focal length to get a FWHM under 2-3 in a stack.  This is your starting point for sampling so somewhere in the 0.9-1.1 range is the realistic target.

 

If you go with something like a mach1 you might get to 0.5 tracking errors, collimate every night then maybe 0.6 diffraction limit and maybe 3-4 night a year with seeing less than 1 arc second.  Then you might go for 0.6-0.7 (but the exposure time will need to be doubled).

 

So to answer this specific question, with the 294 (4.6um pixels), the ideal focal length for resolution will be not more than 1000mm unbinned (to keep about 0.9 arc-sec per pixel) and not more thsn 2000mm binned 2x2 (although that should be done in software for CMOS).    

 

Therefore an 8 inch at f5 is about ideal for this camera (MN180?).  F6 would be the absolute max I would recommend without spending a lot more on a mount and semi-permanent setup.  Even with that setup it will be hard to do more than 2min subs.    If the goal is lucky imaging to reduce the mount requirements then it is ideal to get to f/4 or under - either a RASA or imaging newt - but still not recommended for the ZEQ25.


Edited by bigjy989, 15 September 2019 - 07:56 AM.

  • Stargazer3236 likes this

#6 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23731
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 14 September 2019 - 10:30 PM

I thought binning didn't benefit CMOS cameras? Might as well zoom out.

 

I would take an Edge HD over a Meade ACF (and take the Meade ACF over a regular C8). The Edges are superb scopes. They also can be used at three focal lengths, native F/10, F/7 with a reducer and F/2 with Hyperstar. That makes it superior IMO to the less flexible 8" RC. 

 

With your ASI294, the image scale at F/7 is a very reasonable 0.67. 

 

I know that I've got some very nice images with the Edge8 at F/7 (most yet not posted). I would definitely recommend that scope. 

It is not that binning does not benefit CMOS cameras. Binning with CMOS is currently just digital, so there is no reduction in read noise. There IS, however, still an increase in SNR. Bin 2x2, the combination of 4 samples, increase SNR 2x.


  • RedLionNJ and bobzeq25 like this

#7 Stargazer3236

Stargazer3236

    Soyuz

  • ***--
  • topic starter
  • Posts: 3762
  • Joined: 07 Aug 2010
  • Loc: Waltham, MA

Posted 15 September 2019 - 11:42 AM

Initial question:  How experienced are you at dso imaging?    

 

In general - almost everywhere the best you can hope for is about 1.5-2.0 arc second seeing, short of being on the windward side of the coast or at high elevation.  Combined with tracking errors (1 arc second plus with a large scope on the ZEQ), collimation (essential to be actually diffraction limited but realistically 0.7 ish).   If your sticking with the ZEQ25 it will be hard at the focal length to get a FWHM under 2-3 in a stack.  This is your starting point for sampling so somewhere in the 0.9-1.1 range is the realistic target.

 

If you go with something like a mach1 you might get to 0.5 tracking errors, collimate every night then maybe 0.6 diffraction limit and maybe 3-4 night a year with seeing less than 1 arc second.  Then you might go for 0.6-0.7 (but the exposure time will need to be doubled).

 

So to answer this specific question, with the 294 (4.6um pixels), the ideal focal length for resolution will be not more than 1000mm unbinned (to keep about 0.9 arc-sec per pixel) and not more thsn 2000mm binned 2x2 (although that should be done in software for CMOS).    

 

Therefore an 8 inch at f5 is about ideal for this camera (MN180?).  F6 would be the absolute max I would recommend without spending a lot more on a mount and semi-permanent setup.  Even with that setup it will be hard to do more than 2min subs.    If the goal is lucky imaging to reduce the mount requirements then it is ideal to get to f/4 or under - either a RASA or imaging newt - but still not recommended for the ZEQ25.

Actually, my set up is usually 8" F/10 with the Meade F/6.3 reducer and maybe a spacer to bring it down to F/5, then in Sharpcap, I use 2x2 Binning and my FOV is around 0.54 x 0.38 degrees or roughly 3200" x 2000". So I can fit most DSO's in my FOV with a few exceptions.



#8 bigjy989

bigjy989

    Explorer 1

  • -----
  • Posts: 99
  • Joined: 23 Nov 2010
  • Loc: Saline, MI

Posted 15 September 2019 - 02:28 PM

Great.
If you can already get to f6 or so then the key is to ensure your have the exact spacing to eliminate coma and keep a flat fields over your sensor. Some reducers are better than others and sample to sample variation can be high. Focus is difficult and even low overshoot SCT still have enough to cause problems at critical focus.

The biggest challenge with a 0.5 degree fov is alignment, plate solving is almost mandatory - SGP is a big plus.

I started with a 8” SCT for imaging about 5 years ago and understand the challenges and desire to improve image quality. After a lot of trial and error I moved to a large 5.5” refractor reduced to F5 about 3 years ago and both integration time reduced and quality greatly improved. For most targets even up to a 1.5 ace-arc sampling doesn’t really detract from an image. Only planetary and galaxies really need the resolution but it is really difficult to get enough clear nights to get enough integration for a small sampling ratio - 0.7” is 2x 1.0” which is also 2x 1.4”. I’d much rather do one night at 1.4” than four at 0.7” fwiw.

Finally I put everything on a EQ6+ scope buggy and just roll it out when I when to image. Combined with SGP I have a 45 minute startup time and then 8-10 hrs later check it in the morning. If I keep the sampling between 1.1”and 1.5” then one night is enough for most images from an yellow - orange zone.

Edited by bigjy989, 15 September 2019 - 02:32 PM.


#9 Stelios

Stelios

    Cosmos

  • *****
  • Moderators
  • Posts: 7741
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 16 September 2019 - 01:48 AM

However, I just learned from another topic poster that the Meade ACF can use a regular Reducer Corrector, whereas the EDGE has to use a specifically manufactured reducer which costs up to 10X that price ($80 vs $800).

Huh??? The Edge HD reducer/flattener is $300, not $800.



#10 Stelios

Stelios

    Cosmos

  • *****
  • Moderators
  • Posts: 7741
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 16 September 2019 - 01:49 AM

It is not that binning does not benefit CMOS cameras. Binning with CMOS is currently just digital, so there is no reduction in read noise. There IS, however, still an increase in SNR. Bin 2x2, the combination of 4 samples, increase SNR 2x.

Doesn't just zooming out achieve the same effect? Without the loss of info that binning would? 



#11 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 16934
  • Joined: 27 Oct 2014

Posted 16 September 2019 - 02:03 AM

Doesn't just zooming out achieve the same effect? Without the loss of info that binning would? 

No.  The gain in snr only occurs if you bin.

 

This is one of those deals where people take something too far.  There's nothing wrong about binning CMOS, it works.  CMOS bins similarly to CCD, you just don't get the read noise advantage that you do with a CCD.  But CMOS tends to have lower read noise, so that's less important.

 

You can bin fairly effectively in software.  But it's not as simple as a digital zoom.

 

In the case of the original poster, he's doing very short exposures and total imaging time, what he does strongly resembles EAA.  In his case the increase in snr is very useful.


Edited by bobzeq25, 16 September 2019 - 02:08 AM.


#12 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23731
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 16 September 2019 - 12:47 PM

Doesn't just zooming out achieve the same effect? Without the loss of info that binning would? 

Do you mean downsample? Post-ADC binning or downsampling, the result is the same. It increases SNR. 

 

CCD binning has an ADDITIONAL read noise  benefit because it is charge binning (pre-ADC binning). Charge binning is not completely noise free, but the amp and ADC, which are done after charge combination, add most of the noise, so with charge binning there is an additional benefit that you only get the amp and ADC noise added once per N pixel samples (i.e. 2x2 binning, you would have one hit of read noise per four pixels, plus whatever noise is added during charge transfer).

 

Currently, CMOS cameras just combine the data post ADC, so it is done digitally. There is no additional benefit with read noise, but the standard improvement in SNR always applies. This goes for CCD or CMOS. 




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