Image size - Atik 420 vs Canon 550D

[ozboyz1]

Hi,

I came across this website http://www.skyatnightmagazine.com/field-view-calculatorand was puzzled with the image size produced by Atik 420 and Canon 550D. Please see attached for M104

I thought the canon 550D with almost 18MP would produce larger image than the Atik 420 with only 2MP. Could someone explain how this works?

Much appreciate it.

[rob0971]

It's the physical chip size that determines the image size not the amount of pixels on the chip. The DSLR shows a 2.65 deg fov while the 420 with its smaller chip only gives a fov of 51'05". The amount of pixels (along with their size) on the chip determine its resolution. This is probably a bit simplistic but its how I get my head around it.

[rob0971]

To make it even more simplistic, the bigger the physical size of the chip the smaller the image size (wider field of view) with a given focal length telescope.

[MattJenko]

Not quite same cameras, but close enough : http://astronomy.tools/calculators/field_of_view/?fov[]=3||37||1|1|0&fov[]=3||40||1|1|0&messier=104

[gnomus]

The screencapture that you attached does show that the Canon (~2.7 x 1.8 degrees) has a wider field of view (FOV) than the Atik (51 x 38 minutes).  There are 60 minutes in 1 degree.  As rob said, it is the physical size of the chip that will determine FOV.  The number of pixels on that chip will, of course, be dependent on the size of the pixels (4.4 microns for the Atik, 4.3 microns for the Canon).  The number of pixels will determine resolution, which you could think of as the amount of detail that is available.  

The resolution will come into play when you try to display or print the image.  You will get a larger print from the Canon.  But, you are not getting "more magnification" - because of the similar pixel size (and, therefore, resolution) it is just that the 420 is seeing less of the object than the Canon is.  

(Edit:  Ah .. cross-posted with Matt, whose image shows what I was trying to say more eloquently than I was able to say it.)

[ozboyz1]

Sorry that I'm still struggling to understand the relationship between FOV and image size given that their resolution are almost identical.

I thought FOV gives the area of the sky that can be viewed through the telescope or in in case, can be recorded on the camera sensor. Since the Canon has larger sensor than the Atik, it would cover larger area in the sky (limited by the scope's aperture of course) .... but the size of the object would be the same, given identical focal length.

Much appreciate if you give me some more info here. Thanks heaps 

[Pompey Monkey]

Sorry that I'm still struggling to understand the relationship between FOV and image size given that their resolution are almost identical.

I thought FOV gives the area of the sky that can be viewed through the telescope or in in case, can be recorded on the camera sensor. Since the Canon has larger sensor than the Atik, it would cover larger area in the sky (limited by the scope's aperture of course) .... but the size of the object would be the same, given identical focal length.

Much appreciate if you give me some more info here. Thanks heaps 

You are completely correct.

The confusing aspect of the view calculator is that the simulated galaxy images are completely incorrect in both image scale and field size.

[ngwillym]

M104 is approx 9x4 arc minutes and given that the Atik & the Canon are similar resolution per pixel, this means that the image of M104 will occupy the same/similar amount of pixels, regardless of which chip it is. all that will happen is that more of the surrounding area (i.e. the apparent field of view) will be captured in the Canon

[Uranium235]

Whichever camera has the smaller pixels will produce the "larger" image when both are cropped to the same dimensions (eg: 1200x800).

However, in a fistfight between a DLSR and a CCD, there is only one winner..... and it aint the DSLR

[johnrt]

The representation in the FOV calculator  shows both images as they would be unropped presented at the same size. So the DSLR version shows a small M104 as it is displaying all that surrounding space in the same size image on your monitor as the CCD.

If you crop down the DSLR image to show the same FOV then they would be very close matches - both displaying around 1.89 arcsec/pixel. It is just the uncropped representation in the simulator that gives the huge difference in size.

Of course the mono CCD would win in a fist fight for other reasons, but lets not confuse it too much!

[Chris]

Try both the cameras using this software: 

http://www.12dstring.me.uk/fovcalc.php

The outer box represents the field of view, so with the 420 the field of view is much less than the Canon's, so the small object 'appears' bigger.

[Magnus_e]

Remember that you can always download Stellarium to plan your shots. You can add camera sensor specs and lenses / telescopes and get the correct FOV.

You do have to add the scope / sensor data tough.

[Magnus_e]

First one was the 550D, This is the Atik (If i found the right specs)

[sharkmelley]

However, in a fistfight between a DLSR and a CCD, there is only one winner..... and it aint the DSLR

In a fistfight between a large sensor and a small sensor, there is only one winner.... and it ain't the small sensor

[Uranium235]

In a fistfight between a large sensor and a small sensor, there is only one winner.... and it ain't the small sensor

lol quite right! Perhaps consider a larger mono sensor then (8300).

But even a modestly sized CCD like the 314 still beats the 1000d I have, even though its chip is smaller you are getting much better data and it is not prone to chromatic noise - ie: blue/red blotches where it should be black, thats where the camera has tried to "guess" the colour (it becomes more apparent the more you stretch image data). With a mono CCD, you dont get that problem - and it opens up the lovely world of narrowband too

Edit: Not that im trying to open up a CCD V DSLR debate, that has been a well trodden path already!

[ozboyz1]

This is why I love this forum, all knowledgeable people with positive contribution 

On your remark about the narrowband Rob, what is the minimum exposure length for each colour? Can I try the narrowband filter with my DSLR first then move to CCD later? I have HEQ5 Pro mount, and I understand it has some limit on the long exposure ...

[Uranium235]

In regard to narrowband, the minimum exposure depends on the senistivity of the camera, the speed of the telescope and the brightness of the object in question. For instance, I can quite easily spend 20-30min per sub on something like M16 without blowing any part of the image, but for something like M42 - I can blow the core in as little as 30 seconds.

There is nothing to stop you from using the DSLR for narrowband, but its quite inefficient becuase youre only using one pixel in four (ie: red for Ha), and two in four (greenish for OIII). Plus the sensitivity of a DSLR is lower than a CCD, so you would need lots of long subs (probably 10min each as a minimum). Another problem is that for a DSLR you would need 2" NB filters, which are not cheap im afraid.

You havent mentioned what mount you are using though, as that will be your main limiting factor in regard to sub lengths. Doesnt matter if you have a fancy telescope or camera becuase if it all sits on an unsuitable mount, then you wont be able to reach the full potential of your camera.

[ozboyz1]

Rob, why can't I use 1.25" filter? thanks.

[Uranium235]

1.25" filters with a APS sized sensor (like in a DSLR) will make for a monster vignetting pattern that will prove impossible to correct, you would probably end up having to throw away half of your useable field. There would be no simple way of attaching it either, whilst a 2" would simply screw on to your telescopes corrector, reducer or 2" camera nosepiece without vignetting.

You can get clip-in Ha filters for Canons, but then youre locked into one system (you cant put a clip filter into a filter wheel).

[brrttpaul]

Im glad this topic has come up as I am sure there are many many beginners like my trying to decide whether to take the plunge and buy a CCD,  At the moment I have a 150 reflector but  probably getting a refractor soon (equinox or williams). What would be a good beginners CCD I have seen quite a lot and they vary from less than a £100 to several thousands, just a few suggestions on what a complete beginner should be looking at would be a great help

[ollypenrice]

To make it even more simplistic, the bigger the physical size of the chip the smaller the image size (wider field of view) with a given focal length telescope.

This is too simplistic to be correct. The size of an object on a chip depends only on the focal length of the scope. A larger chip receives an object image of exactly the same size as the same object's image on a smaller chip. The difference is that the larger chip will have more space around the object. Now your PC screen will default to make the image fit the screen so it will, if you like, expand the small chip's image to fill the screen and expand the large chip's image by much less to make that fit the screen as well. But if you zoom in then the large chip's object  can be presented at the same size on the screen.

It's a fairly common error to think that shrinking the chip magnifies the image. This, though, is like thinking that an empty loo roll can be used as a telescope. Restricting FOV and magnifying are not the same thing. 

Olly

[kalasinman]

Knowing that if I go CCD , I'll want to go NB too, and needing great cooling at 25c, a "beginning" CCD for me is a QHY9 mono.

[lensman57]

Sensor size has nothing to do with the size of the M104. it remains the same size regardless of what equipment the human being uses to observe or image it. If you use the same scope or rather more accurately the same FL to image it with different size sensors then the size of the sensor is irrelevant but what changes is the apparent FOV. The larger the sensor the more of the target will appear on the chip. larger sensor of the DSLR can not make the galaxy larger or smaller but your FL will alter your FOV.

A.G

[ozboyz1]

Many thanks Rob, never thought or known of such problem, I guess I was more worried about my pocket .... lol.  , that brings to the next question .... if I decided to go with CCD (let's say Atik 420), what filter size I should buy, 1.25" or 2", given the sensor size is much smaller ?

Thanks again Rob.

[lensman57]

Many thanks Rob, never thought or known of such problem, I guess I was more worried about my pocket .... lol.  , that brings to the next question .... if I decided to go with CCD (let's say Atik 420), what filter size I should buy, 1.25" or 2", given the sensor size is much smaller ?

Thanks again Rob.

Atik 420 is not the best choice for a CCD camera. The sensor is very quiet and clean but the sensitivity leaves a lot to be desired compared to the latest offerings. You can use 1.25" filters up to and including Atik 460.

A.G