Pixel scale Q - Forums

Topic starter

Aug 2, 2023

Hey all!

I was wondering about pixel scale. For example…

At my primary location, seeing + guiding seems to average between 2 and 2.5 ish. Given that… lets say 2.0/3 = max detail of 0.67ish? Does that mean that a scope with a longer reach won’t give any more detail than a shorter scope that can pull off 0.67 pixel scale?

thx!

andrea tasselli

Frequent contributor

Aug 2, 2023

That's correct if aperture is the same. Larger aperture at smaller pixel scale have however the potential to improve resolution in post-processing.

Sean Mc

Topic starter

Aug 2, 2023

Yes I understand that aperture = resolution. Makes sense. Basically what I’m looking at is a smaller aperture refractor vs a larger aperture sct. Obviously the sct is going to be a much longer focal length, but there’s a limit to seeing and guiding so I’m thinking “sharpness” won’t be better than a smaller aperture shorter focal length apo refractor in long exposure imaging?

For visual and lucky imaging the larger aperture would be better. But the question is about 600s exposures. I’ve looked at quite a few images from sct’s and really most aren’t as sharp as I’d like. I’m trying to figure out how to figure out what my “best” setup would be for my location and budget.
Pixel scale for seeing conditions should be a thing shouldn’t it?

Jaymz Bondurant

Aug 2, 2023

It sounds like your desire is to get as much focal length as possible while still obtaining a sharp image. In that case, it would be best to try to match your pixel scale as close to possible to your telescope's resolving power (solely dependent on the aperture). For example, my 8" scope can resolve up to .57". So, in your situation, I would want a pixel scale as close to .57"/px as possible without going over. If you went to, say, .45"/px, you would get a larger image, but it wouldn't have any more detail. Likewise, that's where you start to blur the image. That's not to say you can't achieve nice, sharp images when you go over. But if you want it as sharp as possible right out of the camera, I think that would be your best bet.

Actually, to clarify, you wanna stay a little below that. In my example, the telescope's resolving power is theoretical; basically meaning it can really only achieve that in the absence of an atmosphere. Whatever your scope's resolving power is, you'll want to stay slightly below that limit due to seeing.

andrea tasselli

Frequent contributor

Aug 3, 2023

Yes I understand that aperture = resolution. Makes sense. Basically what I’m looking at is a smaller aperture refractor vs a larger aperture sct. Obviously the sct is going to be a much longer focal length, but there’s a limit to seeing and guiding so I’m thinking “sharpness” won’t be better than a smaller aperture shorter focal length apo refractor in long exposure imaging?

For visual and lucky imaging the larger aperture would be better. But the question is about 600s exposures. I’ve looked at quite a few images from sct’s and really most aren’t as sharp as I’d like. I’m trying to figure out how to figure out what my “best” setup would be for my location and budget.
Pixel scale for seeing conditions should be a thing shouldn’t it?

***
It should and really is. I can't vouch for sct's but that is my experience with long integration.

AdrianC.

Aug 3, 2023

Hey all!

I was wondering about pixel scale. For example…

At my primary location, seeing + guiding seems to average between 2 and 2.5 ish. Given that… lets say 2.0/3 = max detail of 0.67ish? Does that mean that a scope with a longer reach won’t give any more detail than a shorter scope that can pull off 0.67 pixel scale?

thx!

I think this is how it works. The seeing will pretty much impose a limit on resolution, that is why professionals build scopes in good seeing mountain tops.

BUt there is another good advantage of a larger aperture. Even though your resolution won't improve, having a bigger photon collecting area will reduce imaging time. Depending on how many clear nights you have this will be important or not.

I was in the same situation just a few years ago. Like you I have 2 ish FWHM in my images, and I had to decide between a refractor and a SCT. My experience with a Meade 12 inch SCT was not great, too much tinkering and wasted clear nights, so I went the refractor route.

I had a bigger budget so I went crazy on a 8 inch apo refractor. It's a plug and play system and I am very happy with it. Refractors are simply easier to work with

I am working at 0.92"/pixel with the refractor, and with drizzle 2x at 0.4 ish. I did notice a small improvement drizzling but not earth shattering.

You can use this calculator to get an ideea how different systems will perform https://clearskies.go.ro/

Joe Linington

Aug 3, 2023

I run small pixels with a 102mm refractor and sample out at .84 arcseconds. An 8" newt would max me out at around .6 arcseconds. My seeing varies of course but is usually around 2 arcseconds with some nights being better and many nights worse. Everything has to be on point below 1 arcsecond to see the benefits. Guiding, focus and seeing all have to be spot on to enjoy the resolution benefits. Bigger pixels need bigger optics which need bigger mounts. It doesn't matter how you get to the seeing limits but when you get there just know that it starts to become a thin margin of error. The advantage of larger aperture and bigger pixels is time. They image faster. I trade time for cost. My system is thousands cheaper than other options with the same resolution but I need to integrate longer. I can bin2 my camera and get noticeably better SNR, faster but less detail. Like everything, pick your compromise.

Leela.Astro.Imaging

Aug 3, 2023 · Edited: Aug 3, 2023

Perhaps its subjective but I've shot the Crescent Nebula with a 4" f5 refractor and a 5" f15 Mak (the same camera 294MCPro both times, although with the Mak it was binned 2x2 so running at about 1"/pixel). For closer detail the Mak did better (for example I'd never noticed a small comma shaped asterism of stars before). So what I tend to do is choose the rig based on the size of the target. If it's small and deep, I'll tend to go for the Mak.

[EDIT: My seeing is also around 2-3" FWHM, sometimes 3-4"]

Willem Jan Drijfhout

Aug 3, 2023

Yes I understand that aperture = resolution. Makes sense. Basically what I’m looking at is a smaller aperture refractor vs a larger aperture sct. Obviously the sct is going to be a much longer focal length, but there’s a limit to seeing and guiding so I’m thinking “sharpness” won’t be better than a smaller aperture shorter focal length apo refractor in long exposure imaging?

For visual and lucky imaging the larger aperture would be better. But the question is about 600s exposures. I’ve looked at quite a few images from sct’s and really most aren’t as sharp as I’d like. I’m trying to figure out how to figure out what my “best” setup would be for my location and budget.
Pixel scale for seeing conditions should be a thing shouldn’t it?

All plays a role for sure (pixel scale, seeing, aperture, etc). And so is FoV, or which target do you want to go for. The smaller refractor may be better suited for wide(r)field images, while the sct may be better suited for small galaxies and planetary nebulae. Of course sensor-size plays a role in this as well.
If you’re looking for the most versatile option, a 1000mm-ish refractor or newton are great options.
If you’re looking for smaller object targets, a longer reflector could be a good option. But be aware that it can quickly get a snowball effect. To keep exposure times in check you want to add aperture, which adds weight, which is more demanding for mount, etc, etc.

What is ‘best’, always depends on purpose and budget. And ‘best’ is usually determined by the weakest link in the setup. E.g. a premium mount can often add more sharpness to an image than a perfectly matched pixel scale.

vercastro

Aug 4, 2023 · Edited: Aug 4, 2023

Considering camera pixel size when making camera purchasing decisions is not as relevant as in the past. That's mostly because all the best modern sensors have the same pixel size (3.76 micron).

Instead it's better to focus on focal length. It's already been mentioned that under average seeing conditions, 1000mm is a good rule of thumb to stick to these days considering the common pixel size.

How you achieve this focal length is up to you. You can go have a smaller aperture with a slower focal ratio, or you can go big and fast.

However, for a variety of reasons (some to do with software), it is sometimes still worth going for longer focal length even under average seeing. It's not as black and black and white as you might think. And no I'm not talking about lucky imaging.

Something else that's important to understand is how a telescope with a bigger aperture and a larger camera captures more light than a smaller telescope with a smaller camera (assuming the same focal ratio). Yet both systems have the same field of view and the same size pixels (one has more pixels). If the bigger telescope/camera is blurred by seeing, you can simply down sample the image to increase SNR. However if you have a lucky night of seeing, those sub frames can be weighted higher and you gain detail that you simply don't have with the smaller telescope. BlurXExterminator is also an elephant in this proverbial room.

There really isn't a replacement for a bigger aperture, but modern techniques can certainly help to shrink the gap.

John Hayes

Frequent contributor

Aug 5, 2023

Understanding how to optimize pixel size for any given scope taking into account the local seeing conditions can be a little complicated. I outlined how image quality is determined by the MTF of an imaging system, which in turn is determined by the optics, the sensor, and the atmosphere in this presentation:

https://www.advancedimagingconference.com/articles/secrets-long-focal-length-imaging-john-hayes

(The AIC library is open to the public. All you have to do is to register at the AIC site.)

Although the discussion might go through a little more optics than you might like, I distilled the results down to a couple of charts to make it very easy to determine the optimum sampling interval for your scope and average seeing conditions (@ 40:28). I also provided the slides so that you didn’t have to dig through the presentation to find this stuff but it looks like the link to the slides is broken. Either way, if you can find time to check out this presentation, I think that you’ll leave with a little better understanding of how this stuff works.

John

AdrianC.

Aug 5, 2023

John Hayes:
Understanding how to optimize pixel size for any given scope taking into account the local seeing conditions can be a little complicated. I outlined how image quality is determined by the MTF of an imaging system, which in turn is determined by the optics, the sensor, and the atmosphere in this presentation:

https://www.advancedimagingconference.com/articles/secrets-long-focal-length-imaging-john-hayes

(The AIC library is open to the public. All you have to do is to register at the AIC site.)

Although the discussion might go through a little more optics than you might like, I distilled the results down to a couple of charts to make it very easy to determine the optimum sampling interval for your scope and average seeing conditions (@ 40:28). I also provided the slides so that you didn’t have to dig through the presentation to find this stuff but it looks like the link to the slides is broken. Either way, if you can find time to check out this presentation, I think that you’ll leave with a little better understanding of how this stuff works.

John

Really nice charts!. To my surprise the formula I used to match my scope and camera, agrees very well with the value on the chart, at least for my scope.
I can't remember where I found it, but this is an extract from one of my python scripts:

pixelScopeMatch = (1/nyquistCriterium)*meanFWHM/(0.206265/(st.session_state['focalLength']/1000))

Where image FWHM is computed as per this ESO article : Seeing and FWHM
And the python code for it in case anyone needs it:

def imageQuality(seeing, aperture, imageScale, band, angle, guideRMS,L0):
    #L0 wavefront outer scale
    waveLength = band_selection(band)[0]
    airmass = flux(20, float(angle), band)[1]
    
    kolbFactor = (1/(1+300*(aperture/L0))) - 1 
    friedParameter = 0.1 * (seeing**-1) * (waveLength/500)**1.2 * airmass**-0.6
        
    telescopeFWHM = 0.000212 * waveLength / aperture
    atmosphereFWHM = seeing * (airmass**0.6) * ((waveLength/500)**-0.2) * np.sqrt(1+kolbFactor * 2.183 * (friedParameter/L0)**0.356)
    guidingFWHM = guideRMS * 2 * np.sqrt(2 * np.log(2))
        
    finalFWHMrms = np.round(np.sqrt(telescopeFWHM**2 + atmosphereFWHM**2 + imageScale**2 + guidingFWHM**2),3)
    

    return finalFWHMrms

As a side note, calculating the telescope-camera system entendue can also be helpful, but I think one should be careful when comparing two very different systems just using entendue. For example I presume the Vera Rubin telescope has a higher entendue then JWST but they have very different goals/targets, so comparing them like this doesn't make much sense. One is a wide field - survey scope, the other one is a high resolution power horse.
But comparing the JWST with Hubble in terms of entendue makes more sense.