Imaging smaller DSO with a barlow query

[bomberbaz]

Hello all.

I am looking to image smaller DSO, mainly planetary nebula.

I have read a few threads on here (none posted recent) regarding using barlow and generally the answer is no, don't bother but with a few exceptions.

My camera is a ZWO 183mc pro (pixel 2.4) and is mated with a F6 60ed frac. Is it viable to use a barlow on small bright nebula eg M57 ring or is it quicker and more efficient going without and cropping the image?

pro and cons pleaase

Thanks for the help

Steve

 

Edited May 19, 2022 by bomberbaz

[vlaiv]

There is only so much resolution that you can achieve with 60mm scope and you are already over sampling with 2.4um pixel size (probably even when using super pixel mode).

No benefit what so ever in using barlow lens - effect will be the same as when you take regular image without barlow and enlarge it by factor of x2 or whatever barlow you were planning on using.

 

[bomberbaz]

10 minutes ago, vlaiv said:

There is only so much resolution that you can achieve with 60mm scope and you are already over sampling with 2.4um pixel size (probably even when using super pixel mode).

No benefit what so ever in using barlow lens - effect will be the same as when you take regular image without barlow and enlarge it by factor of x2 or whatever barlow you were planning on using.

 

Thanks @vlaiv, one other query, will binning not help with the over sampling or am I getting it wrong?

[vlaiv]

3 hours ago, bomberbaz said:

Thanks @vlaiv, one other query, will binning not help with the over sampling or am I getting it wrong?

Binning does help.

It is just matter of understanding how it is all connected together. I'll try to summarize it into few simple / straight forward points (that we can expand on if you feel the need).

- There is limit to what can be achieved in terms of real resolution of the image - it depends on aperture size, seeing and mount performance, and is expressed in arc seconds per pixel

- Any object has finite size - certain number of arc seconds that it covers / spans.

- From two above points follows that any object that you record will have certain size in pixels.

For example, with 60mm scope and AZ-GTI in regular conditions I would expect 2.5"/px - 3"/px to be realistic resolution of the image.

M57 is about 1'40" at its longest extent. That translates into 100". With 2.5"/px - that is 40px. That is realistic size of that object that you can fully resolve with 60mm scope.

Now, you can use a barlow to make it larger - but it is the same as taking those 40px and enlarging image to make object be say 100px across.

In fact - you can do all sorts of manipulations - you can bin your data, you can use barlow, you can drizzle, you can enlarge / crop - whatever you like - but that won't change maximum resolution of that image which is equivalent to about 40px for object.

Only way to get more detailed image of the object is to get larger aperture scope and better mount and shoot in better seeing.

Even then - you can't expect miracles. Most amateur setups can't go beyond 1"/px (even that is very tall order in 99% of cases).

That does not mean that you can't image at something like 0.5"/px - you can and many people do - it just means that you won't capture any more detail than if you imaged at say 1.5"/px or in best circumstances up to 1"/px.

To put things into perspective - this is what you can hope to achieve with say 8" scope, steady skies and good mount:

That is it.

Most images of M57 that you'll see that were taken by amateur astronomers will:

1) be less detailed

2) will probably be larger by that (be much blurrier and thus oversampled)

For reference, here is my M57 from quite some time ago (it was taken with 5" scope, and is probably far cry from proper image in terms of processing):

Like I said - larger and less detail

On the other hand - if you shoot with 60mm scope and x2 - this is what you can expect to get in ideal circumstances:

while actual image at that resolution (fully resolved) looks like this:

I hope above makes sense.

Bottom line - you can make your target look larger by number of ways - adding barlow, drizzling, enlarging in software - but all of that won't bring in detail as you are ultimately limited by possible resolution of your setup + skies.

 

 

 

 

[ollypenrice]

I agree with Vlaiv but would want to add some thoughts on cropping. I often feel that, when folks talk about cropping, they think they are 'doing something.' But that isn't really true.  To crop is to do very little. Your uncropped image has a certain resolution of detail. If you are neither over nor under sampled the best way to see this resolution is to view the image at 100%, which means 1 camera pixel is given 1 screen pixel. The trouble is, your screen is unlikely to be big enough to look at the whole image this way, so you have to drag it around to see different parts of it at any one time. When you crop, you are just choosing which bit of your image you are going to look at yourself or present to the public. All you are saying is, 'my screen isn't big enough so let's just look at this bit.' But you have done nothing whatever to the bit in question by cropping around it. It is exactly as it always was and is entirely genuine and unadulterated.

Forgive me if you already know this, but I know from experience that many people don't. They feel that cropping is somehow a form of cheating, of inventing resolution. This simply isn't so.

Olly

[bomberbaz]

Thank you @vlaiv for your detailed explanation. That is much clearer to me now but still something is missing in my mind.

I am still not 100% on the physics side of it as in why the barlow at x2 doesn't double the pixels because the way my mind thinks, I am thinking you are doubling the image size on the sensor. Or perhaps I am mixing up different aspects of the image process. 

I will re-read this tomorrow after a good nights sleep.

Thanks again

Steve

[vlaiv]

3 minutes ago, bomberbaz said:

I am still not 100% on the physics side of it as in why the barlow at x2 doesn't double the pixels because the way my mind thinks, I am thinking you are doubling the image size on the sensor. Or perhaps I am mixing up different aspects of the image process. 

It doubles the number of used pixels, but those pixels won't include new / valuable information.

It is really like visual astronomy. You can always add a barlow to scope / eyepiece combination - but if you already have short focal length eyepiece - you will get larger darker image without additional detail.

This is the same thing - you already have very small pixels compared to what scope can deliver (it's like using 2mm eyepiece) and you want to add barlow to that.

[Pitch Black Skies]

9 hours ago, vlaiv said:

you are already over sampling with 2.4um pixel size

How do you calculate this vlaiv?

I know you have issues with the credibility of this calculator but I always use it as I don't know any better.

This is what I get when using it;

[bomberbaz]

9 hours ago, vlaiv said:

It doubles the number of used pixels, but those pixels won't include new / valuable information.

It is really like visual astronomy. You can always add a barlow to scope / eyepiece combination - but if you already have short focal length eyepiece - you will get larger darker image without additional detail.

This is the same thing - you already have very small pixels compared to what scope can deliver (it's like using 2mm eyepiece) and you want to add barlow to that.

I get it now, I was looking at this from wrong angle entirely.

It is what goes in at the front that counts, you can mess around all you like at the other end but that fundamental factor remains unchanged. 

Thanks @vlaiv, you help is greatly appreciated. 

[ollypenrice]

19 minutes ago, bomberbaz said:

I get it now, I was looking at this from wrong angle entirely.

It is what goes in at the front that counts, you can mess around all you like at the other end but that fundamental factor remains unchanged. 

Thanks @vlaiv, you help is greatly appreciated. 

An everyday term for what Vlaiv's describing is 'empty resolution.' The image becomes bigger but has no new information.

Olly

[vlaiv]

11 hours ago, Pitch Black Skies said:

How do you calculate this vlaiv?

Thread where I question accuracy of astronomy.tools contains mathematical details, but it boils down to this:

- when your hands shake - you get blurry picture. Same is with mount - if mount does not track perfectly and none does - there is a bit of blur that comes from that. Guide RMS is measure of how accurate tracking is - and that translates into blur that happens due to mount inaccurate tracking

- seeing creates blur on the image - again there is measure of that (seeing FWHM)

- aperture size limits how much can be resolved - it too creates blur. There is also known way how to calculate it - planetary imagers do it all the time as there is no point going above certain F/ratio (for given pixel size)

Mathematics describes how those three blurs combine to form single blur (convolution).

If we know all those parameters - we can calculate expected level of blur / FWHM and consequently needed sampling rate to capture it properly.

In above estimate I used 60mm of aperture, 1.5" RMS Guide error for AZ-GTI and seeing in 1.5-2" FWHM range (all expected conditions).