Aperture an Pixel Scale

[Rodd]

I was fooling around with a FOV calculator and found that my 4" refractor and ASI 1600 camera and a C11 Hyperstar with the same camera provide identical FOVs and an identical pixel scale, which is 1.46 arcsec/pix.  This got me thinking about resolution.  Correct me if I am wrong, but in this case, the only difference between the two systems would be speed.   In this case, resolution is controlled by the pixel scale.  IF the refractor is able to image at a resolution of 1.46 arcsec/pix (and I know it is), then the same amount of details will be in the images.  The hyperstar image will be brighter--it will have a greater SNR, but the smallest detail visible will be the same--for both systems the smallest feature that could be viewed would be 1.46 arcsecs in size.  It will take longer to reveal it with the refractor, but in the end, the resolution will be the same.

Now--lets take it one step further--lets use a 5" refractor with a pixel scale of 1 arcsec/pix.  Forget the FOV because we can crop the hyperstar image to be the same size.  Based on the pixel scale, finer details aka more resolution would be provided with the 5" scope--not the 11 inch scope using hyperstar.

I always thought that a bigger aperture would provide higher resolution....but this is not always the case with hyperstar.  And, since the 11" scope is affected by poor seeing, tube currents, and wind gusts--in actuality, the 4" refractor just might yield a higher resolution image.

Please let me know if the above is, as they say across the pond, Tosh

[vlaiv]

Have no idea what Tosh means :D, but I have two observations:

1. Hyper star is not diffraction limited system - far from it, it has very large spot diagram - I think that 1.46"/px for hyper star is over sampling.

2. I'm not sure if 5" refractor is able to achieve 1"/px in regular conditions.

In the end, I'd like to add that I agree with you on 5" vs 11" hyper star - just try to make planetary image with 11" hyper star and you'll see how easily it can be out resolved by 5" refractor.

 

[The Lazy Astronomer]

2 hours ago, vlaiv said:

Have no idea what Tosh means

Tosh: rubbish, nonsense - see also: codswallop 😉

[Rodd]

1 hour ago, The Lazy Astronomer said:

Tosh: rubbish, nonsense - see also: codswallop 😉

For a while I thought the post was going to be edited as I was under the impression the gosh meant BS.  I am glad I was wrong

[Rodd]

4 hours ago, vlaiv said:

Have no idea what Tosh means :D, but I have two observations:

1. Hyper star is not diffraction limited system - far from it, it has very large spot diagram - I think that 1.46"/px for hyper star is over sampling.

2. I'm not sure if 5" refractor is able to achieve 1"/px in regular conditions.

In the end, I'd like to add that I agree with you on 5" vs 11" hyper star - just try to make planetary image with 11" hyper star and you'll see how easily it can be out resolved by 5" refractor.

 

That’s true. Though I think established hyper star efficianados use hyper star with smaller pixel cameras.  Not sure. Maybe because that is what is available.  When the scope is a c14 oversampling seems all but guaranteed.  I guess using hyper star to collect wider field shots and cropping stand alone features like I like to do is not really appropriate for hyperstsr  

Edited July 24, 2021 by Rodd

[vlaiv]

6 hours ago, Rodd said:

That’s true. Though I think established hyper star efficianados use hyper star with smaller pixel cameras.  Not sure. Maybe because that is what is available.  When the scope is a c14 oversampling seems all but guaranteed.  I guess using hyper star to collect wider field shots and cropping stand alone features like I like to do is not really appropriate for hyperstsr  

Here is interesting reference:

Quote

With good guiding and careful focus, the HyperStar produces tight, round stars over a large field. With my C14, I typically see minimum FWHM star images in the range of 2 pixels in a stacked image, which translates to 12.8 microns or 3.9 arc-­‐sec with a Canon 6D, which has 6.4 micron pixels.

taken from here: https://www.cloudynights.com/articles/cat/articles/the-amazing-hyperstar-a-guide-to-optimize-perf-r3013

We have 3.9" FWHM stars on 14" telescope - ideal pixel size being 7.8µm (2.43"/px).

[Rodd]

28 minutes ago, vlaiv said:

Here is interesting reference:

taken from here: https://www.cloudynights.com/articles/cat/articles/the-amazing-hyperstar-a-guide-to-optimize-perf-r3013

We have 3.9" FWHM stars on 14" telescope - ideal pixel size being 7.8µm (2.43"/px).

So that is exactly what my 4” refractor is with .6x reducer and 3.8 um pixels.  3.4”/pix

[ollypenrice]

All of these number-based discussions must suppose that the system is diffraction limited. As the Hyperstar isn't, we cannot compare it easily with a system which is. Personally I've never bothered to look at the Hyperstar resolution theoretically because just looking at the pictures tells me I don't want one.  There are, to be fair, tight NB images done through good filters with the larger aperture Hyperstars but I'm not a fan of broadband in this instrument.

We can easily extract the maximum possible resolution for a 5 inch refractor from its Dawes limit. It's 0.97 arcsecs. What is not clear (to me at least) is whether or not this will support a pixel scale of 0.97".  I wonder what Vlaiv thinks?

We can be sure, though, that an instrument with a Dawes limit greater than its pixel scale cannot support that pixel scale. 

All I can really say is that I'm happy using my TEC 140 (Dawes limit 0.8") while sampling at O.9". I'm fairly convinced that the system is seeing-limited.

Olly

[vlaiv]

9 minutes ago, ollypenrice said:

What is not clear (to me at least) is whether or not this will support a pixel scale of 0.97".  I wonder what Vlaiv thinks?

We can't simply equate these two as Dawes limit is an approximation (empirical).

What we can do is use well established formulae.

5in telescope has 127mm of aperture and for that optimum sampling rate for planetary case is 0.41"/px. We have very exact solution to this and it is based on:

https://en.wikipedia.org/wiki/Spatial_cutoff_frequency

combined with Nyquist criteria for sampling of band limited signal.

However, this is case where there is no atmospheric influence.

For long exposure images - it much more depends on sky conditions, mount tracking and other factors. There is simple formula in that case as well and that is FWHM / 1.6.

If your system produces FWHM of 3.9" (like above hyperstar) then you need to sample it at 2.4375"/px. Going in opposite direction - if you sample at 0.9"/px - you better have 1.44" FWHM stars in your image. In my experience - that is not likely in regular seeing that by itself is 1.4" or more (without added optics and mount performance).

 

[Rodd]

It seems to me that it is always seeing dependent for most people.  I would be willing to bet that most systems do not operate at diffraction limited levels due to sky conditions.  So talking about whether a system is diffraction limited is like talking about the maximum speed of a Bugatti on a super windy mountain road.  The car will not travel at its maximum speed because of the road.  

[vlaiv]

19 minutes ago, Rodd said:

It seems to me that it is always seeing dependent for most people.  I would be willing to bet that most systems do not operate at diffraction limited levels due to sky conditions.  So talking about whether a system is diffraction limited is like talking about the maximum speed of a Bugatti on a super windy mountain road.  The car will not travel at its maximum speed because of the road.  

It is sort of like that with small difference - optical blur adds up.

If you take two otherwise identical scopes - one diffraction limited and other having say bad spherical aberration that makes it not be diffraction limited (spherical greater than 1/4th wave) - and place those on the same mount and under same skies (say side by side setup) - one that is diffraction limited will produce tighter stars - smaller FWHM.

 

[Rodd]

5 minutes ago, vlaiv said:

It is sort of like that with small difference - optical blur adds up.

If you take two otherwise identical scopes - one diffraction limited and other having say bad spherical aberration that makes it not be diffraction limited (spherical greater than 1/4th wave) - and place those on the same mount and under same skies (say side by side setup) - one that is diffraction limited will produce tighter stars - smaller FWHM.

 

I see that.  All I am saying is diffraction limited as a benchmark is a bit unrealistic. So back to the original comparison: a 4.2”refractor shooting at 2.46 “/pix and a hyperstar 11( or 14) shooting at the same resolution.  I have 2 questions.  1) will the full FOV (assume same FOV) have more detail in the hyperstar, and 2) will the hyperstar image allow for close in crops of interesting features like I do with the refractor

[ollypenrice]

6 minutes ago, Rodd said:

I see that.  All I am saying is diffraction limited as a benchmark is a bit unrealistic. So back to the original comparison: a 4.2”refractor shooting at 2.46 “/pix and a hyperstar 11( or 14) shooting at the same resolution.  I have 2 questions.  1) will the full FOV (assume same FOV) have more detail in the hyperstar, and 2) will the hyperstar image allow for close in crops of interesting features like I do with the refractor

Out of interest, why the Hyperstar? Why not the RASA, which is optically better?

Olly

[vlaiv]

20 minutes ago, Rodd said:

1) will the full FOV (assume same FOV) have more detail in the hyperstar

This depends on

a) What you mean by more detail

b) if you assume detail in terms of resolution - then it will depend on respective star FWHM. Telescope with smaller FWHM will produce sharper / more detailed image. If by detail you assume "depth" of image or achieved SNR - then HyperStar will win for given imaging time - it will be faster system as it has more aperture on same resolution

12 minutes ago, ollypenrice said:

2) will the hyperstar image allow for close in crops of interesting features like I do with the refractor

If they both sample at the same sampling rate / resolution - then yes, there will be no difference in terms of crops - you can crop the same region and it will have the same size in pixels

 

[Rodd]

3 minutes ago, vlaiv said:

This depends on

a) What you mean by more detail

b) if you assume detail in terms of resolution - then it will depend on respective star FWHM. Telescope with smaller FWHM will produce sharper / more detailed image. If by detail you assume "depth" of image or achieved SNR - then HyperStar will win for given imaging time - it will be faster system as it has more aperture on same resolution

If they both sample at the same sampling rate / resolution - then yes, there will be no difference in terms of crops - you can crop the same region and it will have the same size in pixels

 

I know that.  I am talking about quality.  What I am trying to understand is will the only difference be one of speed?  The adage is aperture controls resolution and detail.  But not in this case

[Rodd]

28 minutes ago, ollypenrice said:

Out of interest, why the Hyperstar? Why not the RASA, which is optically better?

Olly

I am just curious.  I use hyperstar in the discussion because 1) it is commonly used and several highly respected imagers use it at the precise resolution I ask about. It is a real comparison. Using the RASA in this discussion would be more conjecture as I am not familiar with anyone who uses it. 2) my c11edge can be equipped with hyper star ( I understand Version 4 is better than 3), so there is at least  some potential for making a switch if warranted. 

[vlaiv]

2 minutes ago, Rodd said:

 The adage is aperture controls resolution and detail.  But not in this case

Yes, aperture controls resolution when the scope is diffraction limited.  That is always true if we don't include effects of atmosphere - like when we do planetary imaging at critical sampling rate.

Most scopes are diffraction limited on axis - even cheap Chinese mass produced telescopes. However, some astrographs are not diffraction limited.

Why is that? Because they are astrographs and it is much more important that they have large corrected field then diffraction limited field and because they will be used in long exposure where seeing can mask effects of diffraction limit on scope.

If you are going for wide field of view image that you sample at say 3"/px, is it important for your telescope to have spot diagram that fits inside 1.28" (airy disk diameter of 8" scope), or can you have one that maybe fits inside 2.5" circle. Both will be smaller than single pixel. Such wide field scope needs not be diffraction limited.

 

[Rodd]

49 minutes ago, vlaiv said:

Yes, aperture controls resolution when the scope is diffraction limited.  That is always true if we don't include effects of atmosphere - like when we do planetary imaging at critical sampling rate.

Most scopes are diffraction limited on axis - even cheap Chinese mass produced telescopes. However, some astrographs are not diffraction limited.

Why is that? Because they are astrographs and it is much more important that they have large corrected field then diffraction limited field and because they will be used in long exposure where seeing can mask effects of diffraction limit on scope.

If you are going for wide field of view image that you sample at say 3"/px, is it important for your telescope to have spot diagram that fits inside 1.28" (airy disk diameter of 8" scope), or can you have one that maybe fits inside 2.5" circle. Both will be smaller than single pixel. Such wide field scope needs not be diffraction limited.

 

But crops arc different.  If seeing is 2.5 arcsec and a refractor shooting at 2.46”/pix is composed to a hyperstar also shooting at 2.46”/pix, and you crop out a galaxy-which will contain more details?

[ollypenrice]

1 hour ago, Rodd said:

But crops arc different.  If seeing is 2.5 arcsec and a refractor shooting at 2.46”/pix is composed to a hyperstar also shooting at 2.46”/pix, and you crop out a galaxy-which will contain more details?

Only a test will tell you. I do believe that. Theory is fine but it needs perfect data to be correct. The story doesn't end with the capture, either. What processing will the capture allow? That matters.

Olly

Edited July 24, 2021 by ollypenrice

[Rodd]

1 hour ago, ollypenrice said:

Only a test will tell you. I do believe that. Theory is fine but it needs perfect data to be correct. The story doesn't end with the capture, either. What processing will the capture allow? That matters.

Olly

True.  I guess I would compare raw stacks prior to any processing.  Better yet, if the clouds would go away I could forget about such endeavors and get back to work